Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2020 Oct 9;5(41):26347–26356. doi: 10.1021/acsomega.0c02120

Growth and Characterization of Ternary HfxTayOz Films via Nitrogen-Infused Wet Oxidation

Hock Jin Quah 1,*, Farah Hayati Ahmad 1, Way Foong Lim 1, Zainuriah Hassan 1
PMCID: PMC7581075  PMID: 33110962

Abstract

graphic file with name ao0c02120_0013.jpg

Nitrogen-infused wet oxidation at different temperatures (400–1000 °C) was employed to transform tantalum–hafnia to hafnium-doped tantalum oxide films. High-temperature wet oxidation at 1000 °C marked an onset of crystallization occurring in the film, accompanied with the formation of an interfacial oxide due to a reaction between the inward-diffusing hydroxide ions, which were dissociated from the water molecules during wet oxidation. The existence of nitrogen has assisted in controlling the interfacial oxide formation. However, high-temperature oxidation caused a tendency for the nitrogen to desorb and form N–H complex after reacting with the hydroxide ions. Besides, the presence of N–H complex implied a decrease in the passivation at the oxide–Si interface by hydrogen. As a consequence, defect formation would happen at the interface and influence the metal–oxide–semiconductor characteristics of the samples. In comparison, tantalum–hafnia subjected to nitrogen-infused wet oxidation at 600 °C has obtained the highest dielectric constant, the largest band gap, and the lowest slow trap density.

1. Introduction

For the past few decades, significant interest has been focused toward scaling down of silicon (Si)-based micro- and nanoelectronic metal–oxide–semiconductor (MOS) devices to enhance the speed and performance of the devices. One of the main components of sustaining low leakage current in Si-based MOS devices was attributed to the possibility of growing high-quality silicon dioxide (SiO2) as the passivation layer. Unfortunately, the continuous downscaling of low-dielectric-constant (k) SiO2 throughout the years has pushed SiO2 thickness to a limit, wherein a considerably high leakage current governed by direct tunneling mechanism was ensured when the thickness of SiO2 was reduced down to the region of a few nanometers.1−4 To mitigate the influence governed by direct tunneling mechanism, there is a need of replacing the SiO2 with novel materials composed of high k values, such as Al2O3,55 Y2O3,1,2,6 ZrO2,77 La2O3,8 HfO2,99 Ta2O5,10,1110,11 and CeO2,12,13 as the passivation layer for Si-based MOS devices. Of these high-k materials, Ta2O5 has gained increasing attention, owing to the possibility of being employed as a storage dielectric film and a passivation layer for dynamic random-access memories (DRAM)11 and Si-based MOS devices, respectively. The considerably high k value of Ta2O5 was the key driving force toward its realization as a storage dielectric,14−16 but the k value is not the only criterion for the selection of an appropriate material as the passivation layer.

Utilization of Ta2O5 as the passivation layer in Si-based MOS devices has encountered various shortcomings, which include small conduction band offset with respect to Si (∼0.4 eV), moderate energy band gap (4.0–4.5 eV), as well as thermal instability when in contact with Si surface.17,18 Although it was well known that the formation of interfacial layer was unavoidable for high-k materials deposited on Si substrate, the adoption of the Ta2O5/Si structure has been confronted with the issue related to dissociation of the TaSiOx phase formed at the interface into Ta2O5 and SiOx phases during the high-temperature annealing process.17,18 Consequently, the overall k value for the Ta2O5/Si structure would be decreased due to the formation of a low-k SiOx phase. This shortcoming could be circumvented by means of incorporating hafnium oxide (HfO2) into Ta2O5 lattice to mitigate the existence of SiOx compound in interfacial layer, wherein HfO2 would react with SiOx to form HfSiOx that was deemed to be stable compared to TaSiOx. Previous research work divulged a beneficial effect of forming the HfSiOx interfacial layer17,19 in terms of yielding an increase in leakage current, dielectric breakdown field, and channel carrier mobility due to the presence of HfSiOx to assist in reducing dielectric polarization effects. In addition, the formation of ternary HfxTayOz phase would also take advantage of the fascinating properties of HfO2, such as large band gap (∼5.8 eV) as well as conduction band offset with respect to Si (∼1.4 eV),20 that would ultimately abstain the injection of electrons from semiconductor to overcome the barrier layer between the HfxTayOz passivation layer and Si substrate. Hereafter, the realization of ternary HfxTayOz would ensure the fulfillment of criteria pertaining to large band gap and conduction band offset (>1 eV) that was necessary to function as a passivation layer in Si-based MOS devices.

Besides engineering the band gap of Ta2O5 through incorporation of HfO2, the MOS characteristics of HfxTayOz could be also altered depending on the doping concentration of HfO2 being introduced into the Ta2O5 lattice, which would hence control the phase transformation temperature required for transforming amorphous HfxTa1–xOy to polycrystalline HfxTayOz.2121 Generally, an increase in crystallization temperature from 600 to 800 °C was discerned for the HfxTayOz phase compared to the Ta2O5 phase, wherein the formation of polycrystalline Ta2O5 films at temperatures beyond 600 °C has triggered a deterioration in electrical performance due to the formation of grain boundaries. Nevertheless, postdeposition annealing at 700 °C for 180 min on HfxTayOz films comprising of a higher concentration of Hf has triggered partial crystallization, while the amorphous phase of the as-deposited films with a lower concentration of Hf was preserved.22,23 The influence of partial crystallization of HfxTayOz films on MOS characteristics was ascertained, wherein the amorphous phase of HfxTayOz films with a lower concentration of Hf has revealed the existence of a lower density of fixed charge as well as the ability to sustain a higher dielectric breakdown field. Similarly, it was distinguished in recent times that conserving the amorphous phase of HfxTayOz films after undergoing wet oxidation process at 800 °C would contribute to the acquisition of a lower interface trap density and effective oxide charge.24 Nevertheless, one of the shortcomings encountered by this wet oxidized amorphous HfxTayOz films sputtered using a lower tantalum (Ta) sputtering power was the existence of a higher density of slow traps compared to polycrystalline HfxTayOz films sputtered using a higher tantalum (Ta) sputtering power. The outcome of this recent discovery with regard to the formation of amorphous HfxTayOz films with a lower concentration of Ta was not in accordance with previous findings, which has divulged that the existence of a higher concentration of Ta has hindered the transformation from amorphous to polycrystalline HfxTayOz films.

