WorksPhysics and Astronomy
Understanding the evolution of the magnetic ground state in BaNaRuO
Abstract
We report a comprehensive investigation of the quadruple perovskite BaNaRuO, in which we discover a robust spin-lattice coupled ground state characterized by a long-range antiferromagnetic ordering at 257 K. The system's unique structural motif of three symmetrically distinct magnetic ions, including Ru dimers separated by non-magnetic layers, is intimately correlated with its magnetic behavior, as evidenced by temperature-dependent diffraction measurements and specific heat data. The powder neutron diffraction patterns at 13 K showed that the spins within the dimers are antiparallel, leading to a net zero moment contribution and a staggered arrangement of the triangular layers formed by the Ru moments within the corner-shared octahedra along the -axis. The low-temperature specific heat revealed an extra boson peak contribution from optical modes with a maximum vibrational energy of 55cm. The charge transport exhibited variable-range hopping (VRH) behaviour below , with a stronger energy-dependence than expected from the Efros-Shklovskii model, suggesting the presence of multiparticle correlation effects.
Introduction
Perovskites are one of the most celebrated material classes in condensed matter physics due to their unprecedented compositional flexibility and versatility. The ideal perovskite structure, a neatly arranged octahedral network, can be tweaked to generate a variety of complexity, giving rise to genuinely intriguing physics. Substitution of A- or B-sites of the parent structure produces various kinds of layerings of which the hexagonal layered systems are of special interest due to their inherent geometric frustration. Various 4d and 5d double- (1:1 substitution of the B-site) and triple- (1:2 substitution of the B-site) perovskite systems are known to host a plethora of exotic magnetic states [1, 2, 3, 4].
A 1:3 substitution of the A-site generates highly anisotropic crystal systems with very interesting material properties. These systems have recently gained traction due to their excellent dielectric [5] and catalytic properties [6]. They are categorized as distorted cubic perovskites and their structures are strongly influenced by the type of Glazer tilt system [7] involved. For example, large tilts produce the structure where the strong tilting of the B octahedra forces the A′ sites to settle in a square planar geometry. On the other hand, strong tilting in the directions generates the structure possessing an inherent columnar-order [8]. These materials, although with a huge potential for the exploration of symmetry-influenced physical properties, require the use of simultaneous high-pressure and high-temperature environments to be stabilized.
The B-site substituted systems, conversely, show very interesting quasimolecular physics arising from the cation dimers and trimers governing their magnetism. For example, in systems like [9] and [10, 11] (M = Ru, Ir), Ru/Ir trimers isolated by nonmagnetic layers show fascinating frustration-based physics with either long-range order developing at extremely low temperatures or not at all. The relative proximity of the transition-metal (TM) cations within these trimers leads to the formation of molecular orbitals which dictate the physical properties in these systems. The atomic -orbitals of two relatively close TM cations, in a local symmetry, linearly combine to produce bonding and antibonding molecular orbitals. These states are strongly influenced by the geometry of the local field and the electron hopping probabilities between them. This makes -orbital compounds an attractive platform to study instabilities leading to quantum phase transitions [12], due to their ability for direct exchange and high degree of tunability. Additionally, compounds containing isolated groups of magnetic ions as repeating motifs have often surprised physicists with their exotic properties and complex magnetic ground states [1, 13] despite disruptions in the exchange pathway. This highlights the importance of next-nearest neighbour superexchange mediated interactions in these materials.
uniquely stands out in this family of B-site substituted quadruple perovskites in that it is neither a network of isolated or connected Ru/Ir trimers like [13] and [14, 15] nor of isolated Ru dimers like its sister triple-perovskite [1]. The repeating motif of this structure is a set of one octahedra connected via a corner shared oxygen to a dimer separated by layers of nonmagnetic octahedra. To the best of our knowledge, no other layered ruthenate is known to possess such a motif. This material has also been scarcely studied in the literature, with few reports on it from nearly 20 years ago [16, 17, 18], leaving much to be explored, especially its complex magnetism at low . This motivates us to investigate this compound in-depth and explore its ground state’s evolution as a function of temperature and characterize its correlation with lattice and charge degrees of freedom.
Experimental Methods
Polycrystalline samples of BaNaRuO were prepared via solid state synthesis method using stoichiometric amounts of raw materials (BaCO, NaCO and RuO) following the procedure detailed by Battle et al. [16]. To compensate for the volatility of Na, a 15% excess of NaCO was used and RuO was heated for 2 - 3 hours at 250°C before weighing to remove any moisture. The precursors were mixed and ground in a mortar using ethanol until a fine, homogeneous mixture was obtained. The entire mixture was then pelletized to further reduce the loss of Na, placed in an alumina crucible and heated at 800°C for 12 hours in air. Hard, black pellets obtained at the end of the reaction were crushed and checked for phase purity through laboratory x-ray diffraction (XRD) using the BRUKER D8 ADVANCE DIFFRACTOMETER (CuK =1.5406Å). Elemental compositions and their homogeneity was also confirmed by using an energy dispersive X-ray spectrometer (EDS) (Ziess Ultra Plus). The average stoichiometry calculated from the statistical analysis of the EDS data is BaNaRuO, confirming the homogeneity and single phase nature of the sample. Field emission scanning electron microcopy (FESEM) images of hexagonal microstructure are shown in Figure 1
Figure 1. Scanning electron micrographs of BaNaRuO powder samples, taken at 20keV energy, capture the micron-sized hexagonal crystallites displaying a plate-like morphology.
