UHDI Innovations & Next-Gen PCBs with Kunal Shah, PhD.

In this enlightening episode of the OnTrack Podcast, host Zach Peterson chats with Kunal Shah, PhD., President of liloTree. The two have a detailed conversation about the burgeoning field of Ultra High-Density Interconnects (UHDI) and its impact on the future of electronics.

Kunal shares his insights on the latest in UHDI innovation, including its challenges and solutions in electronics manufacturing, especially in high-reliability applications such as defense and medical electronics.

With a focus on the upcoming SMTA panel discussion, this episode is a must-watch for anyone interested in the cutting-edge technologies shaping our world. Don't miss out on this deep dive into UHDI!

Listen to the Episode

Watch the Epsiode

Key Highlights

• Overview of the various of Developments in UHDI
• Discussion of "Traditional" Plating Materials
• Nickel Vs. Copper
• Overplating Compensation?
• Can silver be an UHDI Alternative?

More Resources

• Learn More about liloTree
• Check out this article about Ultra HDI PCB capabilities
• Exclusive 15 Days Free Altium Designer Access

Transcript

Zach Peterson: I think for most people who use silver, they're not operating at that level and so they may not even know about a dendrite issue. The number one concern I see brought up about silver is just the corrosion with tarnishing.

Kunal Shah: Yeah. So that is something at the time of assembly. So I think it is very, very critical because it is referred as a six months shelf life, but when you are in a four or five months range, you start to see tarnishing of cellular and that actually affects your assembly process. So that is sort of an early stage issue with the tarnishing and then the dendrite is sort of after assembly, an application stage issue with dendrite.

Zach Peterson: Hello everyone and welcome to the Altium on Track podcast. I'm your host Zach Peterson. Today we are talking with Kal s Shah, president of Lilo Tree. We've had Kal on the show before and today we're going to be talking to him about an upcoming SMTA panel that he is part of and I'm very excited to learn more about what he will be speaking on. Kal, thanks so much for being back with us today.

Kunal Shah: Absolutely. Zach, again, the last time our discussion was really intriguing and I think it's always great and great and intriguing discussion with you. So thanks for having

Zach Peterson: Me. Well, thank you very much. There have been a lot of, it feels like a whirlwind of developments over the last couple of years and one of the big areas where we've seen a lot of development is in the push towards UHDI, especially here in the United States. And so that is our perfect segue into the broad topic of what's going on with an upcoming SMTA panel on March 26th where you will be speaking. So if you could just give us a brief overview of what's going on at that panel and what you'll be talking about.

Kunal Shah: Yeah, so no, it is very interesting as you brought it up. UHDI, which is ultra high density interconnects in the electronic industry, is becoming sort of one of the fastest growing area, especially as you mentioned in United States. And one of the things also from a perspective of defense and some of the high reliability application UHDI, it is becoming a norm, so to say. So some of the new development, especially from the processing perspective to the material perspective is something that's been going on past couple of years significantly. And I think if you, and I'm slightly digressing from the topic of giving you overview, if you look at a little bit of history, we've been doing a very high line spacing in terms of into a hundred micron to 50 micron line spacing very, very commonly. Now we are entering into HDI and ultra HDI where these dimensions are getting literally into sub 20, sub 10 micron range where a traditional processes are not something that you can use.

So you have to innovate in terms of the processes and material to make these UHDI designs and the fabrication available sort of even possible. Coming back to your specific question, my topic is actually going to be how the surface finishes or plating or a treatment that you do on top, which is a final layer conductive treatment that you do on top of these UHDL line spacing interconnects is what I'm going to be talking about. What are the things one should be aware of? What are the things a traditional technology may not work and why a newer technology will provide you even the possibility of UHTI processing at 20 sub 10 micron range? And also how will you even increase the reliability and sustainability, which is also critical part when you are dealing with high reliability defense or medical electronics, so on and so forth.

Zach Peterson: You had mentioned traditional processes. I'm going to assume this refers to the traditional plating materials that might be used in standard fabrication. So let's say tin lead at the low end or not tin led but immersion tin maybe and then all the way up to emig or hard gold or something at the high end.

