12970 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 This journal is c the Owner Societies 2012

Cite this: Phys. Chem. Chem. Phys., 2012, 14, 12970–12981

Order and disorder effects in nano-ZrO2 investigated by micro-Raman

and spectrally and temporarily resolved photoluminescencew

Carmen Tiseanu,*aBogdan Cojocaru,

bVasile I. Parvulescu,

b

Margarita Sanchez-Dominguez,cd

Philipp A. Primuseand Magali Boutonnet

f

Received 11th June 2012, Accepted 23rd July 2012

DOI: 10.1039/c2cp41946g

Pure and europium (Eu3+) doped ZrO2 synthesized by an oil-in-water microemulsion reaction

method were investigated by in situ and ex situ X-ray diffraction (XRD), ex situ Raman

spectroscopy, high-resolution transmission electron microscopy (HRTEM), steady state and

time-resolved photoluminescence (PL) spectroscopies. Based on the Raman spectra excited at

three different wavelengths i.e. 488, 514 and 633 nm and measured in the spectral range of

150–4000 cm�1 the correlation between the phonon spectra of ZrO2 and luminescence of

europium is clearly evidenced. The PL investigations span a variety of steady-state and time

resolved measurements recorded either after direct excitation of the Eu3+ f–f transitions or

indirect excitation into UV charge-transfer bands. After annealing at 500 1C, the overall Eu3+

emission is dominated by Eu3+ located in tetragonal symmetry lattice sites with a crystal-field

splitting of the 5D0–7F1 emission of 20 cm�1. Annealing of ZrO2 at 1000 1C leads to a

superposition of Eu3+ emissions from tetragonal and monoclinic lattice sites with monoclinic

crystal-field splitting of 200 cm�1 for the 5D0–7F1 transition. At all temperatures, a non-negligible

amorphous/disordered content is also measured and determined to be of monoclinic nature.

It was found that the evolutions with calcination temperature of the average PL lifetimes

corresponding to europium emission in the tetragonal and monoclinic sites and the monoclinic

phase content of the Eu3+ doped ZrO2 samples follow a similar trend. By use of specific

excitation conditions, the distribution of europium on the amorphous/disordered surface or

ordered/crystalline sites can be identified and related to the phase content of zirconia. The role

of zirconia host as a sensitizer for the europium PL is also discussed in both tetragonal and

monoclinic phases.

1. Introduction

At ambient pressure, pure zirconia (ZrO2) has a cubic structure at

high-temperatures, between 1167 1C and 2367 1C it is tetragonal,

and below 1440 1C it is monoclinic. Upon doping with

trivalent cations such as Y3+ or lanthanides (Ln3+), the cubic

and tetragonal phases can be stabilized at lower or even

ambient temperatures.1–10 The anion (oxygen) vacancies sit

on nearest-neighbour (NN) or next-nearest-neighbour (NNN)

sites with respect to the dopant, leading to seven (monoclinic

symmetry) or eight-fold coordination (tetragonal symmetry)

of the dopant, respectively.2 This has a strong impact on the

local coordination and bond lengths of the trivalent dopant

in the zirconia lattice, as the Y3+/Ln3+-oxygen polyhedra

increase their symmetry from the low, C2, (Cs or C1) symmetry

in seven-fold coordination to the more symmetrical eight-fold

coordination with tetragonal, D2d (or D4h), symmetry. The

main structural difference between these zirconia phases is

given by the displacements of the lattice oxygen atoms, there-

fore consideration of the local atomic structure is essential to

aNational Institute for Laser, Plasma and Radiation Physics,P.O. Box MG-36, RO 76900, Bucharest-Magurele, Romania.E-mail: tiseanuc@yahoo.com

bUniversity of Bucharest, Department of Chemical Technology andCatalysis 4 – 12 Regina Elisabeta Bvd., Bucharest 030016, Romania

c Instituto de Quımica Avanzada de Cataluna, Consejo Superior deInvestigaciones Cientıficas (IQAC-CSIC), CIBER en Biotecnologıa,Biomateriales y Nanomedicina (CIBER BBN), Jordi Girona 18-26,08034 Barcelona, Spain

dCentro de Investigacion en Materiales Avanzados, S. C. (CIMAV),Unidad Monterrey; GENES-Group of Embedded Nanomaterials forEnergy Scavenging, Alianza Norte 202, Parque de Investigacion eInnovacion Tecnologica, 66600 Apodaca, Nuevo Leon, Mexico

e Institute of Chemistry, Physical Chemistry, University of Potsdam,Karl-Liebknecht-Str. 24-25, 14476 Potsdam-Golm, Germany

f Kungliga Tekniska Hogskolan (KTH), School of Chemistry,Div. of Chemical Technology, Teknikringen 42, SE-10044,Stockholm, Swedenw Electronic supplementary information (ESI) available: STEM pictures,TGA andDTA curves, textural characterization details and luminescenceexcitation/emission spectra. See DOI: 10.1039/c2cp41946g

PCCP Dynamic Article Links

www.rsc.org/pccp PAPER

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

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

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 12971

fully understand the effect of doping on ZrO2.9 Raman

spectroscopy is intensively used to describe the local structure

of doped zirconia. Raman spectra are locally sensitive to

Zr4+–O bonds in the first (0.1–0.5 nm) and second (0.5–5 nm)

atomic shells and also to the lattice defects. In contrast, the

long-range sensitive conventional X-ray diffraction (XRD) is

able to detect cation order–disorder effects as well as lattice

distortion, but it is less useful when investigating the distor-

tions related to the oxygen sublattice and/or lattice defects.

This is due to the weak scattering factor of oxygen relative to

the heavier Zr4+ cations. The trivalent dopant distribution

between surface, amorphous/disordered, tetragonal or mono-

clinic phases can be discriminated effectively by using the Eu3+

as a local probe. The luminescence properties of europium are

very sensitive to changes of the first coordination sphere, i.e.

Eu3+ shows distinct changes in the emission/excitation spectra

and excited state dynamics with changes in the local symmetry

of the ZrO2 lattice surrounding Eu3+ ions. Several groups have

investigated the luminescence–structure relationships in Eu3+

doped ZrO2 with regard to the site symmetry (phase), dopant

concentration, thermal treatment, surface coating, nanoparticles

shape and size and so on.11–25 It is established that in the ZrO2

polymorphs, the dopant site symmetry plays a major role in the

radiative and nonradiative relaxation mechanisms. It is also

recognized that the emission of Eu3+ located in the tetragonal

sites of ZrO2 displays two lines with relatively close intensities,

one at B590 nm (corresponding to the unresolved two allowed5D0–

7F1 lines) and another one at B606 nm (which is the

strongest among the four allowed 5D0–7F2 lines). Ideally, the

5D0–7F0 related emission should be absent, in accordance with

electric and magnetic dipole transition rules for the D2d sym-

metry. For the monoclinic symmetry, the emission structure of

europium is complex.20 Generally, the emission of Eu3+ is

dominated by the 5D0–7F2 emission between 611 and 618 nm

and one or several weak 5D0–7F0 related emission lines are

observed in the spectral range of 577–580 nm. When high-

resolution emission spectra were reported, a three line structure

corresponding to the 5D0–7F1 is detected with a distinctive

doublet at B597 and 598 nm which may be considered as the

fingerprint for the ordered monoclinic type emission of Eu3+ in

ZrO2. For sol–gel undoped ZrO2, it was shown that for an

initially amorphous sample which crystallizes upon thermal

treatment to the tetragonal phase, a large fraction remains

amorphous and the amorphous/disordered content is of mono-

clinic nature.10 Due to the broad distribution of Eu3+–O2�

environments induced by oxygen vacancies randomly distri-

buted in the anion sublattice,7,20 Eu3+ emission in a multi-phase

ZrO2 has complex emission properties with strongly overlapping

broad and narrow spectral features.

