I'm usiing an Agilent 1100 series for NP purposes, using a quaternary pump. I have used it for a long time with n-heptane and dichloromethane as solvents in isocratic runs, using standard seals (said not to be suited for NP purpose, but I needed to be able to go above 200 bars to keep my method below 2 hours).
Due to shift in retention (decreasing retention time) I tried to dry my solvents to remove water from my system. My first attempt of drying was using molecular sieves (and keeping them in the solvents on the system to remove water added to the solvent from the atmosphere). During my attempt to troubleshoot I have tried to dry by having activated silica at the bottom of my solvent flasks. It all worked rather well for some time, but suddenly I saw great drops of pressure but with different behaviors.
Scenario A) a sharp drop in pressure from 140 bars to below 100 bars, then a large fluctuation and slow build up to 140 again, where it would have larger-than-normal fluctuations (normal was 0.8 bar, new was 2-5 bars) before a new drop Scenario B) a sharp drop in pressure from 140 bars followed by fluctuation at lower pressure and a slow build up to 100 bars where it stays with larger-than-normal fluctuations before a new drop Scenario C) a slower drop in pressure followed by a prolonged time at the lower pressure-plateau with larger-than-normal fluctuations and then a slow build up of pressure
We suspected air caused by evaporation of the solvents in the filters on the lines caused by a localized vacuum if dust from the sieves had clogged the filter. Therefor the filters were exchanged to new ones, but the behavior was the same and the pump started to make noises as if I had a buffer in one of the lines precipitating. I have ended up changing every filter and seal possible in the pump as well as a seal in the ALS. I tried just connecting a long tubing for testing to eliminate the possibility of the problem being somewhere in the system following the pump. I still saw the weird behavior in pressure and by letting air into the line I could see how the pump seemed not to take in solvent (the air bubbles were not mowing) when the pressure dropped, so it is not due to a leak but rather a missing uptake. I have now tried to take a binary pump from one of our other 1100 systems, and saw the exact same behavior as soon as I put pressure on! It worked without any problems in RP-mode a few weeks back.
Both pumps can easily run at 4 mL/min when purging on all lines and with dried n-heptane, dried DCM and MeOH. But as soon as I close the purge valve (at lower flow rates) there seems to be an issue for both pumps.
I hope some of you have experienced something similar and solved it or have an idea for me to try out.
Thank you for your ideas! I will try out a pump pressure test and see how that goes, I would hope/think everything should be working since we've already changed everything in the pump and taken a completely different pump to try it out.
I'm still a bit baffled about this sudden change in behavior, especially if it's the cavitation you talk about @M. Farooq Wahab , since I've used (non-dried, straight out of the bottle) n-heptane and DCM for years with no issues. But I think I will try to use the non-dried again even though the rettention of the system will start to shift once again - that can be handled in some other way for my purpose. Otherwise I will try out stabilizing/replacing n-heptane with something like octane or a larger alkane and stabilize DCM with more MeOH than the vendor has already added or add some IPA to it.
I have seen this type of problem hundreds of times before as we regularly use strongly non-polar solvents such as 98% hexane in Agilent 1100/1200-series Quat and Binary pumps. It should not cause any problems at all, but when it does, here are some reasons why and what to do about it.
Regarding pump piston seals. The standard seals can indeed be used with NP solvents such as Hexane and Heptane. A high pitched noise may be heard at times from the now very dry piston (due to the solvent) coming in contact with the piston seal. This will result in the std piston seals wearing out faster than normal and you should be running frequent pump high-pressure leak tests (using the two tests at 400 bars) to monitor this wear frequently plus replacing the PTFE frit every month too. Seals are consumable items. Test, measure their performance and replace when needed. Bad seals will result in pressure instability in any pump.
