Carrier Transport Mechanism of Pt Contacts to Atomic Layer Deposited ZnO on Glass Substrates

Abstract

We grew ZnO films at different temperatures on glass substrates using thermal atomic layer deposition and investigated the current conduction mechanism of Pt/ZnO junctions. For ZnO samples grown at 46 and 96°C, the current flow through the ZnO layer was not possible at low temperatures such as 298 and 320 K because the ZnO layer under Schottky contacts were wholly depleted. However, the current conduction was observed with increasing the temperature, which was found to be dominated by the Schottky emission at 360 and 380 K and the Poole–Frenkel emission at 380 and 400 K. For ZnO sample grown at 141°C, the strong tunneling current could occur because very thin depletion region was formed because of the high carrier concentration of ZnO layer. The observed difference in the current conduction can provide a guidance how growth temperature of ZnO should be controlled to design the low-temperature grown ZnO based devices built on glass substrates.

Fig. 1 Current density–voltage (J–V) curves measured at 298 K. The inset in (a) shows the schematic diagram of prepared Pt/ZnO/Al devices.

1. Introduction

ZnO is an important wide bandgap semiconductor due to its applications for various devices such as light emitting devices, transparent electrodes, surface acoustic wave (SAW) devices, solar cells, UV photodetectors, chemical detectors, and thermoelectric devices.^1–7) The electrical, optical and structural properties of ZnO are dependent on the growth methods and thus, determining the proper growth method for each application is pivotal to optimize the device performance. The growth of ZnO films is possible by using various growth methods.^8–11) Among these methods, atomic layer deposition (ALD) can produce low-temperature grown ZnO films with its unique growth mechanism of self-limiting and saturated surface reactions that result in high conformity and precise thickness control.¹²⁾

The ALD growth of ZnO films has been done using various substrates such as Si, GaAs, sapphire, and GaN.^12,13) The reported growth window for ZnO varies between 100 and 400°C.^12–15) Makino et al. investigated the effect of Zn-precursors using dimethylzinc (DMZn) and diethylzinc (DEZn) on ZnO films grown on glass by varying the growth temperature between 90∼270°C.¹⁵⁾ They reported that the ZnO films grown using DMZn showed higher electrical resistivity than those grown using DEZn. Chang and Tsai grew ZnO in the temperature range of 90 and 150°C to investigate its effect as an electron collection layer (ECL) in polymer solar cells.¹⁶⁾ Using a 60-nm thick ZnO film deposited at 90°C as the ECL, they obtained power conversion efficiency of 4.1%, low series resistance, and high shunt resistance. Rowlette et al. grew ZnO films by plasma-enhanced ALD at 25∼120°C and reported that the electrical resistivity decreased from 270 Ωcm to 5 Ωcm with the temperature, which are suitable for thin film transistor (TFT) performance.¹⁷⁾ Mauro et al. reported the ZnO films by ALD, grown at temperature down to 40°C, which could be applied to synthesize flexible materials for photocatalysis.¹⁸⁾ Choi et al. studied the encapsulation effect on polymer substrate with ZnO/Al₂O₃ bilayer grown at 60∼250°C.¹⁹⁾ They found that with the increase of ZnO layer from 0 to 60 nm, the physical properties of laminated structures such as film crystallinity, surface roughness, density, and transmittance could be manipulated systematically. However, these works were focused on the characterization of material properties and studies on the ZnO films grown below 100°C, in terms of carrier transport mechanism, are still limited.

Good metal/semiconductor (MS) contacts, consisting of high-barrier Schottky contact and low-contact-resistance ohmic contact, are crucial for realizing high performance devices. Especially, Schottky contacts are frequently used to investigate the electrical properties of ZnO layers.^20–22) For example, Kocyigit et al. investigated Au/ZnO Schottky diodes formed on n-Si substrate in a temperature range from 100 to 380 K and found that the density of interface states increased with the temperature.²²⁾ To understand the properties of ZnO layer itself, it would be necessary to isolate electrically ZnO layer from the substrate. In this regard, we grew ZnO on glass substrate in the temperature range of 46∼141°C and investigated the electrical properties of Pt/ZnO Schottky diodes.

