harvard develops smart knitted textiles that snap between stable shapes to switch and sense

Harvard researchers develop programmable knitted textiles

Researchers at Harvard’s John A. Paulson School of Engineering and Applied Sciences (SEAS) demonstrate that knitted textiles, long valued for their softness and flexibility, can also behave as programmable mechanical systems. Led by Kausalya Mahadevan in collaboration with Katia Bertoldi’s lab, the team develops machine-knitted fabrics that snap between multiple stable three-dimensional configurations without relying on rigid components or complex assemblies. Published in Advanced Functional Materials, the research shows how weft knitting can be used to create mechanical metamaterials using standard industrial knitting machines.

The work centers on multistability, the ability of a structure to remain stable in more than one shape, a behavior that the researchers generate entirely through yarn selection, knitting geometry, and fabrication parameters. The team produces dense textiles by combining highly elastic yarns with a technique known as plating. These naturally curl into three-dimensional forms before snapping between stable states under deformation.

a reconfigurable lamp shade with multistable switches that correspond to different colors of light | all images courtesy of Harvard’s John A. Paulson School of Engineering and Applied Sciences (SEAS)

Conductive knitted textiles function as wearable sensors

The researchers at Harvard exploit the internal geometry of knitted fabric to create programmable motion. By varying horizontal and vertical stripe arrangements, they identify combinations that reliably transition between distinct configurations, much like the mechanical action of a conventional light switch. Computational models further show that the fabrics can be predicted as continuous materials, avoiding the need to simulate every individual loop of yarn.

The approach bridges textile manufacturing with the growing field of mechanical metamaterials, where geometry is the factor that determines how a material behaves. While these systems have typically been fabricated through molding, additive manufacturing, or specialized composite production, the Harvard team demonstrates that similar nonlinear mechanical behavior can occur from a standard industrial knitting process already used throughout the garment industry.

the researchers used highly elastic yarns to create dense textiles that snap into different configurations

Smart fabrics could enable wearable technology

To explore functional applications, the researchers integrate conductive yarns into the textiles, allowing changes in shape to simultaneously alter electrical connectivity. In one prototype, a snapping knitted shell functions as an on-off switch for an LED. Another prototype places the textile over the knee or elbow, where the snapping motion is detected electronically to count movement. A third one transforms a knitted lampshade into an interactive lighting system, with separate stable states activating different colors as the fabric changes configuration.

These demonstrations suggest that knitted structures could evolve into responsive interfaces embedded directly within soft materials. Because the fabrics are produced using commercially available weft knitting machines, the work also points toward practical scalability. Existing textile manufacturing infrastructure could potentially fabricate responsive fabrics without fundamentally changing production methods, offering a pathway toward wearable interfaces, soft robotics, adaptive interiors, and programmable products.

the team develops machine-knitted fabrics that snap between multiple stable three-dimensional configurations | image sourced from ‘Multistable fabrics that snap between different shapes’

the research shows how weft knitting can be used to create mechanical metamaterials using standard industrial knitting machines | image sourced from ‘Multistable fabrics that snap between different shapes’

the work centers on multistability | image sourced from ‘Multistable fabrics that snap between different shapes’

the fabrics can be predicted as continuous materials, avoiding the need to simulate every individual loop of yarn | image sourced from ‘Multistable fabrics that snap between different shapes’