Role of Antisite Disorder, Rare-Earth Size, and ... - ACS Publications

Article Cite This: ACS Appl. Electron. Mater. XXXX, XXX, XXX−XXX

pubs.acs.org/acsaelm

Role of Antisite Disorder, Rare-Earth Size, and Superexchange Angle on Band Gap, Curie Temperature, and Magnetization of R2NiMnO6 Double Perovskites Mohd Nasir,† Sunil Kumar,‡ Nirmalendu Patra,§ Dibyendu Bhattacharya,§ Shambhu Nath Jha,§ Dharma R. Basaula,∥ Subhash Bhatt,∥ Mahmud Khan,∥ Shun-Wei Liu,⊥ Sajal Biring,*,⊥ and Somaditya Sen*,†,⊥

ACS Appl. Electron. Mater. Downloaded from pubs.acs.org by LINKOPING UNIV on 01/07/19. For personal use only.

†

Department of Physics and ‡Metallurgy Engineering and Materials Science, Indian Institute of Technology Indore, Indore 453552, India § Atomic & Molecular Physics Division, Bhabha Atomic Research Centre, Mumbai 400085, India ∥ Department of Physics, Miami University, Oxford, Ohio 45056, United States ⊥ Department of Electrical Engineering, Ming Chi University of Technology, New Taipei 24301, Taiwan ABSTRACT: Homogeneous solid solutions of sol−gel-prepared R2NiMnO6 (R = La, Pr, Nd, Sm, Gd, Tb, Dy, Y, and Ho) double perovskites crystallize in a B-site-ordered monoclinic structure (P21/n space group). Monoclinic distortion enhances with decreasing R3+ ionic radii (rR3+). The magnetic ordering temperature (TC) decreases from 270 K for La2NiMnO6 to 80 K for Ho2NiMnO6 as rR3+ decreases from 1.16 Å (La3+) to 1.02 Å (Ho3+). An additional magnetic anomaly is observed in Nd2NiMnO6, Sm2NiMnO6, Tb2NiMnO6, and Dy2NiMnO6 at lower temperatures, which originates from the 3d−4f coupling between Mn−Ni and Nd3+/Sm3+/Tb3+/Dy3+ magnetic moments. Further, high saturation magnetization is achieved for all samples, indicating that they are atomically ordered and have less antisite disorders. Upon a decrease in the size of R3+, the local structure shows an expansion of NiO6 octahedra and almost unchanged of MnO6 octahedra. X-ray-absorption near-edge spectroscopy reveals a majority of Ni2+ and Mn4+ ions in all samples. Softening of phonon modes results in the elongation of the Ni/Mn−O bond length. Finally, a correlation among lattice parameters, structural distortion, octahedral tilting, superexchange angle, electronic band gap, Curie temperature, and the rare-earth ionic radius is established. KEYWORDS: superexchange interaction, XANES/EXAFS, Curie temperature, antisite disorders, octahedral tilting, rare-earth ion size, band gap, R2NiMnO6, La2NiMnO6 properties can be tailored by changing the radii (rR3+) of the R3+ ions as well as by applying external pressure.13 This enables their usage in fabricating novel magnetic, electronic, and multiferroic superlattices/systems. Depending on the crystallographic site occupancies of Ni and Mn cations, these systems may adopt two different crystal structures. A random arrangement of Ni and Mn ions leads to an orthorhombic crystal structure, while an alternating periodic arrangement of Ni and Mn cations results in the monoclinic symmetry belonging to the P21/n space group.14−16 The R2NiMnO6 double perovskites exhibit high magnetic ordering temperatures.13 The magnetic ordering originates from the virtual hopping of electrons between the half-filled orbital of Ni2+ and the vacant orbital of Mn4+

I. INTRODUCTION Double perovskites R2BB′O6 (R is a rare-earth ion; B and B′ are transition metals) are important materials today because of their rich physical properties and technological application.1−8 Multifunctional properties of these structures include colossal magnetoresistance, magnetocapacitance, and multiferroicity. Spin phonon coupling gives rise to such exotic properties in R2BB′O6 compounds. Such coupling strongly depends on the R and B/B′-site ordering.6−10 Magnetocapacitance was observed in La 2 NiMnO 6 (LNMO).11 Since then, rare-earth-based double perovskites have been intensively investigated. LNMO is a potential ferromagnetic semiconductor with coupled magnetic and electrical properties near room temperature (TC ≈ 280 K).11,12 Ferromagnetic semiconductor oxides, R2NiMnO6 (RNMO), are important materials due to their promising role in the nextgeneration spintronic industry and their rich physics. Their © XXXX American Chemical Society

Received: November 15, 2018 Accepted: December 21, 2018

A

DOI: 10.1021/acsaelm.8b00062 ACS Appl. Electron. Mater. XXXX, XXX, XXX−XXX

Article

ACS Applied Electronic Materials

ground using agate mortar and pestle and thereafter calcined at 600 °C to get rid of trapped polymer and nitrates. The resulting powder was again ground and reheated at 900 °C for 12 h. After that, the powders were mixed with 5% PVA solution and pressed into pellets of 10 mm diameter and 1.2 mm thickness using a uniaxial hydraulic press. A pressure of ∼250 MPa was applied to make pellets. The pellets were sintered at 1150 °C for 12 h to burn the binder (PVA). Finally, another heating at 1350 °C for 24 h was carried out to ensure compactness. The R2NiMnO (RNMO) samples are labeled as LNMO (La2NiMnO6), P N M O ( Pr 2 NiM nO 6 ), NNM O ( Nd 2 N i Mn O 6 ) , S N M O (Sm2NiMnO6), GNMO (Gd2NiMnO6), TNMO (Tb2NiMnO6), DNMO (Dy 2 NiMnO 6 ), YNMO (Y 2 NiMnO 6 ), and HNMO (Ho2NiMnO6). The phase identification and purity of these materials were investigated from powder X-ray diffraction patterns (XRD) collected using a Bruker D2 Phaser X-ray diffractometer with Cu Kα radiation. Rietveld refinement of XRD data using the FullProf software package27,28 was performed to extract the lattice parameters and bond angles. Room temperature Raman spectra for R2NiMnO6 samples were measured with an excitation wavelength of 478 nm using a JobinYvon Horiba LABRAM-HR visible micro-Raman spectrometer. Temperature- and field-dependent magnetization (M−T and M−H) of the samples were examined on a Quantum Design Physical Property Measurement System (PPMS). Temperature-dependent dc magnetization, M(T), in the temperature range of 4−300 K, was performed at a constant magnetic field of 100 Oe for all R2NiMnO6 samples. Samples were cooled in absence of magnetic field from room temperature down to 4 K. At 4 K, a steady magnetic field was applied. As the temperature was increased from 4 K to room temperature at a slow rate of 0.1 K/min, a zero-field cooling, ZFC, M(T) measurement was performed in the stable magnetic field of ∼100 Oe. The samples were further cooled from room temperature to 4 K to record the field-cooled cooling, FCC, M(T) magnetization. Room temperature X-ray-absorption spectroscopy (XAS) measurements in fluorescence mode have been performed at the Energy Scanning EXAFS beamline BL-09 at Indus-2, RRCAT Indore, for measuring the Ni (8333 eV) and Mn (6539 eV) K-edges of RNMO. The electron storage ring operates at 2.5 GeV and 300 mA, while the beamline works in the energy range of 4−25 keV. The energy resolution of the beamline (ΔE/E) was about 10−4. In fluorescence mode, the sample was positioned at 45° to the incident X-ray beam. An ionization chamber measures the incident beam (I0), while the fluorescence signal (If) is measured using a Si drift detector placed at 90° to the incident X-ray beam. The X-ray-absorption coefficient of the sample is determined by μ = If/I0, and the spectrum was obtained as a function of energy by scanning the monochromator over the specified range. For energy calibration, standard metal foils of Ni and Mn were used. The structure EXAFS signal χ(E) was calculated from the absorption coefficient μ(E) defined as