Kunal Shah: Yeah, absolutely. What you brought up is absolutely accurate. Immersion tin, as we know historically has been one of the most popular surface finishes when you talk about nineties and early 2000. But whenever, I know I always go back to history because that actually teaches so many lessons and sort of understand what are the reason for us to innovate. So what happens at early 2000 when these micro BGAs started coming where surface vanity becomes the most critical aspect, an I emergent in and also from I emergent in which from the hassle into I emergent in some of the rojas sort of got kicked in when led tin led was a standard norm and when the rojas came in, the lead sort of started to phase out and then emergent in with lead free was not something reliable and a lot of other issues. So that's when people started predominantly started moving towards enig.

That is electroless immersion goal, just a full form of enig. What are the benefits of enig is one, it gives you extremely good surface linearity because of the nickel layer and the gold layer, the surface softness is extremely planner, the surface of is very low. So that is something from micro BG assembly and all those things was very, very good. And also with the gold-based surface finish, it gives you more breathing room in terms of shelf life. So for example, your boards are being manufactured, one part of the world and assem is happening another part of the world. And especially you have supply chain logistics issues going all over the place. If you are not able to emergent in or OSP or some of the other hassle, you are looking at a shelf life three months, six months roundabout in that range. However, when Enoch got introduced, you could have anywhere from 12 to 24 months shelf life.

It gave you a much more freedom in terms of doing assembly and bare board manufacturing sort of in terms of the planning out how you're going to do logistics. However, as we are moving into UHDI when you have enig, and that's what I'm going to be actually focused in the present in our session on March 26th, which is the part of SMTA on focusing on UHDI because what happens is some of these copper is not traditional copper plating. It's using some sort of SAP semi additive process or M SAP processes as you may have heard. So what happens that it actually puts some sort of a palladium ink or palladium layer wherever you want these feature to populate, and then you have some sort of a copper process that gets deposited on top of these palladium catalyst. What happens that is there is still a, so even if you populate these copper, it's not exactly right. For example, if you have five micron spacing or 10 micron spacing, but your feature, the palladium is slightly bleeding out of your feature. So palladium, palladium and the copper is also slightly bleeding out because palladium is bled out from the defined area of these features, conductive features, which is copper. So when you have a subtractive process because of the laser, it is clear cut removal of copper and palladium underneath, right? However, this is a chemical based additive process. You have this bled out area from the feature itself.

I'm going a little bit of a technical, but it is important for a listener to understand that when the problem when comes that when you are trying to do nickel, it'll look for this copper and it'll plate everywhere. So this bled out area will also be plated, which is almost like over plating. You don't want plating on those areas, but nickel doesn't understand should I be plating onto the actual feature or should I be plating onto the bled out area? Both are identical copper, so wherever it sees copper is going to plate, but when you actually end up seeing under these highly magnified microscopes, it's only when you see see it with naked eye, it looks all perfect. But when you see these magnification lenses, you see these called nickel is being plated in a blade out area also. Now what if your features are so close to each other that the blade out area of one feature and another feature almost coalesce or talk to each other.

So you almost have a bridging issue. Some of these. So nickel plating becomes extremely difficult when you have feature sizes into 20 micron or sub 20 micron 10, 10 or five, 10 or five five is even a substrate like technologies that we have worked with recently with some of the customers, they're actually moving towards five five, which is something a few years down the road for most of the mass manufacturing. But people have already started looking into that. So when you have these features, even Phi micron, you cannot see it in a regular microscope. You actually have to put under the scanning electron like electron microscope of some sort. So when you go into these realm of such a minute feature, understanding what your chemistry is doing and what are these pitfalls. So I may have digressed and started giving a lot of information, but I hope I was making sense.

Zach Peterson: Yeah, just to maybe briefly summarize some of the issues here that arise at UHDI, you're bringing up a line spacing issue where you can have excess plating or over plating and then there's also the polarity issue. Obviously not such of an issue when you have course pitch PGAs, but once you start to going to very fine pitch PGAs where you have high density, high IO count or just maybe it's low IO count, but everything is very dense, sort of the corollary to the plating issue. And then I think there may be a third issue here, which you haven't mentioned, but it's maybe from the SI perspective, which is oh yeah, depending on the plating material you use when you have very thin traces, you have much more skin effect and there's more skin effect in the plating. So now you have a loss issue, especially when you're using nickel

