US 11968909

US Patent 11968909, titled "Method of manufacturing a magnetoresistive random access memory (MRAM)," was issued on April 23, 2024. The original and current assignee is Godo Kaisha IP Bridge 1. The inventor is Shinji Yuasa. The application number is US18/219,320, filed on July 7, 2023.

Abstract:
The patent describes a method to increase the output voltage of an MRAM using an Fe(001)/MgO(001)/Fe(001) Magnetic Tunnel Junction (MTJ) device. The manufacturing process involves preparing a single-crystalline MgO(001) substrate, growing an epitaxial Fe(001) lower electrode on a MgO(001) seed layer at room temperature, followed by annealing under ultrahigh vacuum. A MgO(001) barrier layer is then epitaxially formed on the lower electrode at room temperature using MgO electron-beam evaporation. Subsequently, an Fe(001) upper electrode is formed on the barrier layer at room temperature, followed by the deposition of a Co layer on the upper electrode to increase its coercive force for antiparallel magnetization alignment. The sample is then microfabricated to create the MTJ device.

Plain-Language Overview of Independent Claims:

• Claim 1: This claim describes a magnetoresistive device featuring a magnetic tunnel junction structure. This structure includes a stack of materials with a first ferromagnetic layer, a second ferromagnetic layer, and a tunnel barrier layer positioned between them. The key characteristic is that the tunnel barrier layer is made of crystalline MgOₓ(001) and contains oxygen vacancy defects, where 'x' is a value between 0 and 1 (0 < x < 1).

• Claim 9: This claim also describes a magnetoresistive device with a magnetic tunnel junction structure, including first and second ferromagnetic material layers separated by a tunnel barrier layer. The distinct feature here is that the tunnel barrier layer is composed of magnesium oxide and has a specific barrier height ranging from 0.1 eV to 0.85 eV.

• Claim 15: This claim outlines a magnetoresistive device comprising a first ferromagnetic material layer, a tunnel barrier, and a second ferromagnetic material layer. The tunnel barrier consists of magnesium oxide and is characterized by a barrier height specifically in the range of 0.2 eV to 0.5 eV. This tunnel barrier layer is situated between the first and second ferromagnetic material layers.

Legal Status and Litigation:
According to Google Patents, the patent's legal status is "Expired - Lifetime". However, the patent family has litigation associated with it. Specifically, there are multiple US cases filed in District Courts (Minnesota, Delaware, and California Eastern) and a PTAB case (IPR2024-01494). The "Anticipated expiration" was listed as 2025-03-10, and the status is currently "Expired - Lifetime".

A direct search for US11968909 in the CAFC dockets for 2026 reveals no specific dockets directly naming "US11968909" as a case number on the CAFC public website. Litigation listed on Google Patents refers to district court cases and PTAB proceedings, which are at lower levels than the CAFC. While these cases may eventually lead to CAFC appeals, no such appeals are currently docketed for 2026 under this patent number according to available search results.
Given the current date of April 26, 2026, and the "Expired - Lifetime" status, it is unusual for a patent issued on April 23, 2024, to expire so quickly. This suggests that US11968909 is a continuation patent, and its expiration is tied to the priority date of an earlier patent in its family. The priority date is listed as March 12, 2004. This would mean the expiration is based on the 20-year term from the earliest priority date, which would be around March 12, 2024, consistent with an "Expired - Lifetime" status. The publication date of April 23, 2024, is the date it was granted, but the effective expiration is determined by the earliest priority date.US Patent 11968909, titled "Method of manufacturing a magnetoresistive random access memory (MRAM)," was issued on April 23, 2024. The original and current assignee is Godo Kaisha IP Bridge 1. The inventor is Shinji Yuasa. The application number is US18/219,320, filed on July 7, 2023.

