Dielectric Response and Specific Heat Studies of ... - IEEE Xplore

IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control ,

vol. 60, no. 8,

August

2013

1603

Dielectric Response and Specific Heat Studies of Cd2Nb2O7 Ceramics Obtained From Mechano-Synthesized Nanopowders Maria Połomska, Bożena Hilczer, Ewa Markiewicz, Zbigniew Trybuła, Bartłomiej Andrzejewski, Izabela Szafraniak-Wiza, and Adam Pietraszko Abstract—Cd2Nb2O7 is still an interesting ferroelectric material because of its high permittivity value at helium temperatures and a variety of dielectric relaxation processes, the origin of which is still puzzling. We prepared hot-pressed ceramics, with grain sizes from 100 to 150 nm, obtained from Cd2Nb2O7 nanopowders synthesized by high-energy milling of CdO and Nb2O5 and studied their dielectric response and thermal properties. The nanoceramics were characterized by X-ray diffraction and their dielectric properties were measured at temperatures from 4K to 575K. Dielectric response of the nanoceramics was found to consist of a huge anomaly at ~150K with complex dielectric absorption and three relaxation processes apparent in frequency and temperature dependences of the imaginary part of permittivity in the temperature range from 18K to 145K. The anomaly at ~150K is related to overlapping contributions from the Curie point (shifted downward because of the size effect) and a dielectric relaxation process. The behavior of three relaxation modes observed at temperatures below 145K is discussed, based on the model proposed by Malcherek of polar nanoregions in the orthorhombic phase of Cd2Nb2O7 and the theory of dielectric response of ferroelectric relaxors by Bokov and Ye.

I. Introduction

C

admium pyroniobate (CNO; Cd2Nb2O7), the ferroelectric properties of which have been reported 60 years ago [1]–[4], appears up to now an intriguing material because of high permittivity value at low temperatures, variety of dielectric relaxation processes and extremely interesting sequence of phase transitions the nature of which is still puzzling. Cook and Jaffe reported on the dielectric anomaly at the Curie temperature TC = 170K [1], [2], whereas Shirane and Pepinsky [3] and Hulm [4] located the Curie point between 182K and 188K, depending on the firing temperature of the ceramics. Another, but small, dielectric anomaly suggesting a phase transition was observed by Shirane and Pepinsky at 85K and also by Hulm at 80K. A remarkable lowering of the Curie point, Manuscript received January 31, 2013; accepted May 13, 2013. The authors thank the COST Action MP0904 for support. The work has been also supported by National Science Centre (project No. N N507 229040). M. Połomska, B. Hilczer, E. Markiewicz, Z. Trybuła, and B. Andrzejewski are with the Institute of Molecular Physics, Polish Academy of Sciences, Poznań, Poland ([email protected]). I. Szafraniak-Wiza is with the Institute of Material Science and Engineering, Poznań University of Technology, Poznań, Poland. A. Pietraszko is with the Institute of Low Temperature and Structure Research, Polish Academy of Sciences, Wrocław, Poland. DOI http://dx.doi.org/10.1109/TUFFC.2013.2741 0885–3010/$25.00

to 153K on cooling and to 161K on heating, has been reported for CNO single crystals by Jona et al. [5]; the shift in the Curie temperature has been ascribed by the authors to crystal impurities. Dielectric properties of CNO have been studied repeatedly in many labs and the results of detailed measurements showed that the high-temperature phase transition is complex and, moreover, many dielectric relaxation processes have been reported [6]–[37]. The complexity of the high-temperature dielectric anomaly has been pointed out by the group of Smolensky, who reported in a narrow temperature range 170K to 205K three dielectric anomalies related to an improper ferroelectric transition (201K to 205K), to a ferroelectric-ferroelectric transition (196K to 200K), and to a ferroelectric diffuse phase transition (170K to 196K) [9]–[12]. Later, the transition at 205K was ascribed to an improper ferroelastic phase transition, whereas that at 196K was related to a proper ferroelectric state [38]. Detailed IR and dielectric spectroscopy studies confirmed the assignments and the transition at the Curie point TC = 196K was classified as mixed displacive and order-disorder type [27]. Dielectric relaxation processes below 170K have been studied in more detail by the group from Pennsylvania State University [6], [25], [26], [32], as well as by the group from Ioffe Physico-Technical Institute of the RAS [18], [19], [22]–[24], [29]–[31], [34]–[36], and as a result, CNO was claimed to exhibit relaxor properties [23], [29]–[35]. At lower temperatures, another three dielectric anomalies were reported: at ~80K [3]–[5], [10], [11], [15], [16], [25], [27], [28], [32], at ~45K and at ~18K [10], [20], [33], [36]. The anomaly at ~80K was described as a transition to an incommensurate phase and the anomaly at ~45K as a lock-in phase transition [10], [15], [33]. Existence of optical modes below 80K [27] and lack of modulation of spontaneous polarization [36] exclude, however, the existence of incommensurate phase between 45K and 80K. Similarly, the nature of dielectric anomaly at ~18K is unknown. It has been related to a glass transition [10], [20]; however, Isupov [36] suggested the transition is due to a symmetry change because of the existence of a ferroelectric hysteresis loop down to helium temperatures [3], [4] and crystalline-like behavior of specific heat reported recently [33]. The nature of the phase transitions in CNO has been discussed exhaustively by Isupov based on the main dielectric characteristics of the material [36]. It appears that the most intriguing and ambiguous feature of CNO is the frequency-dependent high permittivity dielectric anomaly

