We have addressed the possibility of surface ferromagnetism in Sr2RuO4 by investigating its surface electronic states by angle-resolved photoemission spectroscopy (ARPES). By cleaving samples under different conditions and using various photon energies, we have isolated the surface from the bulk states. A comparison with band structure calculations indicates that the ARPES data are most readily explained by a nonmagnetic √ 2 × √ 2 surface reconstruction.Following the discovery of superconductivity (SC) at 1 K in the layered perovskite Sr 2 RuO 4 , 1 the exact nature of its SC pairing mechanism has attracted a great deal of interest. While it shares the same structure as the archetypal cuprate parent compound La 2 CuO 4 , RuO 2 planes replace the CuO 2 planes thus resulting in an anisotropic Fermi liquid 2 instead of a strongly correlated charge transfer insulator. Furthermore, there is evidence that Sr 2 RuO 4 exhibits spin-triplet pairing with a p-wave order parameter, 3 as opposed to the spin-singlet, d-wave symmetry found in the cuprates. Although it is now widely believed that the unconventional nature of SC in this compound is mediated by spin fluctuations, the exact nature of this interaction is still unresolved. Originally, it was suggested that ferromagnetic (FM) spin fluctuations were responsible for mediating the SC as inferred from theoretical calculations, 4 NMR measurements, 5 and ferromagnetism in closely related SrRuO 3 . However, more recent evidence has suggested that this simple picture may be incomplete. Antiferromagnetism (AFM) in Ca 2 RuO 4 , the observation of incommensurate peaks at Q = (0.6π, 0.6π, 0) by neutron scattering, 6 and calculations which show strong nesting at Q = (2π/3, 2π/3, 0) 7 all seem to imply AFM correlations should not be neglected, leaving the nature of SC open to speculation.Recently, an analysis of low-energy electron diffraction data from Sr 2 RuO 4 indicated that a √ 2 × √ 2 reconstruction was induced by the freezing of a soft zone boundary phonon into a static lattice distortion at the surface, and comparisons with band structure calculations predicted that the resulting surface was FM. 8 This conjecture also appears conceivable in light of speculation based on recent ARPES results 9 as well as earlier theoretical calculations. 10 If FM does exist on the surface of Sr 2 RuO 4 , such a result should be indicative of strong ferromagnetic tendencies in the bulk and thus possibly relevant to microscopic theories which describe the mechanism of SC. This speculation becomes even more intriguing in light of recent STM measurements 11 which suggest the opening of a superconducting gap with T c = 1.4 K, perhaps hinting that the surface layer may be superconducting, and raises the possibility that the surface of Sr 2 RuO 4 may exhibit the rare coexistence of SC and FM. However, as this proposed surface FM has never been confirmed, it becomes imperative to reinvestigate the surface electronic structure to definitively verify or exclude surface FM.In this paper, we present a ...