Optical and Electrical Properties of α-MnTe Thin Films Deposited Using RF Magnetron Sputtering

Abstract

The optical and electrical properties of MnTe films were investigated to ascertain the feasibility of their use in solar cell applications. Three α-MnTe thin films with different composition, i.e., Mn-47.9 at% Te, Mn-49.2 at% Te, and Mn-50.4 at% Te, were prepared using RF magnetron sputtering. All the films demonstrated a high light absorption coefficient (0.2 × 10⁵–0.8 × 10⁶ cm⁻¹) and an optimal indirect band gap (1.37–1.52 eV) for solar cell applications. Furthermore, all of them exhibited p-type conductivity, with the Mn-47.9 at% Te film demonstrating three to four times higher carrier mobility (5.2 cm²·V⁻¹·s⁻¹) than the Mn-50.4 at% Te film (1.6 cm²·V⁻¹·s⁻¹).

1. Introduction

The development of photovoltaic generation source as the practical one of renewable energy recently attracts considerable attention. A great increase is prompted in the production of solar cells, most of which are based on silicon,¹⁾ that has a band gap of 1.1 eV.²⁾ Such a band gap is close to the theoretical ideal value (∼1.4 eV) for sunlight absorption.^3,4) In general, the maximum theoretical conversion efficiency of Si-type solar cells is ∼30%.⁴⁾ However, Si has a low absorption coefficient (10³–10⁴ cm⁻¹) at the wavelength of sunlight.⁵⁾ Therefore, the thickness of silicon for solar cells must be in micrometers to ensure that sufficient sunlight is absorbed. This issue hinders the miniaturization of Si solar cell devices. Moreover, compound semiconductor-type solar cells are practical photovoltaic generation devices that generally comprise top and bottom electrodes, a window layer, and an active layer. Because of the larger absorption coefficient of CdTe or CuIn_xGa_1−xSe₂ (CIGS) films (>10⁴ cm⁻¹) than that of silicon,^6,7) they are usually used as the active layer. Thus, the preparation of thin solar cell devices is possible, which contribute to the further development of solar cell technology.

CdTe films possess a band gap of 1.5 eV.⁸⁾ Solar cells having CdTe films have been reported to demonstrate a theoretical efficiency as high as 30% that is almost the same with that of Si-type solar cells.⁹⁾ In general, high efficiencies are achieved with long lifetimes of carriers in the absorption layer.¹⁰⁾ Thus, a high carrier mobility, thus, is necessary to enhance the efficiency. CdTe films have a high hole mobility (15–40 cm² V⁻¹ s⁻¹) that contributes to a great efficiency.¹¹⁾ However, Cd is toxic. The development of Cd-free absorption materials, thus, is essential to overcome the potential environmental impact.^12,13)

Werner et al. reported that CIGS-type cells can reach a theoretical efficiency as high as 33%.¹⁴⁾ The band gap can be controlled from 1.04 to 1.68 eV by changing In or Ga composition in the film,¹⁵⁾ which allows the optimization of the band gap value of CIGS. Furthermore, variation in the composition along the depth direction provides a graded band gap structure that can greatly promote carrier migration.¹⁶⁾ However, CIGS-type solar cells generally contain CdS as a buffer layer that enables a suitable band alignment and surface passivation.¹⁷⁾ Some alternative materials, such as TiO₂, Zn(O, S), or (Zn, Mg)O, have been used to replace CdS as buffer layers. A reduction in the efficiency, however, was noticed.^17,18) Therefore, the development of other alternative absorption materials for compound semiconductor-type solar cells continues to be a challenge.

α-MnTe that has a NiAs-type structure is known to be a p-type semiconductor material.^19–24) Because α-MnTe is known as a diluted magnetic semiconductor, it has been extensively studied in the field of spintronics.^25–30) α-MnTe is reported to show an indirect band gap of ∼1.3 eV^21–23,31) that is very close to the theoretical ideal value (∼1.4 eV) for sunlight absorption. Thus, α-MnTe could be a good candidate to act as the light absorption layer of solar cells.

