Synthesis and stability of Sn(II)-containing perovskites: (Ba,SnII)HfIVO3 versus (Ba,SnII)SnIVO3

Article abstract of DOI:10.1016/j.jssc.2021.122419

While Sn(II)-containing perovskite oxides have long drawn attention as Pb(II) substitutes in technologically-relevant dielectric materials, they are also highly thermodynamically unstable and potentially impossible to prepare. Investigations into the new flux-mediated syntheses of metastable Sn(II)-containing hafnate and stannate perovskites were aimed at understanding the key factors related to their synthesizability. The BaHfO₃ perovskite was reacted with SnClF from 250 to 350 ?°C for 12–72 ?h, yielding an unprecedented Sn(II) concentration on the A-site of up to ~70 ?mol%, i.e., (Ba_0.3Sn_0.7)HfO₃ in high purity. Elemental mapping using EDS shows the Sn(II) cations diffuse gradually throughout the crystallites, with two reaction cycles needed to give a nearly homogeneous distribution. In contrast, similar reactions with BaSnO₃ and as little as 10 ?mol% Sn(II) result in decomposition to SnO, SnO₂, and BaSnO₃. The (Ba_1-xSn_x)HfO₃ compositions exhibit a primary cubic perovskite structure (Pm3ˉm; for x ?= ?1/3, 1/2 and 2/3) by powder X-ray diffraction (XRD) methods, with the Sn(II) cations substituted on the A-site. Total energy calculations show the thermodynamic instability versus the ground state (i.e., metastability) for (Ba_1-xSn_x)HfO₃ increases with Sn(II) substitution, reaching a maximum of ~446 ?meV atom^?1 at ~70 ?mol% Sn(II). The decomposition pathway of (Ba_1/3Sn_2/3)HfO₃ was probed by ex situ XRD as well as in situ electron microscopy methods. An onset of thermally-induced decomposition begins at ~350–400 ?°C to give the more stable oxides which are found to segregate out in surface layers. These results help to elucidate the factors underpinning the synthesizability of highly metastable Sn(II)-containing perovskites, which increases with their cohesive energy and with the absence of lower-energy polymorphs or other ground states that can be reached without significant ion diffusion.

Full text of DOI:10.1016/j.jssc.2021.122419

Journal of Solid State Chemistry 302 (2021) 122419
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
II
IV
Synthesis and stability of Sn(II)-containing perovskites: (Ba,Sn )Hf O
3
II
IV
versus (Ba,Sn )Sn O
3
a
a
b
b
a,*
Eric A. Gabilondo , Shaun O'Donnell , Rachel Broughton , Jacob L. Jones , Paul A. Maggard
a
Department of Chemistry, North Carolina State University, Raleigh, NC, 27695-8204, USA
Department of Materials Science and Engineering, North Carolina State University, Raleigh, NC, 27695-8204, USA
b
A R T I C L E I N F O
A B S T R A C T
Keywords:
Perovskite
Metastable solids
Flux synthesis
Metal oxide
While Sn(II)-containing perovskite oxides have long drawn attention as Pb(II) substitutes in technologically-
relevant dielectric materials, they are also highly thermodynamically unstable and potentially impossible to
prepare. Investigations into the new flux-mediated syntheses of metastable Sn(II)-containing hafnate and stannate
perovskites were aimed at understanding the key factors related to their synthesizability. The BaHfO
3
perovskite
ꢀ
was reacted with SnClF from 250 to 350 C for 12–72 h, yielding an unprecedented Sn(II) concentration on the A-
site of up to ~70 mol%, i.e., (Ba0.3Sn0.7)HfO
cations diffuse gradually throughout the crystallites, with two reaction cycles needed to give a nearly homoge-
neous distribution. In contrast, similar reactions with BaSnO and as little as 10 mol% Sn(II) result in decom-
position to SnO, SnO , and BaSnO . The (Ba1-xSn )HfO compositions exhibit a primary cubic perovskite structure
Pm3m; for x ¼ 1/3, 1/2 and 2/3) by powder X-ray diffraction (XRD) methods, with the Sn(II) cations substituted
on the A-site. Total energy calculations show the thermodynamic instability versus the ground state (i.e., meta-
3
in high purity. Elemental mapping using EDS shows the Sn(II)
3
2
3
x
3
(
ꢁ
1
stability) for (Ba1-xSn
mol% Sn(II). The decomposition pathway of (Ba1/3Sn2/3)HfO
x 3
)HfO increases with Sn(II) substitution, reaching a maximum of ~446 meV atom at ~70
3
was probed by ex situ XRD as well as in situ electron
ꢀ
microscopy methods. An onset of thermally-induced decomposition begins at ~350–400 C to give the more
stable oxides which are found to segregate out in surface layers. These results help to elucidate the factors un-
derpinning the synthesizability of highly metastable Sn(II)-containing perovskites, which increases with their
cohesive energy and with the absence of lower-energy polymorphs or other ground states that can be reached
without significant ion diffusion.
1. Introduction
utilizing sufficiently low temperatures for their formation and kinetic
stabilization, such as through soft cation exchange or flux-mediated
The preparation of advanced oxide materials is a scientific challenge
nucleation.
Metastable solids are exemplified in the synthetic pursuit of Sn(II)-
based perovskite oxides [9]. These are of particular interest as alterna-
tives to the isoelectronic Pb(II) analogues because of their enhanced
properties and reduced toxicity. Several Pb(II)-containing perovskite
oxides exhibit technologically-relevant properties, such as ferro- and
that is important for many potential technological applications, such as
for their electronic, magnetic, or thermal properties [1–5]. Many previ-
ous synthetic approaches have been utilized to address solid-state syn-
thetic challenges, including lower-temperature ‘Chimie Douce’
approaches that enable better kinetic control over product formation.
These include hydrothermal, sol-gel, ionic liquids, and molten salt syn-
thetic techniques. The latter approach, also called the flux synthesis
method, facilitates the low-temperature preparation of metastable oxides
in many different chemical systems [6–8]. The effectiveness of this syn-
thetic method arises from the capability to utilize kinetic barriers
inherent to both ion diffusion and nucleation-driven pathways. Suc-
cessful reaction pathways to metastable materials are discoverable by
piezoelectricity in commercial PbZr1-xTi
x 3
O (PZT) [10] or anti-
ferroelectricity in PbHfO [11]. Their detailed and in-depth structural
3
origins have been under investigation for many decades. Their ferroic
properties are currently understood as being largely influenced by the
2
spontaneous polarization, which is induced by the stereoactive 6s
configuration of the Pb(II) cation [11,12]. Unfortunately, the synthesis of
analogous Sn(II)-containing oxides is hindered by synthetic challenges.
Specifically, oxides of the Sn(II) cation undergo facile oxidation to SnO
2
and/or disproportionation-driven decomposition to mixtures of Sn and
ꢀ
*
Corresponding author.
E-mail address: paul_maggard@ncsu.edu (P.A. Maggard).
SnO
at temperatures as low as 250 C [13]. By comparison, PZT is
2
https://doi.org/10.1016/j.jssc.2021.122419
Received 8 May 2021; Received in revised form 7 July 2021; Accepted 8 July 2021
Available online 9 July 2021
0022-4596/© 2021 Elsevier Inc. All rights reserved.

