Wet chemical synthesis of LiBaF3 phosphor

Article abstract of DOI:10.1016/j.jallcom.2013.04.163

LiBaF₃ has great potential applications as X-ray storage phosphor, slow neutron imaging, scintillator, vacuum ultraviolet (VUV) optical lithography, etc. Conventionally, LiBaF₃ is prepared by solid state reaction between the constituent fluorides. However, the preparation of phase pure material and especially single crystals is rather tricky due to incongruent melting. For the first time, a wet chemical preparation of rare earth activated LiBaF₃ is described here. As precipitated powders containing Ce ³⁺ or Tb³⁺ exhibited characteristic luminescence. For observing Eu²⁺ emission, it was necessary to heat the powders in a reductive atmosphere. It is suggested that phosphors prepared by this method may prove useful in applications like OSL, X-ray imaging, etc. which do not require large single crystals.

Full text of DOI:10.1016/j.jallcom.2013.04.163

Journal of Alloys and Compounds 579 (2013) 165–168
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Wet chemical synthesis of LiBaF3 phosphor
a
b
Vartika S. Singh a, , C.P. Joshi , S.V. Moharil
⇑
a Physics Department, Shri Ramdeobaba K.N. Engineering College, Katol Road, Nagpur 440 013, India
b Department of Physics, R.T.M. Nagpur University, Nagpur 440 010, India
a r t i c l e i n f o
a b s t r a c t
Article history:
LiBaF3 has great potential applications as X-ray storage phosphor, slow neutron imaging, scintillator, vac-
uum ultraviolet (VUV) optical lithography, etc. Conventionally, LiBaF3 is prepared by solid state reaction
between the constituent fluorides. However, the preparation of phase pure material and especially single
crystals is rather tricky due to incongruent melting. For the first time, a wet chemical preparation of rare
earth activated LiBaF3 is described here. As precipitated powders containing Ce3+ or Tb3+ exhibited char-
acteristic luminescence. For observing Eu2+ emission, it was necessary to heat the powders in a reductive
atmosphere. It is suggested that phosphors prepared by this method may prove useful in applications like
OSL, X-ray imaging, etc. which do not require large single crystals.
Received 26 January 2013
Received in revised form 19 April 2013
Accepted 22 April 2013
Available online 30 April 2013
Keywords:
Fluoroperovskite
Fluoride
Ó 2013 Elsevier B.V. All rights reserved.
Phosphors
Luminescence
1. Introduction
neutron imaging [27], vacuum ultraviolet(VUV) optical lithography
[18], and other luminescent devices [28].
LiBaF3 was discovered by Ludekens and Welch way back in 1952
[1]. BaLiF3 prepared with solid state reaction and LiF-BaF2 as a raw
material has been reported since the 1970s [2]. It was found to be
incongruently melting and crystals could be grown from melt of
non-stoichiometric compositions [3]. Luminescence of Eu2+ was
studied quite early and found to be interesting as f–f line emission
was noticed [4–6]. Luminescence of other lanthenides like Ce3+
[6,7], Yb2+ [8], Er3+[2], Sm2+ [9,10] has also been studied subse-
quently. Interesting luminescence properties were also obtained
with 3d activators like Ni2+ [11–13], Cr3+ [14], Cu+ [15], Co2+, [3]
Mn2+ [16,17]. Due to matching sizes, 3 d ions can be substituted
at Li+ site. The initial investigations were carried out with the idea
that it is a wide band-gap dielectric [18] and it might be a good host
lattice for incorporating the ions with a lasing effect in the UV as
well as NIR region. During crystal growth of fluorides, PbF2 is used
as scavenger, and emission of Pb+ as well as Pb2+ was studied in
LiBaF3 in which Pb was doped intentionally or otherwise [19,20].
