PTAB

IPR2019-00004

Samsung Electronics Co Ltd v. Invensas Bonding Technologies Inc

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Method for Low-Temperature Wafer Bonding
  • Brief Description: The ’549 patent discloses a method for bonding semiconductor wafers at low or room temperature. The process involves activating a wafer surface via slight etching, terminating the surface with a desired chemical species to form temporary bonds, and then bringing it into contact with another surface to form a permanent chemical bond.

3. Grounds for Unpatentability

Ground 1: Claims 53, 60-63, 67-70, 74-77, 83-86, and 89 are obvious over Li in view of Reiche.

  • Prior Art Relied Upon: Li (Y. Li et al., a 1998 journal article) and Reiche (M. Reiche et al., a 1997 conference proceeding).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Li disclosed all steps of independent claim 53. Li taught a systematic low-temperature process for bonding silicon or SiO2 surfaces that began with a cleaning process followed by surface activation using ammonia or oxygen plasma in a reactive ion etcher (RIE). The wafers were then bonded at room temperature and annealed at 200°C to achieve a bond strength greater than 1000 mJ/m². Petitioner asserted that Li's plasma activation process inherently terminated the surface with species (Si-OH or Si-N-H groups) capable of forming strong chemical bonds, meeting the core limitations of claim 53. To the extent Li did not explicitly teach etching, Petitioner argued Reiche supplied this element. Reiche investigated the effect of RIE in oxygen-containing plasmas on wafer bonding and surface roughness, teaching that this process etches the surface and leads to very strong bonds.
    • Motivation to Combine: A POSITA would combine Li and Reiche because both references address improving low-temperature wafer bonding. Li disclosed using plasma activation, and Reiche provided a detailed analysis of the effects of such RIE plasma treatments, including the impact on surface roughness—a critical parameter for successful bonding mentioned by Li. Reiche’s teachings would thus have been seen as a way to optimize the process disclosed by Li.
    • Expectation of Success: A POSITA would have a reasonable expectation of success because Reiche’s investigation into RIE plasma directly corresponded to the plasma activation step taught in Li. Applying Reiche's findings to control surface roughness during Li's plasma process was a straightforward optimization of known parameters.

Ground 2: Claims 53, 61, 67-68, 74-75, 83-84, and 86 are obvious over Kissinger in view of Reiche.

  • Prior Art Relied Upon: Kissinger (G. Kissinger et al., a 1993 journal article) and Reiche.
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner contended that Kissinger disclosed the key elements of claim 53 by teaching a process to achieve high bond strength at low temperatures. Kissinger used oxygen plasma to activate silicon wafer surfaces, followed by an RCA clean to terminate the surfaces with OH-groups. The wafers were then bonded at room temperature and annealed at 200°C, achieving surface energies of almost 2000 mJ/m². Petitioner argued this process met the claim limitations of exposing to plasma, terminating, bonding, and obtaining a high-strength chemical bond. Reiche was cited to supplement Kissinger’s disclosure regarding the etching nature of the plasma activation and its minimal effect on surface roughness, as required by dependent claim 84.
    • Motivation to Combine: A POSITA would combine Kissinger and Reiche to optimize the plasma activation process. Kissinger taught the benefits of plasma activation for low-temperature bonding, and Reiche provided specific details on how to implement RIE plasma treatments to achieve strong bonds while minimizing changes to surface roughness—a known factor for success that Kissinger also identified as important.
    • Expectation of Success: The combination was expected to succeed because Reiche detailed the precise type of oxygen plasma process Kissinger employed. A POSITA would have understood how to apply Reiche’s teachings on RIE parameters to Kissinger's process to ensure the surface was properly activated and etched without degrading surface smoothness.

Ground 3: Claim 92 is obvious over Li and Reiche in view of Tong (Mar. 1994).

  • Prior Art Relied Upon: Li, Reiche, and Tong (Mar. 1994) (Q. Tong et al., a 1994 journal article).

  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner argued that the combination of Li and Reiche established the base process of bonding wafers as discussed in Ground 1. The additional limitation of claim 92, "storing said wafers in ambient" after bringing them together, was taught by Tong (Mar. 1994). Tong investigated the effects of storing bonded wafer pairs and found a "pronounced increase of interface energy" for wafers stored in air at room temperature.
    • Motivation to Combine: A POSITA seeking to maximize bond strength, a primary goal taught by Li, would be motivated to incorporate the post-bonding storage step from Tong (Mar. 1994). Tong explicitly taught that ambient storage was a simple and effective technique to enhance bond strength for room-temperature bonded wafers, making it a natural and desirable addition to the Li/Reiche process.
    • Expectation of Success: Success was expected because adding a post-bonding ambient storage step is a non-invasive, simple modification to an existing process. Tong demonstrated the technique's effectiveness, giving a POSITA confidence it would improve the bond strength of wafers bonded using the Li/Reiche method.

  • Additional Grounds: Petitioner asserted additional obviousness challenges, including combining Kissinger, Reiche, and Tong (1994) to argue that terminating a surface with nitrogen-containing species was obvious, and combining Kissinger, Reiche, and Tong (Mar. 1994) for claim 92.

4. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 53, 60-63, 67-70, 74-77, 83-86, 89, and 92 of the ’549 patent as unpatentable under 35 U.S.C. §103.