Optical, Electrical Properties and Structure of Multilayer Iron-Doped Indium–Tin Oxide Thin Films Sputtered on Preheated Glass Substrates

Abstract

Multilayer iron-doped indium-saving indium–tin oxide (ML ITO50:Fe₂O₃) thin films with high conductivity and high transmittance in the visible spectrum have been fabricated by sputtering method. Structures consisting of very thin layer of conventional indium tin oxide (90 mass% In₂O₃–10 mass% SnO₂) and iron-doped indium-saving indium–tin oxide layer with reduced content of In₂O₃ (ITO50:Fe₂O₃) to 50 mass% are discussed. By optimizing oxygen flow rate in iron-doped indium-saving indium–tin oxide layer, the lowest volume resistivity of 3.78 × 10⁻⁴ Ω·cm, mobility of 29.8 cm²/(V·s), carrier concentration of 4.60 × 10²⁰ cm⁻³ and transmittance larger than 90% in the visible range have been achieved. ML ITO50:Fe₂O₃ thin films deposited under optimal conditions demonstrated lower volume resistivity and higher transmittance than undoped multilayer indium-saving ITO thin films and iron-doped single-layer thin films obtained under the same oxygen flow rate Q(O₂) = 0.1 sccm. ML ITO50:Fe₂O₃ thin films demonstrated optimal parameters at the lower oxygen flow rate (Q(O₂) = 0.1 sccm) than undoped ML ITO50 thin films. ML ITO50:Fe₂O₃ thin films are crystallized and show In₄Sn₃O₁₂ structure.

Fig. 3 TEM images of as-deposited ML ITO50:Fe₂O₃ thin film: (a) plane view image from 2nd layer, (b) cross-sectional image and (c) plane view image closed to interface between 1st and 2nd layers. SAED patterns of 2nd layer (a) and closed to interface between 1st and 2nd layers (c).

1. Introduction

Indium–tin oxide thin films are widely applied in various technologies, including solar cells, liquid crystal displays, optical solar reflectors, etc.^1–3) Optical and electrical properties of the used materials determine the efficiency of these devices. Indium–tin oxide (ITO) thin films have low resistivity and high transmittance in the visible wavelength region. However, natural indium reserves decrease and its cost is high. Therefore, it is necessary to find more cost-effective material than conventional ITO (90 mass% In₂O₃–10 mass% SnO₂, ITO90). One of the ways to achieve this aim is to sputter indium-saving ITO thin films. ITO thin films with reduced amount of indium oxide or no indium oxide in composition were manufactured in Refs. 4–26). Multilayer thin films with no indium were investigated in Refs. 27–30). Multilayer Al-doped ZnO (AZO/Cu/AZO) thin films showed resistivity of 3.85 × 10⁻⁵ Ω·cm and luminous transmittance of above 82% in the visible range.²⁷⁾ Magnetron sputtered Zn/AZO thin films demonstrated resistivity of 3.36 × 10⁻⁴ Ω·cm and the average transmittance of about 79.1–85.2%.²⁸⁾ FTO/Ag/FTO multilayer thin films showed the average optical transmittance of 95.5% in the visible range of wavelengths and the resistivity of 8.8 × 10⁻⁵ Ω·cm.³⁰⁾ Minami et al.⁸⁾ noticed that F-doped SnO₂ (FTO) and Al-doped ZnO (AZO) films consisting of binary compounds are often limited in their application because the attainable properties are dependent on the intrinsic properties of the material used. For the purpose of optimizing electrical and optical properties for specialized applications, multicomponent oxides systems composed of binary (ZnO–SnO₂ and ZnO–In₂O₃) and ternary (Zn₂SnO₄, ZnSnO₃, Zn₂In₂O₅,) compounds have recently attracted much attention as new materials for TCO films.^31–33) The highly transparent and conductive a ternary compound In₄Sn₃O₁₂ in the In₂O₃–SnO₂ system was prepared by RF magnetron sputtering.¹⁰⁾ Undoped In₄Sn₃O₁₂ and M-doped In_4.5Sn₂M_0.5O₁₂ (M = Nb and Ta),³⁴⁾ In_4+xSn_3−2xSb_xO₁₂,³⁵⁾ In₄Ge₃O₁₂³⁶⁾ compounds are promising as an alternative material for ITO films because of its lower cost resulting from a lower In content. Moreover ITO films in comparison with FTO and AZO films possess better conductivity, are much smoother and more resistant to air and in acid or alkaline environments.^37,38)

This paper introduces the preparation of indium-saving ITO multilayer thin films by DC/RF magnetron sputtering. Recently, our group has investigated the structural, electrical and optical properties of the multilayer (ML) indium-saving ITO thin films (50 mass% In₂O₃–50 mass% SnO₂, ITO50)³⁹⁾ as an alternative to conventional ITO90. It allowed achieving low resistivity and high transmittance in visible range. Within this research, a very thin crystalline layer of ITO90 was deposited as the first layer onto glass substrates preheated at 523 K (PHS523). ITO50 thin films doped with Fe, as Fe₂O₃, (ITO50:Fe₂O₃) were deposited by co-sputtering method as the second layer onto the substrate preheated at 523 K. In order to reduce indium usage in second layer, an amount of indium oxide in target was decreased from 90 mass% to 50 mass%. By using the crystalline ITO90 film as a buffer layer for the Fe-doped ITO50 film, the Fe-doped ITO50 film was expected to crystallize and reduce the resistivity. Fe doping to ITO films is quite interesting, and not only expected to act as a dopant, but is also reported as a transparent conductive film with magnetic properties.^40–42)

