Spintronics-based studies have produced significant attention in the last decade while claiming the observation of room temperature ferromagnetism (RTFM). Nevertheless, there is a lack of consensus on a mechanism responsible for this phenomenon. In this study, we focus on Cu-doped ZnO (ZCO) to understand the microscopic origin of RTFM and the role of different oxidation states of Cu in RTFM. We have performed different spectroscopic techniques using synchrotron facilities. The values of spin-moment obtained from x-ray magnetic circular dichroism sum-rule truly exhibit a ferromagnetic interaction in the nanocrystalline powder of ZCO with ∼0.58 μB for 5% of Cu concentration in the total fluorescence yield mode. Such an enhanced magnetization is attributed to the presence of Cu2+, which is mainly localized in the bulk region. Cu in ZCO is mostly dominated by the presence of Cu2+. This is clearly reflected by the profiles of x-ray photoemission spectroscopy. Consequently, the weakly magnetized total electron yield mode is attributed to a state of magnetic frustration as the majority of Cu3+ is found on the surface. Some of these Cu3+ when come in the vicinity of Cu2+ ions result in a highly correlated state of double exchange mechanism, which is the microscopic origin of RTFM in ZCO. The coupling between Cu2+-Cu3+ is mediated via oxygen vacancies (VO), the presence of which is confirmed through the features of electron energy loss spectroscopy over different edges. The confirmation of VO is also supported by the deconvolution of E2high-phonon in the Raman spectra. Moreover, the defects in the local electronic structures of ZCO are demonstrated by the deconvoluted spectra of Cu L3 x-ray absorption spectroscopy. The images obtained from high-resolution transmission electron microscopy confirm the incorporation of Cu into the wurtzite crystal of ZnO. A clear enhancement in magnetization upon an increase in carriers of Cu in ZCO indicates carrier-induced ferromagnetism. Cu2+ and VO are the two attributes of RTFM in ZCO.

Ever since the prediction of spin degree of freedom of an electron came to reality, a significant advancement has been achieved while incorporating such a breakthrough in quantum theory. Such an intrinsic property of an electron, by its own virtue, is responsible for the inception of the current state of the art, i.e., spintronics.1,2 The birth of spintronics proved to be an indispensable one while dealing with quantum phase transition, highly correlated compounds, and quantum magnetism. None other than the oxides of transition elements have implanted the concept of spin because of their startling nature of multivalency. The quest for securing ferromagnetism (FM) in such transition oxides is triggered by the shallow doping of magnetic impurities, which resulted in a new category of magnetic materials called diluted magnetic oxide semiconductors (DMOS).3,4 A magnetic impurity attributes its substitutional effect on the host lattice site in a manner to form a correlated state of the d-orbitals. This correlation is the genesis of complex semiconducting and magnetic attributes of transition elements. A material with high Curie temperature (TC) is essentially required to establish room temperature ferromagnetism (RTFM).

Zinc-oxide (ZnO) is a wide direct bandgap semiconductor (∼3.37 eV) that crystallizes in the hexagonal structure of wurtzite. Spintronics potential of ZnO has shown to be a consequence of defect in the wurtzite crystal structure.5,6 In particular, Cu-doped ZnO (ZCO) has been extensively studied. From the first principle theoretical approach, Huang et al.7 have shown FM in ZCO. While with a detailed ab initio calculation, Lathiotakis et al.8 exhibited the significance role of Cu-ions for a ferromagnetic state in a co-doped system of ZnO. A microscopic origin of FM on the thin films of ZCO by means of double-exchange (DE) interaction is demonstrated by Herng et al.,9 where the large-sized vacancy orbitals mediated the localized Cu2+ moments in the vicinity of oxygen-vacancies (VO), resulting in the ferromagnetic-coupling between Cu1+-Cu2+. On the contrary, a correlated state of DE-interaction between Cu2+-Cu3+ in the presence of VO is shown by Kataoka et al.10 in the nanowires of ZCO. Imperfection in the crystal structure due to the diffusion of Cu-ions configures the substitutional inducement of bound magnetic polarons (Bmp) that promote the formation and percolation of the ferromagnetic domain in DMOS.11–13 Apart from all these attributes, there is no clear agreement on the exact origin of FM in ZCO from the structural or electronic point of view.

In this study, we have developed a model using various spectroscopic techniques to configure the exact origin of RTFM in ZCO. Structural and morphological information is obtained using various aspects of transmission electron microscopy (TEM), which confirms the attainability of the hexagonal-wurtzite phase upon Cu-doping without segregation. To locate the multivalent nature of Cu in ZCO, a thorough analysis of x-ray photoemission spectroscopy (XPS) at Cu 2p-edge is performed with different spectra in a comparative manner. Magnetic characteristics are shown with the help of vibrating sample magnetometer (VSM) at 300 K. Element-specific local electronic structures in both the bulk-sensitive total fluorescence yield (TFY) and surface-sensitive total electron yield (TEY) modes are investigated using x-ray absorption spectroscopy (XAS), where the discrimination in the helicity of photons resulted in the hyperfine spectrum of x-ray magnetic circular dichroism (XMCD). The onset of bulk magnetization in ZCO is a highly correlated aspect of DE-interaction between Cu2+-Cu3+ via VO, the presence of which is clearly shown by electron energy loss spectroscopy (EELS) and is also confirmed with the deconvolutions over E2high-phonon in Raman and Cu L3 XAS.

Nanocrystalline powder of ZCO with varying atomic-weight concentrations of Cu (Zn1−xCuxO; x = 0.00, 0.01, 0.03, and 0.05) are synthesized via co-precipitation14,15 technique. Crystallinity, phase, and structural-based information of the as-grown powdered samples are obtained using x-ray diffraction (XRD), performed on BL-13 GIXS (SINP beamline) of the Indus-2 synchrotron radiation source of Raja Ramanna Centre for Advanced Technology (RRCAT), Indore, India. The energy of the monochromatic synchrotron radiation source was maintained at 10 keV with a step size of 0.01°. A low-resolution portable Raman Spectrometer (Raman system R-3000) with an optical fiber sampling probe and a solid-state diode laser of wavelength 785 nm was incorporated for the phonon-mediated electronic/vibrational investigation. All the measurements of TEM are performed at IMRE, A*STAR, Singapore. A synchronized source of 1486.4 eV of excitation energy is utilized to measure the XPS data under ultra-high vacuum of ∼10−8 Torr. Magnetic measurements are carried out using VSM, Model 7404/Lakeshore USA. XAS and XMCD measurements at different edges in TEY and TFY modes are performed at the beamline-16A of photon factory (KEK), Japan. It is to be noted that TFY is a bulk-sensitive mode where the probing depth is ∼100 nm while TEY is a surface-sensitive mode where the utmost depth of penetration is ∼5 nm.16,17