Here, we propose a thorough understanding of the local

structure of Eu3+-doped ZrO2 by use of Raman, HRTEM,

XRD (in situ and ex situ), steady-state and time-resolved PL

spectroscopy. Characteristic Eu3+ emission was also extracted

from the Raman spectra by extending the recorded spectral

range from 150–800 cm�1, where the typical phonon modes of

ZrO2 polymorphs lie, up to 4000 cm�1. To discriminate the

emission of intentionally doped Eu3+ from the unintentionally

present impurities,26 three different excitation wavelengths

(lex = 488, 514 and 633 nm) were used. To our knowledge,

there are only a few reports on Raman properties of doped

zirconia measured beyond the phonons spectral region,27–29

with only one study on europium-doped ZrO2 in the amorphous

to tetragonal phase transition.30 Further, only little work has

been done on Eu3+ doped ZrO2 by using temporarily and

spectrally resolved luminescence so far.23,25,34 We have recently

shown that this technique is very efficient in discriminating

spectral dynamics even within emission of Eu3+ in amorphous

ZrO2.34 To infer for the relative contribution of the short- and

long-lived emission and to attribute these to Eu3+ on amorphous/

disordered, tetragonal or monoclinic sites of ZrO2, TRES were

measured under several excitation wavelengths spanning the

UV to Vis spectral range and by using different temporal gates

(delay and widths). Pure and europium-doped ZrO2 were

synthesized by a novel oil-in-water microemulsion reaction

method.30–34 Overall, we show that by combining three local

sensitive techniques such as HRTEM, Raman and lumines-

cence, an in depth correlation between the emission properties

of Eu3+ (such as crystal-field splitting, emission/excitation

shape, and excited state dynamics) and the local phase content

(amorphous, tetragonal and ordered and disordered mono-

clinic) of ZrO2 can be established. Finally, we have investi-

gated the contribution of zirconia host as a sensitizer for the

europium emission in either the monoclinic or the tetragonal

phase in comparison with the O2�–Eu3+ charge-transfer (CT)

absorption band and f–f absorptions of europium.

2. Experimental methods

2.1 Materials and preparation

Synperonics 10/6 was a gift from Croda. Zirconium(IV)

2-ethylhexanoate was from Alfa Aesar; Europium(III) 2-ethyl-

hexanoate was purchased from Strem Chemicals. Hexane

(Suprasolv, for gas chromatography) and ammonia 30% were

purchased from Merck. Isopropanol was purchased from

Carlo Erba. The employed microemulsion system was: water/

Synperonics10/6/hexane. Zirconium(IV) 2-ethylhexanoate

and Europium(III) 2-ethylhexanoate were used as organo-

metallic precursors. The composition used was: 64.5 wt%

water, 21.5 wt% surfactant and 14 wt% oil phase, which

was an hexane solution of Zirconium(IV) 2-ethylhexanoate

for the synthesis of undoped zirconia (ZE), whilst for Eu

bulk-doped zirconia (ZEB) the oil phase was an hexane

solution of Zirconium(IV) 2-ethylhexanoate and Europium(III)

2-ethylhexanoate (0.99 : 0.01 Zr/Eu atomic ratio). The pre-

paration of the nanoparticles by the o/w microemulsion reac-

tion approach was carried out as follows. All the components

were mixed to obtain an isotropic solution (microemulsion) at

a temperature of 35 1C. The pH of the solution was then

increased to 11 by addition of concentrated ammonia under

vigorous stirring and left for 48 h. The obtained zirconia

precipitate was washed by a mixture of ethanol and chloro-

form (1 : 1), dried overnight at 70 1C and grinded using an

agate pestle and mortar. All samples were calcined under air

in situ (during the XRD experiments up to 1000 1C) and ex situ

(at 500, 750 and 1000 1C) at a heating rate of 1–2 1C min�1.

The dried pure and europium-doped ZrO2 were denoted as ZE

and ZEB, respectively. The thermally treated samples are

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

12972 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 This journal is c the Owner Societies 2012

denoted as ZE-T and ZEB-T where T holds for the calcination

temperature (1C).

2.2 Characterization

Particle size and morphology were investigated by High

Resolution Transmission Electron Microscopy (HRTEM).

The sample was prepared as follows: 0.5 mg of zirconia

powder were suspended in isopropanol (2 ml) and sonicated.

For analysis, a drop of this dispersion was deposited onto a

holey formvar/carbon copper grid. Observation was carried

out using a Field Emission Transmission Electron Micro-

scope, JEM-2200FS, 200 kV, with 0.19 nm resolution in

TEM mode and 0.1 nm resolution in STEM mode and

spherical aberration correction in STEM mode. TGA was

recorded on a Thermobalance TGA/SDTA 851 (Mettler

Toledo) using a heating rate of 10 1C, from 25 1C to 850 1C.

The textural characterization was made based on the

adsorption–desorption isotherms of nitrogen at �196 1C

measured with a Micromeritics ASAP 2020 Surface area and

Porosity Analyzer. Before analysis, the samples were out-

gassed at 150 1C for 12 h. From these isotherms the

Brunauer–Emmet–Teller method was applied to determine the

surface area and the Barret–Joyner–Halenda method was used

to calculate the pore size and volume. Powder X-ray diffraction

(XRD) patterns were recorded on a Schimadzu XRD-7000

diffractometer using Cu Ka radiation (l = 1.5418 A, 40 kV,

40 mA) at a scanning speed of 0.101 min�1 in the 6–601 2Yrange inside an in situ cell accessory. In situ calcination of the

samples was carried out using the same apparatus in the

50–1000 1C temperature range. Raman analysis was carried

out with a Horiba Jobin Yvon–Labram HR UV-Visible-NIR

Raman Microscope Spectrometer, at 633, 514 and 488 nm and

a catalytic cell for sample heating. The analysis was carried out

under the following conditions: a beam diameter of 1.0+ 5%mm;

a spot diameter of 0.5 mm for ex situmeasurements and 1.5 mmfor in situ measurements; spatial resolution of 0.35 mm for

ex situ measurements and 0.6 mm for in situ measurements.

2.3 Photoluminescence measurements

The photoluminescence (PL) measurements were carried out

using a Fluoromax 4 spectrofluorometer (Horiba) operated in

both the fluorescence and the phosphorescence mode. The

repetition rate of the xenon flash lamp was 25 Hz, the

integration window varied between 300 ms and 3 s, the delay

after flash varied between 0.03 and 10 ms, and up to 200 flashes

were accumulated per data point. The gate delay (delay after

lamp pulse, dt) was varied form the minimum value of 0.03 ms

up to 10 ms whereas the gate width (GW) was varied from

0.5 to 20 ms. Excitation/emission slits were varied from the

maximum value (29 nm) up to a minimum value of 0.05 nm.

By using specific temporal gates (delay, dt and gate width,

GW) the PL spectra corresponding to short or long-lived

europium species can be easily discriminated. For example,

imposing a relative short delay and narrow gate width, the

contribution of the short-lived emission is enhanced whereas at

long delay, only the long-lived emission is evidenced.

