devices, [10] and the fundamental inspection by theory and experiment.While ferroelectrics constitute wideband gap semiconductors with good insulating properties, their DWs may possess a significantly increased electrical conductivity, as is of central focus in the work here. This so-called domain wall conductivity (DWC) has been reported first for multiferroic bismuth ferrite [11] and lead-zirconate-titanate [12] thin films, followed by research on (improper) erbium manganite [13] and barium titanate [14] bulk ferroelectric crystals. DWC in lithium niobate (LiNbO 3 :LNO) so far was reported to happen under photoexcitation only. [15,16] However, recent experiments by Godau et al. [17] on LNO single crystals (sc) have shown that reshaping the DW to larger inclination angles by applying a dedicated electrical tuning protocol enhances DWC by 3-4 orders of magnitude, and gives rise to DW currents in the upper µA range at room temperature and in the dark.Theoretical approaches applied to explain the increased DWC in these materials include phenomenological Landau [18] and Landau-Ginzburg-Devonshire theory, [19] and more recently also the combination of quantum mechanics with phenomenological Landau theory. [20] All these microscopic theories aim at predicting and quantifying the relevant local-scale domain wall parameters, i.e., the DW formation energy, the free charge distribution, and the DW conductivity. Nevertheless, it is difficult to compare the outcomes of such microscopic theories directly to experimental data, as the latter generally is based on macroscopic quantities. In contrast, our resistor network (RN) approach provides a clue link to experiments, in spite of being nonpredictive (i.e., as it requires some initial experimental input). Hence, our results can be directly compared to the local currents flowing within the DW as measured for instance by conductive atomic force microscopy (cAFM). The simple picture of a RN thus elegantly complements the microscopic theories of DWC.
MethodsFor the modeling a 2D network of linear, Ohmic resistors (see Figure 1a) was used, that was carefully crafted on the basis of the DW's inclination angle distribution. Resistor networks (RN) have already been applied impressively for modeling a Here the concept of a 2D resistor network (2D RN) is applied in order to model the electrical conductivity along sheet-like domain walls (DWs) in single crystalline lithium niobate (sc-LNO). The only input to the RN modeling approach is the DW inclination angle distribution, as measured previously with respect to the polar c-axis. The simulations then show that a 2D network of Ohmic resistors not only adequately accounts for the different boundary conditions envisaged in experiments, but equally well provides a direct link between the local domain wall conductivity (DWC) and the DW inclination angle α. Moreover, the RN simulations can be directly compared to local-scale transport measurements, as obtained by scanning probe techniques. The conceptual simplicity and the low computational e...