There is a need to perform systematic study of optical (transmittance and absorption coefficient) and electrical (resistivity, carrier mobility, and carrier concentration) properties of α-MnTe to discuss on the application of α-MnTe to solar cell. Moreover, according to phase diagram of Mn–Te system, α-MnTe exists in the composition range of 42.5 ≤ Te ≤ 51.0 at%, depending on the temperature.³²⁾ Therefore, in the present work, we investigated the optical and electrical properties of α-MnTe films and their composition dependency. To understand the fundamental properties of α-MnTe film for solar cell applications, α-MnTe films were fabricated in this study by conventional sputtering method and annealing, which enabled to obtain thin films having uniform composition and good crystallinity.

2. Experimental Procedures

MnTe films (thickness of ∼100 nm) were deposited on Si (725 µm)/SiO₂ (100 nm) or glass plate substrates (Corning EAGLE-XG) by RF magnetron sputtering at room temperature, using pure Mn and Te targets. The base pressure was lower than 5.0 × 10⁻⁵ Pa and a fixed flow rate of Ar gas was maintained (15 sccm).

The film thickness was controlled by sputtering time that was estimated from the sputtering rate of each target. The actual film thickness was confirmed by atomic force microscopy (AFM) (VN-8000, KEYENCE). For AFM measurement, the substrate surface was partially covered with a marker and then the film was deposited on it. After deposition, the marker was removed and the step, i.e., the film thickness, was measured using AFM.

MnTe films having three different compositions were prepared by adjusting the RF sputter power ratio of Mn and Te targets. The compositions were measured using scanning electron microscopy equipped with energy dispersive spectroscopy (SEM-EDX; JSM-6500F, JEOL). The compositions of the films were Mn-50.4 at% Te, Mn-49.2 at% Te, and Mn-47.9 at% Te, which were designated as 50.4Te, 49.2Te, and 47.9Te, respectively. A thin surface oxide layer was formed on the surface of the films (see section 3.2), so the obtained composition indicates the overall composition of the obtained film having the thin surface oxide layer. To obtain an α-MnTe structure, the films were heated up to 500°C at a heating rate of 10°C/min, followed by furnace cooling. The heat treatment was conducted under Ar gas flow after evacuating the chamber by rotary pump.

The microstructure was observed using field-emission transmission electron microscopy (FE-TEM; HF-2000EDX, HITACHI) at an accelerating voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) analysis were performed using a Thermo Fisher Scientific Theta Probe to investigate the core-level spectra of the MnTe films using an Al Kα source. The films were sputtered using an Ar plasma before the measurement to remove native oxide and surface contamination.

The transmittance and reflectance were measured using a spectrophotometer (V-630 BIO, JASCO) at the wavelength range of 400–1100 nm. The reflectance was measured with respect to the reflectance of an Al reference mirror.

The dependence of the resistance on temperature was measured by the two-point probe method in the temperature range of 50°C–150°C under an Ar atmosphere while cooling. Moreover, at room temperature, Hall effect measurement was performed to measure the electrical resistivity, Hall mobility, carrier concentration, and carrier type of the films. The W probe electrodes were used for the measurements.

3. Results

3.1 Microstructure

Figure 1 shows the XRD patterns of the obtained MnTe films, indicating that all films had an α-MnTe single phase. Based on the obtained XRD patterns, the grain size D was estimated using Scherrer’s formula³³⁾ as follows:   

\begin{equation} D = 0.94\lambda/\beta\cos\theta \end{equation} (1)

where β is the full width at half maximum and θ is the Bragg angle of the peak. Using β and θ of 101_α, D was estimated to be 63, 60, and 59 nm for 50.4Te, 49.2Te, and 47.9Te films, respectively. All the MnTe films, thus, had almost the same grain size. The lattice parameters of the 50.4Te, 49.2Te, and 47.9Te films were a = 4.124 nm and c = 6.691 nm, a = 4.123 nm and c = 6.688 nm, as well as a = 4.126 nm and c = 6.685 nm, respectively. Three films showed similar values in lattice parameters, because of the small difference of film composition. The observed lattice parameters of the present films are close to reported values of stoichiometric α-MnTe (powder) (a = 4.126 and c = 6.724 nm³⁴⁾ as well as a = 4.13 nm and c = 6.71 nm³⁵⁾). This suggests that less residual stress occurred in the film.