Kunal Shah: Plates. Yes, absolutely. So I think thanks for bringing that up. I was going to get to that point that what happens, that one you have a nickel which is a conductivity. So I give you a little bit of technical background of why nickel is so detrimental for signal integrity is because when you have copper, which is one of the best conductor that we have, and that's why one of the reasons we have that being prevalently used everywhere in entire printed circuit board and semiconductor electronics industry. But then when you plate nickel, usually signal passes through the top layer of your conductor. So when you plate whatever is the top layer and then it has a skin effect. So your top layer is say for gold, but gold is only 15 nanometer. Your skin depth is about a couple of micron. Depending on your frequency range, most of your signal passes through nickel.

Now what if your conductivity of not what if nickel's conductivity is one fourth of the conductivity of copper? So think about what will happen to that signal. It is significantly slower for that exact same reason. It actually makes even worse when you have UHDI, so I'm talking about these standard PCBs where everything is coed, but what if things started come close to each other? Then nickel on top of having a low conductivity is magnetic material. So it actually starts creating a magnetic interference also because these features are so close to each other, a magnetic field of one feature will sort of overlap a magnetic field of another feature. So you have a magnetic, basically a magnetic interference is also being created that is also detrimental to a signal integrity. So something that is completely ruin the performance of the designer when they're designing on the computer. And then when you are actually manufacturing with the surface finish, all these effects will actually detriment the performance of signal integrity that you actually design for.

Zach Peterson: And then regarding the over plating issue, just for a moment, right, I mean if you're at high enough bandwidth, you're going to notice the impedance deviation along that line. But one thing I'm wondering is, is there like an over plating compensation that's performed in manufacturing? Because if you think about with standard a subtractive manufacturing, we do an etch compensation to account for the trapezoidal appearance of the real traces when they're actually printed. I'm wondering is there an over plating compensation that's applied as well?

Kunal Shah: So a lot of these things are, I mean you can always design for compensation, but then things become so difficult because some of these over plating is in the less than one micron or micron and a half and a couple of micron. It is that small level. But when I'm talking about in the scenarios where you are literally the line spacing is five micron. So a one and a half couple of micron over on one side and another one and a half couple of micron overing on other side, there is a chance that things may go out of control and having a bridging issue and something that overpainting may actually, so you can design and have very accurate way of how you put palladium ink or palladium layer underneath. But again, these things that you are talking about managing a couple of micron layer level accuracy, which is extremely, extremely difficult. So yeah, we have had customers especially using msap processes, complain and share some of these issues when they try to plate nickel, they try to do compensation, but again, the technology is still evolving from 20 micron into 10 micron. They're trying to play with it, but it's getting difficult as we move into smaller and smaller feature sizes.

Zach Peterson: So what are some of the solutions here to deal with this? I mean there's such a big drive to go to smaller line widths, smaller trace sizes. We're always trying to fit more stuff into a smaller area and I think packaging is one thing that's going to continue to drive this trend, especially as chips start getting built in 3D. So now you have even more stuff you have to pack into a smaller area. So what are some of the solutions to this? Because it sounds like at some point the old solution in the traditional fabrication world of just space your stuff out a little bit more starts to go out the window.

Kunal Shah: Absolutely. So Zach, I think you brought up a couple of points and I'll get to the solution, but you brought up a good point where as you mentioned that as we are trying to go into more and more dense designs and the features and much smaller features and denser, and we are actually moving. So we are moving as you brought up from a printed circuit boards to almost like a substrate like almost into the packaging world. So what we are seeing is a semiconductor packaging, manufacturing and printed circuit board almost started to overlap where a lot of companies in America have started and have plans to start doing substrate like manufacturing and having these UHDI capabilities available coming back to a surface finish solution that what are the solution for the conventional technology as we talked about that is enig which are available. So one of the things, so I'll get into explanation of what is the background of enig, why do we even use enig, right?

So the best reason why we are using enig is gold. The gold layer is so it's a noble metal, so it doesn't do any chemical reaction with any element. So that's why it's noble. So that's why you have a final layer of gold and that gives you, that's why shelf life for 12 to 24 months because of the gold layer. Also gold provide you extremely planner surface with a very low surface roughness. So that is a second benefit, which is both are extremely beneficial, especially into A-U-H-D-I application. But then you cannot plate gold directly on top of copper because what happens that the copper will start diffusing through the gold because there is no way copper is being stopped and then copper will come to the surface of gold and then it compromises the conductor altogether. And that's why OLA is used as a so-called barrier layer and that is why EIG has become so prevalently popular because they put a layer 2, 3, 3 to six micron of nickel and then put gold.

