In this work, different annealing ambient (nitrogen-oxygen-nitrogen (N₂-O₂-N₂), forming gas-oxygen-forming gas (FG-O₂-FG), and argon-oxygen-argon (Ar-O₂-Ar)) were explored to investigate the feasibility of employing the annealed ternary Ga_xCe_yO_z passivation layer (PL) for development of Si-based metal-oxide-semiconductor (MOS) capacitors. The impact of nitrogen and/or hydrogen in hindering the growth of silicon dioxide (SiO₂) interfacial layer (IL) was quantitatively evaluated. The combination of effects brought by nitrogen attached to oxygen vacancies, nitrogen-silicon bonding, and nitrogen accumulation at the Ga_xCe_yO_z/Si interface effectively minimized the formation of SiO₂ IL. Consequently, among all the samples, the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient exhibited superior MOS characteristics in terms of low effective oxide charge, slow trap density, interface trap density, and interface state density, which have translated into good leakage current density-electric field characteristics.

1. Introduction

The persistent downscaling of complementary metal-oxide-semiconductor (CMOS) devices, which are the foundation of silicon- (Si-) integrated circuits (IC), serves to enhance device packing density and minimizes power consumption of advanced electronic devices [1]. Consequently, the growing demand of high-performance electronic devices has pushed the scaling of device dimension to its limit, whereby the employment of ultrathin silicon dioxide (SiO₂) gate oxide as passivation layer (PL) in Si-based MOS devices would increase the probability of electrons tunnelling through the trapezoidal energy barrier [2]. Inevitably, the occurrence of this direct tunnelling mechanism would promote the undesirable leakage current during the operation of Si-based MOS devices contributing towards the generation of additional heat as well as reliability issues [3]. Therefore, it is necessary to substitute the conventional SiO₂ possessing a low dielectric constant (k = 3.9) with a high-k material as PL for Si-based MOS devices [4]. The employment of high-k material would allow the utilization of a physically thicker PL as well as higher gate capacitance for Si-based MOS devices, which would ultimately lead to the accumulation of more charge carriers at the Si channel [5]. Therefore, the utilization of high-k material as a PL in fabricated Si-based MOS devices could potentially result in the ability to simultaneously endure a higher electric breakdown field and a lower leakage current density [5, 6]. To this day, a vast range of high-k substances, such as Al₂O₃ [7], CeO₂ [8], Yb₂O₃ [9], ZrO₂ [10], HfO₂ [11], Nd₂O₃ [12], La₂O₃ [13], MgO [14], and Tm₂O₃ [15], have been studied as possible alternatives to SiO₂ as the PL for Si-based MOS devices.

Among these investigated high-k materials, CeO₂ has leaped into the forefront to replace conventional SiO₂ as the PL for Si-based MOS devices due to the intriguing features of CeO₂, such as high-k values ranging from 10 to 16 [8, 16], large bandgap ranging from 3.0 to 3.6 eV [8, 17], low interface state density (<10¹¹ cm^-2 eV^-1) [18], low stress-induced defect formation [19], and low lattice mismatch with the Si substrate [20]. The utilization of a physically thicker, high-k CeO₂ PL would effectively reduce the occurrence of leakage current governed by direct tunnelling mechanism that would result in an improvement in the electric breakdown field of the Si-based MOS device with the usage of CeO₂ PL [21]. It was proven in previous research that Si-based MOS capacitor with CeO₂ PL subjected to a combination of post-nitrous oxide (N₂O) plasma treatment and rapid thermal annealing in nitrogen ambient was able to withstand a high electric breakdown field of 24.67 MV/cm [22]. The versatility of CeO₂ to be employed in the applications, such as energy storage, catalysts, and electrochemical devices, was due to its reversible transition from oxidation state of Ce⁴⁺ to reduction state of Ce³⁺, but the feasibility of utilizing this characteristic for passivation layer was deemed to be quite challenging [23, 24]. Earlier studies indicated that although CeO₂ has shown the potential as a PL for Si-based MOS devices, the issue of partial reducibility of Ce⁴⁺ to Ce³⁺ states remained present as a challenge due to the polaron hopping mechanism that occurred between localized states within the forbidden gap, resulting in an increase in leakage current through the CeO₂ PL [25, 26]. The occurrence of this reducibility effect hampering the passivating property of CeO₂ PL was due to the formation of oxygen vacancies [16, 27] that would augment the diffusion of oxygen by modifying the valance state of CeO₂, as observed in prior studies [26, 28]. Consequently, there was an increase in the growth of the low-k SiO₂ IL due to the availability of a larger number of oxygen ions to diffuse to the Si surface [27]. Besides, the formation of oxygen vacancies in the CeO₂ PL would also act as a shallow trap that would promote an increase in leakage current density [29, 30].

In order to circumvent the above-mentioned issue, an initial approach of improving the passivating properties of CeO₂ was through the doping of tetravalent cations, such as Zr⁴⁺ [31], Sn⁴⁺ [32], Si⁴⁺ [33], Ti⁴⁺ [34], and Hf⁴⁺ [35] into the CeO₂ crystal lattice. Unfortunately, prior studies have revealed that the doping of tetravalent cations into the crystal lattice of CeO₂ has reduced the energy required for defect formation, which in turn promoted the creation of additional oxygen vacancies [36, 37]. Thus, the creation of additional deep states within the bandgap of the investigated tetravalent doped CeO₂ would contribute to bandgap narrowing as well as promoting leakage current [27, 36, 38]. Consequently, the use of trivalent cations (M³⁺), such as Mn³⁺, Gd³⁺, Sc³⁺, La³⁺, Ho³⁺, Er³⁺, and Y³⁺ ions as a dopant for CeO₂, has become the focus of research as a lower energy is required to dope the M³⁺ ions into CeO₂ lattice compared to generating oxygen vacancies through CeO₂ phase reduction [39, 40]. Although doping of CeO₂ by M³⁺ leads to the emergence of additional oxygen vacancies to maintain electroneutrality, the ability to forge a strong coulombic interaction between the trivalent ions and oxygen vacancies would give rise to the formation of vacancy dopant defect clusters (M³⁺-V_o) [41, 42]. The generation of M³⁺-V_o defect pair in CeO₂-doped trivalent ions would increase the energy barrier for oxygen migration [43], thereby limiting the accumulation of oxygen ions at the interface. The generation of M³⁺-V_o defect pair has been found to exhibit a beneficial effect on Sc³⁺-, Ho³⁺-, and Er³⁺-doped CeO₂, whereby these M³⁺ ions would serve as scavengers of oxygen vacancies, inhibiting the diffusion of oxygen ions towards the interface [40, 44]. Moreover, the strength of the interaction between M³⁺-V_o defect pair was found to be significantly influenced by the ionic radius difference among the trivalent ions and Ce⁴⁺ ions, indicating that the doping of trivalent ions having smaller ionic radii could lead to the strengthening of the interaction between these ions and oxygen vacancies, ultimately leading to the inhibition of oxygen diffusion [41].

This study proposes to use trivalent gallium (Ga³⁺) ions doped into the CeO₂ crystal lattice to form a ternary Ga_xCe_yO_z, whereby a larger difference in ionic radius between the Ga³⁺ ion (0.62 Å) [45] and the Ce⁴⁺ ion (0.97 Å) [46] would translate to a stronger coulombic interaction between Ga³⁺ and oxygen vacancies, thereby restricting oxygen ions from diffusing to the interface [47]. Additionally, the passivating properties of the Ga_xCe_yO_z PL could be improved due to the excellent properties of Ga₂O₃, which includes the possession of a larger bandgap of 4.9 eV [48] as well as high-k values (10.2-14.2) [49, 50]. Optimizing the postdeposition annealing parameters is crucial to further improve passivating characteristics of the Ga_xCe_yO_z PL, as previous research has revealed that the Ga_xCe_yO_z PL annealed at 700°C in a nitrogen-oxygen-nitrogen (N₂-O₂-N₂) ambient has shown the best MOS characteristics in terms of the minimum leakage current density, maximum k value, and the largest bandgap [51]. Moreover, the process of annealing at a temperature of 700°C in N₂-O₂-N₂ ambient was found to be highly effective in regulating the growth of the SiO₂ IL [51]. This was due to the existence of nitrogen ions which actively participated in attaching to the oxygen vacancies, thereby hampering oxygen ions from diffusing to the interface [51]. It is worth highlighting that oxygen gas was introduced during dwelling stages regardless of the employment of different gases during the heating and cooling stages with the purpose of repairing the oxygen-related defects and broken bonds [52]. It was proven in other research works that a reduction in interface trap density [53, 54] and fixed oxide charge [55, 56] as well as a low leakage current [57, 58] and negligible hysteresis [59, 60] was reported when the high-k materials acting as PL were subjected to postdeposition annealing in oxygen ambient. Nevertheless, postdeposition annealing in oxygen ambient would result in an exaggeration in the formation of SiO₂ IL between the high-k PL and Si substrate [61]. Hence, previous investigation on high-k materials as the PL has diverted towards the usage of nitrogen ambient during postdeposition annealing process for the nitrogen ions to diffuse to the interface between the PL and Si substrate, to accumulate, and to form a barrier layer in order to impede the formation of SiO₂ interfacial layer [62]. Moreover, the utilization of ambient comprising of nitrogen and hydrogen gases during postdeposition annealing would facilitate the reduction of defect states and the improvement of interface quality by passivating the Si dangling bonds with either the hydrogen or the nitrogen, as well as by having the nitrogen to attach to oxygen vacancies, which would eventually improve leakage current density and breakdown field of the Si-based MOS devices [63, 64]. Besides nitrogen and hydrogen gases, the beneficial effects of utilizing argon ambient during postdeposition annealing were also being highlighted for other high-k PL, wherein an improvement in leakage current density, interface trap density, and effective oxide charge, as well as the mitigation in the formation of SiO₂ interfacial layer, was reported when argon ambient was being used [65, 66]. Dissimilar from previous work, this work intends to utilize different combinations of gases during postdeposition annealing process, whereby different gases will be used during the heating and cooling stages while oxygen gas will be kept the same during the dwelling stage of the postdeposition annealing process. Hence, a combination of forming gas-oxygen-forming gas (FG-O₂-FG), nitrogen-oxygen-nitrogen (N₂-O₂-N₂) [51], and argon-oxygen-argon (Ar-O₂-Ar) at 700°C was used to provide a comparative study on the structural, morphological, optical, and metal-oxide-semiconductor (MOS) characteristics of Ga_xCe_yO_z PL deposited on Si substrate, which has not been reported in previous works.

