(238f) Effect of Agitation during Cooling Crystallization on the Unique Regeneration Phenomenon Observed in Paracetamol Crystals Post Breakage

Regeneration is a common phenomenon observed in nature, where fragmented parts from an organism’s body regrow thus making the organism whole again, e.g. a lizard’s tail or starfish’s fragments. But would this same behavior be fathomed for crystals? Although a non-living entity, throughout literature crystals are often described as having a ‘life-cycle’ with three basic stages: nucleation (birth), growth and breakage (death). But the behavior of crystals after breakage i.e. their death is a grey area. Do they really ‘die’ and remain unchanged? do they dissolve? Or do they regrow? Our work has reported for the first time an unconventional ‘regeneration’ phenomenon in paracetamol crystals following breakage along the cleavage plane (010)(1). The broken part was observed to regrow on the parent crystal, restoring the pre-breakage shape prior to demonstrating overall growth, thus suggesting a ‘re-birth’ phase in the crystal life cycle.

This regeneration phenomenon fuels confusion as the cleavage plane which theoretically has the lowest attachment energy (2) and therefore expected to grow the slowest, shows rapid growth compared to all other facets. It becomes extremely crucial to study eccentricities in growth such as the one at hand in order to increase understanding of crystal behavior post breakage and therefore increase the capabilities of process modelling tools for superior prediction of crystal properties. Moreover, elucidating the regrowth behavior also provides pivotal knowledge on the performance of further downstream processes such as filtering, drying and milling. Holistically, given that breakage is a common occurrence in industrial crystallisers especially along cleavage planes since they are the weakest planes in the crystal lattice, unravelling the growth after cleavage can take us steps closer at understanding the ever perplexing crystal structure.

Previous work predominantly used evaporative crystallisation to provide a proof of concept of the regeneration behavior in macroscopic crystals as longer time scales allowed the process to be elongated and therefore be captured clearly using various imaging techniques. This work is an extension that tries to elucidate the effect of operating conditions that are unavoidably used in industrial crystallisation to better understand the applicability of the phenomenon on larger scales – the effect of cooling crystallisation in the presence and absence of agitation has been studied.

Single macroscopic crystals of paracetamol were grown from an ethanolic solution using slow evaporation and broken along the cleavage plane using a scalpel – techniques similar to that reported previously. The cleaved crystal was attached to a borosilicate glass rod using household superglue as it demonstrated to be insoluble in the ethanolic solution after screening multiple binding resins (Figure 1(a)). This process allowed the crystal to be immobilised in solution to prevent it from moving around during agitation. Cleaved crystals were regrown by implementing a cooling ramp whereby a cooling rate of 0.1°C/min was chosen for which the Metastable Zone Width was determined to select the optimum supersaturations to implement. The MSZW was measured by cooling known concentrations of solutions at 0.1°C/min and 200rpm until the first hints of cloudiness was detected using imaging (Figure 2). The immobilised cleaved crystal was inserted in a 30ml vial containing a saturated solution of paracetamol-ethanol at 30°C which was cooled to a supersaturation of 1.1 at a cooling rate of 0.1°C/min, without agitation as well as with agitation using a magnetic stirrer at 200 rpm. The in-house macrophotography rig developed was used for crystal imaging and obtaining growth data (Figure 1(b)). Growth rates of the regenerating crystal will be calculated by measuring the propagation of two characteristic lengths one parallel (length ) and one perpendicular to the cleavage plane (length ).

Comparing the results from evaporative crystallisation to cooling crystallisation without agitation, the latter resulted in the process being 85% quicker than the former for crystals in the similar size range (7-9mm) (Figure 1(c)). While evaporation resulted in the crystal regenerating in around 6 days, the cooling crystallisation completed the process in around 25 hours. This time was further shortened by agitating the solution at 200 rpm on a magnetic stirrer that led to full crystal regeneration in 10 hours, almost 70% faster (Figure 3). The exact growth rates of the agitated vs. the non-agitated system are currently still being investigated however it is predicted that the length propagation of will be atleast 50% faster for the crystal in stirred conditions as it completes regeneration in a shorter period of time.

Although it is intuitive that cooling crystallisation would result in faster regeneration due to a quicker generation of supersaturation that is also higher than that during a constant room temperature evaporation, this has not been reported before in the context of crystal regernation. Moreover, agitation and stirring have shown to enhance crystallisation in literature (3), and it is interesting to see that similar growth rate enhancement is seen during regeneration. This shows that regeneration in the absence of any forced convection of the fluid was not due to the accumulation of solute molecules and depletion of solvent molecules around the cleavage plane. Because even with constant ‘refreshment’ of solution around the cleavage plane due to fluid agitation, rapid growth of the plane was observed. The faster rate than the non-agitated system was postulated to be due to the constant fluid motion which resulted in enhanced mass transfer that allowed solute molecule to deposit on the surface continuously. It is also possible that the fluid shear resulted in crystalline particles being brushed off from the crystal surface that resulted in secondary nucleation on the cleavage plane resulting in faster growth. Finally, homogenous nucleation prior to the regeneration being completed was also observed in the agitated vial which was not seen in the unstirred vial, again reinforcing that stirring enhances the mass transfer and reduces the barrier to nucleation. Further studies are currently ongoing to understand the effect of various supersaturations, 1.05 and 1.01 as well as various agitation rates: 100 rpm, 300 rpm. These will help establish a better understanding of the effect of mass transfer on the crystal regeneration as well as provide a basis for designing experiments for batch scale regeneration.

Overall, it can be seen that agitating the solution during regeneration enhances the mass transfer and makes it possible to achieve regeneration in a 50% shorter time span than that of a non-agitated system. Although the work is in the preliminary stages and concrete conclusions cannot be made, the interesting results branching from the novel work of ‘crystal regeneration’ provide food for thought regarding breakage in crystallisers.

References

1. Bade I, Verma V, Rosbottom I, Heng JYY. Crystal regeneration – a unique growth phenomenon observed in organic crystals post breakage. Mater Horiz. 2023;10:1425-1430
2. SUN CC, KIANG YH. On the Identification of Slip Planes in Organic Crystals Based on Attachment Energy Calculation. J Pharm Sci. 2008;97(8):3456–61.
3. Li X, Heng JYY. The critical role of agitation in moving from preliminary screening results to reproducible batch protein crystallisation. Chemical Engineering Research and Design. 2021;173:81–8