2. Experimental

We prepared soda-lime glass slides (CORNING 2947) with 100% transmittance as starting materials. After ultrasonically cleaning the glass substrates in isopropyl alcohol for 5 min and washing in deionized water for 5 min, they were loaded into a thermal ALD reactor to deposit ZnO films by using diethylzinc (DEZn: Zn(C₂H₅)₂) and H₂O as the Zn and O precursors, respectively. The duration of DEZn feeding, N₂ purge, H₂O feeding, and N₂ purge was 0.2, 15, 0.1, and 15 s, respectively. ZnO films were deposited at substrate temperatures of 46∼141°C. The thicknesses of the ZnO films were calculated from the analysis on the values acquired by using multi-wavelength spectroscopic ellipsometry (FS-1, Film Sense, USA). The electrical parameters of ZnO films were acquired at room temperature by using the Hall-effect measurements (HMS-3000, Ecopia, Korea). Ohmic contacts were made using soldered indium dots placed at four corners of square-shaped samples in a van der Pauw configuration. To electrically characterize by fabricating Pt/ZnO/Al devices, 50-nm-thick circular Pt Schottky (diameter: 300 µm) and large-area 100-nm-thick Al contacts were deposited on the ZnO surfaces grown at 46, 96 and 141°C (denoted as T46, T96 and T141, respectively) by using radio-frequency magnetron sputtering. A schematic diagram of a Pt/ZnO/Al Schottky diode is depicted in the inset of Fig. 1. The electrical characterization took place by using a Keithley 238 current source.

Fig. 1

Current density–voltage (J–V) curves measured at 298 K. The inset in (a) shows the schematic diagram of prepared Pt/ZnO/Al devices.

3. Results and Discussion

Table 1 shows the detailed information on the ZnO films. The prepared samples showed negative Hall coefficients, revealing n-type conductivity. Based on the theoretical calculations, it was suggested that ZnO cannot be p-type doped by acceptor-like defects such as oxygen interstitials (O_I) and Zn vacancies (V_Zn) because donor-like defects such as oxygen vacancies (V_O) and Zn interstitials (Zn_I), with low formation enthalpies, could easily compensate p-type doping at both Zn- and O-rich conditions.²³⁾ Hence, the prepared samples would be unintentionally n-type doped. With increasing the growth temperature up to 119°C, the carrier concentration increased but the mobility decreased. Such opposite tendency can be explained by the impurity scattering. At higher growth temperature of 141°C, however, the mobility increased even with the increased carrier concentration. At lower growth temperatures, more residual OH groups can remain which restrain the formation of oxygen vacancies acting as donors in undoped ZnO.²⁴⁾ These OH groups also act as traps and scattering centers, reducing the mobility.²⁵⁾ Jeon et al. reported that the higher mobility for ZnO samples grown at high temperatures (>110°C) were associated with the polycrystal structures with both c- and a-axis orientations and high oxygen deficiencies owing to fewer OH bonds in ZnO films.²⁶⁾

Table 1 Thickness, carrier concentration, and mobility of ZnO samples.

For about 25-nm-thick ZnO film grown at 96°C, the carrier concentration of ∼3.5 × 10¹⁴ cm⁻³ and the mobility of ∼25 cm²/Vs were obtained. Approximately 45-nm-thick ZnO film grown at 96°C showed the carrier concentration of ∼3.0 × 10¹⁴ cm⁻³ and the mobility of ∼20 cm²/Vs. That is, the carrier concentration and the mobility were similar for 25- and 45-nm thick ZnO films. However, the carrier concentration increased and the mobility decreased for thicker ZnO films. We also observed that the carrier concentration and the mobility for 25-nm-thick ZnO grown at 141°C were to be ∼1 × 10¹⁷ cm⁻³ and ∼31 cm²/Vs, revealing the increased carrier concentration and the decreased mobility for thicker ZnO films. Singh et al. reported an increase in both the carrier concentration and defect concentration (such as oxygen vacancy) with increasing the ZnO thickness.²⁷⁾ The increased mobility for thicker ZnO films was associated with the decreased grain boundary concentration (due to bigger grain size).²⁸⁾ Electron traps are mainly present near the grain boundaries and surface. The density of these traps would decrease for thicker ZnO films, increasing the carrier concentration. When the grain boundary scattering is dominant in determining the mobility, the increased mobility may be found for thicker ZnO.^28,29) Thus, the decreased mobility for thicker ZnO films indicates that not grain boundary scattering but ionized impurity scattering is dominant. However, further investigation requires to understand the exact mechanism.

