Proton conduction: Via lattice water molecules in oxalato-bridged lanthanide porous coordination polymers

Article abstract of DOI:10.1039/c6dt02677j

The proton conducting properties of two different structural types of porous coordination polymers [La₂(ox)₃(H₂O)₆]·4H₂O (1) and [Er₂(ox)₃(H₂O)₆]·12H₂O (2), where ox^2- = oxalate, were investigated. 1 has a two-dimensional layered structure, whereas 2 has a three-dimensional structure. Both 1 and 2 have hydrophilic one-dimensional channels filled by lattice water molecules with hydrogen-bonding networks. The coordinated H₂O molecules are Lewis acidic due to the lanthanoid ions donating protons to lattice-filling H₂O molecules, thereby forming efficient proton conduction pathways. Alternating-current impedance analyses of 1 and 2 indicated significant proton conduction (σ = 3.35 × 10^-7 S cm^-1 at 368 K for 1, 1.79 × 10^-6 S cm^-1 at 363 K for 2 under RH = 100%, with E_a = 0.35 eV for 1 and 0.47 eV for 2), which was attributed to the Grotthuss mechanism via the lattice H₂O molecules.

Full text of DOI:10.1039/c6dt02677j

Dalton
Transactions
View Article Online
View Journal
PAPER
Proton conduction via lattice water molecules in
oxalato-bridged lanthanide porous coordination
polymers†
Cite this: DOI: 10.1039/c6dt02677j
Ryuta Ishikawa,*a Shunya Ueno,a Sadahiro Yagishita,a,b Hitoshi Kumagai,c
Brian K. Breedloved and Satoshi Kawata*a
The proton conducting properties of two different structural types of porous coordination polymers
[La2(ox)3(H2O)6]·4H2O (1) and [Er2(ox)3(H2O)6]·12H2O (2), where ox2− = oxalate, were investigated. 1 has a
two-dimensional layered structure, whereas 2 has a three-dimensional structure. Both 1 and 2 have
hydrophilic one-dimensional channels filled by lattice water molecules with hydrogen-bonding networks.
The coordinated H2O molecules are Lewis acidic due to the lanthanoid ions donating protons to lattice-
filling H2O molecules, thereby forming efficient proton conduction pathways. Alternating-current impe-
dance analyses of 1 and 2 indicated significant proton conduction (σ = 3.35 × 10−7 S cm−1 at 368 K for 1,
1.79 × 10−6 S cm−1 at 363 K for 2 under RH = 100%, with Ea = 0.35 eV for 1 and 0.47 eV for 2), which was
attributed to the Grotthuss mechanism via the lattice H2O molecules.
Received 7th July 2016,
Accepted 19th August 2016
DOI: 10.1039/c6dt02677j
www.rsc.org/dalton
compounds, solid-state PCPs are considered to be an inter-
mediate between amorphous organic polymers and pure in-
Introduction
The development of new proton conductors will lead to clean organic compounds. Moreover, hydrophilic sites can be easily
energy applications, such as solid electrolytes1–3 and proton- incorporated into PCPs via careful design. Hydrophilic sites in
exchange membrane fuel cells,4–6 which are urgently needed PCPs are filled with lattice H2O molecules, which assemble
as energy sources in industrial processes. There are two typical together in hydrogen-bonding networks. These hydrogen-bond
types of solid-state proton conductors, which are (1) amor- networks serve as efficient proton conduction pathways,
phous organic polymers, such as Nafion,7–9 and (2) pure in- mediated by the layers, pores, or channels of the PCPs.
organic compounds, such as metal oxides10,11 and inorganic
solid acids.12 In general, amorphous organic polymers operate ligands, which have played a key role in the development of
at temperatures below 100 °C, whereas pure inorganic com- PCPs. In the present study, we prepared Ln3+–ox2− (Ln3+
Oxalato ions (ox2−
)
are simple bis-bidentate type
=
pounds become more efficient at higher temperatures above lanthanide ion) based PCPs, [La2(ox)3(H2O)6]·4H2O (1) and
600 °C.7–12 So far, both materials have been widely used in fuel [Er2(ox)3(H2O)6]·12H2O (2) with one-dimensional (1D) channels
cell technology with conventional proton conduction.