2. Experimental

The 4-inch n-type Si substrate (100) orientation was sliced into 1 cm² dimension and subjected to standard Radio Corporation of America (RCA) cleaning process. Following that, the RCA cleaned samples were immersed in a diluted hydrofluoric (HF) acid solution to etch the native oxide of SiO₂ from the surface of Si substrate. It was highlighted in previous research works on the importance of chemical purity in the formation of a high-quality passivation layer, wherein all chemicals being used in this work are having a purity level of 99.99% [67, 68]. Initially, 0.191 g of gallium(III) nitrate hydrate (Sigma-Aldrich, 99.99%) was dissolved in 5 ml of methanol (J.T. Baker, analytical grade) and stirred for 15 min at 30°C on a hot plate to form a 0.25 M gallium-containing precursor. Subsequently, 1.0936 g of cerium(III) acetylacetonate hydrate (Sigma-Aldrich, 99.99%) was dissolved in 3 ml of methanol (J.T. Baker, analytical grade) and 6 ml of acetic acid (J.T. Baker, CMOS grade) and stirred for 15 min at 60°C to prepare a 0.25 M cerium-containing precursor. Finally, the cerium- and gallium-containing precursors were stirred continuously for 30 min at 60°C to obtain a ratio of 1 : 1. Then, the resulting Ga_xCe_yO_z precursor was cooled down to room temperature. Subsequently, the Ga_xCe_yO_z precursor was spin-coated on an RCA-clean n-type Si substrate with an rpm of 3000 for 30 s. These as-deposited samples were subjected to postdeposition annealing in different ambient of FG-O₂-FG, N₂-O₂-N₂, and Ar-O₂-Ar for 30 min at 700°C in a quartz tube furnace. During the heating and cooling stages, FG or N₂ or Ar (100 ml/min) was flown into the furnace while O₂ gas with a similar flow rate was employed during the dwelling stage. After completing the postdeposition annealing process, the Al/Ga_xCe_yO_z/Si/Al MOS capacitors were realised by evaporating aluminium (Al) contact on the surface of the Ga_xCe_yO_z PL using a shadow mask (diameter = 0.2 mm) and a layer of Al was deposited as the back contact on Si substrate.

Crystalline phase and orientation of the investigated Ga_xCe_yO_z PL subjected to different postdeposition annealing ambient were characterized using grazing incident X-ray diffraction (GIXRD; Bruker D8 Discover) analysis. The surface morphologies and cross sectional images, as well as elemental composition of the investigated Ga_xCe_yO_z PL, were carried out using field-emission scanning electron microscopy (FESEM; FEI Nova NanoSEM 450) equipped with energy dispersive X-ray analysis (EDX; JSM-6460 LV). The estimation of thickness for the Ga_xCe_yO_z PL and SiO₂ IL was determined based on the acquired X-ray reflectivity (XRR) results measured using Bruker D8 Discover. 3-dimensional topographies as well as root-mean-square (RMS) roughness of the investigated Ga_xCe_yO_z PL were characterized using atomic force microscopy (AFM; Dimension Edge, Bruker). The bandgap of investigated Ga_xCe_yO_z PL was extracted from diffused reflectance spectra acquired using ultraviolet–visible (UVVIS) spectrophotometer (Cary 5000). The capacitance-voltage (C-V) and current-voltage (I-V) characteristics of the investigated samples were measured using Keithley 4200-SCS parameter analyzer.

3. Results and Discussions

3.1. Structural, Chemical, and Optical Characteristics of Ga_xCe_yO_z PL Annealed in FG-O₂-FG, N₂-O₂-N₂, and Ar-O₂-Ar Ambient

The GIXRD pattern as represented in Figure 1 indicates that the Ga_xCe_yO_z PL annealed in FG-O₂-FG, N₂-O₂-N₂, and Ar-O₂-Ar ambient at 700°C have emerged as polycrystalline, with the detection of Ga_xCe_yO_z phase oriented in (111), (200), (220), and (311) planes for all of the Ga_xCe_yO_z PL, which were indexed with the International Centre for Diffraction Data (ICDD) file no. 00-034-0394 for cubic fluorite CeO₂ phase. Additionally, it was also perceived that no additional phases with regard to either cubic Ga₂O₃ (ICDD file no. 00-020-0426) or monoclinic Ga₂O₃ (ICDD file no. 00-041-1103) phases were detected from the attained GIXRD results indicating that no phase separation has taken place. Furthermore, it was observed that the GIXRD peaks associated with the Ga_xCe_yO_z phase exhibited a displacement towards a higher diffraction angle in contrast to the CeO₂ phase documented in the ICDD file no. 00-034-0394. This could be an indication that Ga³⁺ ions having a smaller ionic radius of 0.62 Å than Ce⁴⁺ ions (0.97 Å) have successfully doped into the CeO₂ crystal framework to form ternary Ga_xCe_yO_z phase.

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

GIXRD patterns of Ga_xCe_yO_z PL annealed in different ambient of N₂-O₂-N₂ [51], FG-O₂-FG, and Ar-O₂-Ar.

The lattice parameter a of the investigated Ga_xCe_yO_z PL was determined based on the equation given below [69]:

(1)

where d_hkl, θ, and hkl are the interplanar spacing, diffraction angle, and Miller’s index, respectively. Table 1 presents the calculated lattice parameter a for the investigated Ga_xCe_yO_z PL subjected to annealing at 700°C in different ambient. It has been noted that all Ga_xCe_yO_z PL have undergone lattice contraction in comparison to standard CeO₂, which has a lattice parameter of 0.5411 nm [70]. This observation supports the successful doping of Ga³⁺ ions, which have a smaller ionic radius than Ce⁴⁺ ions, into the CeO₂ crystal lattice, resulting in the formation of a ternary Ga_xCe_yO_z phase. Of these investigated Ga_xCe_yO_z PL, it was noticed that Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient has attained the smallest lattice parameter a (Table 1). The observation at hand may potentially be attributed to the generation of an elevated concentration of oxygen vacancies within the Ga_xCe_yO_z PL during the annealing process in N₂-O₂-N₂ ambient. This, in turn, could result in the attachment of nitrogen ions to the generated oxygen vacancies, hence leading to the release of adjacent oxygen ions from the Ga_xCe_yO_z lattice as a means of achieving charge neutrality [51]. It was noticed that a smaller reduction in the lattice parameter a was attained for Ga_xCe_yO_z PL subjected to annealing in FG-O₂-FG ambient compared to N₂-O₂-N₂ ambient. The aforementioned observation implied that the coexistence of hydrogen and nitrogen ions during annealing in an FG-O₂-FG ambient would result in the hydrogen ions, which possess a smaller ionic radius (0.0290 nm) [71] in comparison to the nitrogen ions (0.1460 nm) [72], diffusing at a faster rate to attach with the oxygen vacancies in the Ga_xCe_yO_z lattice. Subsequently, the nitrogen ions from the FG-O₂-FG ambient that were diffusing at a slower rate would be attached to the unoccupied oxygen vacancies in the Ga_xCe_yO_z lattice in which an excess of negative charges would be generated in the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient. Nevertheless, due to the earlier attachment of hydrogen ions with oxygen vacancies that have contributed to the generation of an excess positive charges in the Ga_xCe_yO_z lattice, charge compensation has taken place and the release of adjacent oxygen ions to achieve charge neutrality in the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient was not required. Upon analysis, it was observed that the peaks corresponding to the Ga_xCe_yO_z phase in the PL annealed in the FG-O₂-FG ambient have shifted towards smaller diffraction angles in comparison to the N₂-O₂-N₂ ambient. This finding serves as substantiating proof that the generation of oxygen vacancies in the Ga_xCe_yO_z PL was reduced in the FG-O₂-FG ambient. Since postdeposition annealing carried out in Ar-O₂-Ar ambient did not involve the incorporation of nitrogen and/or hydrogen ions into the Ga_xCe_yO_z lattice, the attachment of the ions to oxygen vacancies would not happen, and therefore, supposedly the amount of oxygen vacancies present in the Ga_xCe_yO_z PL could be the most significant when compared with the other two ambient, and an abrupt decrease in the lattice parameter a should be expected. Nevertheless, it was noticed that the attained lattice parameter a for this sample (0.5387 nm) was close to the lattice parameter a of the standard CeO₂ sample (0.5411 nm) and to be specific was smaller than CeO₂. This disparity indicated that the acquisition of a similar lattice parameter a was linked to the doping of Ga³⁺ into the CeO₂ crystal lattice, as opposed to the result of either nitrogen or hydrogen ions attached with the oxygen vacancies.

Table 1. Lattice parameter a, crystallite size D, and microstrain ε for Ga_xCe_yO_z PL annealed in different ambient of N₂-O₂-N₂ [51], FG-O₂-FG, and Ar-O₂-Ar as well as lattice parameter a for standard CeO₂.

Annealing ambient	Lattice parameter a (nm)	Crystallite size D (nm)	Microstrain ε
N₂-O₂-N₂	0.5350	15.92	0.0072
FG-O₂-FG	0.5380	13.87	0.0057
Ar-O₂-Ar	0.5387	19.99	0.0109
Standard CeO₂	0.5411	—	—

Further scrutiny was conducted through the utilization of EDX analysis on the scrutinized Ga_xCe_yO_z PL, and the resulting findings are subsequently depicted in Table 2. It was perceived that the sample annealed in FG-O₂-FG ambient attained a higher atomic percentage (at%) of nitrogen when compared to N₂-O₂-N₂ ambient although the employment of forming gas ambient was contained of a lower proportion of nitrogen (90%) in contrast to pure nitrogen ambient. The aforementioned observation has served to bolster the prior postulation that the utilization of FG-O₂-FG ambient would incite the initial attachment of hydrogen ions to the oxygen vacancies, owing to the superior diffusion rate of hydrogen ions as compared to nitrogen ions. Since the ionic radius of hydrogen ions was smaller than that of nitrogen ions, the attachment of hydrogen ions to the oxygen vacancies would provide sufficient space in the lattice for the passage of the nitrogen ions to the interface between the Ga_xCe_yO_z PL and Si substrate when compared with the attachment of nitrogen ions to the oxygen vacancies. Consequently, the outcome of the postdeposition annealing process in FG-O₂-FG ambient was the buildup of a higher concentrations of nitrogen ions at the interface to act as a more potent oxygen diffusion barrier layer, which is highly effective in hindering the formation of SiO₂ IL. The lower value of at% of nitrogen in the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient could be due to the higher likelihood of nitrogen ions attaching to the oxygen vacancies, wherein narrower spacing within the Ga_xCe_yO_z lattice would impede the nitrogen ions from diffusing towards the interface. This indicates that the initially inward diffusing nitrogen ions that were unable to form attachment with oxygen vacancies or diffuse to the interface between Ga_xCe_yO_z PL and Si substrate would subsequently diffuse out of the PL aiding in the acquisition of a lower at% of nitrogen during annealing in N₂-O₂-N₂ ambient. It was noticed that the acquisition of a higher at% of oxygen for Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient when compared with N₂-O₂-N₂ ambient has further reinforced that the simultaneous attachment of hydrogen and nitrogen ions with oxygen vacancies has minimized the release of oxygen ions from Ga_xCe_yO_z lattice. Nevertheless, it was perceived that Ar-O₂-Ar ambient has led to the attainment of the minimum at% of oxygen in Ga_xCe_yO_z PL, but it did not necessarily translate into large amount of oxygen vacancies as evidenced by the close lattice parameter a between CeO₂ and Ga_xCe_yO_z.

