Next Article in Journal
Color Reproduction by Multi-Wavelength Bragg Diffraction of White Light
Previous Article in Journal
Reductive Photodegradation of 4,4′-Isopropylidenebis(2,6-dibromophenol) on Fe3O4 Surface
Previous Article in Special Issue
Radiative and Magnetically Stimulated Evolution of Nanostructured Complexes in Silicon Surface Layers
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Carrier Trap Density Reduction at SiO2/4H-Silicon Carbide Interface with Annealing Processes in Phosphoryl Chloride and Nitride Oxide Atmospheres

1
Łukasiewicz Research Network–Institute of Microelectronics and Photonics, Al. Lotników 32/46, 02-668 Warsaw, Poland
2
Institute of Microelectronics and Optoelectronics, Warsaw University of Technology, Ul. Koszykowa 75, 00-662 Warsaw, Poland
*
Author to whom correspondence should be addressed.
Materials 2023, 16(12), 4381; https://doi.org/10.3390/ma16124381
Submission received: 24 March 2023 / Revised: 2 June 2023 / Accepted: 7 June 2023 / Published: 14 June 2023
(This article belongs to the Special Issue Advanced Semiconductor Materials and Devices 2021)

Abstract

:
The electrical and physical properties of the SiC/SiO2 interfaces are critical for the reliability and performance of SiC-based MOSFETs. Optimizing the oxidation and post-oxidation processes is the most promising method of improving oxide quality, channel mobility, and thus the series resistance of the MOSFET. In this work, we analyze the effects of the POCl3 annealing and NO annealing processes on the electrical properties of metal–oxide–semiconductor (MOS) devices formed on 4H-SiC (0001). It is shown that combined annealing processes can result in both low interface trap density (Dit), which is crucial for oxide application in SiC power electronics, and high dielectric breakdown voltage comparable with those obtained via thermal oxidation in pure O2. Comparative results of non-annealed, NO-annealed, and POCl3-annealed oxide–semiconductor structures are shown. POCl3 annealing reduces the interface state density more effectively than the well-established NO annealing processes. The result of 2 × 1011 cm−2 for the interface trap density was attained for a sequence of the two-step annealing process in POCl3 and next in NO atmospheres. The obtained values Dit are comparable to the best results for the SiO2/4H-SiC structures recognized in the literature, while the dielectric critical field was measured at a level ≥9 MVcm−1 with low leakage currents at high fields. Dielectrics, which were developed in this study, have been used to fabricate the 4H-SiC MOSFET transistors successfully.

1. Introduction

An increasing number of mobile and power electronic devices induces demand for modern, fast chargers and power modules. Silicon carbide is a promising material for producing power electronics elements such as Schottky diodes, bipolar transistors, and metal-oxide field-effect transistors (MOSFETs). These elements are applied to converters for various applications [1], ranging from motor drives for different kinds of vehicles [2], high-temperature [3,4,5,6], and high-frequency applications, to renewable energy generation [7,8]. At the present time, SiC power devices offered on the market are junction field-effect transistors (JFETs), metal–oxide–semiconductor field-effect transistors, bipolar junction transistors (BJT), and bipolar transistors with isolated gate (IGBTs). MOSFETs are particularly important due to their application in high-power and high-frequency switching devices. A major bottleneck in developing these transistors is the low level of channel carrier mobility resulting in high series resistance of a channel region. This is the most important problem in power devices, which often require large currents. The increase in channel resistance is mainly caused by electrically active structural imperfections inside the gate dielectric and at the interface between the dielectric and semiconductor, called carrier traps [9,10]. High trap density of the oxide layer is also a serious problem for the application of such dielectrics as a passivation layer in power device structures, for example in junction terminations. All naturally occurring defects, such as interstitial carbon clusters, Si and C vacancies at the SiC surface, and oxygen vacancies in the oxide gate film [9], are responsible for the formation of electrically active traps [10]. The presence of charges ‘trapped’ in such defects in the oxide layer or at the interface strongly influences the transport of the charge carriers in the conduction channel. Previous studies show that effective channel mobility versus gate voltage is influenced by several physical scattering effects [11]. The high density of interface trap states (Dit) close to the conduction band and charges built in oxide strongly impacts the maximum value of the mobility at low fields. In contrast, the near-interface oxide traps and surface roughness are the main factors affecting the mobility at high fields. The present study focuses on the former problem.
The interface trap states are generated during the formation of the gate oxide film, produced via thermal oxidation of SiC in pure oxygen (dry oxidation) or in oxygen with the addition of water vapor (wet oxidation). Different types of defects are observed in the oxide film depending on temperature and oxidation time, the SiC crystal surface’s orientation, and chemical treatments of the surface before and during the oxidation process [12,13,14,15]. A major source of defects responsible for electrically active trap centers, especially those close to the conduction band edge, are various carbon-related defects such as carbon dimmers and carbon interstitials, as revealed by Density Functional Theory (DFT) simulations [16,17,18]. Some defects can be reduced via nitridation that is expected to lead to the formation of strong Si≡N bonds and passivation of Si dangling bonds at the interface. Moreover, nitridation can help remove C-oxide compounds from the SiO2/SiC interface, which is called nitrogen-assisted carbon removal. Such a thermal process in ambient nitrogen is effective only at very high temperatures, as high as 1350 °C or above [19]. Nitridation of SiO2/SiC interfaces at lower temperatures was extensively studied in an oxynitride or a nitric oxide environment [20,21,22], but different impacts on the density of interface trap states were noted [23]. The nitridation process not only reduces the Dit but can also improve the gate oxide reliability [24]. Experimental works have shown that high-temperature processes at the SiO2/SiC interface are complicated, and the results are not easily repeatable on different SiO2/SiC structures.
Another way to improve channel charge mobility is with the introduction of phosphorous ions (group V elements) into the SiO2/SiC interface via thermal diffusion through the SiO2 film in a vapor of the compounds containing the element. It is the thermal P diffusion process with trichlorophosphate (POCl3) vapors, known in silicon technology. Some papers confirmed improved channel electron mobility in MOSFET, but the oxide film’s parameters changed after the thermal POCl3 process [25,26,27]. Moreover, ions introduced into the oxide film can influence other MOSFET parameters, for example, oxide gate reliability such as dielectric strength, higher leakage at high fields, and thermal instability under high electric fields [28]. Dielectric strength degradation was previously indicated as a weak point of the dielectric layer formation in POCl3 processes. Considering that hi-tech equipment available for the high-temperature processes required in SiC technology has some specific physical conditions, every post-oxidation annealing process should be optimized to reach the expected improvement in SiO2/SiC interface quality. We chose the post-oxidation processes of thermal annealing in POCl3 and nitride oxide atmospheres to verify such treatment of the SiO2/SiC-4H interface on the MOS parameters. A few combinations of possible procedures were applied to verify some reported results [29,30,31,32] and recognize the best one. However, new solutions in gate oxide technology try to eliminate SiC oxidation via deposition of SiO2 or other dielectric metal oxides [33,34]. Post-deposition thermal annealing processes or plasma treatments are still applied to ensure a perfect dielectric-film/SiC interface [34,35,36]. By lowering the density of interface states, we increase reasonable channel mobility. Consequently, a low ON-resistance of the MOSFET can be attained, which is desirable in power transistors. Both applications require high dielectric strength and low leakage current at high electric fields [37]. The present investigations show that, despite previous concerns about the application of POCl3 annealing and oxidation processes, it is possible to realize dielectrics with very low trap state density and high critical fields by using proper technological treatment and a combination of POCl3 and NO annealing processes.

