Shift of retention time and sudden presure drop - help

[Brownie]

Hello,

I use a RP-C18 column and try to separate only two compounds and need to quantitative these. For the separation I would need an isocratic run (80 % ACN and 20% water).

I have following problems:
1) if I only use 100% ACN the runs are reproducible, the pressure and the retention time are over 10 runs constant but as soon as I add only 10% of water the retention times shift per repeated run although I use 10 column volumes to recalibrate the column. Why does the retention time shift between repeatable injections.
2) When I change from 20% Water / 80 % ACN back to 100 % ACN I see the same effect means I let run the system at least 10-15 Column volumes of with only ACN before I start with the injection of my sample. I do 10 times the same run and after 5 runs I see that the retention time start to be more or less the same although the never go back to the value that I had in the beginning of the day with 100% ACN.
3) Sometimes during the run the pressure drops to 0 bar although it seems that the pump is pumping the solvents in the correct amount. We though of a stuck check valve but they just got replaced. And even if we try to purge with Isoprop the pressure does not come back again. Sometimes it just works for several days and before the same mistake with the pressure starts all over again. What can it be ?

What I am doing wrong?

We thought already that maybe the proportionation of the solvents is inconsistent and the Propvalve does not close or open reliably but we made for all four solvent channels the pumproportionation test and they look ok.

We had just the service guy here to change the check valves and we also replaced all filters and solvents. Why do I see this effect.
If anybody has similar problems or can explain it to me please do so, because I do not have further solutions for it.

I am really glad for any kind of advice or help to solve this problems.

[bunnahabhain]

Hello,

2) When I change from 20% Water / 80 % ACN back to 100 % ACN I see the same effect means I let run the system at least 10-15 Column volumes of with only ACN before I start with the injection of my sample. I do 10 times the same run and after 5 runs I see that the retention time start to be more or less the same although the never go back to the value that I had in the beginning of the day with 100% ACN.

Do I understand this correctly: You run one injection with ACH/H2O 80/20, then you flush your column with 10 volumes 100 % ACN and then start the next injection immediately? Then it is obvious that your retention times shift because the column is equilibrated in the wrong solvent (100 % ACN instead of ACN/H2O 80/20).

The pressure drops could be caused by an air bubble trapped in one of the solvent lines. This would also cause ripples in your baseline. Purge the solvent lines and see if you observe bubbles. Maybe one of the fittings is not tight and air is sucked in between degasser/proportioning valve or proportioning valve/pump inlet. Try also tightening the check valve fittings.

Good luck
Jörg

[Brownie]

Hi
thank you for your advice. I try already to check if the system got air inside and tightening the fittings and all connection I have seen.

For the runs: I just wanted to test the reliability of the whole system since I am new in the company and nobody has really done long term quantitative analysis here.
What I did is 10 injections of the same compound with 100% ACN and I got always a retention time of 3.1 min. Than I wanted to know how the separation is when I add 10% water so I run 10 column volumes of the mixture 10% H2O and 90% ACN and injected afterwards again 10 times the same sample but in all 10 samples each time the spectrum shifted a bit further. Startet by 3.4 and ended after 10 runs of the same compound by 4.9 min.
Than I flushed the whole system back to 100% ACN using again 10 CV. But after I injected again the sample even after 10 runs I was not back at my previous retention time of 3,1 min in 100% ACN. And I dont know why.

[LiVD]

Are you sure there are no leaks between your pump and column?

[Brownie]

yes, otherwise the first runs would not have the exact same retention time, or?

[LiVD]

Unless the increased back pressure of ACN/H2O made it worse.

[HPLCaddict]

It's quite a common problem that check valves might get stuck when running 100% ACN. A polymeric film can build on the small balls inside the valve and act as glue . Once it's stuck, it's hard to get it working again just by running the pump. I'd try to clean it by ultrasonication, often this helps. If not, you need to disassemble the valve and then try the cleaning by ultrasonication...

