Methanol vs Acetonitrile

[adam]

I want to ask a question about methanol vs acetonitrile.

In our laboratory when we are trying to optimizing a separation, one of the things we often do is try different columns (C18, polar embedded C8, phenyl, pentafluorophenyl). We find this approach to be pretty easy and effective.

We usually use water and ACN as the mobile phases. However, it has recently occured to me that methanol may be a better choice than ACN in that it would be more likely to generate differences when these different columns are used. Of course you get higher back pressure.

What do you think?

Any thoughts or comments would be much appreciated (main question is what's in bold).

[Uwe Neue]

If you are only looking at C18s, there is a larger difference between the selectivity of methanol versus acetonitrile than between different C18s. If you are looking at other option, such as packing with embedded polar groups, the selectivity differences between methanol and acetontrile are as large as those between C18 and EPG columns.

I can send you some publications on this.

[Rob Burgess]

Hi Uwe - an interesting thread - I'd appreciate a link to these published results as well please....

[Uwe Neue]

Hi Rob, here we go (the first 5 are probably the most relevant ones for this discussion; I have more publications that cover column selectivity):

U. D. Neue, “Stationary Phase Characterization and Method Developmentâ€

[adam]

I read through the papers. One of the interesting things was the unique behavior of C1 columns. They retain more than you would think; and show good selectivity differences vs things like C18 and polar embedded C8.

Would you agree with the general conclusion that although Methanol shows somewhat more variability than ACN; the reality is that changing stationary phase has a greater effect than changing mobile phase. And therefore the following strategy makes sense:

- start with ACN (because it has other benefits) and try the following stationary phases: C18, polar embedded C8, phenyl, C1. Then - if more optimization is needed - take the stationary phase that worked best, and try different pHs or Methanol in the mobile phase.

Any suggestions or comments on my strategy would be appreciated.

[Uwe Neue]

I agree with your overall strategy. There are benefits to start with MeCN, and your column selection is reasonable. However, do not underestimate the solvent choice. You get good differences between MeCN and MeOH.

[adam]

Great - Thanks.

[Rob Burgess]

I would advocate a different strategy perhaps. Start with C18 and AcN first off as a 1st intent (gradient) and if it that fails a C18/MeOH gradient. If that fails then sequntially screen polar embedded and then phenyl?

The key question here is: where do the biggest selectivity benefits come from: mobile phase or stationary phase? can this be answered in terms of the hydrophbic subtraction model as to what gives the bigger affect on selectivity?

For my most recent method development problem, the bigger affect was clearly mobile phase - changing from AcN to MeOH resulted in clear baseline Rs between a key degradant and our main peak.

[Uwe Neue]

With a proper selection of the stationary phases, you can get as much selectivity difference from switching the stationary phase as from switching from methanol to acetonitrile. You will NOT get much of a difference between one C18 compared to another C18, or from a C18 to a C8. Selectivity differences as large as switching from methanol to acetonitrile were obtained when going from a C18 to a packing with an embedded polar group or to a phenyl column.

This is the essence of the study, which included some 450 data points for each comparison and a quantitative measure for the selectivity difference.

[Alex Buske]

[quote]This is the essence of the study, which included some 450 data points for each comparison and a quantitative measure for the selectivity difference.[/quote]
Hi,
I think I read most of the papers. and would have liked a different evaluatiom of the huge data set.
As with most of the "general column selectivity" or "Orthogonality" studies a large number of different compounds is run on a large number of columns under different conditions. The evaluation is done based on all compound.
I am certain that most of the variability accounts for the variability between groups of coumpounds, not within the groups coumpounds.

A few months ago someone told me "Neues study shows that the pH has the highest selectivity impact" Thats true as long as you look at all compounds. If you are focussed on either neutral compound or on weak acids or on a buch of compounds with similar pKA thenthere not influence of the pH.
If a evaluation was restricted on just one "compound groups" at a time conclusions wre much different.

So in method development I am looking at different C18 and C8 columns with ACN and MeOH and I certainly see differences between different C18-Columns. Finetuning is then often done using ACN/MeOH blends.

Alex

[Uwe Neue]

Alex, I do not agree. While most of the overall studies were done with a broad set of compounds, we also looked at subsets of compounds. Specifically, the influence of pH remains strong if one is looking only at basic analytes or only at acidic analytes etc. There is no question whatsoever that pH has the strongest influence on selectivity, even if I only look at a single group of ionizable compounds, be it aromatic and aliphatic acids, phenols, amines, anilines. It also reamins true for the case of analytes with multiple ionic groups.

A classical example of this case is the analysis of tricyclic antidepressants and metabolites, where one can get significant shifts in the elution order as a function of pH, from acidic to neutral to basic.

Even for analytes with an even more similar pKa than the tricyclics, there are significant differences in the selectivity between the ionized form and the non-ionized form.

I do not deny that there are differences between different C18 columns, but they are very small compared to the other options available.

[HW Mueller]

What Alex said about pH and neutral compounds, namely that there pH has no effect, one can say about all other aspects as well. For instance at 100% ACN and 100% MeOH there is no difference for practically all compounds. Or: for highly polar substances all C-18 columns will behave identically, no retention.

[adam]

Since this thread is still going, let me ask another important question.

It seems to me - from the papers previously mentioned - that there is not much of a difference between the performance of PFP columns vs phenyl columns. Some of the companies that sell PFP columns claim that they will show hydrogen bonding effects. But in a reversed phase eluent wouldn't hydrogen bonding effects be swamped out by the large amount of water?

They also claim PFP columns are particularly well suited for separating positional isomers. Is anyone aware of any documentation to suggest this.

Much Thanks!

[hajdaei]

This is one of the more useful threads that I have seen in a while. I want to ask a question about the comments above about how C1 columns behave in a unique and unexpected way. Can anyone explain why it is that C1 columns show more retention that would be expected and behave fundamentally different than, for example, C4 or C8 columns.

Thank You

[tom jupille]

Can anyone explain why it is that C1 columns show more retention that would be expected and behave fundamentally different than, for example, C4 or C8 columns.

Much easier access to residual silanols. I'd be willing to bet that you could use a C1 column with something like hexane/dichloromethane as the mobile phase and it would give normal-phase retention.