Getting the picture of emulsion flows

Along with Unilever, we are employing computational modelling to study detailed dynamics of emulsions to optimise innovative manufacturing processes.

Challenge

Manufacturing and processing of emulsions is an integral part for many industries including oil & gas, waste-water treatment, food and personal-care products. Analysis of emulsions is particularly challenging due to numerous physical mechanisms such as surface chemistry or flow turbulence. Researchers at Unilever are constantly investigating novel devices that can reliably and sustainably deliver improved processing rates. To achieve this, they employ a combination of physical experimentation and computer modelling, seeking accurate insights about the behaviour of emulsions under varying flow conditions.

Approach

The IBM Research team have developed bespoke simulations, tools and work-flows for investigating emulsion dynamics in processing equipment. Using high-performance computing facilities at the Hartree Centre, we have simulated the detailed evolution of large populations of drops transported in turbulent flow fields. Applying data-centric principles, the team has built a tool kit that enables an unprecedented way to study collective and individual behaviour of drops. The team can analyse evolution of drop size distributions, drop break-up rates, shapes and orientations, all of which influence the final quality of process products.

Benefits

The combination of mathematical modelling and data-analytics realises the concept of virtual measurement. It allows researchers to investigate the flow in places where physical instrumentation could not be applied enabling estimates of average drop size or high-resolution photo-realistic flow field visualisation. The computations complemented experiments and domain expertise at Unilever, opening new ways of studying this class of processes. The work has been carried out as part of Innovation Return on Research (IROR) programme, a collaboration between STFC and IBM Research and can now be applied in emulsion research across a broad spectrum of industrial problems.

“This challenge was an extremely difficult but important case and we now have a breakthrough in the simulation.”

Adam Kowalski, Unilever