Fig. 1

XRD patterns of the MnTe films.

Furthermore, we directly observed the microstructures of the α-MnTe films using TEM. The thickness of the MnTe film for the TEM observation was 150 nm. Figure 2(a) shows the cross-sectional bright field TEM image of the 49.2Te film. The grain size was determined to be of ∼140 nm using a simple near-intercept method, which is close to the value estimated from the XRD results. Figure 2(b) shows a selected area diffraction pattern of the 49.2Te film, indicating that the 49.2Te film had the NiAs-type structure of α-MnTe.

Fig. 2

(a) Cross-sectional bright-field TEM image and (b) selected area diffraction pattern obtained for the 49.2Te film, where the top layer is glued to prepare the TEM sample.

3.2 XPS analysis

We also performed XPS analysis to verify the quality of the obtained α-MnTe films. Figure 3 shows the XPS spectra of (a) Mn 2p and (b) Te 3d_5/2 energy range of the 50.4Te film. Figure 3(c) and (d) shows the XPS spectra of Mn 2p for pure Mn and Te 3d_5/2 for pure Te, respectively. The samples were sputtered using an Ar plasma before the measurement to remove native oxide and surface contamination. As shown in Fig. 3(c), the XPS spectrum of Mn 2p of pure Mn sample exhibited two peaks (2p_3/2 and 2p_1/2). The XPS spectrum of Mn 2p of the 50.4Te film also showed two peaks that slightly shifted to high binding energy side and become broad. The two broad peaks had a shoulder at high binding energy side (Fig. 3(a)). These shoulder peaks are satellites of main peaks of Mn–Te bonds.^27,36,37) The intensities of the deconvoluted peaks in Fig. 3(a) indicate that the measured Mn 2p peak was composed of strong peaks originated from Mn–Te bonds and additional weak peaks of MnO_x. The MnO_x would be MnO₂, determined from the binding energy position. The XPS spectrum of Te 3d_5/2 of the 50.4Te film (Fig. 3(b)) shows a strong peak which exhibited lower binding energy side than that of the pure Te (Fig. 3(d)), indicating the formation of Mn–Te bonds. The other two films of 49.2Te and 47.9Te showed almost the same XPS spectra with the 50.4Te film in Mn 2p and Te 3d_5/2 core-level region. These results indicate that no pure Mn- and Te-phase were formed inside the all films. As shown in Fig. 3(a), Mn oxide was detected by the XPS analysis. The TEM observation (Fig. 2) indicates that it was hardly oxidized inside the film. The surface was likely to be natively oxidized, where the surface region (about 11 nm in depth) showed a small grain-structure. Therefore, the MnO_x detected in XPS analysis is supposed to be mainly resulted from the surface oxidation which occurred during heat treatment. However, it is noted that such a preferential surface oxidation of Mn would decrease Mn content inside of the films. The preferential surface oxidation of Mn can be explained from the Gibbs free energy of oxidation for each element, $\Delta G_{\text{MnO}_{2}} = - 375.9$ kJ/mol³⁸⁾ and $\Delta G_{\text{TeO}_{2}} = - 180.9$ kJ/mol³⁸⁾ at 500°C (773 K), indicating that Mn was much easier to be oxidized than Te.