Therefore, it is of interest in this work to investigate the effects of nitrogen-infused wet oxidation at various temperatures (400, 600, 800,24 and 1000 °C) on the structural, optical, morphological, and MOS characteristics of HfxTayOz films grown using a simultaneous co-sputtering of HfO2 and Ta at 160 and 120 W, respectively. It was envisaged that wet oxidation process would repair broken bonds in the HfxTayOz films by occupying oxygen-related defects as well as passivating interface-related defects through the dissociation of H2O into OH– and H+ ions. In addition, this study would also make use of the advantage of nitrogen, following the reported studies pertaining to interface nitridation method for 4H-SiC-based MOS devices during postoxidation annealing in pure nitrogen gas, in which the presence of a sufficient amount of nitrogen at the SiO2/4H-SiC interface could passivate slow traps at the interface. Depending on the oxidation temperature, the interface nitridation was likely to be enhanced.25 A modification was employed in this work by introducing nitrogen as a carrier gas during the wet oxidation process, which has not been reported thus far to minimize the interfacial layer formation.

2. Experimental Procedures

Two-inch n-type silicon substrate oriented in the (100) plane with a resistivity of 1–10 Ω·cm was diced into smaller pieces of 1 cm × 1 cm size, and these samples were cleaned using the standard Radio Corporation of America (RCA) method. After RCA cleaning, these samples were dipped into diluted hydrofluoric acid solution to etch the native oxide (SiO2) formed on the Si substrate. Immediately, these samples were loaded into a dual RF magnetron co-sputtering machine (Edwards A500) to deposit a layer of thin film composed of a mixture of HfO2 and Ta on the Si substrate. The co-sputtering process was carried out using sputtering powers of 120 and 160 W for Ta and HfO2 sputtering targets, respectively, at a working pressure of 2.3 × 10–2 mbar and an argon gas flow rate of 14 sccm. Before the initiation of the co-sputtering process, the chamber of sputtering system was pumped down to a base pressure of 1 × 10–6 mbar, purged with high-purity argon gas (99.999%), and the plasma cleaning process was performed on both HfO2 and Ta sputtering targets for 5 min. After the completion of the sputtering process, the as-deposited tantalum–hafnia were subjected to the wet oxidation process at different temperatures (400, 600, 800, and 1000 °C) for a dwell time of 30 min in a horizontal tube furnace. The wet oxidation process was carried out by heating the distilled water in a flask at 95 °C, and the generated vapor was carried by nitrogen (N2) carrier gas at a rate of 4 centiliter per minute into the quartz tube. Then, aluminum (Al) top electrode (diameter = 0.08 cm) was thermally evaporated using a shadow mask on the wet oxidized HfxTayOz/Si samples at different temperatures. Finally, the back side of the Si substrate was deposited with a blanket of Al to form Al/HfxTayOz/Si/Al metal–oxide–semiconductor (MOS) test structures.

High-resolution X-ray diffraction (HRXRD; PANalytical X’Pert PRO MRD PW3040) and field emission scanning electron microscopy (FESEM; FEI Nova NanoSEM 450) were used to acquire crystalline phases and orientations as well as cross-sectional images, respectively, for the HfxTayOz films subjected to different wet oxidation temperatures. Atomic force microscopy (AFM; Dimension Edge, Bruker) was employed to obtain three-dimensional surface topographies and root-mean-square (RMS) roughness of the investigated films. Fourier transform infrared (FTIR) spectrometer (PerkinElmer Spectrum GX) and a micro Raman spectrometer (Horiba Jobin Yvon HR800UV) operated using an argon-ion laser (514.5 nm) at room temperature were used to acquire chemical functional groups and Raman spectra of the investigated samples, respectively. A UV–visible spectrophotometer (Cary 5000) was used to collect diffuse reflectance spectra of HfxTayOz films subjected to different wet oxidation temperatures. The capacitance–voltage (C–V) characteristics of the fabricated MOS test structures were measured using a Keithley 4200-SCS parameter analyzer.

3. Results and Discussion

Figure 1 shows the HRXRD patterns of the HfxTayOz films that were subjected to different wet oxidation temperatures from 400 to 1000 °C. It was perceived that the HfxTayOz films were in the amorphous phase after undergoing wet oxidation at temperatures from 400 to 800 °C due to nondetectable peaks with regard to the HfxTayOz phase. This indicated that the incorporation of Hf into the Ta2O5 lattice has succeeded in prolonging the crystallization temperature of the Ta2O5 phase, wherein it was reported previously that polycrystalline Ta2O5 films were formed when the oxidation temperature was enhanced to 550 °C.11 In addition, it was disclosed that transformation of the HfxTayOz phase from amorphous to polycrystalline happened at 1000 °C due to the detection of peaks with regard to the HfxTayOz phase that were oriented in the (001), (141), (201), (002), (0220), (0221), (2220), and (2221) planes. These detected peaks for HfxTayOz films grown at 1000 °C were compared to International Centre for Diffraction Data (ICDD) file no. of 00-025-0922 for orthorhombic phase of Ta2O5, and it was perceived that the detected HfxTayOz peaks were shifted to a smaller diffraction angle compared to the diffraction angle for the Ta2O5 phase. The reason contributing to these peak shifts could be related to the substitution of a larger ionic radius of Hf4+ (0.85 Å) with Ta5+ (0.78 Å) that has caused an expansion in lattice parameter. The possibility of attaining the orthorhombic phase of HfxTayOz films at 1000 °C could be supported through foregoing research works that have divulged the formation of the monoclinic phase of Ta2O5 from 550 to 600 °C. This monoclinic phase of Ta2O5 was transformed to hexagonal and orthorhombic phases below and above 700 °C, respectively. The lattice parameters a, b, and c for the orthorhombic phase of HfxTayOz film subjected to wet oxidation at 1000 °C were calculated using the following equations26

3. 1
3. 2

where dhkl is the interplanar spacing, (hkl) is Miller’s index, and θ is the diffraction angle. The calculated lattice parameters a, b, and c for the HfxTayOz film grown at 1000 °C were 0.62784, 0.42586, and 0.38668 nm, respectively, which were very close to the range (a = 0.6198 nm, b = 0.4290 nm, and c = 0.38880 nm) published in the standard reference pattern for Ta2O5 with ICDD file no. of 00-025-0922. In addition, the crystallite size and microstrain of the HfxTayOz film grown at 1000 °C were extracted based on the Williamson–Hall (W–H) approach,24 and the acquired value of crystallite size and microstrain values are 38.31 nm and 1.6743 × 10–3, respectively. This investigated polycrystalline HfxTayOz film subjected to wet oxidation at 1000 °C has also disclosed the acquisition of preferred orientation in the (141) plane, wherein the coefficient of texture, Thkl, of this HfxTayOz films was calculated using the following equation27

3. 3

where Im (hkl) is the measured relative intensity of reflection from the (hkl) plane, Io (hkl) is the intensity from the same plane in a standard orthorhombic-Ta2O5 reference sample, and n is the number of HfxTayOz reflection peaks. However, crystallite size, microstrain, as well as Thkl were not calculated for HfxTayOz films subjected to wet oxidation at/beyond 800 °C due to the existence of these films in the amorphous phase.

