Gorni Nanomaterials 2019

nanomaterials
Article
Transparent Sol-Gel Oxyfluoride Glass-Ceramics with
High Crystalline Fraction and Study of
RE Incorporation
Giulio Gorni 1 , Jose J. Velázquez 2 , Jadra Mosa 1 , Glenn C. Mather 1 , Aida Serrano 1,3 ,
María Vila 3,4 , Germán R. Castro 3,4 , David Bravo 5 , Rolindes Balda 6,7 , Joaquín Fernández 8 ,
Alicia Durán 1 and Yolanda Castro 1, *
1 Instituto de Cerámica y Vidrio, CSIC, 28049 Madrid, Spain; ggorni@icv.csic.es (G.G.); jmosa@icv.csic.es (J.M.);
mather@icv.csic.es (G.C.M.); aida.serrano@icv.csic.es (A.S.); aduran@icv.csic.es (A.D.)
2 FunGlass—Centre for Functional and Surface Functionalized Glass, Alexander Dubček University of
Trenčín, Trenčín 91150, Slovakia; jose.velazquez@tnuni.sk
3 SpLine, Spanish CRG Beamline—European Synchrotron Radiation Facility (ESRF), 38043 Grenoble, France;
maria.vila-santos@esrf.fr (M.V.); german.castro@esrf.fr (G.R.C.)
4 Instituto de Ciencia de Materiales de Madrid, CSIC, 28049 Madrid, Spain
5 Departamento Física de Materiales, Facultad de Ciencias, Universidad Autónoma de Madrid (UAM),
28049 Madrid, Spain; david.bravo@uam.es
6 Departamento Física Aplica I, Escuela Superior de Ingeniería, Universidad del País Vasco (UPV-EHU),
48013 Bilbao, Spain; rolindes.balda@ehu.eus
7 Centro de Física de Materiales, (UPV/EHU-CSIC), 20018 San Sebastian, Spain
8 Donostia International Physics Center (DIPC), 20018 San Sebastian, Spain; xuaco@dipc.org
* Correspondence: castro@icv.csic.es; Tel.: +34-917-355-840

Received: 28 February 2019; Accepted: 26 March 2019; Published: 3 April 2019
Abstract: Transparent oxyfluoride glass-ceramic films and self-supported layers with composition
80SiO2 -20LaF3 doped with Er3+ have been successfully synthesized by sol-gel process for the first
time. Crack-free films and self-supported layer with a maximum thickness up to 1.4 µm were
obtained after heat treatment at the low temperature of 550 ◦ C for 1 min, resulting in a LaF3 crystal
fraction of 18 wt%, as confirmed by quantitative Rietveld refinement. This is the highest value
reported up to now for transparent oxyfluoride glass-ceramics prepared by sol-gel. This work
provides a new synthesis strategy and opens the way to a wide range of potential applications
of oxyfluoride glass-ceramics. The characterization by a wide range of techniques revealed the
homogeneous precipitation of LaF3 nanocrystals into the glass matrix. X-ray absorption spectroscopy
and electron paramagnetic resonance confirmed that the Er3+ ions are preferentially embedded in the
low phonon-energy LaF3 nanocrystals. Moreover, photoluminescence (PL) measurements confirmed
the incorporation of dopants in the LaF3 nanocrystals. The effective concentration of rare-earth ions
in the LaF3 nanocrystals is also estimated by X-ray absorption spectroscopy.
Keywords: glass-ceramic; oxyfluoride; rare-earth; crystallization; sol-gel
1. Introduction
Oxyfluoride glass-ceramics (OxGCs) have attracted great interest in the field of photonics since
the pioneering work of Wang and Ohwaki [1]. OxGCs are polycrystalline materials containing
fluoride nano-crystals (NCs) that crystallize in a glass matrix during a controlled heat treatment of the
precursor glass. Alumina-silicate glass matrices are suitable, due to their excellent mechanical, thermal
and chemical properties compared to phosphate or fluoride glasses. In addition, OxGCs may host
Nanomaterials 2019, 9, 530; doi:10.3390/nano9040530 www.mdpi.com/journal/nanomaterials

Nanomaterials 2019, 9, 530 2 of 16
Rare-Earth (RE) ions in the low-phonon energy fluoride crystals (300–450 cm−1 ), thereby increasing the
luminescence efficiency with respect to the corresponding oxide glasses [2–14]. The melt-quenching
(MQ) method has been routinely used to prepare OxGCs that showed excellent optical properties
and a significant enhancement of photoluminescence with respect to the precursor glasses. Moreover,
the possibility to draw fibers and convert the core into a GC revealed the importance of this method for
the preparation of novel devices with new or improved optical efficiencies [15–20]. However, the high
melting temperatures (1400–1700 ◦ C) greatly affect the fluoride content, limiting the crystal phase
content to 10 wt% or less for some crystal phases, such as LaF3 [7,9,10]. Furthermore, phase separation
and spontaneous fluoride crystallization are usually present, due to fluorine immiscibility at high
temperature in oxide-glass matrices. For these reasons, this method still has to overcome some
problems and efforts are required to improve the quality and processing of these materials. Another
crucial point concerns the incorporation of dopants into the fluoride NCs. In fact, it is known that
in melted glasses there is a portion of RE ions that remain outside the fluoride NCs even after heat
treatment for prolonged periods [10,12]. This phenomenon is explained considering the high glass
viscosity at typical crystallization temperatures (Tg + 20–100 ◦ C) that limits ionic diffusion. Moreover,
due to the depletion of crystal formers in the glass matrix, a viscous Si-enriched shell forms around the
crystals, which further limits ionic diffusion. A challenging goal is to obtain homogeneous materials
with much higher fluoride contents and with higher dopant incorporation in the NCs, thus improving
the optical efficiency.
The sol-gel (SG) route appears to be a promising alternative to prepare OxGC films and
bulk materials for different applications, such as planar waveguides or integrated optics [21–36].
Ballato et al. [35] proposed to use LaF3 into planar photonic devices, such as planar confined
waveguide and Kumar et al. [36] described Er3+ doped LaF3 transparent gels for its use in the infrared
(IR) waveguide amplifiers at 1.5 µm.
The SG process is a cheap and extremely flexible chemical method for obtaining pure and
very homogenous materials, as well as offering a wide range of possible nano/micro-structures at
relatively low sintering temperatures (100–600 ◦ C). Synthesis classically involves the hydrolysis and
polycondensation of metal salts or metal-organic precursors, such as tetraethyl orthosilicate (TEOS),
in alcoholic media [37]. In the late nineties, Fujihara et al. prepared the first OxGC SG materials
in a two-step process [21] involving preparation of silica sol using the classical route followed by
mixing of fluoride and RE-ion precursors (acetates, nitrates, chlorides etc.) dissolved in ethanol or
acidic medium. The mixing of both solutions, with subsequent controlled crystallization, leads to the
precipitation of fluoride crystals in the silica-glass matrix. However, the literature mostly describes
the preparation of OxGC materials with a nominal content of 5 mol% active crystal phase, and with
the final crystal fraction not being estimated [38]. For example, Szpikowska-Sroka reported 3 wt% of
the crystallized phase for a nominal concentration of 5 mol% [34] Such concentrations are even lower
than those obtained by MQ [2,10,12]. Other authors have attempted to increase the crystal content
up to 15 mol% but opaque materials were obtained [21]. Moreover, high treatment temperatures
(800–1000 ◦ C) are used to obtain OxGCs from the amorphous precursors, even though the typical
crystallization temperature of fluorides is between 280–350 ◦ C. To date, the use of sol-gel to obtain
OxGCs has not been widely investigated because no significant improvement in structural and optical
properties has been achieved since the earliest papers. There is still much work to be performed on
correlating the crystallization mechanism of these materials with the structure, dopant incorporation
and optical properties.
In this work, novel OxGC materials with a much higher active crystal phase content and sintered
at a much lower temperature than previously reported have been prepared by taking advantage of
the particular benefits of the SG process. In addition, it was demonstrated how this method leads to
precipitation of RE-doped NCs using short heat treatment times. Moreover, it is shown how most RE
ions are incorporated into the NCs during the crystallization process.
2. Materials and Methods
2.1. Materials Preparation

Sols of composition (mol%) 80SiO2 -20LaF3 doped with 0.1 and 0.5 Er3+ were prepared by
partially replacing TEOS (Sigma Aldrich, St. Louis, MO, USA) with methyl-trimethoxy-silane
(MTES, abcr, Karlsruhe, Germany) as a precursor with a TEOS/MTES molar ratio of 40/60.
The synthesis was performed in two- step process starting from a silica sol with molar ratio:
0.4TEOS:0.6MTES:2.5CH3 CH2 OH:1H2 O(0.1HCl):0.2CH3 COOH (Merck, Darmstadt, Germany) stirred
for 2 h at room temperature. Subsequently, La(CH3 COO)3 (Sigma Aldrich), ethanol (EtOH), H2 O and
trifluoroacetic acid (TFA, Sigma Aldrich) in the molar ratio 1:5:4:4 was stirred for 2 h at 40 ◦ C and then
mixed with the silica sol in the ratio 0.8(TEOS + MTES):0.2La.
Thin films on silica and silicon wafers (111) were prepared by dip coating using withdrawal rates
of 10–35 cm/min. Finally, the films were treated at 550 ◦ C for 1 min and 1 h using a heating rate of
10 ◦ C/min.
Moreover, self-supported layers (bulk-like monoliths 1–2 mm thick) of 80SiO2 -20LaF3 composition
doped with 0.1 and 0.5 Er3+ were prepared with TEOS and TEOS-MTES using the molar ratio
1TEOS:3EtOH:2H2 O(0.1HCl) and 0.4TEOS:0.6MTES:3EtOH:2H2 O(0.1HCl), respectively. An acetate
dissolution, similar to that used for thin films, was added, raising the H2 O content to obtain a total
molar ratio 1TEOS:10H2 O or 1(TEOS+MTES):10H2 O.
Then, the sol was deposited into Petri dishes, covered, sealed and kept at 50 ◦ C for 1 week to
obtain the xerogel samples. GC samples were obtained by heating the xerogels at 550 ◦ C for 1 min
up to 1 h, with an intermediate step at 150 ◦ C to remove water and solvents. Samples prepared with
only TEOS were treated up to 650 ◦ C for 1 h using a heating rate of 1 ◦ C/min. The samples will be
labelled with the composition followed by the silica precursor used for their preparation (TEOS) or
(TEOS-MTES).
2.2. Characterization of Coatings and Self-supported Layers