Concerning your retention time shifts, just some thoughts:
-100% ACN is a very strong eluent and even 80% is still strong. Do you really have enough retention of your analytes under these conditions? If you're analytes are not really retained, all sorts of strange things might happen.
- Are your analytes ionizable, i.e. do they contain any basic and/or acidic groups? If yes, you should use a decent buffer instead of pure water. Running ionizable compounds under non-buffered conditions can lead to, well, retention times that are not really robust .
- Concerning aqueous/organic eluent mixtures, I'd strongly suggest to premix your eluents. This way, you'll take one variable out of the equation as you're not relying on the HPLC's mixing capabilities. If it works flawless with premixed solvents but you see weird results when the HPLC is doing the mixing, chance is something's wrong with the proportioning valves. You should run some kind of gradient step test then...
- 10 column volumes is rather at the low end of the eluent volume required to equillibrate a column. If it needs 20 column volumes in your case, then let it be 20 column volumes. Did you take the system's dwell volume into account?

[Peter Apps]

Doesn't 100% AcN cause phase collapse/dewetting ?, which takes a very long time for the column to recover from.

Peter

[HPLCaddict]

Doesn't 100% AcN cause phase collapse/dewetting ?, which takes a very long time for the column to recover from.

No, it's 100% aqueous that's problematic considering phase dewetting in RP . 100% ACN is no problem for a C18 column.

[asm]

[hijack topic]

It's quite a common problem that check valves might get stuck when running 100% ACN. A polymeric film can build on the small balls inside the valve and act as glue . Once it's stuck, it's hard to get it working again just by running the pump. I'd try to clean it by ultrasonication, often this helps. If not, you need to disassemble the valve and then try the cleaning by ultrasonication...

Hey,

when I'm using a binary pump with Channel A 100% ACN + 1% Formic Acid this can also happen to the outlet ball valve? (It's a Agilent 1100 system)

This week I have an pressure of 92-95 bar at the beginning of my gradient (77% H2O, 23% H2O) instead of 108-110 bar.

In addition the pressure fluctuates more than normal.

This is from mid-November:


This is from this week:


Is it possible that your mentioned problem can cause this?

If yes, there should be no problem to remove the outlet ball valve of the channel A pump and put it in an ultrasonic bath, right?

[Peter Apps]

Doesn't 100% AcN cause phase collapse/dewetting ?, which takes a very long time for the column to recover from.

No, it's 100% aqueous that's problematic considering phase dewetting in RP . 100% ACN is no problem for a C18 column.

Sorry, advanced brain fade So water, which is wet causes de-wetting !? No wonder I'm confused.

Peter

[Gerhard Kratz]

When you clean These check valves and ultrasonic also the Saphir balls, take them between two fingers and make them "dirty" with fat from your fingers. Than they will work better. Good luck

[Arne]

Peter Apps: C18 is hydrophobic and will be wetted only by organic solvents. If the solvent is 100% aqueous, the tails will fold on each other rather than interact with the mobile phase.

Brownie: definitely buffer your mobile phase! I totally agree with that unbuffered mobile phases cause all kinds of peak shifting.

[HPLCaddict]

@asm: I'd say that these pressure fluctuations rather indicate that there's an air bubble trapped somewhere in the system. If it doesn't get better with excessive purging you might want to try flushing the system with isopropanol, this might help. Good luck!

[asm]

Hey,

it's me again.

I purged the system - nothing changed.

Then I tried to look which Channel/Pump causes the problem.

Therefore I wrote a method starting with 100% H2O over 5min, gradiant to 100% ACN over 5min and at least staying at 100% ACN over 5min.

Here is the result:



It looks like the H2O-channel has a problem. So I would reduce the problem to the outlet valve or the pump.

The same method on another system looks like this:



Next thing I'll try is to use the outlet ball valve from the "good" system on the other one.

If anyone of you has another idea - I'm grateful for every suggestion.