Figure 1.

Figure 1

Open in a new tab

HRXRD patterns of HfxTayOz films subjected to different wet oxidation temperatures.

Figure 2 presents the cross-sectional field emission scanning electron microscopy (FESEM) image of the investigated HfxTayOz films. It was deduced that the HfxTayOz films were undergoing film densification process due to a decrease in total oxide thickness from 103.78 to 95.56 nm when wet oxidation temperature was enhanced from 400 to 600 °C. This film densification process could be further verified using the Arrhenius equation as follows28

3. 4

where t is the final thickness, to is the initial thickness (constant), k is the Boltzmann constant, and T is the wet oxidation temperature in Kelvin. Based on the acquired Arrhenius plots (not shown) for the investigated samples, the activation energy (Ea) of HfxTayOz films subjected to wet oxidation temperatures from 400 to 600 °C was estimated from the slope of linearly fitted data in Arrhenius plot, and it was revealed that a negative Ea value of 0.0209 × 10–3 eV was attained, supporting the fact that film densification has taken place. Conversely, a positive Ea value (0.0433 × 10–3 eV) was obtained when the wet oxidation temperature was increased from 800 to 1000 °C, suggesting that the HfxTayOz films were experiencing film growth process that could be related to the formation of interfacial layer at the interface between HfxTayOz films and Si substrate. It was believed that the formation of interfacial layer has contributed to the acquisition of a larger total oxide thickness for HfxTayOz films subjected to wet oxidation temperatures beyond 800 °C (Figure 2). This has been translated into the acquisition of the largest root-mean-square (RMS) roughness of the film, according to Figure 3. The related three-dimensional surface topography images of all of the HfxTayOz films were also presented (Figure 4). It was seen that protrusions were distributed uniformly over the surface of the HfxTayOz films. As the wet oxidation temperature was varied from 400 to 1000 °C, an increase in the RMS roughness of the HfxTayOz films was obtained. It was believed that the nucleation process would happen at the lowest annealing temperature (400 °C) to assist in island growth at the interface between HfxTayOz films and Si substrate. By further increasing the annealing temperature to beyond 400 °C, the oxygen has attained sufficient energy, and thus, more oxygen would be able to diffuse to the interface region to assist in the coalescence of islands to form either larger islands or interfacial layer. It was believed that the acquisition of a larger RMS roughness when the annealing temperature was enhanced from 400 to 1000 °C could be related to the growth of larger islands, coalescence of islands to become a layer, as well as growth of new islands at the interface that would eventually provoke the smoothness of HfxTayOz films. Therefore, it was observed from the three-dimensional surface topography of the HfxTayOz films subjected to annealing at 1000 °C that the formation of protrusions with dissimilar height was achieved, leading to the acquisition of the highest RMS roughness. This could be supporting evidence that the formation of interfacial layer at 1000 °C was due to the growth of larger islands and new islands, which has influenced the smoothness of the interfacial layer. In addition, the process of islands coalescence to become interfacial layer was not ruled out at 1000 °C due to the acquisition of the largest total oxide thickness. As a whole, the process of growing new islands and larger islands could be more dominating than the process of islands coalescence to become interfacial layer from 400 to 800 °C, wherein the increases of total oxide thickness and RMS roughness were less significant compared to the HfxTayOz films annealed at 1000 °C.

Figure 2.

Figure 2

Open in a new tab

Cross-sectional field emission scanning electron microscopy images of HfxTayOz films subjected to wet oxidation at (a) 400, (b) 600, (c) 800,24 and (d) 1000 °C. (c) Adapted by permission from Appl. Surf. Sci., 526, 146722. 2020, Elsevier B.V.

Figure 3.

Figure 3

Open in a new tab

RMS roughness of the HfxTayOz films subjected to wet oxidation at 400, 600, 800,24 and 1000 °C.

Figure 4.

Figure 4

Open in a new tab

Three-dimensional topographies of HfxTayOz films grown at (a) 400, (b) 600, (c) 800,24 and (d) 1000 °C. (c) Adapted by permission from Appl. Surf. Sci., 526, 146722. 2020, Elsevier B.V.

The hastening of interfacial layer formation as a function of wet oxidation temperatures could be further supported through the measured bidirectional high-frequency (1 MHz) capacitance–voltage (C–V) curves of HfxTayOz films that were annealed from 400 to 1000 °C (Figure 5). Based on the acquired accumulation level of capacitance for HfxTayOz films, it was perceived that the increase of wet oxidation temperature from 600 to 1000 °C has triggered a reduction in the accumulation level of capacitance that would translate into a lower dielectric constant (k) value (Figure 6). A substantial reduction of the accumulation level of capacitance for the HfxTayOz films annealed at 1000 °C would propose an exaggeration in the formation of interfacial layer comprising a higher concentration of low-k SiOx, which could be corroborated through the detection of Si–O chemical bonding at 110829 and 1097 cm–130 as well as Si–O–Hf bonding at 872 and 895 cm–1,31 respectively, using Fourier transformed infrared (FTIR) measurements (Figure 7). The Si–O–Hf bonding was observed only for samples subjected to wet oxidation at 600, 800, and 1000 °C, supporting that the interfacial layer formed at a temperature beyond 400 °C consisted of Hf-O-Si and SiOx. The FESEM cross-sectional image shown in Figure 2d has disclosed an acquisition of the largest total oxide thickness by this HfxTayOz film subjected to wet oxidation at 1000 °C. Nevertheless, the influence of the low-k SiOx compound on the overall k values of the HfxTayOz film subjected to wet oxidation from 400 to 800 °C was less significant than that at 1000 °C because of the attainment of a higher accumulation level of capacitance than 1000 °C. This was an indication that nitrogen gas was able to pile up at the interface between HfxTayOz films and Si to efficiently perform nitridation process during wet oxidation at temperatures below 800 °C, wherein the incoming flux of H+ or OH– ions from the dissociation of water (H2O) molecules has been restricted from interacting with Si surface to minimize the formation the SiOx compound. Among the HfxTayOz films annealed from 400 to 800 °C, the lowest k value (16.2) acquired at 400 °C was anticipated to be related to the existence of oxygen vacancies and other related defects. The reason contributing to this anticipation was due to utilization of the lowest wet oxidation temperature of 400 °C, wherein the formation of the SiOx compound was presumed to be least significant and the influence of this compound on the overall k value of HfxTayOz film was minimal. This statement could be further supported through the inability of detecting O–O bond originating from peroxo species (−O22–) at 880 cm–132 that was representing the formation of amorphous SiO2 for HfxTayOz films subjected to wet oxidation at 400 °C (Figure 7). Nonetheless, the increase of wet oxidation temperature to/beyond 600 °C has revealed the detection of O–O bond, suggesting the onset of interfacial layer formation.

