All right. Thanks Don for organizing this event and bringing together this wonderful community. It’s my great pleasure to be sharing today the work that I’ve been doing on bio inpired and biio-based 4D printing towards my goal of creating an adaptive built environment. So currently I’m a assistant professor at Cornell University’s department of design tech. This is actually a recent transition as of 6 months ago. Prior to that I had spent my past eight years actually at the University of Arts Institute for computational design and construction where I was developing my PhD and then later leading the material programming research group.

0:45 And throughout my work at Cornell where I currently lead where I currently direct my new lab called the multime lab and also throughout my past work in Germany. I want to emphasize that the research that I’m doing is very interdisciplinary. I’ve collaborated closely with partners from material science to physics to biology and also industry. And this is important because my work at its core is about learning from nature how to use natural materials more sustainably but also performatively.

1:29 So this work requires that we integrate different disciplines for example developing new biiobased materials to turn properties that we normally consider as disadvantages such as shrinking, swelling, warping into advantages. Also studying biological role models in order to mimic their structural layering and transfer them to emerging technologies like 3D printing and finally how to design and program new types of mechanisms whose shape changes and functions are 100% controlled and powered by the environment.

2:09 So I think by integrating and synthesizing knowledge from these diverse domains, we can then create bioinspired structures that can change shape without consuming any operating energy. And I present this as an alternative to conventional methods of design and making. So before I go into the details of how all that works, I want to share a little bit about where I’m coming from. So I’ve been I consider myself a maker.

2:33 I’ve been making things for practically my entire life. I actually built my first 3D printer about 15 years ago, the original MakerBot Cupcake CNC. And besides 3D printing, I’ve also built a wide variety of projects ranging in size from origami robots to large scale art installations and also buildings when I was working as an architect. And having made so much stuff throughout my life, I became increasingly confronted with the environmental responsibility that comes with the act of making.

So I just want to share this infographic which compares in the year 1900 the earth’s biomass which you see on the right here and comparing that to on the left the anthropogenic mass which is the accumulation of all objects made by humans. And when you consider that humans only make up a tenth of a percent of all of the biomass since 1900 the anthropogenic mass has actually doubled every 20 years and in and according to a recent study the anthropogenic mass has now surpassed the biomass.

So and actually the the biomass has even decreased because of our def because of deforestation and our overall manufacturing activities. So this study really made me realize that the climate problem is really a problem of how we use materials and make things. And a lot of this human-made anthropomass are things like buildings and roads and manufacturing those construction materials are responsible for 9% of global CO2 emissions.

4:20 Building operations like heating and cooling for maintaining comfort in our buildings. That’s 28% of global emissions. So essentially over twothirds of carbon emissions are linked to our overall manufacturing and industrial activities. So in order to address this global problem, I think we need to design our products and buildings to adapt to multiple scenarios to be used and reused into the future. So the question is how do we make structures that are adaptive?

So architects many of us are architects here. I think there is a long history of speculating on ways to make buildings more adaptive and future proof. So for example, the Prada transformer by OMA it has specialized floor plans on all different sides of the of this building surface essentially expanding the usable footprint of one building by rotating the structure with cranes. There’s also the villa sole which was designed to follow the sun’s movement throughout the day.

5:31 But it requires this huge machine to rotate this 1500 ton mass. And of course there’s the Arab Institute, the Arab World Institute, which is a famous example of a kinetic facade with over 16,000 apertures that control the amount of light entering the building. But like the previous examples, it’s very costly and complex to maintain. The the facade itself is no longer adapting autonomously and in the case of the Villa Sole, it’s actually at the risk of demolition.

So so you don’t see a lot of these structures operating over a long term. However, in biology, organisms also move and adapt, but in a very different way. So, plants, respond to their environment by changing their shapes and properties through growth and other mechanisms. I’m really interested in the mechanisms of seed dispersal that you see on the bottom row here. From the folding of pollen grains to, exploding cucumbers on the right.

And these are really interesting because they don’t require any active metabolism to function. So let me explain that with my favorite example, the pine cone. It really nicely in illustrates the example of how the material structure and function are so tightly integrated that it can reliably open its scales without consuming any metabolic energy. And interestingly, it’s actually dead. It’s a dead tissue. It when it falls from the tree.

7:12 And indeed, my partners from biology have actually found a 14 millionyear fossilized cone that can still open and close. So this incredible robustness that we observe in nature often is due to the material structure and you can see in this on the right here the cellulose microfibbrals which are arranged in these two directional layers and this bilayer structure is what causes the pine cone to swell shrink in response to moisture changes.

