[00:00:06]
>> So today Speaker it's a real pleasure to have with us. From the University of Pennsylvania. Got his undergraduate degree in theoretical physics from university. Suffer Moscow State University and then his Ph d. in physics and electrical engineering from Cal Tech and he then did a post doc in France as well as at Stanford before going to the University of Pennsylvania where he is the class of 1965 term assistant professor of Mechanical Engineering and Applied mechanics he's won a number of of awards for both research and teaching including the n.s.f. career ward and the pan engineering teaching award and so it's a real pleasure to have him with us.

[00:01:01]
Thank you David it's so real pleasure and honor to be here and you guys hear me in the back in my class I always ask people to see that over here and I'm going to do it but at least I want to make sure that you guys can hear me in the back everything all right wonderful So I'm here to tell about the work that we've been doing in our in our group for the last several years and we mostly focus on what we call plate mechanical man of materials which are very thin plates that are large enough that you can hold in your hands and have some unusual properties some going to talk about their mechanics and I'm going to talk about their applications these days but mostly focused and excited.

[00:01:41]
About the applications of these structures. Let's see Ok so it's going So who here has heard of mechanical Madam attorney. Some of you have so met metamaterials perhaps more people in general so it's kind of a buzzword great everybody wants to have their own metamaterials their optical electromagnetic ones and so the mechanical engineers want to have their own this well.

[00:02:06]
But as with many buzzwords the definition is not very clear so I kind of make up made up my own and everybody sort of wants to fit what they're doing inside the each buzzword that's out there so here's a definition that I like these these are artificial structures whose novel mechanical properties are defined by their carefully designs and controlled geometry which is often periodic and often at the macro scale but not in it micro scale as you say but not necessarily so you may have seen structures like this this is not my work this has been done several years ago h r l and Cal Tech and u.c. Irvine.

[00:02:49]
Is from jewelers Greer's group website she I guess she was here just last week or so. So these are lattices that a lot of people have been making and exploring their different properties if fabricated out the very thin films the result in structures can be remarkably lightweight about as as lightweight as air and recover from very large defamations.

[00:03:15]
For example you could do a compression test let's see the video works do a compression test go through very large strains that are on the order of one and still have a recovery that's almost complete or complete after you remove the the force that's pressing them in on them.

[00:03:35]
So these are cool properties that you don't normally observe. You're also able to get very high Younes module life for these very low densities right so generally materials that are solid and not structured of all would have densities on the order of thousands of kilograms per cubic meter and have very high module ally and friend you know yes ma So why in the order of hundreds of Giga pass calls but as you go lower you will often find that these fall off very quickly and in fact the theoretical maximum the best you can do is typically to go just linearly with the with the density so.

[00:04:16]
People have been working on getting into this theoretical maximum where things are scaling linearly with them city and these lattices that I was showing you can actually get there pretty well often these are 3 d. printed at the nano scale using tools like nano scribe. And then it's is basically a 2 Foden polymerization process and then they deposit a very thin film on top of the polymer and at your way the polymer.

[00:04:46]
Now that's a process that gives you a lot of control of the structure that you want to make but you made imagine that printing things that the nano scale is not particularly fast so the most of these structures have been limited in their size to less than a millimeter or so and when I was visiting Julio's a few years ago she showed me something that 3 d. printed with a nano scribe that you know was a few millimeters in size and so they took a week to print something like that so what we wanted to do is to focus on slightly different form factor there are many applications out there where you actually want to have a plate rather than a cube and one obvious thing is you know things that fly around for example they need winds this is from.

[00:05:33]
I'm sorry I'm blanking right Rob Wood's group of Harvard's who's making all of these robot bees and this is actually something that was done by my colleague at Penn and we're collaborating with him right now on a new generation of this structure that's called pickle lissome a which is the smallest self powered robot it's only about a quarter in the size of a quarter in size and that's the smallest rotorcraft that can fly on under its own power so many of these would require things that are impermeable plates.

[00:06:04]
And both Manta materials like the lattices that are shown cannot really be easily adapted to applications like this so a few years ago we started to make in such plates which were ultra thin typically we were working with 25 to a 100 nanometers sickness continuous plates that are however big enough to hold in your hands and here is an example.

