3D InCites Members Reveal Breakthrough Technologies at ECTC 2025

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
This episode of the 3D Insights podcast is sponsored
by the IEEE ElectronicComponent Technology Conference,
Organized by the IEEEElectronics Packaging Society.
Ectc brings together the bestin packaging components and
microelectronics systems,science, technology and
education in an environment ofcooperation and technical

(00:20):
exchange.
Learn more at ECTCnet.
Hi there, I'm Francoise vonTrapp, and this is the 3D
Insights Podcast.

(00:43):
Hi everyone, this week we arerecording live from ECTC 2025 in
Dallas and it's the 75thanniversary of this event, and
there are more than 25 membersexhibiting and presenting from
3D Insights, so we've asked someof them to join us on the
podcast to talk about what ECTChas meant to their companies
over the years and then to learnabout what's going on at their

(01:06):
companies.
So first to join me is MonitaPau of Onto Innovation.
Welcome to the podcast, Monita.
Thanks for inviting me,Francoise.
Now, of course, you've been onthe podcast before a couple
times.
Yeah, In fact, I think we hadone episode where you talked a

(01:29):
lot about hybrid bonding andthat's kind of went exploding
off the charts as one of thebest listened to from ECTC.
Yeah, exactly, but if peoplehaven't listened before, can you
just give us a littlebackground your role in the
industry and at Ontu?

Speaker 3 (01:38):
So maybe I'll give a brief introduction of Ontu
Innovation first.
So we are an equipment supplierwith expertise in inspection
and metrology on both front-endof line activities as well as
back-end packaging, but we alsoparticipate in lithography for
advanced packaging as well.
So my role at Onto Innovationis strategic marketing, so I'm

(02:00):
responsible for helping set thestrategy for the company,
particularly for targeting theadvanced packaging in market.

Speaker 1 (02:09):
Okay, so now we were talking a little bit before we
got on the mics about yourexperience at ECTC.
So when was the first time youattended ECTC?
Where was it?
Can you remember where ithappened?

Speaker 3 (02:22):
Yes, so the first time when I attended ECTC is
probably about nine years agoand it was in Vegas.
So I think nobody can forgethaving ECTC in Vegas.

Speaker 1 (02:32):
No, I think I was there yeah.
Yeah, that was when they wereat the Cosmopolitan yes, exactly
Interesting place to have abunch of engineers and then,
like scantily clad guests ontheir way to the pool.
That's what I remember.

Speaker 3 (02:46):
The pool was on the other side of the conference
center Exactly.
You are all ready to go to work, but in the elevator there are
some people who just finishedpartying or, in their flip-flops
, ready to go party some more.

Speaker 1 (02:57):
They've discontinued the Vegas location.
But now we're here in the heartof Dallas, Texas, in this giant
conference center.
I got 8,000 steps in yesterdayand I didn't even work out.
So how do you think ECTC hasevolved over the years since
you've been part of it?

Speaker 3 (03:15):
So definitely a whole lot more people.
And also when I first joinedit's a little bit overwhelming
in a sense because there are somany parallel tracks of
technical informationpresentation, just keeping track
of which talk you want to go tothroughout the day and going
from conference room to aconference room.

(03:36):
It's overwhelming.
But I think over the years Ican definitely see the planning
has improved significantly.
It's much easier for an attendeeto go from one place to another
place, and I think thecommittee also provide a lot of
information right to theattendees as well, helping them
to navigate through theconference, because there are a

(03:56):
lot of things to learn Right,and if you plan it ahead of time
, then definitely you're goingto get a lot of valuable insight
from just attending the talks,as well as networking with all
the other attendees.
And, of course, then there'sthe exhibit hall, right, right,
and I believe that has also gonebigger as well, it has.

Speaker 1 (04:16):
Yeah, so one of the things that I've always thought
about ECTC is you know I attendIMAPS events, I attend SEMI
events.
Always thought about ECTC is.
You know I tend to IMAPS events, I tend semi events, and where
ECTC fits in that technologydevelopment area, I find that if

(04:36):
you want to learn about what'sbeing researched and developed,
you come to ECTC.
If you want to learn aboutwhat's being commercialized, you
go to semi, and if you want tolearn somewhere in the middle,
that's where kind of IMAPS fallsin.

Speaker 3 (04:48):
Yeah, I agree with that and actually I've been a
strong proponent for people Iwork with to come to join this
conference.
Whether you want to learn aboutadvanced packaging or you're
already really deep into it andyou want to know what is the
latest that's going to come, Ithink this is a great venue to
actually get those kind ofinformation.

Speaker 1 (05:08):
I think when I was first starting, I always thought
that something I would see herewould suddenly be in, you know,
high volume manufacturing inlike the next year.
And I've now learned over theyears that what you see here, it
could be 10 years or more, ifit even makes it.
So this is really.
I mean, maybe even earlier onwould be like maybe IEDM, but

(05:30):
this is really where we'relearning about what is coming.
But you were on a panelyesterday and it was what?
High density, ultra highdensity technology, technology.
Okay, can you talk a little bitabout that?

Speaker 3 (05:46):
So for that panel we're looking at what are the
challenges for ultra highdensity interconnect, whether
the supply chain is ready andalso what kind of application
product are driving the need forthis.
So I think the panelists,including myself, share our view
, including myself, share ourview.
So I think that goes hand inhand to other discussion in here

(06:06):
as well is there is a constantdrive for increased performance
and interconnect density andbandwidth.
So we're more looking into isthe supply chain ready?
We understand there's a demandfor it, but is the supply chain
ready to support it?
And I think that it's not justlooking at line and space, right
, because I think the industryhas been also talking about what

(06:28):
can enable high densityinterconnect so you can print it
, but what about the substrateor the carrier or the form
factor you're trying to havethis kind of interconnect on?
So all these things go hand inhand.
It goes from material are thematerial properties suitable for
that kind of thermal behavior?
And then whether the equipment,processing equipment, can

(06:50):
actually do the lithography, todo the patterning as well as
plating it with high reliability.

Speaker 1 (06:56):
Okay, Yep, so now you participated as a panelist in
that.
Did you have any other papersthis week from Onto Innovation
being presented?

Speaker 3 (07:04):
in that?
Did you have any other papersthis week from Onto Innovation
being presented?
From Onto Innovation?
Actually, one of our mycolleague is actually presenting
, as of now on, our lithographysolution targeting panel-level
fan-out not panel-level fan-out,panel-level packaging, as well
as advanced IC substrate.

Speaker 1 (07:19):
Right and I think sometimes people confuse.
We talked so much about fan-outpanel level packaging in the
beginning that I don't know ifwe've explained that all sorts
of wafer level packaging are nowbeing looked at in a panel
approach right because itstarted out being fan out that
was thought to be panelized, solike, can you talk a little bit

(07:42):
about what's driving the push topanel and the work that you're
doing?
Size, because we can see that,because we want to improve the
performance and put more memoryto support the logic die

(08:14):
functioning.

Speaker 3 (08:15):
We are actually making the packages bigger and
bigger and also the drive forco-package optics as well,
that's also another driver forlarge package.
So it makes more sense now toconsider, instead of doing
packaging on a wafer level form,what if we move it to a panel
level form with much higher areautilization versus a circular

(08:37):
wafer Right?

Speaker 1 (08:38):
right.
It's so funny because I've beenin this industry long enough to
have watched packages getsmaller and smaller and smaller,
and thinner, and thinner andthinner, and then stacked, and
even though they're stacked,they're not really that high,
because they're stacking withthin dye right, that was a
smartphone driver, right.
And now that AI is a big driverand we need these larger

(08:59):
package sizes, suddenly thepackages just have been
expanding over the years.
I'd love to see like, maybelike a time lapse of the
different package formats andstructures as the years went on.

Speaker 3 (09:14):
You're exactly right, because previously it was form
factor for mobile.
Everything has to be small,thin so that we can fit it into
the mobile device.
But right now, performance iskey and I think there's more
room for bigger packages as well.
Right for those applications.

Speaker 1 (09:29):
So how does a company like Onto Innovation adapt to
those changing market drivers?

Speaker 3 (09:35):
So we have been involved in panel level
applications, actually evenbefore this sudden search of
demand.
Okay, because we have ourinspection metrology system
developed for a panel levelapplication, as well as our
lithography, so I think we aredefinitely seeing it coming.
That's why we're ready and thatalso tied back into your

(09:57):
previous question ourapplication center of excellence
in our headquarters inWilmington.
So we see that collaborationacross the whole ecosystem is
important and that's why we haveinvested in setting up the
packaging application center ofexcellence at Wilmington to
support our customer as well asthe supply chain player as well,

(10:19):
so that we can accelerate thedevelopment of technology and
help the industry to adopt panellevel application much faster.

Speaker 1 (10:29):
And actually I visited the PACE Center last
October and if you want to seewhat's going on there, the
coverage is on 3D Insights andwe actually had a whole podcast
devoted to what was happeningthere.
So since then, how have thingsprogressed?

