M. Karbowiak, J. Cichos / Journal of Alloys and Compounds 673 (2016) 258e264
259
[
19,20] are based on X-ray diffraction studies of nanofluoride
phases formed in the BaF -YF system during the course of pre-
cipitation from aqueous solution, whereas other authors reported
on nanocrystals of ‘BaYF ’ composition with grain size as small as
e10 nm, obtained by thermolysis of lanthanide trifluoroacetates
2.1.3. Synthesis of ligand-free nanoparticles
Eu 2þxþy nanoparticles without any surface ligands
were prepared according to the method described in Ref. [22].
Appropriate amounts of BaCl and Ln(NO $6H O (2 mmol in total,
e.g.1.00 mmol of BaCl , 0.98 mmol of Y(NO $6H O and 0.02 mmol
were used for preparation of
2.5) were dissolved in 15 cm of water in a teflon
2
3
Ba1ꢀxꢀy
Y
x
y
F
5
2
3
)
3
)
2
5
2
3
3
2
in high boiling organic solvents. The latter non-equilibrium con-
ditions are much different from those during co-precipitation from
aqueous media at room temperature, which may affect structure
and morphology of resulted nanocrystals. However, there are no
systematic studies of barium-yttrium nanofluoride phases obtained
by thermolysis.
One should be aware that whether ‘BaYF
solid state solution is not only a problem of a correct nomenclature.
If the cubic fluorite-type Ba1ꢀx 2þx phase of variable composition
of
3 3 2
Eu(NO ) $5H O
Ba0.50Y0.49Eu0.01F
3
bottle. Afterwards 10% excess (5.5 mmol), relative to stoichiometry,
of NH
4
F was added and the bottle was placed on a heating mantle
ꢁ
with a magnetic stirrer and heated to 90 C. After 30 min the bottle
was sealed with a cap and left for 24 h in these conditions. Fluorides
were collected by means of centrifugation and washed three times
with water and once with ethanol and dried in the air.
5
’ is a compound or a
x
Y F
exists then it is likely that x ¼ 0.5 is not the optimal composition in
regard of morphological or luminescent properties. Yet, to the best
of our knowledge, compositions other than x ¼ 0.5 have not been
2.2. Characterization
XRD experiments were carried out on Bruker D8 Advance
diffractometer equipped with Cu X-Ray lamp and Vantec detector.
5 5
probed in the studies of so-called ‘BaYF ’ or ‘BaGdF ’.
ꢁ
ꢁ
In this paper we report the results of systematic studies of
nanophases formed in BaF -YF system synthesized using the
thermal decomposition method. Besides the X-ray diffraction
XRD) the photoluminescence (PL) spectroscopy was employed for
Patterns were collected in temperature series from 25 C to 1000 C
in argon atmosphere and the samples were placed on Pt/Rh heater.
Lattice parameters were calculated using Bruker TOPAS software.
Corrected emission and excitation spectra were recorded on an
Edinburgh Instruments FLSP 920 spectrofluorimeter.
2
3
(
3
þ
characterization of samples doped with optically active Eu and
3
þ
Er ions. The spectroscopic properties of oleic acid (OA) capped
NCs are compared with those of ligand-free NCs co-precipitated
from aqueous solution. Moreover, we indicate that intensity of
upconversion emission observed under excitation at 980 nm for the
3. Results and discussion
3.1. Structure and morphology
3
þ
3þ
x
Ba1ꢀxY F2þx phase co-doped with Yb and Er ions is strongly
composition dependent and, interestingly, its maximum does not
correspond to x ¼ 0.5. To the best of our knowledge, this study is the
first in which PL spectroscopy is used to identify phases formed in
TEM images shown in Fig. 1 indicate that all samples are
composed from non-aggregated particles with average size ranging
from 5 to 15 nm. At the same time the morphology and dispersity of
NCs synthesized under the same conditions strongly depend on
2 3
BaF -YF system.
3
REF content in the initial composition.