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
ꢀ
prepared by simple high temperature reactions of the binary oxides or in
a molten flux at 700 C and higher temperatures [14,15]. High temper-
reacted in a box furnace at 1500 C for 48 h. The product was ground
ꢀ
finely and reacted a second time as a pellet to obtain a pure, poly-
ature techniques are not compatible with Sn(II)-containing oxides or
similar reactants, as Sn(II)-containing oxides are commonly metastable.
crystalline BaHfO
was synthesized via molten flux techniques. Similarly, a 10% molar
excess of BaCO was finely ground with SnO (Alfa Aesar, 99.9%) and
3 3
powder. The barium stannate perovskite, BaSnO ,
One notable example is SnTiO
and theoretical studies to be metastable with respect to the ilmenite
structure type [16–19]. Further, the ilmenite-type SnTiO structure is
metastable with respect to thermal decomposition to Sn TiO and TiO
3
, which was found in both experimental
3
2
added to a eutectic mixture of NaCl (Fischer, > 99.5%) and KCl (Fischer,
> 99.5%) in a covered alumina crucible. The mixture was reacted in a
3
ꢀ
2
4
2
.
box furnace at 900 C for 24 h. The obtained powder was washed with
ꢀ
Despite these synthetic difficulties, low-temperature synthetic routes
to new Sn(II)-containing oxides have been discovered in recent reported
research efforts [8,17,20,21]. These recent approaches leverage
low-temperature, kinetically-mediated pathways. A notable example,
150 mL of deionized water and dried at 80 C overnight. The Sn(II)-
exchange reactions were performed by mixing the as-prepared BaHfO
or BaSnO phases with a stoichiometric amount of a peritectic SnCl
SnF mixture (Alfa Aesar; 99%, 97.5%, respectively) in a glovebox, until
3
3
2
/
2
metastable Ba1-xSn
Sn(II) cations, was prepared by reacting the Ba-containing perovskite
BZT) with a low melting SnClF peritectic flux [8]. The reaction is driven
thermodynamically by the formation of BaClF from SnClF, enabling the
formation of the metastable products at low reaction temperatures. The
Zr(IV)-richer perovskite compositions were found to have a significantly
higher synthetic Sn(II)-concentration limit than the Ti(IV)-richer perov-
skite compositions, although calculated to have similar metastability
x
Zr1-yTi
y
O
3
(BSZT) perovskite containing up to 60%
homogeneous. Next, the mixed powders were placed in an evacuated,
fused-silica ampule and flame sealed. The reaction ampules were heated
ꢀ
ꢀ
(
in a box furnace at various temperatures and times, from 250 C to 350 C
and from 12 to 72 h. The products were washed with 150 mL of deionized
water to remove the BaClF byproduct and any unreacted SnClF. The
ꢀ
products were dried at 45 C overnight. A typical reaction mixture did
not exceed 100 mg in total mass. Additional reaction cycles followed the
same protocol, with intermediate grinding under argon atmosphere
before loading into a second reaction vessel.
ꢁ
1
(
~250–270 meV atom at 60% Sn(II)). Metastable phases are generally
ꢁ1
considered synthesizable if they lie within ~100–200 meV atom of the
convex hull, but other factors impacting kinetic stability must also be
considered [22,23]. For example, the higher lattice cohesive energy of
2.2. Structural characterization techniques
ꢁ
1
cubic BaZrO
3
(ꢁ8.26 eV atom ) compared to cubic BaTiO
3
(ꢁ7.92 eV
Experimental XRD was measured on a Rigaku R-Axis Spider using a
ꢁ
1
atom ) can act to increase the kinetic barrier to ion diffusion and thus
hinder decomposition by phase segregation. Additionally, while the
sealed X-ray Cu Kα (λ ¼ 1.54056 Å, 40 kV, 36 mA) source in the Debye-
Scherrer geometry with a curved image-plate detector. For the purposes
Zr(IV)-richer compositions decompose by phase segregation to ZrO
SnO, the decomposition of Ti(IV)-richer compositions to ilmenite-type
SnTiO requires significantly less ion diffusion. However, synthesiz-
ability factors such as these remain poorly explored for other possible
Sn(II)-containing oxide perovskites.
Presented herein is an investigation of the synthesis of Sn(II)-
containing hafnate and stannate perovskites starting from pure
3 3
BaHfO and BaSnO . Low-temperature flux reactions were targeted to
2
and
of Rietveld refinement, additional XRD patterns were measured on a
PANalytical Empyrean X-Ray diffractometer using Cu K
α radiation (λ ¼
1
3
1.54056 Å, λ2 ¼ 1.544426 Å, 45 kV, 40 mA) in the Bragg-Brentano ge-
ometry with a step size of 0.0131 in 2θ and a 180 ms count time per step.
High resolution synchrotron diffraction data sets were taken at the 11-
BM-B beam line (λ ¼ 0.457929 Å) of the Advanced Photon Source
(APS) at Argonne National Laboratory. Rietveld refinements of the data
were performed using the General Structure Analysis System-II (GSAS-II)
software to extract crystal structure information such as atomic occu-
pancies and lattice parameters [28].
explore the Sn(II)-concentration limits in these perovskites, with the full
replacement of Ba(II) for Sn(II) cations potentially yielding the meta-
II IV
II IV
stable Sn Hf O
one of the largest calculated lattice cohesive energies (ꢁ8.72 eV atom
and which is hypothesized to yield a greater potential kinetic stabiliza-
tion during these reactions. Conversely, the BaSnO perovskite has a
significantly smaller lattice cohesive energy (ꢁ6.31 eV atom ). Prior
theoretical reports on SnHfO show that it is metastable as a cubic or
3
and Sn Sn O
3
perovskites. The BaHfO
3
perovskite has
Ex-situ thermal decomposition experiments were carried out on a 40%
ꢁ1
)
Sn(II) sample, i.e., (Ba0.6Sn0.4)HfO , prepared by reacting 80 mol% SnClF
3
with BaHfO under standard synthetic conditions listed previously. The
3
3
sample was loaded into an evacuated, sealed, fused-silica ampule and
ꢁ1
ꢀ
ꢀ
annealed in a box furnace at temperatures at 200 C for 1 h with a 10 C
ꢁ1
3
min ramp rate and allowed to cool radiatively. The product was ground
to homogenize before measurement of powder XRD at room tempera-
ture. The process is repeated in 50 C steps as powder XRD data were
2 3
pseudocubic perovskite [24,25]. Additionally, the mixed-valent Sn O
II IV
ꢀ
perovskite, i.e., Sn Sn O
3
, has been calculated to be metastable with
ꢀ
ꢀ
respect to layered structures and to decomposition-driven phase segre-
gation to SnO and SnO [26,27]. Currently, there are no prior reported
syntheses of either of these perovskites. The products of the
Sn(II)-exchange reactions, i.e., (Ba1-xSn )HfO and (Ba1-xSn )SnO , are
collected from 200 C to 500 C.
2
2.3. Electron microscopy and energy dispersive spectroscopy
x
3
x
3
characterized by powder XRD, scanning transmission electron micro-
scopy (S/TEM), and investigated for their synthesizability with
increasing Sn(II), and including their thermal stability and decomposi-
tion pathways.
High resolution images and elemental analyses of BaHfO were per-
formed on a JEOL 6010LA scanning electron microscope (SEM) with an
accelerating voltage of 20 kV. A JEOL EDXS silicon drift detector was
used to determine elemental composition. High resolution images and
3
x 3
elemental analyses of (Ba1-xSn )HfO materials were taken on an FEI
2
. Experimental
Verios 460 L field emission scanning electron microscope (FESEM) and
elemental analysis was measured with an attached Oxford energy
dispersive X-ray spectrometer (EDS). Nanoscale resolution images were
acquired using a ThermoFisher Talos F200X with 200 kV accelerating
voltage. EDS spectra were collected using a 200 pA beam with a Super-X
EDS detector and reported as net atomic intensities.
