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 →Filing growth compares 2021 (24 records) with 2024 (34) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 291 records in scope (CR5), not by the ranked leaders only.
Gallium nitride HEMTs promise lower switching losses than silicon, but dynamic on-resistance drift, current collapse and gate degradation remain the reliability problems standing between lab performance and field deployment. This landscape pulls 291 patent families filed between 2015 and 2026 that combine GaN HEMT or GaN transistor terminology with claims addressing dynamic on-resistance, current collapse reliability or gate reliability specifically — not GaN device fabrication broadly.
Because publication typically lags filing by around 18 months, the 2025 and 2026 counts in the trend chart understate real filing activity; treat the most recent two years as a floor, not a ceiling.
Pick a task. Every answer cites the patents behind it.
The dataset spans 291 published patent families across the 2015-2026 window, filed through patent offices in the United States, China, Europe, under the PCT, and in Germany and Austria.
Filings rose from 18 in 2017 to a peak of 34 in 2024, but the 2022 midpoint of 22 already signalled a plateau rather than sustained growth. The partial 2026 count of 6 is consistent with normal publication lag, not necessarily a real drop-off.
H01L (223 records) and H10D (147) together account for the bulk of the corpus, covering semiconductor device structure generally. Circuit-level classes — H03K pulse technique (24), G01R measurement (23), H02M power conversion (9) — and digital processing (G06F, 2) are comparatively thin, suggesting reliability claims still cluster around the device itself rather than surrounding system or measurement circuitry.
Shares are the percentage of the 291 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
This page is one run against one query. Ask Eureka your own question about gallium nitride hemt reliability and every answer comes back with the patent numbers behind it.
Try EurekaThe filing describes an enhancement-mode GaN transistor built from a substrate, transition layers, a III-Nitride buffer and barrier layer, drain and source contacts, and a gate formed from a III-V compound with acceptor-type dopants paired with a gate metal. The gate compound and gate metal are formed through a single photomask step so that they self-align, with matching dimensions at the gate metal base and gate compound top. A field plate of Ohmic metal is also specified.Filed by Efficient Power Conversion Corporation, published 2010-10-14 — among the most-cited records in this corpus.


| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20100025730A1 | Normally-off Semiconductor Devices and Methods of Fabricating the Same | 298 |
| 2 | US7985986B2 | Normally-off semiconductor devices | 124 |
| 3 | US20100258843A1 | ENHANCEMENT MODE GaN HEMT DEVICE AND METHOD FOR FABRICATING THE SAME | 85 |
| 4 | US20170018617A1 | Field-plate structures for semiconductor devices | 82 |
| 5 | US20140335666A1 | Growth of High-Performance III-Nitride Transistor Passivation Layer for GaN Electronics | 78 |
| 6 | US9761675B1 | Resistive field structures for semiconductor devices and uses therof | 59 |
| 7 | US20120211800A1 | GaN HEMTs with a Back Gate Connected to the Source | 55 |
| 8 | US20140159119A1 | Method for Growing III-V Epitaxial Layers and Semiconductor Structure | 50 |
| 9 | US20110263102A1 | Methods of Fabricating Normally-Off Semiconductor Devices | 46 |
| 10 | US20120235160A1 | Normally-Off Semiconductor Devices | 43 |
Citation counts are drawn from documents inside this searched corpus and skew toward older filings that have had more time to accumulate citations; read them as markers of influence on subsequent filers, not as a measure of current commercial relevance.
Each row carries its publication number; clicking a row searches Eureka by that number.
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 →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 →Three patterns stand out once family counts, classification mix and receiving-office data are read together.
The corpus peaked in 2024 after a flat 2022 midpoint of 22 — this is not a technology still climbing a filing curve. New entrants now compete against a decade of accumulated claims on device structure rather than an open field.
With 118 US-directed records against 63 in China and 44 at the EPO, reliability-specific GaN claims are still filed most heavily where the earliest commercial GaN power devices reached market, though the Chinese share is substantial enough to require freedom-to-operate checks there as a matter of course.
H01L and H10D together cover the large majority of the corpus, while H02M power-conversion and G06F digital-processing classes are in single digits. Reliability improvements are being claimed at the transistor level far more than at the circuit or system level where GaN devices are actually deployed.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to gallium nitride hemt reliability, with the prior art for and against each one.
