Background:
Crystallisation is a widely used processes for separation, purification and particle formation, playing an essential role in a large range of everyday products. It involves a large number of simultaneous physical chemical phenomena: mixing, mass and heat transfer, nucleation, growth, supersaturation and dissolution. While some are easily controlled, controlling nucleation remains a challenge of both academic and industrial/commercial interest, as it impacts directly on industrial processes and outputs.
The use of microfluidics has proven advantageous for the study of crystallization towards the paradigm shift needed in thinking from equipment-based towards physical-chemical predictive design. Droplet-based microfluidic experiments are often preferred over bulk experiments as it is less time-consuming, spends less materials, and enables collection of large datasets with improved statistics to investigate and screen crystallization kinetics. Another reason why microfluidics is a popular method to study crystallization kinetics is the increased mixing and heat exchange properties(1). Many microfluidic setups have been developed and are reported in literature to study the nucleation step of crystallization(2).
While nucleation is the least controlled step in crystallization, it is commonly avoided with the addition of seeds to a supersaturated solution before nucleation takes place. Seeding allows the prediction of the output particle sizes and to obtain the right crystal form, reaching the process requirements. Yet, when seeding one compound in a multicomponent solution, where more than one compound can crystallize, interactions between both growing crystals cause polycrystalline particles of mixed composition (epitaxial growth), enhanced hopper’s growth and increase fluid inclusions. Despite them being desired or not, these interactions are not thoroughly documented in literature, mainly due to experimental limitations. Some seeded microfluidic designs were recently reported, applied to polymorph screening. Sultana & Jensen studied the kinetics and the influence of impurities on the crystallization of different glycine polymorphs in a continuous microfluidic device (3). Garg et al. (4) used a droplet-based approach to study the crystallization of abacavir hemi-sulphate, using a separated inlet for the seeds suspended in glycerol, due to increased viscosity, and other for the supersaturated solutions. Both devices were developed using antisolvent crystallization.

The project:
In this sense, it is proposed to develop a seeded droplet-based microfluidic device to study crystallization. This device would enable the study of the growth kinetics of seeded crystals and the interaction between crystals during the simultaneous crystallization of more than one compound, avoiding the influence of other crystallization phenomena such as secondary nucleation and agglomeration. Using droplet-based microfluidics will allow statistically reliable data. This project will involve the use of basic crystallization concepts, design of the device (using key software as, e.g., SolidWorks), image analysis and, possibly, the implementation of automated control strategies using Arduino protoboards.

References:

1. H. H. Shi et al., Lab Chip. 17, 2167–2185 (2017).
2. N. Candoni, R. Grossier, M. Lagaize, S. Veesler, Annu. Rev. Chem. Biomol. Eng. 10, 59–83 (2019).
3. M. Sultana, K. F. Jensen, Cryst. Growth Des. 12, 6260–6266 (2012).
4. N. Garg et al., Lab Chip. 20, 1815–1826 (2020).