Ultra-Sensitive Iontronic Pressure Sensor with Femtosecond-Laser-Engraved Microstructures for Machine-Learning-Based Tactile Sensing

Abstract

Flexible pressure sensors that mimic human skin are attractive for electronic skin, soft robotics, and healthcare, but it remains difficult to combine ultrahigh sensitivity, wide range, and long-term stability in one device. Here, we present an ultra-sensitive iontronic pressure sensor (USIPS) based on a fingerprint-inspired spiral interpolating electrode, a laserprocessed spacer, and a thermoplastic polyurethane (TPU)/graphene/multi-walled carbon nanotubes (MWCNTs)/ionicliquid composite layer microstructured into cylindrical protrusions by femtosecond-laser engraving. This architecture amplifies contact mechanics and electric-double-layer modulation, delivering ultrahigh sensitivities of ~1.92 × 10⁵ kPa⁻¹ (0-110 kPa), ~6.58 × 10⁴ kPa⁻¹ (110-300 kPa), and ~1.91 × 10⁴ kPa⁻¹ (300-900 kPa), together with a wide detection range (from ~577 Pa to 900 kPa), fast response (~20 ms), and excellent stability under prolonged high-pressure loading. Coupled with a tailored signal-processing and machine-learning pipeline, the USIPS enables quantitative hardness/softness perception and accurate discrimination of representative materials, while also supporting pulse monitoring, robotic grasping, and plantar gait analysis. These results demonstrate a unified iontronic-AI platform for wide-range, high-fidelity tactile sensing and intelligent perception.