Fig. 3

(a) Mn 2p and (b) Te 3d_5/2 core-level spectra of the 50.4Te film, where “sat” in (a) indicates satellite peaks of Mn–Te bonds. (c) Mn 2p and (d) Te 3d_5/2 core-level spectra of pure Mn and pure Te bulk samples, respectively.

3.3 Optical measurement

Figure 4(a) and (b) show the typical transmittance and reflectance spectra of the obtained films as a function of the wavelength, respectively. The optical measurements were carried out for three times for each film. All films showed similar transmittance that drastically increased in the wavelength range of 600–800 nm by increasing wavelength. The transmittance of the α-MnTe films was over 10% in low wavelength range (at around 600 nm), compared with the case of CdTe (at around 800 nm)^39,40) and CIGS (at around 700 nm).^41,42) The reflectance curves also showed similar behaviors in all films, although the positions of the maxima and minima reflectance in the wavelength range of 400–1000 nm for the 50.4Te film shifted to low wavelength compared with those for 49.2Te and 47.9Te films. The reflectance of the α-MnTe films (10%∼50%) was higher than that of CdTe (10%∼30%)⁴³⁾ and CIGS (5%∼20%)⁴⁴⁾ in the wavelength range of 400–1100 nm.

Fig. 4

(a) Transmittance and (b) reflectance curves in α-MnTe films.

Generally, the absorption coefficient α was calculated from the following equation:⁴⁵⁾   

\begin{equation} \alpha = \ln[[(1 - R)^{2} + \{(1 - R)^{4} + 4R^{2}T^{2}\}^{1/2}]/2T]/d, \end{equation} (2)

where d is film thickness, T is transmittance and R is reflectance. Figure 5(a) shows the typical absorption coefficient α as a function of wavelength for α-MnTe films. The α in the wavelength range of 400–800 nm was in the order of 10⁵–10⁶ cm⁻¹, which are slightly higher than that of CdTe⁶⁾ and CIGS.⁷⁾ The band gap E_g can be estimated from the α vs. wavelength curve using the following equation:⁴⁶⁾   

\begin{equation} (\alpha h\nu)^{1/n} = A(h\nu - E_{\text{g}}), \end{equation} (3)

where h is Planck’s constant, v is the frequency, and A is a proportional constant. In this equation, n equaled to 0.5 and 2 for a direct transition and an indirect transition, respectively. Figures 5(b) shows the typical (αhν)^1/2 vs hν curve for α-MnTe films, which is known as the Tauc plot for an indirect transition. The E_g of the 50.4Te, 49.2Te, and 47.9Te films were estimated to be 1.37 ± 0.04, 1.48 ± 0.03 and 1.52 ± 0.02 eV, respectively. The E_g of the α-MnTe films agrees well with the reported value of a MnTe (1.27 ± 0.01∼1.46 ± 0.10 eV).^21–23,31)

Fig. 5

(a) Absorption coefficient in α-MnTe films and (b) Tauc plot under assumption of an indirect bandgap.

As mentioned in section 3.2, we confirmed that a Mn oxide layer existed on the surface of films, implying that the obtained optical data such as absorption coefficient and band gap was affected by the surface oxide layer. To ensure such influence, the surface oxide layer on the 50.4Te film was removed by reverse sputtering to a depth of 30 nm, and the optical properties were then measured. As a result, we confirmed that α and E_g of the film without the surface oxide layer were almost the same with those of the as-annealed film (α = 0.21 × 10⁵∼0.42 × 10⁶ cm⁻¹ (400–800 nm) and E_g = 1.37 eV for the as-annealed 50.4Te film; α = 0.24 × 10⁵∼0.49 × 10⁶ cm⁻¹ (400–800 nm) and E_g = 1.39 eV for the reverse-sputtered 50.4Te film). Thus, the influence of the surface oxide layer on the optical properties of the obtained film was negligible.

