Scale-down characterization of post-centrifuge flocculation processes for high-throughput process development.
Bottom Line: The authors present a novel ultra scale-down (USD) methodology for the characterization of flocculation processes.This USD method, consisting of a multiwell, magnetically agitated system that can be fitted on the deck of a liquid handling robot, mimicked the flocculation performance of a nongeometrically similar pilot-scale vessel representing greater than three orders of magnitude scale-up.Mixing scales (i.e. macromixing, mesomixing or micromixing) modulated the flocs' size and determined the success of some of the scale-up correlations reviewed in the literature.
Affiliation: The Advanced Centre for Biochemical Engineering, Department of Biochemical Engineering, University College London, Torrington Place, London, WC1E 7JE, UK; Lonza Biologics plc, Slough, Berkshire, SL1 4DX, UK.Show MeSH
Related in: MedlinePlus
Mentions: Figure 3A shows that wider floc distributions were gradually obtained with decreasing flocculant addition time in both flocculation scales. However, the spread in the PSD differed depending on the flocculation system and the predominant mixing time scale. At pilot scale, the characteristic time constants were calculated by applying Equation 2009 to Equation 1999 to determine which was the predominant mixing time scale in each of the tadd studied. The results are summarized in Table 1. In the STR, macromixing was thought not to be limiting at the range of tadd studied (i.e. tadd and tm and tc were not of similar magnitude; Baldyga et al., 1993; Baldyga and Bourne, 1999) and tm remained unaffected by the changes in tadd. Therefore, micromixing and mesomixing were the mixing time scales controlling the PSD of the flocs formed at pilot scale. Micromixing governed at low flocculant addition rates (i.e. tadd > 140 s). Across the micromixing regime constant values of d10 and d50 were found, while the value of d90 increased by 8%. A step increase in the values of d50 and d90 took place when tadd was in the 102 s to 45 s range, which corresponded to increments of 4 µm and 9 µm respectively. These larger PSDs may be explained by both micromixing and mesomixing time scales influencing flocculation despite the micromixing time predictions continuing to be larger than those expected for mesomixing (i.e. te > td and ts). This situation changed at tadd < 24 s, when mesomixing was the only controlling mechanism (i.e. td > te). At the USD scale the PSD of the flocs generated gradually decreased with longer tadd. Equation 2009–Equation 1999 could not be applied and the mixing time scales were therefore experimentally characterized (see Section Experimental Characterization of Micromixing).
Affiliation: The Advanced Centre for Biochemical Engineering, Department of Biochemical Engineering, University College London, Torrington Place, London, WC1E 7JE, UK; Lonza Biologics plc, Slough, Berkshire, SL1 4DX, UK.