cations.17 The superexchange interactions between the Ni and Mn ions, which control the magnetic ordering in RNMO, are governed by the Goodenough−Kanamori rule.16,18−20 The magnetic interactions in R2NiMnO6 are also strongly influenced by rR3+. Lattice distribution of Mn and Ni ions determines the superexchange interaction and hence the crystal structure. On decreasing rR3+, the ferromagnetic transition temperature decreases gradually, while the structure is distorted. Additionally, small antiferromagnetic clusters can be formed by the less probable +3 oxidation states of Ni/Mn, antisite disorders, and antiphase boundaries leading to Ni2+−Ni2+ or Mn4+−Mn4+ interactions, which often result in the spin-glass state and additional magnetic transitions. The magnetic exchange interaction is associated with the deviation in octahedral tilting, Ni/Mn−O bond lengths, and Ni−O−Mn bond angles. This kind of mechanism is well established for rare-earth-based perovskites such as RCrO3 and R2CoMnO6.21,22 R2NiMnO6 (R ≠ La) compounds have not been investigated in great detail. The electronic and magnetic properties of these materials, such as the saturation magnetization, magnetic ordering temperature (TC), and the electronic band gap vary considerably from one report to the other.14−16,23 Different conclusions have been drawn regarding probable mechanisms that govern magnetism in these materials. The reason for these differences may be a result of the various synthesis routes that significantly influence the crystallographic structures and the extent of the antisite disorder at the Ni/Mn sites. Also, local atomic structure around Ni and Mn ions is a key factor in determining various physical properties such as magnetic ordering and electronic band gap in manganites.24,25 For RNMO, the local structure and oxidation states of Ni and Mn ions are not well established. The detailed knowledge of the oxidation and spin states of Mn and Ni ions is critical to understanding the broad magnetic response of these systems. To examine the short-range order and the physics of the RNMO series, techniques which directly probe local atomic structure, such as X-ray-absorption spectroscopy (XAS), have been employed.26 Uniformly distributed, homogeneous, and singlephase materials are needed to understand the actual R3+ dependence of magnetization and its mechanism. Hence, nine members of the R2NiMnO6 (R = La, Pr, Nd, Sm, Gd, Tb, Dy, Y, and Ho) family were synthesized by a sol−gel-assisted combustion method. A comprehensive investigation of the effect of the size of the rare-earth ion on the structure is presented in this report, detailing the magnetic ground states. An interesting variation of structure-correlated electronic properties and competing magnetic interactions have been discussed.

χ (E) =

II. EXPERIMENTAL DETAILS

μ(E) − μ0 (E) Δμ0 (E0)

(1)

where, E0 is absorption edge energy, μ0(E0) is the bare atom background, and Δμ0(E0) is the step in the μ(E) value at the absorption edge (difference between the pre-edge and the postedge energy spectra). Finally, the energy-dependent EXAFS function χ(E) was converted to wavenumber-dependent function χ(k) by

Polycrystalline R2NiMnO6 samples were prepared by a sol−gel-assisted combustion method. For each R2NiMnO6 sample, stoichiometric amounts of the R(NO3)3·6H2O (R = La, Pr, Nd, Gd, Dy, Y; Alfa Aesar), Ni(NO3)2·6H2O (Alfa Aesar), and Mn(NO3)2 (50% w/w aqueous solution; Alfa Aesar) were mixed together in doubly deionized water. Sm2O3 (Alfa Aesar), Ho2O3 (Alfa Aesar), and Tb4O7 (Alfa Aesar) were dissolved separately in doubly deionized water with the addition of a few drops of nitric acid. The mixed solution was stirred vigorously for 40 min on a hot plate at ∼50 °C to ensure the through solubility. A gel former was prepared by mixing citric acid and ethylene glycol in a molar ratio of 1:1. This gel former was poured into the clear solution. The resulting solution was heated at 85 °C for 5 h on the hot plate with continuous stirring. After gel formation, the gels were burnt in open air at 100 °C to yield black fluffy powders. The obtained powders were

k = 2m(E − E0)/ℏ2 . After being weighted by k2, which is used to compensate for the damping of the EXAFS amplitude with increasing k, the χ(k) data was Fourier transformed into r space, in the range of 0 ≤ k ≤ 10 Å. The background reduction of the experimental EXAFS data has been done by using the Athena module. The theoretical EXAFS spectra were generated using the FEFF 6.0 code, while the fitting of the experimental EXAFS data with the theoretical spectra have been carried out using the ARTEMIS code of IFEFFIT software packages.29 B


Article


Figure 1. (a) Normalized X-ray diffraction patterns of R2NiMnO6 (R = La, Pr, Nd, Sm, Gd, Tb, Dy, Y, and Ho). All the peaks were indexed to the monoclinic cell belonging to the P21/n space group. (b) Magnified view of peaks in the 2θ ≈ 32−35° range; the peaks (020), (200), and (021) gradually evolve with decreasing rare-earth ion size (rR3+). (c) Intensity ratio (I101/I121) of the Ni−Mn ordering peak (101) to the most prominent peak (112) with decreasing rare-earth ion size (rR3+).

Figure 2. Rietveld refinement of X-ray diffraction patterns of Dy2NiMnO6 double perovskite. Cross symbols (black color): experimental data points; solid line (red color): simulated curve; vertical sticks (Bragg’s reflections); the bottom gray line is the difference between the experimental and the calculated patterns. Inset (i): The enlarged view of the fitted main peak in the 2θ ≈ 32−35° range. Inset (ii): Crystal structure of Dy2NiMnO6, representing two sublattices of NiO6 and MnO6 octahedra alternatively as generated from the Vesta program.33

C


Article


Figure 3. Evolution of the structural parameters of R2NiMnO6 double perovskites as a function of the rare-earth ion sizes, rR3+. (a) Cell parameters a, b, and c; (b) monoclinic angle β and tolerance factor, t; (c) cell volume; and (d) Ni−O−Mn bond angles and tilting angle. The error in the structural parameters is extremely small, even smaller than the symbols.

III. RESULTS AND DISCUSSION A. Crystal Structure. Room temperature powder X-ray diffraction patterns of the R2NiMnO6 series [Figure 1(a)] confirmed that all samples were single phase except Tb2NiMnO6, Ho2NiMnO6, and Y2NiMnO6. The major phase belongs to a monoclinic structure belonging to the P21/n space group. Minor impurity peaks (marked with *) appear at ∼29° for Tb2NiMnO6, Ho2NiMnO6, and Y2NiMnO6, corresponding to Tb2O3, Ho2O3, andY2O3, respectively. Note that the rareearth ionic radius, rR3+, decreases from La3+ (1.16 Å) to Ho3+ (1.02 Å), in the order rLa > rPr > rNd > rSm > rGd > rTb > rDy > rY > rHo. A magnified view of major peaks in the range of 23 < 2θ < 35°, [Figure 1(b)] reveals peak splitting. The splitting becomes quite clear with decreasing rR3+. The gradual shift of the most intense peak (112), toward higher angles, hints at reduction of lattice parameters. However, the peak along (020) shifts toward lower 2θ values, hinting at increment of the b lattice parameter. Further, in the case of HNMO, merging of the (021) and (200) peaks is observed. This emphasizes the stability of the monoclinic symmetry in samples with smaller rR3+. Additional superstructure reflections were observed along (111) for all samples. These reflections become more intense with reducing rR3+. A long-range ordering between Ni2+ and Mn4+ cations is possible in a homogeneous sample. The (101) peak at 2θ ≈ 20° has been discussed in the literature as an indication of long-range ordering of Ni2+and Mn4+ cations in LNMO samples.30 The relative intensity of (101) to (121) peak increases with decreasing rR3+ [Figure 1(c)], revealing an enhanced Ni2+− Mn4+ ordering. Rietveld refinement was performed to estimate structural parameters for each composition. A monoclinic LNMO structure with the P21/n space group was considered as the starting model for refinement.31 A representative refinement profile for DNMO is presented in Figure 2. The main peak at 2θ ≈ 33° shows the quality of fitting achieved for the refinement

