Hemdeep (00:09)
Welcome to Big Ideas in Microscale, the podcast where we explore groundbreaking research happening at the microscale where microinnovations makes a big impact. We're excited to showcase the incredible work being done by our users from around the world who are pushing the boundaries of microfluidics, lab on a chip, organ on a chip, and beyond. Through these conversations, we hope to learn from their experiences, uncover their insight, and bring their big ideas to wider audience. So whether in a lab, On the go, or just curious about the future of microtechnology, join us as we dive into big ideas at Microscale.

Hemdeep (01:01)
Well, thank you for joining us at Big Ideas at Microscale. Today I have the distinct honor of presenting two really outstanding individuals in the field of 3D printing at the Microscale, Greg Norton and Adam Woolley from BYU University. My name is Henry Patel and I'm joined by my co-host Robin. Hi there, Robin.

Robin (01:25)
and hello Greg and Adam.

Hemdeep (01:27)
Hello. So I'm going to give you a bit of my history with Greg and Adam. I want to highlight some of the very key important things that they've done and why I think that they're going to be very impressive speakers during this podcast. I've had the chance of talking to Greg and Adam over the last year, a couple of times. And every time that we spoke, it has always been eye-opening. There's always a learning moment. There has never been a moment where we walked away head scratching. It's almost jaw open, just going, wow, I cannot believe that that is really where the ideas can go and can take you. And when it comes to Greg, I had an encounter with him even prior to that, when we met up at the conference in San Diego. And that again, a very, amazing gentleman at that time. It was very thoughtful when he came over and introduced himself. And then part of that, I saw him on this Ted talk in 2018. That was for us.

It was truly a starting point of just really understanding where technology was and where there was an actual vision of where it could be. At that time, Greg sort of put a, you know, a milestone and put down a flag and said, this is where 3D printing should be when it comes to microfluidics. And so I think that that's probably what I hope we'll be able to sort of explore during this podcast. So Robin, take it away.

Robin (02:53)
Well, thanks, Tami, for giving us the background and your relationship with Adam and Greg. I, for one, have only had one conversation just before this podcast where I sat down with Greg and Adam, kind of learned from them what they do. And since then, I've read a couple of your papers. And I must say, as someone who is still relatively new to the microfluidic field, and I've kind of been learning as I go.

Your papers are so well written and in general are very like simplified for someone who with limited knowledge where I could actually comprehend a lot of what is going on more so on the 3D printing side as opposed to the bio analysis and that sort of thing. think for me, what I'm kind of looking forward to is to pick both of your brains and figure out how you came across some terminology and just overall the concepts that you kind of developed regards to 3D printing. So let's get started and let's talk a bit about like both of your backgrounds, your professional backgrounds, your field of study, how you guys met and how long you guys have been working together.

Greg (04:07)
I'm in an electrical engineering department, but my bachelor's and master's are both physics. And I had intended to do a physics PhD, but found that when I was on that path that my area of interest, which was astronomy and astrophysics, cosmology, I just didn't see how I was going to be able to feed my family. So I needed a plan B. And that meant going across town to University of Southern California to electrical engineering where they had this fantastic optics and photonics program.

And so I swallowed my pride, I thought, and moved away from physics to EE. And that was the best academic decision I ever made. So I come from kind of a different background, but I just love all things technical and working with Adam has just been awesome. That's been a lot.

Adam (04:55)
Thanks. In my background, I'm a chemist by training. I did my undergrad and actually my PhD in chemistry at University of California, Berkeley. So another good California school, Craig. After postdoctoral work, I've been at BYU since 2000. I'm in the Department of Chemistry and Biochemistry, but research interests in microfluidics, I should say some of my initial research as a graduate student, we were developing microfluidic devices as part of the HEMA Genome Project. And so that was...

a very early taste of what the power of microfluidics and also some of the challenges associated with using microfluidics. So Greg was at University of Alabama and then he interviewed to become a faculty member at UIU. And I think I met him at the interview because they said, hey, he's interested in microfluidics. You do microfluidics and work in different departments in different colleges. But I went there and I thought, wow, he's doing some really cool stuff. the interesting thing is the projects that we initially collaborated on had nothing to do with 3D printing. was, mean, Craig, you can talk about it more, but it's about micro cantilevers and sensing. So anyway, it's interesting that we started collaborating on some things that we have completely left behind, and now we're definitely all in for 3D printing.

