US 8680591

I'll search for authoritative information on this specific patent number, including any recent litigation activity.

Let me check specifically for any 2026 CAFC or litigation activity, and for the claim text.

US Patent 8,680,591 — Verified Summary

Important caveat up front: I was able to verify the bibliographic data below against Google Patents, FreePatentsOnline, and Justia, and it is consistent across all three. However, I could not retrieve the verbatim, numbered claim text (with claim numbers and exact wording) from any source in this session. My "plain-language overview of independent claims" is therefore reconstructed from the patent's own Summary of the Invention and description, plus the claim count I could corroborate from PTAB records. Treat the claim overview as high-confidence in substance but not verbatim.

Bibliographic data (verified)

Abstract (as published)

"Photosensitive devices and associated methods are provided. In one aspect, for example, a photosensitive imager device can include a semiconductor substrate having multiple doped regions forming at least one junction, a textured region coupled to the semiconductor substrate and positioned to interact with electromagnetic radiation, and an electrical transfer element coupled to the semiconductor substrate and operable to transfer an electrical signal from the at least one junction. In one aspect, the textured region is operable to facilitate generation of an electrical signal from the detection of infrared electromagnetic radiation. In another aspect, interacting with electromagnetic radiation further includes increasing the semiconductor substrate's effective absorption wavelength as compared to a semiconductor substrate lacking a textured region."

Plain-language overview of the independent claims

The patent was challenged in IPR2016-01910 on claims 1–26, so the document contains at least 26 claims. Based on the specification, the independent claims fall into the following families:

1. Independent device claim (photosensitive imager/pixel). A semiconductor substrate with multiple doped regions forming at least one p-n (or p-i-n) junction; a textured region coupled to the substrate and positioned to interact with incident electromagnetic radiation (the laser-processed "black silicon"-type disordered surface, or an etched/nanoimprinted equivalent); and an electrical transfer element (transistor, sensing node, transfer gate, or combination) coupled to the substrate and operable to transfer an electrical signal from the junction. The textured region is described as increasing the substrate's effective absorption length and enabling detection of infrared radiation (including >1100 nm in silicon).

2. Independent method claim (making the imager). Forming a textured region on a semiconductor substrate that has multiple doped regions forming at least one junction, with the textured region positioned to interact with electromagnetic radiation; then coupling an electrical transfer element to the substrate so that it can transfer an electrical signal from the junction.

3. Independent device claim with a ≥4-transistor requirement. Same semiconductor-substrate/doped-region/textured-region combination, but with at least 4 transistors coupled to the substrate, at least one of which is electrically coupled to the junction (i.e., the 4T active-pixel-sensor architecture).

4. Optional tuning-related subject matter (may appear as an independent claim or as dependent claims): the textured region is configured to selectively absorb, diffuse, or filter a specific wavelength, by way of feature size/dimension, material type/thickness, dopant type and profile, texture placement, and substrate material/thickness.

Representative dependent subject matter I can confirm from the description: textured region opposite vs. adjacent the doped regions; sloping/pyramidal/inverted-pyramidal/spherical/parabolic morphologies; micron- and/or nano-scale features (cones, pillars, pyramids, microlenses, quantum dots, inverted features); texturing by lasing, chemical etching, nanoimprinting, or deposition; a reflective layer; a lens/µlens; trench isolation; low-oxygen-content substrate; and annealing at specified temperatures.

Litigation / PTAB history (verified)

• PTAB IPR2016-01910 — Hamamatsu Corporation v. SiOnyx, LLC, filed Sept. 30, 2016; institution decision Mar. 30, 2017; Final Written Decision Mar. 28, 2018 (Judges Braden, Clements, Ullagaddi; opinion by Ullagaddi). Claims challenged: 1–26. DocketAlarm/UnifiedPatents list the challenged patent as 08680591 with matching inventors.
• District Court (D. Mass.) — SiOnyx, LLC v. Hamamatsu Corp., 1:15-cv-13488-FDS (2015).
• Federal Circuit — Appeal No. 2018-2019 (Sionyx LLC v. Hamamatsu Corporation), argued by Belanger (Pepper Hamilton) for appellant and Simmons (Panitch Schwarze) for appellee; decided April 3, 2019, AFFIRMED (nonprecedential per curiam; Dyk, Clewenger, O'Malley).

On "CAFC 2026 dockets"

I could not confirm any 2026 CAFC docket activity involving US 8,680,591. The only Federal Circuit case I can verify for this patent is the 2018-2019 appeal, terminated/decided in April 2019. I did find ongoing, unrelated SiOnyx-family litigation/PTAB activity around other patents (e.g., IPR2025-00845 referencing U.S. Patent 10,224,359, with Samsung/Apple/Sony and SiOnyx), but those proceedings concern different patent numbers and should not be attributed to 8680591. If you have a specific 2026 CAFC docket number or caption you're matching against, I can attempt to verify it directly — but I will not guess.

Uncertainty flags:

• Verbatim claim language and the exact independent/dependent claim numbering: not verified in this session.
• Whether the "tuning" subject matter is a separate independent claim or only dependent claims: not verified.
• Current assignee of record: sources differ between "Sionyx, Inc." and "SIONYX, LLC"; the assignment record shows an entity name change from SIONYX, INC. to SIONYX, LLC in 2016.

I'll research litigation involving US Patent 8680591. Let me search multiple sources.

Let me search litigation-specific databases.

Let me dig deeper into the specific cases and confirm which patents were asserted.

Litigation involving U.S. Patent No. 8,680,591

Important caveat on scope: The 8680591 patent is in the SiOnyx "black silicon" / photosensitive-imaging family. I confirmed the proceedings below by searching public sources (Google Patents litigation links, Justia, CourtListener, law-firm case notes, PTAB/ITC documents, and docket aggregators). I could not directly query PACER or Unified Patents' case list in this session, so treat docket-level details (especially exact filing dates) as "as reported by secondary sources." I did not return results for any similar-but-different patent number.