Thin films deposited onto a silica and silicon wafer were characterized by spectroscopic
ellipsometry in the range 250–1000 nm using an M-2000UTM ellipsometer (J.A. Woollam Co., Lincoln,
NE, USA). The incident angles were set to 50◦ and 60◦ and the acquisition time to 10 s. The data were
fitted using the CompleteEASE software (version 5.20, J.A. Woollam Co.) representing the film as a
Cauchy layer deposited onto the corresponding substrate.
Differential thermal analysis (DTA) and thermogravimetry (TG) were measured using the SDT
Q600 instrument (TA Instruments, New Castle, DE, USA). Around 20–30 mg of powder materials with
a size of 1–1.25 mm, obtained after drying the sol at 50 ◦ C, were measured in air and Argon using a
heating rate of 10 ◦ C/min.
X-Ray diffraction (XRD) patterns of 80SiO2 -20LaF3 -0.5Er3+ self-supported layers and grazing
incidence XRD (GI-XRD) of 80SiO2 -20LaF3 thin films were obtained at the BM25B-SpLine beamline of
the ESRF (European Synchrotron Radiation Facility, Grenoble, France) [39], using an X-ray wavelength
of 0.619 Å. For GI-XRD measurements, an incidence angle of 0.5◦ was used. The crystal size, φ,
was estimated using the Scherrer’s equation:
0.94λ
φ= q , (1)
cos θ Bm 2 − B2
i
where λ is the wavelength, Bm the full width half maximum of the peak and θ its diffraction angle.
The instrumental broadening Bi was also taken into account. The diffraction-peak parameters were fit
using a pseudo Voigt function.
Rietveld Refinement was performed using the FullProf software (FullProf Suite July 2017) [40],
and the LaF3 crystal fraction was estimated using NaF as the internal weight standard in a similar
manner to that described previously [12]. X-ray powder data were collected over the range 20◦ ≤ 2θ ≤
120◦ in a step width of 0.0289◦ and a counting time of 4s per step employing a Bruker D8 high-resolution
diffractometer (BRUKER, Billerica, MA, USA) equipped with a solid-state rapid LynxEye detector,
and monochromatic Cu Kα1 radiation.
High Resolution Transmission Electron Microscopy (HRTEM) was performed using a JEOL
2100 field emission gun (Akishima, Tokyo, Japan) operating at 200 kV with a point resolution of
0.19 nm. The samples were obtained by scratching the films or milling the self-supported layers and
dispersing in EtOH. Then, one or two drops of suspensions were depositing on the scaled fragments
onto carbon-coated copper grids. The solvent was removed after drying the copper grids under a
UV lamp.
Er3+ L3 edge X-Ray absorption spectroscopy (XAS, Grenoble, France) were used to characterize
the 80SiO2 -20LaF3 0.5Er3+ films treated at 150 ◦ C for 1 h (xerogel) and at 550 ◦ C for 1 min (GC)
employing the BM25A-SpLine beamline (ESRF, Grenoble, France). XAS measurements were collected
in fluorescence mode using a 13 element Si (Li) solid-state detector with the sample surface placed at an
angle of 45◦ to the incident beam. The spectra represent an average of at least six scans; the XAS data
were processed using the software ATHENA (Demeter 0.9.26) [41]. Both ErF3 and Er2 O3 standards
were also measured as references.
Electron Paramagnetic Resonance (EPR) spectra were obtained by a Bruker ESP 300E spectrometer
(BRUKER) working in the X-band with field modulation of 100 kHz. The temperature was controlled
with a continuous-flow helium cryostat ESR 900 (Oxford Instruments, Abingdon, UK) and the
calibration of the resonance magnetic fields and microwave frequencies measured with an NMR
Gauss-meter ER 035 M (BRUKER) and a frequency meter (Hewlett-Packard 5342A, Palo Alto, CA,
USA), respectively. Self-supported layer samples (xerogel and GCs) doped with 0.1 and 0.5 Er3+ were
milled and rolled in Teflon tape to perform the measurements.
The steady-state emission spectra were performed by exciting the 80SiO2 -20LaF3 bulk samples
and thin films doped with 0.1 and 0.5 Er3+ with a tunable Ti-sapphire ring laser (0.4 cm−1 linewidth)
in the 770–920 nm spectral range. The fluorescence was analyzed with a 0.25 m monochromator,
and the signal was detected by an extended IR Hamamatsu H10330A-75 photomultiplier (Hamamatsu,
Shizuoka, Japan) and finally amplified by a standard lock-in technique. The sample temperature was
varied between 9 and 300 K in a continuous flow cryostat.
Lifetime measurements were obtained by exciting the 80SiO2 -20LaF3 -0.1Er3+ (TEOS) bulk sample
treated at 650 ◦ C for 1 h with a Ti-sapphire laser pumped by a pulsed frequency-doubled Nd:YAG
laser, 9 ns pulse width, (BM Industries, Paris, France), and detecting the emission with Hamamatsu
H10330A-75 photomultiplier. Data were processed by a Tektronix oscilloscope (Tektronix Inc.,
Beaverton, OR, USA).
3. Results and Discussion
3.1. Self-supported Layers and Thin Films

Crack-free and homogeneous self-supported layers were obtained before and after heat treatment
of the doped and undoped 80SiO2 -20LaF3 samples at 550 ◦ C for 1 min, Figure 1.
Homogeneous and transparent films with good adhesion were obtained for all the withdrawal
rates used. The good agreement between measurement and fitted data, Figure 2a, demonstrated the
good quality of the films. The inset of Figure 2a shows the variation of the refractive index as a function
of the wavelength in the transparent region of the sample. The corresponding thickness varies almost
linearly with the withdrawal rate, up to a thickness of ~1.4 µm in a single deposition and using a
withdrawal rate of 35 cm/min, Figure 2b. The addition of MTES allows the thickness of films to be
increased, compared to when only TEOS is used. In particular, according to [42], the TEOS/MTES
ratio 0.4/0.6 produces the highest critical thickness. Moreover, the addition of MTES improves the
mechanical strength of the self-supported layers. Such behavior is associated with the presence of
laser, 9 ns pulse width, (BM Industries, Paris, France), and detecting the emission with Hamamatsu
H10330A-75 photomultiplier. Data were processed by a Tektronix oscilloscope (Tektronix Inc.,
Beaverton, OR, USA).
3. Results and Discussion

3.1. Self-supported Layers and Thin Films

CH3 groups which allow the porosity to be increased in the silica network reducing the stress [42].
Crack-free
The amountand
of OHhomogeneous self-supported
groups is also reduced when MTES layers were
is added obtained
to TEOS before
because and after
the surfaces heat
become
treatment of the doped and undoped
more hydrophobic [42,43]. 80SiO 2 -20LaF 3 samples at 550 °C for 1min, Figure 1.
Nanomaterials 2019, 9, x FOR PEER REVIEW 5 of 16
and using a withdrawal rate of 35 cm/min, Figure 2b. The addition of MTES allows the thickness of
films to be increased, compared to when only TEOS is used. In particular, according to [42], the
TEOS/MTES ratio 0.4/0.6 produces the highest critical thickness. Moreover, the addition of MTES
improves the mechanical strength of the self-supported layers. Such behavior is associated with the
FigureFigure
1. (a)1.Xerogel of composition
(a) Xerogel of composition80SiO
80SiO 2-20LaF
2 -20LaF 0.5Er3+3+(tetraethyl
3-3 -0.5Er orthosilicatewith
(tetraethyl orthosilicate withmethyl-
methyl-
presence of CH3 groups which allow the porosity to be increased in the silica network reducing the
◦ C for 24 h; (b) Oxyfluoride glass-ceramics (OxGC) of
trimethoxy-silane, TEOS-MTES) obtained at 150 °C for 24 h; (b) Oxyfluoride glass-ceramics (OxGC)
trimethoxy-silane, TEOS-MTES) obtained at 150
stress [42]. The amount of OH groups is ◦ C also reduced when MTES is added to TEOS because the
of the the
samesame compositionobtained
composition obtainedat
at 550
550 °C for
for 11min.
min.
surfaces become more hydrophobic [42,43].
Homogeneous and transparent films with good adhesion were obtained for all the withdrawal
14 (a) 1400 (b) 550ºC-1h
rates used. The good agreement between measurement and fitted data, Figure 2a, demonstrated the
12 1200
good quality of the films. The inset of Figure 2a shows the variation of the refractive index as a
10
Thickness (nm)
function of the wavelength in the transparent region 1000

of the sample. The corresponding thickness
Ψ (degrees)
8
varies almost linearly with the withdrawal rate, up to a thickness of ~1.4 µm in a single deposition
6 1.49 3+
80SiO2-20LaF3 0.5Er 800
Refractive index n
550°C-1h
4 Meas. 1.48
Fit
2 1.47
600
0 1.46
400 500 600 700 800 900 1000
Wavelength (nm)
-2
400 500 600 700 800 900 1000 10 20 30
Wavelength (nm) Withdrawal rate (cm/min)
2. (a) Ellipsometricmeasurements
measurements and
and fit 3+
Figure
Figure 2. (a) Ellipsometric fitofof80SiO2 -20LaF
80SiO 3 -0.5Er
2-20LaF
(TEOS-MTES) thin film
3-0.5Er3+ (TEOS-MTES) thin film
◦
treated at 550 C for 1 min deposited onto silica substrate using a withdrawal rate of 35 cm/min.
treated at 550 °C for 1 min deposited onto silica substrate using a withdrawal rate of 35 cm/min. The
The inset shows the dispersion relation using the Cauchy model (b) Variation of thickness with the
inset shows the dispersion relation using the Cauchy model (b) Variation of thickness with the
withdrawal rate.
withdrawal rate.
3.2. DTA and Crystallization
3.2. DTA Figure
and Crystallization
3 shows a DTA curve for self-supported layers measured in air with a heating rate of
10 ◦ C/min. The endothermic process at ~100 ◦ C corresponds to water and solvent removal. The first
Figure 3 shows a DTA curve for self-supported layers measured in air with a heating rate of 10
sharp exothermic
°C/min. The endothermic peak at ~280 ◦at
process C is associated
~100 with LaF3 to
°C corresponds crystallization and the removal.
water and solvent correspondingThe first
mass loss (~30%) is due to decomposition reactions, as studied elsewhere [44]. The sharpness of the
sharp exothermic peak at ~280 °C is associated with LaF3 crystallization and the corresponding mass
crystallization peak means that the process occurs very rapidly; in contrast, the LaF3 crystallization
loss (~30%) is due to decomposition reactions, as studied elsewhere [44]. The sharpness of the
peak in the melted glasses is much broader and less intense [12]. Moreover, the Tg of these samples is
crystallization
higher thanpeak
1000 ◦means that the
C, as reported in process
a previousoccurs
papervery rapidly;
[44], and in contrast, of
the crystallization theSGLaF 3 crystallization
OxGCs occurs
peakatinmuch
the melted glasses is much broader and less intense [12]. Moreover, the T g of these
lower temperature than the Tg . The crystallization mechanism is, thus, clearly different with samples
is higher
respectthan 1000
to that of °C,
MQas reported
OxGCs in aThe
[45–49]. previous
secondpaper [44], process
exothermic and theatcrystallization
◦
~575 C, associatedof SGwithOxGCs
occurs at much lower temperature than the T . The crystallization mechanism
oxidation of CH3 groups with a further mass loss of around 10%, does not appear in Ar atmosphere.
g is, thus, clearly
different with respect
Heat treatment to that of higher
at temperatures MQ OxGCs [45–49].
than that of CH3 The secondcauses
elimination exothermic
a sudden process
shrink ofat films
~575 °C,
and self-supported layers, leading to crack of the samples.
associated with oxidation of CH3 groups with a further mass loss of around 10%, does not appear in
Ar atmosphere. Heat treatment at temperatures higher than that of CH3 elimination causes a sudden
shrink of films and self-supported layers, leading to crack of the samples.
6 100
Air
5 90
4 80
Weigth (%)
ΔT (ºC)
3 70
2 60
peak in the melted glasses is much broader and less intense [12]. Moreover, the Tg of these samples
is higher than 1000 °C, as reported in a previous paper [44], and the crystallization of SG OxGCs
occurs at much lower temperature than the Tg. The crystallization mechanism is, thus, clearly
different with respect to that of MQ OxGCs [45–49]. The second exothermic process at ~575 °C,
associated with oxidation of CH3 groups with a further mass loss of around 10%, does not appear in
Nanomaterials 2019, 9, 530Heat treatment at temperatures higher than that of CH3 elimination causes a sudden6 of 16
Ar atmosphere.
shrink of films and self-supported layers, leading to crack of the samples.
6 100
Air
5 90
4 80
Weigth (%)
ΔT (ºC)
Nanomaterials 2019, 9, x FOR PEER REVIEW
3 70 6 of 16
2 60
3.3. XRD, GI-XRD and Crystal Fraction
1 50
The XRD of 80SiO2-20LaF3-0.1Er3+ (TEOS) self-supported layer treated at 650 °C for 1 h is shown
0 40
in Figure 4a. As can be seen, only LaF3 NCs, with a crystal size of 6 nm, are obtained. On the other
-1 30
hand, XRD patterns of 80SiO2-20LaF 0 1003-0.5Er 3+ (TEOS-MTES) self-supported
200 300 400 500 600 700 800 layer treated at 550 °C for
1 min exhibited well-defined peaks consistent with the
Temperature (ºC)crystallization of LaF3 (JCPDS 00-032-0483)
with a crystallite size of 9.0 ± 0.5 nm, demonstrating that only 1 min is sufficient to crystallize the
Figure 3. Differential
Figure 3. Differential thermal analysis
thermal analysis (DTA,
(DTA, red)red) and thermogravimetry
and thermogravimetry (TG, blue)(TG,
curveblue) curve
for 80SiO 2- for
phase, Figure 4b. The difference
3+ (TEOS-MTES) of the crystal size with respect to the TEOS sample treated◦ at 650 °C
80SiO20LaF
2 -20LaF 3 -0.5Er
3-0.5Er 3+ (TEOS-MTES) bulk sample
bulk sample treated in treated
the air in thea air
with withrate
heating of 10 °C/min.
a heating rate of 10 C/min.
is associated with the higher heating rate (10 °C/min) used to crystallize the samples prepared with
3.3.TEOS-MTES.
XRD, GI-XRD Theand crystallization
Crystal Fraction mechanism of similar GCs was extensively studied in previous
papers [44,50], demonstrating how the decomposition reaction of lanthanide fluoroacetates leads to
The XRD of
precipitation of80SiO
NCs once 2 -20LaF -0.1Er3+ (TEOS)
the3crystallization self-supported
temperature layer treated
is reached. at 650 ◦ C formechanism
The crystallization 1 h is shown
in of
Figure
these 4a. As canis,be
materials seen, only
therefore, LaF3 NCs,
completely with afrom
different crystal
thatsize of 6 nm, are
of oxyfluoride obtained.
glasses by MQ Oninthe other
which
hand, XRD patterns
the crystal growth isofa80SiO 2 -20LaF3 -0.5Erprocess.
diffusion-controlled
3+ (TEOS-MTES)
Chemical bondsself-supported layer treated
between fluorine and the 550 ◦ C
atglass
formatrix
1 min were
exhibited
observedwell-defined
using 19FpeaksNuclear consistent
Magnetic with the crystallization
Resonance (NMR) [44]. of Instead,
LaF3 (JCPDS
for the00-032-0483)
xerogel
with crystallite size of 9.0 ±
samples, the fluorine environment is exactly the same as in the precursor TFA acid; almost all fluorinethe
a 0.5 nm, demonstrating that only 1 min is sufficient to crystallize
phase, Figure
is then found 4b. inThethe difference
form ofofLaF the3 crystal
after the sizecrystallization
with respect to the TEOS
process [44].sample treated it
Furthermore, was ◦ C
at 650
is associated with thean higher heating ◦
rate (10 time C/min)
demonstrated that increase of treatment doesusednot to crystallize
increase the samples
the crystal size orprepared
the crystal with
fraction of bulk samples; in contrast, crystal dissolution is
TEOS-MTES. The crystallization mechanism of similar GCs was extensively studied in previous observed for very long heat treatment
times.[44,50],
papers Such results are clearly how
demonstrating inconsistent with a diffusion-controlled
the decomposition reaction of lanthanideprocess. fluoroacetates leads to
Quantitative
precipitation of NCsRietveld
once the refinement
crystallization of the LaF3 crystalline
temperature is reached. fraction with NaF as the
The crystallization internal of
mechanism
standard was carried out for 80SiO -20LaF 0.5Er 3+ bulk GC treated at 550 °C for 1 min following the
these materials is, therefore, completely different from that of oxyfluoride glasses by MQ in which the
2 3
samegrowth
crystal procedure described in [49]. The
is a diffusion-controlled observed
process. diffraction
Chemical bondspattern
between and the difference
fluorine between
and the glass matrix
observed
were observedand using
calculated diffraction
19 F Nuclear patternsResonance
Magnetic are shown in Figure[44].
(NMR) 4b. Weight
Instead, fractions
for the (wt%)
xerogel of samples,
17.8%
theand 82.2% environment
fluorine were determined for thethe
is exactly LaF 3 crystalline fraction and glassy phase, respectively. To our
same as in the precursor TFA acid; almost all fluorine is then
knowledge, this represents the highest crystal fraction reported in the literature for transparent
found in the form of LaF3 after the crystallization process [44]. Furthermore, it was demonstrated that
OxGCs prepared by SG methods and it is the highest crystal fraction reported so far for LaF3-OxGCs
an increase of treatment time does not increase the crystal size or the crystal fraction of bulk samples;
prepared by both MQ and SG [38]. A small shift of the diffraction peaks towards higher 2θ was also
in contrast, crystal dissolution is observed for very long heat treatment times. Such results are clearly
observed for the Er3+-doped GC. This provides evidence of Er3+ incorporation, since the smaller Er3+
inconsistent with a diffusion-controlled process.
ions (1.17 Å) on the sites of the larger La3+ (1.31 Å) ion decrease the unit-cell volume (Table 1).
(a) 650°C-1h
XRD intensity (arb. units)
φ ≈ 6 nm
10 20 30 40 50 60 70 80 90 100 110 120

