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 (32 records) with 2024 (28) — 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 645 records in scope (CR5), not by the ranked leaders only.
GaN-on-silicon buffer and stress engineering addresses a specific, well-known problem: gallium nitride and silicon have mismatched lattice constants and thermal expansion coefficients, so growing crack-free, low-dislocation GaN on a silicon wafer requires graded buffer layers, nucleation control and stress management to avoid wafer bow and breakdown failure. This dataset covers 645 published records filed between 2015 and mid-2026 that combine gallium nitride, composition-graded AlGaN or AlN nucleation-layer language with claims touching wafer bow, crack-free thickness, vertical breakdown, carbon doping, dislocation-density reduction or layer compatibility.
Because publication typically lags filing by around 18 months, the most recent filing year in any trend line understates real activity; treat the last one to two years as a floor rather than a ceiling. The picture below uses patent families rather than raw document counts where possible, since families are the fairer unit for comparing filing strategies across jurisdictions.
Pick a task. Every answer cites the patents behind it.
Two views of the same 645 records: how filing activity has moved year over year, and how those records distribute across IPC subclasses. Because a single record can carry several IPC classes, the composition shares add up to more than 100% of the record total.
Annual filings climbed to 48 in 2018 from 45 in 2017, held near the 2022 midpoint of 27, and fell to 12 by 2026 — a partial year that will revise upward as later publications land, but the multi-year direction is down rather than flat.
H01L (semiconductor devices) appears on 83.7% of the 645 records, as expected for a device-heavy topic. Beneath it, H10D and H10P device sub-classes and C30B crystal growth (16.3%) show that a meaningful share of filers are claiming the substrate and growth process itself, not just the finished transistor or LED structure.
Shares are the percentage of the 645 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 gan on silicon buffer and stress engineering and every answer comes back with the patent numbers behind it.
Try EurekaThe patent covers a HEMT device built from a channel layer stack of alternating undoped and periodically carbon-doped GaN layers beneath a barrier layer, with the doping pattern used to manage trap states and buffer resistivity through the growth stack.Carbon doping is one of the named search terms behind this dataset because it is a common route to increasing vertical breakdown voltage without adding dislocations — this record shows how that mechanism gets reduced to a specific layer structure in claim language.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US6046464A | Integrated heterostructures of group III-V nitride semiconductor materials including epitaxial ohmic contact … | 747 |
| 2 | US5679965A | Integrated heterostructures of Group III-V nitride semiconductor materials including epitaxial ohmic contact,… | 603 |
| 3 | US5670798A | Integrated heterostructures of Group III-V nitride semiconductor materials including epitaxial ohmic contact … | 582 |
| 4 | US20160155629A1 | Formation of heteroepitaxial layers with rapid thermal processing to remove lattice dislocations | 474 |
| 5 | US9929011B2 | Formation of heteroepitaxial layers with rapid thermal processing to remove lattice dislocations | 462 |
| 6 | US6447604B1 | Method for achieving improved epitaxy quality (surface texture and defect density) on free-standing (aluminum… | 353 |
| 7 | US6261929B1 | Methods of forming a plurality of semiconductor layers using spaced trench arrays | 318 |
| 8 | US6255198B1 | Methods of fabricating gallium nitride microelectronic layers on silicon layers and gallium nitride microelec… | 285 |
| 9 | US6521514B1 | Pendeoepitaxial methods of fabricating gallium nitride semiconductor layers on sapphire substrates | 246 |
| 10 | US5585648A | High brightness electroluminescent device, emitting in the green to ultraviolet spectrum, and method of makin… | 237 |
Ranked by citation count within this corpus. Older foundational filings accumulate more citations simply by being available longer, so treat this as a signal of influence on the field rather than 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 →The numbers point to a field with an established core, cooling overall activity, and secondary claim clusters that are less crowded than the headline device claims.
The leading assignee holds 51 records and the fifth-ranked holder 21, with the top five together accounting for 24.5% of the 645 records in scope. That leaves roughly three-quarters of filings spread across a long tail, so freedom-to-operate work cannot stop at the largest few names.
Filings rose to 48 in 2018, sat at 27 by the 2022 midpoint, and had fallen to 12 by 2026, a partial year. Recent-year momentum by individual assignee shows most of the historically active filers recording zero filings in the latest year, with only one showing continued activity.