1. District court case — SiOnyx, LLC v. Hamamatsu Photonics K.K., et al.

This is the only district court case I could confirm in which the '591 patent was actually asserted. The court's summary judgment opinion states plainly: "SiOnyx owns U.S. Patent No. 8,680,591, which it asserts in this lawsuit."

Outcome / status:

• Jury verdict (2019): jury found Hamamatsu breached the NDA; found James Carey a co-inventor of Hamamatsu's disputed U.S. patents; and found willful infringement of the '467 patent. (Note: the reported verdict summaries center on the '467 patent; I could not independently confirm a specific jury finding of infringement of the '591 patent itself, so I flag that as uncertain.)
• Post-trial orders (July 25, 2019): permanent injunction; ownership of the disputed U.S. patents transferred to SiOnyx, LLC; injunctions against HPK/HC practicing the disputed patents or infringing the '467 patent; accounting ordered; pre-judgment interest awarded (reported amounts ≈ $1,091,481 on the breach-of-contract damages and ≈ $660,536 on unjust enrichment).
• Appeal: see §4 below. District court decisions in this case also produced the widely cited IPR-estoppel ruling (SiOnyx, LLC v. Hamamatsu Photonics K.K., 330 F. Supp. 3d 574 / 332 F. Supp. 3d 365 (D. Mass. 2018)).

2. PTAB inter partes review — Hamamatsu Corporation v. SiOnyx, LLC

Outcome: Final Written Decision issued March 28, 2018 (panel included M. Ullagaddi, G. Braden, M. Clements). The record I retrieved lists no claims as invalidated, and the patent remained active after this proceeding; the decision was appealed and affirmed (see §3). I do not have the verbatim holding of the FWD in front of me, so I am not asserting a precise "all claims upheld" conclusion — only that no claims are shown as cancelled and the appeal resulted in affirmance.

3. Federal Circuit appeal of the IPR — Sionyx, LLC v. Hamamatsu Corporation

4. Federal Circuit appeal of the district court case — SiOnyx LLC v. Hamamatsu Photonics K.K.

5. Related 2024–2025 cases — the '591 patent is NOT asserted in these (context only)

Because these are frequently confused with the '591 litigation, note that the newer SiOnyx campaigns involve later family patents, not 8,680,591:

• SiOnyx, LLC v. Apple, Inc., No. 1:24-cv-12354 (D. Mass., filed Sept. 12, 2024) — asserts U.S. 9,064,764 and U.S. 10,224,359 (pixel isolation), not the '591 patent.
• SiOnyx, LLC v. Samsung Electronics Co., Ltd., et al., No. 2:24-cv-00291-JRG (E.D. Tex., filed Apr. 30, 2024; status "Closed" as of May 21, 2025) — asserts 9,064,764; 9,905,599; 10,224,359; 11,069,737; 11,721,714 — not the '591 patent.
• SiOnyx, LLC v. [Samsung Electronics Co.](/litigations/by-defendant/Samsung%20Electronics%20Co.), Ltd., No. 2:24-cv-00408 (E.D. Tex.) — asserts 10,347,682.
• ITC Investigation No. 337-TA-1403, "Certain Sensors with Pixels and Products Containing the Same" (instituted June 5, 2024; respondents Samsung entities) — asserts 9,064,764; 9,905,599; 10,224,359; 11,069,737; 11,721,714 — not the '591 patent.

In the Samsung IPRs (e.g., IPR2025-00064 / IPR2025-00065 re 10,347,682), the '591 patent appears only as a cited prior-art reference ("U.S. Patent No. 8,680,591, Haddad138 Issued Patent"), i.e., as evidence, not as an asserted patent.

Summary

• Yes — litigation involving U.S. 8,680,591 is known. It was asserted by SiOnyx, LLC against Hamamatsu Photonics K.K., Hamamatsu Corporation, and Ocean Optics, Inc. in D. Mass. Case No. 1:15-cv-13488-FDS (filed 2015), together with Harvard's '467 and '446 patents and NDA-based contract claims. The case produced a 2019 jury verdict and post-trial injunctions/ownership transfer, and the patent was also challenged in IPR2016-01910 (FWD Mar. 28, 2018), which was affirmed by the Federal Circuit in No. 2018-2019 (Apr. 3, 2019). The district case was appealed and decided in No. 19-2359 (Fed. Cir. Dec. 7, 2020).
• The '591 patent is not among the patents asserted in the 2024 Apple, Samsung, or ITC matters.

Confidence note: Case parties, court, case numbers, and the identification of the '591 patent as an asserted patent are well grounded in the sources above. The exact filing date of the 2015 D. Mass. case and the precise holdings of the PTAB Final Written Decision and the Federal Circuit's Rule 36 affirmance are stated at the level the sources support; if you need docket-exact dates and the verbatim FWD disposition, those should be pulled directly from PACER and the PTAB (PTACTS) docket.

I'll start by reconciling the structured ODP block (which reports no proceedings) against the patent record itself, which flags a PTAB case, then search for the details.

Found the proceeding. Now let me dig into the merits — institution scope, FWD claim-level outcome, and any companion petitions.

Strong data. Let me pull the actual Final Written Decision and check for any companion petitions or later proceedings.

Let me verify the claim dependencies of the surviving claims before drawing conclusions about their viability.

I'll research the assignment chain and litigation history for this patent before constructing the output.

Excellent — I have litigation context. Now let me dig into the assignment record specifics and SiOnyx's corporate status.

Now let me verify the assignment reel/frame details and the litigation/appeal outcomes.

