Chemical Engineering Science 64 (2009) 304--312 Contents lists available at ScienceDirect Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces Thermal polymerization of styrene in the presence of TEMPO Afsaneh Nabifar a , Neil T. McManus a , Eduardo Vivaldo-Lima b , Liliane M.F. Lona c , Alexander Penlidis a, ∗ a Department of Chemical Engineering, Institute for Polymer Research (IPR), University of Waterloo, Waterloo, Ontario, Canada N2L 3G1 b Departamento de Ingeniería Química, Facultad de Química, Universidad Nacional Autónoma de México (UNAM), Conjunto E, Ciudad Universitaria, México D.F., CP 04510, Mexico c Department of Chemical Processes, School of Chemical Engineering, University of Campinas (UNICAMP), P.O. Box 6066, Campinas, São Paulo, Brazil ARTICLE INFO ABSTRACT Article history: Received 27 June 2008 Received in revised form 3 October 2008 Accepted 11 October 2008 Available online 1 November 2008 Keywords: Polymers Polymerization Controlled/living radical polymerization Styrene polymerization Reaction engineering Kinetics To investigate aspects of the contribution of (thermal) self-initiation in nitroxide-mediated radical poly- merization (NMRP) of styrene, selective styrene polymerizations with 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) in the absence of initiator were carried out at 120 and 130 ◦ C. The results of these experiments (including conversion data, molecular weight averages, polydispersity and molecular weight distribution information) were compared with regular thermal polymerization of styrene and NMRP of styrene in the presence of a bimolecular initiator (benzoyl peroxide; BPO). It was observed that although the ther- mal polymerization of styrene can be controlled to some extent in the presence of TEMPO to provide polystyrene with low polydispersity, the polymerization was never as controlled as that obtained by a BPO-initiated NMRP. © 2008 Elsevier Ltd. All rights reserved. 1. Introduction Controlled radical polymerization (CRP) is one of the most rapidly developing areas of polymer science. Its versatility and ability to produce novel polymer structures (block and gradient copolymers; star, comb, and hyperbranched architectures) are perhaps the main reasons for the increased academic (and potentially industrial) in- terest. Nitroxide-mediated radical polymerization (NMRP) is one of the three currently most popular approaches towards CRP. Polymeric materials synthesized by NMRP have the potential for uses as coat- ings, adhesives, lubricants, gels, thermoplastic elastomers, as well as materials for biomedical applications. It is well known that styrenics exhibit thermal self-initiation at elevated temperatures (Gao and Penlidis, 1996). Since NMRP of styrene is typically conducted at temperatures higher than 100 ◦ C, the contribution of thermal self-initiation is considerable. Formation of thermally initiated radicals is a source of deviation from ideal controlled polymerization when bimolecular or unimolecular initi- ating systems are employed in NMRP. On the other hand, thermal self-initiation helps maintain a reasonable reaction rate because it continuously generates radicals to compensate for loss of radicals due to termination reactions. In addition, radicals produced through styrene self-initiation could be captured by added nitroxides to give ∗ Corresponding author. Tel.: +1 519 888 4567x36634; fax: +1 519 746 4979. E-mail addresses: penlidis@cape.uwaterloo.ca, penlidis@uwaterloo.ca (A. Penlidis). 0009-2509/$ - see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.ces.2008.10.013 “in situ” unimolecular initiators. So the question arises as to whether it would be possible to conduct “controlled” free radical polymer- ization without the presence of “external” initiator, relying only on thermal self-initiation (as the radical source) and the added nitrox- ide to mediate the polymerization. Initial work in thermal self-initiation of styrene in the presence of TEMPO was reported almost simultaneously by Gaynor et al. (1994), Mardare and Matyjaszewski (1994) and Georges et al. (1995). Con- flicting results were obtained, with one group citing polydispersities of 2–2.5 (Georges et al., 1995), while under similar conditions the other group reported polydispersities in the range of 1.2–1.3 (Gaynor et al., 1994; Mardare and Matyjaszewski, 1994). Subsequently, Devonport et al. (1997) revisited the thermal initiation of styrene in the presence of TEMPO at 125 ◦ C. They showed that low polydis- persities and controlled molecular weights could be achieved under these conditions, although the degree of control was not as good as for unimolecular or bimolecular initiating systems. Later, Boutevin and Bertin (1999) conducted a kinetic study on the thermal poly- merization of styrene with TEMPO concentration ranging from 0.02 to 0.1 M at 120 ◦ C. Their calculations demonstrated that regardless of the initial concentration of TEMPO, the total concentration of macromolecular chains generated during polymerization was higher than the initial TEMPO concentration, hence concluding that not all macromolecular chain growth was controlled by TEMPO. Pan et al. (2004) studied thermal self-initiation of styrene in the presence of TEMPO in miniemulsion and compared it to the corre- sponding bulk system. As was established previously for bulk poly- merization (Gaynor et al., 1994; Devonport et al., 1997; Boutevin and