Ethanol as mobile phase

[thohry]

Hi all,

Why don't we use C2H5OH as organic solvent in RP HPLC instead of MeOH or others? It's safer and cheaper.

[bisnettrj2]

A quick search on Sigma Aldrich's website tells me ethanol is not cheaper than methanol. On Sigma-Aldrich, ACS-grade ethanol is $79/L, and LC.MS Chromasolv Methanol is $58.50/L. Maybe it depends on the grade, but I've never seen "LC/MS" grade ethanol.

http://www.sigmaaldrich.com/catalog/Pro ... =SPEC#test

http://www.sigmaaldrich.com/catalog/Pro ... KEY&F=SPEC

Plus, I think most ethanol is not sold 'pure' - it's denatured with isopropanol and methanol (usually) so people aren't tempted (maybe just in the US?) to create a fake laboratory account, buy cheap, pure, tax-free ethanol, and water it down to sell it as moonshine (purely conjecture on my part, but most sites I've seen have a 'taxable' versus a non-taxable price on ethanol, and they are significantly different).

As for safer? I'll give you that. But I think that's the only advantage, for now, in my opinion.

[HW Mueller]

Ethanol is available in HPLC grades, Merck (Darmstadt) for example. I have used it in HILIC as the aminoacids that I separated were supposed to be administered to patients, thus I didn´t want to have ACN anywhere near. Others have reported here to have used EtOH. In HILIC it proved to be a much stronger eluent than ACN. In RP it is probably not so popular, because the propanols allow a greater diff. to MeOH?.

[rhaefe]

Ah, the memories.

While working on reversed phase ion-pairing chromatogarphy of nucleic acids I evaluated several different organic solvents (acetonitrile being the standard). Following the spur of the moment I went down to the nearest BevMo in San Jose and bought a bottle of Stoli vodka, 100proof:
MP A: 0.1mol/L triethylammonium acetate (TEAA), pH=7
MP B: 0.1mol/L TEAA, 25% Stoli

It worked like a charm. To my surprise the vodka was really clean (at 254nm). No baseline noise or ghost peaks. I have to see if I still have the original data somewhere...
I do not recall what happened to the remaining solvent.

So ethanol does work but it is not cheaper than MeOH, mainly due to taxation.

Robert

[JGK]

I've used Ethanol for mobile phases before however, this was in the UK where the use of ethanol for chromatography purposes was not subject to tax and customs and excise duty.

I came to Canada where tax & duty are charged and consequenty the ethanol was prohibitively expensive and rarely used.

[bisnettrj2]

rhaefe - were you able to get reimbursed for the Stoli by the purchasing department? Because New Year's Eve is coming, and I might be able to find a way to run my instruments and make punch at the same time, all on the company dime!

[adam]

Ethanol has about twice the viscosity of methanol. Therefore, twice the backpressure.

[thohry]

Ethanol has about twice the viscosity of methanol. Therefore, twice the backpressure.

Yeah, probably this is the main reason. But now they have UHPLC that can raise pressure up to 1000 bar.

[bisnettrj2]

True, but UPLC doesn't make the solvents cost less (or the instrumentation, for that matter!). I would also imagine that the pressure difference between an HPLC method with water:methanol and water:ethanol would be comparable in UPLC. And being that UPLC uses small particle columns, I bet you'll be testing the upper limit of the pressure range if you use a high viscosity solvent.

[Bintang]

The increased viscosity also reduces column efficiency quite a lot.
Going from Acetonitrile to 1-Propanol in HILIC (80% organic) reduced column efficiency 50-60%. Would likely be less for Ethanol.

[HW Mueller]

Bintang, how did you isolate the influence of viscosity from the other physical differences between 1-propanol and ACN? I mentioned above that EtOH is a much stronger eluent than ACN in HILIC, most likely, propanol is stronger than ACN also. The efficiency is down if one has trouble getting retention.

[Consumer Products Guy]

I have used HPLC grade ethanol occasionally (EMD Reagent Alcohol, HPLC/Omni). The one I've used contains about 5% methanol and 5% isopropyl alcohol as denaturants, so one must be aware that each lot of this might vary slightly in composition. It's 200 proof as that just means all alchols content.

1. So one should not mix different lots mid-stream of an analysis.

2. When left in a reservoir and vented, over time the compsition will change slightly as the components selectively evaporate. Pure methanol, pure ACN, etc., wouldn't have this issue.

I've never used 100% ethanol as the mobile phase, or even "100%" with denaturants other than the alcohols listed above.

[bisnettrj2]

You could combat the pressure increase due to the increase in viscosity by operating at higher temperatures. Obviously, this will be limited by the the temperature restraints placed by your column manufacturer. And, as a stronger eluent, you should be using less ethanol than methanol in your mobile phase. So perhaps it would be a wash in terms of cost?

So, forumers, riddle me this - if ethanol is a stronger eluent than ACN, then why is ACN used in the pharma industry over EtOH, especially in light of its greater safety as a solvent? I'm sure the answer is cost or selectivity, or perhaps USP (or other international regulatory) conventions and requirements for validation, but I am curious.

[HW Mueller]

I am not sure how water-ethanol mixtures behave as far as viscosity is concerned, but I never even had a hint of a viscosity problem.
Evaporation problems are present for all HPLC mixes, not just for an alcohol mixture.
When I was in the US we had no problem getting 95% alcohol (5% H2O, no denaturants), if memory really doesn´t play tricks, then it seems that 100% was also available to chemists.
We had special police come in to our lab at the U of Colorado once. They asked for what we use the LiAlH4 which we ordered all the time. They only knew about its use in a reduction synthesis step toward a drug (LSD??). . . . . . .

[Bryan Evans]

Maybe there's a place in the world for RP using EtOH for 5um or monolith columns. It has a pretty high UV cutoff.
I think very sensitive RP HPLC analysis is unlikely with EtOH.