Pressure increase on UPLC

[Mattias]

This topic is somewhat related to my previous post about the Kinetex 1.7 µm column. In that post the largest problem was found to be that the needle wash (Acquity system) was too strong, which gave really bad chromatography from the first injection.

Now when this is fixed, it is possible to see other problems...

It appears as the pressure is rapidly increasing on some of the Kinetex columns. The mobile phase is 2 mM PIC B-8 in 19% acetonitrile (adjusted to pH 5.5 with ammonium acetate). The mobile phase is filtered and the samples are sterile filtered in production (no sample preparation). Sometimes the pressure increases 1000 psi per hour, just by pumping mobile phase.

Sometimes there is no pressure increase at all, and I can run 300 samples in a row.


I need help to understand what is going on!

- Is there something fundamentally wrong with the mobile phase?
- I used to let the instrument mix acetonitrile and the buffer, but I noticed that the water phase got hazy after a few days. Now I use premixed mobile phase, and it looks clear during the analysis.
- Are there any other common causes of pressure increase on Acquity systems (parts leaking particles etc?)

Thanks again for you help!
Mattias

[bisnettrj2]

Quick question - are you using a pre-column filter or guard? I tend to run a 0.5 micron filter post-mixer/pre-autosampler and a 0.5-micron filter post-autosampler/pre-column to prevent pump seal material and valve seal/injected particulates from clogging my column/system. If you install filters of that sort, it would be interesting to see which are actually causing your issue - the mixing of the mobile phase (post mixing/pre-autosampler filter) or the injection of the samples (post-autosampler/pre-column filter). If it's the column generating increasing backpressure, even with the filters, that's a whole other ballgame, and maybe points to particle breakdown or chemical contamination of the column.

[Mattias]

Thanks,

I do not use any inline filter or guard column at the moment! (maybe I should because this is getting pretty expensive!)

[bisnettrj2]

I would try some filters, at least to see where the contamination is coming from (and if it's physical or chemical). I'm not sure what the porosity of the Kinetex column frits are, but I know Agilent's Poroshell frits are "nominally 2-micron", so 0.5-micron should be sufficient to keep them clear if they are at all similar to the Agilent Poroshell frit porosity. You can try the Phenomenex Ultra SecurityGuard setup or their KrudKatcher filter, or any other direct-connect high-pressure inline column prefilter (I've used the Optimize Technologies' 0.2-micron stem filter setup, but they offer the same setup in 0.5-micron - kind of expensive at ~60 per replacement filter) and any inline low-volume high pressure filter upstream of the autosampler (I use Restek's UltraLine In-Line 0.5-micron filters).

[Klaus I.]

It appears as the pressure is rapidly increasing on some of the Kinetex columns. The mobile phase is 2 mM PIC B-8 in 19% acetonitrile (adjusted to pH 5.5 with ammonium acetate). The mobile phase is filtered and the samples are sterile filtered in production (no sample preparation). Sometimes the pressure increases 1000 psi per hour, just by pumping mobile phase.

Does the pressure increase only when the system pumps a long time at the aqueous starting condition and you did not run the gradient? And after some gradient-runs or washing the column with high organic content, the backpressure is back on the initial value?

[Mattias]

It appears as the pressure is rapidly increasing on some of the Kinetex columns. The mobile phase is 2 mM PIC B-8 in 19% acetonitrile (adjusted to pH 5.5 with ammonium acetate). The mobile phase is filtered and the samples are sterile filtered in production (no sample preparation). Sometimes the pressure increases 1000 psi per hour, just by pumping mobile phase.

Does the pressure increase only when the system pumps a long time at the aqueous starting condition and you did not run the gradient? And after some gradient-runs or washing the column with high organic content, the backpressure is back on the initial value?

The method contains a gradient, and the pressure increases anyhow. It is not possible to wash or backflush the column. A high pressure column is waste!

[Mattias]

Update on the problem:

I have given up on this UPLC method, and converted the method to an HPLC method. The column was exchanged from Kinetex C18 1.7 µm to a Waters XBridge C18 3.5 µm. Same dimensions and mobile phases. Method moved to a 15 years old Waters Alliance.

Selectivity of the two columns is close to identical. The peaks are of course broader and not as high on the 3.5 µm column, but the separation is still good enough (Rs > 2.0 between all peaks).

The interesting part is the the QL of the impurities improved by a factor 3. The reason for this is that the detector noise of the Allaince system is 10 fold lower compared to the Acquity system (which was requalified some weeks ago).

I start to think that the Acquity system is so full of flaws that it cannot be used for routine application:

- The weak needle wash has a huge impact on the chromatography (this is not written anywhere in the manuals, and not even the technicians knows about this!)
- The detector show much higher noise levels that ordinary UV-detectors. You loose all the sensitivity that you gained by the higher peaks
- You cannot use formic/acetic acid in the mobile phase, since then the noise will become completely crazy (hard to see the main peak)
- The capillaries of the column switch needs to be exchanged. The stock parts get stuck all the time.
- The column oven cannot take more than 150 mm columns
- The actual injection volume is very sensitive to the viscosity of the sample. For quantitative analysis you must use full loop injection (this is also not written anywhere).

an so on....

[Klaus I.]

The interesting part is the the QL of the impurities improved by a factor 3. The reason for this is that the detector noise of the Allaince system is 10 fold lower compared to the Acquity system (which was requalified some weeks ago).

Which detector rates/speed have you used with both systems?

- The weak needle wash has a huge impact on the chromatography (this is not written anywhere in the manuals, and not even the technicians knows about this!)

Amazingly this seems to be true, e.g. in the Acquity UPLC System - Operator#s Gide (7150082502/Revision C) it seems that this well-known issue is mentionend only once (Chapter 'Diagnostics and Troubleshooting':

To stop a needle wash routine before it finishes:
...
Caution: Do not abort the sample needle wash sequence. Doing so may leave strong solvent in the sample needle which can adversely affect chromatography."

Maybe a more detailed introduction of the autosampler in the manuals should be done by waters. During the launching of the UPLC, one of the most important topics was always the functioning of this novel autosampler and the requirements (e.g. weak needle wash should have the same compressibilty as the sample solvent).

- The detector show much higher noise levels that ordinary UV-detectors. You loose all the sensitivity that you gained by the higher peaks

I can't believe that this is in general true. According to my experience, most of the problems with noise using a UPLC is caused by a wrong detection rate or problems with mixing of highly uv-active mobile phases.

- The column oven cannot take more than 150 mm columns

I think Waters offers now also special column ovens for this problem which design looks accidently like as known from the Alliance
Also since some years, there is finally a real usefull solvent tray for the old-fashioned Alliance availible, like as known from the UPLC

But in general, doing real sub-2-micron chromatography, there should usually be no need for such long columns. Best starting point in my opinion is a 5 cm length
(and sometimes it may be also to long) and only for special applications a longer column may be considered.

[sketcher]

You should contact Waters Technical Service and they will troubleshoot with you as far as they can from a distance and then can have your local engineer or sales rep bring in a Technical Field specialist who can help you with optimizing your method. They have instrument tech support and chemistry tech support via phone. Try the chemistry group first for method issues and then if they can not help you with everything then they will transfer you to the other group. There is no reason on earth that you should not get huge gains moving from Alliance to Acquity once you understand the differences and intricacies.

[canerist]

Hi,
what is column temperature at the analysis, 40C may be help to your problem, and i offer you to clean or change uplc seal and mixer part, and other advice try different mobile phase line, check the system with all the lines,(A1-A2-B1-B2), work with mixed mobile phase, and if it is possible increase the acetonitrile content in the mobile phase,
good luck