Speaker 3 (10:47):
So we have a lot of collaboration projects that are
ongoing, both requests from somecustomers and our customers, as
well as the supply chaincollaborators as well.
But one of the key things we'reseeing is ultra high density
interconnect.
We have a lot of engagement inthat sense, as well as glass

(11:07):
core substrate.
I think glass core substratehas been, I would say, probably
the buzzword in this currentconference.
Oh yeah, I mean, I guess thereprobably the buzzword in this
current conference room.

Speaker 1 (11:16):
Oh yeah, I mean, I guess there was a session
yesterday that was overflowing.

Speaker 3 (11:19):
Yes, yes, I couldn't even get in and I was standing
by the door and, yeah, it wasamazing and I wish they had have
set that up in a much largerconference room, because I think
that is the hot topic right now.

Speaker 1 (11:32):
Well, it's so funny though, because every year,
thinking back right, I rememberI have pictures of the
overflowing rooms.
Unfortunately, I wasn't hereyet when that was happening, so
I didn't get that photograph ofpeople standing in the hallway.
But there was the year where itwas fan out panel level
packaging.
I remember being in San Diego,everybody falling into the
hallway.
Hybrid bonding has been anotherone, and these continue to be

(11:56):
really hot topics, but there'salways like the darling, and I
think this year the glass coreversus RDL was one, and I'm
guessing co-packaged optics waspretty full yesterday.
And I bet you, next year that'sgoing to be the one that's
dumping into the hallway.
That's my prediction.

Speaker 3 (12:16):
Yeah, yeah, I think there are a lot of activities
along that side right now.
We're definitely seeingcustomers are going to adopt it
much sooner, right, okay?

Speaker 1 (12:26):
Well, it's been great talking to you, as always.
Thanks for joining me.

Speaker 3 (12:29):
Thanks for inviting me here joining me.

Speaker 1 (12:39):
Thanks for inviting me here.
So my next guests are CharlesLee and Mark Gerber of ASE, and
ASE is a platinum sponsor ofECTC 2025.
Welcome to the podcast, guys.

Speaker 5 (12:45):
Thank you.
Thank you for having us.

Speaker 1 (12:48):
So ASE is pretty much all over the place this week,
and here we are the 75thanniversary of ECTC.
Do you guys remember your firstECTC?

Speaker 2 (12:59):
Gosh.
For me it was probably in theearly 2000s.
I don't remember the exactfirst one, but it's been a while
.

Speaker 1 (13:05):
It's been a while, and were you at AAC at the time?

Speaker 2 (13:08):
No, no, at that time I was probably at Motorola, okay
, nxp today.

Speaker 1 (13:14):
Yep, right okay.

Speaker 5 (13:15):
And Charles.
So for me this was a long timeago.
My first ECTC was in 1995.
Oh wow, and that was in LasVegas.
And so this year it will be my30th year with ECTC.

Speaker 1 (13:29):
So how has it evolved and changed since those early
days?

Speaker 5 (13:33):
Well, definitely the size.

Speaker 1 (13:43):
The size, the dynamics and also the topics,
and so you know, this yearthere's been a lot going on, a
lot of discussion about AI andabout reducing energy and, mark,
I know earlier this week youwere on a panel on co-packaged
optics and how co-packagedoptics is going to potentially
help the AI energy issue.
So can you talk a little bitabout that panel and some of the

(14:04):
key takeaways and what ASC hasin that space?

Speaker 2 (14:07):
Yeah, sure.
So the panel specifically wasaround materials and the
importance of differentmaterials in the package optic
space.
Back to your point about power.
So when you look at today, thecloud compute basically
comprises of both the serverside and the processing side.
For the AIML there's a hugechallenge on both the thermal

(14:31):
side due to power and how tomanage this, and the package
plays a very key component intothe overall picture of this.
Yeah, so the module side we'removing from the pluggable
modules to kind of the nextgeneration on both bandwidth and
power.
There's a need for a solution,and so Copac Adoptix kind of

(14:54):
fills that space.
So the discussion was aroundyou know what are the challenges
with Copac Adoptix and how arethe materials playing such a key
role in providing differentsolutions?
To be able to scale that theswitch from the module side as I
mentioned, the pluggable overto CPO.

Speaker 1 (15:13):
It's measured in a picogil per bit type of metric
that tells you kind of thebenefit, and so for each of
these solutions, the main thingthat you're going to find is the
components are being placedcloser together in the solution,
and that's again what CopackAdoptics is bringing the
components closer together toget a better performance metric

(15:37):
out of it I was in that paneldiscussion and one of the things
I was found interesting andmaybe you can talk about this a
little bit is that one of thepanelists called for
standardization of co-packagedoptics technologies or processes
, and I always felt like this isreally what the secret sauce is
for a company, and I understandwhy some companies need it to

(16:01):
be interoperable and thatstandardization helps with that,
but how do companies like ASEfeel about that, when this is
really what differentiates youfrom your competitors?

Speaker 2 (16:12):
This has been one challenge for many years for the
photonics area, because thephotonics die is custom pretty
much to every customer.
So the way that they lay thatout it's not always easy to
package in the same way becausethey're all different.

Speaker 4 (16:26):
Right.

Speaker 2 (16:27):
So the idea is there's something what we call
an optical engine that containsseveral components that are key
to the overall system, and if wecan define some level of
standards of how the dye needsto be laid out for both test and
other areas for integration,then this provides at least a

(16:48):
stepping stone into the areawhere we might have some
consistency in the industry, andthat's one area, and test will
be covered, of course, in thatand other areas, which is really
important, I see.

Speaker 1 (16:58):
Okay, so what are ASC's offerings in co-packaged
optics?
Because that was a bigannouncement earlier this year,
right, I think in April?
Yeah.

Speaker 2 (17:07):
Yeah, so back in, actually 2023, we made an
announcement on a POPpackage-on-package and it
actually was the introduction tothe optical engine integration.
And then, as you mentioned,last quarter we made an
announcement on a CPO that wasan 8 plus 1, eight optical
engines plus the ASIC to kind ofshow the way that this roadmap

(17:33):
and structure is moving forward.
So, yeah, I think when it comesto the package side, there are
many variations because, again,customers are doing things a
little different and there'sdifferent performance metrics,
but we're definitely seeing aroadmap being formed on both
those sides the optical engineside and the CPO side.

Speaker 1 (17:52):
Okay, Now, Charles, there was some other big news
from ASC this week on your FocusBridge technology and that
there's been an enhancement tothat.
So can you both talk a littlebit about what Focus Bridge is
and what the enhancements willbe and why we had to do?

Speaker 2 (18:10):
them.
I'll just start out and thenCharles can talk about some of
the enhancements.
The Focus Bridge it's anadvanced high-density fan-out
solution that allows forembedding or the integration of
other components.
Traditional Focus the Focusplatform.

Speaker 1 (18:26):
It's a fan-out chip on substrate, that's correct
Fan-out chip on substrate.

Speaker 2 (18:30):
So basically what it is is it's a fan out.
Integration of the die Could be, you know, it could be an HBM
with the processor, could be acouple network die that gets
split up.
There's different ways, but itallows the integration of that.
And then that's packaged withagain some essentially fab based
technology that allows you thento integrate that onto a

(18:52):
substrate like a single chip.
And that's the focus.
So the variation of the focusbridge is that integration layer
in the fan out allows you toembed.
So basically we're moving morein the vertical direction and so
that integration really opensup a whole nother level of
performance for differentapplications.

(19:13):
So the announcement this weekI'll let Charles talk about, but
essentially that's kind of thebasic package.

Speaker 1 (19:17):
Okay, and so what's new with the Focus Bridge?

Speaker 5 (19:19):
So what is new in this press release, which is
following our first announcement, is that this time we have
integrated TSV into the BridgeDie, TSV into the bridge die,
and with this advantage of theshorter path we are able to then

(19:41):
enhance the power delivery andalso for the lateral
connectivity to get the higherbandwidth that is needed for the
next generation AI and HPCrequirements.

Speaker 1 (19:51):
So the bridge is an embedded die and is it a certain
functionality, functionalitylike, is it an embedded memory
or is it a logic?
I mean, what's the bridge?
And then you're using a throughsilicon via to shorten the
interconnect.
So is it a stacked package?

Speaker 5 (20:09):
So, essentially, the bridge is nothing but
interconnection die, okay, andso, in this particular case, we
have a true verticalinterconnection, like a true
silicon via, and so this willtake care of the direct access
between the bottom and the topdie and, as well as this, will

(20:29):
shorten the connectivity pathbetween the inter-die
interconnections.

Speaker 1 (20:35):
And so what is the target application for this
technology?

Speaker 5 (20:39):
We foresee this application will be very useful
in AI power data centers, whereyou need to have a lot of
computing power as well as afast bandwidth.

