`_______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________
`
`HAMAMATSU CORPORATION,
`Petitioner,
`
`v.
`
`SIONYX, LLC,
`Patent Owner.
`_______________
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`_______________
`
`Record of Oral Hearing
`Held: October 4, 2017
`_______________
`
`
`Before GEORGIANNA W. BRADEN, MATTHEW R. CLEMENTS,
`and MONICA S. ULLAGADDI Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`APPEARANCES:
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`ON BEHALF OF PETITIONER:
`
`STEPHEN MURRAY, ESQUIRE
`
`John Simmons, Esquire
`
`Keith Jones, Esquire
`
`Panitch Schwarze
`
`One Commerce Square
`
`2005 Market Street, Suite 2200
`
`Philadelphia, PA 19103
`
`smurray@panitchlaw.com
`
`ON BEHALF OF PATENT OWNER:
`
`WILLIAM D. BELANGER, ESQUIRE
`
`GWENDOLYN TAWRESEY, ESQUIRE
`
`Pepper Hamilton, LLP
`
`125 High Street
`
`19th Floor High Street Tower
`
`Boston, MA 02110
`
`belangerw@pepperlaw.com
`
`The above-entitled matter came on for hearing Wednesday,
`October 4, 2017, commencing at the U.S. Patent and Trademark Office,
`600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`
`2
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`
`P R O C E E D I N G S
`JUDGE ULLAGADDI: Good morning. This will be in the matter
`
`of Hamamatsu Corporation, Petitioner, versus SiOnyx, LLC, Patent
`Owner, Review Matter No. 2016-01910. Your panel for the hearing
`today includes myself, Judge Ullagaddi, Judge Braden joining us from
`Texas, and Judge Clements joining us from California.
`
`Good afternoon, Judges Braden and Clements. Are you able to see
`and hear us clearly?
`
`JUDGE BRADEN: Yes. Thank you.
`
`JUDGE CLEMENTS: We are.
`
`JUDGE ULLAGADDI: Thank you.
`
`I'd like to start today by getting appearances of Counsel. Who is
`appearing on behalf of Petitioner?
`
`MR. MURRAY: Good afternoon, Your Honors. My name is
`Steve Murray, appearing on behalf of Petitioner. I have with me lead
`counsel, Mr. John Simmons, and backup counsel, Keith Jones.
`
`JUDGE BRADEN: I'm sorry. But we're unable to hear you unless
`you speak at the podium into the microphone, please. MR.
`MURRAY: Good afternoon, Your Honors. I apologize for that. I am
`Stephen Murray on behalf of the Petitioners. With me is lead counsel,
`Mr. John Simmons, and also back-up counsel, Keith Jones.
`
`MR. SIMMONS: Good afternoon, Your Honors.
`
`JUDGE ULLAGADDI: Thank you.
`
`And who do we have on behalf of Patent Owner?
`
`MR. BELANGER: Good afternoon. It's William Belanger with
`Pepper Hamilton on behalf of the Patent Owner. And with me is
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`3
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`Gwendolyn Tawresey, also with Pepper Hamilton.
`
`JUDGE ULLAGADDI: Thank you, Mr. Belanger.
`
`Thank you all for joining us. I've got a few administrative details
`I'd like to go over today before we get started. Each party will have 45
`minutes to argue their case. We're first going to hear from Petitioner.
`Petitioner, you will present your arguments in chief. Patent Owner, you
`will then be allowed to present your arguments.
`
`Petitioner, would you like to reserve any time for rebuttal today?
`
`MR. MURRAY: Yes, Your Honor, if I may reserve 25 minutes
`for rebuttal.
`
`JUDGE ULLAGADDI: Twenty-five minutes for rebuttal. Thank
`you for that.
`
`One thing I would also like to mention is that when you are
`working with your demonstrative slide sets, if you could remember to
`please identify the slide number that you're referring to. Judges Braden
`and Clements are not able to see the screen from where they are, so if you
`could make sure to refer to the demonstrative slide that you're using, we
`would appreciate that.