Though initial studies were related to optical properties relevant
for obtaining solid state lasers, other applications were also discov-
ered soon. Pure LiBaF3 [21,22], as well as Ce-doped and Rb-doped
crystals, are scintillators, enabling one to discriminate the thermal
neutron and gamma radiations [23]. Fast (ns), cross luminescence
at shorter wavelengths is seen with gamma rays but not neutrons,
while STE emission can be seen in scintillations with gamma rays as
well as neutrons [24]. It is well-established [25] that LiBaF3 has
great potential applications in X-ray storage phosphors [26], slow
Growth of LiBaF3 single crystals is not straightforward as this
compound is incongruently melting and crystals have to be grown
from melt of non-stoichiometric compositions. On the other hand
in several applications like X-ray imaging screens, TL/OSL dosime-
try there is no need to grow single crystals and microcrystalline
powders suffice. LiBaF3 powders are usually prepared by solid state
reaction between LiF and BaF2, though in relatively recent works
novel methods like pyrolysis of metal trifluoroacetates [29] solvo-
thermal [30,25] sol–gel [31] have been attempted. When phos-
phors are prepared by the conventional solid state route
involving heating in air, usually unwanted impurities like OHꢀ,
Oꢀ2 , etc. get incorporated. Such impurities can be quite harmful to
the luminescence processes. Especially, fluorides are highly sus-
ceptible to hydrolysis [32]. Moreover, they are highly reactive
and can attack the crucible material during high temperature pro-
cessing. Use of freshly synthesized powders and inert atmosphere
becomes necessary to prevent the hydrolysis and the reaction with
the crucible. Scavengers like PbF2 have also been used to get rid of
OHꢀ. In recent years, some wet chemical syntheses for preparation
of OHꢀ free fluorides have been described [33–35]. The fact that
the presence of chlorine ions prevents hydrolysis is exploited in
these methods. We have attempted synthesis of LiBaF3 using
similar methods. The results are reported in this paper.
2. Experimental
All the reagents used were of the Analytical Reagent grade. First we tried to
synthesize LiBaF3 using the wet-chemical method developed successfully for other
fluorides [33–35]. The method exploits the fact that chlorine ions prevent hydroly-
sis. Accordingly, aqueous solutions of LiCl and BaCl2 were mixed in equimolar
⇑
Corresponding author. Tel.: +91 712 2283708; fax: +91 712 2249875.
E-mail address: svmoharil@yahoo.com (V.S. Singh).
0925-8388/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jallcom.2013.04.163

166
V.S. Singh et al. / Journal of Alloys and Compounds 579 (2013) 165–168
proportions. Solutions were heated to 85 °C. In a typical experiment, 0.352 g of LiCl
was dissolved in minimum amount of triply distilled, deionized water. Solution of
2 g of BaCl2 was similarly prepared. Two solutions were mixed, stirred and heated
to 85 C and HF (48%) was added drop wise till precipitation completed. Resulting
precipitate was filtered, washed repeatedly with distilled water, dried and used
in further experiments. However, though other double fluorides had been success-
fully synthesized by this method, the method did not yield LiBaF3, BaFCl was
formed instead. It was thought that by eliminating use of chlorides, BaFCl formation
can be avoided and LiBaF3 may be obtained. Accordingly, nitrates were used in place
of chlorides. Though barium nitrate has limited solubility (10.3 g/100 ml), sufficient
quantity could be dissolved in hot distilled water. However, no precipitation re-
sulted from nitrate solutions.
In the second method, metal acetates were freshly prepared by dissolving cor-
responding carbonates in acetic acid. In a typical experiment, 0.66 g of lithium ace-
tate and 2.73 g of barium acetate were dissolved separately in minimum amount of
triply distilled, deionized water. Both acetates are readily soluble (Barium acetate
79.2 g/100 ml, lithium acetate (45 g/100 ml). Rest of the procedure was same as
in method 1. This method met with partial success. LiBaF3 was formed, but BaF2
was the major phase.
In the third method, the precipitant was changed and reverse strike was used.
Hot metal chloride solutions were added to the precipitant solution. KF was used as
the precipitant. KF was freshly synthesized by neutralizing HF with K2CO3. In a typ-
ical experiment, 0.352 g of LiCl was dissolved in minimum amount of triply dis-
tilled, deionized water. Solution of 2 g of BaCl2 was similarly prepared. Two
solutions were mixed, stirred and heated to 85 C. Solutions of dopant chlorides
were also added in the desired concentrations. Amount of dopant refers to this va-
lue and actual concentration incorporated was not measured. This was then added
to hot aqueous solution containing 12 g KF in 50 ml distilled water. This amount is
nearly seven times more than required by stoichiometry. However, this was the
optimum amount for obtaining phase pure LiBaF3. Use of smaller quanitity leads
to contamination of LiBaF3 phase by BaF2. Use of excess KF may have also helped
in prevention of BaFCl formation. Rest of the procedure was same as in method 1.
As prepared Eu-doped sample did not show PL emission. Dopant is not well dis-
persed or it is not incorporated in divalent form. For reducing it to Eu2+ the powder
was heated at 750 C for 1 h in a reducing atmosphere provided by burning charcoal.