2. Experimental Data

N-type 4H-SiC (0001) 4° off substrates from Cree Inc. (Durham, NC, USA) (donor concentration n = 5 × 1018 cm−3) with 20 μm thick, n-type-doped epitaxial layer (n = 1.1 × 1015 cm−3) grown by Showa Denko Materials Co., Ltd. (Tokyo, Japan) were used in our experiments. The substrates were cleaned successively in boiling organic solvents of acetone and propanol. Next, they were etched in Piranha solution (H2SO4:H2O2:H2O, 1:1:1 ratio), rinsed in deionized water, deoxidized in buffered HF solution, and dried in nitrogen blow. Subsequently, for the first series of the samples, high-temperature dry oxidation in an oxygen environment at 1175 °C for 360 or 180 min was conducted. For the second series, wet oxidation in oxygen with water vapors at 1175 °C for 80 min was performed. As a result of these thermal processes, SiO2 layers with thicknesses of about 50 or 70 nm were formed on the 4H-SiC samples. Then, specific post-oxidation annealing processes in POCl3 and NO environments were performed, followed by annealing in N2 for 30 min. POCl3 vapor was supplied from a bubbler by using nitrogen flow. The POCl3 annealing was conducted at 1000 °C for 10 min. Next, annealing was completed in pure N2 at 1000 °C for 30 min or in NO at a temperature of 1000 °C, 1100 °C, or 1175 °C. For chosen samples, annealing in NO at 1000 °C was performed directly after oxidation. Identification symbols of the studied samples are collected in Table 1. Characterization of the Si-oxide film by thickness and refractive coefficient was applied using a spectroscopic ellipsometry technique after every annealing process. The film parameters were calculated based on the fitting of measured ellipsometric parameters using the Sellmeier and Cauchy models of the dielectric film. The next step in MOS fabrication was reactive ion plasma etching using CF4/O2 plasma (rf power p = 100 W, pressure p = 50 mT) to remove silicon dioxide from the backside of the samples. Ti/Ni 10/150 nm thick films were deposited via e-beam evaporation technique on the backside for ohmic contact metallization. The ohmic contacts were formed by annealing in the Ar atmosphere at 1050 °C for 3 min. Next, photolithography and a lift-off technique were used to form capacitance structures. A Ti/Al 10/150 nm metallization was fabricated as metallic circles with diameters of 150, 200, 300, and 750 μm. The last annealing of the structures was performed in Ar at 320 °C for 5 min. Electrical methods determining carrier trap density at the oxide/semiconductor interface are based on capacitance versus voltage measurements (C-V) [38,39,40] of the MOS structure. We applied the high–low frequency method (HI-LO), which replaces an ideal characteristic from the high-frequency method (HF) with a low-frequency measured characteristic (LF) or quasi-static characteristic (quasi-static HI-LO method). This further improves the resolution of the method, eliminating a major source of error—a basic HF method uses ideal C-V characteristics for the calculation of (interface trap) MOS capacitance (Cit), which does not include any fixed or near-interface-trap (NIT) type charges affecting C-V shape during the voltage sweep. The quasi-static HI-LO method depends on applying a frequency with period T higher than for the traps (T > τit) so that every interface trap responds fully to the gate voltage. Then, the interface trap capacitance Cit is determined by the formula:
C i t = 1 C L F 1 C o x 1 1 C H F 1 C o x 1
and Dit can be calculated by:
D i t = Δ C q 1 C H F + Δ C C o x 1 1 C H F C o x 1
where COX is oxide capacitance, ΔC = CLF − CHF, CLF, and CHF are normalized capacitances measured at LF or HF, respectively [41].
Our oxidation and annealing procedures were applied to fabricate a gate in a planar 4H-SiC MOSFET device. The base was formed by deep implantation of Al+ ions with an energy of 450 keV in the 20 μm thick epitaxial n-SiC layer; source and drain regions were donor-doped to the level of 2–5 × 1019 cm−3 by narrow implantations of P+. After forming a graphite cap layer from a 1 μm thick photoresist film on the implanted samples, thermal activation of the dopants was conducted by annealing in a vacuum at a temperature of 1800 °C for 30 min. The mesa structure was formed with photolithography and ion-reactive ion etching to separate the MOSFETs. Wet oxidation and an optimized sequence of the two-step annealing process in POCl3 and next in NO atmospheres were applied for gate SiO2 formation. Ohmic contacts were produced from Ti/Ni and Ti/Al/Ti metallization layers to n-type and p-type regions, respectively. In contrast, ohmic contact formation was performed via annealing at 1050 °C for 2 min in Ar. Ti/Al layers were deposited as the gate electrodes. The channel length (L) and width (W) were designed for 5 and 100 μm, respectively. The field-effect mobility (μFE) was determined from the MOSFET transconductance measurements.

