- American Physical Society Sites
- Meetings & Events
- Policy & Advocacy
- Careers In Physics
- About APS
- Become a Member
By Sophia Chen
Any knitter knows the vast design potential of a single ball of yarn. By using different knitting techniques, that single long strand can be transformed into a scarf, a sweater, a sock, or any number of other shapes. Researchers want to harness knitting’s versatility and incorporate it into new technologies such as soft robotics and wearable electronics. In a session at the 2021 APS March Meeting, researchers shared prototypes, simulations, and theoretical models of knitted objects for investigating the potential role of knitting in technology.
While knitting is an ancient craft, some researchers today are viewing it through a new lens: as a computer algorithm. A knitter follows a prewritten protocol, or pattern, to create a 2D or 3D shape with the desired aesthetic or function. The pattern acts as an algorithm, determining the knitted object’s geometry and elasticity.
Building on the idea of patterns-as-algorithms, researchers are using computer code to operate knitting machines. “It’s a lot more programming than I ever expected,” says Vanessa Sanchez of Harvard University, a fourth-year graduate student who studies textile materials for soft robotics. For her experiments, she uses a knitting machine language called Knitout developed by Carnegie Mellon University.
Knitting could be a useful technique for creating wearable electronics, such as those made of conducting material woven into yarn. Its design advantages are similar to those of 3D printing, in that knitting machines additively create fabrics of composite materials, says Sanchez. Beyond clothing, knitting could also become useful for constructing buildings. In 2018, architecture researchers created a knitted 3D frame that could support 5 tons of concrete.
Knitted fabric has many design advantages over regular fabric. In particular, it can produce a 3D shape without cutting and sewing fabric, eliminating complicated manufacturing steps, thus decreasing the risk of errors. Knitting keeps manufacturing simple, says Sanchez.
Knitting can create 3D structures with a wide range of geometries and mechanical properties.
Sanchez develops knitted materials for making soft robots, such as clothing that contains a motion-tracking sensor to aid people with mobility impairments. “I want to make garments that help people,” says Sanchez.
At the March Meeting, Sanchez presented a sleeve made of knitted yarn, fitted around an inflatable balloon-like pouch, that bends in a desired direction when the balloon inflates. This sleeve could function as an actuator within an assistive glove that helps the wearer grasp objects. As a proof of concept, said Sanchez, the sleeve shows that it’s possible to make an actuator out of knitted garments. “And we can do it all with one yarn,” she adds.
To make the actuator bend in the right direction, Sanchez applied stiffer or stretchier patterns to different parts of the sleeve. In her design, she relied on experiments informed by her prior experience working with textiles. But the physics of these knit projects are complex, and researchers struggle to design materials that consistently produce their desired properties.
The geometry of the yarn, such as how each loop locks into the next, as well as the yarn’s tensile properties and local friction, can result in surprising properties such as fabric curling. To streamline the design process, researchers are developing more physically precise theoretical models to predict fabric mechanical properties.
Xiaoxiao (Catherine) Ding, a third-year graduate student in applied math at Harvard University, presented theoretical research for predicting a knitted fabric’s elasticity. Her group, led by Chris Rycroft at Harvard, has developed a modeling framework that uses yarn properties, such as stiffness and friction, to simulate its characteristics once it is knitted into a fabric.
Ding validated the model by making knitted swatches, performing mechanical tests on them, and comparing their behavior to their simulations. The model can simulate knit patterns made of multiple types of yarn material, such as nylon and plastic. Her group is also working to apply the model in reverse, where it can design a pattern based on a desired material property.
Ding compares knitting to more traditional materials science, in which different chemical combinations yield materials for specific functions. Instead of mixing and matching chemicals, knitters play with geometry to get the material they want, she says.
Some theoretical work discussed in the session applied to materials beyond yarn. Daria Atkinson, a postdoc at the University of Pennsylvania, presented her research on the properties of filament bundles—think a rope consisting of twisted fibers. These geometries occur at scales ranging from DNA strands to spun yarn to steel cables used on suspension bridges.
While a braided rope looks simple, researchers have struggled to theoretically describe how the strands’ geometry, as they twist around each other, gives rise to their mechanical properties. In her work, Atkinson found new constraints on the configurations of the strands as of a rope when it is bent or twisted.
The author is a freelance science writer based in Columbus, Ohio.
Video recordings of these and other textile physics presentations will be available on the meeting website (march.aps.org) until June 19, 2021.
©1995 - 2021, AMERICAN PHYSICAL SOCIETY
APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.
Editor: David Voss
Staff Science Writer: Leah Poffenberger
Contributing Correspondents: Sophia Chen, Alaina G. Levine