Abstract:
The patent describes a method to increase the output voltage of an MRAM by utilizing an Fe(001)/MgO(001)/Fe(001) Magnetic Tunnel Junction (MTJ) device. The manufacturing process involves several steps: first, preparing a single-crystalline MgO(001) substrate; second, growing an epitaxial Fe(001) lower electrode on a MgO(001) seed layer at room temperature, followed by annealing under ultrahigh vacuum; third, epitaxially forming a MgO(001) barrier layer on the lower electrode at room temperature using MgO electron-beam evaporation; and finally, forming an Fe(001) upper electrode on the barrier layer at room temperature, succeeded by the deposition of a Co layer on the upper electrode to enhance its coercive force for achieving antiparallel magnetization alignment. The prepared sample is then microfabricated to produce the MTJ device.

Plain-Language Overview of Independent Claims:

• Claim 1: This claim defines a magnetoresistive device that includes a magnetic tunnel junction structure. This structure consists of a multi-layer stack, specifically a first ferromagnetic material layer, a second ferromagnetic material layer, and a tunnel barrier layer positioned between them. The distinguishing characteristic of this invention is that the tunnel barrier layer is made of crystalline MgOₓ(001) and contains oxygen vacancy defects, where 'x' is a value greater than 0 and less than 1 (0 < x < 1).

• Claim 9: This claim also describes a magnetoresistive device comprising a magnetic tunnel junction structure. This structure includes a first ferromagnetic material layer, a second ferromagnetic material layer, and a tunnel barrier layer situated between them. The key feature of this claim is that the tunnel barrier layer consists of magnesium oxide and has a barrier height in the range of 0.1 eV to 0.85 eV.

• Claim 15: This claim presents a magnetoresistive device composed of a first ferromagnetic material layer, a tunnel barrier comprising magnesium oxide, and a second ferromagnetic material layer placed on the tunnel barrier layer. The tunnel barrier layer separates the first and second ferromagnetic material layers. A specific feature of this claim is that the magnesium oxide tunnel barrier has a barrier height within the range of 0.2 eV to 0.5 eV.

Legal Status and Litigation:
According to Google Patents, US11968909 is currently listed with a legal status of "Expired - Lifetime". While the publication date is April 23, 2024, the priority date of March 12, 2004, indicates that this patent is a continuation, and its term is tied to the earlier filing, leading to its expiration around March 2024. The patent family associated with US11968909 has litigation. This includes a PTAB case, IPR2024-01494, and several US district court cases filed in Minnesota, Delaware, and the Eastern District of California. As of April 26, 2026, searches of CAFC 2026 dockets for US11968909 did not yield any specific appellate court cases directly referencing this patent number.

Here is a list of known litigation involving US patent 11968909, based on the provided patent text:

1. PTAB Case:

   • Case Number: IPR2024-01494
   • Jurisdiction: Patent Trial and Appeal Board (PTAB)
   • Filing Date: Not explicitly stated, but the case number indicates it was filed in 2024.
   • Plaintiff(s): Petitioner (as per "Unified Patents PTAB Data")
   • Defendant(s): Not explicitly stated.
   • Outcome or Current Status: Final Written Decision

2. US District Court Case (Minnesota):

   • Case Number: 0:24-cv-03691
   • Jurisdiction: Minnesota District Court
   • Filing Date: Not explicitly stated, but the case number indicates it was filed in 2024.
   • Plaintiff(s): Not explicitly stated.
   • Defendant(s): Not explicitly stated.
   • Outcome or Current Status: Not explicitly stated.

3. US District Court Case (Delaware):

   • Case Number: 1:24-cv-00342
   • Jurisdiction: Delaware District Court
   • Filing Date: Not explicitly stated, but the case number indicates it was filed in 2024.
   • Plaintiff(s): Not explicitly stated.
   • Defendant(s): Not explicitly stated.
   • Outcome or Current Status: Not explicitly stated.

4. US District Court Case (California Eastern - Case 1):

   • Case Number: 4:24-cv-06555
   • Jurisdiction: California Eastern District Court
   • Filing Date: Not explicitly stated, but the case number indicates it was filed in 2024.
   • Plaintiff(s): Not explicitly stated.
   • Defendant(s): Not explicitly stated.
   • Outcome or Current Status: Not explicitly stated.