3. Results and Discussion

The obtained structures were characterized by current-voltage (I-V), quasi-static, and high-frequency C-V measurements (f = 100 kHz, amplitude of 30 mV), applying polarization from accumulation to depletion and then returning to the accumulation state. Each measurement was preceded by applying a voltage to the sample and forcing accumulation to ensure full trap occupancy at the beginning of the measurement. The length of this initial voltage stress was determined experimentally for each sample by applying stress with increasing duration. The shortest period not resulting in a further C-V characteristic shift was used during measurements. C-V characteristics for dry-oxidized and N2-annealed samples are shown in Figure 1a. One can see that quasi-static C-V is shifted to the left with respect to the theoretical curve.
The C-V results for samples after post-oxidation annealing in POCl3, and NO are shown in Figure 1b. The difference between theoretical and measured C-V curves is significantly smaller than for the samples annealed only in N2. Such figures correspond with the interfacial trap state density in the samples, especially in the energy range near the Fermi level situated close to the conduction band. The calculated Dit vs. energy can be seen in Figure 1c. The Dit values obtained for the d-N2 sample are consistent with the previously reported Dit values for dry oxidation [36,42,43] and, near the conduction band edge, considerably exceed 1 × 1012 eV−1cm−2. Since the sample was not treated with any processes aimed at reducing trap density, it can be used as a reference sample for comparison with others.
It is interesting to compare the dry and wet oxidation in terms of trap density. The kinetics of oxidation suggest that wet oxidation is preferable due to the much shorter time required to form gate oxide [44]. Some results suggest that this process can result in lower interface state densities than conventional dry oxidation [45,46]. Recently, it has been shown that, during oxidation, the wet oxidation process generates lower interface stresses [45,47]. Since hydrogen has been shown to passivate dangling bonds at the SiO2/SiC interface [48], and due to the amorphous structure of the oxide, those defects appear in a variety of trap energies, mostly located close to the conduction band, so wet oxidation can be used as an efficient and better process for SiO2 formation during the oxidation of SiC. Even though our study does not fully confirm this claim, we have been focused on wet oxidation because this process consumes less energy. The wet oxidized sample is characterized by Dit (at 0.2 eV below EC) with an approximately 30% lower value than that for d-POC.
The interface trap densities Dit for the samples after post-oxidation processes are presented in Figure 1c. These are compared for the energy of 0.2 eV below the conduction band and included in Table 2. The trap density Dit behavior of the NO-annealed sample is consistent with previous data [20,24,49]. However, the results for the POCl3-annealed sample must be discussed. Although it is generally agreed that POCl3 post-oxidation annealing improves channel mobility of the MOSFET by reducing density of interface trap states [50], the energy distribution of trap density affected by this process is unclear. Different studies provide dissimilar Dit profiles [51,52,53,54,55,56,57] for POCl3-treated samples. Pascu et al. showed a little reduction in the trap density near the conduction band edge and almost flat Dit profile for deeper levels in the bandgap. However, Okamoto et al. observed that the trap state density near the conduction band edge is reduced by approximately one order of magnitude as compared with a dry-oxidized SiC reference, and it is almost constant for energies deeper in the bandgap. The present study is roughly concurrent with our previously reported findings [27]. This result is also consistent with the theoretical research of Kobayashi et al. [58], which shows via DFT simulations that, at the upper band, close to the conduction band, carbon-related defects play a dominant role in the interface trap density profile [17,59,60,61], and these can be reduced due to the creation of the −O3PCO2 in the oxide film near the interface during POCl3 treatment. A recent study [54], also based on the DFT computation, suggests that phosphorus incorporated near the interface oxide region destabilizes carbon defects, especially carbon dimmers, and some Si defects through structural reconstruction. It was concluded that phosphorus affects mostly traps with energies approximately 0.15–0.4 eV below the conduction band. This is consistent with the curve line shown in Figure 1c for the dry-oxidized sample after the POCl3 treatment, designated as d-POC. The Dit for the traps with energies below 0.4 eV is reasonably lower than for the reference, and almost unchanged at higher energy.
The best results were obtained by combining the post-oxidation processes in the sequence of POCl3 and NO annealing, as seen in Figure 1c. A pronounced slope in the Dit profile is visible here for the samples after wet oxidation, POCl3 annealing, and successive NO annealing. Moreover, the influence of annealing temperature on the Dit profile can be observed, and 1175 °C seems to be the optimal temperature for NO annealing (w-POC-NO-3). The idea behind the sequence was to combine the processes of (1) destabilization of C defects and (2) incorporation of nitrogen into the interface via thermal diffusion. It has been shown [9,16,19] that NO effectively introduces nitrogen into the interface. The mechanism behind improving oxide quality with the annealing process with nitrogen compounds has been theoretically investigated using DFT towards shifting the trap energy associated with a given defect outside the bandgap. In these simulations, carbon-type defects and the dominant defect contributing to the high trap density close to the conduction band edge were considered [62,63,64]. Older as well as newer studies have shown that other possible types of Si-Si bond defects can be removed with nitrogen passivation [65,66,67]. Particularly, NO annealing efficiently removes active states of Si-related defects such as Si-Si bonded atoms, Si dangling bonds, and silicon vacancy [68]. These defects are responsible for introducing trap states nearer the middle of the gap, as was shown by Nakanuma et al. [68]. Numerous Si dangling bonds can be repaired under optimal thermal conditions thanks to the nitrogen reaction with Si bonds at the interface on both sides of the SiC/SiO2 interface.
The obtained values of oxide thickness tox, trap surface density Dit, effective static charge Qeff, and flat-band potential VFB parameters of the MOS structures are collected in Table 2. Different film thicknesses are the results of different thermal and gas conditions. Ellipsometry measurements of the wet-oxidized and POCl3-annealed films are characterized by a thickness in the range of 73.9–86 nm and refractive index n = 1.489, which is increased in relation to the n-index of the model of thermal SiO2 layer on Si. An increase in thickness of approximately 20 nm or a bit above was registered for the dry or wet oxide films after the POCl3 annealing process. Such a rapid oxidation process may be partially responsible for the effective removal of carbon defects from the interface, as previously reported in [27,31,50]. This can be explained by the transformation of the oxide layer into phosphosilicate glass (PSG) during annealing. In our experiment, the process temperature was kept at a relatively low level (1000 °C), so the transition rate to PSG was slower than reported in other works. The oxygen–SiC reaction at this temperature is too slow for efficient SiO2 formation in the timescale used in the experiment. However, it is worth mentioning that the SiC oxidation reaction in the POCl3 ambient has been reported to have lower activation energy than in the dry oxygen [27,69].
The interface trap density Dit at EC − ET = 0.2 eV obtained for the SiO2/4H-SiC samples annealed at different atmospheres and temperatures is illustrated in Figure 2a, while an effective surface charge density (Qeff) is shown in Figure 2b. The most important observation that can be drawn from these results is a change in the value and sign of the effective charge value and in the sign for the sample annealed only in POCl3 (d-POC). These can be attributed to the formation of phosphosilicate glass (PSG) in the volume of the oxide layer, where P-oxide is built into the layer, and such formation has been described earlier [53]. The increased positive effective surface charge suggests a high phosphorus concentration near the interface region, most likely achieved due to the annealing process at a temperature similar to those in experiments described by others [50,51,52,53]. The chemical reaction of POCl3 with SiC at 1000 °C is too slow for quick oxide formation in the timescale used in our experiment. Therefore, a higher phosphorus concentration can be achieved through the diffusion process [57]. The POCl3 interaction with SiC at the interface can be the cause of a significant reduction in the interface trap state density close to the conduction band. Moreover, the samples after POCl3 treatment and subsequently annealed in NO at a higher temperature are characterized by the effective surface charge density (Figure 2b) close to that observed for the reference sample annealed only in N2.
On the samples annealed in POCl3, it was also observed that after annealing in the NO atmosphere, the thickness of the oxide layers slightly decreased. The effect was measured with ellipsometry on additional wet-oxidized SiC samples. Results are presented in Table 3. The difference in thickness (Δtox) increases with NO annealing temperature, while the n-index is still high. The higher temperature causes an amplified effect on the process. Such decrease in thickness may be caused by the evaporation of some P-oxide compounds from the dielectric layer at the applied temperature.
Based on the C-V measurements, calculation of the interface trap density, and oxide thickness measurements, we suggest that the subsequent NO annealing process performed after the POCl3 process can reverse some of the changes in the oxide structure initially created by phosphorus compounds. It leads to a decrease in P concentration in the layer and, consequently, to a decrease in the positive charge accumulated in this layer during POCl3 annealing. Formation of the PSG-like layer during POCl3 annealing can be the most critical step in the application of such a layer to a gate dielectric in MOSFETs, which has been studied in the last decade [25,26,27,28]. Such considerations are mainly due to two reasons: (1) it was shown that introducing phosphorus can decrease the energy barrier for the Fowler–Nordheim current that dominates the conduction mechanism at the high electric field leading to the premature dielectric breakdown [56], and (2) a high concentration of phosphorus ions in the oxide film can lead to thermal instability of the device, as some of the ions in the oxide volume can be mobile under the strong electric fields present in the MOS device during its operation, modulating the device’s threshold voltage.
The low-temperature NO annealing step developed here can reduce both problems by maintaining the desired low interfacial trap density. To support this claim, we have measured the breakdown electric field for our samples, and the results are shown in Figure 3. The critical field for the dry oxides annealed successively in N2 or NO was near 10 MVcm−1. A 30% lower critical field characterized the wet-oxidized structures. This drift has been previously observed [19]. On the other hand, the critical field for oxides annealed in the phosphoryl chloride atmosphere was above 10 MVcm−1. This result is better than reported for high-temperature annealing [56], and it is concurrent with the results reported by Pascu et al. [70], taken on SiO2/SiC samples after the POCl3 annealing process at a similar temperature. The processes applied for reducing the density of the trap states do not significantly affect the breakdown voltage resistance of the oxide films. Annealing processes in the phosphoryl chloride atmosphere followed by a nitric oxide atmosphere at 1175 °C allow obtaining the critical field of ~9 MVcm−1. Moreover, the MOS structures with this two-step annealed dielectric are characterized by a very low leakage current density, <3 × 10−7 Acm−2, at an applied stress voltage of −100 V.
The successful application of the dielectric layer annealed in accordance with the developed procedure was confirmed by electrical measurements of the manufactured planar MOSFET transistors with a gate width of 100 µm, which showed the field effect mobility μFE of the channel electrons at the level of 30 cm2V−1s−1. Such a high mobility is consistent with the results published in the review work for MOSFETs with the dielectric gate passivated via NO annealing at high temperature. This confirms the usefulness of the proposed procedures for the thermal formation of SiO2 dielectrics in SiC MOSFETs. Some papers present μFE mobility with a peak value close to 100 cm2V−1s−1 [25,26,29], suggesting the possibility of further improvement of the channel mobility with additional optimization of technological processes during fabrication of our MOSFET structure, for example by better recovering the mono-crystalline structure of the implanted p-well region. The obtained results are consistent with other studies in this matter, which have been successively published over the last 20 years [18], until now [68,71].

