Optimal design of nonlinear temperature programmed reduction experiments

We propose the application of nonconstant temperature gradients to improve the quality of temperature programmed reduction (TPR) experiments with respect to parameter estimation and model discrimination. This leads to TPR experiments with nonlinear temperature profiles (N‐TPR). To determine optimal profiles for the temperature gradient, optimal control problems are set up and solved numerically. The results show that N‐TPR experiments can be significantly better than traditional linear TPR experiments for many different scenarios. To implement these results in practice, we develop and demonstrate reduced optimization problem formulations, which can be solved faster and more reliably than the original formulation, with very similar results.