Thin and soft Ti3C2Tx MXene sponge structure for highly sensitive pressure sensor assisted by deep learning

Increasing demand for flexible sensors in healthcare has spurred the development of nanomaterial-polymer composite sensors. However, existing pressure sensors suffer from low sensitivity due to poor interactions between the functional filler and the polymer matrix. For example, the direct chemical bonding of nanomaterials to the surface of chemically inert polymers such as polydimethylsiloxane (PDMS) is not easily achieved. Herein, we demonstrate that MXene can significantly promote the sensitivity of a thin and soft piezoresistive sensor based on a surface functionalized PDMS (FPDMS) sponge. Using plasma treatment, we constructed a MXene-FPDMS (MFP) sponge with high mechanical strength via the interlayer hydrogen bonding effect. The formed MFP sponge structure (similar to 450 mu m in thickness) endows the pressure sensor with a high sensitivity of 14.2 kPa(-1) and a low modulus of 9.7 kPa, critical for detecting ultra-low pressures. When operated with deep learning algorithms, the MFP sponge sensor successfully classifies the pronunciation of 26 letters and various polite expressions with an average accuracy of 94 +/- 0.6 %. We envision that this work paves the way for the development of intelligent wearable platforms.