ACS Publications. Most Trusted. Most Cited. Most Read
My Activity
CONTENT TYPES

AgNb7O18: An Ergodic Relaxor Ferroelectric

View Author Information
Department of Physics, University of Warwick, Gibbet Hill Road, Coventry, CV4 7AL, U.K.
Cite this: Inorg. Chem. 2014, 53, 17, 8941–8948
Publication Date (Web):August 12, 2014
https://doi.org/10.1021/ic5007346

Copyright © 2014 American Chemical Society. This publication is licensed under CC-BY.

  • Open Access

Article Views

1338

Altmetric

-

Citations

LEARN ABOUT THESE METRICS
PDF (5 MB)
Supporting Info (2)»

Abstract

AgNb7O18 is a relaxor ferroelectric with a Burns temperature of ∼490 K and an incipient transition to the nonergodic state. The short-range structure is shown by convergent-beam electron diffraction to have the polar space group Im2m, but refinements against powder X-ray diffraction find the long-range structure to have the centrosymmetric space group Immm. Relaxor behavior in AgNb7O18 appears to originate from the partial occupation of large interstices by Ag+ cations. Both cations and oxygen anions are displaced away from zones where NbO6 octahedra are edge-sharing.

Synopsis

AgNb7O18 is an ergodic relaxor ferroelectric at room temperature with an incipient transition to the nonergodic state. Electron diffraction confirms a locally polar symmetry, while X-ray diffraction perceives a nonpolar structure. All ions are repelled away from zones where NbO6 octahedra are edge-sharing.

Introduction

ARTICLE SECTIONS
Jump To

Ferroelectric ceramics are used in an enormous variety of technological applications, and the majority of these materials are based on the perovskite structure. This is for well-established reasons: the prototype structure is simple and particularly flexible, allowing it to accommodate most cations and thus access a wide range of properties which can then be “tuned” to specific applications by control of the ionic content. (1) The research into perovskite-structured materials has developed rapidly but at the price of a lack of research into other ferroelectric structures. The continued development of functional materials and devices requires the identification of crystal structures above and beyond perovskites that can be used to host physical properties and be structurally tuned for specific applications as diverse as temperature-independent dielectrics, (2) thermoelectrics, (3) and multiferroics. (4)
The Ag2O–Nb2O5 pseudobinary system is known to host an unusually large number of distinct crystal structures (5) with five confirmed structures and a number of additional, unconfirmed structures. Among the former are the perovskite-structured ferrielectric AgNbO3, (6) the ferroelectric Ag2Nb4O11, (7) which has the layered natrotantite structure containing edge-shared NbO7 polyhedra, (8) and the tungsten bronze-structured AgNb3O8. (9) The system is rich with functional materials: AgNbO3 is a compound studied for its role in lead-free piezoelectrics, (10) reentrant relaxors, (11) and photocatalysis. (12) Ag2Nb4O11-structured materials are finding value in the study of tunable band gap materials for photocatalyzing the electrolysis of water (13) and in investigations of the anomalously large piezoelectric coefficients of Ta2O5 thin films. (14, 15) Tungsten bronzes are studied for their ferroelectric properties (16) and for their potential use in thermoelectric applications. (17)
However, AgNb7O18 and AgNb13O33 are compounds about which little is currently known, although structural studies of the sodium analogues have been performed. The structure of NaNb7O18 was studied and refined by Marinder and Sundberg, (18) who found that the positions of diffraction peaks were consistent with an orthorhombic body-centered unit cell. The systematic absences limited the choice to four possible space groups: I222, I212121, Imm2, and Immm. The authors chose Immm, but without an explanation for rejecting the other space groups. Of the four symmetries, Immm is the only centrosymmetric option and precludes the existence of functional properties such as piezoelectricity or ferroelectricity.
The crystal structure of NaNb7O18 consists of units formed from 14 corner-sharing NbO6 octahedra in a perovskite-like arrangement. These units are linked together by edge-sharing of the octahedra at the perimeters, resulting in the appearance of tunnels with a large rectangular cross section. Marinder and Sundberg (18) proposed that half of the Na+ ions would be located in these tunnels, and the remainder would be distributed in the cuboctahedral interstices within the perovskite-like units. They recommended studying AgNb7O18 as the larger Ag+ ions would be easier to locate using diffraction methods. A recent study (9) fabricated crystals of AgNb7O18 and showed that the crystal structure is isostructural with that of NaNb7O18, but no refined structures exist for AgNb7O18 and its physical properties are not known.
Nb5+ appears in a significant proportion of ferroelectric oxides (3) in octahedral coordination where it has a tendency to displace off-center. (19, 20) If these Nb5+ displacements in the unit cell are correlated with a common component, the material becomes polarized. If the common component can be switched by an external field, the material is ferroelectric. The construction of AgNb7O18 from perovskite-like units makes it particularly likely to exhibit ferroelectricity. To investigate this possibility, ceramics have been synthesized and the structure and dielectric properties investigated.

