1. Introduction Thanks to the rapid development of microelectronics technology, silicon-based power electronics technology has matured due to the widespread application of new power electronic devices such as high-power field-effect transistors (MOSFETs) and insulated-gate bipolar transistors (IGBTs). Currently, the switching performance of these devices has approached their theoretical limits determined by material properties, thanks to the considerable refinement of their structural design and manufacturing processes. The potential for further improvement and enhancement of power electronic devices and systems relying on silicon devices is quite limited. Therefore, meeting the higher performance requirements of next-generation power electronic devices and systems using new materials was a consensus reached in the power electronics and technical communities before the turn of the century, leading to a surge in research and development of silicon carbide power electronic devices. As a wide-bandgap semiconductor material, silicon carbide not only possesses high breakdown electric field strength and good thermal stability, but also features high carrier saturation drift velocity and high thermal conductivity. It can be used to manufacture various high-temperature resistant, high-frequency, high-power devices, applying them to applications where silicon devices are insufficient, or producing effects that silicon devices cannot achieve in general applications. Using wide-bandgap materials can increase the operating temperature of devices. The band gaps of 6H-SiC and 4H-SiC are as high as 3.0 eV and 3.25 eV, respectively, with corresponding intrinsic temperatures exceeding 800 °C. Even 3C-SiC, with the narrowest band gap, has a band gap of around 2.3 eV. Therefore, devices made of silicon carbide can potentially operate at temperatures exceeding 600 °C. The reverse voltage withstand capability of power switching devices is related to the length and resistivity of their drift region (unipolar devices) or base region (bipolar devices). The on-state resistivity of unipolar power switching devices is directly determined by the length and resistivity of the drift region, and thus inversely proportional to the cube of the breakdown electric field strength of the material. Using materials with high breakdown electric field strength to fabricate high-voltage power switches eliminates the need for excessively high resistivity and excessively long drift or base regions. This significantly reduces the on-state resistivity and increases the operating frequency. Silicon carbide (SiC) has a breakdown electric field strength eight times that of silicon and an electron saturation drift velocity twice that of silicon, which is more conducive to increasing the operating frequency of devices. Therefore, SiC unipolar power switches not only have very low on-state resistivity, but their operating frequencies are generally more than 10 times higher than those of silicon devices. High thermal conductivity allows SiC devices to operate stably at high temperatures for extended periods. Furthermore, SiC is currently the only compound semiconductor that can be produced using thermal oxidation to generate high-quality bulk oxides. This allows it to be used, like silicon, to manufacture devices containing MOS structures such as MOSFETs and IGBTs. Besides power electronics, SiC's main applications include high-frequency electronics, high-temperature electronics, and sensor technology. Therefore, the benefits that power electronics technologies, including microwave power supplies, may gain from the practical application of SiC materials go beyond just improving overall system performance through the use of SiC power switching devices. These benefits also include the material's high-temperature resistance and chemical stability, which, through the integration of signal acquisition and processing systems and intelligent control systems, can improve overall system performance, enabling them to maintain good operating conditions in harsh environments. With the introduction of silicon carbide wafers with a diameter of approximately 30mm around 1990, and the subsequent successful application of high-quality 6H-SiC and 4H-SiC epitaxial layer growth technologies, research and development of various silicon carbide power devices flourished. Currently, various power devices have been proven to be manufactured using silicon carbide. Although issues such as production volume, cost, and reliability still limit its commercialization, the process of silicon carbide devices replacing silicon devices has begun. Cree in the United States and Infineon (Siemens Group) in Germany already offer silicon carbide Schottky barrier diodes with a withstand voltage of 600V and a current of 10A or 