[00:00:05]
>> My name is all of a brand I'm a director of I and the Institute for electronics and on technology I'm really welcoming you to be a high end user day and the distinguished mind to lecture I have only 5 slides I really don't want to take much time before our speaker our speaker can start just to remind you of the agenda.

[00:00:28]
So we're running a little bit late but we give our speaker of course the full one now I hope you can all I hope you can all stay for that hour afterwards we have a poster session the furred 1st morning poster session then we have lunch which would probably be outside I don't even know I don't even know right outside then in the early afternoon we have 3 student presentations then another poster session.

[00:00:54]
And then there is an award presentation at the end of 330 posters out in the in the lobby right if you haven't put up your post or. Wait after until after the talk and then put up your poster. And then just what is this all about right user day so the idea really behind it is that you have the chance to learn a little bit what your fellow clean room and materials characterization users are doing that's really the the idea of this of this day as you probably know we have more than a 1000 users in those 2 core facilities on an annual basis and for sure you don't know what the what the person next to you in the clean room or in the materials characterization facility is doing so this is the idea of today mingle go around the posters learn what the others are doing maybe your good ideas for your own research and then come into the mind to lecture.

[00:01:54]
This distinguished lecture is really an honor of Jim Meindl I don't know how many of you actually have have met him I have met him many many times he was a faculty member at Georgia Tech for 20 years there actually are long career at Stanford already and r.p.i. you see his credentials there he was director of the microelectronics Research Center when he came here then became the founding director of the Nano Technology Research Center and really this building I think it's fair to say we wouldn't have this building without Jim Meindl he has graduated 90 ph d. students.

[00:02:35]
Many of them are in very very high positions such as the president of mit just to give you an example 600 papers and numerous awards and so on so this lecture seriousness really in his honor and I'm extremely excited it to well come 2 days mind the lecture Do you agree or and I do have a cheat sheet with a couple off with a couple of words for her introduction.

[00:03:10]
Professor of materials signs mechanics and medical engineering at Caltech I think you were named that. Over. And it was just recently right that you did you did you would named just a little bit about her background she got her. Bachelor's degree in chemical engineering at mit. With a minor in music and I come back to that in a 2nd.

[00:03:36]
She got her master's and Ph d. degree in materials science and engineering at Stanford University she spent a couple of years in industry Intel during the ph d. is that right. Between muscles and a Ph d. did a post talk at Park Colorado Research Center and in the Bay Area has many many awards and recognitions is really too many too many to too to all spell out here I want to mention she just received the inaugural Heger award by the triple a f m right so the American Association for Advancement of functional materials Heger named after Alan Heger Nobel Laureate in 2000 conducted polymers I think right as the is the topic she received a faculty fellow.

[00:04:27]
Then of the Year event of our bush faculty fellow by the u.s. Department of Defense c n n 2020 vision Arie young she she presented at the World Economic Forum and then many many early career awards I don't think is there anyone left anyone left. That you have not that you have not received company early career award.

[00:04:52]
Nassau early career war Deo your early career award a as I mean early career award and so on just published more than 140 papers I read given more than a 100 invited lecture work is highly cited some 13000 or more than 30000 citations really really pleased to have her here you see the title of her presentation materials by design 3 d. nano architect that made a materials and with that I would like to welcome Dr Greer.

[00:05:23]
Technological challenges so we are all here in the nano technology world right now because it is not thank you. All right so you've been staring at the title waiting for a while so I'll just I'll just go ahead and start Also I really like. Keeping my seminars interactive so please ask questions as we go as opposed to waiting until the end has a lot of information all right Ok So materials by design make sense then you look at the 2nd part of the sentence 3 dimensional nano architected metamaterials and you might be wondering Ok I think I understand each one of these words individually but why they belong in the same sentence might be a little less clear so my goal for today's talk is to convey to you why they do in fact all belong in the same sentence All right so I imagine you're all familiar with a situation like this you go to the store you need to get all these things and you absolutely need to shove them in your little plastic bag then they're just not strong enough to hold all the things that you really need and of course this is not the only example from the wine glass that doesn't quite make it to the toast to the balloons to the I really meant to have 5 children situation so these are all examples of when cereals that are lightweight and because they're so lightweight they're weak and very easy to damage and to tear Now let's look at the opposite end of the spectrum let's look at the materials that are very strong that we know to trust some of us are very familiar with these because I just arrived in one of these so this is a Boeing 747 it weighs a 1000000 pounds so the tremendous expenditures in the airline industry come from the the fuel that's required to propel a 1000000 pound machine through the air so my take it that way and it's election have cost me about 33000 dollars if you do a back in the calculations of thank you for emerging so so right so this is an example of a material that's strong but because of that it's heavy and therefore expensive right so the Dreamliner is already much lighter 50 percent lighter and uses substantially less fuel.

[00:07:20]
Here's another example solar panels right so we in California at least we use these a lot and so if this guy or this panel slipped off the roof accidentally This is really bad news for everyone right so far the the market for solar panels they're made out of single crystalline silicon which is a very brittle material right so it'll shatter so specialized my cereals are heavy and therefore and therefore need to be used in these applications Now remember in the last decade or so we've been talking about this space elevator was far as I know there's still is annoying and the reason why all these problems exist is a very busy plot and I'll walk you through it when i'm glad to hear strength is a function of density and I could have chosen any other mechanical attribute it could be stiffness it could be talking is it could be fracture resistance but it's something that's mechanical mechanically resilient and what you see in these colorful domains are all the materials that we know how to make today and so the picture that emerges right away is that we're very good at making materials that are simultaneously strong and heavy or simultaneously lightweight to me and what we're not so good at is getting into that white space materials that are lightweight so that the payload would be tremendously less and also me Cannick we're resilient so this is a long walk but so even a modest increment within the space already represents a substantial So how do we get into that white space if everything that we already know how to make today is already plotted here.

[00:08:53]
We need to do something we need to have some kind of innovation so the way we do it is we draw our inspiration from architecture so if you look at the largest manmade monument so this is a stone management made monument maybe some of you have been here so the Great Pyramid of Giza stands about 150 meters tall very tall right in that weighs nearly 6000000 tons so a very very heavy weight in contrast to the Eiffel Tower which stands twice as tall 300 meters it weighs 3 orders of magnitude less so i 1000 times less humans just doesn't make any kind of resilience and so what that teaches us is that if you cleverly construct your structure you can get away with using a lot less material but not sacrifice any of the mechanical strength so that's the idea that's the philosophy behind much of our work so we get this work using with developing these micro a lot of his in collaboration with trial so this micro lot is made out entirely of nickel we know nickel to be a strong metal right it's sitting on top of the world's most normal dandelion there's nothing Photoshop here this is not the world's strongest dandelion it's just a regular dandelion and you can see that there might be a lot is hardly perturbing it at all in fact if I were to hold one in one hand and a feather in the other into release them at the same time the feather will fall down faster these are exceptionally lightweight but also very flimsy so to get the simultaneous attainment of the strength and the light weight you need to go down 3 more orders of magnitude in reduction of dimensions and what I'm showing you here are some of the Nano architected materials that we made in our group those of you who are fans of Dr Who Here's a little micro Tartars it's about 110000 of you here the innards I need.