Figure 1(a) demonstrates the XRD patterns of pristine and Cu-doped samples. Hexagonal wurtzite structure of ZnO is confirmed throughout the different levels of doping, however, a slight shift toward the lower value of diffraction angle is observed upon doping. Such a shift is a direct consequence of the mismatch in the ionic radii of Cu and Zn, which indeed suggests the incorporation of Cu into the wurtzite ZnO-matrix. Cu is known to occupy multivalent states, which is the onset of such mismatch in the ionic radii in order to show the effect of substitution, albeit, the mismatch is very minute.18 Average crystallite size of the nanocrystalline powder of ZCO is calculated using Scherrer’s equation and is determined considering the most intense peak of (101) plane. The calculated values within uncertainties are 21.43 ± 0.26, 22.74 ± 0.26, 28.76 ± 0.26, and 32.26 ± 0.26 nm, respectively, for the pristine and the doped-samples with the ascending order of variation in the atomic weight concentration of Cu into the host ZnO-matrix. Evidently, there is a gradual increase in the crystallite size upon the initial stage of doping, which gets significantly enhanced for high Cu-concentration. This is a remarkable illustration of consistency in the crystalline phase of ZnO upon Cu-doping. To confirm the stability of the wurtzite phase over doping, inter-planar spacing (d) and lattice parameters (a, c), concerning the (101) orientation are calculated (see Table I). A linear, smooth but small variation to the parameters is observed upon doping, which is a subtle sign of null segregation in the crystalline phase. It is to be noted that the ratio of the lattice parameters, i.e., c/a ∼ 1.632. A more rigorous analysis of the crystalline structures is demonstrated in the form of selected area electron diffraction (SAED) patterns [Fig. 1(b)]. The formation of concentric rings is an intrinsic property of the randomly oriented mono-crystalline nanoparticles. The similarity in the values of d-spacing for different samples are obtained in a manner of 0.28, 0.26, 0.25, 0.19, and 0.16 nm, in view of (100), (002), (101), (102), and (110), respectively, to the crystallographic planes of ZnO, which makes a solid correlation with the hexagonal phase of wurtzite. Moreover, there is a stark correlation in the obtained values of d101 from XRD and SAED.

FIG. 1.

(a) XRD and (b) SAED patterns of pristine and Cu-doped ZnO.

FIG. 1.

(a) XRD and (b) SAED patterns of pristine and Cu-doped ZnO.

Close modal
TABLE I.

Values of the inter-planar spacing (d) and the lattice-parameters (a, c) associated with the most intense peak of (101) plane in the XRD spectra of Cu-doped ZnO.

ZnOZn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
d101 (Å) 2.495 2.499 2.503 2.507 
a (Å) 3.262 3.267 3.273 3.277 
c (Å) 5.323 5.332 5.341 5.349 
ZnOZn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
d101 (Å) 2.495 2.499 2.503 2.507 
a (Å) 3.262 3.267 3.273 3.277 
c (Å) 5.323 5.332 5.341 5.349 

Figure 2 exhibits the Raman spectra of pristine and ZCO. The two most intense peaks are observed in the vicinity of 99 and 437 cm−1, which represent E2low and E2high-phonons, respectively.17 These two modes are Raman-active nonpolar modes, where E2low is dominated by Zn-sites while E2high is ascribed to the vibration of O-atoms. Two noticeable features are also observed around 200 and 330 cm−1 and are attributed to the overtone of E2low and the superposition of E2high and E2low modes, respectively.19–21 The downshift in frequency and the linewidth-broadening of the E2high-phonon are strong evidences of disorder in the wurtzite crystal structure upon doping, the majority of which in the form of VO [see Fig. 2(b)].22,23 As the appearance of E2high peak is mainly associated with the vibrational motion of O-atoms, the observed linewidth suggests the anharmonic scattering of phonons in the wurtzite crystal due to VO.24,25 For a system like ZnO, the proposed spintronics environment cannot be achieved by either dopant or defect; in other words, this essentially requires the combination of both.

FIG. 2.

(a) Raman spectra of pristine and Cu-doped ZnO. (b) A magnified illustration of E2high-phonon. (c)–(f) Deconvolution of E2high-phonon to exhibit the oxygen-mediated disorder in the hexagonal-wurtzite crystal of ZCO.

FIG. 2.

(a) Raman spectra of pristine and Cu-doped ZnO. (b) A magnified illustration of E2high-phonon. (c)–(f) Deconvolution of E2high-phonon to exhibit the oxygen-mediated disorder in the hexagonal-wurtzite crystal of ZCO.

Close modal

To confirm the creation of oxygen-mediated disorder upon doping in an explicit way, we have performed the deconvolution of E2high branching phonon. Two peaks Gaussian fitting of E2high-phonon is performed using Origin 8.6 software [Figs. 2(c)2(f)]. Such a deconvolution of oxygen-mediated phonon not only corroborates with the creation of VO, rather it also reflects on the oxygen-oriented electronic/vibrational transitions for the different branching phonons associated with the tetrahedron formation of ligands.26 In the wurtzite crystal structure of ZnO, where a single Zn-atom is surrounded by four O-atoms and further each O-atom is surrounded by four Zn-atoms, i.e., a complex formation of tetrahedron, the ubiquitous mode of electronic/vibrational transitions proceed in the following manner: (a) creation of excitons due to the induced ion-beam, where an electron is excited from the valence band to the conduction band, leaving a hole behind; (b) correlation via exchange mechanism between the electron and phonon wavevectors, as a result of transferring their momenta mutually; and (c) the recombinational effect, providing some sorts of transition of lower wavenumber. The nomenclatures of the deconvoluted spectra, E2high(a) and E2high(b) [Figs. 2(c)2(f)], provide solid evidence of vacancy-mediated variation over Cu-substitution, which leave no traces of ambiguity toward the structural-based electronic/vibrational modifications in the ZnO over Cu-inducement.

In order to acquire the chemically sensitive morphological information, a detailed and systematic investigation based on TEM is performed. Figure 3 illustrates the morphology of pristine and Cu-doped samples in terms of varying magnification. A higher magnification clearly exhibits the roughness on the doped-ZnO samples, which is due to the onsite substitutional effect of Cu. As Cu is more electronegative than Zn, this would result in the replacement of Zn-atoms. Moreover, the possibility of any segregation of phase is completely denied by high-resolution transmission electron microscopy (HR-TEM) that measured a value of 0.25 nm for d101, which forms a consensus with the crystallographic orientation of the ascribed plane and maintains the wurtzite phase over doping.27 Change in morphology with doping is analyzed using energy-dispersive x-ray (EDX) mapping in the scanning transmission electron microscopy (STEM) configuration (Fig. 4). The presence of Cu in the doped nanopowder is observed with the raster scanning of it, albeit, a completely dark image can also be seen for the pristine ZnO. Different aspects of the elemental morphology of ZCO are clearly visible, i.e., from annular dark field (ADF) images (extreme left) to the various colored images of RGB (extreme right). EELS is a very sensitive technique to govern the local electronic states where a point probe is utilized to determine the loss in electron energy at a specific edge. The spectra of EELS in STEM configuration are shown in Fig. 5, where tiny colored spheres in Fig. 5(a) represent the point probe to determine the core energy loss of an electron. These nanoprobes resulted in the spectral-edge distribution of Zn L, O K, and Cu L-edges [Figs. 5(b)5(e)]. The splitting of the L-edge by means of spin–orbit interaction locates the features of Zn L (L3 = 1024 eV, L2 = 1047 eV) suggesting the formation of Zn1+.28 Interestingly, for Cu L (L3 = 930 eV, L2 = 950 eV) reflects on the formation of Cu1+ and Cu2+, which strongly suggest the creation of VO.29 Moreover, the observation of pre-edge features in the O K (∼530 eV) spectra, which is mostly associated with the active carrier concentration and the unoccupancy of p-d bands confirm the presence of VO in ZCO.30 

FIG. 3.

