An experimental and numerical investigation of polymerization coupled with an oscillating reaction

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Chemistry and Biochemistry

First Advisor

John A. Pojman

Advisor Department

Chemistry and Biochemistry


Free-radical polymerization was coupled to the Belousov-Zhabotinsky oscillating reaction. Several monomers were tried with acrylonitrile being chosen for further studies. Acrylonitrile, upon polymerization, precipitates out of the aqueous medium, forming a fine powder that is easily dispersed using magnetic stirring. It was found that some monomers, such as methyl methacrylate, will only polymerize if the acid concentration is at least 3 M. Polymerization coupled with oscillations was run in batch, semi-batch and flow reactors. Different behavior was observed and is reported. A mechanism for periodic polymerization of acrylonitrile was proposed based on experimental findings. Acrylonitrile affected the cerium catalyzed oscillating reaction in two ways: (1) inhibited oscillations and (2) increased the induction period. This led to the suggestion that acrylonitrile was removing HOBr from the system. A modified version of the FKN mechanism was used to model qualitatively the periodic behavior of acrylonitrile in the BZ reaction. The model also predicted the increase in induction period in the presence of acrylonitrile. The model was verified by considering three considerations found experimentally: (1) little to no increase in conversion during the induction period, (2) percentage conversion of 52%, and (3) periodic polymerization. Several modes of termination were considered; namely, (1) mutual interaction of propagating radicals, (2) BrO2. termination, and (3) Ce(IV) termination. According to numerical simulations, periodic termination by BrO2. and combination satisfactorily explain periodic behavior with appropriate percentage conversion. Bimolecular termination alone could not explain experimental findings, owing to the large increase in conversion during the induction period. It is found experimentally that the conversion does not increase appreciably during the induction period. At the onset of oscillations, there is an observable increase in conversion. Very low conversion (<1%) was obtained when termination by combination of polymer radicals and Ce(IV) reaction with polymer radicals were considered as possible routes. Only when bimolecular termination and BrO 2 . were considered as possibilities did simulations yield similar behavior to experimental findings. In considering only bimolecular termination of propagating radicals, it is concluded also that the periodic initiation is not sufficient in obtaining periodic polymerization. A variant of the BZ reaction, the R√°cz system, was used. No induction period is observed in the R√°cz system, with the system consisting of a higher malonic acid and acid concentration. Simulations were performed using the Radicalator model with the same rate constants for polymer kinetics used in the previous simulation. Periodic polymerization was identified in the numerical results. However, further considerations were needed to obtain a better fit to experimental findings, namely the two types of malonyl radical present and a reduction in the rate constant for bromine dioxide radical combination. These findings lend support to the validity of the proposed mechanism. Several organic substrates were substituted for malonic acid. Periodic polymerization was found with lactic acid, citric acid, and oxalic acid. Each of the organic substrates behaves differently in the BZ oscillating reaction. Despite the mechanism describing oscillations in each respective system, periodic polymerization was possible. A totally inorganic oscillator was also used. Manganese was used as the catalyst, and periodic polymerization of acrylonitrile was found in this system. These findings show that periodic polymerization is a general phenomenon. The investigation of polymerization coupled with oscillations led to studies done in a semi-batch reactor and continuous flow stirred tank reactor (CSTR). Even though strange behavior was found in the semi-batch reactor, it was evident that the oscillating reaction was controlling the polymerization reaction. A change in temperature followed a change in the potential of the oscillation reaction. Finally, the system was run in a CSTR. At low flow rate, periodic polymerization was shown; however, a steady state was achieved at high flow rate. A comparison of molecular weight distribution showed that the polymer formed during the steady state was of narrower, lower molecular weight than polymer produced during the oscillatory stage.