PhD Symposium

Marangoni-driven pattern transitions in surfactant-covered Faraday waves

When a liquid surface is shaken vertically, it can form waves called Faraday waves. These waves often organise into simple patterns like squares and hexagons, but under certain conditions, they can develop exotic patterns. In this study, we investigate how surfactants-  substances that modify surface tension- affect these patterns through Marangoni stresses, which arise due to variations in surface tension.

We introduce a quantity B to measure how strong these stresses are compared to the wave motion. Our results show that as B increases, more shaking is needed to create the waves- a classical signature of Marangoni-enhanced dissipation of surface waves. When no surfactants are present, the waves form a square pattern. As B increases, the patterns transition from squares to asymmetric squares, then to wavy stripes, and eventually to ridges and hills. 

These hills form due to the bi-directional Marangoni stresses at the narrow regions of the ridges. We discuss the reasons behind these pattern transitions and the formation of these unique shapes with the use of two-phase direct numerical simulations.

Modeling the Solubility of Peptides solutions with the SAFT-g Mie

My PhD work involves developing a predictive tool to deliver accurate predictions of the solubility of active pharmaceutical ingredients, specifically peptide-based drugs, by leveraging recent advancements in molecular science. The goal is to predict solubility early in the drug development process, aiding in the selection of the right solvent for manufacturability, and improving the efficacy of the drug by enhancing its solubility. Following the monumental achievement of discovering insulin to treat diabetes in 1921, development of therapeutic peptide drugs accelerated and they currently account for a significant proportion of the pharmaceutical market with worldwide sales of $48 billion in 2022. The aim of my research is to guide the manufacturing of complex peptides having high specificity and low toxicity. Predicting solubility is a challenging task because it depends on factors like temperature, pressure, and the type of liquid. Further, absence of the melting properties of the peptides at an early stage of development adds another complication. I implement a group-contribution thermodynamic approach, known as SAFT-g Mie, in which molecules are broken down into smaller parts, akin to Lego pieces, to predict properties. This research can significantly advance pharmaceutical manufacturing by improving the drug development process and ensuring more effective medications.

Motion in Stratified Fluids

The transport of particles in density-stratified fluids is a crucial area of study due to its widespread occurrence in nature. In the ocean, particles and marine snow descend through fluids with significant density variations caused primarily by salinity and temperature gradients. This background fluid heterogeneity can markedly influence the settling or rising rates of these particles, often resulting in their accumulation at density transition layers. 
My work investigates how stratification influences the settling of solid particles through three-dimensional Direct Numerical Simulations, with a focus on their transient behaviour and how particle-particle interactions differ from those in homogeneous fluid media. 

GlyCoDA: A compositional data analysis and domain adaptation perspective on disease risk prediction using plasma IgG N-glycans

An Automated High-throughput Workflow to Accelerate the Discovery of Porous Liquids for CO₂ Capture

Porous liquids are materials combining the flowability of a liquid with the functionality of a porous solid. They can be formed by dispersing a microporous solid in a pore-excluded liquid (e.g. ionic liquids, ILs) meaning the porosity of the solid is retained. In CO₂ capture, current aqueous sorbents have high regeneration costs associated with retrieving adsorbed CO₂ and porous liquids have been identified as alternative sorbents which can overcome this issue while maintaining compatibility with current industrial systems.  

However, currently, there is limited fundamental understanding of how porous liquid properties relevant to CO₂ capture (e.g. CO₂ uptake, viscosity, stability) are influenced by its components and their combination. Besides, with more than hundreds of thousands of porous solids and ionic liquids to choose from, the chemical space is too large to be screened using traditional experimental techniques. 

To address this challenge, we have developed an automated, high-throughput workflow to formulate and characterise porous liquids to understand key structure-property relationships which can then be utilised to identify important factors for developing porous liquids as CO₂ capture sorbents. We have used this workflow to study the effect of changing component IL cation size and functionality on porous liquid viscosity, stability and CO₂ uptake.  

Crystal Regeneration: Investigating the Post-Breakage Growth Phenomenon in Organic Crystals

The crystalline structure of organic molecules has long puzzled scientists. As attractive is the highly ordered nature and purity of the crystal products, it is equally difficult to decipher the crystalline lattice, shape and size. Mechanical breakage is a common occurrence in industrial crystallizers, yet there exists limited knowledge on the consequent growth of broken crystals. Hence, process modelling attempts commonly assume pre-breakage growth kinetics for broken crystals, resulting in unreliable predictions of end crystal properties. 
This PhD investigates for the first time an unconventional ‘regeneration’ phenomenon in paracetamol crystals following breakage along the cleavage (weakest) plane (010). The broken part was observed to regrow on the parent crystal, restoring the pre-breakage shape prior to demonstrating overall growth, which was otherwise unexpected – similar to the process in which a lizard regrows its tail after autotomy. The phenomenon has been elucidated through intricate single crystal experiments, supported by advanced imaging setups and automated image analysis via MATLAB to measure growth rates. Crystal regeneration was examined in three solvents—ethanol, acetone, and tetrahydrofuran (THF)—across various supersaturations (1.03 to 1.30) and under agitation.
This PhD systematically explores crystal behaviour post-breakage and proposes mechanisms to explain these phenomena, offering critical insights into previously unexplored post-breakage crystal growth dynamics.