filled with lattice H2O molecules connected into hydrogen-
Porous coordination polymers (PCPs) or metal–organic bonding networks.17,18 In addition, the H2O molecules co-
frameworks are candidates for use as proton conductors.13–16 ordinated to the Ln ions are Lewis acidic. Therefore, both 1
Because PCPs can be flexible, like amorphous organic and 2 should exhibit good proton conduction. Herein we show
polymers, and rigid and crystalline, like pure inorganic through proton conductivity measured by alternating current
(AC) impedance spectroscopy, thermal gravimetric (TG) ana-
lyses, and H2O adsorption and desorption measurements that
both 1 and 2 exhibit good proton conduction.
aDepartment of Chemistry, Fukuoka University, 8-19-1 Nanakuma, Jonan-ku,
Fukuoka 814-0180, Japan. E-mail: ryutaishikawa@fukuoka-u.ac.jp
bDaiichi Kigenso Kagaku Kogyo Co., Ltd, 1-6-38 Hirabayashi-minami, Suminoe-ku,
Osaka 559-0025, Japan
cToyota Central Research and Development Laboratories Inc., 41-1 Yokomichi,
Results and discussion
Structural features
The structures of the PCPs of Ln ions and ox2− ligands depend
on the choice of the Ln ion and the reaction conditions.17,18
The structures of 1 and 2 were determined by using powder
Nagakute, Aichi 480-1192, Japan
dDepartment of Chemistry, Tohoku University, 6-3 Aza-aoba, Aramaki, Aoba-ku,
Sedai, Miyagi 980-8578, Japan
†Electronic supplementary information (ESI) available: Results of Jonscher’s
power law, Le Bail refinements, TG data, IR spectra, and additional AC impe-
dance data. See DOI: 10.1039/c6dt02677j
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
X-ray diffraction (PXRD) analyses on the basis of the corres- crystallographic ac plane (Fig. 1a). The crystal lattice of 1 con-
ponding single-crystal X-ray data, as discussed in the tains four lattice H2O molecules, which are stabilized through
Experimental section. The Le Bail refinement fits to the data hydrogen-bonding interactions with O–H⋯O contacts between
shown in Fig. S1,† and the crystallographic parameters are H2O molecules and coordinated to the La site and lattice H2O
summarized in Table S1.† The experimental PXRD patterns of molecules and between H2O molecules. Thus, each 2D layer
1 and 2 are in agreement with their simulated patterns self-assembles into supramolecular three-dimensional (3D)
obtained from single-crystal X-ray analyses.
networks with 1D channels (Fig. 1b).
For 1, the La center is coordinated by three bidentate ox2−
For 2, the coordination geometry around the Er ions is
ligands and three monodentate H2O molecules to form an O9 similar to that of 1 (Fig. S2b†). The ox2− ligands adopt κ-O,O-
coordination geometry, which is best described as a distorted chelating and μ2-bridging modes, linking the Er centers to
nine-coordinated tricapped trigonal antiprism (Fig. S2a†). The form helical substructures. These helical substructures have a
La centers are connected by ox2− ligands in κ-O,O-chelating three-fold rotation axis along the c axis and show left- or
and μ2-bridging modes, resulting in a neutral two-dimensional right-handed rotation, depending on the Δ- and Λ-confor-
(2D) honeycomb layered structure with 6,3-net topology in the mations of the ErO9 polyhedra (Fig. 2a). Adjacent helices are
Fig. 1 Structural representation of 1. (a) Single 2D honeycomb layer. (b) 1D channels highlighting lattice H2O molecules in the 3D porous frame-
work. Light blue polyhedra represent La ions, and red and gray balls and sticks represent O and C atoms, respectively. H atoms are omitted for
clarity.
