Angle-resolved photoemission is used to determine the energy bands of permalloy (Ni 0.8 Fe 0.2) and compare them to Ni, Co, and Cu. The energy and momentum resolution ͑Ϸ0.01 eV and Ϸ0.01 Å Ϫ1 ͒ is high enough to resolve the magnetically split bands at the Fermi level that are responsible for spin-dependent conductivity and tunneling. For the ⌺ 1 band we find the magnetic exchange splittings ␦E ex ϭ0.27 eV ͑0.23 eV for Ni͒, ␦k ex ϭ0.16Ϯ0.02 Å Ϫ1 (0.12Ϯ0.01 Å Ϫ1 for Ni͒, the Fermi velocity v F↑ ϭ(0.22Ϯ0.02)10 6 m/s (0.28ϫ10 6 m/s for Ni, 0.33ϫ10 6 m/s for fcc Co͒, and the widths ␦k ↑ р0.11 Å Ϫ1 and ␦k ↓ ϭ0.24 Å Ϫ1. Compared to Ni, permalloy features a 27% larger magnetic splitting of the Fermi surface and an extremely short mean free path of 4-8 Å for minority spins.

We have studied the electronic structure of the Ag͑111͒͑ ͱ 3 ϫ ͱ 3͒R30°-Sb surface alloy by angle-resolved photoemission. We find two hybrid surface bands, similar to the isostructural Ag͑111͒-Bi interface. The spinorbit coupling induced spin splitting in momentum space, however, is strongly reduced from the Bi case. First-principles and model band calculations correctly reproduce this difference. The present results illustrate the complex interplay of atomic and structural contributions at the origin of the large spin separation in these systems.

High-resolution angle-resolved photoemission (ARPES) data show that a metal-insulator Mott transition occurs at the surface of the quasi-two dimensional compound 1T -TaSe2. The transition is driven by the narrowing of the Ta 5d band induced by a temperature-dependent modulation of the atomic positions. A dynamical mean-field theory calculation of the spectral function of the halffilled Hubbard model captures the main qualitative feature of the data, namely the rapid transfer of spectral weight from the observed quasiparticle peak at the Fermi surface to the Hubbard bands, as the correlation gap opens up.PACS numbers: 71.30.+h,79.60.Bm,71.45.Lr,71.10.Fd Electronic correlations can modify the electronic structure of solids not only quantitatively, but also qualitatively, inducing new broken-symmetry phases which exhibit charge, spin or orbital-order, and more exotic states in low dimensions. One of the most notable consequences of electronic correlations is the much studied metal-insulator (M-I) Mott transition [1,2]. Recently, new theoretical approaches have considerably extended our understanding of this fundamental problem [2,3].Many physical properties indirectly reflect the dramatic rearrangement of the electronic structure at the transition. Photoelectron spectroscopy, which probes the single-particle spectral function, can provide a direct view of such changes [4,5,6]. However, comparing samples with different compositions faces materials problems like stoichiometry, defects, and disorder. A quantitative analysis is further complicated by the known surface sensitivity of the technique [7][8]. An ideal experiment would record the energy and momentum-dependent spectrum, while tuning the crucial (W/U ) parameter (U is the onsite Coulomb correlation energy; W is the bandwidth) in the same single crystal sample. Remarkably, it is possible to approach this ideal situation exploiting the occurrence of modulated structures (charge-density-waves; CDWs) in appropriate low-dimensional systems. There, the lattice distortion modulates the transfer integrals and therefore modifies the bandwidth. In materials that are close enough to a Mott transition, the reduced bandwidth may lead to an instability. There are strong indications for this scenario in the layered chalcogenide 1T -TaS 2 , which presents a sharp order-of-magnitude increase of the resistivity at T=180 K [9, 10], with a strong rearrangement of the electronic states [11,12,13,14,15]. However, the complex phase diagram of the CDW in 1T -TaS 2 affects the electronic transition, which cannot be considered as a typical Mott transition.Isostructural and isoelectronic 1T -TaSe 2 exhibits a similar CDW, but only one phase below T C =475 K. Its electrical resistivity remains metallic -albeit rather large -to very low temperatures [9], suggesting that the Se compound lies further from the instability than the S analog. Nevertheless, a transition could still occur at the crystal surface, where the U/W ratio is expected to be larger as a result of smaller screening and...

We very much regret that in the above article we failed to cite angle-resolved photoemission spectroscopy of -Mo 4 O 11 by another group. 1-3 In this work, as in ours, nesting and charge density wave ͑CDW͒ gap openings on different Fermi surfaces ͑FSs͒ were studied and conclusions very similar to ours were drawn about the CDW mechanism. Some differences between our data and theirs can be noted. ͑1͒ Better energy resolution in their data enabled detection of a small b band gap opening. ͑2͒ Sharper Fermi surface patterns in our data enabled the observation of small zigzags due to avoided FS crossings of the b band, and hence a more detailed explanation of the nesting vector q c2 . ͑3͒ For unknown reasons, the "P" feature in our data ͓Fig. 2͑a͔͒, i.e., the intensity pile-up at low temperature, is not present in their data.

The electronic properties and growth mode of Sn/InSb(111) interfaces have been investigated by angular resolved photoemission spectroscopy (ARUPS), high energy electron diffraction (RHEED), and work function measurements. For Sn coverages of ∼1 ML the Fermi level is pinned at the bottom of the conduction band. The interfacial growth is heteroepitaxial and extends to a thickness of ∼5 ML of Sn. This is followed by the formation of diamond structured α-Sn up to ∼40 ML. Beyond this thickness three dimensional island growth starts and the films are a mixture of semiconducting α-Sn and metallic β-Sn. Above ∼120 ML Sn, the surface region consists of pure β-Sn while due to a strong substrate stabilization effect the interface region still consists of the α-Sn phase.