`
`JUDGE CLEMENTS: On the topic of demonstratives, will the
`Petitioner be using any?
`
`MR. MURRAY: Petitioner did not plan to use any demonstratives.
`To the extent we need to show anything, we have copies of the exhibits
`in the record that we can refer to.
`
`JUDGE CLEMENTS: Okay. Very good. Thanks.
`
`JUDGE ULLAGADDI: Mr. Murray, you will have 20 minutes for
`your primary case. And you may begin whenever you are ready.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`4
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`MR. MURRAY: So good afternoon again, Your Honors. May it
`
`please the Board, the '591 patent, which we're talking about here today, is
`nothing more than the application of a textured surface to a standard
`image sensor such as the CMOS sensor.
`
`The alleged invention changes nothing about the general image
`sensor structure or its principal method of operation. Incident light is
`stochastically absorbed in the silicon to generate a charge that's let out by
`circuitry for creating an image.
`
`All the '591 patent adds is the texture reading for the stated
`purpose of improving absorption and electrical response of the image
`sensor. But there's no dispute that surface texturing was previously a
`well-known technique for improving such photodetector properties.
`
`Pages 4 and 5 of the petition lay out a brief history of texturing for
`that purpose, including citations to patents and publications by Webb,
`Uematsu, Van Zegh Broeck, Moloney, and, of course, Mazur, which of
`course forms one of the grounds of rejection. So what you end up with
`here is a standard image sensor that now has improved the texturing
`characteristics resulting from application of a technique that was known
`to improve those characteristics. And the Mabuchi Mazur combination
`embodies this unsurprising result.
`
`Patent Owner acknowledges that Mabuchi teaches all of the
`claimed standard CMOS elements, including the substrate PN junction,
`integrated circuitry, and the electrical transfer.
`
`Patent Owner also has not disputed that Mazur teaches laser
`texture and of a light incident surface than silicon. So all of the elements,
`for example, for Claim 1 are present here. And Mazur itself provides an
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`5
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`expressed motivation to utilize its teachings in devices like that of
`Mabuchi.
`
`For example, in Mazur, paragraph 30, recites that microstructured
`samples, such as those composed primary of silicon having a grown
`cone-like structures formed thereon by laser light have many
`applications, which include their use in absorbing devices, such as solar
`cells, photodetectors and other photovoltaics devices.
`
`Mazur further teaches that there are immense benefits to use of its
`texturing technique, including the potential for remarkable -- and that is
`Mazur's own phrase -- remarkable increase in absorption in wavelength
`and silicon that were previously nearly unusable. That can be seen, for
`example, in Mazur Figure 3 in paragraph 36. Mazur also teaches that
`you can get increased quantum efficiency and a response signal. And
`that's Mazur Figure 7 in paragraph 46.
`
`So it's the recitation of these extreme benefits and Mazur's
`instruction that his technique has broad applicability to photodetectors
`and other photovoltaics devices that form the foundation of why one of
`ordinary skill in the art would absolutely have had a reason to combine
`these two references.
`
`So in the face of this reality, Patent Owner attempts to attack the
`reason to combine the references. But there are two important things to
`note; first, Patent Owner has not denied that Mazur contains the
`statements that we're talking about, discussing the broad applicability of
`its texturing technique and the applicability of its texturing technique to
`devices such as that of Mabuchi.
`
`And it is also not disputed that Mazur teaches dramatic
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`6
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`absorptance increases when you use this texture and protocol. So if
`Patent Owner relies on instead is raising the specter of potentially -- and I
`emphasize potentially -- negative side effects that could arise from
`texturing the CMOS device in Mabuchi, but these are overblown in light
`of the record evidence.
`
`For example, Patent Owner and its expert filed drawings that they
`admit are not to scale and were, quote, greatly simplified, which show a
`light ray coming straight down contacting a spike at the edge of a pixel
`and shooting off at a random angle directly into an adjacent pixel. This is
`the phenomenon that's been discussed in this proceeding called optical
`crosstalk.