Fig. 2 Structural representation of 2. (a) Side views of helical strands. M (Λ) helicity (left) and P (Δ) helicity (right). (b) 1D channels highlighting lattice
H2O molecules in the 3D porous framework. Green polyhedra represent Er ions, and red and gray balls and sticks represent O and C atoms, respec-
tively. H atoms are omitted for clarity.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
connected to form a neutral 3D framework with 10,3-net topo- in H2O for a sufficient amount of time for semiactivated 2′,
logy and hexagonal 1D channels running along the c axis. some new peaks appeared in the PXRD pattern of rehydrated 2
Like 1, the 1D channels in 2 are occupied by lattice H2O mole- (Fig. 3b). This may imply that the framework of 2′ is rather dis-
cules, which are stabilized through hydrogen-bonding inter- torted during the semiactivation process, which leads to the
actions (Fig. 2b).
symmetrical change. In other words, the structural symmetry
ˉ
of 2 is higher (R3), whereas that of 2′ would be lower.
Thermo-structural behavior
Therefore, 2′ could be converted back to 2 including a some-
what 2-like structure with a lower symmetry. The dynamic
structural features observed in H2O adsorption and desorption
isotherms for both 1 and 2 indicate that the lattice H2O mole-
cules are tightly held in the framework, which is consistent
with the X-ray, TG and H2O adsorption and desorption studies
(vide supra).
TG analyses were performed on samples of 1 and 2 in the
temperature (T) range of 20–450 °C (Fig. S3†). The gradual
mass losses of 24.38% for 1 and 32.4% for 2 between 20 and
350 °C correspond to the loss of all of the H2O molecules (the
calculated weight losses are 24.95% for 1 and 35.14% for 2,
respectively). Above ca. 380 °C, the frameworks of 1 and 2
began to break down, which was attributed to the decompo-
sition of the ox2− ligand.19,20 Note that the full thermal de-
hydration causes a breakdown of the framework leading to an
amorphous state, which has been previously reported.17
Structural dynamics in H2O adsorption and desorption
processes
Moreover, these amorphous compounds can never be con- H2O adsorption and desorption isotherms of 1′ and 2′ were
verted back to the initial state (compounds 1 and 2) even by acquired (Fig. 3). The H2O adsorption and desorption iso-
soaking in H2O. Nevertheless, semi-activated compounds, therms of 1′ showed that the amount of H2O uptake gradu-
[La2(ox)3(H2O)6] (1′) and [Er2(ox)3(H2O)6] (2′), were sufficiently ally increased with an increase in the relative pressure (P/P0,
obtained by heating 1 and 2 for 3 h at ca. 70 °C for 1′ and at where P0 is the saturation pressure of H2O) up to P/P0 ≈ 0.8.
90 °C for 2′, the temperatures of which were determined from Then, rapid H2O uptake was observed in the higher P/P0
TG analyses (Fig. S3†), where only lattice H2O molecules were ranges from P/P0 ≈ 0.8 to 1. Finally, the maximum H2O
removed from the frameworks.
uptake reached ca. four H2O molecules, whereas the H2O de-
From the PXRD patterns of semi-activated compounds 1′ sorption isotherm (the amount of H2O release) was higher
and 2′, the crystallinity was maintained. However, all of the than that obtained from the H2O adsorption isotherm at the
peaks were very broad, indicating a structural change during same P/P0, indicating a hysteresis loop (Fig. 4a). The H2O
the release of the lattice H2O molecules (Fig. 3). However, in adsorption isotherm of 2′ showed that the amount of
Fourier transform infrared (FT-IR) spectra of 1′ and 2′, the H2O uptake slowly increased with an increase in P/P0 up to
peaks for the CvO groups were unchanged, meaning that the ca. 0.1. Above P/P0 ≈ 0.1, a gradual H2O uptake was observed,
ox2− bridges within the framework remained (Fig. S4†). 1′ and the maximum H2O uptake was ca. twelve H2O mole-
could be fully converted back to 1 by soaking in H2O for a cules. In addition, the H2O adsorption and desorption iso-
sufficient amount of time, which was verified by using the therms of 2′ indicated that there was a large hysteresis loop
PXRD measurement (Fig. 3a). On the other hand, after soaking (Fig. 4b).
Fig. 3 Comparison of PXRD patterns of (a) 1 and (b) 2. Calculated patterns from the corresponding single-crystal X-ray structures.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 4 H2O adsorption and desorption isotherm profiles measured at 25 °C for (a) 1’ and (b) 2’.