[00:10:31]
The rest of them this is to demonstrate that we don't need to make periodic structures that the rest of them are in the in the shape of lattice as a periodic architectures but the but it doesn't have to be so the obvious level of prosody comes from the fact that they're all cellular solids and you can probably see that already right so they're beams they're beam based 3 dimensional networks the less obvious level of prosody comes from that look at this isn't an image here the tubes that comprise these materials are hollow and the wall segments of these tubes can be on the order of 5 to 10 minute meters so when you look at these materials they embody every length scale from some nanometers to hundreds of nanometers to microns many microns and then eventually millimeters and centimeters and because of that they're no longer can be properly described just from a structural perspective or from a materials perspective at these lengths scales materials in structures are no longer independent of one another in this is why it's appropriate to call the metamaterials there you'll see what I mean in just a few a few minutes so I convinced you that they are open in very light weights but I have yet to convince you that they are also strong in so what I'd like to point out is this it's all about the wall thickness being in the ne ne meter range what I'm showing you here is some kind of string is a function of size in a big tribute to Professor Michael here because I did all this work at Stanford and a lot of other people contributed to to this understanding of the smaller is stronger size effect the idea here is this if you take common metals like gold and silver nickel and copper we all know them to be very malleable right in their single crystal in form and now reduce their dimensions to about $200.00 nanometers gold for example is going to become as strong as steel that's just by the virtue of the size reduction.

[00:12:21]
Now take exactly the same metals and deposit them using different techniques you guys all work in the grain right so you sputters use p.c. to use all these other techniques now the same metals lead them Nicole gold copper and other microstructure is nano Crystal and all of a sudden this effect is reversed the smaller dimensions give rise to weaker string this is a more modest size effect is not on the log log plot but it's still important to recognize that when you deposit the same material using a different technique it leads to as it leads to a different atomic level microstructure So the size reduction has the opposite effect so now you have this the tuning knob and strength but the side effect is actually more powerful than that if you look at a glass so you all know what happens if you try to pull on a glass rod or compress a glass rod So what happens this is not a ridge or in fact was right not a rhetorical question so here is some kind of a glass this is a special glass it's a metallic glass and we're going to pull on it this is about 159 meters in diameter and watch what happens here just in this region so it's the glass rod that's been reduced to $150.00 nanometers and watch here it's developing a neck and it's extending by more than a 100 percent before it fails glasses are not supposed to do that right here is my last example boy Ok so the last example is a single unit cell and one of these and a lot is that's made out of titanium nitride very brittle ceramic what happens to these beams It's like bending a piece of chalk I think we all have them to pieces just before and this is not where the piece of chalk would do right in fact if you were to calculate the stress here that means the stress and then to calculate the tensile component is about $1.00 g.p.a. ceramics are not supposed to stay intact when you apply 1.7 get the scale of tension to them but this guy is not only not failing but it's not failing over many many many cycles.

[00:14:13]
So the takeaway message here is that sometimes materials get stronger sometimes they get weaker sometimes you can suppress brittle failure but all of these effects emerge only at the nano scale so the big question was to us at least can we somehow harness all of these beneficial side effects and then proliferate them into the larger scale so that we can make macroscopic materials and that's where the 3 d. architecting comes into play so just a few words on how we write these architectures so I imagine many of you are familiar with u.v. photos out of the right so you have a mask you have light shining through it and then you can define the features because of the wavelength of the light being so small now the problem with your view is obviously is that it has these diffraction cones so yes you can resolve small scale features but no matter what you inevitably limited by the resolution of the Divergent groans in contrast to a 2 step 2 I think that's called 2 photons aggregate So you use a lower you use a much weaker laser in the infrared but you flood the pulse the ball each each pulse is flooded with photons so the constructive interference of exactly 2 photons gives you the precision that is not possible to attain with the u.v. light in so that now you concentrate the light in the intensity all in this tiny Levoxyl in this voxel is then rostered through space in 3 dimensions to write whatever it is that you want to write in 3 d. so I'm choosing to show you this periodic structure here but it doesn't have to be I can write a nano version of each one of you in the same amount of time so complexity comes for free in this technique it's slow but you can write very country complex structures now after we're right the scaffold which here happens to be a lot is but doesn't have to be and we're not necessarily going to work just with this material because usually it's made out of some photo photosensitive resin but we're ultimately interested in some other materials of maybe a ceramic maybe a metal maybe a semiconductor something like that so we then deposit the material we actually care about all around the scaffold and then we expose 2 of the sides.

[00:16:13]
Usually the form factors of our samples or cubes or something like that so I have 6 faces we exposed 2 of them in that provide access into the scaffold and so we etched out the scaffold that was so paid taking We written and so what you end up with is a hollow replica of what we had designed but now the largest dimension of the solid in this material is just the wall thickness so it's just another metres so to you it looks like a solid club that sitting on your hand but anywhere you look it's only 10 meters All right so here's a little video of the real time exposure so this instrument is called bananas banana scribe instrument and so on and scribe is the company that makes them in the process called 2 photons are the maybe you have one do you when you.

[00:16:59]
Forget the video them all right so now everything will work with this kind of plane and even though you can see the I guess some of the samples you can see with your eye or in the optical microscope you can see the detail and you can see the features with your with using any other technique so this is the instrument the in-situ instrument that we generally utilize so without all of these vessels themselves it would just be in a c.m. chamber so it allows you to see the material and then we have a nano mechanical module that's very similar to a nano inventor and then we also have the electrical measurement capability so we can do electro chemical experiments that connected to a potential set in the wheels of a cryo module and all these different Inquisition devices so we can really poke and pull and push and do a variety of different.