[Figure 2, inset (i)]. The refined crystal structure for DNMO is elaborated [Figure 2, inset (ii)]. An ordered unit cell is composed of two alternate sublattices, consisting of cornershared NiO6 and MnO6 octahedra. Evolution of structural parameters with rR3+ is shown [Figure 3]. The unit cell parameters a and c as well as the cell volume decrease with decreasing rR3+ from La to Ho [Figure 3(a) and (c)]. However, variation in b is small as compared to a and c parameters. A similar observation has been discussed and attributed to a typical tilting scheme of the type a−a−c+ of NiO6 and MnO6 octahedra in P21/n double perovskites.32 In such a kind of scheme, the distortion resulting from the reduction in rR3+ leaves b nearly invariant. The value of c/√2 lies between a and b for the entire series. The Ni−O−Mn bond angle decreases almost linearly with rR3+ [Figure 3(d)] from 165° for LNMO to 144° for HNMO. On the other hand, the monoclinic distortion increases (β deviates further from 90°) with reducing rR3+

Figure 4. Evolution of the bond lengths along the R2NiMnO6 series, determined from XRD and EXAFS. D


Article


Figure 5. Normalized unpolarized Raman spectra for R2NiMnO6 samples at 298 K; (a) low-frequency, (b) main, and (c) high-frequency spectral features. For clarity, spectra have been equally stacked vertically. The dotted vertical line is a guide to the eye to show the shift of modes with rR3+. (d) The deconvoluted Raman spectra of GNMO using a Lorentzian line shape function. The shaded areas confirm the presence of three modes as well as a (e) softening of the mode at ∼670 cm−1 with rR3+.

and asymmetry in these modes. This may be due to improper ordering of Ni and Mn ions. Various Ni/Mn−O vibrations have nearly same frequency. Hence, a distribution of bond lengths, related to the (Ni/Mn)O6 octahedra, contribute to the band envelope. Also, domains with different ordering degrees and deviations from stable oxidation states of Ni and Mn contribute to the Ni/Mn−O stretching bands.37 A broad band (1000−1450 cm−1) is also observed [Figure 5(c)]. This high-frequency broad band can be decomposed into three different relatively weak and broad modes, M1, M2, and M3, as in LNMO thin films.41 Modes M1 (∼1068 cm−1) and M3 (∼1340 cm−1) are overtones of fundamental stretching modes Bg and Ag, respectively. On the other hand, the mode M2 (∼1206 cm−1) is a combination of Bg and Ag modes. As a result of the ordering of B/B′ cations, phonons are not restricted to originate from a point near the Brillouin zone center and can originate from anywhere within the Brillouin zone. This accounts for the second-order multiphonon process in the ordered double perovskites. The possibility of the doubly ordered nature of these RNMO samples is strengthened by the appearance of M1, M2, and M3 modes in these samples. Replacement of La3+ by a smaller R3+ in R2NiMnO6 influences the frequency, symmetry, and intensity of Raman peaks. Therefore, Raman spectra are deconvoluted into a number of Lorentzian line profiles in the frequency range of 450−825 cm−1. Three peaks at ∼494, 513, and 653 cm−1 can be fitted for samples with a smaller rare-earth ion size. The splitting of the Bg mode increases with decreasing rR3+, i.e., higher mass number. Note that the deviation of the monoclinic angle (β) (from XRD studies) increases with decreasing rR3+. Hence, the monoclinic distortion may be related to this splitting. A representative fitted spectrum has been presented for GNMO in Figure 5(d). The Bg mode shows a clear softening with decreasing rR3+ [Figure 5(b)]. The evolution of the peak frequencies with rR3+ is illustrated in Figure 5(e) for the Ag mode. A solid may be considered to be composed of atoms of reduced mass, m, joined by springs with the spring constant, k. In this approximation, the mode frequency (ω) is directly proportional to the square root of k, i.e., ω ∝ √(k/m). Hence, softening may happen because of the increase of mass number from La to Ho. A reduction in rR3+ results in decreasing lattice parameters and an increase in bond lengths (from XRD

[Figure 3(b)]. The average Ni/Mn−O bond length increases with decreasing rR3+ [Figure 4]. The size mismatch among R, Ni, and Mn ions results in tilting of the octahedra. The mismatch, i.e., tolerance factor (t), is an estimate of the stability of the structure. For R2NiMnO6, t is expressed as t=

rR + rO ÄÅ r + r ÉÑ Å √ 2ÅÅÅ Ni 2 Mn + rOÑÑÑÑ Ç Ö

(2)

where rR, rNi, rMn, and rO are the effective ionic radii of R, Ni, Mn, and O ions, respectively.34 The tolerance factor, t, is strongly correlated with the magnetic and electronic properties. With decreasing rR3+, the value of t decreases from 0.920 for LNMO to 0.849 for HNMO [Figure 3(b)]. This reduction in tolerance factor is indicative of the tilting of the octahedra. The tilting angle (φ) between the NiO6 and MnO6 octahedral can be calculated as φ = [180 − (∠Ni−O−Mn)]/2. As ∠Ni−O−Mn decreases, φ progressively increases with decreasing rR3+, from 7.1° for LNMO to 17.9° for HNMO [Figure 3(d)]. Hence, cell distortion increases with decreasing rR3+. Thus, the deviation in β from 90° may be related to the tilting of the octahedra. Such variations result in decrease of bond angle, ∠Ni−O−Mn. B. Vibrational Properties. Raman spectroscopy is a robust technique to examine the crystal structure, cation disorder, local/dynamical lattice distortion, spin−phonon coupling, and impurity phases present in the samples. The distortions from the cubic Fm3̅m lattice results in low-symmetry R2NiMnO6-ordered double perovskites. The vibrational modes arise from the (Ni/ Mn)O6 octahedra and R−O bonds. For clarity, the Raman spectra are demonstrated in a stacking configuration with upward baseline shifts with decreasing rare-earth ionic radii, from La to Ho [Figure 5(a−c)]. Two strong modes Bg (∼525 cm−1) and Ag (∼670 cm−1) are observed [Figure 5(b)], consistent with previous reports for ordered monoclinic LNMO, YNMO, RNMO, and RCMO compounds.35−37 The Ag mode at ∼670 cm−1 can be assigned to stretching “breathing” vibrations of MnO 6 and NiO 6 octahedra using lattice dynamics calculations.38−40 On the other hand, the Bg mode at ∼525 cm−1 represents mixed modes of both antistretching and bending motions. These first-order strong modes Ag and Bg are asymmetric. Various factors are responsible for broadness E


Article


Figure 6. Normalized (a) Mn K-edges and (b) Ni K-edges XANES spectra for R2NiMnO6 double perovskites. Inset I: First derivative of XANES spectra. Inset II: Pre-edge region of the XANES spectra for the same compounds. Inset III: Magnified view of main resonance.

resemblance of the spectra implies similar MnO6 octahedral formation for all the RMNO samples. However, subtle changes in the intensity of the main resonance may imply different levels of distortion with change in the rare-earth element. The complicated pre-edge structure clearly shows two peaks labeled as A1 (∼6539 eV) and A2 (∼6547 eV) having partial 3d weight. These are present for the entire series. Such features are a consequence of mixing of quadrupole (1s → 3d) and hitherto forbidden dipole (1s → 3d) transitions because of hybridized 3d and 4p states. A1 is attributed to transitions to majority eg levels, whereas A2 corresponds to minority eg and t2g levels.44 No significant change was observed in either the intensity or the energy position of the pre-edge structures for the RNMO samples. Hence, the hybridization between Mn 3d and O 2p remains steady with varying R3+. Therefore, it is noteworthy that irrespective of the rare-earth element, the MnO6 octahedra remain remarkably unaffected. Normalized Ni K-edge XANES spectra for Ni foil and NiO were performed as a standard along with the R2NiMnO6 samples [Figure 6(b)]. Similar to the Mn K-edge, the Ni K-edge reveals two resonances, B′ and C′. An additional shoulder D′ is observed for some RMNO between B′ and C′. The appearance of the shoulder D′ can be related to changes in the local atomic environment around Ni. The pre-edge regions (A′) of the Ni Kedge data are not as intense as Mn K-pre-edges. Strength of A1′ peak (transitions to majority eg levels) is extremely poor and negligible in most RNMO samples. For TNMO, a considerable A1′ peak is observed [inset (ii), Figure 6(b)], while in other RNMO samples, the peak has a weakened signature. The A2 ′ (transitions to minority eg and t2g levels) feature is more pronounced compared to the A1′ in all RMNO samples. Contrary to the Mn edge, Ni pre-edge features show some change across the RNMO series, which might be a result of the difference in the extent of hybridization between Ni 3d and O 2p states. The white line intensity is different for all samples. The absorption edge energies of Ni K-edges for RNMO samples are closer to edge energies of Ni K-edges for NiO [inset (i), Figure 6(b)]. This emphasizes that Ni is present mostly as +2 valence states. A line broadening of the spectral features is observed with