2θ(°)
Figure4. 4.(a)(a)X-Ray
X-Raydiffraction
diffraction (XRD)
(XRD) ofof 80SiO
80SiO2-20LaF 3-01Er3+ (TEOS)
3+ (TEOS) self-supported layerlayer
treated at
Figure 2 -20LaF3 -01Er self-supported treated
650 °C
◦ for 1 h. (b) Observed (small red circles), calculated (continuous
at 650 C for 1 h. (b) Observed (small red circles), calculated (continuous black line) and black line) and difference
(continuous
difference blue lineblue
(continuous at the
linebottom) XRD profiles
at the bottom) of 80SiO
XRD profiles 2-20LaF3-0.5Er3+ (TEOS-MTES)
of 80SiO 3+ self-
2 -20LaF3 -0.5Er (TEOS-MTES)
supported layer treated at 550 °C mixed with 9 wt.% NaF as the internal standard.
self-supported layer treated at 550 ◦ C mixed with 9 wt.% NaF as the internal standard. Bragg peaks Bragg peaks of of
NaF and LaF3 are indicated by blue and red vertical bars, respectively.
NaF and LaF3 are indicated by blue and red vertical bars, respectively.
Table 1. Lattice constants of 80SiO2-20LaF3-0.5Er3+ (TEOS-MTES) glass-ceramic bulk sample treated

at 550 °C for 1 min.
Sample a (Å) c (Å) Unit Cell Volume (Å3)

Undoped LaF3 [JCPDS]1 7.187 7.350 328
Quantitative Rietveld refinement of the LaF3 crystalline fraction with NaF as the internal standard
was carried out for 80SiO2 -20LaF3 0.5Er3+ bulk GC treated at 550 ◦ C for 1 min following the same
procedure described in [49]. The observed diffraction pattern and the difference between observed
and calculated diffraction patterns are shown in Figure 4b. Weight fractions (wt%) of 17.8% and 82.2%
were determined for the LaF3 crystalline fraction and glassy phase, respectively. To our knowledge,
this represents the highest crystal fraction reported in the literature for transparent OxGCs prepared
by SG methods and it is the highest crystal fraction reported so far for LaF3 -OxGCs prepared by both
MQ and SG [38]. A small shift of the diffraction peaks towards higher 2θ was also observed for the
Er3+ -doped GC. This provides evidence of Er3+ incorporation, since the smaller Er3+ ions (1.17 Å) on
the sites of the larger La3+ (1.31 Å) ion decrease the unit-cell volume (Table 1).
Table 1. Lattice constants of 80SiO2 -20LaF3 -0.5Er3+ (TEOS-MTES) glass-ceramic bulk sample treated at
550 ◦ C for 1 min.
Sample a (Å) c (Å) Unit Cell Volume (Å3 )
Undoped
Er 3+ dopedLaF
GCs 550°C 11m
7.187
3 [JCPDS]
7.16 ± 0.01 7.32 7.350
± 0.01 325328
±1
Er3+ doped GCs 550◦ C 1 m 7.16 ± 0.01 7.32 ± 0.01 325 ± 1
1 JCPDS cards 00-032-0483. Ionic sizes in n-fold coordination for hexagonal phase: 1.31 Å for La3+ and
1 JCPDS cards 00-032-0483. Ionic sizes in n-fold coordination for hexagonal phase: 1.31 Å for La3+ and 1.17 Å
1.17 Å
3+
for Er3+ [51].
for Er [51].
GI-XRD pattern of a thin film was obtained at the BM25B-SpLine beamline of the ESRF [39] and
GI-XRD
is shown pattern
in Figure 5a.ofThe
a thin
XRDfilm
ofwas obtained
the bulk sampleat the
hasBM25B-SpLine
also been acquiredbeamline of the ESRF [39]
for comparison, and5b.
Figure is
shown in Figure 5a. The XRD of the bulk sample has also been acquired for
For the thin film, only a broad band, which could be deconvoluted into two peaks associated with comparison, Figure 5b.
For
LaFthe thin film, only a broad band, which could be deconvoluted into two peaks associated with LaF3
3 crystals (Figure 5a). In this case, the crystallite size was much smaller, ~2.5 ± 0.5 nm. The
crystals
difference(Figure 5a). In this
in crystallite sizecase, the crystallite
between bulk and size
filmwas muchissmaller,
samples, likely to~2.5 ± 0.5 nm.
be related Thesynthesis
to the differenceandin
crystallite size between bulk and film samples, is likely to be related to the synthesis
sintering conditions, since more diluted sols are necessary for film deposition with respect to bulk and sintering
conditions,
materials and since more the
during diluted sols are
sintering thenecessary for film deposition
coating densifies with respect
only by shrinkage to bulk
normal materials
to the and
substrate.
during the sintering the coating densifies only by shrinkage normal to the substrate.
Moreover, the fast evaporation of the solvent during the heat treatment may lead to the precipitation Moreover, the fast
evaporation of the solvent during the heat treatment may lead to the precipitation
of such small NCs. This difference in crystal size between films and bulk-like materials has also been of such small NCs.
This difference
reported in crystal
previously size between
for similar systems, films
butand bulk-like
a clear materials
explanation has
is still also been[52].
necessary reported previously
For comparison,
for
Figure 5b shows the XRD of the bulk samples of the same composition treated under shows
similar systems, but a clear explanation is still necessary [52]. For comparison, Figure 5b the
the same
XRD of the bulk samples of the same composition treated under the same conditions.
conditions.
(a) (b)
LaF3 LaF3
Intensity (a.u)
Intensity (a.u)
φc = 2.5 nm φc = 9.0 nm
λ = 0.619 Å λ = 0.619 Å
(111)
(002)
(110)
8 9 10 11 12 13 8 9 10 11 12 13
2θ (º) 2θ (º)
Figure 5. (a)
(a) Grazing
Grazing incidence
incidence XRD
XRD (GI-XRD)
(GI-XRD) of
of thin
thin film
film and
and (b) XRD of the bulk sample of
composition 80SiO
composition 3+(TEOS-MTES)
80SiO22-20LaF33 0.5Er3+ (TEOS-MTES)treated
treated at
at 550 ◦ C for
550 °C for 11 min.
min. The
The crystal
crystal size is given
together with the LaF33 reference shown with vertical bars (JCPDS 00-032-0483).
3.4. High Resolution