So my personal goal has been to investigate how we can apply strategies such as these from biology to the design and manufacture of our products and buildings. For example, maybe one day we’ll have building skins that autonomously provide shading or ventilation depending on the weather. So my personal dream is to get rid of active heating and air conditioning. So in the next projects that I show today I’m going to be introducing my previous doctoral research in working towards this goal.

8:25 So I’m going to explain fuse filament fabrication is really actually conducive to emulating the fiber structure of biological materials. And this is a process called 40D printing. So even though the printers and materials are very cheap and accessible, one of the core challenges is how do you even design these structures? How do you even 4D print something that can change shape? So how it works is that through the process of extrusion, you leave these anisotropic features on the material layer and when you print that with a hyroscopic material like wood, it swells or shrinks directionally.

9:05 When you print a less swellable layer on top of this existing layer similar to the pine cone structure that we saw earlier the resulting billayer bends in one direction when swelling and the other direction when shrinking. So by programming the extrusion path with a 3D printer with different parameters such as spacing, layer height between the paths, you can control bending in different directions. You can control the amount of bending and you can also control the speed of bending and by combining and stacking together these different blayer building blocks as I call them you can expand the geometric possibilities.

So this extrusion process is kind of like a analog material code that tells the material what shape it should transform into. Through a lot of experimentation I developed a computational fabrication workflow for configuring these assemblies of billayers within the established CAD platform of Rhino and and Grasshopper. And this is allows us to preview the shape change of the combined bilayers and finally generate the the low-level machine code that can create an intricate mezor structure as you see on the on the left here.

10:23 So no STL involved directly machine code. And this allows designers to focus in on the high level design without needing to worry about how to organize each and each of the individual tool paths or needing to learn how to write the machine code. And that allows us to automate the production from a digital design to physical objects. And we can create many types of material transformations that you can see here such as bending expanding, folding, stretching and even double curvature.

Okay. So, the next challenge deals with how do we transfer and upscale these methods I just showed that are on the desktop 3D printer scale, the hobbyist 3D printer scale to a industrial robotic platform. And in terms of the materials, we’re using construction materials here. Wood is a really well-known and easily scalable material which can also be used for self forming billayers with high actuation forces albeit it has a very low resolution of programmability.

11:32 On the other hand 3D printing has enables a high resolution of detail and customizing material properties. So in order to combine both of these materials into a hybrid system, we developed a large scale additive manufacturing platform with multiple integrated endectors which includes a vacuum gripper here for the pick and placement of the wood blayer actuators and also a large format extruder for the 3D printing of metamaterial structures.

12:03 And that allows us to produce this hybrid structure in which the wood and the 3D printed pattern are working together to sell shape up to the meter scale. So in transitioning from the desktop scale to the large scale robotic extrusion, it’s necessary to make some adjustments to things like the extrusion tool path which had to be continuous and non-crossing due to the larger nozzle and the pressure built up inside of the barrel.

12:30 It the 3D printer the 3D printing parameters also had to be recalibrated although the general relationships between depth thickness ratio as geometry remain the same. So using the geometry we can actually modulate and physically program areas with graded stretching areas of high anisotropic stiffness and also program bending in different directions. And so this biomposite structure just by acclimating to the moisture content in the environment can transform from that flat state to a doubly curved geometry without any mechanical force without any subtractive manufacturing without any operating energy.

And so based on this large-scale additive manufacturing platform that we developed my thesis students my master thesis students also explored the unique properties of oxidic metamaterials which when bent form surfaces with various variations in double curvature. So they exploited this property in self- forming bending active shell structures in which the wood is distributed and embedded within a surface with it this honeycomb and inverse honeycomb pattern and by and and depending on the pattern this results in positive and negative gshian curvatures and in other words that’s like a dome-like curvature or a satellite curvature.

14:02 So by modulating these patterns across a functionally graded surface, we were able to fabricate this 3 m long structure with varying gausian curvatures across the surface. And it took 13 hours to print even with all the hiccups as this is our first time printing something this large which is really fast considering the size and curvature of this structure. So I want to give a quick shout out to Laura Karen who were the thesis students who worked on this project.

14:36 So in the next project I want to share how we can translate the working principles from biology to design and engineering applications. And this is a project that was done in collaboration with the plant biomechanics group but also the university hospital at the at Fryberg University. And I use orthotic devices as a motivating example because there’s a need for both customization and adaptation in specific patient pathologies.

15:07 Proper fit is extremely important. And 3D printing has indeed become a very popular technique for customizing personalized splints for a variety of shapes and sizes. However, it doesn’t address the muscular atropy that typically results from immobilization. So, this is when your your limb basically gets thinner if you’re wearing a cast. And the reason why if you’ve ever broken your bone, you had to go back to the doctor every few months to get your cast refitted.

