`
`Taiwan Semiconductor Manufacturing Co.
`v.
`Godo Kaisha IP Bridge 1
`
`Case Nos. IPR2016-01246, -01247
`U.S. Patent No. 7,126,174
`
`Page 1 of 133
`
`TSMC Exhibit 1060
`TSMC v. IP Bridge
`IPR2016-01246
`
`
`
`
`
`BACKGROUND
`
`
`BACKGROUND
`
`2
`
`
`
`
`Page 2 of 133
`
`
`
`References and Instituted Grounds
`REFERENCES*
`Application Date
`
`U.S. Patent No. 5,153,145
`October 17, 1989
`U.S. Patent No. 5,021,353
`February 26, 1990
`U.S. Patent No. 4,506,434
`September 3, 1982
`U.S. Patent No. 5,539,229
`December 28, 1994
`*IP Bridge does not contest status of these references as prior art.
`Lee, Noble
`1–3, 5–7, 9–12, and 14–18
`Lee, Ogawa
`1–3, 5–7, 9–12, and 14–18
`Lowrey, Noble
`1, 4, 5, 8–12, 14, and 16
`Lowrey, Ogawa
`1, 4, 5, 8–12, 14, and 16
`
`GROUNDS
`Prior Art
`
`Publication No.
`
`Claims
`
`3
`
`Inventor
`
`IPR Number
`
`Lee et al.
`Lowrey et al.
`Noble et al.
`Ogawa et al.
`IPR2016-01246
`IPR2016-01246
`IPR2016-01247
`IPR2016-01247
`
`Institution Decision (Paper 8) at 5–6
`
`Page 3 of 133
`
`
`
`
`
`a
`Claim 1 of the ’174 Patent
`Claim 1 of the (174 Patent
`
`
`
`[1.1]
`[1.1]
`[1.2]
`[1.2]
`
`[1.3]
`[1.3]
`[1.4]
`[1.4]
`[1.5]
`[1.5]
`
`[1.6]
`[1.6]
`
`[1.7]
`[1.7]
`[1.8]
`[1.8]
`
`1. A semiconductor device, comprising:
`a trench isolation surrounding an active area of a semi-
`conductor substrate;
`a gate insulating film formed over the active area;
`a gate electrode formed over the gate insulating film;
`first L-shaped sidewalls formed over the side surfaces of
`the gate electrode;
`first silicide layers formed on regions located on the sides
`of the first L-shaped sidewalls within the active area
`an interconnection formed on the trench isolation; and
`second L-shaped sidewalls formed over the side surfaces
`of the interconnection.
`
`4
`\._ £X1001 at 29:39-50.
`EX1001 at 29:39–50.
`
`
`
`
`
`Page 4 of 133
`
`
`
`Claim 1 of the ’174 Patent
`[1.1]
`[1.2]
`
`[1.3]
`[1.4]
`[1.5]
`
`[1.6]
`
`[1.7]
`[1.8]
`
`EX1001 at 29:39–50, FIG. 15(f); 1246 Petition (Paper 2) at 13; 1247 Petition (Paper 2) at 13.
`
`5
`
`’174 Patent, FIG. 15(f)
`
`Page 5 of 133
`
`
`
`Device Isolation
` Isolation
`
`
`“In silicon integrated circuit fabrication it is necessary to isolate
`devices from one another which are built into the same silicon
`matrix. They are subsequently interconnected to create the
`desired circuit configuration.” Schuegraf, EX1009, at 1:11–15.
`“[B]uried insulating layers each . . . surrounds a portion of a
`semiconductor substrate in which elements are fabricated, the
`buried insulating layers functioning to isolate from one another,
`each element fabricated in a chip.” Ogawa, EX1010, at 1:11–15.
`
`
`E.g., EX1009 at 1:11–15; EX1010 at 1:11–15; EX1056 at 79:17–80:5; Response (Paper 14) at 54, 111.
`
`6
`
`Page 6 of 133
`
`
`
`
`
`OBVIOUSNESS
`
`OBVIOUSNESS
`
`7
`
`
`
`
`Page 7 of 133
`
`
`
`
`
`COMBINATIONS
`
`
`COMBINATIONS
`
`8
`
`
`
`
`Page 8 of 133
`
`
`
`Lee Teaches Everything Except Trench Isolation
`
`Lee, FIG. 15 (silicide appended)
`
` Lee uses LOCOS isolation instead of trench isolation
`
`EX1001 at 29:39–50; 1246 Petition (Paper 2) at 13, 17–18; EX1002 at FIG. 15.
`
`9
`
`Page 9 of 133
`
`
`
`Lowrey Teaches Everything Except Trench Isolation
`
`Lowrey, FIG. 12
`
` Lowrey uses LOCOS isolation instead of trench isolation
`
`EX1001 at 29:39–50; 1247 Petition (Paper 2) at 13, 17–18; EX1017 at FIG. 12.
`
`10
`
`Page 10 of 133
`
`
`
`Noble: Trench Isolation in a Semiconductor Device
`
`Noble, FIG. 13
`
`1246 Petition (Paper 2) at 32; 1247 Petition (Paper 2) at 31; EX1015 at FIG. 13.
`
`11
`
`Page 11 of 133
`
`
`
`Ogawa Trench Isolation in a Semiconductor Device
`
`Ogawa, Fig. 5(c)
`
`1246 Petition (Paper 2) at 77; 1247 Petition (Paper 2) at 69; EX1010 at Fig. 5(c).