Time resolved emission spectra (TRES) were recorded at room

temperature using a wavelength tunable Nd:YAG-laser/OPO

system (Spectra Physics/GWU) operated at 20 Hz as

an excitation light source and an intensified CCD camera

(Andor Technology) coupled to a spectrograph (MS257

Model 77700A, Oriel Instruments) as a detection system.

The TRES were collected using the box car technique. The

initial gate delay, dt was set to 500 ns and the gate width was

adjusted to 50 ms. Time resolved emission spectra (TRES) were

also recorded using a nitrogen laser (emission wavelength at

337 nm, frequency of 20 Hz, model VSL-337ND-S from

Spectra-Physics) an intensified CCD camera (Andor Techno-

logy) coupled to a spectrograph (Shamrock 303i, Andor

Technology). The PL decays were analyzed by fitting with a

multiexponential function f(t) using the commercial software

(OriginPro 8): f(t) =P

Ai exp(–t/ti) + B (eqn (1)), where Ai is

the decay amplitude, B is a constant (the baseline offset) and tiis the time constant of the decay i. The average PL lifetimes,

tav, were calculated using the following formula: tav =Aiti(eqn (2)).

3. Results and discussion

3.1 In situ and ex situ X-ray diffraction spectroscopy

The in situ diffractograms of ZEB measured between room

temperature and 1000 1C show broad bands between 2y= 201

and 401 characteristic of the amorphous phase until B450 1C

(Fig. 1(a)). The crystallized metastable t phase (reflection

planes at 2y 29.96 (111), 35.08 (110), 34.18 (200), 34.92 (002),

50.3 (112), 49.7 (220) 58.64 (131) 59.62 (113)) remains the

dominant phase up to 1000 1C where a minor m phase is also

observed (reflection at 50.26 (202)).

The patterns can be regarded as similar for ZEB and ZE,

which shows the non-significant effect of Eu3+ on the long-

range ordering of ZrO2. For the as-synthetized (ZE, ZEB)

samples the particle size has been estimated from HRTEM

images. Its relative constancy up to 400 1C Fig. 1(b) was

crudely extrapolated based on the comparison between the

HRTEM images of as synthetized and calcined samples at

400 1C (not shown). For the tetragonal phase, the crystallite

size was determined from reflection plane at 2y 29.961 (111).

Two steps of increase in the crystallite size were identified with

slower increase from 500 up to 750 1C and a faster increase

from 750 to 1000 1C. Starting with 750 1C a minor contribu-

tion of the monoclinic phase was observed with ZE and ZEB

until 1000 1C. We have also measured the XRD patterns for

samples calcined ex situ at 500, 750, 900 and 1000 1C (heating

rate of 1–2 1Cmin�1, keeping at the maximum temperature for

2 h and subsequently cooled at room temperature) and

represented them in Fig. 1(c). At 500 1C ZEB contains mainly

a tetragonal phase with a crystallite size around 10 nm. At

1000 1C, the contribution of the m phase is largely enhanced

whereas for the in situ calcined samples the contribution of the

monoclinic phase remained low even at 1000 1C. These

differences account for the strong influence of thermal history

on the t to m-phase transition in ZrO2.36 The volume fraction

of the monoclinic m phase was increased from a few percent at

500 1C up to 56% at 1000 1C as estimated by use of the

method of Toraya et al.37 Comparison between crystallite sizes

estimated from the ex situ XRD and the particle sizes obtained

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 12973

from HRTEM shows agglomeration effects which were reduced

upon Eu3+ doping.

3.2 Order and disorder effects in the tetragonal phase of ZrO2

(T = 500 8C)

Fig. 2 shows the TEM and HRTEM pictures of the pure and

europium doped zirconia (ZE-500 and ZEB-500) calcined at

500 1C.

The calcined nanocrystals are much better defined as com-

pared to the as-synthetized sample, although some amorphous

zones were still observed. In addition, the nanocrystals present

structural defects such as dislocations, as well as pits with

diameters of around 2–5 nm. The particle size was larger,

approximately 20–30 nm, although this size should be taken

cautiously as it was difficult to distinguish between individual

particles. The shape of the particles appeared to be slightly

elongated. Upon Eu3+ doping, similar morphology, size and

structure trends were observed, except for smaller particle size

and less agglomeration for the calcined samples (10–20 nm for

ZEB-500; Fig. 2(c) and (d)).

Ex situ Raman spectra of ZEB calcined at 500 and 1000 1C

were measured in the spectral range of 150–4000 cm�1 by

using lex = 488, 514 and 633 nm. At 500 1C, Raman spectra

show several bands in the range of 200 to 700 cm�1 (namely at

265, 318, 460, 600 and 649 cm�1) at similar positions regard-

less of the excitation wavelength. They were unambiguously

assigned to the Raman bands of t-ZrO2.38–40 The crystal

structure of tetragonal ZrO2 is a body-centred lattice with

the space group P42/nmc41 and Zr4+ are eight-fold coordi-

nated to oxygen atoms in D2d (D4h) symmetry. Additionally,

for europium doped ZrO2 (ZEB-500) and for lex = 488 and

514 nm relatively strong and narrow bands whose location

depends on the excitation wavelength were detected in the

spectral range of 2000–4000 cm�1.26

For lex = 514 nm, the relatively strong and narrow Raman

shifts at B2530, 2550, 2936 cm�1 correspond to the 590.8

(5D0–7F1), 591.5 (5D0–

7F1) and 606 nm (5D0–7F2) emission

which are a fingerprint of the tetragonal emission (Fig. 3, see

also Table 1). The 5D0–7F1 transition has two lines (590.8 and

591.5 nm) separated by DE = 20 cm�1 with associated

FWHM values of 15 and 20 cm�1. Furthermore, relatively

broad/disordered spectral features could be detected on the

long-wavelength side of the 5D0–7F2 transition, at 610–613

and 625 nm. Overall, at 500 1C the XRD patterns and Raman

luminescence spectra (Fig. 1 and 3) sustain a long and short-

range ordering characteristic of the tetragonal phase in

europium doped zirconia. Small contribution of the disordered

Fig. 1 In situ XRD patterns with temperature of ZEB nanostructures.

Each plot represents a 50 1C temperature increase. (b) Variation of

the crystallite size of ZE and ZEB with temperature estimated from

the in situ and ex situ XRD (ZEB only). (c) XRD patterns of

ZEB nanocrystals calcined in air at 500, 750, 900 and 1000 1C for

several hours followed by subsequent cooling at room temperature

(* = monoclinic; 1 = tetragonal phase).

Fig. 2 TEM images ZE-500 (a, b) and ZEB-500 (c, d). Insets to (a, c) are HRTEM images. Insets to (b, d) are SAED patterns. HRTEM images of

ZE-500 (e), the white arrows indicate small amorphous zones embedded in the crystalline region; (f) shows an amorphous region.

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

12974 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 This journal is c the Owner Societies 2012

phase is also noticed from the broad luminescence features in

agreement with HRTEM (Fig. 2(e) and (f)).

The PL spectrum of ZEB-500 excited into the O2�–Eu3+

charge-transfer band (at lex = 240 nm)26,29 displays the typical

features of Eu3+ emission in t sites (Fig. 4(a)). The (normalized)

time-resolved PL spectra of ZEB-500 excited show a negligible

dependency on time delay dt after the lamp pulse which means

that at this excitation wavelength only Eu3+ in tetragonal sites

were excited. Some weak 5D0–7F0 emission could also be

noticed at short dt. This implies the presence of minor con-

tribution of amorphous/disordered type emission, as ideally this

transition is forbidden in D2d symmetry. Shifting lex within

270–300 nm range, a relative broader emission could be noticed

along with the t-like emission. To infer for the origin of this

additional emission, two PL spectra were acquired by using

narrow (0.5 ms) and large (5 ms) gate widths (GW).

