Automated Characterization of Aluminum Oxide Nanopore FESEM Images using Machine Learning Algorithms

  • Parashuram Bannigidad, Jalaja Udoshi, C. C. Vidyasagar

Abstract

Anodic aluminum oxide template has gained huge attention in the multidisciplinary nano-applications due to its guided and controlled surface geometry. The proposed study focuses on nano-membrane engineering, imaging and its characterization using an automated system. The nanopore attributes; poresize, wall thickness and porosity in the synthesized membrane for different anodizing parameters are extracted. The study exhibits the change in acidic concentration; 0.1%, 0.2%, 0.3%, 0.4% and 0.5% of the anodization bath with constant time (9 min), temperature (250C) and voltage (20V), the poresize are 83.07nm, 68.52nm, 19.93nm, 207.92nm and 102.43nm, wall thickness are 63.33nm, 75.2nm, 32.06nm, 121.17nm and 89.21nm and porosity is 0.58%, 0.12%, 0.19%, 0.06% and 0.06%. When the anodization time is changed; 3mins, 6mins, 9mins,12mins and 15mins, keeping in the concentration (3%), temperature (250C) and voltage(20V) constant, the poresize were 192.68nm, 237.38nm, 110.98nm, 26.12nm and 95.46nm, wall thickness was 42.19nm, 34.59nm, 34.54nm, 23.79nm and 46.65nm and porosity is 0.14%, 0.66%, 0.24%, 0.75% and 0.09%. Change in temperature 100C, 150C, 200C, 250C and 300C with constant concentration (0.3%), time (9 min) and voltage (20 V) has poresize 29.68nm, 75.72nm, 60.87nm, 37.18nm and 21.57nm, wall thickness 42.33nm, 57.16nm, 43.52nm, 42.21nm and 24.9nm and porosity is 1.68%, 0.37%, 0.06%, 0.04% and 0.24%. The voltage change; 10V, 15V, 20V, 25V and 30V have shown poresize 156.24nm, 64.02nm, 36.09nm and 189.3nm, wall thickness 481.04nm, 157.35nm, 29.87nm and 72.6nm and porosity is 0.01%, 0.03%, 0.22% and 0.49%, when concentration (0.3%), time (9 min) and temperature (250C) was kept constant. The experimental results coincide with the manual results obtained from the chemists and the analysis depict of all the samples in the experiment, only the C and S images are portraying honeycomb structure to an extent and could be apt for the required nano application.

Published
2020-03-31
How to Cite
Parashuram Bannigidad, Jalaja Udoshi, C. C. Vidyasagar. (2020). Automated Characterization of Aluminum Oxide Nanopore FESEM Images using Machine Learning Algorithms. International Journal of Advanced Science and Technology, 29(3), 6932 - 6942. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/7347
Section
Articles