3.4. High Resolution Transmission
Transmission Electron
Electron Microscopy
Microscopy (HRTEM)
(HRTEM)and
andNanostructure
NanostructureCharacterization
Characterization
HRTEM
HRTEM was was used
used to
toreveal
revealthe
thelattice
latticeplanes
planesand
and crystal-size
crystal-size distribution
distribution of LaF
of LaF 3 crystals
3 crystals
in
in the
the thin films (Figure 6a–c) and self-supported layers (Figure 6d–f) treated at 550 ◦ C for 1 min.
thin films (Figure 6a–c) and self-supported layers (Figure 6d–f) treated at 550 °C for 1 min. Small
Small crystals
crystals are homogeneously
are homogeneously distributed
distributed in thematrix.
in the glass glass matrix. The crystal
The crystal size is size is around
around ±
2.0nm
2.0 ± 0.5
and 8.0 ± 0.5 nm for films and self-supported layers, respectively, in agreement with the GI-XRD and
XRD results. A similar structure is observed in both samples, consisting of separated crystals
surrounded by the residual SiO2 matrix. Fourier Transformation of the (111) crystal planes of the LaF3
Tysonite phase was employed to estimate an interplanar distance of 0.33 nm [53]. These results
demonstrate the absence of clusters even for such short heat treatment times and further support the
0.5 nm and 8.0 ± 0.5 nm for films and self-supported layers, respectively, in agreement with the GI-XRD
and XRD results. A similar structure is observed in both samples, consisting of separated crystals
surrounded by the residual SiO2 matrix. Fourier Transformation of the (111) crystal planes of the
LaF3 Tysonite phase was employed to estimate an interplanar distance of 0.33 nm [53]. These results
demonstrate the absence of clusters even for such short heat treatment times and further support
the hypothesis that the crystallization mechanism involves chemical decomposition followed by
crystal precipitation.
Figure
Figure 6.
6. (a)
(a) High
High resolution
resolution transmission
transmission electron
electron microscopy
microscopy (HRTEM)
(HRTEM) micrograph,
micrograph, (b) (b) lattice
lattice
distance
distance and
and (c)
(c) crystal-size
crystal-size distribution of of 80SiO 0.5Er3+3+thin
80SiO22-20LaF33 0.5Er thinfilm
filmtreated
treatedatat550
550°C◦ Cfor
for11min.
min.
(d)
(d) HRTEM micrograph, (e) lattice distance and (f) crystal-size distribution of 80SiO -20LaF
and (f) crystal-size distribution of 80SiO2 -20LaF3
2 3 0.5Er 3+
3+
self-supported
self-supported layer
layer treated
treated at
at 550 ◦
550 °C C for
for 11 min.
min.
3.5. XAS,
3.5. XAS, EPR
EPR and
and RE
RE Environment
Environment
AA crucial
crucial aspect
aspect of of RE-doped
RE-doped OxGCs OxGCs concerns
concerns the the distribution
distribution of of dopants
dopants and and the
the mechanism
mechanism
of incorporation into the fluoride NCs. In fact, RE ions are much more
of incorporation into the fluoride NCs. In fact, RE ions are much more efficient when are embedded efficient when are embedded
in low-phonon-energy
in low-phonon-energy fluoride fluoride crystals.
crystals. Evidence
Evidenceof of RE
RE incorporation
incorporation isis often often obtained
obtained by by optical
optical
spectroscopy but sometimes only indirectly, unless specific selective
spectroscopy but sometimes only indirectly, unless specific selective excitation measurements are excitation measurements are
carried out.
carried out. Here,
Here, thethe RE-ion
RE-ion environment
environmentand andits its role
role in
in transforming
transforming the the xerogel
xerogel to to the
the GCs
GCs have
have
been studied in detail using the local structural techniques
been studied in detail using the local structural techniques XAS and EPR. XAS and EPR.
XAS measurements
XAS measurements of of Er
Er22O O33andandErF
ErF3 3were
wereused usedas asaareference
referenceand andcompared
comparedto to the
the results
results ofof
thin films
films doped
doped withwith 0.5Er 3+ 3+
thin 0.5Er3+, ,Figure
Figure 7. 7. The
The presence
presence of of Er
Er3+ isisclearly
clearly indicated
indicated by by the
the absorption
absorption
edge centered around 8370 eV. However, small shifts of the edge
edge centered around 8370 eV. However, small shifts of the edge are produced by different ligand are produced by different ligand
fields. In general, for the same RE ion, the edge energy increases passing
fields. In general, for the same RE ion, the edge energy increases passing from oxides to fluorides, from oxides to fluorides,
due to
due to lesser
lesser hybridization
hybridization of of the
the chemical
chemical bonds
bonds in in fluoride
fluoride compounds
compounds attributable
attributable to to aa higher
higher field
field
strength [54–56]. In this case, the edge was obtained by the first derivative
strength [54–56]. In this case, the edge was obtained by the first derivative of the absorption spectrum of the absorption spectrum
and the
and the edge
edge shifts
shifts from
from 8356.6
8356.6 to to8359
8359eV eVpassing
passingfrom fromErEr 2O2O3 3to
toErF
ErF33.. More
Moreinsights
insightsare areobtained
obtainedby bya
comparison of the post-edge region that contains information
a comparison of the post-edge region that contains information about the local structure. In about the local structure. In particular,
the X-Ray the
particular, Absorption Near EdgeNear
X-Ray Absorption Structure (XANES),(XANES),
Edge Structure shown inshown Figurein7 Figure
as red 7and black
as red andspectra,
black
for ErF3 for
spectra, andErF
Er32 O 3 , respectively,
and Er2O3, respectively,show slightshowdifferences at the white
slight differences at the line and line
white at resonances of ~8375
and at resonances
and
of 8395and
~8375 eV. 8395
The XANES
eV. The spectra
XANESand derivative
spectra of the xerogel
and derivative and OxGC
of the xerogel andthinOxGC filmthin
(bottom inset of
film (bottom
Figure 7) exhibit an absorption edge centered at ~8359 eV, similar to
inset of Figure 7) exhibit an absorption edge centered at ~8359 eV, similar 3to pure ErF3. Hence, a pure ErF . Hence, a fluorine-rich
environmentenvironment
fluorine-rich is already observed
is alreadyinobserved
the xerogel.in theNevertheless, differences
xerogel. Nevertheless, in the XANES
differences in the XANESregion
(8370–8415 eV) are observed compared to ErF , which may indicate a change in the Er 3+ environment
region (8370–8415 eV) are observed compared 3 to ErF3, which may indicate a change in the Er3+
for the xerogel
environment forandthethe OxGC and
xerogel samples.
the OxGCFourier-Transform calculations indicate
samples. Fourier-Transform a narrowing
calculations of the
indicate a
first shell as the xerogel transforms to the OxGC, which is consistent
narrowing of the first shell as the xerogel transforms to the OxGC, which is consistent with the with the development of a higher
symmetry environment
development of a highercorresponding
symmetry environment to the fluoride crystallineto
corresponding phase.
the fluoride crystalline phase.
A linear combination of XANES spectra in the range 8325–8475 eV was performed for the OxGC
sample using the standard ErF3 and Er2O3 materials. Employing the crystal fraction of LaF3 calculated
by Rietveld refinement, approximate fractions of 91% and 9% were calculated for Er3+ in ErF3 and
Er2O3-like environments, respectively. Thus, the Er3+ ions are likely to be situated near fluoride ions
in the xerogel state, precipitating in fluoride-rich crystals after the heat treatment at 550 °C for 1 min.
These results are explained considering that, in the starting sol, the La3+ and Er3+ ions are surrounded
by TFA cations that are screened and impeded from bonding with Si atoms. Hence, in the xerogel
sample, fluorine is still coordinated as in TFA acid [44] and Er3+ ions are principally coordinated by
ErF3
Der. Normalized μ(E)

8359.0 eV
8356.6 eV
Er2O3
8345 8355 8365 8375

Energy (eV)
Normalized μ(E)
Xerogel 150ºC-1h
White Line GC 550 ºC-1min
Der. Normalized μ(E)

8359.1 eV
8358.8 eV
8345 8355 8365 8375

Energy (eV)
Resonances
8325 8355 8385 8415 8445 8475

Energy (eV)
Figure 7. X-Ray absorption spectroscopy (XAS) spectra and derivatives (inset) of Er2 O3 and ErF3
Figure 7. X-Ray absorption spectroscopy (XAS) spectra and derivatives (inset) of Er2O3 and ErF3
references in black and red, respectively (top), and of XAS spectra and derivatives (inset) of
references in black and red, respectively (top), and of XAS spectra and derivatives (inset) of 80SiO2-
80SiO2 -20LaF3+
3 -0.5Er
3+ (TEOS-MTES) xerogel and GC thin film treated at 550 ◦ C for 1 min, in pink and
20LaF3-0.5Er (TEOS-MTES) xerogel and GC thin film treated at 550 °C for 1 min, in pink and blue,
blue, respectively (bottom). The grey line indicates the white line and the red and black lines indicate
respectively (bottom). The grey line indicates the white line and the red and black lines indicate
resonance positions for ErF3 and Er2 O3 , respectively.
resonance positions for ErF3 and Er2O3, respectively.
A linear combination of XANES spectra in the range 8325–8475 eV was performed for the OxGC
sample Once thethe
using crystal fraction
standard ErF3isand
obtained,
Er2 O3 an indication
materials. of the concentration
Employing of RE ions
the crystal fraction in LaF
of LaF 3 crystals
3 calculated
may
by be obtained.
Rietveld Supposing
refinement, that 91%
approximate of Er ions
fractions
3+
of 91%areand embedded
9% wereincalculated
the crystalfor phase
Er and
3+ thatand
in ErF the
3
crystalline fraction is ~18 wt% (XRD), the effective concentration
3+ of Er 3+ ions in the crystal phase is
Er2 O3 -like environments, respectively. Thus, the Er ions are likely to be situated near fluoride ions
around
in 6.5 mol%,
the xerogel representinginan
state, precipitating order of magnitude
fluoride-rich crystals afterincrease
the heatwith respect
treatment to ◦the
at 550 C fornominal
1 min.
concentration
These results are(0.5 mol%). Similar
explained values
considering were
that, obtained
in the startinginsol, previous
the La3+ studies
and Erof 3+samples
ions are prepared
surrounded by
MQ, albeit with a lower fraction of RE ions incorporated in the crystal
by TFA cations that are screened and impeded from bonding with Si atoms. Hence, in the xerogel phase [57].
sample,Further evidence
fluorine is stillsupporting
coordinated theasmechanism
in TFA acidfor [44]theandincorporation
Er3+ ions are REprincipally
ions is obtained by EPRby
coordinated of
the self-supported
fluorine. During thelayers. Figure 8 shows
heat treatment, the EPR spectra
decomposition promotes of xerogel samples doped
the precipitation of LaFwith 0.5 Er3+
3 crystals in
(mol%)Ermeasured
which 3+ ions become at 5 K. The EPR
trapped. signal
Other of a glass,
techniques, such prepared
as Energy byDispersive
MQ, containing 0.5 Er in an
X-Ray Spectroscopy
3+
(EDXS), demonstrated an Eu enrichment in GdF3 NCs after just 1 min of treatment atare
amorphous fluorine-rich environment is also shown for comparison. Several resonances 550observed
◦ C [50],
for the xerogel
further supporting samples, due to the presence
the aforementioned of organic compounds which remain after heat treatment
conclusions.
at 50Once
°C. Er 3+ resonances are associated with the broad signals appearing in the range 500–1000 G.
the crystal fraction is obtained, an indication of the concentration of RE ions in LaF3
Other authors
crystals may bereported
obtained.EPR measurements
Supposing that 91% of of
Er3+
Erin3+ oxyfluoride
ions are embeddedglasses showing the presence
in the crystal phase and of
broad signals at low magnetic fields (high g values) [58]. Moreover, such
that the crystalline fraction is ~18 wt% (XRD), the effective concentration of Er ions in the crystal resonances
3+ disappear at
room temperature (RT) and are only visible below 20 K, indicating that
phase is around 6.5 mol%, representing an order of magnitude increase with respect to the nominal they are associated with Er3+
ions. Such broad

concentration signals are
(0.5 mol%). clearly
Similar related
values weretoobtained
an amorphousin previousenvironment
studies oftypical
samples ofprepared
glasses and
by
liquids. The EPR signal of an oxyfluoride glass doped with
MQ, albeit with a lower fraction of RE ions incorporated in the crystal phase [57]. 0.5 Er 3+ prepared by MQ is also shown for
comparison. The signal

Further evidence around 2300
supporting theG, observed only
mechanism for the in the xerogel, is attributed
incorporation RE ions is to the presence
obtained by EPR of
organic
of compounds. layers.
the self-supported Interesting
Figureresults
8 shows are obtained for the of
the EPR spectra GCxerogel
samples, whichdoped
samples displaywithresonances
0.5 Er3+
(mol%) measured at 5 K. The EPR signal of a glass, prepared by MQ, containing 0.5 Er inafter
associated with paramagnetic species with effective spin S’ = 1/2 in rhombic or lower symmetry 3+ an
just 1 min of
amorphous heat treatment;
fluorine-rich the resonances
environment appearfor
is also shown at comparison.
g values of gSeveral
1 = 11.2; g2 = 4.9 and g3 = 3.1. On
resonances are observed
comparison
for the xerogelwith the EPR
samples, dueresults obtainedoffor
to the presence an Ercompounds
organic
3+-doped LaF3 single crystal [59,60], it can be
which remain after heat treatment
unambiguously concluded that the resonances in the GC samples are those of Er3+ in LaF3 NCs.
at 50 ◦ C. Er3+ resonances are associated with the broad signals appearing in the range 500–1000 G.
Other authors reported EPR measurements of Er3+ in oxyfluoride glasses showing the presence of
broad signals at low magnetic fields (high g values) [58]. Moreover, such resonances disappear at
room temperature (RT) and are only visible below 20 K, indicating that they are associated with Er3+
ions. Such broad signals are clearly related to an amorphous environment typical of glasses and
liquids. The EPR signal of an oxyfluoride glass doped with 0.5 Er3+ prepared by MQ is also shown for
comparison. The signal around 2300 G, observed only in the xerogel, is attributed to the presence of
organic compounds. Interesting results are obtained for the GC samples, which display resonances
associated with paramagnetic species with effective spin S’ = 1/2 in rhombic or lower symmetry after
just 1 min of heat treatment; the resonances appear at g values of g1 = 11.2; g2 = 4.9 and g3 = 3.1.
On comparison with the EPR results obtained for an Er3+ -doped LaF3 single crystal [59,60], it can
be unambiguously concluded that the resonances in the GC samples are those of Er3+ in LaF3 NCs.
Broader signals
signals areare observed
observed for
for the
the GC
GC doped
doped with
with 0.5
0.5 Er 3+ and associated with stronger relaxation
Er3+
Broader and associated with stronger relaxation
processes, due to a moderately high effective concentration.
processes, due to a moderately high effective concentration.
(a) 0.5Er (b) 80SiO2-20LaF3