Synchrotron x-ray powder diffraction (SXRD) was performed at the Swiss Light Source (PSI) on the ADDAMS beamline [19]. The wavelength of the x-ray was calculated by refining LaB standard to be Å and an instrument-resolution file (IRF) file was created (FWHM (°) vs 2). Using this IRF file, the profile parameters were fixed and only microstructural and atomic parameters were refined for diffraction patterns collected at each K. This allowed us to estimate changes in maximum strain (%) with temperature. Rietveld refinement of the diffraction patterns were performed using the FULLPROF SUITE [20]. Crystal Structures were generated using VESTA [21]. Neutron time-of-flight (tof) measurements were performed at the WISH diffractometer [22] of ISIS Pulsed Neutron and Muon source in the range 13 - 300 K. Data collected on 5 banks were simultaneously refined using FULLPROF and Bank 4-7 (2 =121.66°) was chosen to be the pattern of choice with the most appropriate −range and resolution for plotting. Specific Heat and
Magnetization measurements were performed in a Quantum Design PPMS Dynacool. Resistivity measurements were performed in a Cryogenic Limited Cryogen Free Measurement System (CFMS) using a VTI-based resistivity option.
Results and Discussion
BaNaRuO crystallizes in a rare 8-layered hexagonal structure described by the non-centrosymmetric space group (No. 186) as shown in Figure 2, with lattice parameters Å and Å. The Ba ions have 4 crystallographically distinct sites and occupy corner and edges of the unit-cell as well as the voids between the octahedra. The Ru ions have three unique positions as well, two of them (Ru2, Ru3) forming a face-shared octahedra located on the unit cell edges, while one (Ru1) sits at the centre of an octahedra corner-linking the RuO and the NaO octahedra. This structure is visually presented in Figure 2(a). Viewed along the -axis the 8-layers (2 formula units, so ) forming the unit cell can be seen in Figure 2 (d). Figure 2(b) and (c) show the Ru1(/Na) and Ru2(/Ru3) layers in the -plane, highlighting the triangular arrangement of Ru cations in each layer. The structural parameters obtained from the Rietveld refinement of this structure are tabulated in Tables I. The values obtained are consistent with previous reports [16, 17].
Figure 2. (a) Crystal structure of BaNaRuO. (b) Bond lengths and angles of the exchange path calculated from at room temperature neutron diffraction data. The triangular network formed in the ab-plane by (c) Ru1/Na and (d) Ru2/Ru3 ions viewed along the c-axis. (e) The 8-layered unit cell viewed along the a-axis.
Since the ionic radii of Na (1.02 Å[23]) and Ru (0.565 Å[23]) vastly differ, no site disorder is expected and all ions fully occupy their sites. The refinement results confirm this too and return a total cell content of BaNaRuO. However, since Ba2, Ba3, Ba4, Na and Ru1 - all lie on the same plane [Wyckoff site: 2b (1/3,2/3,z)], the ion positions and thermal parameters required constraints to prevent the fit from diverging. The three Ru sites are inequivalent, but have the same site multiplicity. Figure 3 shows the fit to the structural model in Table I for synchrotron x-ray (top panel) and neutron (bottom panel) powder diffraction data at room temperature. The goodness of fits obtained are reasonable and confirm the accuracy of the model.
Figure 3. Room temperature neutron and synchrotron XRD powder diffraction patterns of BaNaRuO refined using FULLPROF SUITE [20]
The distance between the Ru ions within the RuO dimer is 2.652 Å, shorter than the Ru-Ru metallic bond length (2.6725Å [24]), suggesting a strong antiferromagnetic (AFM) direct interaction between Ru2 and Ru3. The observed average Ru—O bond length Å is very close to that expected for an Ru—O bond (1.964 Å[25]), confirming the pentavalent oxidation state of Ru. The Ru octahedra are also slightly distorted as evidenced by the difference in the distances of the two uniquely located O ions forming each of the Ru octahedra (Ru1—O1 = 1.866 Å, Ru1—O3 = 2.094 Å, Ru2—O3 = 1.943 Å, Ru2—O4 = 2.0 Å, Ru3—O4 = 2.05 Å, Ru3—O2 = 1.876 Å). Ru2 and Ru3 are both farther away from the O connecting them (O4) than those on the opposite side (O3, O2 respectively) due to Coulomb repulsion. Due to the same reason, the Ru2—O3 bond is longer than the Ru3—O2 bond
| Atom | Wyckoff site | x | y | z | |
| Ba1 | 2a | 0 | 0 | 0 | 0.8(3) |
| Ba2 | 2b | 1/3 | 2/3 | 0.123(33) | 0.8(3) |
| Ba3 | 2b | 1/3 | 2/3 | 0.371(03) | 0.8(3) |
| Ba4 | 2b | 1/3 | 2/3 | 0.746(93) | 0.8(3) |
| Na | 2b | 1/3 | 2/3 | 0.559(93) | 0.8(3) |
| Ru1 | 2b | 1/3 | 2/3 | 0.937(03) | 0.8(3) |
| Ru2 | 2a | 0 | 0 | 0.320(31) | 0.8(3) |
| Ru3 | 2a | 0 | 0 | 0.182(41) | 0.8(3) |
| O1 | 6c | 0.488(03) | 0.511(97) | -0.009(27) | 1.0(2) |
| O2 | 6c | 0.844(21) | 0.155(79) | 0.128(71) | 1.0(2) |
| O3 | 6c | 0.834(51) | 0.165(59) | 0.372(51) | 1.0(2) |
| O4 | 6c | 0.847(91) | 0.152(09) | 0.753(21) | 1.0(2) |
Table 1. Room temperature structural parameters calculated from synchrotron x-ray and neutron powder diffraction
which experiences repulsion from Na instead of the more electropositive Ru. Distortion is present for the NaO octahedra as well, likely due to its large size.
The Ru2—O4—Ru3 superexchange () competes with the direct exchange by favoring a weak ferromagnetic (FM) coupling. There will be an AFM exchange across the Ru1—Ru2 path as well () due to the half filled of Ru. These three magnetic ions are separated by non-magnetic NaO layers. Thus, for the exchange path to extend throughout the crystal, a robust next-nearest-neighbour (NNN) exchange between Ru1 and Ru3 is imperative - most likely a super-superexchange interaction mediated by the NaO anions.
