[00:00:45] Speaker 01:
Next case is Lone Star Silicon Innovations versus Andrei Yankov, 2019-1669.

[00:00:57] Speaker 01:
Mr. Maloney.

[00:01:00] Speaker 00:
Thank you, Your Honor.

[00:01:01] Speaker 00:
May it please the court, Tim Maloney for Lone Star Silicon.

[00:01:04] Speaker 00:
The determination in the final written decision that claims two and seven are unpatentable should be reversed because it was based on a ground of obviousness not set forth in Micron's petition.

[00:01:14] Speaker 00:
The evidence and rationale for invalidating those claims set forth in that final written decision deviates significantly from those set forth in the petition.

[00:01:23] Speaker 00:
And much of that evidence and much of that reasoning appeared for the very first time in Micron's reply brief at the late stage of this proceeding.

[00:01:33] Speaker 00:
and the supreme court has confirmed that the statutory scheme for IPRs requires that the petition be guided from beginning to end, that the proceeding be guided from beginning to end by the grounds set forth in the petition and this court has the authority to set aside.

[00:01:52] Speaker 02:
How is this different from QUASO?

[00:01:53] Speaker 02:
As I understand QUASO, there are certain things that are simply unreviewable by us.

[00:01:58] Speaker 02:
the board does at the petition institution level.

[00:02:02] Speaker 02:
And if the board is simply interpreting the petitioner's petition to have a theory for unpatentability and it moves forward with that and institutes on that theory, then I think that's all there is to say.

[00:02:24] Speaker 02:
And that's the end of it.

[00:02:25] Speaker 02:
That's how I understand Quozo.

[00:02:27] Speaker 02:
There's another group.

[00:02:29] Speaker 02:
We have case law, of course, and you pointed out the recently issued Koninklijke case, that where the board itself announces that it's going to create a brand new theory of unpatentability at the institution stage, that it concedes and acknowledges was not raised in the petition, then of course that's a no-no and that's an example of the board going above and beyond

[00:02:58] Speaker 02:
But getting back to Quozzo, if it's an exercise of the board interpreting the contents of the petition to have a certain theory of unpatentability, then I don't know if that's something

[00:03:12] Speaker 02:
for QOZO that we can review.

[00:03:14] Speaker 00:
Okay.

[00:03:15] Speaker 00:
Thank you, Your Honor.

[00:03:16] Speaker 00:
First of all, QOZO is limited to the issue that was raised by QOZO, which is whether or not the petition provided sufficient particularity for the board's decision to institute the IPR proceeding to the trial stage.

[00:03:33] Speaker 00:
That is not the argument we're making here.

[00:03:35] Speaker 00:
We are not challenging the decision to institute this proceeding to the trial stage.

[00:03:40] Speaker 00:
Our position is that once instituted,

[00:03:42] Speaker 00:
The grounds addressed during trial have to be the same grounds asserted in the petition.

[00:03:48] Speaker 00:
In this case, that was not the case.

[00:03:50] Speaker 00:
In Cuozo, very unique facts.

[00:03:53] Speaker 00:
As the court is aware, the assertion of obviousness was made explicitly against a dependent claim, and the petitioner did not explicitly make that same assertion against the broader independent claim.

[00:04:05] Speaker 00:
The Patent Office, the board interpreted that ground to have been implicitly asserted and that's the type of issue that's your understanding.

[00:04:13] Speaker 02:
I mean, you've repeated the facts of Quozzo.

[00:04:15] Speaker 02:
We're familiar with that.

[00:04:18] Speaker 02:
It cannot be that the Supreme Court's holding and principle in that opinion is limited to the specific facts of that case.

[00:04:27] Speaker 02:
The board instituted on additional claims and the specific claim that it was petitioned for.

[00:04:32] Speaker 02:
So what is your takeaway principle of Quozo?

[00:04:37] Speaker 02:
Something broader than the facts of Quozo?

[00:04:39] Speaker 00:
The principle of Quozo is that

[00:04:42] Speaker 00:
Борда решения о том, чьи предпочтения достаточно поддержать решение 314, и идти впервые с тестом, потому что на минимум один клейм, у нас был минимум один клейм, это не вероятно.

[00:04:58] Speaker 00:
Эта решение, и решения в связи с этой решением, в связи с этой решением, не вероятно.

[00:05:04] Speaker 00:
Это не то, что мы здесь имеем.

[00:05:06] Speaker 00:
the petition alleged grounds that were sufficient to at least allow this proceeding to go forward to the trial stage.