Figure 5.

Figure 5

Open in a new tab

Capacitance–voltage curves of HfxTayOz films subjected to different wet oxidation temperatures of 400, 600, 800, and 1000 °C.

Figure 6.

Figure 6

Open in a new tab

Total oxide thickness and dielectric constant for HfxTayOz films as a function of wet oxidation temperatures (400, 600, 800,24 and 1000 °C).

Figure 7.

Figure 7

Open in a new tab

FTIR absorption spectra in the ranges of (a) 370–1000 cm–1 and (b) 1000–4000 cm–1 for HfxTayOz films subjected to different wet oxidation temperatures.

The formation of interfacial layer in the investigated HfxTayOz films during the wet oxidation process would trigger an alteration in the compositional homogeneity of HfxTayOz films through annihilation or generation of more oxygen vacancies as well as other related defects that would eventually regulate the band gap (Eg) of the investigated films. To estimate the Eg of HfxTayOz films subjected to different wet oxidation temperatures, diffused reflectance UV–visible (UV–vis) spectroscopy measurements were performed and the Kubelka–Munk (KM) function was employed to estimate the Eg value based on the following equation33,34

3. 5

where R is the diffuse reflectance and the F(R) function can be multiplied by hv using the corresponding coefficient (n), whereby the n values for direct-band and indirect-band transition were assumed as 1/2 and 2, respectively. Direct Eg (ED) and indirect Eg (EID) values of the investigated HfxTayOz films were extracted by extrapolating the (F(R) × hv)2 and (F(R) × hv)1/2 vs hv plots to zero. Figure 8 presents ED and EID values of HfxTayOz films as a function of wet oxidation temperatures. HfxTayOz films subjected to wet oxidation at 400 °C has attained the smallest ED and EID. The acquisition of the smallest band gap might be due to the existence of the highest density of oxygen vacancies in HfxTayOz films, which might decrease the band gap of the material. However, an energy difference between ED and EID for the HfxTayOz film at 400 °C was 0.39 eV, exceeding the reported energy levels of oxygen vacancies that were located approximately between 0.1 and 0.3 eV35 below the conduction band minimum of HfxTayOz. The reason contributing to this observation suggested that oxygen vacancies might not exist in the HfxTayOz film subjected to wet oxidation at 400 °C due to plausible occupancy of nitrogen at the oxygen vacancies sites. Since nitrogen has been utilized as the carrier gas during the wet oxidation process, the H+ or OH– ions might not have gained sufficient energy at the low temperature of 400 °C to diffuse through the nitrogen barrier layer that was initially formed on the surface of the HfxTayOz film. Therefore, there existed a greater opportunity for the nitrogen to fill up the vacancies, instead of H+ or OH–. A further increase to higher temperatures (600, 800, and 1000 °C) has resulted in a decrease in the energy difference. This observation could be explained in association with an increase in the kinetics of both H+ and OH– ions, whereby the increased temperature would enhance the diffusivity of the ions across the nitrogen barrier layer to an extent that the opportunities of the ions to occupy the oxygen vacancies sites in the HfxTayOz film would be greater. However, owing to the smaller size of the ions with respect to oxygen anions, the chances to fully fill up the oxygen vacancy sites would be less. The acquisition of a decreasing trend in the energy difference from 0.28 to 0.23 eV as the temperature was increased from 600 to 1000 °C suggested that a reduction in the number of oxygen vacancies was achieved, which could be true because the chances for the ions to occupy the oxygen vacancies were increased. In addition, it could be observed that a decreasing trend in both the ED and EID values from 600 to 1000 °C was demonstrated, suggesting the possibility of having additional nitrogen interstitials in the HfxTayOz lattice because the presence of nitrogen interstitials could interrupt the band transition, altering the band gap. According to the previous literature that used density functional theoretical study to investigate the modification of TiO2 by nitrogen and sulfur doping, the presence of nitrogen and sulfur as interstitials in TiO2 exhibited a red shift to the absorption edge of visible light due to the decrease in band gap.36

Figure 8.

Figure 8

Open in a new tab

Direct and indirect band gap values of the investigated HfxTayOz films at 400, 600, 800,24 and 1000 °C.

In fact, the presence of oxygen vacancies could serve as positively charged traps, which would cause a flatband voltage shift (ΔVFB) of the C–V curves to a negative bias (Figure 6). Nonetheless, the shift was not clearly resembled, and thus an estimation of trap density was calculated quantitatively by taking into consideration the effective oxide charge (Qeff) present in the investigated HfxTayOz films using the following equation1

3. 6

where Cox is the maximum accumulation capacitance, q is the electronic charge, ΔVFB is the flatband voltage shift, and A is the capacitor area. The calculated Qeff values of the investigated HfxTayOz films are presented in Figure 9. As the wet oxidation was increased from 400 to 800 °C, an increase in the positive Qeff was obtained, while beyond this temperature range, Qeff was decreased. The lowest Qeff obtained at 400 °C was in agreement with the aforementioned fact, which mentioned about the attainment of the nearly null oxygen vacancies. Moreover, diffusion activities for H+ and OH– were small at this low oxidation temperature, and hence lesser formation of HfxTayOz with oxygen vacancies was attained. As the temperature was increased to 600 °C, an increase in the Qeff value suggested that more H+ and OH– ions were diffusing into the film but were not actively oxidizing the film. Thus, oxygen vacancies were increased. However, as the temperature reached 800 °C, the decrease in oxygen vacancies did not contribute to a decrease in the Qeff value. This finding suggested that besides the formation of oxygen vacancies, an additional factor could have contributed to the changes in the positive Qeff. The increase in Qeff suggested that additional positive charges might come from the H+ ions in the film. The Qeff value has again decreased at 1000 °C, which might be due to the decrease in oxygen vacancies along with the decrease in H+ because these ions could react with the desorbing nitrogen and form a N–H complex. The nitrogen desorption might happen due to its low sticking coefficient. Furthermore, it was highly plausible for some of the N–H complex to have occupied the oxygen vacancies while diffusing away from the interface and hence contributing to reduced positively charged traps.

