Effect of the vanadium(v) concentration on the spectroscopic properties of nanosized europium-doped yttrium phosphates

Article abstract of DOI:10.1039/c2dt30380a

Nanosized rare earth phosphovanadate phosphors (Y(P,V)O₄:Eu ³⁺) have been prepared by applying the organic-inorganic polymeric precursors methodology. Luminescent powders with tetragonal structure and different vanadate concentrations (0%, 1%, 5%, 10%, 20%, 50%, and 100%, with regard to the phosphate content) were then obtained for evaluation of their structural and spectroscopic properties. The solids were characterized by scanning electron microscopy, X-ray diffractometry, vibrational spectroscopy (Raman and infrared), and electronic spectroscopy (emission, excitation, luminescence lifetimes, chromaticity, quantum efficiencies, and Judd-Ofelt intensity parameters). The solids exhibited very intense ⁵D ₀ → ⁷F_J Eu³⁺ transitions, and it was possible to control the luminescent characteristics, such as excitation maximum, lifetime and emission colour, through the vanadium(v) concentration. The observed luminescent properties correlated to the characteristics of the chemical environments around the Eu³⁺ ions with respect to the composition of the phosphovanadates. The Eu³⁺ luminescence spectroscopy results indicated that the presence of larger vanadium(v) amounts in the phosphate host lattice led to more covalent and polarizable chemical environments. So, besides allowing for control of the luminescent properties of the solids, the variation in the vanadate concentration in the obtained YPO ₄:Eu³⁺ phosphors enabled the establishment of a strict correlation between the observable spectroscopic features and the chemical characteristics of the powders.

Full text of DOI:10.1039/c2dt30380a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 6310
www.rsc.org/dalton
PAPER
Effect of the vanadium(V) concentration on the spectroscopic properties of
nanosized europium-doped yttrium phosphates†
Jonathan Carvalhaes Batista, Paulo Cesar de Sousa Filho* and Osvaldo Antonio Serra
Received 17th February 2012, Accepted 26th March 2012
DOI: 10.1039/c2dt30380a
3+
Nanosized rare earth phosphovanadate phosphors (Y(P,V)O :Eu ) have been prepared by applying the
4
organic–inorganic polymeric precursors methodology. Luminescent powders with tetragonal structure and
different vanadate concentrations (0%, 1%, 5%, 10%, 20%, 50%, and 100%, with regard to the phosphate
content) were then obtained for evaluation of their structural and spectroscopic properties. The solids were
characterized by scanning electron microscopy, X-ray diffractometry, vibrational spectroscopy (Raman
and infrared), and electronic spectroscopy (emission, excitation, luminescence lifetimes, chromaticity,
quantum efficiencies, and Judd–Ofelt intensity parameters). The solids exhibited very intense
5
7
3+
D → F Eu transitions, and it was possible to control the luminescent characteristics, such as
0
J
excitation maximum, lifetime and emission colour, through the vanadium(V) concentration. The observed
3+
luminescent properties correlated to the characteristics of the chemical environments around the Eu ions
3+
with respect to the composition of the phosphovanadates. The Eu luminescence spectroscopy results
indicated that the presence of larger vanadium(V) amounts in the phosphate host lattice led to more
covalent and polarizable chemical environments. So, besides allowing for control of the luminescent
3+
properties of the solids, the variation in the vanadate concentration in the obtained YPO :Eu phosphors
4
enabled the establishment of a strict correlation between the observable spectroscopic features and the
chemical characteristics of the powders.
4
Introduction
their very high luminescence efficiencies. Such properties make
these compounds applicable as vacuum ultraviolet phosphors
VUV, λexc < 200 nm), as they can exhibit very high excitabil-
ities around 142 and 172 nm, which are the main lines provided
by the xenon plasma employed in such devices. Therefore, these
properties enable their application in Hg-free fluorescent lamps,
by using rare gas mixtures instead of mercury, or in the develop-
For almost a half of a century, rare earth (RE)-based phosphors
have by far been the most important luminescent materials with
application in many technological fields.
(
1
,2
Since lanthanoid
ions present partially filled 4f orbitals protected from the chemi-
cal environment by the outer 5s and 5p electrons, their unique
line-type f–f electronic transitions, mainly in their trivalent state,
result in practically monochromatic emissions, which has
enabled their use for light generation in several high-perform-
5
ment of large high-resolution plasma display panels. On the
other hand, the luminescence processes in such matrices can be
considerably complex. This is because, although the optical tran-
sitions are only weakly affected by the occupied chemical
environment, the non-radiative transitions are quite dependent on
the characteristics of the matrix, such as composition, electronic
2
ance devices. For example, RE-compounds are still the best
phosphors for the generation of the three primary colours (red,
green, and blue, RGB) in illumination (such as compact fluor-
escent lamps) and visualization (cathode ray tubes, field emis-
6
structure, and particle size. Moreover, in these cases vibrational
3
,4
−1
sion displays, and plasma display panels) technologies.
energies of the matrix are relatively high (>800 cm ), which
can lead to effective vibronic interactions with the metal states,
thereby resulting in intense quenching processes. In this way, the
viability of applying an oxossalt-based phosphor containing RE
ions depends on the study of several spectroscopic, compo-
sitional and structural parameters, so that the advantages and dis-
advantages of their use as light emitters can be weighted.
