Coalescence in Creaming Emulsions. Existence of a Pure Coalescence Zone Sanjeev Kumar, †,‡ G. Narsimhan, § and D. Ramkrishna* ,‡ School of Chemical Engineering and Department of Agricultural and Biological Engineering, Purdue University, West Lafayette, Indiana 47907 Stored emulsions undergo various destabilizing processes including creaming and coalescence of drops. Many experimental studies have been carried out in the past to quantify drop coalescence; however, the presence of two distinct types of coalescence mechanismssone due to the relative motion between the drops and the other due to their permanent proximity to each other in the creamshas not been recognized. Furthermore, the effects of creaming in these investigations could not be eliminated in ways that do not introduce new modes of coalescence or alter the existing ones. The population balance equation for an emulsion in a column in which creaming, Brownian diffusion, and coalescence of drops occur simultaneously is analyzed. The analysis reveals the existence of a dynamic zone at the top of the column in which the net effect of creaming and Brownian diffusion of drops is eliminated. Thus, while the drops cream, diffuse, and coalesce as in an actual emulsion, the measurements from this zone allow the effects of drop coalescence to be isolated from other destabilizing processes. Based on this finding, a new methodology to investigate coalescence is proposed. Experimental support for the proposed theory is also provided. If the size distribution in the uniform zone evolves in a self-similar manner, it is shown that the techniques already available in the literature can be used directly to estimate the coalescence frequency kernel. 1. Introduction A quantitative prediction of the shelf-life of emulsion- based products is important for their successful com- mercial use. In general, the overall appearance and the quality of a stored emulsion deteriorate with time and after a while reach unacceptable levels. The finite time for which an emulsion retains acceptable characteristics while sitting undisturbed on a shelf is taken as its shelf- life. (Clearly, better definitions of shelf-life while do not rely on the interpretation of “what is acceptable?” are needed. We will, however, not dwell on this issue in this paper.) The appearance and quality of a stored emulsion product are affected by three processes: cream- ing of dispersed-phase drops due to buoyancy, their flocculation, and their coalescence. The first two pro- cesses are reversible and harmless by themselves, but they can aid coalescence of drops which is irreversible and eventually leads to complete destruction of the emulsion. A creamed or flocculated emulsion can be shaken to easily restore the original homogeneous emulsion. In comparison, if drops become bigger or oil appears on the top (known as “oiling off”) due to the coalescence process, the emulsion cannot be restored to its original state unless it is homogenized again. Let us consider various stages of the stabilization process. Figure 1 shows the state of a stored emulsion at various times in which creaming and coalescence are the only rate processes present. Figure 1a shows a uniform emulsion at an initial time. Figure 1b shows the same emulsion at a later time when some drops have creamed to the top and have arranged themselves in a tight packing while the others have moved away from the bottom. As the drops move, they collide with each other, coalesce, and form bigger drops. The drops in the tight packing at the top also continue to coalesce with others, and much later, as shown in Figure 1c, free oil appears on the top of the emulsion. It is important to note that the two types of coalescence mechanisms are different in nature and they have different implica- tions for shelf-life. We shall term the former, which is caused by relative motion due to buoyancy or Brownian motion, as “coalescence in bulk” and the latter, which is caused by the permanent proximity of drops to each other in cream as “coalescence in cream”. Coalescence in bulk accelerates the creaming process by forming large drops which accumulate a large amount of dis- persed phase in the cream layer in much shorter times. Furthermore, the cream undergoes accelerated coales- cence because it now consists of large drops to begin with. Thus, increased coalescence in bulk has far more serious implications than the corresponding increase in coalescence in cream. If the two types of coalescence rates are combined into one overall rate of coalescence, the interesting and important implications of the ge- ometry and the height of the container on the shelf-life of emulsion products will not manifest. As pointed out before, coalescence is an irreversible and the most deleterious process for the shelf-life of an emulsion. It is for this reason that coalescence in emulsions has been studied for a long time (King, 1941; * Author to whom correspondence should be addressed. Phone: (317)-494-4066. Fax: (317)-494-0805. e-mail: ramkrish@ecn.purdue.edu. † Present address: Department of Chemical Engineering, Indian Institute of Science, Bangalore 560012, India. ‡ School of Chemical Engineering. § Department of Agricultural and Biological Engineering. Figure 1. Various stages of evolution in a stored emulsion. 3155 Ind. Eng. Chem. Res. 1996, 35, 3155-3162 S0888-5885(96)00014-0 CCC: $12.00 © 1996 American Chemical Society