COLUMN SATURATION

[Maristela]

Hi friends

I am facing a new problem with a brand new column (X-Bridge Shield RP 18 3.5µm 4.6 x 150 mm).
Here are the conditions:
- analyte : micopheniolic acid
- internal standard - naproxen
- matrix: plasma
- volume: 50 µl
- extraction: 3 ml of hexane-diethyether (8:2 v/v) in acid pH (3)
- analysis: LC-MS/MS - API 2000
- mobile-phase: acetonitrile - acetic acid 1mM (8:2 v/v) + 10% isopropanol
- samples are ressuspended with 500 µl of mobile phase and 8 µl are injected.
- flow 0,8 ml/min (split 1:1)

I did a first study with this method using a range of 200 to 30.000 ng/ml and the method works ok for all the samples (2500).

Now I have to quantify the samples in a huge range of 10 to 20.000 ng/ml. I tested different volums to ressuspend the sample and volums of injection, as the follow:

- 200 µl to ressuspend > 25 and 15 µl to inject
- 400 µl to ressuspend > 15 and 10 µl to inject
- 800 µl to ressuspend > 30 and 40 µl to inject

In all cases I have a decreased of 25 - 50% of the analyte response after 50 - 60 samples. If I clean the column with acetonitrile-water (1:1) the response increases again.
So, I know that it is a matrix x column problem, but I do not know how to solve it.
Do I change the mobile phase ??? Maybe try a stronger acid (2 or 5mM) or decrease the organic part (70 or 60%) ?? The ion suppression test was negative.

I hope you can help me.
Thank a lot.
Maristela

[Uwe Neue]

This looks like an accumulation of material on the column that slowly breaks through after X injections. I suggest a more frequent wash of the column, maybe after every 40 injections. A short flush with 5 mL of 100% acetonitrile should do.

[Maristela]

Dear Uwe

thank you for your answer.
It is a good idea, but unfortunatelly is impracticable because after each cleaning step, I have to condition the column again, and it will be difficult to do this in the middle of a set of samples. Each set is composed by 170 samples. We normally clean the column with acetonitrile-water (1:1) after each set.

I was thinking about increase the concentration of the acetic acid in the mobile phase, from 0,06 to 0,1% so the matrix contaminants would run earlier.
What do you think about ????

[Uwe Neue]

OK, I do not agree that it will take a lot of time to reequilibrate the column with the mobile phase. It should not be more than 20 column volumes, possibly as little as 10 column volumes. I do not know what your method requires, but indeed it might be a good idea to do a few intermediate standards before going back to the samples.

Anyway, this was in my opinion the simplest solution. The more complex solution is to remove the stuff from your samples that is building up on the column. Considering your sample prep, I think that these are lipids. You can get rid of lipids easily by a simple reversed-phase SPE after the LLE. You evaporate the sample to dryness, redissolve it in water or water with some organic, load it on a reversed-phase SPE device like Oasis HLB, and reextract the analyte(s) with methanol. The lipids will stay behind.

It's some work and it costs some money. But it is rare to get something for nothing...

[james little]

Some information on removing lipids from column after every injection on my website in plasma LCMS analyses..

http://users.chartertn.net/slittle/

The rough draft of text at

http://littledomain.com/james/files/text.pdf

[Maristela]

I did a cleaning proedure every 7 or 14 samples, but the problems was still there.

I did a new test of ion supression, injecting 15 blank samples, while injecting the standards (infusion in a harvard pump) and I could see that if I use a 5 min run instead of 4 min, the baseline was much more stable and was not increasing or decreasing after the injectiuons.

I run 2 set of CQs and calibration samples (around 100 samples) and I wash the column after each set. I did a set of volunteers samples and almost in the end, I noticed that I had some CQ out of the expected values. I re-injected them and the results were carzy ( in one injection, the IS area was the double and in the other injection of the same sample the analyte area was increased.

I decided to re-calibrate the system with the analyte ( FIA injections) and re-set the capillary position.

I run a set of the 3rd CQ and C samples and the results are very confused, in other words, I still have variation and the accumulation thet was aparently solved reappears again........

Do I have to re-calibrate the system with PPG ??? Or Maybe this is case for the GOD ????????????

Please let me know If you had faced a proble like this.

[promochrom]

Perhaps on-line sample cleanup and enrichment can help? You may look our website: www.promochrom.com for an easy and economical solution.

[Andy Maljevic]

Check your splitter after 20-40 injections. Is the flow still 1:1?

Change youe method from isocratic to gradient, or add one min high organic portion for each run followed with enough time to re-eq. the column to your initial condition.

Then, try to check the drifting (if it is still there) is the mass spec problem or LC problem.

Andy