X-ray diffraction patterns were recorded on a Philips PANalytical X’pert Pro dif-
fractometer. Photoluminescence spectra were recorded on a Hitachi F-4000 spec-
tro-fluorimeter with spectral slit width of 1.5 nm in the spectral range 220–700 nm.
Fig. 1. XRD pattern of attempted LiBaF3 using method
1 (precipitation from
aqueous chloride solution using HF) When precipitation synthesis of LiBaF3 was
attempted from aqueous solutions of LiCl and BaCl2, using HF as precipitating agent,
BaFCl is formed as shown by excellent matching with ICDD data file 76-1374.
3. Results and discussions
Fig. 1 shows the XRD pattern of the compound prepared with
chlorides as starting materials and HF as precipitating agent. XRD
pattern did not match with that for LiBaF3, but resembles that for
BaFCl (ICDD 76-1374). Relative intensities are somewhat different.
It appears that the major phase is BaFCl and small amount of LiBaF3
might have been formed.
The literature method [35] for wet-chemical preparation of
fluorides thus fails for LiBaF3. To avoid formation of BaFCl, we tried
nitrates as the starting materials and HF as precipitating agent,
however no precipitate was obtained. Precipitation occurred when
acetates were used as starting materials. Fig. 2 shows the XRD
pattern of the compound prepared with acetates as starting mate-
rials and HF as precipitating agent (method 2). Stick patterns of
Fig. 2. XRD pattern of attempted LiBaF3 using method
2 (precipitation from
LiBaF3 (ICDD 18-0176) and BaF2 (ICDD 88-2466) are also shown
aqueous acetate solution using HF). When precipitation synthesis of LiBaF3 was
attempted from aqueous solutions of lithium and Barium acetates, using HF as
precipitating agent, mixed phase compound was formed. The phases cane be
identified as BaF2 (marked by crosses, ICDD data file 88-2466), and LiBaF3 (marked
by tick marks, ICDD data file 18-0716).
p
for the comparison. Several lines (marked by ) match with LiBaF3.
However, stronger lines (marked by crosses) match with BaF2
pattern. Best results were obtained for the third method in which
reverse strike was used. Hot metal chloride solutions were added
to the precipitant solution. KF was used as the precipitant. XRD
pattern of the compound so prepared matches excellently with
that for LiBaF3 (ICDD 18-0176). Line at 12.672 characteristic of
BaFCl is totally absent. Use of excess KF may have helped in pre-
vention of BaFCl formation. When minutely observed, very weak
lines of BaF2 can be seen (Fig. 3). All subsequent samples were thus
prepared by method 3.
Having successfully prepared LiBaF3 by co-precipitation, next
step is to attempt preparation of LiBaF3 based phosphors. We at-
tempted incorporation of Ce3+. Eu2+, Tb3+ activators. Fig. 4 (curve
a) shows PL emission of Ce3+ activated LiBaF3. It is remarkable that
characteristic emission is observed in as-precipitated powders
without any further thermal treatment except that used for drying
purpose. The broad emission band peaks at 358 nm. The corre-
sponding excitation spectrum (Fig. 4, curve b) contains prominent
bands at 290 and 261 nm with shoulders around 308, 250 and
238 nm. These values, and in general the shapes of the PL curves
are in excellent agreement with those reported by Tan and Shi
[36]. However, it may be mentioned that different workers have re-
ported various values for emission and excitation bands of Ce3+ in
LiBaF3 (Table 1). Reason for such diversity of reported results can
be understood by considering the crystal structure of LiBaF3. LiBaF3
has an inverted perovskite cubic structure with space group Oh
(Pm3m), where the monovalent Li+ ion is at the center of a F6 octa-
hedron and the Ba2+ divalent ion locates at the corners of a cube

V.S. Singh et al. / Journal of Alloys and Compounds 579 (2013) 165–168
167
case of charge compensation exists, the symmetry of the Ce3+ site
is lowered from cubic to C3v, C4v or even lower. Existence of as
many as five bands/shoulders in the excitation spectrum also indi-
cates lower symmetry. Various local environments are thus possi-
ble for Ce3+ in LiBaF3 host. This may be responsible for observed
variations in literature results on PL of Ce3+. Ce3+ excitation results
2
from transition from ground state F5/2 of 4f1 configuration to ex-
cited states of 5d1 configuration. In crystal, the excited states
may broaden and split into up to 5 bands. Number and positions
of the bands depend on the site symmetry and coordination [37].
