While investigating the instrumental function of a Fabry-Perot interferometer [Appl. Opt. 34, 58 (1995), we noticed some variation in finesse and contrast in the measured spectra when a 1.5-mm-diameter aperture was used at various spots within the standard 8-mm aperture. By comparing experimentally determined finesse versus contrast plots for many such spectra with calculated plots, we found spots on the plates that gave non-Airy-function line shapes over the entire order of interference, unlike the Airy line shape we determined previously by using the entire 8-mm aperture. We have reviewed several models that describe the effects of various types of surface defects, such as Gaussian-height distribution of roughness, curvature and tilt of plates, sinusoidal roughness, and asymmetrical roughness on the finesse and contrast. Our experimental results can be accounted for if we assume that the reflectivity is nonuniform over the Fabry-Perot plates and that there is some reasonable contribution that is due to Gaussian roughness, curvature, or tilt.

We have studied the effect of a multipass Fabry-Perot interferometer (FP) on a scattering line. Here we describe a method that we applied to derive a closed-form expression for a line shape obtained with an ideal, multipass FP. The method reduces the convolution problem between the multipass function and the scattering line to the corresponding single-pass problem. We illustrate the method with a Lorentzian and a damped-harmonic-oscillator line passed through a single-, triple-, and quintuple-pass FP. Furthermore we have applied the method to a study of the effect of the collecting pinhole on a sharp line obtained by multipassing. We show how we used these functions to fit the complete spectra obtained with a single- and triple-pass FP.

We have studied the effect of a multi-pass Fabry-Perot interferometer (FP) on a natural scattering line.

To test the model developed in the preceding paper [Appl. Opt. 34, this issue (1994)] regarding the line shape produced by a Fabry-Perot interferometer system in a multipass mode, we have used three Lorentzian line shapes formed in scattering processes and subjected them to single and triple passes with some variation in the pinhole diameter. In most cases we find good agreement with the calculations with the only adjustable parameter being the single-pass contrast C(1). Where differences occur, plausible explanations are offered.

The quality of vertical roughness produced by the etching of Si in aqueous KOH has been studied by varying several experimental parameters such as molarity, time of etching, temperature, and stirring. We note that at room temperature, unstirred etching is smoother at low and high molarities, and etch rate and roughness both peak near 5–6 M. With no stirring, roughness increases as a function of etch time, then levels off. With stirring, roughness decreases, especially around the peak etch rate near 5–6 M. For a fixed molarity like 5 M, unstirred etching becomes smoother with increasing temperature even as the etch rate increases rapidly. Such results suggest that masking by hydrogen bubbles or silicate etch products is the principle origin of vertical roughness. Bubble properties as a function of molarity and stirring (as determined from electrolysis experiments) are used to suggest a pseudo-masking model to explain some roughness properties.

We present a model to explain the peak observed in the Si etch rate vs. molarity for several alkali-hydroxide solutions. This model requires both free water molecules (H2O)f and hydroxyl ions (OH-) to accomplish etching. As the molarity increases, the OH- concentration increases, while hydration effects steadily reduce the (H2O)f concentration. These two competing effects produce a peak in the etch rate, the location of which is sensitive to the mean hydration number of the particular solution. Results for KOH, NaOH, and LiOH are generally accounted for by the model using mean hydration numbers obtained from independent chemical activity data in the literature. Data for aqueous KOH solutions only (unstirred and stirred) show that stirring shifts the peak etch rate to slightly lower molarity and raises the peak etch rate somewhat, but the behavior is by and large unchanged. Also, at low molarity from 0.1 to 2M the stirred etch rate is measured to be roughly constant, then rises to a peak near 5M. We have attempted to use the present model to account for these more detailed effects and to obtain additional information about the possible etching mechanism applicable at low molarity, where free water alone appears to be the dominant (or rate-limiting) etching species. The very large dependence of the etch rate on crystal orientation at high molarity may be due to geometrical effects in the way the hydration complexes interact with the different crystal surfaces.