Particularly, oxossalt phosphors such as phosphates, vana-
dates, borates, and niobates, among others, are quite important
within the luminescent materials field. This is because they
display very high physico-chemical stability, especially against
high-temperatures or high-excitation energies, not to mention
In this sense, RE phosphate and vanadate phosphors have
been widely applied and studied as commercial luminescent
materials, as in the case of the utilization of the classic green and
Rare Earth Laboratory, Faculdade de Filosofia, Ciências e Letras de
Ribeirão Preto, University of São Paulo, Av. Bandeirantes, 3900,
1
4040-901 Ribeirão Preto, SP, Brazil. E-mail: pcsfilho@usp.br;
Fax: +55 16 3602 4838; Tel: +55 16 3602 4376
Electronic supplementary information (ESI) available: Additional lumi-
3+
3+
3+
red phosphors LaPO :Ce ,Tb and YVO :Eu in cathode ray
4
4
†
tubes and fluorescent lamps. RE vanadates are quite relevant in
the lighting field, since they have been used as phosphors in
nescence spectra, luminescence lifetime decay curves, and additional
SEM micrographs. See DOI: 10.1039/c2dt30380a
6310 | Dalton Trans., 2012, 41, 6310–6318
This journal is © The Royal Society of Chemistry 2012

View Article Online
high-pressure Hg lamps, mainly because of their luminescence
regulated to ∼4. The precursor of orthophosphate ions was tripo-
3,4
efficiency even at high operating temperatures (300 °C). Fur-
thermore, vanadate phosphors have also been successfully
employed in the manufacture of efficient high-power polarizing
lyphosphoric acid (H P O ), which was produced from an
5
3 10
aqueous sodium tripolyphosphate solution (Na P O ·6H O
5
3
10
2
Synth 99%, previously recrystallized five times with methanol)
by using cation exchange resin (Dowex 50W X4
100–200 mesh, J.T. Baker). The H P O solutions were standar-
7
3+ 3,4,8
prisms and active laser materials such as YVO :Nd ,
as
a
4
9
well as scintillators in medical image detectors.
5
3 10
The RE vanadates are isostructural to the RE phosphates, so
they have monoclinic (monazite) or tetragonal (xenotime/zircon)
dized by means of potentiometric titrations with NaOH solutions
immediately before their use. The source of orthovanadate ions
was ammonium metavanadate (NH VO 99.9%, Aldrich). The
procedures were conducted by employing different molar P/V
ratios, namely 100/0, 99/1, 95/5, 90/10, 80/20, 50/50, and 0/100.
10
crystal structures, which makes the development of mixed
phosphovanadate phosphors with a number of advantageous
luminescent properties possible. For instance, the partial replace-
4
3
3
−
3−
ment of PO4 with the VO4 species can afford improved stab-
For this purpose, the desired amount of the RE(NO ) solution
3 3
11
ility and high-temperature luminescence characteristics,
was mixed with citric acid (CA) and ethylene glycol (EG) at a
3+
besides allowing for the control of the emission colour, due to
the higher intensity of the vanadate blue emissions when this
group is diluted, and of the luminescence lifetime. Since the
emission processes are not controlled by the nature of the host
lattice and the activator only, the crystal morphology, size, and
surface area, as well as crystallite defects, agglomeration, and
homogeneity also play important roles in the decay mechanisms
RE /CA/EG molar ratio of 1/20/80, and with the appropriate
volumes of the H P O and/or NH VO solutions, so that the
5
3
10
4
3
3
+
RE /(P + V) = 1.00 molar ratio would be attained. The final
3+
−1
RE concentration was adjusted to 0.02 mol L by addition of
deionized water. The clear and yellowish solution achieved at the
end of this procedure was then heated at ∼100 °C under stirring
for 1 h, which furnished a polymeric resin. The decomposition
of the resin was started in a hot plate at 400 °C and finished with
calcination at 900 °C in air for 4 h, thus yielding the nanosized
powders (Y0.99Eu0.01(P1−xV )O , x = 0.01, 0.05, 0.10, 0.20,
1,4–6
of the excited species.
Thus, in order to control the architec-
ture and spectroscopic features of these compounds, countless
methodologies have been developed for their synthesis, such as
the incorporation of the phosphor into colloidal silica spheres
x
4
0.50, and 1.00).
1
2
or the obtainment of SiO @YVO core–shell systems, their
The morphology of the solids was evaluated by scanning elec-
tron microscopy (SEM) on Zeiss EVO 50 equipment. The struc-
tural analysis was performed by powder X-ray diffractometry
(XRD) on a Siemens D5005 diffractometer. The Fourier trans-
form infrared (FT-IR) and Raman spectra were respectively
acquired on a Shimadzu IRPrestige-21 spectrophotometer (using
KBr pellets, with a 2 cm− resolution) and on a Horiba
LabRAM HR Raman spectrometer (using a He–Ne laser at
632.8 nm as excitation). The luminescence spectra and lifetimes
were recorded on a Horiba Jobin Yvon SPEX Triax 550 Fluoro-
2
4
1
3
preparation by high-temperature solid state reaction, electron
spinning methods, sol–gel and polymeric precursors methodo-
14
1
5–17
18
logies for the obtainment of both powder
and thin film
1
9
phosphors, lithography processed luminescent films, as well as
11,20
21
22,23
hydrothermal,
colloidal, and template-directed
precipi-
1
tations, among others.
Improving the quality of phosphovanadate compounds as
luminescent materials remains an essential research topic, since
the optimization of the matrix components (mainly lanthanum,
yttrium, or gadolinium), the phosphorus/vanadium ratio, and the
activator concentration is crucial for the applicability of these
solids in various devices. In this context, this work aims at the
23
log 3 spectrofluorometer, as previously described. Interference
and long-pass filters were positioned in the excitation and/or
emission beams, in order to improve the quality of the acquired
spectra. The low-temperature luminescence spectra (77 K) were
acquired by placing the samples into quartz tubes in a quartz
Dewar flask filled with liquid N2.
3
+
synthesis and characterization of Eu -doped yttrium phospho-
3
−
vanadates, in order to evaluate the effect of different VO4 con-
centrations in the YPO host on the spectroscopic properties of
4
the obtained powders. To this end, a previously presented
1
7
organic–inorganic polymeric precursors methodology has been
modified and applied to the synthesis of the crystalline solids,
because this technique produces finely divided powders with
high stoichiometric control of the solids. The powders were
characterized with regard to their structural, morphological,
and spectroscopic features, and control of their luminescent
properties was achieved through variation of the vanadate
concentration.