[00:17:46]
Defamation experiments and torture samples as we see fit all right so here's the big hypothesis right Kim we're proliferate the nano size effect on to the larger scale. So what I'm showing you here is an octet Manolis what I mean by an octet is that it's a particular type of cellular solid that's supposed to be very stiff stretching dominated Ok the hollow tube very much very very thin walled So it's 50 nanometers thick walls right it's me not entirely over Lumina we use the deposition technique called ale Vietnam a clear definition in the algae community they say yeah we can absolutely deposit anything you want as long as it's aluminum so this is aluminum it's an aluminum nano nano letters and we're going to compress it right so think of your coffee mug with a severe case of osteoporosis Now if you were going to so what do you think is going to happen Ok you're a professor type person right others and students what do you think is going to happen happen Dr very brittle varies and all material is going to crack right or crash and die so let's see what happens so we're compressing it compressing it and bam it shatters it doesn't recover it does exactly what you said it was going to do we have a little nano seminarian our instruments this guy is dead he's done Ok there's one thing that is in fact positive let's look at the stress train data here the stress exceeds $25.00 mph something that's 99 percent error shouldn't have anything be measured in mega units right it could be k.p.h. could be just as Galv $25.00 mph is not small so there's still strong well it so let's repeat the same experiment but now we're going to reduce the wall thickness by a factor of 5 Ok so the walls are no longer 50 nanometers which is already pretty small by most people's standards but now the walls are 10 and a metre so it's a lower lip relative density it should be a weaker material right it should question the even more so let's see what happens over compressing it in this work is so 5 years ago now so we're compressing it in that's going to crush and does it any moment now we're waiting for it to do it and and wait and it thinks it's a sponge.

[00:19:57]
So this ceramic is going to be very confused but if this doesn't prove to you that there is in fact a size effect I should just fly home right now so you know we sort of fully recovers right but since that time we've done a lot of work to demonstrate that this actually prevails in complex geometries in non-periodic architectures and over multiple length scales so what you're looking at here is this this is a by continuous surface we call it late nano labyrinthine material also very much not too based So there are no more and beams and you can see it fully recovers over many many cycles and it wrinkles here and you would expect it to crack there but it doesn't and the defamation mechanism is different because here the whole structure gets compressed but here you can see the bottom part doesn't even know that it's being deformed in other example is that the density doesn't have to be homogeneous look here the bottom part is clearly much more open so has a lower relative density in the top part is much more dense in so what you can see is that this part deformed 1st it's very predictable very sick when shot so you can pre-build trajectories of where you want your defamation to go so you can actually control it you can imagine that some Same could hold for fracture right so you can control your progress propagation.

[00:21:12]
Here's here's one of these fractal nano lot is where each beam is now comprised of something similar unit cells like a crane or like a truss the Eiffel Tower right so watch what happens when you compress this guy it kinda sucks and it forms these very in a homogeneous homogeneous kingsize you'd expect a crack at any moment now in trying to hurt me every time I watch the video in it's fully recovered so what that tells us is that this size effect is more powerful than the complex it can override the complex tractate the complex architectures in it's not married to period is city so you can in fact utilize it in all the variety in all sorts of different architectures without having to rely on structural.

[00:21:52]
Structural period of city. Now would you like to know why this happened why why do they recovery that we also wanted to know 5 years ago so when one of the different family mechanisms and beams because those lattices were all being based Well this it some teams can recover by this phenomenon called buckling right mechanical phenomenon so but we don't have this or the buckling we don't have this kind of wall fracture what we have is this kind of buckling it's a hollow tube so it's called Shell buckling So that's an elastic instability and when you look at the defamation there are regions in the octets geometry that experience high tensile strength stress which is where you would expect failure to occur so you can set the failure criterion as such as a competition between local buckling stress and the yielding which of course in this case is just from so if you look at each tube here and says Suppose the wall thickness here is that the thickness in the radius is 8 so it turns out that there's a critical ratio of 2 over a basically your soda can right house than the shell is that for this particular material if you exceed this value then your material fracture if you're below this value the new material will buckle so let's see that's what happens so for the 10 meter wall thickness this is a lot of after the compression and you wouldn't be able to tell the difference not only because it's a crappy v.g.a. connector but also because it fully recovers afterwards you can see from the data that it fully recovered each time rate and as we start to increase the wall thickness you can see that affirmation becomes much more stochastic much more jerky and this guy doesn't even recover at all as we approach and exceed the critical wall thickness well so that this right the explanation lies in mechanics so that means that if we were to build a building whose wall thickness is 10 centimeters and whose radius is 3 meters in the whole thing is 3 kilometers by 3 kilometers by 3 kilometers the same thing can happen right because the ratios are scale free me can exist jail free since I'm Jim comes in stuff on a building like this it'll just bring right back up I'm sure there's a building like that in Singapore.

[00:23:59]
You're laughing because hopefully in your mind you're like obviously this would never happen and you're right why not how does the size effect manifest itself here when it's not explicitly built into it why is it that a 10 cent the ratio is preserved right here right we're going to we're going on the ratio What is it about the difference of length scales that leads to to this building never recovering from a giant stepping in any ideas you are to who just said that you're totally right that's exactly it there are so many flaws in something that 10 centimeters thick right 1st of all the probability of finding a gazillion defects and by the fact I mean micro structural micro voids a great grain boundaries phase boundaries inclusion memories anything there are so many of them inside something that 10 centimeters say that upon the application of the global stress or global strength the local stresses in front of these effects will be much higher than the material failure strength so before any of these tubes have a chance to buckle the material will locally fail at those and then the whole structure will crumble in contrast to something that's only 10 m. meters thick 1st of all the size of the largest flaw is centimeters by no means of our weak flaw Secondly the probability of finding a flaw is much much lower so the side effect manifested some here by our ability to make much more perfect materials when they're reduced to the nano size dimension if I were to fill this room with something that's 10 meters thick that's architected to this volume you will have a much higher quality material than if you were to make a monolithic the same material monolithic that fills up this room so that's how the size is that.

[00:25:46]
Emerges here all right so the question was did we get into the white space originally right and how close can we get to the theory well so this is the space from before here's our say radical Maxima and all of these eliminated ellipses here that's all of our data since 2014.

[00:26:04]
So when you can see if you're really paying attention the 1st thing you see is that our scaling is not so good you can see that the strength falls off with density much more rapidly than theory predicts right. Well maybe I'll tell you about the victories 1st hey we made it in the white space so by going hierarchic over here we actually bought ourselves 2 more orders of magnitude reduction in density in relative density that was a big accomplishment in the white space and I already as I mentioned even a small venturing into the white face represents a pretty substantial increase in strength so we certainly we certainly.

