Mobile phase & negative mode LC-MS analysis

[West Coast HILIC]

Hi all,

I've noticed many times that 5 millimolar ammonium acetate and ammonium formate buffers give some loss of sensitivity / signal suppression in negative mode (using electrospray LC-MS), even after re-optimizing the source conditions. In other words, your maximum negative mode ionization efficiency in the 5 millimolar buffered mobile phase is less than what it would be in 0.1% formic or 0.1% acetic acid. I believe this is an unavoidable consequence of the instrinsic nature of the ESI process (see this paper, for example):

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2104782/

but I've always wondered if there are any mobile phase additives that would compensate for this and improve ionization efficiency/sensitivity in negative mode.

Any thoughts? Do you have any tricks for increasing ionization efficiency/sensitivity in negative mode? Any mobile phases that you find work especially well?

[Rndirk]

In ESI (+) you typically go for the [M+H]+ ions, while [M-H]- are the ions you filter in ESI (-).

In reality it's more complicated, but the simple explanation is indeed that it's a matter of pH of the mobile phase versus pKa of the functional groups on your molecule that protonate/deprotonate. It's a good idea to list the pKa's of your target molecules.

You can try more basic mobile phase buffers for increasing sensitivity in ESI (-). Remember to stick to volatile salts, for instance ammonium carbonate.

[Gaetan Glauser]

Some have used NH4F at very low concentrations (ca. 0.5 mM) to increase sensitivity in ESI neg. Otherwise 0.05-0.1% acetic acid seems a good, standard solution for most applications.

[lmh]

Thanks for the reference! I've spent a lot of time (wasted a lot of time) trying to explain to chemists that basic buffers do not improve sensitivity in negative mode, because there is electrochemistry going on in the spray needle and it's not just a matter of adjusting the pH to maximise the ionisation of the compound in solution in the absence of the electrochemistry.
It's easy enough to test, and it may depend a bit on instrument and the particular design of the spray chamber, but I've had very poor experiences of trying to improve negative-mode sensitivity by manipulating pH.
The reference is nice because it provides an explanation.

The message I try to get through to people is to choose the pH/buffer to suit the chromatography, without trying to guess what will happen to the MS (except, of course, that the buffer must be volatile). I'm sceptical of TFA because it really does seem to cause huge suppression, but it's amazing how things ionise when they really shouldn't, based on their pKa and the pH of the buffer.

[Rndirk]

Thanks for the reference! I've spent a lot of time (wasted a lot of time) trying to explain to chemists that basic buffers do not improve sensitivity in negative mode, because there is electrochemistry going on in the spray needle and it's not just a matter of adjusting the pH to maximise the ionisation of the compound in solution in the absence of the electrochemistry.
It's easy enough to test, and it may depend a bit on instrument and the particular design of the spray chamber, but I've had very poor experiences of trying to improve negative-mode sensitivity by manipulating pH.
The reference is nice because it provides an explanation.

The message I try to get through to people is to choose the pH/buffer to suit the chromatography, without trying to guess what will happen to the MS (except, of course, that the buffer must be volatile). I'm sceptical of TFA because it really does seem to cause huge suppression, but it's amazing how things ionise when they really shouldn't, based on their pKa and the pH of the buffer.

I guess my first answer was incorrect. Interesting post.

[carlo.annaratone]

Addition of formic acid, or formate buffers, often leads to formate adducts in negative mode. They can be strong, reproducible, and fragment well (if needed for a tandem MS analysis). Have you tried looking for them?

[tommylife1]

The reference in this post is using a special compound, which is a amide (neutral). Why not use acidic compounds like lactic acid and see how it goes. It is easily detected in negative mode plus high pH mobile phase. No doubt. The classic chemistry works in general cases. The neutral compounds are not.

[lmh]

Honestly, it's the same with acidic compounds; adding a base doesn't improve their ionisation (it generally makes it worse).

The authors suggest a sensible reason:
In negative mode electrospray, in the presence of an acid chromatography additive, the acid rapidly deprotonates. The protons head off to the spray needle, which is attached to the negative side of the power-supply, so they pick up an electron and get reduced (to hydrogen). This leaves an excess negative charge, which is drawn to the surface of the spray droplets, and is made up of the conjugate bases of the acidic additive. These ions are perfect for abstracting hydrogen from any other acid present in the drop, e.g. the analyte.
The excess negative charge is also important in the spray process. In nano-flow it's possible to get a spray without any gas provided the excess negative charge is there to draw droplets out of the spray needle. Even with a nebulizer gas, the excess charge is what causes the droplets to break up as the solvent dries and the charges become concentrated on the surface, repelling one another.

If a base is added instead of an acid, then the base will gain protons, and the base itself will head off to the negative spray needle. Most bases are a lot harder to reduce than lone hydrogen ions (which are rare in basic solution), so it's harder to build up an excess of negative charge, and with no negative charge, there won't be any electrospray process.

[tommylife1]

but,practically, i used pH=8 in lactate, negative mode give best response, positive mode no lactate signal. It is canonical chemistry.

To be honest, I want to understand this really. Welcome to discuss.

[lmh]

Yes, negative mode is generally the best option for detecting acids.

But many people naively believe that making the buffer more alkaline will increase the efficiency of electrospray ionisation of acids in negative mode. This is very often not so. Very often, paradoxically, acids ionise better at low pH in electrospray, a pH where they would be less ionised in solution.

The original reference is an investigation into this, and an explanation of the underlying chemistry.

I would also look for lactate in negative mode. I would expect to see it ionise very nicely even if I used a buffer at pH 3.0 (or lower) - but I can't swear to a particular application.

The basic argument is that if I use dilute formic acid as my running buffer in negative mode for analysis of lactate, the formic acid will ionise, the H(+) will be reduced to hydrogen, leaving an excess of negatively-charged formate ions. These will aid the electrospray process by giving the droplets the surface charge that electrospray needs. They will also grab a hydrogen from any available lactic acid molecule, giving plenty of ionised lactate for me to detect!

I'm not a good enough chemist to know if this is true, but it seems plausible, and personal experience tells me that whenever I try to improve detection of acids in negative mode by using basic buffers, I get a signal that is the same or weaker (as well as having to choose a base-friendly column).

[tommylife1]

Thank you for the reply.

My understanding is for acids (lactic acid), if you add basic buffer (pH=8), the proton from acids will go off to neutralize OH from basic buffer.Thus leaving the excess conjugate form (lactate M-H) in the solution and spray. Does the conjugate lactate (M-H) help the charge-dependent spray formation? Also in the negative model, the electric field is applied opposite as the positive mode.

[Rndirk]

As you can read in the discussion above (especially the posts of lmh) the chemistry in the spray is more complex than simple acid-base pH versus pKa stuff.

We know from experience that acids generally give a much better signal in ESI(-), and amines in ESI(+), for the molecular parent ion.

It is straightforward in the sense that (RCOO)- is much more easily formed than (RCOOH2)+. Similarly, (RNH3)+ is favored over (RNH)-.

The fact that lowering the pH of the aqueous mobile phase can increase the signal for acids in ESI(-) is not straightforward (see above).