Reactivity of solid BaCe0.9Y0.1O3-δ towards melted WO3

Article abstract of DOI:10.1016/j.jallcom.2019.05.010

High-temperature impregnation of solid single-phase Y-doped BaCeO₃ with melted WO₃ was performed towards the evaluation of the possibility of formation of chemically stable material with relatively high electrical conductivity in this system. Impregnation was performed at 1470 °C as a function of time. Due to the relatively high reactivity in solid BaCe_0.9Y_0.1O_3-δ - liquid WO₃ system the reactive and the inner layers were distinguished and analysed. The obtained results clearly indicate the decomposition of Y-doped BaCeO₃ in contact with melted WO₃. X-ray diffraction method together with SEM observation, EDS analysis and EIS measurements results were applied to discuss the mechanism of this multi-stage process and the influence of the reaction time on materials structure and microstructure. As it was observed, partial incorporation of tungsten into the barium cerate structure can process not only on the surface of the materials but also in the inner layer of the synthesised materials. This donor-type doping of the perovskite structure finds reflection in both, materials electrical properties and chemical stability towards CO₂/H₂O. The decrease of proton defects concentration was observed together with the significant increase of the materials chemical stability in comparison to un-modified single-phase BaCe_0.9Y_0.1O_3-δ. The calculated values of activation energy of conductivity confirm the change in the conduction mechanism. Also, the presence of an additional phase in the intergranular voids of doped BaCeO₃ was postulated.

Full text of DOI:10.1016/j.jallcom.2019.05.010

Accepted Manuscript
Reactivity of solid BaCe
Y
O
towards melted WO
3
0
.9 0.1 3-δ
Agnieszka Lacz, Paulina Okas, Radoslaw Lach
PII:
S0925-8388(19)31668-8
DOI:
https://doi.org/10.1016/j.jallcom.2019.05.010
JALCOM 50542
Reference:
To appear in: Journal of Alloys and Compounds
Received Date: 4 February 2019
Revised Date: 26 April 2019
Accepted Date: 1 May 2019
Please cite this article as: A. Lacz, P. Okas, R. Lach, Reactivity of solid BaCe
Y O
.9 0.1 3-δ
0
towards melted WO , Journal of Alloys and Compounds (2019), doi: https://doi.org/10.1016/
3
j.jallcom.2019.05.010.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Reactivity of solid BaCe Y O towards melted WO3
0
.9 0.1 3-δ
Agnieszka Lacz*, Paulina Okas, Radoslaw Lach
AGH University of Science and Technology, Faculty of Materials Science and Ceramics,
al. A. Mickiewicza 30, 30-059 Krakow, Poland
*corresponding author, e-mail: alacz@agh.edu.pl
Abstract
High-temperature impregnation of solid single-phase Y-doped BaCeO with melted WO was
3
3
performed towards the evaluation of the possibility of formation of chemically stable material
with relatively high electrical conductivity in this system. Impregnation was performed at
1
470°C as a function of time. Due to the relatively high reactivity in solid BaCe Y O
-
0
.9 0.1 3-δ
liquid WO system the reactive and the inner layers were distinguished and analysed. The
3
obtained results clearly indicate the decomposition of Y-doped BaCeO in contacted with
3
melted WO . X-ray diffraction method together with SEM observation, EDS analysis and EIS
3
measurements results were applied to discuss the mechanism of this multi-stage process and
the influence of the reaction time on materials structure and microstructure. As it was
observed, partial incorporation of tungsten into the barium cerate structure can process not
only on the surface of the materials but also in the inner layer of the synthesised materials.
This donor-type doping of the perovskite structure finds reflection in both, materials electrical
properties and chemical stability towards CO /H O. The decrease of proton defects
2
2
concentration was observed together with the significant increase of the materials chemical
stability in comparison to un-modified single-phase BaCe Y O . The calculated values of
0
.9 0.1 3-δ
activation energy of conductivity confirm the change in the conduction mechanism. Also, the
presence of an additional phase in the intergranular voids of doped BaCeO was postulated.
3
1

ACCEPTED MANUSCRIPT
Introduction
Solid-state proton conductors area is dominated by materials based on BaCeO , where proton
3
conductivity appears in hydrogen/water containing atmosphere as the consequence of the
following reactions (the Kröger-Vink notation is used):
ꢅ
∙
∙
ꢀꢁ
+ ꢂ2ꢃ ꢂ+ ꢂ2ℎ ꢂ↔ 2ꢃꢀ
ꢄ
∙
∙
ꢅ
∙
ꢆ + ꢀ ꢃ +ꢂꢃ ꢂ↔ 2ꢃꢀ
ꢄ
ꢁ
ꢄ
Oxygen vacancies, which can be treated as the proton defects source, are created as the result
of BaCeO acceptor doping for example with yttrium, ytterbium, neodymium or gadolinium
3
[
1-5]. However, the wide application of these BaCeO -based materials is limited by their low
3
chemical stability in the corrosive atmospheres mainly rich with H O, CO or H S [3, 6, 7].
2
2
2
Two main strategies were applied towards an improvement of the materials chemical stability:
co-doping of acceptor doped BaCeO and synthesis the composites with acceptor doped
3
BaCeO as one of the main phases [8]. The first approach was widely discussed in the
3
literature in terms of acceptor, isovalent and donor-type doping. Generally, it was found that
the introduction of indium or titanium, zirconium or niobium as co-dopant in cerium position
leads to an improvement of the materials chemical stability, however, it also results in
deterioration of the transport properties [7, 9-12].
The second strategy is mainly realized by the formation of dual cell in doped-BaCeO – doped
3
CeO system, where mainly Y, Gd, or Nd was used as the perovskite modifiers and Sm, Gd,
2
Y or Nd were used to modify the fluorite structure [13-18]. The other possibility of formation
of the two-phase material is the direct location of the second phase on the grain boundary of
acceptor-doped BaCeO [19]. The synthesis of materials in systems based on the doped-
3
BaCeO and metal oxide, carbonates or other perovskite structure materials straight for
3
applications in SOFC technology is also undertaken [20-23].
Though the formation of composite materials based on BaCeO is intensive study lately, a lot
3
of challenges are still to be overcome. It was observed that the chemical stability towards
corrosive atmosphere is sufficient. However, changes in the materials chemical composition
and the defects structure as the result of high reactivity between the composite components
were also observed. It leads to a decrease of conductivity and/or changes in the conduction
mechanism [13, 15-17, 23]. Also, the effect of the uneven distribution of an additional phase
on transport properties was observed [20]. Thus, the components mutual reactivity and the
method of second phase introduction are the crucial factors limiting the potential materials
applications.
2

