Humidity sensors, as fundamental wearable electronic devices, offer a non-contact sensing mode, avoiding the drawbacks of contact sensors like mechanical wear and cross-infection. Their potential spans healthcare monitoring, human-machine interfaces (HMIs), sentiment analysis, and more. Existing humidity sensors often use substrates like polyimide (PI), polyethylene terephthalate (PET), and polydimethylsiloxane (PDMS), but these lack breathability, leading to skin irritation. To address this, researchers have explored porous substrates such as paper and fabric, but their thickness hinders conformal skin contact. Electrospinning offers a route to ultrathin, porous substrates. While previous work using electrospun poly(vinyl alcohol) nanofibers (PVA NFs) showed promise, response/recovery times were slow. This study introduces a novel skin-conformal and breathable humidity sensor based on poly(styrene-block-butadienstyrene) nanofibers (SBS NFs) and alkalized MXenes/polydopamine (AMP) composite to overcome these limitations. The porous SBS NFs ensure breathability, while the AMP composite offers high sensitivity due to its large surface area and hydroxyl groups. The Young's modulus of SBS NFs closely matches that of human skin, promoting conformal contact. The sensor aims to achieve high sensing performance, enabling the recognition of breathing patterns and emotional states for advanced non-contact HMI applications.

The existing literature highlights the potential of humidity sensors for various applications but also reveals limitations in current technologies. While substrates like PI, PET, and PDMS offer good humidity sensitivity, their lack of breathability causes skin irritation. Porous substrates such as paper and fabric improve breathability, but their thickness prevents conformal contact with the skin. Electrospinning emerges as a promising technique for creating ultrathin, porous substrates. Studies using electrospun PVA NFs have shown potential, but slow response times remain a challenge. Other studies have explored MXenes for their high surface area and humidity sensitivity, often combined with other materials to enhance performance. This review establishes the need for a humidity sensor that combines high sensitivity, fast response time, excellent breathability, and conformal skin contact—a gap this research aims to fill.

The SAMP-based humidity sensor comprises an AMP composite and interdigital electrodes (IDEs) on an ultrathin SBS NFs substrate. The fabrication process begins with electrospinning SBS NFs to create a porous, flexible, and breathable substrate. Ag NWs-based IDEs are then patterned onto the substrate via air spraying. The humidity-sensitive AMP composite is synthesized by alkalizing MXenes and incorporating polydopamine (PDA). The AMP composite is drop-casted onto the IDEs, completing the sensor assembly. The thickness of the sensor is approximately 26 µm. The Young's modulus of SBS NFs closely matches human skin's, ensuring conformal contact. Various characterization techniques, including SEM, EDS, XRD, XPS, and electrical measurements, are employed to analyze the sensor's morphology, composition, and performance. Water vapor permeation tests assess air permeability. The sensor's response is tested under varying humidity conditions to determine sensitivity, response/recovery times, hysteresis, and long-term stability. For application demonstration, the sensor is used to monitor breathing patterns during exercise and to distinguish emotional states (normal, fear, pain, wonder) using a machine learning algorithm (SVM). A 3x3 sensor array is employed to create a non-contact HMI system to control a robot car.

The SAMP-based humidity sensor demonstrates high sensitivity (S=704) and fast response/recovery (0.9 s/0.9 s). The ultrathin (~26 µm) and porous SBS NFs substrate ensures high air permeability (0.078 g cm⁻² d⁻¹). The sensor successfully monitors breathing patterns during different exercise states (breath-holding, resting, walking, running), showing distinct breathing depths and frequencies. The sensor accurately recognizes four emotional states (normal, fear, pain, wonder) with an 86.7% accuracy using an SVM machine learning algorithm. A 3x3 sensor array is used to develop a non-contact HMI system, effectively controlling the motion of a robot car by detecting finger sliding tracks. Long-term stability and biocompatibility testing confirm the sensor's suitability for long-term on-skin use.

The high sensitivity, fast response, and breathability of the SAMP-based sensor address the limitations of existing humidity sensors. The ability to accurately detect respiration and emotional states opens opportunities for improved healthcare monitoring and personalized feedback systems. The development of a non-contact HMI system using a sensor array demonstrates potential for intuitive and hygienic human-machine interaction. These findings demonstrate significant advancement in wearable sensor technology, potentially impacting areas such as personalized healthcare, emotion-aware computing, and advanced robotic control. The successful integration of MXene-based materials with electrospun nanofibers showcases a versatile approach for creating high-performance wearable sensors.

This research successfully demonstrates a skin-conformal and breathable humidity sensor with exceptional sensitivity and response time, suitable for long-term on-skin use. Its applications in monitoring breathing patterns and emotional states, as well as its integration into a non-contact HMI system, highlight its versatility. Future research could explore integrating this sensor into more complex wearable systems for continuous health monitoring and advanced human-computer interfaces, and investigate the application in other areas like environmental monitoring and industrial process control.

While the sensor demonstrates excellent performance, several limitations should be noted. The current study focused on a limited number of participants in the emotional recognition experiments, necessitating further testing with a larger and more diverse group to validate the findings. The air pressure was found to have a noticeable effect on sensor performance; therefore, further optimization and calibration might be needed for robust real-world applications. Long-term reliability in diverse environments and rigorous clinical trials are needed before widespread clinical adoption.