Fiber Sensors with a Twist

The double helix of DNA may be the most common structure in the world. Stretched out fully, the DNA in a single human cell is about six-and-a-half feet long. If the DNA from all the cells in a single person were lined up, end to end, it would reach 41 times further than the distance between Earth and the Sun. Now consider that there are billions of people, and countless other animals, plants, and single-celled organisms on the planet. That is an awful lot of double helices!

There must be something special about a structure as pervasive as that of DNA. Researchers at Shinshu University in Japan think so, and they have leveraged its unique shape to create a better fiber sensor design. Fiber sensors are surging in popularity in the world of wearable electronics due to their light weight, flexibility, and comfort when worn on the body. But these sensors have a problem — repeated movement causes them to become less accurate, and eventually fail.

To address this problem, the research team developed a new type of fiber sensor that mimics the double-helical structure of DNA. Traditional fiber sensors have electrodes attached at both ends, which tend to weaken or break when placed over joints like fingers or knees — areas that move frequently. In contrast, the new design places both electrodes on the same end, significantly improving durability and reliability during motion.

The novel sensor consists of two coaxial fibers twisted into a double helix. Each fiber has a conductive inner core made from multi-walled carbon nanotubes and a strong insulating shell composed of thermoplastic polyurethane and titanium dioxide nanoparticles. This layered structure is created using a technique called coaxial wet-spinning. After heat treatment, the twisted fibers fuse into a stable double-helix with built-in terminals at one end.

With a diameter of less than 1 mm, the sensor is slim enough to be woven into clothing. Tests show it can stretch to over 300% of its original length and endure more than 1,000 stretching cycles without breaking. These characteristics make it ideal for tracking motion, such as finger gestures, facial expressions, or even breathing patterns.

In one demonstration, the team integrated the sensor into a smart glove. This glove was attached to an Arduino UNO for data collection. Using a convolutional neural network for classification, the glove showed that it could recognize six common hand gestures with an impressive 98.8% accuracy. It could even detect the length of finger movements and use that information to transmit Morse code wirelessly, showing potential for aiding people with communication impairments in the future.

Beyond wearables, the double-helical sensor could also monitor fluid flow, making it useful for a broader set of applications. Its ability to simplify wiring and reduce mechanical failure opens up possibilities for real-time remote health monitoring, sports training, and safety systems in extreme environments like mountaineering. The work could help bring about a new generation of intelligent, reliable, and comfortable wearable devices.