Table 2. Elemental composition for the Ga_xCe_yO_z PL annealed in different ambient of N₂-O₂-N₂ [51], FG-O₂-FG, and Ar-O₂-Ar.

Annealing ambient	O (at. %)	Ce (at. %)	Ga (at. %)	N (at. %)
N₂-O₂-N₂	69.45	16.89	9.80	3.85
FG-O₂-FG	69.84	16.09	9.43	4.64
Ar-O₂-Ar	64.92	24.34	10.73	—

The FESEM images of the examined Ga_xCe_yO_z PL shown in Figure 2 could be utilized to analyze the effectiveness of nitrogen and/or hydrogen ions in hindering the formation of SiO₂ IL. The total oxide thickness of the investigated PL was calculated from 10 distinct points, as illustrated in Table 3. It was deduced that the existence of nitrogen and/or hydrogen ions was effective in minimizing the formation of SiO₂ IL as the average total oxide thickness obtained by both Ga_xCe_yO_z PL subjected to postdeposition annealing in N₂-O₂-N₂ (30.41 nm) and FG-O₂-FG ambient (40.16 nm) was lower than that of Ar-O₂-Ar (53.75 nm). In addition, it was also divulged that FG-O₂-FG ambient was not as effective as N₂-O₂-N₂ ambient in retarding the diffusion of oxygen ions to react with Si surface as a larger average total oxide thickness was attained by the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient. The proposition being put forward posited that the accumulation of hydrogen ions at the interface may be comparatively inadequate in impeding the diffusion of oxygen ions towards the interface, in contrast to the accumulation of nitrogen ions at the same interface. It was hypothesised that the accumulation of hydrogen ions with a smaller ionic radius at the interface and a weaker bond strength between the hydrogen ions and Si [68] would enable the oxygen ions to either bypass or disrupt these bonds, leading to the reaction with the underlying Si surface during annealing in FG-O₂-FG ambient. This reaction would contribute to the development of a thicker SiO₂ IL in comparison to N₂-O₂-N₂ ambient.

Table 3. Thicknesses of Ga_xCe_yO_z PL and SiO₂ IL extracted from XRR measurements for the investigated Ga_xCe_yO_z PL. Total oxide thickness of the investigated Ga_xCe_yO_z PL acquired from XRR and cross-sectional FESEM characterizations. The calculated k values using total oxide thickness from XRR measurement for the investigated Ga_xCe_yO_z PL.

Annealing ambient	XRR: Ga_xCe_yO_z PL thickness (nm)	XRR: SiO₂ IL thickness (nm)	XRR: total oxide thickness (nm)	Cross-sectional FESEM: total oxide thickness (nm)	k value
N₂-O₂-N₂	28.054 [51]	0.958 [51]	29.012 [51]	30.4	11.27 [51]
FG-O₂-FG	36.240	2.321	38.561	40.2	18.24
Ar-O₂-Ar	45.112	6.942	52.054	53.7	15.94

To ascertain the thickness of the Ga_xCe_yO_z PL and SiO₂ IL subjected to annealing in varying postdeposition annealing ambient, the resulting XRR data was subjected to fitting procedures using the Bruker DIFFRAC Leptos software (version 7.10.0.12). The XRR experimental and fitted results for the investigated Ga_xCe_yO_z PL are illustrated in Figure 3. Outcome from the analysis of the XRR findings, it was further reinforced that the accumulation of nitrogen ions at the interface as well as the bonding between the nitrogen ions and Si were less susceptible for the oxygen ions to bypass the accumulated nitrogen ions at the interface as well as breaking the bonding between nitrogen ions and Si. Thus, the lowest thickness of SiO₂ IL (0.96 nm) was attained for the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient. Nonetheless, thicker SiO₂ IL was formed for Ga_xCe_yO_z PL annealed in FG-O₂-FG (2.32 nm) and Ar-O₂-Ar (6.94 nm) denoting that the oxygen ions diffusing towards the interface region were less retarded when compared with N₂-O₂-N₂ ambient. The activation energy (E_a) related to the growth of SiO₂ IL or the densification of Ga_xCe_yO_z PL postdeposition annealed in different ambient could be calculated using the subsequent equation [73]:

(2)

where t is the thickness of SiO₂ IL or Ga_xCe_yO_z PL after postdeposition annealing in different ambient, t_o is the thickness of SiO₂ IL (0.20 nm) or Ga_xCe_yO_z PL (160.64 nm) for the as-deposited sample [51], k is Boltzmann’s constant, and T is the postdeposition annealing temperature in Kelvin. The PL that underwent annealing in Ar-O₂-Ar and FG-O₂-FG ambient had larger E_a values of 2.603 × 10⁻¹ and 2.533 × 10⁻¹ eV, respectively, indicating an exaggerated formation of SiO₂ IL when compared to N₂-O₂-N₂ ambient (1.287 × 10⁻¹ eV). Besides, it was also observed that all of the investigated Ga_xCe_yO_z PL were undergoing densification process due to the acquisition of E_a of 1.462 × 10⁻¹, 1.356 × 10⁻¹, and 0.932 × 10⁻¹ eV for N₂-O₂-N₂, FG-O₂-FG, and Ar-O₂-Ar ambient, respectively. It was postulated that the acquisition of the smallest E_a by Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient could be related to the absence of nitrogen and/or hydrogen ions in which the release of oxygen ions during the incorporation of gallium ions into CeO₂ crystal lattice would either diffuse out from the PL or diffuse inward to react with Si surface, and thus, a less denser PL was formed. Although oxygen gas was being introduced during the dwelling stage for different postdeposition annealing ambient, the existence of nitrogen and/or hydrogen ions at the interface during annealing in N₂-O₂-N₂ and FG-O₂-FG ambient would possibly restrict the oxygen ions from reacting with Si surface, of which the unreacted oxygen ions would possibly diffuse outward and participate in the densification of the Ga_xCe_yO_z PL. It was hypothesised that the acquisition of a lower E_a by the Ga_xCe_yO_z PL annealed in FG-O₂-FG than that of N₂-O₂-N₂ ambient could be associated to the inclusion of additional hydrogen ions into the Ga_xCe_yO_z lattice, which restricted the densification of this PL.

The Williamson-Hall (W-H) approach was adopted to estimate the microstrain (ε) and crystallite size (D) of the Ga_xCe_yO_z PL, as represented by the subsequent equation [74]:

(3)

where λ, θ_i, and β_i are X-ray wavelength, diffraction angle, and integral breadth (in radius 2θ) of the ith Bragg reflection positioned at 2θ_i. Figure 4 presents the typical W-H plot for Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient. The slope and intercept from the constructed W-H plots were used to calculate the ε and D, respectively, for the investigated Ga_xCe_yO_z PL annealed at 700°C in different ambient. Table 1 presents the estimated values of ε and D for the Ga_xCe_yO_z PL. It was discerned that Ga_xCe_yO_z PL annealed in both N₂-O₂-N₂ and FG-O₂-FG ambient have attained smaller D when compared with Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient. This observation could be associated to the existence of nitrogen and/or hydrogen ions in the Ga_xCe_yO_z PL annealed in both N₂-O₂-N₂ and FG-O₂-FG ambient that have restricted the coalescence of crystallites. During annealing in Ar-O₂-Ar ambient, the absence of any additional nitrogen and/or hydrogen ions implies that the oxygen ions that diffused into the PL during dwelling stage could occupy the oxygen vacancies in the Ga_xCe_yO_z lattice and assisted in the coalescence of crystallites, after which these oxygen ions were released to diffuse to the interface and react with the underlying Si surface. Therefore, the largest D was obtained for the Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient. Moreover, the attainment of the largest tensile ε in the Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient could be linked to the formation of oxygen vacancies in the PL. A reduction in tensile ε of the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient has suggested that the attachment of nitrogen ions to the oxygen vacancies would minimize the tensile ε induced on the Ga_xCe_yO_z lattice. When FG-O₂-FG ambient was being employed during postdeposition annealing, the lessening in the release of adjacent oxygen ions due to the attachment of both hydrogen and nitrogen ions to the oxygen vacancies has reduced the concentration of oxygen vacancies being formed in Ga_xCe_yO_z PL, and thus, the lowest tensile ε was obtained for this PL.

The determination of the preferred orientation for the Ga_xCe_yO_z PL was accomplished by determining the coefficient of texture (T_hkl), which was formulated through the subsequent equation [75]:

(4)

where I_o(hkl) is the intensity of the CeO₂ standard reference sample containing the same plane, I_m(hkl) is the measured relative intensity of reflection from the (hkl) plane for Ga_xCe_yO_z phase, and n is the number of Ga_xCe_yO_z reflection peaks. Figure 5 shows the computed T_hkl values of the investigated Ga_xCe_yO_z PL. A reference line of T_hkl value equal to 1 was also included in Figure 5 to denote the preferred growth in the specific plane. It was observed that Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient has attained a preferred orientation along (200) plane. Nonetheless, the dominance of (200) plane has subsided with the emergence of (311) plane with T_hkl value larger than 1 when the Ga_xCe_yO_z PL was annealed in FG-O₂-FG ambient. When Ar-O₂-Ar ambient was being employed during postdeposition annealing, it was noticed that a mixed preferred orientation in (111) and (200) planes was attained with the (111) plane being more pronounced than (200) plane.

The UVVIS measurements in diffused reflectance mode enabled the determination of the direct (E^D) and indirect bandgap (E^ID) of the Ga_xCe_yO_z PL by adopting the Kubelka-Munk (KM) function that could be represented by the subsequent equation [76]:

(5)

where R is the diffused reflectance and the F(R) function is multiplied by the photon energy hv using 1/2 and 2 as the corresponding coefficients (n) for direct band and indirect band transitions, respectively. The direct E_g (E^D) and indirect E_g (E^ID) values of the investigated Ga_xCe_yO_z PL were extracted via extrapolation on the linear region of (F(R) × hv)ⁿ vs. hv plot to zero (not shown). The extracted E^D and E^ID values for the investigated Ga_xCe_yO_z PL are represented in Figure 6. It is noteworthy that the extracted values of E^D and E^ID from the investigated Ga_xCe_yO_z PL, which ranged from 3.87 to 4.07 eV and 2.99 to 3.68 eV, respectively, were within the reported range of E^D (4.85 eV) and E^ID (4.66 eV) for Ga₂O₃, as well as E^D (3.12 eV) and E^ID (2.58 eV) for CeO₂ [73, 77]. This suggests that the Ga³⁺ cation has been successfully doped into the CeO₂ crystal lattice, resulting in the formation of ternary Ga_xCe_yO_z phase. Among the investigated samples, the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient has attained the lowest E^D (3.87 eV) and E^ID (2.99 eV), which has denoted that the incorporation of nitrogen ions into the Ga_xCe_yO_z lattice would trigger the narrowing of E^D and E^ID. Similar observation has also been reported for nitrogen-doped CeO₂ nanoparticles prepared through wet chemical route [78]. The enlargement of E^D and E^ID for Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient was in agreement with previous discussion that a higher concentration of nitrogen ions has diffused to the interface when compared with N₂-O₂-N₂ ambient, wherein the attachment of nitrogen ions to oxygen vacancies could be lower, leading to the acquisition of larger E^D (3.90 eV) and E^ID (3.10 eV). Nevertheless, Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient has attained the largest E^D (4.07 eV) and E^ID (3.68 eV) as no additional nitrogen and hydrogen ions were introduced into the Ga_xCe_yO_z lattice.