[00:06:27]
From the down the line picture we've gone to our own flour so we we may plates that are macroscopic in size that are centimeters scale in size and there's a free standing aluminum oxide plate. If you look closely enough you see that it's not just the flat film that is 45 nanometers fake that's would not work for the reasons that explain in a 2nd but still these are big enough that you can hold in your hands and only about $100.00 at a mistake of the thinnest And so that's would be a 1000 times thinner than things like cling wrap or aluminum foil which were already typically think off as pretty thin films these are corrugated However as I said planar would not work so by corrugated were actually increasing their bend in stiffness if you fabricate a can Cleaver leg this one which is about a millimeter long out of a completely flat film you get something like this it just curled up on itself because however good you control your deposition process is typically you will have some great strain gradients in your deposits film which means that as soon as you release it it will want to curl up and in fact the thinner you make the film The smaller the radius of curvature typically becomes for for that curling that you observe So this actually curls up into essentially an ice cream cone you can even see a little fold here that's pretty much just like here so if you don't want that then you want to make sure that your structure is a lot stiffer and that's what we achieve with this and what we call honeycomb corrugation.

[00:08:06]
Where we essentially have a bunch of he said in the cups sometimes when people look at this image they see something different from what I'm seeing so I want to you know walk you through what I see here is there a 2nd to cups and then on on the top of those cups they're connected by another film that is horizontal so you have a bunch of these cups connected to each other forming a completely continues film without any holes that you can make as large as you want and then use it for various applications that I'll discuss in a little bit so the same corrugation traits of course have been used at the macro scale for a really long time that are well known for mechanical engineers.

[00:08:50]
3 d. pattern and with vertical ribs keeping with a desert scheme or a theme you know if you buy a chocolate candy box in the take it out usually comes with a plastic insert where each week each piece of candy isn't and this one actually has the exact same geometry is what we're used in these are basically cups that are arranged in the Hc sag in the lattice and connected to each other on top we don't really use these fancy additional corrugations but the basic geometry is essentially the same and then of course in things like architecture people use corrugated side in through increase the stiffness.

[00:09:30]
And if you open a random door just like the ones that we have here more than likely it's what we call a sandwich structure in other words you who have to face cheese that are very thin and then in between them you have a very lightweight core that could be foam or it could be again a hexagonal lattice of another thin thin material so these structures are ubiquitous because they are actually optimally engineered to give you lowest mass at the highest bend in stiffness and so they use the laws and transportation aviation even musical instruments.

[00:10:06]
Now it's different that the nano scale is that of course once you start corrugated planes you get an increase in Bend in Stephanus. You get the flatness that you expect but another thing that we found and hopefully we can see it in the video next in the next light is that these structures are also very robust and of course I already showed you these lattices that can recover from very large compressive defamations in the our case since we're focusing on plates the main defamation most that we're after I actually have to do with bending so we can generally recover from very extreme bending defamations and if I ask you what kind of material this was made up without telling you 1st you'd probably guess that this is some kind of a polymer because obviously it's a stain in pretty extreme defamations really well this is in fact aluminum oxide which is a brittle ceramic and the reason that we can we can do is that the reason that these structures do not break easily is the very high aspect ratio that we're using We're talking about sicknesses of 20 nanometers or so and.

[00:11:16]
As you know the period is it is my current So we're talking about mass picked ratio of a 1000 or so and that gives you very low strains if you're in these extreme defamations So there's a probably a famous macro scale optics. Structures that Julie has been making during generally have been more than 100 nanometers thick when they're very large when they're big enough that you can hold them in your hands another common example is the thickness of the so films typically point one to one microns depends on how iridescent or how glassy they appear or if they were destines suggests the 10s or hundreds of micro nanometers thickness because that's where you get the interference.

[00:11:56]
Patterns from. And our plates way as little as a 10th of a gram per square meter and as a result they can actually float in air kind of like mist or smoke and so here's a little video that graduate student that I work with has made where one of the plays that he's been working with has been caused by the ventilation currents in the room so there's no trick here he doesn't have a fan below blowing the blowing it up it's just the h.b.c. currents that exist in any room have basically captured is dragged it up and in fact I heard many stories from a post doc in the student who started doing this work about how they lost the plates that they've been working on and at the beginning it sounded a little bit like the ph d. equivalent of the dog ate my homework but after a video like this it really starts to look convincing so that middle talk a little bit about the mechanics of the corrugation and the structures in particular so we have a pretty good intuition into how stiff a structure like this would be versus versus like this right so what I call the you need erection of corrugation is something that is really relatively easy to understand especially if you have mechanical engineering background mechanical engineers think about these things in terms of moments of area or moments of inertia that directly translates into a bend in stiffness and for a structure like this you centrally have a constant cross-section that is greatly increased in terms of its moment of area compared to a completely planar film for a structure like this you basically don't get any enhancement compared to a completely planar film in fact because there are structures effectively longer it's all would generally be softer than a completed planar film of the same length.