[00:26:38]
Recognize that this really is a way to get into this the white space by combining the material sciences back in the architecture now what you also are saying is that the scaling is not so good because for example for the stretching dominated solids the scaling should be linear for and that's what we're showing you here but is there a cool maximum and here we didn't really know how to address this problem but that it seems like we were able to fix it here so all of these things we can do now much better and perhaps some of you can do much better and we actually can do much better with carbon so it turns out that this is some of our recent work so it turns out that 1st of all we're not so close to the theoretical limit right but what exactly theoretical limit so when you're starting to deal with these materials that are really 5 nanometers they can that are architected you really have to bring to question what exactly this theoretical strengths mean I mean it's linked to the bond strength of course but of those length scales you really can't be comparing something that's bulk to something that's the end in sight so for carbon materials of course with so much work on graphene in carbon the energy of it's actually a range of all these different carbon materials and what we're discovering is that in some of our recent work is that of higher densities we can in fact approach theoretical limit so that was a big That was a nice discovery that we that we just put out there so.

[00:27:56]
Right so we can design new classes of materials now I want to show you that the reason for this scaling not being so good is because when I told you that they recover by a 100 percent I did I might have exaggerated a little bit they actually did recover by 99.2 percent so the little bit of damage accumulation exists in the nodes it's the nodes that hurt you so the junctions right where you have multiple beams coming into coming into play well so this is how we solve this problem so this is very recent work this is all a publish my young student we couldn't specially He didn't even conceived of this I'll tell you I really interesting story we had a random guy who's 76 years old email us with this idea and he said you know I've been through the printing in my garage and I came up with these woven things you're crazy guy but he actually flew out to visit as I felt that he just some random enthusiasm and then my and then we adapted his design and give him full credit and we didn't need a printer that and so now look at these architectures there you can make exactly the same geometry you can make them diamond shaped like that shape but now they're interwoven and look at this there's not a single node not a single node and there's this effectively like a him right and what we can do we can push on a rope so we can now strain these Look they're all they're all interwoven writes their spirals so now we're tensing it in we can tend to do more than 30 percent screen and then we can bring a bed down and then we can compress it in the same experiment so my my really good colleague Dave Barr told me once you know don't ever push on a rope so I can't wait to send this video to him because we're clearly pushing on a rope here so what these architectures show you is that 1st of all they're fully recover Bowl secondly that they provide resistance both in tension and the compression is that in that you cannot destroy them so this is very recent work and we were kind of excited to put it out there.

[00:29:50]
Right so if we were to plot the failure strain as a function of stress so this is this is kind of the boundary so that I guess here so all of the work that's been proposed or that's been put out there up to this point you can see that they fail at the relatively low string you can't really put them in tension very much but these architectures already offer quite a bit a lot more of a chance all the killing.

[00:30:12]
Ok so how many of them especially great of it right perfect All right so how many of you are mechanics people none of them Ok cloudier this is for you. Know so one type of experiments that we're very interested in is fracture because the way people do fracture you just somebody just told us that failure occurs at flaws right well how do you even conceive of fracture in a material that's made out entirely of flaws right so when you zoom out it looks like a material but when you zoom in it's made out entirely of flaws they're all flaws and defects and all sorts of things so we pre-build defects into these architectures and you can see it so here's our flaw this is the not so that's how you do fracture experiments you know we orienting this flaw at different angles with respect to loading direction and we want to see how it feels and so what we're seeing is that the feel your stress is actually behaving pretty well we want to understand can we use continuum theories to describe failure of these materials so they are not just you know a lot is just the strength of course is highest and now we can see that the stress is doing what it's supposed to be doing it's it's falling down falling off and of course it's lowest for the horizontally oriented notch because that's what's nominal loading it so that's predictable.

[00:31:35]
What isn't predictable is the failure trajectory look at where the failure surfaces always are in bring this here they're always they're always orthogonal to the loading direction in the always occur along the nodal plane so. It's both continuum and discrete there's some kind of could this creek is seen in duality because the stresses can be predicted and give you predicted by the models and this is this is the work of our collaborators in Singapore.

[00:32:03]
But the crack trajectory cannot because it's over no matter what the knot orientation is it's always fails along the normal plane and the way you can rationalize this is of course when you 1st start pulling on it it's a true fracture problem because you have these concentrations at the root here but as soon as you start propagating the crack it stops being a fracture problem because at that point you're just rupturing the plane of the node so all you are measuring is the failure strength of these noble planes in so that's that's exactly what it is this is not the best experiment it is hey we tried it so this is not the best fracture experiment but that may explain why fracture always occurs along the no plane so we can do much better now out in the course of doing this work we discovered that actually we had another white space we didn't mean to we didn't want we didn't anticipate too but if you plot specific temple strength so this is tensile strength normalized by density and compared to many other existing natural materials So for example it seems like it's better than balsa wood in so you can start you can imagine being building higher or sports equipment out of these kinds of materials in these are all of our fully dense material so the relative density of ones you can see where Liam here as well.

[00:33:13]
So we had it I had a summer student who decided that this fracture experiment needs to be done much better much more correctly so they designed this single edge not temples that's them and so you can see that now we can we can write these notches directly in the architectures instead of instead of making them in the center and so this is the proper geometry and so you can see we've already made these and in fact we already started started doing some of these experiments and the most challenging part about this is tracking the crack propagation because you need to track the cracked opening displacement in so how can you trace this sojourn So this was a particularly talented undergrad summer undergraduate student so we're finally learning how to do these tractor experiments correctly.

[00:33:56]
Since I have half an hour left this is the time to take a dramatic pause. In a lot of information so far. And what I'm going to do is I'm going to take a vote. So we'll work on a lot of different projects not all of them are mechanics of course and I'm going to offer you a choice of these 3 stories and I can either tell you more mechanics and what I mean by that is so far all the experiments that you've seen are quasi static so everything the compression and the tension is being done slowly so I can show you some of the results of our dynamic impact experiments I can show you more in Fracture I can also tell you a story about batteries and.

[00:34:43]
Various aspects of battery how these are connected materials may serve as battery electrodes for example or just some electric chemistry or I can tell you about additive manufacturing possibly this is this is a very exciting area in our group right now because my students are really becoming chemists and they have figured out how to do materials synthesis of all sorts of different exotic materials from metals to metal oxides to glassy carbon and I'll tell you about that to all these exotic materials for example one of my students just printed listen call that that just 3 d. printed cathode himself.

[00:35:18]
And all the different types of chemically derived materials using using these these kinds of architected approach so why don't you take a look at this and I'm going to take a vote the rules are you have to do it once and only once you can't not vote and you can't vote for more than one thing.