2
.1. Synthetic procedure
3
The barium hafnate precursor, BaHfO , was synthesized using high-
temperature ceramic methods. First, hafnium (IV) oxide nanoparticles
were synthesized by mechanical mixing of HfCl (Acros Organics, 99%)
with excess potassium hydroxide (VWR, 99.5%) under argon and
washing with deionized water. The as-prepared HfO nanoparticles were
intimately mixed with a mortar and pestle with a 10% molar excess of
BaCO (Alfa Aesar, 99.8%), to account for volatilization, and pressed into
a pellet. The reaction mixture was added to an alumina crucible and
4
2.4. Electronic structure and total energy calculations
2
Density functional theory calculations of (Ba1-xSn
x
)MO
3
(M ¼ Sn or
3
Hf) with x ¼ 0.0, 0.25, 0.50, 0.75, and 1.0 were performed starting from
3 ꢂ 3 ꢂ 3 supercells of the cubic perovskite structure using the Vienna Ab
2

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
initio Simulation Package (VASP; ver 4.6) using plane-wave density func-
tional theory [29,30]. For compositions with Ba/Sn statistically mixed on
the A-site, random positions within the supercell were chosen as Ba or Sn
to achieve the desired stoichiometry. Total energy and densities-of-states
calculations used the Perdew-Burke-Ernzerhof functionals in the gener-
alized gradient approximation with the projector augmented wave
method. A 3 ꢂ 3 ꢂ 3 Monkhorst-Pack grid was used to sample the
Brillouin-zone automatically [31].
4.171428 (4) Å and wRp ¼ 7.99%, consistent with prior studies [35,36].
The barium stannate precursor was prepared using a NaCl/KCl eutectic
3
synthesis route and yielded the polycrystalline BaSnO cubic perovskite
phase in high purity as well, with a refined lattice parameter of 4.11755
(3) Å and wRp ¼ 8.24%, consistent with prior reports (JCPDS 150780)
3 3
[37]. Both BaHfO and BaSnO perovskites were reacted separately in the
SnClF flux, aiming at the Sn(II)-richest perovskites having the respective
compositions (Ba1-xSn )HfO and (Ba1-xSn )SnO (vide infra), as repre-
sented in Fig. 1 and in Reaction 4 below:
x
3
x
3
Calculations of the formation energies of the Sn(II)-containing pe-
rovskites followed the same protocol as reported previously by Hautier
et al. and others [8,32,33]. These prior studies have demonstrated a
relatively reliable accuracy in the calculation of formation energies (at 0
BaMO
3
þ x SnClF → ðBa1ꢁxSn
x
ÞMO
3
þ x BaClF ðM ¼ Hf or SnÞ
(4)
ꢁ1
The calculated energy change of the reactions are ꢁ270.8 kJ mol
ꢁ1
K) with a standard deviation of ~24 meV atom , which is much smaller
than for the experimental errors of the measured reaction energies. Using
this method, formation enthalpies of the metal-oxide perovskites were
determined by total energy calculations using VASP at 0 K, in which the
ꢁ1
(
ꢁ2.81 eV/F.U.) for BaHfO
3
→ SnHfO
3
and -285.6 kJ mol (ꢁ2.96 eV/
3 3
F.U.) for BaSnO → SnSnO , demonstrating that both reactions are
thermodynamically favorable. The Sn(II)-halide flux melts beginning at
ꢀ
~
200 C facilitating its reactivity at low temperatures, coupled with the
entropic term is negligible and thus ΔG
enthalpies were obtained from the Open Quantum Materials Database
f
ꢃ ΔH
f
. Binary oxide formation
thermodynamic driving force of the formation of BaClF versus SnClF
ꢁ1
(
ꢁ1028 kJ mol ), as reported previously [8]. Both the SnClF reactant
(
(
OQMD) [18,19]. An example calculation is shown below in Equations
1) and (2).
and BaClF side product are water soluble and can be removed from the
product by washing. A comparison of selected XRD patterns of
x 3
Sn(II)-substituted perovskites is shown in Fig. 2, with the (Ba1-xSn )HfO
BaHfO
3
ðsÞ þ SnClFðsÞ → SnHfO
3
ðsÞ þ BaClFðsÞ
(1)
(2)
series shown on the left and (Ba1-xSn
x
)SnO
3
series shown on the right,
ꢀ
À
Á
À
Á
where x is the mole fraction of reacted SnClF at 250 C for 72 h. As shown
in Fig. 2, the Pm3m perovskite reflections remain nearly constant in each
as the mole fraction of Sn(II) increases. Impurity peaks of tin oxides start
ΔHrxn ¼ ΔHfðSnHfO3Þ þ ΔHfðBaClFÞ ꢁ ΔHfðBaHfO3Þ þ ΔHfðSnClFÞ
Decomposition energies for the solid solution perovskites, i.e., Ba1-
to be detected as low as 10% reacted SnClF with (Ba1-xSn
sponding to immediate phase decomposition. In contrast, decomposition
was not observed in (Ba1-xSn )HfO up to 100% SnClF concentration, but
x 3
)SnO , corre-
x
x 3
Sn HfO , were similarly calculated. Decomposition reactions were
balanced according to the experimental product distributions. For
example, the Sn(II)-containing hafnate perovskite decomposes to SnO,
x
3
only at an excess of 130% reacted SnClF. These XRD results are consistent
with a much higher kinetic stability for the Sn(II)-containing hafnate
perovskite.
HfO
2 3
, and BaHfO , as observed experimentally. An example reaction
equation is shown below in Equation (3).
2
ðBa0:5Sn0:5ÞHfO
3
→ SnO þ HfO
2
þ BaHfO
3
(3)
In this example, the complete decomposition and segregation of SnO
results in the formation of the thermally stable BaHfO . Further heating
3.2. Electron microscopy
3
of SnO, however, results in its disproportionation into Sn metal and SnO
under vacuum at temperatures >300 C. In these cases, the presence of
SnO was used as an indicator of SnO evolution when it was not observed
2
at intermediate temperatures by XRD. All structural models and super-
cells used in this work were generated using the Visualization for Elec-
tronic and STructural Analysis (VESTA) software [34].
2
Electron microscopy techniques were employed to assess the Sn(II)-
content and compositional homogeneity of the (Ba1-xSn )HfO perov-
x 3
skites. SEM and FESEM images of the particles' sizes and morphologies
ꢀ
for x ¼ 0, 0.7 and 1.0 are provided in Fig. S3. These data reveal that the
3
solid-state synthesis of BaHfO produces crystallites with dimensions
ranging from ~1 to 5 μm and relatively smooth surfaces. After reaction
with SnClF, i.e., for x ¼ 0.7 and 1.0, the particles become much smaller
and more nanocrystalline. Some larger particles have formed by
agglomeration, but these also occur together with substantially more
nanoscale particles.
3
. Results and discussion
3
.1. Synthesis and bulk characterization
Energy-dispersive spectroscopy (EDS) on the particles showed a sig-
nificant dependence on the Sn/Ba homogeneity with the chosen reaction
temperature and time. The EDS data for the first set of reaction condi-
The barium hafnate perovskite was prepared by a high temperature
ceramic route, resulting in a nominally pure and polycrystalline cubic
phase with the stoichiometry BaHfO (JCPDS 24–0102). Rietveld
refinement in the Pm3m space group yielded a lattice parameter of
3
ꢀ
tions, i.e., 250 C for 72 h, are shown for x ¼ 0.7 (nominally 70 mol%
Sn(II)) in Fig. 3. The chemical maps reveal a relatively inhomogeneous Ba
3 x 3
Fig. 1. Schematic representation of low temperature flux exchange reaction of BaHfO (left) and relative energy of (Ba1-xSn )HfO products. The barium perovskite
reacts with SnClF to replace Ba(II) and give the (Ba1-xSn
x
)HfO
3
product. As more Sn(II)-cations exchange, the perovskite becomes increasingly metastable.