[00:13:51]
But in the soft direction they have essentially the same stiffness per unit length. As the completely point of plates which scales with the thickness cute but in the stiff direction the guess and enhancements in the bend in stiffness that actually scales. As as the ratio of the height of corrugation to the thickness squared that goes back to these moments of area or moment of inertia so it's a relatively simple physical mechanism for increasing the stiffness now at the same time this and has been factored depends only weakly on the period corrugation the profile whether it's sinusoidal whether it's square though those things matter but not not nearly as much as the high to the thickness ratio multidirectional corrugation that we were after and let me tell you why we're actually would generally trying to create a structure that would be stiff in more than one direction because as you saw if you if you don't corrugated it curls up and if you make it stiff in one direction while yeah it's going to stay flat in that direction that's curled up in the other so we need something that's much more eyes of tropic and that's why we went with the this design where we use in the sag and a lattice and Hc sag and all units.

[00:15:12]
Because that is known to be fairly isotropic in fact from theory it's usually either completely as a tropical or approximately as a tropic. And can be characterized therefore by a single enhancement factor in both directions we could you know we compare it to the plan a plane or abandoned stiffness and what we have found is that there is no theory explaining how this enhanced in fact true with scale with the dimensions of the plate that you're working with and we had to develop our own with a collaborator of mine and Pan who is a theorist in solid mechanics what we found is that above a certain height of corrugation which is about one micron typically for a plate so the enhancement factor depends only weakly on height and instead is determined by the in plane prana Toure's in particular the ratio of the diameter of the hacks advance to the what we call the rib with w. So that's pretty much the exact opposite to what you have in the case of unidirectional corrugation and unidirectional corrugation the inflamed primaries don't matter the out of plain pragmatist is what you want to care about in this case it's the other way around the inflamed crime injuries do matter that's pretty much the only thing that matters whereas the height is something that isn't important once you bulbous or go above a certain height of course if your height to 0 it matters there is and then the show growth but then it quickly saturates and stops mattering and as I said there's no general analytical theories that we have to figure it out and sort of to quickly give you an idea of what it's all about if you want to build a theory one of the simplest thing you can do is to sume that each cup is basically infinitely rigid and then all of the stresses are concentrated in these little triangles where the vertices of the hexagons come together and then as a result the area that these triangles sort or whatever you want to call them the small area that these occupies directly translate into the enhancement of the bend in stiffness the smaller you make that area where the strain is concentrated the higher stiffness you will achieve.

[00:17:24]
This came out a few years ago we got some press out of it we've been working a lot on understand in some of the mechanical properties of the structure such as the sharp bend in that you can see here in that it showed you in the video before as well and we discovered is that the mechanics is actually fairly complex and due to the geometric not only Eric use you would off you would often up pain a non-linear response between the moment that you were bending moment that you're applying how hard you're trying to bend this and the curvature is that you're getting in a standard structure you would have a linear relationship but in a structure like this you actually have multivalued solution and that's where the allows you to get these really crazy bend in profiles so that this came out fairly recently and we've done a lot of theoretical explanations of why these guys can actually undergo a bend in like this another thing that we did is we looked at the 10s how properties of our plates and believe it or not we actually did it in a standard in Strawn machine so we took a structure that is $25.00 nanometers thick and loaded it in and strong and pulled on its until it broke so it turns out that the in strong machines generally do have to sufficient resolution in terms of force and displacement to do such measurements the only thing you have to be careful about is how exactly you mounted without crushing the structure so you can't use clamps you basically basically have to glue them on but otherwise it's very doable so the Corrugated plates of course are softer under tension.

[00:18:59]
Than completely planar plates because these vertical walls give it sort of a little bit of extra give as you're trying to stretch it and at the same time they're about 30 times stiffer and they're bending and this particular ones that they were were tested and so that gives you a what I would call is a matter that's what makes the material method because you have an increase in Bend and stiffness at the same time as a decrease in sense of fitness and that can be advantages give you advantages and some applications where you want the structure to be stretchable but not bendable For example some sensors could benefit from that.

[00:19:38]
And these boys break effective 10 saw strains of about one percent and the peak local strain this incidentally also about one percent now most recently. We were focusing on also trying to make a place that would basically wrapper dues the sandwich plate at the nanoscale how do we make it typically something like a door or something that we're all familiar with these days thanks to Amazon and other online retailers is cardboard corrugated cardboard is an example of a sandwich place where you have to face heats glued to a core that in this case is just another sheet of paper that has been corrugated in a wave pattern.

[00:20:22]
So typically these are glued together and that's nice at the macro scale but if you're trying to glue stains that are and on a scale and thickness that a layer of glue that you can create would be much thicker than the things you're trying to glue so that's obviously not the right approach so instead what we went after was to make monolithic structures that can be used.