Experimental Methods

ARTICLE SECTIONS
Jump To

AgNb7O18 was synthesized by a conventional solid state processing route. Ag2O (Alfa Aesar, 99%) and Nb2O5 (Alfa Aesar, 99%) were weighed out in stoichiometric amounts and ball-milled with 1 cm zirconia media in propan-2-ol for approximately 24 h. This mixture was dried at 70 °C and passed through a 250 μm sieve before being reacted for 10 h at 1050 °C. The reacted material was remilled for a further 24 h, before being dried and sieved as before. This resulted in a fine powder, which was compacted into 13 mm diameter pellets using a load of 1 ton. These pellets were buried in a bed of AgNb7O18 powder in an alumina crucible and sintered for 2 h at 1250 °C. Ceramic densities were calculated from measurements of pellet dimensions and masses.
Dielectric measurements were performed using an HP 4192A LF impedance analyzer. Electrodes were applied to ceramics using silver paint (RS Components), and the ceramics were cooled with liquid nitrogen at a rate of approximately 0.5 °C min–1 during data collection. High-temperature data were obtained with a rig located in a vertical tube furnace using a heating rate of 1 °C min–1 during data collection. A layer of sputtered gold was added before the silver paint to prevent the silver from reacting with the ceramic at high temperatures.
Transmission electron microscopy (TEM) was performed on a JEOL JEM-2100 electron microscope at 200 kV. AgNb7O18 pellets were thinned to ∼30 μm with abrasive paper and perforated with a Gatan PIPS ion mill before being carbon-coated. Digital large-angle convergent beam electron diffraction (D-LACBED) patterns were obtained using recently developed code made freely available. (21) Crystal simulations were performed using CaRine Crystallography v3.1, and electron diffraction simulations were performed using SingleCrystal v2.2.8.
Powder samples were prepared for X-ray diffraction (XRD) by grinding pellets in acetone with a pestle and mortar. XRD was performed using a Panalytical X’Pert Pro with a curved Ge Johansson monochromator excluding all X-rays except Cu Kα1. The scan was performed at 40 kV and 45 mA from 6 to 120° 2θ with a step size of 0.066° 2θ over a period of 17 h and with a Panalytical PIXcel detector. Rietveld refinements were performed with Topas Academic v4.1.

Results and Discussion

ARTICLE SECTIONS
Jump To

Permittivity

Ceramic pellets were found to have a mean density of 4.62 (4) g cm–3. Using the lattice parameters published by Rozier and Szajwaj (9) and assuming no Ag loss, this translates to pellets with densities that are 95.6 (7)% of the theoretical maximum, as defined by the unit cell density.
A structure formed from perovskite-like columns is likely to exhibit displacive phase transitions or functional properties. Dielectric permittivity data can indicate the presence of phase transitions and were collected for AgNb7O18 (Figure 1).

Figure 1

Figure 1. (a) Dielectric permittivity and (b) loss tangent for AgNb7O18 obtained across a range of frequencies at low temperature. The arrow marked “f” indicates the effect of increasing frequency.

The relative permittivity, εr, takes fairly modest values similar to those observed for Ag2Nb4O11. (22) However, the resistivity of AgNb7O18 is 30–40 MΩ·m recorded at 295 K and 1 kHz. These are surprisingly high values, given the presence of rectangular tunnels that might be expected to allow Ag+ to travel with relative ease through the crystal structure.
The permittivity data in Figure 1 show a broad, frequency-dependent peak: behavior characteristic of a relaxor ferroelectric. (23) A relaxor is a material that, like a classical ferroelectric, possesses an electric dipole, but the dipoles do not have long-range order. Instead, the structure is thought to consist of nanoscale clusters of parallel dipoles, called “polar nanoregions” (PNRs), with a size distribution that results in the frequency distribution observed in Figure 1. Below the Burns temperature, TB, the material is in the “ergodic” state, where these dipoles are fluctuating. (24) At lower temperatures, in the “nonergodic” state, the dipoles are frozen and an external field can be used to pole the material and make it behave as a classical ferroelectric. (25)
The Vogel–Fulcher model was originally used to describe the magnetic relaxation in spin-glass systems but has been shown to be applicable to relaxor materials in describing the temperature of permittivity maximum as a function of frequency. (24) The model is(1)where f0 is the “Debye frequency”, the theoretical maximum frequency for the vibration of PNRs, Ea is an activation energy, k is the Boltzmann constant, Tm is the temperature of the permittivity maximum at frequency f, and TF is the freezing temperature that indicates the transition between the ergodic and nonergodic states. By plotting ln(f) as a function of temperature (Figure 2), the Vogel–Fulcher model can be fitted to the data by a least-squares method and the material parameters can be derived.

Figure 2

Figure 2. Log of frequency as a function of the temperature of permittivity maximum encountered on cooling for AgNb7O18. Data obtained from Figure 1a are plotted, along with the Vogel–Fulcher model fitted to these data.

Fitting the Vogel–Fulcher model to the permittivity data gives the following parameters with 95% confidence bounds:
These parameters are useful in comparing the relaxation properties of this material with those of more common relaxors, such as tungsten bronze or perovskite-structured relaxors. It should be noted that the Vogel–Fulcher relationship applied to the maxima in permittivity with temperature may yield values of the parameters that have some dependency on temperature and frequency. (26) The parameters derived from the permittivity data are therefore accompanied by 95% confidence bounds to indicate a range of error from using this analysis.
The values of f0 and Ea are found to be similar in magnitude to materials based on Sr4Nd2Ti4Nb6O30, a tungsten bronze structured relaxor. (27) The relatively large value of Ea obtained here indicates that the polarization fluctuation of a PNR requires more energy than in these tungsten bronzes and in many perovskite-structured relaxors and is therefore slower. (24) The parameter TF of the Vogel–Fulcher model indicates the temperature below which all PNRs are frozen, allowing the material to be poled with an external field and made to behave in a manner similar to classical ferroelectrics. (25) The fit yields a negative value of TF which appears to indicate that some PNRs will continue to fluctuate at absolute zero. Some dielectrics exhibit increasing permittivity on cooling that can be extrapolated to a Curie temperature below 0 K. These are “incipient ferroelectrics” (e.g., Na1/2Bi1/2Cu3Ti4O12 (28)), so it is proposed that the negative value of TF indicates that AgNb7O18 has an incipient transition to the nonergodic state.
It is anticipated that isovalent ions such as K+, Na+, Li+, and Cu+ can be substituted into AgNb7O18 without a change in morphology. (29) Substitutions may freeze-in the dipoles and create a ferroelectric phase, as observed in (SrxBa1–x)4Nd2Ti4Nb6O30 tungsten bronzes. (27) It has also been shown that ferroelectricity can be induced in tungsten bronzes by reducing the size of cations on specific sites. (16) If the role of the Ag+ cation is to make this particular connectivity of the NbO6 framework energetically favorable, it should also be possible to substitute for half the number of divalent cations of similar size, such as Ca2+ or Ba2+. If substitutions result in TF increasing to a value above 0 K, the nonergodic state becomes accessible and an applied electric field on cooling can create a field-induced ferroelectric state.
While TF indicates the lowest temperature in which a relaxor is in an ergodic state, the upper temperature is indicated by TB, often a few hundred degrees above TF. (30) This can be experimentally determined by birefringence (31) but can also be obtained from the permittivity data over a temperature range above that where the dispersion is observed. According to the Curie–Weiss law, (32) above TB, the susceptibility, χ, is related to εr by the equation(2)but below TB, the permittivity deviates from this relationship. For AgNb7O18, the temperature range 300–550 K is sufficiently far above the region of dispersion to obtain a value of TB (Figure 3).