3.4 Temperature dependence of resistance

The E_g of the films was also evaluated from the temperature dependence of the resistance, which for semiconductors can be described using the following equation:   

\begin{equation} R = R_{0}\exp (E_{\text{g}}/2k_{\text{B}}T), \end{equation} (4)

where R is the electric resistance, R₀ is a constant, k_B is Boltzmann’s constant, and T is the temperature. This equation describes the conduction caused by thermal excitation of carriers in semiconductors. The slope of the ln R vs. 1000/T plot corresponds to the half value of E_g.²³⁾ Figure 6 shows the Arrhenius plot of the resistance, R, of the films. As mentioned in section 3.2, the obtained films had a surface oxide layer having thickness of about 11 nm. However, because the Mn oxide layer was thin, the measurement W probes can pierce the thin oxide layer and contact the MnTe film directly. Therefore, the influence of the surface oxide layer on the electrical properties was negligible. The relationship between ln R and 1/T was almost linear for the three films. The value of R increased with Te content. From the data extracted from Fig. 6, the E_g values of the 50.4Te, 49.2Te, and 47.9Te films were estimated as 1.26, 1.36, and 1.33 eV, respectively. These values nearly equal to those obtained from the optical measurements. Determined from the results of optical and electrical measurements, the E_g of the α-MnTe films in the composition range of 47.9 ≤ Te ≤ 50.4 (at%) was less sensitive to film composition.

Fig. 6

ln R vs. 1000/T plots in α-MnTe films.

3.5 Hall effect measurements

We investigated the electric properties of the obtained MnTe films by performing Hall effect measurements at room temperature. Because the Mn oxide layer on the surface of the films was thin, the measurement W probes can pierce the thin oxide layer and contact the MnTe film directly. Figure 7 shows (a) resistivity, ρ, (b) carrier density, n, and (c) hole mobility, μ, of the obtained films as functions of Te content. The α-MnTe films were found to exhibit p-type conduction. As shown from the resistivity plot (Fig. 7(a)), the resistivity increased with Te content, which agrees well with the results obtained from the resistance measurements using the two-point probe method. The resistivity value (∼0.32 Ω·cm) of the 50.4Te film, which is the closest to a stoichiometric MnTe composition, almost corresponds to the value (∼0.4 Ω·cm) for a stoichiometric MnTe film having thickness of 280 nm report by L. Yang et al.⁴⁷⁾

Fig. 7

(a) Resistivity, (b) carrier density and (c) mobility as a function of Te content in α-MnTe films.

Figures 7(b) and (c) show that the carrier density increased with Te content, whereas the carrier mobility decreased by increasing Te content. The decrease in the resistivity with increase in Mn content was caused by the increase in carrier mobility. The obtained carrier density of the α-MnTe films (8.3 × 10¹⁸∼1.2 × 10¹⁹ cm⁻³) is higher than that of α-MnTe bulk samples (3∼7 × 10¹⁸ cm⁻³).^28,48) The film thickness dependence of the carrier density of α-MnTe contributed to this behavior, i.e., the carrier density of α-MnTe film was increased by decreasing the film thickness.⁴⁹⁾ The α-MnTe films showed higher carrier density than CdTe and CIGS, which should cause high reflectance, as mentioned in section 3.3.

It is known that CdTe shows p-type conduction in Te-rich side mainly due to Cd vacancies and shows n-type conduction in Cd-rich side mainly due to Cd interstitial.⁵⁰⁾ In other words, the conduction type of CdTe changes from p to n across the stoichiometric composition. MnTe crystals normally exhibit p-type conduction because of the presence of an excess of Te, leading to Mn vacancies that act as acceptors.²¹⁾ In this study, we observed that α-MnTe showed p-type conductivity in the composition range of 47.9 ≤ Te ≤ 50.4 (at%) across the stoichiometric composition. Considering the XPS results, the surface of the α-MnTe films was oxidized and decreased the Mn content inside the films. Therefore, it is supposed that there were still Mn vacancies in the NiAs-type structure that acted as acceptors even in the 47.9Te and 49.2Te. This speculation is also implied by the fact that the carrier density decreased by increasing Mn content (Fig. 7(b)).

