Roll-to-roll fabrication of touch-responsive cellulose photonic laminates

Hydroxypropyl-cellulose (HPC), a derivative of naturally abundant cellulose, can self-assemble into helical nanostructures that lead to striking colouration from Bragg reflections. The helical periodicity is very sensitive to pressure, rendering HPC a responsive photonic material. Recent advances in elucidating these HPC mechano-chromic properties have so-far delivered few real-world applications, which require both up-scaling fabrication and digital translation of their colour changes. Here we present roll-to-roll manufactured metre-scale HPC laminates using continuous coating and encapsulation. We quantify the pressure response of the encapsulated HPC using optical analyses of the pressure-induced hue change as perceived by the human eye and digital imaging. Finally, we show the ability to capture real-time pressure distributions and temporal evolution of a human foot-print on our HPC laminates. This is the first demonstration of a large area and cost-effective method for fabricating HPC stimuli-responsive photonic films, which can generate pressure maps that can be read out with standard cameras.

: HPC film surface instability associated with knife-over coating. ab, Driven by the motion of the PET substrate, the HPC feed forms a tumbling vortex at subsequent knife shearing, causing flow disturbances which result in streak defects in the coating (c).
Supplementary Figure 4: a, Schematic of peristaltic pump: rotor periodically compresses tube to pump HPC inside. The intermittent pressure generated during fluid transport results in periodic fluctuations ("chatter" defects) in the HPC slot-die coating (b). Pulsation is reduced by adding a pressure dampener along the pump line (depicted in Fig. 1a), ensuring homogenous coating (c). Figure 5: Colour relaxation of a green HPC laminate as the stress in the mesophase from the fabrication process is relieved. Times are from completing the R2R fabrication. Characteristic colour is restored from orange to green in the final state at equilibrium. Figure 6: Force sensor calibration. The force sensor (inset) is calibrated using a set of standard weights. Raw data show a linear trend with standard deviation 0.03. Calibration repeated until data matches the reference.

Supplementary Figure 7:
Integrating sphere spectra with 1s integration (left plot, smoothed) as sample pressure is increased; end-time of each as indicated. Data from force sensor is simultaneously recorded every 400 ms, and force vs time is interpolated (middle plot). For each spectrum, the average force over each spectral acquisition (middle plot, coloured lines) is then converted to a calibrated pressure in kPa. The hue h° is calculated for each spectrum (right plot, coloured points), plotted against pressure (right plot, solid line) and interpolated. This interpolation (right plot, dotted lines) provides the pressure calibration between h° and pressure.

Supplementary Figure 8:
On the camera set-up, video frames are recorded by the camera while the sample pressure is increased (exemplar frames on left). The force sensor reading every 400 ms triggers acquisition of both a video frame and force value (middle plot, showing 3 coloured points corresponding to the 3 photos). The force for each frame is converted to a calibrated pressure in kPa. The average hue H is calculated within the region of interest in each frame, plotted against pressure (right plot, points) and interpolated (right plot, solid blue line). This interpolation (right plot, dotted lines) provides the pressure calibration between H and pressure. (Scale bar: 10 mm)

Supplementary Figure 9:
Colours recorded by the camera consist of three 8-bits values for the three primary channels R (red), G (green) and B (blue). R is set as (255, 0, 0), G is set as (0, 255, 0) and B is set as (0, 0, 255), and a colour with coordinates r on the red channel, g on the green channel and b on the blue channel is represented by (r, g, b) in RGB coordinates, given by r × R + g × G + b × B. On the RGB triangle plot, R, G and B vectors are set at a radius equal to 1 and at an angle of 120° from each other (the red, green and blue diamonds and lines), to make an equal repartition. This is equivalent, in cartesian coordinates, to: = 1 = 0 = cos(120°) = sin(120°) = − cos(120°) = − sin(120°) Every colour recorded in (r, g, b) coordinates is expressed in terms of its weight relative to each channel in cartesian coordinates: Each point representing a colour is therefore the centroid of the three primary colours R = (255, 0, 0), G = (0, 255, 0) and B = (0, 0, 255) normalized to 1 (for example, the purple diamond, with its weights r = 200, g = 50 and b = 70 represented as the black dots).
The closer a point is to one of the three primary points, the more weight this channel has compared to the other. In the alternate HSL representation of RGB colours, H accounts for the hue and represents the shift of information from one channel to the other. In this RGB triangle, this shift is visible as a rotation around the triangle, moving from the red axis to the green axis to the blue axis.