[00:20:48]
Can be made using perforated folds for those of you familiar with micro fab what we're doing is we're taking a silicon oxide s.o.i. wafer and we drill some trenches in the silicon. And then lift off the device layer so we end up with basically a mold with a bunch of trenches in it that go all the way through and put it into ailed the chamber.

[00:21:12]
And deposit it in a very thin veil the film that goes everywhere it goes on top goes on the bottom and it goes inside the trenches and forms these we'll call them cubes essentially or we call them cylinders that are not circular but they go from one face sheet to another and so that's less to create all the necessary components of a sandwich plate than just one shop without any glue whatsoever.

[00:21:41]
If you do however these have these perforations in the face sheet so the face sheets are not quite as stiff as they would be if you had a completely continues film but as we'll say we'll see they also give you some extra advantages. That come you know if you see when you started to form in them so this is what we call nano cardboard because cardboard is probably the sandwich way that most people are familiar with so there's a very stiff hollow plates with nanoscale thickness as I said the bend and stiffness is lower but certainly best 3 times lower than that of an ideal sandwich plate that doesn't have any perforations and that's something you can compensate for actually by going a little bit taller if you go about 70 percent whole or structure you get rid of that factor of 3.

[00:22:31]
The Heights varies between $3.00 to $50.00 microns and that's of course determined by the thickness of the s.o.i. waiver and then thicknesses for between $50.40 nanometers resulting in Bend and stiffness that varies over more than 3 orders of magnitude I guess I skipped over the action part but generally at the end of the process that was showing you have silicon trapped inside your structure so but you know you can basically cut out the form that you're after.

[00:22:59]
An etch away the silicon so that you end up with a hollow structure that is extremely light weight All right so one thing that we had to do deal with as we were pursuing this is that these structures wrinkle up really easily and in fact that initially we were pursuing a simpler geometry where we just had circular cylinders connecting the 2 face sheets and what we found is that once you make plates like this while you make a candle Iraq out of it and it's full of wrinkles that's run along the lines between these cylinders and that turns out to be horrible for experimentation because not only do these wrinkles form in an uncontrollable fashion they actually move as you do the experiments so you you do the magic in the for example of the stiffness of your can leave a spring constant and you get completely non repeatable results.

[00:23:57]
So we wanted to get rid of those wrinkles therefore and that's why we ended up with a geometry that I showed in the previous life where we had trenches rather than just circular cylinders and that what that creates is no paths no straight lines can be. You can run them between those so and there's in there for the wrinkles if they wanted to form would have to do a meandering path that costs a lot of energy that effectively got rid of the wrinkles and all of our structures after that war wrinkle free.

[00:24:35]
You know the thing that we have to deal with in terms of mechanics is. Sheer shared affirmation is something that I actually got used to nor is that I was working on various can deliver is as a Ph d. students for example if you have a solid cam lever and it's long you can forget about your defamation itself going to be about the bend and defamation but it turns out that if you have a hollow structure like the sandwich composite that is not true at all and so we were making structures that were very long they were millimeters long and only if say 50 microns tall and then the same time when you would press at the end of it you would see a death from a sion like this in other words not bend in profile but rather a nice your profile was a whole structure is sugar and as you press on it so that's something that took a little while to figure out but eventually we characterize both the sheer Moggi y. and the bend in modular so this here actually shows the bend in stiffness I'm trying to do an equivalent of the Ashby charge for Puts I showed you an s.p. chart for these lattice materials which was comparing Young's Baggio list to the density right here I'm comparing what matters for plates which is going to be the bend in stiffness versus the aerial density right how much how much does the plate weigh per square metre for example.

[00:25:58]
And what you see here is the performance of our nano cardboard plates compared to solid plates of the same weight. And which we end up with is roughly $4.00 to $5.00 orders of magnitude and hence when compared to a solid plate of the same weight and then of course we still have.

[00:26:20]
What we were hoping for it but we as the end got the same kind of shape recovery that I was shown earlier and let me try to skip to the end of this video so that we don't have to watch the whole thing and see what I see the things that look perhaps the most interesting here is a structure that is pretty tall and looks like a sandwich composites plate and then you go has a new compresses and once he releases you poke some hole with a probe but otherwise it goes up into the original shape now imagine taking a door a sandwich composite snap right it's not going to go back to syringe in no shape it's going to be damaged permanently this came out last year actually so there we go also got some coverage from that.