12A or less; a 4A device currently sells for only $4. The introduction of silicon carbide Schottky barrier diodes to the market has significantly expanded the application range of Schottky barrier diodes from 250V (gallium arsenide devices) to 600V. Simultaneously, its high-temperature characteristics are excellent; from room temperature to the case-limited 175°C, the reverse leakage current shows almost no increase. With appropriate casing, the operating temperature of this new device can exceed 300℃. Currently, many companies have replaced silicon fast recovery diodes with this device in their IGBT frequency converters or inverters, achieving significant improvements in operating frequency and substantial reductions in switching losses. The overall benefits far outweigh the price difference between silicon carbide and silicon devices. In just a few years, the performance of power electronic devices and systems will be greatly improved due to the widespread application of silicon carbide devices. Below, we review several major silicon carbide power electronic devices, from the current level of devices, materials, and manufacturing processes to the main problems encountered. 2. Silicon Carbide Power Electronic Devices In terms of application requirements, power electronic devices, in addition to minimizing static and dynamic losses, must also have the highest possible surge current withstand capability (current several times the steady-state value in tens of milliseconds). Since surge current causes a sudden increase in junction temperature, devices with high on-state resistivity inherently have very low surge current withstand capability. Because the on-state resistance of unipolar power devices increases rapidly with increasing blocking voltage, silicon power MOSFETs only offer a good performance-price ratio for voltage levels not exceeding 100V. Although silicon IGBTs have made significant improvements in this regard, their switching speed is lower than that of power MOSFETs, failing to meet the needs of high-frequency applications. Theoretical analysis shows that power MOSFETs made with 6H-SiC and 4H-SiC can have on-state resistances 100 times and 2000 times lower than comparable silicon power MOSFETs, respectively. This means that if unipolar devices are made of silicon carbide, their on-state voltage drop will be lower than that of silicon bipolar devices even at blocking voltages as high as 10,000V. Since unipolar devices are superior to bipolar devices in terms of operating frequency, research and development of silicon carbide power electronic devices has focused primarily on Schottky barrier diodes and MOSFETs, achieving significant progress. However, bipolar devices such as bipolar transistors and thyristors, especially PIN diodes, are also receiving considerable attention and are progressing rapidly. 2.1 Silicon Carbide Schottky Barrier Diode (SBD) Many metals, such as nickel (Ni), gold (Au), platinum (Pt), palladium (Pd), titanium (Ti), and cobalt (Co), can form Schottky barrier contacts with silicon carbide, with barrier heights generally above 1 eV. It has been reported that the barrier height of Au/4H-SiC contacts can reach 1.73 eV, and that of Ti/4H-SiC contacts is 1.1 eV. The Schottky barrier height of 6H-SiC varies widely, ranging from a minimum of 0.5 eV to a maximum of 1.7 eV. The Power Semiconductor Research Center (PSRC) at North Carolina State University first reported the world's first successful development of a 6H-SiC Schottky barrier diode in 1992, with a blocking voltage of 400V. In their 1994 report, the blocking voltage was increased to 1000V, close to its theoretical design value. Subsequently, research and development of silicon carbide Schottky barrier diodes expanded to Europe and Asia, with the use of 4H-SiC materials increasing and the blocking voltage significantly improved. However, because the Schottky barrier of silicon carbide is thinner than that of silicon at high voltages, further increasing the blocking voltage of silicon carbide Schottky barrier diodes is limited by the reverse leakage current of the tunneling barrier. Calculations show that for a typical silicon carbide Schottky barrier with a height of 1 eV, the barrier width at the highest breakdown voltage corresponding to the critical breakdown electric field of silicon carbide (3 MVcm⁻¹) is only about 3 nm. This is precisely the typical width for electron tunneling. To fully utilize the high critical breakdown electric field strength of silicon carbide, a pn junction Schottky barrier composite structure (JBS or MPS) as shown in Figure 1 can be used to eliminate the limitation of tunneling current on achieving the highest blocking voltage. This structure was originally proposed for silicon devices. Since the barrier height of a pn