I am running a HP 1100 system since 1998 and the corresponding Agilent 1100 system since 2000. Both have a quaternary pump and vacuum degasser. We are typically running Water/Methanol gradients from 50/50 to 0/100 in channel A/B or Water/Acetonitril gradients from 70/30 to 0/100 in channel A/C over about 20 minutes, then going back to starting conditions and 5 mins post time. Flow is 1 ml/min for Water/Methanol and 1.5 ml for Water/ACN.
About 4 months ago both systems started to exhibit spikes in the pressure curve (typically a sharp drop from 200 bar to 150 bar followed by a slow rise over 30 seconds to the old value) but only for Water/Methanol close to 50/50. This changes peak width and area when there is a compound in the DAD cell and makes the analysis void. If this happens between peaks, the baseline stays flat.The ACN gradients are still running fine.
Flushing the whole system with isopropanol helps for 2-3 days, then we have the problems again. Methanol from a different vendor did not help either. One assumption were that we might have algae in the water, we are now adding sodium azide, but why are then the ACN gradients not affected? After four months with sodium azide and many purges with isopropanol, can there still algae survive?
Any help will be appreciated, thanks

K.H.W.

I forgot to mention, substituting channel A by D did not help

K.H.W.

Is the degasser working properly? Mixing methanol with water in low pressure gradient is always critical for bubble formation which appears with maximum usually at 30 to 50% of Methanol.

It sounds like you are getting ar air bubble and then the system recovers. When you mix Methanol and water you get outgassing. Try sonicating the 2 solvents before your run.

Dear KHW;

I have limited experience with the 1100 instrument, but I suspect your problem may have something to do with the pump settings. Methanol and water form mixtures of very different viscosities and compressibilities, and also quite different from those of ACN/H2O mixtures.

There are controls to adjust the pump for different compresibilities of your solvents.
Normally the pump heads are set to different compensation values since it is assumed that one will work mostly with aqueous solvents and the other with organic ones.

The pump-manuals detail some procedures to fine adjust the pump heads according to your situation. Take a look at them.

Good Luck

Benjamin

Thank you all for your feedback! There is now something more we can try to narrow down the error source.
What I still do not understand is the fact that I had reliable systems for 4 years and 2 years respectively and suddenly both systems failed nearly simultaneously. Agilent provided us an offer to replace vacuum chamber, multi channel gradient valve, tubing for ~5000$/system, but as long as we do not know __why__ we have these spikes and how we can avoid them, I see the danger this can happen again.

Thanks again for any help,
K.H.W.