As seen in Fig. 4(a), the two spectra are different: the one

acquired by using large GW is relatively narrow characteristic

of Eu3+ crystalline environments (peaked at 614 nm) whereas

the one acquired by narrow GW is much broader being

characteristic of Eu3+ in disordered environments. The narrow

emission peaked at 614 nm is assigned to Eu3+ in monoclinic

sites and the origin will be further confirmed with a sample

calcined at 1000 1C (ZEB-1000) (see also Fig. S5(a), ESIw).

Fig. 3 Ex situ Raman spectrum of ZEB-500 (lex = 514 nm). Inset 1:

enlarged view in the spectral range of phonon bands. Inset 2: the

emission spectrum obtained under lamp excitation at lex = 240 nm.

Inset 3: tetragonal crystal-field splitting of 5D0–7F1 transition deter-

mined from the Raman spectrum.

Table 1 Assignment of the 5D0 related PL transitions of Eu3+ from Raman spectra illustrated in Fig. 3 and 6

T (1C) lex (nm) Raman shift (cm�1)/emission wavelength (nm)/electronic transition (5D0–7FJ, J = 0, 1,2,4)

500 514 2530/ 2550/ 2936 b3111–3140/ b3455–3518590.8/ 591.5 605.5 610–613 625–6275D0–

7F15D0–

7F15D0–

7F25D0–

7F25D0–

7F2

1000 488 3220/ 3587 3604 3725 3763579 591.5 592 596.5 597.85D0–

7F05D0–

7F15D0–

7F15D0–

7F15D0–

7F1

514 i2021 b2145/ 2530/ 2548/ 2669/ 2708/ 2936 3130/ 3455/ 3557/573.6 576–578/ 590.8/ 591.5/ 596/ 597/ 605.5 612.5/ 625/ 629/

5D0–F05D0–

7F15D0–

7F15D0�7F1

5D0�7F15D0�7F2

5D0�7F25D0�7F2

5D0�7F2

633 1419 1513/ 1619/ 1701/ 1728/ 1774/695/ 700/ 705/ 709/ 711/ 7135D0–

7F45D0–

7F45D0–

7F45D0–

7F45D0–

7F45D0–

7F4b Refers to broad features. i Refers to unintentional impurities (Sm3+).

Fig. 4 Time resolved PL spectra of ZEB-500 at (a) lex = 240 nm and

(b) lex = 394 nm. (c) Dependency of PL decays of ZEB-500 on

excitation and/or emission wavelengths.

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 12975

The time-resolved PL spectra excited into the f–f absorptions

of Eu3+ at lex = 464 nm also evidence for the co-existence of

ordered and disordered environments at europium sites in

ZEB-500 (Fig. 4(b)). At dt r 1 ms the 5D0–7F2 emission is

centred at B611 nm with a pronounced spectral feature at

B625 nm whereas for dt > 4 ms only the t like emission is

measured with the 5D0–7F2 emission centred at 606 nm.

The 611 nm centred emission (assigned in the literature to

Eu3+ in monoclinic sites17,21) could not be detected upon UV

excitation. We assign this emission also to Eu3+ in monoclinic

sites as further confirmed with ZEB-1000. In line with a

multisite distribution of Eu3+ in ZEB-500, the PL decays

are non-exponential with a strong dependency on the excitation

(lex = 240, 270, 300, 310 nm) and/or emission wavelength

(lem = 580, 606 and 614 nm), Fig. 4(c). Due to increased

radiative transition probability of Eu3+ emission in low

symmetry, monoclinic sites compared to the tetragonal sites,

the PL decay measured at 614 nm is shorter than that

measured at 606 nm.

In conclusion, the local structure environment at Eu3+ sites

in ZEB-500 is predominantly tetragonal; however obvious

contributions from the amorphous/disordered and monoclinic

phases have also been noticed. Confirmation for these addi-

tional phases was given by HRTEM investigation (Fig. 2).

Eu3+ luminescence extracted from Raman spectra also

displays some broad/amorphous type spectral features at

610–613 and 625 nm superimposed on the t-like emission of

europium (Fig. 3). Luminescence spectra enhance greatly the

information extracted from the phonons spectra and XRD

patterns which display only the characteristic phonons/reflec-

tions of t-ZrO2.

3.3 Order and disorder effects in the mixed tetragonal and

monoclinic ZrO2 (T = 1000 8C)

At 1000 1C (Fig. 1(b) and (c)), the particle size became much

larger (in the order of 50–100 nm). Nanocrystals were much

better defined, as crystalline zones have evolved into larger

areas with only very few dislocation-like defects (see also

Fig. S1 and S2, ESIw). SAED patterns and FFT analyses of

HRTEM images showed that both tetragonal and monoclinic

structures were present in ZE-500 and ZE-1000 samples, and

that the presence of the latter increased with the temperature

of the thermal treatment. Upon Eu3+ doping, similar mor-

phology, size and structure trends were observed, except for

smaller particle size and less agglomeration for the calcined

samples (10–20 nm for ZEB-500 and 30–70 nm for ZEB-1000)

as shown in Fig. 2 and 5. STEM bright and dark field images

of ZEB-1000 (also ZE-1000) are shown in ESIw (Fig. S2),

confirming the small size and porous texture of all samples.

As expected form the ex situ XRD patterns (Fig. 2c) at

1000 1C, along with the Raman phonon modes of t-ZrO2

Raman phonon modes of m-ZrO2 (9 from the total of 18

predicted for the monoclinic symmetry) at B174, 186, 211,

325, 364, 471, 525, 548, and 613 cm�1 were also detected.38–40

The crystal structure of monoclinic ZrO2 has a space group of

P21/c, where each Zr4+ is 7-fold coordinated with oxygen

atoms with low C1 local symmetry.41 The volume fraction of

the monoclinic phase was estimated from the XRD patterns

around 56% following the method of Toraya et al.37 From

the ex situ XRD patterns, a volume fraction of monoclinic

phase of 56% was measured. Note that the contribution of

m-phase to both phonons and luminescence spectra may be

artificially enhanced since (i) the m phase has a greater

scattering cross-section than the t phase42 and (ii) the radiative

emission rates of europium are larger in the low symmetry,

monoclinic sites compared to the more symmetrical tetragonal

sites.

For lex = 488 and 514 nm, the relatively strong and narrow

bands observed in the spectral range of 2000 and 4000 cm�1

were related to the 5D0–7F0,1,2 transitions (Table 1). The

relative contribution of m (at 170–190 cm�1, 328, 348, 472,

616 cm�1) and t (at 265, 318, 460, 600 and 649 cm�1) Raman

phonon modes parallels that of m and t like emission of

europium. Therefore, an unambiguous correspondence

between the local crystalline phase of ZrO2 host and local

symmetry at europium sites can be established. From the

Raman spectrum excited at 488 nm (Fig. 6(b)) several lines

are detected in the spectral range of the 5D0–7F0 emission

(at 3170 cm�1 (577.3 nm), 3220 cm�1 (579 nm), 3270 cm�1

(580.6 nm), 3297 cm�1 (581.5 nm), 3310 cm�1 (582 nm)

and 3335 cm�1 (582.8 nm)) which may support for the

existence of multiple monoclinic sites and/or the presence of

impurities.

