[00:01:18] Speaker 04:
Okay, I think we're ready, Mr. Kaufman.

[00:01:20] Speaker 04:
Good morning.

[00:01:20] Speaker 02:
Good morning, Your Honors.

[00:01:21] Speaker 02:
May it please the Court?

[00:01:25] Speaker 02:
I intend to focus on two of the key issues that we presented in our briefing today.

[00:01:29] Speaker 02:
The first is the construction of what is referred to as the filling limitation that is found in independent claims 1 and 9 of the 714 patent.

[00:01:39] Speaker 02:
And that term is filling said bit line contact openings, said substrate contact openings, and said gate contact openings

[00:01:46] Speaker 02:
with a conducting layer to complete said forming contacts and fabrication of said integrated circuit device.

[00:01:53] Speaker 02:
And we believe and contend that the board unreasonably construed that term when it engrafted a number of conditions that are nowhere supported by the intrinsic record.

[00:02:05] Speaker 02:
With respect to claim nine, there is no substantial evidence in this record that

[00:02:11] Speaker 02:
The prior art relied upon by Samsung inherently discloses a claim notation found only in independent claim 9, wherein the etched top layer is etched into a lesser.

[00:02:22] Speaker 04:
OK, you're using a lot of time to do an introduction, so why don't you start talking about the specifics of our argument, starting with the one you mentioned with just claim construction.

[00:02:32] Speaker 02:
All right.

[00:02:34] Speaker 02:
The board's construction is unreasonable primarily because it engrafts all sorts of explanations and gloss on the construction that are simply inconsistent with the way the invention is described in the 714 patent.

[00:02:51] Speaker 04:
I may be confused, or maybe I'm thinking about a different case.

[00:02:58] Speaker 04:
construction that grafts these additional requirements.

[00:03:01] Speaker 04:
Am I wrong about that?

[00:03:02] Speaker 04:
You're trying to say that there has to be a single process step, or that all three openings have to be filled with the same substance.

[00:03:11] Speaker 04:
And I'm not seeing that in the claim language.

[00:03:14] Speaker 04:
So have I got the right case and the right facts here?

[00:03:17] Speaker 02:
You have the right case and the right facts, Your Honor.

[00:03:21] Speaker 02:
We submit that the limitation naturally reads as a single filling step.

[00:03:26] Speaker 02:
the three specific openings with a layer to complete the contacts.

[00:03:30] Speaker 02:
And that's a single filling step that occurs.

[00:03:34] Speaker 02:
And that's what's described in the patent specification.

[00:03:38] Speaker 03:
The claim just talks generally about filling, not filling entirely and not filling at a single time.

[00:03:52] Speaker 03:
I think Judge Gross is right that you're arguing limitations that are not there.

[00:03:59] Speaker 02:
With respect, Your Honors, we think that the claim naturally requires filling to complete, and that's something so you get a functional contact.

[00:04:11] Speaker 02:
You have to fill to get a functional contact.

[00:04:14] Speaker 02:
We'll leave aside the degree of fullness, but just you have to fill it, and it has to be a process that fills all three contacts to ultimately yield a functional contact.

[00:04:25] Speaker 04:
But not at the same time and the same step.

[00:04:29] Speaker 02:
We would respectfully disagree, Your Honor.

[00:04:32] Speaker 04:
Where's the language that leads you to that conclusion?

[00:04:35] Speaker 04:
What is it about the claim language that leads you to say it has to be in the same step at the same time?

[00:04:42] Speaker 02:
It's the filling, and then it to completion.

[00:04:46] Speaker 02:
And the filling to completion of the contact suggests to us that it is not the same.

[00:04:50] Speaker 04:
Are we looking at the languages?

[00:04:54] Speaker 04:
Filling said bit line contact openings, said substrate openings, and said gate openings with a conducting layer to complete.

[00:05:03] Speaker 04:
Yes, sir.

[00:05:04] Speaker 04:
OK.

[00:05:07] Speaker 00:
So why does that not indicate that as long as you, at some time or another, put enough conducting material in for the contact to be complete, you're done.

[00:05:19] Speaker 00:
So it doesn't have to be the same conducting material.