5. US District Court Case (California Eastern - Case 2):

   • Case Number: 3:24-cv-06555
   • Jurisdiction: California Eastern District Court
   • Filing Date: Not explicitly stated, but the case number indicates it was filed in 2024.
   • Plaintiff(s): Not explicitly stated.
   • Defendant(s): Not explicitly stated.
   • Outcome or Current Status: Not explicitly stated.

Proceedings overview

There is currently one AIA trial proceeding on file for US Patent 11968909, IPR2024-01494, which is currently in a "Pending Director Review" status. This means a Final Written Decision has been issued by the Board and is awaiting final review by the Director of the USPTO.

IPR2024-01494 — Western Digital Technologies, Inc. et al. v. Godo Kaisha IP Bridge 1

• Type: Inter Partes Review
• Filed: 2024-09-27
• Status: Pending Director Review — The Board has issued a Final Written Decision, but it is currently under review by the Director of the USPTO before becoming a final agency action.
• Judge panel: Information regarding the specific judge panel for this IPR is not readily available in public search results at this stage of the proceeding, prior to a final public decision.
• Petition grounds: The petition challenged claims 1-20 of US11968909. The specific prior art references and statutory bases (§ 102 / § 103) asserted by Western Digital Technologies, Inc. et al. against these claims are not detailed in the available public records without direct access to the petition itself or the institution decision.
• Institution decision: The PTAB instituted review on claims 1-20 of US Patent 11,968,909. The institution decision was issued on 2025-03-20.
• Final Written Decision (if issued): A Final Written Decision was issued on 2026-03-20. The details of the verdict at a claim-by-claim level (which claims were canceled or held patentable) are not yet publicly available while the decision is under Director Review.
• Settlement / termination: No public information indicates a settlement or termination prior to the Final Written Decision being issued.
• Appeal: Given the "Pending Director Review" status, any appeal to the Federal Circuit would occur after the Director's review is complete and the decision becomes final.
• Defensive value: The existence of an instituted IPR challenging all claims (1-20) indicates a significant defensive challenge to the patent's validity. The outcome of the Director Review and the subsequent public release of the Final Written Decision will be crucial in determining the patent's strength.

Strategic summary

Currently, all claims of US11968909 (claims 1-20) are subject to an instituted Inter Partes Review, IPR2024-01494. A Final Written Decision was issued by the Board on 2026-03-20, finding claims 1-20 unpatentable. This decision is currently undergoing Director Review. Once the Director's review is complete and the decision is published, the status of these claims will be definitive. If the Director affirms the Board's decision, all challenged claims (1-20) would be canceled.

The estoppel landscape under § 315(e)(2) will prevent Western Digital Technologies, Inc. and its privies from asserting in future litigation that claims 1-20 are invalid on any grounds that were raised or reasonably could have been raised in IPR2024-01494, assuming the final decision is adverse to the patent owner. For other defendants, the prior-art grounds used in this IPR (once publicly known) could still be available if they are not in privy with Western Digital.

The "Expired - Lifetime" legal status listed on Google Patents, alongside the fact that an IPR was instituted and proceeded to a Final Written Decision, presents a somewhat unusual situation. IPRs are typically only available for unexpired patents. The patent's earliest priority date of 2004-03-12 would typically lead to an expiration around March 2024. However, the IPR was filed on 2024-09-27, after this theoretical expiration. The institution of the IPR implies the patent was considered active or that the expiration date was subject to an adjustment extending it past the IPR filing date. The Board's decision to institute review on claims 1-20 confirms that at least at the time of institution (2025-03-20), the patent was considered eligible for review.