(a)–(d) Exhibit TEM images of the nanocrystalline powder of pristine and Cu-doped ZnO in the left column. A greater magnification of TEM is shown in the middle column. Images of HR-TEM are demonstrated in the right column, confirming the stability in phase over Cu-doping.

FIG. 3.

(a)–(d) Exhibit TEM images of the nanocrystalline powder of pristine and Cu-doped ZnO in the left column. A greater magnification of TEM is shown in the middle column. Images of HR-TEM are demonstrated in the right column, confirming the stability in phase over Cu-doping.

Close modal
FIG. 4.

ADF-mapping of STEM for the different samples of ZCO is exhibited in the extreme left column. The elemental mapping is demonstrated in a manner of zinc (green), oxygen (red), and copper (purple). RGB mapping is shown in the extreme right column.

FIG. 4.

ADF-mapping of STEM for the different samples of ZCO is exhibited in the extreme left column. The elemental mapping is demonstrated in a manner of zinc (green), oxygen (red), and copper (purple). RGB mapping is shown in the extreme right column.

Close modal
FIG. 5.

(a) Bright field (BF) images of STEM for the different variants of ZCO where the tiny colored (red-pristine, orange-1%, blue-3%, and green-5%; of Cu-doping) spheres represent the point-probes to obtain the EELS spectra at Zn L-edge (b), O K-edge (c), and Cu L-edge (d) and (e).

FIG. 5.

(a) Bright field (BF) images of STEM for the different variants of ZCO where the tiny colored (red-pristine, orange-1%, blue-3%, and green-5%; of Cu-doping) spheres represent the point-probes to obtain the EELS spectra at Zn L-edge (b), O K-edge (c), and Cu L-edge (d) and (e).

Close modal

Initiated by the spectra of EELS where Cu is shown to be occupying the states of Cu1+ and Cu2+, it is essentially required to locate the exact oxidation of Cu in ZCO. XPS is an element-specific probing technique where the utmost depth of penetration lies between 5 and 10 nm and is highly sensitive to the local electronic states.31 Cu is very susceptible to the environment that immediately gets oxidized at room temperature, in other words, it is meaningful to compare the spectral features of ZCO with several cuprates to confirm the multivalent states of Cu. A comparative analysis in view of the triple oxidation of Cu is thoroughly governed via the multiplet features of Cu2O,32 CuO,32 and NaCuO233 [Fig. 6(a)]. To locate the different oxidation states, three peaks of Gaussian fitting are performed at Cu 2p3/2 core using the Origin 8.6 software [Figs. 6(b)6(d)]. The percentage-wise weightage of each oxidation state of Cu in ZCO is provided in Table II, where the dominance of Cu2+ is clearly visible with the increase in doping. Cu2+, due to its characteristic single-unoccupancy in the d-shell, plays a crucial role in determining RTFM. A majority of reports have governed ferromagnetic interaction in ZCO based upon the exchange mechanism between the different oxidation states of Cu, such as Cu1+-Cu2+,9 Cu2+-Cu3+,10 and Cu2+-Cu2+.34 Evidently, Cu2+ is an indispensable tool to proclaim RTFM in ZCO. In the present scenario, there is a certain likelihood of DE interaction between Cu2+-Cu3+, mediating via VO.

FIG. 6.

(a) Cu 2p XPS, in comparison with the spectra of Cu2O,32 CuO,32 and NaCuO233 to confirm the multivalency of Cu in ZCO, and is explicitly shown by the Gaussian fitting (b)–(d).

FIG. 6.

(a) Cu 2p XPS, in comparison with the spectra of Cu2O,32 CuO,32 and NaCuO233 to confirm the multivalency of Cu in ZCO, and is explicitly shown by the Gaussian fitting (b)–(d).

Close modal
TABLE II.

Variation of the compositional weightage of triple valency of Cu in ZCO as a consequence of the Gaussian fitting of Cu 2p3/2 XPS.

Zn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
Cu1+ (%) 35.09 40.62 29.78 
Cu2+ (%) 20.61 55.55 64.17 
Cu3+ (%) 44.30 3.83 6.05 
Zn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
Cu1+ (%) 35.09 40.62 29.78 
Cu2+ (%) 20.61 55.55 64.17 
Cu3+ (%) 44.30 3.83 6.05 

VSM is utilized to understand the magnetic properties of ZCO. The possibility of ferromagnetic interaction in ZCO is clearly outlined by the obtained loop of hysteresis [Fig. 7(a)]. The values of saturation magnetization lie in the proximity of 0.21, 0.32, and 0.48 μB/uc (Bohr magneton per unit-cell), respectively, for the different variants of ZCO in an ascending manner of doping. A magnification of the obtained loop of hysteresis clearly exhibits the coercivity, as is shown by Fig. 7(b), where all the samples tend to occupy a value in the neighborhood of 0.028 T. In order to proclaim FM, it is essentially required to locate TC. Considering this fact, a plot for magnetization vs temperature (M-T) is also shown [Fig. 7(c)], which clearly exhibits the existence of TC in the vicinity of 323 K in the Cu-doped samples. It is to be noted that neither Cu nor its oxides are magnetic at the ambient temperature; hence, the observed nature of magnetism is believed to be a typical example of FM because of doping and creation of defects. Moreover, some sort of hysteresis is also detected for the pristine ZnO, inset of Fig. 7(a), and its M-T characteristics clearly underlines the essence of VO.35 VSM provides information as a bulk, which is an implicit way; nevertheless, its outcome is clear enough to support FM in ZCO. However, we are interested in the origin of FM, i.e., a more powerful and sophisticated technique is required to resolve this. XMCD is an elemental precise technique that aims to secure FM in the quantum realm and hence is capable of determining the microscopic origin of FM.36,37 Nowadays, any spintronics-based implementation is a matter of full-scale integration of XMCD technique.38,39

FIG. 7.

(a) Hysteresis shown by the powdered samples of ZCO at 300 K using VSM, and inset is a magnified view of the pristine-ZnO. (b) A magnification of the VSM data to locate the coercivity in Cu-doped ZnO. (c) M-T curves, which clearly exhibit the existence of TC, is well above 300 K in the doped samples.

FIG. 7.

(a) Hysteresis shown by the powdered samples of ZCO at 300 K using VSM, and inset is a magnified view of the pristine-ZnO. (b) A magnification of the VSM data to locate the coercivity in Cu-doped ZnO. (c) M-T curves, which clearly exhibit the existence of TC, is well above 300 K in the doped samples.

