PBL with the Application of Multiple and Nonlinear Linear Regression in Chemical Kinetics and Catalysis

Abstract

In the resolution of problems in chemical kinetics and catalysis the mathematical models relate the independent variable that is usually time, with the dependent variable which is normally the concentration of a reactant. They conform to linear models, whose parameters such as the ordering to origin and the slope are kinetic parameters, applying linear regression to the experimental data can be obtained linear models that are the simplest to be found. On other side, multiple models can be found with two or more independent variables, however, there are models that are not feasible to linearize and therefore you can make use of nonlinear regression. This paper is exemplified by the Problem-Based Learning (PBL) scheme and the use of software such as Stat graphics and Polymath to solve the resulting mathematical models. The results obtained are compared with traditional methods as multiple linearized least squares, in this way the learning teaching process is strengthened, since the student raises the kinetic model and then solves it with computer tools. Through the PBL cycles generated in the resolution of real problems in chemical kinetics and catalysis, significant learnings are achieved on difficult comprehension topics such as initial rate for the determination of the reaction order, half-life times, temperature effect and heterogeneous catalysis.