[00:35:38]
All right Ok who wants to hear more than can I mean Jennings is really popular here who wants to hear about batteries Ok And who wants to hear about that in a manufacturing. All right I think we've made our choice All right those of you who are interested in batteries maybe I can talk to you once a month after I will actually I will tell you a little bit about batteries in the in.

[00:36:07]
Me and the manufacturing here Ok so do you so you have a nano scribe here right so what is the most. Proprietory apart you know they don't charge you for a service contract but they do get you somehow so what is it that you pay for their visits because everybody wants the IP to persist right i.b.s. or is it not if you are clever Cal Tech students because then you say I don't want to do that I want to design my own resist right and this is all acrylic chemistry has a lock really based So when we 1st started I started this work I guess it was only last year but feels like it was a long time ago one of my students Andre questioned the chemistry and he said well wait a minute maybe we could design our own resistance as long as they work with the photo initiator and with a acrylic chemistry maybe we can start making other materials so in this work he try to know if you can see he trapped a nickel ion So there's a metal ions that's trapped in an organic shell and what you can then do is it can undergo a ligand exchange reaction in so that you can what you can do is you can attach the cruelly groups on to the metal ions and then possibly right with that and so this was the idea so you put this nickel actually it and the photo initiator and all this other chemistry into the resin and at that time we were very surprised it weren't so he was able to write a lot is using this kind of resin so this is a nickel containing resin and look at the unit cell size here it's 10 microns right so that's pretty big in the beams are 2 microns and look at really really worked so but what we really wanted was a nickel and a lot as we didn't want a nickel containing the letters we actually wanted something that's made out entirely of nickel How do you do that how do you get rid of all everything else and there was just organic What do you do with organics like when you go camping What do you do with all their gimmicks but you burn that's right so I fancy word for burning in sciences pyrolysis right so we paralyze it under certain conditions for example with no oxygen right so that you can actually burn the mark here what we end up with.

[00:38:10]
It's something that's much smaller much smaller drinks tremendously when you pile I think and look at the n.f.l. size here this is a 2 micron units all sized right and each beam diameter is $500.00 nanometers So my student Andre figured out how to additively manufactured nickel at the resolution that had never been shown before well so then the question is yeah whatever how a nickel is this nickel so then we have to do some serious characterizations of this is some e.d.s. just to reveal the chemistry but the quality of the resolution of the screen is not so good but to show you that the it's 92 percent nickel so it's very high nickel content and there's a little bit of oxygen and a little bit of carbon and then how many of you work with a team how many of you use that do you have a lot of appreciation for preparing samples for a team right how do you take a tiny loan in a lot of input in the t.m. grid so that you can evaluate its microstructure very necessary but so much easier said than done right so sometimes we use a focused I am being right to left things out and glue them but then there's gallium ions in the plot and all that stuff so I'm great of course was clever about this and he said well wait a minute I can write a grid of my own resists on the t.m. grid and just hope that one of the beams will happen to go over one of the holes in the grid so this is a wholly carbon grid and then look one of the really did so this is as unadulterated material as you can get and it's subjected to exactly the same process so what we're able to do with the a.t.m. micro structural analysis on this nickel and what we learned is that it's basically all nano Crystal and nickel for the most part and there's little bit of nickel Carbide in the little bit of nickel oxide that most likely formed after we prepare the team sample to get on the surface this is very nano crystal and you can see the grain size about 20 nanometers.

[00:39:56]
So now let's think about this this is nano Crystal and nickel with some pores in it and some impurities so that is the worst quality material that you can pass as far as metals are going to turn red those of you that like metals I mean you don't want your metals to be Nana gristly because then they're brittle you don't want them to be poor is because then they're brittle and not very strong and you definitely don't want to contain impurities right so let's see we should get some really bad mechanical properties so let's see how this one does.

[00:40:23]
So we're compressing it compressing and you can see the forming very much like a cellular solid wood which is what we would expect so that's good and again if you could see the scale we're in the m.p.a. range so for this relative density of about 30 percent this is good but now let's see how it compares to all the other additive manufacturing processes for metals he vailable today so again if you could see this you would know that this says specific tensile strength in this says beam diameter and this are all the additive manufacturing processes for metals developed today this is actually a very surprising learning for us because where you can see is that as you reduce the dimensions from a millimeter to 100 microns you lose 2 orders of magnitude in strength for all the other processes that don't look at our data so just for all the processes today this is laser most selective laser melting this is laser ablations these are all the different additive manufacturing processes that that are known that are commercial or even in the lab whatever is published in there is available we plotted here and you can see that most techniques really choke when you get down to the microns when you go from a millimeter to a microns to 100 microns and so now here's where our work is so what if we were to follow the scaling we should end up somewhere 7 orders of magnitude below we had to put in a break and here are dimensions here are $500.00 nanometers right so 123.

[00:41:49]
Orders of magnitude below what that what is possible today and look where our strengths are the strengths of these seemingly bad microstructure nickel are as high as something that's that's 500 microns in the additive manufacturing world and so what that tells us is that at the nano scale it's Ok to have defects it's Ok for your microstructure in the classical sense to be not so good because you still don't take a huge sacrifice in strength.

[00:42:19]
So added manufacturing process is much more forgiving as the manufacturing process is much more forgiving in terms of the mechanical problems All right so then once we made this breakthrough saying look we don't have to buy this IP differences we can actually my own my students really went to town so I have this student Darryl who literally become a chemist so he said Well Okrent acrylic chemistry and our colleagues in general they're actually quite toxic in their hard to work with so he developed hot aqueous types and synthesis in hydrogen driven reactions because turns out you could just work with metal salts in a quick solutions working with water much more green much more.

[00:42:57]
Much more environmentally friendly so we started this work was ink oxide So if this is a good painting aqueous photo resin in the question is can we so well in this illusions that that we can swell things in and out of it and so now you have the zinc eye and you have the n o 3 Ion and then this is the develop these are the different developers and photo initiators in this was the big question can we still photo initiate a radical polarisation radical driven free radical driven polymerization So here's the thing containing 3 d. structure and then you can't really call it pyrolysis anymore because you're not available izing there again excite in the same in the same sense but it's still a thermal treatment because then you can undergo the reactions to convert the structure it's an actual chemical reaction to then convert the structure into zinc oxide So let's see so again doing the same kinds of C.D.'s e.d.s. and then x r d and looks like we really did make zinc oxide doing some t.m. on these shows that that it's nano crystal and you can see the 3 rings right here if you are able to see the histogram the grain size here is 5 nanometers this is just a ceramic thing but that is a ceramic so nanocrystals not force very very high quality actually but but not in a crystal Nevertheless in zinc oxide So that's pretty exciting right like we haven't nobody had really demonstrated the hydrogen based synthesis in them 3 d. architecting So what's it good for.