Assignment & Ownership Analysis — US 8680591 B2

Patent: US 8680591 B2, "Photosensitive imaging devices and associated methods"
App. No. 12/885,158 · filed 2010‑09‑17 · issued 2014‑03‑25 · priority 2009‑09‑17 (provisional 61/243,434)
Source of record: Google Patents US8680591B2 legal‑events table (fetched 2026‑09‑28)

Retrieval caveat up front. The Google Patents legal‑events table exposes the conveyance type, parties, and event dates but not the reel/frame numbers or the correspondent of record. I was unable to reach the USPTO Assignment Center / assignment‑api detail pages in this session, so I have not invented reel/frame values. Every place a reel/frame or correspondent would go is marked [not retrieved] rather than guessed. The two events below are corroborated by the patent record; the missing fields should be pulled directly at the verification link in the Verdict section.

Inventors

Employer at time of filing: determinable as SiOnyx, Inc. for all six, because each inventor executed an assignment of rights to SIONYX, INC. (recorded 2010‑11‑29 — see timeline). Five of six reside in the Portland, OR metro; the sixth in Corvallis. That geographic clustering is consistent with a SiOnyx design center, not with independent inventors.

Unusual patterns: none of the departure‑driven fire‑sale pattern (no evidence any inventor left within 12 months of filing; all six signed over rights before publication). Note that SiOnyx's founding technology traces to Eric Mazur's Harvard "black silicon" work, and that a separate group of SiOnyx‑related patents was disputed with Hamamatsu on co‑inventorship grounds (Harvard's Mazur and James Carey). That inventorship fight concerns the Hamamatsu/Harvard patents in suit, not the six inventors listed on the '591 patent.

Original assignee

SIONYX, INC. (Beverly, Massachusetts), original assignee as printed on the face of the patent.

• Primary line of business: silicon‑based photonics — ultra‑low‑light CMOS image sensors and digital night‑vision camera systems (founded 2006; ~45–51 employees; HQ Beverly, MA). Funded by Polaris Venture Partners, Crosslink Capital, Coherent, Vulcan Capital, In‑Q‑Tel, 180 Degree Capital.
• Did they ship a product embodying the claims? Yes. Commercial products include the Aurora consumer digital night‑vision camera, the OPSIN DNVM1 helmet‑mounted color night‑vision monocular (2022), the NIGHTSCAN low‑light IP surveillance camera (2024), plus OEM sensor/camera modules. Public‑sector traction includes a ~$20M U.S. Army award (announced Jan 2019) tied to the IVAS digital night‑vision program. "[P]illar features," backside texturing, and laser‑textured backside absorption described in the '591 specification are the core of this product line.
• Current status: operating, not dissolved and not in bankruptcy. Corporate form changed from SIONYX, INC. to SIONYX, LLC (recorded change of name 2016‑01‑06). The operating business was acquired by DC Capital Partners (announced Aug 2021 per PitchBook/Global Legal Chronicle). It remains active (registered legal name "Sionyx Llc"; trademark filings through 2024) and made its own acquisition of Amigen (Aug 2023). No Chapter 7 or Chapter 11 proceeding surfaced in any source reviewed.

Assignment timeline

Two recorded ownership events exist for this patent. No assignment to any third‑party NPE, aggregator, or licensing LLC appears anywhere in the chain.

• Recorded 2010‑11‑29 (execution date not distinguishable from this source; app. filed 2010‑09‑17) — Reel [not retrieved]/Frame [not retrieved]

  • Conveyance: Assignment (inventor → company)
  • Assignor: Haddad, Homayoon; Jiang, Jutao; McKee, Jeffrey; Miller, Drake; Palsule, Chintamani; Forbes, Leonard
  • Assignee: SIONYX, INC. (Beverly, MA)
  • Correspondent: [not retrieved] — Google Patents does not expose this field.
  • Context: Original employment/obligation assignment of all right, title and interest to the operating company; normal startup practice, recorded ~10 weeks after filing.

• Executed 2016‑01‑06 / recorded 2016‑01‑06 — Reel [not retrieved]/Frame [not retrieved]

  • Conveyance: CHANGE OF NAME (per the USPTO brief field quoted by Google Patents: "CHANGE OF NAME (SEE DOCUMENT FOR DETAILS)")
  • Assignor: SIONYX, INC.
  • Assignee: SIONYX, LLC
  • Correspondent: [not retrieved]. Adjacent data point, not an assignment correspondent: a Power of Attorney filing for SiOnyx dated 2016‑01‑13 was signed by Andrew W. Schultz (Reg. No. 66,869) and cc'd to Pepper Hamilton LLP / SIONYX LLC — i.e., Pepper Hamilton was SiOnyx's prosecution/agency firm at the exact moment the name change was recorded. This is corroborative context for who handled the 2016 paperwork, not proof of the assignment correspondent of record, and it is a single appearance, so it does not meet the recurrence threshold for signal 3.
  • Context: Internal reorganization / change of name only — the operating entity re‑registered as an LLC. No change in beneficial owner, no consideration, no third party.

Subsequent corporate events with NO recorded USPTO assignment: the DC Capital Partners buyout (Aug 2021) generated no patent assignment on the record, which is the signature of an equity/stock purchase rather than an asset or portfolio transfer — the LLC owner of record simply stayed the same. The '591 patent is one member of a large SiOnyx family (US8476681, US9673243, US9911781, US10361232, US10741399) that moves together.

Timeline diagram

timeline
    title Ownership of US 8680591
    2009 : Priority application filed
    2010 : Application filed 17 Sep
         : Inventors assign to SIONYX INC
    2014 : Patent issues 25 Mar
    2015 : SIONYX sues Hamamatsu
    2016 : Change of name to SIONYX LLC
         : Hamamatsu files IPR2016-01910
    2018 : IPR decision partial invalidation
    2021 : DC Capital buys SIONYX business
    2024 : SIONYX LLC sues Samsung

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
The only recorded transfer is SIONYX, INC. → SIONYX, LLC (2016‑01‑06), and the conveyance field is literally CHANGE OF NAME, not an assignment for value. The resulting LLC is not a single‑purpose licensing vehicle: it has ~45–51 employees in Beverly, MA, sells the Aurora/OPSIN/NIGHTSCAN hardware, and holds DoD contracts. No "IP / Patents / Licensing / Holdings / Ventures" successor appears anywhere in the chain.