Results and discussion
Fig. 1 displays the SEM micrographs of some of the obtained
powders; the remaining micrographs are presented in the ESI.†
The images show that all solids have very similar morphology,
as a result of similar conditions employed in the synthesis.
Despite the high annealing temperature, which culminated in a
partial coalescence process, the obtained phosphovanadates
occurred as micrometric aggregates of well-defined spherical
nanoparticles (∼50 nm). Besides, the procedures led to a rela-
tively narrow particle size distribution in all cases, allowing for
further comparisons between structural and spectroscopic proper-
ties of the attained powders.
Experimental
The solids were prepared by using a variation of a previously
1
7
described method starting from polymeric precursors. Briefly,
1
1
.0 mol L− rare earth nitrate solutions (Y0.99Eu0.01(NO ) ) were
Fig. 2 depicts the X-ray diffractograms of the solids obtained
with different vanadium(V) concentrations. Because both pure
and anhydrous YPO and YVO have the same body-centered
3 3
prepared by dissolving the previously calcinated oxides (Y O3
2
9
9.99% Alfa-Aesar and Eu O 99.99% Rhône-Poulenc/Rhodia)
2 3
4
4
in concentrated nitric acid. The pH of these solutions was
zircon/xenotime-type tetragonal structure, with the I4 /amd
1
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6310–6318 | 6311

View Article Online
Fig. 2 X-ray diffractograms of the synthesized phosphovanadates with
different vandium(V) concentrations.
Table 1 Cell parameters and crystallite sizes calculated for the
synthesized tetragonal phosphovanadates
3
a (Å)
c (Å)
V (Å ) T (nm)
3
+
Y(P1−x
x =
x 4
V )O :Eu ,
0
6.879 ± 0.005 6.016 ± 0.001 285
6.878 ± 0.005 6.027 ± 0.003 285
6.883 ± 0.005 6.030 ± 0.001 286
6.901 ± 0.004 6.037 ± 0.002 287
6.930 ± 0.015 6.046 ± 0.003 290
7.032 ± 0.223 6.156 ± 0.089 304
7.104 ± 0.006 6.274 ± 0.004 317
14
24
28
20
14
20
21
0
0
0
0
0
1
.01
.05
.10
.20
.50
.00
vanadium(V) ions on the structure is clearly seen by the shift of
the diffraction peaks to lower 2θ values, a result of larger ionic
3
−
3−
volume of the VO4 groups as compared to that of PO4 . For
better evaluation of this effect, the lattice parameters were calcu-
lated for the synthesized tetragonal solids (Table 1). As pre-
dicted, the introduction of the vanadate groups into the
phosphate lattice results in the progressive increase of these par-
ameters and the cell volumes. Moreover, the cell parameters of
the pure phosphate (a = 6.879, b = 6.016) and vanadate (a =
7.104, b = 6.274) samples are slightly smaller than those of the
YPO and YVO JCPDS standards (a = 6.881, b = 6.017, and
Fig. 1 SEM micrographs of some of the synthesized powders: (a)
3
+
3+
3+
YPO
4
:Eu , (b) Y(P0.80
4 4
V0.20)O :Eu , and (c) YVO :Eu .
o
(
D4h19, n 141) space group, complete incorporation of
vanadium(V) into the YPO4 host lattice should be expected.
Indeed, the diffractograms are in agreement with their standards
4
4
(
YPO JCPDS 01-074-2429 and YVO JCPDS 00-017-0341),
4
4
a = 7.1192, b = 6.2898, respectively). The mean crystallite sizes
of the prepared powders were also estimated (Table 1) from the
X-ray diffractograms by applying the Debye–Scherrer relation
which demonstrates that all solids exhibit the same tetragonal
structure. They also evidence the absence of crystal phases
related to contaminants, as well as good incorporation of the acti-
vator ions into the hosts. The relative intensity of the diffraction
peaks points to the occurrence of preferential growth in the
(eqn (1)):
Kλ
cos θ
(2 0 0) plane of the tetragonal structure for the samples contain-
T ¼ β
ð1Þ
ing intermediate vanadium concentrations. The effect of the
1=2
6312 | Dalton Trans., 2012, 41, 6310–6318
This journal is © The Royal Society of Chemistry 2012

View Article Online
in which λ is the X-ray wavelength, θ is the considered Bragg
diffraction peak, β1/2 is the half-height peak width, and K is the
shape parameter (considered to be equal to 0.9, since the par-
ticles are spherical). Within the limitations of the employed
methodology, the values were close to a mean value of ∼20 nm,
attesting to the nanostructural character of the synthesized
powders. It is also important to mention that the crystallite sizes
are not necessarily equal to the observed particle sizes, since
they correspond to the size of the ordered (crystalline) domains
in the solid structures; thus, they assume values that are less than
typical of tetrahedral groups (orthophosphate and orthovanadate)
distorted to a D symmetry, in accordance with the I4 /amd
2d
1
2
5,26
structure.
In the pure phosphate IR spectrum (Fig. 3), the
broad band with two components centred at ∼1050 cm is
ascribed to the P–O anti-symmetric stretching (ν , B + E),
whereas the doublet with bands at 648 and 524 cm is related
to the anti-symmetric deformation (ν , B + E) in PO4
With the addition of increasing amounts of vanadate
groups, a band at ∼860 cm (ν stretching of VO4 ) appears
and becomes more intense. In the Raman spectra (Fig. 4),
−
1
3
2
−
1
3
−
4
2
2
5,26
groups.
−1
3−
3
24
or equal to the grain sizes observed by microscopy techniques.
besides the ν and ν IR-active vibrations, the symmetric stretch-
3
4
The IR and Raman spectra (Fig. 3 and 4) also show the
achievement of tetragonal solids and display the sets of bands
ing (ν , A ) and symmetric deformation (ν , A + B ) of the tet-
1 1 2 1 1
26,27
rahedral groups under D2d symmetry can also be observed.