4. Conclusions

MOS SiO2/4H-SiC structures were characterized using C-V measurements and ellipsometry techniques. It was pointed out that two-stage annealing of SiO2/4H-SiC structures in POCl3 atmosphere at 1000 °C for 10 min and subsequently in NO atmosphere at 1100 °C or 1175 °C reduced the interface trap state density Dit down to the level of 2 × 1011 cm−2 in the vicinity of the conduction band edge. The reduction in the trap state density is related to the specific phosphorous impact on carbon-associated trap states at the SiO2/SiC interface and to nitrogen attaining and passivating the interface traps. The nitridation process of NO annealing at a temperature of 1100 °C or higher is more effective in combination with phosphorous inclusions in the oxide film. The results show that, by performing two-step annealing at a temperature lower than the threshold value for efficient SiC oxidation in POCl3 and subsequently in NO, one can obtain the dielectric layer that has both low interface state density and high critical electric field, such as the one obtained via simple dry oxidation (the critical field of ~9 MVcm−1). Additionally, it was observed that those layers have very low leakage current (<3 × 10−7 Acm−2 at an applied stress voltage of −100 V), making them suitable for application in SiC device technology. Such layers can be used in high field conditions, where high breaking voltage and small leakage currents are required, for example, as surface passivation for junction termination extension structures, in passivation of SiC IGBT structures, or possibly as gate oxides for MOSFET and IGBT devices.