2. Known asserter in the chain — NOT PRESENT.
Neither SIONYX, INC. nor SIONYX, LLC matches the enumerated NPE lists (Acacia, Marathon, Intellectual Ventures, IPNav, Wi‑LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). Importantly, in IPR2016‑01910 SiOnyx was the patent owner defending, with Hamamatsu as petitioner — the inverse of the NPE posture. SiOnyx's assertions are brought in its own name as an operating company.

3. Repeat correspondent across the chain — UNCLEAR (unverifiable, and no recurrence shown).
Assignment correspondents could not be retrieved, so no recurrence test can be run on the Assignment Center data. The only name‑level data point available anywhere in this chain is Pepper Hamilton LLP / Andrew W. Schultz (Reg. No. 66,869) on a Jan 13, 2016 Power of Attorney — one appearance, in a filing type that is not an assignment. Under the stated rule ("a single appearance is not a finding"), this does not register as a signal. Flagged as a data gap to close, not as a finding.

4. Cascading transfers — NOT PRESENT.
Exactly one substantive change of owner in ~16 years, and it is a name change. No chained LLCs, no shared‑address assignees, no common principals.

5. Pre-litigation transfer — NOT PRESENT.
The 2016 name change is post‑filing relative to the first suit. SiOnyx and Harvard filed D. Mass. No. 1:15‑cv‑13488 against Hamamatsu Photonics K.K., Hamamatsu Corporation and Ocean Optics on Oct 1, 2015 — before the 2016‑01‑06 recording, and under the original SIONYX, INC. identity. No assignment sits in the 6 months before any asserted suit.

6. Bankruptcy fire‑sale — NOT PRESENT.
No Chapter 7/11 filing surfaced for SIONYX, INC./LLC. The 2021 DC Capital Partners transaction is described in deal coverage as an acquisition of the operating business, and the entity continued as a going concern (it made its own acquisition of Amigen in 2023).

7. Privateering — NOT PRESENT.
No operating company transferred this patent to an NPE to assert on its behalf. SiOnyx asserts in its own name: SiOnyx, LLC v. Samsung Electronics Co., Ltd., E.D. Tex. No. 2:24‑cv‑00291 and ITC Inv. No. 337‑TA‑1403 ("Sensors with Pixels").

8. Defensive aggregator — NOT PRESENT.
The chain terminates at an operating manufacturer, not at RPX, AST, LOT Network, Unified Patents, or OIN.

Adversarial note on patent strength (not an NPE signal, but material to valuation)

This patent has been partially invalidated. In IPR2016‑01910 (petitioned by Hamamatsu, Sept 30, 2016) the PTAB's final written decision of 2018‑03‑28 held claims 1, 2, 4–11, 13–18, 21 and 23–25 unpatentable, while upholding claims 12 and 26. SiOnyx noticed appeal (Fed. Cir. No. 18‑2019); I do not have a confirmed outcome for that appeal and will not guess at one. The live claims (12, 26) are what survived. Separately, note that the '591 patent was not among the patents SiOnyx asserted against Samsung/ITC in 2024 — those proceedings concern a later family member (US 11,721,714).

Verdict

Operating-company assertion.

SIONYX, LLC is the current owner of record via a change of name only from SIONYX, INC. (executed/recorded 2016‑01‑06), not via a transfer to a licensing vehicle, and no third‑party NPE, aggregator, or shell entity appears anywhere in the chain. The owner ships products embodying the claimed textured‑silicon imaging architecture (Aurora, OPSIN DNVM1, NIGHTSCAN) and sues actual competitors — Hamamatsu (D. Mass. 1:15‑cv‑13488, filed 2015‑10‑01) and Samsung (E.D. Tex. 2:24‑cv‑00291; ITC 337‑TA‑1403). Zero of the eight NPE signals are present; one (correspondent recurrence) is merely unverifiable due to the retrieval gap noted above, which is a data limitation and not affirmative evidence of shell structure. The patent is nonetheless a weakened asset post‑IPR (claims 1–11, 13–18, 21, 23–25 invalidated 2018‑03‑28; only claims 12 and 26 survived).

Verify: USPTO Assignment Center — search by patent number 8680591 at https://assignmentcenter.uspto.gov/ (legacy interface: https://assignment.uspto.gov/patent/index.html). Pull the reel/frame and the correspondent of record for both records to close signal 3. Supporting public records: Google Patents legal events, PTAB IPR2016‑01910, D. Mass. 1:15‑cv‑13488, Fed. Cir. 18‑2019.

I'll research US 8680591 and its cited prior art references.

Let me get the specific "References Cited" section from the patent.

Let me get the fuller list of cited references.

Scope note and confidence caveats

Before the analysis, three honesty flags — these matter for how you should use this memo:

1. Source. I queried Google Patents, FreePatentsOnline, Espacenet, and Justia, which mirror USPTO/EPO bibliographic data. I could not complete a direct USPTO PatentCenter/Patent Public Search retrieval of the "References Cited" face page before exhausting my search budget. What follows is a reconstruction.
2. The cited-reference list is incomplete. The FreePatentsOnline listing for US 8,680,591 was truncated (it cut off at "780019…"). I have roughly the first third of the "U.S. Patent References" table, plus the references expressly incorporated by reference inside the specification. The "Foreign Patent Documents" and "Other Publications" sections were not retrieved at all.
3. A red flag on the retrieved list. Several items in the retrieved list carry publication dates after this patent's 2009‑09‑17 priority date (e.g., US 2013/0082343, US 2012/0313205, US 2010/0300505, US 2010/0290668). Those cannot be § 102 prior art to this patent as filed; they are either (a) applicant's own family members, or (b) references pulled in during prosecution of later continuations in this family (US 9,673,243 / US 10,361,232) and mixed into the FPO display. I flag each below rather than silently treating them as prior art.