For lex = 633 nm, besides the t and m Raman phonon

modes of ZrO2, a number of lines was detected between 1200

and 2000 cm�1 (Fig. 6(c) and Table 1). These correspond to

the 685–720 nm spectral range of europium emission assigned

to the 5D0–7F4 transition. In comparison with lex = 488 and

514 nm, these luminescence lines are less intense, comparable

to the Raman bands. We have also measured the lamp excited

PL spectra at lex = 488, 514 and 633 nm. The emission spectra

excited by a lamp follow closely the luminescence extracted

from Raman spectra (illustrated in Fig. 6c).

Fig. 5 TEM images ZE-1000 (a, b) and ZEB-1000 (c, d). Insets in (a, c) are HRTEM images. Insets in (b, d) are SAED patterns. HRTEM images

of ZEB-1000 show (e) general view of mesoscale pit defects and the white arrows point at high resolution of pits (f).

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

12976 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 This journal is c the Owner Societies 2012

All excitation wavelengths used in Raman experiments

(lex = 488 and 514 nm) are off-resonant with proximate

europium f–f absorption transitions at 464 nm (corresponding

to 7F0–5D2 absorption) or 525 nm (corresponding to 7F0–

5D1

absorption) (Fig. S4, ESIw). Excitation at 633 nm is also

off-resonant with the proximate absorption at 579 nm (corre-

sponding to the 7F0–5D0 transition) with an energy mismatch

of ca. 1400 cm�1. It is assumed that the excitation occurs via

the upper level 7F2 (located at B1000 cm�1 above the 7F0

ground level) even its fractional population is slightly below

1% at room–temperature.26 We should note that, as the7F0–

5D0 transition is strictly forbidden in D2d symmetry, the5D0–

7F4 related emission extracted from the Raman spectrum

excited at lex = 633 nm (Fig. 6(c)) should relate mainly to

Eu3+ located in the monoclinic sites.

The FWHM of the t-like emission (measured on the highest

energy Stark level of 5D0–7F1 transition) is narrowed from

16 cm�1 (ZEB-500) to 14 cm�1 (ZEB-1000) whereas the

FWHM values of Raman t phonon modes at 266.5 nm decrease

from 26.5 cm�1 (ZEB-500) to 17.6 cm�1 (ZEB-1000). The

results are explained by the improved crystallization, increase

of the crystallite size and reduction of defects in perfect

agreement with the HRTEM and XRD results (Fig. 1

and 2). Generally, only mixed t and m like emission was

obtained under broad steady-state excitation. However, under

specific excitation conditions almost ‘‘pure’’ typical t and m

like PL spectra could be obtained. For example, tetragonal

type emission could be almost selectively excited at lex =

526.7 nm. A t-like emission was obtained by use of pulsed

lamp/laser excitation and long dt (>5–10 ms) at lex = 240 nm,

394 and 464 nm. Monoclinic like emission could be obtained

under narrow excitation at 394 nm or 300 nm. Eu3+ emission

in m sites is dominated by the 5D0–7F2 electric dipole transi-

tion centered at 614 nm with two weaker spectral features at

625 and 631 nm (Fig. 7(a)). In the spectral range of 5D0–7F1

transition, two lines at 597 and 598 nm from the total of three

expected for the C1 symmetry were noticed. The third, highest

energy 5D0–7F1 level of Eu3+, was established by comparing

the Raman spectra around 3600 cm�1 (lex = 488 nm,

Fig. 6(c)) and 2500 cm�1 (lex = 514 nm, Fig. 6(b)) as being

very close to the highest energy 5D0–7F1 level of Eu

3+ in t sites

(i.e. at 590.8 nm). The maximum Stark splitting for 5D0–7F1

emission in m sites can be thus determined at DE = 200 cm�1

which is an order of magnitude greater than the Stark splitting

in tetragonal symmetry (DE = 20 cm�1). Because the maxi-

mum crystal-field splitting of J-manifolds is a linear function

of the CF strength parameters43–45 these values evidence for

the strongly different crystal-field strengths at the monoclinic

and tetragonal sites of ZrO2.

Fig. 7(a) gathers four relevant emission spectra of ZEB-1000

obtained under specific conditions. The t-like emission spec-

trum was obtained either under selective steady-state (SS)

excitation at 526.7 nm (corresponding to the 7F0–5D1 transi-

tion) or under pulsed excitation at 240 nm with dt = 10 ms

(see also Fig. S5(b), ESIw). The broad spectra were obtained

under pulsed excitation and short temporal gate at 240 nm

(plot 3) or SS excitation at 462 nm (plot 4). For comparison,

the PL spectrum of the as synthesized Eu3+ doped ZrO2

(ZEB) was also included (plot 0). It is interesting to note

the similarity between the broad emission spectra of ZEB-

1000 and that of the as-synthesized ZEB (ZrO2 is amorphous

until 450 1C, Fig. 1(a)). The similarity refers to the peak

values, spectral features at B598, 614 and 625 nm and lines

widths.

As concerning the local structure of amorphous ZrO2, there

is still a debate on whether the local symmetry is similar to the

tetragonal or the monoclinic phase.7,8,46–50 Based on our

current or the recently published29 studies which make use

of Eu PL, the local structure of amorphous/disordered sites

seems to be similar to that of the disordered monoclinic phase.

We have also measured time-resolved PL spectra excited at

lex = 464 nm with dt varying from 5 ms up to 10 ms (Fig. 7(b)).

Fig. 6 Ex situ Raman spectra of ZEB-1000. Inset in (a) represent the

enlarged view in the spectral range of phonon bands. Insets in (b)

represent the enlarged view in the spectral range of phonon bands and

the 5D0–7F0 transition. Inset in (c) represents the PL spectrum

obtained under lamp excitation at 633 nm.

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 12977

The emission is initially centred at 611 nm whereas at long dt(>7 ms) the emission maximum is shifted to longer-lived tetra-

gonal type emission peaked at 606 nm. The temporal evolution

of the PL spectra is consistent with Eu3+ distributed on mixed t

andm sites and also confirms that the 611 nm emission measured

with ZEB-500 (Fig. 4(b)) is of monoclinic origin. A 611 nm

centred line was also observed for yttrium-stabilized zirconia

(single crystal) and assigned to Eu3+ in C2v sites of cubic Y2O3.20

The complex mixture between the disordered/amorphous,

t and m like emission in ZEB-1000 is further substantiated by

use of the wavelength selective emission/excitation spectra

(Fig. 8) within the hypersensitive 7F0–5D2 transition. The t

like emission at 606 nm has a relative narrow single peaked

absorption at 465 nm. Conversely, narrow excitation at

465 nm leads to the t like emission with some contribution

from the monoclinic emission centred at 614 nm. A relatively

strong contribution from t like emission was also measured

with excitation on the lower/higher energy side of the 465 nm

centred absorption (at 471 and 457 nm, respectively) where

only weak and broad/featureless absorption is measured. The

contribution of m like emission to the total spectra is enhanced

with excitation at 466 (close to the excitation maximum of t

like emission at 465 nm), 467 and 468 nm. Broad, disordered

like emission was also obtained using either a non-selective

excitation at 465 nm (excitation slit = 20 nm) or selective

excitation at 462 nm (excitation slit = 2 nm). Such complex of

narrow and broad emission spectra is consistent with large

number of configurations of Eu3+ and oxygen vacancies

distributed in the ZrO2 which results in large multiplicity

of Eu3+ substituted Zr4+ sites with low local symmetry.20

Electron–phonon interactions can also contribute to the shape

of 7F0–5D2 excitation spectrum.51

In conclusion, in ZEB-1000, Eu3+ is distributed on t,

ordered and disordered m sites and the related emissions

can be discriminated under specific excitation conditions.