`
`12
`
`Page 12 of 133
`
`
`
`The Asserted Obviousness Grounds
`
` “[Lee/Lowrey] teaches every limitation of the challenged claims except trench
`isolation. A POSITA would have understood that [Noble’s/Ogawa’s] STI was a
`known substitute for [Lee’s/Lowrey’s] LOCOS isolation. The combined teachings
`discussed in this section refer to the teachings of [Lee/Lowrey], with its LOCOS
`isolation replaced by [Noble’s/Ogawa’s] STI.”
`
`Lee, FIG. 15
`
`Noble, FIG. 13
`
`Lowrey, FIG. 12
`
`Ogawa, FIG. 13
`
`13
`
`1246 Petition (Paper 2) at 21, 32, 33, 70, 77; 1247 Petition (Paper 2) at 21, 31, 32, 62, 69.
`
`Page 13 of 133
`
`
`
`The Resulting Combinations (i.e., “How”)
`
`Lee, FIG. 15
`
`Lee, FIG. 15’
`
`Lowrey, FIG. 12
`
`1246 Petition (Paper 2) at 33; 1247 Petition (Paper 2) at 32; Reply (Paper 21) at 20, 25; EX1057 at 57, 87.
`
`14
`
`Page 14 of 133
`
`
`
`
`
`MOTIVATION
`
`MOTIVATION
`
`15
`
`
`
`
`Page 15 of 133
`
`
`
`The LOCOS Bird’s Beak Was a Well-Known Issue for Scaling
`
`Schuegraf, EX1009, at 1:29–43
`
`Adler, EX1025, at 8
`
`1246 Petition (Paper 2) at 4–7, 10, 21–30, 70–76; 1247 Petition (Paper 2) at 4–7, 10, 21–30, 62–68; Reply
`(Paper 21) at 2, 8, 38; EX1009 at 1:30–2:22; EX1025 at 8–10.
`
`16
`
`Page 16 of 133
`
`
`
`The LOCOS Bird’s Beak Was a Well-Known Issue for Scaling
`
`Reply (Paper 21) at 2
`
`Reply (Paper 21) at 2; EX1056 at 76:5–78:17.
`
`17
`
`Page 17 of 133
`
`
`
`LOCOS Isolation Was Becoming Obsolete
`
` LOCOS was “unallowable in the refinement of a semiconductor device
`after the 0.5 μm generation,” and “IBM corporation has introduced the
`trench isolation structure as a 0.5 μm CMOS process for the mass-
`production of an MPU.” ’174 Patent, EX1001 at 1:29–43.
`
`Adler, EX1025, at 9, (Jan./Mar. 1995)
`
`Adler, EX1025, at 10, (Jan./Mar. 1995)
`
`1994 NTRS, EX1054, at 60 Tbl. 6
`
`Adler, EX1025, at 9 Tbl. 3
`
`1246 Petition (Paper 2) at 4–7, 21–30, 70–76; 1247 Petition (Paper 2) at 4–7, 21–30, 62–68; EX1001 at
`1:29–43; EX1025 at 8–10, Tbl. 3; EX1054 at 60, Tbl. 6.
`
`18
`
`Page 18 of 133
`
`
`
`LOCOS/STI Were Interchangeable Functional Equivalents
`
`Adler, EX1025, at 9
`
`Douglas, EX1011, at 4:10–16
`
`Schuegraf, EX1009, at 2:20–22
`
`Ueda, EX1014, at 22:48–52
`
`Thompson, EX1012, at 3:1–10
`
`E.g., 1246 Petition (Paper 2) at 4–7, 21–30, 70–76; 1247 Petition (Paper 2) at 4–7, 21–30, 62–68; Reply (Paper
`21) at 2–17; EX1009 at 2:20–22; EX1010 at 1:24–66; EX1011 at 4:10–16; EX1012 at 3:8–10; EX1014 at 22:49–52;
`EX1015 at 3:35–37; EX1025 at 8–10.
`
`19
`
`Page 19 of 133
`
`
`
`Patent Owner Admits Using STI was not New to the ’174 Patent
`
`Preliminary Response (Paper 7) at 16.
`
`20
`
`Page 20 of 133
`
`
`
`Board Decision
`
`Institution Decision (Paper 8) at 12, 13 n.5.
`
`21
`
`Page 21 of 133
`
`
`
`POSITA’S REASONABLE
`
`EXPECTATION OF SUCCESS
`
`22
`
`Page 22 of 133
`
`
`
`Petition Shows a Reasonable Expectation of Success
`
`
`
`1246 Petition at 25
`
`EX1004 at ¶ 82
`E.g., 1246 Petition (Paper 2) at 25; 1247 Petition (Paper 2) at 25; EX1004 at ¶¶ 82, 198; EX1024 at ¶¶ 93,
`173; EX1014 at 13:14–63.
`
`23
`
`Page 23 of 133
`
`
`
`A POSITA Knew How to Make STI
`
`IPB’s declarant admitted, “The person of ordinary skill in
`the art would know that there are multiple ways how to
`form a LOCOS isolation. The person of ordinary skill in
`the art would also know that there are multiple [ways] to
`form a trench isolation.” EX1056 at 145:11–15; see also
`EX1059 at 2.
`Thompson shows a POSITA knows how to form STI, and
`how it is an alternative to LOCOS isolation. EX1012 at
`3:1–10 (discussing FIG. 1).
`“The isolation trenches may be formed using well-known
`technology. Other isolation technologies such as local
`oxidation of silicon (LOCOS) may be used instead of
`trenches.” EX1012 at 3:8–10.
` EX1012 at 3:1–10; EX1056 at 145:11–15; EX1059 at 2.
`
`24
`
`Page 24 of 133
`
`
`
`A POSITA Knew STI Could Substitute for LOCOS Isolation in 1996
`
`In early 1995, IBM announced it had been replacing LOCOS
`isolation with STI in commercial devices for several years.