ACCEPTED MANUSCRIPT
In this work, the modification of single-phase Y-doped BaCeO by impregnation with melted
3
WO was undertaken in order to evaluate the possibility of formation of the composite in this
3
system. It is known, that tungsten(VI) oxide is an n-type semiconductor and its electrical
conductivity is a function of oxygen content in its structure [24]. WO is considered to be
3
responsible for the enhancement of chemical stability and total electrical conductivity of
ceramic electrolytes and glasses based on phosphates [25, 26]. It was also reported that
tungsten can be rather incorporated into the lattice than agglomerated at grain boundaries
[
25]. The main concern of this study is the chemical reactivity in solid BaCe Y O - liquid
0.9 0.1 3-δ
WO system. This study also shows that characterization of the chemical stability and
3
electrical properties of the received materials can not only be used for evaluation of the
materials properties, as it is typically performed in the literature, but they are also one of the
key factors in the discussion and deliberation of the reaction mechanisms.
Experimental
The investigated materials were synthesized by high temperature impregnation of sintered
BaCe Y O with melted WO . Y-doped barium cerate (BaCe Y O ) was synthesized
0
.9 0.1 3-δ
3
0.9 0.1 3-δ
by solid-state reaction method using reagents delivered by Sigma-Aldrich Chemical Company
Inc. The mixture of barium carbonate (BaCO ) and cerium(IV) oxide (CeO ) was impregnated
3
2
with 0.28M water solution of yttrium nitrate prepared based on Y(NO ) ·6H O. The
3
3
2
precursor’s mixture was dried for 24 hours at 90°C and calcined in air at 1200°C for 24 hours.
After homogenization in the absolute alcohol suspension using a rotation-vibration mill with
ZrO grinding media (φ= 5 mm) follow by drying, the powder was formed in the pellet dies,
2
isostatically pressed at 250 MPa and sintered at 1600°C for 3 hours in air. The surface of the
sintered samples was ground and polished before all further synthesis steps to remove the
CeO -based phases formed on the surface of the materials as the result of BaO evaporation
2
during sintering. The obtained polished pellets in cylinder-shaped form (φ = 10 mm, thickness
2
mm) had a density (estimated based on the mass and geometry of the samples) about 85% of
theoretical density. As prepared BaCe Y O samples were placed in the Al O crucible,
0
.9 0.1 3-δ
2
3
covered on top with proper amount of WO powder (Sigma-Aldrich Chemical Company Inc.)
3
and heated to 1470°C to allow melting of tungsten oxide. The samples were kept for 1, 2 and
4
hours at a selected temperature to let the penetration of liquid WO into BaCe Y O . It
3 0.9 0.1 3-δ
3

ACCEPTED MANUSCRIPT
must be added that for all materials the mass of WO was assumed to achieve molar ratio of
3
0
.95 : 0.05 for BaCe Y O and WO , respectively.
0.9 0.1 3 3
To describe the phase composition and crystal structure of materials after impregnation of
BaCe Y O with WO X-ray diffractometer Philips X’Pert (monochromatized CuKα
0
.9 0.1 3-δ
3
radiation) was used. Rietveld phase analysis was applied to determine the amount of observed
phases and crystallographic parameters. Microstructure and chemical composition were
defined by Scanning Electron Microscopy SEM (Nova Nano SEM 200 FEI & Oxford
Instruments) combined with Energy Dispersive Spectrometry (EDS) (FEI & Oxford
Instruments). XRD measurements were performed on the samples surfaces, while SEM and
EDS analysis were carried out on the samples polished cross-sections.
Chemical stability of BaCe Y O after reaction with WO towards carbon dioxide and
0
.9 0.1 3-δ
3
water vapour was evaluated using thermogravimetry analysis performed on SDT 2960 TA
-1
3
-1
Instruments (mass about 50 mg, heating rate 10°C·min , synthetic air flow of 100cm · min ,
platinum crucibles). The detailed procedure of chemical stability test is given in the Results
and discussion part.
To determine the influence of WO on electrical properties of Y-doped BaCeO the
3
3
Electrochemical Impedance Spectroscopy (EIS) was applied using Frequency Response
Analyser FRA 1260 coupled with the Dielectric Interface DI 1296A (both Solartron
Analytical Inc., USA). The measurements were executed on pellets with Pt porous electrodes
applied centrally from both sides by screen printing (Heraeus Pt LP 11-4493 paste, dried at
1
20°C/24 h and fired at 850°C for 10 minutes). EIS measurements were performed as a
function of temperature (300-650°C) in oxidising (dry synthetic air) and reducing (10% H in
2
-1
6
Ar) atmospheres. The measurements were carried out in 10 -10 Hz frequency range and the
recorded data were analysed using Z’View software (Scribner Asc., USA).
Result and discussion
Fig. 1 presents the results of XRD analysis of the surface of the samples after 1, 2 and 4 hours
reaction at 1470°C between solid single-phase BaCe Y O (orthorhombic, space group
0
.9 0.1
3
Pnma) and liquid WO . The synthesis procedure assumed that melted WO modifier
3
3
penetrates Y-doped BaCeO host material. However, the presence of CeO (cubic, space
3
2
group Fm-3m) as the main phase clearly indicates partial decomposition of Y-doped BaCeO3
in contact with melted tungsten oxide. As the minor phases, tungsten and barium rich phases
(
BaWO and Ba WO ) were also observed suggesting the chemical reactions between WO
4 3 6
3
4

ACCEPTED MANUSCRIPT
and BaO – the second product of the perovskite decomposition. However, concerning that
barium content is much lower that cerium one, the presence of un-reacted BaO in the
amorphous form and/or barium oxide evaporation due the relatively high reaction temperature
is highly probable. Thus, the effect of reaction time finds reflection in the material’s phase
composition (Table 1). Cerium(IV) oxide is a dominant component on the sample surface
regardless to the reaction time. The presence of other components varies with time and can be
used to discuss the reaction mechanism.
Fig. 1. XRD results for the materials synthesised in BaCe Y O – WO systems as a function of
0
.9 0.1
3
3
reaction time.
Table 1. Surface phase composition of BaCe Y O samples impregnated with WO
3
0.9 0.1
3
Mass %
Reaction time
BaCeO3
orthorhombic
BaCeO
3
CeO2
BaWO4
Ba WO
3
6
cubic
1
2
4
hour
hours
hours
89.6 %
88.1 %
91.5 %
6.8 %
-
8.5 %
3.6 %
-
-
1.7 %
-
-
-
10.2 %
-
Barium tungstate BaWO (tetragonal, space group I41/a) is the result of the reaction between
4
BaO and WO , however its relatively small amount (3.6 mas.% = 0.017 mol.%), concerning
3
the initial amount of WO on the synthesis step (molar ratio of BaCe Y O to WO was
3
0.9 0.1
3
3
0
.95 : 0.05), suggests some amount of un-reacted WO probably in the amorphous form. After
3
5