Figure 1 shows the current density–voltage (J–V) data at 298 K. Both the T46 and T96 samples show very low and almost constant current values. Since the carrier concentrations for these samples are low (below 10¹⁶ cm⁻³), the whole ZnO layer under Pt contact would be depleted. Hence, electrons could hardly travel through this region. In contrast, the current values for T141 sample are very high, revealing an ohmic-like behavior. The carrier concentration for T141 sample is as high as 1.4 × 10¹⁹ cm⁻³, and the depletion width would be very narrow. Thus, electrons can tunnel through the thin depletion region easily, increasing the current values.

Figure 2 shows the measured current density–voltage (J–V) curves at different temperatures for T46 and T96 samples. Like room temperature, the current values for 320 K are very low and bias-independent. However, the current values start to increase above 340 K. To understand the current conduction across the Pt/ZnO interface, we analyzed the reverse bias current values using the Schottky emission (SE) and Poole–Frenkel emission (PFE) models, in which the reverse current characteristics according to the electric field (E) are expressed as follows^30–32)   

\begin{align} J_{\textit{SE}}& = A^{**}T^{2}\exp \left[-\frac{q}{kT}\left(\phi_{B} - \sqrt{\frac{qE}{4\pi\varepsilon_{s}^{h}\varepsilon_{0}}}\right)\right],\\ \beta_{\textit{SE}} &= \sqrt{\frac{q}{4\pi \varepsilon_{s}^{h}\varepsilon_{0}}} \end{align} (1)

  

\begin{align} J_{\textit{PFE}} &= BE\exp \left[-\frac{q}{kT}\left(\phi_{t} - \sqrt{\frac{qE}{\pi\varepsilon_{s}^{h}\varepsilon_{0}}}\right)\right],\\ \beta_{\textit{PFE}} &= \sqrt{\frac{q}{\pi\varepsilon_{s}^{h}\varepsilon_{0}}} \end{align} (2)

where A^** is the effective Richardson constant (32 A/cm²K² for n-ZnO), ε₀ is the permittivity of free space, $\varepsilon_{s}^{h}$ is the relative dielectric permittivity of the semiconductor at high-frequency, B is the constant, β_SE is the SE constant, β_PFE is the PFE constant, and ϕ_t is the emission barrier height for electrons from the traps. The theoretical values for ZnO are β_SE = 1.12 × 10⁻⁴ eVV^−1/2 cm^1/2 and β_PFE = 2.24 × 10⁻⁴ eVV^−1/2 cm^1/2.³¹⁾ Figure 3 shows the fitting results to the reverse current characteristics with the SE and the PFE models. As shown in Figs. 3(a) and (b), the current values at 340 and 360 K for both samples are fitted well with the SE model. However, those values at 380 and 400 K did not match well with the SE model. The obtained values from the fittings are shown in Table 2. The barrier heights for both samples exhibited the similar value of ∼0.9 eV. Note that this is similar to the reported value of 0.89 eV for Pt/ZnO Schottky contact obtained from the I–V method.³³⁾

Fig. 2

Current density–voltage (J–V) curves measured at different temperatures for the ZnO samples grown at (a) 46 and (b) 96°C.

Fig. 3

Fitting results to the reverse current characteristics with (a) and (b) Schottky emission (SE) and (c) and (d) Poole-Frenkel emission (PFE) for the ZnO samples grown at 46 and 96°C.

Table 2 Extracted electrical parameters of Pt/ZnO contacts formed on the ZnO samples grown at 46 and 96°C.

