US 9978413

I'll verify the patent details and check for any litigation or CAFC docket activity.

Let me check for any 2026 appellate activity.

US 9,978,413 B2 — Verified Summary

Search scope / confidence note. I searched Google Patents, the USPTO Patent Center/PTAB (PTACTS) public filings, PACER-derived dockets (C.D. Cal., D. Minn.), and the Stanford NPE Litigation Database. I found no Federal Circuit 2026 appeal docket naming 9,978,413. What I did find is active 2025–2026 activity at the PTAB and district-court level involving related patents in the same family — but not the '413 patent itself. Details and one genuine ambiguity are flagged below.

Bibliographic data (from the patent document itself)

Abstract (verbatim, authoritative)

"A multilayer exchange spring recording media consists of a magnetically hard magnetic storage layer strongly exchange coupled to a softer nucleation host. The strong exchange coupling can be through a coupling layer or direct. The hard magnetic storage layer has a strong perpendicular anisotropy. The nucleation host consists of one or more ferromagnetic coupled layers. For a multilayer nucleation host the anisotropy increases from layer to layer. The anisotropy in the softest layer of the nucleation host can be two times smaller than that of the hard magnetic storage layer. The lateral exchange between the grains is small. The nucleation host decreases the coercive field significantly while keeping the energy barrier of the hard layer almost unchanged. The coercive field of the total structure depends on one over number of layers in the nucleation host. The invention proposes a recording media that overcomes the writeability problem of perpendicular recording media."

Independent claims — plain-language overview

Claim 1 — the "ferromagnetically coupled bilayer" claim. A magnetic recording disk built on an essentially non-magnetic substrate, with a two-layer magnetic stack. The two layers are:

• a hard magnetic storage layer (granular, perpendicularly anisotropic) whose own coercive field is H_s, 3–30 nm thick; and
• a hard magnetic nucleation host (granular, perpendicularly anisotropic), deposited on top of the storage layer so that grains line up columnar through the stack, with its own coercive field H_n measured without the storage layer.

The core numerical relationships the claim requires:

1. 0.5 T < H_n < H_s — the host is genuinely "hard" (>0.5 T), but softer than the storage layer;
2. the host's magnetic polarization J_s = μ₀M_s is between 0.3 T and 1.0 T;
3. the two layers are separated by a coupling layer 0.1–3 nm thick, that layer sitting directly between them; and
4. the coupling between host and storage layer is ferromagnetic (not antiferromagnetic).

In plain terms: a hard disk medium with a hard magnetic layer capped by a slightly softer hard magnetic "nucleation" layer, kept in tight ferromagnetic registry through a very thin (sub-3 nm) coupling layer, so that the write head can start reversal in the host and the reversal propagates into the hard storage layer.

Claim 18 — the "angular-dependence" claim. Also a bilayer on a non-magnetic substrate, with a granular hard storage layer (coercive field H_s) and a granular hard nucleation host disposed on it columnar-ly (coercive field H_n), again requiring 0.5 T < H_n < H_s. The distinguishing limitation is the switching-field angular signature:

H_c(α = 20°) < H_c(α = 45°),

where α is the angle between the applied field and the film normal. This is the antithesis of classic Stoner–Wohlfarth behaviour (which minimizes switching field at 45°); the patent attributes it to "pinning magnet"-like domain-wall depinning, where the pinning force varies as 1/cos α. The claim also permits the layers to be either in direct contact or separated by a coupling layer under 5 nm (note: broader than claim 1's 0.1–3 nm, and claim 18 does not itself require ferromagnetic coupling).

Dependent claims (brief): claim 2 (coupling layer exchange constant A > 10⁻¹⁴ J/m); claim 3 (slope k of the normalized hysteresis loop varies by no more than a factor of 3 over −0.7 < M_z/M_s < 0.7, indicating strong coupling); claim 4 (squareness within 10% of the storage layer alone); claim 5/16 (layer compositions — e.g., CoPt, FePt, CoPtCr, CoPtCrB, CoPtCrTa, CoCr alloys); claim 6/7 (the H_c and H_m angular inequalities); claims 8–10 (host materials, 2–10 nm grain diameter, host thickness > 7 nm); claim 11 (perpendicular recording or patterned media); claim 12/25 (a list of characterizing micromagnetic parameters); claims 13/26 and 14/27 (how to determine the storage layer's coercive field and energy barrier — by depositing it alone, removing the host, or micromagnetic calculation); claims 15/28 (coercive field increases when grains are misaligned to the substrate normal); claim 23 (ferromagnetic coupling via coupling layer); claim 17 (exchange coupling enables a domain wall across the hard/host interface during reversal).

(Literal-text note: claim 8 recites "Nr" and "S" where the specification lists "Ni" and "Si"; claim 12 and 25 recite "of 1" / "of 18" without the word "claim". I report these as they appear rather than correcting them.)

Family and litigation status

Family. 9,978,413 is the original parent (app. 11/424,859). Its chain includes U.S. 9,928,864 (division, filed 2009-11-17), U.S. 11,138,997 and U.S. 11,133,031 (continuations), U.S. 12,020,734, U.S. 11,908,500, and abandoned US 2024/0079030 A1 — all sharing one specification and claiming the 2006-06-17 priority date.

⚠️ Identifier ambiguity worth flagging. In the litigation literature, "the '997 patent" almost always means U.S. 11,138,997, not 9,978,413 — and the PTAB petitions confirm the asserted August 2022 Western Digital complaint was against 9,928,864 ('864) and 11,138,997 ('997). However, the Stanford NPE Litigation Database separately records 9,978,413 as involved in MR Technologies, GMBH v. Western Digital Technologies, Inc., No. 2:22-cv-06088 (C.D. Cal.), and Google Patents' litigation metadata lists 9,978,413 in three cases: Texas Western District 6:22-cv-00612, C.D. Cal. 8:22-cv-01599, and C.D. Cal. 2:22-cv-06088. I am not confident which of these actually pleads 9,978,413 as an asserted patent versus merely listing the family. Do not treat this as settled.

Adjudicated matters (family-level, not the '413 patent):

• Jury verdict, 2024-07-26 (C.D. Cal., WD action): the '864 and '997 patents held valid and infringed; $262,388,800 lump-sum royalty; later $117,267,508 prejudgment interest, ~$46,000/day post-judgment interest. Settled May 2025 before post-trial/appellate testing. [D. Minn. order]
• Seagate action, D. Minn. No. 0:25-cv-01460 (motion to dismiss partially denied).
• Toshiba action, C.D. Cal. No. 2:25-cv-00786-JVS-DFM (filed 2025-04-15).
• Resonac DJ action, N.D. Cal. (filed 2025-10-09).

2026 PTAB activity (this is the closest thing to a "2026 docket"): Resonac Hard Disk Corp./Resonac Corp. filed IPRs against the family — IPR2026-00014 (9,928,864) and IPR2026-00016 (12,020,734), plus a related petition against 11,138,997 — with Patent Owner MR Technologies arguing discretionary denial under the March 26, 2025 PTAB workload memorandum. [PTACTS petition record]

Federal Circuit: no 2026 CAFC docket for 9,978,413 was found. The Western Digital verdict was settled before appeal, so it produced no Federal Circuit decision on the merits. I cannot rule out a recently docketed, not-yet-indexed appeal — treat the absence as "not found in the sources searched" rather than affirmative proof of none.

What the patent is really about (one paragraph)

The superparamagnetic limit: smaller grains store more data but become thermally unstable, and raising anisotropy K to fix stability also raises coercivity H_c, making the media unwritable. The patent's insight is that a graded, exchange-coupled nucleation host decouples these two quantities. A domain wall nucleates in the softer host and, during reversal, gets pinned at each anisotropy step on its way into the hard layer — so H_c drops sharply (the patent reports factors of ~4–4.5 for a bilayer, ~7 for a trilayer, up to ~10 for a quadratically graded "G-layer") while the thermal energy barrier, set by the domain-wall energy in the hardest layer, stays essentially unchanged. The specification expressly claims this as a point of departure from Victora/Shen, Inaba, Loxley, and Hagedorn, which treated coercivity and barrier as inseparable.

I'll search for litigation records specific to US patent 9978413.