Author Contributions

Conceptualization, A.T.; Methodology, M.K. and A.T.; Software, A.T. and O.S.; Validation, M.K. and K.K.; Formal analysis, A.T., K.K. and M.G.; Investigation, E.B., M.K. and O.S.; Resources, O.S., K.K. and M.G.; Data curation, E.B., M.K. and O.S.; Writing—original draft, E.B.; Writing—review & editing, K.K. and M.G.; Visualization, E.B., M.K. and M.G.; Supervision, M.G. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Polish National Centre for Research and Development (NCBiR) under the project TECHMATSTRATEG within the framework of “Modern material technologies”, grant No. 1/347452/1/NCBR/2017, SiCharge.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. The data is not publicly available due to internal regulations.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Nee, H.-P.; Kolar, J.; Friedrichs, P.; Rabkowski, J. Editorial: Special issue on wide bandgap power devices and their applications. IEEE Trans. Power Electron. 2014, 29, 2153–2154. [Google Scholar] [CrossRef]
  2. Gong, X.; Ferreira, J. Investigation of conducted EMI in SiC JFET inverters using separated heat sinks. IEEE Trans. Ind. Electron. 2014, 61, 115–125. [Google Scholar] [CrossRef]
  3. Alexandru, M.; Banu, V.; Jordà, X.; Montserrat, J.; Vellvehi, M.; Tournier, D.; Millán, J.; Godignon, P. SiC integrated circuit control electronics for high temperature operation. IEEE Trans. Ind. Electron. 2015, 62, 3182–3191. [Google Scholar] [CrossRef]
  4. Esteve, V.; Jordán, J.; Sanchis-Kilders, E.; Dede, E.J.; Maset, E.; Ejea, J.B.; Ferreres, A. Comparative study of a single inverter bridge for dual frequency induction heating using Si and SiC MOSFETs. IEEE Trans. Ind. Electron. 2015, 62, 1440–1450. [Google Scholar] [CrossRef]
  5. Chen, Z.; Yao, Y.; Boroyevich, D.; Ngo, K.; Mattavelli, P.; Rajashekara, K. A 1200 V, 60 A SiC MOSFET multichip phase-leg module for high-temperature, high-frequency applications. IEEE Trans. Power Electron. 2014, 29, 2307–2320. [Google Scholar] [CrossRef]
  6. Ning, P.; Zhang, D.; Lai, R.; Jiang, D.; Wang, F.; Boroyevich, D.; Burgos, R.; Karimi, K.; Immanuel, V.D. High-temperature hardware: Development of a 10-kW high-temperature, high-power-density three-phase AC-DC-AC SiC converter. IEEE Ind. Electron. Mag. 2013, 7, 6–17. [Google Scholar] [CrossRef]
  7. Zhao, B.; Song, Q.; Liu, W. Experimental comparison of isolated bidirectional DC/DC converters based on All-Si and All-SiC power devices for next-generation power conversion application. IEEE Trans. Ind. Electron. 2014, 61, 1389–1393. [Google Scholar] [CrossRef]
  8. Zhang, H.; Tolbert, L.M. Efficiency impact of silicon carbide power electronics for modern wind turbine full scale frequency converter. IEEE Trans. Ind. Electron. 2011, 58, 21–28. [Google Scholar] [CrossRef]
  9. Cabello, M.; Soler, V.; Gemma, R.; Montserrat, J.; Rebollo, J.; Godignon, P. Advanced processing for mobility improvement in 4H-SiC MOSFETs: A review. Mater. Sci. Semicond. Process. 2018, 78, 22–31. [Google Scholar] [CrossRef]
  10. Li, X.; Ermakov, A.; Amarasinghe, V.; Garfunkel, E.; Gustafsson, T.; Feldman, L.C. Oxidation induced stress in SiO2/SiC structures. Appl. Phys. Lett. 2017, 110, 141604. [Google Scholar] [CrossRef]
  11. Pérez-Tomás, A.; Godignon, P.; Mestres, N.; Millán, J. A field-effect electron mobility model for SiC MOSFETs including high density of traps at the interface. J. Microelectron. Eng. 2006, 83, 440–445. [Google Scholar] [CrossRef]
  12. Palmieri, R.; Radtke, C.; Silva, M.R., Jr. Trapping of majority carriers in SiO2/4H-SiC structures. J. Phys. D Appl. Phys. 2009, 42, 125301. [Google Scholar] [CrossRef]
  13. Pascu, R.; Romanitan, C.; Varesteanu, P.; Kusko, M. A Reliable Technology for Advanced SiC-MOS Devices Based on Fabrication of High Quality Silicon Oxide Layers by Converting a-Si. J. Electron. Devices Soc. 2019, 7, 158–167. [Google Scholar] [CrossRef]
  14. Kaminski, P.; Budzioch, R.; Gaca, J.; Michałowski, P.P.; Kozłowski, R.; Harmasz, A.; Ciuk, T.; Płocharski, J. Effect of oxidation temperature on the heterogeneity of chemical composition and density in SiO2 film grown on 4H-SiC. J. Mater. Chem. C 2021, 9, 4393. [Google Scholar] [CrossRef]
  15. Saito, M.; Li, H.; Inoue, K.; Matsuhata, H.; Ikuharaet, Y. Oxygen atom ordering on SiO2/4H-SiC(0001) polar interfaces formed by wet oxidation. Acta Mater. 2021, 221, 117360. [Google Scholar] [CrossRef]
  16. Kobayashi, T.; Matsushita, Y. Structure and energetics of carbon defects in SiC (0001)/SiO2 systems at realistic temperatures: Defects in SiC, SiO2, and at their interface. J. Appl. Phys. 2019, 126, 145302. [Google Scholar] [CrossRef] [Green Version]
  17. Matsushita, Y.; Oshiyama, A. Structural stability and energy levels of carbon-related defects in amorphous SiO2 and its interface with SiC. Jpn. J. Appl. Phys. 2018, 57, 125701. [Google Scholar] [CrossRef] [Green Version]
  18. Ólafsson, H.Ö.; Allerstam, F.; Sveinbjörnsson, E.Ö. On Shallow Interface States in n-Type 4H-SiC Metal-Oxide-Semiconductor Structures. Mater. Sci. Forum 2001, 389–393, 1005–1008. [Google Scholar] [CrossRef]
  19. 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. [Google Scholar] [CrossRef]
  20. Peng, Z.; Wang, Y.; Shen, H.; Li, C.; Wu, J.; Bai, Y.; Liu, K.; Liu, X. Effects of combined NO and forming gas annealing on interfacial properties and oxide reliability of 4H-SiC MOS structures. Microelectron. Reliab. 2016, 58, 192–196. [Google Scholar] [CrossRef]
  21. Krol, K.; Kalisz, M.; Sochacki, M.; Szmidt, J. The influence of post-oxidation annealing process in O2 and N2O on the quality of Al/SiO2/n-type 4H-SiC MOS interface. Mater. Sci. Forum 2013, 740–742, 753–756. [Google Scholar] [CrossRef]