[00:05:13] Speaker 00:
However, it is very clear that with respect to the two dependent claims, the only grounds asserted in the petition were based on a particular combination of Watatani's Figure 1 embodiment, Watatani's Figure 5 embodiment, which was the supposed evidence of a multi-layer silicon nitride etched up,

[00:05:33] Speaker 00:
And Tanaka's silicon nitride film, which had a lower dielectric constant.

[00:05:37] Speaker 02:
We can see from... Of course, in A105106, the petitioner also talked about the Mvella's concept of one tool as reported out by SST in 1987, and then on the next page it talks about the YM reference and all the good benefits and reasons why you would want to have a multi-layer

[00:05:58] Speaker 00:
And importantly, the claim here does not just require a silicon nitride layer or multiple silicon nitride layers.

[00:06:07] Speaker 00:
It requires an etched-up silicon nitride layer formed of at least two layers of silicon nitride and having this low dielectric constant.

[00:06:17] Speaker 02:
Of course, we responded, but the arguments, I mean, to me, isn't that at least indicia that

[00:06:31] Speaker 02:
you did in fact have notice and you did have an opportunity to respond and you exercised that opportunity to respond based on this additional theory of un-talkability.

[00:06:41] Speaker 00:
Your honor, respectfully, we only responded to the arguments that were set forth regarding SST and Wang and those arguments in particular, first of all, SST 1987 does not at all mention anything about etched-up layers.

[00:06:54] Speaker 00:
It is a much more general teaching about how a piece of equipment was available that could make multilayer structures in general.

[00:07:00] Speaker 00:
and it was never relied on to show the existence in the art of a multi-layer silicon nitrate etched up layer.

[00:07:07] Speaker 00:
Wang is also the same.

[00:07:09] Speaker 00:
It is not an etched up layer at all disclosed in Wang.

[00:07:12] Speaker 00:
It is a totally different type of layer for a totally different type of purpose.

[00:07:15] Speaker 00:
We responded in our preliminary response to the arguments

[00:07:19] Speaker 00:
that SST 1987 demonstrated a knowledge in the art that multi-layer structures could be formed with some confidence that they could be formed with success, and Wang was relied on as motivation to form multi-layer structures supposedly because it reduced pinholes.

[00:07:38] Speaker 00:
That was the use of those references in the petition.

[00:07:41] Speaker 00:
The Watatani Figure 5 embodiment was used to allege that it was known in the art to make multi-layer silicon nitride etched-up layers.

[00:07:52] Speaker 00:
That turned out to be incorrect.

[00:07:54] Speaker 00:
The board interpreted that figure as not disclosing that and the intervener is not challenging that.

[00:08:00] Speaker 00:
So we can see how dramatically the final written decision grounds deviates from the grounds alleged in the petition

[00:08:09] Speaker 00:
the board's own statements to your earlier question that in philips the board acknowledged it was asserting a third ground here the only difference is the board has not acknowledged that what it's done is a new ground but it clearly is the substance of their decision demonstrates this with respect to the etched up layer this multi-layer silicon nitrate etched up the final written decision again acknowledges

[00:08:34] Speaker 00:
That there was this dispute regarding whether Figure 5 of Watatani discloses that.

[00:08:40] Speaker 00:
And in the final written decision, the board says, we no longer have to address that dispute.

[00:08:44] Speaker 00:
Because why?

[00:08:45] Speaker 00:
We're relying on Yoda for that.

[00:08:47] Speaker 00:
Yoda showed up in the reply brief, Micron's reply brief.

[00:08:50] Speaker 00:
It was not in the petition.

[00:08:52] Speaker 00:
And then at Appendix 38, the board goes into detail about what Yoda discloses and another expert declaration from FAIR interpreting that.

[00:09:01] Speaker 00:
and says the board finds that Yoda discloses silicon nitride etched ops.

[00:09:07] Speaker 00:
That was the key element that the petition relied on Watatani for.

[00:09:11] Speaker 00:
At footnote 19, the board says in view of this finding that the multilayer etched up layers were known in the art, citing to Yoda, it is not necessary to address the argument about what Watatani Figure 5 discloses.

[00:09:25] Speaker 00:
And at footnote 20, the board confirmed that the result of this is it entirely changes the analysis of motivations to combine.

[00:09:34] Speaker 00:
At footnote 20, the board says, because we rely on,

[00:09:38] Speaker 00:
do not rely on Watatani to teach the multi-layer silicon nitride etched-up layer.

[00:09:44] Speaker 00:
We do not need to address the argument that this feature is not disclosed by Watatani or patent owners' arguments regarding the lack of motivation and defeating the purpose of Watatani's structure.