ACCEPTED MANUSCRIPT
2
hours of reaction instead of BaWO the complex Ba WO compound was detected. It
4
3
6
implies the progressive reaction in the BaO-WO system. Also, the decrease of the amount of
3
tungsten rich phase is observed suggesting the evaporation occurring in the system. Thus, the
extension of reaction time to 4 hours leads to disappear of the barium and tungsten rich
phases. It must be mentioned that BaWO melts at 1475°C [27] and the melting temperature
4
of WO is 1473°C [28]. Therefore, both can be present as liquid phases in the reaction
3
conditions. Concerning also the high volatility of BaO, the high-temperature evaporation of
barium and tungsten containing compounds after an extension of reaction time is highly
possible. The second minor fraction is barium cerate – BaCeO . An orthorhombic Y-doped
3
BaCeO was initially used as the material for impregnation. After 1 hour reaction, the residue
3
of orthorhombic BaCeO was seen, but the extension of reaction time to 2 hours results in the
3
change of the perovskite structure from orthorhombic to cubic one - the structure with higher
symmetry. It must be noticed that thermodynamic stability of the perovskites is associated, in
approximation, with the Goldschmidt tolerance factor, t, which characterize the distortion of
ABO structure. It is known that almost all perovskites have the t value variety from 0.75 to
3
1
.00 and formally t equal 1 corresponds to the cubic structure [8, 29]. However, the cubic
structure was also observed for perovskites with a smaller value of tolerance factor (t > 0.815)
29]. Generally, the increase of Goldschmidt tolerance factor leads to stabilization of the
perovskite structure and to formation of the structure with higher symmetry. For
[
2
+
BaCe Y O the t factor calculated based on the ionic radii of Ba (0.161 nm, CN =12),
0
.9 0.1
3
4
+
3+
2-
Ce (0.087 nm, CN = 6), Y (0.090 nm, CN = 6) and O (0.140 nm, CN = 6) provided by
Shannon [30] is equal 0.936. It is also known that an introduction of an ion with smaller ionic
radii into cerium position results in an increase of the Goldschmidt parameter. Thus, the
6
+
calculation of t parameter assuming the incorporation of tungsten (W , 0.060 nm, CN = 6)
into cerium position in Y-doped BaCeO leading to formation of double-doped barium cerate
3
(
BaCe0.9-yY W O ) was undertaken. The results, together with the formula used for
0.1 y 3
calculations, are presented in Fig. 2. It can be suggested, that the change of the orthorhombic
structure into the cubic one with the extension of reaction time can be the result of partial
incorporation of tungsten into doped barium cerate structure. The lack of cubic perovskite
after 4 hours of reaction can probably be the effect of progressive decomposition of double-
doped BaCeO . It must be mentioned, that the effects of perovskite cubic structure
3
stabilization by partial incorporation of tungsten into B position were observed for BaCoO3
and SrCoO -based materials previously [31, 32].
3
6

ACCEPTED MANUSCRIPT
Fig. 2. Goldschmidt tolerance factor as a function of mol. fraction of incorporated tungsten
For better understanding the reaction mechanism in solid BaCe Y O - melted WO3
0
.9 0.1 3-δ
system SEM observation and EDS analysis were undertaken on the samples polished cross-
sections (Fig. 3). Based on these results three layers can be distinguished regardless to the
reaction time as marked in Fig. 3C. These data also indicate that reactions in the
BaCe Y O - WO system process mainly on the samples surface since the depth of the
0
.9 0.1 3-δ
3
reactive layers (mainly 1st and 2nd ones) do not show significant differences with time.
However, it must be noticed that an increase of total porosity with the extension of reaction
time is clearly seen indicating the evaporation process in the system. The chemical
composition of the 1st layer after 1 hour of reaction (Fig. 3A) shows the presence of CeO as
2
one of the components. It stays in agreement with XRD results discussed previously. The
relatively high and comparable amount of barium and tungsten suggests the presence of
BaWO , also detected by XRD measurements. However, the existence of WO and/or BaO
4
3
cannot be excluded. It should be noticed, that the chemical composition of this layer changes
with the extension of reaction time. Namely, a gradual decrease of both barium and tungsten
amount is observed (Fig. 3B and Fig. 3C). It implies evaporation of barium and tungsten
containing phases, similar to observation performed based on XRD results. It must also be
noticed, that yttrium clearly seen in EDS analysis is probably incorporated into CeO cubic
2
structure, since Ce Y O materials are widely discussed in the literature even for relatively
1
-x
x
2
high yttrium dopant content (x > 0.33) [33-35]. The second layer can be treated as the
intermediate one. The areas rich with Ce and O (Fig. 3A), together with the areas rich with
Ba, W and O (Fig. 3B) can be observed indicating the presence of CeO (probably doped with
2
7

ACCEPTED MANUSCRIPT
yttrium) and BaWO -based phases. However, the composition of this layer also shows the
4
presence of un-decomposed doped BaCeO . Tungsten clearly observed in this area can be
3
both incorporated into the BaCeO structure, as was previously suggested based on the XRD
3
results and/or can be the component of other phases.
Fig. 3. SEM images of cross-sections of the materials synthesised in BaCe Y O -WO systems as a
0.9 0.1
3
3
function of reaction time (A – 1 h, B – 2 h, C – 4 h) together with the EDS data.
8

ACCEPTED MANUSCRIPT
The chemical composition of the 3rd layer does not show significant differences with reaction
time. The amount of tungsten (3-4 at. %) is comparable after 1 and 4 hours of reaction. It
suggests that the chemical composition of this layer is almost not affected by time. It must be
underlined, that SEM microphotographs do not indicate that this layer is a two-phase material.
However, it cannot be excluded. Also, based on these results it is impossible to state whether
the tungsten is incorporated into the BaCeO structure or it is a component of additional
3
tungsten containing phase. Thus, for a better description of the analysed system, all three
layers were removed by samples polishing to reveal the inner layer of the synthesised
materials. Fig. 4 presents the microstructure and chemical composition of as-received
materials.
Fig. 4. SEM images of the cross-sections of inner part of the materials synthesised in BaCe Y O -
0
.9 0.1
3
WO systems as a function of reaction time (A – 1 h, B – 2 h, C – 4 h) together with the EDS data
3
Microstructure and chemical composition of the inner part of the synthesised materials do not
show significant changes with reaction time. Tungsten content is in the range 3 – 4 at. % and
slight decrease with time, while the total porosity increases from about 14 vol. % to 20 vol.%
for 1 and 4 hours of reaction respectively. Regardless of the reaction time, XRD
measurements performed on the samples after removing of reactive layer clearly show the
presence of single phase BaCeO (orthorhombic, space group Pnma). The lattice parameters
3
for single phase BaCe Y O (as the reference) and the materials after reaction with melted
0
.9 0.1
3
WO are collected in Table 2. The decrease of Y-doped BaCeO lattice parameters was
3
3
observed after 1 hour reaction with WO . Thus, partial incorporation of tungsten into barium
3
cerate structure, probably into the cerium or yttrium position can be postulated, since the ionic
6
+
4+
radii of W (0.060 nm for CN = 6) is significantly smaller than the ionic radii of Ce (0.087
3
+
nm for CN = 6) and Y (0.090 nm for CN = 6). The extension of the reaction time to 2 and 4
hours does not affect the BaCeO lattice parameters significantly. It suggests that there is the
3
9

