uPI Semiconductor Inc v. Force MOS Technology Co Ltd

1. Case Identification

• Case #: IPR2025-00920
• Patent #: 7,812,409
• Filed: April 24, 2025
• Petitioner(s): uPI Semiconductor Corp.
• Patent Owner(s): Fwu-Iuan Hshieh
• Challenged Claims: 1-5

2. Patent Overview

• Title: Trench MOSFET with cell layout, ruggedness, truncated corners
• Brief Description: The ’409 patent discloses a trenched power semiconductor device, specifically a Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET), with a closed-cell layout. The design features trenched gates with rounded corners and a circular trench contact to mitigate parasitic bipolar latch-up, which can cause device failure, and thereby enhance the device's ruggedness.

3. Grounds for Unpatentability

Ground 1: Obviousness over Bulucea and Hshieh - Claims 1-5 are obvious over Bulucea in view of Hshieh.

• Prior Art Relied Upon: Bulucea (Patent 5,072,266) and Hshieh (Application # 2006/0273390).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Bulucea identified the same problems addressed by the ’409 patent: parasitic bipolar latch-up caused by surface breakdown, particularly at the sharp corners of trench gates in closed-cell MOSFETs. Bulucea taught that rounding these corners—either through a photolithography technique ("Litho-rounding") or sacrificial oxidation—would mitigate this issue. Bulucea also taught using a deep p+ dopant region to create a low-resistivity path away from the trench edge, further preventing latch-up. While Bulucea established the problem and the principles for the solution, Petitioner asserted Hshieh taught the specific structural implementation of a circular ("cylindrical shaped") source-body trench contact that penetrates the source and body regions to reduce on-resistance. Hshieh disclosed a complete contact structure, including a Ti/TiN/W metal plug, resistance-reduction layers (Ti), and source metal layers (AlCu alloys), which Petitioner mapped to the limitations of dependent claims 2-5.
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would have been motivated to combine the references. Bulucea provided the foundational teaching on how to solve the parasitic latch-up problem by rounding corners and creating a low-resistivity path via a deep dopant region. Hshieh provided a known, specific implementation of a deep circular trench contact to address the related, and often competing, goal of reducing on-resistance. A POSITA would combine Hshieh's known contact structure with Bulucea's cell design to implement Bulucea’s principles for preventing latch-up while simultaneously optimizing for low on-resistance.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success in combining the known contact structure of Hshieh with the cell design of Bulucea, as it involved applying a known technique to a known device to achieve a predictable improvement in device ruggedness and performance.

Ground 2: Obviousness over Bulucea and Uno - Claim 1 is obvious over Bulucea in view of Uno.

• Prior Art Relied Upon: Bulucea (Patent 5,072,266) and Uno (Patent 6,984,864).

• Core Argument for this Ground:

  • Prior Art Mapping: This ground relied on the same teachings from Bulucea regarding the identification of the parasitic latch-up problem and the solution of rounding trench corners. As a secondary reference, Petitioner cited Uno, which disclosed a trench MOSFET with a circular trench contact that is explicitly a "Deep Body Trench Contact." Uno taught that its contact structure reaches below the silicon surface to connect with a p+ contact dopant region, providing the low-resistivity path that Bulucea identified as a solution to latch-up. Uno's circular contact structure was asserted to teach the limitations of claim 1, including the penetration through source and body regions and separation from the trench gates.
  • Motivation to Combine: The motivation was similar to Ground 1. A POSITA, starting with Bulucea's teachings on preventing latch-up, would have looked to known contact structures to implement Bulucea's proposed deep body diffusion principle. Uno provided a known deep body trench contact designed to improve device characteristics. A POSITA would have been motivated to incorporate Uno's contact into Bulucea's cell layout to achieve the desired ruggedness against latch-up.
  • Expectation of Success: A POSITA would have expected success in applying Uno's established contact design to Bulucea's MOSFET architecture, as it was a straightforward combination of known elements for their intended purposes.

• Additional Grounds: Petitioner asserted additional obviousness challenges against claim 1 over Bulucea in view of Huang (Patent 6,037,628) and against claims 1 and 3-5 over Bulucea in view of Bhalla (Application # 2006/0180855). These grounds relied on the same core argument: that Bulucea identified the problem and rounding solution, while Huang and Bhalla each provided alternative, known "Deep Body Trench Contact" structures that a POSITA would have been motivated to incorporate into Bulucea's design.

4. Key Claim Construction Positions

• "circular trench contact": Petitioner proposed this term be construed as "a trench contact that has a horizontal cross section that is circular or approximately circular." Petitioner argued this construction was necessary to include the patent’s only disclosed embodiment, which is described as being formed by a photolithography process that makes square features "approximately circular."
• "source metal": Petitioner proposed this term be construed broadly as "a metal layer." This term appeared only in the claims and figures, and Petitioner argued a broad construction was appropriate.

5. Arguments Regarding Discretionary Denial

• §325(d) (Advanced Bionics/Becton): Petitioner argued against discretionary denial because none of the prior art references (Bulucea, Hshieh, Uno, Huang, Bhalla) were considered during the original prosecution of the ’409 patent. Further, the arguments presented in the petition were new and distinct from those raised during prosecution or in any other post-grant proceeding.
• §314(a) (Fintiv): Petitioner contended that Fintiv factors strongly favored institution. While acknowledging a co-pending IPR and a district court case involving the ’409 patent, Petitioner asserted it was not a party to those proceedings and there was no overlap in the issues raised. Petitioner argued the merits of the petition were strong, favoring institution under Fintiv factor 6.

6. Relief Requested

• Petitioner requested institution of an inter partes review and cancellation of claims 1-5 of the ’409 patent as unpatentable.