Emission comes from lowest of the excited states of 5d1 configura-
2
tion to F5/2
,
2F7/2 states of 4f1 configuration. The Stokes shift also
strongly depends on the local environment of Ce3+. Thus, we pre-
sume that the diversity of the published results on PL spectra of
LiBaF3:Ce is due to such effects.
From ESR measurements Yamaga et al. [38,39] suggested two
types of defect centers associated with Ce3+ in LiBaF3 which were
designated as T1 and T2. They assigned centers T1 and T2 to Ce3+
at the 12-fold coordinated Ba2+ sites associated with, respectively,
a substitutional Li+ and Ba2+ vacancy (VBa) at the nearest Ba2+ sites
along [001] axis because of charge compensation. Recently, Wen-
Chen et al. [40] questioned this assignment and suggested that
centers T1 and T2 are the substitution of Ce3+ for Ba2+ with, respec-
tively, a nearest interstitial Fꢀ and a nearest Ba2+ vacancy along C4
axis because of charge compensation.
Fig. 3. XRD pattern of LiBaF3 using method 3. An excellent match with ICDD data
file 18-0716 is seen.
It is thus seen that defect centers associated with Ce3+ in LiBaF3
can be of several types and their detailed structures debatable.
However, it is not the objective of the present paper to review
the debate over these structures. The main objective is to show
that LiBaF3 based phosphors can be prepared by simple wet chem-
ical procedure.
Tb3+ can enter LiBaF3: lattice much the same way as Ce3+. In
contrast to LiBaF3:Ce3+, PL emission of Tb3+ is very weak (Fig. 4,
curve c). The reason is that f–d excitation bands of Tb3+ are located
at very high energies lying outside the range of our instrument. Re-
cently, Zhu et al. [41] reported f–d excitation of LiBaF3:Tb3+ at
217 nm. All the same, upon 220 nm excitation, shortest wave-
length possible with our instrument, narrow emission bands
around 543 and 487 nm corresponding to transitions 5D4 ? 7F5,
5D4 ? 7F6 can be seen just above the background noise. Excitation
spectrum (Fig. 4, curve c) shows some f–f structure and tail of the
f–d band. Significant fact is that the as-prepared phosphor shows
characteristic Tb3+ emission, although feeble, without any thermal
treatment.
Fig. 4. a > Ce3+ Emission for 290 nm excitation. b > Excitation for 358 nm emission
of LiBaF3:Ce3+ c > Tb3+ Emission for 220 nm excitation. d > Excitation for 543 nm
emission of LiBaF3:Tb3+ e > Eu2+ Emission for 320 nm excitation. f > Excitation for
410 nm emission of LiBaF3:Eu2+
.
Table 1
In contrast to LiBaF3:Ce3+, and LiBaF3:Tb3+, as-prepared powders
of LiBaF3:Eu did not show Eu2+ emission. Heating in reductive
atmosphere provided by burning charcoal yielded a strongly emit-
ting LiBaF3:Eu2+ phosphor. PL spectra are included in Fig. 4. Eu2+
emission is in form of a narrow line at 358 nm and a broad band
around 411 nm (Fig. 4, curve e). The corresponding excitation spec-
trum (Fig. 4, curve f) shows two prominent bands around 250 and
310 nm with some structure, particularly a staircase structure for
the latter. These results are in excellent agreement with the litera-
ture. It may be suspected that presence of small amount of BaF2
can affect the PL results. However, Eu2+ does not show any PL in
BaF2 due to photoionization. The ground state of the Eu2+ ion is
the spherically symmetrical 8S7/2. LiBaF3, Eu2+ is at 12-coordinated,
Ba2+ substitutional site (Fig. 5). The crystal field splits the Eu2+
4f65d1 electronic configuration into the lower t2g and the upper
eg levels, respectively. Depending on the strength of the crystal
field, the absorption or the excitation spectrum of octahedrally
coordinated Eu2+ ion is characterized by two broad and well-iso-
lated bands. The low and high-energy bands are associated with
the 4f7 [8S7/2]–4f65d1[t2g] and the 4f7 [8S7/2]–4f65d1[eg] optical
transitions, respectively. The latter is responsible for excitation
band around 250 nm in LiBaF3:Eu2+. The low energy excitation
Summary of Reported data on Ce3+ PL in LiBaF3.
Extn. (nm)
Emn. (nm)
Refs.