3

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
Fig. 2. XRD patterns of the (Ba1-xSn
x
)HfO
3
(left) and (Ba1-xSn
x
)SnO
3
(right) perovskite powders, where x indicates the nominal mole fraction of Sn(II). Secondary
2
phases of SnO an SnO are labeled with black circles and squares, respectively.
ꢀ
and Sn distribution, with the separation of Ba-rich and Sn-rich regions
being observed as a result of inhomogeneous diffusion of Sn(II). In-
homogeneity is apparent when comparing the compositions of primary
particles within the larger aggregates. The Sn(II) diffusion typically ter-
minates to within ~100 nm of the surfaces of the larger particles. A
homogenous distribution of the Ba, Hf, Sn, and O elements were obtained
only with the smallest observed particles, i.e., <100 nm in diameter, with
no detectable regions of elemental segregation, shown in Fig. S4. By
contrast, the distribution of Hf and O (not shown) is relatively homoge-
neous across all particle sizes and shapes, supporting the persistence of
the underlying hafnate perovskite structure and the absence of phase
segregation. The EDS results are consistent with the Sn(II)-concentration
being limited by its ion diffusion through the perovskite structure at 250
C.
To test the impact of temperature on the Sn(II) diffusion, the flux-
based reactions were repeated at a higher temperature but shorter time
of 350 C for 12 h, similar to the conditions reported for the analogous
BSZT perovskites. EDS maps of (Ba1-xSn
shown in Fig. 4. At the higher temperature, the Sn(II)-exchange reactions
result in a significantly more homogeneous Ba/Sn distribution for even
the largest particle sizes. While the higher temperature increases the
Sn(II) diffusion, the increased thermal energy also causes significantly
more decomposition. Shown in Fig. 4e (inset), the product formed at the
ꢀ
x
)HfO
3
products for x ¼ 0.7 are
ꢀ
higher reaction temperature of 350 C has turned a dark black color,
whereas the sample prepared at the lower 250 C is yellowish. At this
nominal 70 mol% Sn(II) concentration, the lower temperature reaction
ꢀ
Fig. 3. High-angular dark field images (A–C) and elemental mapping (D–F) of the Ba and Sn contributions, red and yellow colored respectively, within three different
ꢀ
particles after a reaction with a nominal 70% Sn(II) reacted at 250 C for 72 h. (For interpretation of the references to color in this figure legend, the reader is referred
to the Web version of this article.)
4

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
produces no detectable impurities while the high temperature synthesis
yields ~10% of dark black SnO impurity, (Fig. 4e insets). The higher
temperature reaction provides a more homogeneous Sn(II) distribution
in a shorter time, although coming at the expense of increased
decomposition.
of high amounts of impurity phases. Thus, the collective XRD and EDS
results suggest that the synthetic limit of Sn(II) exchange into the
perovskite for these reaction conditions is a conservative ~70%, or
(Ba0.3Sn0.7)HfO
upon further exchange, and is the closest attainable to the highly meta-
stable SnHfO perovskite that can be synthesized under these conditions.
3
, with growing amounts of decomposition occurring
A third set of reaction conditions was tested which involved a second
3
cycle of mechanical homogenization and heating the reaction mixture at
ꢀ
2
50 C for 72 h. The additional reaction cycle was implemented to
3.3. Structural characterization
facilitate the Sn(II) diffusion but at the possible expense of further
product decomposition. EDS data are shown in Fig. 5a for the nominally
Rietveld refinements were performed on powder XRD patterns
measured on the PANalytical Empyrean diffractometer for the Sn(II)-
containing hafnate perovskite compositions with mole fractions of
Sn(II) at x ¼ 1/3, 1/2, and 2/3 as determined by EDS analyses. The
resulting fits to the data for the (Ba2 Sn )HfO and (Ba Sn )HfO
x ¼ 0.7 composition, i.e., Ba0.3Sn0.7HfO
agglomerate with ~10–20 m diameter. Analogous EDS data for the
nominally x ¼ 1.0 product, i.e., SnHfO , are provided in Fig. 5b. A ho-
3
, for a relatively large particle
μ
3
mogeneous distribution of Sn(II) is observed throughout both particles as
shown in the elemental mapping. Fig. 5 shows representative particles
with a high sample purity and homogeneous Sn(II) distribution that can
be achieved at the lower temperature with one additional grinding and
heating step.
/3
1/3
3
1/3 2/3
3
phases are displayed in Fig. 6. The figure shows that the main Bragg
peaks of the cubic perovskite structure are maintained with Sn(II) sub-
stitution. Rietveld refinements were performed in the Pm3m space group
and a summary of selected refined parameters is provided in Table 2.
There is a small decrease in lattice constant with increasing Sn(II)-
substitution and the refinements are in good agreement with cubic
symmetry. The refined atomic A-site occupancy fractions for Sn(II) in-
crease with increasing nominal Sn(II) substitution and are comparable to
the values determined by EDS. To obtain as stable a refinement as
possible, the Ba/Sn fractional occupancies were constrained to sum to
one and were refined independent from the atomic displacement
parameter (Uiso value). In addition to the primary perovskite phase, some
background peaks were observed in the pattern, but were less than ~1%
relative intensity and thus difficult to use for phase identification.
The chemical compositions obtained from EDS analysis are listed in
ꢀ
Table 1 for the lower temperature products at 250 C for 72 h after
3
heating one or two cycles. In each case, the EDS spectrum of BaHfO was
used as an internal standard. The data in Table 1 show a trend with
increasing amounts of Sn(II) in the perovskite with increased nominal
amounts in the reaction and with the number of heating cycles. After a
ꢀ
single reaction cycle with 70 mol% Sn(II) at 250 C, the amount of Sn(II)
exchanged, as evidenced by EDS, is about half of that, ranging from ~30
to 50%, or ~38% on average. At the ~(Ba2/3Sn1/3)HfO composition
3
there are no detectable impurity phases in the XRD patterns, consistent
with a perovskite phase with Ba(II) and Sn(II) cations co-occupying the
3
A-site at about the (Ba2/3Sn1/3)HfO stoichiometry. After two reaction
Synchrotron powder XRD patterns were measured on the (Ba2/3Sn1/3
)
HfO phase to further investigate the main phase and background peaks
3
cycles are performed with a nominal 70 mol% Sn(II), with secondary
grinding, the amount of Sn in the product increases to ~49% with no
with higher resolution, shown in Fig. S9. Between 5 and 15⁰ 2θ, several
small peaks are observed with less than 0.5% relative intensity. Most of
these peaks were identified as trace amounts of impurities of SnO, BaClF,
and BaCl (H O) and their refined phase fractions summed to approxi-
2 2 2
mately 4% of the total pattern. A Rietveld refinement was performed
with a primary cubic perovskite phase combined with the identified
change in the XRD pattern. When the BaHfO
with a nominal 100 mol% Sn, the amount of Sn(II)-cations exchanged by
EDS increases to a relatively larger amount of ~70%. The (Ba0.3Sn0.7
HfO phase is the highest amount of Sn(II) attained in the perovskite
3
precursor is reacted twice
)
3
structure without detection of significant impurity phases in the XRD
impurity phases, resulting in a weighted residual of 8.73% and a reduced
ꢀ
patterns. For the higher temperature reactions at 350 C, the mole frac-
2
χ
of 2.79. The primary peaks agree with the cubic Pm3m space group
tions of Ba/Sn could not be reliably determined because of the formation
with a lattice parameter of 4.173399 (3) Å. A high-resolution pattern of
Fig. 4. High-angular dark field images (A,B) and elemental mapping (C,D) of the Ba and Sn contributions, red and yellow colored respectively, within two different
ꢀ
ꢀ
reactions with a nominal 70% Sn(II) reacted once at 350 C for 12 h. Shown in (E) is the comparative powder XRD of products from reactions run at 250 C for 72 h
ꢀ
and 350 C for 12 h, with 70% Sn(II), with the powder colors shown in the insets. (For interpretation of the references to color in this figure legend, the reader is
referred to the Web version of this article.)