studies). An increase of average Ni/Mn−O bond length with decreasing rR3+ is also verified from EXAFS analysis discussed later. This increase in bond length is a signature of the reduced k value and may also be responsible for softening of this Raman mode. Weak stretching modes, in the low-frequency regime (up arrows), are indications of monoclinic P21/n symmetry [Figure 5(a)]. Hence, the presence of monoclinic P21/n symmetry in the samples is confirmed, consistent with XRD analysis. The lowfrequency vibrations may be attributed to coupled (Ni/Mn)O6 tilting and R−O stretching vibrations. These modes are activated by monoclinic distortions and are more sensitive to changing rR3+. The absence of R2O3 can be ensured by the absence of a common intense Raman mode at ∼377 cm−1 for most rare-earth oxides.42 This is a further confirmation of the phase purity of the samples. C. Electronic Properties: XANES Analysis. The charge state and the local environment of an absorbing constituent element of R2NiMnO6 have been investigated using XAS studies. To understand the charge state of the Mn ions, the room temperature Mn K-edge XANES spectra of a Mn foil as well as MnO, Mn2O3, and MnO2 are plotted as references for Mn oxidation states, along with the RMNO samples [Figure 6(a)]. No difference is observed in the energy position of the absorption edge in RNMO spectra with a change in rare-earth ionic size implying an invariant valence state of Mn [inset (i, Figure 6(a)]. It is observed that absorption energies of these samples are close to those in the standard MnO2 spectrum. This reveals that the charge state of Mn seems to be predominantly Mn4+ with traces of Mn3+. The spectra of RNMO samples seem to be very similar to each other and to the other half manganites25,43 having similar crystal structures. The spectra can be described as composed of two resonance peaks, marked as B (main resonance: white line, Mn 1s → Mn 4p transitions) near the edge and C (second resonance: multiple scattering contributions of MnO6 octahedra) in the postedge region. A complex pre-edge structure (A) [inset (ii), Figure 6(a)] in the energy range of 6537−6552 eV is also observed due to Mn 3d−O 2p hybridization. The F


Article


Figure 7. (a) k2-Weighted EXAFS spectra and (b) moduli of the Fourier transforms along with their fits (red lines) for the R2NiMnO6 double perovskites at the Mn K-edge. The spectra have been shifted vertically for clarity.

reduction factor (s02) was fixed to 0.84 for the whole RNMO series. Bond lengths, Debye−Waller (σ2) factors, and threshold energy were varied during fitting. Only the nearest neighbor oxygen shell was analyzed, with two longer and four shorter

decreasing rare-earth ion size, which can be attributed to the variation in 3d character, the covalent nature of the ground state, and the symmetry across absorbing atoms.45 The broadening increases with decreasing size, which causes the Ni−O bond to expand [inset (iii), Figure 6(b)]. It should be noted that Mn spectral lines show lesser broadening as compared to Ni spectral lines. Thus, the above XANES measurements show that Mn and Ni ions are primarily in +4 and +2 valence states, respectively, and that Mn−O hybridization is more invariant than Ni−O. Note that the ionic radii of Mn4+(VI) is 0.67 Å, while that of Ni2+(VI) is 0.83 Å. A homogeneous solution with such different sizes of octahedra is only possible when a proper alternate arrangement of MnO6 and NiO6 octahedra is possible. This hints at a charge-ordered state in R2NiMnO6. D. Local Structure: EXAFS Analysis. EXAFS analysis estimates the local environmental changes in the (Mn/Ni)O6 octahedra and Mn/Ni−O bond lengths. Note that all the RNMO samples at the Mn K-edges exhibit similar EXAFS oscillations [Figure 7(a)]. The Fourier transform (FT) of k2weighted Mn K-edge EXAFS spectra at room temperature was observed in the range of 0−10 Å−1 using a Hanning window [Figure 7(b)]. The EXAFS spectra exhibit a main peak at ∼1.35 Å for all RNMO samples. This peak corresponds to first Mn coordination shell to the six oxygen atoms surrounding Mn. The intensity and shape of this peak remain almost invariant of rR3+. This indicates that oxygen coordination around Mn remains almost constant. Also, no significant change is noticed in peak positions of oxygen shell, with smaller rR3+ indicating that the Mn−O bond lengths remain almost constant for the whole series. These observations hint at invariant MnO6 octahedra in terms of size and shape. This agrees with the XANES observations. A set of weak and broad features appearing at ∼2.08 and 2.69 Å correspond to multipath Mn−La and Mn−Mn/Ni scattering. To extract further quantitative information on the Mn−O bonds, experimental EXAFS spectra were fitted [Figure 7(b)] with theoretical spectra using crystallographic data of La2NiMnO6. The coordination numbers were not refined and fixed to crystallographic compositions, and the amplitude

Table I. EXAFS Parameters of the Oxygen Coordination Shell for R2NiMnO6 Samples at the Mn K-Edgea Mn−O1

Mn−O2

samples

C.N.

d (Å)

σ2 (Å2)

LNMO PNMO SNMO GNMO TNMO DNMO YNMO HNMO

6 6 4 4 4 4 4 6

1.89(1) 1.85(2) 1.90(2) 1.89(2) 1.90(1) 1.89(1) 1.89(1) 1.89(3)

0.004(2) 0.005(2) 0.002(1) 0.002(2) 0.002(5) 0.002(3) 0.003(4) 0.008(3)

C.N.

d (Å)

σ2 (Å2)

2 2 2 2 2

2.39(4) 2.00(4) 2.02(5) 2.04(1) 1.99(5)

0.002(2) 0.010(9) 0.008(8) 0.011(9) 0.012(9)

a

C.N. is the coordination number, and d is the interatomic Mn−O distance. σ2 is the absolute value for each Debye−Waller factor.

bonds with oxygen atoms.46,47 The multipath Mn−R and Mn− Mn/Ni scatterings were not considered in the fitting process because of their weak and broad nature. Simulated best-fit parameters are summarized in Table I. No significant change is observed in Mn−O bond lengths, suggesting that size and nature of the R ion has little effect on the distortion or shape of MnO6 octahedra. The local environment of Ni cations has also been probed using Ni K-edge EXAFS spectra of all R2NiMnO6 samples. Figure 8(a) shows typical k2-weighted EXAFS experimental data. Fourier transforms (FT) using a Hanning window, between 0 to 10 Å−1, are displayed in Figure 8(b). A prominent peak around 1.44 Å corresponds to the Ni−O coordination shell. Coordination shells between 1.9 to 3.8 Å contribute to other Ni−O, Ni−R, and Ni−Ni/Mn scattering paths. The Ni− O scattering reveals a notable change with decreasing rR3+. With decreasing rR3+, the peak gradually shifts toward higher values, indicating expansion of Ni−O bond lengths [Figure 8(b)]. G


Article


Figure 8. (a) k2-Weighted EXAFS spectra and (b) moduli of the Fourier transforms together with their fits (red lines) for the R2NiMnO6 double perovskites at the Ni K-edge. The spectra have been shifted vertically for clarity.