DC magnetic susceptibility measurements were performed under an applied field of kOe in the zero field-cooled (ZFC), field-cooled cooling (FCC) and field-cooled warming (FCW) modes. A clear cusp at K marks the onset of long-range antiferromagnetic order. A significant hysteresis is seen between the magnetization measured during heating and cooling, suggesting the possibility of the transition being of the first-order. Note that the triple-perovskite version of this compound, BaNaRuO, with RuO as the physically meaningful units, exhibits a first-order charge-ordering transition at 210 K, stabilized by a combination of single-ion and molecular degrees of freedom, and accompanied by a symmetry transformation[1, 26]. The Néel temperature, K is calculated by plotting the derivative of the ZFC susceptibility with respect to temperature. Several subtle features can be seen below , consistent with previous reports. Although, qualitatively the curves look different, both Battle et al. [16] and Kim et al. [17] report anomalies in the magnetic susceptibility below . The former reported a showing an irreversibility below , a plateau between K, and another broad peak centered around K. Kim et al. [17], had performed a magnetic field-dependence analysis and came to the conclusion that all the features except for are suppressed at some critical field.
The inverse molar susceptibility as a function of temperature is shown in Figure 5(b) showing Curie-Weiss (CW) behaviour down to K, below which the system enters the transition region. To ensure the correct estimation of CW parameters, the fitting has been performed across the highest measured range between 350 - 385 K (388 points). The Curie constant was calculated to be emu K molOe and the CW temperature K confirming the dominant antiferromagnetic correlations. Kim et al., [17] had performed high magnetization measurements on this material and shown that above 520 K, a CW fit to
Figure 5. Curie-Weiss fit to the inverse magnetic susceptibility () in the high range 350-385 K. The adj. R value attests to the quality of the fit. obtained from the fit demonstrates the dominance of antiferromagnetic interaction and the Ru moment per formula unit is suppressed to a value of .
yields a K and an effective magnetic moment, , nearly equal to the spin-only moment expected for a Ru state (). They noted a change of slope in below K resulting in a significant reduction in K and . This is attributed to the development of AFM correlations within the RuO dimers which lowers the measured bulk magnetization. Since our measurements were limited to 380 K, we likely probe this already correlated paramagnetic phase and calculate the (reduced) per formula unit to be .
Figure 6 shows the isotherms measured at various temperatures between 2 - 400 K. Overall, an antiferromagnetic behaviour is observed with a linear relationship between and without any saturation up to 8 T. A small loop opens up at low temperatures suggestive of weak ferromagnetic correlations. This is in line with the subtle upturn observed in below 12.5 K. The origin of both the rise in and the loop opening at 2 K could be a weak ferromagnetic contribution contributing to the magnetic response at low temperature. Insets of (b) and (c) show a sharp change in the slope at about kOe for K and 35 K. This feature is suppressed at higher temperatures and for K, 200 K and 225 K, subtle step-like features are visible, reminiscent of metamagnetic behaviour. This also manifests in the in the low-field limit (Fig.6(l)) where distinct features can be seen at Oe which are washed out both at higher and lower temperatures.
Figure 6. (a) isotherms measured in the range [-8 T, 8T] for 8 temperatures between 2 - 400 K, (b) - (i) low field regions showing a finite loop and low-field features, insets of (b) and (c) show in the range [-8 T, 8 T] and (l) for T, 0.1 T], (j) and (k) show the coercivity () and Retentivity (or Remnant Magnetization, ) as function of temperature.
Tracking the variation in the coercivity () and remnant magnetization () with temperature brings out the sharp change in both parameters across and subsequent increase below K. This appearance of symmetric dips in at low fields in an intermediate range of temperatures within the antiferromagnetically ordered phase suggests the possibility of a field-induced domain reorientation or a spin-flop like phenomenon. It will be interesting to perform further magnetic field-dependent measurements on this system to closely study its metamagnetic behaviour.
The analysis of structural parameters at various temperatures through synchrotron x-ray and neutron powder shed
light on the variation in bond parameters that lay the foundation of the exchange pathways within this system. The strong spin-lattice coupling is revealed in Figure 7 and Figure 8 where the lattice parameters and the axial ratio () both show an anomaly across . As shown in Figure 2(e), there are several exchange pathways at play here - the intradimer direct exchange between Ru2 and Ru3, the intradimer superexchange Ru2—O4—Ru3, the Ru1—O3—Ru2 superexchange, and the Ru1—O1—O2—Ru3 super-superexchange.
Figure 7. Evolution of lattice constants , and the unit cell volume as a function of temperature (), obtained from -dependent NPD. The antiferromagnetic transition is marked by the feature at 250 K. The error bars are small and marked on the graph.
Figure 8. Thermal evolution of the axial ratio obtained from -dependent SXRD. The antiferromagnetic transition is again seen as the feature at 250 K, along with subtler changes at lower temperatures K and 10 K.
The -dependence of Ru1—O octahedral bonds is shown in Figure 9. The corner shared Ru octahedra is significantly distorted as visible in the difference between the Ru1—O1 and Ru1—O3 bond lengths and shows marked changes across the 250 K in agreement with the magnetic measurements. Both bonds shrink between 270 and 225 K, plateau between 225 and 175 K, continue to further shrink below 175 K till ∼ 100 K below which no -dependence is observed. The superexchange bond angle within the dimer (Ru2—O4—Ru3) and that between the corner sharing Ru ions (Ru1—O3—Ru2) are shown in Figure 10 (a) and (b), respectively. The AFM transition is clearly highlighted in the intradimer angle (Figure 10 (b)) which closely follows the Ru1—O3 bond angle (Figure 9 (b)). These features can be correlated to the subtle changes in the rate of decline of with as seen in Figure 4. The Ru-Ru overlap strengthens on cooling as evidenced by a monotonic decrease in the Ru2—Ru3 bond length (Figure 11). This bond seems to remain unaffected by the antiferromagnetic transition and no corresponding features are observed in the thermal variation of this bond length within error limits.