A summary of the properties of the α-MnTe films is included in Table 1, along with those of CdTe and CIGS.^{6–8,11,15,51)} The E_g values of α-MnTe were obtained by optical measurements. The absorption coefficient and E_g of the α-MnTe films were similar to those of CdTe and CIGS. However, the carrier mobility of α-MnTe was one order of magnitude lower than that of CdTe and CIGS. This difference was most possibly related to the high carrier density of α-MnTe. To realize the application of this α-MnTe film as an absorption layer in solar cells, its carrier mobility should be enhanced by further increasing Mn content or by performing a grain coarsening to reduce the scattering effect from grain boundaries.⁵²⁾

Table 1 Absorption coefficient, α, Band gap, E_g, resistivity, ρ, carrier density, n, carrier mobility, μ, and conduction-type in MnTe, CdTe and CIGS.

4. Conclusions

In this study, α-MnTe films were deposited by RF magnetron sputtering and subjected to optical and electrical measurements. The observed indirect band gap, E_g, and the absorption coefficient of the α-MnTe films were around 1.37–1.52 eV and 0.2 × 10⁵–0.8 × 10⁶ cm⁻¹ from optical measurements in the wavelength range of 400–800 nm, respectively. All α-MnTe films showed p-type conductivity. The electrical resistivity, ρ, of the α-MnTe films decreased from 0.32 (50.4 at% Te) to 0.14 Ω·cm (47.9 at% Te) by decreasing Te content. The carrier density, n, and the carrier mobility, μ, decreased and increased, respectively, as the Mn content increased. For the α-MnTe film having a Te content of 47.9 at%, the following values were obtained: ρ = 0.14 Ω·cm, n = 8.3 × 10¹⁸ cm⁻³, and μ = 5.2 cm² V⁻¹ s⁻¹.

In this study, it is found that the resistivity of MnTe film qualitatively decreased by increasing Mn content because of the decrease in carrier density. The quantitative composition dependence of the resistivity of MnTe film, however, was still unclear because of the formation of the surface oxide layer. The present results suggest that preferential surface oxidation of Mn in α-MnTe should be carefully prevented to understand the quantitative composition dependence of electrical properties and the conductivity mechanism of α-MnTe, including an off-stoichiometric composition. It will be the subject of a future study.