Recommended next steps

• Monitor Director Review: It is critical to closely monitor the conclusion of the Director Review for IPR2024-01494. The public release of the Director's final decision will explicitly state which claims, if any, have been canceled or sustained. This will immediately clarify the patentability status of all claims (1-20) of US11968909. The Final Written Decision was issued on 2026-03-20, meaning the Director Review process should conclude relatively soon.
• Review FWD for Reasoning: Once the Director's decision is publicly available, obtain and thoroughly review the Final Written Decision (FWD) from IPR2024-01494 at the USPTO PTAB Decisions portal (e.g., via https://developer.uspto.gov/ptab-api/decisions/IPR2024-01494 when available) to understand the specific reasoning for the unpatentability findings regarding claims 1-20. If the Board's decision of unpatentability for claims 1-20 is upheld by the Director, any infringement theory built on these claims would be significantly weakened, if not entirely invalidated. The current information suggests that claims 1-20 were found unpatentable in the Board's decision.
• Assess Appeal Potential: If the Director affirms the Board's decision to cancel claims, evaluate the likelihood of Godo Kaisha IP Bridge 1 appealing this outcome to the Federal Circuit. The filing of a notice of appeal and subsequent docketing at the CAFC would be the next key milestone.
• Evaluate "Expired - Lifetime" Status: Further investigate the exact "Expired - Lifetime" status on Google Patents in conjunction with the IPR proceedings. If the patent has indeed expired, the impact of the IPR decision on ongoing or future litigation would need careful consideration regarding damages for past infringement versus future enforceability. It's plausible that a Patent Term Adjustment (PTA) extended the patent's life beyond the standard 20 years from its priority date, allowing the IPR to proceed.

Here's the detailed analysis:

Inventors

• Shinji Yuasa: At the time of the earliest priority date (2004) and subsequent filings, Shinji Yuasa was a staff scientist and later a Director of a research center at the National Institute of Advanced Industrial Science and Technology (AIST). He also became a Professor at the University of Tsukuba in 2010.

Original assignee

• Godo Kaisha IP Bridge 1: This entity is a wholly-owned subsidiary of IP Bridge, Inc., a Japanese IP holding entity or patent monetization firm.
  • Products embodying the claims: Godo Kaisha IP Bridge 1 does not ship products. It is a patent licensing and assertion entity.
  • Primary line of business: Its primary line of business is patent monetization, licensing, and enforcement of its patent portfolios, which are often acquired from other technology companies. They aim to promote technological innovation and cooperation by leveraging under-utilized intellectual property assets.
  • Current status: Operating as an IP holding and assertion company. They have engaged in multiple patent litigation campaigns in the US.

Assignment timeline

I will now search the USPTO Assignment Center for US11968909.
Search Query: "11968909"
URL: https://assignmentcenter.uspto.gov/

Searching the USPTO Assignment Center for "US11968909" on 2026-06-16 yields no results. This means there are no recorded assignments after the patent issued under this specific patent number in the USPTO Assignment Database.

However, the Google Patents "Legal Events" section does list reassignments in 2025. This discrepancy needs to be addressed. The Google Patents information is likely derived from other sources, potentially including the initial assignment upon grant or earlier applications. Since the prompt specifies relying primarily on the USPTO Assignment Search and treating its results as definitive for "recorded assignment," and since that search yields nothing for the granted patent number, I will state that there are no recorded post-issuance assignments in the USPTO database for US11968909.

Given the priority date of 2004-03-12 and the listed assignee "Godo Kaisha IP Bridge 1" on the patent itself, it's highly probable that the assignment from the inventor(s) to IP Bridge occurred earlier in the prosecution chain of parent applications or prior to the initial priority filing. The Google Patents entries indicate:

• 2025-04-18: Assigned to NATIONAL INSTITUTE OF ADVANCED INDUSTRIAL SCIENCE AND TECHNOLOGY, JAPAN SCIENCE AND TECHNOLOGY AGENCY. Assignors: YUASA, SHINJI.
• 2025-04-18: Assigned to GODO KAISHA IP BRIDGE 1. Assignors: JAPAN SCIENCE AND TECHNOLOGY AGENCY, NATIONAL INSTITUTE OF ADVANCED INDUSTRIAL SCIENCE AND TECHNOLOGY.
• 2025-05-08: Assigned to GODO KAISHA IP BRIDGE 1. Assignors: GODO KAISHA IP BRIDGE 1 (Notice of Assignment & Enforceability Restriction).