Figure 9.

Figure 9

Open in a new tab

Effective oxide charge and slow trap density (STD) of HfxTayOz films subjected to wet oxidation at 400, 600, 800,24 and 1000 °C.

The formation of the N–H complex following the desorption of nitrogen could be attributed to changes in terms of interface quality between the HfxTayOz films and Si substrate, wherein Terman’s method could be employed to calculate the interface trap density (Dit) of the investigated samples using the following equation33

3. 7

where ΔVg = Vg – Vg(ideal) is the voltage shift between the experimental and ideal curves, Vg is the experimental gate voltage, and Φs is the surface potential of Si at a specific gate voltage. Figure 10 shows the calculated Dit values as a function of energy trap level (Ec – Et) for the HfxTayOz films subjected to different wet oxidation temperatures. It was perceived that the HfxTayOz films subjected to wet oxidation at 800 °C have achieved the lowest Dit, followed by 1000, 400, and 600 °C. The formation of interfacial layer composed of more SiOx compound at 1000 °C did not seem to improve the interface quality. This could be related to the desorption of nitrogen accumulated at the interface to react with either incoming H+ ions from the dissociation of H2O or H2 gas generated from the formation of HfxTayOz film to form N–H complex, thus reducing the amount of H available for passivation at the interface. The formation of the N–H complex could be supported through the detection of stretching vibration of NH4+ (2200–3100 cm–1),37 symmetric bending of NH3+ (1413 cm–1),38 NH2 (1608 cm–1),39 and N–H stretching (3360 cm–1)40 using FTIR characterization (Figure 7). As a result, the depletion of nitrogen at the interface would allow the OH– ions and oxygen species released from HfxTayOz films to react with Si surface to form SiOx compound as well as H+ ions and H2 gas to passivate the interface-related defects (Pbx). The effectiveness of H2 molecules in passivating the Pbx through the formation of PbxH bonding has been reported previously. In comparison, the formation of N–H complex was less significant at 400 °C due to detection of NH3+ with the lowest intensity and without the detection of NH4+. The diffusion activities for H+ and OH– dissociated from the water molecules during the wet oxidation process were relatively smaller at 400 °C, and hence filling up of oxygen vacancies was lesser. As the wet oxidation temperature was increased to 600 °C, the highest Dit value was attained. In fact, the increasing diffusion activity of H+ or OH– would create a higher opportunity for the interface defects to be passivated. Nonetheless, the acquisition of the high Dit value suggested that the presence of a large amount of oxygen vacancies in the film as well as nitrogen accumulation on the film could have restricted the passage of the ions to the interface by repelling from each other the positive charges induced by oxygen vacancies and the H+. However, HfxTayOz films subjected to wet oxidation at 400 °C has attained a lower Dit than 600 °C, although the formation of the SiOx compound was deemed to be negligible at 400 °C due to the absence of O–O bonding from FTIR measurements. The nitrogen accumulation would explain the decrease in the total oxide thickness as the temperature was increased from 400 to 600 °C. Moreover, owing to the lack of H+ and OH– ions, inevitably the amount of H+ to reach the interface for passivation of defects would be also limited. Hence, the corresponding interface trap density was increased. The acquisition of the lowest Dit at 800 °C implied that more passivation by hydrogen could have taken place, following the increased desorption of nitrogen from the interface, which would allow more H+ and OH– to diffuse inward to the interface. The increase in H+ passivation as well as reduction in the nitridation activity favored the repairing of interface traps, yielding a larger Qeff. The decrease in nitridation thereafter facilitated film growth, thus contributing to the attainment of a larger total oxide thickness. The corresponding total interface trap density (Dtotal) values of the investigated samples were also calculated based on the area under the graph of Dit vs Ec – Et and are presented in the inset of Figure 10.

Figure 10.

Figure 10

Open in a new tab

Interface trap density and total interface trap density of HfxTayOz films subjected to different wet oxidation temperatures.

In addition, conductance measurements for all of the investigated samples were also carried out to calculate interface trap density (Dit) based on the conductance method, as presented in the following equation41

3. 8

where Inline graphic is the maximum measured conductance, Cox is the capacitance of HfxTayOz films in strong accumulation region, and Cm is the capacitance that corresponds to (G/ω)max. The calculation of Dit for HfxTayOz films subjected to wet oxidation temperatures from 400 to 600 °C was not possible as the acquired conductance–gate voltage (G–Vg) curves (Figure 11) did not reveal an observable Inline graphic peak. It was distinguished that HfxTayOz films subjected to wet oxidation at 800 °C have exhibited a better interface quality due to the acquisition of a lower Dit value of 7.32 × 1012 eV–1 cm–2 than 1000 °C (8.76 × 1012 eV–1 cm–2), which was in agreement with the trend of Dit and Dtotal extracted using Terman’s method (Figure 10). In comparison to the Dit acquired from the conductance method, the HfxTayOz films subjected to wet oxidation at 800 and 1000 °C have demonstrated lower Dit values using Terman’s method. The discrepancy was because of the estimation of Dit using Terman’s method takes into consideration the contribution of slow traps in addition to interface traps in the samples, while the conductance method does not consider slow traps.

Figure 11.

Figure 11

Open in a new tab

Conductance–gate voltage characteristics of HfxTayOz films subjected to different wet oxidation temperatures.

Besides, the FTIR spectra of the investigated HfxTayOz films presented in Figure 7 have also displayed the detection of Hf-O chemical bondings located at 424, 465, 512, 563,32 and 623 cm–1,2929 Ta–O chemical bondings at 524, 648,42,43 1150, and 1184 cm–1,44 as well as Ta–O–Ta chemical bonding at 660 cm–1.4545 Additional chemical bondings with regard to Ta–N and Ta–O–N were detected at 78446 and 392.2 cm–147 as well as at 389.2 cm–1,4747 respectively, for all of the investigated HfxTayOz films. However, weakening of peaks with regard to Ta–N and Ta–O–N was discerned when the wet oxidation temperature was enhanced to 1000 °C, which was in accordance with earlier explanation pertaining to the desorption of nitrogen to react with either H+ ions or H2 gas to form the N–H complex at 1000 °C. The disruption of nitrogen diffusion barrier layer at the interface has acted as a mean for the formation of interfacial layer as well as passivating the Pbx by H+ ions or H2 gas.

