Premium
Monitoring Pyrrol Polymerization Using On‐Line Conductivity Measurements and Neural Networks
Author(s) -
Brusamarello Claiton Z.,
Santos Leila M.,
Amaral Monique,
Barra Guilherme M. O.,
Fortuny Montserrat,
Santos Alexandre F.,
de Araújo Pedro Henrique Hermes,
Sayer Claudia
Publication year - 2013
Publication title -
macromolecular symposia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.257
H-Index - 76
eISSN - 1521-3900
pISSN - 1022-1360
DOI - 10.1002/masy.201300042
Subject(s) - conductivity , artificial neural network , electrical resistivity and conductivity , pyrrole , gradient descent , polymerization , polypyrrole , line (geometry) , materials science , biological system , analytical chemistry (journal) , chemistry , computer science , chromatography , mathematics , composite material , physics , machine learning , organic chemistry , polymer , geometry , quantum mechanics , biology
Summary In this work, the chemical oxidative synthesis of polypyrrole was monitored through on‐line conductivimetry. A group of reactions was carried out at two different temperatures (5 °C and 20 °C) with varying concentrations of pyrrole and oxidant agent (FeCl 3 ) to investigate the effect on conversion and conductivity. The relation of electrical conductivity with conversion was not straightforward. To overcome this, a neural network was proposed to predict conversion based on on‐line conductivity measurements. In this way, the neural networks entry variables were: electrical conductivity, reaction temperature, initial oxidant and pyrrole concentrations. As optimization algorithms Levenberg‐Marquardt and Descent Gradient with momentum term were evaluated. Results obtained by the neural networks based on on‐line conductivity measurements showed a good agreement with off‐line conversion data.