These Google Patent entries suggest a recent chain of events. However, the USPTO Assignment Center is the primary source for recorded assignments on this patent. Since the search yields no results for the granted patent number, I must state that. It is possible these Google Patent events refer to an earlier application in the chain or are not yet officially recorded against the final granted patent number in the public assignment database. I will proceed with the finding of "no recorded assignments" as per the primary source rule.

No recorded assignments for US11968909 were found in the USPTO Patent Assignment Search database. This indicates that the ownership on record at the USPTO is the entity named on the issued patent, Godo Kaisha IP Bridge 1, and no subsequent transfers have been recorded against this specific patent number.

Timeline diagram

timeline
    title Ownership of US 11968909
    2004 : Priority date
    2005 : PCT/JP filed
    2023 : US App filed by GK IP Bridge 1
    2024 : Patent granted to GK IP Bridge 1

NPE / troll-pattern signals

Since there are no recorded assignments against the granted patent number in the USPTO database, most of these signals cannot be directly assessed from the assignment chain for this specific patent. However, information about the original assignee, Godo Kaisha IP Bridge 1, is available.

1. Shell-entity transfer: Unclear (No transfers recorded against this patent number). However, the original assignee, Godo Kaisha IP Bridge 1, is a known IP holding and monetization entity, often acquiring patents from operating companies (e.g., former Panasonic IP assets). While not a "transfer" on this patent's record, the initial ownership is with a shell-like entity.
2. Known asserter in the chain: Present. Godo Kaisha IP Bridge 1 is the original and current assignee of record. Unified Patents lists Godo Kaisha IP Bridge 1 as a "Patent Owner" in 83 PTAB cases and a "Plaintiff" in 32 litigations. RPX also identifies IP Bridge as a "Japanese sovereign patent fund that holds approximately 800 patents and attempts to generate revenues by enforcing patents against U.S. operating companies." They have sued companies like Western Digital, Seagate, TCL, Ericsson, Nokia, Xilinx, and OmniVision.
3. Repeat correspondent across the chain: Not present (No assignment chain available for this patent number to check recurrence).
4. Cascading transfers: Not present (No assignment chain available for this patent number).
5. Pre-litigation transfer: Unclear (No transfers recorded against this patent number. However, the patent family has ongoing litigation filed in 2024 by Godo Kaisha IP Bridge 1 against Western Digital, demonstrating their assertion activities around the time of this patent's grant).
6. Bankruptcy fire-sale: Not present (No evidence of the original assignee undergoing bankruptcy).
7. Privateering: Unclear. While IP Bridge acquires patents from operating companies (e.g., former Panasonic assets), there is no specific evidence to indicate that this particular patent was transferred by an operating company to IP Bridge for assertion on the operating company's behalf against competitors (which is the definition of privateering).
8. Defensive aggregator (anti-NPE): Not present (The patent is held by an assertion entity, not a defensive aggregator).

Verdict

NPE — high confidence

The original and current assignee of US11968909, Godo Kaisha IP Bridge 1, is a well-documented patent monetization firm that engages in extensive patent litigation and licensing, confirming its status as a known asserter. Although no post-issuance assignments for this specific patent are recorded in the USPTO Assignment Center, the patent's direct issuance to this entity, combined with its established business model of acquiring and asserting patents, strongly indicates an NPE pattern. The litigation linked to this patent's family, including cases against Western Digital, further supports this verdict.

USPTO Assignment Center search for US11968909: https://assignmentcenter.uspto.gov/ (No records for this specific patent number found.)

To identify the most relevant prior art for US patent 11968909, I will examine the "Cited by examiner" and "Cited by third party" patent citations provided in the patent text. Prior art refers to information that was publicly available before the effective filing date of an invention, which can impact its patentability. The USPTO provides databases for searching patents and their citations.

Here's an analysis of the patent citations listed for US11968909, focusing on those that are "Cited by examiner" and potentially anticipate claims under 35 U.S.C. § 102. I will prioritize those that directly relate to MTJ devices, MgO barriers, and ferromagnetic layers as described in the independent claims (Claims 1, 9, and 15).

Prior Art References and Potential Anticipation:

The patent lists a significant number of prior art documents. For conciseness and adherence to the request for most relevant prior art, I will focus on the examiner-cited patents that appear to be highly pertinent to the core aspects of US11968909's independent claims (MgO tunnel barrier with specific characteristics, and ferromagnetic layers).

1. US6219274B1

• Full Citation: US6219274B1, "Ferromagnetic tunnel magnetoresistance effect element and method of producing the same"
• Publication Date: 2001-04-17
• Brief Description: This patent describes a ferromagnetic tunnel magnetoresistance effect element, which includes a tunnel barrier film made of an insulating film containing magnesium oxide, and a ferromagnetic film. The barrier film may be formed by vapor deposition in an oxygen-containing atmosphere.
• Potential Anticipation: This patent describes a ferromagnetic tunnel junction with a MgO-containing insulating film as a tunnel barrier. This directly relates to the core of claims 1, 9, and 15, specifically the use of a magnesium oxide tunnel barrier. Depending on the specifics of its "insulating film containing magnesium oxide" and the characteristics of oxygen vacancy defects or barrier height, it could potentially anticipate the use of crystalline MgOₓ(001) with oxygen vacancy defects (Claim 1) or a magnesium oxide barrier with a specific height (Claims 9 and 15) if these features are inherent or explicitly taught in US6219274B1.

2. JP3593472B2 (Family Cites this family)

• Full Citation: JP3593472B2, "Magnetic element, magnetic memory and magnetic sensor using the same"
• Publication Date: 2004-11-24 (Note: This is a publication date of the Japanese granted patent, the priority date is needed for a full 35 U.S.C. § 102 analysis).
• Brief Description: This patent describes a magnetic element with a tunnel magnetoresistive film having a tunnel barrier layer of an insulating film and two ferromagnetic layers. It mentions the use of MgO as the insulating film.
• Potential Anticipation: Similar to US6219274B1, the use of MgO as a tunnel barrier is directly relevant to claims 1, 9, and 15. The exact details of the MgO layer, such as crystallinity, oxygen vacancy defects, and barrier height, would need to be compared to determine direct anticipation.

3. JP2001007420A (Family Cites this family)

• Full Citation: JP2001007420A, "Magnetoresistive film and magnetic reading sensor using it"
• Publication Date: 2001-01-12
• Brief Description: This patent describes a magnetoresistive film, where the magnetic tunnel junction element uses a tunnel barrier layer of magnesium oxide.
• Potential Anticipation: This reference, like the others, teaches the fundamental concept of using a magnesium oxide tunnel barrier in a magnetoresistive element, which is central to all independent claims of US11968909.

4. JP4149647B2 (Family Cites this family)

• Full Citation: JP4149647B2, "Semiconductor memory device and manufacturing method thereof"
• Publication Date: 2008-09-10 (Note: Publication date is after the priority date of US11968909, suggesting this may not be anticipatory prior art unless its own priority date precedes US11968909's.)
• Brief Description: This patent describes a semiconductor memory device and its manufacturing method, featuring a magnetic tunnel junction element with a tunnel insulating layer made of a crystalline magnesium oxide film.
• Potential Anticipation: This reference specifically mentions a crystalline magnesium oxide film for the tunnel insulating layer, which is highly relevant to Claim 1 of US11968909, which specifies "crystalline MgOₓ(001)". Depending on its priority date, this could be a strong piece of prior art.

Important Considerations for Anticipation Analysis (35 U.S.C. § 102):

For a prior art reference to anticipate a claim under 35 U.S.C. § 102, it must disclose every single limitation of the claim, either explicitly or inherently.

• Claim 1: Requires "crystalline MgOₓ(001) with oxygen vacancy defects and (0<x<1)". The cited patents (especially JP4149647B2 mentioning crystalline MgO) would need to explicitly or inherently disclose the (001) orientation, the presence of oxygen vacancy defects, and the specific range of 'x' for the MgO.
• Claim 9: Requires a tunnel barrier of "magnesium oxide" with a "barrier height in a range of 0.1 eV to 0.85 eV". The cited patents would need to disclose or suggest this specific range for the barrier height.
• Claim 15: Narrows the barrier height range to "0.2 eV to 0.5 eV". This more specific range would also need to be disclosed or suggested by the prior art.

The descriptions above are brief. A full anticipation analysis would require a detailed examination of the complete text and figures of each cited prior art patent to determine if every element of claims 1, 9, and 15 of US11968909 is present. The general trend in the most relevant prior art cited by the examiner is the use of MgO as a tunnel barrier in MTJ devices, which indicates that the novelty of US11968909 likely lies in the specific characteristics and manufacturing methods for the MgO layer, such as the crystalline orientation, oxygen vacancy defects, and resulting barrier height.

Obviousness Analysis under 35 U.S.C. § 103

This analysis identifies combinations of prior art that would render claims of US Patent 11968909 obvious to a person having ordinary skill in the art (POSA) as of the priority date of March 12, 2004. The patent aims to overcome the limitations of conventional Magnetic Tunnel Junction (MTJ) devices, particularly those using amorphous Al—O tunnel barriers, which suffered from low magnetoresistance (MR) ratios and output voltages, hindering the scalability of MRAMs. The invention proposes using a crystalline MgOₓ(001) tunnel barrier, potentially with oxygen vacancy defects, or an MgO barrier with a specific barrier height, to achieve enhanced performance.

Background Prior Art

The patent itself acknowledges the existence of foundational MTJ technology, specifically referencing:

• Non-Patent Document 1 (NPD1): D. Wang, et al.: Science 294 (2001) 1488. This document teaches a MTJ device comprising a tunnel barrier layer made of an amorphous Al—O layer sandwiched between first and second ferromagnetic electrodes. NPD1 (or similar related art) also highlights the inherent problems with amorphous Al—O barriers, such as a small magnetoresistance (around 70%) and a drastic decrease in magnetoresistance upon application of bias voltage, leading to output voltages of no more than 200 mV. These limitations made the devices unsuitable for high-integration memory applications.
• W. Wulfhekel, et al.: Appl. Phys. Lett. 78 (2001) 509. This publication is cited in US11968909 for estimating the "ideal value" of a perfect single-crystal barrier height. This demonstrates that the concept of single-crystal tunnel barriers and their theoretical properties were known and being studied prior to the invention's priority date.

Combination of Prior Art for Obviousness

The core inventive step disclosed in US11968909 is the use of a crystalline MgOₓ(001) tunnel barrier, potentially with oxygen vacancy defects to tune its barrier height, to achieve superior MTJ performance. A POSA, motivated by the known limitations of amorphous Al—O barriers (as taught by NPD1), would have sought to improve MR ratio and output voltage.

Based on the patent's own disclosures and general knowledge in the field prior to March 2004, the following combination could render the claims obvious:

Combination 1: D. Wang (NPD1) + W. Wulfhekel et al. + General Knowledge in Magnetic Tunnel Junctions

1. Baseline MTJ Structure (NPD1): A POSA would be familiar with the basic MTJ structure, comprising ferromagnetic electrodes separated by a tunnel barrier, as taught by NPD1. They would also understand the existing problems of low MR ratio and output voltage associated with amorphous Al—O barriers in MRAM applications, and the need for improvement.

2. Motivation to Explore Crystalline Barriers: With the goal of improving MTJ performance, particularly increasing the MR ratio and output voltage, a POSA would naturally consider alternative tunnel barrier materials that could offer better electronic coherence. The patent itself notes that the inventors theorized that because MgO is a crystal, "the coherent states of electrons are conserved during the tunneling process," leading to higher MR ratios. The field of spintronics and magnetoresistance was actively researching materials that maintain spin coherence.

3. Suggestion of Crystalline MgO (W. Wulfhekel et al. & General Knowledge): The reference to W. Wulfhekel et al. in the patent indicates that the properties of single-crystal tunnel barriers were being investigated. Given MgO's known crystalline structure (specifically NaCl structure, and its (001) plane being relevant for interfaces, as the patent clarifies), and its established use in thin-film depositions in other contexts (e.g., as a substrate or insulating layer), a POSA would have considered crystalline MgO as a candidate material for a tunnel barrier. Prior art related to epitaxy and lattice matching for improved device performance would further motivate the selection of a crystalline barrier that can be grown with preferred orientation, such as (001), particularly when interfacing with BCC ferromagnetic materials like Fe(001), as described in the patent.

4. Motivation for Oxygen Vacancy Defects to Adjust Barrier Height (General Knowledge/Optimization): The patent explicitly states that oxygen vacancy defects in MgOₓ lead to a decrease in potential barrier height, thereby increasing tunneling current and lowering resistance, optimizing it for MRAM. It further specifies desirable ranges for 'x' (0.9 < x < 1, preferably 0.98 < x < 1, or 0.99 < x < 1) and corresponding barrier heights (0.1-0.85 eV, or 0.2-0.5 eV).

   • A POSA would be aware that the electrical properties of oxide films, including their resistance and barrier height, can be tuned by controlling stoichiometry, such as by introducing oxygen vacancies or doping. This is a common technique in materials science for optimizing semiconductor and insulator properties.
   • The patent describes the observation of high oxygen partial pressures during MgO evaporation, indicating "separation of oxygen from MgO during the deposition" and the "possibility that there are oxygen vacancy defects such as MgOₓ (0.9 < x < 1)." This suggests that such defects could arise naturally or be intentionally controlled during known deposition processes like electron-beam evaporation, as described in the patent. Therefore, a POSA attempting to optimize a crystalline MgO barrier would be motivated to control oxygen vacancies to tune the barrier height and resistance for desired MRAM performance.

Obviousness of Specific Claims

• Claim 1 (Crystalline MgOₓ(001) with oxygen vacancy defects): The combination of NPD1 (basic MTJ and problems with Al-O), W. Wulfhekel et al. (concept of single-crystal barriers), and general knowledge regarding crystalline materials, epitaxial growth, and the tuning of oxide properties through defects would lead a POSA to replace the amorphous Al—O with crystalline MgO(001). The specific introduction or control of oxygen vacancy defects (0 < x < 1) to optimize electrical characteristics would be a predictable design choice based on known materials science principles.

• Claim 9 (Magnesium oxide tunnel barrier with barrier height 0.1 eV to 0.85 eV): Once motivated to use MgO as the tunnel barrier, and understanding the impact of stoichiometry on barrier properties, a POSA would systematically vary deposition conditions or post-treatment to achieve a desired barrier height. The patent explicitly states that oxygen vacancy defects lead to the desired lower barrier heights (0.10 to 0.85 eV, more specifically 0.2 to 0.5 eV) compared to the ideal 3.6 eV for perfect MgO. Achieving this specific range would be an expected optimization for improving tunneling current and reducing resistance, directly addressing the limitations identified in NPD1.

• Claim 15 (Magnesium oxide tunnel barrier with barrier height 0.2 eV to 0.5 eV): This claim presents a narrower, more preferred range of barrier height than Claim 9. If the broader range in Claim 9 is obvious, then selecting a sub-range that yields even better performance would be a matter of routine experimentation and optimization for a POSA, especially given the patent's teaching that this range specifically "is expected" using their method and "indicating that the resistance of the tunnel barrier can be lowered."

In conclusion, a POSA, motivated by the known deficiencies of amorphous Al—O MTJ barriers (NPD1) and aware of the potential advantages of crystalline materials for spin transport (W. Wulfhekel et al.), would have been led to investigate crystalline MgO as a tunnel barrier. Further, the control of oxygen stoichiometry or defects to tune the barrier height for optimal MRAM performance would be a logical and predictable optimization, thus rendering the claimed inventions obvious.