The trapping and detrapping phenomena of the investigated HfxTayOz films were also investigated from the bidirectional C–V behavior (Figure 6). Of these investigated samples, the HfxTayOz films subjected to wet oxidation at 400 and 800 °C have demonstrated a clockwise hysteresis, while the other HfxTayOz films have exhibited an anticlockwise hysteresis. This indicated that HfxTayOz films grown at 400 and 800 °C were composed of negatively charged scattering centers that would scatter the injected electrons to break the bonding of HfxTayOz films. Therefore, the probability of the HfxTayOz films grown at 400 and 800 °C to undergo the process of trapping and detrapping of charges during bidirectional C–V measurements could be lower than that of the other investigated samples. However, the acquisition of anticlockwise phenomenon for the HfxTayOz films grown at 600 and 1000 °C would propose that the probability of electrons trapping during forward bias and electrons detrapping during reverse bias should be higher than HfxTayOz films grown at 400 and 800 °C. Quantitative analysis was carried out by calculating slow trap density (STD) for all of the investigated HfxTayOz films based on the following equation and is presented in Figure 9(48)

3. 9

where ΔV is the difference between forward- and reverse-bias flatband voltages. It could be observed that STD acquired at 400 and 800 °C was larger than that obtained at 600 and 1000 °C, respectively. This finding supported that negatively charged traps might distract away trapped electrons during the detrapping process as a consequence of repulsion from one another. The negatively charged traps at 400 °C might originate from the nitrogen present on the HfxTayOz film surface, while that at 800 °C was originated from the nitrogen interstitials, which did not take part in forming SiOx but might reside in the film.

4. Conclusions

In this work, hafnium-doped tantalum oxide (HfxTayOz) films were produced after wet oxidation of tantalum–hafnia at various temperatures (400, 600, 800, and 1000 °C) under nitrogen carrier gas flow. The HfxTayOz films remained in amorphous phase at 400, 600, and 800 °C, while beyond these temperatures, the crystalline phase of the film was detected. This was an indication that high temperature (1000 °C) was an onset temperature for crystallization to take place in this work. The onset of crystallization temperature marked also the formation of the thickest total oxide thickness, which was originated from the increasing thickness of SiOx interfacial layer formation. The formation of interfacial layer could be attributed to an enhancement in the diffusivity of H+ and OH– dissociated from the water molecules used during the wet oxidation process to react with the film at the interface. The presence of SiOx was confirmed via FTIR analysis that revealed the presence of Si–O-related bondings in the films. Apart from the formation of the SiOx interfacial layer, the role played by nitrogen carrier gas in terms of controlling the formation of interfacial layer as well as the role of hydrogen in passivating the interface defects were also discussed. As the temperature was increased from 400 to 1000 °C, nitrogen desorption from the interface as well as reaction with the incoming H+ and/or H2 to form N–H complex was possible, as proven via the presence of N–H-related bondings in the samples. In comparison, although the film subjected to wet oxidation at 1000 °C possessed the thickest SiOx interfacial layer, the corresponding Dit was not the lowest, which could be related to a lack of hydrogen at the interface to passivate defects. Owing to the formation of thick SiOx, the dielectric constant became the lowest (4.8). The influence of interfacial layer was deemed to be less significant on the electrical properties of HfxTayOz films subjected to wet oxidation from 400 to 800 °C, wherein the employment of the lowest wet oxidation temperature of 400 °C would mitigate the formation of interfacial layer, but this sample has attained a lower dielectric constant value than the HfxTayOz films subjected to wet oxidation at 600 and 800 °C. This could be an indication that the electrical properties of HfxTayOz films subjected to wet oxidation at 400 °C was controlled by the properties of HfxTayOz films. As for the HfxTayOz films subjected to wet oxidation at 600 and 800 °C, it was believed that both HfxTayOz films and the interfacial layer would influence the electrical performance of these samples because the difference in terms of dielectric constant values acquired by these two samples were comparable but the interface trap density for the HfxTayOz films subjected to wet oxidation at 600 °C was much higher than the interface trap density for the HfxTayOz films subjected to wet oxidation at 800 °C. Although Dit at 600 °C was the highest, its k value and band gap were the highest, while the calculated slow trap density was the lowest and the resulted Qeff was acceptable. Therefore, it is of interest in the near future to extend studies on the approaches to improve the Dit value for the film subjected to wet oxidation at 600 °C.

Acknowledgments

The authors acknowledge the financial support from Malaysia Ministry of Education (MOE) for providing Fundamental Research Grant Scheme (FRGS; 203/CINOR/6711719).

The authors declare no competing financial interest.