coefficient, A is a characteristic parameter independent of the photon energy, Eg is the electronic band gap, and n is a dimensionless parameter. For direct-allowed transitions, n = 1/ 2. The (αhν)2 vs hν plot reveals a linear nature near the absorption edge [Figure 9]. The band gap energy is calculated by extrapolating the linear part of this curve with a straight line to the (αhν)2 = 0 axis. The band gap of RNMO increases with decreasing rR3+ from 1.42 eV for LNMO to 1.68 eV for YNMO [Figure 13(b)]. Theoretical studies support this result.13 In a similar study on R2CoMnO6, such a trend of increasing band gap with decreasing rR3+ has also been reported.49 Omrane et al. reported O 2p states (−5 to 0 eV) below but d states of La, Ni, and Mn above the Fermi level in the energy band structure of La2NiMnO6 perovskites.50 Ma et al. attributed the valence band to the interaction of Ni 3d orbitals with the O 2p orbitals. On the other hand, they pointed out that hybridization between Mn 3d and O 2p orbitals may be responsible for the bottom of the conduction band.51 Hence the band gap, Eg, depends on the overlap of Ni/Mn and O orbitals. The bond angle, ∠Ni−O−Mn, decreases with reducing rR3+. The decrease in bond angle, reduces the degree of overlap of Mn−O and Ni−O orbitals.49 This decrease in orbital overlap results in a larger band gap. Figure 13(a) reveals an increasing relationship between the band gap and cos2∠Ni−O−Mn. Experimental evidence37 in these samples further strengthens the existing results as well as agrees with theoretical observations. Thus, a wide tunability of the band gap by R-site substitution in such perovskites can be achieved as the optical properties are influenced by crystal structures. F. Magnetic Properties. Octahedral tilting, tolerance factor, lattice parameters, and rare-earth ion size greatly influence the magnetic properties of a perovskite/doubleperovskite structure.13,16,49 The orbital overlap of B−O−B′ bonds are the key players in this aspect.49 The ZFC and FCC M(T) data for all the R2NiMnO6 samples are shown in Figure 10, and the results are in good agreement with earlier reports.14,16,20,52 Below the ferromagnetic transition temperature, TC, thermomagnetic irreversibility is observed for all samples in the ZFC and FCC M(T) data. This behavior is

Table II. EXAFS Parameters of the Oxygen Coordination Shell for R2NiMnO6 Samples at the Ni K-Edgea Ni−O1

Ni−O2

samples

C.N.

d (Å)

σ (Å )

C.N.

d (Å)

σ2 (Å2)

LNMO PNMO NNMO SNMO GNMO YNMO DNMO YNMO HNMO

4 4 4 4 4 4 4 4 4

1.97(1) 1.98(3) 2.01(2) 2.01(3) 2.02(1) 2.02(2) 2.03(2) 2.00(2) 2.05(4)

0.005(1) 0.002(3) 0.003(2) 0.003(3) 0.004(2) 0.005(3) 0.004(3) 0.003(2) 0.004(2)

2 2 2 2 2 2 2 2 2

2.02(4) 2.13(2) 2.06(2) 2.06(2) 2.09(5) 2.06(3) 2.09(3) 2.13(5) 2.05(4)

0.005(2) 0.003(9) 0.008(9) 0.008(6) 0.008(6) 0.005(5) 0.007(4) 0.010(9) 0.004(2)

2

2

a

C.N. is the coordination number, and d is the interatomic Ni−O distance. σ2 is the absolute value for each Debye−Waller factor.

A reduction of peak intensity may be related to a larger distortion with reducing rR3+. Distortion may include tilting of the NiO6 octahedra. Structural analysis was performed in the range of 0−4 Å in the r space [Figure 8(b)]. The fitting strategy is similar to the one used for the Mn K-edge. The obtained structural parameters are tabulated in Table II for the first coordination shell. The high value of the Debye−Waller factor for the HNMO sample [Table II] suggests a nonhomogeneous distortion around the Ni atom. Simulation results further confirm that Ni−O bonds increase with reducing rR3+. Expansion of the apical axis, i.e., the longer Ni−O bond length, may be related to an increasing NiO6 tilting. Thus, EXAFS results reveal an increase of the average Ni/ Mn−O bond length with decreasing rR3+ as per XRD results [Figure 4]. Oxygen atoms shared by both R and Ni atoms move toward R atoms, with decreasing rR3+. Also, NiO6 octahedra are larger than MnO6 octahedra. EXAFS, XANES, XRD, and Raman spectroscopy results agree with each other. E. Electronic Band Gap. The R3+-dependent optoelectronic band gap of the RNMO samples is obtained from the UV−vis absorption spectra using the Tauc relation48 [αhν = A(hν − Eg)n]. Here, hν is the incident photon energy, α is the absorption H


Article


Figure 9. UV−vis absorption spectra for R2NiMnO6 double perovskites for direct transition (αhν)2 vs hν: (a) LNMO, (b) PNMO, (c) NNMO, (d) SNMO, (e) GNMO, (f) TNMO, (g) DNMO, (h) YNMO, and (i) HNMO.

Figure 10. Temperature dependence of the field-cooled cooling (FCC) and zero-field cooling (ZFC) dc magnetization for R2NiMnO6 double perovskites under 100 Oe applied magnetic field: (a) LNMO, (b) PNMO, (c) NNMO, (d) SNMO, (e) GNMO, (f) TNMO, (g) DNMO, (h) YNMO, and (i) HNMO.

typical for R2NiMnO6 compounds.20 Bifurcation between FCC and ZFC curves is an indication of competing magnetic interactions or spin geometric frustrations like spin glass or cluster glass at lower temperatures.53 An additional anomaly is observed at lower temperatures for the NNMO, SNMO, TNMO, and DNMO samples. This low-temperature anomaly may be a result of coexisting FM (major phase) and AFM (minor phase) ordering caused by 3d−4f exchange interactions between Mn−Ni and Tb3+/Nd3+/Dy3+/Sm3+ magnetic moments.16,20,54 A negative ZFC M(T) is noticed in the PNMO, NNMO, YNMO, and HNMO samples. However, it is positive for other

samples. This has been discussed in literature to arise from antiphase boundaries consisting of antisite defects at Ni/Mn sites. Microscopic domain structure consists of a number of ferromagnetic clusters and domains. A short-ranged ferromagnetic coupling exists between Ni2+ and Mn4+ ions, and an antiferromagnetic coupling takes place in the antiphase boundary. A low field of 100 Oe is insufficient in the ZFC mode to align the spins in the direction of the applied magnetic field across antiphase boundary. The antiparallel or canted spins get destabilized as the temperature is increased. A similar frustration effect was observed in Sr 2 NiReO 6 and La2−xBixCoMnO6 double perovskites.55,56 I


Article


Figure 11. (a) Inverse susceptibility for R2NiMnO6 and fits to high temperature using the Curie−Weiss law. (b) Variation of the magnetic ordering temperature, TC, of R2NiMnO6 as a function of rare-earth ion size, rR3+.

Table III. Magnetic Parameters for R2NiMnO6 Double Perovskites sample

LNMO

PNMO

NNMO

SNMO

GNMO

TNMO

DNMO

YNMO

HNMO

C (emu·k·mol−1·Oe1−) θWeiss (K) μeff (μB/f.u.) μeff (calc) (μB/f.u.) TC (K) Mr (μB/f.u.) Hc (Oe) M5K,5T (μB/f.u.) Mextrapolated (μB/f.u.) Ms (μB/f.u.)

5.97 273 6.910 5.25 270 1.09 255 4.24 4.39 5.0

4.69 218 6.124 6.97 210 1.10 336 4.92 5.48 11.4

5.12 204 6.400 7.04 195 0.65 688 5.65 7.14 11.6

4.12 174 5.744 4.94 160 0.95 311 5.05 5.21 6.4

13.93 101 10.556 12.23 130 0.92 162 16.21 21.37 19.0

21.74 60 13.187 14.26 110 1.53 287 13.26 15.38 23.0

25.37 42 14.245 15.80 95 1.13 358 13.42 16.28 25.0

3.06 104 4.950 5.25 79 1.24 362 4.85 4.97 5.0

25.40 31 14.254 15.47 80 1.31 429 13.23 15.52 25.0

Figure 12. Magnetization vs magnetic field isotherms for R2NiMnO6 double perovskites at 5 K temperature.