254
287,295
361
312, 328
340
287, 300
367
280, 320
358
Ref. [25]
Ref. [36]
Ref. [40]
Ref. [30]
Ref. [42]
Ref. [41]
Ref. [6]
243, 257, 297
204, 218, 240 and 250
254 and 297
235, 256
265, 299
240
261, 290
This Work
with the 12-fold environment site (Fig. 5), resulting in a different
field interaction from the classic perovskite structure. Ce3+ is triva-
lent, and size wise may be expected to be at Ba substitutional site.
This requires charge compensation which can be achieved in a
variety of ways. Most obvious being incorporation of Li+ vacancy,
Ba2+ vacancies, O2ꢀ substitution at fluorine site or interstitial fluo-
rine. Tan and Shi [36] found that oxygen containing LiBaF3:Ce gives
emission at longer wavelengths (370–379 nm). Considering sizes
and valencies, the occurence of Ba vacancies in A site of perovskite
is more probable. Again, the charge compensating defect may be
located close to Ce3+ or a few lattice spacings away. Whichever

168
V.S. Singh et al. / Journal of Alloys and Compounds 579 (2013) 165–168
Fig. 5. Co-ordination sphere for cations in LiBaF3 8-coordination of Li+ and 12-coordination of Ba2+ are depicted.
band due to 4f7 [8S7/2]–4f65d1[t2g] transition exhibits the charac-
teristic ‘‘staircase’’ spectrum, which retain the character of the se-
ven (Eu3+) 4f6 levels (7F0–6) [42,43]. The staircase spectrum is due
to the transition from the Eu2+ 8S7/2 ground state to the seven 7FJ
(J = 0–6) multiplets of the excited 4f6 [7FJ(J = 0–6)]5d1 electronic
configuration. Observed excitation bands around 288, 298,
312 nm and the shoulder around 332 nm in LiBaF3:Eu2+are due
to such transitions. Band emission around 411 nm is due to transi-
tion from the lowest band of 4f65d1 configuration to the ground
[13] M. Mortier, B. Piriou, J.Y. Buzare, M. Rousseau, J.Y. Gesland, Phys. Rev. B 67
(2003) 115126.
[14] M. Mortier, J.Y. Gesland, B. Piriou, J.Y. Buzare, M. Rousseau, Opt. Mater. 4
(1994) 115.
[15] S. Lizzo, A. Meijerink, G.J. Dirksen, G. Blasse Chem, Phys. Lett. 253 (1996) 108.
[16] A. Trueba, J.M. García-Lastra, M.T. Barriuso, J.A. Aramburu, M. Moreno, Phys.
Rev. B Condens. Matter Phys. 78 (2008) 075108.
[17] Bastian Henkea, Uldis Rogulis, Stefan Schweizer, Radiat. Meas. 43 (2008) 319.
[18] A. Bensalah, K. Shimamura, T. Fujita, H. Sato, M. Nikl, T. Fukuda, J. Alloys Comp.
348 (2003) 258.
[19] L.K. Aminov, S.I. Nikitin, N.I. Silkin, A.A. Shakhov, R.V. Yusupov, R.Yu.
Abdulsabirov, S.L. Korableva, J. Phys. Condens. Matter 18 (2006) 4985.
[20] L. Prado, N.D. Vieira /jr, S.L. Baldochi, S.P. Morato, J.P. Denis, N. Tercier, B.
Blanzat, J. Phys. Chem.Solids 57 (1996) 413.
[21] M.J. Knitel, P. Dorenbos, J.T.M. De Haas, C.W.E. Van Eijk, Nucl. Instrum.
Methods A 374 (1996) 197.
8
state S7/2 state of 4f7configuration. The line emission at 358 is
due to 6Pj ? 8S7/2 transitions between states of 4f7configuration.
[22] C.M. Combes, P. Dorenbos, R.W. Hollander, C.W.E. van Eijk, Nucl. Instrum.
Methods Phys. Res. A416 (1998) 364.
4. Conclusions
[23] P.L. Reeder, S.M. Bowyer, J. Radioanal. Nucl. Chem. 248 (2001) 707.
[24] M.J. Knitel, P. Dorenbos, J.T.M. de Hass, C.W.E. van Eijk, Nucl. Instrum. Methods
A 374 (1996) 197.
[25] Guoxian Zhu, Qu Yang, Xiaobo Shi, Wenzhi Zheng, Yingliang Liu, J. Rare Earths
30 (2012) 985.
[26] C.T. Xia, C.S. Shi, J. Funct. Mater. Dev. 2 (1996) 59.
[27] P. Kulis, U. Rogulis, M. Springis, I. Tale, Nucl. Instrum. Methods Phys. Res. A 509
(2003) 56.