`
`And this is seen in most clearly on page 44 of the Patent Owner
`response, which includes diagrams from the expert Mr. Guidash's
`declaration at paragraph 82.
`
`So when you look at these drawings -- again, aside from the issue
`that they're not to scale and putting aside also the issue of whether there
`should be something in between adjacent pixels, these drawings fail to
`account for several things, including elements that mitigate optical
`crosstalk already seen in Mabuchi, such as a microlens and a light shield.
`
`So when you look at Figure 8 of Mabuchi, you can see there's a
`microlens 870 sitting on top of the device. And the purpose of this
`element is to converge light that is incoming and focus it down towards
`the photo detecting region 750, about two-thirds of the way down the
`drawing.
`
`And you can -- actually Figure 12 gives a nice basic illustration of
`how the light coming in is focused in that area. Figure 12 discusses the
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`7
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`prior art, but it's essentially the same function as the microlens.
`
`JUDGE ULLAGADDI: If I could stop and ask you a question.
`
`MR. MURRAY: Sure.
`
`JUDGE ULLAGADDI: In the combined device of Mabuchi and
`Mazur, what is your position with respect to the probability that light is
`reflected into a neighboring pixel as compared to the probability that
`light is absorbed by the combined device, the device with the textured
`region?
`
`MR. MURRAY: The probability of it being deflected to an
`adjacent pixel compared to the probability it's absorbed in the pixel for
`which it's intended?
`
`JUDGE ULLAGADDI: Yes.
`
`MR. MURRAY: If I understand.
`
`Okay. Our position is that Mabuchi has, as I was discussing, the
`microlens and the light shield layer that make it less likely that you're
`going to have this, because these are going to sit over top of the texture,
`which would be on top of the silicon, and so initially your light coming in
`is all going to be kind of focused down towards the center of the pixel, so
`you're going to have a much longer distance to go to get -- for a photon to
`get to an adjacent pixel.
`
`And there's also other techniques that are known in the art which if
`this were a concern for someone of ordinary skill, they could have used
`to reduce that risk, mitigate it, to the extent necessary.
`
`One example of that would be Exhibit 2006, which Patent Owner
`submitted, and what you have there is a contemporary reference that
`recognized, yes, there's a potential risk of optical crosstalk with texturing
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`8
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`surface, but they go ahead and texture it anyway despite that risk because
`there are known ways to address it. Mabuchi has some. There are others
`known in the art, including Exhibit 2006.
`
`And crosstalk, whenever you have two pixels together, is always
`going to be an issue, just like dark current is always going to be an issue
`when you have a light-absorbing device. It's not something you can run
`away from in this technology.
`
`JUDGE ULLAGADDI: So then am I understanding correctly that
`the absorption resulting from texturing that region outweighs any
`probability that the light would be redirected into a neighboring pixel?
`
`MR. MURRAY: Outweighs, correct, yes.
`
`JUDGE ULLAGADDI: Thank you.
`
`MR. MURRAY: So actually, that --
`
`JUDGE BRADEN: And, I'm sorry, but can you point us where in
`the record where you had that argument about that -- the increased
`absorption being outweighed by the risk of crosstalk? Where is that in
`the record?
`
`MR. MURRAY: Certainly, the reply -- it's an argument that we've
`made throughout, that when you have --
`
`JUDGE BRADEN: That it's outweighed?
`
`MR. MURRAY: Yes. Not just optical crosstalk, but all of the
`negative side effects that Patent Owner advances in the Patent Owner
`response, are outweighed by the absorption that you're going to get from
`the texture and increase in signal and basically a net gain in the
`functionality of the device.
`
`So we talk about how, on page 14 of the reply, that the texturing of
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`9
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`the light incident surface is likely to cause detrimental color mixing and
`crosstalk is based on exaggerations with disregard for aspects that will
`mitigate potential problems. So we talk about the mitigation aspect of it.
