Recently, I received an email from a reader asking why formic acid was specified as an additive for the mobile phase in an HPLC method he was using. Formic or trifluoroacetic acid at 0.1% concentrations are common, especially for LC-MS work. There are a number of reasons for adding an acid at low concentration to the mobile phase. Let’s look at two of these: the influence on the column and the sample.
Silica acts as the support material for most reversed-phase columns. As we saw in earlier articles ('Silica Purity - Metals' and 'Silica Purity - Silanols'), the silica surface can vary depending on its source and how it is treated. However, we can consider the silica surface as somewhat acidic, with pKa values in the 4.8 region often cited for low-purity, type-A silicas, and higher pKa values for the high-purity, type-B silicas in more common use today. Ionized silanol groups interact through ion-exchange with ionized bases, and are a major factor in peak tailing for basic solutes. One way to minimize this problem is to suppress the ionization of the silanol groups. Thus, when the mobile phase pH is <≈3, the silica surface silanols are mostly suppressed, so cation-exchange interactions are minimal.
Carboxylic acids often have pKa values in the 4-5 range. When the pH is above this value, the acidic samples will be ionized. This results in shorter retention times because of the more polar nature of the ionized analytes. With type-A columns, which often contain significant concentrations of metal cations, ionic interactions with ionized acids can occur, causing tailing for acids. At pH <≈3, the ionization of carboxylic acids is suppressed, minimizing the possible ionic interactions with the silica surface and generally increasing retention because of the less polar nature of the non-ionized analytes.
So you can see that low pH favours improved peak shape for both acids and bases, as well as improved retention for acids. Unfortunately, low pH will cause ionization of basic samples and decrease retention, but hopefully there will still be sufficient retention at low pH for bases. We generally limit the use of silica-based columns to 2 < pH < 8. Below pH ≈ 2, the bonded phase tends to hydrolyse and be lost from the surface, whereas above pH ≈ 8, the silica tends to dissolve; either condition will ruin the column. The highest purity type-B columns tend to have wider pH ranges, with a lower limit of pH = 1.5 cited by some manufacturers, and allowable use to pH = 10, albeit with somewhat shorter column lifetimes. And some columns are designed specifically to work at higher pH values, as well.
If we want to work at as low a pH as is possible and not have to worry too much about the design limits of silica columns, we should keep the pH > 2. This means that at 2 < pH < 3, we should be able to get the best results in terms of suppressing silica and acid ionization. This is one reason why phosphate buffers in this range are commonly used as mobile phase additives. In many cases, real buffering is not as critical as having a generally acidic mobile phase. Also, as volatile mobile-phase additives are needed for LC-MS and other evaporative detectors (e.g., ELSD, CAD), so phosphate cannot be used in such cases. For these applications, 0.1% formic acid works quite well. In water it creates a pH ≈ 2.7, nicely within our 2 < pH < 3 target region, it is easily soluble in acetonitrile and methanol, and it is sufficiently volatile to work well with LC-MS. A popular alternative is 0.1% trifluoroacetic acid (TFA), although TFA can suppress ionization in LC-MS, especially in electrospray applications. And TFA can act as an ion pairing reagent, which may or may not be desirable for a particular separation.
So the bottom line is that 0.1% formic acid is the favoured go-to additive in many LC-MS methods. It is easy to formulate, provides sufficient pH control of the mobile phase, and evaporates readily in the LC-MS interface.
This blog article series is produced in collaboration with John Dolan, best known as one of the world’s foremost HPLC troubleshooting authorities. He is also known for his research with Lloyd Snyder, which resulted in more than 100 technical publications and three books. If you have any questions about this article send them to TechTips@sepscience.com