Neural networks for kinetic parameters determination, signal filtering and deconvolution in thermal analysis

Feedforward neural networks have been used for kinetic parameters determination and signal filtering in differential scanning calorimetry. The proper learning function was chosen and the network topology was optimized, using an empiric procedure. The learning process was achieved using simulated thermoanalytical curves. The resilient-propagation algorithm have led to the best minimization of the error computed over all the patterns. Relative errors on the thermodynamic and kinetic parameters were evaluated and compared to those obtained with the usual thermal analysis methods (single scan methods). The errors are much lower, especially in presence of noisy signals. Then, our program was adapted to simulate thermal effects with known thermodynamic and kinetic parameters, generated electrically, using a PC computer and an electronic interface on the serial port. These thermal effects have been generated by using an inconel thread.     

Isothermal and Non-Isothermal Kinetics When Mechanistic Information Available

In the case of a complex mechanism of two parallel independent reactions, peak maximum evolution methods and model-fitting methods give only a mean value of the kinetic parameters, while isoconversional methods are useful to describe the complexity of the mechanism. Isothermal and non-isothermal isoconversional methods can be used to elucidate the kinetics of the process. Nevertheless, isothermal isoconversional methods can be limited by restrictions on the temperature regions experimentally available because of duration times or detection limits. This revised version was published online in August 2006 with corrections to the Cover Date.     

Chemorheological analysis and model-free kinetics of acid catalysed furfuryl alcohol polymerization

The complete curing of furfuryl alcohol (FA), was studied by chemorheological analysis and model-free kinetics under isothermal and non-isothermal modes. Polymerization of FA under acidic catalysis involves complex reactions, with several steps (such as condensations and Diels-Alder cycloadditions). To account for the polymerization complexity, kinetic analysis of DSC data was performed with a model-free isoconversional method. The obtained E(alpha)-dependencies were closely-correlated with the variation of complex viscosity during curing. Linear condensations are predominant during the early curing stage and are followed by two distinct stages of branching cycloadditions. Gelation and vitrification, identified by rheometric measurements, were associated with a decrease of the overall reaction rate that becomes controlled by diffusion of small oligomers. Before vitrification, the rate of crosslinking is limited by the mobility of longer polymer chains and diffusion encounters a large energy barrier due to the cooperative nature of the motions, leading to higher E(alpha) values.     

Isoconversional method to explore the mechanism and kinetics of multi-step epoxy cures

Epoxy-anhydride cure has been studied by differential scanning calorimetry under nonisothermal conditions. The isoconversional method of kinetic analysis yields a dependence of the effective activation energy on the extent of conversion (α) that increases from ≈ 20 to 70 kJ·mol^–1 with increasing α. The occurrence of the dependence suggests a mechanism of competing reactions that are found to be the processes of initiation and propagation of the polymer chain.