Sedimentation and Creaming

A transient gel is a weak gel that collapses after some time. Apparently homogenous and quiescent on bulk scales prior to collapse, it is the sudden, unheralded onset of collapse that makes transient gels interesting.

A typical transient gel system is a colloidal dispersion in which a weak interaction is induced between the particles of the dispersed phase. Our version of this is an oil-in-water emulsion, where the oil is the dispersed phase, aggregated via the addition of a polymer (polyethylene oxide, PEO) that causes a weak attraction between droplets via the depletion effect.

An emulsion containing 30% oil by volume flocculated with 0.1% PEO gives rise to a transient gel. The gel state lasts for around two hours, as can be seen (with care) in the following time-lapse movie where the first 63 riveting hours in the life of a weakly aggregated emulsion are condensed into 63 seconds of video:

Download movie clip [AVI, 280Kb]

To start with, for about two hours nothing seems to happen. Then there is a period of rapid sedimentation (the oil is denser than water) that runs to around the six-hour mark. By this point, a clear layer of continuous phase exists above a layer containing elevated levels of dispersed phase. For the remainder of the movie the clear upper layer grows as the dense lower layer shrinks.

Increasing the polymer concentration increases the strength of the interaction between droplets. The gel becomes stronger, and under otherwise constant conditions the collapse stage may vanish. The following time-lapse movie follows the first 63 hours of a 30% oil emulsion with 0.4% PEO:

Download movie clip [AVI, 270Kb]

The difference between the two is easier to see on a plot of the “clear/cloudy” interface height versus time:

An appropriate scaling of the data shows that creeping gel heights versus time plots (black on the plot below) collapse onto a single master curve, whereas the transient gel system (shown in red) do not, emphasising a difference in the underlying dynamics. This scaling indicates a universality of behaviour in these transient gels.

The “delay time” prior to collapse is relatively short in the 0.1% PEO example. We have other examples in other systems where the delay time is much longer, though the overall behaviour is the same. The 0.4% PEO creeping gel never did collapse: some samples with higher levels of PEO continue to creep for hundreds of hours without any sign of collapse.

We now know that to suggest nothing is happening in a transient gel prior to collapse is wrong. Our confocal microscopy studies of the 0.1% PEO case prior to collapse [LINK to GelCollapseMovies] show that the gel is not in fact dormant but displays complicated dynamics. It seems likely that, with some qualification, a transient gel prior to collapse behaves like a creeping gel.

Simple sedimentation experiments can be quite revealing. Below is another movie of sedimentation in a 30% emulsion with 0.4% PEO. This time, however, there’s a step obstacle sitting on the bottom of the cylinder. It is just visible in the movie as a cloudy rectangle on the left-hand side of container:

Download movie clip [AVI, 270Kb]

As the interface approaches the step it is distorted. This may indicate the presence of a stress-tolerant network, and gives some insight into the distance over which stresses propagate in that network.