Close modal

Figure 8 exhibits the spectral features of XAS and XMCD in the TEY configuration. A comparative analysis of Cu L XAS is demonstrated by referring to the multiplet features of Cu2O, CuO,40 NaCuO2,40 and La2Li0.5Cu0.5O441,42 [Fig. 8(a)]. The comparison corroborates with the outcomes of EELS and XPS, concerning with the triple oxidation of Cu in ZCO. XMCD is the resultant of the discrimination of photon-helicities (μ, μ+), governing handedness of the absorption spectra. This discrimination is clearly visible for the highest doping concentration of 5% [Fig. 8(b)], resulting in the spectrum of XMCD [Fig. 8(c)]. The intensity feature of XMCD confirms the role of Cu in ZCO for the observed ferromagnetic characteristics, as is also shown by the hysteresis of VSM. In a similar manner, a comparison based upon Cu L XAS is also established in the TFY configuration [Fig. 9(a)]. Upon careful inspection, an enlarged difference in the helicities is clearly visible [Fig. 9(b)], resulting in a profound hyperfine spectrum of XMCD [Fig. 9(c)]. Indeed, the intensity feature of XMCD in TFY is 100 times stronger than that in TEY, which can be attributed to the localization of Cu2+ in the bulk regime. Consequently, the weakly magnetized feature of XMCD in TEY reflects the ubiquitous nature of Cu3+ on the surface. In terms of penetration, TFY is capable of providing 100 nm of in-depth while TEY has the maximum depth of 5 nm,16,17 due to which TFY is considered as a bulk-sensitive configuration while TEY is sensitive to the surface only.

FIG. 8.

(a) Cu L XAS in TEY, in comparison with the spectra of Cu2O, CuO,40 NaCuO2,40 and La2Li0.5Cu0.5O441,42 to exhibit multivalent states of Cu in ZCO; (b) and (c), respectively, represents the XAS and XMCD spectrum of Zn0.95Cu0.05O in TEY.

FIG. 8.

(a) Cu L XAS in TEY, in comparison with the spectra of Cu2O, CuO,40 NaCuO2,40 and La2Li0.5Cu0.5O441,42 to exhibit multivalent states of Cu in ZCO; (b) and (c), respectively, represents the XAS and XMCD spectrum of Zn0.95Cu0.05O in TEY.

Close modal
FIG. 9.

(a) Cu L XAS in TFY, in comparison with the spectra of Cu2O, CuO,40 NaCuO2,40 and La2Li0.5Cu0.5O441,42 to exhibit the multivalent states of Cu in ZCO; (b) and (c), respectively, represent the XAS and XMCD spectrum of Zn0.95Cu0.05O in TFY.

FIG. 9.

(a) Cu L XAS in TFY, in comparison with the spectra of Cu2O, CuO,40 NaCuO2,40 and La2Li0.5Cu0.5O441,42 to exhibit the multivalent states of Cu in ZCO; (b) and (c), respectively, represent the XAS and XMCD spectrum of Zn0.95Cu0.05O in TFY.

Close modal

To confirm such an in-depth diversity in magnetism, spin and orbital parts are calculated separately using the XMCD sum rule at 2 T (Table III). These values clearly show a large gap in magnetization for the two yield configurations, which reveal that the surface is weakly magnetized. Furthermore, the calculated values of magnetization (M) are plotted against the magnetic field (H) [Fig. 10(a)]. A linear extrapolation is performed over the M-H curve where there is a clear tendency of having retentivity upon zero field. Also, the nature of magnetization against temperature (M-T) strongly suggests the existence of TC is well above 300 K [Fig. 10(b)]. The implicit nature of FM from VSM is confirmed explicitly via XMCD. Improvement in magnetization with the increase in Cu-content strongly favors carrier-induced ferromagnetism (CIF).43–45 Moreover, it is the DE-interaction between Cu2+ and Cu3+ in the presence of VO that is responsible for the ferromagnetic interaction in ZCO.

TABLE III.

Spin and orbital contributions of magnetization being calculated separately at a field strength of 2 T in TEY and TFY modes using the XMCD sum-rule.

XMCDZn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
Magnetization (μB)TEYTFYTEYTFYTEYTFY
Spin (Sz0.033 0.123 0.258 0.391 0.318 0.577 
Orbital (Lz0.024 0.045 0.063 0.126 0.082 0.132 
XMCDZn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
Magnetization (μB)TEYTFYTEYTFYTEYTFY
Spin (Sz0.033 0.123 0.258 0.391 0.318 0.577 
Orbital (Lz0.024 0.045 0.063 0.126 0.082 0.132 
FIG. 10.

(a) Magnetization vs field (M-H) curves because of Cu L XMCD at 300 K in TFY to exhibit FM in Cu-doped ZnO. (b) Magnetization vs temperature (M-T) curves at a field intensity of 2 T in TFY, which explicitly confirm RTFM in ZCO, as the nature of variation strongly suggests that TC is exceeding 300 K.

FIG. 10.

(a) Magnetization vs field (M-H) curves because of Cu L XMCD at 300 K in TFY to exhibit FM in Cu-doped ZnO. (b) Magnetization vs temperature (M-T) curves at a field intensity of 2 T in TFY, which explicitly confirm RTFM in ZCO, as the nature of variation strongly suggests that TC is exceeding 300 K.

Close modal

In a doped system of ZnO, spintronics properties are dependent on the creation of defects as well as the nature of impurity. In particular, Zn ions have no such significant role in it and the electronic structures mostly remain unaffected over doping.9 The same scenario is also applicable in the present study where a consensus in the spectral features of Zn L XAS is observed with doping of Cu [Fig. 11(a)]. However, XAS spectra of O K-edge are vital for ZCO, which requires a rigorous analytical approach. Five distinct features are clearly observable in the spectra of O K XAS [Fig. 11(b)]. The marked features are found elsewhere, A(533 eV), B(537.5 eV), C(540.8 eV), D(542.5 eV), and E(553.7 eV). Feature-A corresponds to the hybridization between O 2p–Cu 3d and the improvement in its spectral-weight with doping suggests the incorporation of Cu.18 Feature-B is assigned to the hybridization between O 2p–Zn 4s/4p, while feature-C and feature-D, respectively, connote the hybridized states of O 2p–Zn 4p and O 2p–Cu 4s/4p and are vital for the oxygen-mediated disorder in ZCO.46 Feature-E is attributed to the hybridization between O 2p and the higher orbitals of Zn, which, in particular, is independent of doping. Moreover, the observation of the pre-edge feature A is attributed to several interesting activities. It has been strongly claimed that the observation of the pre-edge feature is the reason for the switch in the complex ligand coordination from octahedral to tetrahedral, which helps to secure RTFM in ZCO.18 Pre-edge features are directly linked to the unoccupancy of O 2p–Cu 3d bands that result in the percolation of charge carriers, viz., electrons and holes.47–49 

FIG. 11.

XAS spectra of pristine and Cu-doped ZnO taken at (a) Zn L-edge and (b) O K-edge.

FIG. 11.

XAS spectra of pristine and Cu-doped ZnO taken at (a) Zn L-edge and (b) O K-edge.

