New approach assembles big structures from small interlocking pieces

New approach assembles big structures from small interlocking pieces
Test apparatus with reversibly-assembled cellular composite materials. Credit: © CC-BY-NC-SA Kenneth C. Cheung

MIT researchers have developed a lightweight structure whose tiny blocks can be snapped together much like the bricks of a child's construction toy. The new material, the researchers say, could revolutionize the assembly of airplanes, spacecraft, and even larger structures, such as dikes and levees.

The new approach to construction is described in a paper appearing this week in the journal Science, co-authored by postdoc Kenneth Cheung and Neil Gershenfeld, director of MIT's Center for Bits and Atoms.

Gershenfeld likens the structure—which is made from tiny, identical, interlocking parts—to chainmail. The parts, based on a novel geometry that Cheung developed with Gershenfeld, form a structure that is 10 times stiffer for a given weight than existing ultralight materials. But this new structure can also be disassembled and reassembled easily—such as to repair damage, or to recycle the parts into a different configuration.

The individual parts can be mass-produced; Gershenfeld and Cheung are developing a to assemble them into wings, airplane fuselages, bridges or rockets—among many other possibilities.

The new design combines three fields of research, Gershenfeld says: , cellular materials (those made with porous cells) and additive manufacturing (such as 3-D printing, where structures are built by depositing rather than removing material).

New approach assembles big structures from small interlocking pieces
Part production for reversibly-assembled cellular composite materials, slicing from stock produced by a multiplexed fiber winding method. Credit: CC-BY-NC-SA Kenneth C. Cheung

With conventional composites—now used in everything from golf clubs and tennis rackets to the components of Boeing's new 787 airplane—each piece is manufactured as a continuous unit. Therefore, manufacturing large structures, such as , requires large factories where and can be wound and parts heat-cured as a whole, minimizing the number of separate pieces that must be joined in final assembly. That requirement meant, for example, Boeing's suppliers have had to build enormous facilities to make parts for the 787.

Pound for pound, the new technique allows much less material to carry a given load. This could not only reduce the weight of vehicles, for example—which could significantly lower fuel use and operating costs—but also reduce the costs of construction and assembly, while allowing greater design flexibility. The system is useful for "anything you need to move, or put in the air or in space," says Cheung, who will begin work this fall as an engineer at NASA's Ames Research Center.

The concept, Gershenfeld says, arose in response to the question, "Can you 3-D print an airplane?" While he and Cheung realized that 3-D printing was an impractical approach at such a large scale, they wondered if it might be possible instead to use the discrete "digital" materials that they were studying.

"This satisfies the spirit of the question," Gershenfeld says, "but it's assembled rather than printed." The team is now developing an assembler robot that can crawl, insectlike, over the surface of a growing structure, adding pieces one by one to the existing structure.

New approach assembles big structures from small interlocking pieces
Close-up photograph of reversibly-assembled cellular composite materials. Credit: © CC-BY-NC-SA Kenneth C. Cheung

In traditional composite manufacturing, the joints between large components tend to be where cracks and structural failures start. While these new structures are made by linking many small composite fiber loops, Cheung and Gershenfeld show that they behave like an elastic solid, with a stiffness, or modulus, equal to that of much heavier traditional structures—because forces are conveyed through the structures inside the pieces and distributed across the lattice structure.

What's more, when conventional composite materials are stressed to the breaking point, they tend to fail abruptly and at large scale. But the new modular system tends to fail only incrementally, meaning it is more reliable and can more easily be repaired, the researchers say. "It's a massively redundant system," Gershenfeld says.

Cheung produced flat, cross-shaped composite pieces that were clipped into a cubic lattice of octahedral cells, a structure called a "cuboct"—which is similar to the crystal structure of the mineral perovskite, a major component of Earth's crust. While the individual components can be disassembled for repairs or recycling, there's no risk of them falling apart on their own, the researchers explain. Like the buckle on a seat belt, they are designed to be strong in the directions of forces that might be applied in normal use, and require pressure in an entirely different direction in order to be released.

The possibility of linking multiple types of parts introduces a new degree of design freedom into composite manufacturing. The researchers show that by combining different part types, they can make morphing structures with identical geometry but that bend in different ways in response to loads: Instead of moving only at fixed joints, the entire arm of a robot or wing of an airplane could change shape.

Explore further

The robotic equivalent of a Swiss army knife (w/ video)

More information: "Reversibly Assembled Cellular Composite Materials," by K.C. Cheung; N. Gershenfeld et al Science, 2013.
Journal information: Science

Citation: New approach assembles big structures from small interlocking pieces (2013, August 15) retrieved 27 June 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Aug 15, 2013
Take the concept down a few orders of magnitude, and you have Programmable Matter, Claytronics, Artificial Atoms, or whatever you want to call it.

Aug 15, 2013
Tried that. Unfortunately it's not as straightforward as all that. First off the assembly of small structures is difficult. Second: when you get to very small scales stuff tends to spontaneously stick in ways that you wouldn't expect (electrostatic and even VanderWaals forces suddenly start playing a big part and relying on gravity to keep stuff down becomes tricky).

We had a project with micromanipulation in an electron microscope via robotic tweezers. This turned out to be pretty tricky and more of an artform than an exact science.
I.e. fully autonomous assembly via robotics was a no-go at that scale (a few micrometers to 10 nanometers).
Above that - in the range of pbservation of a microscope - it got easier (10-100 micrometer scael) but still not all that easy.

However we did get the robot to play "Towers of Hanoi" on 0.2mm diameter optic fibres as towers with discs 0.4, 0.6 and 0.8 mm wide fully autonomously.

Aug 15, 2013
So build a very very high tower with it already. That will get people's attention

Aug 15, 2013
Headline news here, and a publication spot in Science, for passive, hand-assembled Tinkertoy units?!