junction is related to the bandgap of the semiconductor, while the Schottky barrier height is determined only by the difference in work function between the metal and the semiconductor, the difference in barrier height can be significant for wide-bandgap semiconductors. Thus, when a JBS device is forward biased, the Schottky barrier region, due to its lower barrier, enters the conduction state first and becomes the dominant region, while the pn junction, due to its higher turn-on voltage, is essentially ineffective. However, in reverse bias, the pn junction effectively utilizes its high barrier, shielding the Schottky barrier from the strong electric field with its rapidly expanding depletion region under high reverse voltage, thereby significantly reducing the reverse leakage current. Like a simple Schottky barrier diode, a JBS is still a majority carrier device, with its reverse recovery time reduced to a few nanoseconds, only one-tenth that of silicon fast diodes and silicon carbide high-voltage pn junction diodes. The current challenge with JBS lies in the difficulty of forming ohmic contacts in p-type silicon carbide, as p-type doping of silicon carbide using ion implantation requires very high annealing temperatures, making it difficult to form p+ regions in silicon carbide. The grooved Schottky barrier diode (TSBS) structure proposed by Baliga achieves similar performance to the JBS, but avoids p-type doping. Two metals with different work functions are used to form Schottky barriers of varying heights on the surface of the silicon carbide epitaxial layer and the surface of the groove. The low-barrier contact is on the surface, and the high-barrier contact is on the groove surface, the latter weakening the reverse electric field of the former. Experiments show that if the barrier heights of these two contacts, as well as the mesa width and groove depth, are properly matched, the reverse leakage current of the device can be significantly reduced. Currently, research and development of high-power silicon carbide Schottky barrier diodes has achieved reverse blocking voltages exceeding 4000 V for small-area devices (diameter less than 0.5 mm) and around 1000 V for large-area devices (diameter greater than 1 mm). For example, a 140A/800V 4H-SiC JBS was reported in mid-2001. In another report from the same year, a 4H-SiC Schottky barrier diode with a reverse voltage as high as 1200V was achieved with a diameter of 3mm and a forward current density as high as 300 Acm-2, while the corresponding forward voltage drop was only 2V. 2.2 Silicon Carbide Field-Effect Devices Silicon carbide power MOSFETs are not structurally significantly different from silicon power MOSFETs, and generally employ DMOS or UMOS structures. However, due to the high critical breakdown electric field strength of silicon carbide, the oxide electric field at the corner of the UMOS groove is often very high, exceeding the range that the oxide layer can withstand, leading to destructive failure. Simultaneously, because the charge density at the SiC-SiO2 interface is higher than that at the Si-SiO2 interface, generally in the range of 7×10¹¹～5×10¹² cm⁻²×eV⁻¹, the channel electron equivalent mobility of silicon carbide DMOS or UMOS is as low as only 1～7 cm²/Vs due to the influence of the SiC-SiO2 interface, making the channel resistance much greater than the drift region resistance, becoming the main component determining the on-state resistivity of the device. Research has found that without addressing this issue, the on-state resistance of silicon carbide MOSFETs is even higher than that of silicon MOSFETs. To address this, Baliga proposed a structural design called ACCUFET, as shown in Figure 3. Here, ACCU stands for accumulation. The key feature of this structure is the creation of an extremely thin depletion region on the n- surface beneath the gate oxide layer using a p+ buried layer. The depth of the buried layer and the impurity concentration in the n- region are carefully chosen to ensure that the n- region between the oxide layer and the buried layer is completely depleted by the built-in potential of the p+n- junction, thus forming a normally closed field-effect device. A positive gate voltage converts the n- depletion region into an electron accumulation region, turning on the device. This structure effectively limits the electric field strength in the oxide layer to a safe range of approximately 1 MVcm⁻² by shielding the semiconductor layer beneath the gate oxide layer through the p+n- junction. Furthermore, the gate oxide layer is formed by deposition rather than thermal growth, significantly reducing the on-state resistance of the device. Using this structure, reports in 2000 indicated that blocking voltages exceeding 2000V, with some reaching 7000V, could