This technique was developed by L. M. Harwood (Aldrichimica Acta, 1985, 18, 25), and can be used as an alternative to flash chromatography. The apparatus simply consists of a parallel-sided vacuum filter funnel. incorporating a porosity 3 sinter and a flask.
Method for running a dry flash column
1. Column packing
Fill the funnel to the lip with TLC grade silica (e.g. Merck Kieselgel 60) and tap gently. then apply suction from a water aspirator. pressing the silica down starting at the circumference and working towards the centre. Continue until a level and firm bed is obtained and there is a head space for sample and solvent addition. The approximate funnel sizes. compared with the quantity of sample to be applied, are given in the table below.
| Funnel Diameter |
Funnel Length |
Silica Weight |
Sample Weight |
Fraction Size |
| 30mm | 45mm | 15g |
15-500mg |
10-15ml |
| 40mm | 50mm | 30g |
0.5-2g |
15-30ml |
| 70mm | 55mm | 100g |
1-5 g |
20-50ml |
2. Pre-elution
Under vacuum, pre-elute the column with a solvent which will give a TLC Rf of about 0.2 for the least polar constituent of the sample. If the silica has been packed correctly, the solvent should run down the column with a horizontal front, but if it channels, the column should be sucked dry and re-packed. Keep the surface of the silica covered with solvent while pre-eluting until solvent starts collecting. then suck it dry.
3. Loading the sample
Load the sample as a solution. in the same solvent as used for pre-elution. in an even layer onto the surface of the silica. Alternatively. if the sample is insoluble in the pre-elution solvent. it can be pre-adsorbed onto a small quantity of silica which is then spread on the surface of the silica in the funnel.
4. Eluting the column
Sequentially add solvent fractions to the funnel according to the quantities indicated in the table, sucking the silica dry in between each fraction, and keeping fractions separate. For each successive fraction increase the solvent polarity by increasing the proportion of the more polar solvent by about 5-10% (e.g. from 50% EtOAc/50% petrol to 55% EtOAc. for a 10% increase). Analyse the fractions by TLC to determine the locations of the components of interest, but as a rough guide. the solvent mixture which would give the compound a tlc Rf value of about 0.5 will probably elute it. As with any chromatographic technique expertise will only come with experience. but given that. you should be able to separate quite closely running compounds quickly using this technique.