REFERENCES

• 1)   S.  Kawanishi,  K.  Matsunaga,  T.  Yoshikawa and  K.  Morita: Mater. Trans. 59 (2017) 1571–1580.
• 2)   R.  Tsu,  H.  Shen and  M.  Dutta: Appl. Phys. Lett. 60 (1992) 112–114.
• 3)   S.  Rühle: Sol. Energy 130 (2016) 139–147.
• 4)   W.  Shockley and  H.J.  Queisser: J. Appl. Phys. 32 (1961) 510–519.
• 5)   M.A.  Green: Sol. Energy Mater. Sol. Cells 92 (2008) 1305–1310.
• 6)   K.  Mitchell,  A.L.  Fahrenbruch and  R.H.  Bube: J. Appl. Phys. 48 (1977) 829–830.
• 7)   S.  Levcenko,  G.  Gurieva,  E.J.  Friedrich,  J.  Trigo,  J.  Ramiro,  J.M.  Merino,  E.  Arushanov and  M.  León: Moldavian Journal of the Physical Sciences 9 (2010) 148–155.
• 8)   A.  Morales-Acevedo: Sol. Energy 80 (2006) 675–681.
• 9)  A. Luque and S. Hegedus: Handbook of Photovoltaics Science and Engineering (Wiley, Hoboken, NJ, U.S.A., 2011) pp. 14–25.
• 10)   B.  Mitchell,  H.  Wagner,  P.P.  Altermatt and  T.  Trupke: Energy Procedia 38 (2013) 147–152.
• 11)   M.  Takahashi,  K.  Uosaki,  H.  Kita and  S.  Yamaguchi: J. Appl. Phys. 60 (1986) 2046–2049.
• 12)   S.  Gedi,  V.R.M.  Reddy,  B.  Pejjaia,  C.  Parka,  C.-W.  Jeon and  T.R.R.  Kotte: Ceram. Int. 43 (2017) 3713–3719.
• 13)   D.  Kindole,  I.  Anyadiegwu,  Y.  Ando,  Y.  Noda,  H.  Nishiyama,  S.  Uehara,  T.  Nakajima,  O.P.  Solonenko,  A.V.  Smirnov and  A.A.  Golovin: Mater. Trans. 59 (2018) 462–468.
• 14)   J.H.  Werner,  J.  Mattheis and  U.  Rau: Thin Solid Films 480–481 (2005) 399–409.
• 15)   C.-H.  Huang: J. Phys. Chem. Solids 69 (2008) 330–334.
• 16)   A.  Morales-Acevedo: Energy Procedia 2 (2010) 169–176.
• 17)   W.  Hsu,  C.M.  Sutter-Fella,  M.  Hettick,  L.  Cheng,  S.  Chan,  Y.  Chen,  Y.  Zeng,  M.  Zheng,  H.-P.  Wang,  C.-C.  Chiang and  A.  Javey: Sci. Rep. 5 (2015) 16028.
• 18)   W.  Witte,  S.  Spiering and  D.  Hariskos: Vakuum in Forschung und Praxis 26 (2014) 23–27.
• 19)   G.  Znamarchi: J. Phys. Chem. Solids 28 (1967) 2123–2130.
• 20)   J.W.  Allen,  G.  Lucovsky and  J.C.  Mikkelsen, Jr.: Solid State Commun. 24 (1977) 367–370.
• 21)   Ch.  Ferrer-Roca,  A.  Sequra,  C.  Reig and  V.  Muñoz: Phys. Rev. B 61 (2000) 13679–13686.
• 22)   W.  Szuszkiewicz,  E.  Dynowsk and  B.  Witkowska: Phys. Rev. B 73 (2006) 104403.
• 23)   Y.  Ren,  Q.  Jiang,  J.  Yang,  Y.  Luo,  D.  Zhang,  Y.  Cheng and  Z.  Zhou: J. Materiomics 2 (2016) 172–178.
• 24)   A.  Tubtimtae,  K.  Arthayakul,  B.  Teekwang,  K.  Hongsith and  S.  Choopun: J. Colloid Interface Sci. 405 (2013) 78–84.
• 25)   M.  Podgorny and  J.  Oleszkiewicz: J. Phys. C 16 (1983) 2547–2557.
• 26)   Y.B.  Li,  Y.Q.  Zhang,  N.K.  Sun,  Q.  Zhang,  D.  Li,  J.  Li and  Z.  Zhang: Phys. Rev. B 72 (2005) 193308.
• 27)   R.J.  Iwanowski,  M.H.  Heinonen and  B.  Witkowska: J. Alloys Compd. 491 (2010) 13–17.