`Numerous references mention the substitutability of STI for
`LOCOS. See ,e.g., EX1057 at 33–34.
`“Although the isolation is composed of the LOCOS film in the
`above embodiments, the present invention is not limited thereto.
`The present invention is also applicable to an isolation of trench
`structure or the like.” Ueda, EX1014 at 22:49–52.
`“Shallow Trench Isolation (STI) is used primarily for isolating
`devices of the same type and is often considered an alternative
`to LOCOS isolation.” Schuegraf, EX1009 at 2:20–22.
`
`1246 Petition (Paper 2) at 4–7, 21–30, 70–76; 1247 Petition (Paper 2) at 4–7, 21–30, 62–68; Reply (Paper 21) at 2–
`17; EX1001 at 1:29–43; EX1009 at 2:20–22, EX1014 at 22:49–52; EX1057 at 33–34.
`
`25
`
`Page 25 of 133
`
`
`
`A POSITA Knew STI Could Substitute for LOCOS Isolation in 1996
`
`“Field oxide 12, preferably silicon dioxide (SiO2), is grown or
`deposited in selected portions of the surface of the substrate
`10 for isolation of active regions from one another according
`to the well known local oxidation (LOCOS) isolation technique;
`of course, other isolation techniques such as trench isolation
`may alternatively be used.” Douglas, EX1011 at 4:10–16.
`“To overcome the foregoing drawbacks [with LOCOS], a
`method wherein each element is isolated from one another by
`buried insulating layers which are grown to fill grooves
`produced along the surface of a silicon (Si) substrate to
`surround each element, has been developed and is presently
`being used.” Ogawa, EX1010 at 1:24–66.
`
`26
`
`1246 Petition (Paper 2) at 4–7, 10, 21–30, 70–76 ; 1247 Petition (Paper 2) at 4–7, 10, 21–30, 62–68; EX1011 at
`4:10–16; EX1010 at 1:24–66.
`
`Page 26 of 133
`
`
`
`A POSITA Knew STI Could Substitute for LOCOS Isolation in 1996
`
`“This technology can be scaled to the 1 Gbit DRAM generation
`with minor modifications, such as replacing LOCOS with
`trench isolation . . . .” Kang, et al., EX1053 at 2, Fig. 1.
`“As the miniaturizing and integration densities of
`semiconductor integrated circuits increase, the conventional
`selectively oxidized film (LOCOS) method used for isolating
`circuit elements has been replaced by the shallow trench
`method.” Sato, EX1034 at 1:60–64.
`“Trench isolation will also be incorporated into more BiCMOS
`structures. Such trench-isolated BiCMOS processes have
`already been reported . . . .” S. Wolf, EX1046 at 30.
`
`Reply (Paper 21) at 2–17; EX1034 at 1:60–64; EX1046 at 30; EX1053 at 2, Fig. 1; EX1057 at 26–50.
`
`27
`
`Page 27 of 133
`
`
`
`A POSITA Knew STI Could Substitute for LOCOS Isolation in 1996
`
` “The purpose of this paper is to present the newly developed field isolation
`technology which overcomes these difficulties [of LOCOS].” Kurosawa,
`EX1052, at 1.
` “LOCOS is not expected to scale significantly beyond 1 μm pitch due to its
`intrinsic limitations such as field oxide thinning, bird’s beak encroachment,
`lack of planarity, and punchthrough. As a result, trench isolation is required
`to meet the demands of ULSI.” Poon, EX1048, at 1.
` “LOCOS-based isolation is used almost exclusively in the fabrication of ICs
`due to its simplicity. However, it is widely recognized that LOCOS-based
`technology cannot be extended to deep submicrometer dimensions because
`of lateral oxide encroachment and field oxide thinning in narrow isolation
`regions. Trench isolation has been proposed as a potential LOCOS
`replacement in scaled and high-performance ULSI.” Fry, EX1047, at 8–10.
`
`Reply (Paper 21) at 2–17; EX1047 at 8–10; EX1048 at 1; EX1052 at 1; EX1057 at 26–50.
`
`28
`
`Page 28 of 133
`
`
`
`A POSITA Knew STI Could Substitute for LOCOS Isolation in 1996
`
`Adler (IBM), EX1025 at 9 (citing Davari, EX1055)
`
`Davari, EX1055, at FIGS. 2(a), 1(c)
`
`EX1025 at 9, 20; EX1055 at 1–3.
`
`29
`
`Page 29 of 133
`
`
`
`A POSITA Knew How to Integrate the STI into Lee’s Device
`
`Trivial Substitute Process Sequence
`
`Reply (Paper 21) at 18–20; EX1057 at 50–57.
`
`30
`
`Page 30 of 133
`
`
`
`The Result of the Combination of Lee and Noble/Ogawa
`
`EX1057 at 57
`
`Reply (Paper 21) at 20; EX1057 at 57.
`
`31
`
`Page 31 of 133
`
`
`
`Integrating the STI Process into Lowrey’s Device
`
` Non-Planar Process
`
` Planar Processes
`
`Reply (Paper 21) at 21
`
`Reply (Paper 21) at 23
`*There should be a “slight jog” under the trenches on the left above, such that the left-hand side of the STI structure is
`slightly lower than the right-hand side of the STI structure. Reply (Paper 21) at 24 n.12; EX2078 at 194:12–17.
`
`Reply (Paper 21) at 20–25 & n.12; EX1057 at 57–87 & n.11; EX2078 at 194:12–17.
`
`32
`
`Page 32 of 133
`
`
`
`The Result of the Combination of Lee and Noble/Ogawa
`
`1247 Petition(Paper 2) at 18
`
`Reply (Paper 21) at 25
`*The non-planar example would have a “slight jog” under the trench, such that the left-hand side of the STI structure is slightly
`lower than the right-hand side of the STI structure. Reply (Paper 21) at 24 n.12; EX2078 at 194:12–17
`
`1247 Petition (Paper 2) at 18; Reply (Paper 21) at 22–25 & n.12; EX1057 at 87; EX2078 at 194:15–17.