Hemdeep (06:10)
If we were to date stamp that, when would that have been?

Adam (06:12)
Greg, when did you interview at BYU? ⁓

Greg (06:14)
So that was 2005 when I interviewed and came to BYU. And then we started collaborating within a year or two of that. And prior to coming to BYU, my research was all focused on micro photonics, diffractive optical elements, micro and nano fabricated devices, MEMS, biosensors. And when I came to BYU, I brought all of that with my PhD students with me. And that's what we continued to do. And long story short,

Adam (06:22)
Yeah, it's been almost 20 years.

Greg (06:44)
ended up moving more into microfluidics and then had the usual clean room issues with making microfluidic devices where yes, you can do it, but it just takes a long time to do. So getting interested in 3D printing is a possible avenue for fabrication. And that's kind how everything started with the 3D printing.

Robin (07:05)
Did BOIU already have 3D printers in place or was that like a technology you happened to come across and you're like, hey, we need this in our lab.

Greg (07:14)
No, I was one of the heavy cleanroom users here at BYU. in fall of 2012, as I was looking at alternatives, kind of the hype cycle had started up on 3D printing with some of the original patents becoming out of date and people being able to start doing things. And so I looked at that and thought, hmm, I wonder what sort of opportunity we could have here with microfluidics. And so January of 2013, I started a five or six senior design project for a semester looking at, hey, let's build a 3D printer. And so we took the scanned laser approach and we built something and started to photopolymerize things and with the goal of microfluidics as the application. But by the end of the semester, it was obvious that that was not the way to go. We needed a different path. And so the DLP or image projection type of stereolithography became much more attractive after having gone down that dead end.

Adam (08:14)
And we did buy some commercial 3D printers of various types. I bought an early, one of the inexpensive ⁓ FDM 3D printers that puts out a ⁓ heated filament and patterns that. And then Greg and I got one of the early SLA dynamic light projection 3D printers that was commercially available. And it was nice to be able to make things, but just not at the microfluidic scale.

Greg (08:42)
Yeah, we did a lot of early work with using several different commercial 3D printers and looking at commercial resins for microfluidics and found that the resolution for the negative features, which are the key thing in microfluidics just was not there. So we thought, well, gee, let's do our own custom resins and maybe that will be good enough. And so we explored that and we found that, no, that's not good enough. you can get down to, I think the best we did was 60 micron tall channels by a over 100 microns wide. And that just wasn't the sky scale that we wanted to be at. And it was difficult to get that to turn out well.

Robin (09:20)
Sorry, this was for the commercial resins or your custom resin.

Greg (09:24)
custom resin with commercial 3D printer. At that point, it was either, well, we just live with it, or we jumped in and make our own 3D printers as well as our own materials and solve the real problems to get down to the size scales we needed to be. And that wasn't necessarily an attractive proposition, but on the other hand, just living with what was available was also not attractive. So just decided to go for it.

Robin (09:27)
Okay.

Adam (09:49)
Yeah, some of the projects we were working on using conventional cleanroom technologies, we were making devices in several different materials and aligning and stacking together five or six different layers. you know, I had a postdoc in my lab at the time who was just really trying to solve this problem. And the problem was that he would spend six and a half days of the week working in the cleanroom, trying to make devices.

for every half day that he could actually test devices. And it was just, it was frustrating that the yields and the abilities of the conventional clean room systems, especially as we went to very complicated fluidic structures with multiple materials and multiple layers, we were not making very good progress. And so that was part of the reason I was open to trying something that was very different, unexplored and not commercially available just because of the potential it had to really allow us to make devices with the exact features we want in 3D space and with resolution that's appropriate for doing microfluidic experiments.

Hemdeep (10:58)
When you were starting to develop the material and the platform, was there any other teams within the United States or anywhere globally that were also trying to find a solution to some of these things that you've described, Adam?

Adam (11:13)
There are a lot of people trying different approaches. I mean, I think the most common one is I'm going to use commercial 3D printers and see how good they can get. And we had also tried that approach. And the challenge was it just, you know, we could not make things small enough to do some of the experiments at the volume scales and with applied voltages and other things that we use to do that. There were groups trying to solve the microfluidic device fabrication challenges with 3D printing, but just the commercial options that were out there at the time.

wouldn't do it. I mean, they couldn't do it. And this was something we published as a review, kind of a trends article about eight years ago that pointed that out that, yes, 3D printing is great, but with the commercial systems, it only gets you so small. And to really address some of the fundamental challenges, you need a better system. then fortunately, right around the time we published that was when Greg had a fantastic student who was able to solve a number of the key technical problems and make it so we could actually make the devices of the scale that we wanted.

