Ellipsometry studies of evaporated aluminum films anodized in 0.6M H~PO4 at constant current indicate that the anodic oxide produced is at first homogeneous with a refractive index of 1.62, as found for barrier anodic oxide of thin film and bulk A1. As the thickness increases, the optical properties change signifying the onset of pore formation. This occurs at increasing thicknesses with increasing current density, however, pore formation always occurs well before the peak in voltage is reached at constant current. Ellipsometry indicates that the mode of development of the oxide prior to pore initiation is similar to that of barrier anodic oxides. However, the efficiency of oxide growth and the electric field required to pass a given ionic current, J, are generally lower and approach values for barrier oxides (of thin film and bulk A1) only at high current densities. The curvature of the log J vs. E curve is in the opposite sense to that normally observed for anodic oxides and it is interpreted in terms of cation and anion motion. After pore formation the efficiency of oxide growth increases and becomes near unity when steady-state formation conditions are reached. These results appear consistent with the model that dissolution at the pore bases is an electric field-aided process.Anodization of evaporated aluminum films is becoming increasingly important in the fabrication of monolithic, large-scale integrated circuits utilizing aluminum metal electrical interconnections where anodization provides both improved processing and reliability (1). In order to arrive at optimum processing, it is necessary to understand the anodization properties of aluminum films. In a previous paper, the barrier anodization properties were studied and found to be essentially the same as those of bulk aluminum (2). The present study is concerned with porous anodization. In particular, the initial stages of oxide growth and the onset of pore formation are quantitatively characterized using ellipsometry. This has not been done previously for porous anodization of aluminum and the results should be valuable in arriving at the mechanism of pore formation.Of the vast literature on porous anodization of aluminum, very little deals with initial stages of film formation. That which does is mostly qualitative because of the difficulties involved with measuring very thin films. The evidence that is presented indicates that a barrierlike film initially forms, followed by pore initiation and porous oxide growth. For example, the anodization characteristics, such as the time development of current at constant voltage (3) or of voltage at constant current, are at first similar to those observed in barrier anodizations. Also, direct observations of the surface with electron microscopy indicate that surface roughness occurs after some growth has taken place (3, 4). Thus, the oxide formed in the initial stages of porous oxide growth and that which remains interposed between the pore-containing layer and the metal is often referred to as a barrier layer...
The dependence of the ionic current density J on the mean field trueE¯ in oxides grown on tantalum in H3PO4 electrolyte was studied using ellipsometry to estimate the thickness. The field in the outer layer which is believed to grow due to metal ion motion is higher because of phosphate incorporation. The field in this layer was estimated by assuming that the field in the inner undoped layer was the same as the mean field at the same current density and temperature in films made in dilute sulfuric acid, which are less affected by electrolyte incorporation. The results indicate that the effect of phosphate on the ionic conductivity may be due to its reducing the permittivity and hence the effective field. It is pointed out that electrolyte incorporation gives two effects which contribute to the nonlinearity of log J vs. trueE¯ plots. Thus, since the metal ion transport number increases with J so does the proportion of the film which contains phosphate and requires a higher field. Also the concentration of phosphate increases with J . The incorporation leads to history effects which are not removed by subsequent formation, since the profile of phosphate records the sequence of current densities used. The time dependence of the field on changing the electrolyte is consistent with the field locally determined by the local film material.
Further ellipsometric data which confirm the two-layer nature of the anodic oxide films grown on tantalum in dilute phosphoric acid are reported. The two layers are believed to arise, respectively, from growth at the oxide/ solution interface due to metal ion transport and from growth at the metal/ oxide interface due to oxygen ion transport. The outer layer has a lower refractive index and dielectric constant due to phosphate incorporation. The effects have been observed in dilute phosphoric acid but not in the previous work in dilute sulfuric acid and citric acid of comparable concentration since phosphate is incorporated relatively strongly during the growth process. A model with a linear gradient of index is shown not to give an adequate representation of the optical results.One of the most unexpected experimental findings in the study of the growth of anodic oxide films on tantalum is that both metal and oxygen ions are mobile to comparable extents. The situation as shown by the most recent tracer studies of Pringle (1) is especially puzzling in that the results are, as he pointed out, quite inconsistent with conventional models of ionic transport. Thus it was found that layers of rare gas marker atoms embedded in thin oxide films were spread out during subsequent growth of the films. The amount of spreading was greater the lower the mass of the rare gas atoms. However, the position of center of the distribution with respect to the metal surface was independent of the mass of the atoms. The amounts of new oxide produced on either side of the marker layer were determined by Pringle and used to calculate the metal ion transport number for anodization in 0.2N sulfuric acid at various current densities. It appears that the film grows simultaneously at the metal/oxide and oxide/solution interfaces: thus, tracer studies by Randall, Bernard, and Wilkinson (2) showed that the outer part of the film contains a uniform and in some cases a quite large amount of species from the electrolyte. The effect was particularly marked with phosphoric acid, and it was shown that the incorporated phosphate decreased the permittivity and ionic conductivity of the oxide compared to films made in dilute sulfuric acid. The oxide produced at the metal/ oxide interface was found to be free from incorporated electrolyte species, as would be expected if the incorporated species are not mobile within the film. Ellipsometric measurements already reported (3) and further extended below give a clear confirmation of this picture of two homogeneous layers for films made in dilute phosphoric acid. In 0.2N sulfuric acid the effect of incorporated sulfate is less and optical methods (4-7) have either indicated a homogeneous film or have detected a deviation from a uniform oxide only in the form of an apparent very thin outer absorbing layer for films made in dilute sulfuric acid (5) although marked optical inhomogeneity through the thickness of the film was found for films made in more concentrated acid. Recent ellipsometric measurements by Muth (...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.