Let me dig deeper into the specific cases.

Litigation Involving U.S. Patent No. 9,978,413 ("Multilayer Exchange Spring Recording Media")

Summary: US 9,978,413 (the "'413 patent") has been asserted in the HDD-media enforcement campaign of its owner, MR Technologies, GmbH ("MRT," Vienna, Austria — assignee of inventor Dieter Suess). It is a member of the same patent family as US 9,928,864 ('864), US 11,133,031 ('031), US 11,138,997 ('997), and US 11,207,334 ('734). Note that the '413 patent was asserted in the complaints below but was dropped before trial in the lead case; the verdicts were on the sibling '864 and '997 patents. I could not find any CAFC appeal decision on the '413 patent itself, and no case in which the '413 patent was the sole or primary patent reaching judgment.

Case 1 — MR Technologies, GmbH v. Western Digital Corporation (first-filed WD action)

Sources: Unified/Google Patents litigation link for 6:22-cv-00612; AO 120 report and complaint (docketalarm/CourtListener, W.D. Tex. 6:22-cv-00612); Stanford NPE database (npe.law.stanford.edu/patent/9978413).

Case 2 — MR Technologies, GmbH v. Western Digital Technologies, Inc. (the lead, tried case)

Key point for your purpose: the '413 patent did not reach the jury. MRT withdrew it. The damages verdict rested on the sibling '864 and '997 patents.

Sources: Google Patents litigation links for 8:22-cv-01599 and 2:22-cv-06088; WD SEC Form 10-K "Legal Proceedings" (June 27, 2025 filing, sec.gov); Reuters, "Western Digital owes $262 mln in hard-drive patent case, jury says" (2024-07-30); Law360 case docket "MR Technologies, GMBH v. Western Digital Technologies, Inc." (case 630d04a7b0223c787146febd); Docket Alarm 8:22-cv-01599; Pretrial Order, Dkt. 481 (July 9, 2024).

Case 3 — MR Technologies, GmbH v. Western Digital (second action)

Source: WD SEC Form 10-K "Legal Proceedings" (June 27, 2025). I could not confirm from my searches whether '413 was asserted in this second WD action, nor its case number — verify on PACER/Unified Patents.

Case 4 — MR Technologies GMBH v. Toshiba (note: '413 NOT asserted)

Sources: PTAB petition papers referencing the Toshiba action and CDCA-8-25-cv-00786 docket entries; PTACTS/USPTO petition documents 1558577.

Other related proceedings (relevant context, not '413-specific)

• Seagate action — PTAB papers refer to "the Seagate and Toshiba Actions" as MRT's serial customer suits. I was unable to confirm a Seagate case number, filing date, or whether '413 is asserted from the sources retrieved.
• Resonac declaratory-judgment action — media supplier Resonac filed a DJ action for non-infringement (asserting its media does not meet the "nucleation host" limitation), and filed IPR petitions at the PTAB (e.g., USPTO petitions 1558576/1558577/1558578) against members of this patent family. These are aimed at the '864/'997/'734 family; I did not confirm an IPR specifically on the '413 patent.
• Litigation flags on Google Patents for US 9,978,413 list three US cases: 6:22-cv-00612 (W.D. Tex.), 8:22-cv-01599 (C.D. Cal.), and 2:22-cv-06088 (C.D. Cal.) — the last two being the same case before and after intra-district transfer.

Bottom line

1. The only known litigation in which US 9,978,413 was actually asserted consists of the MRT v. Western Digital complaints: 6:22-cv-00612 (W.D. Tex., filed 6/13/2022 — voluntarily dismissed 8/26/2022) and 8:22-cv-01599 (C.D. Cal., filed 8/26/2022, originally 2:22-cv-06088).
2. In the C.D. Cal. case, MRT dropped the '413 claim before trial; the $262.3M verdict (July 26, 2024) and $117M prejudgment interest rested on '864 and '997.
3. All MRT–Western Digital disputes were settled globally for $130 million in April 2025 and dismissed, ending any live '413 dispute with WD.
4. No litigation asserting the '413 patent against Toshiba, Seagate, or Resonac is confirmed — those campaigns center on the sibling patents ('864, '997, '734).
5. No CAFC appeal, PTAB IPR, or final judgment on the validity/infringement of the '413 patent specifically was found.

Caveats: Case numbers, dates, and statuses above should be verified against PACER (C.D. Cal. 8:22-cv-01599, 8:25-cv-00786; W.D. Tex. 6:22-cv-00612) and the Unified Patents litigation portal (portal.unifiedpatents.com/litigation/caselist). I could not verify the second WD action's case number, the Seagate action's particulars, or whether '413 appears in any pending IPR — those gaps are flagged rather than filled.

I'll verify the structured ODP data against live sources before writing anything up.

Let me check the litigation-related IPRs and PTAB docket directly.

Critical finding — there do appear to be family IPRs. Let me pin down the proceeding numbers and which patents they target.

⚠️ Headline: There are no PTAB proceedings on US 9,978,413

Before the analysis, the single most important point — and it cuts against the framing of this task:

The canonical ODP/structured "PTAB proceedings on file" block returns zero AIA trial proceedings for US 9,978,413. My web searches were also unable to identify any IPR, PGR, or CBM naming the '413 patent as the challenged patent. The proceedings surfaced by search target sibling patents in the same 2006-06-17 priority family, not the '413 itself. I am flagging them below because they are highly relevant to a defendant's posture, but I will not assign them to the '413 — doing so would be fabrication.

I was also unable to confirm judge panels, FWD text, or appeal dispositions, because no FWD exists for any of these family proceedings (the earliest-institued candidate is an October 2025 filing). Everything below is flagged for confidence.

Proceedings overview

For US 9,978,413 specifically: zero PTAB proceedings — no active trials, no invalidated claims, no sustained claims, no settlements, no institution denials, because nothing was ever filed. The defensive bottom line is therefore not "the patent has survived two IPRs and is hardened," and not "claims 1-5 are canceled." It is a third posture: the '413's validity has never been adjudicated anywhere — Patent Owner MR Technologies dropped the '413 from its only Western Digital case before trial (per WD's SEC Form 10-K), and the family IPRs from Resonac target the '997 continuation.

Caveat on the ODP default: ODP ingest lags. I did search for recently-filed proceedings and found none naming the '413. But I could not exhaustively rule out an unindexed, very recent filing.

Related proceedings in the same patent family (NOT on the '413)

These are siblings, not the patent you asked about. I include them because a defendant facing the '413 needs to know what Patent Owner and its adversaries are actually fighting over.

IPR2026-00015 — Resonac Hard Disk Corp. / Resonac Corp. v. MR Technologies GmbH

• Type: Inter Partes Review. Challenged patent: US 11,138,997 (application 15/925,749, filed 2018-03-19) — a continuation in the same family as the '413, not the '413.
• Filed: 2025-10-08 (per the Petitioner's Sotera stipulation letter and PTACTS docket material).
• Status: Aggregator-reported as "Discretionary Denial," with an Institution Decision Date of 2026-02-03 (ipverse case page). Confidence: medium. I could not retrieve the underlying Board decision text, so treat "denied on discretionary grounds" as reported-but-unverified, not quoted.
• Judge panel: Not obtained. Do not assume a panel.
• Petition grounds (from Resonac's own 2026-01-08 Sotera stipulation, which enumerates its grounds for the '997):

(Only these two grounds were enumerated in the stipulation — which is not necessarily the complete petition. No § 102 or § 112 ground appears.)

• Institution decision: Reported 2026-02-03 as a discretionary (Fintiv-flavored) denial, notwithstanding that Resonac filed a Sotera stipulation on 2026-01-08 promising to drop all patents-and-printed-publications invalidity grounds in the parallel N.D. Cal. case if instituted. If the denial was in fact discretionary despite the Sotera stipulation, that is a notable data point — but I cannot confirm the reasoning.
• Final Written Decision: None. No trial was (apparently) instituted, so no FWD exists.
• Settlement / termination: None reported.
• Appeal: None; nothing appealable.
• Defensive value for the '413: Directly relevant but not dispositive. The Board's art (Takeniori, Li) was mapped against the '997's claims. Because the '413 and '997 share the 2006-06-17 priority specification, that art may be adaptable — but the '413 has a different claim set (see below), so no ground-based outcome transfers automatically.