−
1
The ν signal occurs as a single peak at 1001 cm in the pure
1
−1
phosphate and at ∼891 cm in the pure vanadate; the doublet
corresponding to the ν signal appears as superposed bands at
2
∼
484 cm− in the pure phosphate, and at 377 cm in the vana-
1
−1
date. However, at low vanadium(V) concentrations, a broad back-
ground band is detected in the Raman spectra, which is probably
3
−
a result of the VO4 luminescence. As further discussed, the
vanadate groups present more efficient luminescence when they
are diluted into the YPO lattice. As the vanadate emissions are
4
quenched at higher concentrations, the luminescence becomes
less intense in such cases and does not cause interference in the
Raman spectra.
The luminescence spectra of the synthesized solids were
acquired at 300 and 77 K (liquid N ) and are presented in
2
Fig. 5–7 and in the ESI.† In the excitation spectra of the YPO :
4
3
+
Eu sample, the most prominent signals are attributed to the
Eu f–f absorptions, although a broad charge transfer band
CTB) can be observed at 231 nm. The addition of small
3+
(
vanadium(V) amounts into the YPO host leads to a red-shift of
4
the excitation band, which becomes more intense with regard to
3+
the intraconfigurational Eu absorptions. The increase in the
vanadium(V) concentration results in a progressive bathochromic
shift of the excitation band, as seen in Fig. 5. The position of the
ligand–metal CTB in crystalline host lattices is affected by the
covalence, coordination number, bond polarizability, and anion
charge. The effect of the combination of these factors can be
numerically accounted for in the spectroscopic behaviour
Fig. 3 FT-IR spectra (in KBr pellets) of the obtained phosphate–
vanadate powders with different compositions.
through the so-called environmental factor (h ), which is charac-
e
teristic for each host lattice. The energy relative to the maximum
of the CTB is exponentially related to the environmental factor,
28
as pointed out in eqn (2):
ECTB ¼ A þ Beꢀ
khe
ð2Þ
where A, B and k are experimental parameters that depend only
on the nature of the lanthanoid cation. For trivalent europium,
28
A = 2.804, B = 6.924, and k = 1.256. Because the CTB ener-
gies for the obtained phosphates/vanadates range from 231 nm
3
+
3+
(
5.37 eV) for YPO :Eu to 311 nm (3.99 eV) for YVO :Eu ,
4
4
the calculated environmental factors range from 0.790 to 1.405,
respectively (Table 2). These values are as high as those found
3+
28
for Eu ions in RE oxychlorides or oxybromides. Because
higher h values correspond to more covalent and polarizable
e
environments, it can be concluded that the addition of increasing
amounts of vanadium(V) culminates in a lower ionic character in
the solids. Therefore, the difference in the CTB energies can be
Fig. 4 Raman spectra of the obtained powders for the different vana-
date concentrations.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6310–6318 | 6313

View Article Online
3
+
Fig. 5 Normalized excitation spectra (λem = 616 nm) of the prepared Y(P1−x,V
x
)O
4
:Eu phosphors (inset: excitation spectra in the 380–500 nm
3
+
range showing the main Eu f–f absorptions and their relative intensity with regard to the main spectra).
Fig. 6 Emission spectra of the obtained phosphovanadate powders for
the different vanadium(V) concentrations (λexc = 394 nm for the YPO
Eu sample; the remaining were excited under their CTB maxima).
3
+
4
:
Fig. 7 Emission spectra at 77 K (liquid N
2
) of (a) YPO
:Eu (λexc = 300 nm), and (c) YVO
exc = 310 nm) (insets: amplification of the 400–550 nm range showing
the blue vanadate emissions in the two cases).
4
:Eu (λexc =
3
+
3+
3+
394 nm), (b) Y(P0.95
V
0.05)O
4
4
:Eu
(λ
3+
explained by taking the energy of the first 5d state of Eu and
2
−
of the O ligands in the different cases into account, as illus-
trated in Fig. 8. In the pure phosphate, the energy of the CTB
can be considered as the energy difference between the valence
2−
5+
O
→ V CT absorption, which has a higher molar absorptiv-
ity than the O → Eu CT process. Consequently, excited
intermediate vanadium(IV) species (3d state) are formed, leading
to the population of the Eu emitting states, and giving rise to
very efficient sensitization of the Eu luminescence. For the
orbitals comprising the electronic structure of the solids, because
they are shielded by the outer electrons, the Eu 4f electrons
2
−
3+
7
1
band of the host lattice and the 4f state of the europium ion; i.e.,
3+
the energy necessary for the formation of the bivalent state of
the lanthanoid, since an electron is donated from the host to the
3+
2
−
3+
activator (O → Eu ). On the other hand, in the presence of
3+
vanadate ions, the excitation process is a result of the allowed
6314 | Dalton Trans., 2012, 41, 6310–6318
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 Charge transfer band (CTB) energies and environmental
valence band and reduces the energy difference for a charge
transfer process.
3
+
factors (h
e
) calculated from the excitation spectra of the Eu doped
phosphovanadates
Another aspect of the excitation spectra that agrees with the
5
7
ECTB (eV)
h
e
former conclusions is the relative intensity of the D ← F
2 0
absorption at 464 nm in the different solids, as depicted in
3
+
Y(P1−x
V
x
)O
4
:Eu , x =
5
7
Fig. 5. When compared to the L ← F line, which is the most
6
0
0
5.37
4.28
4.15
4.14
4.10
4.07
3.99
0.790
1.231
1.304
1.310
1.334
1.353
1.405
3+
5
prominent 4f–4f excitation in the YPO :Eu sample, the D ←
F0 transition is progressively intensified with the addition of
vanadate groups. In the pure vanadate, it becomes the most
intense Eu intraconfigurational absorption. The D2
transition is, as well as the D → F transition, hypersensitive
0
0
0
0
0
1
.01
.05
.10
.20
.50
.00
4
2
7
3+
5
7
←
F0
5
7
0
2
to the site symmetry and to the characteristics of the occupied
3
chemical environment. As pointed out in the discussion of the
calculated spectroscopic parameters, the intensification of these
hypersensitive transitions is associated with the occurrence of
transition mechanisms other than the typical forced electric
3−
dipole in the presence of VO4 groups, as a result of the higher
3+
polarizability of the environment around the Eu ions in the
5
7
phosphovanadates. Thus, the behaviour of the D ← F exci-
2
0
tation line corroborates the observation of more covalent and
polarizable sites at higher vanadate concentrations, as high-
lighted in the discussion of the CTB energies.