The typical surface morphology of Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient that was characterized using FESEM technique is shown in Figure 7(d). It was revealed that all of the investigated Ga_xCe_yO_z PL have acquired a smooth surface without visible cracks or voids. Additional assessment was conducted utilizing the AFM characterization, whereby the 3-dimensional surface topographies of the Ga_xCe_yO_z PL were acquired at a scanning area of 10 μm² as presented in Figure 7. It was observed that smaller dimension protrusions with comparable height were uniformly formed throughout the surface of Ga_xCe_yO_z PL annealed in a N₂-O₂-N₂ ambient (Figure 7(a)) leading to the attainment of the lowest root-mean-square (RMS) roughness of 0.793 nm. A substantial increment in RMS roughness to 2.080 nm was noticed when the Ga_xCe_yO_z PL was annealed in FG-O₂-FG ambient, wherein a mixture of protrusions with large and small dimensions was formed throughout the surface of this PL (Figure 7(b)). It was observed that the surface of Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient was comprised of a section with extremely larger protrusion (Figure 7(c)), and thus, the highest RMS roughness of 2.580 nm was attained.

3.2. The Metal-Oxide-Semiconductor (MOS) Characteristics of the Al/Ga_xCe_yO_z/4H-SiC/Al MOS Capacitor

The bidirectional C-V curves measured at 1 MHz with applied voltage from -4 to +4 V for Ga_xCe_yO_z PL subjected to different postdeposition annealing ambient are depicted in Figure 8. The estimation of dielectric constant (k) values has been carried out based on the subsequent equation [79]:

(6)

where C_ox, t_ox, ε_o, and A refer to the value of capacitance at accumulation level, total oxide thickness obtained from the XRR measurements (Table 3), free space permittivity, and metal contact area, respectively. The highest k value of 18.24 was perceived by the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient, despite the formation of a thicker SiO₂ IL (2.32 nm) as evidenced by XRR measurement in comparison to the sample annealed in N₂-O₂-N₂ ambient (0.96 nm). It was reported previously that the formation of thicker SiO₂ IL possessing a lower k value of 3.90 would possibly contribute to the reduction of the overall k value of the investigated PL [78]. It was worth noting that the total capacitance (C_tot) of a MOS structure comprising of Ga_xCe_yO_z PL and SiO₂ IL could be determined by the equation given below [80]:

(7)

where C_ox and C_IL are referring to the capacitance of Ga_xCe_yO_z PL and SiO₂ IL, respectively. Based on this equation, it was deduced that the value of capacitance in the accumulation level for the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient could be dominated by the capacitance of Ga_xCe_yO_z PL rather than the capacitance of SiO₂ IL. Similarly, it was also discovered that the k value of Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient (k = 15.94) was also controlled by the capacitance value of Ga_xCe_yO_z PL, wherein the formation of thicker SiO₂ IL (6.94 nm) did not contribute to the acquisition of a higher k value when compared with Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient (k = 11.27). A comparison between the k values obtained by Ga_xCe_yO_z PL annealed in FG-O₂-FG (k = 18.24) and Ar-O₂-Ar (k = 15.94) ambient has disclosed that the formation of thicker SiO₂ IL would contribute to the reduction in k value, but the impact towards the k value could be less significant when compared to the thickness of Ga_xCe_yO_z PL. Although Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient has achieved a better densification than other PL, the reason of getting the lowest k value for this PL remains unknown. Nevertheless, it was determined in this work that the lowest k value (11.27) obtained by the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient at 700°C was higher than the CeO₂ passivation layers subjected to postdeposition annealing within the range of 600 (k = 9.70) to 800°C (k = 6.30) in N₂-O₂-N₂ ambient [8] and CeO₂ PL annealed at 700°C (k = 8.00) in oxygen ambient [81], suggesting that the incorporation of Ga³⁺ into CeO₂ lattice has contributed to the attainment of a larger k value.

The presented C-V curves for the investigated Ga_xCe_yO_z PL in Figure 8 disclosed that negative flat band voltage shift (ΔV_FB) was attained for all samples denoting the existence of positively charged traps in the investigated Ga_xCe_yO_z PL. It was ascertained that the largest negative ΔV_FB was perceived by the Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient designating the formation of the highest density of positively charged traps, which might be due to the formation of oxygen vacancies in the Ga_xCe_yO_z PL. When FG-O₂-FG ambient were employed, the attachment of nitrogen and hydrogen ions to the oxygen vacancies simultaneously has hindered the released of adjacent oxygen ions from the Ga_xCe_yO_z lattice, wherein a smaller negative ΔV_FB was perceived by this PL when compared with Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient. Moreover, the accumulation of negatively charged nitrogen ions at the interface would also compensate the positively charged oxygen vacancies in the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient leading to the reduction in the overall positive charges in this PL. Furthermore, it was noticed that Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient was comprised of the lowest density of positively charged traps due to the acquisition of the smallest negative ΔV_FB. Nevertheless, GIXRD measurement has disclosed the detection of Ga_xCe_yO_z peaks demonstrating the largest variation when compared with standard CeO₂ sample for Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient. This observation has implied the existence of more positively charged oxygen vacancies in the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient in which the adjacent oxygen ions in Ga_xCe_yO_z lattice were released when nitrogen ions were being attached to the oxygen vacancies. In order to explain this discrepancy, it was postulated that the attachment of nitrogen ions to the oxygen vacancies during annealing in N₂-O₂-N₂ ambient would provide a smaller space in the Ga_xCe_yO_z lattice for the subsequent nitrogen ions to diffuse through. As a result, these unattached negatively charged nitrogen ions could linger in the Ga_xCe_yO_z lattice contributing to the acquisition of the lowest density of positively charged traps for Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient. In order to further verify the density of positively charged traps in the investigated Ga_xCe_yO_z PL, the following equation was used to calculate the effective oxide charge (Q_eff) of the investigated PL [82]:

(8)

where q represents electronic charge. Figure 9 shows the calculated Q_eff values for the investigated Ga_xCe_yO_z PL. In addition, it was also determined that the Q_eff values ranging from 1.83 × 10¹² to 4.39 × 10¹² cm^-2 obtained by the Ga_xCe_yO_z passivation layer annealed in different ambient of N₂-O₂-N₂, FG-O₂-FG, and Ar-O₂-Ar ambient were lower than the previously reported Ta_xGd_yO_z (5.45 × 10¹² cm^-2) [83] and HfO₂ (4.78 × 10¹² cm^-2) [84] PL deposited on the Si substrate.

Further examination on the slow traps formed in the investigated Ga_xCe_yO_z PL was calculated using the following slow trap density (STD) equation [85]:

(9)

where ΔV is the difference between forward and reverse bias flat band voltage. Figure 9 presents the calculated STD of the investigated Ga_xCe_yO_z PL. Among all the investigated Ga_xCe_yO_z PL, the lowest STD was acquired by the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient, which could be related to the negatively charged nitrogen ions accumulated at the interface. When the electrons were being injected during forward biased, these electrons would not be attracted to the negatively charged nitrogen ions accumulated at the interface, and thus, the probability of trapping the electrons would be lower leading to the attainment of the lowest STD for Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient. A slight increment in STD was perceived when the Ga_xCe_yO_z PL was annealed in FG-O₂-FG ambient, wherein the formation of weaker bonding between hydrogen ions and Si could be broken by the electrons during forward bias. Further increment in the forward bias would allow the subsequent injected electrons to be trapped by these positively charged Si dangling bonds, and these electrons would be detrapped during reverse biased. The nonexistence of nitrogen as well as hydrogen ions during annealing in Ar-O₂-Ar ambient would allow more electrons to be captured by the Si dangling bonds as well as the positively charged oxygen vacancies during forward bias. Thus, the highest STD was obtained by this PL as the defects were located nearer to the interface and could be potentially detrapped when reverse biased was being applied.

The interface quality between the investigated Ga_xCe_yO_z PL and the Si substrate was analyzed using the Terman’s method in which the interface trap density (D_it) could be determined based on the subsequent equation [86]:

(10)

where ΔV_g = V_g–V_g(ideal) is the voltage shift between the experimental (V_g) and the ideal curves and Φ_s is the surface potential at a specific gate voltage. Figure 10 shows the calculated D_it values as a function of energy trap level (E_c-E_t) for Ga_xCe_yO_z PL subjected to different postdeposition annealing ambient. It was perceived that the best interface quality was attained by Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient due to the acquisition of the lowest D_it. This could be an indication that the existence of nitrogen ions during annealing in N₂-O₂-N₂ ambient has succeeded in passivating the Si dangling bonds as well as forming an attachment with the oxygen vacancies located near to the interface between Ga_xCe_yO_z PL and Si surface to reduce the D_it of the investigated PL. It was noticed that the D_it value of 4.21 × 10¹² cm^-2 eV^-1 at 0.40 eV for the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient was comparable with Sc₂O₃ (4.80 × 10¹² cm^-2 eV^-1 at 0.40 eV) [87] as well as HfAlO (4.00 × 10¹² cm^-2 eV^-1 at 0.40 eV) [88] PL subjected to annealing in air and vacuum ambient, respectively. The acquisition of a higher D_it by the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient could be related to the existence of unattached nitrogen ions that were accumulated at the interface in which these nitrogen dangling bonds would degrade the interface quality of the investigated PL [89, 90]. In addition, the presence of hydrogen ions with smaller ionic radius during annealing in FG-O₂-FG ambient would be able to diffuse faster to the interface to passivate the Si dangling bond in which more nitrogen dangling bonds could be formed leading to the attainment of a higher D_it when compared with Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient. It was also discovered in this research work that nitrogen and/or hydrogen ions were effective in passivating the interface defects, wherein the employment of Ar-O₂-Ar ambient during postdeposition annealing has contributed to the acquisition of the highest D_it.