© 2013 IEEE

1604


related to relaxor behavior [39], [40]. In classical ferroelectric relaxors, however, the domain structure is apparent only in an external electric field, whereas in CNO, the domains have been observed below 80K without a field applied [30]. Moreover, the crystal does not contain any compositional disorder, which is an essential structural feature of the ferroelectric relaxors, and recent structural studies have concentrated on searching for a local disorder in the CNO lattice [41]–[47]. The room temperature structure of CNO is cubic with space group Fd-3m and consists of two interpenetrating sublattices: a network of corner-sharing NbO6 octahedra (with O(2) oxygen atoms) and a network of O(1)-Cd-O(1) zig-zag chains parallel to 〈110〉 located in empty spaces [3], [5], [41]–[47]. The structure of the ferroelectric phase of CNO in the range from 98K to 192K has been refined to orthorhombic symmetry with Ima2 space group [46]. Although it may be expected that the disorder of Nb atoms decreases because of the transition to the orthorhombic phase (lower number of local minima), thermal displacement parameters of Nb atoms located in chains parallel to the 〈100〉cub were found to be considerably large [44]. This points to some disorder of the Nb ions in the ferroelectric phase. X-ray diffused scattering observed in the room-temperature phase of CNO points, however, to local symmetry breaking [43], [45]. Experimental study of Xray diffuse scattering and calculations revealed a disorder in oxygen atoms of the NbO6 sublattice resulting in a transverse displacement of Nb atoms with respect to the 〈110〉 atomic chains [45]. Based on Monte Carlo simulations of the X-ray diffuse scattering and domain structure modeling, Malcherek proposed the disorder in Nb sites to be the source of relaxor-like properties of CNO [47]. Laterally disordered Nb-O chains within the ferroelectric orthorhombic phase may be considered as polar nanoregions oriented along the 〈110〉cub direction and the dielectric dispersion can be related to the switching of correlated chains of Nb displacement partially disordered between 195K and 100K. Below 85K, the structure is supposed to have the Cc space group with polarization oriented close to 〈100〉cub. Our work was aimed at preparing inexpensive CNO nanoceramics with high permittivity value at helium temperature. We expected, moreover, that it would be possible to get more detailed information on dielectric relaxation processes because the dielectric dispersion and absorption in nanoceramics is not dependent on the direction of applied electric field, as it appears in single crystals. The ceramics were obtained by hot-pressing CNO nanopowder prepared by mechanical activation of CdO and Nb2O5 and characterized by X-ray diffraction (XRD). Thermal behavior of the nanoceramics was measured in the temperature range 2K to 300K. II. Experiment CNO powder was synthesized by high-energy mechanical activation from high-purity CdO and Nb2O5 (Sigma-

vol. 60, no. 8,

August

2013

Aldrich Sp. z.o.o., Poznań, Poland) in stoichiometric ratio. The mechanical synthesis was carried out in a SPEX 8000 Mixer Mill (SPEX SamplePrep LLC, Metuchen, NJ) for 120 h at room temperature in the air; the weight ratio of the stainless steel balls to the oxides was 2:1. Roomtemperature synthesis lowers the fabrication costs and eliminates the undesirable losses of volatile elements [48]. Moreover, grains of nanometer size can be obtained, enabling study of the stability of the ferroelectric phase in a volume with reduced spatial dimensions. CNO ceramics were obtained from the powder by hot-pressing at 1073K under pressure of 200 MPa for 2 h. The structure of the synthesized CNO was controlled by X-ray powder diffraction studies using an X’Pert-PANalytical diffractometer (PANalytical B.V., Almelo, The Netherlands) with CuKα radiation. The XRD data were collected at room temperature in the 2θ range from 10° to 100° after milling and after the hot-pressing process. For dielectric measurements, the ceramic samples (a few millimeters in diameter and ~0.5 mm thick) were coated with gold electrodes. Dielectric response was studied using an Alpha-A high-performance frequency analyzer (Novocontrol GmbH & Co. KG, Hundsangen, Germany) combined with Quatro Cryosystem (Novocontrol GmbH & Co. KG) for the temperature control. The measurements were performed in the temperature range from 125K to 575K on heating at a rate of 1K/min and the frequency varied from 100 Hz to 1 MHz at the oscillation voltage of 1 V. A computer-controlled HP-4284A precision LCR meter (Agilent Technologies Inc., Santa Clara, CA) was used for measurements in the temperature range 4K to 300K. The data were collected and evaluated by WinDETA impedance analysis software and a WinFit v 3.2 program (both Novocontrol GmbH & Co. KG). To characterize the CNO ceramic samples, we measured also the thermal properties by means of a Physical Property Measurement System (PPMS, Quantum Design Inc., San Diego, CA) in the temperature range of 2K to 300K. The temperature was changed at a rate of 2K/min. III. Results A. XRD Studies Fig. 1 shows the XRD pattern of Cd2Nb2O7 powder obtained by mechanical synthesis and that of a ceramic sample produced by hot-pressing of mechanically synthesized CNO powder. The powder sample was obtained by high-energy milling of CdO and Nb2O5 for 120 h and the XRD pattern confirms that the material has CNO structure; however, traces of CdO can be recognized in the form of small reflections at 2θ = 38.4°, 66°, and 69.4°. A small reflection at 2θ = 38.4° is also visible in the diffraction pattern of hot-pressed CNO ceramics. The mean grain size of the CNO powder obtained by high-energy milling for 120 h was assessed to vary from 60 to 90 nm and that of the hot-pressed CNO ceramic sample to increase to

połomska et al.: dielectric response and specific heat studies of Cd2Nb2O7 ceramics

Fig. 1. X-ray diffraction pattern of Cd2Nb2O7 powder produced by mechanosynthesis of CdO and Nb2O5 oxides and that of Cd2Nb2O7 ceramics obtained by hot-pressing the powder.

