NVIDIA Corp v. Ocean Semiconductor LLC

1. Case Identification

• Case #: IPR2022-01288
• Patent #: 7,629,211
• Filed: August 9, 2022
• Petitioner(s): NVIDIA Corporation
• Patent Owner(s): Ocean Semiconductor LLC
• Challenged Claims: 1-5

2. Patent Overview

• Title: Method of Forming a Field Effect Transistor
• Brief Description: The ’211 patent describes a method for fabricating field-effect transistors with strained silicon to improve charge carrier mobility. The method involves creating cavities in the source/drain regions and filling them with a compound material (e.g., SiGe or SiC) where the concentration of the strain-creating element (Ge or C) gradually increases in a vertical direction to reduce crystal defects.

3. Grounds for Unpatentability

Ground 1: Anticipation and Obviousness of Claims 1-5 over Orlowski

• Prior Art Relied Upon: Orlowski (Application # 2006/0166492).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Orlowski anticipates or, in the alternative, renders obvious all challenged claims. Orlowski teaches a method for forming MOSFETs using stress-inducing source/drain (SISD) structures, such as SiGe or SiC, to improve carrier mobility. It explicitly discloses forming recesses (cavities) adjacent to the gate electrode and filling them with these materials. Critically, Orlowski teaches a “graded approach” where the SISD structure is formed in multiple sub-layers (e.g., 141, 142, 143), with each successive vertical layer having a higher concentration of the first chemical element (Ge or C). This directly maps to the ’211 patent’s core limitation of a first concentration ratio in a lower portion being smaller than a second concentration ratio in an upper portion, with the ratio increasing vertically. Orlowski provides specific percentage ranges for Ge and C that satisfy the dependent claims’ requirements for concentration ratios greater than 1% and 10%. Finally, Orlowski describes forming the SISD structures via a selective epitaxial growth process, meeting the limitation of claim 5.
  • Motivation to Combine (for §103 grounds): As an alternative to anticipation, Petitioner argued a person of ordinary skill in the art (POSITA) would have been motivated to apply Orlowski's teachings. The motivation stemmed from the shared goal of improving transistor performance by creating strained silicon while mitigating known problems like crystal defects, which Orlowski’s graded approach directly addresses.
  • Expectation of Success (for §103 grounds): A POSITA would have had a high expectation of success because Orlowski uses known semiconductor fabrication techniques, such as selective epitaxial growth, to achieve its graded concentration profile.

Ground 2: Obviousness of Claims 1-5 over Koontz

• Prior Art Relied Upon: Koontz (Application # 2006/0134873).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner contended that Koontz renders all challenged claims obvious. Koontz is directed to fabricating MOSFETs with "tailored channel strain profiles" by forming composition-controlled recess structures (SiGe or SiC) in the source/drain regions. Koontz explicitly teaches that the strain composition can be varied by layers, linearly, or non-linearly to create different strain profiles. Figures in Koontz depict layered structures and strain profiles that increase in magnitude from the bottom of the recess to the top. Petitioner argued this disclosure would lead a POSITA to implement a structure where the concentration of Ge or C, and thus the claimed concentration ratio, increases vertically from a bottom surface of the cavity. Koontz also provides exemplary concentrations (e.g., 5%, 15%, 25% Ge) that satisfy the numerical limitations of dependent claims 3 and 4.
  • Motivation to Combine (for §103 grounds): The motivation was inherent in Koontz’s disclosure and the background knowledge of a POSITA. The art recognized that abrupt changes in lattice strain caused dislocation defects. Koontz’s teaching of tailoring strain profiles, including profiles that change gradually, provided a direct motivation to implement a smoothly increasing concentration of the strain-inducing element from bottom to top to avoid such defects and optimize transistor performance.
  • Expectation of Success (for §103 grounds): A POSITA would have expected success in implementing Koontz’s teachings because controlling the concentration of reactants during epitaxial growth to create graded layers was a well-known and conventional technique in semiconductor manufacturing.

Ground 3: Obviousness of Claims 1-5 over Koontz in view of Orlowski

• Prior Art Relied Upon: Koontz (Application # 2006/0134873) and Orlowski (Application # 2006/0166492).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground asserted that even if Koontz alone did not render the claims obvious, the combination with Orlowski would. Koontz provides the foundational teaching of creating tailored, multi-layered strain profiles in source/drain recesses. Orlowski reinforced and provided additional detail for implementing a profile where the concentration of the strain-inducing element (Ge or C) explicitly increases from bottom to top.
  • Motivation to Combine (for §103 grounds): A POSITA reviewing Koontz’s flexible approach to tailoring strain profiles would have been motivated to consult a reference like Orlowski, which addresses the same technical problem and provides concrete examples of a vertically graded concentration profile. Orlowski further teaches benefits of this specific profile, such as enabling impurity doping in the upper layers without degrading the stress-inducing characteristics, providing a compelling reason to apply its specific graded composition to Koontz’s method.
  • Expectation of Success (for §103 grounds): The combination involved applying known techniques from analogous art to achieve predictable results, ensuring a high expectation of success.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under Fintiv would be inappropriate. The petition’s merits were described as compelling, which alone favors institution. Furthermore, Petitioner filed a stipulation in the parallel district court litigation to not pursue the same invalidity grounds. The district court case was also in a very early stage, with no claim construction hearing held or significant investment by the court, meaning factors of judicial economy strongly favored institution of the IPR.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-5 of the ’211 patent as unpatentable.