Proton conductivity
shows the degree of proton interactions in the proton
hopping processes and is generally in the range of 0 < n < 1
To study the proton conductivity of 1 and 2, AC impedance and T dependent.21,22 Such σAC behaviours are attributed to
measurements were performed at different T ranges and a rela- the long-range translational motion of protons related to the
tive humidity (RH) of 100% in the frequency range (ν) of 50 jump relaxation model. The parameters obtained from fitting
Hz–5 MHz. σAC plots of 1 and 2 showed two distinct regions the σAC values of 1 and 2 are summarized in Table S2.† The
within the measured ν range (Fig. 5). In the low ν regions, the values of n were determined to be in the ranges of 1.50–1.55
σAC values of 1 and 2 plateaued at every T, meaning that the for 1 and 1.31–1.41 for 2. It has been reported that the value
conductivity is ν independent, i.e., direct-current conductivity of n can be larger than unity and that there is no physical
(σDC). In addition, σDC increased with an increase in T, indicat- argument to restrict the value of n below 1. Indeed, in the
ing that a thermally activated conduction process occurred. cases of ion conducting glasses,23 solid solutions of super-
Moreover, the σAC values abruptly increased with an increase ionic ionic salts,24 multiferroic layer thin films,25 and dis-
in ν. The ν dependence of σAC was estimated by using cotic metallomesogens,26 the n values are within the range
Jonscher’s power law for typical proton conductors of 1 < n < 2. In such cases, the σAC behaviours at higher n
(eqn (1)):21,22
are explained by reorientation (most often a rotational
motion of protons), which leads to intermolecular proton
hopping. Therefore, the observed large n values for 1 and 2
were attributed to transport via proton hopping between
lattice H2O molecules, accompanied by rearrangement of the
protons.
σAC ¼ σDC þ Aωn
ð1Þ
where A is the pre-exponential factor, ω is the angular fre-
quency, which can be converted to the linear frequency (ω =
2πν) and n is the power law exponent, the value of which
Fig. 5 ω-Dependence of the dispersion profiles of σAC for (a) 1 and (b) 2 at a given T. Solid black lines were fitted to the data as described in the text
(eqn (1)).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Real (Z′) and imaginary (Z″) components of the AC impe-
dances were strongly dependent on ν in the T ranges of
368–343 K for 1 and 363–322 K for 2 (Fig. 6 and S5†). Both 1
and 2 showed clear ν dependent maxima (Z″max) in the Z″ com-
ponents of the AC impedance. Each Z″ max shifted to higher ν
with an increase in T, indicating that the observed relaxation
behaviours of 1 and 2 are thermally activated processes and
related to the conduction and the reorientation of mobile
protons. Thus, ν dependent Z″ max data of 1 and 2 could be
used to determine the relaxation time (τ) by using the
Arrhenius law (eqn (2)):
ꢀ
ꢁ
Ea
kBT
τ ¼ τ0 exp ꢀ
ð2Þ
where τ0 is the relaxation time at infinite temperature, Ea is the
activation energy, which corresponds to the energy required to
begin the proton movement, and kB is the Boltzmann con-
stant. From the fits of the data, the values of Ea were deter-
mined to be 0.36 and 0.48 eV with τ0 = 6.61 × 10−11 and 2.57 ×
10−13 s for 1 and 2, respectively (Fig. 7).
The Nyquist plots of 1 and 2 clearly showed a single
semicircle with a tail at lower ν, which arises from grain
interior and boundary contributions (Fig. 8 and S6†). The
real axis intercept of the Nyquist plot is related to the
proton resistance of the compounds, which can be con-
verted to the proton conductivity (σ). The σ values of 1 and
2 were determined to be 3.35 × 10−7 S cm−1 at 368 K for 1
and 1.79 × 10−6 S cm−1 at 363 K for 2 at RH = 100%. The T
dependence of σ generally follows the Arrhenius relation
(eqn (3)):
Fig. 7 ν-Dependent Z’’ for (a) 1 and (b) 2 in the ν range of 50 Hz–5 MHz
at a given T. Solid black lines were fitted to the data as described in the
text (eqn (2)).