[00:09:58] Speaker 00:
This is a concession by the board that the grounds in the final written decision deviate so much from those set forth in the petition

[00:10:05] Speaker 00:
That it no longer needs to address the core issues that the petition raised that we challenged in our preliminary response.

[00:10:13] Speaker 00:
That Watatani Figure 5 doesn't disclose a multi-layer silicon nitride etched up.

[00:10:17] Speaker 00:
Therefore, your proposed combination makes no sense.

[00:10:20] Speaker 00:
And if you did use it, it wouldn't lead to the claimed invention because Watatani's is not two layers of silicon nitride, which is the required element of claims two and seven.

[00:10:31] Speaker 00:
And moreover, our preliminary response and our full patent owner's response

[00:10:35] Speaker 00:
went into detail about why there wouldn't be a motivation to use Watatani Figure 5 in combination with its conventional Figure 1 embodiment in view of Tanaka's low dielectric film.

[00:10:46] Speaker 00:
That was the issue raised by the petition.

[00:10:50] Speaker 00:
The petition cites to SST and to Wang again as evidence of knowledge of the ability to make silicon nitride

[00:10:59] Speaker 00:
is a multi-layer structure, but there's nothing in SST that mentions anything about edge-top layers.

[00:11:05] Speaker 00:
And the same is true of Watatani.

[00:11:06] Speaker 00:
That wasn't a nudge-top.

[00:11:07] Speaker 00:
The board acknowledged in the final written decision, I'm sorry, I think I misspoke, Wang, the board acknowledged in its final written decision that Wang does not disclose edge-top layers either.

[00:11:18] Speaker 00:
That was a misrepresentation, frankly, by Micron.

[00:11:21] Speaker 00:
So by the time we get to the final written decision, the combination is really Watatani Figure 1 with Yoda for the etched-up layers, the multi-silicon nitrate etched-up layers.

[00:11:32] Speaker 00:
Tanaka's still in there, but now Tanaka's being interpreted based on Yoda, Hogan, and Natori, all three of which showed up in the reply brief.

[00:11:42] Speaker 02:
You're talking about reasonable expectation of success, right?

[00:11:45] Speaker 02:
Yes, yes.

[00:11:46] Speaker 02:
But on what we just call motivation combined,

[00:11:48] Speaker 02:
8 39 40 The board is walking through SST and Wang Wang and SST still were included in the board's analysis for motivation to combine It's really a reasonable expectation of success where the board is showing up with Yoda, right?

[00:12:12] Speaker 02:
Yes, the board used Yoda for And it was using Yoda because

[00:12:16] Speaker 02:
I guess in your patent owner response, you must have been saying that, well, even if you were to use the knowledge of the art exemplified in SST and Wang, we have no reasonable expectation of success that there would be a low enough dielectric constant in that outcome.

[00:12:36] Speaker 02:
And then on reply, that's when

[00:12:38] Speaker 02:
We have Yoda who says, well Yoda shows using multilayer edge stop with silicon nitride.

[00:12:48] Speaker 02:
Yodo zeigt, что ты можешь получать довольно низкий диалектический констант, когда ты делаешь это так.

[00:12:58] Speaker 02:
Так что это, в смысле, революция к твоему аргументу, в котором ты должен был делать сервер.

[00:13:03] Speaker 00:
Это правильно?

[00:13:04] Speaker 00:
Я пытаюсь понять это.

[00:13:05] Speaker 00:
Я partly agree with you, but I would like to add a few details.

[00:13:08] Speaker 00:
In our view, Yodo was relied on for two purposes.

[00:13:11] Speaker 00:
One,

[00:13:12] Speaker 00:
to teach the missing disclosure of a silicon nitride etch stop layer, etch stop, very important, for use in a damascene structure.

[00:13:23] Speaker 00:
That was what Figure 5 of Watatani was in the petition for, and that's the only art that was cited in the petition for that key element of the claims.

[00:13:31] Speaker 00:
Yoda was also, to your point, used as part of this new evidence of a reasonable expectation of success in taking

[00:13:39] Speaker 00:
Tanaka's low dielectric silicon nitride material and forming it as a multi-layer structure as opposed to the single layer disclosed in Tanaka and achieving this low below 5.5 dielectric.

[00:13:53] Speaker 00:
Yoda, Kugan and Natori were all cited in a declaration of their expert on the reply for the

[00:14:02] Speaker 00:
It's supporting arguments that it was known that if you reduce the thickness of layers, it lowers the dielectric constant.

[00:14:09] Speaker 00:
And these are very technical arguments.

[00:14:12] Speaker 00:
So that's two reasons why the ground deviated.