Figure 4. DC molar susceptibility measured as a function of in ZFC, FCC and FCW protocols showing the antiferromagnetic transition at K along with other subtler features within the AFM phase.
Figure 9. Thermal variation of the Ru1—O octahedral bond lengths, highlighting the AFM transition at ∼ 250 K. The colours of the curves correspond to the colour coding on the bonds shown in the pictorially represented octahedra.
Figure 10. dependence of the (a) inter- and (b) intra-dimer Ru—O—Ru bond angles, also pictorially represented and colour coded to match the colour of the plotted curves. A marked change across 250 K can be seen in each plot.
Figure 11. dependence of the Ru-Ru bond length (visualized in the dimer unit drawn on the plot). The distance between the two Ru ions within the dimer only decreases on cooling and no distinct features corresponding to the AFM transition are observed.
The appearance of new reflections not allowed by the crystal symmetry confirmed the existence of long-range magnetic order in the system. Figure 12 shows comparison between the diffraction patterns collected at 300 and 13 K, clearly highlighting the two additional reflections at
Figure 12. NPD data collected at K superimposed on room temperature data - highlighting the additional reflections - corresponding to the antiferromagnetic structure. The indices of the planes contributing to these intensities are listed above the peaks.
Å and 2.311 Å which are indexed by the k-vector . These are the two most prominent peaks and no other reflections are observed at higher -spacings.
In order to solve the magnetic structure corresponding to these reflections, we performed magnetic representational analysis using the program SARAh [27]. Although the three
Ru sites are symmetrically unique, their site multiplicities are same (each generates two equivalent positions) which means that their magnetic representations () are the same. For each site, we obtained . The point symmetry and Shubnikov Group corresponding to each basis vector for each surviving irreducible representation (IR) is given in Table II.
| IR | BV | Point Group | Shubnikov Group |
| 6 | P6m’c’ (, BNS #186.207) | ||
| 6 | P6’m’c (, BNS #186.205) | ||
| 6 | C/B | ||
| 3m | P1 (, BNS #1.1) | ||
| 6 | C/B | ||
| 3m | P1 (, BNS #1.1) |
Table II. The point symmetries and Shubnikov space groups associated with each basis vector calculated from the parent space group with . C/B indicates that a change of basis is needed to define the Shubnikov group, i.e. the axis definitions no longer match those of the parent structure used for these calculations. For example, a rhombohedral cell in the hexagonal setting with 3 centring translations being changed to a C-centred monoclinic cell with only the C-centring.
The best fit to the magnetic peaks is obtained using the IR corresponding to the magnetic space group (No. 186.205), implying that the time-reversal symmetry is broken along the and the axes due to magnetic ordering. The Rietveld fit of this model to the observed diffraction pattern at 13 K is shown in Figure 13(a). The accuracy of the model is confirmed by the reasonable magnetic R-factor (}) of 6.33% and the fits to the magnetic reflections (Figure 13(b) and (c)). It is worth noting that the choice of IR was made based on the best fit obtained with freely varying Ru moments. This unconstrained model had a slightly better () but produced a structure assigning nearly identical moment sizes on Ru1 and Ru2 sites while leaving Ru3 with just about half that value. This is physically unlikely since Ru2 and Ru3 are much closer to each other and are likely to have a strong AFM correlation. Constraining all Ru sites to have the same moment values resulted in a structure with ferromagnetic alignment
Figure 13. (a) Rietveld fit of the IR magnetic model (magnetic space group ) to neutron powder diffraction data collected at 13 K. (b), (c) Zoomed plots showing agreement of the model to the magnetic reflections.
within the dimers (), inconsistent with magnetic data. Constraining Ru2 and Ru3 moments to have the same magnitude, keeping the same orientations as obtained from the unconstrained model, resulted in identical visual fits, with only a minor increase in the (5.7 6.3 %) indicative of a physically sensible model. In contrast, for and were found to be 13.5% and 117%, respectively; while led to divergence of the fit.
The magnetic structure (Figure 14(a)) exhibits an A-type arrangement of spins which are ferromagnetically coupled within each layer, while AFM coupling is present between the NN layers. All the spins are aligned collinear to the crystalline easy axis , a testament to the strong influence of the structure on the magnetic order. An interesting aspect of this structure is shown in Figure 14(b) and (c), which is the difference in the packing of the different layers. If the Ru1 and Ru2,3 sublattices are separated out, one can observe that the Ru1 layers have a staggered arrangement owing to the axis running along , while the Ru2,3 have a cubic packing. Similar A-type orderings with spins locked in the -plane, have been found in other layered perovskite systems with cubic packing like BaFe(W,Te)O[28, 29] and NdBaMnO[30]. BaCoTeO also shows an A-type order but with spins canted at an angle of 24.5° with respect to the -axis[31]. BaNdRuO shows a FM order of Nd spins collinear with the -axis below 24 K which cant below 18 K accompanied by an AFM order of the dimer Ru spins[32]. This exemplifies the vast variation in magnetic structures possible and its strong dependence on the underlying crystal structure and the type of TM ions making up the exchange network. We note that this work is the first conclusive solution of the ground state magnetic structure of of a B-site substituted non-trimer quadruple perovskite.
Figure 14. (a) The A-type AFM structure of BaNaRuO at 13 K. (b) The staggered arrangement of the Ru1 triangular layers, AFM coupled to each other along . (c) The cubic packing of the Ru, dimers - characterized by intra-dimer and inter-layered AFM coupling. Both sublattices are intra-layer FM coupled.
