Date of Award


Degree Name

Master of Science


Mechanical and Aerospace Engineering

First Advisor

Daniel Kujawski, Ph.D.

Second Advisor

Muralidhar Ghantasala, Ph.D.

Third Advisor

Sam Ramrattan, Ph.D.


Fatigue, slow strain rate, stress, corrosion

Access Setting

Masters Thesis-Open Access


When designing for any mechanical components or system, the question would arise as to how the material would react to the loads subjected on it? Would the component survive its service load? How would it react to environmental corrosion? To answer these questions, the technique used in this thesis paper is the Slow Strain Rate Test (SSRT) method. Aluminum and steel were chosen as the material to be tested in this paper. Al 7075-T651, and Al 6061-T651 was chosen due to its wide range of application, high strength to weight ratio and ease of machinability. It is highly used in the aerospace industry, for fuselage and wings of airplanes. AISI 4130 steel was chosen for its high strength and low cost. It is highly used in the automotive industry for components such as, chassis and cab A pillar, where high strength is needed for the safety of passengers.

In this thesis paper, when failure occurs due to corrosion, the high stress concentration will be termed as “chemical notch”, kchem. Chemical notch is derived by conducting SSRT, subjecting specimens with various physical notched under corrosive environment of 3.5% NaCl with a pH of 2.5. Chemical notch is then used to predict the corrosion stress-life (S-N) curve of the individual material. S-N curves are generally used in the industry when evaluating the service life of a component. In order to generate an S-N curve, several specimens would be required for testing, and at various stress amplitude with multiple specimen runs for higher accuracy. These are all lengthy processes. This thesis paper seeks to use kchem estimation line in predicting corrosion S-N curve. By using this new and novel approach, it would not only save time but also cost, as the number of specimens needed for testing is far less.