Some technological advances happen by accident. Some happen on purpose. And some are the result of concerted and creative efforts by researchers to explore areas of common interest.
A team of scientists led by Dan Raviv of the Massachusetts Institute of Technology (MIT) has developed a manufacture workflow that combines computationally driven design with printable primitive components, aided by realistic simulations. With this framework, users can design and print non-trivial structures that bend and stretch in response to their environment. They are not unlike the water-capsule children’s toys that blow up into dinosaurs in water. However, these active structures extend the process to more sophisticated shapes and functionalities that can exhibit controllable responses to a greater variety of external stimuli, including electric fields, temperature, and light. The team’s work is described in the December 2014 issue of Scientific Reports (DOI: 10.1038/srep07422).
The design step implemented three basic structural units (or primitives): one bending primitive, and two stretching primitives. These primitives are constituent components of the larger, complex shape being manufactured. The bending primitive relies on differential expansion between fused materials. When stimulated, one side expands at a greater rate than the other side, resulting in a curve. The angle and the plane of the curve are controlled by the dimensions and relative position of the two printed materials.
The two stretching primitives use two different schemes. In the first scheme, the mechanism is similar to the bending primitive, except that there is no differential expansion in the printed object, so the growth is linear. The second stretching scheme uses rings. The rings are printed with different materials on the outside and inside; the inner material expands when stimulated so that the ring elongates, resulting in linear stretching. The mathematical models used in the workflow were developed so the user can design the final form of the active structures, and the computer will print the initial configuration.
The team used a Stratasys Connex 500 multi-material three-dimensional (3D) printer in their work, which enabled them to directly integrate the different materials into the printed structures. The expanding materials were acrylated monomers. When activated—in this case, the materials were hydrophilic, but potentially any stimulus could be used—the monomers assembled into linear chains with a few difunctional acrylate molecules. A hydrogel is created with as much as twice the volume of the original material.
Once they achieved their final form, the printed structures were compared to a simulation. The team used the Project Cyborg design platform, enhanced with Autodesk’s Nucleus system and some additional functionality for modeling the temporal aspect of the deformations. Simulations were restricted to kinematic behavior and did not model atomic movement.
“This was not a one-(person) job. This was truly a multi-author work,” Raviv spoke of the project. There were many parts to the project, across a range of fields: design, mathematics, computation, and materials science. The breadth of the project brought together several laboratories including Raviv’s laboratory at MIT, laboratories at Autodesk Inc., Stratasys Ltd., and Singapore University of Technology and Design. Skylar Tibbits at MIT’s Self-Assembly Laboratory also contributed significantly. Their work anticipates developments in—and increasing use of—soft robotics, where this has many potential applications in vivo. A thermally activated stent, for example, could be printed and inserted into a collapsed artery where it would expand and open the artery on its own. The confluence of materials science, 3D printing technology, and computational engineering heralds a future in which these active, self-evolving structures find many invaluable uses.