550ºC - 1min
3+
Fe
EPR signal (arb.units)
3+
g ≈ 4.3 g1 ≈ 4.9 0.5Er
EPR signal (a.u)
Er in fluorine-rich
droplet
Glass (MQ)
g3 ≈ 3.1
3+
0.1Er
Er in fluorine surrounding
g1 ≈11.2
xerogel (SG)
0 500 1000 1500 2000 2500 3000 0 500 1000 1500 2000 2500 3000
Magnetic field (G) Magnetic field (G)
Figure 8.8.(a) (a)

Figure Electron
Electron Paramagnetic
Paramagnetic Resonance
Resonance (EPR)(EPR) signals
signals for thefor the2-20LaF
80SiO 80SiO23-0.5Er
-20LaF -0.5Er3+
3+3 (TEOS-
3+
(TEOS-MTES). xerogel and of an MQ glass doped with 0.5Er . (b) EPR signals of 80SiO2 -20LaF3
MTES). xerogel and of an MQ glass doped with 0.5Er3+. (b) EPR signals of 80SiO2-20LaF3 (TEOS-MTES)
(TEOS-MTES) GCs doped with 0.1 and 0.5Er3+ (mol%) treated at 550 ◦ C for 1 min.
GCs doped with 0.1 and 0.5Er3+ (mol%) treated at 550 °C for 1 min.
For the composition with a lower concentration of dopant, the hyperfine structure is only partially
For the composition with a lower concentration of dopant, the hyperfine structure is only
resolved, due to the interaction of the electrons with nuclear spins. The low intensity of the hyperfine
partially resolved, due to the interaction of the electrons with nuclear spins. The low intensity of the
structure is due to the quantity of Er isotope nuclei with I = 7/2 (~23%), while the remaining nuclei
hyperfine structure is due to the quantity of Er isotope nuclei with I = 7/2 (~23%), while the remaining
(77%) with I = 0 do not produce any hyperfine interaction [58,61,62]. For the composition doped with
nuclei3+
(77%) with I = 0 do not produce any hyperfine interaction [58,61,62]. For the composition doped
0.5 Er , the hyperfine structure is too broad to be resolved but its presence may be discerned from the
with 0.5 Er3+, the hyperfine structure is too broad to be resolved but its presence may be discerned
shape of the resonance at g1 = 11.2 which presents broader sidebands. Another interesting result is the
from the shape of the resonance at g1 = 11.2 which presents broader sidebands. Another interesting
absence of any further Er3+ resonance other than that of Er3+ in the LaF NCs.
result is the absence of any further Er3+ resonance other than that of Er3+3in the LaF3 NCs.
These results, together with the previous discussion of XAS measurements, indicate that dopant
These results, together with the previous discussion of XAS measurements, indicate that dopant
incorporation occurs with crystal precipitation. In contrast, in melted glasses, residual contributions
incorporation occurs with crystal precipitation. In contrast, in melted glasses, residual contributions
of RE ions in an amorphous environment are still observed even after the crystallization process,
of RE ions in an amorphous environment are still observed even after the crystallization process,
confirming that some RE ions remain in the glass matrix or in phase-separation droplets.
confirming that some RE ions remain in the glass matrix or in phase-separation droplets.
3.6. Photoluminescence (PL)
3.6. Photoluminescence (PL)
The near-infrared emission spectra corresponding to the 4 I13/2 →4 I15/2 transition for 80SiO2 -
The near-infrared
20LaF3 -0.5Er emissionself-supported
3+ (TEOS-MTES) spectra corresponding the 4I13/2at→550
layer heattotreated
4I15/2 transition for 80SiO2-20LaF3-
◦ C for 1 min were obtained at
0.5Er 3+ (TEOS-MTES) self-supported layer heat treated 4at 550 °C
4
room temperature and at 9 K by exciting at 793 nm in the I15/2 → I9/2 absorption for 1 min were obtained
band. at room
After excitation
temperature and at 9 K by exciting at 793 nm in the 4I15/2→4I9/2 absorption band. After excitation of
of this level, the next lower levels are populated by multiphonon relaxations. The room temperature
this level,spectrum
emission the next lower
(Figurelevels are populated
9a) shows a series of by multiphonon
peaks with a maximum relaxations. The room
of around temperature
1540 nm. Figure 9b
emission
shows the excitation spectrum corresponding to the I15/2 → I9/2 transition obtained bynm.
spectrum (Figure 9a) shows a series of peaks 4 with a 4maximum of around 1540 Figure
collecting
9b shows the excitation spectrum corresponding to the I15/2→ I9/2 transition obtained by collecting the
4 4
luminescence spectrum at 1540 nm, revealing well defined peaks which suggest the partial
incorporation of Er3+ ions in the LaF3 NCs.
the luminescence spectrum at 1540 nm, revealing well defined peaks which suggest the partial
incorporation of Er3+ ions in the LaF3 NCs.
4I →4I T= 295 K 4I 4I
(a) (b) T= 295 K
4I13/2→4I 15/2 T= 2→
4I 15/→4I 9/2
(a) 13/2 15/2 λexc295 K nm
= 793 (b) 15/2 9/2
T= 295
λem K
= 1540 nm
0.05
units)
units)
λexc= 793 nm λem= 1540 nm
0.05 Er 0.5%-550ºC-1m
units)
units)
Er 0.5%-550ºC-1m Er 0.5%-550ºC-1m
Er 0.5%-550ºC-1m
(arb.
(arb.
0.02
(arb.
(arb.
0.02
Intensity
Intensity
Intensity
Intensity
0.00 0.00
0.00 0.00
1400 1450 1500 1550 1600 1650 1700 760 770 780 790 800 810 820 830 840
1400 1450 1500 1550 1600 1650 1700 760 770 780 790 800 810 820 830 840
Emission Wavelength (nm) Excitation Wavelength (nm)
Figure 9.9. (a)
(a) Room temperature
temperature emissionspectrum
spectrum of80SiO
80SiO2-20LaF 3-0.5Er 3+ (TEOS-MTES) bulk
3+
Figure 9. (a) Room
Figure Room temperature emission
emission spectrum of of 80SiO22-20LaF -0.5Er
-20LaF33-0.5Er (TEOS-MTES)bulk
3+ (TEOS-MTES) bulk
sampleheat
sample heattreated
treatedatat550
550°C
◦ C for
for 11 min
min obtained
obtained on exciting
on exciting at
at 793
793 nm.
nm. (b)
(b) Low-temperature
Low-temperature excitation
excitation
sample heat treated at 550°C for 1 min obtained on exciting at 793 nm. (b) Low-temperature excitation
spectrumobtained
spectrum obtainedatat1540
1540nm.
nm.
spectrum obtained at 1540 nm.
The
The incorporation
Theincorporation
incorporation of of Er3+
of Er
Er
3+ into LaF
ions into
3+ ions into LaF333 NCs
LaF NCs is is unambiguously
is unambiguouslyconfirmed
unambiguously confirmedby
confirmed bymeasurements
by measurements
measurements
performed
performed
performed at at low
at low temperature.
low temperature.
temperature. FigureFigure 10
10 shows
shows thethe emission
emissionand
andexcitation
excitationspectra,
spectra,respectively,
respectively,
obtained
obtained at 9
9 K.
K. The
The emission
emission spectrum
spectrum shows
shows sharp
sharp peaks superimposed
peaks superimposed
obtained at 9 K. The emission spectrum shows sharp peaks superimposed on a broad band. on a
onbroad
a band.
broad The
band.
The
excitation
The spectrum
excitation spectrumobtained by
obtained collecting
by the
collecting luminescence
the at
luminescencethe emission
at the maximum
emission
excitation spectrum obtained by collecting the luminescence at the emission maximum also shows also
maximum shows
also
very
very well
shows verydefined
well peaks.
well defined
defined The features
peaks. peaks.
The features exhibited
The features
exhibited by the
thelow-temperature
exhibited
by low-temperature
by spectraconfirm
the low-temperature
spectra confirmconfirm
spectra thepartial
the partial
the
incorporation of Er3+ ions3+ in the NCs and provide evidence of the crystalline and amorphous Er 3+
partial incorporation of Er ions in the NCs and provide evidence of the crystalline
incorporation of Er ions in the NCs and provide evidence of the crystalline and amorphous Er3+
3+ and amorphous
3+ environments.
environments.
Er
environments.
4
4II13/2→→44II15/2 T= 99 KK T=9 9K K
(a)
(a) T= (b)
(b) 4I4I15/→44 T=
2→I I9/2
0.04 13/2 15/2
0.04 λλexc== 793
exc
793 nm
nm 15/2 9/2 λem
λem= =1511
1511nm
nm
units)
units)
(arb. units)
(arb. units)
Er 0.5%-550ºC-1m
Er 0.5%-550ºC-1m ErEr0.5%-550ºC-1m
0.5%-550ºC-1m
Intensity (arb.
Intensity (arb.
0.02
0.02
0.02
0.02
Intensity
Intensity
0.00
0.00 0.00
0.00
1400 1450 1500 1550 1600
1400 1600 1650
1650 1700
1700 760
760 770
770 780
780 790
790 800
800 810
810 820
820830
830840
840
Emission Wavelength
Wavelength (nm)
(nm) Excitation
ExcitationWavelength
Wavelength(nm)
(nm)
Figure 10.
Figure 10. (a)
(a) Low temperature
Low temperature emission
temperatureemission spectrum
emissionspectrum
spectrumof of 80SiO
of80SiO
80SiO 2-20LaF 3-0.5Er
2-20LaF
3+
3-0.5Er 3+(TEOS-MTES)
3+ (TEOS-MTES)bulk bulk
Figure 2 -20LaF3 -0.5Er (TEOS-MTES) bulk
sample
sample heat treated at 550◦°C for
for 1min
1min on
on exciting
exciting at
at793
793nm.
nm. (b)
(b)Low
Low temperature
temperature excitation spectrum
excitation
sample heat treated at 550 C for 1 min on exciting at 793 nm. (b) Low temperature excitation spectrum spectrum
obtainedat
obtained
obtained at1511
1511nm.
nm.
Measurementsperformed
Measurements
Measurements performedfor
performed for the
forthe
the 80SiO
80SiO
80SiO 22-20LaF
-20LaF 33-0.1Er
-0.1Er (TEOS)
3+3+3+(TEOS)
(TEOS) bulk
bulk sample
sample heat
heat treated atat
treated at650◦C
650
2 -20LaF 3 -0.1Er bulk sample heat treated 650
°Cfor
°C
for 1for 1h suggest
h 1h suggest
suggest thatthat
that Er3+
3+
Er Er 3+ ions
ions
ions are mainly
aremainly
are incorporated
incorporatedinin
mainlyincorporated LaF
inLaF
LaF 3 NCs.
NCs.As
3 3NCs. Asan
As anexample,
an example,Figure
example, Figure11a
Figure 11ashows
11a shows
shows
the low-temperature
the low-temperature emission
the emission spectrum
emission spectrum obtained
spectrumobtained on exciting
obtainedononexcitingexciting at 793
at at nm.
793793 In
nm.nm. this
In thiscase, sharp
case,case,
In this and
sharpsharp well-
and well-
and
defined peaks
defined
well-defined peaks
peaksare are
are observed.
observed.
observed. Moreover,
Moreover,
Moreover, the excitation
thetheexcitation
excitation spectrum
spectrum
spectrum obtained
obtainedby
obtained bycollecting
by collectingthe the
the
luminescenceatat
luminescence
luminescence atthe
the
the emission
emission
emission maximum
maximum
maximum corresponds
corresponds
corresponds to
to aa crystalline
to a crystalline crystalline environment
for Er3+
environment
environment for
forErEr
ions.
3+ ions.
3+ ions.
Similar
Similar emission
Similar
emission emission
spectra arespectra
spectra are
are obtained
obtained obtained
by exciting by
by exciting
differentdifferent
exciting peaks inpeaks
different peaks ininthe
thespectrum.
the spectrum. spectrum.
These results confirm that, in the 80SiO 2-20LaF3-0.1Er3+3+(TEOS) bulk sample, the Er3+3+ions are
These results confirm that, in the 80SiO2-20LaF3-0.1Er (TEOS) bulk sample, the Er ions are
mainly located
mainly located inin the
the LaF
LaF33 NCs.
NCs. Although,
Although, the the emission
emissionand andexcitation
excitationspectra
spectraofofthe
the80SiO
80SiO 2-20LaF3-
2-20LaF3-
0.5Er 3+(TEOS/MTES) bulk sample also show well-defined peaks, they are much broader and less well
0.5Er (TEOS/MTES) bulk sample also show well-defined peaks, they are much broader and less well
3+
resolved than
resolved than those
those ofof the
the less
less concentrated
concentrated sample, sample,suggesting
suggestingthe thepresence
presenceofofboth
bothcrystalline
crystallineand and
amorphous Er 3+
3+ environments in the 0.5% Er 3+ -doped
amorphous Er environments in the 0.5% Er -doped 80SiO2-20LaF3 sample.
3+ 80SiO 2-20LaF3 sample.
Nanomaterials
Nanomaterials 2019,
2019, 9, 9, x FOR
x FOR PEERREVIEW
PEER REVIEW 12 12
of 16
of 16
0.08
0.08
(a)(a) 4I4I13/2 4 4I15/2 T=99KK (b) 4I4 4
2→→I49/29/
→ T=T=
9 K9 K
13/2→ I15/2 T= (b) I
15/
15/ 2 I 2
λλexc
exc==793
793 nm
nm λem
λ= 1511 nm
em= 1511 nm
Intensity (arb. units)
units)
(arb.units)
Er 0.1%-650ºC-1h 0.10
Er 0.1%-650ºC-1h 0.10 Er 0.1%-650ºC-1h
Er 0.1%-650ºC-1h
Intensity(arb.
0.04
0.04
Intensity
0.05
0.05
0.00 0.00
0.00 0.00
1450 1500 1550 1600 1650 1700 760 770 780 790 800 810 820 830 840
1450 1500 1550
Emission 1600 (nm)
Wavelength 1650 760 770 780 790
Excitation 1700
800 810(nm)
Wavelength 820 830 840
Figure11.11.(a)(a) Low-temperatureemission
Low-temperature emissionspectrum
spectrum of
of the
the 80SiO
80SiO2-20LaF3-0.1Er (TEOS) bulk sample
3+3+
Figure 2 -20LaF3-0.1Er (TEOS) bulk sample
Figure
heat 11. (a)
treated at 650
◦ °C
Low-temperature
for 1h on emission
exciting atspectrum
793 nm. of
(b) the
Low 80SiO -20LaF3-0.1Er
temperature
2
heat treated at 650 C for 1 h on exciting at 793 nm. (b) Low temperature excitation excitation 3+ (TEOS) bulk sample
spectrum
spectrum obtained
obtained
heatat treated
1511 nm. at 650 °C for 1h on exciting at 793 nm. (b) Low temperature excitation spectrum obtained
at 1511 nm.
at 1511 nm.
Another
These resultsimportant
confirmspectroscopic
that, in theparameter to characterize
80SiO2 -20LaF 3+the luminescent behaviorthe of Er3+ ions
