standard area consistancy problem

[Blazer]

I’m working on a normal phase method for Vitamin D and have just recently started to run into problems with my standard. The typical area count has been around 70,000. When I first started using this method, I could simply equilibrate the column and my first injection would be right where it was supposed to be. Now, however, the first injection is typically around 42k, the next at 48k, the next at 51k, and so on. It takes about 8-10 injections for the area count to level off where it’s supposed to be. It does get there eventually, but it seems strange that for the last few runs the above pattern has emerged.

Any help that can be provided would be greatly appreciated. If any further details are needed from me, please let me know. Thank you for your time.

Additional information:

Waters Alliance e2695 system
4.6 x 250mm Supelcosil LC-Si 5u column
25C, 1.0mL/min
Isocratic elution with 0.45% IPA in Hexane
There is a wash step with 80:20 Hexane:IPA for 6 minutes, requilibration for 6 minutes

[tom jupille]

I've seen that happen with proteins. The presumption was that active sites had opened up on the stationary phase surface to which the proteins were adsorbing. The first few injections served to deactivate those sites.

If you have another column handy, you might want to swap it in and see if the problem goes away. Or, simply program a few injections of a high-level standard at the start of your run.

[Blazer]

Thanks for your response Tom. Let me see if I'm understanding correctly as I'm not very practiced with normal phase chromatography.

Looking at figure 6.21 on pg 269 in "Practical HPLC Method Development", you're suggesting that my analyte is adsorbing to active sites ('A') that have opened up in the silica stationary phase. The first few injections of my analyte are serving to deactivate these sites by adsorption. Once these sites are deactivated, this allows the subsequent injections of analyte to properly interact with the stationary phase yielding more consistant area counts. Injections of a high level standard will presumably speed up this process of deactivation.

If this summary is correct, I'd like to try and understand why this process of deactivation of active sites would need to be performed for each of the last few analyses. Basically, how would the deactivated sites become reactivated?

[HW Mueller]

What does it look like after you throughly clean the column? Maybe your "active sites" are actually dirt which has accumulated on the column, the dirt strongly retaining a part of the analyte.

[Blazer]

That's entirely possible, HW. Again, I don't have any experience with normal phase (sans the three weeks I've been working on this project). A quick online search brought up the following cleaning/regeneration procedure for a normal phase column.

• Flush the column with 20 column volumes Heptane
• Flush the column with 5 column volumes Isopropanol
• Flush the column with 20 column volumes Acetonitrile
• Flush the column with 20 column volumes Water
• Flush the column with 20 column volumes Acetonitrile
• Flush the column with 5 column volumes Isopropanol
• Flush the column with 20 column volumes Heptane

Is this procedure considered too much? Also, could I substitute heptane with hexane? I assume that I could, but since I'm new to NPC, I figure I may as well ask all questions that pop in my head, no matter how simplistic they may seem to everyone else.

Thanks again for your time.

[tom jupille]

Actually, I have no idea what the "active sites" might be (and HW's suggestion that they might be adsorbed garbage is a good one). The point is that it looks like some of your sample is sticking to *something*.

Your proposed wash sequence looks OK to me. The general idea in any column "regeneration" is to flush the column with something that is a good solvent for whatever flavor of garbage is stuck there. The old "like sticks to like, and like dissolves like" rule of thumb is useful here. Normal phase columns (which are relatively polar) tend to accumulate polar garbage, which can be washed off (hopefully!) with a polar solvent. The only catch is making sure than whatever solvents you change from/to are mutually miscible.

If cleaning the column doesn't solve the problem, then swap in a new column. If that looks OK, then another possibility is to use a guard cartridge to trap the garbage, and simply change that as appropriate.