5

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
ꢀ
Fig. 5. FESEM/EDS images of representative particles of reactions with a nominal 70% Sn(II) (left) and 100% Sn(II) (right) after two repeated flux reactions at 250 C
for 72 h each. Elemental maps are shown on the right, with Ba denoted as purple, Sn as light blue, Hf as yellow, and O as green. (For interpretation of the references to
color in this figure legend, the reader is referred to the Web version of this article.)
Table 2
Table 1
Summary of refined parameters from laboratory powder XRD measurements for
Sn(II)-hafnate perovskites in the Pm3m space group.
Tabulated results of EDS data for (Ba1-xSn
Sn(II) amount, (x), after heating to 250 C for 72 h for either one or two heating
cycles.
x
)HfO
3
as a function of the nominal
ꢀ
Composition
Ba2/3Sn1/3HfO
3
Ba1/2Sn1/2HfO
3
Ba1/3Sn2/3HfO
3
Element
x
nom ¼ 0.7 (1
x
nom ¼ 0.7 (2
x
nom ¼ 1.0 (2
Ba Fraction
Sn Fraction
a (Å)
V (Å )
wRp (%)
0.71 (2)
0.29 (2)
4.17213 (3)
72.623 (2)
4.80
0.63 (3)
0.37 (3)
4.17206 (6)
72.619 (3)
3.55
0.40 (4)
0.60 (4)
4.17168 (6)
72.600 (3)
3.80
cycle)
cycles)
cycles)
Ba mol %
Sn mol %
Sn:Ba ratio
33.40
20.12
0.602
0.38
25.97
24.98
0.962
0.49
16.49
39.04
2.368
0.70
3
Rp (%)
3.55
6.27
2.65
3.75
2.83
4.10
x
exp (Ba1-xSn
HfO
x
)
Reduced χ2
3
Ba0.62Sn0.38
O
3
Ba0.51Sn0.49
O
3
Ba0.3Sn0.7O
3
trapped at a higher energy than the ground state can be experimentally
probed under elevated temperatures that provide sufficient thermal
diffusion for the ions to diffuse and reach the ground state, i.e., to
decompose to the simpler oxides. Thus, for the current perovskite
products discussed above, the decomposition reaction most typically
the precursor, BaHfO
parameter of 4.173434 (2) Å. While the precursor BaHfO
larger lattice constant after ~1/3 Sn(II)-substitution, this difference is
likely not statistically significant owing to errors inherent in the instru-
mental parameters.
3
, was also measured and refined to a lattice
has a slightly
3
occurs as AMO
3
2
→ AO þ MO . The degree of metastability and the
decomposition pathway has been assessed at increasing temperatures by
heating in a furnace with ex-situ XRD as well as by heating under an
electron beam using in situ S/TEM methods. Decomposition of the stan-
nate perovskites were observed during the Sn(II)-exchange reactions and
described earlier.
3
.4. Metastability and thermal decomposition pathway
x 3
The extent of kinetic stabilization of the (Ba1-xSn )HfO perovskites
was investigated using both ex situ and in situ characterization methods
during heating. As metastable phases, evidence of being kinetically
Shown in Fig. 7 is ex situ powder XRD data in which the (Ba0.6Sn0.4)
Fig. 6. Rietveld refinements of (Ba2/3Sn1/3)HfO
space group.
3
powder XRD data (left) and (Ba1/3Sn2/3)HfO
6
3
(right) according to the cubic perovskite structure in the Pm3m

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
ꢀ
ꢀ
HfO
3
perovskite with 40% Sn(II) cations was heated from 50 C to 500 C
disproportionation of SnO. The ex situ and in situ experiments have
demonstrated, both in the average structure measured from diffraction
and down to a fine-level surface detail in microscopy, a decomposition
pathway of the metastable perovskite that is driven by phase segregation
and is consistent with the following reactions:
ꢀ
under vacuum in 50 C increments with intermediate grinding and XRD
measurements. Powder XRD data were linearly interpolated between
each set. As shown by the primary reflection at ~26⁰, SnO
2
(cassiterite
ꢀ
type) starts forming in growing amounts by 350 C. This diffraction peak
increases in intensity with increasing temperature, along with two of its
next most intense peaks at ~32-34 appearing by ~375 C. Under vac-
ꢀ
ꢀ
Ba1ꢁxSn
x
HfO
3
→ x SnO þ x HfO
2
þ ð1 ꢁ xÞ BaHfO
3
(5)
uum, the formation of SnO
2
is from the concurrent disproportionation of
ꢀ
xSnO → x=2 SnO þ x=2 Sn
(6)
SnO which begins rapidly by ~300 C, in agreement with previous re-
ports [38,39]. The sample also begins to change significantly in color,
starting as a light-yellow powder at room temperature and turning
entirely black colored by 400 C. The color change is the result of the
amorphous Sn metal that also forms during the disproportionation re-
2
The decomposition results are also consistent with the synthetic re-
ꢀ
action conditions. A low reaction temperature of ~250 C is critical to
ꢀ
trap the metastable hafnate perovskites, as slightly higher temperatures
ꢀ
of ~350 C are found to result in significant decomposition. The kinetic
ꢀ
action of SnO. By ~400 C, the XRD patterns also show that HfO
2
crys-
stabilization of the metastable perovskites is thus dependent upon
minimizing ion diffusion within the crystallites, and specifically, the
diffusion of Sn(II) cations. In contrast, the analogous stannate perov-
ꢀ
tallizes out beginning at ~400 C, as evidenced by the intense diffraction
peaks at ~28.1⁰ and ~31.6⁰. Impurity peaks corresponding to residual
SnClF and BaClF also appear in the data, likely because of being adsorbed
to particle surfaces, and which repeated aqueous washing did not effec-
tively remove. Fig. 7 shows that, in the 400–425 C range, the
Sn(II)-exchanged barium hafnate begins decomposing into HfO
skites, i.e., (Ba1-xSn
ditions. The immediate decomposition is
x
)SnO
3
, decomposed under the same synthetic con-
consequence of the
a
ꢀ
underlying perovskite framework being based on Sn(IV) cations, and
2
and
when Ba(II) is exchanged with the Sn(II) cations, there is much less ion
SnO, with the perovskite reflections still present. As the BaHfO perov-
3
ꢀ
II
diffusion required for phase segregation to SnO
local ordering).
2
and SnO (e.g. O–Sn –O
skite is thermodynamically stable and high melting (m.p. ~2620 C), the
decomposition pathway is driven by the removal of Sn(II) cations from
the structure.
Decomposition of the most Sn(II)-rich phase, (Ba1/3Sn2/3)HfO
3
, was
3.5. Calculations of metastability
probed using S/TEM methods to observe fine-level changes at the sur-
faces of individual nano-crystallites that cannot be detected in XRD.
Shown in Fig. 8 are the S/TEM and EDS images of a representative par-
ticle after partial electron-beam induced decomposition at its surfaces.