Figure 3

Figure 3. Susceptibility data for AgNb7O18 obtained at 1 MHz.

Figure 3 shows that TB = 490 (10) K, indicating that AgNb7O18 is an ergodic relaxor ferroelectric at room temperature. Stringer et al. (30) observed a linear relationship between the log values of TBTF and Ea for relaxors with perovskite structures and the values obtained for the nonperovskite relaxor AgNb7O18 closely fit this linear trend. This relaxor behavior is not consistent with the unit cells having the centrosymmetric space group Immm; neither is it consistent with the chiral, nonpolar space groups I222 or I212121. Instead it indicates that, locally, the room temperature structure is one of the three crystallographically distinct Imm2 variants (I2mm, Im2m, or Imm2) derived from the Immm parent structure.

Electron Diffraction

Selected area electron diffraction patterns (SADPs) were obtained from a number of zone axes. All were found to match simulations created using the NaNb7O18 structure determined by Marinder and Sundberg (18) (Figure 4).

Figure 4

Figure 4. SADPs matched with simulations for AgNb7O18 from zone axes (a and b) [001], (c and d) [100], and (e and f) [015]. The simulated patterns are indexed.

The simulated patterns provide a good match to the SADPs obtained from AgNb7O18. These are consistent with the space group Immm, but cannot be used to distinguish between the alternative space groups under consideration as they all produce reflections in the same locations. In order to identify the space group, use has been made of a new convergent-beam diffraction technique. D-LACBED involves digitally combining separate convergent-beam diffraction patterns from individual sets of diffracting planes over a large range of beam tilts obtained from an area with a diameter of ∼15 nm full width at half-maximum (fwhm). (21) The recombined diffraction patterns show the symmetry of the diffracted intensity from individual sets of planes with far greater clarity than can be obtained with the standard convergent-beam technique. The symmetry of the resultant diffraction group is related to the crystal point group, and the techniques set out by Buxton et al. (33) are used to determine the point groups that can give rise to the diffraction symmetry.

Figure 5

Figure 5. D-LACBED pattern obtained from [001] zone axis of AgNb7O18 (BF = bright field; m = mirror).

The bright field (BF) pattern in the center has both vertical and horizontal mirror planes, giving it symmetry 2mm. The whole pattern has a vertical mirror plane but not a horizontal mirror plane, shown most clearly by comparing the ±0 10 0 patterns. Opposing dark field patterns with ±g vectors (e.g., 3̅5̅0 and 350) are not equivalent when translated onto each other, demonstrating that the crystal structure is noncentrosymmetric, at least on the scale of the electron beam. The projection diffraction group of the pattern is therefore m1R. The absence of higher-order Laue zones (HOLZ) from this zone axis means that this projection diffraction group may possibly result from m or mR diffraction groups as well. However, this pattern has been obtained from a ⟨100⟩ zone axis of an orthorhombic crystal, which can only correspond to point group mm2, (33) as predicted from the permittivity data. The absence of a horizontal mirror plane in Figure 5 indicates that there is no mirror plane perpendicular to the y direction, making it the polar axis. The space group must therefore be Im2m in the configuration used by Marinder and Sundberg. (18) The Im2m structure has the same shape and size as the Immm structure, but the absence of the mirror plane perpendicular to y allows cation displacements along ±y, leading to local breaking of the center of symmetry and the formation of two degenerate structures. A larger version of Figure 5, incorporating diffraction information from a far greater number of crystal planes, is in the Supporting Information.

X-ray Diffraction

Rietveld refinements were performed against XRD obtained from the powdered ceramic (Figure 6) using the NaNb7O18 structure published by Marinder and Sundberg (18) as the initial model for a refinement in space group Immm. This model only accounted for half of the Ag+ cations. In their paper, the authors stated that “several perovskite-type sites are available in the structure”, referring to the cuboctahedral interstices similar to those where Ag+ would normally be located in perovskite-structured materials. Ag+ ions were added to these sites in the model and the occupancies were allowed to refine. Thermal parameters were refined with one value for each element. The thermal parameters of the five Nb5+ cations were subsequently allowed to take different values, but when this was extended to Ag+ or O2– species, several values fell below zero so the parameters were kept specific to the species, not the positions. The reflection profile was a Thompson–Cox–Hastings pseudo-Voigt function with an additional modification for anisotropic strain. (34) Once the refinement in space group Immm was completed, the results were used to generate a model for the structure with space group Im2m. However, when refining this lower symmetry model, very little improvement in the fit parameters was observed and the uncertainty in the atomic coordinates became much larger. The results of the Immm refinement are in Tables 1 and 2, and the results of the Im2m refinement are in the Supporting Information.