1605

The relaxation processes are less visible in frequency and temperature dependences of the real part of permittivity ε′( f, T ). Temperature variation in ε′ of the CNO nanoceramics shows a broad maximum at ~151K with rather small dispersion and a shoulder at ~80K. Frequency dependence of the imaginary part of permittivity ε′′( f ) in the vicinity of 151K (relaxation I) appears to be rather complex and is shown in detail in Fig. 3. At low frequencies (1 to 25 kHz) the maximum values ε′′max decrease with increasing f at constant temperature Tε′′max ≈ 151.3K. The contribution we relate to the Curie point which for nanoceramics with reduced grain sizes is diffused and shifted toward lower temperatures with respect to that of conventional ceramics [49], [50]. At higher frequencies, the absorption anomaly becomes broader with a shoulder on the low-temperature side and the ε′′max values increase with increasing frequency. The relaxation processes apparent at lower temperatures and denoted by us as II, III, and IV (Fig. 2) were found to be not overlapped by other contributions and show similar behavior, as shown in Figs. 4 and 5. The processes consist of a part with ε′′max values decreasing with increasing frequency and a part in which the ε′′max values increase with increasing f. Moreover, the temperatures of ε′′max are shifted toward higher temperature with increasing frequency. The previously described ε′′( f, T ) depen-

100 to 150 nm. The room-temperature structure of hotpressed ceramics, refined by using the Rietveld method (starting from single crystal data [41]) and a PANalytical program, yields the space group Fd-3m (number 227) with lattice parameter a = 10.3681(3) Å and the density dceram = 6.2283 g/cm3. To compare the density of hot-pressed ceramics obtained from mechanically synthesized CNO nanopowder with that of CNO single crystal, we used the value dcryst = 6.2470 g/cm3 from our earlier measurements. It appears that the density of hot-pressed CNO ceramics attains 99.7% of the density of CNO single crystal. B. Dielectric Response Dielectric response of cadmium pyroniobate is known to depend strongly on the quality of the crystal as well as on the thermal history of the sample [26], [36]. In the case of conventionally prepared CNO ceramics, only two broad anomalies in the temperature dependence of the real part of dielectric permittivity ε′(T) have been reported, but at temperatures lower than those observed in single crystals. Fig. 2 shows dielectric response in the temperature range 4K to 300K of the ceramics obtained by hot-pressing of CNO nanopowder produced by mechanical synthesis of the respective oxides In the dielectric absorption ε′′( f, T ), one can distinguish at least four dielectric relaxation processes marked with Roman numerals.

Fig. 2. Temperature dependences of (a) real ε′ and (b) imaginary ε′′ parts of dielectric permittivity of Cd2Nb2O7 nanoceramics at various frequencies. Four dielectric relaxations are marked by Roman numerals in the part with ε′′( f, T ) dependences.

1606


vol. 60, no. 8,

August

2013

Fig. 4. Temperature variation of ε′ and ε′′ at various frequencies for Cd2Nb2O7 nanoceramics in the temperature range 60K to 140K.

Fig. 3. Temperature variation of ε′ and ε′′ at various frequencies in the vicinity of the huge dielectric anomaly for Cd2Nb2O7 nanoceramics.

dences in the temperature range 18K to 140K are similar to dielectric relaxation in relaxor ferroelectrics [40], [51]– [54]. It appears also that the relaxations I–IV cannot be considered as simple thermally activated relaxation because temperature variations of the relaxation times τ of the processes instead obey the Vogel–Vulcher relationship:

τ = τ ∞ exp[E a/T − T VF)], (1)

where τ∞, Ea, and TVF are the fitting parameters. We did not determine the activation energies Ea of the processes because the measurements were done in a narrow frequency window (2 to 3 frequency decades) and the results of fitting to Vogel–Fulcher equation would be loaded with a great error. Nevertheless, temperature behavior of the

relaxation times of processes I–IV is similar to that characteristic of ferroelectric relaxors [39], [40], [54]. The small dielectric absorption maxima visible in Fig. 3 at high temperatures are shown in detail in Fig. 6. The relaxation process in the temperature range 200K to 260K, which can be described by Vogel–Fulcher formalism, may be related to a local disorder in CNO. The problem appears with the absorption observed at the same temperature Tε′′max = 290K for frequencies from 1 to 100 kHz, with ε′′max value decreasing with increasing f. The origin of the absorption remains unknown. We observed also a relaxation process in the temperature range 350K to 400K, but with small ε′′ values. A similar process, related to oxygen vacancies, has been reported for CNO single crystals by Samara et al. [37]. It should be noted that the unusual ε′′( f, T ) behavior has been observed in cadmium pyroniobate single crystals and conventional ceramics. Swartz et al. reported a relaxation process in low-frequency ε′′(T ) dependences for CNO ceramics with ε′′max increasing with increasing f in the temperature range 130K to 190K and an ε′′ maximum for f = 10 kHz apparent at ~185K [16]. Similar behavior has been reported by Ang et al. and Yu and Ang for CNO ceramics from 125K to 175K and for single crystals in the temperature range ~160K to 200K (Mode I). The relaxation has been assigned to a dielectric relaxor mode [25], [32]. Moreover, Yu and Ang [32] found three other dielectric relaxations in the temperature range 100K to 150K


1607

Fig. 6. High-temperature anomalies in ε′′(T, f ) dependences for Cd2Nb2O7 nanoceramics.

