Non-photochemical laser-induced nucleation (NPLIN): an experimental investigation via real-time imaging and product analysis
Item statusRestricted Access
Embargo end date07/06/2023
Barber, Eleanor Rose
Non-photochemical laser-induced nucleation (NPLIN) is a technique in which exposure of a metastable system (such as a supersaturated solution) to a pulse of unfocused laser light induces the formation of a new phase (usually crystalline). The mechanism is thought to depend on the heating of impurity nanoparticles by the laser, resulting in the formation of cavities or bubbles which act as nucleation sites. The control, both temporal and spatial, that NPLIN provides over nucleation means that this is a promising technique for the study of nucleation dynamics and crystal growth. Additionally, NPLIN appears to offer control over the polymorph of nucleated crystals; an understanding of such selectivity would be of great benefit for applications in industrial crystallisation, especially, for example, in the pharmaceutical industry. In this work, both of these aspects of control were utilised to provide new insights into the mechanism of NPLIN. Aqueous solutions of various solutes, namely cesium chloride, sodium bromide, sodium acetate, potassium chloride, and sodium chlorate, were chosen for experiments to investigate specific aspects of crystal nucleation and growth, as detailed in the following paragraphs. The temporal and spatial control provided by NPLIN was used to image the growth of cesium chloride crystals from supersaturated aqueous solutions (S = 1.05–1.18) in real time. Crystals were fast-growing and their number showed a non-linear dependence on laser pulse energy. Following the sedimentation of crystals, a convection current was set up in which crystals circled vertically within the sample vial; this type of cyclic convection phenomenon has, to the best of our knowledge, never been studied before and shows the effect of fluid flow around rapidly growing crystals. It was found that crystals nucleated preferentially on the glass walls of the sample vials, rather than in the bulk solution. This behaviour was investigated in samples following various treatments: aging of samples (while hot or cold), filtration, cleaning of vials and ultrasonication of samples. Sample aging increased the proportion of crystals forming on the vial walls, while ultrasonication reduced the proportion of crystals forming on the walls but increased the total number of crystals. When the same sample solution was exposed multiple times to the laser, crystals repeatedly formed in the same locations on the walls. The adhesion of nanoparticles, which act as seeds for nucleation, to the vial walls can explain these observations. In order to make repeat analysis of the same solution, a continuous flow system was developed which produced a supply of supersaturated solution in a capillary tube. Temperature and flow rate measurements were carried out to determine the optimal setup for delivery of cesium chloride solution. NPLIN was successfully carried out within the flowing solution, however, due to the high flow rates required this system may be best suited for other applications and was not used for further experiments. In order to investigate the events that occur in the solution immediately after the laser pulse, high-speed imaging of NPLIN in cesium chloride solutions was carried out with a maximum frame rate of 250,000 frames per second. Objects attributed to thermocavitation bubbles were observed in the solution within 4 µs of irradiation. This is the first time that cavitation bubbles have been directly observed during NPLIN using a unfocused laser pulse. Crystals were later observed to form in the same location as these transient bubbles. Much larger bubbles (maximum diameter 0.5 mm) were also seen forming in the mineral oil in which the sample was immersed as a part of the optical setup. These were also attributed to cavitation and oscillation of the bubble size was observed. NPLIN was also investigated in two salt hydrate systems: aqueous sodium bromide (S = 1.02–1.29) and aqueous sodium acetate (S = 2.72). Here, the focus was on the identity of the crystalline product: for both salts, NPLIN resulted in the formation of unstable anhydrous crystals as opposed to the stable hydrates. In sodium bromide solutions, nucleation was also carried out via seeding, mechanical shock, and ultrasonication, which showed a preference for sodium bromide dihydrate formation; and laser trapping, which selectively formed anhydrous sodium bromide. These results suggest a mechanism for NPLIN based on a hot cavitation event, where crystals form in a dehydrated region and are therefore anhydrous, which is different to the cavitation events that occur during mechanical shock and ultrasonication. In sodium acetate solutions, the addition of the sodium salt of poly(methacrylic acid) (Na-PMAA) (0.25 and 0.73 wt%), which is known to inhibit nucleation of anhydrous sodium acetate, resulted in the formation of bubbles instead of or alongside crystals. The numbers of crystals and bubbles that formed were similar. The number of crystals showed a linear dependence on laser pulse power density before reaching a plateau, and the dependence of the number of bubbles on laser pulse power density was similar to that of crystals at low power densities. This is the first time that numbers of crystals and bubbles have been counted within the same solute/solvent system. The results support particle heating followed by cavity or bubble production as the origin of both crystal and bubble formation. In supersaturated aqueous potassium chloride samples (S = 1.04–1.06), crystals were observed growing on the meniscus of the solution following NPLIN. Imaging of crystal growth was carried out but the source of these crystals could not be identified. When the laser pulse was directed vertically through the solution, passing through the air–solution interface, crystals grew on the meniscus in 100% of samples that nucleated, showing that this is a highly favourable region for nucleation. On the addition of a layer of silicone oil to the surface, this number was reduced to 50%. These results suggest that nucleation at the meniscus is favoured due to evaporation at the interface, which is prevented by the addition of silicone oil. Finally, laser-induced nucleation (LIN) using a focused laser pulse was investigated in sodium chlorate samples, which do not nucleate when exposed to unfocused pulses. Although a violent optical breakdown was observed at the focal volume of the laser, only a few crystals were observed to nucleate in each sample. This demonstrates the usefulness of LIN using a single high-energy pulse for nucleation in systems which do not undergo NPLIN, which could be used for the growth of single crystals for analysis. The effect of right circularly polarized (RCP) and left circularly polarized (LCP) light on the formation of dextrorotatory (d) and levorotatory (l) sodium chlorate crystals was investigated and no correlation was observed. This is consistent with a mechanism involving nucleation of the achiral monoclinic phase III followed by a solid-solid phase transition to the chiral cubic phase I of sodium chlorate. The formation of phase III requires higher supersaturations than the direct formation of phase I, so this result could explain why aqueous sodium chlorate does not nucleate via NPLIN, which will cause lower energy cavitation events than the focused laser. Overall, the results presented here, from both real-time observation of crystal growth and analysis of the resulting crystals, show new evidence in favour of the particle heating mechanism for NPLIN and highlight the usefulness of NPLIN for both the study of crystal growth and the nucleation of desired crystal phases.