1. Patent identification (literal)

Claim architecture (as disclosed in the specification's Summary and the Google Patents claim-term extraction): the independent claims fall into three families — (i) a photosensitive imager/pixel device comprising a semiconductor substrate with multiple doped regions forming at least one junction, a textured region coupled to the substrate and positioned to interact with electromagnetic radiation, and an electrical transfer element (transistor / sensing node / transfer gate) operable to transfer an electrical signal from the junction; (ii) a method of making the same, including laser irradiating to texture; and (iii) a device with at least 4 transistors, at least one electrically coupled to the junction. Dependent terms appearing in the claim set include chemical etching, deposition, quantum dots/nanoparticles, trench isolation, annealing, and silicon. I do not have verbatim claim text, so claim-number assignments below are by family, not by numeral.

2. Prior art expressly incorporated by reference inside the specification (highest relevance)

These are the strongest § 102/§ 103 candidates because the patent itself concedes they describe the core laser-texturing technology:

§ 102 assessment (claims 1-family and method family). Each of these teaches the textured region element and the laser-texturing method step. None, on the face of it, teaches an "electrical transfer element coupled to the semiconductor substrate and operable to transfer an electrical signal from the at least one junction" (transistor/transfer gate/sensing node) — that is the element that distinguishes this patent from the Harvard texturing art. So these are § 103 references (combined with a CMOS imager reference), not clean § 102 anticipators, absent a showing that one of them discloses an integrated transfer element.

3. U.S. patent references retrieved (partial, reverse-chronological as displayed)

(Rows 6, 8, 12, 14, 20, 21 have grant dates after the priority date but may still qualify as § 102(e) prior art based on their filing dates — I have not verified those filing dates and you should not rely on my date assumptions.)

4. Foreign patent documents and non-patent literature

Not retrieved. The face-page "Foreign Patent Documents" and "Other Publications" sections of US 8,680,591 were not recoverable in this session. Based on the sibling family members, I would expect WO 2011/035188 (SiOnyx) to appear, but that is the applicant's own PCT — not prior art — and I am listing it only as an expectation, not a verified citation. Treat this section as open.

5. Bottom-line § 102 assessment

• No retrieved reference appears to anticipate the independent device claim (claim 1 family) in its entirety. Anticipation requires a single reference disclosing (a) a semiconductor substrate with multiple doped regions forming a junction, (b) a textured region interacting with EM radiation, and (c) an electrical transfer element. The Harvard/Mazur-Carey texturing references (US 7,057,256; 7,354,792; 7,442,629; 7,884,446; 7,884,439; 7,816,220; 7,847,253) supply (a) and (b) but not (c). The image-sensor references (US 8,164,126; 7,875,948; 7,847,326; 8,058,615; 8,030,726; US 2011/0073976) supply (a) and (c) but not (b). That is a classic § 103 fact pattern, which is consistent with the patent surviving to grant and with an IPR being filed on it.
• The method claims (laser irradiating to form a textured region + coupling a transfer element) are the most exposed — US 7,057,256 / 7,354,792 / 7,442,629 / 8,207,051 teach the irradiating/texturing step, and the "coupling a transfer element" step is arguably a conventional CMOS integration step.
• The "at least 4 transistors" claim family (3T/4T APS) is the least exposed to the retrieved art; no retrieved reference pairs a textured region with a ≥4‑transistor pixel.
• § 102(e) caveat: For the post‑2009‑publication references (rows 6, 8, 12, 14, 17, 20, 21, 26), § 102(e) prior art status turns entirely on their filing dates, which I did not verify. Do not treat the "date shown" column as a priority date.

6. What I could not confirm

• The complete, verbatim "References Cited" list on the face of US 8,680,591 (U.S., foreign, and NPL).
• Verbatim claim text and exact claim numbering.
• The substantive outcome of IPR2016‑01910 beyond the fact that a Final Written Decision issued.
• Filing dates for the § 102(e) candidates in rows 6, 8, 12, 14, 17, 20, 21, and 26.

If you can supply the USPTO PatentCenter "References Cited" PDF for US 12/885,158, I can complete the per-reference § 102 mapping with exact claim numerals rather than family-level groupings.

I'll research the prior art and litigation context for US 8,680,591 before analyzing.

Let me pull the claim language and the outcome of the IPR/Federal Circuit appeal to ground the analysis.

Let me get the complete claim set from the IPR record so the analysis maps limitations to references precisely.

Obviousness Analysis — U.S. Patent No. 8,680,591 (35 U.S.C. § 103)

1. Framing: law, date, and file wrapper context

Statute. The '591 patent (Application No. 12/885,158) was filed September 17, 2010, with earliest priority to provisional applications 61/243,434 (Sept. 17, 2009); 61/311,004 and 61/311,107 (both Mar. 5, 2010). Because the filing date precedes March 16, 2013, pre‑AIA 35 U.S.C. § 103(a) governs, applied through the Graham v. John Deere framework and KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007). The controlling question is whether the differences between the claimed subject matter and the prior art would have been obvious as of Sept. 17, 2009 to a person having ordinary skill in the art ("POSITA").

POSITA. On this record (petition expert declaration, Ex. 1002, IPR2016‑01910), the level of ordinary skill is a person with a degree in electrical engineering, materials science, or applied physics and roughly 2–5 years of experience with semiconductor photodetectors and CMOS/CCD imager process integration — i.e., familiarity with (a) junction photodiodes in silicon, (b) front‑side vs. back‑side illumination (FSI/BSI) architectures, (c) pixel readout circuitry (3T/4T active pixel sensors, transfer gates), and (d) surface texturing/anti‑reflection techniques. This is my characterization of the level of skill derived from the record; it is not a verbatim quote.

