Investigation of KYF4: Yb, Er // KYF4 nanocrystals - Mechanism of the KYF4 formation

Article abstract of DOI:10.1134/S0036023609120122

This paper presents our investigation of the formation mechanism of cubic phase Yb³⁺ and Er ³⁺ doped nanoscale KYF₄. We generated Yb, Er doped KYF₄ nanoparticles, modified with undoped KYF₄ following

Full text of DOI:10.1134/S0036023609120122

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
ISSN 0036-0236, Russian Journal of Inorganic Chemistry, 2009, Vol. 54, No. 12, pp. 1914–1920. © Pleiades Publishing, Inc., 2009.
Original Russian Text © H. Schäfer, P. Ptacek, K. Hickmann, M. Prinz, M. Neumann, M. Haase, 2009, published in Zhurnal Neorganicheskoi Khimii, 2009, Vol. 54, No. 12,
pp. 1997–2002.
COORDINATION COMPOUNDS
Investigation of KYF4: Yb, Er // KYF4 Nanocrystals – Mechanism
of the KYF4 Formation1
H. Schäfera, P. Ptaceka, K. Hickmanna, M. Prinzb, M. Neumannb, and M. Haasea
a Institute of Chemistry, University of Osnabrück, Barbarastraβe 7, D-49076 Osnabrüeck, Germany
b Department of Physics, University of Osnabrück, Barbarastraβe 7, D-49076 Osnabrüeck, Germany
Received January 14, 2009
Abstract—This paper presents our investigation of the formation mechanism of cubic phase Yb3+ and Er3+
doped nanoscale KYF4. We generatedYb, Er doped KYF4 nanoparticles, modified with undoped KYF4 follow-
ing a procedure described in our previous paper. The lanthanide chlorides, ammonium fluoride and potassium
alkoxide of N-(2-hydroxyethyl)-ethylenediamine in HEEDA were used as precursors. In order to find some
information concerning to the reaction mechanism, we performed possible partial reactions to generate the pos-
sible intermediate products. With the help of high-resolution X-ray photoelectron spectroscopy we additionally
found further evidence for the successful generation of the passivating shell around the lanthanide doped KYF4
particles.
DOI: 10.1134/S0036023609120122
1
1. INTRODUCTION
a diphosphonic acid. We successfully prepared up-con-
version nanocrystals which can be easily dispersed in
water [10]. For bioprobes, the size of the targeted bio-
molecules (proteins, oligonucleotids) in cells and tis-
sues range from several nanometers to tens of nanome-
ters. An optimal universal bioprobe therefore should
have a size below 15 nm.
Our previous synthesis resulted in water soluble
highly efficient upconverting nanoparticles with an
average particle size of approximately 26 nm in diame-
ter. The developed synthetic procedure was a signifi-
cant advancement in the synthesis of efficient upcon-
verting water soluble nanoparticles [10], but neverthe-
less for most possible applications the particles are
oversized. On the one hand, particles have to be small
to be suitable for applications, on the other hand one
essential drawback of small particles is their low
upconversion luminescence efficiency because of the
high surface to volume ratio. In this case defect surface
states as well as coordinating solvent molecules play a
dominating role with regard to optical efficiency.
Quenching of the fluorescence could be reduced by
growing a passivating shell around the particles which
separates the light emitting ions from solvent molecules
and ligands. The properties of passivating inorganic
surface shells were demonstrated by our group [11] as
well as by van Veggel et al. [12], Kynast et al. [13] and
Nann et al. [14]. Very recently we reported on the syn-
thesis and investigation of surface modified KYF4: Yb,
Er nanoparticles [15]. The current paper completes the
investigations described in the previous one [15] and is
focused on the investigation of the mechanism of the
KYF4 formation. It should be mentioned at this point
that we only investigated the formation of the substance
There is a growing interest in nanosized phosphors
owing to their wide application range from solid-state
lasers [1], display devices [2] to commercially used
reporters for nucleic acid microarrays [3, 4]. Lan-
thanide ions are widely used as the emitting species in
many traditional phosphors [5]. In particular the Yb3+
and Er3+ ion couple is utilized in phosphors for the gen-
eration of visible light after excitation in the NIR by so
called photon upconversion. The energy of two IR pho-
tons is transferred to one lanthanide dopant ion, result-
ing in excitation of this ion into a 4f state at high energy.
Visible luminescence is emitted when the ion subse-
quently relaxes to the ground state.
Excitation in the NIR has some advantages, it
induces only a weak autofluorescence background,
avoids photodegradation in biotagging applications and
hence increases the sensitivity of the method. The high-
est upconversion efficiencies known today are observed
for Yb3+/Er3+ and Yb3+/Tm3+ ion couples embedded in
macro crystalline hexagonal phase (β phase) NaYF4
matrix [6–8]. In 2004 we reported the successful syn-
thesis and intense multicolor upconversion emission of
transparently dispersed cubic phase (α-phase) NaYF4
nanocrystals doped with Yb3+/Er3+ and Yb3+/Tm3+ ions
[9]. The use of these nanocrystals directly for biological
applications is limited because of their very low solu-
bility in water. To be suitable for biolabeling applica-
tions, the nanocrystals must be soluble in aqueous
media. Recently we have been able to modify the sur-
face of NaYF4: Yb, Er upconversion nanocrystals with
1
The article is published in the original.
1914

