The design, construction, operation, and performance of a spin polarized electron source utilizing photoemission from negative electron affinity (NEA) GaAs are presented in detail. A polarization of 43±2% is produced using NEA GaAs (100). The polarization can be easily modulated without affecting other characteristics of the electron beam. The electron beam intensity depends on the intensity of the exciting radiation at 1.6 eV; beam currents of 20 μA/mW are obtained. The source is electron optically bright; the emittance phase space (energy-area-solid angle product) is 0.043 eV mm2 sr. The light optics, electron optics, and cathode preparation including the GaAs cleaning and activation to NEA are discussed in depth. The origin of the spin polarization in the photoexcitation process is reviewed and new equations describing the depolarization of photoelectrons in the emission process are derived. Quantum yield and polarization measurements for both NEA and positive electron affinity surfaces are reported. The important considerations for interfacing he polarized electron source to an experiment are illustrated by its application to polarized low energy electron diffraction (PLEED). The advantages of this spin polarization modulated electron gun for PLEED are clearly demonstrated by sample PLEED results for W(100) and ferromagnetic Ni(110). A comparison with other polarized electron sources shows that the GaAs spin polarized electron source offers many advantages for a wide range of applications.

We have used the sharp tip of a surface force microscope to make modifications with submicrometer dimensions on polymer surfaces. In this letter we show three examples: scribed grooves with widths less than 120 nm, raised areas with heights up to 1 nm above the original surface, and pits with depths of 6 nm. We also discuss possible sources of contrast in surface force microscope images that are not due to height variations in the surface topography. Because the surface force microscope can be used for both conducting and nonconducting materials, it has an advantage over the higher resolution scanning tunneling microscope.

We have used electron diffraction to study submonolayers of Se adsorbed on Ni(100); p (2x 2), c(2x2), and disordered phases were observed and the boundaries between them located. Symmetry arguments indicate that the phase diagram belongs to the universality class of the Ashkin-Teller model and allow us to predict the critical behavior near the phase boundaries and at the multicritical point where they meet. Our lattice-gas-model calculations support these results. The predicted critical behavior, and thus the universality arguments, should be tested by additional experiments, in particular synchrotron-x-ray diffraction.

No abstract

We report the first measurements of the critical exponents of a phase transition at the surface of a clean metal, the Au(llO) (1x2)*-^ (1x1) transition. The exponents P,y, and v and the amplitudes were measured by LEED and place the phase transition in the two-dimensional Ising universality class. The critical scattering is observed to peak above Tc^ signaling the breakdown of the Ornstein-Zernike theory, and is in numerical accord with predictions of Fisher and Burford.PACS numbers: 64.60.Fr, 61.14.Hg, 68.20.+ t The prediction^'^ that continuous order-disorder phase transitions can be classified into a small number of universality classes, each one determined by quantities which are not system specific, such as the interaction range, the symmetry of the ordered phase, and the dimensionality of space, implies that some universality classes exist only in two dimensions. While this has prompted many studies of physisorbed and chemisorbed overlayers,^ many metals and semiconductors also show phase transitions at their surfaces which do not occur in the bulk. The purpose of this Letter is to report the measurement of the critical exponents of a phase transition which occurs on the surface of a clean metal, and fits the two-dimensional (2D) Ising universality class.Specifically, we present a study of the phase transition which occurs on the surface of Au(llO). At room temperature on a carefully prepared surface, there is double periodicity in the (11) direction, and hence a (1x2) LEED pattern which has been ascribed to a missing row structure."^ For tempertures greater than -^ 700 K, the y-order spots disappear and an apparently primitive (1x1) structure is observed. It has been suggested"*'^ that the high-temperature phase is disordered in the ''lattice gas" sense, that is, the toplayer atoms are randomly placed on the surface rather than aligned in rows, but remain in the same type of site as in the ordered structure. From symmetry considerations, Bak^ has predicted that the transition might belong to the 2D Ising class. Our highresolution LEED measurements of the critical exponents confirm that this is indeed the case.It has been demonstrated^ that LEED is an ideal technique for studying surface phase transitions. It has high surface sensitivity, and the short interaction times allow every configuration of the system to contribute independently to the diffracted beams. The major drawback of LEED is the limited resolution of conventional diffractometers. We have overcome this limitation with a new kind of high-resolution LEED diffractometer^ which has a sufficiently narrow response in q space compared with the experimental peak width that we are able to extract directly the structure function^ 5(q,n = /(r)8(q-qo)+X(q-qo,T)/xO(T), (1) where /(D8(q -qo) is the long-range order, X(q -qo, T) are the fluctuations in order (short-range order), X^{T) are the fluctuations of a random noninteracting system, and q is the momentum transfer of the electrons. For the Au(llO) (1x2) structure, QiQ=^ {HITT/a,k7r/a).In LEED, multipl...