Besides, different frequencies spanning from 10 kHz to 1 MHz were employed to measure the C-V characteristics of the investigated Ga_xCe_yO_z PL, and the obtained results were used to calculate interface state density (N_ss) based on high-low-frequency C-V method, conveyed by the subsequent equation [91]:

(11)

where C_LF is the lowest value of low-frequency (10 kHz) capacitance and C_HF is the highest value high-frequency (1 MHz) capacitance at voltage corresponding to C_LF. Figure 11 shows the typical different frequencies C-V curves measured for Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient. The calculated N_ss value with regard to V_g for the Ga_xCe_yO_z PL is presented in Figure 12. In contrast, Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient at gate voltage below 0.05 V has procured lower N_ss, possibly due to the presence of fewer nitrogen dangling bonds serving as border traps. It was noticed that when the gate voltage was increased beyond 0.05 V, the N_ss for Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient was higher than Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient, denoting that a higher concentration of border traps was located at a higher energy level. This observation has further supported the earlier explanation regarding the acquisition of a higher D_it using Terman’s method for Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient, which has suggested that more nitrogen dangling bonds were formed at a higher energy level. The acquisition of the highest N_ss for Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient was in agreement with the Terman’s method, wherein the nonexistence of nitrogen and/or hydrogen ions during annealing was not able to passivate the oxygen vacancies located nearer to the interface as well as Si dangling bonds.

Figure 13 illustrates the leakage current density-electric field (J- E) characteristics of the Ga_xCe_yO_z PL annealed at 700°C in different ambient. At E below 0.5 MV/cm, Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient has demonstrated a lower J due to the existence of a higher concentration of border traps located at a lower trap energy level in which high-low-frequency C-V measurement has revealed the acquisition of a lower N_ss at gate voltage below 0.05 V. A deterioration in J was perceived as the E was enhanced to the region of 0.5 to 2.0 MV/cm for the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient, which was due to the presence of a higher density of border traps at a higher gate voltage of 0.05 V as well as a higher D_it. Nonetheless, a sudden improvement in J at E within the region of 2.0 to 3.0 MV/cm for the Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient was due to the formation of thicker SiO₂ IL and the highest k value when compared with Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient. Although Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient has attained the lowest k value, this PL was able to withstand the highest electric breakdown field of ~4.95 MV/cm that could be potentially due to the acquisition of the lowest Q_eff, STD, and D_it. It was revealed that the acquired electric breakdown field (E_B) of 4.95 MV/cm for the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient was better when compared with E_B of bilayer HfO₂/SiO₂ (3.90 MV/cm) PL [92]. Moreover, the attained leakage current density of 2.57 × 10⁻⁶ A/cm^-2 at 1 MV/cm for the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient was lower than the magnetron sputtered ErAlO PL on Si substrate (8.40 × 10⁻⁵ A/cm² at 1 MV/cm) [80]. It was discerned that Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient has demonstrated the lowest electric breakdown field (~2.71 MV/cm) due to the formation of the thickest SiO₂ IL as well as the less dense Ga_xCe_yO_z PL, resulting in a poor J- E characteristic. It was postulated that the formation of the thickest SiO₂ IL during annealing in Ar-O₂-Ar ambient would assist in enhancing the electric breakdown field of this PL with the requirement that a high-quality SiO₂ IL was formed. However, the attainment of the poorest J- E characteristic by this Ga_xCe_yO_z PL annealed in Ar-O₂-Ar ambient has suggested that a low quality of SiO₂ IL was formed contributing to a degradation in E_B of this PL.

4. Conclusion

A comparison study among Ga_xCe_yO_z PL annealed in different ambient (N₂-O₂-N₂, FG-O₂-FG, and Ar-O₂-Ar) was carried out systematically. Hydrogen present in the FG-O₂-FG ambient would attach to oxygen vacancies in the Ga_xCe_yO_z PL, stimulating diffusion of nitrogen from the ambient to the Ga_xCe_yO_z/Si interface. Effectiveness of nitrogen and/or hydrogen in reducing the formation of SiO₂ IL was reported. The extracted E^D and E^ID values (3.87 to 4.07 and 2.99 to 3.68 eV, respectively) were within the range of reported values for CeO₂ and Ga₂O₃, which further reinstated the formation of the ternary Ga_xCe_yO_z phase. The Ga_xCe_yO_z PL annealed in FG-O₂-FG ambient possessed the largest k value of 18.24, followed by the samples annealed Ar-O₂-Ar (15.94) and N₂-O₂-N₂ (11.27) ambient. It was revealed that N₂-O₂-N₂ ambient was effective in passivating Si dangling bonds and oxygen vacancies in addition to the accumulation of nitrogen at the interface, which resulted in the acquisition of a lower D_it, Q_eff, and STD. However, a higher concentration of unattached nitrogen in the Ga_xCe_yO_z PL annealed in FG-O₂-FG has degraded the interface quality when compared with the sample annealed in N₂-O₂-N₂ ambient. Moreover, annealing in Ar-O₂-Ar ambient has led to the acquisition of the lowest E_B (2.71 MV/cm) due to the attainment of the highest Q_eff, STD, D_it, and N_ss. Hence, it was perceived that a low quality of SiO₂ IL was formed during annealing in Ar-O₂-Ar ambient contributing to the attainment of the lowest E_B. Although the Ga_xCe_yO_z PL annealed in N₂-O₂-N₂ ambient has a lower k value than other samples, the acquisition of a high E_B of 4.95 MV/cm could be associated to the acquisition of low Q_eff, STD, N_ss, and D_it, which have supported its compatibility as a potential candidate as a high-k PL for future Si-based MOS applications.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors would like to acknowledge the financial support from the Ministry of Higher Education Malaysia for Fundamental Research Grant Scheme (FRGS) with Project Code FRGS/1/2023/STG05/USM/02/8.