[00:27:11]
And what I want to do in the remaining time is to switch to talking about applications because that's what drives me these days primarily I like mechanics but I like applications even more especially the ones that were pursuing some going to try to address this question that often lingers in the back of my mind when I listen to Speaker.

[00:27:34]
1st one we are collaborating with a company in California that's working on something called thermionic energy converters it's a way of converting heat to electricity that relatively few people know about it operates at very high temperatures and what it does is literally boils electrons you know we can boil water if you go to high enough temperatures you can boil off electrons as it's called a thermionic in fact all of the vacuum tubes that preceded transistors were based on thermionic effect but since then sort of the the thermionic effect and the phenomena associated with thermionic emission have been relatively Nish Arius there are some electronic components that still use them but not as many people are familiar with them now as far as energy conversion is concerned you can build a very simple.

[00:28:28]
Course it's very simple only if you're being superficial but the basic idea is is very simple to understand it says essentially 2 plates right next to each other you apply keys to one of the plates you get a hot enough that it emits electrons those electrons will travel through the gap and be collected by the other electrodes and then can come back to your lows and basically do this many times over and drive in your electrical load with just heat there's no moving parts involved in what we've been doing is you know we had a r.p. project.

[00:29:07]
That we were working with this company on and now we have some follow on funding as well. And these are essentially alternatives to other methods of converting he directly to electricity such as thermoelectric switch you know probably some of you have heard of them maybe most of you have heard of.

[00:29:25]
But they operate a high temperatures they can operate with much higher efficiencies especially if you can get this gap between the 2 plates very small and that's because the electrons are negatively charged they repel each other and the bigger you make that gap the more electrons you have in the gap the more they repel each other and that's called a space charge problem you can get rid of the repellant a fag by making the plates only a few microns apart so that's what we've been working on and the potential applications are things like presidential combined heat and power flex fuel portable power for groans and so on and as I said the efficiency can be greatly improved of the gap is only a few microns maybe even one micron and that eliminates the space charge so how do we do that well we use our plates in a slightly modified form so if we do is we create plays that are big enough that you can cover a macroscopic electrode but if you look closer what you'll see is that these are essentially our hexagonal honeycomb plates with a few modifications Well 1st of all we punch a hole so that the electrons can actually travel from one side to another.

[00:30:41]
They still have this corrugation that. Allows us to get the higher bend in stiffness and then another little thing that we add is the little wave in this that actually gives you the ability to stretch this structure a little bit without a break and. And that's important because when she when you go through about a 1000 degree temperature difference from the room temperature thins expand a lot and so you better accommodate thermal expansion really well so these structures are spacers essentially they keep the 2 electrons apart just by a few microns and have thickness of 50 to 800 nanometers gaffes are now actually down 2 to one we've actually gone below one in some of our structures as well though that hasn't been published yet and they're very thermally insulate and they have effective conduct it's ease of less than air gels well that's because you know just like Arijit else there they don't really fill the entire space with with a solid material they're pretty sparse and they can sustain very high pressures without getting crushed so the applications that were in soon all the Asli are going to be in thermionic devices that was a main driver in collaboration but we're getting to the regime also where it's becoming interesting for something called thermo photovoltaics which is a way of converting heat to electricity use in photovoltaic panels and there you actually want to get to even smaller gaps than one micron to use something called enhanced or near field radiated heat transfer that allows you to get to higher power outputs so our collaborators use this basis to demonstrate their meaning energy with record power densities and the gaps of $2.00 to $3.00 migrants over we continued to push it probably to about one micron or maybe slightly lower but not much.

[00:32:38]
We also made some canned livers a.f.m. candle a verse where you take a silicon can lever and you basically make a hollow shell of it. And this this is demonstrating that they're actually pretty robust you can do things like that and it will recover its original shape. They have roughly the same frequency as the solids once Sergeant should be sold should be solid once.

[00:33:03]
And while the solid ones are sold the whole ones are not yet. Men so you can see that for a solid one you'd have a certain mechanical peak at about 75 kilohertz for a hollow one you get one at slightly lower frequencies just because the Young's logic has the materials that are used in l.d. alumina is not as high as that of silicon but it's comparable right we're not talking about doing orders of magnitude lower frequencies just because you are making a structure that is thinner or lighter that would be the case if you were just making the whole can leave or thinner.

[00:33:39]
But with the greatly reduced mass greatly reduced spring constant and you also get a higher bandwidth in the air which we've done some tests with where you can scan things really fast so high due to high speed Af-Am scanning and you can see that with a solid one there is a lot of blur now the edges with a hollow one you can reduce it quite a bit because your.

