Search

In linear filtering, high-frequency (white) noise is reduced by apodization, which is the attenuation or elimination of high-order Fourier coefficients followed by an inverse transformation. Unfortunately, apodization requires compromises to be made among noise leakage, information loss, and Gibbs oscillations. These shortcomings are avoided with the corrected maximum-entropy (CME) procedure, but this procedure applies only to Lorentzian or approximately Lorentzian features. We develop a generalized maximum-entropy method based on partial Hilbert transforms that allows CME to be applied to any spectrum, thereby eliminating white-noise completely with no deleterious side effects. As Hilbert transforms are exact Kramers–Kronig replicas of the original endpoint-discontinuity-corrected segment, new spectral processing opportunities are also realized. [ABSTRACT FROM AUTHOR]

In spectroscopy, the objective is to obtain information by analyzing spectra that ideally are undistorted and noise-free. In standard Fourier-space filtering, this goal cannot be achieved because of apodization, which forces a trade-off among errors arising from distortion, noise leakage, and Gibbs oscillations. We show that low-order coefficients can be preserved and apodization, and its associated errors eliminated with the corrected maximum-entropy (M-E) filter obtained here. Although the Burg derivation begins as M-E, by making certain assumptions the Burg approach yields a procedure that deconvolves (sharpens) structure in spectra, thereby violating the basic M-E principle of leaving the low-order coefficients intact. The corrected solution preserves these data and projects the trends established by them into the white-noise region in a model-independent way, thereby eliminating apodization and its associated errors. For a single Lorentzian line, the corrected M-E approach has an exact analytic solution, which reveals not only how M-E performs its extension but also why it works particularly well for line shapes resulting from first-order decay processes. The corrected M-E filter is quantitatively superior to any previous filtering method, including recently proposed high-performance linear filters, yet requires only minimal computational effort. Examples, including multiple differentiation, are provided. [ABSTRACT FROM AUTHOR]

Many optical models have been used to construct analytic composition-dependent dielectric functions of AlxGa1-xAs alloys. However, these models incorporate various unphysical assumptions to improve their fits to data. Here, we provide the parameters needed to calculate dielectric functions of AlxGa1-xAs for 1.5≤E≤6.0 eV and 0≤x≤1 by means of the parametric model of Johs et al. [Thin Solid Films 313–314, 137 (1998)], which eliminates these problems. A representative example concerning interface analysis is discussed, where it is necessary to construct a dielectric function of an alloy of essentially arbitrary composition. [ABSTRACT FROM AUTHOR]

The evolution of polariton features with increasing thickness in p-polarized (TM) reflectance spectra of indium tin oxide (ITO) thin films deposited on BK7 glass reveals the nature of plasmons in conducting thin films without interference from band-to-band transitions or the tendency of very thin films to form islands, both of which are complicating factors with the noble metals Au and Ag. Although the dependence on energy, film thickness, and angle of incidence is complex, these features are accurately described by the three-phase (substrate/overlayer/ambient) Fresnel model using only the Drude free-electron representation for the dielectric function of the ITO film. For film thicknesses less than 80 nm the relevant excitation is a one-dimensional screened-bulk plasmon (SBP) that corresponds to charge transfer across the entire film. The associated SBP polariton (SBPP) occurs at the energy of the SBP and is relatively independent of the angle of incidence. For film thicknesses greater than 120 nm, the relevant excitation is the surface plasmons (SP). The associated surface plasmon polariton (SPP) exhibits the usual strong dependence of energy on the angle of incidence. For larger thicknesses this structure gradually weakens, in agreement with theory. No other collective excitations are observed. The optimum thicknesses for the SPP in ITO is 160 nm, whereas the SBPP is observed only when the film thickness is less than 70 nm. The SBPP exhibits many of the features that make the SPP attractive for both science and technology, but has not been observed previously. Our results show that ITO films, in particular, and conducting-metal-oxide films in general provide new opportunities for investigating plasmons in conductors and obtaining new insights into plasmons, plasmon polaritons, and related optical phenomena. [ABSTRACT FROM AUTHOR]

We investigate the effect of incomplete removal of semiconductor overlayers on critical-point (CP) parameters determined from the analysis of ellipsometric spectra. An approximate analytic expression shows that CP energies and broadening parameters should be relatively unaffected for isolated CPs if the dielectric response of the overlayer varies slowly with energy. The results are confirmed by model calculations for InAs, which show that the energies of the E1 and E1+Δ1 CP structures that are commonly used for compositional analysis of semiconductor alloys are relatively unaffected. We also analyze overlayer-removal data for a series of InxGa1-xAs alloy samples. Consistent with the above, the amplitudes and phases are affected significantly for all CPs, while the energies of the well-separated E1 and E1+Δ1 transitions are relatively invariant. The results show that accurate values of composition can be obtained from the analysis of the E1 and E1+Δ1 CP structures, even if complete removal of overlayers is not achieved. [ABSTRACT FROM AUTHOR]