Speaker 1 (20:50):
Right.
And then the shorterinterconnect makes it so that to
achieve that, it needs lesspower to transmit the signal.

Speaker 5 (20:58):
I mean basically it has proven that it will reduce
our resistance by more than 72%and also the inductance by 50%
and we have shown potentiallythis can improve the power loss
reduction by three times.

Speaker 2 (21:17):
And, just to add, one of the key things about this
announcement that's reallyunique is that when you look at
solutions today where you have abridge between the high
bandwidth memory and the ASIC orthe processor in this case,
it's going to be more gearedtowards the GPU side, or the
processor, in this case, it'sgoing to be more geared towards
the GPU side.
That direct connection is oneof the main reasons why you can

(21:38):
have reduced number of layers ofRDL, which again, from a cost
standpoint, is better.
But with this announcement,with the through silicon via on
the bridge, now you're notlosing space where you can have
a power feed through.
Now you're not losing spacewhere you can have a power
feed-through, which is again,from both power delivery and law

(21:59):
standpoint, it provides a muchbetter solution.
So it's a balance between nothaving to use a full 2.5 large
silicon interposer versus theseagain small bridge chips that
provide essentially the sametype of features.
So it's exciting news for us.

Speaker 1 (22:17):
Yeah, excellent.
Well, congratulations on that.
It's been a busy week for ASAall around.
Keep saying you had the lunchpresentation, you were involved
in the panel and there were someother papers presented.

Speaker 2 (22:35):
Yeah, we had Lee Hung was also on a panel, and we
have CP Hung as well, and wehave several different papers
that were also presented thisweek actually a couple today so
a lot of good things going on.
You know, ASC is all over theboard when it comes to different
types of technology and movingthe industry forward, so it's
been an exciting week.

Speaker 1 (22:54):
We'll keep watching and we're excited to see what's
next, and thank you for joiningme today.

Speaker 5 (23:00):
Thank you.

Speaker 1 (23:06):
So I am speaking with Tim Olson from DECA and Scott
Sikorsky from IBM.
They've had some big news thisweek here at ECTC, so we're just
going to dive in.
But before we get started, guys, can you each just tell a
little bit about yourselves andyour role at your companies?
Tim, why don't you?

Speaker 6 (23:25):
Sure, I'm Tim Olson, founder and CEO of DECA
Technologies.
We're now a 15-year-old companyand excited to be working in
the world of advanced packaging15.

Speaker 1 (23:37):
They grow up so fast.

Speaker 7 (23:40):
And I'm Scott Sikorski.
I have responsibility forglobal business development in
the chiplet and advancedpackaging area for IBM.

Speaker 1 (23:47):
Okay, and IBM's way older than 15.

Speaker 7 (23:50):
About 115.
There we go.

Speaker 1 (23:52):
Okay, speaking of years, we're at the 75th ECTC,
so I was trying to remember whenmy first ECTC was, because I
can't remember if I came firstwhen I was with Advanced
Packaging.
I know that I have been comingever since 3D Insights, so that
would be 2009.
When was your first ECTC?

Speaker 6 (24:13):
Very similar, I think in that same time frame, and it
was at the Dolphin and Swan atthe.

Speaker 1 (24:18):
Walt Disney.

Speaker 6 (24:19):
World.
I love that venue.
Oh, I don't.
You didn't like it.

Speaker 1 (24:24):
No, I didn't.
There was one night where Iwent into my hotel room and I
heard drip, drip, drip, drip,drip and there was water
dripping on my pillow from theceiling above and they had to
shift my room no.
I was not a fan.
I'm glad we're not thereanymore, but anyhow, and Scott,
when was your first one?

Speaker 7 (24:42):
Probably even a little earlier than that
actually.
So I'm a lifetime packaging guy, including my undergraduate and
graduate work.

Speaker 1 (24:49):
So you were at ECTC way before you were even at IBM.

Speaker 7 (24:53):
I've always been at IBM.
I've always been at IBM.

Speaker 1 (24:54):
You've always been at .

Speaker 7 (24:55):
IBM.
Yeah, so I did my graduate work, went to IBM, was there 20
years, went to Stats and theOSAT world for 10 years and then
came back.

Speaker 1 (25:05):
Okay, so there was that.
There was a gap.
So there was 10 years where youwere at ECTC with a different
company.
Correct, right.

Speaker 7 (25:17):
So how has ECTC evolved, do you think, from then
till now?
Oh, tremendously.
I mean, it's gone from being arelatively modest conference to
being the centerpiece in theworld for packaging in terms of
gathering the global communityComing into this role.
It was a very easy discussionwithin IBM to be platinum
sponsor, and all of our keyoriginal papers in packaging are

(25:40):
done here, without exceptionactually.
So it's a very important venuefor us to see what's happening
in the packaging world and fornetworking.

Speaker 1 (25:50):
Right 2,500 people this year.
That's like another recordEvery year.
They keep breaking it.
Right.
So I guess we were all rightabout packaging and how
important it is Maybe lucky.

Speaker 6 (26:01):
I think we're smart, I know.

Speaker 1 (26:03):
So let's talk about your big news.
What's the announcement thatDECA and IBM shared this week?

Speaker 6 (26:10):
So we have entered into a TTLA, a technology
transfer and license agreementwith IBM, and we're super
excited about it.
It's clear to everyone in theworld that IBM has this storied
history in advanced packagingand advanced technology in
general.
We've been working with IBMalready for over a year and

(26:30):
we've met all the key people inthe factory sense as well as in
the R&D sense, and I'd saythere's several dozen amazing
technical talents at IBM thatare going to help implement our
technology in the Bromontfactory.
And then what we're reallyexcited about is IBM will be
building future capabilities ontop of the foundation that we

(26:52):
bring.
So we've announced this newrelationship.

Speaker 1 (26:55):
So this is all happening in Bromont.

Speaker 7 (26:58):
Yes.

Speaker 1 (26:59):
Why there?

Speaker 7 (27:00):
Just to take a step back before we get into why
there, why IBM is doing thisanywhere.
So, if you look at IBM, reallya key focus for us now is to
help lead our clients on theirAI journey.
So we're very, very focused onAI, and the unique perspective
we're bringing to our clients isfull stack.

(27:21):
So what do we mean by fullstack?
So that's everything from thetop layer software, where we
have our Watson X, all the waydown to what should the next
generation transistor look likein terms of our Albany Nano
Center and being the first totwo nanometer gate all around,
et cetera.
What we found, though, is toreally optimize AI workloads.

(27:42):
The packaging piece has becomereally critical, and IBM has a
long history, as Tim said, inpackaging, but as we look at AI
moving towards chipletarchitectures and you look at
the standards that that'sdriving, whether it be related
to HBM or UCIE, very quickly youfind that the packaging is the
critical bottleneck and you needa horizontal interconnect

(28:06):
fabric.
And then we look at well, whatare the options?
Today you have siliconinterposer.
We know that's running out ofsteam in terms of physical
scaling as well as otherconstraints, and as we looked at
the market and we could developsomething ourselves.
Time to market is very critical.
Ai is moving incredibly fast.
You see what's happening withlarge language models, with

(28:28):
generative AI and again, we'relooking at this through the
perspective of our clients, themajor companies across the world
, and we've got to move fast.
So time to market is importantand you know, we felt that
leveraging the many years ofexperience and the fact that
this technology that DECA has isin the market and I personally
have looked at this for 10 years, maybe you know through

(28:49):
different lenses, 10 years,maybe you know through different
lenses and felt like this isthe best opportunity to bring a
very high quality state of theart, moving towards leadership
product to our end clients.
So that's why we did it and, asTim said, we spent well over a
year pressure testing thetechnology with our best

(29:11):
technologists across IBMResearch and our infrastructure
group.
Me being convinced is one thing,convincing IBM corporations a
different thing.
So Tim was, I would say,grilled pretty heavily and it
was very patient with us and inthe end, it was clear that this
is the path we should take Now.
Bromont has been IBM's centerof competency for first-level

(29:33):
packaging for 50 years.
We've got a very large facility, 850,000 square feet.
We've got a very seasonedworkforce approaching 1,000
people, with an average of 20years tenure.
It was a very easy decision tosay that this really is the
place because we want to scaleit.
This is not a prototypingfacility.
We want to do mass productionin a secure North American flow

(29:59):
for, you know, largely Americanclients and that's you know.
It fits into our NortheastCorridor.
The last time I had the pleasureof being with you was to talk
about the Northeast Corridor andwhy that is important and this
really puts an exclamation pointon that in terms of the
capability that we're bringingto our clients in the corridor.

Speaker 1 (30:19):
We're talking about DECA's technology.
You have many technologies.
This technology transferlicense agreement is for the
M-series technology also withadaptive patterning and
everything that goes with it.

Speaker 6 (30:33):
Yes.