ACCEPTED MANUSCRIPT
solubility limit of tungsten in the barium cerate structure. Moreover, this limit is probably
reached already after the lowest reaction time. The change of the lattice parameters can also
be the result of BaO evaporation as was discussed in [36], due to the relatively high reaction
temperature. However, since the extension of the reaction time does not lead to the further
decrease of these parameters tungsten incorporation seems to have the major influence on the
lattice parameters of doped BaCeO after reaction with melted WO .
3
3
Table 2. BaCeO lattice parameters for the inner part of the materials synthesised in BaCe Y O -
3
0.9 0.1
3
WO system.
3
a [nm] in BaCeO3 b [nm] in BaCeO3 c [nm] in BaCeO3
BaCe Y O - reference
0
.9 0.1
3
0.6213
0.6184
0.6184
0.6185
0.8772
0.8768
0.8772
0.8767
0.6240
0.6216
0.6215
0.6216
orthorhombic (Pnma)
BaCe Y O -WO
0
.9 0.1
3
3
3
3
1
hour of reaction
BaCe Y O -WO
0
.9 0.1
3
2
hours of reaction
BaCe Y O -WO
0
.9 0.1
3
4
hours of reaction
Both SEM and XRD results clearly indicate that the inner part of the materials synthesised in
solid BaCe Y O – melted WO system were single-phase. Moreover, EDS data and lattice
0
.9 0.1
3
3
parameters values suggest the partial incorporation of tungsten into BaCeO structure and
3
tungsten solubility limit. Furthermore, the materials microstructure and composition seems to
be almost independent on the reaction time.
Single-phase Y-doped BaCeO is a proton conductor since oxide ion vacancies created in the
3
3
+
BaCeO structure as the result of acceptor-type (Y ) doping are considered as the source of
3
the protonic defects in water/hydrogen containing atmosphere [37, 38]. Also, yttrium doped
BaCeO is chemically unstable in carbon dioxide and water vapour rich atmosphere, due to
3
the high reactivity of the perovskite towards CO /H O. As the result of reactions of BaCe
1-
2
2
Y O with the surrounding atmosphere the formation of BaCO and Ba(OH) was confirmed,
x
x
3
3
2
however, absorption of water and/or CO is also possible [3, 39]. Thus, the inner parts of the
2
materials received as the result of the reaction in solid BaCe Y O – melted WO system
0
.9 0.1
3
3
with the microstructure, chemical composition and structure define in Fig. 4 and Table 2 were
submitted for exposition test. Samples were placed in the custom-made degradation chamber
for long term (700 h), low temperature (25°C) exposition to an atmosphere rich with CO2
(10%) and saturated with water to allow the reaction between the inner part of received
1
0

ACCEPTED MANUSCRIPT
materials and the surrounding atmosphere. Samples before and after the exposition test were
submitted to thermogravimetric (TG) measurements (SDT 2960 TA Instruments, sample mass
-1
3
-1
about 50 mg, heating rate 10 °C·min , flow of synthetic air 100 cm ·min ). Figure 5 presents
obtained TG results, together with the data for single-phase BaCe Y O as the reference.
0
.9 0.1 3-δ
The detailed discussion of the processes occurred during the exposition of single-phase
BaCe Y O for CO /H O rich atmosphere on the TG curve is provided in [39]. For un-
0
.9 0.1 3-δ
2
2
modified Y-doped BaCeO sample before the test only single mass loss (ꢀm < 0.03%) in the
3
temperature range 500-750°C is observed (Fig. 5). As was previously reported, this effect can
be attributed to the degradation of protonic defects created in the BaCeO -based materials as
3
the result of acceptor doping [10, 39]. TG curve for un-modified BaCe Y O after the
0
.9 0.1 3-δ
exposition to CO and water vapour rich atmosphere contains several not clearly separated
2
steps leading to 1.4 % as the final mass loss. These steps can be assigned to the desorption of
adsorbed gases and decomposition of Ba(OH) and BaCO formed during the exposition as
2
3
the results of reaction in the Y-doped BaCeO - CO /H O system, since the thermal
3
2
2
decomposition of Ba(OH) and BaCO starts about 410°Cand 900°C, respectively [40, 41]. It
2
3
must be underlined that the exposition does not affect the protonic defects, since they
degradation is still observed for un-modified Y-doped BaCeO sample after the test (Fig. 5).
3
Fig. 5. TG curves for inner part of materials synthesised in BaCe Y O – WO system before and
0.9 0.1
3
3
after the exposition test, together with the results for single-phase BaCe Y O as the reference [39]
0
.9 0.1 3-δ
The shape of TG curves, before and after the test, recorded for the inner part of the materials
synthesised in BaCe Y O – WO system is significantly different while comparing with
0
.9 0.1
3
3
1
1

ACCEPTED MANUSCRIPT
proper TG curves for single-phase un-modified BaCe Y O . Except for slight desorption,
0
.9 0.1 3-δ
there are no signs of other effects clearly seen for single phase Y-doped BaCeO . Moreover, it
3
seems that the exposition to CO /H O rich atmosphere does not affect the materials
2
2
synthesised in BaCe Y O – WO system, since the shape of TG curves is similar for
0
.9 0.1
3
3
materials before and after the test. The most important is the lack of protonic defects in the
materials before the test. It suggests that, impregnation of Y-doped BaCeO with melted WO3
3
results with modification of BaCe Y O defect structure, not only on the material’s
0
.9 0.1 3-δ
6
+
surface but also in the inner part of the samples. Thus, the partial incorporation of W into
the perovskite structure can be postulated. It stays in agreement with the discussion of
BaCeO3 lattice parameters provided above. Moreover, according to the theoretical
calculations introduction of an ion with smaller ionic radii and higher electronegativity (e.g.
tungsten) to the perovskite structure leads to unavoidably decrease of the proton defects
concentration [42]. Also, vanishing of the protonic defects as the result of the change of Y-
doped BaCeO defect structure by co-doping with tungsten was also observed for materials
3
synthesised in BaCe Y O – BaWO system [19]. Additionally, an improvement of the
0
.9 0.1
3
4
materials chemical stability towards water vapour and CO in comparison to single-phase Y-
2
doped BaCeO can also be discussed in terms of the changes in the chemical composition. As
3
was previously shown, co-doping with tungsten leads to an increase of the Goldschmidt's
tolerance factor (Fig. 2), the parameter which direct influence on the enhancement of the
stability of the perovskite structure materials was already confirmed and discussed in the
literature [43, 44]. The electronegativity of the dopant is an additional factor which should be
considered while the chemical stability is discussed. It is known that an introduction of the
element with higher electronegativity in the cerium position in the BaCeO -based materials
3
leads to an increase of the acid characteristics and stabilisation of the perovskite structure,
what finally results with the increase of the chemical stability [12, 45, 46]. According to the
Authors [46], the dopant electronegativity is one of the key parameters for inhibiting the
deterioration of the perovskite structure. It was clearly seen when indium or niobium was
introduced into the Y-doped BaCeO structure [45, 46]. The Allred-Rochow electronegativity
3
of tungsten (1.40) is significantly higher than the electronegativity of cerium or yttrium (1.08
and 1.11, respectively). Thus, concerning both the Goldschmidt tolerance factor and the
dopant electronegativity it can be postulated that the partial incorporation of tungsten into the
Y-doped BaCeO structure seems to be one of the major factors responsible for the
3
improvement of the chemical stability of the inner part the Y-doped BaCeO materials
3
modified with melted WO3.
1
2