Open Research

Data Availability

The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

References

1 Ajayan J., Nirmal D., and Shubham T., Nanosheet field effect transistors-a next generation device to keep Moore’s law alive: an intensive study, Microelectronics Journal. (2021) 114, article 105141, https://doi.org/10.1016/j.mejo.2021.105141.
10.1016/j.mejo.2021.105141
Web of Science® Google Scholar
2 Ranuárez J. C., Deen M. J., and Chen C. H., A review of gate tunneling current in MOS devices, Microelectronics and Reliability. (2006) 46, no. 12, 1939–1956, https://doi.org/10.1016/j.microrel.2005.12.006, 2-s2.0-33748925385.
10.1016/j.microrel.2005.12.006
Web of Science® Google Scholar
3 Wu E. Y., Suñé J., Lai W., Vayshenker A., Nowak E., and Harmon D., Critical reliability challenges in scaling SiO₂-based dielectric to its limit, Microelectronics Reliability. (2003) 43, no. 8, 1175–1184, https://doi.org/10.1016/S0026-2714(03)00169-0, 2-s2.0-0042164631.
10.1016/S0026-2714(03)00169-0
CAS Web of Science® Google Scholar
4 Schlom D. G., Guha S., and Datta S., Gate oxides beyond SiO₂, MRS Bulletin. (2008) 33, no. 11, 1017–1025, https://doi.org/10.1557/mrs2008.221, 2-s2.0-56849087931.
10.1557/mrs2008.221
CAS Web of Science® Google Scholar
5 Sánchez Ahumada D., In-situ low-temperature synthesis of PS-ZrO₂ hybrid films and their characterization for high-k gate dielectric application, Progress in Organic Coating. (2021) 154, article 106188, https://doi.org/10.1016/j.porgcoat.2021.106188.
10.1016/j.porgcoat.2021.106188
Web of Science® Google Scholar
6 Razavieh A., Zeitzoff P., and Nowak E. J., Challenges and limitations of CMOS scaling for FinFET and beyond architectures, IEEE Transactions on Nanotechnology. (2019) 18, 999–1004, https://doi.org/10.1109/TNANO.2019.2942456, 2-s2.0-85073221238.
10.1109/TNANO.2019.2942456
CAS Web of Science® Google Scholar
7 Peng Y. and Wenwu X., Memory behavior of an Al₂O₃ gate dielectric non-volatile field-effect transistor, IEEE Electron Device Letters. (2020) 41, no. 9, 1340–1343, https://doi.org/10.1109/LED.2020.3010363.
10.1109/LED.2020.3010363
CAS Web of Science® Google Scholar
8 Abdul Shekkeer K. M., Cheong K. Y., and Quah H. J., Effects of post-deposition annealing of cerium oxide passivation layer in nitrogen-oxygen-nitrogen ambient, International Journal of Energy Research. (2022) 46, no. 11, 14814–14826, https://doi.org/10.1002/er.8184.
10.1002/er.8184
CAS Web of Science® Google Scholar
9 Pan T. M. and Huang W. S., Physical and electrical characteristics of a high-k Yb₂O₃ gate dielectric, Applied Surface Science. (2009) 255, no. 9, 4979–4982, https://doi.org/10.1016/j.apsusc.2008.12.048, 2-s2.0-60249085426.
10.1016/j.apsusc.2008.12.048
CAS Web of Science® Google Scholar
10 Seon J. B., Cho N. K., Yoo G., Kim Y. S., and Char K., Solution-processed amorphous ZrO₂ gate dielectric films synthesized by a non-hydrolytic sol-gel route, RSC Advances. (2018) 8, no. 68, 39115–39119, https://doi.org/10.1039/C8RA06911E, 2-s2.0-85057176202, 35558330.
10.1039/C8RA06911E
CAS PubMed Web of Science® Google Scholar
11 Ding M. and Liu X., Damage effect of hafnium oxide gate dielectric based metal-oxide-semiconductor structure under gamma-ray irradiation, AIP Advances. (2021) 11, no. 6, article 065304, https://doi.org/10.1063/5.0048080.
10.1063/5.0048080
Web of Science® Google Scholar
12 Pan T. M., der Lee J., Shu W. H., and te Chen T., Structural and electrical properties of neodymium oxide high-k gate dielectrics, Applied Physics Letters. (2006) 89, no. 23, article 232908, https://doi.org/10.1063/1.2402237, 2-s2.0-33845397958.
10.1063/1.2402237
Web of Science® Google Scholar
13 Cheng J. B., Li A., Shao Q. Y., Ling H. Q., Wu D., Wang Y., Bao Y. J., Wang M., Liu Z. G., and Ming N. B., Growth and characteristics of La₂O₃ gate dielectric prepared by low pressure metalorganic chemical vapor deposition, Applied Surface Science. (2004) 233, no. 1-4, 91–98, https://doi.org/10.1016/j.apsusc.2004.03.232, 2-s2.0-2942672871.
10.1016/j.apsusc.2004.03.232
CAS Web of Science® Google Scholar
14 Yan L., Lopez C. M., Shrestha R. P., Irene E. A., Suvorova A. A., and Saunders M., Magnesium oxide as a candidate high-κ gate dielectric, Applied Physics Letters. (2006) 88, no. 14, article 142901, https://doi.org/10.1063/1.2191419, 2-s2.0-33646707862.
10.1063/1.2191419
PubMed Web of Science® Google Scholar
15 Wang J. J., Fang Z. B., Ji T., Ren W. Y., Zhu Y. Y., and He G., Band offsets of epitaxial Tm₂O₃ high-k dielectric films on Si substrates by X-ray photoelectron spectroscopy, Applied Surface Science. (2012) 258, no. 16, 6107–6110, https://doi.org/10.1016/j.apsusc.2012.03.013, 2-s2.0-84862820372.
10.1016/j.apsusc.2012.03.013
CAS Web of Science® Google Scholar
16 Tan K. Y., Megat Abdul Hedei P. H., Mohd Zabidi A. R., Ahmad F. H., Abdul Shekkeer K. M., Lau K. S., Lim W. F., Hassan Z., and Quah H. J., Comparative studies of metal-organic decomposed Ga_xCe_yO_z and CeO₂ based functional MOS capacitor, International Journal of Energy Research. (2021) 45, no. 12, 18257–18261, https://doi.org/10.1002/er.6952.
10.1002/er.6952
CAS Web of Science® Google Scholar
17 Yu T., Lim B., and Xia Y., Aqueous-phase synthesis of single-crystal ceria nanosheets, Angewandte Chemie. (2010) 122, no. 26, 4586–4589, https://doi.org/10.1002/ange.201001521.
10.1002/ange.201001521
Google Scholar
18 Chiu F. C. and Chang S. H., Electrical analyses of charge trapping and stress-induced leakage current in CeO₂ gate dielectric, Journal of the Chinese Institute of Engineers. (2014) 37, no. 3, 407–412, https://doi.org/10.1080/02533839.2013.799927, 2-s2.0-84893963385.
10.1080/02533839.2013.799927
CAS Web of Science® Google Scholar
19 Chiu F. C., Chen S. Y., Chen C. H., Chen H. W., Huang H. S., and Hwang H. L., Interfacial and electrical characterization in metal–oxide–semiconductor field-effect transistors with CeO₂ gate dielectric, Japanese Journal of Applied Physics. (2009) 48, no. 4S, article 04C014, https://doi.org/10.1143/JJAP.48.04C014, 2-s2.0-72449128393.
10.1143/JJAP.48.04C014
Web of Science® Google Scholar
20 Chiu F. C. and Lai C. M., Optical and electrical characterizations of cerium oxide thin films, Journal of Physics D: Applied Physics. (2010) 43, no. 7, article 075104, https://doi.org/10.1088/0022-3727/43/7/075104.
10.1088/0022-3727/43/7/075104
Web of Science® Google Scholar
21 Siddiqui A., Khosa R. Y., and Usman M., High-k dielectrics for 4H-silicon carbide: present status and future perspectives, Journal of Materials Chemistry C. (2021) 9, no. 15, 5055–5081, https://doi.org/10.1039/D0TC05008C.
10.1039/D0TC05008C
CAS Web of Science® Google Scholar
22 Wang J. C., Hung Y. P., Lee C. L., and Lei T. F., Improved characteristics of ultrathin CeO₂ by using postnitridation annealing, Journal of the Electrochemical Society. (2004) 151, no. 2, https://doi.org/10.1149/1.1640630, 2-s2.0-1242264931.
10.1149/1.1640630
Web of Science® Google Scholar
23 Kumar K. D., Kumar Y. A., Ramachandran T., Al-Kahtani A. A., and Kang M., Cactus-Like Ni-Co/CoMn₂O₄ composites on Ni foam: unveiling the potential for advanced electrochemical materials for pseudocapacitors, Materials Science and Engineering: B. (2023) 296, article 116715, https://doi.org/10.1016/j.mseb.2023.116715.
10.1016/j.mseb.2023.116715
Web of Science® Google Scholar
24 Kumar Y. A., Raorane C. J., Hegazy H. H., Ramachandran T., Kim S. C., and Moniruzzaman M., 2D MXene-based supercapacitors: a promising path towards high-performance energy storage, Journal of Energy Storage. (2023) 72, article 108433, https://doi.org/10.1016/j.est.2023.108433.
10.1016/j.est.2023.108433
Web of Science® Google Scholar
25 Ling Y., Liao X., and Liu C., Poly(l-lactide-co-caprolactone)/tussah silk fibroin nanofiber vascular scaffolds with small diameter fabricated by core-spun electrospinning technology, Journal of Materials Science. (2020) 55, no. 16, 7106–7119, https://doi.org/10.1007/s10853-020-04510-z.
10.1007/s10853-020-04510-z
Web of Science® Google Scholar
26 Zhang L., Sun L., Meng Q., Wu J., Hao X., Zhai S., Dou W., Jia Y., and Zhou M., Strain-engineered formation, migration, and electronic properties of polaronic defects in CeO₂, Physica Status Solidi (b). (2021) 258, no. 6, article 2100020, https://doi.org/10.1002/pssb.202100020.
10.1002/pssb.202100020
Web of Science® Google Scholar
27 Mamedov D. and Karazhanov S. Z., Doping-induced modulation of electronic, optical and wetting properties of CeO₂, Journal of Physics and Chemistry of Solids. (2022) 168, article 110820, https://doi.org/10.1016/j.jpcs.2022.110820.
10.1016/j.jpcs.2022.110820
Web of Science® Google Scholar
28 Machida M., Murata Y., Kishikawa K., Zhang D., and Ikeue K., On the reasons for high activity of CeO₂ catalyst for soot oxidation, Chemistry of Materials. (2008) 20, no. 13, 4489–4494, https://doi.org/10.1021/cm800832w, 2-s2.0-47749144374.
10.1021/cm800832w
CAS Web of Science® Google Scholar
29 Brik M. G., Ma C. G., Piasecki M., and Suchocki A., Locating impurity and defect levels in the host band gap by first-principles calculations: pure and Ce^3+-doped YAlO₃, Optical Materials. (2021) 113, article 110843, https://doi.org/10.1016/j.optmat.2021.110843.
10.1016/j.optmat.2021.110843
Web of Science® Google Scholar
30 Choudhury B. and Choudhury A., Ce³⁺ and oxygen vacancy mediated tuning of structural and optical properties of CeO₂ nanoparticles, Materials Chemistry and Physics. (2012) 131, no. 3, 666–671, https://doi.org/10.1016/j.matchemphys.2011.10.032, 2-s2.0-83555163720.
10.1016/j.matchemphys.2011.10.032
CAS Web of Science® Google Scholar
31 Chen Y. C., Chen K. B., Lee C. S., and Lin M. C., Direct synthesis of Zr-doped ceria nanotubes, Journal of Physical Chemistry C. (2009) 113, no. 13, 5031–5034, https://doi.org/10.1021/jp810492s, 2-s2.0-65249190787.
10.1021/jp810492s
CAS Web of Science® Google Scholar
32 Qin W., Xu L., Song J., Xing R., and Song H., Highly enhanced gas sensing properties of porous SnO₂-CeO₂ composite nanofibers prepared by electrospinning, Sensors and Actuators B: Chemical. (2013) 185, 231–237, https://doi.org/10.1016/j.snb.2013.05.001, 2-s2.0-84878367660.
10.1016/j.snb.2013.05.001
CAS Web of Science® Google Scholar
33 Rocchini E., Trovarelli A., Llorca J., Graham G. W., Weber W. H., Maciejewski M., and Baiker A., Relationships between structural/morphological modifications and oxygen storage-redox behavior of silica-doped ceria, Journal of Catalysis. (2000) 194, no. 2, 461–478, https://doi.org/10.1006/jcat.2000.2954, 2-s2.0-0343278994.
10.1006/jcat.2000.2954
CAS Web of Science® Google Scholar
34 Fu M., Wei L., Li Y., Zhou X., Hao S., and Li Y., Surface charge tuning of ceria particles by titanium doping: towards significantly improved polishing performance, Solid State Sciences. (2009) 11, no. 12, 2133–2137, https://doi.org/10.1016/j.solidstatesciences.2009.08.014, 2-s2.0-70450245140.
10.1016/j.solidstatesciences.2009.08.014
CAS Web of Science® Google Scholar
35 Scheffe J. R., Jacot R., Patzke G. R., and Steinfeld A., Synthesis, characterization, and thermochemical redox performance of Hf⁴⁺, Zr⁴⁺, and Sc³⁺ doped ceria for splitting CO₂, Journal of Physical Chemistry C. (2013) 117, no. 46, 24104–24114, https://doi.org/10.1021/jp4050572, 2-s2.0-84888613450.
10.1021/jp4050572
CAS Web of Science® Google Scholar
36 Andersson D. A., Simak S. I., Skorodumova N. V., Abrikosov I. A., and Johansson B., Theoretical study of CeO₂ doped with tetravalent ions, Physical Review B. (2007) 76, no. 17, article 174119, https://doi.org/10.1103/PhysRevB.76.174119, 2-s2.0-36648998537.
10.1103/PhysRevB.76.174119
Web of Science® Google Scholar
37 Tang Y., Zhang H., Cui L., Ouyang C., Shi S., Tang W., Li H., Lee J. S., and Chen L., First-principles investigation on redox properties of M-dopedCeO2(M=Mn,Pr,Sn,Zr), Physical Review B. (2010) 82, no. 12, article 125104, https://doi.org/10.1103/PhysRevB.82.125104, 2-s2.0-77957745032.
10.1103/PhysRevB.82.125104
Web of Science® Google Scholar
38 Manibalan G., Murugadoss G., Thangamuthu R., Kumar M. R., and Kumar R. M., Facile synthesis of CeO₂-SnO₂ nanocomposite for electrochemical determination of L-cysteine, Journal of Alloys and Compounds. (2019) 792, 1150–1161, https://doi.org/10.1016/j.jallcom.2019.04.127, 2-s2.0-85064396243.
10.1016/j.jallcom.2019.04.127
CAS Web of Science® Google Scholar
39 Keating P. R. L., Scanlon D. O., and Watson G. W., Computational testing of trivalent dopants in CeO₂ for improved high-κ dielectric behaviour, Journal of Materials Chemistry C. (2013) 1, no. 6, 1093–1098, https://doi.org/10.1039/C2TC00385F, 2-s2.0-84876917303.
10.1039/C2TC00385F
CAS Web of Science® Google Scholar
40 Shehata N., Meehan K., Hudait M., and Jain N., Control of oxygen vacancies and Ce⁺³ concentrations in doped ceria nanoparticles via the selection of lanthanide element, Journal of Nanoparticle Research. (2012) 14, no. 10, 1–10, https://doi.org/10.1007/s11051-012-1173-1, 2-s2.0-84865959088.
10.1007/s11051-012-1173-1
PubMed Web of Science® Google Scholar
41 Minervini L., Zacate M. O., and Grimes R. W., Defect cluster formation in M₂O₃-doped CeO₂, Solid State Ionics. (1999) 116, no. 3-4, 339–349, https://doi.org/10.1016/S0167-2738(98)00359-2.
10.1016/S0167-2738(98)00359-2
CAS Web of Science® Google Scholar
42 Gerhardt Anderson R. and Nowick A. S., Ionic conductivity of CeO₂ with trivalent dopants of different ionic radii, Solid State Ionics. (1981) 5, 547–550, https://doi.org/10.1016/0167-2738(81)90313-1, 2-s2.0-0019621205.
10.1016/0167-2738(81)90313-1
CAS Web of Science® Google Scholar
43 Kilic M. E., Lee J. H., and Lee K. R., Oxygen ion transport in doped ceria: effect of vacancy trapping, Journal of Materials Chemistry A. (2021) 9, no. 24, 13883–13889, https://doi.org/10.1039/D1TA00926E.
10.1039/D1TA00926E
CAS Web of Science® Google Scholar
44 Gerhardt R., Lee W., and Nowick A. S., Anelastic and dielectric relaxation of Scandia-doped ceria, Journal of Physics and Chemistry of Solids. (1987) 48, no. 6, 563–569, https://doi.org/10.1016/0022-3697(87)90053-9, 2-s2.0-0023213155.
10.1016/0022-3697(87)90053-9
CAS Web of Science® Google Scholar
45 Shekkeer K. M. A., Cheong K. Y., and Quah H. J., Growth of quaternary Ga_xCe_1-xO_yN_z passivation layer for silicon based metal-oxide-semiconductor capacitor, Materials Chemistry and Physics. (2022) 290, article 126549, https://doi.org/10.1016/j.matchemphys.2022.126549.
10.1016/j.matchemphys.2022.126549
Web of Science® Google Scholar
46 Patwe S. J. and Tyagi A. K., Solubility of Ce⁴⁺ and Sr²⁺ in the pyrochlore lattice of Gd₂Zr₂O₇ for simulation of Pu and alkaline earth metal, Ceramics International. (2006) 32, no. 5, 545–548, https://doi.org/10.1016/j.ceramint.2005.04.009, 2-s2.0-33646018631.
10.1016/j.ceramint.2005.04.009
CAS Web of Science® Google Scholar