Fig. 5. Temperature variation of ε′ and ε′′ at various frequencies for Cd2Nb2O7 nanoceramics in the temperature range 4K to 60K.

(Mode II), in the range 50K to 80K (Mode III), and below 50K (Mode IV), but they did not discuss the physical nature of the processes. Dielectric relaxation characterized by increasing ε′′max values with increasing frequency in the temperature range 100K to 200K has been published also by Kolpakova et al. for CNO single crystals and ceramics [22], [23], [29]–[31], [34], [35]. Kolpakova et al. discussed the sequence of disordered states in cadmium pyroniobate and claimed its relaxor-like behavior. C. Specific Heat Temperature variation of the specific heat Cp of CNO nanoceramics is shown in Fig. 7. The measurements did not reveal in the nanoceramics any anomaly in the vicinity of 200K, which has been reported for CNO single crystals by Tachibana et al. [33]. At low temperatures (the inset of Fig. 7) the dependence Cp/T 3 versus log T shows a departure from the Debye law below 5K and a maximum at 18K. Similar maximum at 18K has been reported earlier for CNO conventional ceramics by Lawless [55] and for CNO single crystals by Tachibana [33] and described by an Einstein term with ω = 53 cm−1. At temperatures below 5K, the Debye contribution of CNO single crystals was reported by Tachibana to be constant, whereas a sharp upturn below ~5K can be observed for CNO nanoceramics. The difference may be due to the size effect [49].

Fig. 8 shows temperature dependence of the excess heat ΔCp over the lattice heat Cp of the nanoceramics. A small excess heat ΔCp, of ~3% of the Cp value in this temperature range can be revealed in the vicinity of 150K, where the huge dielectric absorption, related by us to an overlapping of the Curie point contribution and the dielectric relaxation process, is apparent (Figs. 2 and 3). The broad and small ΔCp maximum observed by us for CNO nanoceramics at ~150K is in agreement with the size effect on the specific heat of ferroelectrics reported earlier [49]. An excess heat (of ~2% of the Cp value) appears also between 175K and 240K, where the dielectric absorption related by us to a local disorder in CNO was observed (Fig. 6). IV. Discussion Though the phase transitions and the dielectric response of cadmium pyroniobate have been extensively studied for more than half a century a clear understanding of the dielectric anomalies is still not available. One of the difficulties in resolving the problem was the narrow temperature range 170K to 205K, in which overlapping contributions of various transitions to the dielectric response of CNO have been observed. The second obstacle was due to the lack of the crystal structure data below room temperature. Only recently has the structure of CNO been determined in the temperature range 98K to 192K using synchrotron radiation and refined in the Ima2 space group starting from the structure previously obtained by the ab initio method [46]. Ab initio calculations of the stability of the CNO structures suggested also a sequence of the phase transitions in the crystal: Fd-3m–Ima2–Cc [44]. The origin of dielectric relaxation processes in cadmium pyroniobates still appears to be the most puzzling problem because the material does not contain any compositional disorder, which is a structural feature of ferroelectric relaxors. The results of X-ray diffuse scattering studies pointed to a local disorder in CNO lattice [42]–

1608


Fig. 7. Temperature dependence of the specific heat of Cd2Nb2O7 nanoceramics; departure from the Debye law at low temperatures is shown in the inset.

[46], but only recently Malcherek proposed a model of polar nanoregions (PNR) in the ferroelectric orthorhombic phase [47]. He studied the order-disorder contribution to the ferroelectric properties of CNO by Monte Carlo simulation and found that the relaxor-like behavior of CNO can be due to low-frequency switching of correlated chains of Nb displacement. He obtained that the Ima2 structure contains 2.5 double short Nb=O bonds for each Nb atom, which results in a disorder at the Nb1 site, located close to the (0, 0, 0) position. Correlated Nb-O chains of laterally disordered Nb displacement can be considered as PNRs in the orthorhombic matrix of CNO and a low-frequency electric field can reorient permanent dipoles within the nanoregions between the [110]cub and the [100]cub polarization states. Below ~80K, the CNO structure is expected to have monoclinic symmetry and higher number of short Nb=O bonds per Nb atom in comparison with that in the orthorhombic symmetry. This can be achieved by creating local short-range order (PNR) in the ferroelectric lattice and relaxor-like dielectric response. A similar approach can be used to explain the dielectric response below ~45K, where another ferroelectric phase is expected to appear [36]. Malcherek stressed also that frequency dependences of the permittivity of CNO single crystals should be strongly dependent on the direction of the applied electric field. We consider that differences in the relaxation processes reported by various authors were related to different sample orientations and our CNO nanoceramics, in which the grain sizes of 100 to 150 nm provide excellent spatial averaging, has an advantage in dielectric relaxation studies. Although the model proposed by Malcherek shows the origin of relaxor-like behavior in CNO, the specific feature of dielectric relaxation modes (a part with decreasing ε′′max and a part with increasing ε′′max values with increasing frequency) has not been explained. An explanation yields the theory of dielectric response of ferroelectric relaxors [40], [51]–[54] which considers the behavior of ε′ and ε′′max with increasing frequency. In the case when ε′

vol. 60, no. 8,

August

2013

Fig. 8. Excess heat ΔCp over the lattice specific heat versus temperature for Cd2Nb2O7 nanoceramics.