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
INVESTIGATION OF KYF4
1915
KYF4 and we can not give no particular information
concerning the particle formation. In addition X-ray
photoelectron spectroscopy was performed which
brought further evidence for the core-shell nature of the
particles.
2. RESULTS AND DISCUSSION
2.1. Investigation of the Core-Shell Structure
Using XPS
IR-to-visible upconversion fluorescent nanocrystals
of cubic phase KYF4: 28.4% Yb, 2.9% Er, and KYF4:
28.4% Yb, 2.9% Er nanocrystals modified with
undoped KYF4 were synthesized in a high boiling sol-
vent described in an early report [15]. The topological
and optical properties were investigated via X-ray pow-
der diffraction analyses in combination with a Rietveld
refinement of the data, respectively emission spectros-
copy (Up-Conversion Luminescence Measurements)
[15]. Growing undoped KYF4 on the surface of the
KYF4: Yb3+, Er3+ nanocrystals did increase the upcon-
version efficiency by more than a factor of 20 (Fig. 1).
In addition, the XRD data of these particles, displayed
a slight increase of the mean particle size from 13 to
15.5 nm [15]. To sum up we can say that we have
already found evidence for the shell formation.
Fig. 1. Photography of the upconversion luminescence in
1 wt % colloidal solutions of KYF : 28.4% Yb, 2.9% Er
4
core nanocrystals (right) and of KYF : 28.4% Yb, 2.9% Er
4
// KYF nanocrystals (left) in tetraethylene glycol. Excita-
4
–2
tion at 978 nm with a power density of about 20 W mm .
Overall laser power: 2.5 W. The laser is positioned on the
right side.
atomic concentration. Figure 2b demonstrates the XPS
spectrum of the region corresponding to the binding
energy range of 190–180 eV which presents the Er 4d
peak. Compared with the peak measured in case of the
unmodified sample, (curve A) the intensity of the peak
corresponding to the core-shell sample (curve C) is
drastically quenched, obviously upon shell growth. The
50% mixture of both samples (curve B) shows a peak
height in between the peak height from curve A and
curve C, belonging to core and core-shell particles,
respectively. The intensities cannot be explained by the
formation of a mixture of Yb, Er doped KYF4 and
undoped KYF4 particles during surface modification of
the particles. We rather rate these results as a direct
proof that a shell made of undoped KYF4 grows on Yb,
Er doped KYF4 core particles.
A significant problem especially of this matter is
that the lattice constants of Yb3+, Er3+ doped KYF4– and
undoped KYF4 materials are identical, so the core of
these particles cannot be distinguished from the shell
by means of transmission electron microscopy. Mainly
because of this reason we decided to characterize our
samples additionally via high-resolution X-ray photo-
electron spectroscopy (XPS). Photoelectrons created in
the core of a core-shell structured particle should be
severely attenuated as they have to pass through the
core and the shell before they escape into the vacuum
for their kinetic energy analysis. In case of the core-
shell type of nanostructures this attenuation is well
known and up to now some papers have been published
concerning this theme [11, 16–19]. In order to provide
further evidence that a shell was grown on the core par-
ticles and that no intermixing of Yb, Er doped KYF4–
and undoped KYF4 particles took place, we investigated
the unmodified particles (KYF4: 28.4% Yb, 2.9% Er
core particles) and the surface modified particles
(KYF4: 28.4%Yb, 2.9% Er core nanoparticles modified
with undoped KYF4) together with a 50% mixture made
of modified and unmodified particles via XPS. Due to
numerous strong overlaps of the XPS core levels of F,
K, Y and the rare earth metals, the analysis of the XPS
data is limited to the Y3p and the satellite of the Er4d
2.2. Investigation of the Reaction Mechanism
for the Formation of KYF4 in the High
Boiling Solvent HEEDA
Several different synthesis procedures have already
been carried out in the high boiling solvent HEEDA for
the generation of (cubic and hexagonal) NaYF4 [9, 10],
KYF4 core and core-shell particles [15], RbY2F7 and
CsY2F7 particles [20, 21]. However a possible synthesis
mechanism for this type of synthesis procedure, partic-
ularly for the generation of KYF4: Yb, Er (1) using lan-
core-level line. Figure 2a displays the XPS spectrum of thanide chlorides, ammonium fluoride and potassium
the region corresponding to the binding energy range of alkoxide of N-(2-hydroxyethyl)-ethylenediamine in
317–297 eV which includes the Y 3pl/2 and Y 3p3/2 HEEDA (Scheme 1) has not been reported up to now.
peaks respectively. A stoichiometric analysis was per- Straight after combining the transparent solutions of
formed and yields for the three samples a constant Y the reactants we obtained a cloudiness, which disap-
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 54 No. 12 2009