Er3+
3 -0.1Er (TEOS) bulk sample, ions
areis mainly
the decaylocated
time of in thespectroscopic
emitting
LaF3level. The Although,
lifetime
to of the
theI13/2 levelthewas obtained at 9K by exciting
Another important parameter characterize
4
luminescent behavior of Erof3+ at
ions
the NCs. emission and excitation spectra the
is 793 decay
the
80SiO nm and
-20LaF collecting
time of the
-0.5Er the luminescence
emitting
3+ (TEOS/MTES) level. The atlifetime
bulk 1511 nm.ofalso
sample The
the decay for
4I13/2 level
show the
was80SiO 2-20LaF
obtained
well-defined at3-0.1Er
peaks, 9K they (TEOS)
by3+exciting
are much at
2 3
793bulk
nm sample
and (Figure 12)
collecting thedeviates from a at
luminescence single
1511 exponential
nm. The function
decay for and
the the lifetime
80SiO 2-20LaF is 3unexpectedly
-0.1Er 3+ (TEOS)
broader and less well resolved than those of the less concentrated sample, suggesting the presence of
short.
bulk The average
sample (Figure lifetime, calculated
12) deviates 3+ aby:
from single exponential function 3+ and the lifetime is unexpectedly
both crystalline and amorphous Er environments in the 0.5% Er -doped 80SiO2 -20LaF3 sample.
short.Another
The average lifetime,
important calculatedparameter
spectroscopic by: 𝑡𝐼 𝑡 𝑑𝑡
to characterize the luminescent behavior of Er3+ ions
〈𝜏〉 = 4 , (2)
is the decay time of the emitting level. The lifetime of 𝐼 the
𝑡𝐼 𝑡𝑡 𝑑𝑡𝑑𝑡I13/2 level was obtained at 9K by exciting at
793 nm and collecting the luminescence at 〈𝜏〉 1511= nm. The decay , for the 80SiO2 -20LaF3 -0.1Er3+ (TEOS) (2)
where I(t) represents the luminescence intensity at time 𝐼 𝑡 t𝑑𝑡 corrected for the background, is around 0.79
bulk sample (Figure 12) deviates from a single exponential function and the lifetime is unexpectedly
ms. This lifetime is nearly one order of magnitude shorter than those observed in bulk samples
where I(t) represents
short. The the luminescence
average lifetime, calculated by:intensity at time t corrected for the background, is around 0.79
prepared by MQ [13]. Similar behavior has also been observed in Nd3+-doped 80SiO2-20LaF3 thin
ms. This lifetime is nearly one order of magnitude shorter than those observed in bulk samples
films, and attributed to the fact that the effectiveR ∞ REtI3+(-iondtconcentration in the NCs differs by up to
prepared by MQ [13]. Similar behavior has alsoR0 beent)observed
hτ i(or
= , in Nd3+-doped 80SiO2-20LaF3 thin (2)
an order of magnitude from the theoretical nominal)
∞ value [12,52].
0REI (-ion
films, and 3+ t ) dt concentration in the NCs differs by up to
The attributed to the fact
room temperature that the
emission effective
spectra of 80SiO 2-20LaF3 (TEOS-MTES) thin films doped with
an0.1
order
andof0.5magnitude from the
Er3+ heat treated theoretical
at 550 °C for 1 (or nominal) value
h corresponding to [12,52].
the 4I13/2→4I15/2 transition obtained by
where I(t)room
The represents the luminescence
temperature emission intensity
spectra of at time
80SiO t corrected for the background, is around
2-20LaF3 (TEOS-MTES) thin films doped with
exciting at 980 nm reveal a glass-like behavior. Figure 13 shows the emission spectra for thin films
0.79 ms.
0.1doped
and 0.5This lifetime
heat
Er3+0.1 is nearly one order
°C for of magnitude shorter than those
→4Iobserved in bulk samples
with andtreated
0.5% Erat3+.550
In both 1 h corresponding
samples the spectrum to the 4Ian
shows 15/2 transition
13/2unresolved structure
3+ -doped 80SiO
obtained
with aby
prepared
exciting by
at 980 MQ [13].
nm reveal Similar behavior has also been observed in Nd -20LaF
2for thin thin
3films
maximum at around 1540a nm
glass-like behavior.the
which indicates Figure
absence
3+
13 of shows the emission
a crystalline spectrafor
environment Er3+ ions.
films, and
doped withattributed
0.1 and 0.5%to theErfact that
3+. In the samples
both effective the
RE spectrum
-ion concentrationshows aninunresolved
the NCs differs by upwith
structure to ana
order of magnitude from the theoretical
-2 (or nominal) value [12,52].
maximum at around 1540 nm which indicates the absence of a crystalline environment for Er3+ ions.
T= 9 K
λem= 1511 nm
-2-4
Ln [I(t) (arb. units)]
λT= =K
exc9 793 nm
λem= 1511 nm
-4-6
Ln [I(t) (arb. units)]
λexc= 793 nm
-6-8
-10
-8
0 1 2 3 4 5 6 7 8
-10 Time (ms)
Figure 12. Semi-logarithmic plot of the fluorescence decay of the 4I13/2 level for the 80SiO2-20LaF3-
0 1 2 3 4 5 6 7 8
0.1Er3+ (TEOS) bulk sample doped with 0.1% Er3+Time obtained
(ms)
by exciting at 793 nm and collecting the
luminescence at 1511 nm.
Figure 12. Semi-logarithmic
Figure 12. Semi-logarithmic plot
plot of
of the
the fluorescence
fluorescencedecay
decayof ofthe
the4 I413/2
I13/2 level for the
level for the 80SiO
80SiO22-20LaF
-20LaF33--
0.1Er (TEOS)
0.1Er 3+
3+ (TEOS)bulk
bulksample
sample doped
doped with
with 0.1%
0.1% Er
Er obtained
3+
3+ obtainedby byexciting
excitingat at 793
793 nm
nm and
and collecting
collecting the
the
luminescence
luminescence atat 1511
1511 nm.
nm.
The room temperature emission spectra of 80SiO2 -20LaF3 (TEOS-MTES) thin films doped with
0.1 and 0.5 Er3+ heat treated at 550 ◦ C for 1 h corresponding to the 4 I13/2 →4 I15/2 transition obtained by
exciting at 980 nm reveal a glass-like behavior. Figure 13 shows the emission spectra for thin films
doped with 0.1 and 0.5% Er3+ . In both samples the spectrum shows an unresolved structure with a
maximum at around 1540 nm which indicates the absence of a crystalline environment for Er3+ ions.
4I 4 T= 295 K
13/2→ I15/2 λexc= 980 nm
0.4
Er 0.1%-550ºC-1h
Er 0.5%-550ºC-1h
0.2
0.0
1400 1450 1500 1550 1600 1650 1700
Emission Wavelength (nm)
Figure 13.
Figure Room temperature
13. Room temperature emission
emission spectrum
spectrum of of 80SiO -20LaF33 (TEOS-MTES)
80SiO22-20LaF (TEOS-MTES) thin
thin films
films doped
doped
with 0.1% (black line) and 0.5% (red line) Er 3+ heat treated at 550 ◦ C for 1 h obtained on exciting at
with 0.1% (black line) and 0.5% (red line) Er heat treated at 550 °C for 1 h obtained on exciting at 980
3+
980 nm.
nm.
4. Conclusions
4. Conclusions
For the first time the preparation of transparent oxyfluoride glass ceramics of composition
For the first time the preparation of transparent oxyfluoride glass ceramics of composition
3+ -dopant have
80SiO2 -20LaF3 containing Er been prepared by sol-gel using TEOS and MTES as
80SiO2-20LaF3 containing Er - dopant have been prepared by sol-gel using TEOS and MTES as silica
3+
silica precursors. XRD and HRTEM showed the precipitation of LaF3 nanocrystals, with a size of
precursors. XRD and HRTEM showed the precipitation of LaF3 nanocrystals, with a size of 9 and 3
9 and 3 nm, after heat treatment at 550 ◦ C for 1 min for bulk and thin film, respectively. Rietveld
nm, after heat treatment at 550 °C for 1 min for bulk and thin film, respectively. Rietveld refinement
refinement indicated a LaF3 crystal fraction of ~18 wt% for the bulk glass ceramic. To our knowledge,
indicated a LaF3 crystal fraction of ~18 wt% for the bulk glass ceramic. To our knowledge, this
this represents the highest crystal fraction reported in the literature for transparent oxyfluoride glass
represents the highest crystal fraction reported in the literature for transparent oxyfluoride glass
ceramics prepared by both melt-quenching and sol-gel methods. XAS and EPR shown a fluorine-rich
ceramics prepared by both melt-quenching and sol-gel methods. XAS and3+ EPR shown a fluorine-rich
amorphous environment in the xerogel and confirmed the presence of Er 3+ in the LaF3 nanocrystals
amorphous environment in the xerogel and confirmed the presence of Er in the LaF3 nanocrystals
after crystallization, with a concentration around one order of magnitude higher than the nominal value.
after crystallization, with a concentration around one order of magnitude higher than the nominal
No XAS and EPR results have previously been reported for RE-doped oxyfluoride glass-ceramics
value. No XAS and EPR results have previously been reported for RE-doped oxyfluoride glass-
prepared by sol-gel. The incorporation of Er3+ ions into LaF 3 nanocrystals was unambiguously
ceramics prepared by sol-gel. The incorporation of Er3+ ions into LaF3 nanocrystals was
confirmed by measurements performed at low temperature. For doping levels higher than 0.1 mol%,
unambiguously confirmed by measurements performed at low temperature. For doping levels higher
contributions from crystalline and amorphous environments are observed. A clear difference is found
than 0.1 mol%, contributions from crystalline and amorphous environments are observed. A clear
between the emission spectra of thin films and bulk samples, likely related to the different crystal size.
difference is found between the emission spectra of thin films and bulk samples, likely related to the
different crystal size. Conceptualization, G.G., J.J.V., J.M., R.B., J.F., A.D. and Y.C.; methodology, G.G., J.J.V.,
Author Contributions:
A.S, M.V., G.R.C., D.B. and R.B.; software, G.G., J.J.V., G.C.M., A.S., M.V. and R.B.; formal analysis, G.G., J.J.V.,
Author Contributions:
A.S., M.V., Conceptualization,
G.R.C., D.B. and R.B.; investigation,G.G., J.J.V.,
G.G., J.M.,
J.J.V., R.B.,
A.S., J.F.,G.R.C.,
M.V., A.D. andD.B.Y.C.;
and methodology,
R.B.; resources,G.G.,
A.D.,J.J.V.,
Y.C.,
A.S, M.V.,
G.R.C., G.R.C.,
D.B., D.B. J.F.;
R.B. and anddata
R.B.;curation,
software,G.G.,
G.G.,J.J.V.,
J.J.V.,A.S.,
G.C.M.,
M.V.,A.S., M.V.D.B.
G.R.C., and and
R.B.;R.B.;
formal analysis, G.G., J.J.V.,
writing—original draft,
G.G., M.V.,
A.S., J.J.V., J.M., A.S.,
G.R.C., M.V.
D.B. and and R.B.;
R.B.; writing—review
investigation, G.G.,and editing,
J.J.V., A.S., G.G.,
M.V.,J.J.V., G.C.M.,
G.R.C., D.B. G.R.C.,
and R.B.; D.B., R.B., J.F.,A.D.,
resources, A.D.
and Y.C.;
Y.C., visualization,
G.R.C., G.G.,
D.B., R.B. and J.J.V.,
J.F.; dataA.S., M.V., G.G.,
curation, D.B., R.B.
J.J.V.,and J.F.;M.V.,
A.S., supervision, R.B.,and
G.R.C., D.B. J.F.,R.B.;
A.D.writing—original
and Y.C.; project
administration, R.B., J.F., J.M., A.D. and Y.C.; funding acquisition, R.B., J.F., A.D. and Y.C.
draft, G.G., J.J.V., J.M., A.S., M.V. and R.B.; writing—review and editing, G.G., J.J.V., G.C.M., G.R.C., D.B., R.B.,
Funding:
J.F., ThisY.C.;
A.D. and research was funded
visualization, by J.J.V.,
G.G., Spanish National
A.S., projects
M.V., D.B., R.B.MAT2017-87035-C2-1-P/2-P
and J.F.; supervision, R.B., J.F.,(AEI/FEDER, UE),
A.D. and Y.C.;
MAT2016-75362-C3-1-R,
project administration, R.B., andJ.F.,
Basque
J.M.,Country
A.D. andUniversity
Y.C.; funding PPG17/07 and GIU17/014.
acquisition, R.B., J.F., A.D. and Y.C..
Acknowledgments: We acknowledge The European Synchrotron Radiation Facility (ESRF), MINECO, and CSIC
Funding: This research was funded by Spanish National projects MAT2017-87035-C2-1-P/2-P (AEI/FEDER, UE),
for the provision of measurements using the BM25-SpLine beamline. This paper is a part of dissemination
MAT2016-75362-C3-1-R,
activities of project FunGlass.and Basque Country
This project University
has received PPG17/07
funding from and GIU17/014.
the European Union’s Horizon 2020 research
and innovation programme under grant agreement No 739566. A.S. acknowledges the financial support from the
Acknowledgments: We acknowledge The European Synchrotron Radiation Facility (ESRF), MINECO, and CSIC
Comunidad de Madrid for an “Atracción de Talento Investigador” contract (No. 2017-t2/IND5395).
for the provision of measurements using the BM25-SpLine beamline. This paper is a part of dissemination
Conflicts of
activities of Interest: The authors
project FunGlass. Thisdeclare no has
project conflict of interest.
received funding from the European Union´s Horizon 2020
research and innovation programme under grant agreement No 739566. A.S. acknowledges the financial support
from the Comunidad de Madrid for an “Atracción de Talento Investigador” contract (No. 2017-t2/IND5395).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Wang, Y.; Ohwaki, J. New transparent vitroceramics codoped with Er3+ and Yb3+ for efficient frequency
upconversion. Appl. Phys. Lett. 1993, 63, 3268–3270. [CrossRef]
2. Dejneka, M.J. The luminescence and structure of novel transparent oxyfluoride glass-ceramics.
J. Non-Cryst. Solids 1998, 239, 149–155. [CrossRef]
3. Tanabe, S.; Hayashi, H.; Hanada, T.; Onodera, N. Fluorescence properties of Er3+ ions in glass ceramics
containing LaF3 nanocrystals. Opt. Mater. 2002, 19, 343–349. [CrossRef]