be achieved with 4H-SiC, exhibiting an on-state resistivity 250 times lower than that of silicon ACCUFETs. Silicon carbide MESFETs and JFETs represent another class of highly distinctive and potentially valuable field-effect devices. Due to the absence of a SiC-SiO2 interface, these devices exhibit high equivalent carrier mobility in their channels, reaching 300 cm²/(V×s) for both 6H-SiC and 4H-SiC. Therefore, silicon carbide MESFETs are being developed as microwave devices. Early theoretical calculations suggested that silicon carbide MESFETs could potentially generate microwave power up to 10 GHz and 65 W (4W/mm²). Recent research and development have proven this goal entirely achievable. Buried-gate JFETs, lacking the Schottky contacts that significantly limit operating temperature, can operate at higher temperatures and can be used as high-temperature, high-power devices. In 2000, the development level of the 4H-SiC JFET reached 1800 V. This was a 1.5 A vertically conductive device with a chip area of ​​2.3 mm² and an on-state resistivity of 24.5 mW/cm². This device achieved its normally closed state by using a depletion layer generated in the drift region by a buried gate and the built-in electric field of the p+n-junction. Applying a positive bias voltage to both the upper and lower gates simultaneously widened the conductive channel, thus reducing the on-state resistivity. The experimental device had a chip size of 1.9 × 1.9 mm², an active region area of ​​2 × 10⁻² cm², an n-epitaxial layer impurity concentration of 7 × 10¹⁴ cm⁻³, and a thickness of 75 mm. This device achieved a blocking voltage as high as 5.5 kV, and an on-state resistivity of only 218 mW/cm² when the gate voltage was applied to 2.6 V. Silicon carbide electrostatic induction transistors (SITs), belonging to the same type as JFETs, were also a major focus of research and development for microwave power devices, primarily used for microwave heating. In 1998, there were reports of a 1.3 GHz frequency and a pulse output power of 400 W. 2.3 Silicon Carbide Power Bipolar Devices Silicon carbide can be used to manufacture bipolar devices with very high blocking voltages, such as high-voltage pin diodes and thyristors. Theoretically, to design a silicon carbide pin diode with a reverse blocking voltage of 25 kV, its n-region impurity concentration only needs to be as low as 5 × 10¹³ cm⁻³, its thickness only 0.2 mm, and its minority carrier lifetime only 20 ms. If the same device were made using silicon, its n-region impurity concentration would need to be as low as 10¹² cm⁻³, its thickness at least 2 mm, and its minority carrier lifetime would need to be as high as 400 ms. Clearly, it is impossible to make such a high-voltage device using silicon, but it is not difficult with silicon carbide. Silicon carbide pn junction diodes are usually made into p+nn structures using liquid phase epitaxy or vapor phase epitaxy, and are divided into planar and mesa types. Conventionally, they are also called pin diodes. Currently, the commonly reported silicon carbide pin diodes use 6H-SiC and 4H-SiC materials, with some using heteroepitaxial 3C-SiC on silicon substrates. However, devices fabricated with 6H-SiC and 4H-SiC still offer the highest reverse voltage withstand capability. High-voltage silicon carbide pin diodes also require termination protection, but in principle, all termination techniques suitable for high-voltage silicon devices are also applicable to silicon carbide. Currently, silicon carbide pin diodes approaching 20 kV have been reported. The Sugawara Laboratory in Japan has used JTE (Junction Termination Extension) termination technology to fabricate 12 kV and 19 kV mesa-type pin diodes using 4H-SiC. The n-region impurity concentrations of these two devices are 2 × 10¹⁴ cm⁻³ and 8 × 10¹³ cm⁻³, respectively, with thicknesses of 0.12 mm and 0.2 mm, respectively. These experimental data are quite close to the theoretical expectations mentioned above, indicating that the blocking voltage of practical silicon carbide diodes is mainly limited by lightly doped thick epitaxial technology. With the widespread application of silicon power MOS and IGBTs, silicon high-power bipolar transistors (BJTs) have gradually faded from the power electronics application stage. However, the surge in silicon carbide device research and development has also sparked the interest of some researchers in developing silicon carbide BJTs, because BJTs, unlike MOSFETs, do not suffer from the problem of oxide layer quality severely affecting device characteristics. The basic structure of silicon carbide BJTs: Early work mainly used 6H-SiC and 3C-SiC materials, while in recent years there has been a preference for 4H-SiC. This is mainly because the substrate problem of 3C-SiC has not yet been