Analysis of W1/O/W2 double emulsions with CLSM: Statistical image processing for droplet size distribution

Abstract

Droplet size distribution is an important measure in processing of (food) emulsions, due to its influence on relevant product characteristics as e.g. taste, mouth feel and texture. Analysis of droplet sizes in more complex structures like double emulsions is currently a challenge for common measuring techniques (e.g. laser diffraction). Apart from nuclear magnetic resonance, optical microscopy shows potential to measure the encapsulated droplets of (double-) emulsions. Statistical image processing can be applied to get particle or droplet size distributions from confocal laser scanning microscope (CLSM) images. CLSM is well suited for the characterisation of emulsion systems and has already been applied to single emulsions. In this study, it is combined with statistical image processing with the aim to obtain droplet size distributions of droplets, dispersed in droplets (double-emulsions). The question is addressed whether an accurate droplet size distribution can be extracted from a set of images in spite of the systematic errors of the measurement method itself. Algorithms for error correction and for estimation of a statistically relevant number of objects are presented. The results from image data processing are compared with data from other measuring techniques on single and double emulsions as well as glass particle suspensions. Corrected CLSM data are in good agreement with measured values from the laser diffraction on single emulsions and glass particle suspensions. A simultaneous representation of the inner and outer phase of a double emulsion is possible.