On the contrary, the current values at 380 and 400 K exhibit stronger electric field dependence (i.e., bias-dependence) than lower temperatures, which were matched better with the PFE model as shown in Figs. 3(c) and (d). The SE emission can be limited by the injection rate from the electrode (i.e., contact-limited transport mechanism) whereas the PFE is involved with the transport of carriers via traps in the insulator (i.e., bulk-limited transport mechanism).³⁴⁾ Further, PFE can be dominant at high temperatures, when the carriers are thermally detrapped from the traps. At lower temperatures (340 and 360 K), electrons from the electrode can overcome the energy barrier of Pt/ZnO interface by obtaining thermal energy, resulting in the SE-dominated current transport. At higher temperatures (380 and 400 K), carriers trapped in the ZnO layer can be detrapped thermally and the current conduction via these traps occurs by the PFE effect. As shown in Table 2, the values of ϕ_t were found to be 0.60∼0.68 eV. It was suggested that blue luminescence (BL) in ZnO was related with the transition from oxygen vacancy (V_O) located at 0.6 eV below the conduction band to valence band.³⁵⁾ Based on the result of thermally annealed ZnO, Kang et al. reported that V_O-related deep levels are located at 0.8∼1.0 eV below the conduction band.³⁶⁾ Therefore, we conjecture that V_O-related defects may be associated with the PFE current conduction.

According to the temperature-dependent electrical conductivity for ZnO nanoparticles, the activation energies at lower and higher temperature stages (corresponding to 300–540 K and 540–620 K, respectively) were obtained as 0.22 and 0.47 eV, respectively.³⁷⁾ Similarly, Dinesha et al. reported the activation energies for ZnO nanoparticles at lower and higher temperature stages (corresponding to 300–400 K and 400–550 K, respectively) to be 0.47 and 0.97 eV, respectively.³⁸⁾ The activation energy at lower temperature stage (300–400 K) was associated with one of the following two donor ionization processes involving oxygen vacancy (V_O → V_O⁺ + e⁻) or Zn interstitial (Zn_i⁺ → Zn_i⁺⁺ + e⁻).^37,38) As discussed before, the PFE model in Figs. 3(c) and (d) was associated with V_O-related defects. The ionized oxygen vacancy at high temperatures (380 and 400 K) might increase the conductivity, which in turn contributed to the PFE conduction.

We also measured the J–V values at different temperatures for T141 sample, which are shown in the top panel of Fig. 4(a). The current values changed little with the temperature, indicating that tunneling current is involved strongly. The bottom panel of Fig. 4(a) shows the J–V data obtained from different contact configurations. Since the different electrode metals (Al and Pt) exhibit the similar current values, the barrier is thin enough for electrons to tunnel through it. This assures again that the field emission (FE) is mainly involved. In addition, the linear region in the J–V curves are very narrow and hence, we could not apply the typical thermionic emission (TE) model to analyze the I–V data. Therefore, we used other method utilizing reverse biased I–V data as suggested by Mori et al.,³⁸⁾ to extract the barrier height. In this method, current density can be described as^39–45)   

\begin{equation} J = J_{0}\exp (-qV/nkT)[1 - \exp (-qV/kT)] \end{equation} (3)

  

\begin{equation} J_{0} = A^{**}T^{2}\exp (-q\phi_{B}/kT) \end{equation} (4)

where J₀ is the reverse-bias saturation current density, n is the ideality factor, and ϕ_B is the effective barrier height. The ln(J₀) was extrapolated from the intercept of logarithmic plots of J/[1 − exp(−qV/kT)] vs. V as shown in Fig. 4(b). Wahid et al.,⁴⁰⁾ Su et al.,⁴³⁾ and Puneetha et al.⁴⁵⁾ also used this method to obtain the barrier height for ohmic contacts to p-GaN. The saturation current density ranges from 5.26 mA at 298 K to 6.92 mA at 400 K. The weak temperature dependence of saturation current density is another indicator that the tunneling current is the main transport mechanism for T141 sample.⁴²⁾ From these values, the barrier height at each temperature was calculated.

Fig. 4

(a) Temperature dependent current density–voltage (J–V–T) curves (top panel), J–V data at 298 K with different contact configurations (bottom panel) and (b) logarithmic plot of J/[1 − exp(−qV/kT)] versus V for the ZnO sample grown at 141°C.