Close modal

XAS is highly sensitive to the local electronic states and is vital to understand the local electronic structures.50–52, L-edge XAS is comprised of the five well-known lobes of the d-shell, i.e., dxy, dyz, dxz, dx2-y2, and dz2, in which the first three are the lobes of t2g while the last two lobes represent eg. It is very difficult to deconvolute t2g due to the symmetrical arrangement of the electron clouds that oppose any modification. On the contrary, the bi-spallation of eg is suitable to underline the defect-oriented modification in the crystal structure. Considering all these facts, we have performed Gaussian fittings of Cu L3 XAS in TEY and TFY modes using Origin 8.6 software [Figs. 12(a)12(f)]. In dealing with transition-oxides, unoccupancy of the d-shell is essentially required. It is seen that the L-edge XAS of the transition element is directly proportional to the unoccupancy of the d-shell.53,54 Furthermore, the peak profile of t2g is sensitive to the valency.55 There is a smooth increase in the spectral weight of t2g and eg in both the TEY and TFY modes upon doping [Figs. 12(g)12(h)]. Nonetheless, the extent of variation of t2g in TEY is comparatively low to that in TFY. On the contrary, the extent of variation of eg in TEY is comparatively large to that in TFY. Due to this the startling nature of variation in the spectral weight ratio of t2g/eg is reported, where there is a clear degradation in TEY but a stark improvement in TFY is observed upon doping [Fig. 12(i)]. In a tetrahedrally coordinated compound, the majority of unoccupancy would lie in t2g while eg is mostly occupied. Evidently, Figs. 12(g)12(i) confirm that the majority of Cu2+ is localized in the bulk while the surface is mostly occupied by Cu3+. Moreover, the lift in the degeneracy of eg into dx2-y2 and dz2 is attributed to the distortion in the wurtzite crystal of ZnO.56,57 The ascribed values of the constituent parameters of Cu L3 XAS due to the Gaussian fittings are represented in Table IV.

FIG. 12.

Deconvolution of Cu L3 XAS into eg (dx2-y2, dz2) and t2g: (a)–(c) TEY; (d)–(f) TFY. Variation of the deconvoluted parameters over Cu-doping (g)–(i).

FIG. 12.

Deconvolution of Cu L3 XAS into eg (dx2-y2, dz2) and t2g: (a)–(c) TEY; (d)–(f) TFY. Variation of the deconvoluted parameters over Cu-doping (g)–(i).

Close modal
TABLE IV.

Compositional spectral-weight of dx2-y2, dz2, eg, t2g, and t2g/eg over Cu-doping, as a consequence of the deconvolution of Cu L3 XAS in TEY and TFY.

Zn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
XMCDTEYTFYTEYTFYTEYTFY
dx2-y2 0.003 0.013 0.07 0.019 0.09 0.023 
dz2 0.02 0.06 0.17 0.12 0.31 0.19 
eg 0.023 0.073 0.24 0.139 0.40 0.213 
t2g 0.004 0.02 0.005 0.05 0.006 0.08 
t2g/eg 0.174 0.274 0.021 0.360 0.015 0.376 
Zn0.99Cu0.01OZn0.97Cu0.03OZn0.95Cu0.05O
XMCDTEYTFYTEYTFYTEYTFY
dx2-y2 0.003 0.013 0.07 0.019 0.09 0.023 
dz2 0.02 0.06 0.17 0.12 0.31 0.19 
eg 0.023 0.073 0.24 0.139 0.40 0.213 
t2g 0.004 0.02 0.005 0.05 0.006 0.08 
t2g/eg 0.174 0.274 0.021 0.360 0.015 0.376 

The oxidation states of host and dopant are at the focal point to the understanding of the spintronics properties of any doped transition oxide. But for a system of ZCO, inclination remains toward governing the states of Cu rather than Zn, which is mostly associated with the arrangement of electrons in the d-shell. As an element, neither Cu nor Zn has any vacancy in its d-shell. Nonetheless, there is possibility of Cu being oxidized at room temperature and is a common practice. The valence shell electronic arrangement of Cu is 4s1 3d10, which creates a favorable environment for the d-electron to be in a state of correlation when being exposed to another element of a similar arrangement, like Zn, due to a single unoccupancy in the s-orbital. Cu2+ is energetically more plausible than Cu1+, which respectively, forms a coordination with O as CuO and Cu2O. But Cu2O has a major drawback of completely filled d-orbital, which is an undesired constraint in the smooth percolation of carriers for the exchange mechanism. The number of unpaired electrons in the d-orbital is vital for any magnetic-based characterization. Considering the formation of Cu2O3 or Cu3+, the valence shell arrangement is 3d8 where the number of unpaired electrons is two for the high spin configuration. But for the low spin configuration, there are no unpaired electrons for Cu3+. Consequently, there is a single unpaired electron for Cu2+, which is independent of high/low spin configuration. This makes Cu2+ unique in defining FM in ZCO.

The possibility of an exchange mechanism between Cu1+(d10)–Cu2+(d9) cannot be ignored, which is ferromagnetic but due to the level of percolation threshold and the amount of Cu1+ carriers in ZCO are not significant to establish a correlated state of d10-d9.58,59 On the contrary, a highly correlated state of Cu2+(d9)–Cu3+(d8) is established by means of DE-interaction and is mediating in the presence of VO (Fig. 13). In a tetrahedral complex, defect in the form of vacancy is more favorable than interstitial due to its chiral structure. This chirality switches the lobes of t2g and eg in such a manner that t2g occupies the anti-bonding state. The observation of the pre-edge feature in the spectra of O K XAS, which is common for a system of transition-oxides considers the switching action of ligands in a more rigorous way and attributes this action to favor the exchange interaction. The dependency of oxidation on t2g and the number of holes that mainly lie in t2g for a tetrahedrally coordinated system reveal a strong hybridization between Cu 3d–O 2p in ZCO. The formation of cuprates is a process of the rearrangement of electron-affinity, which ultimately results in VO.

FIG. 13.

(a) An illustration of the oxygen-mediated disorder in the hexagonal-wurtzite crystal structure of ZnO due to Cu-doping, being created using VESTA 3.5.8 software. (b) A schematic of the microscopic origin of DE-interaction between Cu2+-Cu3+ in the presence of VO to confirm RTFM in the nanocrystalline powder of ZCO.

FIG. 13.

(a) An illustration of the oxygen-mediated disorder in the hexagonal-wurtzite crystal structure of ZnO due to Cu-doping, being created using VESTA 3.5.8 software. (b) A schematic of the microscopic origin of DE-interaction between Cu2+-Cu3+ in the presence of VO to confirm RTFM in the nanocrystalline powder of ZCO.

Close modal

Magnetic properties of ZCO completely rely on Cu2+ and defects in the form of VO. However, a competitive scenario between Cu1+(d10)–Cu2+(d9) and Cu2+(d9)–Cu3+(d8) result in a state of magnetic frustration.60 Such a state of magnetic frustration is one of the reasons for the weakly magnetized TEY mode as most of Cu2+ is deep inside the bulk-regime. There is a considerable lag in the itinerancy of Cu1+ as compared to Cu3+ because of the absence of any unpaired spin. Low itinerancy of Cu1+ acts as a barrier for the smooth percolation of correlation between d10-d9. This is typically shown by the strange nature of variation of t2g/eg over Cu doping [Fig. 12(i)], regardless of the increase in the spectral weight of t2g and eg, individually. Somehow, the formation of Cu1+ and Cu3+ might form another correlated state of d10-d8, which is antiferromagnetic by means of super-exchange interaction.58,59 Moreover, the state of magnetic frustration is mostly exhibited by TEY whereas TFY is strongly magnetized by the virtue of DE-interaction between d9-d8, which is ferromagnetic. In a complex tetrahedron of ZCO where the lobes of eg are completely filled, its deconvolution is directly related to the creation of defect and its confirmation as VO is attributed to the deconvolution of E2high-phonon.