[00:34:02]
You have a hollow structure that prints down a lot faster in air while at the same time still has the sensitivity that you want in other application that's what I'm after is probably the one that has been the most exciting so far and this hasn't been published yet but.

[00:34:20]
We have submitted the paper. It's levitation how many of you have seen a structure like this. Call the crooks radiometer. It's a popular toy has been around for more than a 100 years a bunch of famous physicist like Einstein and Max will have written papers about it took a while to figure out how these work but I'm going to give you the simplest possible explanation I can in 30 seconds but generally you have these these a vain space for veins on a low friction spin the spin though and then each vein is painted black on one side to put it out in direct sunlight and the black side gets about one degree warmer than the white side because it's absorbed in sunlight and then the molecules that hit the black size they will absorb that heat and leave with a higher velocity then they came in with so there's a momentum exchange between the gas and the paper veins and that creates a reaction force that acts on the black side now some people think that this structure is actually rotating due to light pressure that's a very common misconception in the radiometric radiometer all these terms certainly leads to that misconception very naturally turns out that the light pressure is there but it's about 2 orders of magnitude smaller and also it's going to point in the opposite direction because if you have a white surface is essentially a reflector Once you have a reflection you have twice the momentum of change of absorption so it's this force should actually the light pressure of course should actually act on the white side push on the white side whereas in reality we see is that the forces acting on the black side so anyway this has been around for a while but the force is small it's about towards a magnificent simply again too small to overcome gravity for paper so you can make it rotate on a low friction bearing but you can't make it fly around.

[00:36:23]
As I said but we can make structures that is specifically engineers to maximize this force and also can be a lot lighter than paper so this is what we did These are basically engineers pieces of nano cars for that have a bunch of these. Trenches more than what what I showed in the previous life and what we do is also we paint one side of them black What's the lightweight pays that we can use while we just basically put some carbon nanotubes on it on them and let them dry out that creates something that absorbs the majority of the incident light bulb being very very thin and you can probably guess what's going to happen once we have a structure like this as mysterious pressure it starts to float is essentially is a little hovercraft if you look from the side you'll see a gap that's about a 3rd of a millimeter and the lies that we're producing this coming from and they only deal lab and it could come from either direction could come from the bottom from the top it doesn't matter because what is important this that they absorb are those carbon nanotubes are only on one side so no matter what where the light is coming from is going to only get absorbed in that bottom layer so the mechanism is similar to have a craft in in that the air is pumped below the plates to create an air cushion or sometimes it's called an air bearing.

[00:37:55]
And why is the air pump while that goes back to another physical effect that's. Not a lot of people are familiar with that's called thermal creep. If you have a flu that's right next to a solid wall were normally used to the fact that the fluid would basically be stationary right next to the wall Well it turns out that's true if you don't have any temperature gradient a lot of that if you do have a temperature gradient you will have a uniform flow of fluids through that channel and that is called thermal creep sometimes also called thermal transpiration that's in effect that also is fairly well known but only you know if you're actually dealing with those types of devices and that creates a loss it is proportional to the temperature gradient gas is always flowing from the cold side to the hot side and so if you put your absorbers on the bottom of these type of plates the gas will flow down and create overpressure below the plate and that's basically a hovercraft operational principle right great and over fresh and you're floating on that aircraft.

[00:39:06]
That overpressure forces something that we modeled then you get. Reasonable agreements with experimental results the issue is often that the place or not perfectly plates even though we try to make them stiff they still curl up a little bit and so you get some disagreements but you get the nice scaling that you would expect from the theory and then the more interesting thing that happens this apologized by about the video quality I'm going to show that in the 2nd but this is video just from just a few days ago so what we did is you can reduce the pressure and do measurements not at atmospheric pressure which is $100.00 killed Pascoe's that's something that's a lot slower in the order of $100.00 Pascals and now you mean free path is a lot larger now you create philosophies are not about millimeters to centimeters per 2nd that you have that mystery pressure but rather you go to tens of meters per 2nd century you created jets of air to choose from below your plate and as a result of these structures can now start to fly around that sea so I can get that started.

[00:40:14]
To cross but here and so you have a plate here that is going to take off and starts kind of doing a little patrol of the area. And what you have below is a mesh it's a very sparse metal mesh to minimize what people call the ground effect and make sure that the mechanism here is very different from the hovercraft mechanism that was above the last flight so this one can can levitate them in there and we've done a lot of the experiments by now the tricky thing was to make sure that you can actually trap it and we have to use multiple Alie the light sources to make sure that the overall beam shape that we can create is is sort of forming an optical.