References

  1. Quah H. J.; Cheong K. Y. Effects of annealing time on the electrical properties of Y2O3 gate on silicon. J. Exp. Nanosci. 2015, 10, 19–28. 10.1080/17458080.2013.781689. [DOI] [Google Scholar]
  2. Quah H. J.; Cheong K. Y. Effects of post-deposition annealing time in oxygen ambient on Y2O3 film deposited on silicon substrate. Mater. Res. Innov. 2014, 18, S6-495–S6-498. 10.1179/1432891714Z.0000000001032. [DOI] [Google Scholar]
  3. Lim W. F.; Cheong K. Y.; Lockman Z. Physical and electrical characteristics of metalorganic decomposed CeO2 gate spin-coated on 4H-SiC. Appl. Phys. A 2011, 103, 1067–1075. 10.1007/s00339-010-6039-8. [DOI] [Google Scholar]
  4. Lim W. F.; Cheong K. Y. Study of molar ratio on the characteristics of metal-organic decomposed LaxCe1-xOz film as a metal reactive oxide on Si substrate. J. Alloys Compd. 2013, 581, 793–800. 10.1016/j.jallcom.2013.07.173. [DOI] [Google Scholar]
  5. Vitanov P.; Harizanova A.; Ivanova T.; Dimitrova T. Chemical deposition of Al2O3 thin films on Si substrates. Thin Solid Films 2009, 517, 6327–6330. 10.1016/j.tsf.2009.02.085. [DOI] [Google Scholar]
  6. Quah H. J.; Lim W. F.; Wimbush S. C.; Lockman Z.; Cheong K. Y. Electrical properties of pulsed laser deposited Y2O3 gate oxide on 4H-SiC. Electrochem. Solid State Lett. 2010, 13, H396–H398. 10.1149/1.3481926. [DOI] [Google Scholar]
  7. Vitanov P.; Harizanova A.; Ivanova T.; Trapalis C.; Todorova N. Sol-gel ZrO2 and ZrO2-Al2O3 nanocrystalline thin films on Si as high-k dielectrics. Mater. Sci., Eng. B 2009, 165, 178–181. 10.1016/j.mseb.2009.09.002. [DOI] [Google Scholar]
  8. Jun J. H.; Kim H. J.; Choi D. J. Effect of hydration on the properties of lanthanum oxide and lanthanum aluminate thin films. Ceram. Int. 2008, 34, 957–960. 10.1016/j.ceramint.2007.09.072. [DOI] [Google Scholar]
  9. Pradhan D.; Das S.; Dash T. P. Study of strained-Si p-channel MOSFETs with HfO2 gate dielectric. Superlattices Microstruct. 2016, 98, 203–207. 10.1016/j.spmi.2016.08.019. [DOI] [Google Scholar]
  10. Frunză R. C.; Kmet B.; Jankovec M.; Topic M.; Malic B. Ta2O5-based high-k dielectric thin films from solution processed at low temperatures. Mater. Res. Bull. 2014, 50, 323–328. 10.1016/j.materresbull.2013.11.025. [DOI] [Google Scholar]
  11. Atanassova E.; Spassov D.; Paskaleva A.; Koprinarova J.; Georgieva M. Influence of oxidation temperature on the microstructure and electrical properties of Ta2O5 on Si. Microelectron. J. 2002, 33, 907–920. 10.1016/S0026-2692(02)00120-9. [DOI] [Google Scholar]
  12. Quah H. J.; Hassan Z.; Yam F. K.; Ahmed N. M.; Salleh M. A. M.; Matori K. A.; Lim W. F. Effects of ammonia-ambient annealing on physical and electrical characteristics of rare earth CeO2 as passivation film on silicon. J. Alloys Compd. 2017, 695, 3104–3115. 10.1016/j.jallcom.2016.11.339. [DOI] [Google Scholar]
  13. Quah H. J.; Cheong K. Y.; Hassan Z.; Lockman Z.; Jasni F. A.; Lim W. F. Effects of postdeposition annealing in argon ambient on metalorganic decomposed CeO2 gate spin coated on silicon. J. Electrochem. Soc. 2010, 157, H6–H12. 10.1149/1.3244214. [DOI] [Google Scholar]
  14. Chaneliere C.; Autran J. L.; Devine R. A. B.; Baland B. Tantalum pentoxide (Ta2O5) thin films for advanced dielectric applications. Mater. Sci. Eng., R 1998, 22, 269–319. 10.1016/S0927-796X(97)00023-5. [DOI] [Google Scholar]
  15. Kingon A. I.; Maria J. P.; Streiffer S. K. Alternative dielectrics to silicon dioxide for memory and logic devices. Nature 2000, 406, 1032–1038. 10.1038/35023243. [DOI] [PubMed] [Google Scholar]
  16. Chaneliere C.; Four S.; Autran J. L.; Devine R. A. B. Dielectric permittivity of amorphous and hexagonal electron cyclotron resonance plasma deposited Ta2O5 thin films. Electrochem. Solid-State Lett. 1999, 2, 291–293. 10.1149/1.1390814. [DOI] [Google Scholar]
  17. Wilk G. D.; Wallace R. M.; Anthony J. M. High-k gate dielectrics: Current status and materials properties considerations. J. Appl. Phys. 2001, 89, 5234. 10.1063/1.1361065. [DOI] [Google Scholar]
  18. Hubbard K. J.; Schlom D. G. Thermodynamic stability of binary oxides in contact with silicon. J. Mater. Res. 1996, 11, 2757–2776. 10.1557/JMR.1996.0350. [DOI] [Google Scholar]
  19. Wong H.; Sen B.; Filip V.; Poon M. C. Material properties of interfacial silicate layer and its influence on the electrical characteristics of MOS devices using hafnia as the gate dielectric. Thin Solid Films. 2006, 504, 192–196. 10.1016/j.tsf.2005.09.123. [DOI] [Google Scholar]
  20. Robertson J. High dielectric constant oxides. Eur. Phys. J. Appl. Phys. 2004, 28, 265–291. 10.1051/epjap:2004206. [DOI] [Google Scholar]
  21. Barquinha P.; Pereira L.; Goncalves G.; Martins R.; Kuscer D.; Kosec M.; Fortunato E. Performance and stability of low temperature transparent thin-film transistors using amorphous multicomponent dielectrics. J. Electrochem. Soc. 2009, 156, H824–H831. 10.1149/1.3216049. [DOI] [Google Scholar]
  22. Lu J.; Kuo Y.; Tewg J. Y. Hafnium-doped tantalum oxide high-k gate dielectrics. J. Electrochem. Soc. 2006, 153, G410–G416. 10.1149/1.2180647. [DOI] [Google Scholar]
  23. Yu X. F.; Zhu C. X.; Li M. F.; Chin A.; Du A. Y.; Wang W. D.; Kwong D. L. Electrical characteristics and suppressed boron penetration behaviour of thermally stable HfTaO gate dielectrics with polycrystalline-silicon gate. Appl. Phys. Lett. 2004, 85, 2893–2895. 10.1063/1.1795369. [DOI] [Google Scholar]
  24. Lim W. F.; Quah H. J. Wet oxidation growth of hafnium doped tantalum oxide films with different composition deposited on silicon substrate. Appl. Surf. Sci. 2020, 526, 146722 10.1016/j.apsusc.2020.146722. [DOI] [Google Scholar]