`
`33
`
`Page 33 of 133
`
`
`
`Similarity of the LOCOS and STI Processes
`
`EX2078 at 218:5–22, 243:5–19.
`
`34
`
`Page 34 of 133
`
`
`
`The LOCOS Process
`
`Ogawa, Fig. 1
` The LOCOS Process (see, e.g., EX1005 at 10; EX1008 at 3, Fig. 2;
`
`EX1008 at Fig. 2a
`
`EX1017 at 7:57–8:16, FIGS. 1–2):
` Deposit optional pad oxide
` Deposit silicon nitride
` Define isolation region (by etching nitride only)
` Form isolation (by thermal oxidation)
` Remove nitride
` Remove pad oxide
`
`
`EX1005 at 10; EX1008 at 3, Fig. 2; EX1010 at 1:33–42; Fig. 1; EX1017 at 7:57–8:16, FIGS. 1–2; EX2078 at
`218:5–22.
`
`35
`
`Page 35 of 133
`
`
`
`The STI Process—Admitted Prior Art
`
` The STI Process (see, e.g., EX1001 at 4:16–41 (admitted prior art)):
` Deposit optional pad oxide
` Deposit polish/etch-stop (e.g., silicon nitride) (IBM calls it a planarization block mask (PBM))
` Define isolation region (by etching polish/etch-stop, pad oxide, and substrate)
` Form isolation (by CVD oxidation + planarization)
` Remove polish/etch-stop (e.g., nitride)
` Remove pad oxide
`
`
`EX1001 at 4:16–41, 21:33–23:6, 26:36–45; EX1057 at 26–50; EX2078 at 218:5–22, 243:5–19.
`
`’174 Patent, FIGS. 13(a)–13(b), 20(a)–20(b)
`
`36
`
`Page 36 of 133
`
`
`
`Similarity of the LOCOS and STI Processes
`
`The LOCOS Process
`
` Deposit optional pad oxide
` Deposit silicon nitride
` Define isolation region
` (by etching nitride)
` Form isolation
` (by thermal oxidation)
` Remove nitride
` Remove pad oxide
`
`The STI Process
`
` Deposit optional pad oxide
` Deposit stopper (e.g., silicon nitride)
` Define isolation region
` (by etching stopper, pad, and substrate)
` Form isolation
` (by CVD oxidation + planarization)
` Remove stopper (e.g., nitride)
` Remove pad oxide
`
`EX1001 at 4:16–41, 21:33–23:6, 26:36–45; EX1005 at 10; EX1008 at 3, Fig. 2; EX1010 at 1:33–42; Fig. 1;
`EX1017 at 7:57–8:16, FIGS. 1–2; EX1057 at 26–50; EX2078 at 218:5–22, 243:5–19.
`
`37
`
`Page 37 of 133
`
`
`
`Board Decision
`
`Institution Decision (Paper 8) at 14.
`
`38
`
`Page 38 of 133
`
`
`
`OTHER CLAIM LIMITATIONS
`
`39
`
`Page 39 of 133
`
`
`
`PATENT OWNER MISCONSTRUES
`LOWREY’S L-SHAPED SIDEWALLS
`
`40
`
`Page 40 of 133
`
`
`
`Meaning of “L-Shaped Sidewalls”
`
`TSMC and the Board felt the plain and ordinary meaning sufficed
`
`Petitions (Paper 2) at 16
`
`IP Bridge advocates the construction the district court adopted,
`but that appears to be the plain and ordinary meaning
`
`Institution Decision (Paper 8) at 7
`
`Response (Paper 14) at 42
`
`Petitions (Paper 2) at 16; Institution Decision (Paper 8) at 7; Response (Paper 14) at 42.
`
`41
`
`Page 41 of 133
`
`
`
`IP Bridge Implicitly Adds Another Requirement
`
`IP Bridge suggests “L-shaped” sidewalls must be
`“distinguishable” by a specific experimental technique (SEM).
`
`Response (Paper 14) at 104
`
`Response (Paper 14) at 42, 102–04.
`
`Response (Paper 14) at 103
`
`42
`
`Page 42 of 133
`
`
`
`Lowrey Discloses L-Shaped Sidewalls
`
`1247 Petition (Paper 2) at 59
`
`1247 Petition (Paper 2) at 37
`
`1247 Petition (Paper 2) at 59
`1247 Petition (Paper 2) at 35–38, 42–44, 58–59; Reply (Paper 21) at 40–42; EX1017 at 8:58–9:12, FIGS. 6–8, 12;
`EX1057 at 87–92.
`
`43
`
`Reply (Paper 21) at 40
`
`Page 43 of 133
`
`
`
`Lowrey’s L-Shaped Sidewalls Are Separate
`
`Reply (Paper 21) at 40
`
`Reply (Paper 21) at 41
`
`Reply (Paper 21)at 42
`1247 Petition (Paper 2) at 35–38, 42–44, 58–59; Reply (Paper 21) at 40–42; EX1051 at 3–4, Figs. 1, 5; EX1057 at
`87–92.
`
`44
`
`Page 44 of 133
`
`
`
`PATENT OWNER’S CRITICISM OF
`LEE IGNORES LEE’S DISCLOSURE
`
`45
`
`Page 45 of 133
`
`
`
`Lee Discloses Silicide Regions and LDD Doping
`Self-Aligned Silicide
`LDD Doping
`
`Lee, FIG. 5
`
`Lee, FIG. 9
`
`Reference Figure (see text)
`
`EX1002 at 7:13–35
`
`Lee, FIG. 6
`Lee, FIG. 14
`1246 Petition (Paper 2) at 37–40; Reply (Paper 21) at 20, 42–44; EX1002 at 3:49–4:3, 4:53–6:30, 7:13–35, FIGS.