After its partial decomposition, four separate regions are apparent as a
result of phase segregation, labeled A to D. In region A, the EDS data yield
Using methods established previously by Hautier and others, total
energy calculations were performed using density functional theory in
order to assess the extent of metastability of the (Ba1-xSn )SnO and (Ba
x
3
1-
x
Sn )HfO perovskites with increasing Sn(II) content [8,32,33]. Results
x
3
of the calculations show that as Sn(II) is increasingly incorporated into
2
an approximate composition of (Ba1/2Sn1/2)HfO
3
, which is slightly
the A-site the total energy decreases nearly linearly (R ¼ 0.995) with the
ꢁ
1
Sn(II)-deficient as compared to the starting composition. Extending
outward, the elemental mapping (colored) in regions B and C show ev-
Sn(II) concentration, from ꢁ8.722 eV atom in BaHfO (0% Sn(II)) to
3
ꢁ1
ꢁ8.205 eV atom in SnHfO (100% Sn(II)). The calculated total energy
3
idence for phase segregation, respectively, into BaHfO
mixture of SnO and HfO . Only Sn and O are observed in the outermost
region D, corresponding to either a mixed-valent tin oxide phase or to a
mixture of SnO and elemental Sn. The latter interpretation is consistent
3
and then a
for BaHfO3 agrees with prior published data (online) in the Open
ꢁ1
2
Quantum Materials Database of ꢁ8.715 eV atom
[18,19]. While
BaHfO is shown to be thermodynamically stable against decomposition,
3
2
the thermal decomposition corresponding to Reaction 5 (given above)
with the powder XRD results described above, as owing to the
becomes increasingly favorable with increasing Sn(II) amounts for
(
x 3
Ba1-xSn )HfO beyond ~10 mol% Sn. For example, for only ~25% Sn (x
¼
0.25), the perovskite composition is already highly metastable by
ꢁ1
about ꢁ156 meV atom towards decomposition by phase segregation,
e.g., 4 (Ba3/4Sn1/4)HfO → 3 BaHfO . For even higher
þ SnO þ HfO
Sn(II) amounts, the decomposition energies increase to ꢁ312 meV
3
3
2
ꢁ1
ꢁ1
ꢁ1
atom , -479 meV atom and -647 meV atom , for x ¼ 0.5, 0.75, and
.0, respectively. The trend in decomposition energies is consistent with
1
the experimentally observed decomposition reactions. It is generally re-
ported that metastable compounds can potentially be synthesized that
ꢁ1
fall within ~100–200 meV atom of the ground state [22,23]. The
current results show that this range is greatly exceeded for the highest
3
synthesized Sn(II)-containing perovskite, Ba0.3Sn0.7HfO with a calcu-
ꢁ1
lated metastability of about ꢁ446 meV atom . The extremely high
ꢀ
melting point of the reactant perovskite, BaHfO
3
(m.p. ~2620 C), sug-
gests that ion diffusion and therefore phase segregation is highly
inhibited at low temperatures, as discussed below.
The stannate perovskites are similarly found to become increasingly
thermodynamically unstable towards decomposition by phase segrega-
tion. The B-site cation is fixed as Sn(IV) and the flux reaction causes
substitution of the Sn(II) cations onto the A-site. The total energy of (Ba1-
x
Sn
x
)SnO
3
increases linearly with the Sn(II) amount on the A-site from
ꢁ
1
ꢁ1
ꢁ
6.306 eV atom for 0% Sn(II) in BaSnO
00% Sn(II) in Sn Sn O in the cubic perovskite structure. The
3
3
to ꢁ5.819 eV atom for
ꢀ
Fig. 7. Ex situ XRD of Ba0.6Sn0.4HfO
3
at temperatures between 25–500 C.
II IV
1
Diffraction data shown for the 20–44⁰ 2θ range. Perovskite reflections from
Ba0.6Sn0.4)HfO /BaHfO are labeled by stars with hkl indices, HfO by squares,
BaClF by circles, SnO by triangles, and SnClF by diamonds. BaClF and SnClF are
left over from the formation reaction and were not sufficiently removed
via washing.
decomposition pathways were modeled analogous to the hafnate pe-
rovskites, i.e., as (Ba1-xSn )SnO → (1-x) BaSnO . The
results show that the Sn(II)-substituted perovskites again become sub-
stantially more metastable as the amount of Sn(II) increases with respect
(
3
3
2
x
3
3
þ x SnO þ x SnO
2
2
7

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
Fig. 8. S/TEM image of perovskite nano-crystallite after electron-beam induced phase segregation. EDS data show that the composition within the (marked) region A
is the (Ba,Sn)HfO perovskite, region B is BaHfO , region C is HfO and SnO, and region D is Sn and SnO
3
3
2
2
.
to decomposition by phase segregation. The decomposition energy for
the Sn(II)-exchange limit increases proportionally from <10% to ~70%,
more closely following the trend with increasing lattice cohesive energy
than do the melting points. The two perovskites with the smallest
cohesive energies exhibit very low synthesizability of their metastable
Sn(II)-containing analogues under these conditions. Notably, both can
also decompose by routes that involve minimal ion diffusion, such as
ꢁ
1
BaSnO
modynamically stable. For increasing amounts of Sn(II) of x ¼ 0.25, 0.5,
.75, and 1.0, the decompositions energies were found to be ꢁ122 meV
3
is calculated to be þ183 meV atom , showing that it is ther-
0
ꢁ
1
ꢁ1
ꢁ1
ꢁ1
atom , -244 meV atom , -381 meV atom , and -519 meV atom
.
Thus, the barium stannate perovskites are all less metastable than the
corresponding hafnate perovskites, despite not being as nearly synthe-
sizable. Reactions intended at each of these compositions decompose
nearly entirely to the phase-segregated products, clearly suggesting sig-
nificant differences in the kinetic barriers of their decomposition
pathways.
3
with the occurrence of ilmenite-type SnTiO that competes for formation
in the (Ba1-xSn )TiO perovskite system. Thus, favorable conditions to
x
3
prepare the most metastable Sn(II)-containing perovskites include both
high lattice cohesive energies as well as the existence of lower-energy
competing phases that require maximal ion diffusion to reach.
Shown in Fig. 9 are schematic diagrams of the energetics of the
decomposition pathways of the (Ba1-xSn
x
)HfO
3
and (Ba1-xSn
x
)SnO
3
as a
4. Conclusions
function of the amount of Sn(II), as well as the empirically estimated
relative activation barriers based upon the observed decomposition of
Low temperature flux-mediated reactions of SnClF show that the ex-
BaHfO
each. The (Ba1-xSn
vation barrier against decomposition as compared to (Ba1-xSn
reflecting the greater kinetic stabilization and synthesizability. One fac-
tor leading to the seemingly absent activation barrier of the (Ba1-xSn
SnO perovskite is the greater overall concentration of Sn, i.e., containing
3
and BaSnO
3
at the synthetic limits of the Sn(II) concentrations in
)HfO is shown to have a significantly higher acti-
)SnO , as
change of Sn(II) for Ba(II) cations in BaHfO and BaSnO can yield sig-
3
3
x
3
nificant Sn(II)-cation incorporation only in the former, yielding (Ba1-
Sn )HfO at up to x ~ 0.7. Electron microscopy imaging shows that the
x
3
x
x
3
highest concentration of Sn(II) cations of ~70% with a homogeneous
distribution can only be achieved at low reaction temperatures of ~250
x
)
ꢀ
3
C with two consecutive reactions and intermittent grinding. Analogous
both Sn(IV) and Sn(II) cations, and that reduces the diffusion lengths
needed for phase segregation to occur. However, another prominent
cation-exchange reactions for the stannate perovskite only resulted in its
decomposition. The metastable (Ba1-xSn )HfO3 compositions exhibit a
x
factor is the significantly higher cohesive energy of the reactant BaHfO
versus BaSnO perovskite. A recent study of the analogous BSZT perov-
3
decomposition pathway that proceeds at its surfaces via phase segrega-
tion into the simpler oxides, beginning at only ~350 C. Total energy
ꢀ
3
skite system showed a higher Sn(II)-exchange limit for the Zr-richer
compositions, which was hypothesized to be the result of a trend to-
ward a greater lattice cohesive energy. The calculated difference in lat-
calculations demonstrate the highest metastability of the ~70% Sn(II)
ꢁ1
perovskite, (Ba0.3Sn0.7)HfO , reaches a remarkable ~ -446 meV atom
3
against decomposition, which is significantly higher than previously
considered for being synthesizable. The additional kinetic stabilization of
the hafnate perovskites is posited to result from its much higher high
cohesive energy, melting point, and the absence of lower-energy poly-
morphs or a ground state that could potentially be reached without
requiring significant ion diffusion. The presented results have helped to
elucidate the relationship between the synthesizability and metastability
of Sn(II)-containing perovskites that have been long desired as a new
class of technologically relevant materials. Discovery of the key factors
leading to the successful synthesis of Sn(II)-containing metastable pe-
rovskites will hopefully lead to the synthesis of the first high-purity
ꢁ1
ꢁ1
tice energies were quite small, i.e., ꢁ39.6 eV atom and -40.9 eV atom
for BaTiO and BaZrO , respectively. By comparison, BaHfO and BaSnO
3
3
3
3
have much larger and smaller cohesive energies, respectively. These
energies are listed in Table 3 along with the observed limit of Sn(II)
concentration for each perovskite system.
