(344g) Reaching Optimal Productivity in Batch Viedma Ripening Processes Using Minimal Data

In 2004 Viedma reported that nearly racemic mixtures of L-sodium chlorate and D-sodium chlorate undergo a deracemization process in the presence of an attrition source, i.e., the initially racemic mixture reaches chiral purity within several days of simply stirring the suspension^[1]. This discovery, subsequently termed Viedma ripening (VR), was subsequently shown to also be applicable to organic molecules that exhibit a conglomerate phase diagram. It was successfully applied to a host of different molecules, among them amino acids^[2,3] and pharmaceuticals^[4]. VR thus represents a new and promising way to arrive at enantiomerically pure products.

One of the key observations made in early studies on VR was that the rate towards enantiomeric purity proceeds in an autocatalytic fashion, i.e., starting from a low enantiomeric excess there is considerable lag time before the conversion rate speeds up. The rate is finally slowing down again towards the end of the process.

The overall Viedma ripening process was later identified to consist of a complex interplay between racemization in solution, enantioselective agglomeration, attrition/breakage, and size-dependent solubility/Ostwald ripening.^[5,6] Being a rather slow process, attempts at optimizing deracemization processes focused on shortening the time needed to reach enantiomeric purity and hence to increase productivity.^[7,8] However, no study so far has attempted a thorough optimization of process conditions with the goal of maximizing productivity.

The incentive to optimize a process towards optimal productivity and shortened batch times is evident, but the flexibility to exchange equipment on existing plants is limited. Consequently, the focus of process optimization should be shifted towards the remaining process conditions, such as amounts of seeds (and enantiomeric purity of seeds), seed distribution, and suspension density. Hence, the key objective of this contribution is to introduce a methodology that achieves optimal productivity (maximizes mass of pure enantiomer produced per time) for batch Viedma ripening processes with the use of as little experimental data as possible.

The methodology presented here seeks to optimize a batch-after-batch production of enantiopure crystals. Within such a processing strategy, the current batch can be conveniently run by re-using the mother liquor from the previous batch (minus appropriate purging to avoid impurity build up), adding a defined amount of a racemic mixture of solids and by seeding the process with some enantiomerically pure seeds (part of the product from the previous batch). Due to VR, such a batch will reach enantiopurity after some time. The goal of this methodology is to keep this time as short as possible. It can be shown that such a processing strategy is mathematically equivalent to an ideal continuous plug flow reactor with a recycle stream consisting of part of its effluent stream. Since Viedma ripening exhibits autocatalytic features, this analogy can be used to find an optimal seeding policy for batch Viedma ripening processes. It will be shown that the only data needed for this strategy is a curve of enantiomeric excess versus time, obtained from a single experiment by standard techniques (e.g. HPLC). It will further be shown that an optimal suspension density can be established without additional data.

The strategy is first shown to work based on extensive model results employing population balance equation models and various kinetic expressions (to ensure a wide range of applicability) and then exemplified using an experimental study on Naproxenethylester crystallized from ethanol, which is racemized in solution using a base (sodium ethoxide in this case).

References:

[1] Viedma, "Chiral symmetry breaking during crystallization: complete chiral purity induced by nonlinear autocatalysis and recycling", Phys. Rev. Lett., 2005, 94, 065504

[2] Viedma et al., "Evolution of Solid Phase Homochirality for a Proteinogenic Amino Acid", J. Am. Chem. Soc., 2008, 130, 15274â??15275.

[3] Noorduin et al., "Emergence of a Single Solid Chiral State from a Nearly Racemic Amino Acid Derivative", J. Am. Chem. Soc., 2008, 130, 1158â??1159.

[4] Noorduin et al., "Fast attritionâ?enhanced deracemization of naproxen by a gradual in situ feed", Angew. Chem. Int. Ed., 2009, 48, 4581â??4583.

[5] Iggland and Mazzotti, "A population balance model for chiral resolution via Viedma ripening", Cryst. Growth Des., 2011, 11, 4611â??4622.

[6] Gherase et al., "Experimental and Theoretical Study of the Emergence of Single Chirality in Attrition-Enhanced Deracemization", Cryst. Growth Des., 2014, 14, 928â??937.

[7] Iggland et al., "Complete solid state deracemization by High Pressure Homogenization", Chem. Eng. Sci., 2014, 111, 106-111.

[8] Noorduin et al., "Scaling Up Attrition-Enhanced Deracemization by Use of an Industrial Bead Mill in a Route to Clopidogrel (Plavix)", Org. Process Res. Devel., 2010, 14, 908-911.