How do I choose between Normal- or Reversed-phase flash column chromatography for my compound purification?

How to choose between normal- and reversed-phase flash column chromatography is an excellent question and one that my readers often ask.  Those who use column chromatography know that as long as the reaction products or compounds are fairly non-polar and near neutral pH they will have successful purifications.  However, when your mixture’s chemical characteristics are more challenging (polar, non-polar, basic, acidic) there are other options that are available to successfully separate pure compounds.

In this posting I will discuss the criteria you can use to guide your choice between normal- or reversed-phase flash chromatography.

There is nothing more important to purification success with chromatography as compound or sample solubility. Concentrating on just the most popular solvent system of hexane and ethyl acetate may send you down the wrong purification path (is nothing ever easy with chromatography?).  Well, in this case it can be easy with just a little chemistry knowledge. If you need to purify a synthetic reaction mix, then you already know your product’s structure as well as those of your starting materials. So, the main question is – in which solvent did you perform the synthesis? The answer to this question can guide your flash purification decision.

Organic Synthesis

How, you may ask. Think about it.  If your synthesis is performed in DCM, CHCl3 or other relatively non-polar solvent, then your compounds are likely lipophilic but with some polar functionality and amenable to normal-phase chromatography.  If your reaction solvent is DMSO or DMF, the products created may be lipophilic but getting them out of the polar solvent, may be a challenge so why not simplify the purification by using reversed-phase?  If the reaction products are polar, definitely use reversed-phase. As always, try normal-phase TLC with various solvent systems to verify your the proper purification mode.  If your compound of interest has a normal-phase TLC Rf under 0.5, then it is likely amenable to silica flash column chromatography.  Above 0.5, I would look at reversed-phase.

RxN solventTLC Rf <0.5TLC Rf >0.5
DCMNormal-phaseReversed-phase
CHCl3Normal-phaseReversed-phase
ACNNormal-phaseReversed-phase
DMFN/AReversed-phase
DMSON/AReversed-phase
NMPN/AReversed-phase

A good rule of thumb is if the sample or compound is soluble in DCM, EtOAc, or similarly water-immiscible solvents then normal-phase is likely the best approach. If the compounds are soluble in water-miscible solvents then reversed-phase is likely the best way forward. Alkane soluble compounds – what do you think? (answer later in the post)

Natural Products

In the natural product world, base your flash purification choice on extraction solvent choice – DCM, EtOAc, and other non-water miscible solvent extracts will most likely contain compounds best purified by normal-phase. Alcohol and water extracts are best purified by reversed-phase. Hexane extracts will likely contain the most lipophilic compounds which, in many cases, will not retain and separate well on silica so I typically use reversed-phase, sometimes even non-aqueous reversed-phase to separate these compounds, e.g. carotenoids.

Extraction solventTLC Rf <0.5TLC Rf >0.5
Hexane (or similar hydrocarbon)Normal-phaseReversed-phase
DCMNormal-phaseReversed-phase
EtOAcNormal-phaseReversed-phase
AcetoneNormal-phaseReversed-phase
MeOHNormal-phaseReversed-phase
H2ON/AReversed-phase

In the case where non-polar or even moderately polar solvents (e.g. acetone) are used as extraction solvents, again, let TLC be your guide.  If any of the extracted compounds run with the solvent front or have Rf values > 0.5 in 100% hexane (or heptane, cyclohexane, etc.) then consider reversed-phase.  If you achieve separation of compounds in 50% or less EtOAc (in hexane, heptane, or other similarly lipophilic co-solvent) then pursue a normal-phase purification solution.  Above 50% polar modifier, I suggest reversed-phase.  Remember if it doesn’t stick to silica it will stick to C18 (a commonly utilized reversed-phase sorbent). Likewise, if it sticks to silica too well, reversed-phase will likely do a better job of separating the compound from its impurities. As a reference I have summarized these suggestions in the table below.

Normal-phase - Reversed-phase selection criteria

RxN/Extraction solventTLC Rf <0.5 (normal-phase)TLC Rf > 0.5 (normal-phase)
AcetoneNormal-phaseReversed-phase
AcetonitrileNormal-phaseReversed-phase
Dichloromethane (DCM)Normal-phaseReversed-phase
Dimethylformamide (DMF)N/AReversed-phase
Dimethylsulfoxide (DMSO)N/AReversed-phase
EtherNormal-phaseReversed-phase
Ethyl acetate (EtOAc)Normal-phaseReversed-phase
Hexane (or other alkane)Normal-phaseReversed-phase
Methanol or other alcoholsReversed-phaseReversed-phase

What approach do you take to chromatographic mode selection?

Is there anything you have learned about choosing either reversed- or normal-phase that can help educate other readers?

Published by

Bob Bickler

Technical Specialist, Biotage

5 thoughts on “How do I choose between Normal- or Reversed-phase flash column chromatography for my compound purification?”

  1. I have a collaborator who has an Isolera Flash purification system that he uses for normal phase separations. He mentioned the C18 Snap cartridge for reversed phase purification when we were discussing my search for a semi-prep HPLC system in our department (I found none). He could not answer two questions: on an Isolera system how much backpressure would develop using 10g and 25g RP C18 cartridges? How hard is it, and how harmful to the system is it, if you switch from normal phase (ethyl acetate and hexanes) to reversed phase and back again (assume acetonitrile-water for the RP solvent system)? What do you need to watch out for? Thanks.

    1. Hi Louis,

      First, many thanks for reading my blog and second, thank you for the really good questions.
      To answer your first question on pressure for 10 and 25 gram C18 cartridges will generate higher pressures than equivalent sized normal phase cartridges; this is due to the higher viscosity of the water-based solvents as I am sure you know. In studies I conducted in 50:50 MeOH/H2O on these size cartridges with our KP-C18-HS I noticed the back pressures below (acetonitrile/water solutions will generate lower pressure).

      12 g 25 g
      25 mL/min 34 psi 21 psi
      50 mL/min 55 psi 32 psi
      75 mL/min 72 psi 46 psi
      100 mL/min 59 psi

      To answer your second question, switching between modes is quite simple. You mentioned hexane/ethyl acetate for NP and acetonitrile/water for RP. All you need to do to switch is
      1. go to setup and press prime
      2. Prime without cartridge 50 mL of EtOAc
      3. Prime with 50 mL of CH3CN
      4. Prime with 50 mL of your starting mobile phase

      To go back to normal phase
      1. Prime with 50 mL of CH3CN
      2. Prime with 50 mL of EtOAc
      3. Prime with your starting mobile phase

      The design of our pump allows this solvent switching and will not harm the Isolera. The only thing to watch out for is ensuring solvents are miscible with each other as you move from one priming solvent to another.

      Bob

  2. Hi Bob,

    This is very helpful specially to know that if Rf > 0.5, should choose reverse phase. I had an issue with this kind of purification and after reading your blog I tried reverse phase and it did work for me.
    Thanks for posting good knowledge.
    Aaliya

    1. Hello Aaliya,

      Many thanks for your reply. I am happy that my post provided useful information. If you think about the relationship is between TLC Rf values and flash chromatography elution volumes or column volumes (CV) you will see why this suggestion makes. As you know the number of column volumes required to elute a compound is equal to 1/Rf and if your compound has an Rf of 0.5 then it will elute within 2 CV with an isocratic method. This tells us that there is very little retention on silica for this compound and that something different needs to be tried; reversed-phase is usually a good alternative.

      Bob

Leave a Reply

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