Two–three additional family IPRs (numbers NOT confirmed)

A Resonac filing states that "three IPRs are currently awaiting the Director's decision on discretionary denial, involving three patents in the same family" and references "the '864 and '997 Patents," a "Resonac DJ Action," and a WD jury verdict (PTACTS petition filing). I can only confirm the proceeding number for IPR2026-00015 ('997). I could not independently confirm the numbers or the specific patents of the other two, and I will not guess them. A defendant should pull the Resonac IPR set from PTAB E2E before relying on this.

Why the '413 is different from its litigated siblings — and why that matters

The '413 is the 2006 parent (US 11/424,859, filed 2006-06-17, granted 2018-05-22, patent term adjusted to 2029-08-22). Its claim set as granted has 28 claims, and claim 1 is a coupling-layer claim:

"…the nucleation host and the hard magnetic storage layer are separated by a coupling layer between 0.1 nm and 3 nm thickness; the coupling layer is directly between the nucleation host and the storage layer; and the coupling between the nucleation host and the hard magnetic storage layer is ferromagnetic." (claim 1, as granted)

That is materially narrower and more structurally specific than the graded-anisotropy subject matter MRT successfully asserted on the '864 and '997. Independent claim 18 is the other independent claim, keyed to the Hc(α=20°) < Hc(α=45°) angular signature plus direct-contact-or-<5 nm coupling layer.

Strategic summary

What is canceled vs. sustained vs. untested on the '413. Nothing on the '413 is canceled, sustained, or tested. Not by the PTAB — there is no proceeding. And not by a court: MR Technologies asserted the '413 (along with the '864, '031, and '997) in MR Technologies GmbH v. Western Digital, C.D. Cal. No. 8:22-cv-01599, filed 2022-08-26, but dropped the '413 and the '031 before trial, per Western Digital's SEC Form 10-K disclosure ("MRT dropped its claims with respect to U.S. Patent Nos. 9,978,413 and 11,133,031, and the case proceeded to trial in July 2024 on the remaining two patents"). The July 2024 verdict — $262,388,800 in lump-sum damages, ~$380M with $117M prejudgment interest, entered 2024-08-15 — covers only the '864 and '997. The widely-circulated claim that "four patents were validated" is inaccurate; the jury reached two. So if a demand letter or complaint today asserts the '413, a defendant is facing a claim the patent owner has already walked away from once, in the only case where it was tested.

Estoppel landscape. Because no IPR was instituted against the '413 (and no FWD issued), no § 315(e)(2) estoppel attaches to the '413 at all. That cuts both ways. It means no one — not Western Digital, not Resonac, not any privy — is barred from raising § 102/§ 103 grounds against the '413 at the PTAB or in court. It also means the '413 is the cleanest target in the family for a fresh IPR petitioner, unencumbered by any prior institution decision or estoppel record. By contrast, the whole family is now crowded with a § 315(b) time-bar problem: anyone served with a complaint asserting the '413 more than one year ago is time-barred from filing its own IPR. That one-year clock has long since run for the 2022-dated cases; newer defendants (the 2025 Seagate and Toshiba actions) may still be within it. Also note: the Resonac Sotera stipulation never took effect, because if institution was denied, its condition precedent ("if the PTAB institutes the IPR") failed.

Pattern signals. (1) No defensive aggregator. Nothing in the record shows Unified Patents or an RPX-type entity in the chain on the '413 — the challenges are coming from a competitor/supplier, Resonac. (2) The petitioner dynamic is upstream-supplier driven. MRT has been criticized on the record for "suing customers while avoiding naming or involving" the media suppliers; Resonac's filings describe a deliberate strategy to sue downstream HDD makers (WD, Seagate, Toshiba) rather than the upstream media supplier, which prompted Resonac to file IPRs plus a parallel declaratory judgment action in N.D. Cal. (Resonac Hard Disk Corp. v. MR Technologies GmbH, No. 3:25-cv-08631). (3) No confirmed patent-owner PTAB appeal history on this family. (4) The Western Digital matter ended in a global April 2025 settlement of $130M, dismissing both the 2022 case and a second 2024-08-22 suit; WD's announced plan to appeal the verdict was mooted by the settlement, so do not assume a live Federal Circuit appeal. I could not locate any CAFC docket number.

Recommended next steps

1. If you are a defendant on the '413: attack validity, not just infringement. The '413 has never been through an IPR and never been to a jury. It is the family member with the least defensive scar tissue. Pull the Resonac '997 petition and its Takeniori/Li art as a starting point only — then re-map against the '413's narrower coupling-layer claim 1 (0.1–3 nm coupling layer, ferromagnetic, directly between the layers) and its claim-18 angular-dependence limitation.

2. Check your § 315(b) clock first. If MRT served you with a '413 complaint more than one year before your intended petition date, you are barred and must litigate validity in district court. If you were served recently, your window is open — and given the reported 2026-02-03 discretionary denial in the sister '997 IPR, file a Sotera stipulation early and cite it hard. The intervening 2025 USPTO restoration of the Iancu-era Fintiv/Sotera regime means parallel-litigation discretion is very much in play.

3. Do not quote a "claims canceled" position. There is no FWD to quote. If opposing counsel or a licensing letter implies PTAB outcomes on the '413, that is wrong — say so. Verify any asserted proceeding number against USPTO PTAB E2E / PTACTS before relying on an aggregator's status field.

4. Verify the family IPR set independently. My confirmation ceiling was IPR2026-00015 ('997). The "three IPRs… in the same family" statement in Resonac's brief is a lead, not a fact I verified patent-by-patent. If the '413 is among them, my analysis flips — confirm before you strategize.

Confidence summary: (low/medium/high)

• ODP returns no PTAB proceedings on the '413 — high (canonical structured source).
• No web-surfaced IPR names the '413 as challenged patent — medium-high (search-limited; not exhaustive).
• MRT dropped the '413 and '031 from the WD case pre-trial — high (WD SEC Form 10-K).
• $262.4M lump-sum verdict + $117M interest, later settled globally for $130M (April 2025) — high (SEC Form 10-K corroborated by trade press).
• IPR2026-00015 targets the '997, filed 2025-10-08, Resonac petitioners, grounds as tabulated — high (PTACTS-fetched Sotera letter).
• IPR2026-00015 denied institution on discretionary grounds on 2026-02-03 — medium (single aggregator status field; Board paper not retrieved).
• Existence of two additional family IPRs and their numbers — low (unconfirmed).

I could not obtain: any PTAB judge panel names, any FWD text (none exists), any Federal Circuit docket number, or the Board's actual institution-decision reasoning. Those are genuine gaps, not omissions — say so rather than filling them.

I'll research this patent's assignment chain and litigation history before assembling the report.

Let me dig into the assignment record details and the inventor's affiliations.

I hit my research step limit before I could pull the live Assignment Center record, so I'll flag precisely what is confirmed versus not retrieved rather than guess.

Assignment & Ownership Report — US 9,978,413 B2

Patent: US 9,978,413 B2 — "Multilayer exchange spring recording media"
Application: 11/424,859 · Filed: 2006-06-17 · Granted: 2018-05-22 · Adjusted expiry: 2029-08-22
Current assignee of record (Google Patents legal events): MR Technologies GmbH (Austria)

Retrieval note / data caveat: I could not open the live USPTO Assignment Center record for this patent before my research step limit was reached. The reel/frame number, the execution date (as distinct from recordation date), and the correspondent of record on the assignment cover sheet were not retrieved and are marked as such below. Everything stated as fact is sourced to the Google Patents legal-events feed and the litigation record; nothing is inferred.

Inventors

Unusual patterns:

• Single inventor, filed as an individual. The application transmittal for the family (docket SUSP02US3) shows the "Assignment Papers" box unchecked and "Applicant asserts small entity status" checked. The issued patent lists Original Assignee: Individual. Suess filed personally, not through TU Wien and not through any corporate assignee — so no university or employer took title at filing.
• No co-inventors. Unusual for a device/media patent of this scope.
• A third-party aggregator (PatentLeaderboard) lists Suess under "11 Patents at Infineon Technologies AG." That aggregation appears to mis-attribute this individual filing to Infineon; treat it as unverified. It does suggest a possible prior/parallel Infineon (Villach, Austria) affiliation worth confirming in the assignment record.
• No "all inventors departed within 12 months" pattern applies — there is only one inventor and he never assigned away during the prosecution period.