The Norde method^46,47) was also applied to the forward I–V data to obtain the barrier height. The plots of F(V) vs. V for each temperature, described as following equation, is shown in Fig. 5(a).   

\begin{equation} F(V) = \frac{V}{2} - \frac{kT}{q}\ln \left(\frac{J}{A^{**}T^{2}}\right) \end{equation} (5)

The barrier height can be calculated using the following equation   

\begin{equation} \phi_{B} = F(V_{\min}) + \frac{V_{\min}}{2} - \frac{kT}{q} \end{equation} (6)

where F(V_min) is the minimum value of F(V) and V_min is the corresponding voltage. The barrier heights from I–V and Norde methods are shown in Fig. 5(b). Although there is a slight difference between the I–V and the Norde methods, the barrier height was observed to enhance with increasing the temperature, related with the inhomogeneous Schottky barrier.^44,48) Compared to the reported value of 0.89 eV for Pt/ZnO Schottky contact,³³⁾ the acquired barrier height is relatively low. Hara et al. reported in SiC Schottky contact that the barrier lowering by image-force (Δϕ_B) is about 0.2 eV for the carrier concentration of 1.8 × 10¹⁹ cm⁻³.⁴⁹⁾ The intrinsic barrier height (ϕ_B0 = ϕ_B + Δϕ_B) of ∼0.93 eV in Au/n-GaN Schottky diodes was found to be lowered by about 0.06 eV for high doping of 7.8 × 10¹⁸ cm⁻³, which was explained by the onset of FE.⁵⁰⁾ Hence, both the image-force and FE effects lowered the barrier height for T141 sample. Then, the corrected barrier height at room temperature would be about 0.7∼0.75 eV, which is comparable to the value of 0.77 eV obtained from the Pt contact to 30-nm-thick ALD grown ZnO.⁵¹⁾

Fig. 5

(a) Plots of F(V) vs. V at each temperature and (b) obtained barrier heights for the ZnO sample grown at 141°C.

The X-ray diffraction (XRD) results in previous work revealed a wurtzite structure for ALD grown ZnO films.⁵²⁾ From XRD patterns, it was also observed that the (002) direction was dominant for the ZnO films grown at 46°C and 73°C. With increasing the temperature, the (100) direction became dominant for the ZnO films grown at 96°C and 119°C and the (002) direction became stronger than (100) direction for the ZnO film grown at 141°C. Due to the low surface energy of the [002] plane in ZnO, the (002) direction is the thermodynamically favorable growth direction.⁵³⁾ Higher crystallinity and film density could be achieved for the ZnO films grown at higher temperatures because the gas molecules became more reactive with sufficient thermal energy for surface reactions.⁵²⁾ Similarly, it was shown that ZnO films grown at 70 and 90°C had strong (002) orientations with cylindrical, columnar crystal structures, and n-type carrier concentration of ∼10¹⁶ cm⁻³ whereas ZnO films grown at high temperatures (>110°C) showed wedge-shaped crystal structures having c- and a-axis growth directions, high oxygen deficiencies, and high n-type carrier concentrations up to 10²⁰ cm⁻³.²⁶⁾ Therefore, the lower carrier concentration and conductivity were obtained for T46 and T96 samples, which produced different current transport mechanism compared to T141 sample.

Finally, we drew the band diagrams schematically for T46 and T96 (Fig. 6(a)) and T141 samples (Fig. 6(b)) according to the observed results. The whole region below the Pt contact is depleted due to the low carrier concentration at low temperatures (298 and 320 K) for T46 and T96 samples. At intermediate temperature (340 and 360 K), electrons at the Pt electrode gain thermal energy to overcome the barrier height of Pt/ZnO contact, associated with the SE. At high temperatures (380 and 400 K), some defects in ZnO layer can be thermally detrapped, causing the PFE current conduction. Note that we could not obtain the depletion width from the capacitance–voltage measurement for T141 sample because the reverse current was very leaky. Mosbacker et al. suggested that the depletion width of Au/ZnO contact corresponds to about 10 nm with the carrier concentration of ∼2 × 10¹⁸ cm⁻³.⁵⁴⁾ Then, we may assume that the depletion width of T141 sample is less than 10 nm, based on the carrier concentration of 1.44 × 10¹⁹ cm⁻³. Moreover, some defects in ZnO would be activated with the temperature, increased the effective carrier concentration. This narrowed the depletion width further, enhancing the tunneling probability. In such case, FE due to strong tunneling current will occur dominantly, possibly with the weak thermionic field emission (TFE).

Fig. 6

Schematic energy diagrams for (a) the ZnO samples grown at 46 and 96°C and (b) the ZnO sample grown at 141°C.