Fitting the model to NPD patterns collected at different temperatures allows us to track the development of the magnetic phase as a function of temperature. The rise in moment values at each magnetic site with cooling, is shown in Figure 15. Both Ru moments record a qualitatively similar curve with Ru1 saturating at and Ru2 (= Ru3) at . We fit the growth of moment below to a power-law function to ascertain the magnetic order parameter exponent and the temperature of the phase transition. The mean-field equation is:
Figure 15. Growth in Ru magnetic moment on cooling. Moment values at each obtained from fitting -dependent NPD data to the magnetic space group . curves fitted to mean field equation (shown on plot) to obtain the critical exponent and .
where is the magnetic moment at , the temperature of the phase transition and the critical exponent capturing the behaviour of the order parameter () at criticality. For Ru1, the parameters obtained are: =2.38±0.03 and for Ru2 (=Ru3), =1.61±0.03, elucidating the effect of Ru-Ru direct interaction within the RuO facilitating the suppression of the individual Ru values. The critical exponent for Ru1 sites is = 0.19 ± 0.02, and that for Ru2,3 sites is = 0.24 ± 0.02 (using =250.18 K from the fit) placing BaNaRuO well within the universality window of two-dimensional critical exponents[33], with its lower limit marked by the Ising 2D model ( = 1/8 = 0.125). Values of for 3D magnetic models is expected to be much larger (), while Fe(NCS)(pyrazine), a 2D molecular antiferromagnet is reported to have the exact same value of [34]. Similar values of and are reported for PdCrO, an triangular lattice antiferromagnet[35], and LaCoO, a 2D layered material with longitudinal Ising spin correlations[36]. This attests to the layered, quasi-2D nature of the magnetism in BaNaRuO with ferromagnetic triangular layers of Ru ions are antiferromagnetically linked along the easy-axis () with out-of-plane collinear spins.
The specific heat measured as a function of temperature () is shown in Figure 16(a). A clear -type anomaly is visible at 255 K, confirming the antiferromagnetic transition at . The small magnitude of the peak likely suggests that the entropy change associated with this transition is small. This observation also matches up with the absence of any symmetry lowering in the crystal structure of the compound as a function of temperature. Inset (i) of Figure 16 shows the data in the full range highlighting the absence of any more features at lower temperature. This probably means that the octahedral distortions only affect the magnetic ground states and the overall energy of the state remains impervious to those local changes. Inset (ii) of Figure 16 highlights the region where Debye behaviour is observed, and a linearity is captured in the vs plot. At high temperature can be seen to approach the Dulong-Petit Limit, as expected.
Figure 16. (a) Temperature dependence of specific heat capacity of BaNaRuO in the vicinity of . Inset (i) shows for the whole range (2 - 300 K) and inset (ii) shows a plot of vs showing the range of temperatures for which the Debye term is dominant. (b) A plot of vs with the linear fit along with the parameters obtained from the fit. (c) A graph showing the variation in as a function of .
Figure 16(b) shows the low-temperature linear fit to vs , to estimate the electronic and lattice contributions to the specific heat. We obtain a Debye Temperature K using
consistent with reports for other layered perovskites [37, 38]. Here, is the number of atoms per formula unit (20 for quadruple perovskites) and the gas constant. The values obtained for mJ molK and mJ molK are both minuscule, implying that most of the entropy of the system is lost in approaching low temperatures. This is further corroborated in the plot of vs where a peak in is observed at K. This excess phonon specific heat is a common feature in many perovskite systems and is attributed to the presence of low-lying optical (Einstein) modes that cause a departure from the phononic density of states and promote a pure dependence of at . The position of the peak is said to be a manifestation of the boson peak [39] and can be used to estimate the vibration energy of the largest contribution to at the peak temperature under the dominant phonon approximation [40]. For the observed peak in at K, we obtain a vibrational energy of cm (using )[40].
Figure 17(a) shows the semiconducting nature of the material along with a small discontinuity observed in the derivative at . The charge transport shows a thermally activated behaviour above with a linear against . Figure 17(b) shows the linear fit in the range 333 - 287 K. Deviation from this linearity begins close to the magnetic transition
Figure 17. Transport behaviour of BaNaRuO. (a) vs , showing a semiconducting behaviour, and the derivative shows the discontinuity at . (b) Linear fit to vs for . Insets show the deviation from linearity and the sharp change in the slope below . (c) Linear fit to the Efros-Shklovskii Variable Range Hopping (ES-VRH) model below .
and is depicted more clearly in the zoomed inset of Figure 17(b) and the derivative plot vs marking the onset of a declining slope below . The linear fit returned an activation energy of meV and the minimum resistivity m-m.
Below , when was found to be non-linear with respect to Mott’s Variable Range Hopping (VRH) behaviour [41], we employed the generalized VRH law given by:
where and define the minimum resistivity at , and characteristic temperature of the hopping behaviour, respectively. is the hopping exponent, which takes the value 1/3 and 1/4 in Mott’s law [41] for 2D and 3D systems, respectively, having noninteracting electrons and constant density of states . However, in real materials, electronelectron interactions are often important and their inclusion modifies the Mott-VRH to the Efros-Shklovskii (ES) VRH [42] with for both 2D and 3D systems. The density of states in this model has a quadratic dependence on the energy close to the Fermi energy () given by[43, 44]:
with in a 3D semiconductor where, is the dielectric constant of the material and encapsulates the effect of Coulomb interaction between the electrons. This leads to the opening of a soft gap in the single particle density of states, the width of which is known as the Coulomb gap. Fitting the to returns a reasonably linear fit across 7-8 orders of magnitude, however, the value K obtained is significantly high. Similar results, with even higher values of have been reported for the 5d double perovskites AMnReO (A = Ba, Sr, Ca)[45].
