Dancing Crystals
The motion of the crystals has several potential sources:
Ø Increase in surface tension on the surface sites due to
a) increase in naphthalene concentration where evaporation of acetone occurs
b) lowering of temperature via evaporative cooling of acetone
c) absorption of water from the air
Ø Reduction in surface tension as the solid forms and grows due to
a) increase in temperature with exothermic heat of Crystallization
b) release of diluted low surface tension solution
A device to direct air to at a point on the surface can be made by connecting a plastic thin tipped pipette to a hose with a very slow flow of compressed air. The accumulation of motionless crystals suggests evaporation effects can be strong.
Pass the mouse over the box below to show a video of dancing crystals.
Visualizing Marangoni Flows
Flow patterns are observed which indicate local Marangoni currents exist in acetone solution. The patterns are illustrated in the picture and video below. These are observed in both the naphthalene solution and plain acetone. This is accomplished by adding a scoopula tip full of phenol red solid so that there is an excess of powder in solution (other dyes like thymol blue work as well). Both the solution and pure acetone show the tears of wine effect characteristic of Marangoni flow. Looking at these flows it is easy to see how they could contribute to the motion of the crystals.
Pass the mouse over the box below to show a video of phenol red patterns
Some Pertinent observations (read This)
If dancing crystals and phenol red (in acetone) in separate containers are set side by side and each covered with a Petri dish to reduce evaporation, both show reduced speed and motion but when the naphthalene crystals first form there is a sudden burst of speed and fast motion. This eventually settles down to a slow meandering (sometimes oscillatory) motion that seems to be only slightly more energetic than the phenol red flows observed in the other container. If the dancing crystal solution and phenol red are placed together in the same watch glass the crystals start out with a burst of speed and move with the flows for a few seconds and then have another burst of speed. The flows accelerate when the crystals are about to form. Some crystals exhibit flows away from the solid. The effect of the solid formation and crystal growth does appear to play a role in the initial motion and bursts of speed. The Marangoni flows appear to be a contributor later on to continue motion. It's hard to say whether the bursts of speed are due temperature effects, release of solution, fresh solution cycling from underneath or some other source.
The role of moisture absorption by the solution
Some of the changes in surface tension may be because acetone absorbs moisture from the air at the surface. This would explain in part the variance with humidity. (Another reason for humidity effects may be that water decreases the solubility of naphthalene). It wouldn’t be surprising if a surface that has cooled to 14 oC would condense and absorb moisture. This can be illustrated by adding 0.003g/ml of thymol blue to dried acetone. The dye is almost colorless in dry acetone but is a bright red when water is present. When placed on a watch glass a red coloration migrates from the surface to the periphery. Eventually a red coloration is seen in the body of the liquid.
Lycopodium Patterns
Another interesting effect that may result from local currents is observed if lycopodium powder is added to a solution of 7.25 g of naphthalene per 20ml of dried acetone (without water added) the powder will fall to bottom. After allowing a minute for evaporation, the powder can be swirled back to the surface and allowed to fall. Patterns are observed which may indicate these local currents. This is illustrated in the pictures and videos below.
Pass the mouse over the box below to show a video of lycopodium patterns
The Camphor Dance
Since camphor is a surface active agent, the camphor motion may be due to localized surface effects brought about by camphor itself. The spreading camphor decreases the surface tension which then increases when the camphor vaporizes at the edge of the open space. This creates surface tension gradients. Naphthalene is not prone to this type of motion since it is a weak surface coating agent on acetone. The picture and video below show the surface action and motion of camphor.Pass the mouse over the box below to show a video of the camphor dance.
Dancing Ghosts
One explanation could be that as the alcohol spreads over the surface it evaporates creating strong short range surface tension gradients which produce the motion. The gradient is sustained by alcohol pushed along on the surface as the drop moves. In fact looking at the picture below we see a short range surface disturbance very near the drop. One explanation for the motion when distorted to the crescent shape is provided by C. Bates et al at the web site for the Washington State University surface dynamics lab. (wsu.edu/~std/presentations/chris/page1.html). At the front of the crescent the alcohol radiates out to a wider area and thins. Whereas in the rear of the crescent the alcohol radiates to a small area and concentrates. This leads to a surface tension gradient.
Removal of alcohol from the surface by evaporation and dissolving would fuel these flows. In fact if you cover the dancing ghosts with a Petri dish they will stop moving immediately and start right back when the dish is removed. Also when the ghosts start to slow down adding a little more water will speed them up. This creates a more diluted solution below and accelerates the dissolution of the alcohol.
P.S. The radiation of material as explained for the crescent shape of the dancing ghosts also helps explain the marangoni flows. As solution spreads on a wider arc of a disc on the watch glass the increased surface area facilitates evaporation and increases surface tension leading to flows.
A short cut to the long range motion of Dancing Ghosts
Add 20 drops of allyl alcohol (with a dropper that provides 44 drops/ml of allyl alcohol) to 25 ml of water at 40 oC in a closed vial and shake to dissolve. You can also add the allyl alcohol to cold water and heat it later. Pour this into a Petri dish and add 9 drops more of the allyl alcohol. The rapid long range motion should start quickly.
Pass the mouse over the box below to show a video of Dancing Ghosts
Thunderstorm
In all of these experiments we have essentially a climate system is a dish. This just takes it to a wide range disturbance and creates a “thunderstorm”. The highly surface active and volatile ethyl acetate spreads quickly and evaporates quickly creating strong sudden surface tension disturbances.
Pass the mouse over the box below to show a video of the thunderstorm.
Additional Experiments I have not done.