• 28)   D.  Kriegner,  K.  Výborný,  K.  Olejník,  H.  Reichlová,  V.  Novák,  X.  Marti,  J.  Gazquez,  V.  Saidl,  P.  Němec,  V.V.  Volobuev,  G.  Springholz,  V.  Holý and  T.  Jungwirth: Nat. Commun. 7 (2016) 11623.
• 29)   M.  Krause and  F.  Bechstedt: J. Supercond. Nov. Magn. 26 (2013) 1963–1972.
• 30)   K.  Akabli,  Y.  Magnin,  M.  Oko,  I.  Harada and  H.T.  Diep: Phys. Rev. B 84 (2011) 024428.
• 31)   C.  Reig,  V.  Muñoz,  C.  Gómez,  Ch.  Ferrer and  A.  Segura: J. Cryst. Growth 223 (2001) 349–356.
• 32)   M.E.  Schlesinger: J. Phase Equilibria 19 (1998) 591–596.
• 33)   B.J.  Babu,  S.  Velumani,  A.  Kassiba,  R.  Asomoza,  J.A.  Chavez-Carvayar and  J.  Yi: Mater. Chem. Phys. 162 (2015) 59–68.
• 34)   G.I.  Makovetskii,  A.I.  Galyas,  G.M.  Severin and  K.I.  Yanushkevich: Inorg. Mater. 32 (1996) 846–849.
• 35)   I.N.  Odin,  M.V.  Chukichev and  M.E.  Rubina: Inorg. Mater. 39 (2003) 345–348.
• 36)   R.J.  Iwanowski,  M.H.  Heinonen and  E.  Janik: Chem. Phys. Lett. 387 (2004) 110–115.
• 37)   W.  Kim,  I.J.  Park,  H.J.  Kim,  W.  Lee,  S.J.  Kim and  C.S.  Kim: IEEE Trans. Magn. 45 (2009) 2424–2427.
• 38)  E.A. Brandes and G.B. Brook: Smithells Metals Reference Book, Seventh Edition, (Butterwirth-Heinemann Ltd, Oxford, 1992) chapter 8, pp. 26–27.
• 39)   Y.O.  Choi,  N.H.  Kim,  J.S.  Park and  W.S.  Lee: Mater. Sci. Eng. B 171 (2010) 73–78.
• 40)   F.  de Moure-Flores,  J.G.  Quiñones-Galván,  A.  Guillén-Cervantes,  J.S.  Arias-Cerón,  A.  Hernández-Hernández,  J.  Santoyo-Salazar,  J.  Santos-Cruz,  S.A.  Mayén-Hernández,  M. de la L.  Olvera,  J.G.  Mendoza-Álvarez,  M.  Meléndez-Lira and  G.  Contreras-Puente: J. Cryst. Growth 386 (2014) 27–31.
• 41)   S.H.  Jung,  S.J.  Choi and  C.W.  Chung: Thin Solid Films 570 (2014) 189–193.
• 42)   S.H.  Moon,  S.J.  Park,  Y.J.  Hwang,  D.-K.  Lee,  Y.  Cho,  D.-W.  Kim and  B.K.  Min: Sci. Rep. 4 (2014) 4408.
• 43)   E.R.  Shaaban,  N.  Afify and  A.  El-Taher: J. Alloys Compd. 482 (2009) 400–404.
• 44)   S.Y.  Han,  C.  Pan,  D.H.  Kim and  C.h.  Chang: RSC Advances 5 (2015) 24712–24717.
• 45)   R.  Vahalová,  L.  Tichý,  M.  Vlček and  H.  Tichá: Phys. Status Solidi A 181 (2000) 199–209.
• 46)   J.  Tauc,  R.  Grigorovici and  A.  Vancu: Phys. Status Solidi 15 (1966) 627–637.
• 47)   L.  Yang,  Z.H.  Wang and  Z.D.  Zhang: J. Appl. Phys. 119 (2016) 045304.
• 48)   W.  Xie,  S.  Populoh,  K.  Gałązka,  X.  Xiao,  L.  Sagarna,  Y.  Liu,  M.  Trottmann,  J.  He and  A.  Weidenkaff: J. Appl. Phys. 115 (2014) 103707.
• 49)   A.  Goswami and  A.B.  Mandale: Jpn. J. Appl. Phys. 17 (1978) 473–478.
• 50)   M.A.  Berding: Phys. Rev. B 60 (1999) 8943–8950.
• 51)   A.  Belghachi and  N.  Limam: Chin. J. Phys. 55 (2017) 1127–1134.
• 52)   J.N.  Roy,  S.  Basu and  D.N.  Bose: J. Appl. Phys. 54 (1983) 847–852.