decreases and ε′′max increases with increasing frequency, the main contribution to the dielectric response is due to flipping of the dipole moments between allowed directions in the PNRs, whereas a decrease in ε′ and ε′′max with increasing frequency is related to the contribution of fluctuations of the boundaries of PNRs. Thus, in the dielectric relaxation mode II of CNO (Fig. 4), the low-temperature and low-frequency part of the response may be related to the contribution of fluctuations of PNR boundaries, whereas the high-temperature contribution is due to reorientation of dipoles in PNRs between pseudocubic 〈110〉 directions in the orthorhombic phase of CNO. Following the consideration, one can state that in relaxation processes III and IV (Fig. 5) the dielectric response at low-frequencies is mainly due to the fluctuations of PNR boundaries, but at higher frequencies, the contribution from flipping of PNR dipole moments between the allowed directions becomes dominant. It should be also noted that the CNO nanoceramics exhibit a huge and broad ε′(T ) maximum and a complex ε′′(T, f ) dependence (Fig. 3) in the vicinity of 150K. The dielectric anomaly we describe as due to overlapping contributions from the ferroelectric–paraelectric phase transition, which in nanoceramics with reduced grain sizes is broadened and shifted downwards [49], [50], and to a relaxation process (the lowest part of Fig. 3), the origin of which should be clarified. V. Conclusions Cd2Nb2O7 nanoceramics was obtained by hot-pressing of CNO nanopowders synthesized by mechanical activation of CdO and Nb2O5 oxides. The synthesis is not expensive (does not required high temperatures) and the permittivity value of the nanoceramics (with mean grain sizes 100 to 150 nm) is very high even at helium temperature, where ε′ at the frequency of 100 Hz is equal to 540, and at 100 kHz the permittivity ε′ = 525. The dielectric anomaly (I) of the CNO nanoceramics is related by us


to an overlapping of the contributions to the dielectric response from the ferroelectric–paraelectric transition, which because of the size effect was found to be diffused and shifted toward lower temperatures by ~50K [49], [50], and from a dielectric relaxation process of unknown origin. Three other relaxation modes were found on cooling the nanoceramics to helium temperature, and their relaxorlike frequency behavior was discussed based on the model of polar nanoregions in CNO proposed by Malcherek [47] and the theory of frequency dependence of the dielectric response of ferroelectric relaxors [40], [51]–[54].

References [1] W. R. Cook Jr. and H. Jaffe, “Ferroelectricity in oxides of fluorite structure,” Phys. Rev., vol. 88, p. 1426, Dec. 1952. [2] W. R. Cook Jr. and H. Jaffe, “Ferroelectricity in oxides of facecentered cubic structure,” Phys. Rev., vol. 89, pp. 1297–1298, Mar. 1953. [3] G. Shirane and R. Pepinsky, “Dielectric properties and phase transitions of Cd2Nb2O7 and Pb2Nb2O7,” Phys. Rev., vol. 92, p. 504, Oct. 1953. [4] J. K. Hulm, “Low-temperature dielectric properties of cadmium and lead niobates,” Phys. Rev., vol. 92, pp. 504–505, Oct. 1953. [5] F. Jona, G. Shirane, and R. Pepinsky, “Dielectric, X-ray and optical study of ferroelectric Cd2Nb2O7 and related compounds,” Phys. Rev., vol. 98, pp. 903–1909, May. 1955. [6] A. de Bretteville Jr., F. A. Halden, T. Vasilos, and L. Reed, “Dielectric studies in the system CdO-Nb2O5,” J. Am. Ceram. Soc., vol. 40, pp. 86–89, Mar. 1957. [7] V. A. Isupov, G. I. Golovshchikova, and I. E. Mylnikova, “Peculiarities of dielectric polarization of cadmium pyroniobate,” Sov. Phys. Solid State, vol. 13, pp. 2349–2352, Aug. 1971. [8] V. A. Isupov, G. I. Golovshchikova, and I. E. Mylnikova, “Unusual ferroelectric properties of cadmium pyroniobate,” Ferroelectrics, vol. 8, no. 1, pp. 507–509, 1974. [9] N. N. Kolpakova, G. I. Sinii, M. Polomska, and R. Margraf, “Phase transitions in Cd2Nb2O7,” Sov. Phys. Solid State, vol. 24, pp. 1729– 1738, Jun. 1982. [10] G. A. Smolenskii, N. N. Kolpakova, and S. A. Kizhaev, “Instability of the dielectric constant of Cd2Nb2O7,” JETP Lett., vol. 36, pp. 361–363, Oct. 1982. [11] F. M. Salaev, L. S. Kamzina, N. N. Krainik, E. S. Sher, and G. A. Smolenskii, “Domain structure and dielectric permittivity relaxation in cadmium pyroniobate single crystals,” Sov. Phys. Solid State, vol. 25, pp. 163–168, Jan. 1983. [12] G. A. Smolenskii, F. M. Salaev, L. S. Kamzina, N. N. Krainik, and S. N. Dorogovtsev, “Anomalously narrow peak in the dielectric constant of Cd2Nb2O7 near the diffuse phase transition,” JETP Lett., vol. 37, pp. 303–306, Mar. 1983. [13] N. Yasuda, S. Fujimoto, K. Tanaka, and T. Hachiga, “Pressure and temperature dependence of dielectric properties of cadmium niobate,” J. Phys. D, vol. 17, pp. 2069–2079, Oct. 1984. [14] G. A. Smolensky, N. N. Krainik, L. S. Kamzina, F. M. Salaev, E. A. Tarakanov, and E. S. Sher, “The investigation of successive phase transitions in cadmium pyroniobate crystals,” Jpn. J. Appl. Phys., vol. 24, suppl. 24-2, pp. 820–822, 1985. [15] G. A. Smolensky, N. N. Kolpakova, S. A. Kizhaev, and E. S. Sher, “Thermal hysteresis of dielectric constant and anomalous behavior of the Raman soft mode damping of Cd2Nb2O7,” Ferroelectrics, vol. 73, no. 1, pp. 161–169, 1987. [16] S. L. Swartz, C. A. Randall, and A. S. Bhalla, “Dielectric and pyroelectric properties of cadmium niobate ceramics,” J. Am. Ceram. Soc., vol. 72, pp. 637–641, Apr. 1989. [17] N. N. Kolpakova, L. Bobrowicz, A. Pawłowski, and W. Nawrocik, “Dielectric permittivity of Cd2Nb2O7 single crystals under hydrostatic pressure,” Solid State Commun., vol. 75, pp. 461–463, May 1990.

