Estimating Drop Size in Turbulent, non-Coalescing Liquid-Liquid Systems
Immiscible liquid-liquid dispersions and emulsions are ubiquitous in the chemical, petrochemical and pharmaceutical process industries and form the basis of many intermediates and consumer goods, such as coatings, foods, lotions, gels and injectable/topical medications. During formulation, the ingredients are agitated vigorously and the rheology builds as the drop size decreases. Since the quality/stability of the formulation depends upon the initial degree of dispersion, it is important to predict the dispersion/emulsion drop size under turbulent conditions. Often, the product is stabilized against coalescence.
This presentation will consist of several parts. First, we discuss the motivation for our work and provide some background in turbulence fundamentals. Next, we discuss development and validation of mechanistic theories to predict equilibrium drop size in turbulent liquid-liquid systems that are dilute or otherwise stabilized against coalescence. It will be argued that breakage phenomena depend upon microscopic flow details that also determine the ultimate drop size. As a result, prediction and correlation methods are independent of device geometry and a common basis exists to assess device performance. Consideration will be given to a variety of contacting geometry (pipes/static mixers, stirred tanks, high shear mixers, etc.) to demonstrate that dispersion mechanisms and predictive tools are device independent. An important factor to be discussed is the size of drops relative to the various scales of turbulence.
Ultimate drop size depends on energy transfer rates on the drop or microscale, but most velocity field simulation and measurement techniques focus on the macro or device scale features of the flow. In this part, we discuss how we use Computational Fluid Dynamics (CFD) and measurement techniques such as Laser Doppler Anemometry (LDA) and Particle Image Velocimetry (PIV) to estimate device dependent quantities, such as local energy dissipation rate and the extent of the dispersion zone, to aid in drop size prediction and to assess device dispersion performance. If time permits, we will discuss crystal wet milling of active pharmaceutical ingredients (API) to demonstrate that a common basis exists for predicting solid particle and immiscible liquid drop size in turbulent fluids.