Interestingly, the carrier-induced aspect of FM in ZCO has not been discussed to this date. In this study, different carrier concentrations of Cu not only play a significant role in the substitution of Zn rather they have shown to increase the magnetization. The present study identifies the long-range room temperature ferromagnetism in the nanocrystalline powder of Cu-doped ZnO. This long-range FM is due to the Cu2+ ions in ZCO with the presence of the unpaired electron spin in the 3d-orbital. A microscopic indirect double exchange interaction proposed by Herng et al.9 also suggests the necessity of oxygen vacancies (VO) for the smooth correlation to form a long-range ferromagnetic state. The unoccupied 3d-shell of Cu and its mediation in the presence of VO to form a correlated state of Bmp is responsible for the long-range FM, which is observed in the ZCO. A similar tendency is observed in both TEY and TFY modes. Due to the high electronegativity of Cu, it is supposed to replace Zn from the tetrahedrally coordinated wurtzite crystal. As the atomic-weight concentration of Cu is increased, ferromagnetically aligned carriers of Cu resulted in the enlarged magnetization, which is obvious from the hyperfine spectra of XMCD. Evidently, the shift in saturation-magnetization for undoped ZnO to 1% of Cu-content in VSM and further its enhancement is strong evidence of CIF.26 

Experimental results point to the spin polarized conducting carriers in Cu-doped ZnO. This is expected due to the formation of correlated states between the different Cu-ions in the presence of VO. A state of magnetic frustration between AFM and FM is caused due to the interaction between Cu2+-Cu2+ and Cu2+-Cu3+, respectively, and affects the formation of Bmp. A study by Ye et al.61 confirms the half-metallic ferromagnetism in Cu-doped ZnO based on the density functional calculations. Furthermore, the availability of the unpaired spin in the d-orbital of Cu2+ is free from the arrangement based upon high/low spin configuration but there is a restriction of having the high spin arrangement of Cu3+ state. In a complex tetrahedron of ZCO, the holes mainly lie in the t2g ligand. All these attributes indicate that only the Cu2+ state is responsible for the effective creation and percolation of Bmp in the presence of distortion, which is seen in the form of VO. Apart from exhibiting a state of magnetic frustration in TEY, a small but detectable increase in magnetization over an increase in carrier concentration of Cu is also observed. The prediction of exact oxidation of transition metal is a common issue, which is also evident from the triple oxidation of Cu in ZCO. Nevertheless, XPS truly underlines that the majority of Cu is shown to occupy the oxidation of Cu2+, which is the major reason for ferromagnetic alignment in a system of ZCO.

We developed a model to investigate the exact origin of coupling between different oxidation states of Cu in ZCO and found that it is a highly correlated aspect of coupling between Cu2+-Cu3+ in the presence of VO, which is responsible for RTFM. The confirmation of VO is clearly exhibited by (a) deconvolution of E2high-phonon Raman; (b) Cu L-edge spectra of EELS; and (c) deconvolution of eg into dx2-y2 and dz2 in the Cu L3 XAS. Furthermore, the heavily magnetized TFY configuration and localization of Cu2+ therein is strongly shown by the startling nature of variation in t2g/eg upon doping. Indeed, the results obtained from XMCD clearly demonstrate that the intensity features are 100 times stronger in TFY than in TEY. Moreover, weakly magnetized TEY mode is an attribute of magnetic frustration due to the incapability of active Cu-ions to percolate smoothly on the surface. The ferromagnetic coupling in ZCO is observed in the form of CIF. Any modeling to locate FM in ZCO must consider the oxidation of Cu2+ and defect in the form of VO, simultaneously.

A. Kumar acknowledges CUSB Gaya/UGC, New Delhi, India, for the financial support to carry out this work. The experiment at the Photon Factory was approved by the Program Advisory Committee (Proposal No. 2021G501). The TEM work was supported by Singapore’s National Research Foundation (NRF) under the Competitive Research Program (Grant No. NRF-CRP23-2019-0001).

The authors have no conflicts to disclose.

A. Kumar: Data curation (lead); Formal analysis (lead); Investigation (lead); Methodology (lead); Software (lead); Validation (lead); Writing – original draft (lead); Writing – review & editing (lead). T. Ghosh: Data curation (supporting); Formal analysis (supporting); Methodology (supporting); Writing – review & editing (supporting). Z. Aabdin: Formal analysis (supporting); Funding acquisition (supporting); Investigation (equal); Writing – review & editing (supporting). J. Roy: Methodology (equal); Writing – review & editing (supporting). V. K. Verma: Formal analysis (supporting); Validation (equal); Writing – review & editing (equal). A. Ghosh: Formal analysis (supporting); Methodology (supporting). S. K. Sahoo: Formal analysis (supporting); Investigation (supporting); Writing – review & editing (supporting). R. Urkude: Data curation (supporting); Formal analysis (supporting); Writing – review & editing (supporting). S. Bhunia: Investigation (supporting); Methodology (supporting); Writing – review & editing (supporting). U. K. Goutam: Data curation (supporting); Investigation (supporting); Writing – review & editing (supporting). K. Amemiya: Data curation (supporting); Formal analysis (supporting); Writing – review & editing (supporting). A. Kandasami: Data curation (supporting); Formal analysis (supporting); Validation (equal); Writing – review & editing (equal). V. R. Singh: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Software (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

