SiC MOSFET Patents: Who Leads, Where the Gaps Are 2026
- Filing activity peaked in 2022 at 8 families and has not been exceeded since, suggesting the core trench and planar architecture claims were staked out early rather than being an emerging wave.
- China receives more than half of all filings with 14 of 27 families routed through CNIPA versus 8 through the USPTO, so freedom-to-operate work has to start with the Chinese docket, not the US one.
- No single assignee shows recent-year momentum every top-cited organisation in the ranking shows zero filings in the latest year, pointing to a pause in new architecture claims rather than a consolidating leader.
What this landscape covers
This dataset tracks 27 patent families published between 2015 and mid-2026 that combine silicon carbide MOSFET terminology with architecture-specific claim language: trench MOSFET design, planar SiC structures, superjunction geometry and double-trench layouts, filtered against the core IPC classes for semiconductor device structure. It is a narrow, claim-language-defined slice of the much larger SiC power semiconductor field, built to isolate architecture claims from process, packaging or circuit-level filings that happen to mention SiC MOSFETs in passing.
Because publication lags filing by roughly 18 months, the 2025-2026 counts in any trend line understate real filing activity; treat the last one to two data points as a floor, not a ceiling.
Filing trend and technology composition
Two views of the same 27 families: how filing volume has moved year over year, and which IPC subclasses the claims actually sit in.
A flat-to-declining trend after a single peak year
Filings rose from nothing in 2017 to a high of 8 in 2022, then did not repeat that level through the most recent complete years. Read against the publication lag, this looks less like a cooling field and more like a claim space that got staked out in a short window and has since gone quiet at the architecture-specific level captured by this search.
Concentrated in core semiconductor device structure classes
All 27 records sit in H01L (semiconductor devices), with 8 also carrying the newer H10D general semiconductor device classification and a single record touching H10P. The near-total overlap with H01L confirms this corpus is squarely device-structure claims rather than a mix of process, packaging and circuit filings.
Shares are the percentage of the 27 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Silicon Carbide MOSFET Transistor Architecture with Eureka
This page is one run against one query. Ask Eureka your own question about silicon carbide mosfet transistor architecture and every answer comes back with the patent numbers behind it.
Try EurekaThe most-cited records in this corpus
Planar SiC MOSFET device and manufacturing method (EP4492473A1)
The filing describes a planar silicon carbide MOSFET that routes channel current parallel to the (0001) crystal plane while using shallow trenches to expose higher-mobility planes such as (11-20), (1-100) or (03-38), reducing channel resistance without an additional ion-implanted protection layer for the source and well regions.Abstract condensed from the original filing language.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US11152503B1 | Silicon carbide MOSFET with wave-shaped channel regions | 13 |
| 2 | CN117038453A | MOSFET结构及工艺方法 | 5 |
| 3 | CN116259542A | 平面型SiC MOSFET器件的制造方法 | 3 |
| 4 | CN115799303A | 一种具有浅P+结构的SiC MOSFET结构 | 2 |
| 5 | CN115938944A | 适用于平面型SiC MOSFET器件的自对准工艺 | 2 |
| 6 | CN115132823A | 一种集成反向续流二极管的平面SiC MOSFET | 2 |
| 7 | CN116259644A | 平面型SiC MOSFET器件 | 2 |
| 8 | US11410990B1 | Silicon carbide MOSFET with optional asymmetric gate clamp | 2 |
| 9 | CN116387348A | 一种精确控制短沟道的平面型SiC MOSFET及其制造方法 | 1 |
| 10 | WO2021019086A1 | OPTIMIZED RETROGRADE CHANNEL IMPLANT FOR PLANAR SiC MOSFET | 1 |
Citation counts inside a bounded search corpus like this one reward older, earlier-filed records more than they reflect current commercial importance — read them as a map of influence on later filings, not a ranking of present-day relevance.
Patent titles are shown in the language they were filed in, not translated, so that each record stays verifiable against the original filing — a translated title will not match in Eureka or in any national register. Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the numbers mean for a filing decision
Three read-throughs from the trend, citation and jurisdiction data that matter more than the raw counts on their own.
A single peak, not a ramp
Volume climbed from zero in 2017 to 8 in 2022 and has not matched that level since. For a 27-family corpus this is a small enough base that one or two active filers can move the whole trend line, so treat the shape as directional rather than a market-scale signal.
China is the primary filing venue
More than half of the corpus was filed through CNIPA, with the USPTO a distant second and Europe, South Korea and the PCT route each carrying a handful of records. Any clearance search that stops at the USPTO is missing the majority of the documented architecture claims.
One record carries most of the citation weight
The top-cited record draws more than double the citations of the next closest family, with the remainder clustered in single digits. That gap usually marks a foundational claim that later filers had to reference or design around, rather than a crowded, evenly-cited field.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to silicon carbide mosfet transistor architecture, with the prior art for and against each one.
