, , , ,

*It’s time for the series of posts that cannot possibly go wrong.  Every day this week I’ll be talking about a new aspect of flash chromatography.*

AFC - Column Chromatographysk five organic chemists for chromatography advice and you’ll get at least six contradictory opinions.  Everyone has their own way of doing things.

With that said, I’m proud of my method.  It’s designed to get things done in the bare minimum amount of time, without sacrificing resolving power.  I’ve yet to find someone who looks forward to their weekly allotment of columns, so I think that’s about the best you can hope for.

For those familiar with flash chromatography, I slurry pack my columns, adding compounds via wet load.  I learned the basics from Kyle Finchsigmate (1, 2, 3, 4, without swearing here), the Advanced Practical Organic Chemistry book, and the original Still publication.  Today we’ll cover the essentials of loading and running a flash column, moving into the finer details later in the week.  If you have an improvement or find something unclear let’s hear it about in the comments.

Setting up the Column

Cotton Plug and Glass Rod. The pipette bulb is shown for scale.

Mouseover for text.

Begin by measuring out fifteen centimetres (6 inches) of flash silica into an empty column.  Pour the silica into a 250mL Erlenmeyer, then add in enough low polarity solvent to cover the silica [1].  With a long glass rod from the glassblower’s shop and a small piece of cotton, lightly plug the end of the column.  This plug should have no effect on solvent flow rates, and is simply there to prevent sand/silica from flowing through the stopcock.

Plug and Sand layer.  Sand should stop when the walls of the column are parallel.With clamps at the top and bottom, anchor the column to a stand or (preferably) the fumehood’s gridwork.  Take care to ensure that the column is perfectly vertical; this will be important later.  Add in enough coarse sand to cover the bell of the column, then fill the column roughly 5 cm high.  Tap the outside of the column with a cork ring to resettle the sand [2].  While swirling the Erlenmeyer flask pour in your silica, then rinse the flask and add any stubborn residue to the top of the column. Put the flask under the column’s spout to collect solvent waste.

With 1-3 PSI of pressure bring down the surface of the solvent until it rests just above the layer of silica (use a pressure adaptor) [3].  If the silica is uneven or slanted correct the angle of the column and then tap the outside near the surface of the silica with a cork ring.  In extreme cases you can use the long glass rod to agitate the top 2 cm or so of the silica.

Adding Your Sample

Wet load.  Note the dimple in the silica.  This occurs when solvent is added too quickly.Dissolve your sample in the lowest polarity solvent (mixture) that it is soluble in (ex. 9:1 Hexanes/EtOAc for a EtOAc/Hexanes column).  If the compound has only moderate solubility in the column solvent use DCM or toluene, provided the Rf in these solvents stays below 0.3.  If those solvents give a high Rf you’ll have to dry load (check back tomorrow).

Using a long Pastuer pipette, slowly add your dissolved compound to the top of the column.  To avoid perturbing the silica layer, bring the tip of the pipette as close to the silica layer as possible and rotate around the inner column wall during addition.  To prevent the compound from sinking into the silica before it’s evenly dispersed keep the column’s stopcock closed, and start with a thin layer of pure solvent above the silica (1-2 mm).  Once all the compound has been added let it absorb onto the silica under simple gravity flow, then rinse the flask and walls of the column twice.

Add a 5 mm layer of sand to the top of the column, then via pipette add solvent until you have a 3-10 cm (1-4 inches) head [4].  Once the solvent is high enough you can simply pore in more, as needed.

Running the Column

With the column at just the right height test tubes can be easily removed and replaced.  More convenient than moving the TT rack.Under pressure elute the column at a rate of 5 cm (2 inches) of solvent per minute, collecting fractions every thirty seconds.  Your compound should start eluting at around 1/Rf column volumes, where a column volume is height of your silica plug [5].  Spot fractions onto a TLC plate as you fill each test tube, and run plates two at a time in the TLC chamber.  Track the elution of your compound(s) with either ultraviolet light or a staining solution.

When your compound has eluted, increase the pressure to ~10 PSI and allow the column to drain dry.  As solvent-silica bonds are broken the silica will cool, and when the silica at the bottom of the column is cold (~15 min), silica will pour easily into the solid waste bin.


[1]  By “low polarity solvent” I mean the less polar solvent in your biphasic mixture.  If you’re running a hexanes/ethyl acetate column this would be hexanes; dichloromethane/methanol, dichloromethane.  The solvent is added early to prevent water from inactivating the silica, and may be overkill.

[2]  If your column comes with a coarse glass frit there’s no need for the cotton/sand.  The sand is there because in the bell portion of the column compounds traveling along the walls will have farther to travel than those in the middle, decreasing the effective resolution.  Fritted columns aren’t as good for DCM/MeOH columns though.

[3]  I’ve heard if allowed to settle slowly the silica will form bands of differing density, reducing resolution.  I haven’t empirically tested this though.

[4]  Larger columns require a higher head, while simultaneously taking significantly more solvent to get there.  To speed things up put down a thick sand layer (~1.5 cm) and add solvent via wash bottle.

[5]  Ie. One column volume is 15 cm of solvent.  This is distinct from “void volume” the amount of solvent contained in the silica.  Move of a rule of thumb than an exact measurement.