Chromatography Forum

Hi, all -

C18 column.
Mobile Phase A: H20/AcN/TFA (98/2/0/1)
Mobile Phase B: H2O/AcN/TFA (10/90/0.9)
Gradient: 5%B to 40%B
Flow Rate: 1 ml/min.
Pressure: 99bar.
Wavelength: 215.

After months of fine running and normal-looking baselines, I suddenly (after running a different method on this HPLC and switching back) have craziness.

While running a mobile phase A blank, I was seeing an increasing baseline, about 150 mAU, that increased with the gradient in B. Pressure seemed normal. I washed the system in MeOH. During the wash (with no column, but monitoring baseline) I saw all kinds of peaks coming off. Eventually, the MeOH baseline stabilized.

Today I try my gradient again on a blank, and now the baseline is going NEGATIVE as the B increases ... AND ... the pressure increases to ~150 bar!!

Something is wrong! Help! These are old, used systems, and who knows what's inside them.

Do I need to passivate??

THanks for any and all suggestions.

EM

I had some crazy results like this a long time ago when I bought a new deioniser. The water produced was low conductivity, passed all the routine tests, and appeared to be suitable for most purposes, but was a nuisance in several circumstances. In later life I had to deal with a RO unit that gave similar problems.

I never did determine the actual causative agent, I just got pragmatic about the matter.

Test the water by using it to prepare a small amount of saturated potassium iodide solution. If it goes yellow-brown in less than 30min (2I' -> I2 +2e), your water has "the lurgi".

In the original situation, I still had my old, low output deioniser, and I used that for mission critical work. I found that the output of my new deioniser could be improved by using a small inline 8 micron candle filter, into which it was possible to fit a coil of pure iron wire (ie. any rust formed stayed inside the filter).

Hope that helps ...

It's possible you have some previous injection residue still stuck int he injector port (or even septum pieces if it's ant automatic inj). You can take the inj port apart and clean it well. I have had to sonicate the parts of one of our older machines to get all the junkity junk out. Normal wear and tear and normal junk from buffers and such...

Good luck!!!

probably purity of your mobile phases - try using new ACN, TFA, cleanest water you can get, maybe from different place

Thanks, everyone!

By the way, I had tried multiple mobile phase preps, and they all exhibited the problem - even the 2 mobile phases that had worked fine the day before.

Here's the unsatisfying* result:
I had to get these samples on, bad baseline be damned. So I put on an overnight run with a bunch of blank and standard injections at the beginning. The blanks and first standard all had the baseline slope, and the first standard also had my peak of interest eluting about 4 minutes too late.

The next injection, and the subsequent 70+ injections, were all normal, just as if it had all been a bad dream.

I think there was some kind of contamination that worked its way out, luckily.

Thanks again -

EM

* - I say unsatisfying because I still don't know the cause. The way the situation turned out was very satisfying, from a get-the-samples-done perspective.

"I think there was some kind of contamination that worked its way out, luckily. " - it seems so, I would carefully look at the second method used (the one after problem occured)

seems like flushing with water helped to clean "it"

was it the same column or only the system?

Don't forget the resevoirs and sinkers. If you change the mobile, but not the tubing and sinker you just "inocculate" your water with whatever was happily growing in the old jug. I had some RT issues last week, changing the sinker helped Of course it was a 100% water... Don't laugh - it can happen.

Living in it?
One place I worked, we had a pink yeast that grew in every crevice of the lab water system. When we went on holiday for a fortnight, it actually produced mycelia in the pH meter "keeper beaker". That thing lived on NOTHING!

We replaced the carboys, threw things away, disinfected with bleach and tried all manner of things. We ended up working on stills to be sure that it wasn't hiding in the deioniser - it still showed up all over the place, even after we replastered the ceiling and repainted the walls.

That experience convinced me that sometimes demolishing the building is not bad thing ...

That pink fungus is the WORST! we had that too in my last lab (biochemistry)!! Had to start adding an antifungal to the cell media to keep it from spoiling all our protein factories!!

Well, it's back. I have taken all your suggestions. I've removed individual lines throughout the system, I have changed reservoirs, changed lines & filters, prepared new mobile phases, opened fresh solvents ... nothing's changed.

As soon as my gradient kicks in, the baseline drops about 250 mAU over 7 minutes. This doesn't sound like much, but it's significant for smaller peaks that elute on the slope.

SOMETIMES, when I've changed lines and mobile phases, the slope is INCREASING instead of decreasing. Very often, as my organic mobile phase increases, the pressure ALSO increases, which is very strange.

If you ask me, it's more evidence that one should buy new instruments, not used.

Anyway, any further ideas? It seems like it HAS to be a contaminant, but I've changed out almost everything (except detector parts) and the problem persists.

What is most strange to me is that it was fine one day, then had this problem the next. Then miraculously went away (see above) then returned!

Desperately,
EM

Is it possible that he is seeing the negative baseline due to the TFA concentration reducing over the gradient? In A, TFA is 1%, in B TFA is 0.9%.

(I know that 0.1% isn't a huge change, but when you are doing things analytically, it sure can be. )

Also, did you try to 'regenerate' the column using a good long rinse in ACN?

Kerri

Hi, Kerri -

Thanks for the suggestion. The difference in TFA concentration in the mobile phases is actually there to try to keep this sort of thing (drifting baseline) from happening! I was thinking of playing around with TFA concentrations, but I don't suspect this is the cause, as the baseline was always fine before, the problem arose so suddenly, and even when I re-prep, it remains.

As for the column, I might have forgotten to mention: this phenomenon happens with and WITHOUT a column, so it's not column-related.

Thanks for your advice. Any and all is welcome.

I'm currently simply running samples on a crappy baseline, 'cause I gotta get the samples out the door.

EM

PS - holy moley, I wrote the wrong mobile phase up above! "A" should read "0.1% TFA" and "B" should read "0.09% TFA."

The only obvious thing that comes to mind is that you have an unconsidered stabiliser in the mix somewhere.

I've had problems with BHT in THF, because some moron returned stabilised solvent to the "wrong" bottle, and with commercial alkylsulphonic acid solutions which also contained 0.05% of acetic acid as stabiliser. And sulphuric acid which contained tiny amounts of mercury as stabiliser ...

Sometimes you can have problems you don't know are there ...

"PS - holy moley, I wrote the wrong mobile phase up above! "A" should read "0.1% TFA" and "B" should read "0.09% TFA."" - I read it as 0.1% and 0.09%

baseline changes can be due to TFA concentration changes when you do mobile phase, I once played with this about 10% less TFA in ACN but it was not worth the hassle, I'm adding the same amounts to both mobile phases now

"It seems like it HAS to be a contaminant, but I've changed out almost everything (except detector parts) and the problem persists" - I would examine the detector then, at least see if it looks clean

"Very often, as my organic mobile phase increases, the pressure ALSO increases, which is very strange. " - strange indeed, did you do system qualification, lamp energy, gradient and flow checks?

Pressure changes are not always remarkable.

You mix 500 ml of methanol with 500 ml of water in a litre measuring cylinder. Stopper and mix thoroughly.

Does it get warm, indicating that something has happened? Yes.
If you let it cool, is there a litre of liquid in the cylinder? No, there is less than a litre.
If you know what you put in the cylinder, and you have not lost anything, you can deduce that the mixture is denser than it would be if there was no interaction - so some pressure variation on mixing is not unexpected, especially if significant heat is generated.