Procedural posture (this matters for how to read the art). The Google Patents page for US 8,680,591 links to Family has litigation → PTAB IPR2016‑01910 (Hamamatsu Corp. v. SiOnyx, LLC, filed Sept. 30, 2016; instituted Mar. 30, 2017; Final Written Decision Mar. 28, 2018) and to Fed. Cir. No. 18‑2019 (affirmed, nonprecedential, Apr. 3, 2019). The IPR record is the authoritative source for which combinations of art were actually asserted and how they fared, and I use it below.

Prior-art set of record. The references cited on the face of the '591 patent and those asserted in the IPR give the following § 102(b) art (all published more than one year before the Sept. 17, 2009 priority date):

Also of record on the page: U.S. Pat. Nos. 7,057,256; 7,354,792; 7,442,629 (Carey/Mazur, Harvard) — expressly incorporated by reference into the '591 specification; plus US 8,164,126 (Moon) (BSI CMOS image sensors), US 7,875,948 (Hynecek) and US 7,847,326 (Park) (backside‑illuminated sensors), US 7,847,253 (Carey) and US 7,816,220 (Mazur) (laser structuring), and a set of non‑patent light‑trapping/grating references (Yamamoto, NIR Sensitivity Enhancement by Laser Treatment for Si Detectors, NIM‑A, Mar. 31, 2010; Zaidi, Diffraction Grating Structures in Solar Cells; Senoussaoui, Thin‑Film Solar Cells with Periodic Grating Coupler; Carey, Femtosecond‑Laser‑Assisted Microstructuring of Silicon Surfaces).

Important caveat on the incorporated‑by‑reference documents. Because US 7,057,256 / 7,354,792 / 7,442,629 are incorporated by reference into the '591 specification, they are properly treated as part of the '591 disclosure rather than as independent prior art against it. Their subject matter remains probative of what a POSITA knew, but the cleaner § 103 references here are the stand‑alone publications (Mazur 2003/0029495, Yamamoto 2010, etc.).

2. The claim set and what actually has to be proven

From the district court opinions in SiOnyx, LLC v. Hamamatsu Photonics K.K. (D. Mass. No. 1:15‑cv‑13488), the operative independent claims are:

Claim 1 (device):

1. a semiconductor substrate having a substantially planar surface and multiple doped regions forming at least one junction;
2. a textured region coupled to the semiconductor substrate on a surface opposite the substantially planar surface and positioned to interact with electromagnetic radiation;
3. integrated circuitry formed at the substantially planar surface; and
4. an electrical transfer element coupled to the substrate and operable to transfer an electrical signal from the at least one junction.

Claim 13 (method): forming a textured region on a substrate that has a substantially planar surface opposite the texture and multiple doped regions forming a junction; forming integrated circuitry on the substantially planar surface; coupling an electrical transfer element operable to transfer a signal from the junction.

Claim 23 (device): substrate + multiple doped regions forming a junction + textured region + at least 4 transistors, at least one electrically coupled to the junction.

Claims 12 and 26: claim 12 adds a trench isolation; claim 26 depends from claim 12 (this dependency is my inference from the ground structure below — see § 6). Claims 19 and 20 add the "forming surface features to have dimensions that selectively diffuse or selectively absorb a desired wavelength" limitation. Claim 10 adds a lens optically coupled to the substrate; claim 23 recites a transistor "electrically coupled to" a junction (Markman opinion, ECF 251).

Adjudicated outcome (IPR2016‑01910 FWD, Mar. 28, 2018): claims 1, 2, 4–11, 13–18, 21, 23–25 unpatentable; claims 12 and 26 upheld; claims 3, 19, 20, 22 not reached. Affirmed at Fed. Cir. No. 18‑2019.

3. Primary combination: Mabuchi + Mazur

This is the combination that carried the day before the PTAB (instituted and largely sustained), and it is the strongest § 103 theory against claim 1, claim 13, and claim 23.

3.1 What each reference supplies

Mabuchi (US 2003/0214595) supplies the device architecture: a solid‑state imaging device built in a semiconductor substrate with a planar processed surface, multiple doped regions forming a photodiode junction, integrated readout circuitry formed at that planar surface, and a charge‑transferring element for moving the collected signal out of the pixel. In other words, Mabuchi supplies claim elements 1, 3, and 4 of claim 1 essentially in full.

Mazur (US 2003/0029495) supplies the textured region. Mazur teaches microstructuring a silicon surface with femtosecond laser pulses to produce "cone‑like structures" ("a quasi‑ordered array of sharp conical microstructures up to fifty (50) micron high … about 0.8 micron wide near the top and up to ten (10) micron wide near the base"), expressly for the purpose of enhancing light absorption and IR response of silicon. Mazur supplies the "textured region … positioned to interact with electromagnetic radiation" limitation.

3.2 The element not supplied by either alone

Neither reference alone places the texture on the surface opposite the substantially planar (circuit‑bearing) surface. That is the only genuine gap. The combination fills it because:

• placing the texture opposite the circuitry is the only geometric option once the device is backside‑illuminated, and BSI was a well‑developed architecture by 2009 (Akahori; Hynecek US 7,875,948; Park US 7,847,326; Moon US 8,164,126 — all of record); and
• the '591 specification itself concedes the equivalence: "[t]he laser treated region can be associated with the surface nearest the impinging electromagnetic radiation, or the laser treated surface can be associated with a surface opposite in relation to impinging electromagnetic radiation" ('591 col. 11).