Original assignee

The original assignee on the issued patent is the inventor himself, recorded as "Individual" (Dieter Suess, personally).

• Product embodying the claims? No. Suess is an academic physicist (later Professor and head of the Physics of Functional Materials department, University of Vienna). He has never manufactured HDD media. His "business" is research and, since ~2022, patent enforcement through his own company.
• Primary line of business: academic research in micromagnetics; commercial arm is the Austrian private company MR Technologies GmbH ("MRT"), which a PTAB filing describes as "a privately held company based in Austria." MRT does not sell a product — it licenses and litigates.
• Current status: The individual inventor is active. The patent is owned by MR Technologies GmbH, active and litigating (W.D. Tex., C.D. Cal., D. Minn., N.D. Cal.). The underlying technology, by the prevailing account, was adopted industry-wide.

Assignment timeline

Only one post-issuance assignment is reflected in the Google Patents legal-events record for US 9,978,413. There is no record of any assignment at filing (2006), consistent with the transmittal form showing no assignment papers submitted.

• 2022-06-13 (date per Google Patents legal events — the record is labeled "Assigned to MR TECHNOLOGIES, GMBH … ASSIGNMENT OF ASSIGNORS INTEREST"; I could not separately confirm execution vs. recordation date) / recorded 2022-06-13 — Reel [NOT RETRIEVED]
  • Conveyance: Assignment (assignment of assignors' interest)
  • Assignor: Dieter Suess (individual; per the Google Patents event, "Assignors: SUESS, DIETER")
  • Assignee: MR Technologies GmbH (Austria)
  • Correspondent: [NOT RETRIEVED] — the attorney/firm of record on the assignment cover sheet could not be pulled from the Assignment Center. Do not assume it matches litigation counsel (Russ August & Kabat); that is litigation counsel, not the recording correspondent, and conflating them would be an error.
  • Context: Transfer-to-asserter — the sole inventor conveyed the patent to his own Austrian company, executed/recorded the same day the first infringement suit naming this patent was filed (W.D. Tex. 6:22-cv-00612, MR Technologies GmbH v. Western Digital, filed 2022-06-13).

No other recorded assignments, security interests, mergers, name changes, licenses, or releases appear in the legal-events feed for this patent. The chain is therefore a single link (inventor → MRT) over a 16-year gap. If the Assignment Center shows additional records (e.g., a patent-security-agreement or a family-level assignment covering 11/424,859 together with its continuations), they were not visible to me and should be checked directly.

Cross-references:

• Family (all sharing the 2006-06-17 priority, all naming Suess): US 9,928,864 (div. of 11/424,859); US 11,138,997 (cont. of 12/619,849); US 11,133,031; US 11,908,500; US 12,020,734. The '864 and '997 were the two patents tried against Western Digital.
• Litigation indexed to this patent: 6:22-cv-00612 (W.D. Tex., filed 2022-06-13); 8:22-cv-01599 (C.D. Cal., filed 2022-08-26 — WD jury verdict $262.3M, 2024-07-26/30, plus ~$117M prejudgment interest); 2:22-cv-06088 (C.D. Cal.); 8:25-cv-00786 (C.D. Cal. v. Toshiba, filed 2025-04-15); 0:25-cv-01460 (D. Minn. v. Seagate, 2025); Resonac DJ action, N.D. Cal. (filed 2025-10-09).
• Asserter directory: Stanford NPE Litigation Database entry for patent 9978413 lists MR Technologies, GMBH v. Western Digital Technologies, Inc. (2:22-cv-06088) and codes the asserter category as "8 – Product company." That coding is a notable counter-signal (see below).

Timeline diagram

timeline
    title Ownership of US 9978413
    2006 : Filed by Dieter Suess as individual
         : No assignment recorded at filing
    2007 : Application published 2007 12 20
    2018 : Patent US 9978413 issued
    2022 : Assigned to MR Technologies GmbH
         : Suit filed against Western Digital
    2024 : Jury awards 262M against Western Digital
    2025 : Suits filed against Toshiba and Seagate

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. The trigger is a move from an operating assignee to a licensing-only LLC with an anonymizing name. Here the transferor was the individual inventor, not an operating company; the assignee is his own named company ("MR Technologies GmbH"), not a Delaware/Texas single-purpose LLC, with no "IP / Patents / Licensing / Holdings / Ventures" suffix and no registered-agent-service address. The only qualifying aspect — the assignee is non-practicing — is real but is the inventor's own vehicle, not a shell. (Reel/frame not retrieved; call based on the recorded parties and the absence of any cascade.)

2. Known asserter in the chain — UNCLEAR (weakly present). MR Technologies GmbH does not appear on the enumerated NPE lists (Acacia, Marathon, IV, Wi-LAN/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Round Rock, etc.). It is, however, a serial plaintiff: five matters asserting this family since 2022-06-13 (WD in W.D. Tex. and C.D. Cal.; Seagate in D. Minn.; Toshiba in C.D. Cal.; plus a Resonac DJ action). It is surfaced by the Stanford NPE Litigation Database — but coded as "8 Product company," not an NPE category, which cuts against a troll characterization.

3. Repeat correspondent across the chain — NOT DETERMINABLE / NOT PRESENT within the chain. With only one recorded assignment, there is no recurrence of an assignment correspondent to detect. Flag for a follow-up pull: the litigation firm Russ August & Kabat (Marc Fenster, Reza Mirzaie, Paul Kroeger, Matthew Aichele, Jacob Buckzo, Dale Chang, Minna Chan, Brian Ledahl) recurs as plaintiff's counsel across MRT's campaigns (WD 2022, Toshiba 2025, Seagate 2025, and appearances in the 2025 N.D. Cal. DJ action). RAK is a known high-volume plaintiff-side patent firm. This is litigation counsel, not the assignment correspondent — the two should be checked independently before drawing any "shell LLCs change, the lawyer doesn't" conclusion.

4. Cascading transfers — NOT PRESENT. One transfer across sixteen years (2006 filing → 2022 assignment). No chained LLCs, no transfers within 24 months of each other.

5. Pre-litigation transfer — PRESENT. The conveyance to MR Technologies GmbH is dated/recorded 2022-06-13 — the same day the first suit naming this patent was filed (W.D. Tex. 6:22-cv-00612, MR Technologies GmbH v. Western Digital, 2022-06-13), followed by C.D. Cal. 8:22-cv-01599 on 2022-08-26 and 2:22-cv-06088. Same-day assignment + filing is the strongest tell in the chain: it moved the patent from the inventor personally into the corporate name to establish standing and to select venue.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11, no estate sale, no backruptcy-adjacent assignor or assignee.

7. Privateering — NOT PRESENT. There is no operating-company-to-NPE transfer in which the operating company asserts through a proxy against rivals. The posture is closer to the inverse: an academic inventor enforcing against the three HDD manufacturers that (per the record and trial testimony) adopted his exchange-spring architecture.

8. Defensive aggregator — NOT PRESENT. The chain terminates at MR Technologies GmbH, an active litigant. It does not end at RPX, AST, LOT Network, Unified Patents, or OIN. This patent has not been neutralized.

Verdict

NPE — moderate confidence.

MR Technologies GmbH is a non-practicing Austrian licensing/assertion vehicle — it ships no product embodying the claims, and the record shows one conveyance (inventor → MRT, dated/recorded 2022-06-13) landing on the same day the first infringement suit naming this patent was filed in W.D. Tex. (6:22-cv-00612), which is a strong pre-litigation-transfer signal. That is one strong signal, so the call is moderate rather than high: there is no shell cascade, no anonymizing LLC naming, no NPE-list match, and no purchased-patent provenance, and the entity is controlled by the inventor himself — indeed the Stanford NPE Litigation Database codes MRT as a "Product company." I could not retrieve the reel/frame or the correspondent of record, which are the two fields most likely to either confirm or defeat signal #3; those should be pulled directly before this verdict is relied upon.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (and the legacy UI at https://assignment.uspto.gov/patent/index.html), searching patent number 9978413 / application 11/424,859.