[00:14:15] Speaker 01:
Yoda is being used not only for... I thought you wanted to save some rebuttal time.

[00:14:20] Speaker 00:
It's almost gone.

[00:14:21] Speaker 00:
I will, Your Honor, if I could make one point.

[00:14:23] Speaker 00:
We're not arguing...

[00:14:24] Speaker 00:
We're not arguing that the reason for reversal is that we didn't get sufficient notice, although we do believe the notice of a seven-page sir reply in response to all this new evidence was completely unfair.

[00:14:37] Speaker 00:
But our main point is the grounds were new and the board doesn't have the authority to entertain a new grounds.

[00:14:44] Speaker 00:
Thank you, Your Honors.

[00:14:47] Speaker 01:
—Mr.

[00:14:49] Speaker 01:
Kasdan, as an intervenor.

[00:14:50] Speaker 03:
—Yes.

[00:14:52] Speaker 03:
May it please the court.

[00:14:54] Speaker 03:
This case, as my friend explained, involves two issues.

[00:14:58] Speaker 03:
One is the scope of the IPR, and then one is the method.

[00:15:02] Speaker 03:
In terms of the scope, I think Judge Chen explained it correctly, these are the facts of—or these are analogous to the facts of close-up.

[00:15:10] Speaker 03:
The question is here, is what did the petition mean

[00:15:13] Speaker 03:
When I kept talking about SST 1987's multi-layer silicon nitride.

[00:15:19] Speaker 03:
And they're arguing that it didn't mean that that's where you should learn the element from, but that's a question of what the petition meant.

[00:15:28] Speaker 03:
And the Supreme Court already said in closure what the petition means is something that

[00:15:32] Speaker 03:
The board is left to decide and this court cannot review that.

[00:15:36] Speaker 03:
That's the distinction about which particular statutory provision was raised to challenge it.

[00:15:42] Speaker 03:
I don't think is a meaningful one.

[00:15:43] Speaker 03:
I think Quozo raised the best one because the statute says the petition needs to explain things with particularity.

[00:15:49] Speaker 03:
That's trying to tell you what needs to be in the petition.

[00:15:52] Speaker 03:
Instead, I think they're using SAS, but SAS isn't the statutory provision, but it's more a broader notion of notice, but they had the notice, as Strachan noted, and then it's unclear exactly even what the statutory provision is.

[00:16:06] Speaker 03:
But the challenge is the same.

[00:16:08] Speaker 03:
What did the petition mean?

[00:16:09] Speaker 03:
And Screamfort said that's not reviewable.

[00:16:12] Speaker 02:
Regardless... What about our recent Conan Clique versus Google opinion?

[00:16:16] Speaker 02:
—We did take a look at what the Board did at Institution, and we concluded that at least a portion of the Institution was incorrectly instituted, and we ended up repudiating that ground of rejection.

[00:16:36] Speaker 03:
—So, as Your Honor noted earlier, this Court explained it was undisputed that— —Does the PTO disagree with that case?

[00:16:46] Speaker 03:
No.

[00:16:46] Speaker 03:
I mean, it's... No, I mean... So you agree with that case?

[00:16:52] Speaker 02:
Yeah.

[00:16:52] Speaker 02:
Okay.

[00:16:52] Speaker 02:
So we got that right.

[00:16:53] Speaker 02:
Yes.

[00:16:53] Speaker 03:
And the board went too far and Google... I mean, what's confusing there is this very same reference was in there as a knowledge of the art.

[00:17:02] Speaker 03:
uh... ground which this court accepted and as a separate ground which this court didn't and the briefing makes the point that's not so clear why that matters except for the fact that philips in that case was trying to argue knowledge of the art can't be used at all but ultimately the the petition came in but in that case it was this is what this court wrote it is undisputed that google's petition advanced only two grounds the petition did not allege the third so that's a sass case where

[00:17:30] Speaker 03:
We know what the petition means.

[00:17:32] Speaker 03:
Nobody's arguing about that.

[00:17:33] Speaker 03:
And the question is, can the board leave the petition?

[00:17:36] Speaker 03:
And this court said, and the Supreme Court said in SAS, no, the petition means what it means.

[00:17:41] Speaker 03:
And so, sorry, this petition defines the scope of the proceeding.

[00:17:47] Speaker 03:
And then you, board, cannot leave that petition.

[00:17:51] Speaker 03:
But it's not a fight about what the petition means.

[00:17:53] Speaker 03:
Here we have a fight about what the petition means.

[00:17:55] Speaker 03:
And so that's what's unreviewable, under quotes.