Speaker 1 (30:33):
So M-sereries was not your flagship product, it was
the second product, right, or itis your flagship?

Speaker 6 (30:40):
It's always been the flagship.
We started the factory runningconventional wafer-level
packaging to prove our companyas another capable, high-quality
, short cycle time, good valuesupplier.
But we did that only for thatpurpose to create a business.
But in the initial founding ofDECA my goal in the end was to

(31:01):
create the world's best fan-outtechnology.
I had some experience in my pastroles at Amcor and other
opportunities, so we created andlaunched the M-Series fan-out
in 2013, so it's over 10 yearsold.
Created and launched theM-series fan-out in 2013, so
it's over 10 years old.
And it really hit the big timein the 2017-2018 time frame

(31:22):
because Qualcomm adopted thetechnology in a big way.
They invested in DECA.
They asked us to teach ASE inTaiwan how to build it.
So we did our first TTLA withASE in Kaohsiung and today ASE
continues to run very, very highvolume of the technology.
It is the highest volume fanout technology today in the
industry over 7 billion devices.
If you have a latest iPhone,latest Samsung phone, you

(31:46):
definitely have a handful ormore parts in there with the
DECA M-series technology wrappedaround them.

Speaker 1 (31:52):
Okay, and so you have license agreements with a bunch
of different companies that aremanufacturing Skywater is also
on that and you've got IBM andyou've got ASC.
So it's no wonder that it's oneof the most high-volume
products on the market today.
Is there any like issues withcompeting companies licensing

(32:16):
the technology?

Speaker 6 (32:17):
Well, obviously we'd like every company to have their
own unique value, and I thinkIBM clearly has the greatest
differentiated value of anyoneelse.
Today they are going to becoming up with the most advanced
version of our technology thathas not yet gone to production,
which is the M-Fit, which is avery high-end interposer using

(32:38):
the M-Series platform, and soIBM will be helping us take off
the rough edges in the end ofdevelopment to bring that into
high volume.
So we're very much lookingforward to that help and that
will position IBM, I'd say atleast a number of years ahead of
anyone else.
So we're excited that they havethat business opportunity.

(32:59):
We're really excited about themhelping and making the
technology robust and highvolume for the highest
technology packagingapplications in the world.

Speaker 1 (33:08):
So MFIT F-I-T stands for.

Speaker 6 (33:10):
M-series fan-out interposer technology.

Speaker 1 (33:13):
Fan-out interposer technology.
Fan-out interposer technology.
Are we using organicinterposers?

Speaker 6 (33:17):
It's an interposer itself, so it may or may not
have a substrate.
Typically we'll have anadvanced build-up substrate
underneath it, but not always.
But in the interposer sensethat's how it works.
And the interposer is theM-series, it's the molded
organic, and when you sayorganic people think substrate
or PCB.
But mold compound structuresare also organic.

(33:40):
So, this is a organicinterposer built from the
M-Series fan-out.

Speaker 7 (33:45):
Yeah, with RDL, fine-pitched RDL.

Speaker 1 (33:47):
And so this is an exclusive with IBM's, the only
one who's going to be buildingthe MFIT.

Speaker 6 (33:54):
For now, yes.

Speaker 1 (33:56):
From IBM's perspective.
Do you already have clientsinterested in that series?

Speaker 7 (34:02):
Yeah, this is a big undertaking.
As Tim said, it's not proven inthe market.
Hence we did the very extensivepressure test technically, with
a lot of input from IBMResearch.
We think one of the reasons whywe have a good chance of being
a success is, as Tim said, thecompany DECA has quite a bit of
experience with foundationaltechnology that's in the market

(34:24):
today.
This definitely takes it to awhole different level in the
market today, this definitelytakes it to a whole different
level and we think, with theexpertise we have in IBM
research, that you know, wereally have a good opportunity
to differentiate a solution toour end clients.
And, as he said, you know thisis basically an interposer, so

(34:44):
the die will sit on top of thisand it creates a very attractive
horizontal integration fabricfor those die to talk to each
other.
So is this a chiplet structure?
Yes, exactly.

Speaker 1 (34:54):
So when we hear about different companies have their
different terms for theirinterposer technology on a
chiplet, it's all some sort offabric.

Speaker 7 (35:02):
Correct.

Speaker 1 (35:03):
It's an interconnect fabric, so there's active
components in the interposer.

Speaker 7 (35:12):
Not necessarily Not that we rule that out in the
future, but the activecomponents are on top.
That creates kind of a datasuperhighway that you can't do
in a base substrate at thispoint, and so that's really the
advantage of the interposer.

Speaker 1 (35:25):
Okay, so how long do you think it's going to take to
bring this up to production?

Speaker 7 (35:29):
Yeah, great question.
So we've already started thework.
Obviously you need a dedicatedcleanroom environment.
We've got space at IBM Bromontthat we're now retrofitting to
have the appropriate cleanroomenvironment.
So that work has started.
We're in the final, late phasesof tool selection.
Essentially, we'll be bringingin tools next year, bring up the

(35:52):
process late 26, early 27, andwe intend to be ready for
production in late 27.

Speaker 1 (35:58):
And the tools are the secret sauce still for.

Speaker 6 (36:00):
DECA.
They're part of it.

Speaker 1 (36:02):
I mean because I've never had you actually name.
You know, when you started offwith DECA, it was very much part
of the secret sauce, becauseit's panel-level packaging
before everybody else wasstarting to do panel.
And it had been technology thatwas formerly used in solar

(36:23):
right.

Speaker 6 (36:24):
A few pieces, yeah.

Speaker 1 (36:25):
So has that changed.

Speaker 6 (36:26):
No, there's still a few process steps that actually
were born of SunPower, one ofthe number leading solar
companies in the world, backthen.
So the equipment is part of thesecret sauce and the reason we
treat that very confidentiallyis it's a tremendous amount of
development to get the machinesto run the various formats, from
different size silicon wafersto the plastic panel wafers and,

(36:50):
in the future, of course,square and large square things.
So yeah, so one interestingthing about the equipment set,
the DECA.
There's multiple suppliersinvolved and in most process
steps there's at least a couplesuppliers that we've developed
that can produce the capability.
But if you look at whathappened with our first
implementation at ASE, throughour process we basically have

(37:12):
the suppliers remove all theirIDs from the tool.
They paint the tools either acertain off-white color or
they're stainless steel.
So when you walk in any factorywith that kind of technology
you will not know where themachines came from, including
ASE, who has hundreds andhundreds and hundreds of these
machines.
We keep that confidentialbecause equipment suppliers are
a source of folks pulling tradesecrets and other know-how.

Speaker 1 (37:36):
Yeah, so I was wondering if that was still the
case.
So you're building up a cleanroom with DECA-authorized tool
sets that would be differentfrom anywhere else in your clean
room Correct.
Or if they're not different, wedon't know it because they're
white-labeled.

Speaker 7 (37:53):
Also correct.

Speaker 1 (37:54):
Okay, do they all say DECA on them?

Speaker 6 (37:56):
No, no, no.
So DECA is not an equipmentcompany and we don't want to be.
We want to provide advancedpackaging technology and, as
mentioned, equipment is a keypart.
But there's materials, there'sprocesses there's design systems
, design rules.

Speaker 1 (38:08):
Right right.

Speaker 6 (38:09):
So it's a whole integrated solution.
But without the equipment, ofcourse, it can't be produced.
So the equipment is what is akey piece.

Speaker 1 (38:16):
All right, well, is there anything else that you
want to add?

Speaker 7 (38:20):
I think we've covered the key points.

Speaker 1 (38:21):
Okay, well, we'll be watching, and super excited for
you guys and you know, just keeptalking about this.

Speaker 7 (38:27):
Yeah, stay tuned for future announcements.
That's all I can say.

Speaker 1 (38:30):
Okay, thanks so much for joining me.

Speaker 4 (38:34):

Speaker 1 (38:38):
So my next guest is Roland Rittenmeyer, now with
Schmidt Group, and he waspreviously with Evatech, so he's
been on the podcast before, sowhy don't you update us on your
new role and the companies thatyou're at?

Speaker 4 (38:51):
Oh yeah, thank you for having us, or having me,
here today.
My new role at Schmid is ChiefSales Officer, so I joined
Schmid in early March as CSO.
So Schmid is producing wetprocessing tools, but also
plasma deposition and edge toolsfor panel-level packaging.

Speaker 1 (39:12):
So has it come from other industries?
Was it in flat panel display orsomething like that?

Speaker 4 (39:17):
Schmid is about 160 plus years old and the roots of
the company and the processesare basically printed circuit
board and IC substratemanufacturing.
The processes are basicallyprinted circuit board and IC
substrate manufacturing, butalso in our company's history we
did a lot of photovoltaic panelprocessing, so panel-level

(39:37):
packaging is part of our DNA.

Speaker 1 (39:40):
All right, that makes sense.
It's a good transition fromphotovoltaics into advanced
packaging space.