Figure 6

Figure 6. Full range of XRD data for AgNb7O18 with the calculated profile, positions of Bragg peaks, and the difference between the data plotted. The inset shows the data range that contains the three most intense peaks. The intensity is plotted on a linear scale.

Table 1. AgNb7O18 Refinement Parameters
crystal systemorthorhombic
space groupImmm (71)
a (Å)14.33158(15)
b (Å)26.15102(31)
c (Å)3.83624(3)
V (Å3)1437.766(26)
density (g cm–3)4.839(5)
refined parameters68
Rexp (%)19.859
Rp (%)16.383
Rwp (%)23.744
RBragg (%)4.825
χ21.196
Table 2. Unit Cell Contents of AgNb7O18, Refined in Space Group Immm
atomic sitesite symxyzfractional occupationUiso2)
Ag12a0000.946(7)0.0169(17)
Ag22b1/2000.320(6)0.0169(17)
Ag38n0.3075(13)0.1053(9)00.120(3)0.0169(17)
Ag44h00.2872(10)0.50.139(5)0.0169(17)
Nb18n0.11970(17)0.10269(12)0.510.0020(8)
Nb28n0.30630(21)0.20720(12)0.510.0037(9)
Nb34f0.31413(31)00.510.0141(15)
Nb44g00.18972(16)010.0022(11)
Nb54g00.39491(19)010.0128(12)
O18n0.7005(11)0.5550(7)010.0080(13)
O28n0.9027(11)0.5577(7)010.0080(13)
O38n0.7104(11)0.6598(6)010.0080(13)
O48n0.9050(11)0.6616(6)010.0080(13)
O58n0.9020(10)0.7666(6)010.0080(13)
O68n0.1006(10)0.1192(6)010.0080(13)
O78n0.2924(11)0.2237(7)010.0080(13)
O84h00.0640(9)0.510.0080(13)
O94h00.1627(9)0.510.0080(13)
O104h00.3887(10)0.510.0080(13)
O114e0.3037(16)0010.0080(13)
Weak peaks at 22.6, 29.8, and 32.3° 2θ that are not accounted for by the model correspond closely to strong reflections expected from AgNb3O8. (9) The proportion of AgNb3O8 refines to less than 3% of the total volume.
The collected results require careful interpretation to resolve the discrepancy between the different symmetries observed by X-ray and electron diffraction. In a relaxor in the ergodic state, the polar axis of the unit cell does not have long-range correlation across the sample. However, XRD has a long correlation length and is collected from a large sample with grains of size ∼1 μm over a period of many hours. It therefore perceives a time-averaged structure which is nonpolar, in common with relaxor materials such as the perovskite PbMg1/3Nb2/3O3, where X-ray and neutron diffraction perceive no deviation from the cubic prototype even at temperatures below TF; ∼400 K below TB. (31, 35)
As the PNRs are expected to be of the order of 10 nm, with rapid fluctuations in polarization direction, (36) electron diffraction from a relaxor material would normally be expected to also display nonpolar symmetry. However, it is important to consider that there will always be a distribution in PNR size, in which the largest regions fluctuate more slowly. Viehland et al. (24) showed that the frequency of the fluctuations depends very strongly on PNR size; their calculations show regions of size 45 and 55 Å with fluctuation frequencies that are different by 4 orders of magnitude. Moreover, several recent studies have shown that a significant fraction of PNRs are static in ergodic relaxors, far above TF. (37-39) Although our electron diffraction information was collected from an area the size of the electron beam (∼15 nm diameter fwhm), this was located within a large region that was free from defects. We conclude that the frequency of polarization fluctuation in this region was sufficiently slow that the D-LACBED technique perceived a structure that retained some polar character.
The AgNb7O18 unit cell derived from the high-symmetry refinement is depicted in Figure 7. To a first approximation, the unit cell has the dimensions 5ac/√2 × 9ac/√2 × ac, where ac is the lattice parameter of a pseudocubic perovskite unit cell. The refined lattice parameters give this unit cell the dimensions 4.05 Å × 4.11 Å × 3.84 Å, indicating that the octahedra are slightly extended along the polar direction and significantly compressed parallel to the c direction. The building block defined by the corner-shared NbO6 octahedra and resembling the perovskite structure is depicted in Figure 8.

Figure 7

Figure 7. Refined AgNb7O18 unit cell, depicted as a network of NbO6 octahedra. The unit cell is indicated by the dashed border. Colors indicate the NbO6 octahedra that are related by symmetry. Arrows on octahedra indicate the direction and relative magnitude of the Nb5+ displacements from ideal locations. Labels for Ag atoms and NbO6 octahedra correspond to their numbering scheme in Table 1. The darkness of the Ag+ cations indicates the occupation of those sites (light = lowest occupation; dark = greatest occupation), and arrows next to Ag+ cations indicate the direction and relative magnitude of their displacements from ideal (undistorted) locations. In order for cation displacements to be clear in the figure, they have been exaggerated by a factor of approximately 2.

Figure 8

Figure 8. “Perovskite-like” building block of AgNb7O18. See Figure 7 caption for explanation of details.