Robin (12:13)
In general, what were the sizes that you were looking to achieve? Like sub 100, was there a specific number that you have always been shooting for?

Adam (12:20)
Well, sub 100 micron was the target, but even if you're at 99 by 99 microns, there's still some challenges. In my lab, we do a lot of separation experiments using a technique called capillary electrophoresis, where you apply high voltages across a channel. And the larger the channel cross section, the more current that flows through and therefore the more heating. so optimal sizes are in the ⁓ 50 to 60.

micron cross sections or smaller. And the work that Greg's fantastic student, ⁓ Hua Gong, was able to do put us at where we could get down to about 20 by 20 microns. So even under maybe a suboptimal day, we were still able to print things that were well within the tolerances that we need.

Greg (13:07)
Yeah, that particular result, 20 by 20 microns, was our 18 by 20 microns is what we published. That was done with our very first custom 3D printer with our own custom resin. That involved a lot of detailed work to understand the kind of everything that goes into negative feature resolution. So in a three-dimensional space, you have the XY plane, which is kind of the plane of the layers.

And there the key thing is what the projected pixel size is for these ultraviolet images that you project into the resin. And that's kind of a straightforward thing. mean, the smaller the pixels, the better at the potential resolution you could have. But the one that is more difficult and that most people stumble on is the resolution in Z, which is orthogonal to the layers. And there it's all about how deeply does the light penetrate the resin.

And if you don't control that and minimize that, then there's no way you can get good Z resolution. And so that paper, together with the analysis in the 2015 paper that we did, really showed how you design your resin in conjunction with the 3D printer system to get high resolution in both X, Y, and Z. And so that Z dimension was really the crucial missing factor.

Even though we did, I think, 7.6 micron pixels, which was much smaller than what others were doing, which was more like 27 microns in the best case, but usually more like 50 microns. And so our XY resolution was really excellent compared to what else was out there. But the real kicker was the Z resolution, being able to master that and get that to work out really well. Like I said, it's all about controlling the optical penetration depth.

That means in your resin, it's all about what UV absorber you put into the resin to limit the penetration of the light. And then the key factors there are that that absorber, its absorption spectrum has to fully cover the emission spectrum of your source, which is typically an LED, either 405, 385 or 365 nanometer LED. And so a choice of UV absorber is really what it boils down to with that Z resolution.

Hemdeep (15:26)
describing the actual solution that you came to. What were the stop starts that you had while you got there? Because that must have been ⁓ far more an important learning experience of getting to the actual final solution in this case.

Adam (15:39)
Yeah, it's amusing. mean, the paper that we published that showed that, I think that was the one, Greg, where Bryce went through 20 different UV absorbers. So we had a student who was just like, OK, here are some ones that match the optical properties we need. And then it's like, but it doesn't dissolve in our resin, or it's fluorescent. And so we have this wonderful flow diagram that's in the paper that kind of went from this pool of 20 possibilities down to one that actually worked well for that initial paper. And we've since come up with a couple others that are suitable, in fact, better than that initial one that we identified. that was certainly a part of it. Yeah, I don't know, Greg, what else?

Greg (16:21)
Yeah, that choice of UV absorber was just crucial. to begin, we started with Sudan 1 because we were using a 405 nanometer source. And the Sudan 1, while it's very effective, it's just this ugly orange color and everything in the lab turns to orange because the students spill it. I mean, it just gets everywhere. It's just nasty. And so I just really wanted to get rid of that Sudan 1. I didn't want it in my lab anymore.

And we shifted to 3D5 nanometers where there was a greater variety of potential absorbers. And so this student that Adam mentioned was an undergraduate, Bryce Pickham, and he started attending our research group meetings and he wanted something to do. And in one meeting, he piped up and said, yeah, what can I do? How can I help? I told him, well, you know, in the basement of the library, there is this, I don't know, 20 volume set.

each of them three inches thick, that just compiles the absorption spectra of all these materials. Why don't you start pulling those out and just go through the absorption spectra and here are the criteria we're looking for. This is the kind of absorption we want. And so he went through these volumes and just page after page, and he would note down the ones that had an absorption spectrum that could potentially match ⁓ what we were looking for.