`
`JUDGE BRADEN: So I guess, part of my question is -- and this
`kind of brings up, when you point to page 14 of your reply, that you talk
`about the deactivation of the P positive type acceptors, the dopant
`gradients, and if you look at that, you make the statement on top of page
`15 of your reply, talking about how if someone was concerned about the
`texture affecting this, they could switch to an alternative method. And I
`believe that later on in your reply you talk about that in regards to why
`you would combine with Mazur.
`
`Could you explain that a little bit further to me why someone
`would -- what would lead them to change to this in order to be able to get
`to this texturing?
`
`MR. MURRAY: When you say, "would change to this," you mean
`--
`JUDGE BRADEN: Because you say they could change to it --
`
`MR. MURRAY: Could change to the --
`
`JUDGE BRADEN: -- could do a lot of -- could change the type of
`
`gradient that they're using.
`
`MR. MURRAY: Well, that's -- that argument is really only to the
`effect that if someone was overly concerned about optical crosstalk from
`the texture.
`
`JUDGE BRADEN: Right.
`
`MR. MURRAY: There are alternatives stated within Mabuchi
`which, you know, if -- if you're going to be concerned about deactivating
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`10
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`these P positive acceptors, Mabuchi contains other teachings within it
`which allow you to, if it doesn't suit your needs, make an adjustment to
`balance and compensate.
`
`One of ordinary skill in the art is not an automaton, so you know,
`this -- I'm sure the Board is familiar with In Re: Keller, it's not whether
`these things bodily can be incorporated into one other, it's what the
`teachings are.
`
`So one of ordinary skill in the art would have flexibility to the
`extent there's some issue with the texturing that may be slightly caused P
`dopant gradient changes may cause optical crosstalk, may raise dark
`current. There are things which can easily be done to mitigate those
`factors and still get the immense benefit that you're going to get from
`texture.
`
`JUDGE BRADEN: Okay.
`
`MR. MURRAY: So hopefully that answers your question.
`
`JUDGE BRADEN: Okay. Thank you.
`
`MR. MURRAY: So like I said, Patent Owner throws out a number
`of potential concerns, dark current being one, recombination from surface
`damage being another. All of which are shown by the evidence in the
`record to be exaggerated based a lot on exhibits in the record, such as
`Moloney, and Mazur itself in particular.
`
`But the evidence that we have demonstrates that the benefit
`attributable to texturing simply outweighs any of these alleged negative
`consequences. Again, you're getting a net gain in performance,
`particularly at wavelengths in which you would not have seen any real
`result previously.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`11
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`And Patent Owner's approach is to basically highlight worst-case
`
`scenarios, "Well, a photon could hit the edge of a pixel and deflect over
`to an adjacent pixel." It could, but is that enough of a risk to make
`someone of ordinary skill in the art throw up their hands and say, "I give
`up"? And the Petitioner's position is that is not the case here.
`
`So unless there's any other questions on the Mabuchi combination,
`I was going to transition to Nakashiba.
`
`Okay. So as it turns out, the physical structure of the claims in the
`'591 patent already existed in fingerprint sensors prior to 2009. And that,
`of course, is Nakashiba. Again, we have no dispute that Nakashiba
`discloses an imager with all of the standard imager elements discussed
`earlier.
`
`And there's no dispute that Nakashiba included a textured region.
`And this textured region is coupled to the semiconductor surface opposite
`to a planar surface and opposite a plurality of dopant gradients.
`
`So the texture region in Nakashiba is in the same spot as the
`embodiments of '591, but Patent Owner says because of the language
`“positioned to interact with electromagnetic radiation,” Nakashiba is
`excluded.
`
`This is about location, not result. And the Board, we think,
`correctly found that because this textured region in Nakashiba was
`located at the light incident surface of the device and received the
`incoming light, it was positioned to interact with, as construed by the
`Board, electromagnetic radiation. And I'm referring to the decision at 19,
`light can get to it.
`
`Patent Owner obviously disagreed and maintains two alternative
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`12
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`rationales for why Nakashiba doesn't anticipate independent Claims 1
`and 13. And I'll take these in reverse order from the briefing.
