A design of experiment (DoE) approach to optimize the inner geometry - PowerPoint PPT Presentation

A design of experiment (DoE) approach to optimize the inner geometry of baffled meso-scale tubes for continuous crystallization Louisa Ejim (Loughborough University) Supervisor: Dr. Nuno Reis 2 nd Supervisor: Prof. Chris D. Rielly

Figure 1 Aim & Objectives Glass tubes; Sharp edged baffle type (SEPC) Smooth edged baffle type (SPC) Baffle spacing ( l ) Orifice Inner tube diameter diameter F A ( d o ) ( 𝑒 = 10 𝑛𝑛 ) • To determine the optimal B G geometry • By characterising the tubes using H solid-liquid residence time C distribution (RTD) • DoE approach to investigate the I D effect of baffle spacing ( 𝑚 ), orifice diameter ( 𝑒 𝑝 ) and baffle type on RTD E J • PVC particles are used because of their similarities in flow properties to crystals.

Batch Suspension Studies (critical amplitude: centre-to-peak) Conclusions: • Increased oscillation frequency provides a lower critical amplitude required to suspend particles • Sharp edged-baffles of the same dimensions proved to require higher critical amplitude • Tube A ( 𝑚 = 30 mm, 𝑒 𝑝 = 3.5 mm) proved to require the lowest amplitude for suspension.

Continuous Liquid-solid RTD 𝐷 𝑝𝑣𝑢 (𝑢) 𝑢 𝑋 𝑢 = , 𝐺 𝑢 = 1 − 𝑋 𝑢 , 𝜄 = Conclusions: 𝐷 0 𝑢 Smooth-edged baffle type : • 1.2 1.2 Gives lower energy input SPC b SEPC a requirement (e.g critical amplitude) 1.0 1.0 • F Narrower RTD (close to plug 0.8 0.8 A flow behaviour) G B F( Ө ) F( Ө ) 0.6 0.6 C H • Uniform mixing D I 0.4 0.4 Sharp- edged baffle type : E J 0.2 0.2 • Chaotic mixing • Broader RTD 0.0 0.0 • Bubble retention/trapping. 0.0 1.0 2.0 3.0 4.0 0.0 1.0 2.0 3.0 4.0 Ө [ -] Ө [ -]

Plug flow with axial dispersion (D/uL)  Levenspiel Open- Open boundary condition: Smooth edged baffles (SPC) Sharp edged baffles (SEPC) 0.09 0.08 0.07 0.06 D/uL [-] 0.05 Fitting parameters: 0.04 • Superficial velocity (m/s) 0.03 0.02 • Axial dispersion 0.01 coefficient (m 2 /s). 0.00 A B C D E F G H I J Tube geometry

Plug flow with axial dispersion DoE plots Conclusion The open baffle area,  identified as the dominant design parameter in controlling solids backmixing and • batch suspension of particles, with small values of  = 0.12 resulting in minimised axial dispersion Strong eddy vortices generated at lower values of  , that presumably led to trapping of particles, an effect • not previously observed in OFRs • Tube A ( l/d = 3.0) showed clear flow characteristic advantages over the other meso-tubes studied.

Thank You!