3.3 Motivation to combine (the KSR rationales)

1. Same field of endeavor and same problem. Both references address silicon photodetection; Mazur states the problem in the identical terms the '591 patent identifies — silicon's poor absorption of longer wavelengths — and proposes texturing as the solution. Where the secondary reference solves the very deficiency the primary reference's device suffers from, the motivation is inherent in the problem, not hindsight.
2. Predictable use of a known technique to improve a known device. KSR, 550 U.S. at 417. Texturing a silicon photosurface to reduce reflectance and lengthen the optical path is a standard, well‑understood step; the expected result (increased effective absorption length, more multiple internal reflections, higher quantum efficiency) follows predictably from Mazur.
3. Design incentive / market pressure. Demand for NIR/SWIR sensitivity in low‑light, night‑vision, security, and biometric imaging created a concrete incentive to improve IR response in standard silicon imagers rather than move to InGaAs (which the '591 background itself identifies as expensive).
4. Finite number of predictable solutions. A POSITA could place the texture on the front, back, or sides. Picking the back surface for a BSI pixel, where the front is occupied by circuitry, is the obvious selection.
5. Reasonable expectation of success. Mazur demonstrates the texturing empirically on silicon; the process is a surface treatment applied after or before device fabrication, so it does not inherently destroy the planar circuitry surface recited in claim 1.

3.4 Claim‑by‑claim result under Mabuchi + Mazur

Why the combination succeeds where a motivation‑only attack would fail. The patent owner's preliminary response argued that Petitioner "fails to establish that a person skilled in the art would have been motivated to combine Mabuchi and Mazur." The PTAB rejected that argument as to most claims, which shows the motivation showing above does carry the day for the genus claims. Where the PTAB did side with SiOnyx (claims 12, 19, 20, 26), it was because of a missing limitation, not a missing motivation — a crucial distinction.

4. Secondary combination asserted but rejected: Akahori + Mazur

This ground was asserted against claims 1–5, 7–9, 11, 13–22, 24–26, and it is instructive precisely because it failed.

• Akahori (US 6,204,506) discloses a back‑illuminated photodetector — "a back‑illuminated charge coupled device (CCD) or back‑illuminated active pixel sensor (APS)" — with a thinned semiconductor substrate, junctions, and readout circuitry. On its face this is an even better structural fit for claim 1's "textured region … on a surface opposite the substantially planar surface," because the illuminated (back) surface is by definition opposite the circuitry‑bearing surface.
• However, the PTAB declined to institute on this ground because Akahori explicitly discourages texturing on the key surface (ECF 485‑8 at 27–28). That is a teaching away, and under In re Gurley / DePuy Spine a teaching away that is directed to the very modification proposed defeats the motivation prong.

Practical takeaway: Akahori + Mazur is only viable if the petitioner can (i) show the discouragement is limited in scope (e.g., addressed to a different surface or a different reason), or (ii) rely on a different secondary reference for the back‑surface texture (Mazur in combination with a BSI reference that does not teach away, such as Hynecek US 7,875,948 or Moon US 8,164,126). As asserted, it fails.

5. Claim 6: Mabuchi + Mazur + Uematsu

Claim 6 was held unpatentable over Mabuchi + Mazur in further view of Uematsu (JP H06‑244444). The three‑way combination works because Uematsu supplies the additional structural feature recited in claim 6 (a further reflector/texture arrangement) that completes the claim without altering the principle of operation of the Mabuchi device, and the motivation to add it follows the same optical‑path‑lengthening rationale that motivates Mazur.

Disclosure of uncertainty: I was unable to retrieve the verbatim text of claim 6 from the materials reviewed. I therefore cannot state precisely which limitation Uematsu supplies; the statement above reflects the ground as it appears in the IPR record (Petition Ground 4; FWD holding claim 6 unpatentable), not my independent reading of the claim.

6. Claims 12 and 26 (trench isolation): the combination that fails

• Ground: claims 12 (and, in the related ground, 26) obvious over Mabuchi + Mazur in further view of Furukawa (US 2006/0086956).
• Outcome: the PTAB rejected the ground. SiOnyx successfully argued both (i) that Furukawa fails to disclose a trench isolation, and (ii) that Petitioner failed to establish a motivation to combine Mabuchi, Mazur, and Furukawa. The FWD accordingly upheld claims 12 and 26, and the Federal Circuit affirmed.

Analytical consequence. Claims 12 and 26 are the residual, defensible claims of the '591 patent, and a § 103 challenge to them must come from a different trench‑isolation reference. The candidates of record that were not used in the IPR are the natural place to look:

• Moon, US 8,164,126 — "CMOS image sensors including backside illumination structure" (cited on the face of the '591): BSI CMOS with isolation structures; a stronger fit than Furukawa for combining with the Mabuchi+Mazur backside‑textured pixel because it is in the same BSI pixel‑isolation context;
• Hynecek US 7,875,948 and Park US 7,847,326 — backside‑illuminated sensors with isolation;
• and the '591 specification's own discussion of "shallow (FIG. 7) or deep (FIG. 12) trench isolation" as conventional, including the express teaching that "trench isolation can include dielectric materials, reflective materials, conductive materials, and combinations thereof, including textured regions and other light diffusing features."

If a POSITA would recognize trench isolation as a standard measure "to maintain pixel to pixel uniformity by reducing optical and electrical crosstalk" — which the specification itself says — then the combination with a same‑field BSI reference supplies both the missing element and the motivation. That ground was never presented to the PTAB and would not be subject to § 315(e) estoppel (cf. the district court's estoppel analysis regarding Adkisson).

7. Claims 19 and 20 (wavelength‑selective feature sizing)

The PTAB declined to institute, and did not reach, claims 19–20, because Mabuchi and Mazur did not teach "forming surface features to have dimensions that selectively diffuse or selectively absorb a desired wavelength of electro‑magnetic radiation" (ECF 485‑8 at 37–38).