Speaker 4 (39:47):
Yes, we do see now, let's say, pcb, ic substrate,
blending in also glass, glasscore with organic buildups, so
we can benefit from our processtechnology in glass processing,
whether that's cleaning oretching of the glass or
metalizing glass as such, plusour huge experience in PCB and

(40:11):
IC substrate, where we handleorganic load, where we handle
organic buildups, where we dothe different processes of any
kind, horizontal or vertical,for different kinds of
applications.

Speaker 1 (40:22):
Okay, so we're here at ECTC.
It's the 75th anniversary, soin honor of that, I'm asking
everybody when was the firsttime you were at ECTC?

Speaker 4 (40:32):
It is years back, I presume it's about 10 years back
and it was in San Diego.
I remember that.
And first time, ectc I mean.
You see, for me it was veryimpressive seeing ECTC as the
lead conference for advancedpackaging on the planet.
I would say there is no suchevent like ECTC.

(40:53):
The insights you gain in somany different fields, from
photonic application topackaging application, is
phenomenal.
It's even for one personjoining ECTC it's not enough.
You need to be at least three,four people to cover the
interest.
And for me ECTC is the leadconference.

(41:14):
So I was introducing this toSchmidt Group and now my
colleague Dian Chang and I arehere to scout, even for next
year's even exhibition, handingin papers, participating also by
contributing next year, so itwas your first time here with
Evatech.
My first time here was withEvatech.
That was San Diego.

Speaker 1 (41:36):
Okay, so you were there with Evatech before, so
now you're here with Schmidt,and this is their first.

Speaker 4 (41:40):
It's the first time for Schmidt to engage with ECTC.
We are three people now hereOur local sales is joining and,
as I said, my colleague Deon andI from the C-Level team and,
yes, I think we are alreadyconfident right now that we want
to do more next year.

Speaker 1 (41:56):
Yeah.
So why did you choose ECTC tointroduce Schmidt to the
industry?

Speaker 4 (42:05):
Well, ECTC is the meet and greet of the advanced
packaging industry.
There is some private platformsevolving.
There are some semi-eventsevolving.
We are already participating inthe Semicon shows, exhibiting
there or contributing to that.

Speaker 1 (42:20):
How about?

Speaker 4 (42:20):
IMAPS.
Imaps is also evolving verynicely and this is another event
we will contribute and show up,but ECTC is still the leading
conference.

Speaker 1 (42:32):
Especially for developing technologies.
It's kind of thepre-competitive research space
where you really learn aboutwhat's up and coming.

Speaker 4 (42:44):
Schmid in some cases is a pretty disclosed company,
so we develop a lot of thingsbut we do not really talk about
it in public.
We have a lot of know-how anddevelopments done for Glasgow
substrates with organic buildupsand this is to be published

(43:04):
soon and we will hand in paperthen, especially with this focus
, most likely for next year'sECTC.

Speaker 1 (43:11):
So have you attended some of the sessions here?

Speaker 4 (43:13):
Absolutely yes.

Speaker 1 (43:14):
Okay, so what are you hearing as far as some of the
key takeaways from this year'sevent, maybe in comparison with
previous years that you've beento?

Speaker 4 (43:21):
Well, we have already seen in previous years that
co-packaged optics and opticalsignal transfer will be a major
functionality the industry needsin the future.
I was amazed how much power isput into a server rack of AI
computing.
We talk about 0.5 to 1 megawattelectrical power for one server

(43:42):
rack and then you need to movethe bits and bytes from one rack
to the other, from one CPU tothe other CPU.
So co-packaged optics, opticalsignal transition, will be one
of the key things that will movethe packaging industry in the
future.

Speaker 1 (43:58):
And that will help reduce the power consumption of
the AI chips.

Speaker 4 (44:04):
For certain distances , yes, so the shorter distances,
I think, will still remain oncopper, but for longer distances
you will need the fiber then totake over the functionality of
the signal transfer.

Speaker 1 (44:15):
I can see that.
So that's taking the signal offfrom one package to another.

Speaker 4 (44:19):
then yes, even so yes , Even on package-to-package or
server-to-server communication.

Speaker 1 (44:25):
Right, right, Because within the package you're going
to want the copper, Do you see?
Don't see optical interconnectswithin the package.

Speaker 4 (44:32):
Well, there is different approaches and
different beliefs in that.
I think that the opticaltransfer makes sense from a
certain length of transfer.
So we talk about severalcentimeters.
Several millimeters might stillremain on copper lines.

Speaker 1 (44:51):
Yeah, so what can we expect to see from Schmid next
year, here and just in thefuture in general?
As you break into this space,what's going to be your core
competency?

Speaker 4 (45:01):
Our core competence is clearly in the panel level
packaging field, so Schmid isnot doing anything on wafer
scale.
So we are a panel levelpackaging company by our DNA, by
our history, so you can expecta state-of-the-art panel level
packaging technology on wetprocesses, but also dry
processes like sputterdepositions, sputter edge deep

(45:23):
rectifier on edge in vacuum, avery, very novel approach for
glass core buildups, like adamascene process for glass core
and ABF buildup.
So we have a very novelapproach here, or an
IP-protected approach here forembedded traces, as we call it.
So it's an infinity board,gives you infinite possibilities

(45:46):
for horizontal and verticalinterconnects.

Speaker 1 (45:51):
Any other equipment company I know in this space
that is developing panel leveltools and processes started out
in the wafer but from what I'mhearing from you, Schmidt has
always been focused on panellevel.

Speaker 4 (46:05):
Yes, that's right.
So our equipment is designedfor panels since years.
We have never touched the waferspace, but so are others like
when we look into the material.
Let's pick anjinomoto finetechno right right they, they
are solely coming.
They're mainly coming from icsubstrate business they're very
limited footprint in wafer.

(46:25):
Also, when you look at someplating companies, let's pick
SensuScore, let's pick Atotecthey're also basically evolving
from PCB IC substrate into thepen-level packaging space.

Speaker 1 (46:38):
I see okay, orbotec maybe, which is now part of.

Speaker 4 (46:42):
KLA right, okay.

Speaker 1 (46:44):
So any company that started in the PCB space is just
having to develop finer-tunedprocesses to match what's needed
in the advanced packaging space.
Because it's a lot of the same.
I remember going to visit a PCBcompany and realizing that all
of the terminology, all of thetechnology was the same.

(47:08):
Realizing that all of theterminology, all of the
technology was the same.
It just was at a differentaspect ratio or higher.
They were talking millimetersinstead of microns.

Speaker 4 (47:15):
Precisely In the PCB or in the IC substrate industry,
you have a kind of let's callit a pyramid.
On the lower end you have theprice battle, where you have the
commodity products it'smulti-layer, it's simpler
products or more simple productsand you have the performance
battle on the very top notch ofthe of the ic substrate, pcb,

(47:35):
advanced packaging panel level.
Packaging products talk aboutminiaturization in terms of line
space, going towards fivemicron, two micron space,
talking about higher aspectratio through glass, via glass
core being part of the panel.
So this is kind of if you arean innovative company, if you

(47:56):
are a company which is based inEurope, based in equipment maker
, based in Europe, based in theUS, you have to push for the
higher end, you have to push forthe performance battle.
You will never win the prizebattle.

Speaker 1 (48:10):
Okay, well, it was great to have you stop by and we
look forward to speaking withyou next time.

Speaker 4 (48:16):
Thank you very much, Frances.

Speaker 1 (48:22):
So Simon McElroy.

Speaker 10 (48:23):
Howdy ma'am.

Speaker 1 (48:24):
It's always good to see you.

Speaker 10 (48:26):
Good to see you again.

Speaker 1 (48:27):
Here we are again, and it was the last time we
podcasted together on well, wedid an exclusive.

Speaker 10 (48:33):
Yeah, that's right.

Speaker 1 (48:33):
But then last year ECTC.

Speaker 10 (48:36):
We did.

Speaker 1 (48:37):
We introduced Liquidics to the world.

Speaker 10 (48:38):
Yes, we did.
Liquidics was about a month old, I think at the time.

Speaker 1 (48:42):
Well, the new brand?
Yeah, and we talked aboutinterconnectology.
So here we are at the 75th ECTC.

Speaker 10 (48:50):
Amazing.

Speaker 1 (48:51):
And one of the things I've been asking people this
week is when was your first ECTC?

Speaker 10 (48:56):
I think pre-millennial, but let's just
say 25 years ago.

Speaker 1 (49:00):
So out of the 75, you've been here for a third of
them.

Speaker 10 (49:04):
On and off, but mostly because I've been in and
out of the industry.
Oh, that's, true, so I getthese snapshots about every few
years.

Speaker 9 (49:11):
That's kind of cool.

Speaker 10 (49:12):
Yeah, it is interesting to see because I
mean gosh talk about exponentialgrowth in every part
technically and people and justinterest.
It's just huge.