[00:05:21] Speaker 00:
It doesn't have to be all done at once.

[00:05:23] Speaker 00:
And it doesn't have to go up to the lip.

[00:05:25] Speaker 02:
First, Your Honor, the specification just describes a single filling step.

[00:05:32] Speaker 02:
The holes are formed in a single etch, and then they're filled in a single filling process.

[00:05:39] Speaker 02:
And that process, that overall process, is designed to reduce the number of steps that are used during the formation of the semiconductor process, and that's a goal of this.

[00:05:50] Speaker 03:
Well, the specification just describes one way of accomplishing this result, but the claim is much broader.

[00:06:02] Speaker 03:
You're entitled to claim it more broadly as long as it's described and enabled.

[00:06:08] Speaker 03:
You can't read the limitations from the specification into the claim.

[00:06:12] Speaker 02:
And I would submit that we're not intending to read the limitations into the claim.

[00:06:16] Speaker 02:
We're attempting to show that the board's construction is inconsistent with the invention described in the specification.

[00:06:23] Speaker 02:
There's no disclosure

[00:06:24] Speaker 02:
in the specification of filling the holes separately, which would require wholly additional masking processes that aren't disclosed.

[00:06:35] Speaker 02:
This isn't like something where you can just fill an individual hole and then fill a separate individual hole.

[00:06:43] Speaker 02:
The wafer is in a processing chamber and the

[00:06:47] Speaker 02:
conductive material is applied across the wafer.

[00:06:49] Speaker 02:
So there's not some process that's described that would suggest you could fill a hole, do a bunch of other stuff, fill a second hole, do a bunch of other stuff, and then fill the third hole.

[00:07:02] Speaker 02:
That's not what's talked about at the 714 patent.

[00:07:05] Speaker 02:
That's not what the invention is describing.

[00:07:08] Speaker 02:
And that's not what is appropriate in this context.

[00:07:13] Speaker 02:
The board got hung up on the use of the word concurrently, which appears earlier in the claim regarding the etching steps.

[00:07:21] Speaker 02:
And that was necessary because the rate of claim was drafted.

[00:07:25] Speaker 02:
There was a first etching step and a second etching step in the clause.

[00:07:29] Speaker 02:
And to refer to them occurring at the same time, they used the term concurrently.

[00:07:33] Speaker 02:
But that doesn't mandate the need for that term in the filling step.

[00:07:40] Speaker 04:
Well, it suggests it doesn't mandate, perhaps.

[00:07:44] Speaker 02:
I would say that it doesn't even suggest it, in the sense the filling to completion is sufficient.

[00:07:51] Speaker 02:
And this matters because in the Okawa prior art, which is the primary reference relied upon by Samsung and the board, there's actually a multi-step process to

[00:08:05] Speaker 02:
form Okawa's holes that consist of a number of intervening steps, and that Okawa's initial filling step would not complete the three contacts in Okawa.

[00:08:18] Speaker 02:
And that is because if you consider figures 12a, 12b, and 12c, which is what was relied on below,

[00:08:27] Speaker 02:
After the first etch, contact holes 21A and D are ready for filling.

[00:08:34] Speaker 02:
They are exposed to substrate in the gate, and if you fill them, they would form a fully functional contact.

[00:08:40] Speaker 02:
However, what is identified as 21B over bit line 5A, there's a layer 7A on top of that.

[00:08:48] Speaker 02:
It's a silicon nitride layer described by the patent.

[00:08:52] Speaker 02:
described by the Okawa Pat, and if you filled that hole, you would not have a functional contact.

[00:09:00] Speaker 02:
And indeed, Okawa goes on to describe subsequent steps after its first filling step, where it conducts additional etches to form sidewalls in 21D, and then it conducts an etch step to etch through 7A.

[00:09:18] Speaker 02:
And only then does

[00:09:21] Speaker 02:
is a filling step conducted that would actually complete what is now referred to somewhat confusingly as 21C in figure 12C.

[00:09:30] Speaker 02:
Only after that step would you have all three contacts full.

[00:09:36] Speaker 02:
And what the board's construction does is it in effect says, if at any time you have three full contacts, then this claim limitation is met.