ACCEPTED MANUSCRIPT
Fig. 6 shows the example of EIS spectra (using Nyquist representation) for dry air atmosphere
recorded for the inner part of the Y-doped BaCeO materials after impregnation with melted
3
WO as a function of reaction time. Spectra consist of two semicircles, where the high-
3
frequency semicircle can be attributed to the grain interior and the low-frequency semicircle
to the grain boundary properties. It must be mentioned, that this type of materials response
suggests the presence of ionic-type conductivity. A small decrease of the grain boundary
resistance with an extension of the reaction time can be the result of a decrease of the grain
boundary surface area due to an increase of the grain size. It is a consequence of the
progressive sintering as the temperature of the reaction is relatively high. This effect was
already discussed in the literature [47]. Obtained results were fitted using equivalent circuit
consists of two connected in parallel R1 - CPE1 and R2 - CPE2 elements placed in series
using Z'View 2.0 software (Scribner Asc., USA). The calculated resistance and CPE values
were attributed to the grain interior and the grain boundary properties concerning the
-11
-9
determined CPE parameters (10 F and 10 F) and the literature data [48].
Fig. 6. EIS spectra of the inner part of the BaCe Y O materials modified with WO for dry air at
0
.9 0.1
3
3
400°C as a function of reaction time.
Based on the obtained resistances and the geometry of the samples the electrical
conductivities of the grain interior and the grain boundary were calculated and presented in
Fig. 7 in Arrhenius coordinates together with the results for single-phase BaCe Y O and
0
.9 0.1 3-δ
two-phase BaCe Y O - BaWO materials as the references [19, 39]. It must be mentioned
0
.9 0.1
3
4
that the detailed discussion of electrical properties of BaCe Y O - BaWO material was
0
.9 0.1
3
4
1
3

ACCEPTED MANUSCRIPT
performed in [19]. Briefly, it was found that impregnation of sintered single-phase
BaCe Y O with melted BaWO resulted in formation of two-phase material, with
0
.9 0.1 3-δ
4
perovskite phase as the major one and BaWO phase randomly distributed in the intergranular
4
voids of doped-BaCeO as the minor one. It was also shown that during this synthesis co-
3
doping of perovskite phase with tungsten occurs.
The significant decrease of the conductivity of grain interior and grain boundary for Y-doped
BaCeO materials modified with tungsten oxide, almost independent on to the reaction time,
3
was observed for dry air atmosphere (Fig. 7).
Fig. 7. Conductivity of the grain interior and the grain boundary of BaCe Y O materials modified
0
.9 0.1
3
with WO for air atmosphere presented in Arrhenius coordinates.
3
For single-phase un-modified BaCe Y O in dry air, the holes and the oxide ion vacancies
0
.9 0.1 3-δ
are dominant charge carriers. Thus, the decrease of the conductivity suggests the change in
the defects structure of barium cerate modified with WO compare to un-modified
3
BaCe Y O . For the discussion of the conduction mechanism, the activation energies of
0
.9 0.1 3-δ
conductivity were calculated and collected at Table 3, together with the proper results for the
1
4

ACCEPTED MANUSCRIPT
reference materials. The activation energy E for the grain interior conductivity significantly
a
increases after the exposition of single-phase BaCe Y O to melted WO reaching the
0
.9 0.1 3-δ
3
values comparable with BaCe Y O - BaWO material where co-doping of the perovskite
0
.9 0.1
3
4
structure with tungsten was observed. Thus, the value of E of the conductivity is another
a
6+
parameter suggesting the partial incorporation of W into the BaCeO structure, as was also
3
postulated based on the XRD, EDS and TG analysis results. It must be noticed, that
6
+
incorporation of W into cerium position in BaCeO structure is donor-type doping. Thus, the
3
compensation of electron-hole pairs should be observed, where the electrons are generated as
the results of donor doping and the holes are the consequence of the introduction of yttrium
(
acceptor dopant) into the BaCeO structure. Also, an introduction of a donor dopant into a
3
material with a relatively high content of oxygen vacancies should lead to an annihilation of
the oxygen vacancies, due to the defects compensation. Thus, the decrease of the conductivity
of BaCe Y O modified with WO is observed in comparison to un-modified single-phase
0
.9 0.1 3-δ
3
BaCe Y O (Fig. 7). Concerning also the fact, that oxide ion vacancies are considered as
0
.9 0.1 3-δ
the source of the protonic defects in water/hydrogen containing atmosphere, the decrease of
these defects content could lead to the disappearance of the protonic defects, as was observed
in TG measurements results.
Table 3. Activation energy E of grain interior and grain boundary conductivity in dry air for
a
BaCe Y O -based materials
0.9 0.1
3
Activation energy E (eV)
a
BaCe Y O -based materials
0
.9 0.1
3
grain interior
0.48±0.03
1.04±0.02
1.03±0.04
1.16±0.01
grain boundary
0.76±0.05
BaCe Y O
0
.9 0.1 3-δ
BaCe Y O - WO (1 hour)
1.45±0.02
0
.9 0.1
3
3
BaCe Y O - WO (4 hours)
1.49±0.04
0
.9 0.1
3
3
BaCe Y O - BaWO
4
1.41±0.01
0
.9 0.1
3
Both, XRD measurements and SEM observations suggest that inner part of Y-doped BaCeO3
material after reaction with melted WO is single-phase. However, the value of activation
3
energy of the grain boundary conductivity for BaCe Y O modified with WO and
0
.9 0.1 3-δ
3
BaWO are comparable. Thus, the presence of an additional phase, probably in the amorphous
4
form, located in the intergranular voids of doped BaCeO after reaction with melted WO3
3
cannot be excluded. Especially that, this additional phase can have the blocking effect on
1
5