`5–6, 9, 15; EX1057 at 56–57, 92–94.
`
`46
`
`Page 46 of 133
`
`
`
`Lee Discloses LDD Doping Using L-Shaped Sidewalls
`
`Lee, FIG. 5
`
`EX1057 at 56
`
`EX1057 at 56
`Lee, FIG. 6
`1246 Petition (Paper 2) at 37; Reply (Paper 21) at 20; EX1002 at 3:49–4:3, 4:53–6:30, 7:16–22, FIGS. 5–6, 15;
`EX1057 at 56.
`
`47
`
`Page 47 of 133
`
`
`
`Lee Discloses Self-Aligned Silicide Regions
`
`Lee, FIG. 9
`
`EX1057 at 56
`
`Lee, FIG. 15
`
`EX1057 at 57
`
`1246 Petition (Paper 2) at 38–40; Reply (Paper 21) at 20, 42–44; EX1002 at 4:53–6:30, 7:22–28, FIGS. 9, 15;
`EX1057 at 56–57, 92–94.
`
`48
`
`Page 48 of 133
`
`
`
`RAISED SHALLOW TRENCH
`ISOLATION (STI)
`
`49
`
`Page 49 of 133
`
`
`
`Noble Discloses Raised STI (Claims 9 and 10)
`
`1246 Petition (Paper 2) at 56
`
`Noble at 4:14–19
`
`Noble at 4:39–50
`1246 Petition (Paper 2) at 52–57; Response (Paper 14) at 86–88; Reply (Paper 21) at 49–50; EX1015 at 3:49–54,
`4:14–19, 4:39–50, 5:49–52, 6:13–24, FIGS. 4–5, 9–13.
`
`50
`
`Page 50 of 133
`
`
`
`
`
`RESPONSE TO PATENT OWNER
`RESPONSE TO PATENT OWNER
`ARGUMENTS
`ARGUMENTS
`
`
`51
`
`
`
`
`Page 51 of 133
`
`
`
`PATENT OWNER HAS NOT
`CHANGED POSITIONS
`
`52
`
`Page 52 of 133
`
`
`
`Petitions
`
`1246 Petition (Paper 2) at 21, 62.
`
`53
`
`Page 53 of 133
`
`
`
`Petitions
`
`1247 Petition (Paper 2) at 21, 62.
`
`54
`
`Page 54 of 133
`
`
`
`Institution Decision
`
`Institution Decision (Paper 8) at 12.
`
`55
`
`Page 55 of 133
`
`
`
`Institution Decision
`
`Institution Decision (Paper 8) at 20
`
`56
`
`Page 56 of 133
`
`
`
`Institution Decision
`
`Institution Decision (Paper 8) at 23, 24, 27.
`
`57
`
`Page 57 of 133
`
`
`
`
`
`Reply
`
`In its Petitions, TSMC explained how Lee and Lowrey teach every limitation
`
`of the challenged claims, and why a POSITA would have wanted to substitute
`
`Noble’s and Ogawa’s functionally equivalent shallow-trenchisolation (““STT’)
`
`
`
`
`
`structures for Lee’s and Lowrey’s LOCOS isolation. One reason was such
`
`substitutions allow increased device density. TSMC even showedseveral
`
`examples how a POSITA would have known to make the STI structures in Noble
`
`and Ogawa.
`
`IPB doesnot challenge this evidence.
`
`58
`\ Reply (Paper 21) at 1. TS
`
`Reply (Paper 21) at 1.
`
`
`
`Page 58 of 133
`
`
`
`PATENT OWNER ATTACKS
`COMBINATIONS PETITIONER NEVER
`ASSERTED
`
`59
`
`Page 59 of 133
`
`
`
`The Asserted Combinations Do not Use Noble or Ogawa Gates
`
`60
`
`Patent Owner did not argue a distinction between “embedded
`STI” and “non-embedded STI” until the Sur-Reply.
`No reference refers to an embedded gate or a non-
`embedded gate.
`Petitioner never suggested using the gate stack of Noble or
`Ogawa in any combination. 1246 Petition (Paper 2) at 21, 70;
`1247 Petition (Paper 2) at 21, 62; Institution Decision
`(Paper 8) at 12, 21, 24, 27.
`Noble and Ogawa use a standard process for forming STI.
`Reply (Paper 21) at 15–17.
`How the gate stack is made does not affect the STI. EX1004 at
`¶¶ 82, 198; EX1024 at ¶¶ 93, 173.
`
`1246 Petition (Paper 2) at 21, 70; 1247 Petition (Paper 2) at 21, 62; Institution Decision (Paper 8) at 12,
`21, 24, 27; Reply (Paper 21) at 15–17; Sur-Reply (Paper 37) at 3–4, 6, 10–11, 23–24, 26; EX1004 at ¶¶
`82, 198; EX1024 at ¶¶ 93, 173.
`
`Page 60 of 133
`
`
`
`The Asserted Combinations Do not Use Noble or Ogawa Gates
`
`To show likelihood of success, Petitioner referred to
`prior art that did not use the gate as a polish-stop or
`etch-stop.
`Ueda (E.g., EX1014 at 13:14–63)
`Mandelman (E.g., EX1016 at 3:55–65)
`Admitted Prior Art in ’174 Patent (E.g., EX1001 at 4:16–39)
`
`E.g., 1246 Petition (Paper 2) at 5–7, 10–11, 21–26, 31, 70–71; 1247 Petition (Paper 2) at 5–7, 10–11, 21–26, 30–31,
`62–63; EX1001 at 4:16–39; EX1014 at 13:14–63; EX1016 at 3:55–65.