1609

[18] Z.-G. Ye, N. N. Kolpakova, J.-P. Rivera, and H. Schmid, “Optical and electric investigations of the phase transitions in pyrochlore Cd2Nb2O7,” Ferroelectrics, vol. 124, no. 1, pp. 275–280, 1991. [19] N. N. Kolpakova, R. Margraf, and M. Połomska, “Dielectric relaxation phenomena in Cd2Nb2O7 ferroelectric-ferroelastic,” J. Phys. Condens. Matter, vol. 6, pp. 2787–2798, Apr. 1994. [20] N. N. Kolpakova, B. Hilczer, and M. Wiesner, “Incommensurable and glassy behavior of Cd2Nb2O7 oxide pyrochlore single crystals,” Phase Transit., vol. 47, no. 1–2, pp. 113–124, 1994. [21] N. N. Kolpakova, L. Szczepańska, I. L. Shulpina, M. Wiesner, and P. Piskunowicz, “Superstructure and thermal hysteresis effects at the TCurie in Cd2Nb2O7 pyrochlore,” Ferroelectrics, vol. 190, no. 1, pp. 173–178, 1997. [22] N. N. Kolpakova, M. Wiesner, G. Kugel, and P. Bourson, “Relaxation processes in the ferroeletric phase of Cd2Nb2O7 far below TCurie,” Ferroelectrics, vol. 190, no. 1, pp. 179–184, 1997. [23] N. N. Kolpakova, M. Wiesner, G. Kugel, and P. Bourson, “Relaxor-like behavior of pyrochlores,” Ferroelectrics, vol. 201, no. 1, pp. 107–115, 1997. [24] N. N. Kolpakova, M. Wiesner, A. O. Lebedev, P. P. Syrnikov, and V. A. Khramtsov, “Influence of accidental impurities and defects on the low-frequency dielectric relaxation in the ferroelectric Cd2Nb2O7,” Tech. Phys. Lett., vol. 24, pp. 679–682, Sep. 1998. [25] Ch. Ang, R. Guo, A. S. Bhalla, and L. E. Cross, “Dielectric relaxation processes in Cd2Nb2O7 compound,” J. Appl. Phys., vol. 87, pp. 7452–7456, May 2000. [26] Ch. Ang, L. E. Cross, R. Guo, and A. S. Bhalla, “Cluster polarization of Cd2Nb2O7 compound,” Appl. Phys. Lett., vol. 77, pp. 732– 734, Jul. 2000. [27] E. Buixaderas, S. Kamba, J. Petzelt, M. Savinov, and N. N. Kolpakova, “Phase transitions sequence in pyrochlore Cd2Nb2O7 studied by IR reflectivity,” Eur. Phys. J., vol. B19, pp. 9–16, Jan. 2001. [28] Ch. Ang, A. S. Bhalla, R. Guo, and L. E. Cross, “Effect of dc bias on dielectric properties of Cd2Nb2O7 ceramics,” J. Appl. Phys., vol. 90, pp. 2465–2468, Sep. 2001. [29] N. N. Kolpakova, P. P. Syrnikov, A. O. Lebedev, P. Czarnecki, W. Nawrocik, C. Perrot, and L. Szczepanska, “2-5 pyrochlore relaxor ferroelectric Cd2Nb2O7 and its Fe2+/Fe3+ modifications,” J. Appl. Phys., vol. 90, pp. 6332–6340, Dec. 2001. [30] N. N. Kolpakova, “Recent understanding the dielectric relaxation processes, domain evolution and disordering in Cd2Nb2O7 pyrochlore,” Ferroelectrics, vol. 291, no. 1, pp. 65–73, 2003. [31] N. N. Kolpakova, P. Czarnecki, W. Nawrocik, M. R. Shcheglov, and L. Szczepanska, “Recent understanding the dielectric relaxation processes and dipolar disordering in high-symmetry relaxor system Cd2Nb2O7,” Ferroelectrics, vol. 302, no. 1, pp. 233–239, 2004. [32] Z. Yu and Ch. Ang, “Dielectric relaxor behavior of Cd2Nb2O7,” Appl. Phys. Lett., vol. 85, pp. 801–803, Aug. 2004. [33] M. Tachibana, H. Kawai, and T. Atake, “Calorimetric investigation of successive transitions in Cd2Nb2O7,” Phys. Rev. B, vol. 70, art. no. 064103, Aug. 2004. [34] N. N. Kolpakova, P. Czarnecki, W. Nawrocik, M. P. Shcheglov, P. P. Syrnikov, and L. Szczepańska, “Crossover from glassy to ferroelectric polarization behavior under a dc bias electric field in relaxor ferroelectrics,” Phys. Rev. B, vol. 72, art. no. 024101, Jul. 2005. [35] N. N. Kolpakova, P. Czarnecki, R. P. Syrnikov, M. P. Shcheglov, and L. Szczepanska, “Nano-scale frustrated Cd2Nb2O7 ferroics,” Ferroelectrics, vol. 318, no. 1, pp. 155–161, 2005. [36] V. A. Isupov, “On the nature of the phase transition in cadmium pyroniobate,” Sov. Phys. Solid State, vol. 47, pp. 2119–2129, Nov. 2005. [37] G. A. Samara, E. L. Venturini, and L. A. Boatner, “Dielectric properties and phase transitions of Cd2Nb2O7: Effects of pressure and temperature,” J. Appl. Phys., vol. 100, art. no. 074112, Oct. 2006. [38] N. N. Kolpakova, S. Waplak, and W. Bednarski, “Gd3+ EPR spectroscopy and optical microscopy study of ferroic states in Cd2Nb2O7 pyrochlore,” J. Phys. Condens. Matter, vol. 10, pp. 9309–9316, Oct. 1998. [39] L. E. Cross, “Relaxor ferroelectrics: An overview,” Ferroelectrics, vol. 151, no. 1, pp. 305–320, 1994. [40] A. A. Bokov and Z.-G. Ye, “Recent progress in relaxor ferroelectrics with perovskite structure,” J. Mater. Sci., vol. 41, pp. 31–52, Jan. 2006. [41] K. Łukaszewicz, A. Pietraszko, J. Stepień-Damm, and N. N. Kolpakova, “Temperature dependence of the crystal structure and dy-