Actual link w/supplementary info:


Image: http://www.eureka...0640.php

This is a great example of why MIT kept begging us Whitesides group postdocs down the street at Harvard to make stuff for them other than just toys.

Aug 15, 2013
This sounds very promising.

Aug 16, 2013
Beware that MIT "researchers" already stole in Nature journals and Nano Letters (American Chemical Society) both the ideas and money of taxpayers. It confirmed by the numerous swindlers from David H. Koch Institute for Integrative Cancer Research and Department of Chemical Engineering, also with Department of Chemistry and Chemical Biology and School of Engineering and Applied Science of Harvard University at .
Their plagiaristic "masterpieces" titled Macroporous nanowire nanoelectronic scaffolds for synthetic tissues (DOI: 10.1038/NMAT3404) and Outside Looking In: Nanotube Transistor Intracellular Sensors ( were funded by NIH Director's Pioneer Award (1DP1OD003900) and a McKnight Foundation Technological Innovations in Neurosciences Award, also a Biotechnology Research Endowment from the Dep. of Anesthesiology at Children's Hospital Boston and NIH grant GM073626, and NIH grants DE013023 and DE016516.

Aug 16, 2013
rsklyar's real correspondence is here:

Aug 16, 2013
In aircraft structures there are 3 requirements not one. These are 1. high strength/weight ratio, 2. stiffness tuned away from flutter modes and 3. crack-resistance to cycles of loading (fatigue). Usually, the structural designer obtains a compromise between all 3 kinds in the actual design of the structures, when using conventional materials and attachment systems. But this proposal is likely to be unsatisfactory in at least one of them, whilst more easily satisfying the rest.

Aug 16, 2013
MIT's new computer: http://postimg.or...ofz3a0v/
Their new sustainable car: http://postimg.or...0rbay75/
Their new aeroplane: http://postimg.or...dcdzfpv/
Their new smart textile: http://postimg.or...ii4lnbt/

Now with headline-grabbing 3D printed carbon fiber!

The "reversible-assembly" process is detailed here: http://postimg.or...iw3rkw5/

"Reality is obsolete." - Unofficial motto of the MIT Media Lab

Aug 16, 2013
gershenfeld is not merely a genius at invention, he's a genius at incubating other talented geniuses---which is perhaps what he will ultimately be remembered for.

gershenfeld is truly a visionary genius. the 21st century practical version of buckminster fuller, only smarter , better, and less lost in his own vision, and far more attuned to those of others'.

Aug 16, 2013
Gershenfeld's TED talk reveals the sad face of Media Lab inspired MIT "engineering":

Here, he has taken Bucky Fuller's "Octet Truss," also a favorite of Alexander Gram Bell during his kite design days ( http://postimg.or...w0sgpz9/ ), which is the simple structure you get from the closest packing of marbles in a box, and he's simply built it in manual TinkerToy fashion out of 3D printed squares:


There are more than a handful of options on how to do this exactly, but claiming a new revolution in aircraft design is just rent seeking hype in a world in which private astronaut Burt Rutan has been machine-winding carbon fiber around cheap soluble foam cores for years.

What Fuller proposed way back in the 50s was a fractal that made each strut of such a truss out of a smaller version of the truss and so on, down to atomic scale, both with solid trusses and airy tensegrity units:


Aug 16, 2013

Wouldn't this kind of design be difficult to inspect for fatigue?

Composite construction for high priced high tech stuff is cool, but kinda low hanging fruit, since those applications don't have much constraint on price or sophistication. Aerospace companies don't mind spending money if it makes their product work better.

It would be more impactful if they created a system that could more affordably construct residential and commercial structures, or third world housing, or emergency shelters.

lol Nik. I like MIT.

Aug 16, 2013
Parts made by robots, transported by robots, assembled by robots, for use by robots. and ultimately designed by robots.

-There is a bank vault Id like to try this out on-

Aug 16, 2013
Wouldn't this kind of design be difficult to inspect for fatigue?

If it's printed one could print a sensor right on top of it (a simple resistive strip would be enough).

It also reminds me very much of finite elemens from FEM/FEA. So one could do something fun like an 'inverse' finite element modeling: The computer mesh for the FEM would be equal to the real mesh of the elements. And knowing the specs of one element one could then test predicted vs. actual pliability.
Reverse solution of the error term would give you the position(s) of fatigue within the system.

Hmm..I think I'll spend a couple minutes with pen and paper on this. That idea migh actually be worth something.

Aug 17, 2013
Before you expend a lot of paper and ink you may want to do a little research to find out if it hasn't already been thought of.

Aug 19, 2013
If it's printed one could print a sensor right on top of it

I suppose, but that's a lot of interconnected parts to keep in contact with each other, unless you're thinking of wireless. That would get really complicated.

Yeah, I'm sure you would want to calculate the stress foci ahead of time, and build something into those areas, like a planned void or a different geometry.

I would worry about the failure mode being like a zipper or even worse would be like glass cracking in a fractal pattern. I'm sure they'd work that all out through testing though.

Aug 19, 2013
I suppose, but that's a lot of interconnected parts to keep in contact with each other,

Since they're all the same and you know where they will interlock creating a structure that will automatically give you an adressable array (like in a LCD-monitor array) would actually be fairly simple. You'd need to print a resistor, a diode and a transistor on each part (all of which is already feasible with conductive ink).

It seems that the failure mode is more aking to wood splintering/ripping rather than the sudden/catastrophic break you get from current materials. At least that way you have a chance that it will remain functional until repairs can be made (if the stress isn't too great).

The modular feature also has the charm that you can take it apart easily if some of them break. You also wouldn't have to manufacture specialized sections - you could print replacement cells on the spot.

I think this is an idea that has a bright future.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more