\\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Title-R1d.png PRACTICAL CONSIDERATIONS DIFFERENCES BETWEEN LABORATORY AND LARGER SCALE PROCESSES Petr Beňovský \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Title-R1d.png DIFFERENCES BETWEEN ACADEMIC AND PROCESS CHEMISTRIES Academic – Discovers, reveals, disputes, confirms, brings new knowledge. Small amount of material. Process – Selects, optimizes, seeks for efficiency, defines control points, considers efficiency and environment (also safety). Role of chemical engineers. Relatively large amount of material. \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS Laboratory (medicinal) chemistry (mg – g) has to be diverse and flexible; chromatography is very common; Can be done by a synthetic chemist only; Process (up-scaled) pathway (kg – 1000 kg) must provide reliable results, the procedure is expected to be robust, repeatable, simple, economic, focuses on safety (both operators and patients); chromatography is to be avoided; Must be done in mutual cooperation of a synthetic chemist (thinks in steps) and a chemical engineer (thinks in unit operations) \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS What is scale-up? Transferring a lab-scale chemical process to pilot or commercial equipment with: • same yield • same selectivity • same quality Scale-up is NOT a simple linear increase in geometric dimensions \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS What is scale-up? MORE • understanding of critical parameters • • how to control them • • ability to predict performance at any scale • • \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS • Laird, T. – How to Minimise Scale Up Difficulties 1.Appropriate conditions 2. 2.Correct dosing time 3. 3.Hazards 4. 4.Mass transfer issues 5. 5.Solvent extractions 6. 6.Optimising using statistical methods 7. Laird, T. Chemical Industry Digest, p. 51, July 2010 \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS • The best way to minimize the scale-up problems is by important data gathering and detailed process understanding. \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Ideally, dimensions in geometry, velocities of the components, forces on the system, temperatures and concentrations should be kept constant between different scales •Surface area per volume ratio (serious consequences for heat removal and heat input during process scale-up); •Kinematic similarity – they exists when two systems have the same shape and the ratios of the velocities between corresponding places are also equal; Fluid dynamics – the Reynolds number (Re) – it increases during scale-up at a constant stirrer speed as the diameter of the stirrer increases; •Hydrodynamic similarity – they exists when the ratios of forces between corresponding places are also equal in both systems; \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS •Thermal similarity – temperature differences between corresponding places in a system have a constant ratio with one another (temperature profile, heat transfer area); • •Chemical similarity – concentration differences between corresponding places in two systems have a constant ratio to one another (ratio between the chemical conversion rate, rate of molecular diffusion). \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Reactor Size [L] Surface Area [m2] Surface Area / Volume [m2/L] Factor Lab Scale 0.5 0.02 0.04 28.6 10 0.20 0.02 14.3 Pilot Plant Scale 380 2.32 0.0061 4.4 Large Production 38 000 53.0 0.0014 1 Maintaining geometrical similarities for various scales is not practical in batch processing as the jacket heat-exchange area per unit reduces significantly with a scale \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Reynolds number - gives a measure of the degree of turbulence or the ratio of inertial force to viscous force (the higher Reynolds number the higher turbulence of the system) Re = ND2 ρ/μ (mostly for homogeneous systems) Re … Reynolds number N … rotational speed (revolutions per second) D … diameter of the stirrer (m) μ/ρ … kinematic viscosity (m2/s) \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Maintaining total similarity of all possible scale-up parameters on different scales cannot be established and, in fact, is almost impossible; A reliable batch process scale-up cannot be simulated in generally applicable mathematical models without a clear understanding of all process and reaction mechanisms; Regime analysis – significance/trade-off of particular similarities – can be done in an early stage of process development; \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Regime analysis can be done in an early stage of the process development: •If heat effects are relative small, then the thermal similarity will be easily maintained; •If reaction rate is slow compared to the mixing time, a turbulent regime is not that relevant anymore; •For very rapid reactions any limitation in diffusion might be the rate-controlling step and the chemical reaction is a subject to a hydrodynamic regime and the energy input should get priority; •if in a heterogeneous reaction the particle size and, therefore, the dissolution rate is an important process parameter, the chemical regime might dictate a stirrer rate on large scale where all particles are free from bottom of the reactor; \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Stirrer rate and diameter of the stirrer are important parameters to play with in the early stages of process