Getting liquid perfection

From gypsum to chocolate – particle size is vital in the formation of desirable properties in a wide range of liquid suspensions

AUTOMATED image analysis enables complex particle characterisation from two-dimensional images. The technique facilitates the calculation of a wide range of size and shape parameters, allowing identification and quantification of even the most subtle differences. Applications are wide ranging with many industries now using these automated systems to obtain valuable particulate data at all points in the manufacturing process, from research and development through to final product quality control.

This article looks at the use of a fully automated, microscopy-based image analysis system for characterising particles in liquid suspensions. It illustrates the use of a new large volume wet cell to address the challenges traditionally associated with characterising liquid samples.

Automated image analysis offers high resolution measurement of particle size and particle shape through direct image capture. It delivers excellent results for a wide range of particle sizes, typically from below 1µm up to several thousand, and is ideal for many sample types including those, such as dry powders, creams and gels, that may be difficult to characterise by other means.

The emergence of fully automated systems has made it possible to rapidly examine the very large numbers of particles needed to obtain statistically significant size distributions, while at the same time removing the subjectivity inherent in manual microscopy techniques. However, achieving this statistical significance when examining liquid samples with conventional microscopy instrumentation presents a challenge, the major obstacle being the limited sample volume that can be accommodated under a coverslip.

A novel solution to the problem of measuring liquid samples is seen on the Morphologi G3 system from Malvern Instruments. Here the development of a large volume wet cell addresses the problem of obtaining statistically relevant data. Its 2 to 6ml interior volume and large scan area enable more representative sampling, reducing the possibility of sampling error and ensuring the availability of a sufficiently high number of particles for analysis.

The wet cell (shown in Figure 1) consists of two glass windows separated by one or more 250µm gasket spacers, and held together by a magnetic clamping assembly that prevents sample evaporation during analysis and allows easy cleaning between measurements. Its use in the analysis of two very different types of sample - chocolate and gypsum - illustrates the breadth of application and its ability to greatly extend the utility of automated microscopy-based techniques for particle characterisation.

Particle size is an established parameter for testing the product quality of

Figure 1: Large volume wet cell installed on the Morphologi G3

chocolate. Particle size measurements obtained by laser diffraction, for example, are known to correlate well with subjective tasting panel observations such as ‘mouth feel’. Now, the ability to characterise particle shape and intensity as well as size, offers manufacturers the potential for even greater product understanding and fine-tuning. To investigate this further, three samples of milk chocolate - a market-leader, a supermarket brand and a luxury brand - were analysed using the wet cell on the Morphologi G3. Each chocolate was dispersed in sunflower oil in an ultrasonic bath for 10 minutes to produce a suspension of the non fat-soluble solid components. Following dilution to a concentration suitable for analysis, 2 ml of the sample was injected into the wet cell.

Standard operating procedures define all the software and hardware variables. Thus the 5X objective was selected and the system was set to scan in a defined area of the wet cell until 30,000 particles had been measured. Automatic calculation of multiple size and shape parameters for every particle measured revealed the most significant parameters for the chocolate samples to be particle size distribution, circularity and elongation

Volume weighted particle size distributions expressed in terms of Circle Equivalent (CE) diameter (Figure 2) indicated that while samples had similar median particle sizes there were differences in the size distributions. In particular, the luxury chocolate had a significantly smaller proportion of very large particles. This is consistent with expectations for a luxury brand, as desirable attributes such as ‘smoother’ mouth feel are associated with finer particles.

Figure 2: Calculation of CE diameter

Another parameter, high sensitivity (HS) circularity, is a measure of how perfectly circular the 2D projection of a particle image is. The leading brand showed the highest values, indicating a more spherical particle morphology, another factor associated with improved mouth feel. Finally, elongation provides an indication of the length/width ratio of the particle. Again, the luxury brand was distinctive, with a distribution different from the other samples, suggesting that the chocolate contains ingredients of a different origin or was processed differently.

Since all particle images (in this case 30,000 per measurement) are stored, it is easy to compare size and shape parameters and to filter results to concentrate on specific criteria. A post-analysis filter to exclude particles with a CE diameter of less than 20 µm, for example, showed interesting trends in mean intensity. Particles in the market leader brand were darker and had a higher standard deviation while the luxury brand had the lowest SD, which probably reflects more homogeneous ingredients. The supermarket chocolate had the lightest particles overall, almost certainly due to a higher sugar content.

Gypsum is a common mineral used in a variety of industries, but perhaps most

Figure 3: Parameter Variability Chart indicating that Aspect ratio is the parameter that varies most across the sample sets.

usually associated with building materials. For many applications the shape of gypsum particles, particularly aspect ratio (width/length) and elongation, is very important because symmetrical particles offer better packing properties than do elongated ones. Gypsum processing involves the production of a concentrated slurry, and particle characterisation at this stage delivers a better understanding of the production process.

To demonstrate the use of the wet cell in this application, four samples were prepared for analysis by diluting approximately 1mg of wet slurry into 20ml of saturated gypsum solution. Dispersions were agitated and ‘pumped’ several times with a syringe before 2ml was extracted and injected into the cell.

SOP-controlled analysis involved the measurement of a specific area of the wet cell using the 5X magnification objective. Particle images with a surface area of less than 100 pixels were filtered from the final results and a convexity filter was applied to remove touching particles. After applying the filter, each analysis involved between 20,000 and 30,000 particles. Measurement times were just 20 minutes.

Figure 4: Dendrogram in terms of aspect ratio which groups the samples into two sets: 1 with 3 and 2 with 4.

Using the data comparison tool in the instrument software it was clear that aspect ratio was the most variable parameter (Figure 3). Further analysis showed the relationship between the four samples for this parameter, indicating two sets – samples 1 & 3 and samples 2 & 4. A trend graph (Figure 5) showed how the mean aspect ratio varies across these samples. This can be useful in determining a threshold value for distinguishing between the two different sets of samples, for example in a pass/fail quality control application.

Particle characterisation using automated microscopy-based techniques is rapidly superseding time-consuming and operator-dependent manual microscopy and opening up many new areas of application. Until now, however, these automated techniques have been largely confined to dry dispersions and some semi-solids, but the new large volume wet cell for the Morphologi G3 broadens the applications considerably, enabling the quick and easy analysis of liquid suspensions. Multiparameter analysis delivers comprehensive particle size, shape and intensity information, and the system stores images of every particle analysed for future reference.

Figure 5: Trend plot showing the mean aspect ratio for each sample