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
1916
SCHÄFER et al.
(a)
(b)
C
B
C
B
A
A
316 312 308 304 300
Binding energy (eV)
190 188 186 184 182 180
Binding energy (eV)
Fig. 2. (a) XPS Y 3p core-level spectrum of A: KYF : Yb, Er core particles. B: 50% Mixture of KYF : Yb, Er core particles and
4
4
undoped KYF particles. C: Surface modified KYF : Yb, Er particles. (b) XPS Er 4d core-level spectrum of A: KYF : Yb, Er core
4
4
4
particles. B: 50% Mixture of KYF :Yb, Er core particles and undoped KYF particles. C: Surface modified KYF :Yb, Er particles.
4
4
4
Δ
pears within a few minutes. The end product KYF4:Yb,
Er is also transparently solved in the solvent HEEDA
[14]. Due to these observations we come to the assump-
tion that the reactants do not simply dissociate and reas-
semble directly into KYF4: Yb, Er. There should be
intermediates, insoluble in the solvent HEEDA.
YCl3:Yb, Er + NH4F
YF3:Yb, Er + NH4Cl
2
Scheme 2
It is known that solid state reactions using the start-
ing materials YF3 and NaF or KF, lead under drastic
reaction condition to KYF4, respectively NaYF4
[22, 23]. We therefore assumed that in case of our KYF4
synthesis YF3 and KF solids might be possible interme-
diate products and hence we tried to generate these
intermediate species with the same compounds we had
used for the KYF4 synthesis. We achieved to synthesize
YF3: Yb, Er with YCl3: Yb, Er and NH4F precursors
according to Scheme 2. Figure 3 shows the powder pat-
tern of the reaction product together with the calculated
line pattern from orthorhombic YF3 (PDF 00-032-
1431) and the Rietveld fit of the measured powder data.
The diffraction peaks can be easily indexed to pure
orthorhombic phase YF3: (PDF 00-032-1431). The best
agreement between observed and calculated profiles
was obtained to the predefmition of elongated particles
in 010 direction. The diffractogram is well fitted by
Rietveld method yielding a value of a = 6.35 , b = 6.87
and c = 4.42 Å for the orthorhombic unit cell and a
mean crystallite size of 6 × 20 nm.
YCl3:Yb, Er + NH4F
–
+ K+ OHC2H4NHC2H4NH2
Δ
KYF4:Yb, Er + NH4Cl
1
Scheme 1
Usually the synthesis is carried out for 15 h at 185–
200°C. Unfortunately we were not able to isolate inter-
mediates just by reducing the reaction time. To check
the progress of the reaction by means of upconversion
luminescence measurements is not suitable in this case,
because only after 15 h duration time we noticed effi-
cient upconversion emission in solution. We were also
not successful in trapping intermediate stages. We did
sampling after combining the reactants and investigated
the samples by means of XRD powder diffractometry.
Even straight after combining the three reactants and
heating up to 185°C only the final product KYF4:Yb, Er
can be proven in the reaction mixture. Therefore we
decided to strike out in a different non classical direc-
tion.
–
NH4F + K+ OHC2H4NHC2H4NH2
KF
Δ
3
Scheme 3
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 54 No. 12 2009