[28] P.L. Reeder, S.M. Bowyer, J. Radioanal. Nucl. Chem. 248 (2001) 707.
[29] S. Fujihara, Y. Kishiki, T. Kimura, J. Solid State Chem. 177 (2004) 1032.
[30] Ruinian. Hua, Bingfu. Lei, Demin. Xie, Chunshan. Shi, J. Solid State Chem. 175
(2003) 284.
[31] Shinobu Fujihara, Kazuaki Tokumo, J. Fluorine Chem. 130 (2009) 1106.
[32] N. Kodama, T. Hoshino, M. Yamaga, N. Ishizawa, K. Shimamura, T. Fukuda, J.
CrystalGrowth 229 (2001) 492.
[33] A. Mech, M. Karbowiak, L. Kepinski, A. Bednarkiewicz, W. Strek, J. Alloys Comp.
380 (2004) 315.
A simple co-precipitation method using KF as precipitating
agent is developed for the preparation of LiBaF3. The method is
much simpler compared to the conventional solid state reaction,
especially in view of the fact that LiBaF3 is incongruently melting.
Activation by rare earth ions is also demonstrated. For Ce3+ and
Tb3+
, characteristic emission was observed in as-precipitated
powders, without taking recourse to any thermal treatment. For
Eu2+ activation, heating in reductive atmosphere was necessary.
Phosphors prepared by this method may prove useful in applica-
tions like OSL, X-ray imaging, etc. which do not require large single
crystals.
[34] M. Karbowiak, A. Mech, A. Bednarkiewicz, W. Strek, J. Alloys Comp. 380 (2004)
321.
References
[35] P Boutinaud, R Mahiou, Jean-Claude Cousseins, M Bouderbala, J. Mater. Chem.
9 (1999) 125.
[36] Y. Tan, C. Shi, J. Solid State Chem. 150 (2000) 178.
[37] P. Dorenbos, J. Lumin. 91 (2000) 155.
[38] M. Yamaga, M. Honda, N. Kawamata, T. Fujita, K. Shimamura, T. Fukuda, J.
Phys. 13 (2001) 3461.
[39] M. Yamaga, M. Honda, K. Shimamura, T. Fukuda, T. Yosida, J. Phys. 12 (2000)
5917.
[40] Wen.-Chen. Zheng, Yang. Mei, Hong.-Gang. Liu, Yu.-Guang. Yang, J. Alloys
Comp. 551 (2013) 224.
[41] Guoxian Zhu, Qu Yang, Xiaobo Shi, Wenzhi Zheng, Yingliang Liu, Opt. Mater., in
press. doi.: 10.1016/j.optmat.2013.01.044.
[42] M.J. Freiser, S. Methfesskl, F. Holtzberg, J. Appl. Phys. 39 (1968) 900.
[43] F.M. Ryan, W. Lehmann, D.W. Feldman, J. Murphy, J. Electrochem. Soc. 121
(1974) 1475.
[1] W.L.W. Ludekens, A.J.E. Welch, Acta Cryst. 5 (1952) 841.
[2] J. Yan, Z. Li, L. Wang, L. Cao, S. Wang, C. Shi, J. Rare Earth 26 (2008) 48.
[3] R. Leckebusch, A. Neuhaus, K. Recker, J. Cryst. Growth 16 (1972) 10.
[4] B. Tanguy, P. Merle, M. Pezat, C. Fouassier, Mat. Res. Bull. 9 (1974) 831.
[5] J.L. Sommerdijk, J.M.P.J. Versteegen, A. Bril, J. Lumin. 10 (1975) 411.
[6] M. Yamaga, T. Imai, K. Shimamura, T. Fukuda, J. Cryst. Growth 229 (2001) 487.
[7] M. Nikl, E. Mihokova, Z. Malkova, A. Vedda, M. Martini, K. Shimamura, T.
Fukuda, Phys. Rev. B66 (2002) 184101.
[8] M. Henke, J. Perbon, S. Kuck, J. Lumin. 87 (2000) 1049.
[9] A. Meijerink, G.J. Dirksen, J. Lumin. 63 (1994) 189.
[10] A. Gros, F. Gaume, J.C. Gacon, J. Solid State Chem. 36 (1981) 324.
[11] E. Martins, N.D. Vieira Jr., Si. Baldochi, S.P. Morato, J.Y. Gesland, J. Lumin. 62
(1994) 281.
[12] M. Mortier, J.Y. Gesland, M. Rousseau, F. Auzel, D. Meichenin, Radiat. Eff.
Defects Solids 136 (1995) 65.