`
`So the first position is that the position to interact limitation must
`somehow require an effect on the incident radiation, and particularly one
`that differs from planar silicon.
`
`Patent Owner's expert acknowledges that refraction and reflection
`will occur at material boundaries where the materials have differing
`indices of refraction. This is at paragraph 68 of Mr. Guidash's
`declaration. And he reiterates it at pages 14 through 16 and 18 of his
`deposition.
`
`And Dr. Souri has provided an illustrative example to demonstrate
`that Nakashiba discloses embodiments for every refraction and reflection
`can be measurably different than at a planar surface.
`And I refer to page 6 of the reply, which takes Dr. Souri's image
`from his declaration of paragraphs 30 to 33. And what you see on the
`right-hand side on page 6 is planar silicon, you have light coming straight
`down vertically, and at the surface, one of two things will happen --
`actually both can happen -- part of that light will go straight into the
`silicon in the same downward vertical direction and some of it will reflect
`back vertically upward. That arrow is not shown here because it overlaps
`with the incoming light.
`
`But when you look on the left, what you have here is an
`embodiment of Nakashiba where you have a five to one width to height
`ratio for each feature. And so the same incoming light vertically
`downward hits the surface, and it's actually hitting at about 11 degrees
`from the normal of that surface, and so part of the light will reflect.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`13
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`And it reflects out at the same angle it comes in, so what you have
`
`is 11 degrees coming in, 11 degrees the other way coming out, you have
`22 degrees of reflection -- 22 degrees change as it reflects, and then going
`into the silicon, part of the light will refract.
`
`And what's shown here, taking air as the upper material, and I
`believe it's 550 nanometer wavelength light we're talking about, so the
`calculation is done that your light direction will change by about 9
`degrees when you enter the silicon. Clearly different from planar silicon.
`
`JUDGE ULLAGADDI: Can I ask you a question about how
`Nakashiba shows the finger actually on top of the roughened surface? Is
`the index of refraction with the finger the same as air and does it differ
`from the silicon?
`
`MR. MURRAY: There's no evidence either way what it is. Dr.
`Souri estimates it would actually be closer to water. But two things on
`that issue: One is this is an example, and what we're showing is basically
`the general broader concept that when you have two different materials of
`different indices of refraction, with planar and with the features that are
`cited in Nakashiba, you will have a difference; and, second, as Dr. Souri
`testified, contact with the finger is not going to be conformable, there will
`be air pockets.
`
`And actually when you look at Figure 3 of Nakashiba, there's --
`due to the curve of the thumb, you're going to see there are photo-
`detecting elements, which sit beneath an air pocket underneath the
`thumb. So the air example is completely legitimate under what's shown
`in Nakashiba.
`
`So hopefully that answers your question.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`14
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`JUDGE ULLAGADDI: Thank you.
`
`You have about a minute remaining.
`
`MR. MURRAY: Okay. All right. So the other argument that they
`
`have is on the construction of “positioned to interact” being construed to
`require enhanced response of filtering of received radiation. I don't think
`there's any need to go into detail on this. The Board has already twice
`found this construction to be improper. The Board correctly determined
`that the phrase "interact" in particular was not defined by the '591 patent,
`nor was it equivalent to Patent Owner's requested construction.
`
`And indeed, further to that point, the term interact is not even used
`consistently in the '591 patent, because in the background section, it's
`used to refer to the prior art, which didn't exhibit any of these enhanced
`effects. That's column 1, lines 15 to 16.
`
`Again, this is about location on the substrate, not a functional
`result of the placement. So Petitioner submits that the Board continue to
`decline to adopt Patent Owner's overly narrow construction.
`
`Unless there's any further questions, I reserve the rest of my time
`for rebuttal.
`
`JUDGE ULLAGADDI: Judge Braden, Judge Clements, any
`questions?
`
`JUDGE BRADEN: I have no further questions at this time.