Assignee landscape
The ranking is a long tail: no assignee in this corpus shows filing activity in the most recent year, and the one identified co-assignee pair links two entities that appear to be related organisational entries rather than separate competitors.
No current momentum leader
Every organisation that appears in the recent-momentum data, from research-linked entities to corporate assignees, shows zero filings in the latest year. This is consistent with a field where the architecture-level claim space was filled during the 2022 peak and has gone quiet since, at least at the publication stage visible today.
Filing is mostly solo
Only one co-assignee relationship shows up across 27 families, and it links two closely related entries under the same Nanjing third-generation semiconductor innovation centre naming, not an arm's-length joint filing between competitors. Cross-company co-filing is effectively absent from this dataset.
Chinese entities dominate volume
The receiving-office split points to Chinese research institutes, foundries and specialist SiC firms as the most active filers by count, with US, European and Korean assignees making up the remainder in smaller numbers.
| Assignee | Recent year | YoY |
|---|---|---|
| Semikron | 0 | — |
| Hitachi Energy Switzerland AG | 0 | — |
| United Nova Technology -Yuezhou (Shaoxing) Corp. | 0 | — |
| Zhizhan Technology (Shanghai) Co., Ltd. | 0 | — |
| Nanjing Third Generation Semiconductor Technology Innovation Center Co., Ltd. | 0 | — |
| Shanghai Dingyangtong Semiconductor Technology Co., Ltd. | 0 | — |
| University of Electronic Science and Technology of China | 0 | — |
| Renesas Design UK Limited | 0 | — |
Where to take this
The dataset answers what has been filed; these are the practical next steps for turning it into a filing or clearance decision.
Map claims against your own architecture
Run your specific trench, planar or superjunction design against the most-cited records first, since those carry the claim language later filers had to navigate around.
Explore the full patent set in EurekaWatch for the next filing wave
Because publication lags filing by around 18 months, a real pickup in 2024-2025 activity would not be visible in this data yet. Re-check the trend in two to three quarters.
Set up monitoring in EurekaLook past the citation leaders
The under-claimed sub-areas sit next to, not inside, the dense citation cluster. A first-to-file position there is still available on the evidence gathered so far.
Search white space in EurekaCommon questions on SiC MOSFET architecture patents
This dataset identifies 27 patent families published between 2015 and mid-2026 that combine SiC MOSFET terminology with architecture-specific claim language such as trench design, planar structures, superjunction geometry and double-trench layouts, filtered against core semiconductor device IPC classes. That is a narrow, claim-language-defined figure, not a count of every SiC MOSFET patent in existence; broader searches on process, packaging or circuit-level claims would return a larger and different set. Filing peaked in 2022 at 8 families and has not been matched since in the data available so far.
The corpus shows a long tail rather than one dominant assignee: the most-cited individual record, US11152503B1 on wave-shaped channel regions, draws noticeably more citations than any other family, but no organisation in the recent-momentum data shows filings in the latest year. Chinese entities account for the majority of receiving-office filings, with China taking 14 of 27 against 8 for the United States, so Chinese research institutes and specialist SiC firms are worth tracking alongside the more visible US filer. Treat any single-company narrative with caution given the small base size.
China, by a clear margin at the receiving-office level: 14 of the 27 families in this dataset were filed through CNIPA compared with 8 through the USPTO, with Europe, South Korea and the PCT route accounting for the remainder. A freedom-to-operate review limited to US filings would miss more than half of the documented architecture claims. This does not necessarily mean Chinese entities hold the most commercially valuable claims, only that they filed the most in this jurisdiction and claim-language window.
In this corpus, planar SiC MOSFET claims typically focus on channel orientation relative to the crystal plane and on reducing channel resistance without extra ion-implanted protection layers, as in the representative EP4492473A1 filing. Trench and double-trench claims more often centre on channel geometry itself, such as the wave-shaped channel regions in the top-cited US filing. Both approaches are actively represented in the highest-cited records, so neither can be read as clearly obsolete or clearly dominant from citation volume alone.
The IPC data shows the corpus is almost entirely concentrated in core H01L semiconductor device structure classes, with only partial overlap into H10D and a single record in H10P, meaning most filings sit squarely on device architecture rather than adjacent process or packaging angles. Sub-areas such as self-aligned planar process integration, crystal-plane-selective channel routing and specific double-trench superjunction combinations appear less densely claimed than the core trench and planar geometry claims. A first filing in these adjacent branches would sit outside the densest citation cluster rather than directly on top of it.
Research Silicon Carbide MOSFET Transistor Architecture in depth with Eureka
Go past this page: query the whole silicon carbide mosfet transistor architecture corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.