Development and demonstration of photoelectrochemical reactors for solar hydrogen production

For commercially viable photoelectrochemical hydrogen production, considerable progress must still be made in the development and demonstration of up-scaled devices. My PhD project seeks to investigate the challenges in scaling up materials and devices for water splitting. The first challenge of my project is to develop scalable routes to fabricate promising photoelectrodes. The second challenge is the development of up-scaled devices. The development of up-scalable photoelectrodes must be addressed in parallel with device development and field testing, to ensure material compatibility with the systems they are used in. Finally, we envisage what a commercial photoelectrochemical water splitting system might look like. I have already developed a prototype system, which I tested in South Africa under natural concentrated sunlight in 2024. The system was able to spontaneously split water into hydrogen and oxygen using only photon energies. The next step is to make design improvements to the system based on this experience, with the aim to carry out further field tests in London this summer.

Linking Zeolite Physical and Chemical Properties to Flow Behaviour

Powder flow is a major challenge in zeolite manufacturing. While studies have largely explored particle properties related to flowability in pharmaceuticals [1-3], similar research on zeolites remains limited [4], challenging manufacturing efficiency. This study investigates the impact of zeolites' physicochemical properties on their powder flowability. ZSM-5 pre- and post-calcination, ALD-modified ZSM-5, and commercial samples were examined. Characterisation included inverse gas chromatography, dynamic vapour sorption, XPS, XRF, nitrogen adsorption, PSA, and SEM, while flowability was evaluated using basic flowability energy and the flow function coefficient.

Calcination increased surface area, packing porosity (~20%), and surface energy (~98 mJ/m²). Improved flowability was associated with particle properties, including increased packing porosity (r = 0.83) and reduced bulk density (r = -0.84). ALD modification reduced surface area (-13%) and moisture uptake (-18%), enhancing flow. Correlations between reduced surface area (r = -0.88), micropore volume (r = -0.83), and moisture uptake (r = -0.94) indicated reduced interparticle friction and interparticle forces increasing flowability.

In conclusion, processing history significantly influences powder properties and flow. Calcination improved flowability by altering packing porosity and bulk density. Conversely, where size and density mainly remained unaffected by ALD, reductions in surface area, micropore volume, and moisture sorption primarily drove flow enhancement.

[1] U. V. Shah, V. Karde, C. Ghoroi, J. Y.Y. Heng, Influence of particle properties on powder bulk behaviour and processability, International Journal of Pharmaceutics, 518(1–2) (2017) 138–154.
[2] V. Karde, S. Panda, C. Ghoroi, Surface modification to improve powder bulk behavior under humid conditions, Powder Technology, 278 (2015) 181–188.
[3] C.G. Jange, R.P.K. Ambrose, Effect of surface compositional difference on powder flow properties, Powder Technology, 344 (2019) 363–372.
[4] S. Zinatlou Ajabshir, C. Gucuyener, V. Vivacqua, D. Gobby, H. Stitt, D. Barletta, M. Poletto, Flow behaviour of zeolite powders at high process temperatures, Powder Technology, 409 (2022) 117818.

Upstream Process Parameter Optimization for Product Quality: A Comparative Study of Statistical Modeling and Probabilistic Design Space Identification

Development of Peptide Biosensor and Microneedle-Based Diagnostics for Melanoma Skin Cancer

Melanoma is the deadliest form of skin cancer, and current diagnostic methods rely on biopsies—an invasive and time-consuming procedure. My research focuses on developing a minimally invasive diagnostic platform that detects melanoma-specific biomarkers in skin fluid.

I designed and synthesized a novel peptide-based bioreceptor that detects S100B, a key melanoma biomarker, at extremely low concentrations. The molecule was engineered using peptide synthesis and optimized for sensitivity through a fluorescent labelling strategy that responds to protein binding (established proof of concept for cancer detection). Additionally, I developed a microneedle-based platform capable of extracting interstitial fluid (skin fluid) for biomarker detection. I validated this platform using 3D tissue-engineered melanoma skin models, assessing its biocompatibility, sampling kinetics, mechanical robustness, and penetration efficiency.

This work lays the foundation for a future point-of-care test that could replace traditional biopsies, enabling earlier melanoma detection with a painless and rapid approach.

Studying monoclonal antibody aggregation under flow using ATR-FTIR spectroscopic imaging

Biopharmaceuticals such as therapeutic monoclonal antibodies (mAbs) have shown to be effective drugs for the treatment of a range of diseases, including various types of cancer. During the life cycle of a therapeutic mAb, from the production, purification and formulation down to transportation, storage and clinical use, the mAb product is exposed to different types of physical and chemical stress conditions. This may result in undesired changes to the protein’s structure, increasing the probability of protein unfolding and subsequent aggregation. Since these phenomena directly impact the efficacy and safety of the drug, it is of vital importance that the structural integrity of monoclonal antibody products is investigated under various stress conditions and monitored during manufacturing. Fourier-transform infrared (FTIR) spectroscopy is a powerful tool to study protein secondary structures based on the absorption of infrared light. By applying FTIR spectroscopic imaging, chemical images are generated which provide both spatial and chemical information on the sample, allowing for the study of dynamic and multi-component systems. In this research, macro-ATR FTIR spectroscopic imaging is combined with microfluidic technology to study monoclonal antibodies under flow. Using this technique, the behaviour of mAb formulations was studied (i) at the air-liquid interface of air bubbles, and (ii) in low pH elution buffer as part of protein A chromatography.