Fig. 4: Obtaining the contact angle switching times by the wet exposure method.

a Profile image of the droplet, sample, and LED collimation lens (above) from the contact angle goniometer with self-explanatory graphics added to show the key elements in the wet exposure experiment. b Example of apparent contact angles measured during wet exposure by UV light with a wavelength of 365 nm (for illumination with 300 and 275 nm UV light, see Supplementary Fig. 5). The apparent contact angles switch to lower values in steps. The contact angle values right after the switches are marked with black dots and used for fitting to Eq. 4. Three traces for each MOx layer thickness are used to generate a fit of one switching time \(\tau\). In (I) the fitting is shown for a ZnO/Ti heterojunction with a ZnO layer thickness of 75 nm, whereas in (II) the fit is shown for a TiO₂/Si heterojunction with a TiO₂ layer of 35.5 nm. The onset of photoinduced switching happens with a delay time \(\triangle t\), where a smooth decay is observed, which is associated with photocatalytic oxidation of adsorbed hydrocarbon groups on the surface. \(\triangle t\) is fitted alongside the switching time constant and the limiting contact angles \({\theta }_{0}\) and \({\theta }_{1}\). c Validation of linearity of switching rate \(1/\tau\) with light intensity \({I}_{0}\). Shown data are for the same samples as in (b) and also illuminated at 365 nm. Error bars in switching rate represent one STD to the fit of switching time constant as shown in (b) with a minimum of three samples, while error bars in light intensity represent accumulated errors from intensity variation during the illumination and uncertainty in the determination of illumination area (see Methods).