Beyond the dominant H01L device classification, C30B crystal growth appears on 105 of 645 records and C23C coating/deposition on 38, indicating that buffer and nucleation-layer process claims are pursued as a filing strategy separate from finished-device claims.
The United States receives the largest share of filings at 332 records, ahead of the EPO at 80, WIPO/PCT at 68, China at 56, India at 31 and Singapore at 20 — a distribution consistent with a technology whose commercial device manufacturing and litigation exposure both concentrate in the US.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to gan on silicon buffer and stress engineering, with the prior art for and against each one.
The ranked leaders span universities, pure-play GaN foundries, IDMs and equipment makers, reflecting a field where both device houses and materials/equipment suppliers file. Recent-year momentum has dropped to near zero for most of the historically active names.
The leading assignee's 51 records, combined with strong co-assignee pairings with individual named inventors, point to a long-running academic research program that has been a foundational source of buffer and nucleation-layer IP rather than a single commercial filer.
Of the assignees tracked for recent-year momentum, only one recorded filings in the latest year; the others, including several ranked among the most active historically, show zero. This is consistent with the broader decline from the 2018 peak rather than a single company's retreat.
The top 10 combined account for 36.7% of the 645 records in scope, meaning the majority of filings sit with entities outside the most active ten — a pattern typical of a mature, widely practiced process technology rather than one locked up by a handful of players.
| Assignee | Recent year | YoY |
|---|---|---|
| Texas Instruments Inc. | 2 | — |
| North Carolina State University | 0 | — |
| Qromis Inc. | 0 | — |
| Micron Technology Inc. | 0 | — |
| Nichia Corporation | 0 | — |
| Soitec SA | 0 | — |
| Veeco Instruments Inc. | 0 | — |
| PGT International Inc. | 0 | — |
The dataset points to specific questions worth running deeper diligence on before committing engineering or legal resources.
With three-quarters of records held outside the top five assignees, a freedom-to-operate check needs to cover the ranked tail, not just the most visible names.
Explore assignee detail in EurekaCrack-free thickness and wafer-bow compensation claims are thinner in this dataset than core device claims, which may leave room for a well-drafted first claim.
Run a claim-scope search in EurekaFiling has declined from its 2018 peak, but the most recent year is a partial one; a fresh pull closer to 2027 will show whether activity is stabilising or continuing to fall.
Set up monitoring in EurekaThe ranked leaders in this 645-record dataset include a university research program, GaN materials and equipment suppliers, and device IDMs, with the top assignee holding 51 records. Concentration is moderate rather than extreme: the top five hold 24.5% of all records in scope and the top ten hold 36.7%, leaving most filings spread across a long tail of smaller filers. This means competitive monitoring should extend well beyond the handful of largest names.
Filing activity peaked in 2018 at 48 records and has declined since, falling to 12 by 2026, though that most recent year is still partial and publication lag of roughly 18 months means it will revise upward. The 2022 midpoint of 27 filings confirms this is a genuine multi-year slowdown rather than a single-year dip. Recent-year momentum data shows most previously active assignees recording no filings in the latest tracked year, suggesting the core buffer-engineering approaches may be reaching claim saturation.
Carbon doping is used to control trap states and increase the resistivity of buffer layers, which supports higher vertical breakdown voltage without introducing additional dislocations. A representative record, US9608103B2, claims a HEMT built from alternating undoped and periodically carbon-doped GaN layers within the channel stack beneath a barrier layer. Because carbon doping appears as a named search criterion in this dataset, it is a well-represented but specific mechanism, and claim language tends to focus on the doping periodicity and layer arrangement rather than the doping chemistry alone.
The IPC composition shows H01L device claims dominating at 83.7% of the 645 records, while process-oriented classes like C30B crystal growth (16.3%) and C23C coating and deposition (5.9%) are comparatively thinner. Specific sub-areas such as crack-free thickness thresholds, wafer bow compensation for larger-diameter silicon, and composition-graded AlGaN transition schemes show less density than core device claims. These are reasonable starting points for a novelty search before drafting, though a thin IPC share is not proof of an open claim — it needs confirming against the actual prior art.
The United States receives the largest share by a wide margin at 332 records, followed by the European Patent Office at 80, WIPO under the PCT route at 68, China at 56, India at 31, and Singapore at 20. This distribution reflects where GaN power and RF device manufacturing and enforcement activity are concentrated. A filer targeting global coverage should expect the US and PCT routes to carry the densest prior art to search against.
Go past this page: query the whole gan on silicon buffer and stress engineering 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.