Reviving an HPLC system

[TimB]

So, I have a Waters Alliance system sitting in my lab that my group wants to dedicate solely to our projects, rather than split time with people in the other labs. Problem is, it has been dormant for about 2 years. So here's the story:

I have the LC moved from its current location to my lab, and begin to hook everything up. Computer still works, Empower is still loaded and I even manage to track down the username and password. (That was a project itself). I hook up the HPLC and everything comes online. No malfunctions which makes me happy.

I fill solvent bottles with ACN and Water so I can get liquid running through the system again. I dry prime by wetting the lines with a syringe and switching to 100% composition.

Long story short, no matter how long I dry prime, in line A i get only about 145 psi, and 45 psi in line B. The pressure doesn't remain very constant either, it will jump around signficantly.

I contacted waters, and they're supposed to have a service engineer call me (The guy I originally tried to call no longer works at Waters) so they can set me up with a service plan and come out and do a PM.

In the mean time though, is there anything I am missing? Do I just need to continue to prime? In the past when i had low pressure on a system It was due to a leak somewhere, or my solvent bottle ran dry overnight and there was air in the system. That however was on an Agilent and those leaks were easy to trace. Is it possible there is a leak I can't see? Is there anything else I can try?

Thanks again for the help and support.
Tim

[Noser222]

Did you try a wet prime at > 5 mL/min?

[TimB]

Did you try a wet prime at > 5 mL/min?

Yea, I did a wet prime with 100% water, 100% ACN, and then 50/50 at about 7.5 mL/min for 15-20min. I get a slight increase in pressure, but not much.

[Fat Jonny]

Take the pump apart and replace all the seals

[rhaefe]

and clean/replace the check valves.

[TimB]

Put the service request into waters. Any guides out there if I ever wanted to do it myself?

[rhaefe]

Once the field tech is doing the service watch closely and ask questions. I have not worked on a Waters Alliance but on quite a few other systems. Changing pistons, pump seals, check and proportional valves is not "rocket science" as they say.

[leadazide]

I had a similar issue on an older Gilson 306 pump that had been dormant for close to 5 years.. It turned out the but was fine but the degasser had a cloged line.. So the pump couldn't draw suffecient liwuid through..

So try checking the solvent filter (maybe remove them while you test) and if you have a degasser on the system route around that as a test..

Hope it comes up and running..

[lcguy1]

Hate to ask this, but do you have a column n place? Otherwise the pressure you quote is not out of the ordinary. Try disconnecting the black line coming into the inlet check valve from the gradient prop valve. Do you gt flow there? Remove the check valves and sonicte them for 15 minutes and try again. While you are at it, shake them and see if you can hear the ruby balls rattling inside. If you are not getting flow from the black line, try hooking a syring up to it and SLOWLY drawing from it. IF you do not get any MP from A, try B, C, and then D. Also, put a flow of 1 ml/min in and 25% each solvent and listen to the GPV clicks, do they all sound the same? If not, one or more may not be opeinig/closing correctly. Let us know what you find.

[Hollow]

Long story short, no matter how long I dry prime, in line A i get only about 145 psi, and 45 psi in line B. The pressure doesn't remain very constant either, it will jump around signficantly.

Do I get it right, the pressure differences are with the same solvent in line A and B? (otherwise this would be due to viscosity differnces)

What is the pressure reached by the degasser? does it reach and hold below 1 psia?

in addition to lcguy's tipps:
- also clean/replace the frits in the solvent reservoir in the same way as the check valves. Clean the frits/check valves with 100% water first to disolve dried buffer residues, and then with 100% of MeOH or IPA.

- beside of the pressure fluctuations, do you get a constant and accurate flow?
Are there any differnces between each solvent line at 100% and all four at 25% (and two lines at 50%/50%)?
(Messure the flow with a stopp watch and a volumetric flask.)

After cleaning of the check valves and the solvents frits, I would prime the system with differents solvents, same solvent for all lines
1.) Water 100%, ca. 50-100 ml each line at 100%, flow about 1-2 ml/min

1.a.) (optional: cleaning/passivation with a whole of ca. 50 ml HNO3 6N, and thoroughly rinsing (about 300 ml, till neutral) of water 100%; collect the HNO3 in a separate waste container!)

2.) MeOH 100%, same as above

- do also prime the injector (path and syringe with water and MeOH)

Afterwards check if the backpressure is same and constant for each line with the same solvent and check the flowrate for each line at diff flowrates
If not, I think it's time for waters service engineers.

I would also do additional qualification tests before routine use (injection reproducibility/accuracy and linearity, gradient proportioning valve test, detector's wavelength accuracy and linearity)

hope this could help

[lmh]

Can I add a comment on the previous poster's HNO3 suggestion.

If anyone reading this thread does ever decide to wash/passivize with HNO3, note carefully Hollow's very correct advice about collecting the HNO3 in a different container. It is strongly oxidising, and mixed with typical HPLC waste, can very easily explode, often later, at an unexpected moment. This isn't just theory - it really happens.

Fortunately it's very rarely necessary to resort to HNO3.

[TimB]

Update:

Waters came and did everything that pretty much everyone mentioned. I actually got to learn a lot. The problem is, i'm still only getting about 700-800 psi, which to me still seems low. I ran a sample and the response is very very low.

Is that pressure normal? I always thought I wanted to be above 1000 psi. I can't detect any leaks anywhere and there is no air in my system since we purged everything.

[Hollow]

if Waters had reviewed and maintained the system and everything is ok I wouldn't concern the pressure item any longer. Furthermore 700-800 psi for high ACN mobile phase seems ok to me, it also depends on your flow and column diameter/lenght and particle size.
H2O/MeOH 50/50 mobile phases produces much higher pressure.

About your low response:
- what is your detector cell dimensions?
- do you detect at your analyts absorbance maximum?
- does your analyt realy have a good UV sensitivty?
- concentration of your analyt?

[TimB]

if Waters had reviewed and maintained the system and everything is ok I wouldn't concern the pressure item any longer. Furthermore 700-800 psi for high ACN mobile phase seems ok to me, it also depends on your flow and column diameter/lenght and particle size.
H2O/MeOH 50/50 mobile phases produces much higher pressure.

About your low response:
- what is your detector cell dimensions?
- do you detect at your analyts absorbance maximum?
- does your analyt realy have a good UV sensitivty?
- concentration of your analyt?

Thank you Hollow. I realize now that there are a lot of things i didn't take into account.

I'm running a normal phase experiement for monosaccharides. Mobile Phase is 80% ACN, 20% H2O. (Isocratic at this point) I am using a supelcosil LC-NH2 Column about 250 x 4.0mm ID (5um particle size) with a 2.0 cm x 4.0 mm NH2 guard column. Absorbance is set at the very bottom of the scale at 190-210nm. The compounds don't have extremely good UV sensitivity, but it's good enough for us. We haven't settled on a final concentration yet as the method is still being developed.

Tim