HRTEM results also evidenced for monoclinic and tetragonal

phase with some content of the amorphous/disordered phase

(Fig. 5), whereas from XRD patterns (Fig. 2a and c) and

Raman spectra (Fig. 6) such a disordered phase could not be

evidenced.

3.4 Evolution of the emission properties with the monoclinic

phase content

We have also investigated the PL of Eu3+-doped ZrO2 at

intermediary temperatures, i.e. 750 and 900 1C. Based on the

method of Toraya et al.37 the monoclinic phase content

estimated form the XRD patterns gathered in Fig. 2(c)

increased from a few percent (500–900 1C) up to 56%

(1000 1C). In contrast with the evolution of the phase content,

we found that the differences between the PL spectra of ZEB-

500, 750 and 900 1C are not significant; nevertheless, by use of

Fig. 7 (a) Tetragonal, ordered and disordered monoclinic like emis-

sion of ZEB-1000; (b) time-resolved PL spectra of ZEB-1000 measured

at lex = 394 nm. Inset shows the increase in the relative contribution

of 606 to 611 nm line intensity with increasing dt.

Fig. 8 PL emission (a) and excitation (b) spectra of ZEB-1000. Excitation was performed within 7F0–5D2 absorption transition whereas emission

was selected at specific wavelengths within the 5D0–7F1,2 transitions.

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

12978 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 This journal is c the Owner Societies 2012

the time-resolved PL spectra acquired with a short temporal

gate the enhancement of the monoclinic (614 nm) relative to

the tetragonal type emission (606 nm) with the calcination

temperature is evident (Fig. 9(a)).

The PL decays corresponding to the t (606 nm) and m

(614 nm) like emissions of Eu3+ were also the measured

function of the calcination temperature/crystallite size

(Fig. 9(b), see also Fig. S6, ESIw). The decays excited at

240 nm show an obvious shortening of the PL decays with

the increase of the calcination temperature. All curves were

non-exponential and were reasonably fitted by using a three

exponential function (eqn (1)). The average PL lifetimes

values, tav, at lem = 606 (eqn (2)) yield the following values:

3.62, 3.3, 2.4 and 2.22 ms for the calcination temperature on

500, 750, 900 and 1000 1C, respectively (see also Fig. 9(c)). The

average PL lifetimes values are shorter than those reported for

Eu3+ in t-ZrO211,17 (4–5 ms). The crystallite size of the t-ZrO2

is determined at B10, 16, 19 and 23 nm for T = 500, 750, 900

and 1000 1C, respectively (Fig. 1(b)).

The average PL lifetimes, tav, measured at lem = 614 yield

the following values: 2.34, 2.25, 1.95 and 1.01 ms for the

calcination temperature on 500, 750, 900 and 1000 1C, respec-

tively (see also Fig. 9(c)). The crystallite size of the m-ZrO2 is

determined at B11, 14, 20 and 19 nm for T = 500, 750, 900

and 1000 1C, respectively (Fig. 1b). The crystallite sizes are

smaller than the particle sizes determined with HRTEM

(10–20 nm at 500 1C up to 30–70 nm at 1000 1C) which

ascertain for some agglomeration effects. For a single phase

nanocrystal host, the PL properties of europium Eu3+ are

considered to be mainly determined by the crystallinity and the

local symmetry.52 Thus, the PL intensities are expected to

increase with increasing crystallite size and particle size with

the effect of the crystallite size on PL properties being claimed

to be stronger than that of particle size.53

If only the nonradiative and radiative processes are con-

sidered to predominate in the depopulation of the 5D0 state,

the experimental lifetime t, the radiative (Arad), and non-

radiative (Anrad) transition rates are related through the

following equation: 1t ¼ Arad þ Anrad (3). The shape of the

emission spectrum of Eu3+ to its radiative lifetime, trad, isrelated via the equation 1/trad = Arad =A01 � n3 � Itot/I01(4)54 where A01 is the Einstein coefficient of spontaneous

emission between the 5D0 and 7F1 levels, n is the refractive

index of the medium and Itot and I01 represent the integrated

intensities corresponding to the D0–7F1 and total emission,

respectively. It is well-known that the ED 5D0–7F2 transition is

hypersensitive to local symmetry at Eu3+ sites whereas the

MD 5D0–7F1 transition is much less sensitive being used as an

internal standard in the quantitative estimation of Eu3+ f–f

intensities.55–59 The intensities ratio corresponding to two

transitions (asymmetry ratio, R) has been extensively used as

an indicator of the local asymmetry and/or degree of the

covalency of the Eu3+–oxygen coordination polyhedron.

Generally, the lower the R value, the higher the local sym-

metry at Eu3+ sites. It was used to investigate the differences

between zirconia crystalline phases, role of surface coating,

effects of calcination temperature, europium concentration

and the incorporation method, the shape effect on the color

purity and so on.11–25,58 The evolution with the delay time, dt,after the laser/lamp pulse of the asymmetry ratio R can be also

Fig. 9 Evolution with calcination temperature of (a) 614 nm centred monoclinic like emission; (b) PL decays measured at 606 and 614 nm;

(c) monoclinic phase content (black symbols) and average PL lifetimes measured at 606 and 614 nm; (d) tetragonal like emission.

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 12979

used as an indicative for the existence of multiple luminescent

species or distribution effects.35,59 The illustration of the

evolution with time of the asymmetry ratio measured at

different excitation wavelengths is included in Fig. S7 (ESIw)for ZEB-500 and ZEB-1000.

The comparison between the t-like emission of Eu3+ doped

ZrO2 calcined at 500, 750, 900 and 1000 1C (Fig. 9(d)) denotes

little change with calcination temperature/crystallite size of the

emission shape with the asymmetry ratio varying within

1.2–1.4 range. As the intensity related to the 5D0–7F3–6

transitions do not change much the overall intensity it results

that the radiative transition rate Arad has less effect on tav andthe decrease of the tav with calcination temperature/crystallite

size should be connected with the increase of non-radiative

transition rate Anrad. Nonradiative relaxation can be enhanced

due to high surface to volume ratio at small particle sizes

combined with the presence of surface defects and OH/CH

impurities (from the Eu3+/Zr4+ precursor). These defects still

present at 500 1C26 were greatly reduced at 1000 1C (according

to FT-IR spectra not included) and, in consequence, the

tav values were expected to increase with increasing calcina-

tion temperature/crystallite size. Fig. 9(c) illustrates similar

dependencies of the PL lifetimes measured at 606 and 614 nm

and the monoclinic content on the calcination temperature.

This could explain that the decrease of the PL lifetimes

corresponding to the t-like emission with calcination tempera-

ture might be related to loss of selectivity at 606 nm due

to some interference from monoclinic type emission. The

decrease of the PL lifetimes corresponding to the m-like

emission with calcination temperature/crystallite size may be

related to Eu3+–Eu3+ interactions (increase of the non-

radiative relaxation rate) and/or distribution on additional

low symmetry monoclinic sites (increase of the radiative

relaxation rate).21 Further work is under way to elucidate

the dependency on the calcination temperature of the PL

lifetimes corresponding to Eu3+ in the monoclinic and tetra-

gonal sites of ZrO2.

