Tailing increasing on an ODS column?

[Apharmd Battler]

Got an old method that needs some work, but that comes later. Now I'm trying to get some work done using the method as is. It's a potency method for a large molecule using an XPertek SP-ODS2 4.6X100mm (3um) column. Mobile phase is a 0.2M sulfate buffer with ACN.

My tailing criteria for my main peak is 1.5. I was getting borderline values all last month when running some assays (i.e. 1.48, 1.49, 1.51 etc.) with different mobile phase preps. I thought I would come back later and troubleshoot. I rinsed the column with 80:20-ACN:H20 last week (it sat idle on insturment). I prepared new mobiles phases this last night and let my system equilibrate overnight. I came in this morning and I'm getting tailing from 1.56 to 1.59. When I called the lab that has the expertise with this method, they normally just switch columns until they get one that works. They have never really tried to investigate issues (24 hour in process lab) and this method is known to be problematic. I had on my docket later this year to look at some of these methods, but right now I need to run it as is for a comparison study.

Sorry for the long post but I was trying to set the stage. Any suggestions on how to get my tailing under control? I really can't modify the analytical method yet. I was thinking along the lines of active sites (on column) interferring with my chromatography. A possible dilute TEA rinse may help, but my backgrond is mostly small molecule and I wasn't sure if this would apply to large molecule seperation. Any tips would be appreciated. This method needs some work, but hopefully I can get to that down the road. Right now I need to work with it as is (if I can).

[Mark Tracy]

You didn't state the pH of the mobile phase, but sulfate buffers are all around 2. This is the bottom end of the stability range for most C18 columns, and you are probably seeing an increase in silanol active sites with time. The corresponding loss of bonded phase is probably too small to set off any alarms. Given your constraints of time, I would take the advise of that other lab and just buy some columns. TEA in the mobile phase is effective for suppressing silanol activity, but it does wash out of the column, especially at low pH. One other possibility is iron contamination; a wash with a chelating solution (EDTA at pH 4 and 30% acetonitrile) would help in this case.

[Apharmd Battler]

Sorry the 0.2M Sulfate buffer is pH adjusted to 2.3 with phosphoric (I'll have to check my method to be sure tomorrow morning at my desk). The the mobile phase is made by combining the buffer with ACN (75buffer:25ACN). To save time I run 50% pump A (80:20 mobile phase) and 50% pump B (70:30 mobile phase) and adjust pump composition as necessary to meet my retention time window.

Yeah,I know about small amounts of free metal and using EDTA. That was an issue I remember dealing with a purity method with that as an issue. I'm not sure how to prep the EDTA soln. for this particular method though (i.e what concentration, and I guess I'll pH with the same acid as I use for my buffer). I'm under the gun, so I put a new column (same lot of stationary phase though) on my system and letting it equilibrate with mobile phase at slow flow overnight. We see what my test injections look like in the morning. Thanks for the tips though, it's given me some food for thought.

[Uwe Neue]

Tailing can be caused by a few different things, from injection issues to sample overload. It also can be a problem with silanol activity, which is implied in your answer.

Whether one runs into tailing problems due to silanols or not depends on the quality of the packing material. It is usually less of a problem with a modern packing based on high-purity silica.

That said, I have been in HPLC for some time now, and on the older packings tailing of basic analytes can be suppressed by adding a hydrophobic base to the mobile phase. Commonly, triethylamine is used. If this does not do the job, a longer chain amine is recommended. In the old times, I used cetyltrimethylammonium bromide and never saw a tailing peak on anything that I injected into a mobile phase with this stuff in it, even with unendcapped C18 packings.

[HW Mueller]

It seems that I reduced/eliminated tailing due to interaction with silanols via the additin of 0.1 mole or more of anions (have to review a whole lot of data lying around before I am certain) and certainly by using low pH. Both are the case in AB´s example, so it would be interesting to know what the analytes are.

[Apharmd Battler]

Well, the new column is giving me a tailing of 1.6. I've checked the pH of the buffer and it's right on the money. I'm going to prepare fresh mobile phase, the only thing I can think of is that my organic to aqueous ratio could be off. If that doesn't do it then I'll prepare a fresh buffer solution tomorrow and start from scratch. Other than that I'm fresh out of ideas. Looks like this project work is going to be delayed.

[Victor]

I would check the tailing factor of a simple neutral compound first. Manufacturers ship columns with a range of "acceptable" tailing factors according to their product specifications, which are often based on simple neutral compounds, although there may be a "batch" test of the material using more complex compounds. Any tailing due to the complexity of your solute is likely to be superimposed on the value of that for a simple neutral compound. It sounds to me that your method is on the edge of your own acceptability criteria and that different new columns will just exceed or fall within those criteria due to column to column variations.
I guess that's why your "expert" lab just switches columns until they find one that works.
If your method allows you to substitute a different make of C18 column you might find one which gives you better tailing factors. The inherent column variabilities may then not be a problem if for example you got tailing factors of 1.2-1.4 for this new column. However, you do not tell us what your solute is so it is difficult to recommend another.

[Apharmd Battler]

Sorry.....

The sample is dissolved in 0.01N HCL. The standards/samples come in an lyophilized vial and are diluted to the appropriate concnentration. The method states to use Spherisorb ODSII, 3um, 10cm X 4.6mm ID, or a properly validated equivalent. So I may try a different manufacturuer or vendor.

Victor, I agree that this method is on the 'edge'. I know other labs have had their share of issues with this method and I'm trying to piggy back on the tail end of some previous work. If I decide to switch and order a new column any suggestions on vendor or manufacturer? I've also got some small vendors who can put something together in a snap if need be.

Hey guys thanks for all the help, all your suggestions have got me thinking and that's what I like. My job is to work on getting the labs a better method, and I can tell that this method has been given the proper attention. Thanks for all the input.

[Uwe Neue]

If somebody got the method to work on Spherisorb ODS-2, you should have no problem on a decent modern C18. Unless you are overloading the column...