Silicon Photonics Modulator Patents: Leaders, Trends & Gaps 2026
- Filing has cooled since 2022. The peak year in this dataset is 2022 at 10 families, and counts have not returned to that level since.
- Germanium photodetector claims dominate citation influence. The five most-cited records in the corpus are all germanium-on-silicon detector architectures, several dating to the early 2010s.
- H01L carries the bulk of the claim weight. 60 of 73 records sit in H01L (semiconductor devices), versus 25 in G02B and just 3 in G02F, the class nominally dedicated to optical modulation.
What this dataset covers
This landscape draws on 73 patent families published between 2015 and mid-2026 that combine silicon photonic modulator, germanium photodetector, or silicon photonics device terminology with claim-level language on modulation bandwidth, insertion loss, thermal sensitivity, dark current, or CMOS compatibility, filtered to IPC classes G02F1, H01L31, and G02B6. The scope deliberately centres on the CMOS-integration boundary: where photonic function meets standard semiconductor processing.
Publication lags filing by roughly eighteen months, so 2025 and 2026 counts in the trend chart are undercounts, not a real drop-off. The 2022 peak is the most reliable recent read on filing intent.
Filing trend and technology composition
Two views of the same 73 families: how filing has moved year over year, and which IPC subclasses carry the claim density.
A flat-to-declining filing curve
Filings rose from 6 in 2017 to a peak of 10 in 2022, then eased off. With a midpoint year at the peak, the shape reads as flat-to-declining rather than accelerating — consistent with a field where the core architectures were claimed early and later activity is incremental.
Semiconductor device claims outweigh dedicated optics classes
H01L (semiconductor devices) accounts for 60 of 73 records, more than double G02B (optical elements and systems) at 25, and twenty times G02F (optical control and modulation) at 3. G01S (radar, sonar and positioning) shows up with 13 records, pointing to LiDAR and ranging applications sharing claim territory with photonic detector structures. The imbalance suggests most protection is being built around the semiconductor fabrication of the device, not the optical modulation function itself.
Shares are the percentage of the 73 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Silicon Photonics Modulators and Detectors with Eureka
This page is one run against one query. Ask Eureka your own question about silicon photonics modulators and detectors and every answer comes back with the patent numbers behind it.
Try EurekaThe most-cited records in the corpus
Integrated germanium photodetector with silicon nitride launch waveguide
The filing describes a germanium photodetector coupled to a silicon nitride launch waveguide rather than a silicon waveguide directly: a silicon layer carries a germanium layer, contacted through conductive vias and metal contacts, with the silicon nitride waveguide extending over the silicon layer to form a coupling region into the germanium absorber.Abstract condensed from the published filing; see the full record for claim scope.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | CN106328751A | 硅基锗光电探测器 | 42 |
| 2 | US8178382B2 | Suspended germanium photodetector for silicon waveguide | 42 |
| 3 | US20140027826A1 | GERMANIUM PHOTODETECTOR SCHOTTKY CONTACT FOR INTEGRATION WITH CMOS AND Si NANOPHOTONICS | 40 |
| 4 | US20100038736A1 | Suspended germanium photodetector for silicon waveguide | 31 |
| 5 | US20120001283A1 | Germanium Photodetector | 26 |
| 6 | US7902620B2 | Suspended germanium photodetector for silicon waveguide | 25 |
| 7 | US20200124791A1 | Germanium photodetector coupled to a waveguide | 24 |
| 8 | US20120288992A1 | Germanium Photodetector | 24 |
| 9 | WO2017038072A1 | Photodetector | 23 |
| 10 | US20140134789A1 | Germanium Photodetector | 18 |
Citation counts are drawn from a searched corpus and skew toward older filings; treat them as a signal of influence on later work, not of current commercial 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 say about where this field stands
Three patterns fall out of the filing trend, the IPC mix, and the citation table together.
Growth already happened
The filing curve peaked in 2022 and has not exceeded that level since. For a technology area this narrow, a flat-to-declining curve after a single peak year usually means the foundational architectures — waveguide-coupled germanium absorption, CMOS-compatible modulator stacks — are already claimed, and new filings are refining rather than founding.
Fabrication claims outnumber optical-function claims
Semiconductor device structure (H01L) carries the large majority of claim volume, while G02F — the class most directly tied to optical modulation — has only 3 records. That gap implies the contested ground is how the device is built into a silicon process, not how the modulation effect itself is achieved.
Older germanium detector filings anchor the field
The most-cited records in the corpus are all germanium photodetector designs, several from the early 2010s, cited up to 42 times. New entrants building detector claims are almost certainly citing, or designing around, this small cluster.
