Intrinsically Stretchable Semiconducting/Conducting Nanocomposites for AI-Aided Visual and Tactile Sensing Systems

The human sensory systems, particularly vision and tactile, play a crucial role in our daily lives. For individuals who have lost or are born without these sensory capabilities, the world can be a different and challenging place. Developing electronic devices capable of effectively perceiving and transducing these sensory modalities is an important challenge to improve the quality of life for such individuals. Especially, the need for stretchable electronic devices that can conform to the movements of the human body has driven extensive research in this area. Two representative approaches have been explored to achieve stretchable electronics. i) One method is to adopt high-performance inorganic materials together with strain-dissipative designs, which can maintain their electrical performance under mechanical deformations. However, the large area occupied by the structural designs can limit the device density. ii) The alternative method is to adopt intrinsically stretchable materials, such as elastic conductors and semiconductors using a mixture of organic/polymeric materials and elastomers, which does not require the use of special device structures. However, there are still challenges in terms of the electrical performance and durability of materials and devices under severe repetitive mechanical deformations.<br/>In this presentation, we introduce two studies that use intrinsically stretchable semiconducting/conducting nanocomposites to overcome these challenges, mimicking the functionality of visual and tactile sensing systems. First, we report on the development of a retina-like stretchable multiplexed arrays. This new class of the intrinsically stretchable artificial retina consists of organic-inorganic hybrid semiconducting nanocomposites, crack-based gold nanomembrane electrodes, and elastomeric substrate. To enable effective light-sensing capabilities, the composite incorporates vertically phase-separated, size-tunable quantum dots that facilitate efficient charge transfers. The individual phototransistor cells can be either multiplexed or stacked in a misaligned manner using a transfer-printing method, allowing for high areal density. Additionally, a deep learning algorithm is utilized to fully compensate for optical distortions during repetitive mechanical deformations of the devices. This enables the accurate recognition of specific color patterns, such as red, green, and blue, even under harsh strain conditions, thereby mimicking the unique biological functionality of the retina.<br/>Second, we report on the development of a skin-like intelligent wearable system, a bioinspired stretchable sensory-neuromorphic system, comprising an artificial mechanoreceptor and artificial synapse. These biomimetic functionalities correspond to a stretchable capacitive pressure sensor and a resistive random-access memory, respectively. This system features a rigid-island structure interconnected with a sinter-free printable conducting nanocomposite optimized by controlling the evaporation rate of solvent. The proposed design enhances both areal density and structural reliability while preventing the thermal degradation of heat-sensitive stretchable electronic devices. In addition, the system is able to accurately recognize different patterns of stimuli through an artificial neural network with training and inferencing functions, even in the range of skin deformation.