Chair: Dr. Karen Robertson

10:15-10:45:  How Crystal Surfaces Shape the Behaviour of Molecular Materials

Crystallization does not only produce a crystal structure; it also defines which surfaces are exposed, how particles grow, and how the resulting material behaves. For molecular crystalline solids, small differences in packing, intermolecular interactions, and crystal habit can lead to large differences in performance during processing, storage, and use. This talk will explore how crystallographic information can be used to understand these effects at the molecular and surface level. I will discuss how crystal packing, interaction motifs, facet-specific energetics, surface topology, and predicted interaction patterns can be combined to interpret the behaviour of different solid forms. Rather than treating crystal structures as static endpoints, the focus will be on using them as a basis for understanding material properties. Examples from pharmaceutical and molecular materials will illustrate how structure-derived descriptors can help rationalize differences in stability, transformation pathways, dissolution behaviour, and particle performance. The talk will highlight crystal-surface analysis as a bridge between molecular crystallography, crystallization, and the practical behaviour of crystalline materials.

Meet the Speaker

Eliska Zmeskalova , FZU – Institute of Physics of the Czech Academy of Sciences

Ing. Eliška Zmeškalová, Ph.D., is a scientific researcher at the Institute of Physics of the Czech Academy of Sciences and at the University of Chemistry and Technology, Prague. She obtained her Ph.D. in Chemical Technology of Pharmaceuticals from UCT Prague in 2016, followed by postdoctoral research at University College Cork, Ireland. Since 2019, she has led the Laboratory of Single-Crystal Diffraction at the Institute of Physics of the Czech Academy of Sciences. Her research focuses on crystallographic and solid-state analysis of pharmaceutical molecular materials, especially structure–property relationships in polymorphs, salts, solvates, and cocrystals. She collaborates extensively with the pharmaceutical industry and was awarded the Otto Wichterle Prize in 2025.

10:45-11:05:  Dehydration of pharmaceutical hydrates

Active pharmaceutical ingredients (APIs) often encounter water/moisture during various stages of pharmaceutical manufacture and formulation, resulting in the formation of hydrate. In fact, crystalline hydrates are prevalent and nearly 50% of drug substances that can form hydrates are utilized in their hydrate forms. However, the manufacture of hydrate crystals presents its own challenge. It is crucial to avoid the loss of bound moisture when the crystals are exposed to heat, low humidity conditions or high-energy mechanical processing, e.g. during drying, tableting and storage. Therefore, a good understanding of the dehydration kinetics is required for the control of manufacturing process and storage conditions to maintain the integrity of the hydrate crystals. In this work, we investigated the dehydration of two hydrates, carbamazepine dihydrate (CBZD) and theophylline monohydrate (TPh) at different humidity and temperature conditions. The dehydration of CBZD mainly follows Avarami-Erofe’ev 2D or 3D nucleation and growth model but shifts to a diffusion-controlled mechanism at high humidity. CBZD consistently transforms to Form III when dehydration is diffusion controlled. However, when dehydration is nucleation-controlled, the polymorphic outcome depends on the specific dehydration conditions. Different dehydration pathways and mechanisms were proposed for CBZD based on the findings. TPh dehydration also follows predominantly the Avrami-Erofe’ev 2D or 3D nucleation and growth model. However, an exception was observed at 25 °C and 0% RH, where the data fit best to the F1 model (first-order kinetics). This behavior is likely due to the formation of a metastable anhydrate. PXRD analysis of TPh crystals stored at 0% RH and 25 °C confirmed the initial formation of this metastable form, which subsequently transformed into the stable anhydrate (Form II). However, the metastable anhydrate converted fully to stable Form II after 4 months of storage.

Meet the Speaker

Ann (Pui Shan) Chow , Institute of Sustainability for Chemicals, Energy and Environment, Agency for Science, Technology and Research, Singapore

Dr Ann Chow has over 25 years of research experience in crystallization science, modelling and control, formulation science and technology. She is currently the group leader of Sustainable Product Development Group and deputy director of Specialty Chemicals & Bio-Technologies Division in the Institute of Sustainability for Chemicals, Energy and Environment. Her research interests include crystallization fundamentals, crystallization process development, control and monitoring of crystallization process using process analytical technology (PAT) tools, and formulation development. She has led multi-disciplinary research projects in numerous collaborations with industry (e.g. GSK, MSD, Syngenta, AstraZeneca, P&G, Nestle) and overseas universities, including University of Illinois Urbana- Champaign (UIUC), Imperial College London and University of Manchester. She has published more than 130 journal papers and several books and book chapters.