Speaker 1 (49:25):
Do you think this is a testament not just to the
conference, but but to the factthat advanced packaging is
really it's a rising star?
It has been.

Speaker 10 (49:31):
I think it's 100% to it.
Why?
Because you know, even 10 yearsago, those of us that were out
banging the drum for packagingwe weren't so much banging the
drum because it's what we did wecould foresee that it was going
to be the bottleneck, so wehoped that the big semis would
show up and you know, we'relooking at their banners all

(49:54):
over the place right now.
Um, and now they're herebecause it's a pain point and
but it's.
It just lifts the level of theconversations hugely right, so
let's talk about liquid metalink again we talked about it
last year, but I think peoplealways need to be reminded.

Speaker 1 (50:11):
You know, you listen to so much in the middle and
maybe they didn't catch it lastyear.
So let's just briefly talkabout what it is that Liquidix
provides.
What's the solution?

Speaker 10 (50:23):
Okay, well, we call it PVD in a bottle, because it's
basically atomic metalizationof any substrate you can think
of, from FR4, printed circuitboard materials to Teflon, to
advanced substrates, to all theway today to glass glass
substrates and what we do is oursolution is in a bottle, it's

(50:44):
an ink, it's a bottle.
It's an ink, it's a liquid.
The two classic ways tometalize are vacuum processing,
dry processing, which isextremely expensive, as used in
the semiconductor industry, orplating.
Plating is used in printedcircuit boards, substrates, etc.
Which is pretty much 100%aqueous or water-based.
The challenge with water isit's just one solvent and most

(51:09):
of the substrates arehydrophobic.
They hate water.
You spend a lot of your timeputting in additives and trying
to change the fundamentals ofthe chemistry to make the
chemistry work.
Plating chemistries areubiquitously aqueous, so
water-based chemistries and allthe substrate materials are
typically water-repelling orhydrophobic.

(51:31):
So part of the game is to makethe metallization chemistry work
, you have to add all theseadditives.
We approach it from using anorganic solvent or a mix of
organic solvents, so it justopens up an infinite number of
possibilities of differentsolutions you can use to stick

(51:51):
to different substrates.
So, for example, when we applymetal to ABF.
We don't have to use harshdesmears, we don't have to do
all these preparation steps tomake the water-based chemistry
stick.
Same with glass we can justcreate a chemistry that adheres
directly to and tailor it to thesubstrate.

(52:11):
So we don't try to remove glass, we don't try to etch glass, we
just stick to the glass.
So we just concentrate on that.

Speaker 1 (52:18):
So there's no etching to create the traces.

Speaker 10 (52:23):
You don't have to roughen the surface.
There's so much work that getsdone in order to accommodate
either PVD, for example, oraqueous-based plating
chemistries, just because that'sthe traditional way of doing
things.
However, in other industries,for example just taking glass as
an example organics and glassare prevalent throughout, for

(52:46):
example, contact lenses.
They're treated with organicsolvents, but in the
semiconductor industry, it's aconservative industry.
As you know, there are ways ofdoing things.
So you know, the challenge forany startup and I love startups
is you do have to disrupt.

Speaker 6 (53:01):
Right.

Speaker 10 (53:02):
And disrupting a conservative industry is more
difficult, Right, but once youhit a big pain point and that's
why we have a huge group ofpeople here and a focus on the
next generation, which is glassthen you really start to get
attention, and that's wherewe're getting our attention.

Speaker 1 (53:18):
There was a plenary session on glass, I think the
first day, and apparently I meanI wasn't here yet, but it was
out into the hallway.
So if you have a solution foran application that has some
challenges that you can solveNow I had not heard this before
from you about you're basicallyjust depositing straight onto

(53:39):
the glass.
So we're talking about likeyou're doing PVD for like a
whole well, for like trace, isit patterning?
You're doing PVD for like awhole well, for like trays, is
it patterning?

Speaker 10 (53:49):
Well, the first step of metallization for glass is
let's say, you want to ship aglass core with TGVs, a sea of
TGVs, but that should bemetallized.
It should be electrolysiscopper on the surfaces and on
the surfaces of the vias To dothat.
Today, some of the proposedsolutions, some of the equipment

(54:15):
that's being tested, is like a20-step process where you start
by preparing the glass.
Things don't stick to glass,naturally.

Speaker 1 (54:19):
That's why we drink out of glasses.

Speaker 10 (54:20):
It doesn't you know, nothing will penetrate the
glass.
So you have to roughen theglass, you have to treat the
glass to allow anything to stickand then, typically, you have a
series of adhesion promoters,things that are borrowed from
the decorative glass industry.
People have been putting youknow, the Italians were doing,
the Romans were doing, but thereare much simpler ways to do

(54:42):
things.
So if you just design yourchemistry, your metal chemistry,
to be adhesive to glass, thenyou can skip 19 out of 20 steps.
And you can also do it usingthe same process you would use
for a printed circuit board youdip a panel in a tank, bring it
out and that's it Dry it.

Speaker 1 (55:01):
So how much disruption is there in the sense
of putting other processes outof business?

Speaker 10 (55:09):
well, the other processes aren't in business yet
, but the proposed solutions arevery expensive and multiple
steps and multiple differenttechnologies in the same
equipment.
So the thing about thisindustry the technologies that
get adopted are the ones thatthey can be disruptive, but they
do have to fit the process flowor the equipment set.

(55:31):
So you know, for example, ifyou have the kind of chemistry
that goes into a like a lambsaber tool, then it's so much
easier to get that adopted.
Or if it's a tank in a systemthat already has 12 tanks
because you're doing otherprocesses it's.
it's so much easier, so we dohave to design it to be
integrated, which we have.

Speaker 1 (55:52):
Okay.

Speaker 10 (55:53):
Otherwise you've got to take it out and put it in a
vacuum system and do PVD Right.
For example, trying to dophysical vapor deposition of
palladium and copper in a reallydeep glass via and they're
trying to get the 20 to 1 aspectratio is impossible.

Speaker 1 (56:10):
Right, you're not going to get a conformal coating
, you're going to get, like,built up in the end.
It's a line of sight.

Speaker 10 (56:14):
You're literally shooting the atoms and trying to
hit the sidewalls right, andyou've got a 1 to 2 degree taper
on the sidewalls.
So what we do is we justimmerse it.
It's very wetting.
It immediately wets.
The reaction just takes placeatomically in a liquid phase as
opposed to vacuum, which meansthat the only trick you've got

(56:34):
to play is you've got to makesure that the chemistry gets to
all the surfaces Right.
So it has to be really wetting.

Speaker 1 (56:39):
Right, okay.

Speaker 10 (56:40):
And then you just let the reaction happen for a
few minutes.
So our challenge as a businessis always startups have a hard
time in this industry.
I went to the startupcompetition last night and it's
great that it's being run andRosalia and Farhang are going to
run it every year.
I think there are very few andit's very hard for them to break

(57:02):
in.
So you see them sort of headingtowards the software side, the
AI of finite element, or towardsmetrology.
That's where you're seeing thestartups.
It's really quite hard to justbreak into equipment process
chemistries.

Speaker 1 (57:21):
Right, so you're providing the chemistries, so
are you looking for equipmentpartners to work with?

Speaker 10 (57:26):
Yes, we do work with equipment partners, but we're
constantly looking for andtrying to work with the biggest.

Speaker 1 (57:31):
So what brought you here as an exhibitor to ECTC
this year?

Speaker 10 (57:37):
Well, it's the center of the universe for the
industry.
Now, right, and so the greatthing is now the equipment,
semi-equipment folks areexhibiting.
They are hunting for new tech.
You know, most of our meetingsare with people at the top of
the supply chain or at the rightpart of the supply chain.

(57:58):
It used to be that you wouldhave to go to multiple shows.
Right, you'd go to Semicon forthat, right?
Yeah, now you can actually findeverybody, from the basic
materials to the, you know, tothe top of the food chain
driving the whole market, whichis why this is such a
well-attended show.

Speaker 1 (58:16):
I always think about and I've said this before, that
will probably pop up in otherareas of this episode that I
always find that ECTC is whereall of the pre-competitive
research is being presented.
Where all of thepre-competitive research is
being presented, the potentialprocesses.
We might not see these enterthe market for another 10 years,
not in the exhibit hall rightbut in the sessions it's all

(58:36):
very deeply technical content.

Speaker 6 (58:39):
It is.

Speaker 1 (58:43):
And then if you go to because you mentioned other
events then IMAPS is a littlebit further towards
commercialization of technologyand then Semi is really where
it's a smaller group yeah, Semiis sort of it's more of a
traditional trade show.
It's like I'm selling my stuffI have.
I mean, they have content right, but the focus there is to be
on the trade show floor.