We have fabricated Si (100) membranes about 2 μm thick by several different techniques using open‐circuit and biasing potentials with high molarity aqueous solutions. These membranes are characterized by visible‐infrared transmittance and reflectance, scanning with focused Ar+ ion laser beam, and by Raman scattering in order to access thickness uniformity, roughness and strain properties. These techniques may lead to better control of membrane structure for masked ion beam and x‐ray lithography applications.

We use the modulation techniques of electrolyte electroreflectance and photoreflectance to locally probe the semiconductor surface potential of various semiconductor/electrolyte interfaces. Changes in the surface potential are important in determining the rate of charge transfer in both electrode and electrodeless photochemical etching.

The bias‐dependent etching of Si in has been studied in detail. The quantity of Si etched and evolved has been measured at various anodic and cathodic potentials. These quantities have been related to the current flow in the electrochemical cell in such a way as to shed light on the mechanisms responsible for bias‐dependent etching of Si.

We use optical emission to study chemically assisted ion-beam etching. Ar-ion and Cl 2 -gas beams are used separately and together to etch Al and Si. The atomic and electronic optical emission is measured with a multichannel analyzer spectroscopic system. The relationship between optical emission and etch rate is examined. We observe surface roughening caused by Ar sputtering of Al which is minimized by the addition of small amounts of O 2 to the etching chamber. This is ascribed to the smoothing out of ion-channeling effects by the amorphous oxide which grows on the surface and is itself sputtered off. 1 O 2 pressure determines the oxidation rate, while Ar pressure determines the sputter rate. As these two processes compete, the intensity of neutral Al (398.3 nm) varies, depending on whether the Ar ions are sputtering Al from Ai or Al 2 O 3 . Similar results are obtained for neutral Si (288.1 nm) for Si and SiO 2 . The addition of the Cl 2 beam begins to alter the emission characteristics, since in addition to sputtering we expect chemical etching to form volatile Al and Si chlorides.

The etch‐stop phenomenon which occurs in both p‐ and n‐Si when the doping density exceeds 1019 cm−3 has been studied in situ with ellipsometry. Biasing the samples both anodically and cathodically from open‐circuit potential causes layers to grow which can then be observed to etch back when the applied potential is released. We are lead to the conclusion that p‐Si etch stops because of spontaneous passivation which produces a thin oxide‐like layer, while n‐Si shows a tendency to etch stop, probably owing to formation of a prepassive layer.

AbstractEllipsometrische Messungen werden gleichzeitig mit Zellstrom‐ und Spannungsmessungen in 2 M KOH an Si‐Proben unterschiedlicher kristallographischer Orientierung und Dotierungsart während des Siüx‐Oberflächen?lmwachstums unter anodischer und kathodischer Vorspannung durchgeführt.

Basic properties of plasma-deposited amorphous and microcrystalline layers are summarized. Limitations for solar cell performance, which are a result of these basic properties are pointed out. Efficiencies and other solar cell parameters obtained for the best laboratory cells are presented and the resulting implications for future research discussed. Large-area Industrial modules and their typical key parameters are then given. Conclusions for the future market position of thin-film silicon PV as very low-cost renewable source of electricity are drawn up.

Ellipsometric measurements (in conjunction with observations of reflectance and cell current and voltage) have been performed in for Si wafers, of various crystallographic orientations and dopant types, during growth of surface films under anodic and cathodic bias and during open‐circuit etch back. The optical effects have been modeled in terms of the formation and removal of thin films on the Si surface, with contributions, in some cases, from surface roughness. Changes in the thickness and stoichiometry of the surface phase during various treatments provide a basis for a model of the Si etching chemistry and for the orientation dependence of the etch rate.

The electric-field modulation of the internal-reflection response of a Si-electrolyte interface is shown to give information both about accumulation of free carriers, interface states, and about molecular species in the oxide or in the Helmholtz or Gouy layer. Measurements were made in either a 0.1-M KOH or 0.1-M H2SO4 solution and evidence of electromodulation of vibration bands of H3O+, OH, and H2O was obtained.