Now, the hopping exponent can be a priori extracted from the vs curve by plotting vs , the slope of which returns [46]. As shown in Figure 18(a), the temperature regimes and are defined by two different slopes. For , we find , as discussed earlier, and for , , which is higher than the ES value. In a linearized plot (Figure 18(b)), this hopping exponent yields a significantly better fit. Generally, the exponent describing the dependence of on () is related to the hopping exponent as: [47], so that for the ES model, when , and . In the present case, the obtained value of yields the relation , a far stronger dependence than proposed by ES-VRH. This scenario is likely if multi-particle excitations due to correlation effects are considered instead of the single-particle consideration of the ES model. [48] Numerical calculations also have suggested that the Coulomb gap can have a quasiexponential energy dependence in presence of multi-electron transitions [49]. This attests to the strongly insulating low temperature state, where the total energy available to the system is greatly reduced and thus, so are the density of states close to Fermi level, leaving the system with little possibility to conduct. Another interesting aspect here is the lack of existence of the Mott regime. Generally, a gradual crossover from Mott to ESVRH regime is observed on cooling, and the Mott and ES characteristic temperatures and are related to the Coulomb gap width () as
Figure 18. (a) log-log plot of showing the two distinct transport regimes characterized by different slopes (). (b) vs yielding an improved linear fit with a lower
[50]. But here, we bypass the Mott regime entirely to reach the ES hopping regime.
Conclusions
A thorough investigation of BaNaRuO via structural, magnetic and thermal and electronic transport measurements reveals a robust spin-lattice coupling in its ground state. The system consists of a unique motif uncommon in layered perovskite systems - of three symmetrically unique magnetic ions occupying the centre of a corner octahedra connected to a dimer, isolated by a non-magnetic layer. Despite this separation of magnetic entities, it exhibits a long-range magnetic ordering at a fairly high temperature of K which is corroborated in both specific heat as well as temperature-dependent diffraction measurements, attesting to the strong coupling between the magnetic and lattice degrees of freedom, and the robustness of the exchange mechanism. The magnetically ordered ground state is visualized by the neutron powder diffraction measurements at 13 K and the Ru spins are seen to couple antiferromagnetically along axis and ferromagnetically within the plane. Thermal dependence of structural parameters indicates the presence of substantial spin-lattice coupling. The magnetization isotherms show an interesting metamagnetic feature in the intermediate -range of 125 - 225 K, concurring with the region where the Ru1—O6 show the maximum change - pointing to some type of spin reorientation. This places Ba4NaRu3O12 as an interesting material candidate for a theoretical investigation of the competing interactions influencing its ground state and the reason behind the strength of the antiferromagnetic exchange interaction. The low- specific heat response is dominated by Einstein (optical) mode vibrations attested to by the peak seen in plots and the diminished electronic and Debye terms ( and ). The charge transport shows a transition from a thermally activated behaviour above to a VRH transport with , implying the presence of a soft Coulomb gap with a stronger energy-dependence than the quadratic dependence suggested by the Efros-Shklovskii model with . This suggests that multiparticle effects might be active in this system due to strong correlation effects.
Acknowledgments
Ajit Patra, Abhinav Kumar Khorwal and Sujoy Saha are acknowledged for resistivity and magnetic measurements on the PPMS facility at the Central University of Rajasthan. Vikram Singh and Arun Kumar are acknowledged for helping with performing resistivity measurements and synchrotron XRD measurements respectively. Dibyata Rout is acknowledged for helping with heat capacity measurements. SC and SN acknowledge funding from an Air Force Research Laboratory Grant (FA2386-23-1-4046).
References
References
- [1]S. A. Kimber, M. S. Senn, S. Fratini, H. Wu, A. H. Hill, P. Manuel, J. P. Attfield, D. N. Argyriou, and P. F. Henry, Charge order at the frontier between the molecular and solid states in Ba₃NaRu₂O₉, Physical Review Letters 108, 167201 (2012).DOI
- [2]D. Ziat, A. A. Aczel, R. Sinclair, Q. Chen, H. D. Zhou, T. J. Williams, M. B. Stone, A. Verrier, and J. A. Quilliam, Frustrated spin-1/2 molecular magnetism in the mixed-valence antiferromagnets Ba₃MRu₂O₉ (M=In, Y, Lu), Physical Review B 95, 184424 (2017).DOI
- [3]Q. Chen, A. Verrier, D. Ziat, A. J. Clune, R. Rouane, X. Bazier-Matte, G. Wang, S. Calder, K. M. Taddei, C. R. d. Cruz, A. I. Kolesnikov, J. Ma, J.-G. Cheng, Z. Liu, J. A. Quilliam, J. L. Mulfeldt, H. D. Zhou, and A. A. Aczel, Realization of the orbital-selective Mott state at the molecular level in Ba₃LaRu₂O₉, Physical Review Materials 4, 064409 (2020).
- [4]I. Terasaki, T. Igarashi, T. Nagai, K. Tanabe, H. Taniguchi, T. Matsuita, N. Yada, A. Takata, T. Kida, M. Hagiwara, K. Kobayashi, H. Sagayama, R. Kumai, H. Nakao, and Y. Murakami, Absence of magnetic long range order in Ba₃ZnRu₂O₉: A spin-liquid candidate in the S = 3/2 Dimer Lattice, Journal of the Physical Society of Japan 86 (2017).DOI
- [5]M. Ahmadipour, M. F. Ain, and Z. A. Ahmad, A short review on copper calcium titanate (ccto) electroceramic: synthesis, dielectric properties, film deposition, and sensing application, Nano-Micro Letters 8, 291 (2016).DOI
- [6]W. Liu, K. Kawano, M. Kamiko, Y. Kato, Y. Okazaki, I. Yamada, and S. Yagi, Effects of A-site Cations in Quadruple Perovskite Ruthenates on Oxygen Evolution Catalysis in Acidic Aqueous Solutions, Small 18, 2202439 (2022), https://onlinelibrary.wiley.com/doi/pdf/10.1002/smll.20220439.