[00:44:20]
What is good for what is a very common use of things proper some property can somebody say yes who just said that yes it's be a Selectric right so ph electric materials are used in memes they're very exciting and everything but what do you know that Haas to be true for zinc oxide to B.P.'s electric or for beef or anything what does it what do we what has to happen for its microstructure crystalline but not just whistlin it has to be single crystalline right because the way the A's are interested in works is that when you apply a strain or mechanical strain of some kind of creates a polarization effect so that that you're polarizing the crystal structure such that you now have the the positive charges on one side the negative charges on the other side and creates a field so for the strain to induce voltage that has to be a single crystal in the other way around by applying a bias right it shrinks and then strength it has to be single crystal and did you kind of process that this is so not single crystal and it's not even funny this is a 5 nanometer range that we absolutely should not see any of these electricity here right Ok so here's the zinc oxide resin so don't look at the bottom part yet this is just the resin think iron containing resin nothing outside yet and so what we're doing is this experiment where we contact you can with an enemy chemical tip and then we're applying a displacement and we're measuring the voltage so this is the displacement it's a step function so we're holding it and then we're compressing it and here's the voltage and nothing happens just like we would expect and then once we converted to zinc oxide we are doing exactly the same loading profile we do the displacement and the voltage so it's very much b.s. electric So again at Banana scale when you make these architected materials you can override think it's a lot more forgiving so does anybody want to offer any ideas why we might still get these electricity so this this is the schematic that shows you why is it still be a selector given those nonexistent.

[00:46:22]
It shouldn't be right did everyone understand the argument it should be because it's not a single crystal but it's nanograms than right so all the green should cancel themselves out this train feels like they don't align the Green think it's a very very tiny displacement it's very much a last week so nothing rotates nothing new nothing nothing changes orientation as a result of this this step is.

[00:46:52]
Something like 29 meters 10 to 20 is very small and in my right 20200 nanometers still very small for something that's microns tall 200 nanometers very very last low strain yet 200. So one of the hypothesis actually was that we falsely believe that 1st was that maybe so he's electricity in zinc oxide occurs only along the c.x. so we thought that maybe this t.m. image means that there are randomly oriented out of plane but see axis aligned along the plane because they were deforming were straining it vertically right so maybe we have the c.s. is all aligned in these greens and so maybe that's why in so upon further analysis we discovered that that's not true either they're actually truly randomly oriented so I asked my student Rebecca who took over this project to just do a simple math experiment what if you took this very small.

[00:47:46]
Diameter so each beam has a very small diameter something like one micron or 5 microns or something that in just populated with a randomly oriented green the average should be 0 right the average change in voltage should be 0 but now we know what the what the bulk values of p.s. Electric how much Milli how many millivolts you should.

[00:48:04]
Millivolts per nanometer displacement right so how much of the electric response in voltage you should get in so what ends up happening is that the average is 0 but it's plus or minus depending on the randomly oriented grains right so in the integral it ends up being 0 but there is a plus and minus air and so it turns out the as you reduce the dimensions it actually goes from bulk value down to a much much much lower value in this value of about 40 millivolts is in fact what we would expect for something that's one micron diameter that contains 200000 grains so 200000 randomly oriented grains will naturally lead to some of them being aligned favorably in some other ones not so it's literally the rule of mixtures that's all that so no alignment but the randomness actually gives you just the smallest signal so we were we were pretty happy with the street we do in it with Yes that's right in so it's everywhere between minus 40 and plus 40 yet because you can also do tensions in that that was exactly so it was very satisfying to finally understand this and then to confirm it yeah.

[00:49:03]
All right so now I told you about the complex oxides re the hydrogen of a synthesis I told you about like in exchange reaction for metals now the simplest of this whole thing is really what if we don't add anything into it and just paralyze the material you can do this with a feed it so what we end up with is carbon right and so the carbon is actually very nice because we can 3 d. printed as well so this is a 3 d. printer not through $4.00 Odyssey type of a micro lattice and you can also do 2 photon without graffiti in the idea is that if you just burn it if you just fire lies an inert atmosphere with no oxygen you convert all of the organic polymer into a type of carbon and of course there are so many different kinds of carbon so this is some kind of a glass the carbon amorphous plastic garbage so we we worked on 2 different geometries So this is octet that the one that seems to be the going the most popular geometry and this is an iso truss geometry that's supposed to be the this differs the strongest in every way and then looked at the microstructure this is the a.t.m. image of one of these the beam dimensions after pyrolysis about 500 meters 400 so what does this image tell you you look at it and you say what is this any ideas what does it look like.

[00:50:21]
Look the more of a story those of you who are familiar with with carbon actually it looks glassy So they're little ribbons if you if you could resolve it well Ok so more of the glass the carbon Ok what do we know about carbon is very much not deformable right so let's go back to the basics let's make some pillars just to understand the material strength here so one thing that carbon really really really doesn't like to do is to be compressed right we do not compress the carbon So here's one of these carbon micro pillars so it's about 2 microns in diameter and we're compressing the carbon compressed into carbon we're expecting it to shatter at any point.

[00:50:58]
And look at the stress here this is we're already at 5 g.p.a. g.p.a. and he really really is not enjoying himself at the moment he's really not interested in being compressed but he's not feeling and he's not failing until you get to about 30 percent screen this is carbon very very stiff glassy carbon we get to about 7 g.p.a. in a fully recovers this carbon things that it's rubber Ok I don't know what well actually I do know why it's things that it's rubber but but this is very surprising please tell me that you're surprised so when you take your pencil pencil lead right you're not going to be able to compress it like this so something's happening so terms of the glassy carbon is actually a very interesting material because when you look at the micro structure it looks glossy an amorphous but when you zoom in you actually see that there are a little curled graphene f.p. to hybridize sentence fragment and they are about 3 angstroms apart they're curled and they provide a little bit of free space here so those of you who are familiar with metallic glasses it's the distribution of free volume here and so we work with our collaborators that chin going University who are able to model this a molecular dynamics and you can actually create exactly that they reproduce this microstructure turns out to be about 80 percent as b. to hybridize carbon and 20 percent as the 3 hybridized carbon and you can create that you can actually replicate these little grafting segments in create a replica of this Mike.