Now since we discussed that having nickel is such a detrimental from a surface signal integrity perspective and also from a perspective of plating and over plating and even managing UHDI plating a nickel on top of those structures are extremely difficult. So what is a solution? So solution is actually Lilo three has innovated one of the products called barrier layer, which is not a nickel-based barrier layer. It's an organic solution treatment which actually is not adding a layer on top of copper, but it actually ate the copper layer on the top. So what I was suggesting that why is nickel used? Because nickel is used as a barrier layer, the copper is prevented to diffuse out of nickel into the gold. Lilo is barrier layer treatment is actually a treatment conducted on top of copper that actually does exact same thing without the need for a nickel layer. So that is actually very, very beneficial because you're not adding a layer. The plating process is extremely, extremely stable and you don't have to have this issue of over plating and so on and so forth. It actually only treats the copper layer wherever it is defined.

The best part is it does not compromise the copper conductivity. So pretty much you are getting and when you have only 15 nanometer of gold after the treatment is conducted on top of copper, so basically you are getting all the benefits of enig without the need of enig because the barrier layer is acting as a nickel for providing that barrier for that copper diffusion. But on top of it, the benefits are signal integrity is as good as a best conductor, which is copper because you have only 15 nanometer, but your skin depth is about two three microns. So most of your signal passes through copper, which is the ideal thing that you want. And then there are no plating issue or over plating issues with the barrier layer and the gold. So those are some of the benefits because you are not using nickel, but it's this organic based solution treatment is actually providing that barrier layer but giving you all the benefits that how the nickel is playing detrimental effects off.

Zach Peterson: So has anyone looked into some of the other plating options for UHDI such as silver or OSP? I bring up silver because this is one area where I've had a little bit of experience, which is where you have a digital section and a really high frequency RF section in the same board. And with rf, one of the go-to options is usually silver because we're trying to reduce that insertion loss that would come from something like enig, but maybe we want a little bit longer shelf life than something like OSP.

Kunal Shah: Absolutely. So wonderful point. I mean emergent silver has been a go-to surface finish or a plating option for anything RF or high frequency reason why silver actually has even higher conductivity than copper. So that actually gives you the best signal integrity possible, like the best you can actually get with silver. The problem with emergent silver is the overall environmental reliability. So for example, if your surface finish or a portion of your surface finish is left behind in the assembly, if any of the pad or any of, so even if your pad is, say for example this size, I'll do some hand gesturing. Your pad is this size and your component this size, there is an exposed surface finish left behind a slight or few micron. Silver actually has a tendency to create or do chemical reaction with sulfur and form something called silver sulfide and they actually then starts forming dendrites.

So within year or two years in the environment out in open this dendrite will start forming and start growing and then it may actually create exactly the bridging a dendrite from another pad and dendrite for this pad will merge and create a bridging issue. So then you'll start having a malfunctioning issue because of these dendrites. So these environmental corrosion issues with the S is something that one has to keep in mind and that is why I think it's a great discussion. We had this discussion last time that when we talk to a designer and the materials company and the fabrication material that are used for the PCB and the assembly, it is important for a collaboration because not knowing how and where it is going to be used in the end application, it actually can be in a very tricky situation because you are designing for one reason, but then when it's applied all the factors that one has to pay attention to that oh, where is it going to be used?

What are the environmental condition? For example, when you are using somewhere in Asia or some parts of the Europe, the environmental condition, the levels of these gases including sulfur are much higher than other parts of the world. So understanding all of these issues and then the temperature and the humidity bias also with silver can actually lead to these dendrites growth even faster. So understanding these corrosion issues and that is why a gold-based surface finish is still a prevalent of course as you brought it up, enig has been prevalent, but when you come to RF people is like, oh, just do immersion silver. But you also have to understand some of the reliability concerns with immersion silver that it brings about.