2. Experiment

Deposition of both ITO90 and ITO50:Fe₂O₃ thin films were performed in a commercial sputtering system ULVAC, CS-200. The very thin layer of ITO90 (≈ 12–14 nm) was sputtered using (90 mass% In₂O₃–10 mass% SnO₂) target onto the glass substrates (Corning EAGLE 2000, surface: 50 mm × 50 mm, thickness: 0.7 mm). The second ITO50:Fe₂O₃ layer was obtained by co-sputtering of ITO50 (Mitsui Mining & Smelting, In₂O₃–48.9 mass% SnO₂) and Fe₂O₃ (Kojundo Chemical Laboratory, 99.9 mass%) targets. DC plasma power for sputtering of ITO90 and ITO50 targets was kept at 100 W. The RF plasma power for sputtering of Fe₂O₃ (W_RF(Fe₂O₃)) target was changed from 0 to 40 W. In this investigation ML ITO50:Fe₂O₃ thin films were deposited onto glass substrates preheated at 523 K (PHS523). Substrate holder rotated with the speed of 40 rpm in order to obtain a homogeneous deposition. The argon gas flow rate was kept constant at Q(Ar) = 50 sccm, oxygen flow rate Q(O₂) for deposition of ITO90 layer was set at 0.2 sccm and for sputtering of ITO50:Fe₂O₃ layer it varied from 0 to 0.4 sccm. Sputtering time for multilayer thin films was set at 3 min for ITO90 as the first layer in order to obtain 12 nm thickness. Such thickness was chosen since ITO90 thin films of 12 nm thickness demonstrated the best optical and electrical properties.³⁹⁾ Sputtering time for the second ITO50:Fe₂O₃ layer with 138 nm thickness was 28–30 min. Total thickness of multilayer films was 150 nm. Single layer (SL) ITO50:Fe₂O₃ (W_RF(Fe₂O₃) = 20 W, Q(Ar)/Q(O₂) = 50 sccm/0.2 sccm), and ITO90 (Q(Ar)/Q(O₂) = 50 sccm/0.2 sccm) thin films were sputtered within 32 and 37 min respectively to get the same thickness as that of multilayer film. The deposited films were subject to heat treatment in air at 523 and 623 K (HT523 and HT623) within 60 min and cooled down at room temperature. The composition of the film in the second ITO50:Fe₂O₃ layer was determined by EDX (energy-dispersive X-ray) analysis in different places of the sample. Table 1 shows the chemical analysis of the film averaged over seven measurements.

Table 1 Elements’ and oxide compositions of as-deposited ML ITO50:Fe₂O₃ thin film sputtered at Q(Ar)/Q(O₂) = 50 sccm/0.1 sccm and W_RF(Fe₂O₃) = 20 W.

The measurements of volume resistivity ρ_v were carried out with a resistivity meter (Mitsubishi chemical analytech, Loresta GP Model MCP-T610) by using a 4-terminal method in accordance with Japanese Industrial Standards JIS K 7194-1994. Optical transmittance τ was measured in the 200∼2600 nm wavelength range of using the Spectrophotometer (Hitachi High-Tech, U-4100). X-ray diffraction (XRD) film profiles were recorded using the Rigaku Rint-2000 diffractometer with Cu-Kα (0.15418 nm) radiation. Surface analysis was carried out during deposition of ML ITO50:Fe₂O₃ thin films in comparison to SL ITO50:Fe₂O₃ and conventional SL ITO90 thin films by using a scanning probe microscope (SPM, SII, L-trace II) under the dynamic force mode (DFM). Microstructural observations of ML thin films were carried out using a transmission electron microscope (TEM, Hitachi High Technology, H-9000NAR and Titan 80-300 with Image Corrector, FEI) operated at 300 kV.