47 Li Z. P., Mori T., Zou J., and Drennan J., Defects clustering and ordering in di- and trivalently doped ceria, Materials Research Bulletin. (2013) 48, no. 2, 807–812, https://doi.org/10.1016/j.materresbull.2012.11.073, 2-s2.0-84871719340.
10.1016/j.materresbull.2012.11.073
CAS Web of Science® Google Scholar
48 Dakhel A. A., Investigation of opto-dielectric properties of Ti-doped Ga₂O₃ thin films, Solid State Sciences. (2013) 20, 54–58, https://doi.org/10.1016/j.solidstatesciences.2013.03.009, 2-s2.0-84876153188.
10.1016/j.solidstatesciences.2013.03.009
CAS Web of Science® Google Scholar
49 Onuma T., Fujioka S., Yamaguchi T., Higashiwaki M., Sasaki K., Masui T., and Honda T., Correlation between blue luminescence intensity and resistivity in β-Ga₂O₃ single crystals, Applied Physics Letters. (2013) 103, no. 4, article 041910, https://doi.org/10.1063/1.4816759, 2-s2.0-84885012509.
10.1063/1.4816759
Web of Science® Google Scholar
50 Li X., Lu H., and Yang J., Chemical, optical, and electrical characterization of Ga₂O₃ thin films grown by plasma-enhanced atomic layer deposition, Current Applied Physics. (2019) 19, no. 2, 72–81, https://doi.org/10.1016/j.cap.2018.11.013, 2-s2.0-85057474240.
10.1016/j.cap.2018.11.013
Web of Science® Google Scholar
51 Abdul Shekkeer K. M., Cheong K. Y., and Quah H. J., Alteration in growth temperatures of metal-organic decomposed Ga_xCe_1-xO_yN_z passivation layer in nitrogen/oxygen/nitrogen ambient, Ceramics International. (2023) 49, no. 9, 14760–14770, https://doi.org/10.1016/j.ceramint.2023.01.072.
10.1016/j.ceramint.2023.01.072
CAS Web of Science® Google Scholar
52 Yang Y. L., Cheng C. Y., and Yeh W. K., Reliability improvement of 28-nm High-k/metal gate-last MOSFET using appropriate oxygen annealing, IEEE Electron Device Letters. (2012) 33, no. 8, 1183–1185, https://doi.org/10.1109/LED.2012.2202089, 2-s2.0-84864429933.
10.1109/LED.2012.2202089
Web of Science® Google Scholar
53 Tan S. Y., Control of interface traps in HfO₂ gate dielectric on silicon, Journal of Electronic Materials. (2010) 39, no. 11, 2435–2440, https://doi.org/10.1007/s11664-010-1323-0, 2-s2.0-78149283819.
10.1007/s11664-010-1323-0
CAS Web of Science® Google Scholar
54 Stahlbush R. E., Hughes H. L., and Krull W. A., Reduction of charge trappings and electron tunneling in SIMOX by supplemental implantation of oxygen, IEEE Transactions on Nuclear Science. (1993) 40, no. 6, 1740–1747, https://doi.org/10.1109/23.273484, 2-s2.0-0027851453.
10.1109/23.273484
CAS Web of Science® Google Scholar
55 Dimitrova T., Arshak K., and Atanassova E., Crystallization effects in oxygen annealed Ta₂O₅ thin films on Si, Thin Solid Films. (2001) 381, no. 1, 31–38, https://doi.org/10.1016/S0040-6090(00)01569-8, 2-s2.0-0035155840.
10.1016/S0040-6090(00)01569-8
CAS Web of Science® Google Scholar
56 Atanassova E., Novkovski N., Paskaleva A., and Pecovska-Gjorgjevich M., Oxygen annealing modification of conduction mechanism in thin rf sputtered Ta₂O₅ on Si, Solid State Electronics. (2002) 46, no. 11, 1887–1898, https://doi.org/10.1016/S0038-1101(02)00134-X, 2-s2.0-0036839485.
10.1016/S0038-1101(02)00134-X
CAS Web of Science® Google Scholar
57 Yang H., Wang Y. Q., Wang H., and Jia Q. X., Oxygen concentration and its effect on the leakage current in BiFeO₃ thin films, Applied Physics Letters. (2010) 96, no. 1, article 012909, https://doi.org/10.1063/1.3291044, 2-s2.0-75749108208.
10.1063/1.3291044
Web of Science® Google Scholar
58 Tagui K., Nakamura K., Ogawa M., Saito K., and Suzuki S., The Electrical Property of CeO₂ Films Deposited by MOCVD on Si(100), Electrochemical and Solid-State Letters. (2007) 10, no. 7, article D73, https://doi.org/10.1149/1.2733802, 2-s2.0-34248506790.
10.1149/1.2733802
Web of Science® Google Scholar
59 Tye L., El-Masry N. A., Chikyow T., McLarty P., and Bedair S. M., Electrical characteristics of epitaxial CeO₂ on Si(111), Applied Physics Letters. (1994) 65, no. 24, 3081–3083, https://doi.org/10.1063/1.112467, 2-s2.0-0000051481.
10.1063/1.112467
CAS Web of Science® Google Scholar
60 Qu M., Chang C. H., Meng T., Zhang Q., Liu P. T., and Shieh H. P. D., Stability study of indium tungsten oxide thin-film transistors annealed under various ambient conditions, Physica Status Solidi (a). (2017) 214, no. 2, article 1600465, https://doi.org/10.1002/pssa.201600465, 2-s2.0-85007105157.
10.1002/pssa.201600465
Web of Science® Google Scholar
61 Lin Y.-S., Puthenkovilakam R., Chang J. P., Bouldin C., Levin I., Nguyen N. V., Ehrstein J., Sun Y., Pianetta P., Conard T., Vandervorst W., Venturo V., and Selbrede S., Interfacial properties of ZrO₂ on silicon, Journal of Applied Physics. (2003) 93, no. 10, 5945–5952, https://doi.org/10.1063/1.1563844, 2-s2.0-0037974737.
10.1063/1.1563844
CAS Web of Science® Google Scholar
62 Kim H. D., Jeong S.-W., You M. T., and Roh Y., Effects of annealing gas (N₂, N₂O, O₂) on the characteristics of ZrSi_xO_y/ZrO₂ high-k gate oxide in MOS devices, Thin Solid Films. (2006) 515, no. 2, 522–525, https://doi.org/10.1016/j.tsf.2005.12.287, 2-s2.0-33748743891.
10.1016/j.tsf.2005.12.287
CAS Web of Science® Google Scholar
63 Zoolfakar A. S., Study of zirconium dioxide (ZrO₂) dielectric charges, Materials Research Innovations. (2009) 13, no. 3, 161–164, https://doi.org/10.1179/143307509X437509, 2-s2.0-70149086320.
10.1179/143307509X437509
CAS Web of Science® Google Scholar
64 Simoen E., Rothschild A., Vermang B., Poortmans J., and Mertens R., Impact of forming gas annealing and firing on the Al₂O₃/p-Si interface state spectrum, Electrochemical and Solid-State Letters. (2011) 14, no. 9, H362–H364, https://doi.org/10.1149/1.3597661, 2-s2.0-79960268771.
10.1149/1.3597661
CAS Web of Science® Google Scholar
65 Yamada N. and Yamada-Kaneta H., Drastic improvements of gate oxide reliability by argon annealing compared with hydrogen annealing, Journal of the Electrochemical Society. (1998) 145, no. 10, 3628–3631, https://doi.org/10.1149/1.1838854, 2-s2.0-0032187416.
10.1149/1.1838854
CAS Web of Science® Google Scholar
66 Quah H. J. and Cheong K. Y., Deposition and post-deposition annealing of thin Y₂O₃ film on n-type Si in argon ambient, Materials Chemistry and Physics. (2011) 130, no. 3, 1007–1015, https://doi.org/10.1016/j.matchemphys.2011.08.024, 2-s2.0-80054017643.
10.1016/j.matchemphys.2011.08.024
CAS Web of Science® Google Scholar
67 Kumar Y. A., Koyyada G., Ramachandran T., Kim J. H., Sajid S., Moniruzzaman M., Alzahmi S., and Obaidat I. M., Carbon materials as a conductive skeleton for supercapacitor electrode applications: a review, Nanomaterials. (2023) 13, no. 6, https://doi.org/10.3390/nano13061049, 36985942.
10.3390/nano13061049
PubMed Web of Science® Google Scholar
68 Kumar Y. A., Koyyada G., Ramachandran T., Kim J. H., Hegazy H. H., Singh S., and Moniruzzaman M., Recent advancement in quantum dot-based materials for energy storage applications: a review, Dalton Transactions. (2023) 52, no. 25, 8580–8600, https://doi.org/10.1039/D3DT00325F, 37096427.
10.1039/D3DT00325F
PubMed Web of Science® Google Scholar
69 Tan K. Y. and Quah H. J., Growth of metal-organic decomposed ternary Ga_xCe_yO_z films by nitrogen-infused wet oxidation for metal-oxide-semiconductor capacitor, International Journal of Energy Research. (2022) 46, no. 5, 5756–5770, https://doi.org/10.1002/er.7520.
10.1002/er.7520
CAS Web of Science® Google Scholar
70 Ramasamy V. and Vijayalakshmi G., Effect of Zn doping on structural, optical and thermal properties of CeO₂ nanoparticles, Superlattices and Microstructures. (2015) 85, 510–521, https://doi.org/10.1016/j.spmi.2015.05.015, 2-s2.0-84935860038.
10.1016/j.spmi.2015.05.015
CAS Web of Science® Google Scholar
71 Patel K. J., Panchal C. J., Desai M. S., and Mehta P. K., An investigation of the insertion of the cations H⁺, Na⁺, K⁺ on the electrochromic properties of the thermally evaporated WO₃ thin films grown at different substrate temperatures, Materials Chemistry and Physics. (2010) 124, no. 1, 884–890, https://doi.org/10.1016/j.matchemphys.2010.08.021, 2-s2.0-77957231778.
10.1016/j.matchemphys.2010.08.021
CAS Web of Science® Google Scholar
72 Chen J., Shen S., Wu P., and Guo L., Nitrogen-doped CeO_x nanoparticles modified graphitic carbon nitride for enhanced photocatalytic hydrogen production, Green Chemistry. (2015) 17, no. 1, 509–517, https://doi.org/10.1039/C4GC01683A, 2-s2.0-84948757493.
10.1039/C4GC01683A
CAS Web of Science® Google Scholar
73 Muller R. S. and Theodore T. I., Device electronics for integrated circuits, 2002, John Wiley & Sons.
Google Scholar
74 Gupta M., Kumar A., Sagdeo A., and Sagdeo P. R., Doping-induced combined fano and phonon confinement effect in La-doped CeO₂: Raman spectroscopy analysis, Journal of Physical Chemistry C. (2021) 125, no. 4, 2648–2658, https://doi.org/10.1021/acs.jpcc.0c09133.
10.1021/acs.jpcc.0c09133
CAS Web of Science® Google Scholar
75 Suresh R., Ponnuswamy V., Mariappan R., and Senthil Kumar N., Influence of substrate temperature on the properties of CeO₂ thin films by simple nebulizer spray pyrolysis technique, Ceramics International. (2014) 40, no. 1, 437–445, https://doi.org/10.1016/j.ceramint.2013.06.020, 2-s2.0-84887624488.
10.1016/j.ceramint.2013.06.020
CAS Web of Science® Google Scholar
76 Sane P. K., Tambat S., Sontakke S., and Nemade P., Visible light removal of reactive dyes using CeO₂ synthesized by precipitation, Journal of Environmental Chemical Engineering. (2018) 6, no. 4, 4476–4489, https://doi.org/10.1016/j.jece.2018.06.046, 2-s2.0-85049248849.
10.1016/j.jece.2018.06.046
CAS Web of Science® Google Scholar
77 Tatar B., Sam E. D., Kutlu K., and Ürgen M., Synthesis and optical properties of CeO₂ nanocrystalline films grown by pulsed electron beam deposition, Journal of Materials Science. (2008) 43, no. 15, 5102–5108, https://doi.org/10.1007/s10853-008-2750-7, 2-s2.0-47849088999.
10.1007/s10853-008-2750-7
CAS Web of Science® Google Scholar
78 Mao C., Zhao Y., Qiu X., Zhu J., and Burda C., Synthesis, characterization and computational study of nitrogen-doped CeO₂ nanoparticles with visible-light activity, Physical Chemistry Chemical Physics. (2008) 10, no. 36, 5633–5638, https://doi.org/10.1039/b805915b, 2-s2.0-52449090739, 18956099.
10.1039/b805915b
CAS PubMed Web of Science® Google Scholar
79 Islam M. M., Saha J. K., and Bukke R. N., Solution-processed La alloyed ZrO_x high-k dielectric for high-performance ZnO thin-film transistors, IEEE Electron Device Letters. (2020) 41, 1021–1024.
CAS Web of Science® Google Scholar
80 Zhu Y. Y., Fang Z. B., and Tan Y. S., Structural and electrical characteristics of amorphous ErAlO gate dielectric films, Chinese Physics Letters. (2012) 29, no. 8, article 087701, https://doi.org/10.1088/0256-307X/29/8/087701, 2-s2.0-84864925570.
10.1088/0256-307X/29/8/087701
Web of Science® Google Scholar
81 Fukuda H., Miura M., Sakuma S., and Nomura S., Structural and electrical properties of crystalline CeO₂ films formed by metalorganic decomposition, Japanese Journal of Applied Physics. (1998) 37, no. 7R, 4158–4161, https://doi.org/10.1143/JJAP.37.4158.
10.1143/JJAP.37.4158
CAS Web of Science® Google Scholar
82 Pascu R., Romanitan C., Varasteanu P., and Kusko M., A reliable technology for advanced SiC-MOS devices based on fabrication of high quality silicon oxide layers by converting a-Si, IEEE Journal of the Electron Devices Society. (2019) 7, 158–167, https://doi.org/10.1109/JEDS.2018.2886373, 2-s2.0-85058630075.
10.1109/JEDS.2018.2886373
CAS Web of Science® Google Scholar
83 Li S., Wu Y., Li G., Yu H., Fu K., Wu Y., and Li X., Ta-doped modified Gd₂O₃ film for a novel high k gate dielectric, Journal of Materials Science. (2019) 35, 2305–2311, https://doi.org/10.1016/j.jmst.2019.05.028, 2-s2.0-85069961840.
10.1016/j.jmst.2019.05.028
CAS Web of Science® Google Scholar
84 Gao J., He G., Zhang J. W., Deng B., and Liu Y. M., Annealing temperature modulated interfacial chemistry and electrical characteristics of sputtering-derived HfO₂/Si gate stack, Journal of Alloys and Compounds. (2015) 647, 322–330, https://doi.org/10.1016/j.jallcom.2015.05.157, 2-s2.0-84933500128.
10.1016/j.jallcom.2015.05.157
CAS Web of Science® Google Scholar
85 Quah H. J., Hassan Z., and Lim W. F., Passivation of silicon substrate using two-step grown ternary aluminium doped zirconium oxide, Applied Surface Science. (2019) 493, 411–422, https://doi.org/10.1016/j.apsusc.2019.07.023, 2-s2.0-85068562352.
10.1016/j.apsusc.2019.07.023
CAS Web of Science® Google Scholar
86 Schroder D. K., Semiconductor Material and Device Characterization, 2005, John Wiley & Sons, Inc., https://doi.org/10.1002/0471749095, 2-s2.0-84890769835.
10.1002/0471749095
Google Scholar
87 Belosludtsev A., Yakimov Y., Mroczyński R., Stanionytė S., Skapas M., Buinovskis D., and Kyžas N., Effect of annealing on optical, mechanical, electrical properties and structure of scandium oxide films, Physica Status Solidi (a). (2019) 216, no. 18, 1900122–1900129, https://doi.org/10.1002/pssa.201900122, 2-s2.0-85069804292.
10.1002/pssa.201900122
Web of Science® Google Scholar
88 Gao J., He G., Zhang J., Chen X., Jin P., and Sun Z., Annealing temperature dependent electrical properties and leakage current transport mechanisms in atomic layer deposition-derived Al₂O₃-incorporated HfO₂/Si gate stack, Journal of Nanoscience and Nanotechnology. (2016) 16, no. 8, 8075–8082, https://doi.org/10.1166/jnn.2016.12791, 2-s2.0-84978716188.
10.1166/jnn.2016.12791
CAS Web of Science® Google Scholar
89 Lee E. C., Nitrogen-induced interface defects in Si oxynitride, Physical Review B. (2008) 77, no. 10, article 104108, https://doi.org/10.1103/PhysRevB.77.104108, 2-s2.0-41449101420.
10.1103/PhysRevB.77.104108
Web of Science® Google Scholar
90 Ko L. H., Cho B. J., Nga Y. B., and Chan L. H., The effect of nitrogen incorporation into the gate oxide by using shallow implantation of nitrogen and drive-in process, Proceedings 1998 Hong Kong Electron Devices Meeting (Cat. No.98TH8368), 1998, Hong Kong, China, 32–35, https://doi.org/10.1109/HKEDM.1998.740182, 2-s2.0-85051770567.
10.1109/HKEDM.1998.740182
Google Scholar
91 Yildiz D. E., Apaydin D. H., Kaya E., Altindal S., and Cirpan A., The main electrical and interfacial properties of benzotriazole and fluorene based organic devices, Journal of Macromolecular Science, Part A. (2013) 50, 168–174, https://doi.org/10.1080/10601325.2013.741864, 2-s2.0-84873395646.
10.1080/10601325.2013.741864
CAS Web of Science® Google Scholar
92 Tao F., Yang H., Tang B., Tang Z., Xu Y., Xu J., Wang Q., and Yan J., TDDB characteristic and breakdown mechanism of ultra-thin SiO₂/HfO₂ bilayer gate dielectrics, Journal of Semiconductors. (2014) 35, no. 6, 064003–064009, https://doi.org/10.1088/1674-4926/35/6/064003, 2-s2.0-84902664227.
10.1088/1674-4926/35/6/064003
CAS Google Scholar

