Advanced Computing in the Age of AI | Thursday, March 28, 2024

4D Printing: Manufacturing Intelligent Machines 

<img style="float: left;" src="http://media2.hpcwire.com/dmr/4dprinter.jpg" alt="" width="95" height="54" />Each week, 3D printing seems to be behind yet another breaking news headline. In light of recent breakthroughs in stem cell and live-tissue printing, it feels like we've only just now realized the depth of additive manufacturing's potential. But a new idea from MIT may have given us a glimpse of the shape of things to come: not 3D, but 4D printing.

Each week, 3D printing seems to be behind yet another breaking news headline, revolutionizing industries from aerospace to biomedical research. In light of recent breakthroughs in stem cell and live-tissue printing, it feels like we've only just now realized the depth of additive manufacturing's potential.

But a new idea from MIT may have given us a glimpse of the shape of things to come: not 3D, but 4D printing.

You may be asking yourself “Well, what is the fourth dimension you can print in?,” The answer? Time.

Okay, so maybe you can't print “time.” But what a team of MIT researchers has been able to print is a strand of materials that will continue to evolve into different forms after the printing process is completed, without any human intervention. This evolution of the object over time adds to the product the extra dimension that gives this technology its name.

MIT's 4D printing process actually begins with a Stratasys 3D printer that produces a string of multi-layered materials. Each part is comprised of a rigid plastic inner layer, topped by an outer layer of absorbent materials. When submerged in water, the outer layer absorbs water to expand and distort until the strand folds in on itself at certain joints to form a predetermined shape.

We've now seen these self-assembling objects arrange themselves into cubes and even the MIT logo, but the eventual goal is to manufacture objects that can continue to change form multiple times. This would effectively extend the temporal lifespan of the 4D printing process from a matter of minutes to hours, days, or perhaps even years.

This research was presented at this year's TED conference in Long Beach by MIT professor and director of the university's Self-Assembly Lab, Skylar Tibbits, who hopes to one day print materials that will take advantage of their extended longevity to self-assemble and self-repair.

For anyone who remembers Bio 101, this process may sound familiar on a number of levels. Not only would these products be able to adapt to their environment, but the act of self-assembly bears resemblance to protein folding. In both cases the environment acts on the flattened self-folding strand (either a chain of amino acids or a 3D printed string) which then reacts by folding itself into a specific origami-like structure.

This demonstrates what that extra dimension – time – brings to the table that 3D printing has overlooked: allowing technology to evolve to the point where it resembles natural systems. In a recent TED interview, Tibbits defined the technology in terms of basic instinctual drives that are inherent in all organisms but almost universally lacking in man-made systems.

Next -- Bringing Intelligence to the Machines >

Bringing Intelligence to the Machines

Tibbits explains, “Natural systems obviously have this built in – the ability to have a desire. Plants, for example, generally have the desire to grow towards light and they generate energy from the translation of photosynthesis, carbon dioxide to oxygen, and so on.

“This is extremely difficult to build into synthetic systems – the ability to “want” or need something and know how to change itself in order to acquire it, or the ability to generate its own energy source. If we combine the processes that natural systems offer intrinsically (genetic instructions, energy production, error correction) with those artificial or synthetic (programmability for design and scaffold, structure, mechanisms) we can potentially have extremely large-scale quasi-biological and quasi-synthetic architectural organisms.”

But Tibbits was quick to point out that 4D printing is not copying directly from nature. He noted that outright mimicry can lead to fundamental design flaws, because natural systems have evolved over millions of years to function under highly specific conditions. One cannot simply translate these systems to other scales and functions and expect the same results.

Instead of trying to reproduce living systems, Tibbits says that his approach involves embedding “dead” materials with information to give it more “active characteristics.”

“I’m trying to discover how much information can they store, how can they replicate inherently, how can they move and assemble themselves, and so on. None of these properties are necessarily found within the materials themselves. Rather, it’s a different way of looking at the materials and at the way we build things.”

Tibbits' ultimate goal is to create 4D printed components that will serve as building blocks in architecture, although this project is still a ways off from reaching the skyscraper scale. In the meantime, his efforts are centered around seeding growth in defining constraints in his 4D objects in hopes of creating an even more dynamic system.

But why do we need self-assembling pieces? Why not just directly print the desired end result? According to Tibbits, there was no other way to build certain structures at small scales, especially when wandering into the realm of nanotechnology. Instead, they had to work on the materials' terms, which meant self-assembly.

Furthermore, by using materials that can generate their own directives and act of their own volition, we are opening the door for objects and even large-scale structures such as buildings that can one day construct and repair themselves.

As for why self-assembly came about in the first place in the face of all the possibilities offered by 3D printing, Tibbits sees two main possibilities. It's possible that we've pushed ourselves to the limits of existing materials and processes, forcing the creation of a new technology such as his. Or this simply came about to serve designers in need of something novel to spark their inspiration.

Either way, what is clear from this technology is that while 4D printing may have a significant presence in the future of manufacturing, it by no means replaces 3D printing. Instead, each technology will likely have its own spheres of influence, and will build upon and complement one another. So no matter how close to nature, this shouldn't become a case of 'survival of the fittest.'

EnterpriseAI