15:41 So together with my biologist partners, we found this very interesting plant the air potato plant which has a very clever strategy of climbing smooth supports without slipping off. And it generates high squeezing forces onto the host structure through these stipules which are located at the base of its leaves and when they grow they expand thereby tensioning the system. So we we studied and abstracted this tensioning mechanism to 4D printing.

16:15 The squeezing behavior basically happens in two phases. First the stem helix loosely winds about an existing structure and then it pushes outwards with the stipules in in order to stiffen the stem helix. And then we so we abstracted this and transferred them into a twisting billayer strip called the helix mechanism. And the second part is this blayer flap that what that we call the pocket mechanism. And that creates a pocket of space between itself and the helix on top of which it sits.

So here’s an overview over overview of the biomedic process where we took that tensioning strategy from the air potato plant and transferred it to 4D printing through the design of the helix and pocket mechanisms. These two mechanisms can be individually tuned and combined into a single material system and then we could evaluate it for applications where the fit and grip are important. So we tested the 40D printed material system that we created and evaluated the ability for it to stay gripping onto a support pole.

17:25 So we we here you can see the the pure helix mechanism gripping onto a 30 diameter millimeter diameter pole and self-releasing after 40 minutes. I’m not sure why the videos are not working, but in car in comparison, the helical system without the pocket mechanisms can actually stay gripping onto even smaller diameters and stay gripping for over 3 hours during moisture uptake. Okay. We also measured the squeezing forces of the 4D printed material system and compared the different combinations of the two mechanisms.

18:08 So here you can see on the bottom the helix mechanism by itself without anything else generated pretty low forces and the addition of these pocket mechanisms in low and high densities generated increased forces to varying degrees but the most effective strategy for generating squeezing force was to delay the actuation of the pocket mechanisms. Now, the doctors who would actually use this self-tightening mechanism are likely not experts in digital modeling.

18:41 So, we created this design workflow in which physicians could physically prototype designs directly onto the patient’s body, 3D scan and digitally analyze the curvatures and then generate the fabrication data for 4D printing. And this is important for adopt adoption outside of the lab. So this is the design of the splint with the bioinspired mechanism which self-tightens around the wrist and forearm at varying curvatures. But what’s the most interesting takeaway for me is that we were able to generate similar ranges of forces as the air potato plant while mimicking the stipules in low quantities which is how they are found in nature.

But if we increase the number of these stipules and delay their actuation, we can actually measure higher forces. And so it was exciting to see that computational design and 4D printing allows us to extend beyond what nature can achieve. Okay. In this last project, I’m going to talk about whether or not we can apply bio inpired 4D printing to the real world in actual buildings, which is something that I kind of talked about in the very beginning.

19:51 So this is a collaboration with the cluster of excellence in university of fyberg the liftmats and also the plastics technology institute at the university of shikart and the goal was to adapt the foundational scientific research that we developed over the previous projects that I that I introduced to a real building demonstrator application. So there are new challenges in integrating the self-shaping design with real environmental and weather conditions and this happens across a longer time scale than we were used to.

20:30 But also if we can even scale up the 4D printing manufacturing process at at the level of a building facade. So using this research building project that we had at the ICD which is called the Livmats biomedic shell building we use this as a case study to design the self-shaping adaptive facade according to its very specific building conditions. And we worked with the global design team of the building project to position the building on site so that the south facing orientation of the building allows us to gather the maximum amount of solar heat possible.

21:12 And that means that during the winter months, we needed to design the 4D printed shading elements to open to allow solar heat to enter the building. But in the summer months, then they should close to block out the high heat loads from entering the building. So, how do we program shading for the different seasons using a humidity reactive material? So, in temperate climates such as in Germany, but but most likely here in the Netherlands as well, there’s a reliable link between temperature and humidity.

21:48 So if you take a look at a temper a typical winter day, you can see that the humidity in blue remains quite high while the temperature in red maintains a very low trend. And on a typical summer day on the right, you can see that there’s much more daily fluctuations in temperature, but the humidity always remains that always maintains this inverse relationship. And so when we designed our shading elements to so that means we had to design our shading elements to curl and open when humid and then flatten and shade when dry.

And we also needed to match the response time of the material to the sky time scale of weather changes in Fryberg where the building would be. And so together with material scientists, we co-developed a highly responsive 3D printable filament material which was consisting of cellulose locally sourced from a company in Germany. We went through then a very intensive testing process in the lab using an active climate micro generator to cycle many times alternating high and low humidities.

23:00 There we discovered that our LED adhesives are not really robust against high humidities. But we were happy to see that our 4D printed elements could continue to operate against these disturbances and phased by multiple attacks. Now, in order to test the performance under real weather conditions, I built a facade mockup in Strickart, which mimics the same orientation and height as the target building in Fryberg. There’s very similar weather patterns, and there the 40 printed elements were then exposed to the full range of humidity, temperature, and UV effects over all of the seasons for over a year.