2
. Experimental section
The selected area diffraction patterns (SAED, Fig. 1) and the XRD
x 2þx
patterns (Fig. S1, Supplementary data) confirm that all Ba1ꢀxY F
2.1. Chemicals and materials
(x ¼ 0.20e0.90) samples are composed of single cubic fluorite type
phase only. In Fig. 2 the lattice parameters a determined for OA-
Lanthanide oxides were purchased from Stanford Materials and
3
capped NCs are presented as a function of YF content and
were of 99.99% purity grade. Oleic acid (90%, technical), octadec-1-
en (90%, technical) and all other chemicals (ACS purity grade) came
from SigmaeAldrich and were used for samples preparation
without any purification.
compared with those previously reported [20] for ligand-free
samples precipitated from aqueous solution. The plot for OA-
caped NCs clearly reveals two regions of linear dependence be-
3
tween a and YF content (x): the a constant decreases mono-
tonically between x ¼ 0.20 and x ¼ 0.40, as well as between x ¼ 0.5
2
.1.1. Preparation of trifluoroacetates precursors
Trifluoroacetic acid salts were prepared by dissolution of
and x ¼ 0.7, but for x ¼ 0.50 it is larger by 0.048 Å then for x ¼ 0.40.
3
For samples with relative YF content from 45 to 75 mol% the
appropriate lanthanide oxide (or barium hydroxide octahydrate) in
excess of trifluoroacetic acid at 90 C. After total reconstitution of
substrates the excess of acid was evaporated near to dryness and
linear trend obtained for OA-caped NCs matches very well that
recognized for NCs precipitated from aqueous solution [20]. By
analogy to results of Fedorov et al. [20] the cubic phases corre-
ꢁ
the beakers were placed in vaccum desiccator over P
2
O
5
for 2 days.
sponding to the first (YF
second (YF content from ~47 to 48e~75 mol%) region of linear
dependence in Fig. 1 will be denoted as F-phase and F -phase,
respectively. If YF content is larger than 80 mol% formation of F
phase is accompanied by precipitation of the second cubic phase
depicted in Ref. [20] as (H O)Y 10$nH O. Some discrepancy be-
tween our results for OA-caped NCs and those presented therein
[20] for ligand-free NCs occurs only in the composition region up to
3 2
30 mol% of YF . This is due to the fact that in aqueous solutions BaF
3
content up to ~47e48 mol%) and the
3
2
.1.2. Synthesis of oleic acid (OA) capped nanoparticles
Appropriate amounts of trifluoracetates (2 mmol in total; e.g.
63.0 mg of Ba(CF CO , 289.1 mg of Y(CF CO $3H O, 203.6 mg
CO $3H O and 22.4 mg of Er(CF CO $3H O were used
for preparation of Ba0.50 2.5) were placed in a three
1
3
1
-
3
3
2
)
2
3
2
)
3
2
of Yb(CF
3
2
)
3
2
3
2
)
3
2
3
3
F
2
Y
0.30Yb0.18Er0.02
F
3
neck round bottom flask and 20 cm of a mixture of oleic acid and
octadec-1-en (1:1 vol) was added. The flask was placed in a heating
ꢁ
mantle and the solution was degassed in vacuum at 100e110 C for
precipitates as the second phase besides the F-phase, the latter
1
h with periodic Ar refills. Afterwards the flask was placed in a
having the constant lattice parameter value a ¼ 5.933 Å. Since
heating salt bath (KNO
3
:NaNO
3
, 1:1 weight ratio) preheated to
2
formation of BaF is not observed in the procedure used by us for
ꢁ
3
60 C. 30 min after trifluoroacetates decomposition (which occurs
synthesis of OA-capped NCs the lattice parameters for single cubic
fluorite type phases formed in this region vary monotonically and
the obtained linear correlation matches well the trend found using
ꢁ
at 280e290 C) the flask was pulled out of the bath and left to cool
down naturally. Nanoparticles were precipitated with excess of
ethanol, centrifuged and washed five times by means of dissolution
in toluene and precipitation with ethanol (about 1:10e1:8 v/v).
data of [20] for co-precipitated samples with 43 and 35 mol% YF
3
,
and data of [21] for sample with 25 mol% YF synthesized by solid
3
Karbowiak, Miros?aw