Citing Literature

All articles

Effects of Varying Annealing Ambient towards Performance of Ternary Ga_xCe_yO_z Passivation Layers for Metal-Oxide-Semiconductor Capacitor

Abstract

1. Introduction

2. Experimental

3. Results and Discussions

3.1. Structural, Chemical, and Optical Characteristics of Ga_xCe_yO_z PL Annealed in FG-O₂-FG, N₂-O₂-N₂, and Ar-O₂-Ar Ambient

3.2. The Metal-Oxide-Semiconductor (MOS) Characteristics of the Al/Ga_xCe_yO_z/4H-SiC/Al MOS Capacitor

4. Conclusion

Conflicts of Interest

Acknowledgments

Open Research

Data Availability

References

Citing Literature

Figures

References

Information

About Wiley Online Library

Help & Support

Opportunities

Connect with Wiley

Effects of Varying Annealing Ambient towards Performance of Ternary GaxCeyOz Passivation Layers for Metal-Oxide-Semiconductor Capacitor

Abstract

1. Introduction

2. Experimental

3. Results and Discussions

3.1. Structural, Chemical, and Optical Characteristics of GaxCeyOz PL Annealed in FG-O2-FG, N2-O2-N2, and Ar-O2-Ar Ambient

3.2. The Metal-Oxide-Semiconductor (MOS) Characteristics of the Al/GaxCeyOz/4H-SiC/Al MOS Capacitor

4. Conclusion

Conflicts of Interest

Acknowledgments

Open Research

Data Availability

References

Citing Literature

Figures

References

Related

Information

Effects of Varying Annealing Ambient towards Performance of Ternary Ga_xCe_yO_z Passivation Layers for Metal-Oxide-Semiconductor Capacitor

3.1. Structural, Chemical, and Optical Characteristics of Ga_xCe_yO_z PL Annealed in FG-O₂-FG, N₂-O₂-N₂, and Ar-O₂-Ar Ambient

3.2. The Metal-Oxide-Semiconductor (MOS) Characteristics of the Al/Ga_xCe_yO_z/4H-SiC/Al MOS Capacitor