`
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`____________
`
`TAIWAN SEMICONDUCTOR MANUFACTURING CO., LTD,
`Petitioner,
`
`v.
`
`GODO KAISHA IP BRIDGE 1,
`Patent Owner.
`____________
`
`IPR2017-018411
`Patent 7,893,501
`____________
`
`SUR-REPLY DECLARATION OF ALEXANDER D. GLEW
`PURSUANT TO JULY 20, 2018 ORDER (PAPER NO. 26)
`
`1 Case IPR2017-01842 has been consolidated with this proceeding(cid:15916)
`IP Bridge Exhibit 2024
`TSMC v. Godo Kaisha IP Bridge 1
`IPR2017-01841
`
`
`
`I, Alexander D. Glew, declare:
`
`1.
`
`2.
`
`My compensation and credentials are provided in Exhibit 2007.
`
`Agata and Rashed do not support Dr. Shanfield’s assertion that “the
`
`claim requirement that ‘the MISFET includes: an active region’ is met by the prior
`
`art relied upon in the grounds” (Ex. 1027, ¶¶ 10, 17-18) for the reasons below.
`
`3.
`
`Neither Agata nor Rashed refers to a MISFET (or other transistor)
`
`that “includes: an active region” as required by claim 1 of the ’501 patent.
`
`4.
`
`Agata’s “active region 2” that Dr. Shanfield refers to (Ex. 1027, ¶17)
`
`is a region where a multi-transistor “semiconductor device” (more specifically a
`
`“sense amplifier”) is formed. Ex. 1025, 5:9-18. Agata does not refer to any of the
`
`transistors (MOSFETs) in Agata’s device as “including” the active region 2. To
`
`the contrary, Agata states that it is the “sense amplifier [that] includes … [the]
`
`active region 2.” Id., 5:9-18; Abstract (“[t]he device includes … an active
`
`region”). The sense amplifier also “includes” the “isolation regions 3 for isolating
`
`the active region 2 from other active regions for other devices.” Id., 5:9-18.
`
`5.
`
`Rashed similarly describes a multi-transistor “device [that] includes a
`
`continuous active region.” Ex. 1026 at Abstract, 2:55-56. The active region is
`
`“defined … by one or more isolation structures.” Id. at 1:51-55. Rashed does not
`
`refer to any transistor as including an active region.
`
`6.
`
`Dr. Shanfield’s assertion that Agata and Rashed illustrate that “more
`
`1
`
`
`
`than one transistor can exist in an active region” (Ex. 1027, ¶¶ 17-18) is misleading
`
`and irrelevant to the issue in this proceeding, which is whether either MISFET in
`
`the multi-transistor device of modified Igarashi Fig. 12 in the Petition “includes”
`
`the active region the Petition alleges is formed by isolation in that device. That
`
`multiple transistors exist in an active region of a multi-transistor device does not
`
`establish that an individual MISFET in that device somehow “includes” the
`
`device’s active region. For the reasons stated in Ex. 2007, VIII.C.2. in connection
`
`with Igarashi’s multi-transistor device, no transistor in the multi-transistor devices
`
`of Agata and Rashed “includes” an active region because no such transistor
`
`“includes” a region bounded by isolation.
`
`7.
`
`Thus, neither Agata nor Rashed refutes my opinion that the claim
`
`language “MISFET includes: an active region” requires that an entire region
`
`bounded and defined by isolation be part of (i.e., included in) the MISFET, i.e.,
`
`there is only one MISFET in an active region that the “MISFET includes.” See id.,
`
`VIII.C.2.c-d. Petitioner has not cited a single document that describes a MISFET
`
`as “including” an active region where that active region is shared with another
`
`transistor, or that refutes my testimony that a POSA would have understood that an
`
`active region a MISFET “includes” is dedicated to that MISFET. Id.
`
`8.
`
`To the extent Agata and Rashed are relevant at all, they support my
`
`opinion that a MISFET only “includes” an active region if the active region is
`
`2
`
`
`
`dedicated to the MISFET. Agata, Rashed, and the ’501 patent refer to a structure
`
`(respectively, “sense amplifier”, “device,” and “MISFET”) that “includes” an
`
`active region where the active region is dedicated to the structure that “includes” it.
`
`9.
`
`Dr. Shanfield’s assertion that “all functional MOSFET transistors
`
`have an active region” is unsupported and wrong—an area not defined by isolation
`
`is not an active region. See Ex. 2007 at VII.A, VIII.C.1.b. All transistors must
`
`have a region in the substrate where they are formed, but as the ’501 patent makes
`
`clear this is a “formation region.” Ex. 1001 at 3:20-28, Fig. 1.
`
`10.
`
`Dr. Shanfield mischaracterizes my deposition testimony which
`
`addressed the term “comprise” and not “includes.” Ex. 1027, ¶31; Ex. 1024 at
`
`94:13-95:7. The open ended “comprising” transition in claim 1 does not eliminate
`
`the requirement that the “MISFET includes: an active region,” which the grounds
`
`do not meet. See Ex. 2007 at VIII.C.
`
`11. Dr. Shanfield’s suggestion that the ’501 patent does not show 1-to-1
`
`correspondence between the active regions and MISFETs because the figures are
`
`cross sections (Ex. 1027, ¶¶ 14-15) is wrong with respect to Fig. 9A, and refuted
`
`by the ’501 specification. ’501 patent at 3:24-28 (each MISFET “formation region
`
`… includes the active region”). Fig. 9A is a “plane view of an MISFET” (3:8-10,
`
`14:42-45) and illustrates isolation region 2 bounding the active region in which the
`
`transistor is formed, i.e., defining the boundary of the active region on all sides.
`
`3
`
`
`
`and the like
`false statements
`and have been warned that willful
`I understand
`
`that
`declare
`§ l 00 l ). I
`or both (18 U.S.C.
`by fine or imprisonment,
`are punishable
`
`all statements
`are true and that all statements
`of my own knowledge
`made herein
`
`made on information
`that these
`and further,
`to be true,
`are believed
`and belief
`
`and the like
`false statements
`that willful
`were made with the knowledge
`statements
`
`so made are punishable
`or both, under§ I 00 I of title
`18
`by fine or imprisonment,
`
`Code. l declare
`States
`of the United
`that the foregoing
`is
`of pe1jury
`under penalty
`
`true and correct. �!)&&c
`Dated: Jt4-t � r
`2oJK
`
`)
`
`D. Glew
`Dr. Alexander
`
`4
`
`