The Cloudy Blue Bottle Experiment
A demonstration of semiconductor – photocatalysis
Theory:
Heterogeneous photochemistry is dominated by semiconductor photocatalysis, in which light of energy, hט, greater or equal to that of the bandgap of the semiconductors absorbed by the semiconductor is absorbed by the semi-conductor, generating electron – hole pairs (e-, h+). These photogenerated electrons and holes that reach the semi-conductor’s surface reduce and oxidise adsorbed chemical species respectively. TiO2 is used as a semiconductor to photosensitise the reduction of oxygen on its surface. It is known that photogenerated electrons are able to reduce the dye, methylene blue, MB, to its colourless form, leuco methylene blue, LMB, whilst at the same time it is capable of oxidising a sacrificial electron donor such as glucose, GluH.
TiO2 → TiO2 (e-, h+) (1)
TiO2 (e-, h+) + GluH → TiO2 (e-) + oxidation products (2)
TiO2 (e-) + MB → LMB (3)
If dissolved oxygen is present – introduced into the solution by shaking the reaction solution with the air headspace – then the LMB is oxidised to MB and the original blue colour returns:
2LMB + O2 + 2H+ → 2MB + 2H2O (4)
The simple irradiation of a still dispersion of TiO2 particles in a solution containing both glucose and MB with UVA light will photo bleach the MB, via reactions (1) – (3) and the shaking of the reaction vessel with regenerate the original blue colour solutions (4).
The kinetics of photobleaching of this system is well described by a lLangmuir-Hinshelwood type equation:
Rate (~ 1/ttb) = kIθK[MB]/(1+K[MB])(K’[glycerol]/(1+K’[glycerol]))
Where ttb = the time to bleach the solution, I = incident light intensity, θ = power term – typically 0.5-0.6 and k, K and K’ are empirical constants.
For further information see the paper:
Mills, A.; Lawrie, K.; McFarlane M. Blue Bottle light: Lecture demonstrations of homogeneous and heterogeneous photo-induced electron transfer reactions, Photochemical and Photobiological Sciences, 2009, 8, 421-425.
Materials:
Stock Solutions
- [MB]stock = 2.5 x 10-3 M; 50 mL
Dissolve 0.046 g methylene blue in 50ml distilled water
- [glucose]stock = 0.25 M; 100 mL
Dissolve 147.5 g of alpha-D-Glucose in ~110 ml of water, using a sonic bath, then make up to 250ml mark in a volumetric flask.
- [TiO2]stock = 0.038 M; 100 mL
Disperse 0.3 g TiO2 (P25) powder in 100 ml distilled water.
Equipment
- 28 ml McCartney Bottles can be purchased from VWR (see: http://uk.vwr.com/app/catalog/Product?article_number=215-0861)
- UVA Lamps (2 × 8W BLB) as the irradiation source can be purchased from Vilber Lourmat (see: http://www.vilber.com/products/UVlamps0.html)
- A glass sheet is placed over the face of the lamp to provide a surface for the bottle to rest on.
- This glass sheet is, in turn, covered by a piece of black card with a hole cut in it for the bottle to sit in so that light emerging from the lamp is directly incident on the bottle’s base and there is very little stray light.
The video demonstration
To a 25 ml volumetric flask add: 1 ml [MB] stock, 15 mL [glucose] stock and 3 ml [TiO2] stock then make up the mark with water and mix by vigorous shaking. Place 12.5 ml of this solution into a McCartney bottle and screw on the top. Irradiate the solution with UVA light using the UVA light photoreactor and note the time taken for the solution to photobleach, ttb. Shake the sample 20 times and note the time taken for the original blue colour to return during this dark step, t(dark).
Figure 1: Photobleaching from 0 – 10 minutes. The video shows both the photobleaching step and the slow dark recover step
Further Experiments
Experiment (2)
Prepare a standard cloudy solution as in experiment (1) but this time place a 5 layer stack of plastic filters underneath it to attenuate light. Irradiate the solution and determine ttb. Repeat the experiment with stacks of 10, 15 and 20 plastics filters. Note down ttb and Irel values.
Irel = (X)n / 100 where X is % reduction of light intensity by filters and n is the number of
- stacks of 5 filters.
It was found that as more filters where stacked underneath the reaction vessel the time for photobleaching increased from ca. 6 minutes to ca. 15 minutes. This was due to the fact that the filters attenuate the light so with less light getting through it takes longer to photobleach. In this system the incident light intensity is high and so electron-hole recombination is a significant process; thus, a plot of rate (1/ttb)/(min) vs Irelθ, where θ = ca. 0.5-0.6 generates a good straight line.
Experiment (3)
Prepare a standard cloudy solution as in experiment (1) but this time add only 0.25mL of the glucose stock solution. Irradiate the solution and determine ttb of the solution. Repeat the experiment with 0.50, 0.75, 2.3 and 11.4 mL of glucose solution. On each occasion ttb is noted down.
It was found that when only very small quantities of glucose were used in the reaction the reaction solution bleached very slowly, ca 18 minutes, due to the formation of LMB, but bleaching time decreased to ca 5 minutes when the quantity of glucose was increased. The kinetics fit a Langmuir-Hinshelwood equation and so a plot of ttb vs 1/[GluH] generates a good straight line.