PREDICTING CATASTROPHIC PHASE INVERSION ON THE BASIS OF DROPLET COALESCENCE KINETICS

Citation
Gej. Vaessen et al., PREDICTING CATASTROPHIC PHASE INVERSION ON THE BASIS OF DROPLET COALESCENCE KINETICS, Langmuir, 12(4), 1996, pp. 875-882
Citations number
30
Categorie Soggetti
Chemistry Physical
Journal title
ISSN journal
07437463
Volume
12
Issue
4
Year of publication
1996
Pages
875 - 882
Database
ISI
SICI code
0743-7463(1996)12:4<875:PCPIOT>2.0.ZU;2-6
Abstract
A predictive model for catastrophic phase inversion, based on the kine tics of droplet breakup and coalescence, is presented here. Two invers ion mechanisms can be distinguished, depending on the direction of the phase inversion process. With the surfactant predominantly present in the dispersed phase, the coalescence rate is high and ''easy'' phase inversion takes place at relatively low volume fractions. Going in the other direction, surfactant is predominantly present in the continuou s phase. The coalescence rate is dramatically lowered because of the G ibbs-Marangoni effect, and ''difficult'' inversion will not take place up to relatively high volume fractions. Experiments were carried out in a stirred vessel, where phase inversion was detected by a jump in e mulsion conductivity. Easy inversion points were found on the order of 20-50% volume fraction of the dispersed phase. Difficult inversion wa s not detected up to 97% dispersed phase. The easy inversion point inc reases with dispersed phase addition rate and is independent of the st irrer speed below a stirrer speed of 1500 rpm. A simple model based on the breakup and coalescence rate of emulsion droplets in the easy inv ersion regime allows us to calculate the stationary droplet size as a function of the volume fraction of the dispersed phase, as well as the evolution of the droplet size in time under addition of dispersed pha se. The stationary droplet size diverges above a critical volume fract ion of 26.4%, indicating phase inversion. This model can qualitatively describe hysteresis and the phase inversion point dependence on stirr er speed and dispersed phase addition rate, as found in our experiment s.