`
`JUDGE CLEMENTS: No questions.
`
`JUDGE ULLAGADDI: Thank you.
`
`MR. MURRAY: Thank you, Your Honor.
`
`MR. BELANGER: Good afternoon. I'd like to start the
`presentation directing you to slide 9 of our demonstratives, which I
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`15
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`believe is an illustrative of some of the questions that you asked Counsel
`in Counsel's initial argument, which is whether it would have been
`obvious to one of skill in the art to combine the Mazur reference with the
`Mabuchi reference in order to create a functional device that would meet
`each and every limitation of the claim.
`
`I would start with the basic observation that it is Petitioner's burden
`to show unpatentability. And it is clear on this record that they have not
`done so for several reasons. The primary reason being, as the Board had
`acknowledged in the institution decision, Mabuchi's entire focus and
`purpose is to avoid color mixing that would be caused by optical
`crosstalk. And so the key benefit that Mabuchi is directed to is avoiding
`color mixing.
`
`The way in which Mabuchi achieves this benefit, according to
`Mabuchi, is by including a gradient or alternative embodiments, applying
`an electrical field to create an electrical gradient so that electrons are
`moved from the surface of the semiconductor substrate towards a PN
`junction.
`
`In Mabuchi, his motivation for including this electric gradient is --
`the reference clearly states that for manufacturing concerns, he wants to
`create a substrate that is 10 micrometers or greater. So he's looking to
`create a device that has a thick epitaxial layer, a thick substrate where the
`minimal dimension that is acceptable in Mabuchi is 10 micrometers. And
`he suggests that it should be either 10 micrometers or greater.
`
`He also has as his goal to create smaller pixels which are closer
`together. And he states that preferably the pixels will be a few microns
`square or less. And so what we've shown in demonstrative page 9, is an
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`16
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`illustration responding to Dr. Souri's critique that, in fact, the proposed
`combination would create this exact type of crosstalk or color mixing that
`Mabuchi is intending to avoid.
`
`So according to Dr. Souri's deposition testimony, adding the
`texture to the Mabuchi device would cause certain light rays to be
`reflected at a 45-degree angle, and given the dimensions stated in
`Mabuchi, those photons would travel across at least one, if not two,
`adjacent pixels.
`
`And what we've illustrated here is the most favorable dimensions
`from the combination. So, for example, Mabuchi states that the pixels
`should be at a pitch of two and a half to three microns or less, and so
`we’ve illustrated a pitch of three microns. He states that thickness should
`be 10 micrometers or more, and so we've shown the dimension as 10
`micrometers.
`
`And Dr. Souri testified at his deposition that the infrared light that
`is traveling through the Mabuchi device would travel at least 15
`micrometers, but potentially as far as 50 micrometers. And so if there's a
`45-degree angle, using the minimal dimensions in this combination, there
`would be crosstalk across at least two pixels, if not more in the proposed
`combination.
`
`JUDGE ULLAGADDI: Before you continue, Counsel, I would
`like to ask you the same question that I asked Petitioner's Counsel, and
`that is, is it your position that the probability that this incident light
`travels to this neighboring pixel outweighs the probability that the
`incident light was absorbed by the textured region?
`
`MR. BELANGER: Yes. And I would say that there is clearly no
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`17
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`evidence provided by Petitioner, other than the argument that he made to
`the contrary. So Dr. Souri does generally say that there is increased
`absorption, but there is absolutely no analysis of the relevant -- the
`relative increase in absorption versus decrease in performance due to
`crosstalk.
`
`So there's been no analysis done by the Petitioner that could
`support his statement that the benefit of absorption outweighs the
`detriment of crosstalk.
`
`JUDGE ULLAGADDI: Can you point to somewhere in the record
`where that argument -- that you have that argument that the crosstalk --
`or the crosstalk outweighs the absorption?
`
`MR. BELANGER: Yes.
`
`JUDGE ULLAGADDI: Because in this figure, the green arrow
`comes into the textured region and then it looks like all of the incident
`light is traveling at a 45-degree angle, it doesn't appear that any incident
`light is being absorbed.