4. Conclusion

We grew ZnO films on soda-lime glass substrates via thermal ALD at different growth temperatures and electrically characterized contact properties of Pt/ZnO junctions. At room temperature, the ZnO sample grown at 46 and 96°C did not show proper current conduction whereas ohmic-like I–V behavior was found for the ZnO sample grown at 141°C. With increasing the measurement temperatures, the current conduction transferred from the SE at 340 and 360 K to the PFE at 380 and 400 K for the ZnO samples grown at 46 and 96°C. Possibly the detrapping of defects from traps in ZnO would affect the current transport via PFE mechanism. Very thin depletion width was formed for the ZnO sample grown at 141°C because of high carrier concentration, caused the strong tunneling current.

Acknowledgments

This study was supported by the Research Program funded by the Seoul National University of Science and Technology (Seoultech).

REFERENCES

• 1)   Ü.  Özgür,  Ya.I.  Alivov,  C.  Liu,  A.  Teke,  M.  Reshchikov,  S.  Doğan,  V.  Avrutin,  S.  Cho and  H.  Morkoç: J. App. Phys. 98 (2005) 041301. doi:10.1063/1.1992666
• 2)   Y.  Liu,  Y.  Li and  H.  Zeng: J. Nanomater. 2013 (2013) 196521. doi:10.1155/2013/196521
• 3)   Y.  Choi,  J.  Kang,  D.  Hwang and  S.  Park: IEEE Trans. Electron Devices 57 (2010) 26. doi:10.1109/TED.2009.2033769
• 4)   V.  Raj,  A.  Nimal,  Y.  Parmar,  M.  Sharma,  K.  Sreenivas and  V.  Gupta: Sensor Act. B 147 (2010) 517. doi:10.1016/j.snb.2010.03.079
• 5)   J.  Park,  S.  Choi and  S.  Kim: Nanoscale Res. Lett. 5 (2010) 353. doi:10.1007/s11671-009-9487-3
• 6)   K.  Matsubara,  P.  Fons,  K.  Iwata,  A.  Yamada,  K.  Sakurai,  H.  Tampo and  S.  Niki: Thin Solid Films 431–432 (2003) 369. doi:10.1016/S0040-6090(03)00243-8
• 7)   M.  Ohtaki,  K.  Araki and  K.  Yamamoto: J. Electron. Mater. 38 (2009) 1234. doi:10.1007/s11664-009-0816-1
• 8)   R.  Triboulet and  J.  Perriere: Prog. Cryst. Growth Charact. Mater. 47 (2003) 65. doi:10.1016/j.pcrysgrow.2005.01.003
• 9)   C.  Ahn,  Y.  Kim,  D.  Kim,  S.  Mohanta and  H.  Cho: J. Appl. Phys. 105 (2009) 013502. doi:10.1063/1.3054175
• 10)   Z.  Ng,  K.  Chan,  C.  Low,  S.  Kamaruddin and  M.  Sahdan: Ceram. Int. 41 (2015) S254. doi:10.1016/j.ceramint.2015.03.183
• 11)   S.  Särkijärvi,  S.  Sintonen,  F.  Tuomisto,  M.  Bosund,  S.  Suihkonen and  H.  Lipsanen: J. Crys. Growth 398 (2014) 18. doi:10.1016/j.jcrysgro.2014.04.006
• 12)   T.  Tynell and  M.  Karppinen: Semicond. Sci. Technol. 29 (2014) 043001. doi:10.1088/0268-1242/29/4/043001
• 13)   J.  Faugier-Tovar,  F.  Lazar,  C.  Marichy and  C.  Brylinski: Condens. Matt. 2 (2017) 3. doi:10.3390/condmat2010003
• 14)   D.  Kim,  H.  Kang,  J.  Kim and  H.  Kim: Appl. Surf. Sci. 257 (2011) 3776. doi:10.1016/j.apsusc.2010.11.138