4.
T.
Dietl
,
H.
Ohno
,
F.
Matsukura
,
J.
Cibert
, and
D.
Ferrand
,
Science
287
,
1019
(
2000
).
5.
Q.
Wang
,
Q.
Sun
,
G.
Chen
,
Y.
Kawazoe
, and
P.
Jena
,
Phys. Rev. B
77
,
205411
(
2008
).
6.
D.
Kim
,
J.-h.
Yang
, and
J.
Hong
,
J. Appl. Phys.
106
,
013908
(
2009
).
7.
L. M.
Huang
,
A. L.
Rosa
, and
R.
Ahuja
,
Phys. Rev. B
74
,
075206
(
2006
).
8.
N. N.
Lathiotakis
,
A. N.
Andriotis
, and
M.
Menon
,
Phys. Rev. B
78
,
193311
(
2008
).
9.
T. S.
Herng
,
D.-C.
Qi
,
T.
Berlijn
,
J. B.
Yi
,
K. S.
Yang
,
Y.
Dai
,
Y. P.
Feng
,
I.
Santoso
,
C.
Sanchez-Hanke
,
X. Y.
Gao
,
A. T. S.
Wee
,
W.
Ku
,
J.
Ding
, and
A.
Rusydi
,
Phys. Rev. Lett.
105
,
207201
(
2010
).
10.
T.
Kataoka
,
Y.
Yamazaki
,
V. R.
Singh
,
A.
Fujimori
et al,
Phys. Rev. B
84
,
153203
(
2011
).
11.
P. A.
Wolff
,
R. N.
Bhatt
, and
A. C.
Durst
,
J. Appl. Phys.
79
,
5196
(
1996
).
12.
A. C.
Durst
,
R. N.
Bhatt
, and
P. A.
Wolff
,
Phys. Rev. B
65
,
235205
(
2002
).
13.
J. M. D.
Coey
,
M.
Venkatesan
, and
C. B.
Fitzgerald
,
Nat. Mater.
4
,
173
179
(
2005
).
14.
S.
Gautam
,
S.
Kumar
,
P.
Thakur
,
K. H.
Chae
,
R.
Kumar
,
B. H.
Koo
, and
C. G.
Lee
,
J. Phys. D: Appl. Phys.
42
,
175406
(
2009
).
15.
M.
Sajjad
,
I.
Ullah
,
M. I.
Khan
,
J.
Khan
,
M. Y.
Khan
, and
M. T.
Qureshi
,
Results Phys.
9
,
1301
1309
(
2018
).
16.
H.
Wadati
,
A. J.
Achkar
,
D. G.
Hawthorn
,
T. Z.
Regier
,
M. P.
Singh
,
K. D.
Truong
,
P.
Fournier
,
G.
Chen
,
T.
Mizokawa
, and
G. A.
Sawatzky
,
Appl. Phys. Lett.
100
,
193906
(
2012
).
17.
D.
Asakura
,
E.
Hosono
,
Y.
Nanba
,
H.
Zhou
,
J.
Okabayashi
,
C.
Ban
,
P.-A.
Glans
,
J.
Guo
,
T.
Mizokawa
,
G.
Chen
,
A. J.
Achkar
,
D. G.
Hawthron
,
T. Z.
Regier
, and
H.
Wadati
,
AIP Adv.
6
,
035105
(
2016
).
18.
R.
Bhardwaj
,
A.
Bharti
,
J. P.
Singh
,
K. H.
Chae
, and
N.
Goyal
,
Nanoscale Adv.
2
,
4450
(
2020
).
19.
T. C.
Damen
,
S. P. S.
Porto
, and
B.
Tell
,
Phys. Rev.
142
,
570
(
1966
).
20.
J.
Serrano
,
A. H.
Romero
,
F. J.
Manjón
,
R.
Lauck
,
M.
Cardona
, and
A.
Rubio
,
Phys. Rev. B
69
,
094306
(
2004
).
21.
M. S.
Jang
,
M. K.
Ryu
,
M. H.
Yoon
,
S. H.
Lee
,
H. K.
Kim
,
A.
Onodera
, and
S. A.
Kojima
,
Curr. Appl. Phys.
9
,
651
(
2009
).
22.
K. A.
Alim
,
V. A.
Fonoberov
, and
A. A.
Balandin
,
Appl. Phys. Lett.
86
,
053103
(
2005
).
23.
G. T.
Du
,
Y.
Ma
,
Y. T.
Zhang
, and
T. P.
Yang
,
Appl. Phys. Lett.
87
,
213103
(
2005
).
24.
J.
Serrano
,
F. J.
Manjón
,
A. H.
Romero
,
F.
Widulle
,
R.
Lauck
, and
M.
Cardona
,
Phys. Rev. Lett.
90
,
055510
(
2003
).
25.
A.
Ghosh
and
R. N. P.
Choudhary
,
J. Appl. Phys.
105
,
124906
(
2009
).
26.
A.
Kumar
,
M.
Zzaman
,
A.
Kumari
,
J. B.
Franklin
,
S.
Srivastava
,
V. K.
Verma
,
K.
Amemiya
,
Y.
Miura
,
A.
Kandasami
, and
V. R.
Singh
,
J. Appl. Phys.
134
,
035304
(
2023
).
27.
V.
Rajendar
,
C. H.
Shilpa Chakra
,
B.
Rajitha
,
K.
Venkateswara Rao
,
M.
Chandra Sekhar
,
B.
Purusottam Reddy
, and
S.-H.
Park
,
J. Mater. Sci.: Mater. Electron.
28
,
3272
3277
(
2017
).
28.
Q.
Yuanshen
,
Y.
Kauffmann
,
A.
Kosinova
,
A. R.
Kilmametov
,
B.
Straumal
, and
E.
Rabkin
,
Mater Res. Lett.
9
,
58
64
(
2021
).
29.
H.
Kabbara
,
J.
Ghanbaja
,
C.
Noël
, and
T.
Belmonte
,
Mater. Chem. Phys.
207
,
350
358
(
2018
).
30.
S.
Krishnamurthy
,
C.
McGuinness
,
L. S.
Dorneles
,
M.
Venkatesan
,
J. M. D.
Coey
,
J. G.
Lunney
,
C. H.
Patterson
,
K. E.
Smith
,
T.
Learmonth
,
P.-A.
Glans
,
T.
Schmitt
, and
J. H.
Guo
,
J. Appl. Phys.
99
,
08M111
(
2006
).
31.
J. B.
Gilbert
,
M. F.
Rubner
, and
R. E.
Cohen
,
Proc. Natl. Acad. Sci. U. S. A.
110
,
6651
6656
(
2013
).
32.
J.
Ghijsen
,
L. H.
Tjeng
,
J.
van Elp
,
H.
Eskes
,
J.
Westerink
,
G. A.
Sawatzky
, and
M. T.
Czyzyk
,
Phys. Rev. B
38
,
11322
(
1988
).
33.
P.
Steiner
,
V.
Kinsinger
,
I.
Sander
,
B.
Siegwart
,
S. H.
Pfner
,
C.
Politis
,
R.
Hoppe
, and
H. P.
Muller
,
Z. Phys. B
67
,
497
(
1987
).
34.
N.
Ali
,
B.
Singh
,
Z. A.
Khan
,
A. R.
Vijaya
,
K.
Tarafder
, and
S.
Ghosh
,
Sci. Rep.
9
,
2461
(
2019
).
35.
D. C.
Agarwal
,
U. B.
Singh
,
S.
Gupta
,
R.
Singhal
,
P. K.
Kulriya
,
F.
Singh
,
A.
Tripathi
,
J.
Singh
,
U. S.
Joshi
, and
D. K.
Avasthi
,
Sci. Rep.
9
,
6675
