The effect of confinement on thermal frontal polymerization† Preeta Datta, Kirill Efimenko and Jan Genzer * Received 11th August 2012, Accepted 7th September 2012 DOI: 10.1039/c2py20640d In thermal frontal polymerization, the interplay between heat diffusion and Arrhenius type reaction kinetics gives rise to a prop- agating reactive front that is effectively sustained through a system positive feedback mechanism. We seek to understand how spatial confinement affects the system dynamics and overall polymerization kinetics. We find that with increasing confinement of the system, the front propagation velocity decreases and the nearly ‘‘parabolic’’ front profile flattens out. While in bulk or unconfined systems the convective heat and mass transfer effects are responsible for higher heat generation rates and higher temperature gradients, leading to faster front propagation, the convection is significantly suppressed in highly confined systems. Consequently, a smaller number of radicals are generated, resulting in a slower propagation step and long polymer chains. Thermal frontal polymerization involves spontaneous conversion of monomer units to polymers via a localized exothermic radical reac- tion initiated by an external heat source, which self-propagates in an unstirred reaction vessel. 1,2 Front propagation occurs through posi- tive feedback mechanisms as a result of the interplay between heat diffusion and Arrhenius type reaction kinetics. 3,4 The reactive front and its stability are affected by various factors, i.e., initiation temperature, initiator concentration, monomer concentration in solution, and reactor geometry. 5–7 Most previous frontal polymeri- zation studies have employed a vertical reactor in order to obtain a stable propagating front. 8,9 In contrast, in horizontal channels, convection gives rise to front instabilities and ‘‘fingering’’. 10–13 Here we report the first investigation of the effect of confinement on frontal polymerization in horizontal channels. Fig. 1 depicts a schematic diagram of the thermal frontal poly- merization process in a horizontal reaction chamber. Since the front propagation properties are strongly correlated with the heat gener- ated in/removed from inside the polymerization reactor, it is impor- tant to estimate the heat losses in the system/mechanism during heat transfer processes and find an appropriate correlation with the overall system geometry. Previous studies of ‘‘bulk’’ systems have suggested that predominant heat loss to the ambient environment occurs due to heat convection through the reaction media–glass interface. 5,14 This heat loss is proportional to the total surface area of the reaction chamber. For long channels with a high aspect ratio cross-section, heat loss takes place primarily through the top and bottom surfaces, and can thus be assumed to be independent of the channel height. It is important to point out that the exothermic heat generated in the confined system depends on the reaction volume and is therefore proportional to the channel height. Thus, by tuning the vertical confinement of the reaction channel, we can control the diffusion of heat into the unreacted monomers and thereby tune the reaction kinetics. The reaction chamber consists of a Teflon jacket sandwiched between two thick (z6 mm) borosilicate glass plates (to minimize heat loss); the thickness of the Teflon jacket is varied from 0.4 to 5 mm to achieve different degrees of vertical confinement. Acrylamide (>99%, Sigma-Aldrich) monomer is dissolved in dimethyl sulfoxide (DMSO, Fisher) with potassium persulfate (Sigma-Aldrich) acting as the reaction initiator. We have chosen this system due to its well established polymerization mechanism without detrimental side reactions, including, non-volatile side products (i.e., no gas bubbles), formation of polyacrylamide gels in DMSO (i.e., fronts are not destroyed by ‘‘fingering’’), low volatility of DMSO (i.e., no vapors, hence no bubbling), and visibility of a distinct front due to density differences. 15 Trace amounts of hydroquinone (Acros) are used as an inhibitor to eliminate spontaneous polymerization and thus to ach- ieve a better control of the polymerization process. For all the experiments, a 1 : 750 ratio of initiator to monomer (by moles) is used. Infrared (IR) thermal images as well as regular photographic snapshots of the system are captured at regular time intervals to record the temperature profiles and polymerization front position, using FLIR A325sc (FLIR) and Canon Rebel iX (Canon) cameras, respectively. Typical examples of the temperature profiles are shown in Fig. 2. An hour after initiation, the assembly is dismantled and the Fig. 1 Schematic diagram of the positive feedback mechanism in a frontal polymerization system. Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, NC 27695-7905, USA. E-mail: Jan_Genzer@ ncsu.edu † Electronic supplementary information (ESI) available. See DOI: 10.1039/c2py20640d This journal is ª The Royal Society of Chemistry 2012 Polym. Chem., 2012, 3, 3243–3246 | 3243 Dynamic Article Links C < Polymer Chemistry Cite this: Polym. Chem., 2012, 3, 3243 www.rsc.org/polymers COMMUNICATION