Analysis of corrosion inhibition of Kleinhovia hospita plant extract aided by quantification of hydrogen evolution using a GLCM/SVM method

Abstract

During a corrosion inhibition test, a combination of common electrochemical corrosion test methods with an in-situ quantification of H2 evolution could provide a comprehensive analysis of the effectiveness of an organic inhibitor. This work analyzes the corrosion inhibition efficiency of Kleinhovia hospita plant extract on carbon steel specimens polarized in 1 M HCl, based on acquired H2 bubbles images, by using gray level co-occurrence matrix (GLCM) and support vector machine (SVM) classification. A conformity was established between the classified-algorithm models and the corrosion test results obtained by potentiodynamic polarization test and electrochemical impedance spectroscopy. Hydrogen rate and corrosion rate show the same lowest trends at the addition of 3000 mg/L of KH extract. The inhibitor addition led to 99% of maximum inhibition efficiency. Based on the polarization data, KH extract is a mixed type inhibitor. Supported by Langmuir calculation for adsorption isotherm, a physisorption is stated as the main inhibition mechanism. The feature extraction using GLCM was able to distinguish changes in H2 bubbles characteristics where the addition of inhibitor affected the corrosion rate. The GLCM/SVM method applied as a linear kernel function and showed 88% accuracy with d = 5 for image data testing. Remarkable changes in H2 gas bubbles characteristic were observed in the specimen immersed in the solution with 3000 mg/L inhibitor addition, signified by 99% inhibition efficiency.