Shaud G. Tavakoli

Columbia University

Project Title: Fabrication of Porous Anodic Alumina Templates with Sub-20nm Pores
Advisor:Professor Tim Sands

Introduction

An oxide film can be grown on certain metals, including aluminum, niobium, tantalum, titanium, tungsten, and zirconium, via anodization. Aluminum and titanium are unique, however, in that a thick oxide coating with a high density of pores is possible. With other metals, only a barrier oxide layer is formed. Here we will focus on porous anodic alumina, or PAA.

PAA is electrically insulating, optically transparent over a wide energy band, and exhibits chemical and thermal stability. In the past, PAA has been fabricated with pore diameters of between 4 and 250nm, pore densities ranging from 108 to 1012 pores/cm2, and thickness of up to 300 micrometers. Porous anodic alumina is particularly interesting because of its usefulness in the development of nanostructured materials such as electronic and optoelectronic devices, chemical sensors, biochemical membranes, carbon nanotubes, and metallic/semiconducting nanowires and nanorods.

Ideally, PAA is fabricated with pores arranged in a close-packed array of columnar hexagonal cells, with a central pore normal to the substrate (aluminum) in each cell. Hexagonal organization represents the greatest number of pores per unit area, while also ensuring uniformity of pore diameters. Thus in my research I will strive to create hexagonal order.

Project Objectives

  • determine optimal conditions for anodization in sulfuric acid
  • find maximum voltage for production of sub-20nm pores in sulfuric acid
  • attempt two-step anodization in two different electrolytes at different voltages
  • determine effectiveness of pore shrinking via heat treatment
  • produce sub-20nm pores with highest possible order

Approach

  • anodize aluminum in sulfuric acid varying concentration, anodization time, temperature, and voltage
  • perform two-step anodization in two different electrolytes at different voltages
  • 104V in phosphoric acid (1M), 10V in sulfuric acid (0.313M)
  • 25V in sulfuric acid (0.313M), 10V in sulfuric acid (0.313M)
  • 40V in oxalic acid (0.3M), 10V in sulfuric acid (0.313M)
  • 30V in oxalic acid (0.3M), 10V in sulfuric acid (0.313M)
  • experiment with three- and four-step anodization process
  • perform heat treatment in boiling water on hexagonally ordered pores

Findings

The optimal sulfuric acid concentration for anodization was found to be 0.3M. In sulfuric acid, sub-20nm pores were only produced in samples anodized at under 15V. When anodizing for over 20 hours in sulfuric acid, greater pore order was not observed; on the other hand, the average pore diameter of samples anodized over 20 hours seemed to be slightly larger than that of samples anodized for shorter periods, indicating potential etching at the surface.

When performing two-step anodizations with different electrolytes and voltages for the first and second anodization, the size of cells formed during the first anodization was proportional to the voltage applied during the first anodization, regardless of electrolyte. For example, cells formed in phosphoric acid at 104V were, on average, 280nm in size. Cells formed in sulfuric acid at 25V were an average of 60nm in size.

Additionally, the number of pores per cell was proportional to voltage of the first step (assuming constant second-step conditions). For example, with a first step anodization in phosphoric acid at 104V and a second step in sulfuric acid at 10V, approximately 100 pores per cell were observed. With a first step anodization in sulfuric acid at 25V and a second step in sulfuric acid at 10V, approximately 5 pores per cell were observed. Again we see a proportionality between the voltage for the first anodization – or cell size – and the number of pores per cell after the second anodization.

From these results it was evident that cells and pores could be tuned in such a way as to form cells with 7 pores per cell, potentially producing a hexagonal structure (optimal pore order). Our first goal was to generate cells which were perfectly hexagonally structured; this goal was achieved by way of a three-step anodization process – anodizing twice at the higher voltage in the first electrolyte bath, then once in the final electrolyte bath (generally sulfuric acid at 10V, for sub-20nm pores).

Next, the cell size was adjusted such that, on average, 7 pores were formed in each cell. These results were obtained by way of a three-step anodization, in which the first two steps were performed in oxalic acid at 30V, while the third anodization was carried out in sulfuric acid at 10V. The average pore diameter was approximately 15nm. Unfortunately, hexagonal organization was not observed.

Finally, pore shrinking was employed on templates formed using optimal conditions in oxalic acid. A 20% pore diameter reduction was seen on a sample anodized in 0.3M oxalic acid (two-step) at 40V which was treated in boiling water for 1 min. The same sample was ruined when treated for periods of 5, 10, and 20 minutes.

Pore shrinking by way of heat treatment in boiling water was not further explored due to time constraints. In the future, samples formed at optimal conditions in sulfuric acid should be heat treated for times ranging from 5 to 240 seconds, and cross-sections of these samples should be imaged.


Three-step anodization. First two steps in 0.3M oxalic acid at 30V and 4 degrees C. Third step in 0.3M sulfuric acid at 10V and 4 degrees C.
Two-step anodization in 0.3M oxalic acid at 40V and 4 degrees C. Heat treated in boiling water 1 min.

Contact me: sgt2102@columbia.edu