23:49 And then with the assurance from our extensive testing, we then rented a total of four 3D printers to fabricate almost 450 bespoke shading elements. So each of them were unique. And because of this very material efficient structure, each of these only took on average 20 20 minutes to produce. And the whole process of 4D printing the facade took 17 days and less than 6 kg of material.

24:20 So all of these 4D printed elements were then preassembled at our lab in Strickart onto thin aluminum strips through this integrated pressfit 3D printed connection. And each of the windows then went through a short testing phase just to make sure there’s no defects before we actually put it into the building. So, in the end, we had these pre-assembled windows, very extremely lightweight that we could then pack it into one minivan.

24:52 We could drive it onto the construction site and then unload and install it onto the roof without any cranes or heavy equipment. I also want to mention that we designed this facade system with operable glass doors which means that we can always access the window cavity if we need to repair if we need to even remove the facade system if needed. And also that the windows are vented on the top and the bottom which allows those materials to acclimate to the temperature outside.

25:31 And so there it is. The real the real thing. And this is to my knowledge the first 40D printed weather responsive facade. I’m proud to say that it’s also a public building which is open to students, faculty, visitors at the University of Fryberg. And here’s a view from the inside. This is just a couple days after an installation. I was very happy to see that the facade was working immediately and you can see on the morning where it’s relatively cool, they’re open and then as the afternoon heats up they they’re closed and and I also want to share an anecdote which is that there are sensors on the top and the bottom of each of these windows where I’m still monitoring the results and the electronics actually took over a year to get working.

26:27 It took over a year to get AC power onto the windows whereas u the material just started working immediately after installation. Yeah and so this is just a kind of a video of that same day. A time lapse that I took and again it requires zero operating energy demonstrating resilience against things like power outages or the complex engineering and constant maintenance that’s required for comparable electromechanical systems that I showed in the very beginning of my presentation.

27:01 And what’s exciting to me is that this solution has the potential to be economically scalable because we were able to scale up the 4D printing process to the level of building facade using very commonly available, very affordable 3D printing equipment and the most abundant biop-olymer on Earth, which is cellulose. And now I want to move a little bit into my future and ongoing work. At Cornell University.

27:32 So one of the limitations of such a weather responsive system is that many people who are in buildings want to be able to control their their surroundings. We’re so used to things like the shades that we can just open and close. So overriding this passive functionality is is a is definitely a challenge. And currently I’m looking into modeling the dynamics of self-shaping materials and their interactions with human behavior with the weather cycle and occupant comfort over multiple time scales.

28:09 And still I I have this dream of completely getting rid of aluminum in my facade or the steel in the structure of buildings. So I’m currently also looking beyond facades using a monomaterial system enabled by advanced robotic manufacturing methods and multiscale additive fabrication. So imagine that we have this industrial robot here with nano to microscale fabrication all integrated in one workflow. Now aside from the building scale I’m also really interested in body scale applications.

28:46 We saw a lot of work from the apparel industry, the footwear footwear industry and I think adaptation for the body requires a much faster response time than in buildings and so using the human body as a stimulus for example human sweat poses some really interesting questions about adapt adaptability and working on the scale of wearables will force us to improve these axes of performance. The scale and application is also much closer to transfer to the market.

29:18 So I think that having some projects with a direct industry partner is really important for keeping the needs of the real world and the transfer to actual products in focus. And I’ve talked a lot about material programming but for a truly adaptive and futureproof environment we need to talk about material reprogramming. So what if we could unlock functionality at vastly different time scales to respond to different stimuli conditions?

29:47 And so imagine we have a building facade that grows like a climbing ivy, passively regulates comfort in the building, learns and adapts to climate change and finally self-deal I’m really interested in studying amplification mechanisms like curved crease origami structures and bstable geometries in order to create materials with advanced computation and logic. So, I’m going to just restart this because I really like this video and I hope it shows this time.

30:23 Okay. So, there’s so many lessons to to learn from nature on how materials u natural materials can be performative yet sustainable. And that’s the mission that I’m continuing at the multime lab at Cornell to pinpoint how we can harness the inherent characteristics of natural materials to perform even more advanced functions and showcase that bioinspired design and making is a critical a credible pathway to overcoming the competing resources between nature and technology towards a future where our technologies are no different from nature.

31:01 So, I’m really excited to hear your thoughts and and possibly ideas about this work. Thanks a lot for your attention. To learn more about the CDFAM computational design symposium series, to see the archives of previous presentations, and to learn about future events, visit CDFAM.com.