The emission spectra (Fig. 6) of the obtained solids under UV
5
7
3+
excitation exhibit the characteristic D → F Eu transitions,
0
J
which occur as a set of sharp lines. Although all solids possess
the same crystal structure, as demonstrated by the XRD analysis,
3+
the elevation in the vanadium(V) concentration changes the Eu
emission profile drastically. Because the synthesized solids have
the xenotime-type I4 /amd structure regardless of the presence
1
3+
of vanadium(V), it is expected that the Eu ions occupy D2d
sites in all cases. In such situations, as well as in the case of
5
7
higher site symmetries, the D → F transition tends to be non-
0
0
observable due to the lower degree of relaxation of the spin,
Fig. 8 Simplified energy level diagram illustrating the energy differ-
ence which gives rise to the charge transfer absorptions in the phosphate
and vanadate phosphors.
3,4
parity and ΔJ ≠ 0 ↔ 0 selection rules. This is in agreement
with the observations from the emission spectra, since in all the
5
7
cases the D → F transition has very low intensity. Neverthe-
0
0
less, as seen in the low-temperature emission spectra (Fig. 7 and
poorly contribute to the formation of the valence and conduction
ESI†), intermediate vanadium(V) concentrations give rise to
3
+
3+
bands, and a higher contribution is found for the Eu 5d states.
Thus, if the 5d states lie at higher energies, the generated valence
band will lie at lower energies, and the energy of the CTB will
increase. The introduction of the vanadate groups into the phos-
phate lattice results in a higher covalence, thereby culminating in
a larger ligand field splitting of the 5d levels. As a consequence,
the first 5d states will be at lower energies at higher vanadium(V)
more than one chemical environment for the Eu ions in the
3
+
3+
synthesized solids. Whereas for YPO :Eu and YVO :Eu
4
4
5
7
only two components in the D → F transition are observed,
0
1
in agreement with the occupation of a single D site by the
2
d
3+
Eu ions, the intermediate samples clearly display four signals
for such transition when higher spectral resolutions are utilized.
The same occurs for the D → F transition. Thus, although
5
7
0
2
2
−
concentrations. Moreover, the energy of the ligands (formally O
YVO and YPO can form solid solutions with a unique crystal
4 4
3
−
ions) will also be affected by the differences between the PO
structure in all studied proportions, the microenvironments
4
3−
3+
and VO4 groups. Because they have higher bond energy, lower
around Eu are slightly inhomogeneous for the intermediate
bond length, and higher force constant, P–O bonds are much
more stable than the V–O bonds. Hence, the orbitals in the O
cases, thereby providing two distinct D2d sites for the cations in
2
−
3+
the Y(P1−xV )O :Eu solids.
x
4
ions that participate in the formation of the valence and conduc-
tion bands will lie at higher energies in the presence of larger
vanadium(V) concentrations. The differences between the CTB
energy in the prepared phosphovanadates result from the differ-
ent position of the valence bands with regard to the first charge
Another important observation is that, even under CTB exci-
2−
5+
tation, involving the O → V electron transfer absorption,
3
−
there is no significant blue luminescence of the VO4 anions at
room temperature when the vanadium(V) concentration is higher
than 10%. This attests to an efficient energy transfer process
7
2−
3+
3+
transfer state, which is predominantly 4f (O → Eu ) in the
between these groups and the Eu ions. At 300 K, only in the
1
2−
5+
pure phosphate, whereas it is 3d (O → V ) in the phospho-
vanadates. The rise in the vanadium(V) concentration leads to
higher covalence in the solids, which reflects in a high lying
less concentrated samples (1% and 5% V) a vanadate blue emis-
sion is observed. In these cases, a low degree of concentration
quenching occurs since the absorbed energy cannot migrate
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6310–6318 | 6315

View Article Online
5
5
7
Table 3 Luminescence lifetimes, radiative and non-radiative decay rates, quantum efficiencies ( D
0
level), intensity ratio between the D
0
→ F
2
and
5
7
3+
a
D
0
→ F
1
transitions, Judd–Ofelt intensity parameters, and chromaticity coordinates of the Eu doped phosphovanadates
Chromaticity
RAD (s−
1
)
A
NRAD (s−1
)
η (%)
I
02/I01
Ω
(×10−20 cm2)
Ω
4
(×10−20 cm2)
x
y
τ (ms)
A
2
3
+
Y(P1−x
V
x
4
)O :Eu , x =
0
2.8
2.7
2.2
1.9
1.4
1.2
1.0
259
334
384
440
662
669
943
99
36
70
87
62
165
98
72
90
85
84
91
80
91
0.9
1.8
2.4
3.0
5.1
5.3
8.4
1.6
3.0
4.0
5.1
8.5
9.0
13.7
2.2
1.9
1.7
1.9
2.4
2.2
2.4
0.662
0.428
0.486
0.553
0.629
0.645
0.667
0.336
0.272
0.281
0.302
0.313
0.316
0.317
0
0
0
0
0
1
.01
.05
.10
.20
.50
.00
a
Luminescence lifetimes are accurate within 5%; the calculated parameters are accurate within 10%.
between the VO4 − groups. Due to elimination of the vibrational
effects, at 77 K the vanadate emission becomes prominent in all
cases (ESI†). The occurrence or not of the vanadate lumines-
cence in the room temperature emission spectra is quite relevant
with respect to the purity of the red emission colour of the
solids, which would be lower with the broad blue vanadate emis-
sion. The chromaticity coordinates (CIE 1931) have been calcu-
3
relation to the orange D0
5
→
7
F1 ones, as also depicted in
Table 3 (I /I ratio). As a result, the purity of the red emissions
of the obtained powders can be continuously augmented from a
“magenta-type” colour in Y(P0.99
0.272) to very pure red in YVO :Eu (x = 0.667, y = 0.317) by
the addition of different vanadium(V) concentrations.