I'll start by searching for the specific patent and its cited references to ground my analysis.

I have substantial information. Let me verify details on several of the cited references to ground the analysis.

Prior-Art Analysis — US 9,978,413 B2 ("Multilayer exchange spring recording media")

0. Identification and scope (read this first)

• Patent searched: US 9,978,413 B2, application 11/424,859, filed/priority 2006-06-17, granted 2018-05-22, inventor Dieter Suess, current assignee MR Technologies GmbH. Title: Multilayer exchange spring recording media. (Source: https://patents.google.com/patent/[US9978413B2](/patent/US9978413B2)/en)
• Number-interpretation note (per your strict rule): I have interpreted the number exactly as given — 9,978,413. Do not confuse this with US 11,138,997 ("the '997 patent," from Ser. No. 15/925,749), which is a different patent in the same family (that one is a later continuation). Much of the public prosecution/litigation commentary on the web refers to "the '997 Patent" (= 11,138,997) and "the '864 Patent" (= 9,928,864, a division of 11/424,859). US 9,978,413 is the root/original application of the family.
• Tooling caveat: the "USPTO database" is mirrored by the Google Patents/PatentCenter/PTActs records I retrieved. I was able to verify the patent text, claims, front-page citations and dates directly. Characterizations of references whose full text I could not retrieve in this session are based on the '413 specification's own descriptions, the reference titles, and front-page data — flagged where relevant. I did not find any evidence that would let me "auto-correct" any identifier.

Independent claims at issue (the §102 anchor points):

• Claim 1 — magnetic bilayer: granular hard magnetic storage layer (⊥ anisotropy, coercive field H_s without another magnetic layer, thickness 3–30 nm) + granular hard magnetic nucleation host (⊥ anisotropy, disposed on the storage layer in a columnar manner, coercive field H_n) wherein 0.5 T < H_n < H_s; host J_s = μ₀M_s between 0.3 T and 1.0 T; the two layers separated by a coupling layer 0.1–3 nm directly between them; and the coupling is ferromagnetic.
• Claim 18 — similar, but the distinguishing feature is H_c(α=20°) < H_c(α=45°), with the layers either in direct contact or separated by a coupling layer <5 nm.

Dependent claims 2–17 depend from claim 1; claims 19–28 depend from claim 18 (26–28 via claim 25).

Bottom-line §102 conclusion up front: No single one of the twelve front-page patent citations discloses every limitation of claim 1 or claim 18. Under a strict §102 (anticipation) analysis, none of them anticipates the '413 claims. They are relevant as §103 obviousness references, and as teaching individual limitations. The single closest structural reference is US 7,550,210 B2, and the closest overall "prior art" is the inventor's own non-patent literature and the examiner-cited Fullerton/Takenoiri/Girt references that appear in the family prosecution history rather than on the '413 face.

1. The twelve patent citations on the face of US 9,978,413

(1) US 5,583,727 A — IBM

• Full citation: US 5,583,727 A, Multiple data layer magnetic recording data storage system with digital magnetoresistive read sensor, Int'l Business Machines Corp.
• Dates: priority 1995-05-15; published/granted 1996-12-10.
• Description: A storage system using multiple magnetic recording data layers read by a magnetoresistive read sensor. It is a multi-record-layer magnetic stack (independent, separately readable data layers) — not an exchange-spring hard/soft coupled bilayer. Cited in the background as general multi-layer media context.
• Potential §102 claims: None. It lacks the granular hard-magnetic nucleation host, the perpendicular-anisotropy bilayer, the 0.5 T < H_n < H_s relationship, the 0.1–3 nm ferromagnetic coupling layer, and the columnar granularity. Not anticipatory of any of claims 1–28; at most background/supports a §103 rationale for "a plurality of magnetic layers."

(2) US 5,851,643 A — Hitachi, Ltd.

• Full citation: US 5,851,643 A, Magnetic recording media and magnetic recording read-back system which uses such media, Hitachi, Ltd. (Honda, Inaba, Suzuki, Futamoto, Awano et al.).
• Dates: priority 1993-11-11; published/granted 1998-12-22. (A related Hitachi case describes it as "Vertical magnetic recording media with multilayered magnetic film structure.")
• Description: Multilayered vertical (perpendicular) magnetic recording media with a plurality of magnetic film layers; a read-back (magneto-resistive) system using the media. Establishes a multi-layer perpendicular recording layer, but with no soft/hard exchange-spring pinning architecture, no nucleation-host coercive-field relation, and no defined thin ferromagnetic coupling layer.
• Potential §102 claims: None. Discloses a multi-layered perpendicular recording stack (relevant to the general "magnetic bilayer/multilayer" concept in claim 1 / claim 18) but not the combination; not anticipatory.

(3) US 6,280,813 B1 — IBM

• Full citation: US 6,280,813 B1, Magnetic recording media with antiferromagnetically coupled ferromagnetic films as the recording layer, Int'l Business Machines Corp.
• Dates: priority 1999-10-08; published/granted 2001-08-28.
• Description: The "AFC" media concept — the conventional single recording layer is replaced by two ferromagnetic films antiferromagnetically coupled across a non-ferromagnetic spacer film; used to reduce the demagnetizing field of bits (longitudinal recording). Explicitly referenced in the '413 background.
• Potential §102 claims: None as to claim 1. It teaches two magnetic layers + a thin intervening layer, which maps onto the "coupling layer" limitation, but the coupling is antiferromagnetic, whereas claim 1 requires ferromagnetic coupling (and claim 18 requires direct contact or <5 nm coupling with H_c(20°) < H_c(45°)). It also lacks perpendicular granularity, a nucleation host, and the 0.5 T < H_n < H_s relation. Useful only as a §103 teaching for the "two coupled magnetic layers / spacer layer" element.

(4) US 6,468,670 B1 — IBM

• Full citation: US 6,468,670 B1, Magnetic recording disk with composite perpendicular recording layer, Int'l Business Machines Corp. (Ikeda et al.).
• Dates: priority 2000-01-19; published/granted 2002-10-22.
• Description: A composite perpendicular recording layer — a hard CoCrPt-type granular perpendicular layer with a continuous ferromagnetic overlayer to improve SNR. The '413 background specifically cites this as "a continuous ferromagnetic overlayer … to increase the Signal to Noise Ratio (SNR)."
• Potential §102 claims: None. It teaches a hard perpendicular layer plus an overlayer (mapping loosely onto "hard magnetic storage layer" + a second magnetic layer), but the overlayer is continuous, not a granular columnar nucleation host, and there is no teaching of 0.5 T < H_n < H_s, J_s = 0.3–1.0 T, or a 0.1–3 nm ferromagnetic coupling layer. Good §103 material for the "composite perpendicular bilayer" concept.

(5) US 6,383,668 B1 — IBM

• Full citation: US 6,383,668 B1, Magnetic recording media with antiferromagnetically coupled host layer for the magnetic recording layer, Int'l Business Machines Corp.
• Dates: priority 2000-03-27; published/granted 2002-05-07.
• Description: An antiferromagnetically coupled "host layer" used with the magnetic recording layer (a synthetic-antiferromagnet-type stabilizer). Notably this is the one face citation that uses the term "host layer" — the same term family as the '413's "nucleation host" — but the coupling is antiferromagnetic and the structure is a longitudinal stabilization scheme, not a perpendicular exchange-spring nucleation host.
• Potential §102 claims: None. Its "host layer" is antiferromagnetically coupled (opposite to claim 1's ferromagnetic coupling) and lacks the granular perpendicular nucleation host and the H_n/H_s and J_s constraints. Relevant as a §103/terminology reference only.

(6) US 2001/0051287 A1 — Akira Kikitsu

• Full citation: US 2001/0051287 A1, Magnetic recording medium and magnetic recording apparatus, Akira Kikitsu.
• Dates: priority 2000-06-12; published 2001-12-13.
• Description: A magnetic recording medium and apparatus (multi-layer magnetic recording layer generally). General background.
• Potential §102 claims: None. No teaching of the claim-1/claim-18 combination.

(7) US 2004/0191576 A1 — Showa Denko K.K.