11:05-11:25: Chord Length Distribution Inversion Strategy in Crystallization Monitoring Using Adaptive Tikhonov Regularization

Inline particle characterization via Focused Beam Reflectance Measurement (FBRM) generates chord length distributions (CLD) that serve as the primary real-time observable in crystallization process monitoring. Converting CLD data to crystal size distributions (CSD), which is eventually required by population balance model (PBM) for mechanistic prediction and control, constitutes an ill-posed inverse problem. Mathematically, this inversion is expressed as minimizing ∥Ax-b∥^2+λ∥Lx∥^2, where A is the transformation matrix, x is the model parameter to be estimated, b is the observed noisy data, L is the weight matrix. The regularization parameter λ critically governs the trade-off between fidelity to measured CLD data and solution smoothness. Although Tikhonov regularization has been used to stabilize CLD inversion, the selection of λ remains largely heuristic in crystallization applications, while standard strategies such as the Morozov Discrepancy Principle, the L-curve criterion, and Generalized Cross-Validation have mostly been discussed in the broader inverse-problems literature rather than tailored to FBRM-based monitoring. This study presents an adaptive regularization parameter selection strategies tailored to the CLD inversion problem in crystallization. Three candidate selection frameworks are systematically developed and compared: the Morozov Discrepancy Principle, the L-curve criterion, and Generalized Cross-Validation. The sensitivity of each method to FBRM measurement noise level, particle size distribution shape, and crystal aspect ratio is analyzed through simulation studies using CLD datasets. Ill-conditioning of the transformation matrix A is quantified through condition number analysis across particle size ranges relevant to crystallization. In addition, an online noise-estimation scheme tailored to provide CLD measurements is proposed and coupled with Morozov-based parameter selection to support real-time implementation. The study aims to identify a robust inversion strategy for stable CSD reconstruction under realistic measurement uncertainty and to provide a foundation for subsequent integration with population balance model-based monitoring and control.

Meet the Speaker

Abhishe Dutta , National Institute of Technology Calicut, Kerala, India

Abhishek Dutta is an Associate Professor of Chemical Engineering at İzmir Institute of Technology, Türkiye. His work combines process modelling, population balance modelling, CFD and Deep Learning techniques to address challenges in sustainable process engineering. He is particularly interested in the interface of crystallization science, process modelling, and artificial intelligence for improved understanding and control of crystal formation processes.

11:25-11:45: Simultaneous Polymorph Control and Uniform Spherulite Growth for Enhanced Particle Properties of Cefuroxime Sodium

Cefuroxime Sodium (CS) is a low-toxicity and highly effective antibiotic drug with a wide application market. Its polymorph form β has the advantage of thermal stability and resistance to degradation. However, poor powder properties cause filtration difficulties, severe electrostatic phenomena and difficulties in automated dispensing, which limit its application. In order to improve the powder properties and stability of CS from the source, we used a one-step crystallization method to prepare CS (form β) with high sphericity and uniform particle size. The spherulites of form β of CS (β spherulites) have excellent powder properties, with high flowability (32.8◦), good sphericity (90.2%), low C.V. (45.6%), and adjustable particle size distribution. Thanks to improved powder properties, the β spherulites exhibits 40% higher thermal stability and maximum 41% faster dissolution rate in the compatible solutions, compared to commercially available products, which implies the β spherulites is more competitive in powder formulations.

Meet the Speaker

Jingwei Cai , Tianjin University

Jingwei Cai is a PhD Student at the National Engineering Research Center of Industry Crystallization Technology, Tianjin University. His research interests mainly include pharmaceutical crystallization, crystal engineering and polymorph control. His recent work focuses on cefuroxime sodium, particularly the simultaneous control of polymorphic transformation and uniform spherulite growth to enhance pharmaceutical particle properties. At BACG 2026, he will present his work entitled “Simultaneous Polymorph Control and Uniform Spherulite Growth for Enhanced Particle Properties of Cefuroxime Sodium.”