development and that in any realistic process scale-up the larger scale-reactor will always represent higher tip speed and longer circulating and mixing times than on the smaller scale. For heterogeneous processes this fact might have serious consequences. \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Widely used scale-up rule is the equal power per unit of volume criterion and has given accurate results in many cases. This rule has been concluded to be the best in almost any scale-up problem. For non-laminar flow (high Reynolds number), with constant geometry and the same stirrer type P/V = P0 x r x N3 x D2 = constant P0 … power number of the stirrer r … density N … stirrer speed D … diameter of the stirrer \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png DIMENSIONAL ANALYSIS Parameter Power P/V Q/V Tip Speed Reynolds number Equal P 1.0 10-3 0.0215 0.215 2.15 Equal P/V 103 1.0 0.215 2.15 21.5 Equal N 105 102 1.0 10 102 Equal Tip Speed 102 0.1 0.1 1.0 10 Equal Reynolds numner 0.1 10-4 10-2 0.1 1.0 Table represents effects of various scale-up strategies from 1 L to 1000 L Q/V … the liquid pumping capacity of the stirrer per volume \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS Why is scale-up so difficult? • There are no standard approaches for doing quantitative process scale-up; • Textbooks on scale-up are limited; • Scale-up practice largely depends upon individual experience; • There is a shortage of people with the right experience; •The success of process scale-up depends to a great extent on the communication and transfer of information between the chemists and the chemical engineers; •There are no systematic ways for a chemical engineer to ask a chemist what information is required for process scale-up and vice versa; •Companies and chemical engineering community are not learning from the success and failures that are occuring on a daily basis throughout the industry; •There is a gap between how chemical engineers and chemists want the process to run in the plant and how the operators actually run it, due to lack of training or involvement. \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS Process development should be defined as the process of converting a synthetic route into an optimum, robust, safe and economic process for manufacturing the chemical of desired quality at the desired ultimate scale within a reasonably desired period of time; \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS •SAFETY • •TEMPERATURE CONTROL • •TEMPERATURE RANGE • •MOBILE (TRANSFERABLE) STREAMS • •INCREASE EFFECTIVITY (MINIMIZE SOLVENTS, INCREASE CONCENTRATION WHERE POSSIBLE) • •STABILITY OF COMPONENTS DURING REACTION AND HOLD ONS • •SIMULATE LARGE SCALE CONDITIONS IN LABORATORY (prolonged additions, heat accumulation, stability of reactants and products, etc.) • •GET INFORMATION ABOUT PROPERTIES (solubilities, pH tolerance, …) • \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png BASIC CONSIDERATIONS •DETERMINE CONTROL POINTS (in process controls) • •KEEP IT SIMPLE • •ANTICIPATE FATE OF VOLATILE REAGENTS • •DEVELOP EFFICIENT AND STRAIGHTFORWARD WORK UP PROCEDURE • •CONSIDER INERT ATMOSPHERE TO AVOID THE PRESENCE OF MOISTURE AND OXYGEN • •ASSUME SCRUBBING FOR ANNOYING OR TOXIC OFF-GASES • •SUGGEST RESISTANT MATERIAL \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png CHARGING •Weighing of reagents (differential, reactors mounted on a load cell); • •Charging of liquids (by weight or by volume) – use the same approach in the laboratory – density of liquids will change slightly with temperature; • •Recommended accuracy (tolerance) - volumes + 5%, weights + 2%; • •Different transfer times considering a laboratory scale and the production (large) scale; • \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png SOLVENT CONSIDERATIONS Watch out hydrocarbon solvents with even number of carbons (toxicity, electrostatic buildup); Classification of solvents – ICH Harmonised Guideline Q3C – Impurities: Guideline for Residual Solvents Class 1 – solvents to be avoided (known human carcinogens, strongly suspected human carcinogens, and/or environmental hazards, e.g. carbon tetrachloride (concentration limit 4 ppm), 1,2-dichloroethane (5 ppm), 1,1,1-trichloroethane (1500 ppm), benzene (2 ppm)) Class 2 – solvents to be limited (non-genotoxic animal carcinogens, agents of irreversible toxicity, e.g. acetonitrile (410 ppm), chlorobenzene (360 ppm), chloroform (60 ppm), N,N-dimethylformamide (880 ppm), hexane (290 ppm), methanol (3000 ppm), N-methylpyrrolidone (530 ppm), toluene (890 ppm)) Class 3 – solvents with low toxic potential (permissible daily exposure 50 mg or more per day, e.g. acetic acid, acetone, ethyl acetate, heptane, 2-propanol, triethylamine) Solvents for which no adequate toxicological data was found – a manufacturer is asked to supply justification for residual levels of these solvents (e.g. diisopropyl ether, petroleum ether, trifluoroacetic acid) \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP IN PROCESS CONTROLS (IPCs) Off-line analysis In-line analysis On-line analysis WORK UP Efficiency – e.g. crystallization directly from the reaction mixture; – labor cost is very important; Extractions are generally preferred over filtration to remove impurities; Column chromatography is rare and very expensive; \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP •Includes operations after the reaction was declared complete; •Such operations include quenching the