1610


namic disorder of cadmium in Cd2Nb2O7 and Cd2Ta2O7,” Mater. Res. Bull., vol. 29, pp. 987–992, Sep. 1994. [42] M. T. Weller, R. W. Hughes, J. Rooke, C. S. Knee, and J. Reading, “The pyrochlore family—A potential panacea for the frustrated perovskite chemist,” Dalton Trans., vol. 2004, pp. 3032–3041, Jul. 2004. [43] T. Malcherek, “Anisotropic diffuse x-ray scattering in early transition metal oxide structure,” J. Phys. Condens. Matter, vol. 19, art. no. 275208, Jun. 2007. [44] M. Fischer, T. Malcherek, U. Bismayer, P. Blaha, and K. Schwarz, “Structure and stability of Cd2Nb2O7 and Cd2Ta2O7 explored by ab initio calculations,” Phys. Rev. B, vol. 78, art. no. 014108, Jul. 2008. [45] M. Pasciak, M. Wołcyrz, and A. Pietraszko, “Local structure in the paraelectric phase of Cd2Nb2O7 determined from x-ray diffuse scattering, by means of ab initio molecular dynamics and Monte Carlo modeling,” Phys. Rev. B, vol. 81, art. no. 014107, Jan. 2010. [46] T. Malcherek, U. Bismayer, and C. Paulmann, “The crystal structure of Cd2Nb2O7: Symmetry mode analysis of the ferroelectric phase,” J. Phys. Condens. Matter, vol. 22, art. no. 205401, Apr. 2010. [47] T. Malcherek, “The ferroelectric properties of Cd2Nb2O7: A Monte Carlo simulation study,” J. Appl. Cryst., vol. 44, pp. 585–594, Apr. 2011. [48] L. B. Kong, T. S. Zhang, J. Ma, and F. Boey, “Progress in synthesis of ferroelectric ceramic materials via high-energy mechanochemical technique,” Prog. Mater. Sci., vol. 53, pp. 207–322, Feb. 2008. [49] M. D. Glinchuk and P. I. Bykov, “The peculiarities of the specific heat and dielectric permittivity related to the grain size distribution in ferroelectric nanomaterials,” J. Phys. Condens. Matter, vol. 16, pp. 6779–6788, Sep. 2004. [50] J. M. Gregg, “Ferroelectrics at the nanoscale,” Phys. Stat. Solidi A, vol. 206, pp. 577–587, Apr. 2009. [51] A. A. Bokov and Z.-G. Ye, “Low-frequency dielectric spectroscopy of the relaxor ferroelectric Pb(Mg1/3Nb2/3)O3–PbTiO3,” Phys. Rev. B, vol. 65, art. no. 144112, Apr. 2002. [52] A. A. Bokov and Z.-G. Ye, “Universal relaxor polarization in Pb(Mg1/3Nb2/3)O3 and related materials,” Phys. Rev. B, vol. 66, art. no. 064103, Aug. 2002. [53] A.A. Bokov and Z.-G. Ye, “Double freezing of dielectric response in relaxor Pb(Mg1/3Nb2/3)O3 crystals,” Phys. Rev. B, vol. 74, art. no. 132102, Oct. 2006. [54] A. A. Bokov and Z.-G. Ye, “Dielectric relaxation in relaxor ferroelectrics,” J. Adv. Diel., vol. 2, art. no. 1241010, Jun. 2012. [55] W. N. Lawless, “Specific heats of lead and cadmium niobates at low temperature,” Phys. Rev., vol. B19, pp. 3755–3759, Apr. 1979.

Maria Połomska received the M.Sc. and Ph.D. degrees in physics in 1966 and 1974, respectively, from the Adam Mickiewicz University, Poznań, Poland. Until 1971, she worked at the Institute of Physics, Adam Mickiewicz University. Since 1971, she has been employed in the Department of Ferroelectrics of the Institute of Molecular Physics of the Polish Academy of Sciences (PAS) in Poznan. Since 2009, she has been a professor at the Institute of Molecular Physics of PAS. Her main fields of interest are ferroic and superprotonic phase transitions; ferroic domain structure; and molecular dynamics of multiferroics, superprotonics, and ferroics studied by Raman spectroscopy. She is a member of the International Steering Committee.