ACCEPTED MANUSCRIPT
materials conductivity, thus the high value of activation energy across grain boundaries can be
observed.
EIS spectra recorded for hydrogen containing atmosphere were similar for those observed for
dry synthetic air, with clearly separated two semicircles which were also assigned to the grain
interior and the grain boundary response. The analogous procedure of the conductivity
calculation was undertaken and the results for BaCe Y O samples after 1 hours reaction
0
.9 0.1 3-δ
with melted WO together with the data for oxidising atmosphere, as the reference, are
3
presented in Fig. 8.
Fig. 8. Comparison of conductivity of grain interior and grain boundary of BaCe Y O after 1 hour
0
.9 0.1
3
reaction with WO for oxidising and reducing atmosphere presented in Arrhenius coordinates.
3
The increase of the conductivity, for about 1-2 orders of magnitude, is observed for both grain
interior and grain boundary in comparison to air atmosphere. It must be noticed, that the
reaction leading to a creation of the oxide ion vacancies together with the generation of
electrons is privileged in an atmosphere with low oxygen partial pressure due to the oxygen
desertion from the perovskite structure. Thus, the observed increase of the conductivity is
probably the result of the formation of new current carriers in the material. However, the
4
+
6+
possibility of partial reduction of Ce and W concerning the reducing nature of a hydrogen
atmosphere cannot be excluded. Calculated values of the activation energy of grain interior
and grain boundary conductivity for 10% H in Ar atmosphere were 0.84±0.03 and 1.16±0.05,
2
respectively. The first value is comparable with the reported values of the activation energy of
1
6

ACCEPTED MANUSCRIPT
conductivity for perovskite materials where the oxide ion conductivity was dominant [49, 50].
The activation energy of grain boundary conductivity is higher, thus the presence of an
additional phase in the intergranular voids of doped BaCeO , with blocking effects on
3
materials conductivity, cannot be excluded, similar as was postulated based on the results of
EIS measurements performed in air.
Summary and conclusions
The possibly of application of diffusion controlled process of introduction of melted WO into
3
the solid single-phase Y-doped BaCeO was undertaken. The impregnation was performed as
3
a function of time (1, 2 and 4 hours) at temperature 1470°C which was regulated by the
melting temperature of tungsten(VI) oxide. The direct location of WO in the intergranular
3
voids of BaCe Y O was hindered by high chemical reactivity between the components of
0
.9 0.1 3-δ
the potential composite. Thus, the analysis of the received materials was performed for
reactive and the inner layer separately.
As it was shown, solid BaCe Y O undergoes decomposition in the contact with melted
0
.9 0.1 3-δ
WO already after 1 hour reaction. The extension of the reaction time did not lead to an
3
increase in the depth of the reactive layer, however interesting changes in chemical and phase
composition were observed. The main identify phase was CeO , nevertheless, the presence of
2
minority products supported the interpretation of the reaction mechanism. BaWO and
4
Ba WO were the results of reactions in BaO – WO system. However, due to the relatively
3
6
3
high volatility of the compounds at the reaction temperature they could be treated as
intermediate and unstable.
The inner layer properties were discussed concerning chemical and phase composition. Partial
incorporation of tungsten into barium cerate structure was observed. Thermogravimetric
analysis showed that the change in defects structure of doped barium cerate as the
consequence of tungsten co-doping results in disappearing of proton defects existed in the
initial BaCe Y O . The change in the conduction mechanism was also postulated since the
0
.9 0.1 3-δ
increase of activation energy of the grain interior and the grain boundary conductivity
compare to un-modified single-phase BaCe Y O was observed. In contrast, materials
0
.9 0.1 3-δ
modified with WO had significant chemical stability towards CO and water vapour rich
3
2
atmosphere, probably as the results of structure stabilisation by tungsten doping. Also, the
presence of an additional phase in the intergranular void of doped-BaCeO was postulated.
3
1
7

ACCEPTED MANUSCRIPT
Acknowledgments
This work was financially supported by the National Science Centre Poland, Grant No.
017/01/X/ST5/00789.
2
References
[
1] G. Ma, T. Shimura, H. Iwahara, Ionic conduction and nonstoichiometry in Ba Ce0.90Y0.10O , Solid
x 3-δ
State Ion. 110 (1998) 103-110. https://doi.org/10.1016/S0167-2738(98)00130-1.
[
2] M. Oishi, S. Akoshima, K. Yashiro, K. Sato, J. Mizusaki, T. Kawada, Defect structure analysis of
B-site doped perovskite-type proton conducting oxide BaCeO : Part 2: the electrical conductivity and
3
diffusion coefficient of BaCe M O , Solid State Ion. 179 (2008) 2240-2247.
0
.9 0.1 3-δ
https://doi.org/10.1016/j.ssi.2008.08.005.
3] K.H. Ryu, S.M. Haile, Chemical stability and proton conductivity of doped BaCeO –BaZrO solid
[
3
3
solutions, Solid State Ion. 125 (1999) 355-367. https://doi.org/10.1016/S0167-2738(99)00196-4.
4] M. Oishi, K. Yashiro, K. Sato, J. Mizusaki, N. Kitamura, K. Amezawa, T. Kawada, Y. Uchimotod,
Oxygen nonstoichiometry of the perovskite-type oxides BaCe M O (M = Y, Yb, Sm, Tb, and
[
0
.9 0.1 3-δ
Nd), Solid State Ion. 179 (2008) 529-535. https://doi.org/10.1016/j.ssi.2008.03.038.
5] D. Medvedev, A. Murashkina, E. Pikalova, A. Podias, A. Demin, P. Tsiakaras, BaCeO : materials
[
3
development, properties and application, Prog. Mater. Sci. 60 (2014) 72-129.
https://doi.org/10.1016/j.pmatsci.2013.08.001.
[
6] D. Medvedev, J. Lyagaeva, S. Plaksin, A. Demin, P. Tsiakaras, Sulfur and carbon tolerance of
BaCeO -BaZrO proton conducting materials, J. Power Sources 273 (2015) 716-723.
3
3
https://doi.org/10.1016/j.jpowsour.2014.09.116.
7] L. Jingde, W. Ling, F. Lihua, L. Yuehua, D. Lei, G.J. Hongxia, Chemical stability of doped
BaCeO -BaZrO solid solutions indifferent atmospheres, J. Rare Earths 26 (2008) 505-510.
[
3
3
https://doi.org/10.1016/S1002-0721(08)60127-1.
8] D.A. Medvedev, J.G. Lyagaeva, E.V. Gorbova, A.K. Demin, P. Tsiakaras, Advanced materials for
[
SOFC application: Strategies for the development of highly conductive and stable solid oxide proton
electrolytes, Prog. Mater. Sci. 75 (2016) 38-79. http://dx.doi.org/10.1016/j.pmatsci.2015.08.001.
[
9] F. Zhao, Q. Liu, S. Wang, K. Brinkman, F. Chen, Synthesis and characterization of
BaIn0.3-xY Ce O (x = 0, 0.1, 0.2, 0.3) proton conductors, Int. J. Hydrog. Energy 35 (2010) 4258-
x
0.7 3-δ
4
263. https://doi.org/10.1016/j.ijhydene.2010.02.080.
10] P. Pasierb, E. Drozdz-Ciesla, R. Gajerski, S. Labus, S. Komornicki, Chemical stability of
Ba(Ce Ti ) Y O proton-conducting solid electrolytes, J. Therm. Anal. Calorim.96 (2009) 475-480.
[
1-x x 1-y
y
3
https://doi.org/10.1007/s10973-008-9829-x.
1
8