Rietveld refinement has been used to establish the locations of the Ag+ cations for the first time. There are four distinct Ag+ sites which, if they were all fully occupied, would yield a material with the formula Ag4Nb7O18, clearly a problem for maintaining charge neutrality. However, the occupancies of the Ag sites were refined to give a material with the empirical formula Ag1.012(18)Nb7O18. This confirms that no significant quantities of Ag volatilized during synthesis and that Ag is present as Ag+. The refined occupancies show that the Ag+ has a strong preference for the largest interstice in the structure—the rectangular tunnel located at the center and corners of the unit cell—and a weaker preference for the cuboctahedral “perovskite-like sites”. It has long been noted that relaxor behavior in perovskites and tungsten bronzes is associated with mixed species sharing a single crystallographic site, resulting in local heterogeneity (e.g., Viehland et al. (24)). In AgNb7O18, the heterogeneity is introduced by the partial occupation of the Ag sites.
In order to clarify and interpret the cation displacements in the structure, their magnitudes and directions have been calculated relative to both their “ideal” positions and the centers of the oxygen octahedra. Ideal positions are found by placing atoms on the nodes of a grid with the same dimensions as the unit cell, where a is divided into 10 parts, b is divided into 18 parts, and c is divided into 2 parts of equal length. The data are given in Table 3, and the cation displacements from the ideal positions can be seen in Figures 7 and 8.
Table 3. Cation-Related Data Obtained from Structural Refinement
cationdisplacement from ideal position (Å)direction of displacement (clockwise)displacement from oxygen polyhedron center (Å)direction of displacement (clockwise)
Ag100
Ag200
Ag30.186(22)–54.4(2)° from [100]0.144(23)–62.0(3)° from [100]
Ag40.246(24)[010]0.119(27)[010]
Nb10.358(3)–37.954(2)° from [100]0.377(7)–47.67(2)° from [100]
Nb20.403(3)–77.06(2)° from [100]0.306(8)–77.36(10)° from [100]
Nb30.203(4)[100]0.170(9)[100]
Nb40.603(4)[010]0.469(9)[010]
Nb50.157(5)[010]0.133(10)[010]
Figure 8 shows that the structure of AgNb7O18 can be viewed as columns of perovskite-like material connected together by edge-sharing of the peripheral octahedra, which experience significant distortion. Statistically, in each layer of each column, one Ag+ is located in one of the cuboctahedral sites, displaying a preference for the central site. The cation displacements are all directed toward the center of the perovskite-like unit, and the displacements increase in magnitude the further the cations are from the center of the unit. This indicates that the cations are being repelled from the edge-sharing zones, since if they were attracted to the center of the perovskite unit, the displacements would be largest for those cations closest to the center of the unit. Additionally, with the sole exception of Nb1, the displacements of all cations from their ideal positions are greater than their displacements from the center of their respective oxygen coordination polyhedra. This results from the oxygen framework moving away from the ideal position in the same direction as the cations. It is proposed that this repulsion is in mitigation of the locally high density in the edge-sharing zones. This suggests that the structure may experience high local stresses and indicates that this connectivity of octahedra is not particularly stable.

Conclusions

ARTICLE SECTIONS
Jump To

AgNb7O18 has been fabricated as dense ceramics without Ag loss. Permittivity data show a frequency-dependent response that is a classic indicator of a relaxor ferroelectric. Fitting the Volger–Fulcher equation to the data reveals an incipient transition to the nonergodic state, and high-temperature permittivity data indicate that the Burns temperature is ∼490 K. The local structure revealed by electron diffraction has the polar space group Im2m, but X-ray diffraction perceives a structure with the centrosymmetric space group Immm, apparently isomorphous with NaNb7O18. Rietveld refinements show that Ag+ cations partially occupy all cuboctahedral sites but occupy the largest, rectangular sites almost completely. Both anions and cations appear to be repelled away from zones of edge-sharing octahedra.

Supporting Information

ARTICLE SECTIONS
Jump To

Crystallographic parameters for AgNb7O18 in CIF format. Complete D-LACBED data set (Figure S1) and results from Rietveld refinement in space group Im2m (Tables S1 and S2). This material is available free of charge via the Internet at http://pubs.acs.org.

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

ARTICLE SECTIONS
Jump To

  • Corresponding Author
    • David I. Woodward - Department of Physics, University of Warwick, Gibbet Hill Road, Coventry, CV4 7AL, U.K. Email: [email protected]
  • Author
    • Richard Beanland - Department of Physics, University of Warwick, Gibbet Hill Road, Coventry, CV4 7AL, U.K.
  • Notes
    The authors declare no competing financial interest.

Acknowledgment

ARTICLE SECTIONS
Jump To

The PANalytical MPD diffractometer used in this research was obtained though the Science City Energy Futures Project: Hydrogen Energy, with support from Advantage West Midlands (AWM) and partly funded by the European Regional Development Fund (ERDF). R.B. was supported by EPSRC Grant EP/J009229/1. For crystallographic assistance and encouragement, it is a pleasure to thank Dean Keeble, Wook Jo, Antonio Feteira, and Ljuba Schmitt.

References

ARTICLE SECTIONS
Jump To

This article references 39 other publications.

