Yes, thank you Dan for the invitation, kind introduction. Hello everyone, my name is Tuo. I’m a researcher at Princeton University. So, I’m also looking at interested in architected assembly structure at metamaterials. The way we are looking at it, we want to create a way to deform those structures in large information. Okay. So we got a lot of inspiration from different concept for example corality differential geometry specifically we like origami.

0:29 So folding origami can be is art. Okay the process can be very complex but the result is beautiful. So this is one of the my favorite art piece origami folded by Eric. So you are looking at 11 musician and then 11 instruments and total number paper that you he was using was 11 + 11 22. Okay. And then so the the I show I just showed you the most one of the most compact origami folding but there’s another way to fold it.

1:06 Some pattern can be very very simple. For example, Miss Crestling who is a a French artist, she was folding this so-called crestling origami. So just using two tubes and one piece of paper in the middle and then she utilized this twist to generate a spawning spontaneous folding and then you can see you have a folding because of buckling you have this crease pattern in the middle. Okay.

1:33 So this pattern is essentially the main motivation for our research and then the way we are looking at this cresting origami essentially a shellbased model but we are looking at different way we consider masses. So then you are looking at the trusses we can get the key geometric information of the trust and then we can build up a simplified analytical model. Then if you look at look at the look at the screen this quadratic form essentially we consider all the bars as spring and then although this analytical solution is very simple we can get some insights from it.

2:10 The insight we get is the geometry this trust is bstable. So means that it has two configuration. These two configuration you don’t need to apply any load it just stand there unstable. Okay. And then from the equation we can draw the twisting angle and also the elastic energy envelope showing on the bottom left. Then you see that this two local minimum indicates the this is has two stable states.

2:36 Okay. And then also from this anal simplified analytical solution we can we can know that how to play with the energy barrier between these two stable state. Energy barriers means like you need to put certain energy to break so that you can transit between these two stable states. One way is to utilize the geometry. For example, we can utilize four gong six g a gong and then assuming we build up this geometry with same material but you can see the energy envelope the amount of energy you input this structure to deform it is very different.

3:10 Okay. And then another way more straightforward we just changing the material we build up this geometry then we can get tunable mechanical property in this case. So now let’s look at some experiment. Okay. So in this array I’m using the building up all the cells with same material but different geometry. And then you can see that when I load it the two cells with the agon geometry default first.

3:38 Why? Because among this all the geometry agon has the lowest energy barrier and then as we keep loading it the six gun geometry start to deform. Okay. Because the six gun has a little bit higher energy barrier than the agon. And then as we keep loading it in the end the foreground geometry will deform as it has energ highest energy barrier among all the three design and then you will see some interesting snapping behavior happens here.

4:06 Okay. And then this we are looking at the geometry in fact to control the folding sequence. We can also explore how the material property can change the folding capability. For example, assuming that we have a total amount of unisells essentially we have seven red, seven white and seven yellow. The the different color means they have the build the the material you build them is a little bit different.

4:35 Okay. Although they have different geometries but because the material different and then the mechanical property of each unit cell is different and then now we are looking at how do we distribute those 21 unit cells in a certain design domain to get different type of assembly with different mechanical property. Okay. So here are four represented cases. I’m still looking at the two of them. If you see if you top row and bottom row there are two assemblies.

5:03 They are made of same amount of material but the way we we assign the unit cell is different and you can see the mechanical property can be quite different. In the first row if you look at the displacement load curve we will see that the peak forces they are gradually increasing. Okay. But then you can second row the peak force are almost same. Then if you compare the store energy with these two assembly, they are almost the same which is making sense because they essentially build up the same material.

5:36 And then let me show you a video how this experiment has been conducted. Then you get a more intuitive idea how this works. For example, same idea, same amount of material. We just change the way resemble them. This is the first case. As you can see as this loading you will see the snapping behavior indicating this assembly has multiple stable states and then you will see that in this case the peak force of this assembly as deforms it gradually increasing.

6:05 Okay. And then in the second scenario I just show you as we load this structure the peak force will be almost the same. Okay. So depends on different type of application. Maybe this case for from a energy dissipation perspective when you have a peak constant peak load that may that may be ideal. Okay. And then I’m showing just play with different like the kids playing Legos. We can play different ways to them to them and then in this case I can show that such that the second the third peak force they are almost the same and the first one be lower.

6:46 And then in the last example I’m going to show you we can play with the design to get different response with different mechanical property. As you can show here I want to have the first and second pick load almost same and the third one will be higher than them. And again I want to emphasize that because I build up those material with the same amount of material the total store energy they’re almost identical.

7:15 Okay. So now you can see that from the Zambi the system is highly modular. We also utilizing this modularity concept to create some structure that can deform in a very interesting way and the specific mechanism we are looking at is coupled twist with actual deformation. Okay in the literature especially a paper published 2017 they’re looking at this concept. So when you compress the structure you want to twist.

7:42 Okay. But the literature has a limitation means that this type of deformation is in is is very small. For the original paper the twisting angle is smaller than 3°. So our interested to expand this concept in large deformation. Okay. So then we achieve a a 3D design from left to right. It will twist up to 90 degree and then impan it can contract or expand up to 25% and then the vertically the high change will be more than 50%.

8:21 This highlights the large deformation idea we are looking at. Okay, we are also we creating this two type of actuation scheme. One involving compress with free translate free rotation. The other one with we are twist assembly but allow free translation. Okay, let me show you some experimental result to highlight this concept. This is the first boundary condition. We are going to apply a twist but allow free translation in the vertical direction.