Zach Peterson: Yeah, and for all the times that I've used silver, it's never been in A-U-H-D-I application, it's been in applications where you have BGAs but certainly not sub one mill, let's say line width and spacing. When you get to that level, I can really see dendrites being an issue. I think for most people who use silver, they're not operating at that level and so they may not even know about a dendrite issue. The number one concern I see brought up about silver is just the corrosion with tarnishing.

Kunal Shah: Yeah. So that is something at the time of assembly. So I think it is very, very critical because it is referred as a six months shelf life, but when you are in a four or five months range, you start to see tarnishing of cellular and that actually affects your assembly process. So that is sort of an early stage issue with the tarnishing. And then the dendrite is sort of after assembly and application stage issue with dendrites. So sort of from a both perspective, corrosion before assembly and corrosion and dendrite in the application after assembly is something that one has to keep in mind in terms of immersion silver, and that's why we always prefer we as a chemical supplier, but whenever we have to recommend for these high high application high reliability, when I say high application, high reliability application, especially now going into UHDI because when the dendrites become even a bigger issue, a goal-based surface finishes something that we recommend because one of the reasons is exactly because there are no dendrites, gold will remain as is.

So even you brought up OSP also, so I'll touch upon OSP as well. It's exactly that. OSP shelf life is only about three months is what is rated. And the second thing with OSP and emergent silver, both from immersion silver tarnishing, but even from an assembly perspective, how many reflow cycle that you have to are going to be incurring with OSP, it is a polymeric layer on top of copper. So when you put 265 degrees Celsius in an assembly first or secondary flow, the OSPs more or less sort of evaporated or decomposed, however you term it, but it gets compromises what I'm saying. So underneath copper is being exposed at high temperature and it can get and likely to get oxidized. So when you go into third or fourth reflow cycle, your surface is already compromised and you expect it to wet and perform the assembly operation on those compromised surfaces.

Yeah, the likelihood of failures could be high. So OSP has that issue of how many reflow cycle you can run, even emergent silver, it gets tarnished and maybe after second or third reflow, it may not give you a similar performance with emergence silver as you were getting on the first reflow. So those are some of the issues with OSP and immersion silver also, one of the other aspect one has to also understand with OSP because we are working with one of the customers and their requirement is, hey, the pads have to be conducted. So surface finish is not an area where they have to be surface mounted. There are a lot of other application, a lot of areas where the surface finish remains as a conductor staying as exposed on the printed circuit board for whatever application and reasons. But if you have OSP, the pad becomes nonconductive because it's a polymer layer on top of your PCB. So that is something that you also have to keep in mind with respect to OSP.

Zach Peterson: Yeah, I see. I think the number of reflow passes is definitely something where the designers don't really think about it because they're not looking at it from the assembly standpoint. They don't know how they're going to plan it out. I do think a lot of designers will just click the tin led button, or not the TIN led, but the immersion tin button on their quote form online or they'll click the email button and they'll just say, yeah, go for it.

Kunal Shah: And I don't know, maybe if I can make this comment, they click on a button, I don't know from a designer perspective or from placing an order at PCB Fab House, whatever is cheapest, right? So let's pick the one which is the cheapest because nowadays everything is online application where you're filling out all the forms with the drag down options and whichever is the cheapest. Let's pick that. But yeah, I mean some of the thing is yes, one has to know the reflow passes as you brought up. Second is the cost, but you have to be very judicious of cost because there are other surface finishes that I can bring up like PIC if you may be aware of, because a lot of time with the higher reliability application palladium layer is put between nickel and gold. And one of the reasons is the plaque pad with enig is historically corrosion between nickel and gold layer at that interface.

And to prevent that palladium layer was introduced, and that is why ick is what is called electro electrodes, palladium immersion gold is the full form of that. Now the cost will be even extremely higher, exponentially higher is because the palladium layer, because of the precious metal like palladium, which is 1.5 times the cost of gold. So you are not adding the cost of gold, but on top of it you are adding the cost of palladium but not necessarily you get all the reliability with eick. There are issue with signal integrity that are issue with some of the reliability depending on the palladium layer thickness and so on and so forth. So it's not that you pay the highest money, you get the best product and it's not that if I select the cheapest and I'll get away with it. So one has to understand the pros and cons of each aspect and spend wisely to get the optimal performance that one should get, especially understanding where it's going to be applied, who is your customer, so on and so forth.