In PCPs 1 and 2, the hydrogen-bonding networks of
lattice H2O molecules act as efficient proton conduction
pathways. Protons, which hop between neighboring lattice
H2O molecules, diffuse into the layers of 1 and the pores of
2 along with the reorientation of protons. Moreover, co-
ordinated H2O molecules are rendered to be Lewis acids by
the Ln ions (Ln = La for 1 and Er for 2) to contribute
protons to the lattice H2O molecules. This proton transfer
process is called structural diffusion or better known as the
Grotthuss mechanism.4,6 In other words, the proton conduc-
tion of 1 and 2 occurs via the Grotthuss mechanism, and the
Ea values of 1 and 2 are consistent with the related proton
conduction materials following this mechanism. Proton con-
ꢀ
ꢁ
σ0
Ea
kBT
σ ¼ exp ꢀ
ð3Þ
T
duction may also occur through molecular diffusion, where
+
where σ0 is a pre-exponential factor. From a fitting of the the protons diffuse together with H3O+ and NH4 , which is
data, the activation energies of 1 and 2 were estimated to be called the vehicle mechanism.4,6 In some cases, a mixed con-
0.35 and 0.47 eV, respectively (Fig. 9), which are within duction mechanism, which is a combination between the
the range of those previously reported for ox2− bridged vehicle mechanism and the Grotthuss mechanism has been
PCPs.27–34
observed.
Fig. 6 ν-Dependent Z’’ for (a) 1 and (b) 2 in the ν range of 50 Hz–5 MHz at a given T. Solid black lines serve as a guide to the eye.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 8 Nyquist plots of (a) 1 and (b) 2 in the ν range of 50 Hz–5 MHz at a given T. Solid black lines serve as a guide to the eye.
lattice H2O molecules were strongly captured in the frame-
work. 1 and 2 showed significant proton conduction with σ =
3.35 × 10−7 S cm−1 at 368 K and 1.79 × 10−6 S cm−1 at 363 K
under RH = 100% and Ea = 0.35 and 0.47 eV for 1 and 2,
respectively, which were attributed to the Grotthuss mechan-
ism via lattice H2O molecules.
Experimental
Materials and methods
Reagents and solvents purchased from Tokyo Kasei Co. Ltd
and Wako Pure Chemical Industries Ltd. The reagents were of
reagent grade and used without further purification. All
manipulations were performed under aerobic conditions.
Preparation of [La2(ox)3(H2O)6]·4H2O (1). This compound
was prepared using a published procedure.17,18 LaCl3·7H2O
(0.74 g, 2 mmol) and dimethyl oxalate (0.57 g, 4.8 mmol) were
dissolved in water (40 mL) and benzene (10 mL), respectively.
The two clear solutions were mixed with vigorous magnetic
stirring for 1 day at 15 °C, during which time a pale pink fine
crystalline powder gradually appeared. The powder was col-
lected by filtration and then air-dried. Yield: 88%. IR (KBr,
ν/cm−1): 3340(s), 1620(s), 1364(w), 1316(s), 800(m), 733(w),
582(w), 493(m).
Fig. 9 T dependences of σT for 1 and 2. Solid black lines were fitted to
the data as described in the text (eqn (3)).
Conclusions
Two lanthanide–oxalato based porous coordination polymers,
[La2(ox)3(H2O)6]·4H2O (1) and [Er2(ox)3(H2O)6]·12H2O (2),
which were characterized by using TG, PXRD, H2O adsorption
and desorption studies, were shown to be good proton conduc-
tors via AC impedance proton conduction studies. 1 and 2
have hydrophilic 1D channels, which are filled with lattice
H2O molecules connected into hydrogen-bonding networks.
PXRD analyses indicated that the flexibility of the frameworks
of 1 and 2 depended on the number of H2O molecules.
Removing the lattice H2O molecules from 1 and 2 by heating
caused the PCPs to convert to semi-activated compounds,
Preparation of [Er2(ox)3(H2O)6]·12H2O (2). This compound
was prepared using the same method as that for 1, except
ErCl3·6H2O was used instead of LaCl3·7H2O. Yield: 85%.
IR (KBr, ν/cm−1): 3370(s), 1640(s), 1364(s), 1324(s), 813(m),
660(w), 530(w), 489(m).