4. Lavin, V.; Iparraguirre, I.; Azkargorta, J.; Mendioroz, A.; González-Plata, J.; Balda, R.; Fernández, J. Stimulated
and upconverted emissions of Nd3+ in a transparent oxyfluoride glass-ceramic. Opt. Mater. (Amst.) 2004,
25, 201–208. [CrossRef]
5. Mattarelli, M.; Tikhomirov, V.K.; Seddon, A.B.; Montagna, M.; Moser, E. Tm3+ -activated transparent
oxy-fluoride glass–ceramics: Structural and spectroscopic properties. J. Non-Cryst. Solids 2004, 346, 354–358.
[CrossRef]
6. Chen, D.; Wang, Y.; Yu, Y.; Huang, P.; Weng, F. Near-infrared quantum cutting in Ho3+ /Yb3+ codoped
nanostructured glass ceramic. Opt. Lett. 2011, 36, 876–878.
7. de Pablos-Martín, A.; Durán, A.; Pascual, M.J. Nanocrystallization in oxyfluoride systems: Mechanisms of
crystallization and photonic properties. Int. Mater. Rev. 2012, 57, 165–186. [CrossRef]
8. Wei, Y.; Li, J.; Yang, J.; Chi, X.; Guo, H. Enhanced green upconversion in Tb3+ -Yb3+ co-doped oxyfluoride
glass ceramics containing LaF3 nanocrystals. J. Lumin. 2013, 137, 70–72. [CrossRef]
9. de Pablos-Martín, A.; Ferrari, M.; Pascual, M.J.; Righini, G.C. Glass-ceramics: A class of nanostructured
materials for photonics. Riv. Nuovo Cimento 2015, 38, 311–369.
10. Fedorov, P.P.; Luginina, A.A.; Popov, A.I. Transparent oxyfluoride glass ceramics. J. Fluor. Chem. 2015,
172, 22–50. [CrossRef]
11. Pascual, M.J.; Garrido, C.; Durán, A.; Pascual, L.; de Pablos-Martín, A.; Fernández, J.; Balda, R. Optical
Properties of Transparent Glass–Ceramics Containing Er3+ -Doped Sodium Lutetium Fluoride Nanocrystals.
Int. J. Appl. Glass Sci. 2016, 40, 27–40. [CrossRef]
12. Gorni, G.; Velázquez, J.J.; Mather, G.C.; Durán, A.; Chen, G.; Sundararajan, M.; Balda, R.; Fernández, J.;
Pascual, M.J. Selective excitation in transparent oxyfluoride glass-ceramics doped with Nd3+ . J. Eur.
Ceram. Soc. 2017, 37, 1695–1706. [CrossRef]
13. Gorni, G.; Balda, R.; Fernández, J.; Pascual, L.; Durán, A.; Pascual, M.J. Effect of the heat treatment on the
spectroscopic properties of Er3+ -Yb3+ - doped transparent oxyfluoride nano-glass-ceramics. J. Lumin. 2018,
193, 51–60. [CrossRef]
14. Velázquez, J.J.; Balda, R.; Fernández, J.; Gorni, G.; Pascual, L.; Chen, G.; Sundararajan, M.; Durán, A.;
Pascual, M.J. Transparent oxy fl uoride glass-ceramics with NaGdF4 nanocrystals doped with Pr3+ and
Pr3+ -Yb3+ . J. Lumin. 2018, 193, 61–69. [CrossRef]
15. Samson, B.N.; Tick, P.A.; Borrelli, N.F. Efficient neodymium-doped glass-ceramic fiber laser and amplifier.
Opt. Lett. 2001, 26, 145–147. [CrossRef]
16. Reben, M.; Jaglarz, J. Nd3+ -Doped Oxyfluoride Glass Ceramics Optical Fibre with SrF2 Nanocrystals.
Opt. Appl. 2012, 42, 353–364.
17. Lisiecki, R.; Augustyn, E.; Ryba-romanowski, W.; Zelechower, M. Er-doped and Er, Yb co-doped oxyfluoride
glasses and glass–ceramics, structural and optical properties. Opt. Mater. 2011, 33, 1630–1637. [CrossRef]
18. Kang, S.; Fang, Z.; Huang, X.; Chen, Z.; Yang, D.; Xiao, X.; Qiu, J.; Dong, G. Precisely controllable fabrication
of Er3+ -doped glass ceramic fibers: Novel mid-infrared fiber laser materials. J. Mater. Chem. C 2017,
5, 4549–4556. [CrossRef]
19. Gorni, G.; Balda, R.; Fernández, J.; Iparraguirre, I.; Velázquez, J.J.; Castro, Y.; Pascual, L.; Chen, G.;
Sundararajan, M.; Pascual, M.J.; et al. Oxyfluoride glass–ceramic fibers doped with Nd3+ : Structural
and optical characterization. CrystEngComm 2017, 19, 6620–6629. [CrossRef]
20. Kang, S.; Yu, H.; Ouyang, T.; Chen, Q.; Huang, X.; Chen, Z.; Qiu, J.; Dong, G. Novel Er3+ /Ho3+ -codoped
glass-ceramic fibers for broadband tunable mid-infrared fiber lasers. J. Am. Ceram. Soc. 2018, 101, 3956–3967.
[CrossRef]
21. Fujihara, S.; Mochizuki, C.; Kimura, T. Formation of LaF3 microcrystals in sol-gel silica. J. Non-Cryst. Solids
1999, 244, 267–274. [CrossRef]
22. Biswas, A.; Maciel, G.S.; Friend, C.S.; Prasad, P.N. Upconversion properties of a transparent Er3+ -Yb3+
co-doped LaF3 -SiO2 glass-ceramics prepared by sol-gel method. J. Non-Cryst. Solids 2003, 316, 393–397.
[CrossRef]
23. Fujihara, S.; Koji, S.; Kimura, T. Structure and optical properties of (Gd,Eu)F3 -nanocrystallized sol–gel silica
films. J. Mater. Chem. 2004, 14, 1331–1335. [CrossRef]
24. Ribeiro, S.J.L.; Araújo, C.C.; Bueno, L.A.; Gonçalves, R.R.; Messaddeq, Y. Sol-gel Eu3+ /Tm3+ doped
transparent glass-ceramic waveguides. J. Non-Cryst. Solids 2004, 348, 180–184. [CrossRef]
25. Luo, W.; Wang, Y.; Bao, F.; Zhou, L.; Wang, X. Crystallization behavior of PbF2-SiO2based bulk xerogels.
J. Non-Cryst. Solids 2004, 347, 31–38. [CrossRef]
26. Zhou, L.; Chen, D.; Luo, W.; Wang, Y.; Yu, Y.; Liu, F. Transparent glass ceramic containing Er3+ : CaF2
nano-crystals prepared by sol–gel method. Mater. Lett. 2007, 61, 3988–3990. [CrossRef]
27. Velázquez, J.J.; Yanes, A.C.; Del-Castillo, J.; Méndez-Ramos, J.; Rodríguez, V.D. Optical properties
of Ho3+ -Yb3+ co-doped nanostructured SiO2 -LaF3 glass-ceramics prepared by sol-gel method.
Phys. Status Solidi 2007, 204, 1762–1768. [CrossRef]
28. del-Castillo, J.; Yanes, A.C. Bright luminescence of Gd3+ sensitized RE3+ -doped SiO2 -BaGdF5 glass-ceramics
for UV-LEDs colour conversion. J. Alloys Compd. 2017, 695, 3736–3743.
29. Szpikowska-Sroka, B.; Pawlik, N.; Goryczka, T.; Pietrasik, E.; Bańczyk, M.; Pisarski, W.A. Lead fluoride
β-PbF2 nanocrystals containing Eu3+ and Tb3+ ions embedded in sol-gel materials: Thermal, structural and
optical investigations. Ceram. Int. 2017, 43, 8424–8432. [CrossRef]
30. Szpikowska-Sroka, B.; Pawlik, N.; Goryczka, T.; Pisarski, W.A. Influence of silicate sol-gel host matrices
and catalyst agents on the luminescent properties of Eu3+ /Gd3+ under different excitation wavelengths.
RSC Adv. 2015, 5, 98773–98782. [CrossRef]
31. Secu, C.E.; Bartha, C.; Polosan, S.; Secu, M. Thermally activated conversion of a silicate gel to an oxyfluoride
glass ceramic: Optical study using Eu3+ probe ion. J. Lumin. 2014, 146, 539–543. [CrossRef]
32. Secu, C.E.; Negrea, R.F.; Secu, M. Eu3+ probe ion for rare-earth dopant site structure in sol–gel derived LiYF4
oxyfluoride glass–ceramic. Opt. Mater. 2013, 35, 2456–2460. [CrossRef]
33. Santana-Alonso, A.; Méndez-Ramos, J.; Yanes, A.C.; Del-Castillo, J.; Rodríguez, V.D. Up-conversion in sol-gel
derived nano-glass-ceramics comprising NaYF4 nano-crystals doped with Yb3+ , Ho3+ and Tm3+ . Opt. Mater.
2010, 32, 903–908. [CrossRef]
34. Szpikowska-Sroka, B.; Zur, L.; Czoik, R.; Goryczka, T.; Swinarew, A.S.; Zadlo, M.; Pisarski, W.A. Long-lived
emission from Eu3+ :PbF2 nanocrystals distributed into sol–gel silica glass. J. Sol-Gel Sci. Technol. 2013,
68, 278–283. [CrossRef]
35. Ballato, J.; Riman, R.E.; Snitzer, E. Sol–gel synthesis of fluoride optical materials for planar integrated
photonic applications. J. Non-Cryst. Solids 1997, 213–214, 126–136. [CrossRef]
36. Kumar, G.A.; Riman, R.; Snitzer, E. Solution synthesis and spectroscopic characterization of high Er3+ content
LaF3 for broadband 1.5 µm amplification. J. Appl. Phys. 2004, 95, 40–44. [CrossRef]
37. Hench, L.L.; West, J.K. The sol-gel process. Chem. Rev. 1990, 90, 33–72. [CrossRef]
38. Gorni, G.; Velázquez, J.J.; Mosa, J.; Balda, R.; Fernández, J.; Durán, A.; Castro, Y. Transparent Glass-Ceramics
Produced by Sol-Gel: A Suitable Alternative for Photonic Materials. Materials 2018, 11, 212. [CrossRef]
39. Rubio-Zuazo, J.; Ferrer, P.; López, A.; Gutiérrez-León, A.; Silva, I.; Castro, G.R. Nuclear Instruments and
Methods in Physics Research A The multipurpose X-ray diffraction end-station of the BM25B-SpLine
synchrotron beamline at the ESRF. Nucl. Inst. Methods Phys. Res. A 2013, 716, 23–28. [CrossRef]
40. Rodríguez-Carvajal, J. Recent advances in magnetic structure determination by neutron powder diffraction.
Phys. B Condens. Matter 1993, 192, 55–69. [CrossRef]
41. Ravel, B.; Newville, M. ATHENA, ARTEMIS, HEPHAESTUS: Data analysis for X-ray absorption spectroscopy
using IFEFFIT. J. Synchrotron Radiat. 2005, 12, 537–541. [CrossRef] [PubMed]
42. Innocenzi, P.; Abdirashid, M.O.; Guglielmi, M. Structure and properties of sol-gel coatings from
methyltriethoxysilane and tetraethoxysilane. J. Sol-Gel Sci. Technol. 1994, 3, 47–55. [CrossRef]
43. Ma, Y.; Lee, H.R.; Tsuru, T. Study on Preparation and Hydrophobicity of MTES Derived Silica Sol and Gel.
Adv. Mater. Res. 2012, 535–537, 2563–2566. [CrossRef]
44. Gorni, G.; Pascual, M.J.; Caballero, A.; Velázquez, J.J.; Mosa, J.; Castro, Y.; Durán, A. Crystallization
mechanism in sol-gel oxyfluoride glass-ceramics. J. Non-Cryst. Solids 2018, 501, 145–152. [CrossRef]
45. Yu, Y.; Chen, D.; Cheng, Y.; Wang, Y.; Hu, Z.; Bao, F. Investigation on crystallization and influence of Nd3+
doping of transparent oxyfluoride glass-ceramics. J. Eur. Ceram. Soc. 2006, 26, 2761–2767. [CrossRef]
46. Bhattacharyya, S.; Bocker, C.; Heil, T.; Jinschek, R.; Höche, T.; Rüssel, C.; Kohl, H. Experimental Evidence of
Self-Limited Growth of Nanocrystals in Glass Experimental Evidence of Self-Limited Growth of Nanocrystals
in Glass. Nano Lett. 2009, 9, 2493–2496. [CrossRef]
47. Bhattacharyya, S.; Höche, T.; Hemono, N.; Pascual, M.J.; van Aken, P.A. Nano-crystallization in
LaF3 -Na2 O-Al2 O3 -SiO2 glass. J. Cryst. Growth 2009, 311, 4350–4355. [CrossRef]
48. de Pablos-Martín, A.; Muñoz, F.; Mather, G.C.; Patzig, C.; Bhattacharyya, S.; Jinschek, J.R.; Höche, T.;
Durán, A.; Pascual, M.J. KLaF4 nanocrystallization in oxyfluoride glass-ceramics. CrystEngComm 2013, 15,
10323–10332. [CrossRef]
49. de Pablos-Martín, A.; Hémono, N.; Mather, G.C.; Bhattacharyya, S.; Höche, T.; Bornhöft, H.; Deubener, J.;
Muñoz, F.; Durán, A.; Pascual, M.J. Crystallization kinetics of LaF3 nanocrystals in an oxyfluoride glass.
J. Am. Ceram. Soc. 2011, 94, 2420–2428. [CrossRef]
50. Velázquez, J.J.; Mosa, J.; Gorni, G.; Balda, R.; Fernández, J.; Pascual, L.; Durán, A.; Castro, Y. Transparent
SiO2 -GdF3 sol–gel nano-glass ceramics for optical applications. J. Sol-Gel Sci. Technol. 2019, 89, 322–332.
[CrossRef]
51. Jia, Y.Q. Crystal radii and effective ionic radii of the rare earth ions. J. Solid State Chem. 1991, 95, 184–187.
[CrossRef]
52. Gorni, G.; Balda, R.; Fernández, J.; Velázquez, J.J.; Pascual, L.; Mosa, J.; Durán, A.; Castro, Y. 80SiO2 -20LaF3
oxyfluoride glass ceramic coatings doped with Nd3+ for optical applications. Int. J. Appl. Glass Sci. 2018,
9, 208–217. [CrossRef]
53. Li, F.; Li, C.; Liu, X.; Bai, T.; Dong, W.; Zhang, X.; Shi, Z.; Feng, S. Microwave-assisted synthesis and up–down
conversion luminescent properties of multicolor hydrophilic LaF3 :Ln3+ nanocrystals. Dalt. Trans. 2013,
42, 2015–2022. [CrossRef]
54. Agarwal, B.K.; Verma, L.P. A rule for chemical shifts of X-ray absorption edges. J. Phys. C Solid State Phys.
1970, 3, 535–537. [CrossRef]
55. Choi, Y.G. Covalence of chemical bonds and white-line intensity of an L3-edge X-ray absorption near-edge
structure of rare earth elements embedded in glass. Met. Mater. Int. 2009, 15, 993–999. [CrossRef]
56. Figueroa, A.I.; van der Laan, G.; Harrison, S.E.; Cibin, G.; Hesjedal, T. Oxidation Effects in Rare Earth Doped
Topological Insulator Thin Films. Sci. Rep. 2016, 6, 22935. [CrossRef] [PubMed]
57. de Pablos-Martín, A.; García, M.A.; Muñoz-Noval, A.; Castro, G.R.; Pascual, M.J.; Durán, A. Analysis
of the distribution of Tm3+ ions in LaF3 containing transparent glass-ceramics through X-ray absorption
spectroscopy. J. Non-Cryst. Solids 2014, 384, 83–87. [CrossRef]
58. Dantelle, G.; Mortier, M.; Vivien, D. EPR and optical studies of erbium-doped β-PbF2 single-crystals and
nanocrystals in transparent glass–ceramics. Phys. Chem. Chem. Phys. 2007, 9, 5591–5598. [CrossRef]
[PubMed]
59. Baker, J.M.; Rubins, R.S. Electron Spin Resonance in Two Groups of Lanthanon Salts. Proc. Phys. Soc. 1961,
78, 1353–1360. [CrossRef]
60. Schulz, M.B.; Jefries, C.D. Spin-Lattice Relaxation of Rare-Earth Ions in LaF3 . Phys. Rev. 1966, 149, 270–288.
[CrossRef]
61. Budoyo, R.P.; Kakuyanagi, K.; Toida, H.; Matsuzaki, Y.; Munro, W.J.; Yamaguchi, H.; Saito, S. Electron
paramagnetic resonance spectroscopy of Er3+ :Y2 SiO5 using a Josephson bifurcation amplifier: Observation
of hyperfine and quadrupole structures. Phys. Rev. Mater. 2018, 2, 11403. [CrossRef]
62. Yang, S.; Evans, S.M.; Halliburton, L.E.; Slack, G.A.; Schujman, S.B.; Morgan, K.E.; Bondokov, R.T.;
Mueller, G.; Yang, S.; Evans, S.M.; et al. Electron paramagnetic resonance of ions in aluminum nitride
Electron paramagnetic resonance of Er3+ ions in aluminum nitride. J. Appl. Phys. 2009, 105, 23714–23715.
[CrossRef]
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Gorni Nanomaterials 2019