• Full citation: US 2004/0191576 A1, Magnetic recording medium, method manufacture therefor, and apparatus for magnetic reproducing and reproducing recordings, Showa Denko K.K.
• Dates: priority 2001-02-28; published 2004-09-30.
• Description: Magnetic recording medium and its manufacture; conventional perpendicular/longitudinal layer stacks (seed/underlayer/magnetic layer). Background.
• Potential §102 claims: None. No exchange-spring nucleation-host structure or coercive-field relationship.

(8) US 2005/0058855 A1 — Seagate Technology LLC

• Full citation: US 2005/0058855 A1, Anti-ferromagnetically coupled perpendicular magnetic recording media with oxide, Seagate Technology LLC.
• Dates: priority 2001-11-30; published 2005-03-17.
• Description: Perpendicular recording media in which two magnetic layers are antiferromagnetically coupled across a thin spacer, with oxide grain-segregated (granular) magnetic layers.
• Potential §102 claims: None. This is the closest of the "AFC" references on two claim-1 elements (perpendicular anisotropy + granular oxide media + thin coupling layer), but the coupling is antiferromagnetic, and there is no granular columnar nucleation host with 0.5 T < H_n < H_s or a specified J_s. Strong §103 reference on the coupling-layer + granular-perpendicular limitations.

(9) US 2004/0053078 A1 — Kabushiki Kaisha Toshiba

• Full citation: US 2004/0053078 A1, Magnetic recording medium and magnetic recording/reproducing apparatus, Kabushiki Kaisha Toshiba.
• Dates: priority 2002-06-28; published 2004-03-18.
• Description: Magnetic recording medium and drive; conventional perpendicular layer stacks (potentially composite/multi-layer). Background.
• Potential §102 claims: None disclosed. Background/§103 only.

(10) US 2004/0057156 A1 — Samsung Electronics Co., Ltd.

• Full citation: US 2004/0057156 A1, Perpendicular magnetic recording media, [Samsung Electronics Co.](/litigations/by-defendant/Samsung%20Electronics%20Co.), Ltd.
• Dates: priority 2002-09-19; published 2004-03-25.
• Description: Perpendicular magnetic recording media layer stacks. Background.
• Potential §102 claims: None as established by the retrieved material. Background only.

(11) US 7,550,210 B2 — Hitachi Global Storage Technologies Netherlands B.V. ← closest structural reference of the twelve

• Full citation: US 7,550,210 B2, Perpendicular magnetic recording medium with multiple exchange-coupled magnetic layers having substantially similar anisotropy fields, Hitachi Global Storage Technologies Netherlands B.V. (Berger et al.).
• Dates: priority 2006-03-09; published/granted 2009-06-23. Because its filing/priority (2006-03-09) predates the '413 filing (2006-06-17), it is potentially prior art under pre-AIA §102(e) (and would be §102(a)(2)/§102(a)(1)-type art under the AIA framework).
• Description (verified from the patent): A multilayer recording layer (RL) of at least two ferromagnetically exchange-coupled magnetic layers (MAG1, MAG2), each with perpendicular magnetic anisotropy, separated by a ferromagnetic coupling layer (CL), where the CL is preferably 0.2–1.5 nm thick (<2.0 nm) and provides a tunable ferromagnetic coupling. The magnetic layers are granular Co-Pt-Cr-(oxide) alloys (oxide grain segregation). MAG1/MAG2 have substantially similar anisotropy fields (the lower H_k ≥ ~70–90% of the higher), and either or both may be superparamagnetic/thin, relying on coupling for stability. See Fig. 3 and the passage: "the CL may have a thickness of less than 2.0 nm, and more preferably between about 0.2 nm and 1.5 nm."
• §102 assessment vs. claim 1: This reference maps onto claim 1's hard magnetic storage layer + second (nucleation-host-type) granular perpendicular magnetic layer + thin ferromagnetic coupling layer (0.1–3 nm) elements very closely, and it is why this is the most structurally relevant citation of the twelve. However, it does not clearly disclose (i) the specific coercive-field relationship 0.5 T < H_n < H_s (it instead requires substantially similar H_k, with "MAG2" possibly superparamagnetic, i.e., H_n could be ≈0 → below the 0.5 T floor); (ii) the nucleation host J_s = μ₀M_s = 0.3–1.0 T limitation; or (iii) the granular hard magnetic nucleation host disposed in a columnar manner with the recitation of the storage layer thickness 3–30 nm as claimed. Consequently, anticipation of claim 1 is arguable at best and not clean; US 7,550,210 is best characterized as a primary §103 reference that, alone or with a secondary reference, discloses most of claim 1's structure. Same conclusion for claim 18 (it does not disclose the H_c(20°) < H_c(45°) angular behavior).
• Claims most affected: claim 1 (elements: two ⊥-anisotropy magnetic layers; granular oxide media; ferromagnetic coupling layer ~0.2–1.5 nm), claim 2 (ferromagnetic coupling-layer exchange constant), claim 17 (domain-wall formation across the hard/soft interface — consistent with its non-coherent reversal/"exchange-spring" description), claims 5, 8, 16 (materials — CoPtCr/Cr-based granular alloys), claim 9 (grain size), and claim 11/24 (perpendicular recording use).

(12) US 2007/0292720 A1 — Dieter Suess

• Full citation: US 2007/0292720 A1, Multilayer Exchange Spring Recording Media, Dieter Suess.
• Dates: priority 2006-06-17; published 2007-12-20.
• Description: This is the pre-grant publication of the '413 application itself (Ser. No. 11/424,859). It contains the identical disclosure and claims.
• Potential §102 claims: None. A reference cannot be prior art against its own application; this is the applicant's own publication. (Any apparent "anticipation" here is an artifact of the same disclosure.)

2. Non-patent literature cited on the face of US 9,978,413

These are the references most directly on-point for the inventive concept (and several are not the inventor's own work, making them the more probative prior art):

These are printed publications and are available as §102(a)/(b) prior art (pre-AIA). They individually disclose hard/soft or composite exchange-spring perpendicular media and the domain-wall-pinning mechanism, but none discloses the full claim-1 combination (especially 0.5 T < H_n < H_s, J_s 0.3–1.0 T, and the specific coupling-layer thickness with ferromagnetic coupling), so—like the patents—they are §103-type art, not clean anticipations of claim 1.

3. The actually most relevant prior art (prosecution-history references not on the '413 face)

Public prosecution/PTAB records for this patent family show the examiner relied on references that do not appear among the '413 face citations but are the true closest art. These are the references I would treat as the most relevant prior art for the '413 family:

1. "Fullerton 1" — US 2007/0243418 A1 (Fullerton et al.) — discloses an exchange-coupled magnetic multilayer structure with a hard magnetic storage layer (MAG1) and a "nucleation host" (MAG2) of ferromagnetic layers with differing H_k (and, the examiner reasoned, increasing anisotropy constant K) from layer to layer. This is the reference the examiner used in §103 rejections against the family's claims; the applicant ultimately narrowed around it. (See the Office Action text reproduced at ptacts.uspto.gov petition 1558578 and 1558577.)
2. "Fullerton 2" — US 2003/0108721 A1 — supplies the perpendicular-anisotropy materials (L1₀-ordered FePt/CoPt, FePt–X/CoPt–X, FePtC, FePt–ZrO/MgO/B₂O₃ granular composites) relied on for the material claims. Relevant to claims 5, 8, 16.
3. Takenoiri — US 8,329,321 B2 (filed 2005-07-01; issued 2012-12-11) — a hard magnetic recording layer + softer magnetic recording layer coupled by a coupling layer, with three-or-more magnetic recording layers; recognized as §102(a)/(e) art against the family. Directly relevant to claims 1, 2 (coupling layer) and claim 18.
4. Girt (Girt et al.) — cited for a thin (~2–20 Å) antiferromagnetic coupling layer between two Fe magnetic layers; used in §103 combinations for the coupling-layer limitation.
5. D. Suess et al., "Optimization of Exchange Spring Perpendicular Recording Media," Intermag Nagoya, Apr. 2005 — cited by the examiner as teaching a hard magnetic storage layer (H_c > 0.5 T) with a gradient-anisotropy nucleation host; relevant to claims 1, 3, 9, 10, 13–14.