`
`61
`
`Page 61 of 133
`
`
`
`IPB Never addressed TSMC’s Combination; It Attacked Different One
`
` IP Bridge included Noble’s/Ogawa’s gate stack and interconnection,
`with the trench isolation, and only attacked that combination.
`
`Response (Paper 14) at 55
`
`Response (Paper 14) at 65
`
`Response (Paper 14) at 55; 65, 111.
`
`Response (Paper 14) at 111
`
`• But TSMC never proposed this combination. 62
`
`Page 62 of 133
`
`
`
`Petitioner Only Used the Noble STI
`Noble
`
`Portion in Petitioner’s combination
`
`Portion in Patent Owner’s combination
`
`Response (Paper 14) at 55; 65, 111; EX1015 at FIG. 13.
`
`63
`
`Page 63 of 133
`
`
`
`Patent Owner Admits it Uses a Different Combination
`
` IP Bridge seeks to incorporate features not part of the alleged obviousness
`combination (i.e., raised source/drains).
`
`Response (Paper 14) at 87
`
` This is inconsistent with the claim language.
`
`Claim 1
`
`64
`
`1246 Petition (Paper 2) at 52–57; Response (Paper 14) at 86–88; Reply (Paper 21) at 49–50.
`
`Page 64 of 133
`
`
`
`Petitioner Only Used the Ogawa STI
`
`Ogawa
`
`Portion in Petitioner’s combination
`
`Portion in Patent Owner’s combination
`
`Response (Paper 14) at 55; 65, 111; EX1010 at Fig. 5(c).
`
`65
`
`Page 65 of 133
`
`
`
`PATENT OWNER ARGUMENTS
`BASED ON MISREPRESENTATION OF
`PROPOSED COMBINATIONS
`
`66
`
`Page 66 of 133
`
`
`
`Patent Owner Arguments Based on Faulty Combination
`
`Response
`
`Patent Owner Argument
`Lee and Noble/Ogawa Are Not Combinable To Arrive At The Claimed
`Invention
`Lee And Ogawa Are Not Combinable To Arrive At the Claimed
`Invention
`Substituting The Trench Isolation of Noble/Ogawa Into Lee Conflates
`Two Contradictory Designs
`
`Substituting The Trench Isolation Of Noble/Ogawa Into Lowrey
`Conflates Two Contradictory Device Designs
`
`Source
`
`POR at 50, et seq.
`POR at 59, et seq.
`POR at 63, et seq.
`POR at 111, et seq.
`
`67
`
`Page 67 of 133
`
`
`
`NO NEED TO DISCLOSE PROCESS
`
`68
`
`Page 68 of 133
`
`
`
`Process Not Required to Invalidate a Structure Claim
`
`The law treats structure claims (all of those in
`issue) differently from process claims
`IPB cited no law in its papers requiring disclosure
`of the actual process sequence of making the
`combination.
`
`
`
`Reply (Paper 21) at 26–29; Institution Decision (Paper 8) at 17.
`
`69
`
`Page 69 of 133
`
`
`
`Process Not Required to Invalidate a Structure Claim
`
`“The patentability of a product does not depend on its method of
`production.” In re Thorpe, 777 F.2d 695, 697 (Fed. Cir. 1985).
`“The method of manufacture, even when cited as advantageous,
`does not of itself convert product claims into claims limited to a
`particular process.” Vanguard Prods. Corp. v. Parker Hannifin
`Corp., 234 F.3d 1370, 1372-73 (Fed. Cir. 2000).
`“Courts must generally take care to avoid reading process
`limitations into an apparatus claim.” Baldwin Graphics Sys., Inc. v.
`Siebert, Inc., 512 F.3d 1338, 1344 (Fed. Cir 2008).
`“Appellants claim a gene sequence. Accordingly, the obviousness
`inquiry requires this court to review the Board’s decision that the
`claimed sequence, not appellants’ unclaimed cloning technique, is
`obvious in light of the abundant prior art.” In re Kubin, 561 F.3d
`1351, 1356 (Fed. Cir. 2009).
`
`Reply (Paper 21) at 18 n.9, 26–30.
`
`70
`
`Page 70 of 133
`
`
`
`Process Not Required to Invalidate a Structure Claim
`
`IPB cases involve process claims, not structure claims
`IPB cases discussing “how or why” an obviousness
`combination would be made do not require a process
`for making a claimed structure
` Kinetic Concepts—rev’d JMOL because “[t]he record [wa]s
`devoid of any reason someone would combine the[]
`references”
` Innogenetics—aff’d exclusion of expert testimony that had
`only a “stock phrase” concluding the claims were obvious
` Personal Web—“how” described the Board’s failure to
`explain how combination satisfied claims; not a reference to
`failure to show a process for making a claimed structure
`
`
`
`Reply (Paper 21) at 26–29.
`
`71
`
`Page 71 of 133
`
`
`
`Board Decision
`
`Institution Decision (Paper 8) at 16–17.
`
`72
`
`Page 72 of 133
`
`
`
`Board Decision
`
`Institution Decision (Paper 8) at 17 n.6.
`
`73
`
`Page 73 of 133
`
`
`
`PROCESS DETAILS UNNECESSARY
`BECAUSE A POSITA KNEW HOW TO MAKE
`STI
`
`74
`
`Page 74 of 133
`
`
`
`Forming an STI is not Complex
`
`’174 Patent at 2:47–60
`
`’174 Patent FIG. 18(a)
`
`EX1001 at 2:47–60, FIG. 18(a).