8:52 Then you see the experimental result. As you can see that now we are twisting this geometry and then in the counterclock direction up to 83°. Then you can see implant the abs is contract is contracting. Okay. But vertically is keep shrinking and then we reverse the twisting direction to clockwise and then keep twisted until it reach the least volume configuration. And then this is demonstrated that for the first boundary condition twist with free translation we can achieve this type of deformation.

9:32 And then if you look at some data from the experiment we compare we compare this kinematic parameters for example high change edge lens change and twisting angle and then compare those experimental result with our legal solution. You see you see the result match quite well. Okay. And then this come to the second boundary condition. We start exactly the same geometry but now we are applying a vertical displacement allow free rotation.

10:03 As you can see that as we compress this geometry at the very beginning there’s almost no twist only vertical deformation that is because a dipole contains two different clarity they are deforming only vertically. Okay. And then as only one layer they start to twist. Okay, again with this boundary condition we can also reach this lowest volume configuration. So we also compare look at some key geometric parameters and also we plot some representative configurations.

10:41 I want to highlight that with these two type of boundary condition although in the end the end configuration are the same but in during the transition the transition configuration they are different because the boundary condition is different. Okay. And then we also created a surrogate simulation model to understand the behavior of the assembly. We love this reduce order modeling because comparing to the standard shell elements in finite element in general just building up the mesh it takes hours and render the simulation sometime takes days but this with this reduce auto modeling we can get the behavior very fast.

11:21 The the reduce auto modeling essentially based on bar and hinge essentially we are getting the key geometry parameters and then simulate them as a springs. Okay. And then if you’re looking at for example the figure in the central bottom we compare the experimental result with a simulation from this reduced order modeling. You see that we overall we can predict the trend of the curve but we admit that so this reduce order modeling is not perfect because we cannot get fully information comparing for example standard FM but we from a global kinematics perspective we love it because can get we can insight from this simulation and then I’m show you some simulation result in the simulation we also implement this two type of boundary condition right one is compression with free rotation in this case the cority of the cell doesn’t matter because we only compress it and then free rotate and the cority in general dedicates to the twisting direction so this is the simulation result we can also get this bstable multistable behavior with different type of peak force and then the stanking bond condition we apply twist but with free translation.

12:40 But you see here because twistity plays a big role. Then you can see that the we have a two arrays differentity then the mechanical property can be very different. Okay. So this is simulation at unisell level. We can also simulate at a 3D. You see this is the first type of a boundary condition that’s very similar boundary condition exactly same as we apply experiment instead of simulation can help us to verify our experimental result and do to get insight from it and then this is the twist boundary condition as you can see and then you can see that we can with the simulation we can play with the shifted twisting direction we can get this quite a euro type of a torque envelope.

13:34 Okay. Then what we can do out of this assembly. The first concept we’re looking at so is soal noncomitive state transition. In the phys in the physics textbook we learn about for example if you have rubber you apply left twist right twist and then if you reverse the twisting direction to right and left essentially you get the same result. But here is different. Here our assembly the deformation depends on the a sequence of loading.

14:04 Let me give you example. Okay. Let’s start with this fully deployed configuration. And then the first boundary can the first loading sequence we have here is sorry. Okay. Now the first loading scenario is we want to apply a counterclock twist and then clockwise twist. Okay. This is scenario number one. And then you can see that after we apply these two type of twist, the Zambi goes back to the original configuration.

14:37 Okay, the which is fully deployed. But then in the second scenario, I’m going to reverse this tweet in direction. Then what we’re going to do is we apply the clockwise twist first then counterclockwise twist. Let’s see what will happen. Okay, so we are applying a clockwise twist. The zombie is twisting, contracting and shrinking in height. And then now we reverse the twisting direction to counterclockwise. As you can see the zombie jump to different configuration which is shorter.

15:08 Okay, this highlight that our Zambi depends on looking the the deformation is highly dependent looking sequence. Okay, this is one application. The other application is like we want to create some robot of bit. The way we are doing it is a proof concept. We are integrating some magnetic response material on top of Z assembly and then using by control the magnetic field direction we can control the folding sequence or enable local motion assembly.

15:38 First of all I’m going to using magnetic field to control the loading folding sequence. Now you can see that we we want to fold the red layer first. As you can see this is the demo of the experiment. Now if we reversed the magnetic direction then we can fold the Y layer. Just to confirm our actuation scheme robust I want to fold it twice twice and then if we keep changing magnetic magnetic direction we can make it move.

16:16 For example, we can showcase some very preliminary local motion. For example, forward, backward, steer, left and right and then come back. Okay. So the last application I’m going to show you regarding the thermal regulation. If we incorporate some optical material on this assembly and then we can fully utilizing the different configuration to achieve the so-called pass cooling or heating. In central as a proof concept we just utilizing the black and white paper to build up the unit cells and then we expose this assembly under the sun.

16:54 Okay. And then depend depends on configuration you can see the heat map can be very different. Advantage of this scheme is that because the fabrication is so straightforward material is so cheap. We can actually skills up. For example, we can be a facade or roof on building to save energy for thermal regulation. Okay. So with this I would like to thank my wonderful collaborators Shanging Continuous Shishi GT M and my supervisor Princeton Glaus.

17:24 Okay. And then hopefully enjoy the talk. If you have any questions here my contact I like to connect. Thank you so much. To see the full recording of this and previous presentations, as well as information about future CDF events, visit CDFAM.com.