(These appear in the family's prosecution exhibits and Notice of Allowance reasoning; the examiner noted that "the prior art of record does not teach or suggest [(i)] the coercive field relationship between the nucleation host and hard magnetic storage layer and [(ii)] a nucleation host comprising multiple ferromagnetic layers with increasing anisotropy constant K from layer to layer.")

4. Consolidated §102 verdict

Why claim 1 is not anticipated: §102 requires a single reference to disclose every limitation, including (a) 0.5 T < H_n < H_s with the host still "hard," (b) J_s = 0.3–1.0 T, (c) the granular host disposed in a columnar manner, and (d) the 0.1–3 nm ferromagnetic coupling layer. No face citation supplies all four; the AFC references (US 6,280,813, US 6,383,668, US 2005/0058855) supply antiferromagnetic, not ferromagnetic, coupling; US 7,550,210 supplies the ferromagnetic coupling layer and two granular perpendicular layers but with substantially similar anisotropy fields (not 0.5 T < H_n < H_s) and no explicit J_s range.

Practical takeaway: The most relevant prior art for US 9,978,413 is, in order, (1) US 7,550,210 B2 and the exchange-spring/composite-media non-patent literature (Victora 2005; Wang 2005; Inaba 2005; Hagedorn 1970; Loxley 2001; Dobin & Richter 2006; and Suess's own 2005–2006 papers), and (2) the prosecution-history references Fullerton 1 (US 2007/0243418), Fullerton 2 (US 2003/0108721), Takenoiri (US 8,329,321) and Girt, which the examiner used. None is a clean §102 anticipation; the validity dispute over the '413 family has therefore been fought, and won at the district-court level (MR Technologies GmbH v. Western Digital, C.D. Cal. 8:22-cv-01599, $262M verdict; see also NPE database entry for 9978413 and the Texas Western case 6:22-cv-00612), primarily as a §103/claim-construction battle rather than a pure §102 anticipation.

Sources

• US 9,978,413 B2 full text and front-matter citations: https://patents.google.com/patent/US9978413B2/en
• US 7,550,210 B2 text: https://patents.google.com/patent/US7550210
• Family prosecution materials (Office Actions/Notice of Allowance, Fullerton/Takenoiri/Girt): https://ptacts.uspto.gov/ptacts/public-informations/petitions/1558577 and .../1558578
• NPE/litigation record for 9978413: https://npe.law.stanford.edu/patent/9978413

Confidence notes: (1) The face-citation dates and titles are taken directly from the patent record and are high-confidence. (2) For US 5,583,727, US 5,851,643, US 6,383,668, US 2001/0051287, US 2004/0191576, US 2004/0053078 and US 2004/0057156, I could not retrieve full text in this session; their descriptions rest on titles, the '413 specification's own characterizations, and the front-page data — treat those specific descriptions as provisional. (3) The "USPTO database" mirror I used is Google Patents/PatentCenter/PTActs; I found no basis to alter any patent number.

I'll ground this analysis in the prior art actually listed on the patent page, verifying key reference dates and disclosures.

Obviousness Analysis — U.S. Patent No. 9,978,413 B2

"Multilayer exchange spring recording media"

Scope note: This is a technical invalidity-risk assessment, not a legal opinion. All claim and reference identifiers are reproduced literally as they appear on the patent page at https://patents.google.com/patent/[US9978413B2](/patent/US9978413B2)/en and in the "Prior Art" sections (Patent Citations (12), Non-Patent Citations (21), Family Cites Families (43)). Some of these references are cited by the family rather than in it; where a reference's own date falls after 2006‑06‑17 it is identified as not prior art to the '413 patent.

I. Prior-art window

• Earliest effective date of the '413 claims: 2006‑06‑17 (filing = priority; application 11/424,859, filed 2006‑06‑17).
• Pre-AIA §102 applies. Qualifying art includes US patents granted on applications filed before 2006‑06‑17 (§102(e)) and printed publications before that date (§102(a)/(b)).
• Concrete qualifying references from the page's prior-art lists:
  • US 6,468,670 (Ikeda, IBM) — composite perpendicular recording layer, 2002‑10‑22
  • US 6,280,813 (Carey, IBM) — AFC recording layer, 2001‑08‑28
  • US 6,383,668 (Fullerton, IBM) — "antiferromagnetically coupled host layer," 2002‑05‑07
  • US 5,583,727 (Parkin, IBM) — 1996‑12‑10; US 5,851,643 (Honda, Hitachi) — 1998‑12‑22
  • US 7,550,210 (Berger, Hitachi GST) — cited on the face of '413; priority 2006‑03‑09
  • US 2005/0058855 A1 (Gitt, Seagate) — AFC perpendicular media with oxide, 2005‑03‑17
  • US 7,687,157 B2 (Berger/Fullerton, Hitachi GST) — "Perpendicular recording media having an exchange-spring structure," app. filed 2005‑02‑04 (published as US 2006/0177704 A1; granted 2010‑03‑30) — §102(e) art. Note: US 7,989,096 B2 is the same family continuation.
  • US 7,846,564 B2 (Li et al., Seagate) — "magnetic anisotropy/coercivity gradient and local exchange coupling," app. 11/235,208 filed 2005‑09‑27 (published US 2007/0072011 A1)
  • US 7,488,545 B2 (Hitachi GST) — laminated recording layers of exchange-coupled ferromagnetic layers, filed 2006‑04‑12
  • US 2006/0147758 A1 (Jung, Hitachi GST), filed 2005‑01‑06; US 2006/0210834 A1 (Do, Hitachi GST), filed 2005‑03‑18
  • US 2002/0192506 A1 (IBM, "Thermal Spring" media), 2001‑06‑04; US 2003/0108721 A1 (Fullerton), 2001‑12‑11; US 6,773,112 B1 (Seagate), 2004‑08‑17
• NPL (all pre-2006‑06‑17 unless noted):
  • Suess et al., "Exchange spring media for perpendicular recording," Appl. Phys. Lett. 87, 012504 (2005) — expressly incorporated by reference into the '413 specification and cited by the examiner.
  • Suess et al., "Exchange spring recording media for areal densities up to 10 Tbit/in²," JMMM 290‑291 (2005) 551‑554
  • Suess et al., "Optimization of Exchange Spring Perpendicular Recording Media," Intermag Nagoya, Apr. 2005
  • Victora & Shen, "Composite Media for Perpendicular Magnetic Recording," IEEE Trans. Magn. 41(2), Feb. 2005, 537‑542
  • Wang et al., "Composite media (dynamic tilted media) for magnetic recording," APL 86, 142504 (2005)
  • Inaba et al., "(CoPtCr/NiFe)—SiO₂ Hard/Soft-Stacked Perpendicular Recording Media," IEEE Trans. Magn. 41(10), Oct. 2005, 3136‑3138
  • Hagedorn, "Analysis of Exchange-Coupled Magnetic Thin Films," J. Appl. Phys. 41(6), May 1970, 2491‑2502
  • Loxley et al., IEEE Trans. Magn. 37(4), Jul. 2001, 2098‑2100
  • Thiele et al., APL 82(17), 2003, 2859‑2861; Dobin & Richter, arXiv:cond‑mat/0605368 v1, 2006‑05‑15
• Not prior art to the '413 claims: US 7,582,368, US 7,588,841, US 7,578,526 (2007 priorities), US 9,142,240, US 8,163,405, US 2008/0211689 — these cite the Suess family rather than predating it.

II. Claim 1 — element-by-element mapping

Result: Every element of claim 1 is disclosed or rendered obvious by US 7,687,157 in combination with Suess APL 2005, with US 7,846,564 supplying the numerical ranges.

III. Combinations that would render the claims obvious

Ground 1 (strongest — claim 1)

US 7,687,157 (Berger et al.) in view of Suess et al., APL 87, 012504 (2005), further in view of US 7,846,564 (Li et al.).