Zach Peterson: Yeah, you used a term that I think is often misunderstood, which is best product, right? Best always comes with a big asterisk because when you look at it from enig, best really means what highest reliability, whereas you look at it from the silver perspective, best means signal integrity and not necessarily reliability. So I guess best really takes some consideration here. And then also I think as we get into UHDI, we're pushing more and more and more into the higher frequency range. So below let's say a gigahertz, you're probably not going to notice the loss difference between enig and tin. You just care about the reliability. But once you get into the many gigahertz bandwidth range, now you definitely notice

Kunal Shah: Absolutely Zach, so you brought up an amazing point because even not even a gig, even up until five to 10 gigahertz, you may not see a major detrimental effect between emergent silver or any other surface finishes. They would be all from surface finish perspective, they would be all being same in terms of how much loss that you are getting. However, as you go into 10 gigahertz to 25 gigahertz, that is where the high bend of 5G is. The 77 gigahertz is the automotive frequency where those are the typical frequency in the automotive application. And then some of the RF is a hundred plus gigahertz. So exactly what you brought up that when you go into 10 plus gigahertz, you will actually start seeing the effects of if you put enig versus if you put emergence silver. And that's when you have to realize that hey, for reliability, should I go enig?

But even there are concerns from the reliability side with enig, that is a different story altogether, but it is still more reliable from a gold layer perspective and environmental corrosion perspective because it's at the end of the day, it's a gold outer layer, right? But when you put emergent silver with high frequency, then reliability is major concern from a perspective of environmental corrosion. So those are the things that one has to understand. And exactly that is when our solution, I'll discuss in terms of you are removing the nickel, you are putting this barrier layer treatment, so it gives you the performance of nickel in terms of barrier layer for that copper atoms, but it gives you an outer layer goal. So you get the best environmental corrosion protection from a reliability perspective as you brought up. But from a signal integrity perspective, it's very similar to emergent silver performance because your signal is passing through either gold and most of it is passing through copper. So in that perspective, your signal integrity is as good as it can get comparable to silver. But the reliability is always good because you have outer layer of copper and it is protected by underneath bare layer treatment.

Zach Peterson: So as someone who is much more of an expert on platings than I am, I'm sure you've done a lot of digging into the research literature and you probably have found all sorts of ways people have tried to overcome this problem and eliminate nickel and still ensure that we have a highly reliable surface plating. You've gone one direction, which is using passivation to create a barrier layer. What are some of the other approaches that maybe haven't worked out or that others are trying to work on to then help us get to this next level with UHDI?

Kunal Shah: Yeah, absolutely, Zach. So people have actually tried this. This direction is not completely new, exactly what you brought up. We took this path, but it has been explored previously. So there are two or three ways or two ways mainly people have explored is one is it's something called DIG, which is direct immersion gold. What they do is remember in the beginning of our conversation I said that you cannot put immersion gold at that thin layer because if you don't have nickel, copper will diffuse out to the top of the gold, the top surface of the gold layer because it's only 15 nanometer. But direct immersion goal actually plate as I as 150, 200 nanometer. So the idea is even if it diffuses, we hope that for 200 nanometer it would not go all the way out. And then our application in terms of assembly or our application is not as if it takes in terms of an application, if you do a simulation or in real scenario that if copper takes five years to come out to gold for 200 nanometer, that's good enough because we need reliability only for two or three years or four years.

So we will be fine. Let's put 200 nanometer, 250 nanometer of gold. So that is one approach people have taken. The second approach people have taken, hey, instead of using NICO as burial layer, let's use palladium as barrier layer like E pig, but not put nickel in it, put electrodes palladium directly on top of copper and then or directly or they put some sort of a seed layer of gold, but mainly put electrodes, palladium, and then put immersion gold. So they call EEG or eag that process. Now again coming back to DIG, there is something I like to mention is with DIG, instead of 15 nanometer, you're putting 200 or 215 nanometer of gold. So your cost of plating almost automatically becomes four or five times more. So that is a major, major drawback when you do mass volume manufacturing for your products. But the second thing is also when you have such a high gold count or gold layer with a micro B, g, A, when you're trying to do a solder application that so much of gold make and may cause, I'm not saying it's the case for every scenario, but may cause such a small areas, such a thick gold will get dissolved into solder may cause gold in brittle.

