# Quasi‐Periodic Porous Structures‐Based Temperature and Pressure Dual‐Mode Electronic Skin for Material Cognition

## TL;DR

A new electronic skin with a special porous structure can accurately detect temperature and pressure to identify different materials with high accuracy.

## Contribution

A quasi-periodic porous structure enables precise control of temperature and pressure sensing for superior material cognition.

## Key findings

- The electronic skin can distinguish 33 materials with 97.64% accuracy using thermoelectric and piezoresistive signals.
- Quasi-periodic structures allow better performance than random structures in material identification.
- The system can identify materials with similar properties, such as cotton fabrics and alloys.

## Abstract

It is of great significance to prepare a temperature and pressure (T‐P) dual‐mode electronic skin (DMES) with a controllable porous structure and use it to achieve material cognition. In this work, a quasi‐periodic porous structure‐based T‐P DMES was proposed, exhibiting excellent performance in material cognition. The quasi‐periodic porous structure of the electronic skin was prepared through the interaction of confined two‐dimensional bubbles and polydimethylsiloxane (PDMS). After attaching graphene, PEDOT: PSS, and Bi2Te3 to the porous PDMS, the obtained electronic skin can precisely detect and distinguish T‐P stimuli without crosstalk. Benefiting from the quasi‐periodic porous structure, the temperature and pressure sensing performance of the electronic skin can be precisely constructed and optimized by changing the size of the porous structure, which is impossible for electronic skin with a random porous structure. By analyzing the thermoelectric and piezoresistive signals of the electronic skin and combining them with a convolutional neural network, the electronic skin can identify 33 different materials, including materials with similar softness or thermal conductivity, cotton fabrics with different textures, and even diverse alloys, with an accuracy of 97.64%. The proposed T‐P DMES significantly outperforms both existing electronic skins and human skin in terms of material cognition.

A quasi‐periodic porous structure‐based temperature and pressure dual‐mode electronic skin was proposed. Benefiting from the quasi‐periodic porous structure, the temperature and pressure sensing performance of the electronic skin can be precisely constructed and optimized by changing the size of the porous structure. By analyzing the thermoelectric and piezoresistive signals of the electronic skin and combining them with a convolutional neural network, the electronic skin can identify 33 different materials, including materials with similar softness or thermal conductivity, cotton fabrics with different textures, and even diverse alloys, with an accuracy of 97.64%.