[00:09:44] Speaker 02:
And we would submit that that would cover anything in the prior art, and that's unreasonably broad.

[00:09:53] Speaker 02:
If there are no further questions, I would turn to Claim 9.

[00:09:58] Speaker 02:
Claim 9 includes the further limitation regarding differential etching in terms of the etching at the top of the gate versus the bottom of the other two contact holes.

[00:10:12] Speaker 02:
And the board made no findings that this was explicitly disclosed in the art.

[00:10:16] Speaker 02:
And indeed, Samsung advanced a theory of inherency

[00:10:21] Speaker 02:
that this was disclosed inherently by the art, not explicitly by the art.

[00:10:27] Speaker 02:
In response, Your Honors, we presented evidence from our expert, Mr. Maltiel, as well as siding to patents that Samsung had cited for the state of the art, particularly the multi-patent, which shows that in some cases,

[00:10:44] Speaker 02:
the Samsung's time-based etching might occur, but in some cases it would not.

[00:10:52] Speaker 02:
And particularly under the conditions described by Nolte, where Nolte describes process conditions where you do not obtain the differential etching, that it would not necessarily occur in the way that Samsung was proposing.

[00:11:09] Speaker 02:
The board made no factual findings about this.

[00:11:12] Speaker 02:
The board did not discuss nullity.

[00:11:14] Speaker 02:
The board did not make any conclusions about the fact that this did or did not necessarily occur that are adequate to support its obviousness finding.

[00:11:24] Speaker 02:
And so we would submit, Your Honor, that claim nine, independently, whatever this court does with claim one and the filling step, independently, claim nine is patentable because there's no substantial evidence of the differential

[00:11:40] Speaker 02:
that the differential steps.

[00:11:41] Speaker 04:
Well, what about the board's discussion of this is at 26 and 27, right?

[00:11:45] Speaker 04:
Right.

[00:11:47] Speaker 02:
That's correct, Your Honor.

[00:11:48] Speaker 02:
And it's a very summary discussion.

[00:11:50] Speaker 04:
Well, I mean, that's in the eye of the beholder, I guess.

[00:11:53] Speaker 04:
I mean, it is a discussion.

[00:11:55] Speaker 04:
There's a long paragraph in there that's quite detailed that has citations to the expert declarations, to the reference.

[00:12:04] Speaker 04:
And it kind of explains why

[00:12:07] Speaker 04:
their alternative aspect ratio construction claim term.

[00:12:12] Speaker 04:
They disagree with that.

[00:12:13] Speaker 04:
So I mean, I guess one can always say somebody could have said more or gone into more detail.

[00:12:19] Speaker 04:
But you think that this discussion somehow lacks?

[00:12:25] Speaker 02:
So you're on with the episode.

[00:12:27] Speaker 02:
Go ahead.

[00:12:28] Speaker 02:
So there are two aspects here.

[00:12:29] Speaker 02:
One is the aspect ratio, and that was an argument presented below and the board disposed of it.

[00:12:34] Speaker 02:
The other was that we said, look, there are these conditions where what they say is happening in the time-based process, and specifically the conditions of multi,

[00:12:43] Speaker 02:
with these specific etching gases and conditions where you won't obtain the differential etching even though you have a gate electrode that is higher than the three-dimensional space and therefore is exposed, as I understand, some time longer to the etchant,

[00:13:01] Speaker 02:
And we presented that argument, and the board says nothing about that aspect of it or why or how this would necessarily occur and thus satisfy the inherency problem.

[00:13:12] Speaker 02:
So that's what we're focused on, Your Honor, is there's nothing about NOLTE in this paragraph.

[00:13:18] Speaker 02:
And there's nothing about NOLTE in this discussion.

[00:13:21] Speaker 02:
The board didn't engage with the fact that sometimes you may get the etching and sometimes you won't.

[00:13:26] Speaker 02:
And if that's the case and our Continental can, there's no inherency and there's no disclosure by inherence.

[00:13:31] Speaker 02:
And I see that I'm in my rebuttal time.

[00:13:59] Speaker 01:
Good morning.

[00:14:00] Speaker 01:
Good morning, Your Honor.

[00:14:01] Speaker 01:
May it please the Court?

