A two-dimensional melting transition has been observed in a freely expanding colloidal monolayer lattice formed of 1.01-jum polystyrene microspheres in an aqueous suspension between two parallel optical flats. The correlation functions computed from digitized images of the particle distribution within a fixed sampling area are consistent with a continuous two-step melting transition. However, direct observations of defect creation and evolution reveal that the system may melt via a first-order process.PACS numbers: 64.70.Dv, 61.25.Hq, 82.70.Dd Whether the melting transition in two dimensions (2D) is first order or continuous as proposed by Kosterlitz, Thouless, Halperin, Nelson, and Young (KTHNY) 1 " 3 has been a challenging problem for over a decade. 4 In KTHNY theory, between the solid (S) and the liquid (L) phases, there exists an intermediate hexatic (H) phase which has short-range translational order and quasi-long-range orientational order.Melting occurs in a two-state process governed by dislocation pairs unbinding at the S-H transition and by disclination pairs unbinding at the H-L transition. These two transitions are characterized by separate divergences in the translational and orientational correlation lengths, respectively. Colloidal monolayers are an especially good experimental system for studying this problem because the charged colloidal micorspheres (CCM) in aqueous suspensions are microscopically observable. In particular, a colloidal monolayer system in a small wedge geometry, first introduced by Pansu, Pieranski, and Strzelecki 5 and by Clark, Ackerson, and Hurd, 6 has been successfully employed by Murray and Van Winkle (M-VW) 7 in the study of 2D melting.In this Letter we describe the results of a study of free-expansion melting of 2D CCM solids formed in films of highly uniform thickness. Two-dimensional populations of 1.01-jum-diam polystyrene-sulphated microspheres suspended in water were established between fused-silica optical flats in a parallel-plate film cell. With this arrangement, the particle-number density gradient inherent to the wedge method could be eliminated. Most importantly the free-expansion melting is time dependent, which allows observation of the time evolution of defects as a given sample of the monolayer melts.The films are formed between a 1.6 mm x 3.2 cm diam flat which defines their lower boundary, and a 3.2 mmx 1.27 cm diam flat which is glued to a 3.2 mm x3.2 cm flat. These are fitted in an anodized aluminum cell which is sealed by tightening alignment screws in order to press the two larger flats against a Viton rubber O ring; the space -1.2 cm 3 between the O ring and the periphery of the smaller flat serves as a 3D reservoir. Interferometry reveals that the lower flat flexes slightly concave when the cell is sealed so that the gap thickness decreases -0.011 jum from the cell center out to a ra-dius of 500 /im. This distortion gives rise to an inward directed electrostatic wall-particle compressive force. A three-nested-ring stepped annular channel was plas...

No abstract

We have used an electrochemical sample chamber to modulate the surface charge of Ag(111) in spectral measurements (over a range of the incident photon energy from 1.4 to 2.23 eV) of the isotropic contribution to the second-harmonic reflectance. When the charge modulation was positive, and the harmonic energy was below the onset for interband transitions, the results compared favorably with a previously published prediction of the same phenomenon which was based upon time-dependent localdensity-functional theory for jellium having the same bulk electron density as Ag. The correlation between this theory and our experiment was poor for negative charge modulation. However, this was not unexpected, since the jellium model makes no allowance for the influence of the d bands. We have tentatively assigned a localized feature in the spectrum for negative charge modulation, which appears at a harmonic energy of 3.4 eV, as a two-photon resonance involving crystal-potential and image-potential surface states. INTRODUCTIONOptical second-harmonic generation (SHG) has emerged as a promising interfacial diagnostic technique. 'However, progress has been delayed by inadequate understanding of the mechanisms, particularly those governing SHG at metal surfaces. Theoretical efforts have concentrated on the isotropic, free-electron character of metals.In that case, the problem separates into two parts. The first involves local mixing of optical fields, in the electric dipole approximation, through a surface susceptibility, as well as higher-order multipole contributions driven by local fields in the bulk. The second, also dipole allowed, is intrinsically nonlocal. It is related to the field and electron-density variations in a direction along the surface norma1. The local-field part is adequately described by electron hydrodynamics applied to a jellium model, with an abrupt termination at the surface. By contrast, the nonlocal part (whose magnitude depends upon a parameter aas) has been the object of controversy 6 Rudnick and Stern, as well as others, ' originally concluded that~a Rs~= 1. This seemed to be confirmed by an early experiment which sampled the Ag-glass interface. However, more recent models, which involve time-dependent local-density-functional (LDF) concepts, have suggested that metal surfaces are more polarizable.They predict aRs = -36 -9i for Al with an incident photon energy of A'co = 1.17 eV. This was confirmed in experiments on Al(100) and Al(111). This test was conducted at one value of R~, a common characteristic of SHG studies. Most of the fundamenta1 data which relate to the suitability of the models have been obtained with one or only a few incident photon energies.' It is fully expected from the LDF prediction that there should be some significant cu dependence. ' Features should appear at 2co-= 0=0.8co and 0=4/th' (where co is the bulk-plasmon frequency and N is the work function).The first effect is associated with the character of the nonlocal fields below, yet near to, the bulk-plasmon frequency. The seco...

No abstract

We have studied the optical second-harmonic reflectance of Ag(lll) as a function of the incident photon energy and the interfacial potential in an electrochemical environment. The behavior of the anisotropic contribution is consistent with a potential-induced surface reconstruction from Ci v to Ci under conditions of positive surface charge.

No abstract

No abstract

No abstract