We critically test the capabilities of the effective-medium approximation (EMA) and alloy models to describe multilayer samples with gradual interfaces by analyzing spectroscopic ellipsometric (SE) data of two AlGaAs samples grown expressly for this purpose. The dielectric functions [variant_greek_epsilon] of the interfaces are calculated in the EMA and alloy models, and the interfaces themselves simulated either as a single layer of Al0.5Ga0.5As or a stack of layers of AlxGa1-xAs with x increasing or decreasing between 0.1 and 0.9 in increments of 0.1. The EMA essentially fails completely for either interface representation. For the alloy model the stepwise-graded representation is significantly better, not only simulating the data more accurately but also yielding thicknesses in essential agreement with those obtained by cross-sectional transmission electron microscopy. The results highlight the types of errors that are encountered with the different models, and show that the analysis of SE data can provide information about these interfaces. [ABSTRACT FROM AUTHOR]

We obtain a solution of the three-phase model in the limit d/λ<1, where the complex refractive index ñ=n+iκ and thickness d of an isotropic film on an isotropic substrate are given analytically at any single wavelength λ from polarimetric data Δρ/ρ and ΔR/R, where ρ is the complex reflectance ratio and R is either the p- or s-polarized reflectance. We describe several procedures for extending the range of validity of the solution. Analysis of correlations shows that the uncertainty δ(ΔR/R) of ΔR/R is significantly more important than the δ(Δρ/ρ) of Δρ/ρ, which allows us to obtain an expression for the uncertainties δn, δκ, and δd of n, κ, and d, and to identify conditions that optimize the determination of the layer parameters. We find that the relative uncertainties δn/n and δd/d are not equal, as would be expected if they were determined by the optical thickness nd measured by ellipsometry, but that ΔR/R breaks the connection. We verify our results by measurements of H2O reversibly physisorbed on oxidized GaAs, finding, for example, that for our conditions δκ is determined more accurately than δn, and δn more accurately than δd. These data and model calculations show that fluctuations in parameters, particularly d, are asymmetric, leading in principle to inaccurate average values. However, we show that the importance of the ΔR/R data together with the remaining high correlation between n and d allows us to define a characteristic curve that can be used to correct the results for this nonlinearity. Finally, we extend our analysis to determine the orthogonal linear combinations of n, κ, and d that the data actually determine, which explains why the data fit the characteristic curve so well. Our results will be useful in various contexts for the analysis of films less than 1 nm thick, for example, in applications involving preparation of next-generation electronic and optoelectronic devices with complicated multilayer structures, real-time control of deposition, and the identification of physisorbed and chemisorbed layers on the monolayer scale. [ABSTRACT FROM AUTHOR]

Room-temperature pseudodielectric function spectra 〈ε(ω)〉=〈ε1(ω)〉+i〈ε2(ω)〉 of AlxGa1-xSb (0.00<=x<=0.39) alloys were measured from 1.5 to 6.0 eV by spectroscopic ellipsometry. Effects of surface overlayers on the measured spectra have been minimized as far as possible by performing in situ chemical etching on samples immediately before the measurements rather than correcting the measured data mathematically for the presence of overlayers. Analysis of second energy derivatives numerically calculated from the measured data yielded the critical-point energies of the E1, E1+Δ1, E0′, E2, and E1′ interband transitions. Dependence of the critical-point energies on composition x is obtained. [ABSTRACT FROM AUTHOR]

Reciprocal-space analysis offers several advantages for determining critical point parameters in optical and other spectra, for example the separation of baseline effects, information, and noise in low-, medium-, and high-index Fourier coefficients, respectively. However, endpoint-discontinuity artifacts can obscure much of the information when segments are isolated for analysis. We developed a procedure for eliminating these artifacts and recovering buried information by minimizing in the white-noise region the mean-square deviation between the Fourier coefficients of the data and those of low-order polynomials, then subtracting the resulting coefficients from the data over the entire range. We find that spectral analysis is optimized if no false data are used, i.e., when the number of points transformed equals the number of actual data points in the segment. Using fractional differentiation we develop a simple derivation of the variation of the reciprocal-space coefficients with index n for Lorentzian and Gaussian line shapes in direct space. More generally, we show that the definition of critical point energies in terms of phase coherence of the Fourier coefficients allows these energies to be determined for a broad class of line shapes even if the direct-space line shapes themselves are not known. Limitations for undersampled or highly broadened spectra are discussed, along with extensions to two- or higher-dimensional arrays of data. © 2001 American Institute of Physics. [ABSTRACT FROM AUTHOR]

Focuses on a study which determined the microstructures that optimize the near-IR-visible normal-incidence optical transmittance of electrically conducting metal films. Photoelectrochemical reduction of protons in perchloric acid solutions; Nature of the microstructural inhomogeneities in metal film contacts; General mechanisms that give transparency to metal films.

Deals with a study which examined the room-temperature pseudodielectric and related optical function data binary alloy series. Information on binary alloy series; Methodology of the study; Results and discussion.

Deals with a study which determined the results of single and multiple angle of incidence spectroscopic ellipsometric measurements in the visible to near ultraviolet on polished molybdenum mirrors which have been implanted with molybdenum ions. Methodology of the study; Results and discussion.

Focuses on a study which examined the strain and disorder of In[subx]Ga[sub1-x]/(GaAs, graded gallium phosphide), using spectroscopic ellipsometry and Raman spectroscopy. Details of the samples and characterization; Review of strain dependence of E [sub1] gap and of optical phonons; Results.

Provides information on a study that measured the critical points of (Al[subx]Ga[sub1-x])[sub0.5]In[sub0.5]P semiconductor alloys grown by metal-organic chemical vapor deposition and lattice matched to GaAs using spectroscopic ellipsometry. Experimental procedures; Results and discussion on the study; Conclusions.