Continuous Emulsion Styrene-Butadiene Rubber (SBR) Process: Computer Simulation Study for Increasing Production and for Reducing Transients between Steady States Roque J. Minari, Luis M. Gugliotta, Jorge R. Vega, and Gregorio R. Meira* INTEC (UniVersidad Nacional del Litoral and CONICET), Gu ¨emes 3450, (3000) Santa Fe, Argentina The industrial production of two styrene-butadiene rubber (SBR) grades is optimized by means of a representative mathematical model. The emulsion process involves a train of seven continuously stirred-tank reactors. In the simulations, intermediate feeds of comonomers and chain transfer agent (CTA) are admitted. The steady-state (SS) productions can be increased by ∼8%, while maintaining the required final values of conversion, particle diameter, and molecular characteristics (M h n , M h w , and the average number of branches per molecule). For the same rubber grade, the polymer production can be changed with negligible generation of intermediate off-specs, by regulating the global calorimetric conversion. For changes of grade at a fixed polymer production, transient profiles of the intermediate CTA feeds are proposed. For changes of production and grade, a sequential (rather than a simultaneous) transition seems preferable. Introduction Styrene-butadiene rubber (SBR) is a general-purpose com- modity that is mainly used in the tire industry. It is produced by copolymerizing styrene (S) and butadiene (B) in emulsion processes that are performed in up to 15 continuously stirred tank reactors (CSTRs) in series. Compared to the batch or semi- batch processes, continuous processes offer improved efficiency and better product consistency. 1-3 The reagents are continuously fed into the first reactor, and the product (a synthetic latex) is removed from the last reactor. The unreacted monomer then is recuperated from the latex, and the rubber is precipitated and dried. “Hot” SBR grades are produced at ∼50 °C with persulfate initiators, whereas “cold” grades are synthesized at 5-10 °C, using redox couples. The rubber quality is determined by the molecular weights, chemical composition, and degree of branch- ing. Cold grades are preferable to hot grades, for their reduced levels of branching, cross-linking, and low-molecular-weight material. 2 In industry, the rubber quality is measured by the Mooney viscosity (ASTM D 1646), which is indirectly related to the molecular characteristics and the possible incorporation of (low-molecular-weight) additives. 4,5 In the investigated cold copolymerizations, the reactivity ratios are relatively similar to unity (r B ) 1.7 and r S ) 0.44), and this determines a moderate compositional drift, that slightly increases the mass fraction of S in the copolymer. The most common cold SBR grades are known as 1712 and 1502, with the former exhibiting higher molecular weights. Also, grade 1712 includes a mineral oil, whereas no oil is added onto grade 1502. For both grades, the mass fraction of S is ∼23.5%. Molecular branching is due to reactions between the growing free radicals and B repeating units. The molecular weights and branching are controlled by the addition of a chain transfer agent (CTA) or “modifier”, and by limiting conversion to ∼70%. The limitation of conversion also limits the gel formation and the compositional drift. 2 The mathematical modeling of emulsion copolymerizations has been reviewed by Saldı ´var et al. 6 Emulsion models have been classified according to the way in which they treat the free-radical compartmentalization. 7 The simplest (pseudo-bulk) models estimate the molecular weight distribution (MWD) as in a bulk process, 8 calculating the total number of free radicals from the product between the total number of particles (N p ) and the average number of radicals per particle (n j). If most of the dead polymer is generated by chain transfer to the modifier, then pseudo-bulk models have proven adequate for predicting the produced MWDs. 7,9-17 The steady-state (SS) control of continuous emulsion pro- cesses has been investigated by Poehlein and Dougherty, 3 who intended to increase the polymer production of a single CSTR, by finding the optimal mean residence time (θ) that maximizes N p . At low θ values, N p is mainly determined by the initiation rate (R I ), and it is essentially unaffected by the emulsifier concentration. At high θ values, the polymer particles grow to larger sizes, which reduces the total number of soap micelles, and the N p results are strongly affected by the emulsifier concentration, but unaffected by R I . The N p -θ relationship presents a maximum for systems exhibiting Case II kinetics. 18,19 Since practically no polymerization occurs in the monomer droplets phase, it is possible to increase the polymer production by reducing the (almost inert) monomer phase in the first re- actors of the train. The total monomer phase in the first reactors can be reduced by splitting the feed of comonomer mixture into two or more reactors. 20-23 Thus, Hamielec and MacGregor 20 proposed to split the (unemulsified) monomer feed between the first and fourth reactor in an emulsion copolymerization of S and B that was conducted in a six-reactor train. For splitting ratios 90/10 and 80/20, the polymer production was increased by 3% and 6%, respectively. Also, the branching frequency was practically unaffected, but the average molecular weights were slightly augmented. 20 In addition to splitting the comonomers feed, Penlidis et al. 21 suggested to simultaneously control molec- ular weights and branching by also splitting the total CTA feed. Furthermore, the copolymer composition could be made more uniform, via intermediate addition of the most reactive comono- mer (B). 21 Kanetakis et al. 22 also investigated the increase of productivity by splitting the comonomers feed, showing that the degrees of branching could be reduced when increasing the total inlet flow rate at a fixed monomer conversion. In industrial practice, the changes of grade and/or of the level of production are rather frequent, for which reason a large * To whom correspondence should be addressed. Tel.: 0054-342- 456-5882. Fax: 0054-342-455-0944. E-mail: gmeira@ceride.gov.ar. 245 Ind. Eng. Chem. Res. 2006, 45, 245-257 10.1021/ie0504755 CCC: $33.50 © 2006 American Chemical Society Published on Web 11/25/2005