A viable § 103 case for those claims would require an additional reference teaching dimension‑selected surface features, and the page's prior‑art listings supply several:

• Yamamoto et al., "NIR Sensitivity Enhancement by Laser Treatment for Si Detectors," NIM‑A (Mar. 31, 2010) — laser treatment of Si detectors tuned for NIR response;
• Zaidi, "Diffraction Grating Structures in Solar Cells" and Senoussaoui, "Thin‑Film Solar Cells with Periodic Grating Coupler" — periodic/dimensioned grating texturing that couples specific wavelengths;
• Adkisson, US 2010/0264473 — dimensioned anti‑reflection structures for CMOS image sensors (of record but never asserted);
• Baggenstoss, US 2006/0081900 — grated/blazed interface texturing (used against a SiOnyx sibling patent in a different IPR and therefore available as a § 103 secondary reference here).

Caveat: the "tuning by feature dimension" limitation is the one place the record shows the asserted art falling short, and one of the asserted secondary references (Furukawa) was found inadequate. A claim‑19/20 theory should be treated as untested, not established.

8. Objective indicia (secondary considerations) and why they are unlikely to save the genus claims

The patent owner has factual material that could support objective indicia: SiOnyx's commercial products and Hamamatsu's adoption of laser‑textured "black silicon" photodiodes. The problem is nexus. The district court found, and the Federal Circuit affirmed, that the disputed Hamamatsu patents and products arose from SiOnyx's confidential information disclosed under the 2007 NDA — the jury found Hamamatsu breached the NDA by using that information, and SiOnyx was awarded sole ownership of the Hamamatsu U.S. patents. Evidence of an alleged copier's success that is attributable to a breach of confidence and misappropriation of the patentee's own disclosure does not establish that the claimed subject matter as a whole was non‑obvious over Mabuchi + Mazur; it tends instead to corroborate that the combination was valuable and accessible, i.e., that it would have been obvious to a POSITA who knew the Mazur and Mabuchi disclosures. At minimum, the § 103 analysis above is unaffected by this record.

9. Additional observations and limits on this analysis

1. Obviousness‑type double patenting with US 8,476,681. US 8,476,681 ("Photosensitive imaging devices and associated methods," Haddad et al., July 2, 2013) appears in the '591 patent's own "US Patent References" listing. It is a same‑family Haddad patent with overlapping disclosure. A separate, non‑§ 103‑prior‑art attack on the '591 claims could arise from obviousness‑type double patenting over the '681 claims if the two patents are not commonly owned or terminally disclaimed. This is a distinct doctrine and I flag it, not assert it.
2. Anticipation by Nakashiba was asserted but was not the basis for the surviving FWD. Nakashiba (US 2007/0237504) was asserted to anticipate claims 1, 4, 5, 7–9, 13, 24, 25, and the PTAB reversed institution as to claims 9, 24, and 25 on rehearing. It does not figure in the validity holdings that survived appeal; I therefore treat it as secondary to the Mabuchi+Mazur theory.
3. I could not verify (a) the verbatim text of claims 2–11, 14–22, and 24–26; (b) the exact publication dates of Nakashiba, Mabuchi, Mazur, and Furukawa beyond what their numbers and the IPR record indicate; and (c) the precise limitation supplied by Uematsu for claim 6. Where I have relied on the IPR record rather than the claim text, I have said so. The claim language for claims 1, 12, 13, and 23 is quoted or paraphrased from the district court opinions, which is a reliable source.

10. Bottom line

• Claims 1, 2, 4–11, 13–18, 21, 23–25: rendered obvious by Mabuchi (US 2003/0214595) in view of Mazur (US 2003/0029495), with the motivation supplied by (i) Mazur's express solution to silicon's poor long‑wavelength absorption — the same problem the '591 patent addresses, (ii) the predictable use of a known surface‑texturing technique to improve a known photodetector, and (iii) the design necessity of putting the texture on the back surface of a backside‑illuminated pixel whose front surface carries the integrated circuitry. Claims 6 and 23 are further supported by Uematsu (JP H06‑244444) and by the conventionality of 4T APS architectures, respectively. This is the combination the PTAB sustained and the Federal Circuit affirmed.
• Akahori (US 6,204,506) + Mazur is structurally attractive (real BSI disclosure) but fails on motivation because Akahori teaches away from texturing the key surface; it should be replaced by a non‑teaching‑away BSI reference if that theory is to be pursued.
• Claims 12 and 26 (trench isolation) are the defensible claims. The asserted combination with Furukawa (US 2006/0086956) failed for lack of both disclosure and motivation. Any renewed challenge must substitute a same‑field trench‑isolation reference such as Moon, US 8,164,126, Hynecek US 7,875,948, or Park US 7,847,326 — grounds that were not raised in IPR2016‑01910 and therefore are not exposed to § 315(e) estoppel.
• Claims 19 and 20 require a dimensioned/wavelength‑selective texturing reference (Yamamoto 2010, Adkisson US 2010/0264473, Baggenstoss US 2006/0081900, or the diffraction‑grating solar references); that theory is untested on this record.

Primary sources: US 8,680,591 full text and prior‑art listings — https://patents.google.com/patent/US8680591/en and https://www.freepatentsonline.com/8680591.html; IPR2016‑01910 petition and Final Written Decision — https://portal.unifiedpatents.com/ptab/case/IPR2016-01910; SiOnyx, LLC v. Hamamatsu Photonics K.K., No. 1:15‑cv‑13488‑FDS (D. Mass.) — https://www.govinfo.gov/content/pkg/USCOURTS-mad-1_15-cv-13488/pdf/USCOURTS-mad-1_15-cv-13488-6.pdf; Fed. Cir. No. 18‑2019 (affirmed Apr. 3, 2019) — https://cases.justia.com/federal/appellate-courts/cafc/18-2019/18-2019-2019-04-03.pdf.