Speaker 10 (59:03):
Yeah, this is.
I mean, a lot of people callthis a trade show, but it's
really a technology.
It's a technology conference.
And what's really interestingto me, I mean just sitting here
with you and staring at some ofthese posters.
10, 15 years ago, we used todraw pictures of these advanced
packaging technologies and, youknow, people quite often laughed
.
It was, you know, the island ofmisfit toys of packaging.

(59:26):
It was like we'll put this oneunder this one and this one.
You know, the island of misfittoys of packaging.
It was like we'll put this oneunder this one and this one.
And now they are actuallyproduction products because it
was the only way to get there.

Speaker 1 (59:34):
Right, so isn't it fun to be right.

Speaker 10 (59:40):
Yes, How's that for an answer?
Well, yeah, and you know, thegood thing is there were very
smart people Just coming back toyour point about it can be 10
years away people that havecommercialized it.
Weren't the people coming upwith the ideas 10, 15 years ago?
Because, quite frankly, whenyou're in a big company, you're
not really incentivized to takean enormous amount of risk, so

(01:00:03):
the risk is usually being takenoutside of the company and then
it gets adopted.

Speaker 1 (01:00:08):
Well, thank you, as always, for joining me, and
we'll talk to you next time.
It's wonderful to see you again.

Speaker 10 (01:00:13):
Thanks.

Speaker 1 (01:00:18):
So my next guest today is Hanan Zhang from ACM
Research.
Welcome to the podcast, Hanan.

Speaker 8 (01:00:24):
Thank you.

Speaker 1 (01:00:25):
So now, this is the first time we've ever met, and
the first time you've been onthe podcast, so can you just
give us a little of yourbackground and your role at ACM
Research?

Speaker 8 (01:00:34):
Right.
Hi everyone.
My name is Hanan.
My background is I was acleaning guy in the past 15
years.
After school in New York, Istarted my first job at Intel,
working in Portland.
Then I'm working for a fewequipment companies and now I'm
the general manager for ACMResearch.
I'm responsible for cleaningand plating business and

(01:00:55):
focusing US and Europe.
So where are you based?
I'm working from home.
I'm based in New York forcleaning and the plating
business and focusing on the USand Europe.

Speaker 1 (01:00:59):

Speaker 8 (01:01:01):
I'm working from home and based in New York.

Speaker 1 (01:01:03):
Okay, well, I think there's a lot of ACM's team is
spread out, but I know there's aphysical facility in Oregon.

Speaker 8 (01:01:11):
ACM research.
The headquarter is based at aFremont, california.

Speaker 1 (01:01:16):
Okay.

Speaker 8 (01:01:16):
Right, but our biggest operational office is
based at Fremont, california.
Okay, but our biggestoperational office is based at
Hillsborough.

Speaker 1 (01:01:21):
Oregon.

Speaker 8 (01:01:22):
But we are growing the team and, starting from this
year, we have engineers basedat Florida and also we are
hiring more people based atPhoenix, arizona.

Speaker 1 (01:01:34):
Okay, so this is the 75th ECTC conference right, yes.
So I've been asking people youknow, when was your first ECTC?

Speaker 8 (01:01:43):
This is my first ECTC .
This is your first ECTC.
Well, before we continue thetopic, let's congratulate the
75th anniversary for ECTC.

Speaker 1 (01:01:50):
That's a long time right.

Speaker 8 (01:01:51):
It is, it is.

Speaker 1 (01:01:52):
That's actually longer than the transistor has
been.
I believe, so Right.

Speaker 8 (01:01:57):
It's a really fascinating experience for me,
since this is my first time.
So I see a lot of greatpresentations in the customers
and potential customers, so it'sproductive.

Speaker 1 (01:02:08):
Okay, that's great.
So this is really big this year, and I can tell you that it
hasn't always been this big.
I see, but it just is atestament to the importance of
advanced packaging in thismicroelectronic space that we're
in.
Exactly, and I know that ACMhas just been tipping its toe
and getting into advancedpackaging in the past few years.

(01:02:28):
Its origins were in front-endprocessing.

Speaker 8 (01:02:31):
Right, you're right.

Speaker 1 (01:02:32):
But it's really growing in the advanced
packaging space.

Speaker 8 (01:02:35):
It is really growing.
So ACM research was focused onwafer-level packaging and more
recently we developed ourbusiness in panel-level
packaging.
So we have three products doingpanel-level at a different
panel size and we believe theyare the technical solutions for
the customers, especially theyare in the AI packaging industry

(01:02:59):
.

Speaker 1 (01:02:59):
Right, especially AI, which is becoming a
larger-sized package.

Speaker 8 (01:03:02):
Exactly.

Speaker 1 (01:03:03):
Then it used to be what was driving the packaging
industry the small form factorfor smartphones.

Speaker 8 (01:03:09):
Right.

Speaker 1 (01:03:09):
And now that AI is a big driver, I'm seeing that the
packages are getting biggeragain with more and more
functionality on them, and sonow we're going into the panel
space.

Speaker 8 (01:03:18):
Exactly.
This is where our engineeringteam will concentrate on Right
right.

Speaker 1 (01:03:23):
So ACM just had a really big milestone of its own.
I saw pictures on LinkedIn.
I think you just celebratedyour 20th anniversary.

Speaker 8 (01:03:32):
Yes, it's the 20th anniversary of ACM Research
Shanghai.

Speaker 1 (01:03:36):
Okay, so let's talk about some of you know.
You just had your Q1 earningsreport.
So, how are things going for?

Speaker 8 (01:03:44):
you.
I think this is another verystrong quarter for ACM Research
and we have revenue for $172million.
It's 13% year-to-year growthand we forecast in 2025 that
revenue will be between $850million to $950 million.

Speaker 1 (01:04:02):
Do you know how much of that is going to be driven by
the advanced packaging space?

Speaker 8 (01:04:06):
I don't really know.
So about 75% of revenue is bycleaning, 15% is by plating and
9-10% is the revenue frompackaging.

Speaker 1 (01:04:16):
Okay, so what are some of the milestones that you
have accomplished?

Speaker 8 (01:04:21):
We have several milestones this year, so one
thing we want to highlight firstis our high-temperature SPM
single wafer tools.
So our high-temperature SPMsingle wafer tools has been
qualified by a key devicecustomers in China and we
developed a patent nozzle.
The nozzle can reduce the acidmist.

Speaker 1 (01:04:43):
So what does SPM stand for?

Speaker 8 (01:04:45):
SPM stands for sulfuric acid peroxide mix.
Once the two chemicals mixtogether, it has a strong acid.
You have to have an exhaustcontrol.

Speaker 1 (01:04:55):
Okay.

Speaker 8 (01:04:56):
So that's something we focus on.
Okay, and we have a patentnovel can deliver the chemicals.
At the same time, the tool hasreally high availability and
good particle control and verylow maintenance.
So far, we deliver the systemto 13 customers already.

Speaker 1 (01:05:16):
Okay, great, any other milestones?

Speaker 8 (01:05:19):
Another milestone is relevant to 3D Insights.
We won the 3D InsightsTechnology.

Speaker 1 (01:05:26):
Award.
You did.

Speaker 8 (01:05:27):
Yes, so because our focus on the packaging, we have
a platform called ECP AP Tools.
It's doing the copper and othermetal depositions.
We believe that's the firstcommercial available tools.
Who is going to introduce thewafer horizontally?

Speaker 1 (01:05:47):
Okay, oh yeah, all right, yeah, instead of
vertically.

Speaker 8 (01:05:51):
Right, you're right.

Speaker 1 (01:05:52):
Yeah.

Speaker 8 (01:05:52):
And we believe our system can help our customers to
address their technicalchallenge.
That will be contributed to thepackaging society greatly in
the future.

Speaker 1 (01:06:04):
Okay, and so let's talk about.
You mentioned Hillsborough,Oregon.

Speaker 8 (01:06:08):
Yes.

Speaker 1 (01:06:09):
And I believe that there's been expansion there.
Right, so can you talk a littlebit about your Oregon facility.

Speaker 8 (01:06:15):
Right, definitely.
So we decided to continue toinvest in the US, particularly
in Hillsborough area.
So we purchased a facility Ibelieve it's 40,000 square feet
with a 5,200 square feet ofclean room.
So our executive team, theydecided to bring our R&D and

(01:06:36):
demonstration capability to theUS and also including onshore
initial manufacturing, to bettersupport our US customers and
eventually support our globalcustomers.

Speaker 1 (01:06:46):
Okay, so up until now most of the manufacturing and
R&D has happened in China.

Speaker 8 (01:06:52):
Up to now, most of the R&D and manufacturing
happened in China or happened inKorea In.

Speaker 1 (01:06:57):
Korea.
Okay, all right, and so now,what's the motivation to bring
it to the US?

Speaker 8 (01:07:03):
I think we want to engage very closely with the US
customers, so we think the morewe can engage with customers,
the better we can co-develop andcollaborate on the project.