Prior theoretical studies have found the cohesive energy is an
important variable to calculate the average amplitude of thermal vibra-
tions within the Lindemann criterion [40,41], which posits that melting
occurs, and hence significant ion diffusion, when the lattice vibrations
exceed a threshold value that generally increases with the lattice cohe-
sive energy. As listed in Table 3, as the lattice cohesive energies increase
Sn(II)-only SnMO perovskites in the future. For example, reactions to
3
from the low extreme of BaSnO
eV atom . These generally correspond to relatively lower and higher
melting points with some deviations from the expected trend. However,
3
to the high extreme of BaHfO
3
of -43.6
target a more complete Sn(II)-cation exchange are currently under
investigation using synthetic approaches which can increase both the
extent and rates of cation diffusion at lower temperatures. These
ꢁ1
8

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
x 3 x 3
Fig. 9. A schematic of the calculated energy above ground state (i.e., metastability) for (left) (Ba1-xSn )HfO and (right) (Ba1-xSn )SnO . The energy of decomposition
increases with %Sn(II), becoming metastable at x ꢃ 0.15 in the stannates and x ꢃ 0.1 in the hafnates. The relative Ea,f was estimated based upon the concentration of
Sn(II) that exchanges before the onset of decomposition.
Acknowledgements
Table 3
Listing of perovskite cohesive energies and melting points compared to the
The authors acknowledge support of this work by the National Sci-
synthesizable Sn(II)-containing perovskites and the extent of their metastability
ence Foundation (DMR-2004455). A portion of this research used re-
compared to the ground state.
sources of the Advanced Photon Source, a U.S. Department of Energy
Perovskite
Composition
Cohesive
Melting
Point (K)
Maximum Sn(II)
Concentration(mol
%)
ΔEDecomp at
Max
(
DOE) Office of Science User Facility operated for the DOE Office of
Energy (eV
ꢁ
1
Science by Argonne National Laboratory under Contract DE-AC02-
06CH11357. Extraordinary facility operations were supported in part by
the DOE Office of Science through the National Virtual Biotechnology
Laboratory, a consortium of DOE national laboratories focused on the
response to COVID-19, with funding provided by the Coronavirus CARES
Act. Additional components of this research were performed in part at the
Analytical Instrumentation Facility (AIF) at North Carolina State Uni-
versity, which is supported by the State of North Carolina and the Na-
tional Science Foundation (Award ECCS-2025064). The AIF is a member
of the North Carolina Research Triangle Nanotechnology Network
atom
)
Sn(II)(meV
ꢁ
1
atom
)
BaSnO
3
ꢁ31.5
ꢁ39.6
ꢁ40.9
ꢁ43.6
2128
1898
2923
2893
<10
10
50
ꢄ0
ꢄ0
BaTiO
BaZrO
BaHfO
3
3
þ225
þ446
3
70
strategies include increasing the concentration of A-site and/or B-site
cation vacancies as well as decreasing the particle sizes.
(
RTNN), a site in the National Nanotechnology Coordinated Infrastruc-
Supporting information
ture (NNCI).
Additional experimental data are provided including powder XRD
Appendix A. Supplementary data
data of the BaHfO
reaction Sn(II)-substituted BaHfO
two reaction cycles, and results of the multi-phase Rietveld refinement of
synchrotron XRD results. Additional S/TEM images of (Ba1-xSn )HfO are
3
and BaSnO
3
perovskite precursors, before and after
3
, the (Ba1-xSn )HfO products after
x
3
Supplementary data to this article can be found online at https://do
i.org/10.1016/j.jssc.2021.122419.
x
3
provided, including Hf and O elemental mapping, EDS of electron-beam
induced phase segregation, and selected area electron diffraction (SAED)
images of the perovskite and secondary phase.
References
[
1] B.L. Musico, D. Gilbert, T.Z. Ward, K. Page, E. George, J. Yan, D. Mandrus,
V. Keppens, The emergent field of high entropy oxides: design, prospects,
challenges, and opportunities for tailoring material properties, Apl. Mater. 8 (1–8)
CRediT authorship contribution statement
(2020), 040912.
[
[
2] R. Cava, Oxide superconductors, J. Am. Ceram. Soc. 83 (2000) 5–28.
3] K.N.D.K. Muhsen, R.A.M. Osman, M.S. Idris, The effects of Ca, Zr, and Sn
Eric A. Gabilondo: Investigation, sample synthesis, computation,
DFT, Writing – review & editing. Shaun O'Donnell: Conceptualization,
reviewing. Rachel Broughton: Structure determination, Conceptuali-
zation, reviewing. Jacob L. Jones: Conceptualization, Methodology,
reviewing. Paul A. Maggard: Conceptualization, Methodology, writing,
reviewing.
3 3 3
substitutions into a ternary system of BaTiO -BaSnO -BaZrO towards its dielectric
and piezoelectric properties: a review, J. Mater. Sci. Mater. Electron. (2020).
[4] B. Yan, M. Jansen, C. Felser, A large-energy-gap oxide topological insulator based
on the superconductor BaBiO
[5] N. McLamb, P.P. Sahoo, L. Fuoco, P.A. Maggard, Flux growth of single-crystal
Na Ta 11 particles and their photocatalytic hydrogen production, Cryst. Growth
3
, Nat. Phys. 9 (2013) 709–711.
2
4
O
Des. 13 (2013) 2322–2326.
[
6] S.M. Schmuecker, D. Clouser, T.J. Kraus, B.M. Leonard, Synthesis of metastable
chromium carbide nanomaterials and their electrocatalytic activity for the
hydrogen evolution reaction, Dalton Trans. 46 (2017) 13524–13530.
7] J. Boltersdorf, P.A. Maggard, Silver exchange of layered metal oxides and their
photocatalytic activities, ACS Catal. 3 (2013) 2547–2555.
Declaration of competing interest
[
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[8] S. O'Donnell, C. Chung, A. Carbone, R. Broughton, J.L. Jones, P.A. Maggard,
Pushing the limits of metastability in semiconducting perovskite oxides for visible-
light-driven water oxidation, Chem. Mater. 32 (7) (2020) 3054–3064.
[9] J. Gardner, A. Thakre, A. Kumar, J.F. Scott, Tin titanate-the hunt for a new
ferroelectric perovskite, Rep. Prog. Phys. 82 (9) (2019) 1–14, 092501.