ACCEPTED MANUSCRIPT
[
11] M. Talimi, V. Thangadurai, Electrical conductivity and chemical stability of perovskite-type
BaCe0.8-xTi Y O , Ionics 17 (2011) 195-200. https://doi.org/10.1007/s11581-011-0522-x.
x
0.2 3-δ
[
12] E.D. Bartolomeo, A. D’Epifanio, C. Pugnalini, F. Giannici, A. Longo, A. Martorana, S. Licoccia,
Structural analysis, phase stability and electrochemical characterization of Nb doped BaCe Y O
0.9 0.1 3-δ
electrolyte
https://doi.org/10.1016/j.jpowsour.2011.10.038.
13] D. Lin, Q. Wang, K. Peng, L.L. Shaw, Phase formation and properties of composite electrolyte
BaCe Y O -Ce Gd O for intermediate temperature solid oxide fuel cells, J. Power Sources 205
for
IT-SOFCs,
J.
Power
Sources
199
(2012)
201-206.
[
0
.8 0.2 3-δ
0.8
0.2 1.9
(
[
2012) 100-107. https://doi.org/10.1016/j.jpowsour.2012.01.033.
14] D. Hirabayashi, A. Tomita, T. Hibino, M. Nagao, M. Sano, Design of a reduction-resistant
Ce Sm O electrolyte through growth of a thin BaCe Sm O layer over electrolyte surface,
0.8
0.2 1.9
1-x
x
3-α
Electrochem. Solid State Lett. 7 (2004) 318-320. https://doi.org/10.1149/1.1786231.
15] M. Khandelwal, A. Venkatasubramanian, T.R.S. Prasanna, P. Gopalan, Correlation between
microstructure and electrical conductivity in composite electrolytes containing Gd-doped ceria and
Gd-doped barium cerate, J. Eur. Ceram. Soc. 31 (2011) 559-568.
https://doi.org/10.1016/j.jeurceramsoc.2010.10.027.
16] D. Medvedev, E. Pikalova, A. Demin, A. Podias, I. Korzun, B. Antonov, Structural,
thermomechanical and electrical properties of new (1-x)Ce Nd O -xBaCe Nd O composites,
[
[
0
.8
0.2 2-δ
0.8
0.2 3-δ
J. Power Sources 267 (2014) 269-279. https://doi.org/10.1016/j.jpowsour.2014.05.070.
17] R. Gawel, K. Przybylski, M. Viviani, Chemical stability and electrical properties of
[
BaCe0.85Y0.15O -Ce0.85Y0.15O composite bulk samples for use as central membrane materials in dual
3-δ
2-δ
PCFC-SOFC
fuel
cells,
Mater.
Chem.
Phys.
147
(2014)
804-814.
https://doi.org/10.1016/j.matchemphys.2014.06.024.
18] S. Ricote, A. Manerbino, N.P. Sullivan, W.G. Coors, Preparation of dense mixed electron- and
[
proton-conducting ceramic composite materials using solid-state reactive sintering:
BaCe Y M O -Ce Y M O (M=Y, Yb, Er, Eu), J. Mater. Sci. 49 (2014) 4332-4340.
0
.8 0.1 0.1 3-δ
0.8 0.1 0.1 2-δ
https://doi.org/10.1007/s10853-014-8129-z.
19] A. Lacz, P. Pasierb, Electrical properties and chemical stability of BaCe Y O -BaWO
4
[
0
.9 0.1
3
composites synthesized by co-sintering and impregnation method, Solid State Ion. 302 (2017) 152-
57. https://doi.org/10.1016/j.ssi.2016.12.006.
20] K. Zagorski, S. Wachowski, D. Szymczewska, A. Mielewczyk-Gryn, P. Jasinski, M. Gazda,
1
[
Performance of a single layer fuel cell based on a mixed proton-electron conducting composite, J.
Power Sources 353 (2017) 230-236. https://doi.org/10.1016/j.jpowsour.2017.04.0.
[
21] Y. Hei, J. Huang, C. Wang, Z. Mao, Novel doped barium cerate - carbonate composite electrolyte
material for low temperature solid oxide fuel cells, Int. J. Hydrog. Energy 39 (2014)14328-14333.
http://dx.doi.org/10.1016/j.ijhydene.2014.04.031.
1
9

ACCEPTED MANUSCRIPT
[
22] V.B. Vert, C. Solís, J.M. Serra, Electrochemical Properties of PSFC–BCYb composites as
cathodes for proton conducting solid oxide fuel cells, Fuel Cells 11 (2011) 81–90.
http://dx.doi.org/10.1002/fuce.201000090.
[
23] L. Zhang, R. Lan, S. Tao, An intermediate temperature fuel cell based on composite electrolyte of
carbonate and doped barium cerate with SrFe Mn Mo O cathode, Int. J. Hydrog. Energy 38
0.7
0.2
0.1 3-δ
(
[
2013) 16546-16551. http://dx.doi.org/10.1016/j.ijhydene.2013.06.029.
24] V.O. Makarov, M. Trontelj, Sintering and electrical conductivity of doped WO , J. Eur. Cer.
3
Soc.16 (1996) 791-794. https://doi.org/10.1016/0955-2219(95)00204-9.
25] X. Lu, F. Meng, S. Huang, D. Zhao, Enhanced ionic conductivity and chemical stability of
Li Al Ti (PO ) by doping of WO3, Mater. Lett. 230 (2018) 177–179.
https://doi.org/10.1016/j.matlet.2018.07.103.
26] L. Pavic, A. Santic, J. Nikolic, P. Mosner, L. Koudelka, D. Pajic, A. Mogus-Milankovic, Nature
of mixed electrical transport in Ag O-ZnO-P O glasses containing WO and MoO , Electrochim. Acta
[
1.3
0.3 1.7
4 3
[
2
2
5
3
3
2
76 (2018) 434-445. https://doi.org/10.1016/j.electacta.2018.04.029.
27] I.S. Voronina, L.I. Ivleva, T.T. Basiev, P.G. Zverev, N.M. Polozkov, Active Raman media:
SrWO :Nd3+, BaWO :Nd3+. Growth and characterization, J. Optoelectron. Adv. M. (2003) 887–92.
[
4
4
[
28] T. Kosuge, Y. Benino, V. Dimitrov, R. Sato, T. Komatsu. Thermal stability and heat capacity
changes at the glass transition in K O-WO -TeO glasses, J. Non-Cryst. Solids 242 (1998) 154-164.
2
3
2
https://doi.org/10.1016/S0022-3093(98)00800-X.
29] L.M. Feng, L.Q. Jiang, M. Zhu, H.B. Liu, X. Zhou, C.H. Li, Formability of ABO cubic
[
3
perovskites, J. Phys. Chem. Solids 69 (2008) 967–974. https://doi.org/10.1016/j.jpcs.2007.11.007.
30] R.D. Shannon, Revised effective ionic radii and systematic studies of interatomic distances in
halides and chalcogenides, Acta Crystallogr. A32 (1976) 751-767.
https://doi.org/10.1107/S0567739476001551.
31] J. Zhang, Z. Zhang, Y. Chen, X. Xu, C. Zhou, G. Yang, W. Zhou, Z. Shao, Materials design for
[
[
ceramic oxygen permeation membranes: Single perovskite vs. single/double perovskite composite, a
case study of tungsten doped barium strontium cobalt ferrite, J. Membr. Sci. 566 (2018) 278-287.
https://doi.org/10.1016/j.memsci.2018.09.004.
[
32] E.V. Artimonova, O.A. Savinskaya, A.P. Nemudry, Effect of B-site tungsten doping on structure
and oxygen permeation properties of SrCo Fe O perovskite membranes, J. Eur. Cer. Soc. 35
0.8
0.2 3−δ
(
[
2015) 2343-2349. http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.011.
33] J. van Herle, T. Horita, T. Kawada, N. Sakai, H. Yokokawa, M. Dokiya, Sintering behaviour and
ionic conductivity of yttria-doped ceria, J. Eur. Ceram. Soc. 16 (1996) 961-973.
https://doi.org/10.1016/0955-2219(96)00012-X.
[
34] M. Hartmanova, E.E. Lomonova, V. Navratil, P. Šutta, F. Kundracik, Characterization of yttria-
doped ceria prepared by directional crystallization, J. Mater. Sci. 40 (2005) 5679–5683.
https://doi.org/10.1007/s10853-005-2795-9.
2
0

