Emulating artificial mechanoreceptor functionalities from SiO2-based memristor and PDMS stretchable sensor for artificial skin applications

The development of robust artificial sensory systems that will be able to detect, memorize and respond to external stimuli is attracting a considerable amount of interest over the last years towards emulating artificial skin properties. Skin is the largest sensory system of the human body that carries out complex procedures by employing a variety of biological receptors. However, it is quite difficult to acquire stretchable artificial skin functionalities by combining the present tactile sensors. Along these lines, we report here a sensory system that has the ability to detect various strain loads with excellent sensitivity and at the same time respond and store the applied stimuli. In order to attain this goal, we used a bipolar switching memory device made on poly-ethylene naphthalate (PEN) substrate and Pt nanoparticles (NPs)-based sensor on poly-dimethylsiloxane (PDMS). The memristor operates only under strain conditions and retains the induced resistance changes even when the applied stain gauge is removed with <20% resistance attenuation. The sensing element exhibits a colossal gauge factor (GF) of 10
⁹
, which is utilized in order to produce sharp and distinct responses. Moreover, the memristive element exhibits also a quite large memory window (∼10
⁵
) which can be leveraged in order to assimilate various external stimuli. Additionally, paired-pulse facilitation (PPF) and long-term plasticity (LTP) effects are demonstrated indicating the potential perspective of our approach to performing neuromorphic computations with low power consumption.