  22. Fujihira, K.; Tarui, Y.; Imaizumi, M.; Ohtsuka, K.; Takami, T.; Shiramizu, T.; Kawase, K.; Tanimura, J.; Ozeki, T. Characteristics of 4H–SiC MOS interface annealed in N2O. Solid-State Electron. 2005, 49, 896–901. [Google Scholar] [CrossRef]
  23. Chowdhury, S.; Yamamoto, K.; Hitchcock, C.W.; Chow, T.P. Characteristics of MOS capacitors with NO and POCl3 annealed gate oxides on (0001), (11-20) and (000-1) 4H-SiC. Mater. Sci. Forum 2015, 821–823, 500–503. [Google Scholar] [CrossRef]
  24. Tsui, B.-Y.; Huang, Y.-T.; Wu, T.-L.; Chien, C.-H. Time-dependent dielectric breakdown of gate oxide on 4H-SiC with different oxidation processes. Microelectron. Reliab. 2021, 123, 114186. [Google Scholar] [CrossRef]
  25. Fiorenza, P.; Giannazzo, F.; Vivona, M.; La Magna, A.; Roccaforte, F. SiO2/4H-SiC interface doping during post-deposition-annealing of the oxide in N2O or POCl3. Appl. Phys. Lett. 2013, 103, 153508. [Google Scholar] [CrossRef]
  26. Okamoto, D.; Yano, H.; Hatayama, T.; Fuyuki, T. Development of 4H-SiC MOSFETs with Phosphorus-Doped Gate Oxide. Mater. Sci. Forum 2012, 717–720, 733–738. [Google Scholar] [CrossRef]
  27. Krol, K.B.; Sochacki, M.; Strupiński, W.; Racka, K.; Guziewicz, M.; Konarski, P.; Miśnik, M.; Szmidt, J. Chlorine-enhanced thermal oxides growth and significant trap density reduction at SiO2/SiC interface by incorporation of phosphorus. Thin Solid Films 2015, 591, 86–89. [Google Scholar] [CrossRef]
  28. Fiorenza, P.; Swanson, L.K.; Vivona, M.; Giannazzo, F.; Bongiorno, C.; Frazzetto, A.; Roccaforte, F. Comparative study of gate oxide in 4H-SiC lateral MOSFETs subjected to post-deposition-annealing in N2O and POCl3. Appl. Phys. A 2014, 115, 333–339. [Google Scholar] [CrossRef]
  29. Yano, H.; Araoka, T.; Hatayama, T.; Fuyuki, T. Improved stability of 4H-SiC MOS device properties by combination of NO and POCl3 annealing. Mater. Sci. Forum 2013, 740–742, 727–732. [Google Scholar] [CrossRef]
  30. Yano, H.; Kanafuji, N.; Osawa, A.; Hatayama, T.; Fuyuki, T. Threshold Voltage Instability in 4H-SiC MOSFETs with Phosphorus-Doped and Nitrided Gate Oxides. IEEE Trans. Electron. Devices 2015, 62, 324–332. [Google Scholar] [CrossRef]
  31. Fiorenza, P.; Giannazzo, F.; Roccaforte, F. Characterization of SiO2/4H-SiC Interfaces in 4H-SiC MOSFETs: A Review. Energies 2019, 12, 2310. [Google Scholar] [CrossRef] [Green Version]
  32. Tachiki, K.; Kaneko, M.; Kobayashi, T.; Kimoto, T. Formation of high-quality SiC(0001)/SiO2 structures by excluding oxidation process with H2 etching before SiO2 deposition and high temperature N2 annealing. Appl. Phys. Express 2020, 13, 121002. [Google Scholar] [CrossRef]
  33. Taube, A.; Guziewicz, M.; Kosiel, K.; Gołaszewska-Malec, K.; Król, K.; Kruszka, R.; Kaminska, E.; Piotrowska, A. Characterization of Al2O3/4H-SiC and Al2O3/SiO2/4H-SiC MOS structures. Bull. Pol. Acad. Sci. Tech. Sci. 2016, 64, 447–451. [Google Scholar] [CrossRef] [Green Version]
  34. Kim, Y.-J.; Baek, R.-H.; Chang, S.K.; Choi, K.K. Effect of hydrogen plasma treatment on the electrical properties for SiC-based power MOSFETs. Microelectron. Eng. 2022, 258, 111769. [Google Scholar] [CrossRef]
  35. Kobayashi, T.; Suda, J.; Kimoto, T. Reduction of interface state density in SiC (0001) MOS structures by post-oxidation Ar annealing at high temperature. AIP Adv. 2017, 7, 045008. [Google Scholar] [CrossRef] [Green Version]
  36. Yano, H.; Katafuchi, F.; Kimoto, T.; Matsunam, H. Effects of wet oxidation/anneal on interface properties of thermally oxidized SiO2/SiC MOS system and MOSFET’s. IEEE Trans. Electron. Devices 1999, 46, 504–510. [Google Scholar] [CrossRef]
  37. Godignon, P.; Montserrat, J.; Rebollo, J.; Planson, D. Edge Terminations for 4H-SiC Power Devices: A Critical Issue. Mater. Sci. Forum 2022, 1062, 570–575. [Google Scholar] [CrossRef]
  38. Nicollian, E.H.; Brews, J.R. MOS (Metal Oxide Semiconductor) Physics and Technology; John Willey and Sons: New York, NY, USA, 1982. [Google Scholar]
  39. Afanas’ev, V.V.; Stesmans, A.; Bassler, M.; Pensl, G.; Schulz, M.J. Shallow electron traps at the 4H-SiC/SiO2 interface. Appl. Phys. Lett. 2000, 76, 336–338. [Google Scholar] [CrossRef]
  40. Shroder, D. Oxide and interface trapped charges, oxide thickness. In Semiconductor Material and Device Characterization; John Wiley & Sons, Inc.: New York, NY, USA, 2006; pp. 319–387. [Google Scholar]
  41. Sze, S.M.; Kwok, K.N. Physics of Semiconductor Devices; Wiley Interscience, John Wiley & Sons: New York, NY, USA, 2006. [Google Scholar]
  42. Afanas’ev, V.V.; Bassler, M.; Pensl, G.; Schulz, M.J. Intrinsic SiC/SiO2 Interface States. Phys. Status Solidi A 1997, 162, 321–337. [Google Scholar] [CrossRef]
  43. Okamoto, M.; Tanaka, M.; Yatsuo, T.; Fukuda, K. Effect of the oxidation process on the electrical characteristics of 4H-SiC p-channel metal-oxide-semiconductor field-effect transistors. Appl. Phys. Lett. 2006, 89, 023502. [Google Scholar] [CrossRef]
  44. Hosoi, T.; Nagai, D.; Shimura, T.; Watanabe, H. Exact evaluation of interface reaction-limited growth in dry and wet thermal oxidation of 4HSiC (0001) Si-face surfaces. Jpn. J. Appl. Phys. 2015, 54, 098002. [Google Scholar] [CrossRef]
  45. Indari, E.D.; Yamashita, Y.; Hasunuma, R.; Oji, H.; Yamabe, K. Relationship between electrical properties and interface structures of SiO2/4H-SiC prepared by dry and wet oxidation. AIP Adv. 2019, 9, 105018. [Google Scholar] [CrossRef] [Green Version]
  46. Fukuda, K.; Kato, M.; Kojima, K.; Senzaki, J. Effect of gate oxidation method on electrical properties of metal-oxide-semiconductor field-effect transistors fabricated on 4H-SiC C(000-1) face. Appl. Phys. Lett. 2004, 84, 2088–2090. [Google Scholar] [CrossRef]