3. Results and Discussion

3.1 Structure of ML ITO50:Fe₂O₃ thin film deposited under optimal conditions

Figure 1 shows XRD profile of as-deposited and heat-treated ML ITO50:Fe₂O₃ thin films sputtered under optimal conditions. XRD spectrum of ML ITO50:Fe₂O₃ thin film was compared to the spectra of ML ITO50 thin film and SL ITO50:Fe₂O₃ thin film deposited under optimal conditions. As-deposited ML ITO50:Fe₂O₃ thin films shows polycrystalline structure similar to undoped ML ITO50 and in contrast to amorphous SL ITO50:Fe₂O₃ thin film. ITO90 layer promotes crystallization of the ITO50:Fe₂O₃ layer as it was observed for ML ITO50 thin film.³⁹⁾ Main peaks of XRD pattern of ML ITO50:Fe₂O₃ thin film could be referred to the rhombohedral lattice of indium–tin oxide In₄Sn₃O₁₂ (PDF card No. 01-088-0773).^17,43,44) As shown in works^8,45) the In₄Sn₃O₁₂ phase was reported to be observed in the composition range of 47.9–59.3 mol% SnO₂ or with Sn contents between 40 and 60 mol%. Iron oxide or any other secondary impurity phases were not revealed both: in the ML ITO50:Fe₂O₃ and in the SL ITO50:Fe₂O₃ XRD patterns,^17,18) it means that 2.4 ± 0.7 mol% Fe embedded in ML ITO50 thin film (in In₄Sn₃O₁₂ crystal) and formed ordered substitutional sites within the lattice and iron doping didn’t change the structure of film. The result reported by Xu and Li⁴⁶⁾ suggested that 5 mol% Fe-doping could not change the structure of ZnO thin films. Table 2 shows the performance XRD parameters of as-deposited and heat treated ML ITO50:Fe₂O₃ thin films. It is obvious that the full width at half maximum (FWHM) decreased for the second intense In₄Sn₃O₁₂ crystal peak (21-2) and crystalline size was increased from 32 to 38 nm when doped with Fe and heat treated in ML ITO50:Fe₂O₃ thin films at temperatures 523 and 623 K (HT523 and HT623), implying that the crystal quality was improved. It is shown in Table 2 that the diffraction angle shifts to larger angle direction, but it is lower than the reference angle value 35.338°⁴⁴⁾ for the strain free In₄Sn₃O₁₂ crystal. The interplanar distance of iron doped ML ITO50:Fe₂O₃ thin film is slightly lower than that of the undoped ML ITO50. It might be due to some defects that exist in the Fe-doped ML ITO50:Fe₂O₃ thin film which could cause strain and consequently affecting the normal growth of In₄Sn₃O₁₂ crystals. The difference of the above results might be associated with the valence state of Fe ions in ML ITO50:Fe₂O₃ film. Based on the previous XPS results of single-layer Fe-doped indium tin oxide ITO50 thin films¹⁷⁾ we speculate that Fe ions in ML ITO50:Fe₂O₃ thin films prepared in this study exist mainly in the form of Fe³⁺. The effective ionic radii of Fe⁺³ are 0.55 nm at low-spin state or 0.645 nm at high-spin state and the ionic radii of Sn⁺⁴ and In⁺³ are 0.069 and 0.080 nm, respectively.⁴⁷⁾ Since the radius of iron cations (Fe³⁺) is comparable to the radius of tin cations (Sn⁴⁺), then it can be assumed that Sn⁴⁺ will be replaced by Fe³⁺ in In₄Sn₃O₁₂ crystal according to the formula In_4+xSn_3−2xFe_xO₁₂.³⁴⁾ Due to less ionic radius of Fe⁺³ than that of Sn⁴⁺ it will lead to tensile strain in the ML ITO50:Fe₂O₃ thin film. Accordingly, a manifestation of this situation in XRD patterns is that the (21-2) peak shifts towards bigger angle direction, but the interplanar distance becomes smaller. In addition, further heat treatment of the ML ITO50:Fe₂O₃ thin films leads to an enlargement of crystallite size and a shift of diffraction peaks towards larger angles.

Fig. 1

XRD profiles of ML ITO50:Fe₂O₃ thin film deposited at W_RF(Fe₂O₃) = 20 W and Q(O₂) = 0.1 sccm, ML ITO50 thin film (Q(O₂) = 0.3 sccm),³⁹⁾ SL ITO50:Fe₂O₃ thin film (Q(O₂) = 0.2 sccm and W_RF(Fe₂O₃) = 20 W).¹⁷⁾

Table 2 The XRD parameters of ML ITO50:Fe₂O₃ and ML ITO50 thin films.

Figure 2 shows the surface analyses for the ML and SL as-deposited films obtained under optimal conditions. All studied films had the same thickness of 150 nm. Images of ML ITO50:Fe₂O₃ thin film deposited at Q(O₂) = 0.1 sccm and W_RF(Fe₂O₃) = 20 W (Fig. 2(a)) were compared with those of SL ITO50:Fe₂O₃ thin film sputtered at Q(O₂) = 0.2 sccm, W_RF(Fe₂O₃) = 20 W (Fig. 2(b)) and SL ITO90 thin film sputtered at Q(O₂) = 0.2 sccm (Fig. 2(c)). Arithmetical mean height (S_a) and root mean square height (S_q) of ML ITO50:Fe₂O₃ as-deposited thin film are somewhat larger than those of SL ITO50:Fe₂O₃ as-deposited thin film due to crystallization as it was observed for ML ITO50 thin film³¹⁾ (Fig. 2(d)), but much smaller than those of as-deposited SL ITO90 thin film (Table 3).