[00:41:02]
Now what's interesting is that this force is maximized in a certain pressure range that generally is between one and a 100 Pascoe or thereabouts. And these pressures exist naturally in a bunch of places for example in on the earth if you go high enough you'll eventually reach the region of the atmosphere that's called the mesosphere typically considered to be between 50 and 80 kilometers same altitude and the pressure there is exactly that now the interesting thing is about that region of the atmosphere is that it's not very accessible it's too high for airplanes and balloons both of those top out at about 50 kilometers if you look at the records of altitude and this too low for satellites the legs don't really go below roughly 100 kilometers maybe can go to 80 or so but then it's going to basically slow down and burn in just a few days.

[00:41:58]
So this is a mechanism where you can launch things and keep things flying a that's range of the atmosphere and the interesting thing is that you don't need anything for it except on light as your power source on these structures you know the flux that we're creating with the l.e.d.s. here is the same order as the.

[00:42:20]
The Solar Light flux about a 1000 watts per meter squared. Ok so that's kind of the vision that we've been working on trying to make things that can fly around in the mesosphere eventually we would like to get to this point this more of a dream right now and I've talked to some of you about how we can get to create in very high velocity jazz even that's mysterious pressure so that that's more of a dream this is more of a reality although the structures could because look the way I mean something like this can fly in them as far as well and this that it can be powered by solar flux or an external laser beam would could also provide the steering capabilities so you can actually tell it where you wanted to go the payload capacity is relatively small in the milligram scale but these days you can do a lot with a milligram milligrams is what you can get with a cubic millimeter of silicon and there's a whole lot of research that used to go by the name smart dust which focused on capabilities that you can achieve with systems that are just one cubic millimeter in size that weighs a few milligrams if you're using silicon and these days they can put batteries in there you can put sensors they can be communication electronics all sorts of plans available these days if you are willing to integrate a few layers of silicon.

[00:43:49]
And then finally the last thing this is currently given through contract negotiation hopefully we're going to start it soon I hope it doesn't get killed in the process but this is a private foundation that's called the breakthrough foundation and if anybody has heard of it they. They give us prices in physics they call the Breakthrough Prize this and I think also in your sciences or something like that but they also have.

[00:44:17]
A research program and their goal is to launch something to another star in other words to actually make humanity travel or send things I guess not travel send things not only within the solar system but to other stars. And of course the Earth start to us is Alpha Centuri And the question is how can you get there before your sponsors die because this director foundation is all privates a bunch of internet billionaires and they got interested in this and they're providing the funding but who knows how long that's going to last so you have to propose to them something that would work within tens of years not thousands of years because that's how long it would take to launch a conventional rocket to Alpha Centaurian.

[00:45:05]
And so the best idea that's out there right now is to launch what are called Star chips these are tiny silicon chips or chips of some some electronic on them that would weigh about a gram and would be attached to a light sail that would also weigh about a gram you don't want your light sails to behaviorism you payload.

[00:45:31]
And then you would be a lot of laser power on to that flight sail create maybe 30 Newtons of force or so. But acting on some on a gram or a couple of grams that creates thousands of G.'s of acceleration and if you're able to maintain it for 15 minutes you actually get to a 5th of the speed of light.

[00:45:52]
And once you get there you can you will travel for 20 years reach out a centaury hopefully do something interesting maybe snap some pictures I'm not sure what can actually be done. And then of course you get aim really well but even then there's a chance you're going to miss So the idea is to not launch one of them like we do with Voyager is for example but rather launch a 1000 of them and hold the least south some of them will will get his All right point Ok so you get there you maybe take some pictures do some measurements.

[00:46:27]
Radio them back you wait another 4 years for the radio 6 to get here and then maybe publish some pictures before the sponsors die so I know that's that's kind of the grand vision for the project it's really wacky I could tell you a lot of reasons why is doing work that's current currently trying to address but it's fun to think about and of course what our role in this is on the light sail side because you need something that's 10 square meters but weighs a gram How do you get to such low weights Well we actually are spacers that I was showing you earlier are later than that so they have a then city that is less than 1.

[00:47:12]
110th of a gram per square meter and we can go up to an order of magnitude below that even now for us the spacers by themselves are not particular collective So what you need is an additional layer and that could be very thin lived in them that itself is a risk or something similar to it maybe maybe various things so we can as well that would actually be a reflector that can create something like a 4 repro resonator and reflect the lights back and create those propulsion force that's active so we have something that's very lightweight something that's mechanical or robust can be integrated with the tonic layers and that's something that we're hoping to start working on very soon so I'm going to wrap it up here I'm going to.

