Büro für Altlastenerkundung und Umweltforschung

Dr. Rainer Haas

Stadtwaldstr. 45a, D-35037 Marburg, Tel.: 06421/93084, Fax: 06421/93073

email: haasr@gmx.net

Analogies of a pattern forming chemical reaction with

Raleigh-Benard convection

Rainer Haas1, Ingo Schreiber2

1: Stadtwaldstrasse 45a, D-35037 Marburg

2: Badener Dorfstr. 19, D-28832 Achim-Baden


In this paper some aspects of a pattern forming chemical reaction -that is of the formation of a iodine-iodite-starch-complex by UV-irradation of a potassium-iodite/starch solution- are described. Anologies with the Raleigh-Benard convection are discussed.

1 Results

Several examples of pattern forming chemical reactions have been described in scientific publications. Inspirated by some of those (1-5) we examined a simple chemical reaction, which forms clearly visible structures. By modification of some parameters we tried to investigate the multiplicity of patterns, which can be obtained by this reaction. In the course of these experiments we observed some unexpected and estonishing effects.

In a Petri dish (diameter: 10 cm, filling heigth: 0,5 cm) solutions of 1,5 % potassium iodite and 1 % to 2,5 % starch in water were irradated with UV-light. By UV-irradation free iodine ist formed, that reacts with starch to a blue iodine-iodite-starch complex.

After only a few seconds of irradation blue-violett, hair-like lines of the starch-complex appeared on the surface of the solution, rapidly growing and beginning to ramify. After one minute of irradation a net of convex polygons with constantly thickening and strengthening sides had appeared. Near the sides of the Petri dish the polygons opened and their lateral sides met the glas in a right angle.

About three minutes after the beginning of irradation the blue coloured product, which so far had accumulated in the polygon sides, began to sink slowly downwards. From the netlines of the surface structure of the solution the blue product built up threads, which condensed to something like vertical walls. As soon as the walls had reached a specific depth, a projection-picture of the surface net was formed on a deeper layer of the solution. The largest portion of the product accumulated in the polygon-sides and in the vertical walls.

After a reaction time of about 15 minutes a blue layer began to appear on the surface, still leaving the primary polygonic structure visible (Figs. 1-4).

This reaction product is a iodine-iodite-starch-complex, which has an absorption maximum of 560 nm. A complex formed by the direct reaction of iodine, potassium iodite and starch in aqueous solution by stirring has its absorption maximum by 600 nm. Control tests proved that by UV-irradating of a solution of potassium iodite iodine is built.

In the case of not UV-irradating the whole surface of the solution in the Petri dish, the main reaction stayed the same: in the irradated zone the above described structures built up as usual whereas the not irradated zone did not show any immediate reaction. But as soon as the polygonic structures started to grow towards the bottom of the dish, they, in a deeper layer, also developed blue threads which reached through the not irradated zone towards the sides. Each of these threads met the glas in a right angle. There these threads developed vertically upward until they met the surface. There they built up a structure similar to their structure in the deeper layer.

This experiment shows, that the described structures are convection cells. The long-streched cells towards the sides of the Petri dish were detected in all experiments (Fig. 5).

At the filling height of 1 cm a two-phase system developed, consisting in the 0,5 cm deep upper layer of a convecting blue coloured cellular structure and a clear layer beyond, where no reaction and convection are visible. The two-phase system remained stable for 6 hours after stopping the irradation. Then the upper layer cell structure collapsed by forming many threads, which dissolved in 30 minutes (Fig. 6).

By varying the strength of the starch solution the viscosity was modified. The diameters of the convection cells depended on the viscosity of the solution. Only in a certain range of viscosity, at a starch concentration from 1 % to 2,5 %, this specific cell formation was possible. Beyond this interval the motion of the reaction product depended on diffusion.

For the examination of the interdependence of viscosity and cell structure potassium-iodite solutions of 1,5 % and a filling hight of the Petri dish of 1 cm were used. At a higher viscosity, that is a starch concentration of 2,5 %, small cells with a diameter of 1 mm appeared and the depth of the convection layer was about 2 mm. At a lower viscosity, at a starch concentration of 1 %, the diameter of the cells and the depth of the convection layer were about 5 mm.

At a filling height of 5 cm (cylindric glas beaker, diameter 5 cm) blue-coloured lines appeared on the surface of the solution after having started the UV-irradation. All the lines started at one single point on the surface and subsequently the product sank down to the bottom of the beaker in a diffused manner. A formation of convection cells was not observed.

2 Discussion

The observed cellular convection layers showed anologies with the Raleigh-Benard-convection:

1) The cross-sections of the convection cells were convex polygons (6).

2) At the sides of the convection cells there was downward motion.

3) The convection cells which touched the sides of the reaction basin or any other obstacle were perpendicular to them.

4) A thin layer at the suface of the solution did not take part in the convection. The uppermost surface consisted of a thin film of homogenious complex, which cracked at some places (see Fig. 4). This is in analogy to the statement of Berge and Dubois that border layers exist in the Raleigh-Benard convection where no convection takes place and where heat ist transported by diffusion (7).

Fine dust of aluminium scattered on the surface of the solution showed no sign of motion as the convection cells below appeared. This confirmed the existence of motionless thin layers (less than 1 mm) at the surface.

Even through the described phenomenia are similar to those of the Raleigh-Benard convection, they depend on totally different physical conditions. In the case of the Raleigh-Benard convection there is a temperature gradient opponently to gravity. The energy needed for the formation of the convection cells is supplied by the steady heating of the bottom of the reaction basin. In the case of our system the driving force of the convection cannot be a temperature gradient. By irradating the surface of the solution energy is indeed supplied. But as the surface ist heated, this energy is working against convection.

In this case the driving force of convective motion could be the growing accumulation of the iodine-iodite-starch complex at the surface of the solution. Iodine-iodite-starch has a higher molecular weight than the starch itself. So the upper solution with the reaction product could have a considerably higher specific weight than the solution below. Thus there would be a gradient of specific weight, which might cause the constant downward motion and cellular convection of the product as long as iodine-iodite-starch is produced at the surface.

3 References

1) P. Möckel: Naturwissenschaften 64 (1977) 224.

2) M. Orban: J. Am. Chem. Soc. 102 (1980) 4311.

3) M. Kagan; A. Levi; D. Avnir: Naturwissenschaften 69 (1982) 548

4) M. Gimez: Naturwissenschaften 70 (1983) 90.

5) L. Weissenborn; R. Bausch: Naturwissenschaften 70 (1983) 307.

6) S. Chandrasekhar: Hydrodynamic and Hydromagnetic Stability; Oxford 1961.

7) P. Berge; M. Dubois: Contempt. Phys. 25 (1984) 535.


Fig. 1: Reaction time 2 min, starch concentration 1 %; formation of big cells, opened cells at the side

Fig. 2: Reaction time 4 min, starch concentration 1,5 %; homogenious layer at the surface, veils caused by sunk down reaction-product

Fig. 3: Reaction time 1 min, starch concentration 2 %; formation of small cells at high viscosity, transformation of the lines to polygons

Fig. 4: Reaction time 12 min, starch concentration 1,5 %; cracked film at the surface, opened cells at the side, visible structures in the depth of the fluid

Fig. 5: Structure of the convection cells in the shadow zone

Fig. 6: 30 minutes after stopping irradation; transformation of the cellular structure

Fig. 7: Surface structure in a cylindric glas beaker at 5 cm filling height

There are no figures in that article. Reprints of the publication „Analogies of a pattern forming chemical reaction with Raleigh-Benard convection“ you can order from the author by email.