[00:18:01] Speaker 03:
But regardless, in this case I think there's good evidence that the petition actually meant to include SST.

[00:18:10] Speaker 03:
So starting in the background, this court says it was known in the prior art, I mean sorry, the petition says it was known in the prior art to form multi-layer silicon nitride etched stop films.

[00:18:20] Speaker 03:
And the first thing... APPX 81.

[00:18:23] Speaker 03:
This is in the background.

[00:18:24] Speaker 03:
And it starts with

[00:18:27] Speaker 03:
SST 1987, and then it goes to the Watatani, which is also one of the petition bases for learning multilayer edge stops, but it wasn't the only one.

[00:18:40] Speaker 03:
My friend is certainly right that there were a lot of arguments that were based on the Watatani itself disclosing multilayer, and I think the next appeal will discuss that, but that wasn't the only basis Micron raised.

[00:18:51] Speaker 03:
And then going to the argument itself, the petition writes, the integrated circuit was known, multi-layer circuits were known.

[00:19:00] Speaker 03:
It says it was taught in Watatani, and then it goes, such multi-layer silicon nitride layers were well known in the art.

[00:19:07] Speaker 03:
Indeed, in its normal operation, the SST 1987 did just that.

[00:19:11] Speaker 03:
So that's a separate place.

[00:19:13] Speaker 03:
So even if we get to the merits, if that's reviewable,

[00:19:16] Speaker 03:
The fact is, the petition did explain it.

[00:19:18] Speaker 03:
I think, on point, there's a case that they cite, which is Arioso, where the background section of the petition discussed the reference.

[00:19:27] Speaker 03:
I think there was exhibit 1010.

[00:19:29] Speaker 03:
And then the board said, it was unclear what the board was doing with that.

[00:19:33] Speaker 03:
But this court wrote, to the extent you're going to ignore it just because it wasn't formally laid out in the grounds, we're not, we're going to, that's reversible error.

[00:19:41] Speaker 03:
So go back and tell us what you meant to do there.

[00:19:44] Speaker 03:
And that's at best what they have here, that there was a very long discussion of what SST 1987 meant.

[00:19:51] Speaker 03:
In terms of the merits, this seems like a standard KSR case.

[00:19:55] Speaker 03:
We have a reference that tells you that using multiple layers helps create uniformity, it helps avoid pinholes,

[00:20:05] Speaker 03:
And then it would just be obvious to apply that where you had a single layer.

[00:20:10] Speaker 03:
You might want those benefits to avoid a pinhole through which the etching chemicals could come.

[00:20:16] Speaker 03:
So it's just the standard KSR case.

[00:20:20] Speaker 03:
I don't have anything further.

[00:20:22] Speaker 03:
So I'm happy to cede back my time.

[00:20:24] Speaker 03:
I don't want to cut off any questions.

[00:20:26] Speaker 01:
That's fine.

[00:20:27] Speaker 01:
You don't get any demerits for that.

[00:20:30] Speaker 01:
Mr. Maloney, you've got a minute and 17 seconds to chip away at the patented petition.

[00:20:37] Speaker 00:
Thank you very much, Your Honor.

[00:20:38] Speaker 00:
I would direct the Court to Appendix at page 42, which is the final written decision, where the Board tells us that the institution and final written decision rely on SST 1987 in the same way as it was relied on in the petition.

[00:20:53] Speaker 00:
as evidence that multi-layer silicon nitride layers were known in the art.

[00:20:58] Speaker 00:
Not edge stops.

[00:20:59] Speaker 00:
There's nothing in SST having to do with edge stops.

[00:21:02] Speaker 00:
And consistent with that, as I pointed out earlier, the final written decision at Pedics38 says there is substantial evidence that it was known to use multi-layer silicon nitride layers as an edge stop layer in a damascene structure.

[00:21:17] Speaker 00:
That's the relevant element.

[00:21:19] Speaker 00:
And the evidence that the board cites at Appendix 38 is Yoda and Dr. Fair's reply declaration regarding Yoda.

[00:21:28] Speaker 00:
So this is clear, even though the board would not acknowledge that it had created a new ground, the substance of what it did makes that quite clear.

[00:21:38] Speaker 00:
In terms of the case law and reviewability, it can't be the case that the board can avoid reviewability by this court by simply characterizing what it's done as an interpretation of the petition.

[00:21:49] Speaker 00:
SES makes clear that CUSO does not limit, and in fact CUSO makes clear, it does not limit this court from reviewing actions by the PTO that exceed its statutory authority.

[00:21:59] Speaker 00:
Thank you very much.