well resolved, while the large-size crystal growth technology of 6H-SiC and 4H-SiC has developed rapidly, although the electron mobility of 6H-SiC is not as high as that of 4H-SiC. The main problem in developing silicon carbide BJTs is improving the current gain. Early 6H-SiC BJTs had a current gain of only about 10, mainly limited by carrier recombination in the base region. Shortening the base region to meet the short-lifetime carrier transport requirements, however, increases the lateral resistance of the base region. A promising solution is to use a wide bandgap material as the emitter and employ an effective heterojunction structure to improve minority carrier injection efficiency while maintaining low base resistance. Since silicon carbide inherently has various homomorphic forms with different bandgap widths, realizing a heterojunction should not be difficult. For example, 6H-SiC can be epitaxially grown on 3C-SiC using liquid-phase epitaxy, or 4H-SiC can be epitaxially grown on 6H-SiC to create a wide bandgap emitter. In 2001, the current gain of a 4H-SiC BJT with a blocking voltage as high as 1800 V reached 20. Similar to silicon thyristors, if the n+ collector region in the structure shown in Figure 6 is replaced with a p+ thin layer to create a pnpn structure, a silicon carbide thyristor is formed. This type of device best utilizes the material advantages of silicon carbide in balancing switching frequency, power handling capability, and high-temperature characteristics. Compared to silicon carbide power MOSFETs, silicon carbide GTOs (Gate Turn-Off Thyristors) offer several orders of magnitude higher on-state current density for blocking voltages above 3000V, making them particularly suitable for AC switching applications. For DC switching applications, however, silicon carbide GTOs excel. The first report on thyristor fabrication using 6H-SiC was published in 1994. This research used n+ type 6H-SiC as the substrate and epitaxially grew n-type or p-type long base regions. Due to the limitations of materials at the time, devices with a 6.5 mm thick n-type long base region with a doping concentration of 2.7 × 10¹⁵ cm⁻³ could only withstand a forward blocking voltage of 98 V, while devices with an 8 mm thick p-type long base region with a doping concentration of 1.8 × 10¹⁶ cm⁻³ could withstand a forward blocking voltage of 600 V. The first report on thyristor fabrication using 4H-SiC was published in 1995, also with a blocking voltage of 600 V. Due to the current high demand for GTOs with blocking voltages above 4500 V, recent research and development activities on silicon carbide thyristors have begun to focus on GTOs. In 2000, there were reports of 4H-SiC GTOs with blocking voltages as high as 3100 V and a turn-off gain of 41 at 50℃. Research and development of silicon carbide IGBTs started later, with the first reports appearing in 2000. 3. Materials and Manufacturing Processes of Silicon Carbide Devices In the development of semiconductor science and technology, the research and development of silicon carbide materials and devices started relatively early, but its initial progress was very slow, mainly due to the unique characteristics of silicon carbide crystal growth technology. After this problem was initially solved around 1990, a period of rapid development for silicon carbide devices ensued, largely due to the high compatibility and borrowing between silicon carbide device processes and silicon device processes. Therefore, once the material preparation process matures, silicon carbide devices and integrated circuits will develop faster than other compound semiconductor devices and integrated circuits. 3.1 Silicon Carbide Material Preparation Because silicon carbide is difficult to melt under normal pressure and sublimates at around 2400°C, it is difficult to prepare single crystals like other crystals through slow growth of seed crystals in a melt. Most methods use sublimation to grow seed crystals directly in silicon carbide vapor. This is naturally much more difficult than the preparation of commonly used semiconductors such as germanium, silicon, and gallium arsenide. As a result, the silicon carbide crystal and wafer market has long been dominated by Cree, with high-density defect wafers around 30mm in diameter selling for over $1000 each. Although other companies in Europe, Japan, and the United States (such as Sterling-ATMI and Litton-Airtron) can now also produce and sell silicon carbide wafers, Cree still primarily supplies the 4H-SiC and 6H-SiC wafers used in the research and production of silicon carbide devices worldwide. The price remains high, although the diameter has increased to 40-50mm, and the defect density has been significantly reduced. Cree showcased 100mm diameter 4H-SiC and 6H-SiC wafer