  47. Li, X.; Lee, S.S.; Li, M.; Ermakov, A.; Medina-Ramos, J.; Fister, T.T.; Amarasinghe, V.; Gustafsson, T.; Garfunkel, E.; Fenter, P.; et al. Effect of nitrogen passivation on interface composition and physical stress in SiO2/SiC(4H) structures. Appl. Phys. Lett. 2018, 113, 131601. [Google Scholar] [CrossRef]
  48. Wang, T.; Liu, G.; Li, Y.; Hou, H.; Xu, Z.; Wang, M.; Qiao, G. Native point defects on hydrogen-passivated 4H–SiC(0001) surface and the effects on metal adsorptions. J. Chem. Phys. 2017, 147, 024707. [Google Scholar] [CrossRef]
  49. Kil, T.-H.; Kita, K. Anomalous band alignment change of SiO2/4H–SiC (0001) and (000-1) MOS capacitors induced by NO-POA and its possible origin. Appl. Phys. Lett. 2020, 116, 122103. [Google Scholar] [CrossRef]
  50. Okamoto, D.; Yano, H.; Hirata, K.; Hatayama, T.; Fuyuki, T. Improved Inversion Channel Mobility in 4H-SiC MOSFETs on Si Face Utilizing Phosphorus-Doped Gate Oxide. IEEE Electron. Device Lett. 2010, 31, 710–712. [Google Scholar] [CrossRef]
  51. Pascu, R.; Craciunoiu, F.; Kusko, M.; Mihaila, M.; Pristavu, G.; Badila, M.; Brezeanu, G. SiO2/4H-SiC interface states reduction by POCl3 post-oxidation annealing. In Proceedings of the 2015 International Semiconductor Conference (CAS), Sinaia, Romania, 12–14 October 2015; pp. 255–258. [Google Scholar] [CrossRef]
  52. Yano, H.; Hatayama, T.; Fuyuki, T. POCl3 Annealing as a New Method for Improving 4H-SiC MOS Device Performance. ECS Trans. 2013, 50, 257. [Google Scholar] [CrossRef]
  53. Sharma, Y.K.; Ahyi, A.C.; Isaacs-Smith, T.; Modic, A.; Park, M.; Xu, Y.; Garfunkel, E.L.; Dhar, S.; Feldman, L.C.; Williams, J.R. High-mobility stable 4H-SiC MOSFETs using a thin PSG interfacial passivation layer. IEEE Electron. Device Lett. 2013, 34, 175–177. [Google Scholar] [CrossRef]
  54. Jayawardena, A.; Shen, X.; Mooney, P.M.; Dhar, S. Mechanism of phosphorus passivation of near-interface oxide traps in 4H–SiC MOS devices investigated by CCDLTS and DFT calculation. Semicond. Sci. Technol. 2018, 33, 065005. [Google Scholar] [CrossRef]
  55. Jiao, C.; Ahyi, A.C.; Xu, C.; Morisette, D.; Feldman, L.C.; Dhar, S. Phospho-silicate glass gated 4H-SiC metal-oxide-semiconductor devices: Phosphorus concentration dependence. J. Appl. Phys. 2016, 119, 155705. [Google Scholar] [CrossRef]
  56. Ju, Y.; Bouvet, D.; Stark, R.; Woerle, J.; Grossner, U. 4H-SiC Power VDMOSFET Manufacturing Utilizing POCl3 Post Oxidation Annealing. Mater. Sci. Forum 2020, 1004, 559–564. [Google Scholar] [CrossRef]
  57. Xu, Y.; Xu, C.; Liu, G.; Lee, H.D.; Shubeita, S.M.; Jiao, C.; Modic, A.; Ahyi, A.C.; Sharma, Y.; Wan, A.; et al. Concentration, chemical bonding, and etching behavior of P and N at the SiO2/SiC(0001) interface. J. Appl. Phys. 2018, 118, 235303. [Google Scholar] [CrossRef] [Green Version]
  58. Kobayashi, T.; Matsushita, Y.; Okuda, T.; Oshiyama, A. Microscopic mechanism of carbon annihilation upon SiC oxidation due to phosphorus treatment: Density functional calculations combined with ion mass spectrometry. Appl. Phys. Express 2018, 11, 121301. [Google Scholar] [CrossRef]
  59. Kimoto, T.; Watanabe, H. Defect engineering in SiC technology for high-voltage power devices. Appl. Phys. Express 2018, 13, 120101. [Google Scholar] [CrossRef]
  60. Devynck, F.; Alkauskas, A.; Broqvist, P.; Pasquarello, A. Defect levels of carbon-related defects at the SiC/SiO2 interface from hybrid functionals. Phys. Rev. B 2011, 83, 195319. [Google Scholar] [CrossRef] [Green Version]
  61. Kaneko, T.; Tajima, N.; Yamasaki, T.; Nara, J.; Schimizu, T.; Kato, K.; Ohno, T. Hybrid density functional analysis of distribution of carbon-related defect levels at 4H-SiC(0001)/SiO2 interface. Appl. Phys. Express 2018, 11, 011302. [Google Scholar] [CrossRef]
  62. Gavrikov, A.; Knizhnik, A.; Safonov, A.; Schebinin, A.; Bagatu’yants, A.; Potapkin, B.; Chatterjee, A.; Matocha, K. First-principles-based investigation of kinetic mechanism of SiC(0001) dry oxidation including defect generation and passivation. J. Appl. Phys. 2008, 104, 093508. [Google Scholar] [CrossRef]
  63. Deak, P.; Knaup, J.M.; Hornos, T.; Thill, C.; Gali, A.; Frauenheim, T. The mechanism of defect creation and passivation at the SiC/SiO2 interface. J. Phys. D Appl. Phys. 2007, 40, 6242–6253. [Google Scholar] [CrossRef]
  64. Salemi, S.; Goldsman, N.; Akturk, A.; Lelis, A. Density Functional Theory Based Investigation of Defects and Passivation of 4H-Silicon Carbide/SiO2 Interface. In Proceedings of the International Conference on Simulation of Semiconductor Process and Devices, SISPAD 2012, Denver, CO, US, 5–7 September 2012. [Google Scholar]
  65. Pennington, G.; Ashman, C. Nitrogen Passivation of (0001) 4H-SiC Dangling Bonds. Mater. Sci. Forum 2009, 600–603, 469–472. [Google Scholar] [CrossRef]
  66. Salemi, S.; Goldsman, N.; Ettisserry, D.P.; Akturk, A.; Lelis, A. The effect of defects and their passivation on the density of states of the 4H-silicon-carbide/silicon-dioxide interface. J. Appl. Phys. 2013, 113, 053703. [Google Scholar] [CrossRef]
  67. Wang, Z.; Zhang, Z.; Shao, C.; Robertson, J.; Liu, S.; Guo, Y. Defects and Passivation Mechanism of the Suboxide Layers at SiO₂/4H-SiC (0001) Interface: A First-Principles Calculation. IEEE Trans. Electron. Devices 2020, 68, 288–293. [Google Scholar] [CrossRef]
  68. Nakanuma, T.; Iwakata, Y.; Watanabe, A.; Hosoi, T.; Kobayashi, T.; Sometani, M.; Okamoto, M.; Yoshigoe, A.; Shimura, T.; Watanabe, H. Comprehensive physical and electrical characterizations of NO nitrided SiO2/4H-SiC(11 2 ¯ 0) interfaces. Jpn. J. Appl. Phys. 2022, 61, SC1065. [Google Scholar] [CrossRef]
  69. Zheng, Z.; Tressler, R.E.; Spear, K.E. The effects of Cl2 on the oxidation of single crystal silicon carbide. Corros. Sci. 1992, 33, 557–567. [Google Scholar] [CrossRef]