Speaker 1 (01:07:17):
And if you're actually building the tools here
and your customers are here, itmakes sense to be the shipping
it's going to reduce the amountof distance that the tools have
to be traveling.

Speaker 8 (01:07:26):
Exactly right, I think.
From strategic level, thisdecision is very customer
oriented.

Speaker 1 (01:07:31):
Right, okay, all right, well, so what's next for
you this week?
What are you focusing on insome of the sessions that you're
attending?

Speaker 8 (01:07:40):
This is a very good question.
I'm very looking forward tomore technical exchanges with my
customers to better understandwhat they need and what we can
offer and help the team toexplore more opportunities,
because we have the solutionalready and we want to customize
our product for our customersdepending on what they need.

(01:08:03):
So I just want to hear fromcustomers, receive the feedback
and then work closely with them.

Speaker 1 (01:08:10):
So, beyond ECTC, when people listen to this, how can
they find you?

Speaker 8 (01:08:15):
They can find me by checking our ACM Research link
in, or they can find me bychecking our ACM research link
in, or they can find me by email.

Speaker 1 (01:08:22):
Okay, well, I can put links to that in our show notes
.

Speaker 8 (01:08:25):
Are you on LinkedIn?
Yeah, linkedin is working, sowe'll connect to you on LinkedIn
.

Speaker 1 (01:08:29):
Okay, well, thanks so much for joining me today.

Speaker 8 (01:08:31):
Thank you for having me.

Speaker 1 (01:08:36):
So my next guest is Evelyn Wang from ERS Electronic,
one of our premium members.
Welcome to the podcast, Evelyn.

Speaker 9 (01:08:43):
Thank you.

Speaker 1 (01:08:44):
So you are brand new to the company.

Speaker 9 (01:08:46):
Yes, I just joined in March.

Speaker 1 (01:08:49):
Okay, so how do you like it so far?

Speaker 9 (01:08:50):
I like it.
I like the role.
It's challenging and a lot tolearn, but it's an interesting
role.

Speaker 1 (01:08:55):

Speaker 9 (01:08:57):
I'm based in Taiwanwan, but I might be
relocated to germany in thefuture.

Speaker 1 (01:09:01):
Oh, that wouldn't be terrible.
Germany is a nice place and Ilove germany, so have you been
to to visit?

Speaker 9 (01:09:06):
yes, and it is always my dream to work in germany,
actually.
Oh well, there you go yeah.

Speaker 1 (01:09:12):
Well, here we are at ectc.
It's the 75th anniversary andI've been asking everyone when
was your first ECTC?

Speaker 9 (01:09:21):
This year is my first ECTC.
That's amazing.
It's really impressive to seeso many people from the industry
around the world gathering here.

Speaker 1 (01:09:30):
And it's all about advanced packaging.

Speaker 9 (01:09:33):
Oh yeah, it's very focused.

Speaker 1 (01:09:35):
So in your previous roles you didn't have the
opportunity to come here, no.

Speaker 9 (01:09:40):
I started to work in the industry as the sales
account manager, okay, but I wasbased in Taiwan.
So, I mainly deal with theTaiwanese customer and that is
my territory.
Okay, and by the time, onlymany management or marketing.
They will have moreopportunities to go abroad to
attend the conferences and so on.

Speaker 1 (01:10:00):
So where were you before ERS then?

Speaker 9 (01:10:03):
I was with EOS Engineering.
Okay, it's an American-basedcompany, right, they're actually
a member of 3D Insights as well.
Yes, I saw that.

Speaker 1 (01:10:10):
So what was your role there?

Speaker 9 (01:10:12):
I was also the product manager in EOS
Engineering.
Okay, but based in Taiwan,always Based in Taiwan, but I
travel back and forth a lotbecause my customers at yield
engineering are most in America.

Speaker 1 (01:10:25):
Okay, so what is your background then?

Speaker 9 (01:10:28):
I'm actually the business marketing background.
I don't have the technicalbackground.

Speaker 1 (01:10:32):
Well, neither do I.
This is great.
We don't have to talk abouttechnology at all.

Speaker 2 (01:10:36):
Oh, that's good.

Speaker 1 (01:10:37):
What brought you into the semiconductor industry,
then what interested you aboutit?

Speaker 9 (01:10:42):
I think that is the coincidence.
I joined Merck after I finishedmy master's Okay, and it was
the coincidence to join Merck.
And then I started to take careof the materials for optical
electronics and thesemiconductor.
And then semiconductor getsmore interest and attractions

(01:11:02):
around the world, so I keep mycareers in this industry.

Speaker 1 (01:11:07):
That's fantastic.
So you've learned a lot then, Iwould imagine, since you've
been.

Speaker 9 (01:11:11):
Yes.

Speaker 1 (01:11:12):
You know if you started with Merck, or do we
have to say Merck KGAA Darmstadt, germany?
Yes, german company, so you'rebased in Taiwan, but you're
potentially going to berelocating.
Yes, can you talk about that?

Speaker 9 (01:11:27):
Well, when I had the interview with the general
manager at Merck, he told mewell, only management team has
opportunity to go to Germany.
But I still have some hope thatI can visit there one day.
But I stayed in Merck for likeseven years, eight years.
I didn't get any chance tovisit Germany at all.

Speaker 1 (01:11:47):
Oh, no, well, now you have, yeah, now.

Speaker 9 (01:11:50):
I've been to our headquarter in Gammachin and
also our factory in Babingalready.
We have really good colleaguesthere.

Speaker 1 (01:11:58):
So have you been to any of the presentations since
you've been here this week?

Speaker 9 (01:12:02):
oh yeah, I went to the session introducing the
advanced, advanced packagingbecause I think semiconductor is
a big, big, big industry andthey cut so many fragmented
pieces so it is hard to get theoverview for the whole processes
and the technology and give methe good chance to recap and
relearn the whole process.

Speaker 1 (01:12:24):
So was that one of the professional development
courses.

Speaker 9 (01:12:26):
Yes.

Speaker 1 (01:12:27):
You know, that's how I started learning about this
industry.
I remember my very, very first.
This is way back in 2005.
And I went to IMAPS and I did aprofessional development course
in Wirebond.
I went to IMAPS and I did aprofessional development course
in Wirebond and it turned out tobe a very good introductory as
just packaging, but it got meinto the understanding about why

(01:12:47):
packaging is important and whatthe processes are and just kind
of at a high level and it'shelpful.

Speaker 9 (01:12:55):
It helps a lot and since then, well for a for a
semiconductor is too many piecesso you are only familiar with
what you are doing right butthat give you the overviews from
what others are doing as wellso what has been your favorite
part of the conference so far?
networking is good and thecourses are good and the.

(01:13:17):
The thing I like most is thefocus, because people come from
here.
They're from more or lesssimilar background, looking for
the solutions for advancedpackaging, so we more or less
have the similar language andsimilar target to work for this
industry.

Speaker 1 (01:13:35):
Well, I think it was great for them to send you to
this, especially if you're justlearning about the advanced
packaging sector.
You've been involved in thesemiconductor side of things
when you were at Merck, but thisis really a little bit
different and there's lessstandardization and there's a
lot to learn.

Speaker 9 (01:13:50):
Yeah, especially now everybody's developing their own
technology, so it's sodifferent that you have too many
things to learn.

Speaker 1 (01:13:58):
So now, what are you showcasing here in the exhibit
hall this year?

Speaker 9 (01:14:02):
For ERS.
We have thermal chalk systemand also the machine for
debonding, including the nextgeneration photonic debonding we
call it photothermal debondingand also some warpage correction
and warpage measurement system.
These are critical machines foradvanced packaging in the
future because we see in theindustry many of the customers

(01:14:24):
are facing such problem and weare offering the solutions for
the customers to make theirprocess more smooth and increase
their yield.

Speaker 1 (01:14:33):
Well, thank you so much for joining me today.
Thank you.

Speaker 9 (01:14:35):
Thank you for inviting me.
Thank you so much for joiningme today.
It was nice to meet you.
Thank you for inviting me.

Speaker 1 (01:14:41):
That's a wrap on ECTC 2025, the 75th anniversary
edition.
We hope you enjoyed it andwe'll be taking a break for a
few weeks and then be back withan all-new episode featuring an
interview with ASE's Yin Chang.
Stay tuned there's lots more tocome, so tune in next time to

(01:15:02):
the 3D Insights Podcast.
No-transcript.

All Episodes

Episode Transcript

Popular Podcasts

On Purpose with Jay Shetty

The Breakfast Club

The Joe Rogan Experience

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}3D InCites Members Reveal Breakthrough Technologies at ECTC 2025

Episode Transcript

Popular Podcasts

.css-r6mb8g{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:1;overflow:hidden;}On Purpose with Jay Shetty

The Breakfast Club

The Joe Rogan Experience

All Episodes

3D InCites Members Reveal Breakthrough Technologies at ECTC 2025

On Purpose with Jay Shetty