Bożena Hilczer graduated with an M.Sc. degree in 1958 and received her Ph.D. degree in physics in 1965 from Adam Mickiewicz University, Poznań, Poland. In 1974, she received the Dr.Habil. degree in solid-state physics from the Institute of Physics, Polish Academy of Sciences (PAS), Warsaw, and in 1981, the title of Professor in Physics. Since 1958, she has been working at the Institute of Physics, PAS, and later at the Institute of Molecular Physics, PAS, in Poznań, currently as Professor Emeritus. She was a Supervi-

vol. 60, no. 8,

August

2013

sor of more than 30 M.Sc. students in physics and 9 Ph.D. students in physics. Her main professional activity is related to phase transitions in ferroics, multiferroics and superionics; dielectric response in electroactive composites; and charge storage in dielectrics (electrets, piezopolymers, polymer relaxors). She is a member of the Editorial Board of Ferroelectrics and Ferroelectrics Letters.

Ewa Markiewicz received her M.Sc.Eng. degree in electronics from the Technical University in Wroclaw, Poland, in 1982. From 1982 to 2000, she was employed as design engineer in Loudspeaker Factory TONSIL in Wrzesnia. Her intensive cooperation with the Institute of Molecular Physics, Polish Academy of Sciences (PAS), in the field of electrets microphones resulted in changing her employment. Since 2000, she has been employed in the Ferroelectrics Department of the Institute of Molecular Physics PAS in Poznan. In 2006, she received a Ph.D. degree in material science from the Institute of Electronic Materials Technology in Warsaw. She has developed a specialty in studies of the dielectric properties of ceramics, polymers, and composites, as well as the synthesis of ceramic powders. She has also been involved in the determination of piezoelectric, elastic, and dielectric coefficients of crystals.

Zbigniew Trybuła was born May 21, 1957, in Pleszew, Poland. He received his M.Sc. degree in physics from Adam Mickiewicz University in 1981 and his Ph.D. degree in physics from the Institute of Molecular Physics (IMP), Polish Academy of Sciences (PAS), in 1985. Since 2012, he has been an Ordinary Professor at IMP PAS. From 1995 to 2011, he was a head of the Low-Temperature Physics Division of IMP PAS in Odolanów. During his sixth term (2005 to 2007) he was the senator of the Republic of Poland. He participated in organization of the Low-Temperature Physics Laboratory of the Institute of Molecular Physics, Polish Academy of Sciences in Odolanów. His main fields of interest is low-temperature investigations in proton glass of RADA-type crystals by dielectric and EPR methods, dielectric study of nominally pure and doped incipient ferroelectric KTaO3, and low-temperature dielectric investigation in the temperature range from 0.3K to 300K. He received training at Montana State University, Bozeman, MT, in 1989 and August 1993. He received the Award of the Director of the Institute of Molecular Physics Polish Academy of Sciences in 1987 and the Award of the Scientific Secretary of the Polish Academy of Sciences in 1989. He is a member of the Polish Physical Society.

Bartłomiej Andrzejewski was born in Bydgoszcz, Poland, in 1967. He received his M.S. degree in experimental physics from Adam Mickiewicz University in Poznan in 1992 and his Ph.D. degree from the Institute of Molecular Physics, Polish Academy of Sciences (PAS), in Poznan in 1999. One year later, he joined the CRISMAT laboratory in Caen, France, where, as a postdoctoral researcher, he worked on high-temperature superconductivity, surface vortex pinning, and magnetic properties of bismuth granular superconduc-

tors. Currently, he is a Professor and a head of the Department of Ferroelectrics at the Institute of Molecular Physics in Poznan. His research interests focus on the properties of nanostructured ferroelectrics and multiferroics and on the microwave synthesis of these materials with desired size and shape. He has received several awards, including the Foundation for Polish Science award for young scientists, European Science Foundation visiting fellowship, and the award of the Polish Scientific Network. He is an author of more than 100 papers in national and international journals.

połomska et al.: dielectric response and specific heat studies of Cd2Nb2O7 ceramics Izabela Szafraniak-Wiza received the M. Sc. and Ph.D. degrees from Adam Mickiewicz University, Poznań, Poland, in 1995 and 2001, respectively, in experimental physics (in the field of ferroelectric materials). Between 2001 and 2003, she worked at the Max Planck Institute of Microstructure Physics, Halle, Germany, as Marie-Curie postdoctoral fellow. Her research interest was focused on preparations and characterizations of ferroelectric nanostructures. Since 2003, she has worked at Poznan University of Technology, Institute of Materials Sciences and Engineering. Her current scientific interests are in the fields of nanoscience and nanotechnology, synthesis, and physical property investigations of perovskite oxides. She has been a member of the Management Committee of COST 539 and MP0904 Action and the International Advisory Board of the European Meeting of Ferroelectricity. She is a referee of several leading journals and of scientific projects of the EU.

1611

Adam Pietraszko graduated in 1966 from the University of Wrocław, Poland, and received his Ph.D. and Dr.Habil. degrees in physics from the Institute of Low Temperature and Structure Research, Polish Academy of Sciences (PAS), in Wrocław in 1973 and 1992, respectively. In 2000, he was granted the title of Professor in Physics. From 1971 to 1998, he was the Secretary, and from 2000 to 2008 the President of the Committee of Crystallography, of the PAS. He is heading the Department of Crystallography at the Institute of Low-Temperature and Structure Research, PAS, in Wrocław and is involved in crystal structure studies of ferroics, multiferroics, and superionics, as well as in X-ray diffuse scattering and local structure studies.