How do I purify ionizable organic amine compounds using flash column chromatography?

For most organic reaction mixture purifications the process is fairly straightforward. Use hexane/ethyl acetate or, for polar compounds, DCM/MeOH.  But what do you do if this doesn’t work and your compounds are basic organic amines?

In this post, I re-examine the options available to achieve an acceptable organic amine purification when typical separation methods are insufficient.

In chromatography, the most important parameter to optimize is selectivity.  Selectivity is the amount of separation between any pair of sample compounds. Therefore, it is beneficial to maximize selectivity (or separation) to ensure the best fraction purity and yield. Without selectivity, there is no separation.

Normally, the differences in sample compound chemistry along with the column media chemistry (stationary phase) and elution solvent chemistry (mobile phase) is enough to create selectivity and therefore a separation. However, sometimes the chemistry differences between individual sample compounds is slight or the compounds’ functionality is reactive or ionizable.

These situations complicate the separation and when they arise, method changes are needed.  These changes can be in the form of the solvent system (changing solvents), the stationary phase (changing stationary phase), or both.

Organic amines, especially heterocycles, tertiary, and secondary amines that are basic can be quite challenging to separate and purify.  If using typical normal-phase chromatography with silica columns the biggest problem is an acid-base interaction between silica (a BrØnsted acid) and the basic amines. This interaction can cause compound degradation, yield loss, and/or increased bandspreading.  To counteract these problems you can employ one of the three options listed below.

A.      Add a competing amine to the mobile phase

By incorporating a base in the elution solvent for both equilibration and the actual purification the acidic silica silanols are “neutralized” by the small, competing amine mobile phase modifier.  Competing amines include ammonia, triethylamine, pyridine, ammonium hydroxide, etc in a mobile phase of DCM/MeOH. This, however, can be challenging to get to work due to the strong displacement effects of both methanol and the added base, so I prefer option B.

B.      Change column media

To stay in a normal-phase purification mode you can replace your silica column with one packed with either basic alumina or an amine-functionalized silica.

Amino-silica has the advantage of maintaining the same particle size as standard silica (alumina typically has a much larger particle size) but because it is bonded with an organic amine, the surface is now “basic” and organic amines separate using “softer and safer” solvents such as hexane/ethyl acetate or ethyl acetate/IPA – no added base is required.

In Figure 1, we show an example of a DCM/MeOH purification of three tricyclic amines. The elution gradient originally programmed (0-20% MeOH in DCM) was based on TLC data which showed a separation of the three compounds without any amine modifiers. So, using a silica column and no any modifier I used the suggested gradient. Well, the results are not what was expected or needed as nothing eluted using the programmed method. So, to try to elute the compounds I increase the methanol % which eventually eluted the compounds in a single peak.

Figure 1. DCM-MeOH gradient of three basic tricyclic organic amines based on TLC data at 5% MeOH/95% DCM provides no separation of the compounds.

That result prompted me to switch from silica to an amine-functionalized silica. As can be seen below, this column chemistry works by providing a chromatographic environment much more conducive the organic amine purification and I was able to use hexane with ethyl acetate as my solvents, Figure 2.

Figure 2. An amine-functionalized silica column (Biotage® KP-NH) successfully separates the three basic, tricyclic organic amines using a hexane/ethyl acetate gradient.

C.      Use reversed-phase

Reversed-phase HPLC or UPLC is typically used to monitor a reaction’s progress.  That being said, if it works on an analytical scale why not use it for preparative flash chromatography? It can be useful for polar, ionizable, and lipophilic compound purification.

Basic amine compounds are best retained and separated when the mobile phase pH is alkaline and in the free-base form.  At high pH, these compounds become more lipophilic and therefore more retentive increasing the probability of a successful purification. To achieve this goal I like to employ the 2 pH rule which states to deprotonate an amine, adjust the solution pH to two units above the amine’s pKa. For this I like to use a volatile base (easier to evaporate) at the 0.1% level in both the weak and strong solvents. Examples of volatile bases include…

i.     Ammonium hydroxide

ii.     Diethylamine (DEA)

iii.    Triethylamine (TEA)

In Figure 3, we show an example where three organic amines, 4-dimethylamino antipyrine (pKa 9.6), phenylbenzyl amine (neutral), and amitriptyline (pKa 9.4) are separated using a gradient incorporating TEA as the base. By increasing mobile phase pH compound hydrophobicity is improved and retention increases compare to a neutral pH mobile phase or an acidic mobile phase. Another benefit to purifying amines in an alkaline solvent is you use more organic solvent (because of the increased compound hydrophobicity) which then is easier to evaporate post purification.

Figure 3. Organic amines purified using reversed-phase gradient flash chromatography. Solvent used were acetonitrile, water and 0.1% TEA. Peak 1 dimethylamino antipyrene, Peak 2 phenylbenzyl amine, Peak 3 amitriptyline.

So, the moral of the story is to create purification methods best suited for organic amines, choose an appropriate stationary phase (e.g. amino-functionalized silica, reversed-phase) and the right solvent conditions (pH, buffers, solvent chemistry/functionality) and you will improve your purification success rate.

Do you have other tips for purifying organic amines?

For more information on flash method development and optimization, click on this link and register.

Also, please take our survey regarding this blog; I’d love to see your thoughts.



Published by

Bob Bickler

Technical Specialist, Biotage

4 thoughts on “How do I purify ionizable organic amine compounds using flash column chromatography?”

  1. Hey Bob,

    What do you think might be the reason that your TLC separation didn’t translate to the column?
    Regarding normal phase mode chromatography, I often use 2M ammonia in MeOH (anhydrous) mixed with DCM at various ratios (typically 1-10%). Many times it works wonders.
    When running RP mode. Addition of formic acid also gives you usually good separation. And this is a generic method we use for LC-MS so you already have an idea how your separation will look like. And there is also an ion-pair mode with TFA. Then you end up with TFA salt, but SCX afterward can take care of that.
    Great blog. Keep up the good work.

    1. Hi Tomasz,

      You ask a really good question. In my experience, DCM/MeOH solvent blends used with TLC often do not translate to flash. I believe this is due to the fact that there is no pre-wetting of the silica on the TLC plate whereas with flash the columns are usually pre-wetted. So, when TLC is performed the methanol is preferentially adsorbed to a greater extent on the dry TLC silica as compared to the wetted flash silica. So, with the methanol removed (or its concentration greatly reduced in the TLC solvent), the compounds are better able to interact with the TLC silica and separate. That is my theory anyway.

  2. Amine additives are very useful, but not so long ago I made the mistake of storing a pre-mixed solution of triethylamine in dichloromethane. The grams of wonderful white solid I isolated after chromatography seemed like overkill for my milligram-scale reaction. Alkylation of amines (at least triethylamine) with dichloromethane to give ammonium salts occurs slowly over time.

    1. Chris,

      You bring up a great point about storing base (and likely acid) modified solvents – reactions can occur. This is one reason I like to have the pH modifier solution as a totally separate solvent and add it isocratically while running a gradient with the other two solvents.


Leave a Reply

Your email address will not be published. Required fields are marked *