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
INVESTIGATION OF KYF4
1917
(b)
PDF 0036–1458
(a)
70
20
30
40
50
2θ (°)
60
5
10 15 20 25 30 35 40 45 50 55 60 65 70
2θ (°)
Fig. 3. Observed X-ray powder diffraction profiles of
Fig. 4. (a) Line pattern of cubic KF (PDF 00-0036-1458).
(b) Powder pattern of the KF sample gained from the
HEEDA synthesis.
nanocrystalline KYF : 28.4% Yb, 2.9% Er (grey line),
4
Rietveld fit (black line) and residuum.
KYF4: Yb, Er were also obtained in this pattern. Again
there are no relevant differences concerning to peak
positions and peak intensities between this pattern and
that of the reference sample (Fig. 6b). To sum up we can
say that these experiments provide at least hints that
YF3 and KF are intermediates in the synthesis of KYF4
starting from the lanthanide chlorides, ammonium flu-
oride and potassium alkoxide of N-(2-hydroxyethyl)-
ethylenediamine in HEEDA. We present herein a very
simple approach to achieve information about the
mechanism of the KYF4 formation.
YF3:Yb, Er + KF
2
3
Scheme 4
The second possible intermediate compound, KF
had been gained starting from NH4F and a solution of
potassium alkoxide of N-(2-hydroxyethyl)- ethylenedi-
amine in HEEDA (see Scheme 3). The powder pattern
of the sample together wilh the calculated line pattern
of cubic KF (PDF 00-0036-1458) can be taken from
Fig. 4. The formation of almost pure KF can be
deducted from this comparison. The following attempt
to generate the tetrafluoride 1 from the addition of the
compounds 2 and 3 was unsuccessful (Scheme 4). No
chemical reaction took place in HEEDA at 205°C, even
after 16 h. Nevertheless compounds 2 and 3 can be
intermediates in the synthesis according to Scheme 1,
but obviously no direct addition of the compounds
takes place. In order to confirm the assumption that the
trifluoride 2 is an intermediate in the synthesis reaction
according to Scheme 1, we performed in a third step the
synthesis of compound 1 choosing the lanthanide triflu-
oride 2, a solution of the potassium alkoxide of
N-(2-hydroxyethyl)-ethylenediamine in HEEDA and
ammonium fluoride as precursors (Scheme 5). The
XRD powder pattern of the reaction product according
to Scheme 4 can be taken from Fig. 5a. Figure 5b shows
the XRD powder pattern of the sample gained from the
original synthesis procedure according to Scheme
1.The peak positions agree very well, almost pure
KYF4: 28.4%Yb, 2.9% Er has been generated. In a last
experiment (see Scheme 6) we wanted to verify if
potassium fluoride 3 could also be an intermediate
product in the synthesis reaction according to Scheme
1. We tried to achieve the desired compound 1 via a pro-
cedure according to Scheme 6. The corresponding
XRD powder pattern of the sample is shown in Fig. 6a.
All the peaks we had found in the reference pattern
(Fig. 6b) and we therefore had been refered to cubic
–
YF3:Yb, Er + K+ OHC2H4NHC2H4NH2
2
Δ
+ NH4F
KYF4:Yb, Er
1
Scheme 5
(b)
(a)
60
20
30
40
50
70
2θ (°)
Fig. 5. (a) Observed X-ray powder diffraction profiles of
KYF : 28.4%Yb, 2.9% Er gained from the synthesis proce-
4
dure according to Scheme 5. (b) Observed X-ray powder
diffraction profiles of KYF : 28.4% Yb, 2.9% Er gained
4
from the original synthesis procedure according to Scheme 1.
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 54 No. 12 2009