  1. 1
    Mitchell, R. H. Perovskites: Modern and Ancient; Almaz Press Inc.: Thunder Bay, ON, Canada, 2002.
  2. 2
    Cava, R. J.; Krajewski, J. J.; Roth, R. S. Mater. Res. Bull. 1998, 33, 527 532
  3. 3
    Lee, S.; Bock, J. A.; Trolier-McKinstry, S.; Randall, C. A. J. Eur. Ceram. Soc. 2012, 32, 3971 3988
  4. 4
    Keeney, L.; Maity, T.; Schmidt, M.; Amann, A.; Deepak, N.; Petkov, N.; Roy, S.; Pemble, M. E.; Whatmore, R. W. J. Am. Ceram. Soc. 2013, 96, 2339 2357
  5. 5
    Brusset, H.; Gillier-Pandraud, H.; Belle, J.-P. Bull. Soc. Chim. Fr. 1967, 7, 2276 2283
  6. 6
    Yashima, M.; Matsuyama, S.; Sano, R.; Itoh, M.; Tsuda, K.; Fu, D. Chem. Mater. 2011, 23, 1643 1645
  7. 7
    Woodward, D. I.; Thomas, P. A. Appl. Phys. Lett. 2011, 98, 132904
  8. 8
    Masó, N.; Woodward, D. I.; Thomas, P. A.; Várez, A.; West, A. R. J. Mater. Chem. 2011, 21, 2715 2722
  9. 9
    Rozier, P.; Szajwaj, O. J. Solid State Chem. 2008, 181, 228 234
  10. 10
    Fu, D.; Endo, M.; Taniguchi, H.; Taniyama, T.; Koshihara, S.-Y.; Itoh, M. Appl. Phys. Lett. 2008, 92, 172905
  11. 11
    Lei, C.; Ye, Z.-G. J. Phys.: Condens. Matter 2008, 20, 232201
  12. 12
    Kato, H.; Kobayashi, H.; Kudo, A. J. Phys. Chem. B 2002, 106, 12441 12447
  13. 13
    Palasyuk, O.; Maggard, P. A. J. Solid State Chem. 2012, 191, 263 270
  14. 14
    Audier, M.; Chenevier, B.; Roussel, H.; Vincent, L.; Peña, A.; Lintanf Salaün, A. J. Solid State Chem. 2011, 184, 2033 2040
  15. 15
    Le Gallic, M.; Roussel, H. J. Solid State Chem. 2013, 200, 143 149
  16. 16
    Stennett, M. C.; Reaney, I. M.; Miles, G. C.; Woodward, D. I.; West, A. R.; Kirk, C. A.; Levin, I. J. Appl. Phys. 2007, 101, 104114
  17. 17
    Lee, S.; Wilke, R. H. T.; Trolier-McKinstry, S.; Zhang, S.; Randall, C. A. Appl. Phys. Lett. 2010, 96, 031910
  18. 18
    Marinder, B.-O.; Sundberg, M. Acta Crystallogr., Sect. B 1984, 40, 82 86
  19. 19
    Cohen, R. E.; Krakauer, K. Ferroelectrics 1992, 136, 65 84
  20. 20
    Hill, N. A. J. Phys. Chem. B 2000, 104, 6694 6709
  21. 21
    Beanland, R.; Thomas, P.; Woodward, D. I.; Thomas, P. A.; Roemer, R. Acta Crystallogr., Sect. A 2013, 69, 427 434
  22. 22
    Masó, N.; West, A. R. J. Mater. Chem. 2010, 20, 2082 2084
  23. 23
    Cross, L. E. Ferroelectrics 1987, 76, 241
  24. 24
    Viehland, D.; Jang, S. J.; Cross, L. E.; Wuttig, M. J. Appl. Phys. 1990, 68, 2916 2912
  25. 25
    Bokov, A. A.; Ye, Z.-G. J. Mater. Sci. 2006, 41, 31 52
  26. 26
    Tagantsev, A. K. Phys. Rev. Lett. 1994, 72, 1100 1103
  27. 27
    Zhu, X. L.; Wu, S. Y.; Chen, X. M. Appl. Phys. Lett. 2007, 91, 162906
  28. 28
    Ferrarelli, M. C.; Nuzhnyy, D.; Sinclair, D. C.; Kamba, S. Phys. Rev. B 2010, 81, 224112
  29. 29
    Palasyuk, O.; Palasyuk, A.; Maggard, P. A. Inorg. Chem. 2010, 49, 10571 10578
  30. 30
    Stringer, C. J.; Shrout, T. R.; Randall, C. A. J. Appl. Phys. 2007, 101, 054107
  31. 31
    Burns, G.; Dacol, F. H. Solid State Commun. 1983, 48, 853 856
  32. 32
    Viehland, D.; Jang, S. J.; Cross, L. E.; Wuttig, M. Phys. Rev. B 1992, 46, 8003 8006
  33. 33
    Buxton, B. F.; Eades, J. A.; Steeds, J. W.; Rackham, G. M. Philos. Trans. R. Soc., A 1976, 281, 171 194
  34. 34
    Stephens, P. W. J. Appl. Crystallogr. 1999, 32, 281 289
  35. 35
    Bonneau, P.; Garnier, P.; Calvavrin, G.; Husson, E.; Gavarri, J. R.; Hewat, A. W.; Morell, A. J. Solid State Chem. 1991, 91, 350 361
  36. 36
    Bokov, A. A.; Ye, Z.-G. J. Adv. Dielectr. 2012, 2, 1241010
  37. 37
    Bokov, A. A.; Rodriguez, B. J.; Zhao, X.; Ko, J.-H.; Jesse, S.; Long, X.; Qu, W.; Kim, T. H.; Budai, J. D.; Morozovska, A. N.; Kojima, S.; Tan, X.; Kalinin, S. V.; Ye, Z.-G. Z. Kristallogr. 2011, 226, 99 107
  38. 38
    Kholkin, A.; Morozovska, A. N.; Kiselev, D.; Bdikin, I.; Rodriguez, B. J.; Wu, P.; Bokov, A.; Ye, Z.-G.; Dkhil, B.; Chen, L.-Q.; Kosec, M.; Kalinin, S. V. Adv. Funct. Mater. 2011, 21, 1977 1987
  39. 39
    Xie, L.; Li, Y. L.; Yu, R.; Cheng, Z. Y.; Wei, X. Y.; Yao, X.; Jia, C. L.; Urban, K.; Bokov, A. A.; Ye, Z.-G.; Zhu, J. Phys. Rev. B 2012, 85, 014118

Cited By

ARTICLE SECTIONS
Jump To

This article is cited by 8 publications.