- [7]A. M. Glazer, The classification of tilted octahedra in perovskites, Acta Crystallographica Section B 28, 3384 (1972).DOI
- [8]A. A. Belik, D. D. Khalyavin, Y. Matsushita, and K. Yamaura, Triple A-site cation ordering in the ferrimagnetic Y₂CuGaMn₄O₁₂ perovskite, Inorganic Chemistry 61, 14428 (2022).DOI
- [9]L. T. Nguyen, T. Halloran, W. Xie, T. Kong, C. L. Broholm, and R. J. Cava, Geometrically frustrated trimer-based mott insulator, Phys. Rev. Mater. 2, 054414 (2018).DOI
- [10]Y. Shimoda, Y. Doi, M. Wakeshima, and Y. Hinatsu, Magnetic and electrical properties of quadruple perovskites with 12 layer structures Ba₄LnM₃O₁₂ (Ln= rare earths; M= Ru, Ir): The role of metal–metal bonding in perovskite-related oxides, Journal of Solid State Chemistry 183, 1962 (2010).
- [11]Y. Shimoda, Y. Doi, Y. Hinatsu, and K. Ohoyama, Synthesis, crystal structures, and magnetic properties of new 12l-perovskites ba₄lnru₃o₁₂ (ln= lanthanides), Chemistry of Materials 20, 4512 (2008).DOI
- [12]B. Yuan, B. H. Kim, Q. Chen, D. Dobrowolski, M. Azmanska, G. Luke, S. Fan, V. Bisogni, J. Pelliciari, and J. Clancy, Highly Tunable Ru-dimer Molecular Orbital State in 6H-perovskite Ba₃MRu₂O₉, arXiv preprint arXiv:2405.09418 (2024).
- [13]T. Fang, F. Y. Wei, Q. Zhang, Y. Fang, and X. Ke, Complex magnetic structure in Ba₅Ru₃O₁₂ with isolated Ru₃O₁₂ trimer, Phys. Rev. Mater. 4, 114401 (2020).DOI
- [14]Y. Klein, G. Rousse, F. Damay, F. Porcher, G. André, and I. Terasaki, Antiferromagnetic order and consequences on the transport properties of Ba₄Ru₃O₁₀, Phys. Rev. B 84, 054439 (2011).DOI
- [15]G. Cao, H. Zheng, H. Zhao, Y. Ni, C. A. Pocs, Y. Zhang, F. Ye, C. Hoffmann, X. Wang, M. Lee, M. Hermele, and I. Kimchi, Quantum liquid from strange frustration in the trimer magnet Ba₄Ir₃O₁₀, npj Quantum Materials 5, 26 (2020).DOI
- [16]P. Battle, S. Kim, and A. Powell, The crystal structure and electronic properties of Ba₄Ru₃O₁₀ (M = Li, Na, Mg, Zn), Journal of Solid State Chemistry 101, 161 (1992).
- [17]I.-S. Kim, T. Nakamura, and M. Itoh, Magnetic Properties of 8H-Type Hexagonal Ba(Na₁/₄Ru₃/₄)O₃, Bulletin of the Chemical Society of Japan 71, 2247 (1998).DOI
- [18]K. E. Stitzer, W. R. Gemmill, M. D. Smith, and H.-C. zur Loye, Crystal growth of a novel oxygen-deficient layered perovskite: Ba₃Li₃Ru₄O₂₀, Journal of Solid State Chemistry 175, 39 (2003), special Issue on Metal Oxides - From Proceedings of the ACS Symposium on Metal Oxides.
- [19]P. Willmott, D. Meister, S. Leake, M. Lange, A. Bergamaschi, M. Böge, M. Calvi, C. Cancellieri, N. Casati, A. Cervellino, et al., The materials science beamline upgrade at the swiss light source, Journal of synchrotron radiation 20, 667 (2013).DOI
- [20]J. Rodriguez-Carvajal, Fullprof: a program for rietveld refinement and pattern matching analysis, in satellite meeting on powder diffraction of the XV congress of the IUCr, Vol. 127 (Toulouse, France, 1990).
- [21]K. Momma and F. Izumi, VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data, J. Appl. Crystallogr 44, 1272 (2011).
- [22]L. C. Chapon, P. Manuel, P. G. Radaelli, C. Benson, L. Perrott, S. Ansell, N. J. Rhodes, D. Raspino, D. Duxbury, E. Spill, et al., Wish: The new powder and single crystal magnetic diffractometer on the second target station, Neutron News 22, 22 (2011).DOI
- [23]R. D. Shannon, Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides, Acta Crystallographica Section A 32, 751 (1976).DOI
- [24]A. Jain, S. P. Ong, G. Hautier, W. Chen, W. D. Richards, S. Dacek, S. Cholia, D. Gunter, D. Skinner, G. Ceder, and K. A. Persson, The Materials Project: A materials genome approach to accelerating materials innovation, APL Materials 1, 011002 (2013).DOI
- [25]O. C. Gagné and F. C. Hawthorne, Bond-length distributions for ions bonded to oxygen: results for the transition metals and quantification of factors underlying bond-length variation in inorganic solids, IUCrJ 7, 581 (2020).DOI
- [26]K. E. Stitzer, M. D. Smith, W. R. Gemmill, and H.-C. zur Loye, Novel mixed-valent (v/vi) triple perovskite ruthenates: Observation of a complex low-temperature structural and magnetic transition, Journal of the American Chemical Society 124, 13877 (2002).