μeff, is obtained from the Curie−Weiss fit using the relation μeff = 2.828√C. The values for θWeiss are larger than TC for LNMO, PNMO, NNMO, and SNMO and smaller for GNMO, TNMO, DNMO, YNMO and HNMO. The theoretical effective magnetic moments are calculated using the relation, μeff(calc) = [2μB(R3+)2 + μB(Ni2+)2 + μB(Mn4+)2]1/2.16 The effective ground state paramagnetic moments used for calculations are 3.58 μB for Pr3+, 3.62 μB for Nd3+, 0.85 μB for Sm3+, 7.94 μB for Gd3+, 9.72 μB for Tb3+, 10.65 μB for Dy3+, 10.60 μB for Ho, 3.87 μB for Mn4+, and 2.83 μB for Ni2+. The parameters C, θWeiss, μeff,

The Curie−Weiss law defines the paramagnetic component above TC: χ = C/(T − θWeiss), where C is the Curie constant, and θWeiss is the paramagnetic Curie temperature (Weiss constant). Above TC, a linear behavior of inverse FCC susceptibilities with temperature has been observed for most compounds [Figure 11(a)]. Plots for DNMO, HNMO, and TNMO show some deviation from linearity at high temperature. However, the Taxis intercept still remains positive for all samples [Figure 11(a)]. Hence, the Weiss temperature is positive for all samples, confirming the predominant ferromagnetic nature of magnetic interactions in R2NiMnO6. The effective paramagnetic moment, J


Article


Figure 13. Electronic band gap and Curie temperature TC of R2NiMnO6 as a function of (a) superexchange angle, ∠Ni−O−Mn, and (b) ionic radii rR3+.

and μeff(calc) are documented in Table III. The μeff and μeff(calc) values agree with each other for all RNMO samples. As shown in Figure 11(b), the TC (estimated from the derivative of M(T) data) decreases with decreasing rR3+. This monotonic decrease of TC with decreasing rR3+ hints at the correlation of magnetism with Ni/Mn−O bond length and the Ni−O−Mn bond angle. The mechanism of magnetism is mainly governed by superexchange process as per the Goodenough− Kanamori rule. The superexchange coupling arises from the virtual hopping between half-filled Ni2+ (d8: t2g6eg2) and empty Mn4+ (d3: t2g3eg0) orbitals. This gives rise to a set of ferromagnetic ordering tailored by Ni/Mn−O distance and the Ni−O−Mn bond angle. X-ray diffraction revealed that ∠Ni−O−Mn bond angles decrease with decreasing rR3+, from 165° for LNMO to 144° for HNMO. This reduction in ∠Ni−O−Mn and increase of average Mn/ Ni−O bond distances (from EXAFS and XRD studies) with reducing rR3+ reduce the overlap between orbitals, i.e., the Mn/ Ni−O covalent character. This also matches with the reduction in tolerance factor from LNMO to HNMO.15 This in turn lowers the strength of the superexchange interaction between the Ni2+ and Mn4+ ions. Thus, the Curie temperature decreases from 270 K for LNMO to 80 K for HNMO. Hence, TC is being controlled by the nature and size of rare-earth ion. Isothermal field-dependent magnetization hysteresis loops for R2NiMnO6 at 5 K [Figure 12] reveal a ferromagnetic nature for all samples [Figure 12, inset]. The M−H curves do not saturate at 5 T. The behavior of M−H curves depend on the magnetic nature of R3+ ions. The magnetization increases continuously for GNMO, TNMO, DNMO, and HNMO samples up to a field of 5 T. This might arise from the polarization of Gd3+/Tb3+/Dy3+/ Ho3+ moments. The theoretical saturation moments (Ms), the measured (M5T,5K), and corresponding extrapolated moment values (Mextrapolated) are listed in Table III. The Ms for all samples has been calculated using, Ms = [2gJ + 5.0] μB/f.u, where gJ is the rare-earth saturated moment, and 5 μB/f.u is the saturated moment for Ni2+ and Mn4+ ions. The extrapolated moment values (Mextrapolated) are extracted from M vs 1/H extrapolated to 1/H = 0. The measured magnetic moments are smaller than the theoretical moments. Note that these samples have higher magnetization than previously experimentally reported values. This indicates that these samples have better magnetic ordering and less antisite disorders than previously reported samples. A better crystalline nature and homogeneity may be the only

reason for a better magnetization. However, a perfect lattice is yet not achieved in spite of the relatively better crystallinity. A perfect collinear canting is not possible. Therefore, one cannot compare to a theoretical value in which Ni2+/Mn4+ spins and R3+ moments are perfectly ordered. Antisite disorders largely influence magnetization, which in turn depends on synthesis routes, sintering temperature, reaction time, etc. For example, in the present study, the LNMO sample has a saturation moment of 4.39 μB/f.u., emphasizing a more ordered state than reported previously for LNMO with a saturation moment of 3.2 μB/f.u.30 The saturation moment of 5 μB/f.u. is expected for a perfectly ordered LNMO system, because Ni2+ contributes two unpaired electrons, while Mn4+ contributes three unpaired electrons. To calculate the net magnetism of these samples in the presence of antisite disorder, one needs to assess the antisite fraction, x, where a Mn site can now be estimated as composed of [(1− x) Mn + x NiMn] ions, while the Ni site can be estimated as composed of [(1− x) Ni + x MnNi] ions. NiMn and MnNi are antisite disorders at Mn and Ni sites, respectively. The Mn−Mn and Ni−Ni interactions may be estimated as AFM coupled. Hence, these do not contribute to the net magnetic moment. Therefore, only (1−2x) ions of both Mn and Ni contribute to the net magnetic moment. In the case of LNMO, it can hence be derived that Mextrapolated = [(1 − 2x)(μB(Mn) + μB(Ni))]. Thus, ∼6.1% antisite defect is estimated from a net magnetic moment of 4.39 μB/f.u. compared to a theoretical moment of 5 μB/f.u. Thus, less antiste defects in the LNMO sample hint at high degree of ordering, resulting in large magnetization. From this trend in LNMO and consistently significant larger values of remnant magnetization (Mr) and coercivity (Hc), it may be inferred that a higher degree of ordering and stability of ferromagnetic coupling is possible in all other RNMO samples in this study. The nearly linear correlation [Figure 13] between Eg, TC, and cos2θNi−O−Mn indicates that rR3+ affects the band gap and Curie temperature mostly by changing the tilting angles. Thus, rR3+ and thereby superexchange angle is possibly responsible for the variation in band gap and Curie temperature.

IV. CONCLUSIONS Single-phase R2NiMnO6 perovskites crystallize in a monoclinic P21/n structure. Lattice parameters decrease due to decreasing rare-earth ionic size (rR3+) from LNMO to HNMO. Because of the decreasing rR3+, structural distortions like octahedral tilting K


