3D-printed microstructured ionic hydrogel and nano-friction generator for self-powered electronic skin and deep learning character recognition

Developing flexible sensing materials that integrate high conductivity, mechanical robustness, self-healing capabilities, and environmental stability remains a critical challenge for next-generation wearable electronics. However, conventional hydrogels often suffer from intrinsic trade-offs between these properties and are prone to freezing-induced deactivation. Herein, a multifunctional POAL hydrogel was engineered via a synergistic strategy combining the double-helix network of agar, dynamic Schiff base crosslinking of oxidized sodium alginate (OSA), and ionic coordination of LiCl. This unique multi-network architecture endows the hydrogel with outstanding comprehensive properties, including a tensile strength of 111.9 kPa, an elongation at break of 158%, and a self-healing efficiency of 48.9%. Notably, the hydrogel exhibits exceptional ionic conductivity (2.25 S/m and anti-freezing tolerance down to −42.1 °C, ensuring reliable operation in harsh environments. Leveraging its excellent rheological properties, micro-triangular pyramidal arrays were precisely fabricated via Digital Light Processing (DLP) 3D printing to construct a high-performance triboelectric nanogenerator (POAL-TENG). By maximizing the effective contact area, the device achieved a peak output voltage of 239.9 V and a superior pressure sensitivity of 89.14 mV Pa −1 , effectively harvesting mechanical energy. Furthermore, to realize intelligent sensing, a hybrid Deep Learning model (ResNet-18 + Bidirectional LSTM) was integrated to process handwriting signals, achieving a 97% recognition accuracy. This work presents a comprehensive strategy for designing robust, freeze-tolerant, and intelligent self-powered sensing systems, expanding the horizons of human-machine interaction (HMI). • Multifunctional POAL hydrogels integrate high ionic conductivity (2.25 S/ m), robust self-healing, and anti-freezing tolerance (−42.1 °C). • DLP 3D-printed micro-pyramidal TENGs achieve high voltage (239.9 V) and sensitivity (89.14 mV Pa -1 ) for efficient energy harvesting. • Hybrid Deep Learning model (ResNet-18 + BiLSTM) enables intelligent handwriting recognition with 97% accuracy for advanced HMI.