Fig. 2

Surface analysis results for (a) the as-deposited ML ITO50:Fe₂O₃ thin films sputtered at Q(O₂) = 0.1 sccm and W_RF(Fe₂O₃) = 20 W in comparison to (b) the as-deposited SL thin films sputtered under optimal conditions: ITO50:Fe₂O₃ thin film (Q(O₂) = 0.2 sccm, W_RF(Fe₂O₃) = 20 W)¹⁷⁾ and (c) as-deposited ITO90 thin film (Q(O₂) = 0.2 sccm)³⁹⁾ and (d) as-deposited ML ITO50 thin film (Q(O₂) = 0.3 sccm).³⁹⁾

Table 3 Arithmetical mean height (S_a) and root mean square height (S_q) of as-deposited ML ITO50:Fe₂O₃ and SL ITO50:Fe₂O₃¹⁷⁾ thin films compared to those of SL ITO90 and ML ITO50 thin films.³⁹⁾

TEM images and SAED patterns of ML ITO50:Fe₂O₃ as-deposited thin film are presented in Fig. 3. Figure 3 shows two layers in the multilayer film. Thickness of the first layer is about 14 nm, while the second layer thickness is about 132 nm. As it can be seen from Fig. 3, the SAED patterns show spots making up rings corresponding to polycrystalline structure of In₄Sn₃O₁₂ (PDF #01-088-0773) for second layer (plane surface observation depth area), and those of In₄Sn₃O₁₂ (PDF #01-088-0773) and In₂O₃ (PDF #06-0416) an area closed to the interface between first and second layers.

Fig. 3

TEM images of as-deposited ML ITO50:Fe₂O₃ thin film: (a) plane view image from 2nd layer, (b) cross-sectional image and (c) plane view image closed to interface between 1st and 2nd layers. SAED patterns of 2nd layer (a) and closed to interface between 1st and 2nd layers (c).

Figure 4(a) shows the cross-sectional TEM image of as-deposited ML ITO50:Fe₂O₃ thin film. Two layers are clearly seen in Fig. 4(a). The second layer clearly shows the vertical columnar growth. Figure 4(b) shows cross-sectional high-resolution transmission electronic microscope (HRTEM) image of as-deposited ML ITO50:Fe₂O₃ thin film. It was performed the detailed investigation of the both film layers in two areas – the square No. 1 (Fig. 4(b)) near the glass substrate and area of individual columns (the square No. 2, Fig. 4(b)) using the Fast Fourier Transformation (FFT). The both layers showed the digital diffraction spots from crystal planes with the different interplanar distances (Fig. 5). The interplanar distances (0.70 nm, 0.305 nm, 0.27 nm, 0.20 nm, 0.185 nm) from the square No. 1 of the first ITO90 layer can corresponds to the crystal hkl planes (011) 0.7155 nm, (311) 0.3051 nm, (321) 0.27040 nm, (431) 0.1984 nm, (422) 0.2066 nm, (521) 0.1848 nm for the bcc lattice of indium oxide In₂O₃ (PDF card No. 06-0416). The interplanar distances (0.44 nm, 0.35 nm, 0.27 nm) from the square No. 2 of the second ITO50:Fe₂O₃ layer can corresponds to hkl planes (002) 0.4429 nm, (021) 0.3718 nm, (300) 0.2731 nm for the rhombohedral lattice of indium–tin oxide In₄Sn₃O₁₂ (PDF card No. 01-088-0773).

Fig. 4

(a) Cross-sectional bright field TEM image and (b) cross-sectional HRTEM image of as-deposited ML ITO50:Fe₂O₃ thin film.

Fig. 5

(a) and (b) digital diffractograms computed by the Fast Fourier Transformation for the images surrounded by squares No. 1 and No. 2 in the Fig. 4(b), respectively.

3.2 Optical and electrical properties

Figure 6 demonstrates volume resistivity of as-deposited and heat-treated ML ITO50 (In₄Sn₃O₁₂):Fe₂O₃ thin films, depending on RF sputtering power of Fe₂O₃ target. Figure 6 shows that the increase of W_RF(Fe₂O₃) leads to the increase of the volume resistivity for both as-deposited and HT523 conditions at Q(O₂) = 0.2 sccm and HT523 film deposited at Q(O₂) = 0.1 sccm. However, the as-deposited ML ITO50:Fe₂O₃ thin films sputtered at Q(O₂) = 0.1 sccm and W_RF(Fe₂O₃) = 20 W shows minimal value of volume resistivity 5.06 × 10⁻⁴ Ω·cm.

Fig. 6

Volume resistivity of as-deposited and heat-treated ML ITO50:Fe₂O₃ thin films as a function of RF sputtering power of Fe₂O₃ target.

Figure 7 shows volume resistivity of the ML ITO50:Fe₂O₃ thin films compared to that of SL ITO50: Fe₂O₃ and ML ITO50 thin films depending on the oxygen flow rate. Volume resistivity of both as-deposited ML and SL thin films decreased with increasing of oxygen flow rate to 0.1 sccm and then increased. All data for volume resistivity of different films are collected in the Table 4, and the lowest volume resistivity 3.78 × 10⁻⁴ Ω·cm of as-deposited ML ITO50:Fe₂O₃ thin film was measured at Q(O₂) = 0.1 sccm, W_RF(Fe₂O₃) = 20 W (the optimal conditions) that is decreased to one-third compared to the volume resistivity 11.5 × 10⁻⁴ Ω·cm of as-deposited SL ITO50:Fe₂O₃ thin film sputtered at the optimal conditions (Q(O₂) = 0.2 sccm and W_RF(Fe₂O₃) = 20 W) and that is lower than the volume resistivity 6.87 × 10⁻⁴ Ω·cm of ML ITO50 deposited at the same oxygen flow rate Q(O₂) = 0.1 sccm (Table 4).³¹⁾ It is due to good crystallinity of the as-deposited ML ITO50:Fe₂O₃ thin film and carrier density value raised by 3.5 times compared to the as-deposited SL ITO50:Fe₂O₃ thin film (Table 5).

Fig. 7

Volume resistivity of as-deposited and heat-treated ML ITO50:Fe₂O₃ and SL ITO50:Fe₂O₃¹⁷⁾ thin films compared to as-deposited ML ITO50.³⁹⁾ Thin films were deposited at W_RF(Fe₂O₃) = 20 W.

Table 4 Volume resistivity of ML ITO50:Fe₂O₃ and SL ITO50:Fe₂O₃¹⁷⁾ thin films deposited at different sputtering power W_RF(Fe₂O₃), oxygen flow rate Q(O₂), and heat treatment temperatures compared to that of ML ITO50 thin films.³⁹⁾

Table 5 Volume resistivity, transmittance at 550 nm, mobility and carrier density of as-deposited ML ITO50:Fe₂O₃, SL ITO50:Fe₂O₃¹⁷⁾ and ML ITO50³⁹⁾ thin films obtained under optimum conditions.

Mobility of as-deposited ML ITO50:Fe₂O₃ thin film is lower than that of as-deposited SL ITO50:Fe₂O₃ thin film (Table 5), since the interface between the ITO90 and ITO50:Fe₂O₃ film may act as a defect to carrier movement.⁴⁸⁾ The lowest resistivity 4.89 × 10⁻⁴ Ω·cm for HT523 conditions was also obtained for ML ITO50:Fe₂O₃ thin film sputtered at Q(O₂) = 0.1 sccm and W_RF(Fe₂O₃) = 20 W. This is decreased by almost a half compared to HT523 SL ITO50:Fe₂O₃ thin films (9.39 × 10⁻⁴ Ω·cm) (Table 4).

Though, the volume resistivity of 3.78 × 10⁻⁴ Ω·cm for ML ITO50:Fe₂O₃ thin film is almost two times lower in comparison with the volume resistivity of 6.87 × 10⁻⁴ Ω·cm for ML ITO50 at the oxygen flow rate Q(O₂) = 0.1 sccm, a further increase in oxygen flow rate from 0.1 to 0.4 sccm causes an increase in volume resistivity from 378 to 20.2 × 10⁻⁴ Ω·cm for ML ITO50:Fe₂O₃ thin films at W_RF(Fe₂O₃) = 20 W, while in undoped ML ITO50 films the volume resistivity decreases from 6.87 × 10⁻⁴ to 2.75 × 10⁻⁴ Ω·cm (Table 4). The increase of the volume resistivity of ML ITO50:Fe₂O₃ thin films with increasing oxygen flow rate is obviously related to the presence of additional oxygen atoms in films which act as carrier traps.⁴⁹⁾

Figure 8 shows the optical transmittance spectra of ML ITO50:Fe₂O₃ deposited in optimal conditions compared to SL ITO50:Fe₂O₃ thin films and undoped ML ITO50. ML ITO50:Fe₂O₃ thin film demonstrates higher transmittance than SL ITO50:Fe₂O₃ thin film at wavelength under 610 nm (Fig. 8(a)), whereas SL ITO50:Fe₂O₃ thin film showed slightly higher transmittance than ML ITO50:Fe₂O₃ thin film above 610 nm. As it can be seen from Fig. 8(b), introduction of oxygen leads to noticeable increase of the transmittance of as-deposited ML ITO50:Fe₂O₃ thin films at wavelength longer than 1500 nm. Whereas, transmittance of as-deposited ML ITO50:Fe₂O₃ thin films changed insignificantly in the 500–1500 nm wavelength optical range. The Table 5 presents collected data of volume resistivity, transmittance at 550 nm, mobility and carrier density of as-deposited ML and SL ITO50:Fe₂O₃ thin films obtained in optimal conditions. The as-deposited ML ITO50:Fe₂O₃ and SL ITO50:Fe₂O₃ thin films showed τ = 97.7 and 91.9% at λ = 550 nm and Q(O₂) = 0.1 sccm, respectively (Fig. 8(a)). Our investigations showed that sputtering power W_RF(Fe₂O₃) does not significantly affect the transmittance of ML ITO50:Fe₂O₃ thin films deposited at Q(O₂) = 0.1 sccm. Thus, ML ITO50:Fe₂O₃ thin films demonstrated optimal conditions at lower oxygen flow rate Q(O₂) = 0.1 sccm than undoped ML ITO50 thin films, although undoped ML ITO50 thin films deposited in optimal conditions (at oxygen flow rate Q(O₂) = 0.3 sccm) showed somewhat better optical and electrical properties than ML ITO50:Fe₂O₃ thin films sputtered in optimal conditions (Q(O₂) = 0.1 sccm and W_RF(Fe₂O₃) = 20 W).

Fig. 8

Optical transmittance spectra of (a) as-deposited ML ITO50:Fe₂O₃ thin film sputtered at W_RF(Fe₂O₃) = 20 W and Q(O₂) = 0.1 sccm compared to ML ITO50 thin film sputtered at Q(O₂) = 0.1 and 0.3 sccm³⁹⁾ and SL ITO50:Fe₂O₃ thin film sputtered at W_RF(Fe₂O₃) = 20 W and Q(O₂) = 0.1 sccm,¹⁷⁾ (b) as-deposited ML ITO50:Fe₂O₃ thin film sputtered at W_RF(Fe₂O₃) = 20 W and Q(O₂) = 0–0.4 sccm.

In₄Sn₃O₁₂ films obtained in present work showed better electrical properties than films manufactured in Refs. 12, 34) and enhanced optical properties in comparison with Refs. 8, 12) (Table 6). In work¹²⁾ In₄Sn₃O₁₂ films prepared by pulsed laser deposition showed a maximum transmission of 85% of the incident radiation over the visible wavelengths.

Table 6 Volume resistivity and average transmittance in the visible range of In₄Sn₃O₁₂ films.

4. Conclusions

Multilayer (ML) indium-saving iron-doped ITO (ITO50:Fe₂O₃) thin films were deposited on glass substrates preheated at 523 K by sputtering method at different oxygen flow rates and subsequently heat treated at 523 and 623 K. An amount of indium oxide in the target was decreased from 90 to 50 mass%. Decrease of volume resistivity of as-deposited ML ITO50:Fe₂O₃ thin film (3.78 × 10⁻⁴ Ω·cm) compared to as-deposited SL ITO50:Fe₂O₃ film (10.1 × 10⁻⁴ Ω·cm) in optimal conditions was observed due to good crystallinity of the as-deposited ML ITO50:Fe₂O₃ thin film and raised value of carrier density – 3.5 times compared to the as-deposited SL ITO50:Fe₂O₃ thin film. A transmittance above 90% in the visible range was obtained for as-deposited ML ITO50:Fe₂O₃ thin film. The ML ITO50:Fe₂O₃ thin film shows polycrystalline In₄Sn₃O₁₂ structure. The XRD analysis suggests that the 2.4 ± 0.7 mol% Fe-doping and heat treatment can improve the crystalline quality of multilayer indium saving indium–tin oxide thin films. The cross-sectional TEM image of as-deposited ML ITO50:Fe₂O₃ thin film indicates two layers, the ITO50:Fe₂O₃ layer shows vertical columnar growth. The HRTEM investigations confirmed the XRD analysis results and showed a high crystallinity of as-deposited ML ITO50:Fe₂O₃ thin films. ML ITO50:Fe₂O₃ thin films demonstrated better optical and electrical properties than undoped ML ITO50 thin films and SL ITO50:Fe₂O₃ thin films obtained at the same oxygen flow rate 0.1 sccm, though undoped ML ITO50 thin films deposited at higher oxygen flow rates Q(O₂) = 0.3 sccm showed somewhat better optoelectronic properties than ML ITO50:Fe₂O₃ thin films. ML indium saving ITO thin films show an economical advantage over the use of conventional ITO90 thin films due to the lower material cost.

Acknowledgments

The present research was supported by New Energy and Industrial Technology Development Organization (NEDO), Japan. Part of the Fig. 7 was reprinted from Ref. 17) with permission of Elsevier.

Parts of the Figs. 1, 2, 7, 8 were reprinted from Ref. 39) with permission of Elsevier.

REFERENCES

• 1)   J.  Hotovy,  J.  Hüpkes,  W.  Böttler,  E.  Marins,  L.  Spiess,  T.  Kups,  V.  Smirnov,  I.  Hotovy and  J.  Kováč: Appl. Surf. Sci. 269 (2013) 81–87. doi:10.1016/j.apsusc.2012.10.180
• 2)   K.P.  Sibin,  N.  Swain,  P.  Chowdhury,  A.  Dey,  N.  Sridhara,  H.D.  Shashikala,  A.K.  Sharma and  H.C.  Barshilia: Sol. Energy Mater. Sol. Cells 145 (2016) 314–322. doi:10.1016/j.solmat.2015.10.035
• 3)   U.  Betz,  M.  Kharrazi Olsson,  J.  Marthy,  M.F.  Escolá and  F.  Atamny: Surf. Coat. Technol. 200 (2006) 5751–5759. doi:10.1016/j.surfcoat.2005.08.144
• 4)   P.K.  Biswas,  A.  De,  K.  Ortner and  S.  Korder: Mater. Lett. 58 (2004) 1540–1545. doi:10.1016/j.matlet.2003.10.023
• 5)   P.K.  Biswas,  A.  De,  N.C.  Pramanika,  P.K.  Chakraborty,  K.  Ortner,  V.  Hock and  S.  Korder: Mater. Lett. 57 (2003) 2326–2332. doi:10.1016/S0167-577X(02)01220-X
• 6)   M.  Thirumoorthi and  J.  Thomas Joseph Prakash: J. Asian Ceram. Soc. 4 (2016) 124–132. doi:10.1016/j.jascer.2016.01.001
• 7)   K.  Utsumi,  H.  Iigusa,  R.  Tokumaru,  P.K.  Song and  Y.  Shigesato: Thin Solid Films 445 (2003) 229–234. doi:10.1016/S0040-6090(03)01167-2
• 8)   T.  Minami,  Y.  Takeda,  S.  Takata and  T.  Kakumu: Thin Solid Films 308–309 (1997) 13–18. doi:10.1016/S0040-6090(97)00530-0
• 9)   T.  Minami,  T.  Miyata and  T.  Yamamoto: Surf. Coat. Technol. 108–109 (1998) 583–587. doi:10.1016/S0257-8972(98)00592-1
• 10)   T.  Minami,  T.  Kakumu,  K.  Shimokawa and  S.  Takata: Thin Solid Films 317 (1998) 318–321. doi:10.1016/S0040-6090(97)00547-6
• 11)   T.  Minami: J. Vac. Sci. Technol. A 17 (1999) 1765–1772. doi:10.1116/1.581888
• 12)   D.H.  O’Neil,  V.L.  Kuznetsov,  R.M.J.  Jacobs,  M.O.  Jones and  P.P.  Edwards: Mater. Chem. Phys. 123 (2010) 152–159. doi:10.1016/j.matchemphys.2010.03.075
• 13)   S.  Li,  X.  Qiao and  J.  Chen: Mater. Chem. Phys. 98 (2006) 144–147. doi:10.1016/j.matchemphys.2005.09.012
• 14)   M.  Epifani,  R.  Diaz,  J.  Arbiol,  P.  Siciliano and  J.R.  Morante: Chem. Mater. 18 (2006) 840–846. doi:10.1021/cm0522477
• 15)   L.  Voisin,  M.  Ohtsuka,  S.  Petrovska,  R.  Sergiienko and  T.  Nakamura: Optik 156 (2018) 728–737. doi:10.1016/j.ijleo.2017.12.021
• 16)   L.  Voisin,  M.  Ohtsuka and  T.  Nakamura: Mater. Trans. 51 (2010) 503–509. doi:10.2320/matertrans.MBW200907
• 17)   M.  Ohtsuka,  R.  Sergiienko,  S.  Petrovska,  B.  Ilkiv and  T.  Nakamura: Optik 179 (2019) 19–28. doi:10.1016/j.ijleo.2018.10.130
• 18)   M.  Ohtsuka,  R.  Sergiienko,  S.  Petrovska,  B.  Ilkiv and  T.  Nakamura: Metallofiz. Noveishie Tekhnol. 41 (2019) 941–952. doi:10.15407/mfint.41.07.0941
• 19)   L.T.C.  Tuyen,  S.-R.  Jian,  N.T.  Tien and  P.H.  Le: Materials 12 (2019) 1665. doi:10.3390/ma12101665
• 20)   B.L.  Zhu,  F.  Liu,  K.  Li,  K.  Lv,  J.  Wu,  Z.H.  Gan,  J.  Liu,  D.W.  Zeng and  C.S.  Xie: Ceram. Int. 43 (2017) 10288–10298. doi:10.1016/j.ceramint.2017.05.058
• 21)   R.  Hussin,  S.Z.S.  Anuar,  S.A.  Ibharim,  Z.  Kamdi,  A.R.  Ainuddin and  Z.  Harun: AIP Conf. Proc. 2339 (2021) 020198. doi:10.1063/5.0044520
• 22)   K.-H.  Seo,  D.-H.  Park,  J.-H.  Lee and  J.-J.  Kim: Solid State Ionics 177 (2006) 601–605. doi:10.1016/j.ssi.2005.12.020
• 23)   J.-A.  Jeong and  H.-K.  Kim: Sol. Energy Mater. Sol. Cells 93 (2009) 1801–1809. doi:10.1016/j.solmat.2009.06.014