Table 3 also lists the other spectroscopic parameters obtained
02 01
3+
V
)O :Eu (x = 0.428, y =
0.01 4
3+
4
2
9
3+
5
lated from the room temperature emission spectra, and are
presented in Table 3 and Fig. 9. In the case of YPO :Eu , a
for the Eu -doped samples. The luminescence lifetimes ( D
0
3+
5
7
level, monitored at the D → F transition) were attained by
0 2
4
very pure red colour is achieved (x = 0.662, y = 0.336), despite
fitting the experimental decays with monoexponential decay
curves. The radiative and non-radiative decay rates (A and
5
7
the high relative intensity of the orange D → F emissions. In
0
1
RAD
5
this case, the absence of broad band emissions below 570 nm
ANRAD, respectively) as well as the D quantum efficiencies
0
5
7
and the high relative intensity of the D → F transitions in the
were computed from the emission spectra and luminescence life-
0
4
5
7
deep red region result in a good overall chromaticity for this
phosphate. In the phosphovanadate powders, the increase of the
vanadium(V) concentration leads to a reduction of the blue emis-
times by taking the magnetic dipole allowed D → F tran-
0 1
sition as a reference, as described in previous works and in eqn
23,30
(3)–(8).
In summary, the emission quantum efficiency (η) of
5
7
sions and to a progressive rise in the red D → F emissions in
a particular excited state is defined as the ratio between the rate
0
2
of radiative deactivation and the total rate of deactivation:
ARAD
η ¼ A
ð3Þ
RAD
þ ANRAD
The total rate of excited state deactivation (ARAD + ANRAD
)
is equal to the reciprocal of the experimental excited state life-
time (τ):
ARAD þ ANRAD ¼ τꢀ
So, the quantum efficiency can be written as
η ¼ ARADτ
1
ð4Þ
ð5Þ
5
3+
When the D level of Eu is considered as the emitting state,
0
the radiative rate of excited state deactivation is equal to the sum
5
7
of the particular rates for each of the D → FJ (J = 0–6)
0
transitions:
X
ARAD
¼
A
ð6Þ
0ꢀJ
J
5
7
Because of its magnetic dipole character, the D → F tran-
0
1
sition has a calculable rate of radiative deactivation and can be
used as a reference for the determination of the other A0−J rates,
and its value is given by
Fig. 9 CIE 1931 chromaticity diagram showing the dependence of the
emission colour with regard to the vanadium(V) concentration in the syn-
thesized powders.
A0ꢀ1 ¼ 0:31 ꢁ 10ꢀ11n3σ3
ð7Þ
0ꢀ1
6316 | Dalton Trans., 2012, 41, 6310–6318
This journal is © The Royal Society of Chemistry 2012

View Article Online
where n is the refractive index of the material and σ
is the
− 1
0
7
energy baricentre of the F level (calculated from the emission
1
spectra). The refractive index considered for our calculations was
n = 1.8, the value usually found for xenotime. The other A0−J
radiative rates are calculated from
S0
S0
ꢀJ 0ꢀ1
σ
A0 ¼ A0ꢀ1
ð8Þ
is
ꢀJ
σ
ꢀ1
0ꢀJ
7
where σ0−J is the energy baricentre of the F level and S
J
0−J
5
7
the area corresponding to the D → F transition in the emis-
0
J
sion spectra.
Finally, the Judd–Ofelt intensity parameters (Ω ) are calcu-
λ
3
0–33
lated by applying eqn (9):
2
3
2
2 X
λ¼2;4
ꢀ
ꢀ
ꢀ
ꢀ
4
e ω nðn þ 2Þ
7
5
ðλÞ
2
A0
ꢀJ
¼
Ω k D
λ
0
U
FJ l
ð9Þ
5
ꢀ
3
0
Fig. 10 Luminescence decay curves ( D level, monitored at the
3
hc
9
5
7
3+
D
0
→
F
2
transition) of the Y(P1−x
V
x
)O
4
:Eu powder phosphors
with different vanadium(V) concentrations.
where e is the elementary charge, c is the speed of light, and ħ is
the Planck constant over 2π (the three in cgs units). ω is the
angular frequency of the considered transition (in cm ) and n is
−1
around 87 ± 7%. So it can be inferred that the shorter excited
state lifetimes are not a consequence of the relative increase of
non-radiative decay mechanisms, such as multi-phonon deactiva-
tion or interactions with structural defects. Therefore, the
enhancement in the total decay rate without reduction in the
quantum efficiency can be interpreted as the result of a higher
relaxation of the selection rules in the presence of higher
vanadium(V) contents. This can be explained by considering the
dynamic coupling mechanism for 4f–4f transitions. This mech-
anism is related to the occurrence of local transient electric
dipoles, which are induced by the incident radiation on the elec-
trons of the ligands, and these induced fields around the lantha-
noid ions can explain the intensification of some hypersensitive
4f–4f transitions. The interaction between these induced dipoles
and the metal transition quadrupole moment results in non-zero
transition moments in such cases. Although this effect is
obviously dependent on the site symmetry, even being forbidden
in the presence of inversion symmetry, the higher is the induced
electric dipole, the higher is the contribution of the dynamic
coupling for the overall intensity of the spectra. Because the
intensity of the local induced fields is proportional to the polariz-
ability of the occupied chemical environment, more polarizable
sites result in higher contributions of the dynamic coupling
mechanism. In the present case, the introduction of larger
amounts of vanadate groups into the phosphate host culminates
the refractive index. The square reduced matrix elements
5
(2)
7
2
5
(4) 7
2
<
D kU k F > and < D kU k F > have tabulated values
0
2
0
4
5 7
which are, respectively, 0.0032 for the D → F transition and
0
2
5
7
30–33
0
.0023 for the D → F transition.