[00:14:03] Speaker 01:
The Board correctly construed the filling limitation, especially given the broadest reasonable interpretation standard that applies here, and substantial evidence simply supports all of the Board's findings.

[00:14:14] Speaker 01:
Let me jump right into the filling limitation construction.

[00:14:18] Speaker 01:
So as the Court recognizes, PROMAS is trying to import

[00:14:22] Speaker 01:
three limitations into this very simple limitation from our perspective.

[00:14:27] Speaker 01:
They want to say that it's one step that is concurrently.

[00:14:31] Speaker 01:
It's the same material, and you make the contact full.

[00:14:36] Speaker 01:
We submit this intrinsic evidence simply does not limit the invention in such a manner.

[00:14:42] Speaker 01:
So if we start with the concurrently limitation of the single step, starting with the claim language, it is a broad step filling.

[00:14:49] Speaker 01:
It includes sub-steps, as the board recognized.

[00:14:52] Speaker 01:
And if you look at their other claim limitations, specifically the concurrently etching limitation, they knew how to claim concurrently.

[00:15:00] Speaker 01:
And I heard my colleagues say that this is the whole point of the invention, that filling is the whole point of the invention.

[00:15:07] Speaker 01:
That's simply not true.

[00:15:09] Speaker 01:
Their invention is about concurrently etching.

[00:15:12] Speaker 01:
And they claimed that everywhere in the specification when they wanted to say something was concurrently, they said so, but not so with respect to the filling limitation.

[00:15:21] Speaker 01:
And with respect to the same layer concept, the same material, again, the claim language is broad here.

[00:15:27] Speaker 01:
It does not support that.

[00:15:29] Speaker 01:
And if you look at our expert's testimony, he showed how this language is broad and submitted the plain and ordinary meaning.

[00:15:37] Speaker 01:
And that's in appendix 1473.

[00:15:39] Speaker 01:
And then finally, in terms of the completely full limitation that they want to inject into the claim, again, that's simply not supported.

[00:15:46] Speaker 01:
Both of the experts here submitted testimony that said,

[00:15:50] Speaker 01:
that a contact does not have to be 100% full for it to be completely full.

[00:15:56] Speaker 01:
And you can find testimony from Samsung's expert at Appendix 1471.

[00:15:59] Speaker 01:
And you can find testimony from Mr. Maltillo, their expert, at Appendix 1446, for example.

[00:16:07] Speaker 01:
And in fact, he said a contact could be 85% full and still meet this limitation.

[00:16:13] Speaker 01:
So their claim construction simply should be rejected.

[00:16:16] Speaker 01:
Unless there are any questions on the claim construction issue, I'd like to go to claim nine.

[00:16:20] Speaker 01:
Turning to the substantial evidence issues, I think it basically comes down to they want a better written decision.

[00:16:28] Speaker 01:
But as this court is aware, as long as this court can reasonably discern the board's fact findings and the path it took to arrive at its decision, this court should affirm.

[00:16:39] Speaker 01:
And we submit the decision does that.

[00:16:41] Speaker 01:
So if you look at claim nine, I heard counsel say that, well, the board didn't say anything about NALTY.

[00:16:46] Speaker 01:
I think we should look at the board's decision.

[00:16:47] Speaker 01:
And I'll show you what the board said about NALTY, or at least it referred to the arguments from us.

[00:16:55] Speaker 01:
And the way the board set up its decision was they started with Samsung's arguments.

[00:17:00] Speaker 01:
They went through them.

[00:17:01] Speaker 01:
They sided to the expert testimony.

[00:17:02] Speaker 01:
And with respect to claim nine, we can see that at appendix 20, the board's final decision.

[00:17:08] Speaker 01:
And that's where this limitation is discussed at the bottom.

[00:17:11] Speaker 01:
The board starts with the arguments that Samsung made on appendix 20.

[00:17:16] Speaker 01:
and basically shows that Samsung made two arguments, one based on Okava and why it would be inherent.

[00:17:23] Speaker 01:
Again, it cites to the expert testimony as well as Samsung's petition.

[00:17:28] Speaker 01:
It also recognized that Samsung actually made an alternative argument, obviousness based on Okava and Jenga, and you can see that at appendix 21 at the top.

