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 (275 records) with 2024 (169) — 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 2,886 records in scope (CR5), not by the ranked leaders only.
This dataset tracks patent filings and publications at the intersection of ferroelectric memory and hafnium zirconium oxide dielectrics — the materials system that revived ferroelectric switching as a scalable memory candidate after decades of lead-zirconate-titanate (PZT) stagnation. The search combines core device terms with the specific physics vocabulary practitioners use to describe the technology's known failure modes and enablers: wake-up effect, polarization fatigue, coercive field, orthorhombic phase, depolarization field, and scaling of thickness.
Coverage runs from 2015 through the mid-2026 data cut-off across 2,886 published records. Because publication trails filing by roughly 18 months, the most recent one to two years in any trend understate real filing activity and should be read as provisional rather than declining.
Two views of the same 2,886-record corpus: how filing volume has moved year over year, and which IPC subclasses carry the claim density.
Annual filings rose from 54 in 2017 to a peak of 290 in 2022, then eased to 169 by 2024 — a 39% decline from the 2021 level of 275. Figures for 2025 and 2026 are still filling in under the standard publication lag and should not be read as a continued fall.
H01L (semiconductor devices) covers 66.9% of records and H10B (memory device manufacture) covers 52.6%; G11C (static and digital memories) and H10D (semiconductor devices, general) follow at 33.1% and 23.5%. Records commonly carry more than one class, so these shares sum above 100%.
Shares are the percentage of the 2,886 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 ferroelectric memory and hafnium oxide and every answer comes back with the patent numbers behind it.
Try EurekaA device structure is formed by depositing a first metallic electrode, a node dielectric of amorphous hafnium zirconium oxide (Hf(1-α)Zrα O2, α between 0.50 and 0.90), a second metallic electrode, and annealing the amorphous layer into a crystalline layer engineered as a mixture of tetragonal and orthorhombic phases.Filed by Taiwan Semiconductor Manufacturing Company, published 2026-07-30 — illustrates continued claim activity on phase-fraction control rather than bulk material composition alone.


| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20180269057A1 | Formulation for Deposition of Silicon Doped Hafnium Oxide as Ferroelectric Materials | 445 |
| 2 | US4809225A | Memory cell with volatile and non-volatile portions having ferroelectric capacitors | 268 |
| 3 | US5903492A | Semiconductor memory device and various systems mounting them | 231 |
| 4 | US6067244A | Ferroelectric dynamic random access memory | 228 |
| 5 | US6249014B1 | Hydrogen barrier encapsulation techniques for the control of hydrogen induced degradation of ferroelectric ca… | 211 |
| 6 | US5418389A | Field-effect transistor with perovskite oxide channel | 211 |
| 7 | US6656748B2 | FeRAM capacitor post stack etch clean/repair | 210 |
| 8 | US6924997B2 | Ferroelectric memory and method of operating same | 207 |
| 9 | US5146299A | Ferroelectric thin film material, method of deposition, and devices using same | 184 |
| 10 | US7029925B2 | FeRAM capacitor stack etch | 176 |
Citation counts favour older filings that have had more time to accumulate citations within the searched corpus; treat them as a signal of influence on the field, not of current commercial weight.
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 read-throughs from the filing pattern, the citation table and the class distribution.
With the five leading assignees holding 1,181 of 2,886 records, freedom-to-operate work on the base hafnium-zirconium oxide stack and its standard deposition routes should start from those portfolios, not from a blank-slate search.
The peak of 290 filings in 2022 was followed by a documented decline to 169 by 2024. That is a real three-year contraction, not a data-lag artefact, since 2024 is the last year treated as complete in this dataset.
The most-cited record concerns silicon-doped hafnium oxide deposition chemistry; several of the next most-cited date to earlier ferroelectric DRAM and hydrogen-barrier encapsulation work, underlining how much of today's activity still builds on process and reliability fixes rather than net-new materials.
Two-thirds of records sit in general semiconductor devices and over half in memory device manufacture specifically, meaning most claim activity addresses integration and process flow rather than the standalone material or circuit-level classes further down the list.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to ferroelectric memory and hafnium oxide, with the prior art for and against each one.
