Recent theoretical and experimental studies have proposed a two-step mechanism for crystal formation in which crystal nucleation is preceded by formation of disordered molecular assemblies. Here, we investigated whether similar intermediates might also form as crystals dissolve, effectively the reverse process. A model system of glycine in water was studied, and the resultant solutions were characterized using small-angle X-ray scattering, dynamic light scattering, and nanoparticle tracking analysis. Invariably, dissolution of glycine crystals into water was observed to produce scattering nanospecies with liquid-like properties and a mean diameter of about 250 nm, at near saturation concentration. The nanospecies persisted indefinitely in solution in the presence of excess glycine crystals and were identified as glycine-rich nanodroplets with an equilibrium population of about 10 9 per mL. The time to appearance of glycine crystals from quiescent supersaturated solution (S = 1.1) containing either a low population of nanodroplets (nanofiltered) or a high population of nanodroplets (unfiltered) was indistinguishable with typically only a single crystal forming after about 30 h. However, a very significant acceleration of crystal formation was observed whenever a gently tumbling stirrer-bar was introduced into the vial; thousands of microcrystals appeared after an incubation period of only 3−5 h. The possibility of this being caused by factors such as secondary nucleation, bubbles, or glass splinters or scratches was eliminated via control experiments. Further investigation of the glycine solution, just prior to appearance of microcrystals, revealed an additional subpopulation of extremely large glycine-rich nanodroplets (diameter >750 nm), not observed in quiescent solutions. It is proposed that productive nucleation of glycine crystals occurs exclusively within these larger glycine-rich nanodroplets because a critical mass of glycine is required to form nascent crystals large enough to survive exposure to bulk more dilute solution. We hypothesize that nucleation occurs frequently but nonproductively within subcritical mass nanodroplets and infrequently but productively within very rare critical mass solute-rich nanodroplets. Such a model provides a new compelling way of bridging classical mechanisms of crystal nucleation with the more recently proposed two-step processes.

Aqueous solutions of highly soluble substances such as small amino acids are usually assumed to be essentially homogenous systems with some degree of short range local structuring due to specific interactions on the sub-nanometre scale (e.g. molecular clusters, hydration shells), usually not exceeding several solute molecules. However, recent theoretical and experimental studies have indicated the presence of much larger supramolecular assemblies or mesospecies in solutions of small organic and inorganic molecules as well as proteins. We investigated both supersaturated and undersaturated aqueous solutions of two simple amino acids (glycine and DL-alanine) using Dynamic Light Scattering (DLS), Brownian Microscopy/Nanoparticles Tracking Analysis (NTA) and Cryogenic Transmission Electron Microscopy (Cryo-TEM). Colloidal scale mesospecies (nanodroplets) were previously reported in supersaturated solutions of these amino acids and were implicated as intermediate species on non-classical crystallization pathways. Surprisingly, we have found that the mesospecies are also present in significant numbers in undersaturated solutions even when the solute concentration is well below the solid-liquid equilibrium concentration (saturation limit). Thus, mesopecies can be observed with mean diameters ranging from 100 to 300 nm and a size distribution that broadens towards larger size with increasing solute concentration. We note that the mesospecies are not a separate phase and the system is better described as a thermodynamically stable mesostructured liquid containing solute-rich domains dispersed within bulk solute solution. At a given temperature, solute molecules in such a mesostructured liquid phase are subject to equilibrium distribution between solute-rich mesospecies and the surrounding bulk solution.

In this work DLS, NTA, SAXS and NMR were used to investigate populations, size distributions and structure of clusters in undersaturated aqueous solutions of glycine. Molecular and colloidal scale (mesoscale) clusters with radii around 0.3-0.5 nm and 100-150 nm, respectively, were observed using complementary experimental techniques. Molecular clusters are consistent with hydrated glycine dimers present in equilibrium with glycine monomers in aqueous solutions. Mesoscale clusters previously observed in supersaturated glycine solutions appear to be indefinitely stable, in mutual equilibrium within mesostructured undersaturated solutions across all glycine concentrations investigated here, down to as low as 1 mg/g of water.

Efficient processing of particulate products across various manufacturing steps requires that particles possess desired attributes such as size and shape. Controlling the particle production process to obtain required attributes will be greatly facilitated using robust algorithms providing the size and shape information of the particles from in situ measurements. However, obtaining particle size and shape information in situ during manufacturing has been a big challenge. This is because the problem of estimating particle size and shape (aspect ratio) from signals provided by in-line measuring tools is often ill posed, and therefore it calls for appropriate constraints to be imposed on the problem. One way to constrain uncertainty in estimation of particle size and shape from in-line measurements is to combine data from different measurements such as chord length distribution (CLD) and imaging. This paper presents two different methods for combining imaging and CLD data obtained with in-line tools in order to get reliable estimates of particle size distribution and aspect ratio, where the imaging data is used to constrain the search space for an aspect ratio from the CLD data