PTAB

IPR2025-01450

SK hynix Inc. v. Advanced Memory Technologies, LLC

Log In

1. Case Identification

2. Patent Overview

  • Title: Booster Circuit
  • Brief Description: The ’018 patent relates to a two-row booster circuit, also known as a charge pump, designed for use in integrated circuits. The technology uses one or more analog comparison circuits to compare potentials from the two rows of boosting cells and output a well bias potential to a deep N-well region of the circuit’s switching elements, with the stated goal of suppressing current consumption and minimizing the body effect.

3. Grounds for Unpatentability

Ground 1: Anticipation by Cordoba - Claims 1, 3, 4, and 7 are anticipated by Cordoba.

  • Prior Art Relied Upon: Cordoba (Patent 7,382,177).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Cordoba discloses every limitation of the challenged claims. Cordoba teaches a two-row booster circuit (charge pump) with a single boosting stage. It includes an analog comparison circuit that compares the input potentials of the two rows and outputs a well bias potential to the deep N-well of the circuit's n-channel transistors. This structure, featuring a P-well within a deep N-well on a P-type substrate, was asserted to meet the multi-well limitations of claim 1. For claim 4, Petitioner contended that Cordoba’s comparison circuit is explicitly designed to output the lower (more negative) of the two input potentials to the deep N-well to prevent forward biasing. The structure of Cordoba’s comparison circuit transistors within the multi-well structure was also argued to anticipate the limitations of dependent claim 7.

Ground 2: Anticipation by Meng - Claims 1-3, 9, 10, and 12 are anticipated by Meng.

  • Prior Art Relied Upon: Meng (Patent 6,501,325).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner asserted that Meng teaches a two-row, cross-coupled charge pump with multiple boosting stages, anticipating the structure of claim 1. Meng’s circuit includes an analog comparison circuit, formed by diode-connected transistors, that compares potentials from each row and outputs a well bias potential to a common deep N-well shared by the switching elements in both rows. Petitioner contended that this diode-based comparison circuit inherently selects and outputs the higher of the two potentials, thus anticipating claim 3. The petition noted that its interpretation of "input potential" as the output potential of a preceding stage is consistent with the Patent Owner's infringement contentions in related district court litigation. Further, Meng’s disclosure of diode elements between the boosting stage outputs and the common well region was argued to anticipate claim 9, while the shared deep N-well structure anticipated claim 10.

Ground 3: Obviousness over Meng and Park - Claims 1-10 and 12 are obvious over Meng in view of Park.

  • Prior Art Relied Upon: Meng (Patent 6,501,325) and Park (Patent 6,914,791).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground was presented as an alternative to Ground 2. Meng provides the fundamental two-row booster circuit with an analog comparison circuit and multi-well regions. Petitioner argued that if the Board were to find Meng does not explicitly teach applying the generated well bias potential to the first well region (the deep N-well) of the switching element, Park supplies this missing element. Park expressly teaches applying a well bias potential, derived from a boosted voltage, to both the P-well and deep N-well of a charge transfer device to reduce the body effect.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine Meng and Park to improve the performance of Meng’s circuit. Park’s explicit teaching of applying a well bias to reduce the body effect directly addresses a known problem in charge pumps and aligns with the stated objectives in Meng, such as improving efficiency. A POSITA would have been motivated to incorporate Park’s proven technique into Meng’s efficient two-row architecture to achieve a predictable improvement.
    • Expectation of Success: Petitioner asserted a POSITA would have had a high expectation of success. Both references operate in the same field of charge pump design, utilize the same set of well-known circuit components, and address the same underlying technical challenges. Integrating Park’s well-biasing scheme into Meng’s structure would have been a straightforward application of known principles to achieve a predictable result.

4. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-10 and 12 of the ’018 patent as unpatentable.