Article


3d- and 5d-Sublattice Magnetism in the Double Perovskite Substitution Series La2Zn1‑XCoxIro6. Phys. Rev. B: Condens. Matter Mater. Phys. 2018, 97, No. 035155. (9) Kayser, P.; Alonso, J. A.; Muñoz, A.; Fernández-Díaz, M. T. Structural and Magnetic Characterization of the Double Perovskites R2NiRuo6 (R = Pr-Er): A Neutron Diffraction Study. Acta Mater. 2017, 126, 114−123. (10) Sarkar, T.; Manna, K.; Elizabeth, S.; Anil Kumar, P. S. Investigation of Multiferroicity, Spin-Phonon Coupling, and Unusual Magnetic Ordering Close to Room Temperature in LuMn0.5Fe0.5O3. J. Appl. Phys. 2017, 121, No. 084102. (11) Rogado, N. S.; Li, J.; Sleight, A. W.; Subramanian, M. A. Magnetocapacitance and Magnetoresistance near Room Temperature in a Ferromagnetic Semiconductor: La2NiMnO6. Adv. Mater. 2005, 17, 2225−2227. (12) Choudhury, D.; Mandal, P.; Mathieu, R.; Hazarika, A.; Rajan, S.; Sundaresan, A.; Waghmare, U. V.; Knut, R.; Karis, O.; Nordblad, P.; Sarma, D. D. Near-Room-Temperature Colossal Magnetodielectricity and Multiglass Properties in Partially Disordered La2NiMno6. Phys. Rev. Lett. 2012, 108, 127201. (13) Zhao, H. J.; Liu, X. Q.; Chen, X. M.; Bellaiche, L. Effects of Chemical and Hydrostatic Pressures on Structural, Magnetic, and Electronic Properties of R2NiMnO6 (R = Rare-Earth Ion) Double Perovskites. Phys. Rev. B: Condens. Matter Mater. Phys. 2014, 90, 195147. (14) Booth, R. J.; Fillman, R.; Whitaker, H.; Nag, A.; Tiwari, R. M.; Ramanujachary, K. V.; Gopalakrishnan, J.; Lofland, S. E. An Investigation of Structural, Magnetic and Dielectric Properties of R2NiMnO6 (R = Rare Earth, Y). Mater. Res. Bull. 2009, 44, 1559−1564. (15) Yang, W. Z.; Liu, X. Q.; Zhao, H. J.; Lin, Y. Q.; Chen, X. M. Structure, Magnetic, and Dielectric Characteristics of Ln2NiMnO6 (Ln = Nd and Sm) Ceramics. J. Appl. Phys. 2012, 112, No. 064104. (16) Retuerto, M.; Muñoz, Á .; Martínez-Lope, M. J.; Alonso, J. A.; Mompeán, F. J.; Fernández-Díaz, M. T.; Sánchez-Benítez, J. Magnetic Interactions in the Double Perovskites R2NiMnO6 (R = Tb, Ho, Er, Tm) Investigated by Neutron Diffraction. Inorg. Chem. 2015, 54, 10890−10900. (17) Dass, R. I.; Yan, J. Q.; Goodenough, J. B. Oxygen Stoichiometry, Ferromagnetism, and Transport Properties of La2−XNiMnO6+Δ. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 68, No. 064415. (18) Goodenough, J. B. Theory of the Role of Covalence in the Perovskite-Type Manganites [La, M(Ii)] Mn O3. Phys. Rev. 1955, 100, 564−573. (19) Chakraborty, T.; Nhalil, H.; Yadav, R.; Wagh, A. A.; Elizabeth, S. Magnetocaloric Properties of R2NiMnO6 (R = Pr, Nd, Tb, Ho and Y) Double Perovskite Family. J. Magn. Magn. Mater. 2017, 428, 59−63. (20) Neenu Lekshmi, P.; Vasundhara, M.; Varma, M. R.; Suresh, K. G.; Valant, M. Structural, Magnetic and Dielectric Properties of Rare Earth Based Double Perovskites RE2NiMnO6 (RE = La, Pr, Sm, Tb). Phys. B 2014, 448, 285−289. (21) Zhou, J. S.; Alonso, J. A.; Pomjakushin, V.; Goodenough, J. B.; Ren, Y.; Yan, J. Q.; Cheng, J. G. Intrinsic Structural Distortion and Superexchange Interaction in the Orthorhombic Rare-Earth Perovskites RCrO3. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 81, 214115. (22) Blasco, J.; García-Muñoz, J. L.; García, J.; Subías, G.; Stankiewicz, J.; Rodríguez-Velamazán, J. A.; Ritter, C. Magnetic Order and Magnetoelectric Properties of R2NiMnO6 perovskites (R = Ho, Tm, Yb, and Lu). Phys. Rev. B: Condens. Matter Mater. Phys. 2017, 96, No. 024409. (23) Zhang, G.; Li, G.; Liao, F.; Fu, Y.; Xiong, M.; Lin, J. Crystal Growth and Magnetic Properties of the Double Perovskites R2NiMnO6 (R = Pr, Sm and Ho) by a Hydrothermal Route. J. Cryst. Growth 2011, 327, 262−266. (24) Louca, D.; Egami, T. Local Lattice Distortions in La1−XSrxMno3 Studied by Pulsed Neutron Scattering. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59, 6193−6204. (25) Cuartero, V.; Lafuerza, S.; Rovezzi, M.; García, J.; Blasco, J.; Subías, G.; Jiménez, E. X-Ray Absorption and Emission Spectroscopy

and monoclinic distortions become stronger. The changes in the octahedral tilting happen because of the changing orbital overlap between O2− and R3+ ions (with decreasing rR3+). This results in decreasing superexchange angle (∠Ni−O−Mn) from 165° (LNMO) to 144° (HNMO), followed by phonon softening and Ni(Mn)−O bond length elongation. Hence, a decrease in superexchange interaction between Ni2+ and Mn4+ ions is observed, which leads to reduction of TC, from 270 K (LNMO) to 80 K (HNMO). Magnetic ordering, originating from 3d−4f exchange interactions between Mn−Ni and Tb3+/Nd3+/Dy3+/ Sm3+ ions, occurs at low temperature, giving rise to a magnetic anomaly. Antisite disorder is reduced because of extremely homogeneous distribution and regular arrangement of Ni2+ and Mn4+ ions. A significant B-site ordering is also observed. The electronic band gap increases with decreasing rR3+. Thus, the magnetic and electronic properties are able to be manipulated by rR3+, so these materials are potential candidates for nextgeneration spintronics applications.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (S.S.) *E-mail: [email protected] (S.B.) ORCID

Shun-Wei Liu: 0000-0002-3128-905X Somaditya Sen: 0000-0002-6129-8344 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are thankful to IIT Indore for providing laboratory space, consumables, and the XRD and UV−vis facility. Dr. Pankaj R. Sagdeo is acknowledged for providing UV−vis spectra. Mohd. Nasir gratefully acknowledges financial support received from UGC, New Delhi under the Maulana Azad National fellowship. Dr. Sajal Biring acknowledges the financial support from the Ministry of Science and Technology, Taiwan (MOST 105-2218-E-131-003 and 106-2221-E-131-027).

■

REFERENCES

(1) Vasala, S.; Karppinen, M. A2BB’O6 Perovskites: A Review. Prog. Solid State Chem. 2015, 43, 1−36. (2) Lee, A. T.; Marianetti, C. A. Structural and Metal-Insulator Transitions in Rhenium-Based Double Perovskites Via Orbital Ordering. Phys. Rev. B: Condens. Matter Mater. Phys. 2018, 97, No. 045102. (3) Kitagawa, Y.; Hiraoka, Y.; Honda, T.; Ishikura, T.; Nakamura, H.; Kimura, T. Low-Field Magnetoelectric Effect at Room Temperature. Nat. Mater. 2010, 9, 797. (4) Azuma, M.; Takata, K.; Saito, T.; Ishiwata, S.; Shimakawa, Y.; Takano, M. Designed Ferromagnetic, Ferroelectric Bi2NiMnO6. J. Am. Chem. Soc. 2005, 127, 8889−8892. (5) Sariful Sheikh, M.; Ghosh, D.; Dutta, A.; Bhattacharyya, S.; Sinha, T. P. Lead Free Double Perovskite Oxides Ln2NiMno6 (Ln = La, Eu, Dy, Lu), a New Promising Material for Photovoltaic Application. Mater. Sci. Eng., B 2017, 226, 10−17. (6) El Rhazouani, O.; Slassi, A.; Ziat, Y.; Benyoussef, A. Magnetocaloric Effect in Sr2criro6 Double Perovskite: Monte Carlo Simulation. J. Phys. Chem. Solids 2017, 104, 32−35. (7) Andrews, R. L.; Heyns, A. M.; Woodward, P. M. Raman Studies of A2MWo6 Tungstate Double Perovskites. Dalton Trans. 2015, 44, 10700−10707. (8) Vogl, M.; Corredor, L. T.; Dey, T.; Morrow, R.; Scaravaggi, F.; Wolter, A. U. B.; Aswartham, S.; Wurmehl, S.; Büchner, B. Interplay of L


