“…The measurement was performed by keeping the (1 × 1) surface at 1500 K, the operating temperature of the Sc-O/W(100) emitter, in the oxygen atmosphere. Since the speed of the transition from the (1 × 1) to (2 × 1)-(1 × 2) surfaces strongly depends on the oxygen partial pressure, [12] i.e. the higher the partial pressure is, the faster the transition is, the oxygen partial pressure was set at ∼5 × 10 −7 Pa, which was low enough to measure the Sc-LMM and O-KLL spectra during the transition.…”