Getting their inspiration from the biological structure of actual human skin, the researchers embedded a grid of molded square carbon nanotube (CNT) pyramids into a polyurethane (PU) matrix to form the top electrodes of an array of capacitors. The bottom electrodes were also lined up with CNTs, forming a two-dimensional array of molded hills which mimicked the spinosum layer in human skin. These two electrode layers separated by a thin-film dielectric layer allowed the formation of conformable multi-pixel capacitors (a 5x5 top grid centred over each bottom hill) whose electrical response is highly dependent on the skin's deformations under external pressure.
The e-skin's configuration is such that for each hill about 1mm in diameter and 200μm high corresponds 25 tiny capacitors each only 90μm2 in size, with one capacitor precisely located at the top of the hill, four on the slopes, four on the "corners" and 16 surrounding the hill. This high density of mechanoreceptor-like sensors ensured the e-skin would respond differently depending on the pressure's direction or even based on drag deformation, being able to measure and discriminate in real time between normal and shear forces.
In various experiments including one with robot fingers grabbing a delicate raspberry, the researchers were able to establish a capacitance map around each hill, which allowed them to differentiate several types of applied forces with a response time in the order of the microsecond, adapting its grip force to the detection of shear forces (to adjust against slip).
"It is possible, by looking back at a recorded signal, to evaluate the nature of an unknown stimulus based on the combination of amplitude, shape, and frequency of the signal by referring to a previously known library of stimuli response curves", the authors report.