samples at the International Conference on Silicon Carbide and Related Materials (ICSCRM) in 1999, and began selling 75mm diameter wafers in October of that year, though it still primarily sells 50mm wafers. However, its micropipe defect density has been decreasing, now below 100cm⁻², with high-quality wafers achieving a micropipe density of no more than 15cm⁻². From a device manufacturing perspective, further improvements in silicon carbide crystal growth technology are needed to meet the requirements for producing high-quality ingots with diameters exceeding 100mm, micropipe densities below 0.5cm⁻², and dislocation densities below 10⁴cm⁻². Micropipes are macroscopic defects visible to the naked eye, and their density directly determines the effective area of ​​silicon carbide devices. Until silicon carbide crystal growth technology advances to the point where micropipe defects can be completely eliminated, high-power power electronic devices such as large-diameter diodes and thyristors will be difficult to manufacture using silicon carbide. However, micropipes may simply be a defect unique to crystals like 4H-SiC and 6H-SiC, which have a mixed cubic and hexagonal structure. While no pure cubic 3C-SiC ingots have yet been produced, Hoya Corporation of Japan has claimed to be able to grow 6-inch dislocation-free wafers up to 2mm thick, and no micropipes were found in these wafers. If micropipes are indeed an intrinsic defect related to crystal structure and not so much to the growth process, then the significance of developing 3C-SiC crystal growth technology for the development of silicon carbide power electronic devices and power electronics technology as a whole is self-evident. Currently, the manufacture of silicon carbide power electronic devices mainly uses 4H-SiC or 6H-SiC wafers as substrates, with a high-resistivity epitaxial layer as the reverse voltage blocking layer. Therefore, high-resistivity thick epitaxial technology has become a key focus of silicon carbide epitaxial process research and development. Vapor isotropic epitaxy of silicon carbide generally requires temperatures above 1500°C. Due to sublimation issues, the temperature cannot be too high, generally not exceeding 1800°C, resulting in a relatively low growth rate. Liquid phase epitaxy has lower temperatures and higher rates, but lower yields. Currently, silicon carbide homoepitaxial growth can generally only achieve impurity concentrations below 10¹⁵ cm⁻³ and thicknesses not exceeding 50 mm. 3.2 Silicon Carbide Device Processes Although silicon carbide device processes and equipment are highly compatible with silicon devices, they cannot be directly copied. Compared to silicon, silicon carbide device processes generally require much higher temperatures. Silicon carbide wafers are small, fragile, transparent, and expensive, making them difficult for large companies' production lines to adapt to. University laboratories, however, are more flexible and have become the main force in developing silicon carbide device processes. Doping is the most fundamental device process. Since the diffusion coefficient of general impurities in silicon carbide is as low as in SiO₂, at temperatures suitable for effective impurity diffusion in silicon carbide, SiO₂ loses its masking effect on impurities. Furthermore, silicon carbide itself is unstable at such high temperatures. Therefore, diffusion doping is not suitable; instead, ion implantation and accompanying doping during material preparation are mainly used to meet the needs of manufacturing silicon carbide devices. In the vapor-phase growth of silicon carbide materials, n-type doping typically uses electron-grade pure nitrogen as the dopant, while p-type doping generally uses trimethylaluminum. Nitrogen is also commonly used as the impurity in n-type ion implantation. The lattice damage caused by nitrogen ion implantation is relatively easy to eliminate with annealing. Aluminum is also commonly used as the impurity in p-type ion implantation. Because aluminum atoms are much larger than carbon atoms, the lattice damage caused by implantation and the inactive state of the impurity are more severe, often requiring high substrate temperatures followed by annealing at even higher temperatures. This leads to problems such as silicon carbide decomposition and silicon atom sublimation on the wafer surface. If the residual carbon can form a graphitic carbon film, it can help prevent further surface decomposition. Therefore, boron, with a size comparable to carbon, has also become a commonly used p-type implantation impurity. Currently, p-type ion implantation still faces many challenges; a series of process parameters, from impurity selection to annealing temperature, still need optimization. Furthermore, p-type ion implantation is