reaction both to remove impurities and facilitate product isolation and to allow safe handling of process streams, even after product isolation. • •Quenching reactive species •pH adjustment •Filtration •Precipitation •Extractions •Concentration (including azeotropic distillation) •(Chromatography) \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP Typical time and money saving technique TELESCOPING The reaction proceeds further without full isolation of an intermediate, with advantage even without any quench; Pushing a reaction to completion – removal of side products \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP Gallou, F. et al J. Org. Chem. 70, 6960 (2005) \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP QUENCHING REACTIONS Safe decomposition of excessive reagents stops a reaction \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP QUENCHING REACTIONS Careful with halogenated solvents (e.g. dichloromethane) in the presence of azides (diazidomethane !!) VERY CAREFUL Recent issue with the formation of nitrosamines (e.g. limit in Valsartan (320 mg dose) will be 0.03 ppm in 2021 \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP EXTRACTIONS •Used to separate neutral compounds from water soluble components; •Solid Phase Extraction (SPE) – separate compounds of significant different polarity; •Solubility or miscibility of organic solvents with water; •Separation of layers •pH value adjustment – extra opportunities; •Ionic strength; •Solubility at higher temperatures; • \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP EXTRACTIONS Zegar, S. et al Org. Process Res. Dev. 11, 747 (2007) \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP EXTRACTIONS Convenient Aqueous Solutions for Extractions Solvent pH of 0.1 N solution Relative Solubility in Organic Solvents Comments HCl 1.1 High Corrosive, volatile H2SO4 1.2 Low AcOH 2.9 High Weak acid Na2HPO4 8.5 Low NaHCO3 8.4 Low NH3 11.1 Moderate Volatile Na2CO3 11.6 Low Na3PO4 12.0 Low \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP FILTRATION Polish Filtration – an operation to remove trace amounts of insoluble impurities before other operations – passing a process stream through in-line filters with different porosity; Very important for crystallizations, avoiding emulsions; Ultrafiltration – protein separation through membranes; \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP PERVAPORATION Pervaporation through membranes – specific for some solvents – a processing method for the separation of the mixtures of liquids by partial vaporization through a nonporous or porous membranes; Separation of components is based on a difference on a transport rate of individual components through the membrane; Pervaporation is effective for solutions containing traces or minor amounts of the component to be removed; Hydrophilic membranes for dehydration of alcohols containing small amount of water, hydrophobic membranes for removal of traces of organic compounds from aqueous solutions; \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP PERVAPORATION Hydrophilic membranes – commercially most successful membranes are formed from polyvinyl alcohol or polyimides; Hydrophobic membranes – based on polydimethylsiloxane PRINCIPAL The pressure difference on sides of a membrane (usually atmospheric vs. vacuum), permeate goes through a membrane, retentate does not go through and thus it is separated. \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP PERVAPORATION \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png WORK UP CHROMATOGRAPHY Best to avoid, technically difficult and expensive on large-scale, but still used in special cases (preparative chromatography); Solid Phase Extraction Simulated Moving Bed Chromatography https://www.youtube.com/watch?v=Harx2khTuEc \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png EFFICIENT PROCESS DEVELOPMENT •Anticipate and avoid problems • •Do experiments at minimum and maximum ranges to confirm robustness/sensitivity in cases where a particular parameter is significant • •Identify critical impurities within the whole process and their fate • •Get maximum allowed level of critical impurities • •How to proceed if the specification criteria in IPCs are not met? • •Pay attention to details, observe unusual changes • •Avoid systematic errors • •Take into account future process validation \\DROBO-FS\QuickDrops\JB\PPTX NG\Droplets\LightingOverlay.png Droplets-HD-Content-R1d.png PROCESS VALIDATION •The cumulative effort to demonstrate reliable processing and product quality; • •The fruition of the labor of process chemists and engineers, the ultimate tests of how well one understands the process; • •Before 1970s little attention has been paid to efficient process development; • •1987 – FDA – Guideline on General Principles of Process Validation (validation is defined as establishing documented evidence which provides a high degree of assurance that a specific process will consistently produce a product meeting its predetermined specifications and quality attributes); • •2008 – FDA – process validation is the collection and evaluation of data, from the process design stage throughout production, which establishes scientific evidence that a process is capable of consistently delivering quality products; the quality is built up into the product through process understanding and cannot be tested in batches – quality by design (QbD) Anderson, N.G. et al Org.Process Res.Dev. 15, 162 (2011)