  1. Ye Tian, Panpan Song, Giuseppe Viola, Jindou Shi, Jing Li, Li Jin, Qingyuan Hu, Yonghao Xu, Wanyin Ge, Zhongna Yan, Dou Zhang, Nadezda V. Tarakina, Isaac Abrahams, Xiaoyong Wei, Haixue Yan. Silver niobate perovskites: structure, properties and multifunctional applications. Journal of Materials Chemistry A 2022, 10 (28) , 14747-14787. https://doi.org/10.1039/D2TA00905F
  2. R.L. Nayak, Yajun Zhang, Sadhwi S. Dash, M.P.K. Sahoo. Intermediate ferroelectric phase driven ferroelectric-relaxor crossover and superior storage properties in phase boundary engineered BCZT ceramics. Ceramics International 2022, 48 (8) , 10803-10816. https://doi.org/10.1016/j.ceramint.2021.12.296
  3. Yinjie Qian, Jian Wang, Sana Ullah Asif, Jiyang Xie, Wanbiao Hu. Origins of dielectric relaxations in AgNb7O18 ceramic. Ceramics International 2020, 46 (14) , 23021-23026. https://doi.org/10.1016/j.ceramint.2020.06.078
  4. Ranjit Pattanayak, Simanchalo Panigrahi. Coupling of two types of polar regions in Na0.5Bi0.5TiO3: An impedance spectroscopic study. Chemical Physics Letters 2020, 741 , 137087. https://doi.org/10.1016/j.cplett.2020.137087
  5. M. Jeyakanthan, Uma Subramanian, R. B. Tangsali, Roshan Jose, K. Venkata Saravanan. Relaxor like colossal dielectric constant in CoWO4 and CoWO4/PbWO4 nanocomposites. Journal of Materials Science: Materials in Electronics 2019, 30 (15) , 14657-14668. https://doi.org/10.1007/s10854-019-01837-5
  6. XuanXuan Liu, Chuanxiang Qin, Lei Cao, Yongyi Feng, Yanlin Huang, Lin Qin, Hyo Jin Seo. A silver niobate photocatalyst AgNb7O18 with perovskite-like structure. Journal of Alloys and Compounds 2017, 724 , 381-388. https://doi.org/10.1016/j.jallcom.2017.07.048
  7. S. Ramesh, Jerald V. Ramaclus, Edgar Mosquera, B. B. Das. Sol–gel synthesis, structural, optical and magnetic characterization of Ag 3(2+x) Pr x Nb 4−x O 11+δ (0.0 ≤ x ≤ 1.0) nanoparticles. RSC Advances 2016, 6 (8) , 6336-6341. https://doi.org/10.1039/C5RA24925B
  8. David I. Woodward, Richard Beanland. ChemInform Abstract: AgNb 7 O 18 : An Ergodic Relaxor Ferroelectric.. ChemInform 2014, 45 (42) https://doi.org/10.1002/chin.201442005
  • Abstract

    Figure 1

    Figure 1. (a) Dielectric permittivity and (b) loss tangent for AgNb7O18 obtained across a range of frequencies at low temperature. The arrow marked “f” indicates the effect of increasing frequency.

    Figure 2

    Figure 2. Log of frequency as a function of the temperature of permittivity maximum encountered on cooling for AgNb7O18. Data obtained from Figure 1a are plotted, along with the Vogel–Fulcher model fitted to these data.

    Figure 3

    Figure 3. Susceptibility data for AgNb7O18 obtained at 1 MHz.

    Figure 4

    Figure 4. SADPs matched with simulations for AgNb7O18 from zone axes (a and b) [001], (c and d) [100], and (e and f) [015]. The simulated patterns are indexed.

    Figure 5

    Figure 5. D-LACBED pattern obtained from [001] zone axis of AgNb7O18 (BF = bright field; m = mirror).

    Figure 6

    Figure 6. Full range of XRD data for AgNb7O18 with the calculated profile, positions of Bragg peaks, and the difference between the data plotted. The inset shows the data range that contains the three most intense peaks. The intensity is plotted on a linear scale.

    Figure 7

    Figure 7. Refined AgNb7O18 unit cell, depicted as a network of NbO6 octahedra. The unit cell is indicated by the dashed border. Colors indicate the NbO6 octahedra that are related by symmetry. Arrows on octahedra indicate the direction and relative magnitude of the Nb5+ displacements from ideal locations. Labels for Ag atoms and NbO6 octahedra correspond to their numbering scheme in Table 1. The darkness of the Ag+ cations indicates the occupation of those sites (light = lowest occupation; dark = greatest occupation), and arrows next to Ag+ cations indicate the direction and relative magnitude of their displacements from ideal (undistorted) locations. In order for cation displacements to be clear in the figure, they have been exaggerated by a factor of approximately 2.

    Figure 8

    Figure 8. “Perovskite-like” building block of AgNb7O18. See Figure 7 caption for explanation of details.

  • References

    ARTICLE SECTIONS
    Jump To

    This article references 39 other publications.

