- You are here:
Dissertation Defense – Bedi Aydin Baykal
MSE Grad Presentation
Thursday, March 29, 2018 - 2:00pm
MRDC 3515, Hightower Conference Room
Committee Members: Prof. Preet M. Singh, Advisor, MSE Prof. Faisal M. Alamgir, MSE Prof. Victor Breedveld, ChBE Prof. Hamid Garmestani, MSE Prof. Arun M. Gokhale, MSE
"Flow-Assisted Erosion-Corrosion of Steel in Pulping Liquors: Effect of Mechanical Properties of Fluids and Addition of Hard Particles"
Erosion-Corrosion is a type of corrosion where mechanical (erosive) and chemical (corrosive) effects combine to accelerate material loss due to corrosion. Erosion-corrosion can cause a higher rate of wastage compared to the sum of its components through its negative impact on the reformation of a thin, protective oxide layer on the surface of metals. The most important industrial occurrences of erosion-corrosion are in the pulp, paper, nuclear, petrochemical and pharmaceutical industries. Most of these industries employ alkaline streams such as pulping liquors, where erosion-corrosion has been observed.
While erosion-corrosion maps have been produced and modeling has been done in certain environments, the effect of fluid properties has not been explored and this study aims to fill this gap using electrochemical (linear polarization resistance-LPR) testing of C1018 carbon steel, 304L and 316L austenitic stainless steel and 2205 duplex stainless steel in pulping liquors, a highly alkaline chemical mixture used in pulping. Flow regimes and their effects within white liquor were characterized and a method to control viscosity in white liquor through addition of agar powder was developed. It was found that acceleration of material loss due to erosion-corrosion happened at a critical Reynolds number and not a critical speed, pointing out that viscosity was an important factor in the erosion-corrosion mechanism. Further results demonstrated that the a fluid with higher viscosity caused higher erosion-corrosion rates at the same speed as a fluid with lower viscosity.The wall shear stress acting on the passive film and the erosion-corrosion rate were found to be directly proportional. The entire dataset generated was used to create a predictive model using a support vector machine technique.