Electronic textiles that can measure movements and forces hold great promise in the seamless monitoring of human health, the intuitive interactions of humans with machines and virtual worlds, and the proprioception and exteroception of soft robots. Imagine, for instance, a hospital bed that can measure your respiration without the need of obtrusive cables attached to your body, or, in another example, a robot equipped with a sensorized skin that you can naturally interact with through touch. Unfortunately, the technological realization of this futuristic vision is more than challenging, because the surfaces in question are very large (in the example of the bed ~2 m2), and because textiles are usually exposed to all sorts of loads and movements, necessitating the devices to be flexible and often even stretchable. Past developments in electronics, based on fragile and dense arrays of small sensors, work wonderfully in smartphones but are evidently not a good fit for textiles.
In our lab at EPFL, we take a different approach: rather than relying on point sensors that were designed for small and hard devices, we impart functionality within the fibers that make up the very textiles. With these fibers, which usually boast hundreds of meters of length and are flexible by nature, we can easily functionalize large surfaces with a minimum number of contact points. By combining ever new combinations of materials in complex microstructures, we have extended our reach from fibers that sense light to fibers that repel water, fibers that generate electricity, fibers that deliver drugs, and even fibers that can be eaten (if this research intrigues you, you can find more information in our review paper). However, in the case of fiber-based sensors that detect deformations, we have always faced ever-present challenges: how do we determine both the intensity and the position of a mechanical load? How do we measure multiple loads at once? How can we differentiate between different types of loads, such a pressing and stretching?
Examples of soft transmission lines. Photograph of a line featuring two liquid metal microchannels in a thermoplastic elastomer cladding (left). Micrograph of the cross-section of a coaxial line exhibiting 39 liquid metal microchannels (right).
In our recent work, we found that all these questions can be answered with specialized fibers that act as soft transmission lines. We selected a thermoplastic elastomer dielectric and a liquid metal conductor as the constituting materials because they are inherently soft and elastic, enabling the reversible application of large strains under low loads. To enable radio frequency transmission with effective shielding and low losses, we had to meticulously arrange liquid metal microchannels, up to 39 in number, within the elastomer and flawlessly maintain the coaxial structure over extended lengths of fiber. We found that the processing of these unusual materials, with the necessary cross-sectional integrity and at a relevant scale, could be achieved through the thermal drawing technique—the same process than the one used in the optical fiber industry.
Once fabricated, we discovered that the soft transmission lines featured unprecedented sensing functionality when interrogated by time-domain reflectometry—a time-of-flight technique in which high-frequency pulses are reflected at discontinuities in impedance-controlled transmission lines. For one, pressures applied on the soft transmission lines can be accurately measured in terms of both intensity and position. Additionally, the lines can undergo large levels of stretching, and the events quantified and localized. Best of all, the diverse mechanical stimuli can all be simultaneously detected on a single line. To demonstrate that our concept exhibits higher functionality than existing solutions that require hundreds of point sensors and electrical connections, we created a large electronic textile with a single soft transmission line interfaced through a single contact point to a custom reflectometer. Based on the range of outputs that the soft transmission line produces, the actual configuration of the electronic textile, which may undergo convoluted changes in the form of pressures and stretches, can be successfully reconstructed. In future work, we hope to scale down the size and price of peripheral electronics to make this new paradigm in fiber-based sensing more accessible.
An electronic textile for multimodal deformation sensing. A soft transmission line interfaced through a single contact point is in integrated on a stretchable textile and exposed to stretching and pressing (left). The output of the line is processed in real-time to a map indicating changes in length and loads (middle) and to a measure of line elongation (right).