ACCEPTED MANUSCRIPT
[
35] M. Dudek, J. Molenda, Ceria-yttria-based solid electrolytes for intermediate temperature solid
oxide fuel cell, Mater. Sci. 24 (2006) 45–52.
36] P. Babilo, T. Uda, S.M. Haile, Processing of yttrium-doped barium zirconate for high proton
conductivity, J. Mater. Res. 22 (2007) 1322-1330. https://doi.org/10.1557/jmr.2007.0163.
37] K. Takeuchi, C.K. Loong, J.W. Richardson Jr., J. Guan, S.E. Dorris, U. Balachandran, The crystal
[
[
structures and phase transitions in Y-doped BaCeO : their dependence on Y concentration and
3
hydrogen doping, Solid State Ion. 138 (2000) 63-77. https://doi.org/10.1016/S0167-2738(00)00771-2.
[
38] M. Oishi, S. Akoshima, K. Yashiro, K. Sato, J. Mizusaki, T. Kawada, Defect structure analysis of
B-site doped perovskite-type proton conducting oxide BaCeO Part 1: The defect concentration of
3
BaCe M O
(M
https://doi.org/10.1016/j.ssi.2008.12.025.
39] A. Lacz, Effect of microstructure on chemical stability and electrical properties of
BaCe Y O , Ionics 22 (2016) 1405-1414. https://doi.org/10.1007/s11581-016-1665-6.
=
Y
and Yb), Solid State Ion. 180 (2009) 127–131.
0
.9 0.1 3−δ
[
0
.9 0.1 3-δ
[
40] G.M. Habashy, G.A. Kolta GA, Thermal-decomposition of hydrates of barium hydroxide, J.
Inorg. Nucl. Chem. 34 (1972) 57–67. doi:10.1016/0022-1902(72)80361-0.
41] A.K. Galwey, M.E. Brown, Thermal decomposition of ionic solids. Amsterdam: Amsterdam
Elsevier Science B.V; 1999.
42] K.D. Kreuer, Proton-Conducting Oxides, Ann. Rev. Mater. Res 33. (2003) 333-359.
https://doi.org/10.1146/annurev.matsci.33.022802.091825.
43] S.M. Haile, G. Staneff, K.H. Ryu, Non-stoichiometry, grain boundary transport and chemical
[
[
[
stability of proton conducting perovskites, J. Mater. Sci. 36 (2001) 1149–1160.
https://doi.org/10.1023/A:1004877708871
2
+
4+
[
44] E. Takayama-Muromachi, A. Navrotsky, Energetics of compounds (A B O ) with the
3
perovskite structure. J. Solid State Chem. 72 (1988) 244–56. https://doi.org/10.1016/0022-
596(88)90028-X
45] A. Radojkovic, M. Zunic, S.M. Savic, G. Brankovic, Z. Brankovic, Chemical stability and
electrical properties of Nb doped BaCe Y O as a high temperature proton conducting electrolyte
4
[
0
.9 0.1 3-δ
for IT-SOFC, Ceram. Int. 39 (2013) 307-313. https://doi.org/10.1016/j.ceramint.2012.06.026
46] A. Radojković, M. Žunić, S.M. Savić, S. Perać, D. LukovićGolić,Z. Branković, G. Branković,
Co-doping as a strategy for tailoring the electrolyte properties of BaCe Y O , Ceram. Int. 45
[
0
.9 0.1 3–δ
(
[
2019) 8279-8285. https://doi.org/10.1016/j.ceramint.2019.01.134
47] S.M. Haile, D.L. West, J. Campbell, The role of microstructure and processing on the proton
conducting properties of gadolinium doped barium cerate, J. Mater. Res. 13 (1998) 1576–1595.
https://doi.org/10.1557/JMR.1998.0219
[
48] E. Barsoukov, J.R. Macdonald, Impedance Spectroscopy Theory, Experiment, and Applications,
Second ed., Wiley-Interscience, 2005.
2
1

ACCEPTED MANUSCRIPT
[
49] K.D. Kreuera, Th. Dippel, Yu.M. Baikov, J. Maier, Water solubility, proton and oxygen diffusion
in acceptor doped BaCeO : A single crystal analysis, Solid State Ion. 86–88 (1996) 613–620.
3
https://doi.org/10.1016/0167-2738(96)00221-4
[
50] A.K. Azad, J.T.S. Irvine, High density and low temperature sintered proton conductor
BaCe Zr Sc Zn O , Solid State Ion. 179 (2008) 678–682.
https://doi.org/10.1016/j.ssi.2008.04.036
0
.5 0.35
0.1
0.05 3–δ
2
2

ACCEPTED MANUSCRIPT
Highlights
Decomposition of BaCe Y O in the contact with melted WO
3
0.9 0.1
3
Incorporation of tungsten into Y-doped BaCeO structure
3
The increase of the activation energy of conductivity - conduction mechanism change