Back in October we presented a talk at Gulf Coast Conference that concerned the prediction of the chemical and physical properties of heavy petroleum feeds being converted to higher value product in a residual catalytic cracker (RCC). Over the years we have analyzed these materials by 300 and 60 MHz NMR and obtained good PLS-regression models that can adequately predict properties for real-time process control and optimization in a petroleum refinery. With the advent of a large number of new benchtop NMR systems we have been convincing ourselves that these types of analyses can be performed by systems such as the Magritek Spinsolve 43 MHz. We ran a series of samples that had been sitting around our lab for 15 years by dissolving them at about 50 volume% in a 50/50 CDCl3/CS2 solvent system. For each sample we had laboratory test data for a number of chemical and physical properties of interest to process engineers. We regressed the lab data variability against the variability in the Magritek 43MHz 1H NMR spectra and obtained cross-validated PLS models. The presentation material is given here at this link – Gulf Conference Presentation – 43 MHz RCC Feedstream Regression Models