Using density-functional theory (DFT), the Gibbs free energy of RuO₂(110) surfaces in equilibrium with oxygen-rich environments is calculated to determine the lowest-energy structure. The traditionally assumed stoichiometric termination is only favorable at low oxygen chemical potentials (low pressure/high temperature). At realistic oxygen pressures, the surface is predicted to have excess oxygen atoms, forming a polar surface. However, the prevalent ionic model, which dismisses such terminations on electrostatic grounds, is not valid for RuO₂(110). This finding is consistent with previous results for corundum-structured oxides, indicating that non-stoichiometric terminations are a general phenomenon on transition metal oxide surfaces. The surface free energy is calculated using the Gibbs free energy, which depends on the oxygen chemical potential. The range of allowed oxygen chemical potentials is determined by the Gibbs free energy of formation of the oxide. The oxygen-poor limit is defined as the point where all oxygen would leave the sample, while the oxygen-rich limit is where gas-phase oxygen would start to condense. The calculated range of oxygen chemical potentials is between ½ΔG_f(0,0) and ½E_O₂_total. The surface free energy is calculated using DFT total energies, and the vibrational contribution is considered. The vibrational contribution is found to be small compared to the total energy, and the calculated surface free energy is dominated by the total energy. The surface free energy is found to depend on the oxygen chemical potential, with the stoichiometric termination having a constant surface free energy, while the polar termination becomes more favorable at higher oxygen chemical potentials. The results show that the polar termination of RuO₂(110) is stabilized at higher oxygen chemical potentials, indicating that the surface can have either a stoichiometric or a high-pressure termination depending on the experimental conditions. The rejection of polar surfaces on electrostatic grounds is not valid, as the strong dipole moment can be reduced by surface relaxation and electron rearrangement. The surface is viewed as a new material with structural and electronic degrees of freedom that allow significant modification of the properties of the atoms in the bulk. The results for RuO₂(110) are analogous to previous findings for corundum-structured oxides, supporting the argument that polar terminations are a more general phenomenon on transition metal oxide surfaces. The stability of polar terminations is important for understanding the function of oxide surfaces under realistic environmental conditions. The different properties of polar terminations should be considered when modeling high-pressure applications like catalysis.Using density-functional theory (DFT), the Gibbs free energy of RuO₂(110) surfaces in equilibrium with oxygen-rich environments is calculated to determine the lowest-energy structure. The traditionally assumed stoichiometric termination is only favorable at low oxygen chemical potentials (low pressure/high temperature). At realistic oxygen pressures, the surface is predicted to have excess oxygen atoms, forming a polar surface. However, the prevalent ionic model, which dismisses such terminations on electrostatic grounds, is not valid for RuO₂(110). This finding is consistent with previous results for corundum-structured oxides, indicating that non-stoichiometric terminations are a general phenomenon on transition metal oxide surfaces. The surface free energy is calculated using the Gibbs free energy, which depends on the oxygen chemical potential. The range of allowed oxygen chemical potentials is determined by the Gibbs free energy of formation of the oxide. The oxygen-poor limit is defined as the point where all oxygen would leave the sample, while the oxygen-rich limit is where gas-phase oxygen would start to condense. The calculated range of oxygen chemical potentials is between ½ΔG_f(0,0) and ½E_O₂_total. The surface free energy is calculated using DFT total energies, and the vibrational contribution is considered. The vibrational contribution is found to be small compared to the total energy, and the calculated surface free energy is dominated by the total energy. The surface free energy is found to depend on the oxygen chemical potential, with the stoichiometric termination having a constant surface free energy, while the polar termination becomes more favorable at higher oxygen chemical potentials. The results show that the polar termination of RuO₂(110) is stabilized at higher oxygen chemical potentials, indicating that the surface can have either a stoichiometric or a high-pressure termination depending on the experimental conditions. The rejection of polar surfaces on electrostatic grounds is not valid, as the strong dipole moment can be reduced by surface relaxation and electron rearrangement. The surface is viewed as a new material with structural and electronic degrees of freedom that allow significant modification of the properties of the atoms in the bulk. The results for RuO₂(110) are analogous to previous findings for corundum-structured oxides, supporting the argument that polar terminations are a more general phenomenon on transition metal oxide surfaces. The stability of polar terminations is important for understanding the function of oxide surfaces under realistic environmental conditions. The different properties of polar terminations should be considered when modeling high-pressure applications like catalysis.