`
`75
`
`Page 75 of 133
`
`
`
`Forming an STI is not Complex
`
`Ueda, EX1014, at 22:48–52
`
`Kang, EX1053, at 2
`
`Kang, EX1053, at Fig. 1
`
`EX1014 at 22:48–52; EX1053 at 2; EX1057 at 33–34.
`
`76
`
`Page 76 of 133
`
`
`
`Forming an STI is not Complex
`
`Konaka, EX1032, at 3:65–4:13, FIGS. 5, 6
`
` Filed June 17, 1985
` Claims priority to an application filed October 27,
`1981, which contains FIGS. 5 & 6.
`
`Reply (Paper 21) at 3–5, 17, 38; EX1032 at 3:65–4:13, FIGS. 5,
`6; EX1033 at 3; EX1057 at 27–29, 35–36, 41, 45, 47–48, 52.
`
`77
`
`Page 77 of 133
`
`
`
`Forming an STI is not Complex
`
`Douglas, EX1011, at 4:6–16
`
`Douglas, EX1011, FIG. 1
`
`EX1011 at 4:6–16; EX1057 at 33–34.
`
`78
`
`Page 78 of 133
`
`
`
`Forming an STI is not Complex
`
`Thompson, EX1012, at 3:1–11
`
`Thompson, EX1012, FIG. 1
`
`EX1012 at 3:1–10; EX1057 at 33–34.
`
`79
`
`Page 79 of 133
`
`
`
`A POSITA KNEW HOW TO FORM STI
`WITHOUT THE TRANSISTOR GATE
`
`80
`
`Page 80 of 133
`
`
`
`The STI Structure Is Independent of the Gate Process
`
`Noble, FIG. 9
`
`Ogawa, Fig. 4(c)
`
`Mandelman, FIG. 1d
`
`Sato, Fig. 1D
`
`Kodera, Fig. 29C
`
`Pierce, Fig. 1
`
`Deleonibus, Fig. 1
`
`Davari, Fig. 1(c)
`
`Dash, Fig. 6
`
`Reply (Paper 21) at 2–17; EX1010 at 5:57–6:59, Fig. 4(c); EX1015 at 3:64–4:19, 4:40–49, 5:49–57, FIG. 9;
`EX1016 at 3:27–4:22, Fig. 1d; EX1034 at 4:30–5:49, Fig. 1D; EX1035 at 26:62–28:33, FIG. 29C; EX1042 at 4–5,
`Fig. 1; EX1043 at 2–3, Fig. 1; EX1055 at 1–3, Fig. 1(c); EX1057 at 26–50; EX1058 at 2:48–4:28, Fig. 6.
`
`81
`
`Page 81 of 133
`
`
`
`Making STI Before the Gate Stack Was Routine
`
`Dash, Fig. 6
`
`Noble at 3:35–37
`
` Noble incorporates Dash by reference.
`
`Dash at 4:17–28
`
`Reply (Paper 21) at 3–17; EX1015 at 3:35–37; EX1057 at 34–44; EX1058 at 4:17–28, Fig. 6.
`
`82
`
`Page 82 of 133
`
`
`
`Making STI Before the Gate Stack Was Routine
`
`Mandelman at 3:55–65, 5:40–47, FIG. 1d
`
`Reply (Paper 21) at 3–17; EX1016 at 3:55–65, 5:40–47, FIGS. 1d, 5; EX1057 at 34–44.
`
`83
`
`Page 83 of 133
`
`
`
`Making STI Before the Gate Stack Was Routine
`
`Sato, Fig. 1C
`
`Sato, Fig. 1D
`
`Reply (Paper 21) at 5
`
`Sato at 5:36–49
`Reply (Paper 21) at 3–17; EX1034 at 5:36–49, Figs. 1C, 1D; EX1057 at 34–44.
`
`84
`
`Page 84 of 133
`
`
`
`Making STI Before the Gate Stack Was Routine
`
`Pierce, at 2–3, Fig. 1
`
`Reply (Paper 21) at 3–17; EX1043 at 2–3, Fig. 1; EX1057 at 34–44.
`
`85
`
`Page 85 of 133
`
`
`
`Making STI Before the Gate Stack Was Routine
`
`Deleonibus, at 4–5, Figure 1
`
`Reply (Paper 21) at 3–17; EX1042 at 4–5, Figure 1; EX1057 at 34–44.
`
`86
`
`Page 86 of 133
`
`
`
`ADDITIONAL REFERENCES CAN
`SHOW REASONABLE EXPECTATION
`OF SUCCESS
`
`87
`
`Page 87 of 133
`
`
`
`No Need to Limit Analysis to References
`
`The Board recognized other references besides those
`used for the obviousness combination could be used to
`show reasonable expectation of success.
` Genzyme Therapeutic Prods. Ltd. v. Biomarin Pharm. Inc.,
`825 F.3d 1360, 1367–68 (Fed. Cir. 2016) (rejecting an
`argument that references not among the combinations of
`references on which the Board granted review could not be
`used to show a reasonable expectation of success).
`“Art can legitimately serve to document the knowledge that
`skilled artisans would bring to bear in reading the prior art
`identified as producing obviousness.” Ariosa Diagnostics v.
`Verinata Health, Inc., 805 F.3d 1359, 1365 (Fed. Cir. 2015).
`
`
`Institution Decision (Paper 8) at 13 n.5.
`
`88
`
`Page 88 of 133
`
`
`
`No Need to Limit Analysis to References
`
`As KSR established, the knowledge of such an artisan is part
`of the store of public knowledge that must be consulted when
`considering whether a claimed invention would have been
`
`“By narrowly focusing on the four prior-art references cited by the
`Examiner and ignoring the additional record evidence Randall
`cited to demonstrate the knowledge and perspective of one of
`ordinary skill in the art, the Board failed to account for critical
`background information that could easily explain why an
`ordinarily skilled artisan would have been motivated to combine
`or modify the cited references to arrive at the claimed inventions.