    1. 1
      Mitchell, R. H. Perovskites: Modern and Ancient; Almaz Press Inc.: Thunder Bay, ON, Canada, 2002.
    2. 2
      Cava, R. J.; Krajewski, J. J.; Roth, R. S. Mater. Res. Bull. 1998, 33, 527 532
    3. 3
      Lee, S.; Bock, J. A.; Trolier-McKinstry, S.; Randall, C. A. J. Eur. Ceram. Soc. 2012, 32, 3971 3988
    4. 4
      Keeney, L.; Maity, T.; Schmidt, M.; Amann, A.; Deepak, N.; Petkov, N.; Roy, S.; Pemble, M. E.; Whatmore, R. W. J. Am. Ceram. Soc. 2013, 96, 2339 2357
    5. 5
      Brusset, H.; Gillier-Pandraud, H.; Belle, J.-P. Bull. Soc. Chim. Fr. 1967, 7, 2276 2283
    6. 6
      Yashima, M.; Matsuyama, S.; Sano, R.; Itoh, M.; Tsuda, K.; Fu, D. Chem. Mater. 2011, 23, 1643 1645
    7. 7
      Woodward, D. I.; Thomas, P. A. Appl. Phys. Lett. 2011, 98, 132904
    8. 8
      Masó, N.; Woodward, D. I.; Thomas, P. A.; Várez, A.; West, A. R. J. Mater. Chem. 2011, 21, 2715 2722
    9. 9
      Rozier, P.; Szajwaj, O. J. Solid State Chem. 2008, 181, 228 234
    10. 10
      Fu, D.; Endo, M.; Taniguchi, H.; Taniyama, T.; Koshihara, S.-Y.; Itoh, M. Appl. Phys. Lett. 2008, 92, 172905
    11. 11
      Lei, C.; Ye, Z.-G. J. Phys.: Condens. Matter 2008, 20, 232201
    12. 12
      Kato, H.; Kobayashi, H.; Kudo, A. J. Phys. Chem. B 2002, 106, 12441 12447
    13. 13
      Palasyuk, O.; Maggard, P. A. J. Solid State Chem. 2012, 191, 263 270
    14. 14
      Audier, M.; Chenevier, B.; Roussel, H.; Vincent, L.; Peña, A.; Lintanf Salaün, A. J. Solid State Chem. 2011, 184, 2033 2040
    15. 15
      Le Gallic, M.; Roussel, H. J. Solid State Chem. 2013, 200, 143 149
    16. 16
      Stennett, M. C.; Reaney, I. M.; Miles, G. C.; Woodward, D. I.; West, A. R.; Kirk, C. A.; Levin, I. J. Appl. Phys. 2007, 101, 104114
    17. 17
      Lee, S.; Wilke, R. H. T.; Trolier-McKinstry, S.; Zhang, S.; Randall, C. A. Appl. Phys. Lett. 2010, 96, 031910
    18. 18
      Marinder, B.-O.; Sundberg, M. Acta Crystallogr., Sect. B 1984, 40, 82 86
    19. 19
      Cohen, R. E.; Krakauer, K. Ferroelectrics 1992, 136, 65 84
    20. 20
      Hill, N. A. J. Phys. Chem. B 2000, 104, 6694 6709
    21. 21
      Beanland, R.; Thomas, P.; Woodward, D. I.; Thomas, P. A.; Roemer, R. Acta Crystallogr., Sect. A 2013, 69, 427 434
    22. 22
      Masó, N.; West, A. R. J. Mater. Chem. 2010, 20, 2082 2084
    23. 23
      Cross, L. E. Ferroelectrics 1987, 76, 241
    24. 24
      Viehland, D.; Jang, S. J.; Cross, L. E.; Wuttig, M. J. Appl. Phys. 1990, 68, 2916 2912
    25. 25
      Bokov, A. A.; Ye, Z.-G. J. Mater. Sci. 2006, 41, 31 52
    26. 26
      Tagantsev, A. K. Phys. Rev. Lett. 1994, 72, 1100 1103
    27. 27
      Zhu, X. L.; Wu, S. Y.; Chen, X. M. Appl. Phys. Lett. 2007, 91, 162906
    28. 28
      Ferrarelli, M. C.; Nuzhnyy, D.; Sinclair, D. C.; Kamba, S. Phys. Rev. B 2010, 81, 224112
    29. 29
      Palasyuk, O.; Palasyuk, A.; Maggard, P. A. Inorg. Chem. 2010, 49, 10571 10578
    30. 30
      Stringer, C. J.; Shrout, T. R.; Randall, C. A. J. Appl. Phys. 2007, 101, 054107
    31. 31
      Burns, G.; Dacol, F. H. Solid State Commun. 1983, 48, 853 856
    32. 32
      Viehland, D.; Jang, S. J.; Cross, L. E.; Wuttig, M. Phys. Rev. B 1992, 46, 8003 8006
    33. 33
      Buxton, B. F.; Eades, J. A.; Steeds, J. W.; Rackham, G. M. Philos. Trans. R. Soc., A 1976, 281, 171 194
    34. 34
      Stephens, P. W. J. Appl. Crystallogr. 1999, 32, 281 289
    35. 35
      Bonneau, P.; Garnier, P.; Calvavrin, G.; Husson, E.; Gavarri, J. R.; Hewat, A. W.; Morell, A. J. Solid State Chem. 1991, 91, 350 361
    36. 36
      Bokov, A. A.; Ye, Z.-G. J. Adv. Dielectr. 2012, 2, 1241010
    37. 37
      Bokov, A. A.; Rodriguez, B. J.; Zhao, X.; Ko, J.-H.; Jesse, S.; Long, X.; Qu, W.; Kim, T. H.; Budai, J. D.; Morozovska, A. N.; Kojima, S.; Tan, X.; Kalinin, S. V.; Ye, Z.-G. Z. Kristallogr. 2011, 226, 99 107
    38. 38
      Kholkin, A.; Morozovska, A. N.; Kiselev, D.; Bdikin, I.; Rodriguez, B. J.; Wu, P.; Bokov, A.; Ye, Z.-G.; Dkhil, B.; Chen, L.-Q.; Kosec, M.; Kalinin, S. V. Adv. Funct. Mater. 2011, 21, 1977 1987
    39. 39
      Xie, L.; Li, Y. L.; Yu, R.; Cheng, Z. Y.; Wei, X. Y.; Yao, X.; Jia, C. L.; Urban, K.; Bokov, A. A.; Ye, Z.-G.; Zhu, J. Phys. Rev. B 2012, 85, 014118
  • Supporting Information

    Supporting Information

    ARTICLE SECTIONS
    Jump To

    Crystallographic parameters for AgNb7O18 in CIF format. Complete D-LACBED data set (Figure S1) and results from Rietveld refinement in space group Im2m (Tables S1 and S2). This material is available free of charge via the Internet at http://pubs.acs.org.


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

You’ve supercharged your research process with ACS and Mendeley!

STEP 1:
Click to create an ACS ID

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

MENDELEY PAIRING EXPIRED
Your Mendeley pairing has expired. Please reconnect