Physical measurements
[La2(ox)3(H2O)6] (1′) and [Er2(ox)3(H2O)6] (2′). The conversion FT-IR spectra were acquired as KBr disks in the range of
was reversible, and 1 and 2 could be fully recovered by soaking 400–4000 cm−1 at room temperature on a JASCO Model FT/IR
1′ and 2′ in H2O for an appropriate amount of time. 410 spectrophotometer. PXRD data were collected on
a
Dehydration and hydration occurred via crystalline-to-crystal- RIGAKU MultiFlex (50 kV/32 mA) or a RIGAKU SmartLab (40
line transformations. In other words, the frameworks were kV/100 mA) X-ray diffractometer using Cu-Kα radiation (λ =
stable even after removing the lattice H2O molecules. H2O 1.5406 Å) in the 2θ range of 3°–60° with a step width of 0.02°
adsorption and desorption studies of 1′ and 2′ showed that at room temperature. The lattice parameters were determined
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
by using the Le Bail refinement35 on the PXRD patterns with 13 P. Ramaswamy, N. E. Wong and G. K. H. Shimizu, Chem.
the Rigaku software package PDXL2.36,37 TG measurements
Soc. Rev., 2014, 43, 5913–5932.
were carried out on a Seiko Instruments SSC5200 Thermo 14 S. Horike, D. Umeyama and S. Kitagawa, Acc. Chem. Res.,
Analyzer with a heating rate of 10 K min−1 in the temperature
2013, 46, 2376–2384.
range of 293–450 K under an N2 atmosphere. H2O adsorption 15 T. Yamada, K. Otsubo, R. Makiura and H. Kitagawa, Chem.
and desorption isotherms were acquired by using BELSORP- Soc. Rev., 2013, 42, 6655–6669.
max volumetric adsorption equipment (BEL Japan Inc.). AC 16 M. Yoon, K. Suh, S. Natarajan and K. Kim, Angew. Chem.,
impedance measurements were carried out by using electro- Int. Ed., 2013, 52, 2688–2700.
chemical impedance spectroscopy on a Hioki 3532-50 LCR 17 D. Kustaryono, N. Kerbellec, G. Calvez, S. Freslon,
Hitester. The ν range was 50 Hz–5 MHz, and an oscillating
voltage of 0.1 V was used. Prior to the measurements, the pow-
C. Daiguebonne and O. Guillou, Cryst. Growth Des., 2010,
10, 775–781.
dered samples were pressed into small pellets (2 cm diameter 18 M. Camara, C. Daiguebonne, K. Boubekeur, T. Roisnel,
and 200 mm thickness) and inserted between Pt electrodes.
The conductivity was calculated by using the formula σ = (1/R)/
Y. Gérault, C. Baux, F. L. Dret and O. Guillou, C. R. Chim.,
2003, 6, 405–415.
(L/A), where R is the resistance, L is the thickness of the elec- 19 L. De Almeida, S. Grandjean, N. Vigier and F. Patisson,
trolyte between the electrodes, and A is the cross-sectional Eur. J. Inorg. Chem., 2012, 4986–4999.
contact area of the electrodes. Measurements were conducted 20 A. K. Galwey and M. E. Brown, Thermal Decomposition of
with a relative humidity of 100% at the desired temperature.
Ionic Solids: Chemical Properties and Reactivities of Ionic
Crystalline Phases, Elsevier, Amsterdam, 1999.
21 A. K. Jonscher, J. Mater. Sci., 1978, 13, 553–562.
22 A. K. Jonscher, Nature, 1977, 267, 673–679.
23 C. Cramer, S. Brunklaus, E. Ratai and Y. Gao, Phys. Rev.
Lett., 2003, 91, 266601.
Acknowledgements
This work was partially supported by JSPS KAKENHI (Grant-in-
Aid for Research Activity Start-up and Grant-in-Aid for Scientific
Research (C)) Grant Numbers 26888017 (R. I.) and 16K05735
(S. K.), respectively. The authors also appreciate the financial
support from Daiichi Kigenso Kagaku Kogyo Co., Ltd.
