Abstract
The surfaces of rutile transition-metal oxides (TMO2) are widely investigated for catalysis, photoelectrochemical solar cells, memristors, and supercapacitors, but their structures have remained controversial. Here we employ density functional theory to predict that a universal behavior of metallic TMO2 surfaces, i.e., the stoichiometric TMO2 surfaces, exhibit a contrast reversal in simulated scanning tunneling microscopy (STM) images at different scanning biases. The predictions are verified by experimental STM imaging of RuO2(110) surfaces and this feature is shown to enable accurate determinations of the TMO2(110) surface structures under various conditions. This work provides different insights into the electronic properties of TMO2(110) surfaces and offers an effective method to directly map the surface structure and point defects using bias-dependent STM.