(
2019
).
36.
T.
Funk
,
A.
Deb
,
S. J.
George
,
H.
Wang
, and
S. P.
Cramer
,
Coord. Chem. Rev.
249
,
3
30
(
2005
).
37.
G.
van der Laan
and
A. I.
Figueroa
,
Coord. Chem. Rev.
277–278
,
95
129
(
2014
).
38.
T.
Ueno
,
H.
Hino
,
A.
Hashimoto
,
Y.
Takeichi
,
M.
Sawada
, and
K.
Ono
,
npj Comput. Mater.
4
,
4
(
2018
).
39.
B.
Dieny
,
I. L.
Prejbeanu
,
K.
Garello
,
P.
Gambardella
,
P.
Freitas
,
R.
Lehndorff
,
W.
Raberg
,
U.
Ebels
,
S. O.
Demokritov
,
J.
Akerman
,
A.
Deac
,
P.
Pirro
,
C.
Adelmann
,
A.
Anane
,
A. V.
Chumak
,
A.
Hirohata
,
S.
Mangin
,
S. O.
Valenzuela
,
M. C.
Onbaşlı
,
M.
d’Aquino
,
G.
Prenat
,
G.
Finocchio
,
L.
Lopez-Diaz
,
R.
Chantrell
,
O.
Chubykalo-Fesenko
, and
P.
Bortolotti
,
Nat. Electron.
3
,
446
459
(
2020
).
40.
D. D.
Sarma
,
O.
Strebel
,
C. T.
Simmons
,
U.
Neukirch
,
G.
Kaindl
,
R.
Hoppe
, and
H. P.
Müller
,
Phys. Rev. B
37
,
9784
(
1988
).
41.
Z.
Hu
,
G.
Kaindl
,
S. A.
Warda
,
D.
Reinen
,
F. M. R.
de Groot
, and
B. G.
Müller
,
Chem. Phys.
232
,
63
(
1998
).
42.
C. T.
Chen
,
L. H.
Tjeng
,
J.
Kwo
,
H. L.
Kao
,
P.
Rudolf
,
F.
Sette
, and
R. M.
Fleming
,
Phys. Rev. Lett.
68
,
2543
(
1992
).
43.
Y.
Yamada
,
K.
Ueno
,
T.
Fukumura
,
H. T.
Yuan
,
H.
Shimotani
,
Y.
Iwasa
,
L.
Gu
,
S.
Tsukimoto
,
Y.
Ikuhara
, and
M.
Kawasaki
,
Science
332
,
1065
(
2011
).
44.
V. R.
Singh
,
Y.
Sakamoto
,
T.
Kataoka
,
M.
Kobayashi
,
Y.
Yamazaki
,
A.
Fujimori
,
F.-H.
Chang
,
D.-J.
Huang
,
H.-J.
Lin
,
C. T.
Chen
,
H.
Toyosaki
,
T.
Fukumura
, and
M.
Kawasaki
,
J. Phys.: Condens. Matter
23
,
176001
(
2011
).
45.
V. R.
Singh
,
K.
Ishigami
,
V. K.
Verma
,
G.
Shibata
,
Y.
Yamazaki
,
T.
Kataoka
,
A.
Fujimori
,
F.-H.
Chang
,
D.-J.
Huang
,
H.-J.
Lin
,
C. T.
Chen
,
Y.
Yamada
,
T.
Fukumura
, and
M.
Kawasaki
,
Appl. Phys. Lett.
100
,
242404
(
2012
).
46.
S.
Kumar
,
P.
Vats
,
S.
Gautam
,
V. P.
Gupta
,
K. D.
Verma
,
K. H.
Chae
,
M.
Hashim
, and
H. K.
Choi
,
Mater. Res. Bull.
59
,
377
381
(
2014
).
47.
Z.
Hu
,
S.-L.
Drechsler
,
J.
M’alek
,
H.
Rosner
,
R.
Neudert
,
M.
Knupfer
,
M. S.
Golden
,
J.
Fink
,
J.
Karpinski
,
G.
Kaindl
,
C.
Hellwig
, and
C.
Jung
,
Europhys. Lett.
59
,
135
(
2002
).
48.
M.
Karppinen
,
M.
Kotiranta
,
T.
Nakane
,
H.
Yamauchi
,
S. C.
Chang
,
R. S.
Liu
, and
J. M.
Chen
,
Phys. Rev. B
67
,
134522
(
2003
).
49.
K.
Asokan
,
J. C.
Jan
,
K. V. R.
Rao
,
J. W.
Chiou
,
H. M.
Tsai
,
S.
Mookerjee
,
W. F.
Pong
,
M.-H.
Tsai
,
R.
Kumar
,
S.
Husain
, and
J. P.
Srivastava
,
J. Phys.: Condens. Matter
16
,
3791
(
2004
).
50.
J.
Yano
and
V. K.
Yachandra
,
Photosynth. Res.
102
,
241
254
(
2009
).
51.
C. L.
Chen
,
C.-L.
Dong
,
C.-H.
Chen
,
J.-W.
Wu
,
Y.-R.
Lu
,
C.-J.
Lin
,
S.
Ya Hsuan Liou
,
C.-M.
Tseng
,
K.
Kumar
,
D.-H.
Wei
,
J.
Guo
,
W.-C.
Chou
, and
M.-K.
Wu
,
Phys. Chem. Chem. Phys.
17
,
22064
(
2015
).
52.
K.
Wojtaszek
,
W.
Blachucki
,
K.
Tyrala
,
M.
Nowakowski
,
M.
Zajaç
,
J.
Stępień
,
P.
Jagodziński
,
D.
Banaś
,
W.
Stańczyk
,
J.
Czapla-Masztafiak
,
W. M.
Kwiatek
,
J.
Szlachetko
, and
A.
Wach
,
J. Phys. Chem. A
125
,
50
56
(
2021
).
53.
K.-S.
Yang
,
Y.-R.
Lu
,
Y.-Y.
Hsu
,
C.-J.
Lin
,
C.-M.
Tseng
,
S. Y. H.
Liou
,
K.
Kumar
,
D.-H.
Wei
,
C.-L.
Dong
, and
C.-L.
Chen
,
J. Phys. Chem. C
122
,
6955
(
2018
).
54.
A.
Bhogra
,
A.
Masarrat
,
R.
Meena
,
D.
Hasina
,
M.
Bala
,
C.-L.
Dong
,
C.-L.
Chen
,
T.
Som
,
A.
Kumar
, and
A.
Kandasami
,
Sci. Rep.
9
,
14486
(
2019
).
55.
V.
Kumar
,
A.
Bhogra
,
M.
Bala
,
S. C.
Haw
,
C. L.
Chen
,
C. L.
Dong
,
K.
Asokan
, and
S.
Annapoorni
,
Phys. Rev. B
103
,
024104
(
2021
).
56.
V. R.
Mastelaro
,
P. P.
Neves
,
S. R.
de Lazaro
,
E.
Longo
,
A.
Michalowicz
, and
J. A.
Eiras
,
J. Appl. Phys.
99
,
044104
(
2006
).
57.
H.-W.
Kuo
,
C.-J.
Lin
,
H.-Y.
Do
,
R.-Y.
Wu
,
C.-M.
Tseng
,
K.
Kumar
,
C.-L.
Dong
, and
C.-L.
Chen
,
Appl. Surf. Sci.
502
,
144297
(
2020
).
58.
J. B.
Goodenough
,
Phys. Rev.
100
,
564
(
1955
).
59.
J.
Kanamori
,
J. Phys. Chem. Solids
10
,
87
(
1959
).
60.
O.
Mustonen
,
S.
Vasala
,
T.-L.
Chou
,
J.-M.
Chen
, and
M.
Karppinen
,
Phys. Rev. B
93
,
014405
(
2016
).
61.
L.-H.
Ye
,
A. J.
Freeman
, and
B.
Delley
,
Phys. Rev. B
73
,
033203
(
2006
).