  70. Pascu, R.; Craciunoiu, F.; Pristavu, G.; Brezeanu, G.; Kusko, M. Oxide trap states versus gas sensing in SiC-MOS capacitors—The effect of N- and P- based post oxidation processes. Sens. Actuators B 2017, 245, 911–922. [Google Scholar] [CrossRef]
  71. Zhang, Z.; Guo, Y.; Robertson, J. Mobility degradation in 4H-SiC MOSFETs and interfacial formation of carbon clusters. Solid-State Electron. 2021, 183, 108051. [Google Scholar] [CrossRef]
Figure 1. C−V characteristics of MOS SiO2/4H-SiC structures formed by using oxide formation processes: dry oxidation and N2 annealing—(a), successive POCl3 and NO annealing—(b); interface trap state density Dit versus energy distance from conductance band edge calculated for the SiO2/4H-SiC structures formed by using dry or wet oxidation, and post-oxidation annealing processes in N2, POCl3, and/or NO environments—(c).
Figure 1. C−V characteristics of MOS SiO2/4H-SiC structures formed by using oxide formation processes: dry oxidation and N2 annealing—(a), successive POCl3 and NO annealing—(b); interface trap state density Dit versus energy distance from conductance band edge calculated for the SiO2/4H-SiC structures formed by using dry or wet oxidation, and post-oxidation annealing processes in N2, POCl3, and/or NO environments—(c).
Materials 16 04381 g001
Figure 2. Interface trap state density Dit at EC − ET = 0.2 eV obtained for gate oxide thermally formed on 4H-SiC(0001) at different atmospheres and post-oxidation annealing processes—(a), surface charge density Qeff measured on the MOS structures—(b).
Figure 2. Interface trap state density Dit at EC − ET = 0.2 eV obtained for gate oxide thermally formed on 4H-SiC(0001) at different atmospheres and post-oxidation annealing processes—(a), surface charge density Qeff measured on the MOS structures—(b).
Materials 16 04381 g002
Figure 3. Breakdown electric field measured on gate oxide thermally formed on 4H-SiC(0001) at different atmospheres and post-oxidation annealing processes.
Figure 3. Breakdown electric field measured on gate oxide thermally formed on 4H-SiC(0001) at different atmospheres and post-oxidation annealing processes.
Materials 16 04381 g003
Table 1. Parameters of the thermal process used for thermal formation of SiO2/4H-SiC structures (oxidation temperature of 1175 °C) and the identification symbols (ID) of the studied samples. Post-oxidation annealing processes in POCl3, NO, and N2 environments were performed at temperature T for time intervals t of 10 min, 30 min, and 30 min, respectively.
Table 1. Parameters of the thermal process used for thermal formation of SiO2/4H-SiC structures (oxidation temperature of 1175 °C) and the identification symbols (ID) of the studied samples. Post-oxidation annealing processes in POCl3, NO, and N2 environments were performed at temperature T for time intervals t of 10 min, 30 min, and 30 min, respectively.
Process Type,
Parameter
Values
Oxidation,
Time t (min)
DryDryDryWetWetWetWet
36018036080808080
POCl3
T (°C)
--1000-100010001000
N2
T (°C)
1000-1000-100010001000
NO
T (°C)
-1000-1000100011001175
sample identification,
ID
d-N2d-NOd-POCw-NOw-POC-NO-1w-POC-NO-2w-POC-NO-3
Table 2. Parameters of the MOS structures obtained from C−V measurements after thermal processes used to develop SiO2/4H-SiC interface: oxide film thickness tox, trap surface density Dit calculated for energy near the conduction band level EC − ET = 0.2 eV, and flat band potential VFB.
Table 2. Parameters of the MOS structures obtained from C−V measurements after thermal processes used to develop SiO2/4H-SiC interface: oxide film thickness tox, trap surface density Dit calculated for energy near the conduction band level EC − ET = 0.2 eV, and flat band potential VFB.
ParameterSample Identification, ID
d-N2d-NOd-POCw-NOw-POC-NO-1w-POC-NO-2w-POC-NO-3
tox (nm)80507853587373
Dit × 1011 (eV−1cm−2)
(EC − ET = 0.2 eV)
487.65.4224.53.12.0
VFB (V)2.373.97−1.153.522.473.323.82
Qeff × 1011 (cm−2)−3.2−126.3−10−5.4−6.5−7.9
Table 3. The thickness tox of oxide films formed by wet oxidation and POCl3 annealing, just before and after NO annealing at different temperatures TNO; refractive index η (at λ = 650 nm) and thickness difference Δtox. The tox and η-index calculated based on ellipsometry measurements.
Table 3. The thickness tox of oxide films formed by wet oxidation and POCl3 annealing, just before and after NO annealing at different temperatures TNO; refractive index η (at λ = 650 nm) and thickness difference Δtox. The tox and η-index calculated based on ellipsometry measurements.
Before NOTemperatureAfter NO
tox (nm)TNO (°C)tox (nm)η (λ = 650 nm)Δtox (nm)
75100066.41.472−8.6
75110062.11.487−12.7
86117572.61.488−13.4
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Brzozowski, E.; Kaminski, M.; Taube, A.; Sadowski, O.; Krol, K.; Guziewicz, M. Carrier Trap Density Reduction at SiO2/4H-Silicon Carbide Interface with Annealing Processes in Phosphoryl Chloride and Nitride Oxide Atmospheres. Materials 2023, 16, 4381. https://doi.org/10.3390/ma16124381

AMA Style

Brzozowski E, Kaminski M, Taube A, Sadowski O, Krol K, Guziewicz M. Carrier Trap Density Reduction at SiO2/4H-Silicon Carbide Interface with Annealing Processes in Phosphoryl Chloride and Nitride Oxide Atmospheres. Materials. 2023; 16(12):4381. https://doi.org/10.3390/ma16124381

Chicago/Turabian Style

Brzozowski, Ernest, Maciej Kaminski, Andrzej Taube, Oskar Sadowski, Krystian Krol, and Marek Guziewicz. 2023. "Carrier Trap Density Reduction at SiO2/4H-Silicon Carbide Interface with Annealing Processes in Phosphoryl Chloride and Nitride Oxide Atmospheres" Materials 16, no. 12: 4381. https://doi.org/10.3390/ma16124381

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop