Western Digital Technologies v. Godo Kaisha IP Bridge 1

1. Case Identification

• Case #: IPR2025-00701
• Patent #: 10,367,138
• Filed: March 6, 2025
• Petitioner(s): Western Digital Technologies, Inc.
• Patent Owner(s): Godo Kaisha IP Bridge 1
• Challenged Claims: 1-14

2. Patent Overview

• Title: Magnetic Tunnel Junction Device
• Brief Description: The ’138 patent discloses a magnetic tunnel junction (MTJ) device for use in technologies like MRAM. The invention centers on a structure comprising first and second ferromagnetic electrodes made of a crystallized Cobalt-Iron-Boron (CoFeB) alloy, separated by a polycrystalline Magnesium Oxide (MgO) barrier layer with a preferred crystal orientation.

3. Grounds for Unpatentability

Ground 1: Obviousness over Parkin, Wang, and Soukup - Claims 1-4, 7-10, and 12-13 are obvious over Parkin in view of Wang and Soukup.

• Prior Art Relied Upon: Parkin (Patent 7,270,896), Wang (a 2004 IEEE conference paper), and Soukup (a 1970 journal article).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that the primary reference, Parkin, teaches the core structure of the claimed MTJ device. Parkin disclosed an MTJ with a polycrystalline, highly textured (100) MgO tunnel barrier disposed between two ferromagnetic layers. Parkin’s top electrode was made of an amorphous CoFeB alloy, and Parkin taught that annealing this structure caused the CoFeB layer adjacent to the MgO barrier to crystallize, satisfying the “entirely crystallized” limitation of independent claims 1, 7, and 12. Parkin also taught that its MgO layer composition could differ from stoichiometric, suggesting the presence of oxygen vacancy defects (MgOx where 0<x<1). To meet the limitation of both electrodes comprising CoFeB, Petitioner pointed to Wang, which taught that replacing conventional CoFe electrodes with CoFeB electrodes on both sides of a barrier layer significantly increases the Tunnel Magnetoresistance (TMR). For claims reciting a specific barrier height (0.2 to 0.5 eV), Petitioner relied on Soukup, which taught that nonstoichiometry (i.e., oxygen vacancies) in MgO films lowers the barrier height and disclosed an experimental barrier height of "about 0.33 eV," squarely within the claimed range.
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine Parkin with Wang for the explicit and predictable purpose of improving TMR, a primary goal in MTJ research at the time. A POSITA, knowing from Parkin that the MgO layer could be non-stoichiometric, would have been motivated to consult a reference like Soukup to understand and control the effect of such defects. Soukup provided the direct teaching that oxygen vacancies lower the barrier height, motivating a POSITA to control this known parameter to optimize device performance and facilitate tunneling current.
  • Expectation of Success: Petitioner asserted that a POSITA would have a high expectation of success. Combining Parkin and Wang involved substituting a known, superior material (CoFeB) into a known device structure (Parkin’s MTJ) to achieve an expected improvement in a known metric (TMR). Applying Soukup’s teachings to achieve a specific barrier height was presented as routine optimization of a known variable.

Ground 2: Obviousness over Bowen, Wang, and Soukup - Claims 1-14 are obvious over Bowen in view of Wang and Soukup.

• Prior Art Relied Upon: Bowen (a 2001 journal article), Wang (a 2004 IEEE conference paper), and Soukup (a 1970 journal article).

• Core Argument for this Ground:

  • Prior Art Mapping: Petitioner presented Bowen as an alternative starting point, which described a single-crystal Fe/MgO/FeCo MTJ. Petitioner argued it would have been obvious to modify Bowen in several ways. First, a POSITA would replace Bowen's single-crystal MgO barrier with a polycrystalline MgO barrier—a known and predictable alternative used to improve manufacturability, as taught by references like Parkin. Second, following the teachings of Wang, a POSITA would replace Bowen’s Fe and FeCo electrodes with CoFeB electrodes on both sides of the barrier to achieve higher TMR. Petitioner argued that upon making these substitutions, a POSITA would have routinely annealed the device, as taught by both Bowen and Wang, resulting in the "entirely crystallized" CoFeB layers claimed. As in the first ground, Soukup was cited to teach that oxygen vacancies in the MgO barrier would lower the barrier height to the claimed 0.2-0.5 eV range. Bowen itself acknowledged that its observed low barrier height (0.9 eV) could be due to "stoichiometric...inhomogeneities," i.e., oxygen vacancy defects.
  • Motivation to Combine: The motivations were analogous to Ground 1. A POSITA would be driven to combine the teachings of Bowen and Wang to improve the primary performance metric of TMR. The motivation to use a polycrystalline MgO layer stemmed from known advantages in manufacturing flexibility. Finally, the motivation to apply Soukup's teachings was to optimize a known device parameter (barrier height) by controlling a known variable (oxygen vacancy defects).
  • Expectation of Success: Success was predictable because each step involved combining familiar elements using known methods to yield predictable results. Replacing electrodes with CoFeB was a known strategy for improving TMR, and using polycrystalline MgO was a common design choice.

• Additional Grounds: Petitioner asserted an additional obviousness challenge against claims 5-6, 11, and 14 based on Parkin in view of Wang, Soukup, and Bowen. This ground added Bowen for its teaching on using the well-known Simmons' formula to calculate barrier height by fitting J-V characteristics, as recited in those specific claims.

4. Arguments Regarding Discretionary Denial

• Fintiv Factors: Petitioner argued that discretionary denial under §314(a) is unwarranted. The parallel district court litigation was stayed at a very early pleading stage pending the outcome of this and other IPRs. Given the 48.9-month median time-to-trial in the relevant district (N.D. Cal.), a trial is highly unlikely to occur before a final written decision (FWD) from the PTAB.
• §325(d) Factors: Petitioner argued denial under §325(d) is also inappropriate because the petition raises new art and arguments not previously considered by the USPTO examiner. Specifically, Soukup was never presented to the examiner. Furthermore, Petitioner contended the examiner committed a material error by initially allowing the claims based on a belief that the prior art did not teach the claimed barrier height or oxygen vacancy defects. The same examiner later reversed this position when examining related applications, finding that other prior art did teach these features, just before the Patent Owner abandoned those applications. This reversal was presented as evidence that the initial examination was flawed and warrants a new review.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-14 of Patent 10,367,138 as unpatentable.