Collaboration is limited and academic-industrial
Only 10 co-assignee pairs appear across the dataset, and the strongest pair links a university and a microelectronics research centre rather than two commercial rivals. Most filings in this space are single-assignee.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to silicon photonics modulators and detectors, with the prior art for and against each one.
Who holds the claim space, and where the gate sits
Recent-year momentum across the named assignees in this dataset is uniformly flat: none of the leading holders show new filings in the latest year, reinforcing the flat-to-declining trend read from the aggregate curve.
A concentrated but not runaway field
With 73 families across roughly a decade, this is a moderately sized landscape. Filing offices skew heavily to the United States (46 records), with EPO, WIPO, Japan, Canada and China each contributing single digits.
No leading assignee is currently active
Every one of the assignees tracked for recent-year momentum — spanning IBM, NTT, Cisco, GlobalFoundries, Ghent University and Aurora — shows zero filings in the latest year of the dataset. That is consistent with the field cooling from its 2022 peak rather than any single company pulling back.
The densest link is academic-industrial
The strongest co-assignee pair in the dataset connects Ghent University with imec (the interuniversity microelectronics research centre), sharing 4 families. IBM appears in two further pairs with named individual inventors, suggesting internally-driven rather than jointly-owned R&D.
| Assignee | Recent year | YoY |
|---|---|---|
| IBM | 0 | — |
| Nippon Telegraph and Telephone Corporation (NTT) | 0 | — |
| Aurora Operations, Inc. | 0 | — |
| Cisco Technology, Inc. | 0 | — |
| GlobalFoundries Inc. | 0 | — |
| Ghent University | 0 | — |
| Interuniversity Microelectronics Centre (imec) | 0 | — |
| Elenion Technologies | 0 | — |
Where to take this analysis
The aggregate numbers point to specific follow-up questions rather than a single conclusion.
Check freedom-to-operate against the germanium detector cluster
The highest-cited records in this corpus are concentrated in a small set of germanium-on-silicon detector architectures. Any new detector filing should be checked against that cluster specifically, not just the broader H01L class.
Run a claim chart in EurekaWatch the G02F gap
Only 3 of 73 records sit in the class dedicated to optical modulation itself. That could mean the modulation function is being claimed as part of broader device structures elsewhere, or that it is genuinely under-filed — worth a targeted search before assuming either.
Explore G02F filings in EurekaRe-test momentum after the publication lag clears
Because 2025 and 2026 filings are still publishing, the apparent post-2022 decline should be re-checked in twelve to eighteen months once the record set is complete.
Set a monitoring alert in EurekaCommon questions on this landscape
The dataset's leading assignees by family count include IBM, NTT, Cisco, GlobalFoundries, Ghent University and Aurora, alongside imec as a frequent academic-industrial co-filer. None of these assignees show filings in the most recent year tracked, which is consistent with the broader flat-to-declining trend rather than any single company exiting the space. Ghent University and imec together form the strongest co-assignee pair in the corpus, with four shared families, pointing to a sustained joint research programme rather than one-off collaboration.
Based on this dataset, no — filings rose from 6 in 2017 to a peak of 10 in 2022 and have not exceeded that level since. Because publication lags filing by around eighteen months, the very latest years will always look lower than they eventually turn out to be, so a firm read on 2025-2026 needs another year or two of data. But the pattern through the confirmed years is flat-to-declining, not accelerating.
The five most-cited records in this corpus are all germanium-on-silicon photodetector designs, several dating to the early 2010s, with citation counts up to 42. Citation counts inside a searched corpus naturally favour older filings because they have had more time to accumulate citing references, so this reflects historical influence on the field's foundational architecture rather than current commercial dominance. Newer filings building detector claims are very likely citing, or designing around, this cluster.
60 of the 73 records sit in H01L, the general semiconductor device class, compared with only 3 in G02F, the class specifically covering optical modulation. That gap suggests most patent protection in this field is built around how the photonic device is fabricated into a silicon process — layer stacks, contacts, waveguide coupling — rather than around the modulation mechanism itself. A filing strategy focused narrowly on modulation physics may be entering thinner-claimed ground than one focused on integration structure.
The IPC composition points to silicon nitride waveguide coupling, thermal sensitivity compensation, and dedicated modulation-effect claims under G02F as comparatively thin relative to the dense H01L fabrication and germanium-detector clusters. The G01S overlap — 13 records tying photodetector structures to radar and ranging applications — also suggests LiDAR-oriented integration is an active but not yet saturated adjacency. Any of these should still be checked against the specific highly-cited records before filing, since thin IPC counts do not guarantee an open claim, only a less crowded one.
Research Silicon Photonics Modulators and Detectors in depth with Eureka
Go past this page: query the whole silicon photonics modulators and detectors 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.