So your solder may be subject to a brittle failure of some sort because too much gold at that interface and everything gets dissolved at the time of assembly. So there is a reliability concern with DIG, but also major is the cost concern. Now let's talk about eeg, EEG electrodes, palladium immersion gold as I mentioned, not also that actually add your cost even to another level because palladium layer, as I mentioned, is 1.5 times the cost of gold. So you are replacing nickel with even more expensive precious metal compared to gold. So it makes your assembly even more expensive. That is one. And second is even palladium has a concern of signal integrity. So if you are removing nickel, it performs definitely better than nickel palladium is, but it's not the ideal like gold or copper or silver. So you do still have signal integrity concern, especially if you are going into higher frequency into 20 30, 50, 70 gigahertz.

So in those scenario, it's not the ideal replacement as signal integrity because your performance is not as good as I mentioned, likes of gold likes of copper or likes of silver, yet the cost is exorbitant because you're replacing nickel with even higher precious metal than gold. So yeah, those are some of the alternatives which are out there. And again, coming back to the alternative in terms of nickel free is either you go completely goal free, which is A OSP and emergent silver, but we talked about some of their drawbacks from that perspective on the reliability side of things. So yeah,

Zach Peterson: So of course you being president of Lilo Tree, I have to ask how much of a positive response have you seen to your solution versus some of these other solutions that you mentioned? I get that DIG super cost prohibitive unless you're at low volume epic, probably also not as cost prohibitive but still cost prohibitive. So it sounds like maybe one of Lilo tree's advantages is from the cost perspective.

Kunal Shah: Yeah, so actually a couple of advantages I'll bring up is with all the benefits of ease of plate and for USDI with signal integrity we talked about, but the cost is the OT threes nickel free solution is actually cheaper 20 to 25% cheaper than enig. So that actually makes us a very, very attractive proposition from a cost perspective because it's 20 to 25% cheaper than enig. And the second benefit is the typical gold plating happens with the cyanide based gold source molecule or chemistry, whatever you call. So it's a cyanide based solution. Our solution, our gold plating solution is completely cyanide free. It's more stable and it's actually cheaper to operate than a cyanide based gold. From a nickel free perspective, I think we are, we call it as a chosen one where whoever wants to do nickel free, they always use or opt Lilo three's process compared to any other nickel free options out there.

However, we are also, I mean sort of broadening the scope, there are some of the fab houses we are talking to and we are actually in the discussion where they're considering to even replace enig with a standard boards not necessary UHDI, not necessary the high frequency application, even a standard board with a low frequency application and not necessary UHDI are thinking that, hey, why are we using nickel or standard enig, which is even more expensive in the first place and has a reliability concerns at the interface of nickel and gold as I mentioned, the black pad. And then also the third thing, getting a little bit technical is your solder joint is a tin nickel inter metallic versus the nickel free option you actually got get copper tin, which is much more stronger and if the nickel free cheaper provides better reliability and if it is sustainable, which is cyanide free, why should we even use traditional Enoch in the first place? So that is where we are actually four into so far. Definitely from a signal coming back to your question to answer for a signal integrity and USD application, definitely Lilo threes nickel freeze, sort of a premier selection option, but we are also being considered as a traditional enig replacement solution as well for our traditional low frequency non UHDI boards as well.

Zach Peterson: Well, this is all extremely informative. We're just about out of time, but I want to thank you so much because I always feel like I learned something new whenever we chat. So thank you so much for coming on the podcast.

Kunal Shah: Thanks, Zach, it was great talking to you for sure, as

Zach Peterson: Always, and to everyone that's listening out there, make sure to head over to the Peoria Sports Complex in Peoria, Arizona. If you happen to be in Peoria, Arizona on March 26th, 2024 for the ultra high density Interconnect Symposium, sponsored or put on by SMTA, you'll get to see Kunal Shaw at the symposium talking about all of the stuff that we've been talking about here today. To everyone that's out there listening or watching on YouTube, make sure to hit the subscribe button and hit the like button. You'll be able to keep up with all of our podcast episodes and tutorials as they come out. And last but not least, don't stop learning, stay on track and we'll see you next time. Thanks everybody.