The ranked leaders span integrated-device manufacturers, memory specialists and a materials/IP licensing house whose portfolio underpins several of the largest co-filing relationships in the dataset.
The top-ranked assignee's 460 records sit well above fifth place at 138 and tenth place at 70, giving it a filing footprint roughly matching the next several competitors combined.
The leading assignee recorded only 3 filings in the latest year, down 90% year over year; several other leading assignees show zero filings in the latest year, consistent with the general 18-month publication lag rather than a real stop in R&D.
The strongest of the ten identified co-assignee pairs links a materials-and-licensing specialist with a large Japanese electronics manufacturer across 36 shared records, pointing to a technology-licensing relationship rather than an internal joint venture.
| Assignee | Recent year | YoY |
|---|---|---|
| Taiwan Semiconductor Manufacturing Co., Ltd. | 3 | -90% |
| Thin Film Electronics ASA | 0 | — |
| SanDisk Technologies LLC | 0 | -100% |
| Symetrix Corp | 0 | — |
| Intel Corp | 0 | -100% |
| Samsung Electronics Co., Ltd. | 0 | -100% |
| Texas Instruments Inc | 0 | — |
| Canon Inc. | 0 | — |
The dataset points to specific next steps depending on whether the goal is freedom-to-operate, sourcing or technology scouting.
With 40.9% of records held by five assignees, a targeted claim chart against those portfolios will surface overlap faster than a full-corpus review.
Explore assignee portfolios in EurekaThe 2021-2024 decline may not be uniform across H01L, H10B and G11C; breaking the trend out by subclass shows where activity is holding up.
Run a class-level trend query in EurekaThin filing density in interfacial engineering and 3D FeFET integration is where new entrants are more likely to find open claim space.
Set up a white-space alert in EurekaThe assignee ranking in this dataset is led by a single filer with 460 records, well ahead of fifth place at 138 and tenth place at 70. The five leading assignees together account for 40.9% of all 2,886 records in scope, meaning a large share of the field's core claims sit with a small group of integrated device manufacturers and memory specialists. Anyone doing freedom-to-operate work should treat those five portfolios as the first checkpoint rather than searching the corpus at large.
Filings peaked at 290 in 2022 and had declined to 169 by 2024, a documented 39% drop from the 2021 level of 275. That is measured against 2024 as the last complete year in the dataset; 2025 and 2026 figures are still filling in because publication typically lags filing by about 18 months. The honest read is a real pullback from a 2022 high, not a technology in decline, since process and integration challenges around the material remain active research areas.
Hafnium zirconium oxide is a thin-film dielectric that, in specific orthorhombic crystalline phases, exhibits ferroelectric switching compatible with standard CMOS deposition processes — unlike older ferroelectric materials such as lead-zirconate-titanate, which do not scale well below certain thicknesses. This scalability is why the search terms in this dataset centre on phase control, coercive field and thickness scaling: these are the specific technical levers filers are patenting to make the material commercially viable at advanced nodes.
The wake-up effect describes an increase in a ferroelectric hafnium-oxide device's remanent polarization over its first several switching cycles, before it stabilizes or begins to fatigue. It is one of the field's best-documented reliability problems, which is why it appears as a search term alongside polarization fatigue and depolarization field in this dataset. Patents addressing it typically claim specific doping profiles, electrode materials or annealing schedules intended to shorten or eliminate the wake-up period before deployment in production memory.
Filing density is comparatively thin in interfacial layer engineering at the electrode-to-hafnium-oxide boundary, in wake-up-free doping schedules, and in three-dimensional vertical FeFET integration schemes, relative to the dense claim activity on the base dielectric stack itself. These are technically specific gaps rather than broad categories, and they sit adjacent to — not inside — the crowded core covered by the leading five assignees' 40.9% share of records. A first claim in these branches would need to specify the exact layer stack, doping species or integration geometry to clear the existing prior art.
Go past this page: query the whole ferroelectric memory and hafnium oxide 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.
Help us improve
Something wrong, out of date, or missing on this page? Tell us and it goes straight to the team that maintains it.
This page