crucial for improving the channel mobility of power MOSFETs. Interface defects between the gate oxide and silicon carbide have a significant impact on the channel mobility of power MOSFETs, making the growth or deposition of the gate oxide crucial. Besides thermal oxidation similar to silicon, silicon carbide can also be grown using a combustion method for the gate oxide, which produces a lower interface state density. Thermal oxidation of the gate oxide in NO can also reduce the interface state density. For the same gate oxide growth method, 6H-SiC has a slightly higher channel mobility than 4H-SiC; however, in terms of carrier mobility in the bulk material, 4H-SiC is higher than 6H-SiC. This indicates that the oxide interface defect problem is more severe in 4H-SiC. Using a high-temperature rapid annealing method at 1400 °C, ohmic contacts of n-type and p-type 4H-SiC can achieve contact resistances as low as 10⁻⁵ W/cm², using Ni and Al electrode materials, respectively. However, the thermal stability of this contact is poor above 400 °C. Using Al/Ni/W/Au composite electrodes for p-type 4H-SiC can improve thermal stability to 600 °C for 100 hours, but its contact resistivity is as high as 10⁻³ W/cm². Similar effects can be achieved using TaC and AlSi alloy electrodes. 6H-SiC is easier to obtain low-resistance ohmic contacts than 4H-SiC, with a contact resistivity as low as 10⁻⁶ W/cm². Most termination and passivation techniques used in high-voltage silicon devices, such as field plates, field rings, and junction terminations, are also applicable to silicon carbide devices. In addition, implanting large doses of Ar or B at the junction termination, by damaging the lattice to form a high-resistivity region, plays a role similar to that of semi-insulating polycrystalline silicon (SIPOS) in silicon power devices, and also has significant effects. If Ar or B ion implantation is followed by annealing at 600 °C, the reverse characteristics of the device will be further improved. Plasma etching for trenching on semiconductor surfaces has played a positive role in the development of many novel silicon devices. This method is also an important approach for the development of silicon carbide power devices. However, using high-energy ions to etch grooves on the silicon carbide surface often results in high-density defects on the groove wall surface. These defects intensify surface scattering of charge carriers, which is the main reason for the severe decrease in channel electron mobility in UMOS and silicon carbide devices with similar structures. Simultaneously, the roughness of the groove walls also causes problems such as gate voltage drop and excessive gate leakage current. 4. Conclusion The superior performance of silicon carbide materials and the emerging excellent characteristics and greater potential advantages of silicon carbide devices inspire undiminished enthusiasm and hope. Research and development of silicon carbide power electronic devices has therefore flourished, deepened, and progressed at an increasingly rapid pace. Taking silicon carbide MOSFETs as an example, since the development of this device began in 1992, its blocking voltage basically doubled every 13 months in the first three or four years, and subsequently almost doubled every six months. Of course, for power electronic devices, the material advantages of silicon carbide are not limited to improving the device's withstand voltage. For silicon carbide (SiC) power electronic devices to truly compete with silicon devices in the market, a crucial factor lies in their potential to significantly reduce power consumption. This is evidenced by commercially available SiC Schottky barrier diodes and other SiC power devices still in the laboratory. This is the entry point for SiC as a new material for manufacturing power electronic devices, allowing the energy-saving advantages of power electronics technology to be fully realized. Another advantage of SiC over silicon in the field of power electronics is its ability to balance device power and frequency, as well as high-temperature resistance. These are precisely the basic requirements for devices in the further development of power electronics technology, areas where silicon and gallium arsenide have significant limitations. With further improvements in SiC crystal growth and device manufacturing technologies, various SiC power electronic devices will see significant improvements in yield, reliability, and price in the coming years, thus entering a stage of widespread application. This is highly likely to trigger a new revolution in power electronics technology. Therefore, the birth and development of SiC power electronic devices represents a revolutionary advancement in power electronics technology at the turn of the century.