• 24)   J.-W.  Lim,  S.-I.  Na and  H.-K.  Kim: Sol. Energy Mater. Sol. Cells 98 (2012) 409–416. doi:10.1016/j.solmat.2011.11.054
• 25)   H.C.  Ma,  D.  Putri,  M.  Putri,  J.H.  Cheon,  J.H.  Kim and  H.Y.  Lee: J. Korean Phys. Soc. 61 (2012) 1994–1999. doi:10.3938/jkps.61.1994
• 26)   M.  Kawamura,  T.  Kiba,  Y.  Abe and  K.H.  Kim: J. Phys.: Conf. Ser. 987 (2018) 012014. doi:10.1088/1742-6596/987/1/012014
• 27)   S.  Yu,  Y.  Liu,  H.  Zheng,  L.  Li and  Y.  Sun: Opt. Lett. 42 (2017) 3020–3023. doi:10.1364/OL.42.003020
• 28)   Y.  Liu,  S.  Zhu and  B.  Song: Results Phys. 13 (2019) 102286. doi:10.1016/j.rinp.2019.102286
• 29)   J.  Kang,  D.  Park,  D.  Lee,  M.  Kamiko,  S.-J.  Kim,  S.-K.  Lee and  J.-H.  Koh: J. Nanosci. Nanotechnol. 21 (2021) 1971–1977. doi:10.1166/jnn.2021.18913
• 30)   S.  Yu,  L.  Li,  X.  Lyu and  W.  Zhang: Sci. Rep. 6 (2016) 20399. doi:10.1038/srep20399
• 31)   T.  Minami,  H.  Sonohara,  S.  Takata and  H.  Sato: Jpn. J. Appl. Phys. 33 (1994) L1693–L1696. doi:10.1143/JJAP.33.L1693
• 32)   T.  Minami,  H.  Sonohara,  T.  Kakumu and  S.  Takata: Jpn. J. Appl. Phys. 34 (1995) L971–L974. doi:10.1143/JJAP.34.L971
• 33)   H.  Enoki,  T.  Nakayama and  J.  Echigoya: Phys. Status Solidi A 129 (1992) 181–191. doi:10.1002/pssa.2211290116
• 34)   A.  Jiamprasertboon,  A.  Waehayee,  N.  Chanlek,  N.  Yong,  S.  Suthirakun and  T.  Siritanon: J. Alloy. Compd. 783 (2019) 28–36. doi:10.1016/j.jallcom.2018.12.288
• 35)   J.  Choisnet,  L.  Bizo,  M.  Allix,  M.  Rosseinsky and  B.  Raveau: J. Solid State Chem. 180 (2007) 1002–1010. doi:10.1016/j.jssc.2006.12.031
• 36)   Y.G.  Zhang and  Y.X.  Wang: J. Electron. Mater. 40 (2011) 1501–1505. doi:10.1007/s11664-011-1642-9
• 37)   D.-J.  Kwak,  B.-H.  Moon,  D.-K.  Lee,  C.-S.  Park and  Y.-M.  Sung: J. Electr. Eng. Technol. 6 (2011) 684–687. doi:10.5370/JEET.2011.6.5.684
• 38)   H.  Agura,  A.  Suzuki,  T.  Matsushita,  T.  Aoki and  M.  Okuda: Thin Solid Films 445 (2003) 263–267. doi:10.1016/S0040-6090(03)01158-1
• 39)   L.  Voisin,  M.  Ohtsuka,  S.  Petrovska,  R.  Sergiienko and  T.  Nakamura: J. Alloy. Compd. 827 (2020) 154378. doi:10.1016/j.jallcom.2020.154378
• 40)   D.  Chakraborty,  K.C.  Kumar,  S.  Kaleemulla and  I.  Omkaram: AIP Conf. Proc. 2221 (2020) 110001. doi:10.1063/5.0003080
• 41)   D.  Chakraborty and  S.  Kaleemulla: J. Supercond. Nov. Magn. 32 (2019) 729–737. doi:10.1007/s10948-018-4745-1
• 42)   F.F.H.  Aragón,  J.A.H.  Coaquira,  S.W.  da Silva,  R.  Cohen,  D.G.  Pacheco-Salazar and  L.C.C.M.  Nagamine: J. Alloy. Compd. 867 (2021) 158866. doi:10.1016/j.jallcom.2021.158866
• 43)   W.J.  Heward and  D.J.  Swenson: J. Mater. Sci. 42 (2007) 7135–7140. doi:10.1007/s10853-007-1569-y
• 44)   N.  Nadaud,  M.  Nanot,  J.  Jove and  T.  Roisnel: Key Eng. Mater. 132 (1997) 1373–1376. doi:10.4028/www.scientific.net/KEM.132-136.1373
• 45)   H.  Enoki,  J.  Echigoya and  H.  Suto: J. Mater. Sci. 26 (1991) 4110–4115. doi:10.1007/BF02402954
• 46)   L.  Xu and  X.  Li: J. Cryst. Growth 312 (2010) 851–855. doi:10.1016/j.jcrysgro.2009.12.062
• 47)   R.D.  Shannon: Acta Crystallogr. Sec. A 32 (1976) 751–767. doi:10.1107/S0567739476001551
• 48)   Y.S.  Kim,  J.H.  Park,  D.H.  Choi,  H.S.  Jang,  J.H.  Lee,  H.J.  Park,  J.I.  Choi,  D.H.  Ju,  J.Y.  Lee and  D.  Kim: Appl. Surf. Sci. 254 (2007) 1524–1527. doi:10.1016/j.apsusc.2007.07.080
• 49)   K.H.  Kim: J. Ceram. Process. Res. 8 (2007) 19–21.