Issue link: https://www.e-digitaleditions.com/i/1499609
PharmTech.com Trends in Manufacturing eBook May 2023 Pharmaceutical Technology ® 23 continuous manufacturing ers are designed to be much faster than the react ion, the effect of scale on module performance will be minimal. If necessary, the approach can be easily modified to include three or more reactants in any molar ratio desired. Module 2: Tubular reactor with multiple heat exchange stations If the reactive stream is pre-heated to target tempera- ture and pre-mixed quickly, the main function of the tubular reactor is to provide the system with enough time to reach optimum conversion, and with enough heat exchange capacity to maintain the desired tar- get temperature(s) along the reactor length. Figure 3 shows a schematic of the reactor module. Conversion of reactants to desired product is measured and is con- trolled by manipulation of the overall mass f low rate and temperature, which is measured and controlled at multiple points utilizing one or more heating ele- ments. This general approach enables the modeling, control, and scale-up of many types of reactions. The basic requirement of this module is that it needs an accurate chemical kinetics model describing the rel- ative rates of all competing chemical reactions, and information regarding the relevant transport param- eters describing material behavior. At the very small scales used for chemical process design, tubular reactors operate in the laminar re- gime. For each reactant and product, the predictive model is the simultaneous solution in radia l and a x ia l coord i n ates of t he heat a nd mu lt i-species ma ss t ra nsfer equat ions, i nclud i ng t he ef fec t of chemical reactions. However, as the system is scaled up beyond 1000x, most cases of interest will take place in t urbulent f low, which achieves ef f icient local mixing. This allows for design of the reactor using a simplified one-dimensional (1D) convection- d i s per sion equ at ion , where t he on ly mode l pa- rameter is the dispersion coefficient, which can be measured experimentally and/or estimated from available correlations. Likewise, a 1D heat balance is enough to model a nd cont rol t he temperat ure profile. Importantly, this transition from laminar to turbulent f low prevents the traditional approach to process scale-up based on matching dimensionless numbers. It is precisely in this situation where pre- dictive process models are most useful; under such conditions, reliable scale-up can be achieved using models that can accurately represent the behavior of the manufacturing process at all the relevant scales, where the small-scale experimental system is used to rapidly and conveniently select the appropriate reaction pathway, and the model is then used to pre- dict the performance of the scale-up system when implemented at the manufacturing scale. In such a situation, the dynamic system model, accurately predicting system performance at multiple scales, becomes the essential scaling-up tool. Module 3: Separation and purification Chromatographic separations (Figure 4) proceed by pulsating two f lows through a chromatographic packed bed. The first pulse contains the solution of the species to be separated, which have different affinities for the chromatographic packed bed, so they absorb and desorb at dif ferent rates. This is followed by a pulse of clean solvent. By capturing the eff luent stream at different times, the desired species can be separated from other materials in the incoming solution. A mathematical model of chromatographic separation has been developed. While chromatography is widely used at the small scale and even at the industrial scale, the general problem of isolat i ng a nd pu r i f y i ng a n A PI may require other separation operations (e.g., liquid–liq- uid separation, centrifugation, precipitation, crystal- lization, filtration, drying). ! '' ! ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( !"#$%&'()'*'+,&-".,&'/"01&20"2#34&56"2#37"-"2#'78/$,&'5.,&'68'7"-'698' %&5:65260'"2'52;'/&0"%&/'1%818%6"82'52/'/&,"<&%'64&7'50'5'4878#&2&8$0' 06%&57'4&56&/'68'5'/&0"%&/'6&71&%56$%&=' !"#$%&' (( !"#$%&' )( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( !"#$%&'()'*'+,&-".,&'/"01&20"2#34&56"2#37"-"2#'78/$,&'5.,&'68'7"-'698' %&5:65260'"2'52;'/&0"%&/'1%818%6"82'52/'/&,"<&%'64&7'50'5'4878#&2&8$0' 06%&57'4&56&/'68'5'/&0"%&/'6&71&%56$%&=' !"#$%&' (( !"#$%&' )( ( ( ( ( ( ( ( Figure 4: schematic representation of the chromatographic separation module ! "#$% & ' ( ( ( Figure 3: Schematic representation of the PFR reactor module displaying the monitoring of conversion and temperature. FIGURE 3. Schematic representation of the plug flow reactor module displaying the monitoring of conversion and temperature. A systematic framework of methods and tools are needed to develop, adapt, and evaluate the flexible modular CM plant for API manufacturing with reduced time and resources.