Close Product safety notice for gas cookers. Please click here.
Self Assembly

When objects create themselves

We want to make things smarter, without turning them into a robot!

The Lord of the things: Skylar Tibbits, a trained architect and computer scientist, co-directs the Self-Assembly Lab at MIT with Jared Laucks. He is developing objects that can assemble themselves and materials that are programmable.

Skylar Tibbits Skylar Tibbits at MIT’s Self-Assembly Lab

Mr. Tibbits, whenever I have to assemble something – an IKEA closet for example – I am rarely successful but it always brings me close to a nervous breakdown. So when I heard about the principle of self-assembly I thought: “At last! There is hope!” Is there? Our main drive is not to take on IKEA assembly problems, but that is a typical understanding of a problem when you have complex parts that need to come together in complex ways. And one of the goals of our self-assembly research is to find ways for those materials to assemble themselves. It would help us to build complex and sophisticated things in easier ways.

Besides the IKEA closet – why does the world need self-assembling things? There's a lot of effort, energy, materials, money, and time going into thinking about how to put things together in better ways in order to make better products. So if we could have a principle that would allow materials to assemble themselves, we could be faster at assembling complex, creative and awesome things – because the materials can do it with us and for us. We could also build in complex environments where it is difficult to get people or machines/materials or build complex structures that are impossible today with traditional means.

How does it work? There’re three principles or ingredients to self-assembly. The first is geometry, the second is how to connect and the third is energy.

How exactly do these ingredients work together? So in terms of geometry, if you change the shape or the geometric structure of the material, it'll produce different global structures. Then you need connections. You can use magnets, you can use surface tension, velcro or lock and key mechanisms for that. Very important: you want these things to come together and stick in just the right places. They should disconnect when they are not right.

Connecting through activation energy Connecting through activation energy

That means there needs to be some kind of error correction? Exactly! And then you need energy to somehow shake the pieces around and tumble them in order to put them in contact with one another. You can do that in different ways in different environments: underwater with pumps or with fans in the air. It does not matter. You just need some type of energy to get them all excited and find one another, connect in just the right places with just the right geometry, and then they'll assemble into precise structures.

Exciting – but still very theoretical. In what field or industry are we going to see the first self-assembling objects? Well, I think construction is probably where it might have the greatest impact and has the greatest potential but I don't think by any means it will be the first. It's quite a slow industry in a way, compared to the medical sector, the military, aviation or space travel. I can imagine that disaster relief is a sector that could be interesting. There you have the funding, you have a drive for innovation and a real need to get materials to assemble quickly without a lot of labor, without a lot of complexity and in difficult environments.

Will kitchens soon be self-cleaning?

BSH is also already working and collaborating on methods of self-assembly – but on a molecular level. At the Center of Corporate Innovation and Technology in Zaragoza (Spain), researchers and developers are examining how molecules can assemble themselves in such a way that materials with completely new properties are created.

“We are developing materials that bind with each other at molecular level in such a way that they create a functional layer. We are developing glass that is for example, extremely water-repellent, and even self-cleaning surfaces.”

However, if you were already looking forward to ordering an entirely self-cleaning kitchen, you are going to have to be a little patient. “We are still in the development phase”, says Dr. Artal, “but in a not so distant future, such kitchens would be a reality.”

Do you already have partners on board you work with? No, not yet. This research is still very much basic or fundamental research. We are still discovering what the potential and the limitations of self-assembly are on a more experimental level. But programmable materials is a topic of research that we are working on with a lot of companies already.

What are programmable materials? Look around the world and you will see we're surrounded by physical materials, whether that's products, clothing, buildings, cars, planes, shoes. All of these are based on materials. At the same time there's clearly a direction to make all these things smarter and smarter. Today the only way to make something smarter, we typically think, is to make it more robotic or more like a computer. The problem with that is that it costs a lot of money, it has a lot of potential failure, it has a lot of energy, there's a lot more assembly time. So our principle is that we want to program the materials themselves so they become smarter without turning them into robots or computers.

Light folding carbon layers Light folding carbon layers

How do you program material? How does it work? So there are three principles to programmable materials, one is the material that you're using so maybe you're going to use wood or you're going to use metal or plastics or foams or whatever. The second one is energy, you want to link up the material properties with the activation energy. With wood you might use moisture to activate the material, if you use plastics or metals you may use temperature. And if you put the materials in just the right way, using multiple materials or multiple structures like micro and macro structures, the geometric information encodes a transformation. So the third principle is that you want some kind of transformation.

What may these transformations look like? It can be a fold or a curl or an expansion or contraction and you do that by linking the material structure with just the right activation energy. So you can make wood that can fold, curl, twist in the right way or you can make textiles that respond to temperature or carbon fiber that transforms based on light.

What specific products or applications are you working on? We have developed an engine component with Airbus and Carbitex made of programmable carbon fiber. It curls up at a certain temperature or wind speed so you can change the ventilation and aerodynamics – the composite becomes the sensor and the actuator and it has no motors, no power, no electronics, no devices, it's just carbon-fiber with printed polymers. So there is less potential for errors. We have also worked on car spoilers that use the same principle. With textiles, the idea is that you want them to respond to the body or to the environment so textiles that become breathable or change color based on body temperature or moisture or sunlight. And there is also a manufacturing application you can do with textiles or composites – the Active Shoe is a project that we developed based on this principle. The idea is that you print on a stretched textile. Then when you release the textile it jumps into the 3D shape of the shoe based on the printed pattern. That's interesting because it minimizes the cost of production. And looking at furniture: we did a table with a company called Woodskin where you have a flat sheet of wood that ships in a flat box. It jumps into the shape of a table by itself when you unbox it.

Wow! Well, now we are talking! Yes, I know. But you cannot buy that at IKEA yet.

Interview: Lars Gaede

Skylar Tibbits is a trained architect, designer, computer scientist, and artist whose research focuses on developing self-assembly technologies for large-scale applications. He is an Assistant Professor in the Department of Architecture at MIT and the co-director and founder of the Self-Assembly Lab housed at MIT’s International Design Center. Tibbits received numerous awards and mentions. For example, he was named R&D Magazine's 2015 Innovator of the Year, 2015 National Geographic Emerging Explorer and Inaugural WIRED Fellow in 2014. Previously, he has worked at a number of renowned design offices including: Zaha Hadid Architects, Asymptote Architecture and Point b Design. Skylar Tibbits is also the founder and principal of a multidisciplinary design practice, SJET LLC.

For more information about the project here.

Photos and Video © Roderick Aichinger

Media cart Contact us Back to top