`obvious.” Randall Mfg. v. Rea, 733 F.3d 1355, 1362–63 (Fed. Cir.
`2013) .
`“The record shows the well-known and reliable nature of the
`cloning and sequencing techniques in the prior art, not to mention
`the readily knowable and obtainable structure of an identified
`protein. Therefore this court cannot deem irrelevant the ease and
`predictability of cloning the gene that codes for that protein.” In
`re Kubin, 561 F.3d 1351, 1360 (Fed. Cir. 2009).
`
`
`Institution Decision (Paper 8) at 13 n.5.
`
`89
`
`Page 89 of 133
`
`
`
`PATENT OWNER IGNORES THE LAW
`IN LIMITING THE PRIOR ART TO A
`SPECIFIC EMBODIMENT
`
`90
`
`Page 90 of 133
`
`
`
`A POSITA Must Consider a Reference for All it Teaches
`
` “A reference must be considered for everything that it teaches, not
`simply the described invention or a preferred embodiment.” In re
`Applied Materials, Inc., 692 F.3d 1289, 1298 (Fed. Cir. 2012).
` “A reference must be considered for everything it teaches by way of
`technology and is not limited to the particular invention it is describing
`and attempting to protect. On the issue of obviousness, the combined
`teachings of the prior art as a whole must be considered.” Belden Inc. v.
`Berk-Tek LLC, 805 F.3d 1064, 1076 (Fed. Cir. 2015).
` “It is well-established that a determination of obviousness based on
`teachings from multiple references does not require an actual, physical
`substitution of elements. . . . Rather, the test for obviousness is what the
`combined teachings of the references would have suggested to those
`having ordinary skill in the art.” In re Mouttet, 686 F.3d 1322, 1332–33
`(Fed. Cir. 2012).
`
`91
`
`Reply (Paper 21) at 30–32.
`
`Page 91 of 133
`
`
`
`A POSITA Has Ordinary Creativity
`
`“[O]ne of ordinary skill is also one of ‘ordinary creativity’ that knows
`how to combine familiar prior art elements to achieve the same
`functions.” Tyco Healthcare Grp. LP v. Ethicon Endo-Surgery, Inc.,
`774 F.3d 968, 978 (Fed. Cir. 2014).
`“[W]e do not ignore the modifications that one skilled in the art
`would make to a device borrowed from the prior art. One skilled in
`the art would size the components from Teague appropriately for
`Icon’s application, therefore producing an embodiment meeting
`Icon’s claims.” In re ICON Health & Fitness, Inc., 496 F.3d 1374,
`1382 (Fed. Cir. 2007) (citation omitted).
`
`Reply (Paper 21) at 6.
`
`92
`
`Page 92 of 133
`
`
`
`
`
`RESPONSE TO SUR-REPLY
`RESPONSE TO SUR-REPLY
`
`
`93
`
`
`
`
`Page 93 of 133
`
`
`
`PETITIONER’S REPLY DID NOT
`RAISE NEW ISSUES
`
`94
`
`Page 94 of 133
`
`
`
`TSMC’s EX1025 through EX1059 Respond to IPB Arguments
`
`Corroborating Evidence to
`Show Substitutability of STI
`for LOCOS Isolation
`
`Evidence Showing That
`Making STI Without Using
`the Gate Stack Was Known
`
`Other Rebuttal Evidence(e.g.,
`L-Shaped Sidewalls, SEM)
`
`• EX1025 (Adler)
`• EX1046 (Wof & Tauber, vol. 2)
`• EX1047 (Fry)
`• EX1048 (Poon)
`• EX1052 (Kurosawa)
`• EX1053 (Kang)
`• EX1054 (NTRS)
`
`• EX1036 (Chen)
`• EX1026/EX1027 (Sumi)*
`• EX1038 (Ma)
`• EX1028/EX1029 (Horiguchi)*
`• EX1039 (Manukonda)
`• EX1030/EX1031 (Ueda JP)†
`• EX1040 (Hiroki)
`• EX1032/EX1033 (Konaka and
`• EX1041 (Kusunoki)
`JP counterpart)
`• EX1044 (Chang & Sze)
`• EX1034 (Sato)
`• EX1045 (Wolf & Tauber, vol. 1)
`• EX1035 (Kodera)
`• EX1049 (Clement)
`• EX1037 (Gasner)
`• EX1050 (Pantel)
`• EX1042 (Deleonibus)
`• EX1051 (Servanton)
`• EX1043 (Pierce)
`• EX1055 (Davari)
`• EX1058 (Dash)‡
`* Cited on the face of the ’174 patent.
`† This is the July 21, 1995, publication of JPH 05-284820, a priority document for Ueda (EX1014), which
`confirms the portions of Ueda cited in the Petition had been published earlier.
`‡ Incorporated by reference by Noble for teaching “STI and processes for forming STI.” EX1015 at 3:35–37.
`
`Petitioner’s Opposition to Motion to Exclude (Paper 40) at 9.
`
`
`95
`
`Page 95 of 133
`
`
`
`TSMC’s EX1025 through EX1059 Respond to IPB Arguments
`
`Response (Paper 14) at 64
`
`Reply (Paper 21) at 18 n.9
`
`Response (Paper 14) at 10, 64–65, 114–16; Reply (Paper 21) at 3–18 & nn.8, 9.
`
`
`96
`
`Page 96 of 133
`
`
`
`TSMC’s EX1025 through EX1059 Respond to