3.5 Tetragonal and monoclinic ZrO2 host sensitization of

Eu3+ emission

To our knowledge, until recently,60 no energy transfer from

ZrO2 nanocrystals (either in tetragonal or monoclinic phase)

to europium was reported. To decipher whether the ZrO2 host

can sensitize the europium PL either in the tetragonal or the

monoclinic phase we have performed steady state and time-

resolved emission and excitation measurements as described

below.

For europium-doped ZrO2 calcined at 500 1C (ZEB-500),

relatively strong host related fluorescence centred at 430 nm

(broad, FWHM around 100 nm) with some weak europium

related emission minor was observed upon steady-state (SS)

excitation at 300–370 nm. The fluorescence is short-lived

(ms range or less) and the corresponding excitation spectrum

(lem = 430 nm) is centred at B350 nm (Fig. 10(a)). The

excitation spectrum measured on t-like emission of europium

(lem = 606 nm) is peaked around 240 nm (in both ZEB-500

and ZEB-1000) with some contribution from a broad absorp-

tion peaked at 270–280 nm. The latter contribution absorption

band is strongly reduced following time delay, dt = 5 ms,

which suggests that it is related to short-lived emission of

Eu3+ europium located in amorphous sites. No contribution

from the 350 nm peaked band absorption could be measured

in the excitation spectrum of t-like emission of europium,

which means that Eu3+ PL is not sensitized by tetragonal

ZrO2 (Fig. 10(b)).

For europium-doped ZrO2 calcined at 1000 1C (ZEB-1000),

a relative strong and broad (FWHM value of ca. 100 nm) blue-

green fluorescence related ZrO2 peaked at 480 nm along with

some weak europium related emission was observed upon SS

Fig. 10 (a) ZrO2 related emission and excitation spectra (ZEB-500).

(b) PL excitation spectra of europium in tetragonal sites in ZEB-500

and ZEB-1000 (lem = 606 nm) measured at short and long dt. (c) ZrO2

related emission and excitation spectra (ZEB-1000) and steady-state

and time-resolved excitation spectra of Eu3+ (ZEB-1000).

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

12980 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 This journal is c the Owner Societies 2012

excitation at 280–310 nm (Fig. 10(c)). The excitation spectrum

corresponding to the fluorescence maximum (lem = 480 nm)

is peaked at 305 nm with a FWHM value of 40 nm. It is

interesting to note that both emission and excitation bands of

ZrO2 follow closely those reported in ref. 61. Strong confirma-

tion for the energy transfer from the monoclinic ZrO2 host to

Sm3+ was demonstrated based on the excitation spectra and

the effects induced by Sm3+ acceptor concentration on the

ZrO2 donor emission intensity and decays.61

Fig. 10(c) also shows the steady-state (SS) excitation spectra

corresponding to europium emission in monoclinic sites

(lem = 614 nm) and zirconia emission (lem =480 nm). The

presence of ZrO2 related excitation band in the excitation

spectrum of europium may be strong evidence for the existence

of energy transfer from the zirconia host to the europium

dopant. However, the broad band observed in the excitation

spectrum of europium emission is peaked around 290 nm

which is blue shifted by 15–20 nm relative to the absorption

band centred at 305 nm. Furthermore, in the time-resolved

excitation spectrum (lem = 614 nm, dt = 1 ms) the broad

band centred around 290 nm is strongly reduced relative to the

f–f absorptions of europium. The data suggest that the

presence of 290 nm centred band in the SS excitation spectrum

of europium emission is due to the superposition of ZrO2 and

europium related emissions at 614 nm rather to an efficient

energy transfer from zirconia host to europium.

4. Conclusions

Pure and europium doped ZrO2 nanoparticles were syn-

thesized by the oil-in-water microemulsion reaction method.

The local structure environment at Eu3+ sites in ZEB- 500 was

predominantly tetragonal; however obvious contributions

from the amorphous/disordered and monoclinic phases have

also been noticed. Confirmation for these additional phases

was given by HRTEM investigation and luminescence

extracted from Raman spectra or excited in the steady-state

and time-resolved mode. At 1000 1C, the emission was

assigned to Eu3+ located in both tetragonal and monoclinic

sites. A non-negligible amorphous/disordered content was

measured for all calcination temperatures (500, 750, 900 and

1000 1C) being determined to be of monoclinic nature. In the

tetragonal phase there is no energy transfer from the zirconia

host to the europium dopant. In monoclinic phase, the

efficiency of zirconia host sensitization of europium emission,

if any, is weaker than direct excitation into the absorption

transitions of europium.

Acknowledgements

CT acknowledges financial support from UEFISCDI (grant IDEI

PN-II-ID-PCE-2011-3-0534). M.S.D and M.B acknowledge

the COST D43 and COST D36 actions. Financial support

from Ministerio de Ciencia e Innovacion (MICINN, Spain,

grant number CTQ2008-01979) and Generalitat de Catalunya

(Agaur, grant number 2009SGR-961) is acknowledged.

M.S.D is grateful to NaNoTeCh, the National Nanotechnology

Laboratory of Mexico and Cesar Leyva (CIMAV, S.C.) for the

HRTEM/STEM measurements and assistance.

References

1 E. C. Subbarao,Advances in Ceramics, Vol. 3, Science and Technologyof Zirconia, ed. A. H. Heuer and L. W. Hobbs, American CeramicSociety, Westerville, OH, 1981.

2 P. Li, I.-W. Chen and J. E. Penner-Hahn, Phys. Rev. B: Condens.Matter Mater. Phys., 1993, 48(14), 10063; J. Am. Ceram. Soc.,1994, 77, 118.

3 M. Yashima, T. Kato, M. Kakihana, M. A. Gulgun, Y. Matsuoand M. Yoshimura, J. Mater. Res., 1997, 12, 2575.

4 S. Shukla and S. Seal, Int. Mater. Rev., 2005, 50, 45.5 E. C. Subbarao and H. S. Maiti, Solid State Ionics, 1984,11(4), 317.

6 R. C. Garvie, J. Phys. Chem., 1965, 69, 1238.7 P. Li and I-W. Chen, J. Am. Ceram. Soc., 1994, 77, 23.8 X. Zhao, D. Ceresoli and D. Vanderbilt, Phys. Rev. B: Condens.Matter Mater. Phys., 2005, 71, 085107.

9 M. S. Khan, M. S. Islam and D. R. Bates, J. Mater. Chem., 1998,8, 2299.

10 A. V. Chadwick, G. Mountjoy, V. M. Nield, I. J. F. Poplett,M. E. Smith, J. H. Strange and M. G. Tucker, Chem. Mater., 2001,13, 1219.