0 4
The first consideration with regard to the results summarized
in Table 3 is the large increase in the I /I ratio with the
addition of higher vanadium(V) amounts. Many authors try to
interpret this ratio as an indication of the centrosymmetric char-
acter of the occupied site, since the intensity of the D → F
transition is virtually constant (magnetic dipole allowed), and
0
2 01
34
5
7
0
1
5
7
the D → F transition is hypersensitive to the presence of an
inversion centre, which results in low intensities in such cases.
0
2
3,4
This is in fact correct, but it must be borne in mind that not only
the symmetry affects this ratio. The present results are a clear
example that the symmetry alone does not govern the I /I
2 01
ratio, because all the solids display the same tetragonal structure,
and thus D2d sites are occupied in all cases. If the I02/I01 ratio
depended only on the centrosymmetric character, it would be
high when the P/V ratios were close to unity, since these con-
ditions correspond to the most inhomogeneous environments,
and it would be low at both pure compositions. However, this is
not observed. The results clearly demonstrate that the I /I
2 01
intensity ratio is also intrinsically related to the chemical nature
of the occupied sites, thus being affected by the polarizability
and covalence of the ligand–metal bonds. In this sense, the com-
parison of I /I ratios in terms of the symmetry is only valid
when ligands with very similar properties are around the Eu
ions, which seems to be not the case here.
0
0
3+
in an increased polarizability of the Eu sites, which reflects in
a larger dynamic coupling contribution, thereby relaxing the
selection rules for the involved transitions and increasing the
total decay rates.
0
2 01
3+
The behaviour of the luminescence lifetimes and of the radia-
tive and non-radiative decay rates in relation to the composition
of the solids also points to more polarizable chemical environ-
The proposed conclusions are in agreement with the calculated
Judd–Ofelt intensity parameters (Ω
). The Ω parameter is
2
λ
related to the degree of covalence and polarizability of the
3+
3+
ments around Eu at higher vanadium(V) concentrations. As
chemical environment experienced by the Eu ion; higher Ω
2
5
31–35
depicted in Fig. 10, the D luminescence lifetimes are reduced
values point to more covalent and polarizable environments.
In the prepared solids, there is a progressive increase in the Ω
0
3
+
from 2.8 ms in the YPO :Eu sample to 1.0 ms in the YVO4:
2
4
3
+
Eu sample, thus revealing an increase in the overall decay rate
of this excited state. On the other hand, the D0 quantum
values with the increase in the vanadium(V) concentrations, thus
confirming a higher covalence and polarizability for larger vana-
date contents. For the highest vanadium(V) concentrations, for
5
efficiencies are practically independent of the vanadium(V) con-
centration, within the experimental error, and present values
3+
example, the Ω
values are as high as those found for some Eu
2
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6310–6318 | 6317

View Article Online
3
2,33
This is because the PO43− groups are harder
β-diketonates.
4 Phosphor Handbook, ed. W. M. Yen, S. Shinoya and H. Yamamoto, CRC
Press, Boca Raton, 2nd edn, 2007.
3
−
than the VO4 ones, due to the higher stability of the P–O
bonds. So, higher vanadium(V) concentrations result in more
polarizable chemical environments and in more covalent RE–O
5
(a) B. Moine and G. Bizarri, Mater. Sci. Eng., B, 2003, 105, 2;
b) K. Toda, J. Alloys Compd., 2006, 408–412, 665; (c) T. Jüstel and
(
H. Nikol, Adv. Mater., 2000, 12(7), 527.
6
7
C. Ronda, Luminescence: From Theory to Applications, Wiley, Wein-
heim, 2008.
C. H. Huang, G. Zhang, M. Wei, L. X. Huang, X. J. Huang and H.
Y. Shen, Opt. Commun., 2003, 224, 1.
bonds. On the other hand, the Ω parameters are smaller than the
4
Ω ones and have close values in all cases. Some authors have
2
attempted to relate the value of the Ω parameter with the rigid-
4
ity of the occupied chemical environment, which seems to be
8 X. Meng, L. Zhu, H. Zhang, C. Wang, Y. Chow and M. Lu, J. Cryst.
Growth, 1999, 200, 199.
3
+
32,33,35
correct for many Eu coordination compounds.
In the
9
G. Panayiotakis, D. Cavouras, I. Kandarakis and C. Nomicos, Appl. Phys.
A, 1996, 62, 483.
present case, the occurrence of crystalline solids and the struc-
tural similarity between all the obtained powders attest to very
10 L. Niinistö and M. Leskelä, in Handbook on the Physics and Chemistry
of Rare Earths, ed. K. A. Gschneidner, Jr and L. Eyring, Elsevier,
Amsterdam, 1987, ch. 59.
close site rigidities in all cases. This reflects on close Ω values,
4
which are practically the same within the experimental error.
11 H. D. Nguyen, S. Mho and I. H. Yeo, J. Lumin., 2009, 129, 1754.
1
2 (a) J.-F. Dechézelles, G. Mialon, T. Gacoin, C. Barthou, C. Schwob,
A. Maître, R. A. L. Vallée, H. Cramail and S. Ravaine, Colloids Surf., A:
Physicochem. Eng. Asp., 2011, 373, 1; (b) M. Yu, J. Lin and J. Fang,
Chem. Mater., 2005, 17(7), 1783.
Conclusions
In summary, the polymeric precursors method has been success-
fully applied for the obtainment of nanosized phosphovanadate
1
3 D. C. Yu, M. Y. Peng, Q. Y. Zhang, J. R. Qiu, J. Wang and
L. Wondraczek, Sol. Energ. Mater. Sol. Cells, 2011, 95, 1950.
3
−
red phosphors with different VO4 concentrations. Despite the
expected alterations introduced by the vanadate groups to the
structural parameters, such as the increase in the unitary cell
dimensions, all solids have the same tetragonal structure, and
YPO and YVO could form solid solutions in all cases. More-
14 (a) X. Li, M. Yu, Z. Hou, G. Li, P. Ma, W. Wang, Z. Cheng and J. Lin, J.