Next, let me suggest the most common reason for the Quaternary pump failing to stabilize while using dry solvents such as Hexane. The cause is often the multi-channel gradient valve (MCGV) seals drying out (i.e. Channels A/B/C/D). Long term use of very dry solvents (Hexane and Heptane) results in the solenoid seal, inside each MCGV channel position, drying out and leaking internally. This results in a loss of pressure. The "fix" is to re-hydrate the now dry seal(s). We use pure IPA for this purpose. Flush all MCGV positions with pure IPA for 20 to 30 minutes, per channel, then let the system 'sit' in IPA for the same amount of time. Finally, resume use to see if the performance improves. This often fixes the issue and makes the seal swell allowing the pump to maintain stable pressure. Please note that the seals inside the MCGV do not last forever and replacement of the entire MCGV will eventually be needed to restore 100% performance. Most of the 1100-Series pumps are now 20 years old so it is not unreasonable to think that some maintenance is needed now and then. If not abused, the MCGV usually last ~ 10+ years.
- BTW: With these pumps, inlet "check valves" rarely get stuck as those unfamiliar with the design of the Agilent HPLC pumps (1050, 1100, 1200....) may not be aware of Agilent's special use of an electronic, computer controlled inlet valve (AIV) which rarely has any "sticking" issues (regular check valves which do stick are common in other brands of HPLC pumps such as Waters).
DEGASSING: One other item we should briefly discuss is the use of an HPLC degasser. I will assume that you have an Agilent inline vacuum degasser. Hopefully it is the model G1322A (with the large internal vacuum channels of 12 mLs) for your application and not the incompatible model G1379A or B degasser. The G1379-Series is NOT compatible with the solvents you are using (esp. DCM which may dissolve the internal parts of the degasser, which may in turn contaminate the entire flow path of your HPLC and the column. A long list of problems may occur in that case, too much to list here). If you are using a G1322A degasser, which is compatible with Hexane and DCM, then you will want to make sure it is clean and being regularly serviced at least every 5 years. Proper degassing, even of NP solvents, is critical to maintaining a stable flow rate (and stable baseline). Pumps perform at their best when degassed liquid is introduced to them. No matter what kind of HPLC pumping system you use (high pressure mixing or low pressure mixing, Quat, Binary or Ternary), all of them greatly benefit from proper mobile phase degassing. Be sure and FULLY purge ALL lines of the system every day, before use. *Gas re-dissolves back in the PTFE tubing overnight. Be sure and replace those solvent pickup frits too and always use filtered, HPLC grade mobile phase solutions.
Lastly, as a fine tuning step, PUMP SOLVENT COMPRESSIBILITY VALUES should be set for each channel (in the case of a binary pump). More info at this link: https://hplctips.blogspot.com/2011/10/hplc-pump-solvent-compressibility.html
Another item not addressed which also may cause pressure fluctuations could be your column. If you have not done so already, it may be time to dispose of your column and install a new one. Columns are another consumable item and far cheaper than your time or most of the solvents used in HPLC. Try running tests and checks using a calibrated pressure restriction capillary (which provides similar back-pressure to the column you use) instead of the actual column. That will help rule it in or out as a possible cause of the problems observed.
Here are a few free article links which offers common reasons for HPLC pressure fluctuations too (normal fluctuation on an 1100 Binary or Quaternary would be 0.1 to 0.2%).
Diagnosing & Troubleshooting HPLC Pressure Fluctuation Problems (Unstable Baseline) : https://hplctips.blogspot.com/2014/01/diagnosing-troubleshooting-hplc.html
HPLC Retention Time Drift, Change, Variability or Poor Reproducibility. Common Reasons for it : https://hplctips.blogspot.com/2015/11/hplc-retention-time-drift-change.html
The HPLC Restriction Capillary; Troubleshooting, Qualification and Running Without A Column: https://hplctips.blogspot.com/2018/04/the-hplc-restriction-capillary.html
"Tips and Advice for Priming your HPLC PUMP (or similar pumps, FPLC or UHPLC Pump)"; https://hplctips.blogspot.com/2020/09/tips-and-advice-for-priming-your-hplc.html
William Letter thank you so much for such thorough answers! I have tried using a long piece of tubing which I know the backpressure of at certain settings from earlier part of my project, and I could see the issue was situated in the pump or earlier on. I will try and get a new degasser (I use the wrong one) and change the MCGV, as I think that has never been done and we already suspected the issue could be there.
- Badges
- Science topic
- Similar topics
- Engineering
- Materials Engineering
- Powders