11 A. Parma, I. Freris, P. Riello, F. Enrichi and A. Benedetti,J. Lumin., 2010, 130(12), 2429.

12 T. Lopez-Luke, E. De la Rosa, V. H. Romero, C. Angles-Chavezand P. Salas, Appl. Phys. B: Lasers Opt., 2011, 102, 641.

13 L. Yu, H. Liu and M. Nogami, Opt. Mater., 2010, 32, 1188.14 K. Smits, L. Grigorjeva, D. Millers, A. Sarakovskis, A. Opalinska,

J. D. Fidelus and W. Lojkowski, Opt. Mater., 2010, 32(8), 827.15 P. Ghosh, K. R. Priolkar and A. Patra, J. Phys. Chem. C, 2007,

111(2), 571.16 H.-Q. Liu, L.-L. Wang, S.-G. Chen and B.-S. Zou, J. Alloys

Compd., 2008, 448(1–2), 336.17 T. Montini, A. Speghini, L. De Rogatis, B. Lorenzut, M. Bettinelli,

M. Graziani and P. Fornasiero, J. Am. Chem. Soc., 2009,131, 13155.

18 S. Gutzov, M. Bredol and F. Wasgestian, J. Phys. Chem. Solids,1998, 59(1), 69.

19 L. Chen, Y. Liu and Y. Li, J. Alloys Compd., 2004, 381(1–2), 266.20 J. Dexpert-Ghys, M. Faucher and P. Caro, J. Solid State Chem.,

1984, 54(2), 179.21 P. Ghosh and A. Patra, Langmuir, 2006, 22(14), 6321.22 Z. W. Quan, L. S. Wang and J. Lin, Mater. Res. Bull., 2005,

40(5), 810.23 A. Gedanken, R. Reisfeld, E. Sominski, O. Palchik, Yu. Koltypin,

G. Panczer, M. Gaft and H. Minti, J. Phys. Chem. B, 2000,104(30), 7057.

24 B. Julian, R. Corberan, E. Cordoncillo, P. Escribano, B. Viana andC. Sanchez, Nanotechnology, 2005, 16(11), 2707.

25 X. Qu, H. Song, G. Pan, X. Bai, B. Dong, H. Zhao, Q. Dai,H. Zhang, R. Qin and S. Lu, J. Phys. Chem. C, 2009, 113(15), 5906.

26 C. Tiseanu, V. I. Parvulescu, B. Cojocaru, M. Sanchez-DominguezandM. Boutonnet, J. Phys. Chem. C, 2012, DOI: 10.1021/jp3040538.

27 M. Sanchez-Dominguez, M. Boutonnet and C. Solans,J. Nanopart. Res., 2009, 11(7), 1823.

28 M. Sanchez-Dominguez, L. F. Liotta, G. Di Carlo, G. Pantaleo,A. M. Venezia, C. Solans and M. Boutonnet, Catal. Today, 2010,158, 35.

29 C. Tiseanu, V. I. Parvulescu, M. Sanchez-Dominguez andM. Boutonnet, J. Appl. Phys., 2011, 110, 10352135.

30 P. Fornasiero, A. Speghini, R. Di Monte, M. Bettinelli, J. Kaspar,A. Bigotto, V. Sergo and M. Graziani, Chem. Mater., 2004,16, 1938.

31 T. Otake, H. Yugami, H. Naito, K. Kawamura, T. Kawada andJ. Mizusaki, Solid State Ionics, 2000, 135, 663.

32 M. Maczka, E. T. G. Lutz, H. J. Verbeek, K. Oskam, A. Meijerink,J. Hanuza and M. Stuivinga, J. Phys. Chem. Solids, 1999, 60, 1909.

33 M. Sanchez-Dominguez, M. Boutonnet and C. Solans, J. Nanopart.Res., 2009, 11(7), 1823.

34 M. Sanchez-Dominguez, L. F. Liotta, G. Di Carlo, G. Pantaleo,A. M. Venezia, C. Solans and M. Boutonnet, Catal. Today, 2010,158, 35.

35 C. Tiseanu, V. I. Parvulescu, M. Boutonnet, B. Cojocaru,P. A. Primus, C. M. Teodorescu, C. Solans and M. S. Dominguez,Phys. Chem. Chem. Phys., 2011, 13(38), 17135.

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 12970–12981 12981

36 D. A. Ward and E. I. Ko, Chem. Mater., 1993, 5, 956.37 H. Toraya, M. Yoshimura and S. Samiya, J. Am. Ceram. Soc.,

1984, 67, C-119.38 A. Feinberg and C. H. Perry, J. Phys. Chem. Solids, 1981, 42(6), 513.39 E. F. Lopez, V. S. Escribano, M. Panizza, M. M. Carnasciali and

G. Busca, J. Mater. Chem., 2001, 11, 1891.40 C. H. Perry, D. W. Liu and R. P. Ingel, J. Am. Ceram. Soc., 1985,

68, C184.41 CJ. Howard, RJ. Hill and BE. Reichert, Acta Crystallogr., Sect. B:

Struct. Sci., 1988, 44, 116.42 B.-K. Kim, J.-W. Hahn and K. R. Han, J. Mater. Sci. Lett., 1997,

16, 669.43 F. Auzel and O. L. Malta, J. Phys. (Fr.), 1983, 44, 201.44 O. L. Malta, E. Antic-Fidancev, M. Lemaitre-Blaise and

A. Milicic-Tang, and M. Taibi, J. Alloys Compd., 1995, 228, 41.45 C.-G. Ma, M. G. Brik, V. Kiisk, T. Kangur and I. Sildos, J. Alloys

Compd., 2011, 509(8), 3441.46 G. Yanwei, G. Gagherazzi and S. Polizzi, J. Mater. Sci., 1995,

30, 2153.47 O. Stachs, Th. Gerber and V. Petkov, J. Non-Cryst. Solids, 1997,

210, 14.48 D.-Y. Cho, H.-S. Jung, J. H. Kim and C. S. Hwang, Appl. Phys.

Lett., 2010, 97, 141905.49 M. J. Winterer, J. Appl. Phys., 2000, 88(10), 5635.

50 J. Livage, K. Doi and C. Mazieres, J. Am. Ceram. Soc., 1968, 51, 349.51 E. Hemmer, K. Soga, T. Konishi, T. Watanabe, T. Taniguchi and

S. Mathur, J. Am. Ceram. Soc., 2010, 93(11), 3873.52 S. Fujihara and K. Tokumo, Chem. Mater., 2005, 17, 5587.53 W. N. Wang, W. Widiyastuti, T. Ogi, I. W. Lenggoro and

K. Okuyama, Chem. Mater., 2007, 19, 1723.54 H.Martinus, V.Werts, Y. Ronald, T. F. Jukes andW. Jan Verhoeven,

Phys. Chem. Chem. Phys., 2002, 4, 1542.55 G. Blasse, A. Brill and W. C. Nieuwpoort, J. Phys. Chem. Solids,

1966, 27, 1587.56 C. Gorller-Walrand, L. Fluyt, A. Ceulemans and W. T. Carnall,

J. Chem. Phys., 1991, 95, 3099.57 E. W. J. L Oomen and A. M. A. van Dongen, J. Non–Cryst. Solids,

1989, 111, 205.58 T. Ninjbadgar, G. Garnweitner, A. Boarger, L. M. Goldenberg,

O. V. Sakhno and J. Stumpe, Adv. Funct. Mater., 2009,19(11), 1819.

59 C. Tiseanu, V. A. Lorenz- Fonfria, V. A. A. Gessner, M. Kumkeand V. I. Parvulescu, J. Appl. Phys., 2008, 104(1–9), 033530.

60 X. Bai, A. Pucci, V. T. Freitas, R. A. S. Ferreira and N. Pinna,Adv. Funct. Mater., 2012, DOI: 10.1002/adfm.201200759.

61 E. De la Rosa-Cruz, L. A. Diaz-Torres, P. Salas, R. A. Rodriguez,G. A. Kumar, M. A. Meneses, J. F. Mosino, J. M. Hernandez andO. Barbosa-Garcia, J. Appl. Phys., 2003, 94, 3509.

Dow

nloa

ded

by L

inko

ping

s un

iver

site

tsbi

blio

tek

on 2

9 O

ctob

er 2

012

Publ

ishe

d on

24

July

201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2CP4

1946

G

View Online