  25. Chanthaphan A.; Hosoi T.; Shimura T.; Watanabe H. Study of SiO2/4H-SiC interface nitridation by post-oxidation annealing in pure nitrogen gas. AIP Adv. 2015, 5, 097134 10.1063/1.4930980. [DOI] [Google Scholar]
  26. Pelleg J.; Elish E.; Mogilyanski D. Evaluation of average domain size and microstrain in a silicide film by the Williamson-Hall method. Metall. Mater. Trans. A 2005, 36, 3187–3194. 10.1007/s11661-005-0089-0. [DOI] [Google Scholar]
  27. Zhang S.; Sun D.; Fu Y. Q.; Du H. J.; Zhang Q. Effect of sputtering target power on preferred orientation in nc-TiN/a-SiNx nanocomposite thin films. J. Metastable Nanocryst. Mater. 2005, 23, 175–178. [Google Scholar]
  28. Quah H. J.; Cheong K. Y. Study on gallium nitride-based metal-oxide-semiconductor capacitors with RF magnetron sputtered Y2O3 gate. IEEE Trans. Electron Dev. 2012, 59, 3009–3016. 10.1109/TED.2012.2212903. [DOI] [Google Scholar]
  29. Zhang X. Y.; Hsu C. H.; Cho Y. S.; Lien S. Y.; Zhu W. Z.; Chen S. Y.; Huang W.; Xie L. G.; Chen L. D.; Zou X. Y.; Huang S. X. Simulation and fabrication of HfO2 thin films passivating Si from a numerical computer and remote plasma ALD. Appl. Sci. 2017, 7, 1244. 10.3390/app7121244. [DOI] [Google Scholar]
  30. Lobo I. P.; da Cruz R. S. Synthesis and characterization of novel ZrO2-SiO2 mixed oxides. Mater. Res. 2014, 17, 700–707. 10.1590/S1516-14392014005000046. [DOI] [Google Scholar]
  31. Chen T. C.; Peng C. Y.; Tseng C. H.; Liao M. H.; Chen M. H.; Chern M. Y.; Tzeng P. J.; Liu C. W. Characterization of the ultrathin HfO2 and Hf-silicate films grown by atomic layer deposition. IEEE Trans. Electron Dev. 2007, 54, 759–766. 10.1109/TED.2007.892012. [DOI] [Google Scholar]
  32. Yan K.; Yao W. Q.; Zhao Y. Y.; Yang L. P.; Cao J. L.; Zhu U. F. Oxygen vacancy induced structure change and interface reaction in HfO2 films on native SiO2/Si substrate. Appl. Surf. Sci. 2016, 390, 260–265. 10.1016/j.apsusc.2016.08.051. [DOI] [Google Scholar]
  33. Nicollian E. H.; Brews J. R. M. O. S.. Metal Oxide Semiconductor Physics and Technology; John Wiley & Sons: New York, 1982. [Google Scholar]
  34. López R.; Gomez R. Band-gap energy estimation from diffuse reflectance measurements on sol-gel and commercial TiO2: a comparative study. J. Sol. Gel Sci. Technol. 2012, 61, 1–7. 10.1007/s10971-011-2582-9. [DOI] [Google Scholar]
  35. Quah H. J.; Hassan Z.; Lim W. F. Passivation of silicon substrate using two-step grown ternary aluminium doped zirconium oxide. Appl. Surf. Sci. 2019, 493, 411–422. 10.1016/j.apsusc.2019.07.023. [DOI] [Google Scholar]
  36. Chen H.; Li X.-C.; Wan R.-D.; Kao-Walter S.; Lei Y.; Leng C.-Y. A. DFT study on modification mechanism of (N,S) interstitial co-doped rutile TiO2. Chem. Phys. Lett. 2018, 695, 8–18. 10.1016/j.cplett.2018.01.044. [DOI] [Google Scholar]
  37. Lin H. J.; Yan H.; Liu B.; Wei L. Q.; Xu B. S. The influence of KH-550 on properties of ammonium polyphosphate and polypropylene flame retardant composites. Polym. Degrad. Stab. 2011, 96, 1382–1388. [Google Scholar]
  38. Villalba M. E. C.; Echeverria G.; Aymonino P. J.; Varetti E. L. Crystal structure and vibrational spectra of [Ru(NH3)4(OH)(NO)]ZnCl4. J. Chem. Crystallogr. 2011, 41, 508–513. 10.1007/s10870-010-9910-8. [DOI] [Google Scholar]
  39. Pandey R. K.; Patil L. S.; Bange J. P.; Gautam D. K. Growth and characterization of silicon nitride films for optoelectronics applications. Opt. Mater. 2004, 27, 139–146. 10.1016/j.optmat.2004.02.028. [DOI] [Google Scholar]
  40. Srikanth C. S.; Chuang S. S. C. Spectroscopic investigation into oxidative degradation of silica-supported amine sorbents for CO2 capture. ChemSusChem 2012, 5, 1435–1442. 10.1002/cssc.201100662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Xiao H.; Huang S.-H. Frequency and voltage dependency of interface states and series resistance in Al/SiO2/p-Si MOS structure. Mater. Sci. Semicond. Process. 2010, 13, 395–399. 10.1016/j.mssp.2011.05.009. [DOI] [Google Scholar]
  42. Ismail A. A.; Faisal M.; Harraz F. A.; Al-Hajry A.; Al-Sehemi A. G. Synthesis of mesoporous sulfur-doped Ta2O5 nanocomposites and their photocatalytic activities. J. Colloid Interface Sci. 2016, 471, 145–154. 10.1016/j.jcis.2016.03.019. [DOI] [PubMed] [Google Scholar]
  43. Cetinorgu-Goldenberg E.; Klemberg-Sapieha J. E.; Martinu L. Effect of postdeposition annealing on the structure, composition, and the mechanical and optical characteristics of niobium and tantalum oxide films. Appl. Opt. 2012, 51, 6498–6507. 10.1364/AO.51.006498. [DOI] [PubMed] [Google Scholar]
  44. Mannequin C.; Tsuruoka T.; Hasegawa T.; Aono M. Identification and roles of nonstoichiometric oxygen in amorphous Ta2O5 thin films deposited by electron beam and sputtering processes. Appl. Surf. Sci. 2016, 385, 426–435. 10.1016/j.apsusc.2016.04.099. [DOI] [Google Scholar]
  45. Meidanchi A.; Jafari A. Synthesis and characterization of high purity Ta2O5 nanoparticles by laser ablation and its antibacterial properties. Opt. Laser Technol. 2019, 111, 89–94. 10.1016/j.optlastec.2018.09.039. [DOI] [Google Scholar]
  46. Jiang G. Q.; Zhang W.; Zang S. Y.; Zhang W. L. Rh nanospheres anchored TaON@Ta2O5 nanophotocatalyst for efficient hydrogen evolution from photocatalytic water splitting under visible light irradiation. Int. J. Hydrogen Energy 2019, 44, 24218–24227. 10.1016/j.ijhydene.2019.07.166. [DOI] [Google Scholar]
  47. Nurlaela E.; Harb M.; del Gobbo S.; Vashishta M.; Takanabe K. Combined experimental and theoretical assessments of the lattice dynamics and optoelectronics of TaON and Ta3N5. J. Solid State Chem. 2015, 229, 219–227. 10.1016/j.jssc.2015.06.029. [DOI] [Google Scholar]
  48. Schroder D. K.Semiconductor Material and Device Characterization; John Wiley & Sons: New York, 1998. [Google Scholar]

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES