How do I decide between normal- or reversed-phase flash column chromatography?

Choosing a good purification strategy is an important for successful crude compound purification. A major factor in your strategy is choosing between normal-phase or reversed-phase chromatography.  How do you choose?

In this post, I will provide some simple guidance on helping determine which route to take.

In my role at Biotage, I am frequently supplied with compound structures from chemists and then asked – how do I purify this? So, with that question I open a dialog with the chemist to understand purification goals and learn more about the compound’s properties.

Where I like to start is with the question of solubility – in which solvents is your compound most soluble.  This typically tells me whether normal- or reversed-phase flash chromatography is most likely to work.

A rule of thumb I use is if the sample is organic solvent soluble (DCM, EtOAc, toluene, ether, etc.) then try normal-phase.  If the crude material is soluble in polar solvents (alcohols, DMSO, DMF, acetonitrile, etc.) then I first suggest reversed-phase.  If the sample is hydrocarbon soluble (hexane, heptane, cyclohexane) then I then to think reversed-phase, but normal-phase may also work. Table 1 provides my suggested purification routes based on crude sample solubilty.

Table 1. Sample solubility and flash purification route

SolventNormal-phaseReversed-phase
DMSOX
DMFX
WaterX
MethanolX
AcetonitrileX
EthanolX
PropanolXX
AcetoneXX
THFXX
Ethyl acetateXX
EtherXX
DCMXX
TolueneXX
Hexane (and other saturated hydrocarbons)XX

The reason I lean towards reversed-phase for the lipophilic compounds is that they usually have little functionality other than C and H, which can make them difficult to retain and separate on silica. Compounds that are organic soluble are also candidates for reversed-phase chromatography but should be dry loaded.

Organic amines provide a different challenge and, as many of you know, these compounds, especially if basic, are quite difficult to purity.  With organic soluble 2° and 3°organic amines and N-heterocycles, I steer away from silica to amine-functionalized silica as the chromatography is usually simpler and better.  For primary amines and quaternary amines, these often are soluble in polar solvents so I suggest reversed-phase for their purification.

As with most anything, there are always exceptions to the rule, but more times than not following these solubility guidelines will provide you with the right purification route.

How do you determine which purification route to use?

Published by

Bob Bickler

Technical Specialist, Biotage

8 thoughts on “How do I decide between normal- or reversed-phase flash column chromatography?”

  1. I look at the TLC, if I see nice round spots then it’s normal phase. If the spots are streaking, even with the addition of gHOAc or Et3N, yet the LC/MS shows sharp peaks, then I use reverse phase. A lot of our acidic compounds come in this category, don’t tend to make basic ones. Of course it’s a lot harder to evaporate reverse phase fractions. I only have a rotary evaporator but if I can see globules of product after the CH3CN has gone then I extract with CH2Cl2:MeOH, 90:10 and pass it through a phase separating cartridge.
    Sometimes a crude reaction is incredibly crude but I have to isolate super pure material for a particular purpose. It’s usually normal phase to isolate all the product plus some by-products, followed by reverse phase to isolate clean product. It’s unlikely that any particular by-product will co-elute by both techniques. Some people say they are orthogonal methods but I think they are just different.
    If a compound is water soluble and very polar then it’s almost always reverse phase. Only once in my life have I needed to resort to Aqueous Normal Phase with CH3CN (weak) and water (strong). I used a good quality normal phase silica and ran from 95% CH3CN/5% water to 50% water. The crude was dissolved in aqueous methanol, dry loaded and placed on top of a my column which had been equilibrated as if it was a poor man’s HILIC.
    I would guess 85% of my purifications are normal phase.

    1. Hi Derek,

      Thanks for sharing your experiences with us again, really good information. I am curious to know what type of compounds needed your “poor-man’s HILIC” approach.

      Bob

      1. Bob

        Can’t send you the structure but it was a small molecule with two amides and 4 hydroxyls. Was unretained on C18AQ with 100% water. Much less retention than uracil. It was starting material for the final product which could be purified by C18AQ, only just.
        The silica was YMC 50um spherical for the ANP.

        Derek

        1. Derek,

          I appreciate you sharing the information. I worked for YMC prior to moving to Biotage (after Waters bought YMC Inc. in 1997) so I know the capabilities AQ has for retaining many, but not all, polar molecules. Your approach using silica and aqueous solvents makes perfect sense to me. I have successfully used aqueous normal-phase (silica) to separate uracil and cytosine.

          Bob

  2. Another question, I want to do purity check of a C-18 Non-endcapped columns packing material. The column i have is used for many times. Is there any way or method i can check the purity. Some articles say by GC-MS i can check the purity but i did not found the MOA.And also the volatile content measurement of the C-18 phase column.

    Thank you.

    1. I assume you want to see if the column has some contamination that could potentially cause problems with subsequent separations. If the column has been used my guess is it has some contamination but sometimes simple solvent flushes can remove contamination.

      C18-bonded phases can tolerate strong organic solvents so I will wash my columns with a series of miscible solvents starting with the past method’s organic solvent (usually methanol or acetonitrile) and follow that solvent with acetone, ethyl acetate, and finally DCM then back to the method’s organic solvent. I find, at least for flash chromatography, that if the column is still usable even if it does not get totally clean. For your analytical column I would follow this protocol and analyze the collected solvent, as well as the pure solvent before flushing, by GC-MS and compare the data. If you find significant impurity levels perhaps a new column is in order. You may want to consider using a guard column to trap strong-binding compounds, this technique works well with flash as well.

      Bob

  3. Dear Bob, a wonderful article for newcomers like me in this field. I want to ask a basic question, don’t know whether its right to ask or not.
    If i have a unknown sample and to be extracted by flash chromatography, what are the first things i need to do?(I have a sample named as MK5, but i don’t know about its physical or chemical properties)

    Thanks in advance.

    1. This is an excellent question. For me, I test the crude sample’s solubility. If the sample is soluble in polar solvents like water, methanol,DMSO, DMF, I will use reversed-phase – this is also true if the samples are soluble in alkanes. However, if the crude mixture has solubility in DCM, ethyl acetate, acetone, or other organic solvents I will use normal-phase.

      Bob

Leave a Reply

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