Recent-year momentum data shows a striking pattern: several of the assignees with the deepest historical filing records in this corpus show no filings at all in the latest year, some with a documented -100% year-on-year change.
Assignees including large research labs, universities and power semiconductor specialists in this corpus register zero filings in the latest year, several with a -100% year-on-year change. This does not necessarily mean withdrawal from GaN reliability work — it may reflect a pivot to trade-secret protection or a lag in publication.
Only nine co-assignee pairs appear in the dataset, with the strongest repeated pairing involving Cambridge Electronics alongside named individual inventors across three separate filings — a pattern consistent with a research spin-out still filing jointly with its founding scientists.
The receiving-office split shows the United States as the largest single venue, but China's 63 records plus a run of Chinese entities in the assignee list confirm that reliability-focused GaN filing is no longer a US-and-Europe-only conversation.
| Assignee | Recent year | YoY |
|---|---|---|
| HRL Laboratories, LLC | 0 | — |
| Wolfspeed Semiconductor Co., Ltd. | 0 | — |
| The Hong Kong University of Science and Technology | 0 | -100% |
| Apogan Inc. | 0 | — |
| Power Integrations, Inc. | 0 | -100% |
| Huawei Technologies Co., Ltd. | 0 | — |
| École Polytechnique Fédérale de Lausanne (EPFL) | 0 | — |
| Texas Instruments Incorporated | 0 | -100% |
The dataset points to specific next steps depending on whether the goal is freedom-to-operate, white-space filing, or competitive tracking.
With H01L and H10D covering the large majority of records, any new enhancement-mode or field-plate design should be checked against the dense prior art in these classes before filing.
Explore in Patsnap EurekaSeveral historically active assignees show zero recent filings — worth confirming through non-patent literature whether this reflects a strategy shift rather than genuine exit from the field.
Explore in Patsnap EurekaThin coverage in H02M and G06F suggests circuit-level and software-based reliability compensation for GaN HEMTs remains comparatively open.
Explore in Patsnap EurekaDynamic on-resistance is the increase in a GaN HEMT's on-state resistance that occurs after high-voltage switching, distinct from its resistance under static DC conditions. It is caused primarily by electron trapping effects in the buffer and barrier layers, sometimes called current collapse. Because it directly affects switching efficiency and long-term reliability in power conversion applications, it is one of the most heavily claimed reliability problems in this corpus, appearing alongside gate reliability as a defining search term for the field.
The corpus includes research labs, universities and power semiconductor specialists among the historically active assignees, several of which show zero filings in the most recent full year. Efficient Power Conversion Corporation holds one of the most-cited records in the dataset, an enhancement-mode GaN HEMT filing from 2010. Because several leading historical filers show no recent activity, current standing should be checked against recent product and conference disclosures rather than patent counts alone.
The trend data shows filing activity peaked in 2024 at 34 families after a flat 2022 midpoint of 22, indicating the growth phase has already passed rather than being underway. The 2026 count of 6 reflects a partial year and normal publication lag rather than a confirmed collapse in activity. Overall the pattern is one of a maturing claim space, not an emerging one, which matters for anyone deciding whether to file broad structural claims versus narrower improvement claims.
US20100258843A1, assigned to Efficient Power Conversion Corporation, claims an enhancement-mode GaN transistor built with a self-aligned gate formed from a III-V compound with acceptor-type dopants and a matching gate metal produced through a single photomask step, plus a field plate of Ohmic metal. It is one of the most-cited records in this corpus, meaning many later filings build on or design around its specific gate-formation and field-plate approach. Anyone filing enhancement-mode GaN HEMT claims involving self-aligned gate structures should review it directly rather than relying on this summary.
Classification data shows claims concentrated heavily in device-structure classes H01L and H10D, while power-conversion (H02M) and digital-processing (G06F) classes each have single-digit record counts. This points to comparatively open claim space around system-level current-collapse compensation, in-situ dynamic on-resistance monitoring circuits, and predictive-maintenance approaches for GaN modules — areas adjacent to the core device claims but not yet densely covered by them.
Go past this page: query the whole gallium nitride hemt reliability 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.