`
`MR. BELANGER: Correct. So the first point of reference would
`be the Mabuchi reference itself. So in paragraph 38 and paragraph 40 of
`the Mabuchi reference, he describes dimensions that would be used for
`infrared radiation.
`
`And just to back up and give foundation. So the Petitioner has
`suggested that the motivation to combine Mazur with Mabuchi has to do
`with the increased absorption of infrared wavelengths of light provided
`by Mazur. So the benefit that they're pointing to is infrared wavelengths
`as opposed to visible light wavelengths. So that's why we have
`illustrated here the performance with infrared wavelengths.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`18
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`In Mabuchi, at paragraph 35 -- starting at paragraph 35, he
`
`describes that, "From view of light frequency characteristics desirable
`form thicknesses range from 5 microns to 15 microns for visible light,
`and from 15 microns to 50 microns for infrared light."
`
`And so if you then look at Dr. Souri's testimony, the Petitioner's
`expert, which we've cited in the demonstrative, page 184, line 12 to 16,
`and page 192, line 7 to 24, he is relying on that range of 15 microns to 50
`microns to say what Mabuchi teaches as to where the infrared
`wavelengths of light will be absorbed.
`
`And so Mabuchi's teaching is that in order to absorb infrared
`radiation rather than applying texture, which is not taught or suggested
`anywhere in Mabuchi, you would increase the thickness of the substrate
`to increase the possibility of absorption.
`
`So rather than adding texture, what Mabuchi suggests is that for
`infrared light, you create a substrate that is between 15 microns and 50
`microns deep. And so the support for the infrared wavelengths being
`absorbed between 15 microns and 50 microns in the Mabuchi device
`comes directly from Mabuchi. And that is supported by Dr. Souri's
`testimony, which we've also cited in the demonstratives.
`
`JUDGE ULLAGADDI: Actually, my question was, was there
`support for the position that the crosstalk outweighs the absorption?
`
`MR. BELANGER: Yes. And I'll -- that is in our response at pages
`38 to 47. And that is citing Mr. Guidash's testimony in support of that.
`
`JUDGE ULLAGADDI: Thank you.
`
`JUDGE CLEMENTS: Counsel, this deposition testimony from
`Dr. Souri is in the context of not having a light shield, and I heard
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`19
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`Petitioner argue just now that what you've illustrated here on slide 9 isn't
`the necessary result of the combination of Mabuchi and Mazur, because a
`person of ordinary skill in the art would know how to do things like this
`light shield taught in Mabuchi to prevent the extreme crosstalk illustrated
`there. So I think -- what does Patent Owner say in response to that?
`
`MR. BELANGER: Sure.
`
`And to explain my answer, I'd like to direct you to slide No. 6 of
`the demonstratives. And what slide No. 6 reproduces is a figure from
`Mr. Guidash's declaration next to an annotated figure from Figure 8 of
`Mabuchi. And Figure 8 is the figure that Counsel referred to in his
`argument as illustrating a solution to crosstalk.
`
`So the first problem is this was not a configuration or combination
`that Petitioner advanced in their petition, this is something that has been
`raised in reply. And neither Petitioner, nor their expert, has actually
`explained how one of skill in the art could apply texture in a functional
`process to the device shown in Figure 8.
`
`As you can see from Figure 8, there are multiple components, 830
`and 840, that would need to be redesigned and reconfigured in order to
`accommodate texture. And there's no evidence in the record that that
`would be within the capability of one of skill in the art to form those
`layers over a layer that had been created using the texture described in the
`Mazur reference. So there's no evidence in the record that that would be
`within the capacity of one of skill in the art.
`
`But even if it were, what Counsel just --
`
`JUDGE BRADEN: I want to make sure I under your argument,
`because I was -- so despite the fact that Mabuchi teaches the use of this
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`20
`
`
`
`Case IPR2016-01910
`Patent 8,680,591 B2
`
`microlens and