- [27]A. Wills, A new protocol for the determination of magnetic structures using simulated annealing and representational analysis (SARAh), Physica B: Condensed Matter 276, 680 (2000).DOI
- [28]Y. Tang, R. P. Sena, M. Avdeev, P. D. Battle, J. Cadogan, J. Hadermann, and E. C. Hunter, Magnetic properties of the 6H perovskite Ba₃Fe₂TeO₉, Journal of Solid State Chemistry 253, 347 (2017).DOI
- [29]S. Ivanov, S.-G. Eriksson, J. Erikssen, R. Tellgren, and H. Rundlof, Nuclear and magnetic structure of Ba₃Fe₂WO₉, Materials research bulletin 39, 615 (2004).DOI
- [30]D. Akahoshi, Y. Okimoto, M. Kubota, R. Kumai, T. Arima, Y. Tomioka, and Y. Tokura, Charge-orbital ordering near the multicritical point in A-site ordered perovskites SmBaMn₂O₆ and NdBaMn₂O₆, Physical Review B—Condensed Matter and Materials Physics 70, 064418 (2004).
- [31]S. Ivanov, P. Nordblad, R. Mathieu, R. Tellgren, and C. Ritter, Neutron diffraction studies and the magnetism of an ordered perovskite: Ba₂CoTeO₆, Dalton Transactions 39, 5490 (2010).arxiv.org/abs/1011.6637
- [32]M. S. Senn, S. A. Kimber, A. M. Arevalo Lopez, A. H. Hill, and J. P. Attfield, Spin orders and lattice distortions of geometrically frustrated 6H-perovskites Ba₃B′Ru₂O₉ (B′ = La³⁺, Nd³⁺, and Y³⁺), Physical Review B—Condensed Matter and Materials Physics 87, 134402 (2013).
- [33]A. Taroni, S. T. Bramwell, and P. C. Holdsworth, Universal window for two-dimensional critical exponents, Journal of Physics: Condensed Matter 20, 275233 (2008).
- [34]H. N. Bordallo, L. Chapon, J. L. Manson, J. Hernández-Velasco, D. Ravot, W. M. Reiff, and D. N. Argyriou, s = 1/2 Ising behavior in the two-dimensional molecular magnet Fe(NCS)₂(pyrazine)₂, Phys. Rev. B 69, 224405 (2004).
- [35]H. Takatsu, H. Yoshizawa, S. Yonezawa, and Y. Maeno, Critical behavior of the metallic triangular-lattice Heisenberg antiferromagnet PdCrO₂, Physical Review B—Condensed Matter and Materials Physics 79, 104424 (2009).DOI
- [36]K. Yamada, M. Matsuda, Y. Endoh, B. Keimer, R. J. Birgeneau, S. Onodera, J. Mizusaki, T. Matsuura, and G. Shirane, Successive antiferromagnetic phase transitions in single-crystal La₂CoO₄, Phys. Rev. B 39, 2336 (1989).pubmed.ncbi.nlm.nih.gov/9948473
- [37]T. Igarashi, R. Okazaki, H. Taniguchi, Y. Yasui, and I. Terasaki, Effects of the Ir Impurity on the Thermodynamic and Transport Properties of Ba₄Ru₃O₁₀, Journal of the Physical Society of Japan 84, 094601 (2015), https://doi.org/10.7566/JPSJ.84.094601.
- [38]G. Radtke, A. Saúl, Y. Klein, and G. Rousse, Magnetism of Ba₄Ru₃O₁₀ revealed by density functional calculations: Structural trimers behaving as coupled magnetic dimers, Phys. Rev. B 87, 054436 (2013).
- [39]G. D’Angelo, G. Carini, C. Crupi, M. Koza, G. Tripodo, and C. Vasi, Boson peak in alkaline borate glasses: Raman spectroscopy, neutron scattering, and specific-heat measurements, Phys. Rev. B 79, 014260 (2009).
- [40]R. Berman, Thermal Conduction in Solids, Oxford studies in physics (Clarendon Press, 1976).
- [41]N. F. Mott, Conduction in non-crystalline materials: III. Localized states in a pseudogap and near extremities of conduction and valence bands, Philosophical Magazine 19, 835 (1969).
- [42]A. L. Efros and B. I. Shklovskii, Coulomb gap and low temperature conductivity of disordered systems, Journal of Physics C: Solid State Physics 8, L49 (1975).DOI
- [43]B. Shklovskii and A. Efros, Electronic Properties of Doped Semiconductors, Solid-State Sciences Series (Springer-Verlag, 1984).DOI
- [44]B. I. Shklovskii, Half century of Efros-Shklovskii Coulomb gap. Romance with Coulomb interaction and disorder (2024), arXiv:2403.19793 [cond-mat.mtrl-sci].arxiv.org/abs/2403.19793
- [45]B. Fisher, J. Genossar, K. B. Chashka, A. Knizhnik, L. Patlagan, and G. M. Reisner, Variable range hopping in A₂MnReO₆ (A=Ca,Sr,Ba), Journal of Applied Physics 104, 033716 (2008).DOI
- [46]J. Massey and M. Lee, Direct observation of the coulomb correlation gap in a nonmetallic semiconductor, Si: B, Physical Review Letters 75, 4266 (1995).pubmed.ncbi.nlm.nih.gov/10059861
- [47]M. Pollak, A percolation treatment of dc hopping conduction, Journal of Non-Crystalline Solids 11, 1 (1972).
- [48]M. Pollak, Interaction effects on low temperature hopping, Journal of Non-Crystalline Solids 35, 83 (1980).DOI
- [49]J. Davies, P. Lee, and T. Rice, Properties of the electron glass, Physical Review B 29, 4260 (1984).DOI
- [50]R. Rosenbaum, Crossover from Mott to Efros-Shklovskii variable-range-hopping conductivity in InₓOᵧ films, Phys. Rev. B 44, 3599 (1991).pubmed.ncbi.nlm.nih.gov/9999988