PTAB

IPR2017-00392

Micron Technology Inc v. Flamm Daniel

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Process Optimization In Gas Phase Dry Etching
  • Brief Description: The ’849 patent discloses methods for optimizing gas phase dry etching processes used in fabricating semiconductor devices. The core of the invention is a method of modeling gaseous diffusion and surface chemical reactions to predict and improve etch rate uniformity by extracting and then using a "surface reaction rate constant" derived from empirical etch profile data, thereby avoiding conventional trial-and-error approaches.

3. Grounds for Unpatentability

Ground 1: Claims 1-29 are obvious over Alkire in view of Kao.

  • Prior Art Relied Upon: Alkire (J. Electrochem. Soc., Mar. 1985) and Kao (J. Electrochem. Soc., Mar. 1990).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that the combination of Alkire and Kao taught all limitations of the independent claims (e.g., claim 1). Alkire disclosed the initial steps: providing a plasma etching apparatus with a substrate and etching to create a non-uniform profile. Alkire's model conceptually defined etch rate data via an equation for film thickness over time at a given radial position. However, Alkire did not teach experimentally measuring this data or extracting a rate constant from it. Petitioner asserted that Kao supplied these critical missing limitations. Kao taught using a profilometer to measure etch depths at multiple spatial coordinates, defining empirical etch rate data. Crucially, Kao then used this data in a best-fit analysis with its mathematical model to extract a "surface reaction rate constant" (ke). Finally, both references taught using the model and its derived constants for device fabrication, with Alkire focused on finding "optima" for throughput and Kao focused on improving circuit yields by reducing non-uniformity.
    • Motivation to Combine: A POSITA would combine Kao with Alkire to improve Alkire’s theoretical model with real-world experimental data, thereby increasing its predictive capability for process development and reactor design. This would allow for optimizing throughput and yield while avoiding costly physical trial-and-error prototyping. Kao explicitly identified prior work like Alkire as providing models that lacked experimental validation, directly suggesting the combination to test and confirm the theoretical model’s accuracy.
    • Expectation of Success: Success was expected because both references address the same problem (etch non-uniformity) in the same field (plasma etching) using compatible modeling assumptions, such as diffusion-limited, first-order reactions. Applying Kao's established measurement and calculation techniques to Alkire's model would yield predictable and useful results.

Ground 2: Claims 1-29 are obvious over Alkire in view of Kao and Flamm.

  • Prior Art Relied Upon: Alkire (J. Electrochem. Soc., Mar. 1985), Kao (J. Electrochem. Soc., Mar. 1990), and Flamm (J. of Applied Physics, May 1981).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner asserted this ground as an alternative, specifically targeting a potential claim construction of "surface reaction rate constant" that requires a temperature dependence defined by an Arrhenius relationship. While Alkire and Kao taught a temperature-dependent constant, Flamm explicitly disclosed an experimentally validated Arrhenius equation that describes the surface reaction rate for etching silicon with fluorine atoms as a function of temperature. This equation separates the etchant concentration from the temperature-dependent portion, which Petitioner argued is the claimed "surface reaction rate constant" defined by an Arrhenius relationship. Petitioner contended this combination met every claim limitation even under a narrower claim construction.
    • Motivation to Combine: The motivation to add Flamm stemmed directly from Kao. Kao explicitly compared its calculated surface reaction rate constant at a specific temperature to the rate constant disclosed in the Flamm paper. This direct reference would have prompted a POSITA to consult Flamm to understand the temperature dependency more fully and incorporate Flamm's Arrhenius expression to enhance the Alkire/Kao model. Doing so would make the combined model more robust and broadly applicable over a range of operating temperatures, a clear and desirable improvement.
    • Expectation of Success: A POSITA would have a high expectation of success in incorporating Flamm's Arrhenius equation, as it provided a well-known mathematical formula to describe a physical parameter (temperature dependence) that was already a known variable in the Alkire and Kao models.

4. Key Claim Construction Positions

  • The petition’s arguments centered on the construction of the term "surface reaction rate constant," which appeared in all challenged claims.
  • Primary Proposed Construction: Petitioner proposed that the term be construed as "a temperature-dependent reaction rate constant for the chemical reaction between a gas phase etchant and the surface of an etchable material." This construction was based on the patent's specification and prosecution history, where the patentee described the constant (ks) as being predominantly dependent on temperature.
  • Alternative Proposed Construction: As a fallback, Petitioner proposed a narrower construction: "a reaction rate constant having a temperature dependence defined by an Arrhenius relationship..." This alternative was offered to address specific language in the specification and provided the basis for the secondary obviousness ground including the Flamm reference, which explicitly taught such a relationship.

5. Relief Requested

  • Petitioner requested the institution of an inter partes review and the cancellation of claims 1-29 of Patent 5,711,849 as unpatentable under 35 U.S.C. §103.