• '157 supplies: non-magnetic substrate; granular perpendicular hard recording layer; a softer-but-still-magnetic exchange-spring layer (the claimed "nucleation host"); a 2 nm coupling layer directly between them; and expressly ferromagnetic interlayer coupling with H_exchange-spring < H_recording layer.
• Suess APL 2005 supplies the enabling physics and the motivation: a hard/soft bilayer in which "a magnetic domain wall is created next to the hard/soft interface," yielding a coercive-field reduction "by a factor of 6" with retained/increased thermal stability, plus working numerical parameters (A = 1×10⁻¹¹ J/m; soft-layer J_s = 0.9 T).
• Li '564 supplies the layer-thickness, grain-size, spacer-thickness and local-exchange ranges (6–25 nm / 3–15 nm sub-layers; 4–10 nm grains; ≤3 nm spacer; local coupling ~10⁻²–10⁻⁹ erg/cm), which read directly onto the 3–30 nm hard layer, 0.1–3 nm coupling layer, and 2–10 nm host grains.

Ground 2 (claim 1 / claim 18)

Victora & Shen 2005 + Inaba et al. 2005 + US 7,846,564.

• Victora & Shen disclose a composite perpendicular medium of hard and soft regions within each grain with an "exchange layer that moderates the interaction between the two regions," and quantify the energy-barrier-to-switching-field trade-off.
• Inaba discloses the physical hard/soft stacked (CoPtCr/NiFe)—SiO₂ granular perpendicular structure with columnar grain alignment.
• '564 supplies columnar grains with a designed H_k gradient and local exchange control.
• Note the motivation: Victora/Wang teach moderating the interlayer exchange to lower coercivity; the claimed 0.1–3 nm ferromagnetic coupling layer is the same structural expedient with a different coupling strength — a predictable engineering choice once Suess taught that strong coupling with domain-wall formation also reduces coercivity.

Ground 3 (claims 1, 3, 17)

Suess APL 2005 + Hagedorn 1970 + Loxley et al. 2001 + Dobin & Richter 2006.

• Hagedorn and Loxley are the analytic domain-wall-pinning/domain-wall-nucleation teachings cited on the face of the patent. Hagedorn explicitly treats the finite-anisotropy soft layer (supporting 0.5 T < H_n), and Loxley teaches a ~4× coercive-field reduction in a two-section magnetic structure. Dobin & Richter extend domain-wall-assisted recording to recording media.
• This ground directly attacks claim 17 ("exchange coupling … enables a formation of a domain wall across an interface … during reversal") and claim 3 (the k ≤ factor-of-3 slope criterion, which the '413 spec itself derives from the same Hagedorn pinning formula).

Ground 4 (claims 6, 7, 15, 18, 28 — angular dependence)

Suess APL 2005 (abstract: "the lower angular dependence of coercivity of exchange spring media will improve the signal to noise ratio") + Hagedorn.

• The claimed H_c(α=20°) < H_c(α=45°) signature is the well-known "pinning magnet" behavior. The '413 specification itself states that H_c = 1/cos(α) for pinning magnets, and admits this is the known behavior being exploited. A claim to an inherent, predictable consequence of the selected structure is obvious under In re Kao / MPEP 2144.04.

IV. Motivation to combine (KSR/MPEP 2143 rationales)

1. Same field / same problem. Every reference addresses the same problem expressly identified in the '413 Background: the superparamagnetic limit versus writeability. '157 states "One of the primary challenges to increasing areal densities … is overcoming the constraints imposed by the superparamagnetic effect."
2. Known problem, known solution, predictable result. Suess APL 2005 reports that exchange-spring bilayers achieve "high thermal stability without increase of coercive field." Victora & Shen report an energy-barrier/switching-field ratio "similar to the previously proposed tilted media, while avoiding some of the difficulties," and note fabrication is "significantly easier." A POSITA would have expected success, not uncertainty.
3. Combining known elements with known functions (KSR). Substrate + granular perpendicular storage layer + exchange-spring layer + thin coupling layer were all known elements performing their known functions; the combination yields nothing more than the expected aggregate.
4. Obvious optimization of result-effective variables (MPEP 2144.05). Layer thicknesses (3–30 nm; 0.1–3 nm), grain diameters (2–10 nm), polarization (0.3–1.0 T) and the 0.5 T < H_n < H_s window are process/materials parameters that the art itself taught to tune. The '413 specification concedes the numbers are tunable ("parameters are for illustrative purposes only and can widely vary").
5. Arbitrary-range logic. The claimed J_s range (0.3–1.0 T) brackets the Suess APL 2005 value of 0.9 T; the claimed hard-layer range (3–30 nm) brackets '564's 6–25 nm; the claimed coupling range (0.1–3 nm) brackets '157's CoRu ≈ 2 nm.

V. Counter-arguments a defender would raise (and their weaknesses)

1. Teaching away from strong coupling. Victora & Shen and Wang et al. both concluded that a decoupling/coupling layer was needed to reduce the exchange coupling to lower coercivity. The '413 specification itself highlights this "contrast" with Suess. Rebuttal: a teaching of one operable coupling regime does not teach away from a second operable regime, especially where Suess (applied to the same media and expressly incorporated by reference, dated the same month as Victora) showed strong exchange coupling with domain-wall formation works. "Two possible options" is not "teaching away" (In re Fulton).
2. Li '564's domain-wall disclaimer. Li claim 8 requires total thickness "less than the local exchange coupling distances … whereby domain walls are not present." This is a genuine teaching-away argument specifically against claim 17 (domain-wall formation) and arguably against the G-layer/continuous-anisotropy embodiments. Rebuttal: Li's other embodiments (3+ sub-layers) rely on "incoherent rotation" reversal propagating through the grain, which is functionally the same physics; and Li is not the primary reference.
3. Inaba's uniform-magnetization model. Inaba requires a sufficiently thin soft magnet so magnetization stays uniform in both layers, and the '413 spec uses Inaba's model to argue a uniform bilayer has no writeability benefit. Rebuttal: this attacks Inaba as the sole reference, not the combination; it does not undercut the '157/Suess ground.
4. Secondary considerations. The family is in active litigation (e.g., Texas Western District Court case 6:22‑cv‑00612; four C.D. Cal. cases, including case 8:22‑cv‑01599), and per Unified Patents / IPR filings a Texas jury found sibling patents '864 and '997 not invalid in July 2024, with testimony that the technology became foundational to modern PMR media. If MR Technologies can show a nexus between the claimed subject matter and that commercial adoption (e.g., industry adoption of the claimed nucleation-host/thin-coupling-layer combinations), that is the strongest non-obviousness evidence and would likely defeat Grounds 1–3 on claim 1's specific ranges. Note, however, that the '864/'997 claims are broader than '413 claim 1; a verdict on those claims does not establish non-obviousness of claim 1's narrower numerical recitations.

VI. Dependent-claim disposition

VII. Bottom line

• Claim 1 is, on the face of the '413 record, in significant §103 jeopardy. The single strongest combination is US 7,687,157 (Berger et al.) + Suess APL 87, 012504 (2005) + US 7,846,564 (Li et al.), with Hagedorn supplying the finite-anisotropy soft-layer teaching required by "0.5 T < H_n < H_s."
• Claim 18 (independent, broader — direct contact or <5 nm coupling layer, no thickness/J_s recitations) is even more exposed to '157 + Suess APL 2005 alone.
• Claim 17 and the angular-limitation claims (6, 7, 15, 28) are vulnerable on the "inherent/predictable result" theory, with Suess's own 2005 abstract supplying the angular-dependence teaching.
• The realistic defense is claim-differentiation via the numerical range recitations plus secondary considerations (industry adoption, the 2024 jury verdict on the sibling '864/'997 patents, and the pending Texas/C.D. Cal. litigation). Absent a demonstrated nexus between that adoption and the specific 0.5 T < H_n < H_s / 0.1–3 nm / 0.3–1.0 T windows, I do not see a technical basis in the cited art for distinguishing claim 1 from the exchange-spring/composite-media literature that was already mainstream in 2005–2006.

Key URLs: https://patents.google.com/patent/US9978413B2/en · https://patents.google.com/patent/US7687157 · https://www.freepatentsonline.com/[7687157](/patent/7687157).html · https://patents.google.com/patent/US7846564 · https://aipp.silverchair.com/aip/apl/article-abstract/87/1/012504/[925370](/patent/925370)/ · https://www.semanticscholar.org/paper/Composite-media-for-perpendicular-magnetic-Victora-Shen/5480f7a5b864ebf9d52894741c328c3b5a906591