In this work, we present structured capillaries that were inspired by the microstructures of the external scent efferent system as found in different European true bug species (Pentatomidae and Cydnidae). These make use of small, orientated structures in order to facilitate fluid movement towards desired areas where defensive substances are evaporated. Gland channels and microstructures were investigated by means of scanning electron microscopy and abstracted into three-dimensional models. We used these models to create scent channel replicas from different technical substrates (steel and polymers) by means of laser ablation, laser structuring and casting. Video analysis of conducted fluid-flow experiments showed that bug-inspired, artificial scent fluid channels can indeed transport different fluids (water solutions and oils/lubricants) passively in one direction (velocities of up to 1 mm s −1 ), while halting the fluid movement in the opposite direction. At the end of this contribution, we present a physical theory that explains the observed fluid transport and sets the rules for performance optimization in future work.
We introduce a model to account for the dielectric response of doped organic semiconductor devices. In addition to the phenomena observed for undoped devices, mobile charge carriers created by doping can alter the dielectric function of the organic material and hence the dielectric response of the devices. These extrinsic charges may be trapped and contribute to the capacitance on re-emission. We directly model the real part of the dielectric function based on this phenomenon. The imaginary part is obtained via the application of the Kramers-Kronig transformation. We use oxygen-doped poly(3-hexylthiophene):[6,6]-phenyl-C61-butyric acid methyl ester\char21{} (P3HT:PCBM) based organic solar cells as a model system to test our approach and hence contribute to the understanding of oxygen-induced degradation in these devices. We fit our equations to the measured dielectric data and compare it to Debye relaxation as well as two widely used equivalent circuit models. Together with the device resistance determined from the steady-state current-voltage characteristic around 0V an excellent agreement between the experimental data and our model is achieved for both the real and the imaginary part of the dielectric function over a frequency range covering five orders of magnitude. Unlike the Debye relaxation model or the equivalent circuit approach, our model yields important device parameters such as the dopant concentration.