Solid State Chem., 2011, 184, 141; (b) Z. Hou, P. Yang, C. Li, L. Wang,
H. Lian, Z. Quan and J. Lin, Chem. Mater., 2008, 20(21), 6686.
1
5 M. Yu, J. Lin, Y. H. Zhou, M. L. Pang, X. M. Han and S. B. Wang, Thin
Solid Films, 2003, 444, 245.
6 Y.-S. Chang, F.-M. Huang, Y.-Y. Tsai and L.-G. Teoh, J. Lumin., 2009,
1
4
4
129, 1181.
over, the luminescence characteristics of the obtained solids,
such as emission colour, red-to-orange ratio, luminescence life-
time, and excitation maxima, can be controlled by the vanadium(V)
concentration in the yttrium phosphate host. This is quite impor-
tant with regard to the possible applications of these phosphors
in visualization or illumination devices, so that the luminescent
properties can be adapted to the required features in each case.
Furthermore, these spectroscopic observations could be intrinsi-
cally correlated to the chemical characteristics in each case, and
it can be concluded that the introduction of higher vanadate
amounts into YPO4 leads to a more polarizable chemical
1
1
7 P. C. de Sousa Filho and O. A. Serra, J. Fluoresc., 2008, 18, 329.
8 M. Yu, J. Lin and S. B. Wang, Appl. Phys. A, 2005, 80(2), 353.
19 (a) W. Wang, Z. Cheng, P. Yang, Z. Hou, C. Li, G. Li, Y. Dai and J. Lin,
Adv. Funct. Mater., 2011, 21(3), 456; (b) M. Yu, J. Lin, Z. Wang, J. Fu,
S. Wang, H. J. Zhang and Y. C. Han, Chem. Mater., 2002, 14(5), 2224.
2
2
2
0 C. Li, Z. Hou, C. Zhang, P. Yang, G. Li, Z. Xu, Y. Fan and J. Lin, Chem.
Mater., 2009, 21(19), 4598.
1 V. Buissette, M. Moreau, T. Gacoin, J.-P. Boilot, J.-Y. Chane-Ching and
T. Le Mercier, Chem. Mater., 2004, 16(19), 3767.
2 J. Fang, Y. Guo, G. Lu, C. L. Raston and K. S. Iyer, Dalton Trans., 2011,
4
0, 3122.
23 P. C. de Sousa Filho and O. A. Serra, J. Phys. Chem. C, 2011, 115(3),
36.
24 Y. Dieckmann, H. Cölfen, H. Hofmann and A. Petri-Fink, Anal. Chem.,
009, 81(10), 3889.
6
3
+
environment around Eu and to more covalent RE–O bonds.
2
2
2
5 R. Kijkowska, E. Cholewka and B. Duszak, J. Mater. Sci., 2003, 38, 223.
6 K. Nakamoto, Infrared Spectra of Inorganic and Coordination Com-
pounds, Wiley, New York, 4th edn, 1986.
Acknowledgements
2
7 G. M. Begun, G. W. Beall, L. A. Boatner and W. J. Gregor, J. Raman
Spectrosc., 1981, 11(4), 273.
The authors thank the Brazilian agencies CAPES, CNPq (Proc.
4
80003/2009-2, O.A.S.; 146809/2011-4, J.C.B.), inct-INAMI,
28 L. Li and S. Zhang, J. Phys. Chem. B., 2006, 110(43), 21438.
29 P. A. Santa-Cruz and F. S. Teles, Spectra Lux Software v.2.0, Ponto Quân-
tico Nanodispositivos/RENAMI (Brazil), 2003.
and FAPESP (Proc. 2008/09266-5, P.C.d.S.F.) for financial
support and scholarships. The authors are also grateful to pro-
fessors J.M.A. Caiut and S.J.L. Ribeiro for the Raman spectra,
and Drs R.F. Silva and C.M.C.P. Manso for helpful discussions.
30 I. L. V. Rosa, P. C. de Sousa Filho, C. R. Neri, O. A. Serra, A.
T. Figueiredo, J. A. Varela and E. Longo, J. Fluoresc., 2011, 21, 1575.
31 C. A. Kodaira, H. F. Brito, O. L. Malta and O. A. Serra, J. Lumin., 2001,
01, 11.
1
32 G. F. de Sá, O. L. Malta, C. de Mello Donegá, A. M. Simas, R.
L. Longo, P. A. Santa-Cruz and E. F. Silva, Jr, Coord. Chem. Rev., 2000,
Notes and references
196, 165.
1
(a) C. Feldmann, T. Jüstel, C. Ronda and P. Schmidt, Adv. Funct. Mater.,
003, 35, 511; (b) T. Jüstel, Nanoscale, 2011, 3, 1947.
H. A. Höppe, Angew. Chem., Int. Ed., 2009, 48, 3572.
(a) G. Blasse and B. C. Grabmaier, Luminescent Materials, Springer,
Berlin, 1994; (b) J.-C. G. Bünzli, in Lanthanide Probes in Life, Chemical
and Earth Sciences, ed. J.-C. G. Bünzli and G. R. Choppin, Elsevier,
Amsterdam, 1989, ch. 7.
33 H. F. Brito, O. L. Malta, M. C. F. C. Felinto and E. E. S. Teotônio, in The
Chemistry of Metal Enolates, ed. J. Zabicky, Wiley-Blackwell, West
Sussex, 2009, ch. 3.
34 O. L. Malta and L. D. Carlos, Quim. Nova, 2003, 26, 889; S. F. Mason,
R. D. Peacock and B. Stewart, Chem. Phys. Lett., 1974, 29, 149.
35 V. Divya, R. O. Freire and M. L. P. Reddy, Dalton Trans., 2011, 40,
3257.
2
2
3
6318 | Dalton Trans., 2012, 41, 6310–6318
This journal is © The Royal Society of Chemistry 2012