[00:52:21]
Structure and so it turns out that when you model this invention we did Thameslink oppressive experiment it makes sense why you get that much tensile and compressive that Filippi because it's these local regions that undergo small shear transformations locally to accommodate the globally applied stress amazingly enough they're just riveted all throughout the filler volume to accommodate the global stress and strain and to to enable the mechanism where you have that much expendability intention and also in compression and he was a collaborator was so excited that they chose his image for the cover so I have to show it.

[00:52:56]
Ok Now there's a new I have 5 minutes left 7 Awesome thank you I just want to tell you one last chapter of the story Ok so I told you I was going to tell you about batteries but the story is kind of exciting and very relevant right now and in fact it's our collaboration with Professor dealer who's sitting right there so I had one student who is now a doctor Dr shouting Shaw who is very curious about these materials and their character and batteries and all of this is the culmination of a lot of years of work and so he said well I'm going to write the Xana lettuces but I'm going to be much more clever about this because we know a lot about them so he wrote a track and all micro letters and he wrote it out of this resist called pos which is effectively a glass like resist them key coated with a copper layer and then he coated with silicon now so it kind of course is the active material for batteries for lithium ion batteries for lithium ion transport in the copper can serve as a current collector and so these types of geometries lend themselves to lift the ations and use the Asians so let's see what what happens so this is a tracking all micro lattice you can see the scale by here is 50 microns right we're looking at the top down one of these in all charging is going to do is now listed a particular rate case a look at what happens so here's a video.

[00:54:12]
And while it forms these crazy buckling patterns so to me they look like violin patterns because I'm a musician you guys can think of them as sinusoidal pattern so what you see here here's the top you imagine here's a tilted image sort of top of an image and fills that image you can see that there vertical post and so they're subject to a little bit of torque but in each plane we see these violence I knew so little type patterns that are self-consistent in the chair right and so in response to lift the edition it seems like globally there's no expansion or contraction which is a big problem with silicon based was a my batteries right so it expands and contracts so by depositing 100 nanometers sent sick nano ribbon of silicon all around by wrapping it effectively around this architecture we can actually bypass any kind of failure so 1st of all the anode It's effectively an anode dozens doesn't crack as a result of litigation and it accommodates all of this less the ation through local buckling it so any This really this really is kind of interesting discovery All right so just to show you these are not just pretty pictures this is in fact a battery so when you cut off all the voltages at about point 6 balls it actually works pretty well so this is a voltage versus specific capacity and so you can see that each one of these points were actually plotting to give providing images of litigation and deal with the ation in this is what the structure looks like after 10th litigation deal of ph and I think it took it to over 50 cycles and as long as you're kind of voltages at about point 6 volts we're Ok so it looks like we're losing just a little bit of specific capacity so it's a spectral battery it's not a great battery it's not a terrible battery it's Ok and Ok battery.

[00:55:51]
So because it was just an Ok battery and tell that students don't like Ok anything he said forget the battery part so instead we look to our collaborators here in fact a cloudy deal it was group and we said Alright what can we learn from these so what we learn from these is that sometimes these nodes cooperate in they rotate in opposite directions which leads to this really nice buckled back and pattern where you have opposite concavity right so you can form the violin and sometimes they get frustrated because they're random defects right any fabricate you all know you fabricate things any fabrication process leads to the fact that a lot of these small imperfections may lead to the 2 nodes co-rotating and when they're correlating this beam gets really frustrated and then it undergoes a 2nd mode buckling and so Claudio's team develops these very clever models that include the kinetics that include billet the ation and the different elastic properties of litigated silicon and what do you know he was able to model these as a function of lithium concentration to demonstrate that there's actually a parameter space that describes these and predicts the yielding conditions as a function of living with concentration and the buckling profile as your material yields for example is of plastic buckling not elastic buckling So it's not once you remove it it's not going to snap back like an elastic instability but what challenging was really interested in is understanding these 2 different phenomena so when they cooperative loo buckle you form almost one domain right here and when they don't collaborate and they buckle in the in the frustrated pattern it forms a domain boundary so what these are our actual digitized images of the s.c.m. images of the different domain formations so you can see that a different litigation rates as we increasing the rate we're forming smaller domains right in so it's almost like don't mean switching in response to our rate so those anybody speak physics or.

[00:57:43]
So what do we what does that remind you of if anybody what does that remind you of when you have domain switching under differing rate which is translates into a thermal parameter yes in what is a model called. The icing model right so the n.c.a. pheromone not expecting right so what tell me she looked at is the coupling parameter between any 2 adjacent nodes and then he looked at the probability distribution of what kind of switching you would expect depending on the particular distribution of defect so this is the real data experimental data in these Monte Carlo simulations where you now have in this icing model instead of using the rate is your parameter we now have k.t. is a parameters of course is the thermal normally activated type domain switching and it seems like the trend is very much well preserved so this seems like there is that it has a similar effect that the litigation rate in the thermal ramping rate have a similar effect on the domain formation and so what we learned is that when you do that you can actually now play around with these architectures and you can start with the tracking Alatas and then as it converts into a list and 3 silicon material it undergoes this transformation and so now when we plot the dispersion the acoustic dispersion relations we can see that therefore Nanuk been gaps that are emerging So now you can start working with a critic frequency and modify your materials like that and then joshing really went to town he said well you know what you can actually induce all these other defamation mechanisms in fact you can make rotation occur you can have out of plane buckling you can have expansion you can have bending and so this is what's so so he really demonstrated that the degrees of freedom can be fully controlled and give me programs So speaking of programming this work was really all about defects what challenging managed to show is that if you understand the role of defects and if you properly account for them you can use them to your advantage to program because now you understand how your material response to the distribution of effects so what I'm going to show you is the same track going to micro lattice.

[00:59:46]
But now he programs some defect into them so the one that you saw before with all the buckling it was there was nothing programmed in there they were just randomly distributed defects in that's what led to all these different domain from agents once he understood how the demands were formed he's now playing with these effects and materials you know are just like people it's the imperfections that make them interesting so something is programmed in there and I'm going to show you the Lucien All right so here we go so now it's Lizzie ating and it's Lucy ating and it's buckling in hopefully you can now see what's emerging here that it bears a lot of resemblance to the little logo here on the right so anyway and we're excited actually because this work is coming out in nature next week next Wednesday I think or something that's so I hope that you will find it interesting all right so this is a good time to finish so if you walk out of this room saying she showed us all these images and videos and have no idea what that was all about please remember this is the message does that take a message if you are clever about 3 things just 3 things.

