Optimization of the photochemically initiated polymerization of methyl methacrylate

• Published in: Chemical engineering and processing Vol. 45; no. 11; pp. 1001 - 1010
• Main Authors: Silvares, Adriano F.M., Nascimento, Claudio A.O., Oliveros, Esther, Bossmann, Stefan H., Braun, André M.
• Format: Journal Article
• Language: English
• Published: Lausanne Elsevier B.V 01.11.2006; Elsevier
• Subjects: Applied sciences; Benzoin; Chemical engineering; Exact sciences and technology; Gel permeation chromatography; Gravimetry; Methylmethacrylate; Optimal experimental design methodology; Photochemically initiated polymerization; Polydispersity; Polymethylmethacrylate; Pre-polymers; Rates of conversion; Reactors; Rates of conversion; Polymethylmethacrylate; Optimal experimental design methodology; Photochemically initiated polymerization; Pre-polymers; Methylmethacrylate; Polydispersity; Benzoin; Gel permeation chromatography; Gravimetry; Gravimetric analysis; Molecular weight distribution; Conversion rate; Polymerization; Photochemical reactor; Modeling; Optimization; Batchwise; Experimental design; Production
• ISSN: 0255-2701; 1873-3204
• DOI: 10.1016/j.cep.2006.03.011

The photochemically initiated polymerization of methyl methacrylate (MMA) using benzoin as an initiator was investigated for the production of a liquid pre-polymer (PMMA) of narrow molecular weight distribution. The experiments were carried out in a batch procedure using a tubular photochemical reactor with negative irradiation geometry and connected to a reservoir. The products were characterized using gel permeation chromatography (GPC) and gravimetric analysis. Initial initiator concentration and flow rate were identified as primary process parameters effecting the molecular weight distribution of the pre-polymer. Conditions for minimizing polydispersity were determined on the basis of the experimental design methodology. A series of designed experiments (Doehlert matrix) were performed for calculating the coefficients of the corresponding quadratic polynomial model. Results show that compromises have to be made between polydispersity and conversion rate.