[00:17:36] Speaker 01:
Then, if you can jump to appendix 26,

[00:17:42] Speaker 01:
The board then carefully went through and addressed Promas's arguments.

[00:17:45] Speaker 01:
So if you look at Appendix 26, under subheading C, the differential etching of the etch stop layer limitation.

[00:17:52] Speaker 01:
You can see there, the first full paragraph, the board carefully steps through Promas's arguments.

[00:17:59] Speaker 01:
And what I want to focus on is the last sentence of that, or the last couple of sentences in that paragraph.

[00:18:05] Speaker 01:
And I'll read them for the court.

[00:18:07] Speaker 01:
The board said, patent owner further argues that OKAVA fails to teach such an aspect, ratio-dependent etch, and that such a feature is not inherent to the process of OKAVA.

[00:18:17] Speaker 01:
And the board cites to their patent owner response, and that's appendix 839 to 44.

[00:18:23] Speaker 01:
And that's where they raise the argument on null T with respect to inherency.

[00:18:28] Speaker 01:
And then the board goes on and says, well, they made additional arguments with respect to Zheng's etch process.

[00:18:33] Speaker 01:
And again, that goes to the alternative arguments they made.

[00:18:36] Speaker 01:
or in response to our alternative argument about obviousness.

[00:18:40] Speaker 01:
So the board did carefully look at the arguments.

[00:18:45] Speaker 01:
Then after that, at the bottom of appendix 26, the board recognizes.

[00:18:49] Speaker 01:
They say, well, petitioner response.

[00:18:51] Speaker 01:
And they cite to the reply pages from our submission, including the expert declaration of Dr. Shanfield.

[00:18:58] Speaker 01:
And you can see that on 26 to 27.

[00:19:00] Speaker 01:
And then 27 to 28, the board says they agree with Samsung.

[00:19:04] Speaker 01:
We think the board's decision is supported.

[00:19:06] Speaker 01:
And I think what's telling here is appendix 28.

[00:19:09] Speaker 01:
It's the conclusions paragraph where the board said, we have reviewed

[00:19:14] Speaker 01:
Petitioners' explanations and supporting evidence as to how proffered combinations teach the limitations of independent claims one and nine, as well as its explanations as to how one afforded skill in the art would have combined the relevant teachings of the references.

[00:19:28] Speaker 01:
They go on and say, we've reviewed the patent owner's argument regarding this ground and the independent claims, and then say, we agree with the petitioner's analysis, and we do not agree with the patent owner's arguments.

[00:19:38] Speaker 01:
I think under Chenery, under this court's case law, for example, PACE,

[00:19:42] Speaker 01:
This is more than enough for you to cross this court apart.

[00:19:44] Speaker 01:
Unless you have any other questions, that's all I have.

[00:19:47] Speaker 04:
Thank you.

[00:19:47] Speaker 01:
Thank you.

[00:20:09] Speaker 02:
Going in reverse order, Your Honor.

[00:20:11] Speaker 02:
with Claim 9.

[00:20:13] Speaker 02:
We believe that the discussion, the simple recitation of what we might have said, what they might have said, is not a sufficient factual finding under January of this Court's precedent for Claim 9 because it doesn't address the second issue with respect to the differential etching question, which is different from the aspect ratio, dependent etching.

[00:20:34] Speaker 02:
And with respect to the

[00:20:35] Speaker 02:
The second obviousness reference that Mr. Modi referenced, which we believe we understand to be that Zhang uses the same materials and therefore it would necessarily happen.

[00:20:49] Speaker 02:
We submit there have been insufficient findings on that because the existence and the fact that they use the same material is not in and of itself enough to give rise to the same etching effects, their temperature, pressure, lots of other factors that come into play.

[00:21:06] Speaker 02:
With respect to the claim construction issue, we believe that the board's construction is simply overbroad in an effort to capture the prior art and to include any possible steps that might occur in any intervening processes, and we don't think that the claim, as read in light of the specification and the plain language of the claim term, encompasses such a broad understanding.

[00:21:32] Speaker 02:
So, unless there are further questions.

[00:21:34] Speaker 04:
Thank you.