As I said yesterday, I prefer to dissolve my compounds in a relatively non-polar solvent before adding them to the flash column. Unfortunately this isn’t always possible, especially when running in hexanes/ethyl acetate . The solution is to dissolve your solvent in something quite polar (ie. acetone), then adsorb it onto silica. The silica can then be added to the top of the flash column, and the setup run as normal.
The process in detail:
Step 1. Dissolve your compound in a minimum of solvent.
Straight-forward enough. As best you can try to keep the volume of solvent to a minimum.
Step 2. Add in flash silica.
The objective is to form a thick slurry of silica/solvent/compound. As the solvent evaporates the compound and silica will be in close proximity, ensuring adsorption. There doesn’t seem to be a standard silica:compound ratio, so measure qualitatively (next step).
Step 3. Remove the solvent.
Using a rotovap and medium vacuum (~15 Torr) remove the solvent . Avoid the use of a hivac, as fine silica particles will run through the line directly into the pump.
Once the solvent is removed lightly tap the RBF against a stiff surface (ie. the bench). With a good silica:compound ratio the silica will slide of the walls and rest in a free flowing pile (this will appear to boil under reduced pressure). If the majority remains on the wall, resuspend in solvent and add a little more silica. Keep the amount of added silica below ~1.5 cm of column height, as with more than that you’ll start compromising the resolving power of the system.
Step 4. Add the silica to the column.
With a spatula scrape off any stubborn silica that is clinging to the RBF, then with the help of a small funnel pour everything onto the column. Rinse the RBF with your less-polar solvent and add the suspension via pipette. Once the silica has been solvated sand can be added and the column run as normal.
When I was first learning flash chromatography I generally got better results with columns that were dry loaded (likely because of excess wet loading solvent). There are those in the lab who still dry load the vast majority of their compounds, regardless of solubility, but in the end this process is simply too time consuming for my tastes. Even in skilled hands it will still add at least five minutes to the chromatography process, time better spent elsewhere.
 Be careful here. If the solubility of your sample in the column solvent is below the volume of 1-2 test tubes the product will elute over multiple fractions, simply because it won’t be able to dissolve into the column. From experience, this usually means 10+ fractions.
 Removing the solvent can moderately activate the intrinsically acidic silica gel. If you are concerned about acidic degradation it is generally a good idea to add a weak base like triethylamine to the mixture, or use an inert solid like florisil or celite.