He developed this gigantic list of materials and then we sorted through it for toxicity to get rid of the toxic ones. And in the end, we ended up with these 20 different candidates that we purchased and we tried. as Adam said, we had this ⁓ set of criteria that they had to pass, first being soluble in the monomer we were using. And of course, many of them were not soluble, so that got rid of them. then...

And then there were other things having to do with measuring the absorption spectrum as a solubalized absorber in the resin, make sure everything was good there. And then there were other issues with different ones like Adam mentioned. Some of them were fluorescent themselves, which kills your resolution. Some of them also created polymer matrices that were just crumbly. They just didn't have mechanical properties that were needed. And so in the end, of these 20 candidates, one of them worked out really well. And so that's what we ran with for a number of years.

Robin (18:44)
And one of your papers, you mentioned, especially for the clear material, you mentioned that the yellowing of the device wasn't desirable. Did you find that to be a big issue when you were developing your custom resin?

Adam (18:56)
The initial absorber that we identified actually has sort of a light yellow color to it. And that wasn't, I mean, it wasn't a showstopper for a lot of things. I mean, it was, we had to kind of work around the feature of the resin when we were doing fluorescence detection. I would say, you know, it was a challenge, but not insurmountable. And then, you know, with the better UV absorber, the Avobenzone, which by the way, is just an ingredient in a lot of sunscreens.

That's it's a good UV absorber because it's a good UV absorber anyway it that one it doesn't have the coloration and I guess there's There's a little bit of a something that happens post cure under certain conditions But I think that's actually the photo initiator and not the not the UV absorber itself. So yes ideally the less ⁓ Stuff that interferes with optical measurement in your device the better I mean if you want to measure small quantities of things in a channel within a device You don't want to lose your light to anything that absorbs. you know, when we, when we went to this better, clearer resin with it, with the improved UV absorber, that's really the advantage. I mean, you could work with one that does have ⁓ a tint to it, obviously going from the Sudan one to the ⁓ NPS to Avobenzone. We went from, you know, incredible amounts of light absorption to, you know, moderate amounts to essentially none. So at least in the, in the finished device. So that's. I think that was helpful.

Robin (20:26)
But guess at the end of the day, didn't really affect, I guess, the resolution or the final channel of negative void.

Greg (20:32)
Yeah. It doesn't really affect the resolution. Also, just to be clear, doing any sort of microscope observation of the 3D prints was always fantastic because the final 3D printed surface was very flat and smooth, so you could look through it with a microscope, no problem. We always print on top of glass slides, so you could look through the glass slide and look in the other direction. And it's just perfectly clear and easy to observe.

the structures that you fabricate. And the only issue really is whether that tint and coloring ⁓ affects whatever you're trying to do. In many cases, it doesn't. But as Adam said, we've shifted over to this ava benzo, so there really isn't much of a tint even. But in terms of microscope observation, that's never an issue. It's just always beautifully pristine.

Hemdeep (21:24)
So this is the material side that you're sort of giving us a good sense of what that pathway looked like. I think we'll have to leave the specifics of the printer until next week. That's a wrap for today's episode of Big Ideas at Microscale.

Robin (21:36)
A huge thank you to Greg and Adam for sharing their backgrounds and their pursuit of building their own high resolution 3D printing solution for microfluidic applications. We're curious to hear more and hope you are too.

Hemdeep (21:48)
Next week, we'll continue our conversation with Greg and Adam as they walk us through the custom-built 3D printing system that they've developed, the challenges they face, and how their unique approach is opening up new ideas and possibilities for the microfluidic research community.

Robin (22:05)
Thanks for tuning in to Big Ideas of Microscale. If you enjoyed the episode, make sure to follow us and stay up to date. You can listen on Apple Podcasts and Spotify, or watch the full video on YouTube. You can also follow us for more updates and behind the scenes content on LinkedIn, Instagram, Blue Sky, and X. We're Cadworx3D across the board. Let's spell C-A-D-W-O-R-K-S-3D. for show notes, paper references, and bonus resources on today's topic, visit our website, catworks3d.com. That's spelled C-A-D-W-O-R-K-S 3D.com.

Hemdeep (22:47)
Thank you for tuning in and as always stay curious, keep exploring and never stop asking the big questions that are shaping our world. Whether you're in the lab, on the go or just curious about the future of technology, join us as we continue to dive into big ideas at Microsoft.