24 B. Louati, M. Gargouri, K. Guidara and T. Mhiri, J. Phys.
Chem. Solids, 2005, 66, 762–765.
25 N. Ortega, A. Kumar, P. Bhattacharya, S. B. Majumder and
R. S. Katiyar, Phys. Rev. B: Condens. Matter, 2008, 77, 014111.
26 C. Cuerva, J. A. Campo, M. Cano, J. Sanz, I. Sobrados,
V. Diez-Gómez, A. Rivera-Calzada and R. Schmidt, Inorg.
Chem., 2016, 55, 6995–7002.
27 H. Ōkawa, M. Sadakiyo, K. Otsubo, K. Yoneda, T. Yamada,
M. Ohba and H. Kitagawa, Inorg. Chem., 2015, 54, 8529–
8535.
28 M. Sadakiyo, T. Yamada, K. Honda, H. Matsui and
H. Kitagawa, J. Am. Chem. Soc., 2014, 136, 7701–7707.
29 S. Miyatsu, M. Kofu, A. Nagoe, T. Yamada, M. Sadakiyo,
T. Yamada, H. Kitagawa, M. Tyagi, V. García Sakai and
O. Yamamuro, Phys. Chem. Chem. Phys., 2014, 16, 17295–
17304.
30 S. Tominaka and A. K. Cheetham, RSC Adv., 2014, 4,
54382–54387.
31 H. Ōkawa, M. Sadakiyo, T. Yamada, M. Maesato, M. Ohba
and H. Kitagawa, J. Am. Chem. Soc., 2013, 135, 2256–
2262.
Notes and references
1 R. M. Dell and A. Hooper, Solid Electrolytes, Academic
Press, New York, 1978.
2 S. Chandra, Superionic Solids: Principles and Applications,
Amsterdam, North-Holland, 1981.
3 P. G. Bruce, Solid State Electrochemistry, Cambridge
University Press, Cambridge, 1995.
4 K.-D. Kreuer, S. J. Paddison, E. Spohr and M. Schuster,
Chem. Rev., 2004, 104, 4637–4678.
5 M. A. Hickner, H. Ghassemi, Y. S. Kim, B. R. Einsla and
J. E. McGrath, Chem. Rev., 2004, 104, 4587–4612.
6 P. Colomban, Chemistry of Solid State Materials 2. Proton
Conductors, Cambridge Univ. Press, Cambridge, U.K., 1992.
7 Z. Li, G. He, B. Zhang, Y. Cao, H. Wu, Z. Jiang and 32 M. Sadakiyo, H. Ōkawa, A. Shigematsu, M. Ohba,
Z. Tiantian, ACS Appl. Mater. Interfaces, 2014, 6, 9799–9807.
8 Y. Sone, P. Ekdunge and D. Simonsson, J. Electrochem. Soc.,
1996, 143, 1254–1259.
9 K. A. Mauritz and R. B. Moore, Chem. Rev., 2004, 104,
4535–4585.
T. Yamada and H. Kitagawa, J. Am. Chem. Soc., 2012, 134,
5472–5475.
33 M. Sadakiyo, T. Yamada and H. Kitagawa, J. Am. Chem.
Soc., 2009, 131, 9906–9907.
34 H. Ōkawa, A. Shigematsu, M. Sadakiyo, T. Miyagawa,
K. Yoneda, M. Ohba and H. Kitagawa, J. Am. Chem. Soc.,
2009, 131, 3144–3145.
10 L. Malavasi, C. A. J. Fisher and M. S. Islam, Chem. Soc. Rev.,
2010, 39, 4370–4387.
11 K.-D. Kreuer, Annu. Rev. Mater. Res., 2003, 33, 333–359.
35 H. M. Rietveld, J. Appl. Crystallogr., 1969, 2, 65–71.
12 P. Knauth and M. L. Di Vona, Solid State Proton 36 Rigaku Corporation, Rigaku J., 2012, 28, 29–30.
Conductors: Properties and Applications, in Fuel Cells, ed. 37 A. Sasaki, A. Himeda, H. Konaka and N. Muroyama, Rigaku
P. Knauth and M. L. Di Vona, Wiley, Chichester, 2012.
J., 2010, 26, 10–14.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.