Statistical modelling of the amination reaction of cyclohexanol to produce cyclohexylamine over a commercial Ni-based catalyst

• Type: Article in International Scientific Journal
• Date: 2021

Authors

Rui Churro (Author) Other; Fernando Mendes (Author) Other; Paulo Araújo (Author) Other; Filipa Ribeiro (Author) Other; Joana Peres (Author) FEUP; Luís M. Madeira (Author) Other; Luis Madeira (Author) FEUP

Journal

Title: Chemical Engineering Research and Design

Vol. 170

Pages: 189-200

ISSN: 0263-8762

Publisher: Institute of Chemical Engineers

Indexing

ISI Web of Knowledge

ISI Web of Science

Scopus - 0 Citations

Other information

Authenticus ID: P-00T-TYK

DOI: 10.1016/j.cherd.2021.03.029

Abstract (EN): Multivariate linear regression (MLR) models between the operating conditions and the performance parameters for the amination reaction of cyclohexanol (CHOL) with NH3 as aminating agent and in the presence of H2, using a commercial Ni-based catalyst, were established. The regressors were chosen according to a stepwise variable selection approach so that only significant and non-redundant input variables were selected. The obtained models (related to CHOL conversion and yields for the more relevant products) were statistically significant with good fittings. Moreover, the models also reveal to act as good predictors, by cross validation and external validation. Nevertheless, models fitted to cyclohexyldeneaniline (CHONE) formation still present opportunities for improvement, given their poorer prediction performance. Sensitivity analysis of the developed models confirm the tendencies observed by Exploratory Data Analysis (EDA) indicating that the established relationships are able to translate the experimental observations. Therefore, the models can be used for process optimization purposes and scale-up methodologies. © 2021 Institution of Chemical Engineers

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 12