[01:00:48]
1st is the atomic level micro structure of your material right is that amorphous is it single crystalline is an anarchist and what is it's made out of in terms of the atomic arrangement the 2nd one is how your particular material response to the nano size affects all materials exhibit size effects but is very unique in specific to the class of materials that you're working with and you need to really know that and the 3rd one of course is the specific architecture How are you going to arrange these nano building blocks into what geometry do you does it give you an advantage or not if you're clever about those 3 things you can create entirely new classes of materials with not only on precedent properties but with unprecedented combinations of properties that were never possible before in the reason why you may want to do that the so that we our kids live in the world where we don't have the hearing aid because we can just write one we can just write a cochlear bone and put it directly in your ear so and then we've already written actually bone scuffles to create bone tissue so that your i Phone $83.00 can hold its charge for a year without needing to be recharged because we can somebody can make much better batteries out of these of course the balloons we don't need to be to be filling them up with helium anymore because that's a actually a pretty precious resource.

[01:02:02]
Because vacuum is lighter than air so we can just evacuate them right so as soon as we figure out how to make our analyzers not port we can start making those who know them. Christmas ornaments course will never shatter any more right the pressing brittle failure and I think that this is probably the most the most exciting pursuit so we all like Brown.

[01:02:23]
So our latest pursuit is chocolate nano lettuce is for John 100 percent taste 99.9 percent air and point one percentile or. So without a huge thanks to you this is my this is the part of my group you can see to work very hard on this beach in Malibu.

[01:02:41]
And all the wonderful funding sources and thank you all very much. I'm happy to answer questions if those are. True. Flew I don't play the food I played a. Couple of questions yes and. I think you should use them I. Think you just shout. Thank you. For. Your.

[01:05:07]
Thank you I really appreciate actually both questions very much so the 1st one is you're absolutely right it's a bad idea to build airplanes out of very light materials because they're too light you're absolutely the airplane was an example of a material that's very heavy but the true applications of these materials lie in small scale components something where precision is very important by medical devices for example and where size can be relatively small I think that they're not meant to be structural material.

[01:05:37]
Unless you're making a small device you're absolutely right we can't have winter been better to light began our airplanes that are too light so absolutely you're right they're not macro scale materials. For the thermal France where we actually have a paper that shows that the thermal conductivity can in fact be controlled and changed in these nano architected materials So for example it shows that for that they have comparable thermal Comic-Con activity as the space shuttle tiles which need to be extremely similar insulating at a much higher stiffness so the uncoupling that happens in the thermal properties is that so we did the omega 3 Omega measurements right so where you resist of heat them in so you can control your soul conduct of a t. But now as a much stiffer in a much stiffer structure so something that would have the same stiffness would have a much lower conduct city rate or something that would have a much lower stiffness which would have a higher productivity so it's the combination of of the mechanical stiffness write like a ceramic tile versus a very lightweight nano lattice and manipulate the thermal conductivity through that here thank you.

[01:06:59]
You're. Lucky you got it. Of doctor to. Work. On Me. And. I hope they're right and we're really on this machine. Or. Yes thank you that's also a great question so then the scribes are terrific terrific instrument for research. But it's really not manufacturable and you can't use this technology to manufacture samples so what we're pursuing now is how to scale the most is that precisely the question and you can't really use the 2 photon was obviously process for that because by the time you reference single voxel in 3 dimensional space no matter how many little galvanic mirrors you're using no matter which objective you're using you're just not going to be able to do that so we're using this but we developed this process for the for this particular approach using interference lithography a metal surface masks so the Nano scribe doesn't use any mask right and so that's where they have you know the objective come in and the patterning comes in that's why it's Ok to use it for very small samples complex by very small samples and so what we're doing in our group is we're actually using these methods are 1st mask that my colleague Professor fire owns group makes and we form a 3 d. interference pattern directly underneath in the planes in the space that's directly underneath the mask and you can transfer that pattern into a photo initiated will resist and then in a single laser exposure instead of making just one box where you actually make an entire field and then you separate through but because the Nano scribe is wonderful for little things yes if you want to manufacture a new thing it's not going to work very well but stick stick with your instrument it's often.

[01:09:20]
I really appreciate questions about defect Thank you. Any Like I said any manufacturing process leads to a natural distribution of defect so we actually do the systematic study if you take out one b. is that a defect does the structure know that or does your man architect the material know that what if you take out to be what if you take out an entire known so we actually did 1st of all for the mechanical defamation we did a study to show at what point will the defamation of the entire material be susceptible to to the presence of these types of defects and so we took out random things in other summer soon took out a beam took out 2 beams took out and beams right to go out the nose so what I'm showing the videos that I showed you I would say.

[01:10:01]
At this point there are the better samples of course but they're not the best one and that's the only one we're able to make so they're not better I mean obviously if something contains defects or bubbles or explosions we don't I'm not going to show you these results if we don't we learn from these results but we don't show them and then months we're able to perfect the process then those of the these are the typical results that would say and I'm not making this joke when in the paper you say typical what I mean by the only one that works.

[01:10:28]
They are pretty resilient against defects especially against manufacturing defects because if you think about it how many units all do you have in the structure rate so if you were to make a capacitor out of them where you want to minimize the distance between the plates and you only have a single layer I suspect that then you will be more susceptible to defect so I think you just make it specific to what what you are and use application is yeah but for the mechanical properties they seem to be particularly robust again even.

[01:11:06]
I guess. Yeah that's a very good observations there is quite a bit of there's quite a bit of history says anytime you compress these it's not elastic right so it's recoverable but there is always some kind of energy dissipation So we suspect that the dissipated comes from microfracture events in the nodes and so what ends up happening is that I'm the 1st defamation cycle you actually crack formed little micro cracks in all the nodes or maybe not in all of them but in many of them and then what happens after that is that they just open and close there are also some critical so they don't destroy the integrity of the entire structure but they certainly contribute to the energy dissipation right just because with each one you know you accumulate just a tiny bit more damage and then you're opening and closing the crack so some of it some energy is probably dissipated through through other mechanisms but most of the history says that you see is because upon loading your clothes in the cracks in the pond unloading that so that the Depending on the particular a local stress state that each node experiences right it undergoes it undergoes the crack opening and closing and that's why that's exactly right that's why this feeling was so poor That's exactly right because when you have the hollow tubes then you open up these defects and cracks it doesn't affect the can recover ability because globally your mentor it's still undergoes buck the buckling but your strength deteriorates at a higher rate than theoretically predicted That's why we went to the woman architectures so in the woman architectures you still hit see the history says which is now coming through friction and through other processes but now we don't see the scaling deteriorating so rapidly very good point.

[01:13:24]
Here I'll be here for the whole day so I'm happy to chat with all of you for. Your. Own Words. Thank you thanks.