Article

ACS Applied Electronic Materials Study of Mn and Co Valence and Spin States in TbMn1−XCoxO3. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 94, 155117. (26) Lee, P. A.; Citrin, P. H.; Eisenberger, P.; Kincaid, B. M. Extended X-Ray Absorption Fine Structure - Its Strengths and Limitations as a Structural Tool. Rev. Mod. Phys. 1981, 53, 769−806. (27) Rietveld, H. A Profile Refinement Method for Nuclear and Magnetic Structures. J. Appl. Crystallogr. 1969, 2, 65−71. (28) Roisnel, T.; Rodríquez-Carvajal, J. Winplotr: A Windows Tool for Powder Diffraction Pattern Analysis. Mater. Sci. Forum 2001, 378− 381, 118−123. (29) Newville, M.; Ravel, B.; Haskel, D.; Rehr, J. J.; Stern, E. A.; Yacoby, Y. Analysis of Multiple-Scattering Xafs Data Using Theoretical Standards. Phys. B 1995, 208, 154−156. (30) Pal, S.; Govinda, S.; Goyal, M.; Mukherjee, S.; Pal, B.; Saha, R.; Sundaresan, A.; Jana, S.; Karis, O.; Freeland, J. W.; Sarma, D. D. Effect of Anti-Site Disorder on Magnetism in La2NiMno6. Phys. Rev. B: Condens. Matter Mater. Phys. 2018, 97, 165137. (31) Bull, C. L.; Gleeson, D.; Knight, K. S. Determination of B-Site Ordering and Structural Transformations in the Mixed Transition Metal Perovskites La2CoMnO6 and La2NiMnO6. J. Phys.: Condens. Matter 2003, 15, 4927. (32) Glazer, A. Simple Ways of Determining Perovskite Structures. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1975, 31, 756−762. (33) Momma, K.; Izumi, F. Vesta 3 for Three-Dimensional Visualization of Crystal, Volumetric and Morphology Data. J. Appl. Crystallogr. 2011, 44, 1272−1276. (34) Kakarla, D. C.; Jyothinagaram, K. M.; Das, A. K.; Adyam, V. Dielectric and Magnetodielectric Properties of R2NiMnO6 (R = Nd, Eu, Gd, Dy, and Y). J. Am. Ceram. Soc. 2014, 97, 2858−2866. (35) Bezerra Macedo Filho, R.; Ayala, A. P.; William de Araujo, C. W. Spin-Phonon Coupling in Y2NiMno6 Double Perovskite Probed by Raman Spectroscopy. Appl. Phys. Lett. 2013, 102, 192902. (36) Guo, H. Z.; Burgess, J.; Ada, E.; Street, S.; Gupta, A.; Iliev, M. N.; Kellock, A. J.; Magen, C.; Varela, M.; Pennycook, S. J. Influence of Defects on Structural and Magnetic Properties of Multifunctional La2NiMnO6 Thin Films. Phys. Rev. B: Condens. Matter Mater. Phys. 2008, 77, 174423. (37) Bull, C. L.; McMillan, P. F. Raman Scattering Study and Electrical Properties Characterization of Elpasolite Perovskites Ln2(BB′)O6 (Ln = La, Sm···Gd and B,B′=Ni, Co, Mn). J. Solid State Chem. 2004, 177, 2323−2328. (38) Iliev, M. N.; Guo, H.; Gupta, A. Raman Spectroscopy Evidence of Strong Spin-Phonon Coupling in Epitaxial Thin Films of the Double Perovskite La2NiMnO6. Appl. Phys. Lett. 2007, 90, 151914. (39) Iliev, M. N.; Abrashev, M. V.; Litvinchuk, A. P.; Hadjiev, V. G.; Guo, H.; Gupta, A. Raman Spectroscopy of Ordered Double Perovskitela2comno6 Thin Films. Phys. Rev. B: Condens. Matter Mater. Phys. 2007, 75, 104118. (40) Truong, K. D.; Singh, M. P.; Jandl, S.; Fournier, P. Influence of Ni/Mn Cation Order on the Spin-Phonon Coupling in Multifunctional La2NiMnO6 Epitaxial Films by Polarized Raman Spectroscopy. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 80, 134424. (41) Guo, H.; Burgess, J.; Street, S.; Gupta, A.; Calvarese, T. G.; Subramanian, M. A. Growth of Epitaxial Thin Films of the Ordered Double Perovskite La2nimno6 on Different Substrates. Appl. Phys. Lett. 2006, 89, No. 022509. (42) Panitz, J.-C.; Mayor, J.-C.; Grob, B.; Durisch, W. A Raman Spectroscopic Study of Rare Earth Mixed Oxides. J. Alloys Compd. 2000, 303−304, 340−344. (43) Sánchez, M. C.; García, J.; Blasco, J.; Subías, G.; Perez-Cacho, J. Local Electronic and Geometrical Structure of LaNi1−XMnxo3+Δ Perovskites Determined by X-Ray-Absorption Spectroscopy. Phys. Rev. B: Condens. Matter Mater. Phys. 2002, 65, 144409. (44) Bridges, F.; Booth, C. H.; Anderson, M.; Kwei, G. H.; Neumeier, J. J.; Snyder, J.; Mitchell, J.; Gardner, J. S.; Brosha, E. Mn K-edge Xanes Studies of La1‑XAxo3 Systems (A = Ca, Ba, Pb). Phys. Rev. B: Condens. Matter Mater. Phys. 2001, 63, 214405.

(45) Abbate, M.; de Groot, F. M. F.; Fuggle, J. C.; Fujimori, A.; Strebel, O.; Lopez, F.; Domke, M.; Kaindl, G.; Sawatzky, G. A.; Takano, M.; Takeda, Y.; Eisaki, H.; Uchida, S. Controlled-Valence Properties of La1−XSrxFeo3 and La1−XSrxMno3 Studied by Soft-X-Ray Absorption Spectroscopy. Phys. Rev. B: Condens. Matter Mater. Phys. 1992, 46, 4511−4519. (46) Monesi, C.; Meneghini, C.; Bardelli, F.; Benfatto, M.; Mobilio, S.; Manju, U.; Sarma, D. D. Quantitative Structural Refinement of Mnk Edge Xanes in LaMno3 and CaMno3 Perovskites. Nucl. Instrum. Methods Phys. Res., Sect. B 2006, 246, 158−164. (47) Guo, Y.; Shi, L.; Zhou, S.; Zhao, J.; Wei, S. Local Valence and Hole-Doping Effect on Magnetic Properties in Double Perovskite La2NiMnO6. J. Supercond. Novel Magn. 2013, 26, 3287−3292. (48) Tauc, J.; Grigorovici, R.; Vancu, A. Optical Properties and Electronic Structure of Amorphous Germanium. Phys. Status Solidi B 1966, 15, 627−637. (49) Zhou, H. Y.; Chen, X. M. Structural Distortions, Orbital Ordering and Physical Properties of Double Perovskite R2NiMnO6 Calculated by First-Principles. J. Phys.: Condens. Matter 2017, 29, 145701. (50) Ullah, M.; Ayaz Khan, S.; Murtaza, G.; Khenata, R.; Ullah, N.; Bin Omran, S. Electronic, Thermoelectric and Magnetic Properties of La2NiMnO6 and La2CoMnO6. J. Magn. Magn. Mater. 2015, 377, 197− 203. (51) Lan, C.; Zhao, S.; Xu, T.; Ma, J.; Hayase, S.; Ma, T. Investigation on Structures, Band Gaps, and Electronic Structures of Lead Free La2NiMnO6 Double Perovskite Materials for Potential Application of Solar Cell. J. Alloys Compd. 2016, 655, 208−214. (52) Lin, H.; Shi, X. X.; Chen, X. M. LaRNiMno6 (R = Pr, Nd, Sm) Double Perovskites with Magnetodielectric Effects. J. Alloys Compd. 2017, 709, 772−778. (53) Bhowmik, R. N.; Ranganathan, R. Magnetic Order and Electrical Conductivity Scaling of the Spinel Oxide Mn0.5Ru0.5Co2O4. Phys. Rev. B: Condens. Matter Mater. Phys. 2006, 74, 214417. (54) Nair, H. S.; Swain, D.; N., H.; Adiga, S.; Narayana, C.; Elzabeth, S. Griffiths Phase-Like Behavior and Spin-Phonon Coupling in Double Perovskite Tb2NiMno6. J. Appl. Phys. 2011, 110, 123919. (55) Retuerto, M.; Martínez-Lope, M. J.; García-Hernández, M.; Fernández-Díaz, M. T.; Alonso, J. A. Crystal and Magnetic Structure of Sr2MReo6 (M = Ni, Co, Zn) Double Perovskites: A Neutron Diffraction Study. Eur. J. Inorg. Chem. 2008, 2008, 588−595. (56) Bai, Y.; Xia, Y.; Li, H.; Han, L.; Wang, Z.; Wu, X.; Lv, S.; Liu, X.; Meng, J. A-Site-Doping Enhanced B-Site Ordering and Correlated Magnetic Property in La2−XBixCoMno6. J. Phys. Chem. C 2012, 116, 16841−16847.

M


Role of Antisite Disorder, Rare-Earth Size, and ... - ACS Publications

Recommend Documents