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gorni Nanomaterials 2019

Uploaded by

Copyright:

Available Formats

nanomaterials

Keywords: glass-ceramic; oxyfluoride; rare-earth; crystallization; sol-gel

Nanomaterials 2019, 9, 530; doi:10.3390/nano9040530 www.mdpi.com/journal/nanomaterials

2. Materials and Methods

2.1. Materials Preparation

2.2. Characterization of Coatings and Self-supported Layers

3. Results and Discussion

3.1. Self-supported Layers and Thin Films

3. Results and Discussion

3.1. Self-supported Layers and Thin Films

Nanomaterials 2019, 9, x FOR PEER REVIEW 5 of 16

function of the wavelength in the transparent region 1000

10 20 30 40 50 60 70 80 90 100 110 120

Table 1. Lattice constants of 80SiO2-20LaF3-0.5Er3+ (TEOS-MTES) glass-ceramic bulk sample treated

Sample a (Å) c (Å) Unit Cell Volume (Å3)

3.4. High Resolution

Der. Normalized μ(E)

8345 8355 8365 8375

Der. Normalized μ(E)

8345 8355 8365 8375

8325 8355 8385 8415 8445 8475

ions. Such broad

comparison. The signal

(a) 0.5Er (b) 80SiO2-20LaF3

Figure 8.8.(a) (a)

Conflicts of Interest: The authors declare no conflict of interest.

You might also like