• 15)   H.  Makino,  A.  Miyake,  T.  Yamada,  N.  Yamamoto and  T.  Yamamoto: Thin Solid Films 517 (2009) 3138. doi:10.1016/j.tsf.2008.11.088
• 16)   C.  Chang and  F.  Tsai: J. Mater. Chem. 21 (2011) 5710. doi:10.1039/c0jm04066e
• 17)   P.  Rowlette,  C.  Allen,  O.  Bromley,  A.  Dubetz and  C.  Wolden: Chem. Vap. Depos. 15 (2009) 15. doi:10.1002/cvde.200806725
• 18)   A.  Di Mauro,  M.  Cantarella,  G.  Nicotra,  V.  Privitera and  G.  Impellizzeri: Appl. Catal. B 196 (2016) 68. doi:10.1016/j.apcatb.2016.05.015
• 19)   D.  Choi,  S.  Kim,  J.  Lee,  K.  Chung and  J.  Park: Curr. Appl. Phys. 12 (2012) S19. doi:10.1016/j.cap.2012.02.012
• 20)   R.  Schifano,  P.  Dluzewski,  W.  Zajkowska,  B.  Kurowska,  T.  Krajewski,  R.  Jakiela,  G.  Luka,  B.  Witkowski,  D.  Jarosz,  R.  Minikayev,  A.  Wierzbicka,  K.  Goscinski,  K.  Kopalko,  E.  Guziewicz and  P.  Smertenko: Appl. Surf. Sci. 552 (2021) 149067. doi:10.1016/j.apsusc.2021.149067
• 21)   H.  Hwang,  K.  Chang,  W.  Yoo,  C.  Shim,  S.  Lee,  J.  Yang,  S.  Kim,  Y.  Lee,  C.  Cho and  B.  Lee: Nanoscale 9 (2017) 2442. doi:10.1039/C6NR08829E
• 22)   A.  Kocyigit,  I.  Orak,  Z.  Çaldıran and  A.  Turut: J. Mater. Sci. Mater. Electron. 28 (2017) 17177. doi:10.1007/s10854-017-7646-3
• 23)   S.  Zhang,  S.  Wei and  A.  Zunger: Phys. Rev. B 63 (2001) 075205. doi:10.1103/PhysRevB.63.075205
• 24)   N.  Huby,  S.  Ferrari,  E.  Guziewicz,  M.  Godlewski and  V.  Osinniy: Appl. Phys. Lett. 92 (2008) 023502. doi:10.1063/1.2830940
• 25)   S.  Kwon,  S.  Bang,  S.  Lee,  S.  Jeon,  W.  Jeong,  H.  Kim,  S.  Gong,  H.  Chang,  H.  Park and  H.  Jeon: Semicond. Sci. Technol. 24 (2009) 035015. doi:10.1088/0268-1242/24/3/035015
• 26)   S.  Jeon,  S.  Bang,  S.  Lee,  S.  Kwon,  W.  Jeong,  H.  Jeon,  H.  Chang and  H.  Park: J. Electrochem. Soc. 155 (2008) H738. doi:10.1149/1.2957915
• 27)   T.  Singh,  T.  Lehnen,  T.  Leuning,  D.  Sahu and  S.  Mathur: Appl. Surf. Sci. 289 (2014) 27. doi:10.1016/j.apsusc.2013.10.071
• 28)   Y.  Wang,  K.  Kang,  M.  Kim and  H.  Park: Appl. Surf. Sci. 529 (2020) 147095. doi:10.1016/j.apsusc.2020.147095
• 29)   K.  Bergum,  H.  Fjellvåg and  O.  Nilsen: Appl. Surf. Sci. 332 (2015) 494. doi:10.1016/j.apsusc.2015.01.124
• 30)   H.  Zhang,  E.  Miller and  E.  Yu: J. Appl. Phys. 99 (2006) 023703. doi:10.1063/1.2159547
• 31)   A.  Farag,  W.  Farooq and  F.  Yakuphanoglu: Microelectron. Eng. 88 (2011) 2894. doi:10.1016/j.mee.2011.03.016
• 32)   Z.  Khurelbaatar,  K.  Shim,  J.  Cho,  H.  Hong,  V.  Rddy and  C.  Choi: Mater. Trans. 56 (2015) 10. doi:10.2320/matertrans.M2014263
• 33)   S.  Kim,  H.  Kim and  T.  Seoung: Appl. Phys. Lett. 86 (2005) 112101. doi:10.1063/1.1862772