[00:47:58]
Acknowledge the help of all the people that actually did the actual work. And I'll be happy to take any questions you may have. Thank. You. Jack. So. I mean you could argue that they're already at the macro level in the sense that centimeter is a macro can you go even larger to get the higher payloads so that the tricky thing about the photo credit levitation radiometric forces is that there is a relationship between the size and the pressure with which it operates best Basically you want to you want to be at the size that corresponds to the means path at the pressure that you're at and that means that if you want to go bigger it is going to work at even higher altitudes so that it's possible to do that but you have to keep that relationship in mind another way to get to higher payloads without necessarily change in the size of the plates that we're working with this to connect a bunch of them and so we've been thinking about you know how how do you string them together and how do you arrange them with respect to each other to make sure that you can maximize the lift force that makes sense.

[00:49:39]
Or. Right so I mean at this point we can manufacture things that are as big as a way for. If we wanted to make them even bigger you could the could either try to use other substrates that are you know kind of more roll to roll for example.

[00:49:56]
Or stitch them together so for for this. Star shot project. We were thinking about both both approaches potentially you know the 1st step is actually demonstrated something that works at the millimeter a centimeter scale so that's what we're focusing on right now. Yeah it's because it's not only Mir It's actually a history addict so once a kink formed it's not easy to move it so you have to release the whole structure and then start over if you start over in a different place then you can get a kink happening in a different location that makes sense.

[00:51:06]
So I mean if if they do break they will generally form fractures along the fold line if you do actually exceed their yield strains so they will often actually even after they develop fracture they will still get back into their original shape they just won't be there at the original Yawar that makes sense they don't chatter in other words.

[00:51:39]
Yeah I think the fact probably make the fractures more likely where they are. But again that doesn't really typically make the structure fail catastrophic Kelly it. Just you know makes makes a local fracture point and then the structure is not a stiff as it used to be. So it depends on the pressures at this very pressure we had these hovercrafts rights and those are typically typical gaps are only about a 3rd of a millimeter So if you you know $300.00 microns or so.

[00:52:30]
But as you go to lower pressures you can start flying in midair and if you're talking about the atmospheres as I said those pressures exist at 50 to 80 kilometers an hour that addresses your question. So the light pressure is definitely there however it's again the Kapoor is a magnitude smaller than the force that we're focusing on is a fact for our structure is probably even smaller than 2 orders of magnitude so we don't we don't we don't we usually take that into account our structure is is not designed to maximize the light forces at all.

[00:53:24]
And then of that addresses your question you. Know I think you could you could hence the thermal effect with some optical resonators for example and that's something that we're very interested in so if you can get to high temperatures by absorbing narrow band light for example. And create faster jets then that would be quite interesting because it would allow us to operate these things at atmospheric pressure not just in the hovercraft mode but rather in the.

[00:54:03]
Kind of mid-air flyer. Yeah we're hoping to do that one day so you know we got to the 1st point where the thing is. Hovering it's not because it's you know if I was given this talk couple weeks ago the best I could show you is basically these things taken off and then fallen off by the wayside because it was we didn't really have the trappin potential that would allow it to be trapped laterally and turned out to be a harder problem than we expected because it turns out that the thermal time constants of these places on the order of seconds so.

[00:54:49]
You cannot trap it unless it takes a 2nd to travel or longer to travel to the area higher intensity so we started out with relatively small traps in the small vacuum chamber and we basically get out and overshoot. And wouldn't be trapped and that was really frustrating and so we we got a bigger chamber made out of acrylic build a whole bunch of l.e.d.s. in the right configuration and now it's actually able to do this patrolling in a relatively confined area.

[00:55:22]
But the next step would be to figure out how you can actually create a laser beam that not only just wraps it in one place but makes you makes it go where you wanted to go. Not micro wires so I mean these are insulated. So I look at the mock site is is pretty good insulator and so it's possible I think to put other things on top or make them out of other materials like platinum is another material that's used so you could probably probably do that how long you can make them.

[00:56:15]
Again you know our car in fact process is based on the way for so we can make something that's the size of a way for you could probably cut it into strips and trade maybe connect them my if you wanted to. But that's that's kind of our limits right now it's.

[00:56:40]
Just. The silicon oxide have is it like going to court. Sort. Of think oxide. As a stone idea I think the most of the materials and we're using nail d for a deposition to make sure that it's highly conformal they go down amorphous so you wouldn't necessarily get the p.s. electric in fact that signal is a significant out of the morph a smith serial.

[00:57:15]
I think you probably end Neal them and get to the public Crystal in structure then and then poll them and get peers electric effect but that's not something we have explored yet. Well. Here's. Your Brain and. I should look it up yeah so we we haven't been up to God yet.

[00:57:59]
Thank you.