9

E.A. Gabilondo et al.
Journal of Solid State Chemistry 302 (2021) 122419
[
[
10] E. Sawaguchi, Ferrolectricity versus antiferroelectricity in the solid solutions of
PbZrO and PbTiO , J. Phys. Soc. Jpn. 8 (5) (1953) 615–629.
[25] D. Chenine, Z. Aziz, Theoretical investigation of structural, elastic and electronic
3
3
3
properties of SnHfO compound: Ab initio calculations, in: Proceedings of the Third
International Symposium on Materials and Sustainable Development, 2018,
pp. 75–80.
11] M.A. Knyazeva, D.A. Andronikova, G.A. Lityagin, I.A. Bronwald, P. Paraskevas,
A. Majchrowski, K. Roleder, A.V. Filimonov, R.G. Burkovsky, Phase transitions in
lead hafnate under high pressure, Phys. Solid State 61 (10) (2019) 1806–1812.
12] A. Walsh, D.J. Payne, R.G. Egdell, G.W. Watson, Stereochemistry of post-transition
metal oxides: revision of the classic lone pair model, Chem. Soc. Rev. 40 (2011)
[26] J. Wang, Mixed valence tin oxides as Novel van der Waals materials: theoretical
predictions and potential applications, Adv. Energy Mater. 6 (1–11) (2016)
1501190.
[27] A. Seko, A. Togo, F. Oba, I. Tanaka, Structure and stability of a homologous series of
tin oxides, Phys. Rev. Lett. 100 (1–4) (2008), 045702.
[28] B.H. Toby, R.B. Von Dreele, GSAS-II: the genesis of a modern open-source all
purpose crystallography software package, J. Appl. Crystallogr. 46 (2) (2013)
544–549.
[29] G. Kreese, J. Hafner, Efficiency of Ab initio total energy calculations for metals and
semiconductors using a plane-wave basis set, Comput. Mater. Sci. 6 (1996) 6–15.
[30] G. Kreese, J. Furthmuller, Efficient iterative schemes for Ab initio total-energy
calculations using a plane-wave basis set, Phys. Rev. B 6 (1996) 11169–11186.
[31] J.P. Perdew, L. Burke, M. Ernzerhof, Generalized gradient approximation made
simple, Phys. Rev. Lett. 77 (1996) 3865–3868.
[
[
4
455–4463.
13] C.M. Campo, J.E. Rodriguez, A.E. Ramirez, Thermal behavior of romarchite phase
SnO in different atmospheres: a hypothesis about the phase transformation, Heliyon
2
(5) (2016) e00112.
[
[
[
14] F. Bortolani, R.A. Dorey, Molten salt synthesis of PZT powder for direct write inks,
J. Eur. Ceram. Soc. 30 (2010) 2073–2079.
15] S.S. Chandratreya, R.M. Fulrath, J.A. Pask, Reaction mechanisms in the formation
of PZT solid solutions, J. Am. Ceram. Soc. 64 (7) (1981) 422–425.
16] K.C. Pitike, W.D. Parker, L. Louis, S.M. Nakhmanson, First-principles studies of lone-
pair-induced distortions in epitaxial phases of perovskite SnTiO
Rev. B 91 (2015), 035112.
3 3
and PbTiO , Phys.
[
[
[
[
[
17] L. Diehl, S. Bette, F. Pielnhofer, S. Betzler, I. Moudrakovskit, G.A. Ozin,
R. Dinnebier, B.V. Lotsch, Structure-directing lone pairs: synthesis and structural
characterization of SnTiO , Chem. Mater. 30 (24) (2018) 8932–8938.
18] S. Kirklin, J.E. Saal, B. Meredig, A. Thompson, J.W. Doak, M. Aykol, S. Rühl,
C. Wolverton, The open Quantum materials Database (OQMD): assessing the
accuracy of DFT formation energies, Npj. Compu. Mater. 1 (September) (2015).
19] J.E. Saal, S. Kirklin, M. Aykol, B. Meredig, C. Wolverton, Materials design and
discovery with high-throughput density functional theory: the open Quantum
materials Database (OQMD), JOM (J. Occup. Med.) 65 (11) (2013) 1501–1509.
20] S. O'Donnell, A. Hamilton, P.A. Maggard, Fast flux reaction approach for the
[32] G. Hautier, S. Ping Ong, A. Jain, C.J. Moore, G. Ceder, Accuracy of density
functional theory in predicting formation energies of ternary oxides from binary
oxides and its implication on phase stability, Phys. Rev. B 85 (2012) 155208.
[33] A. Benisek, E. Dachs, The accuracy of standard enthalpies and entropies for phases
of petrological interest derived from density-functional calculations, Contrib.
Mineral. Petrol. 173 (90) (2018) 1–11.
[34] K. Momma, F. Izumi, VESTA 3 for three-dimensional visualization of crystal,
volumetric and morphology data, J. Appl. Crystallogr. 44 (2011) 1272–1276.
[35] A. Dias, V.S.T. Ciminelli, Electroceramic materials of tailored phase and
morphology by hydrothermal technology, Chem. Mater. 15 (2003) 1344–1352.
[36] S. Ullah, T. Iqbal, G. Rehman, I. Ahmad, Electronic and optical properties of group
IIA-IVB cubic perovskite oxides: improved TB-mBJ study, Chem. Phys. Lett. 757
(2020) 137887.
3
2 4
preparation of Sn TiO : tuning particle sizes and photocatalytic properties,
J. Electrochem. Soc. 166 (5) (2019) 3084–3090.
21] J. Boltersdorf, I. Sullivan, T.L. Shelton, Z. Wu, M. Gray, B. Zoellner, F.E. Osterloh,
P.A. Maggard, Flux synthesis, optical, and photocatalytic properties of n-type
[37] S. Upadhyay, O. Parkash, Preparation and Characterization of barium stannate
BaSnO3, J. Mater. Sci. Lett. 16 (1997) 1330–1332.
Sn
2 4
TiO : hydrogen and oxygen evolution under visible light, Chem. Mater. 28 (24)
(
2016) 8876–8889.
[38] H. Giefers, F. Porsch, G. Wortmann, Kinetics of the disproportionation of SnO, Solid
State Ionics 1–2 (176) (2005) 199–207.
[39] M.S. Moreno, G. Punte, G. Rigotti, R.C. Mercader, A.D. Weisz, M.A. Blesa, Kinetic
study of the disproportionation of tin monoxide, Solid State Ionics 1–2 (144) (2001)
81–86.
[
22] W. Sun, S.T. Dacek, S.P. Ong, G. Hautier, A. Jain, W.D. Richards, A.C. Gamst,
K.A. Persson, G. Ceder, The thermodynamic scale of inorganic crystalline
metastability, Sci. Adv. 2 (2016) 1600225.
23] M. Aykol, S.S. Dwaraknath, W. Sun, K.A. Persson, Thermodynamic limit for
synthesis of metastable inorganic materials, Sci. Adv. 4 (2018), 0148.
24] Q. Mahmood, B. Ul Haq, M. Yaseen, S.M. Ramay, M.G.B. Ashiq, A. Mahmood, The
[
[
[40] J. Tateno, An empirical relation on the melting temperature of some ionic crystals,
Solid State Commun. 10 (1972) 61–62.
first-principle study of mechanical, optical and thermoelectric properties of SnZrO
and SnHfO for renewable energy applications, Solid State Commun. 292 (2019)
7–23.
3
[41] S.-N. Luo, Vibrational density of states and Lindemann melting law, J. Chem. Phys.
122 (1–5) (2005) 194709.
3
1
1
0