• 34)   E.  Lipp,  Z.  Shahar,  B.  Bittel,  P.  Lenahan,  D.  Schwendt,  H.  Osten and  M.  Eizenberg: J. Appl. Phys. 109 (2011) 073724. doi:10.1063/1.3573036
• 35)   Y.  Lin,  C.  Tsai,  Y.  Lu and  C.  Liu: J. Appl. Phys. 99 (2006) 093501. doi:10.1063/1.2193649
• 36)   H.  Kang,  J.  Kang,  J.  Kim and  S.  Lee: J. Appl. Phys. 95 (2004) 1246. doi:10.1063/1.1633343
• 37)   S.  Amin and  A.  Sedky: Mater. Res. Express 6 (2019) 065903. doi:10.1088/2053-1591/ab08c6
• 38)   M.  Dinesha,  H.  Jayanna,  S.  Ashoka and  G.  Chandrappa: J. Alloy. Compd. 485 (2009) 538. doi:10.1016/j.jallcom.2009.06.022
• 39)   T.  Mori,  T.  Kozawa,  T.  Ohwaki,  Y.  Taga,  S.  Nagai,  S.  Yamasaki,  S.  Asami,  N.  Shibata and  M.  Koike: Appl. Phys. Lett. 69 (1996) 3537. doi:10.1063/1.117237
• 40)   S.  Wahid,  N.  Chowdhury,  M.  Alam and  T.  Palacios: Appl. Phys. Lett. 116 (2020) 213506. doi:10.1063/5.0010699
• 41)   Y.  Zhou and  S.  Ramanathan: J. Appl. Phys. 113 (2013) 213703. doi:10.1063/1.4807922
• 42)   K.  Çınar,  N.  Yıldırım,  C.  Coşkun and  A.  Turut: J. Appl. Phys. 106 (2009) 073717. doi:10.1063/1.3236647
• 43)   H.  Su,  T.  Zhang,  S.  Xu,  J.  Lu,  H.  Du,  H.  Tao,  J.  Zhang and  Y.  Hao: Appl. Phys. Lett. 120 (2022) 222101. doi:10.1063/5.0090693
• 44)   B.  Moon,  G.  Han,  H.  Kim,  H.  Choi,  J.  Bae,  J.  Kim,  Y.  Jin,  H.  Jeong,  M.  Joo,  Y.  Lee and  S.  Lim: ACS Appl. Mater. Interfaces 9 (2017) 11240. doi:10.1021/acsami.6b16692
• 45)   P.  Puneetha,  M.  Reddy,  Y.  Lee,  S.  Jeong,  R.  Lokanadham,  C.  Park,  A.  Pradeep and  V.  Reddy: J. Mater. Sci. Mater. Electron. 28 (2017) 16903. doi:10.1007/s10854-017-7609-8
• 46)   H.  Norde: J. Appl. Phys. 50 (1979) 5052. doi:10.1063/1.325607
• 47)   V.  Reddy,  L.  Rao,  V.  Janardhanam,  M.  Kang and  C.  Choi: Mater. Trans. 54 (2013) 2173. doi:10.2320/matertrans.M2013281
• 48)   İ.  Taşçıoğlu,  U.  Aydemir,  Ş.  Altındal,  B.  Kınacı and  S.  Özçelik: J. Appl. Phys. 109 (2011) 054502. doi:10.1063/1.3552599
• 49)   M.  Hara,  S.  Asada,  T.  Maeda and  T.  Kimoto: Appl. Phys. Express 13 (2020) 041001. doi:10.35848/1882-0786/ab7bcd
• 50)   S.  Noh and  P.  Bhattacharya: Appl. Phys. Lett. 78 (2001) 3642. doi:10.1063/1.1377848
• 51)   M.  Shen,  T.  Muneshwar,  K.  Cadien,  Y.  Tsui and  D.  Barlage: IEEE Trans. Electron Devices 65 (2018) 4513. doi:10.1109/TED.2018.2866598
• 52)   J.Y.  Park,  Y.B.  Weon,  M.J.  Jung and  B.J.  Choi: Arch. Metall. Mater. 67 (2022) 1503. doi:10.24425/amm.2022.141082
• 53)   N.  Yuan,  S.  Wang,  C.  Tan,  X.  Wang,  G.  Chen and  J.  Ding: J. Crys. Growth 366 (2013) 43. doi:10.1016/j.jcrysgro.2012.12.024
• 54)   H.  Mosbacker,  Y.  Strzhemechny,  B.  White,  P.  Smith,  D.  Look,  D.  Reynolds,  C.  Litton and  L.  Brillson: Appl. Phys. Lett. 87 (2005) 012102. doi:10.1063/1.1984089