We have a problem with the multi channel gradient valve in a quaternary pump (Serial No. DE62976920) but before replacing it, I would like to know if anyone faced this problem before and could somehow solve it without replacing it.
Our problem is that channel A doesn't take the set volume, so we decided to use the rest of the channels and avoid channel A. However, even when channel A is set at 0%, it is still taking up volume, what ruins the set gradient. Did anyone tried to physically block the channel to prevent it from taking volume or any other more creative solution?
Thanks in advance!
This is a very simple problem which is often caused by the solenoid valve getting stuck in a specific channel, 'A' in your case. It often occurs when the HPLC pump has: (A) been sitting around unused for long periods; (B) When the system is left in water, buffer or other inappropriate liquids over time or (C) when the MCGV is just worn out ('A' is the most used channel).
BTW: One of the first things to check are the four (4) solvent pickup frits. You are replacing them at regular intervals, right? You are keeping them clean too? If one of them develops a restriction, then it can create a vacuum on the valve resulting in internal leakage and poor performance. Be sure to flush down the entire HPLC with fully degassed liquid before doing any troubleshooting.
OK, Options:
(1) Replace the MCGV. Is has a lifetime and it may be that it is over (your pump is 13 years old). You could run a formal gradient mixing test to check the operation of each channel to be sure [https://hplctips.blogspot.com/2014/05/gradient-mixing-test-for-your-hplc-pump.html ].
(2) Flush/Clean the MCGV. Fill a beaker with HPLC grade IPA/Water (50/50). Place ALL FOUR of the clean solvent pickups in it. Open the prime purge valve. Flush, one-at-a-time, each channel in 100% IPA. Make sure you account for the delay volume in the low pressure tubing and degasser too (that could be 30+ mls per line if you have a G1322A!). Do this for 'A', 'B' and so on. Flush 50-100 mls of IPA at 2 mL/min to waste, per channel. Do this for ALL channels. Stop the flow, set 'A' back to 100% again, purge with 100% IPA and set the flow at 1 mL/min, let some IPA run through it for > 10 minutes, then set flow to 0. Let it sit in IPA for ~1 hour. Return and flush all channels out with your own mobile phases and test the valve again. *Repeat this for each "sticking" valve channel. What happens and why this sometime helps is the seal inside of the valve dry out, then re-hydrates from extended contact with the alcohol. Sometimes it reforms back into shape after the soak in liquid. Plus the IPA rinses and washes away any particulate matter. This may work to restore some MCGVs that have been ignored, abused or contaminated. If it fails, then it is time to buy a new MCGV valve.
This is a very simple problem which is often caused by the solenoid valve getting stuck in a specific channel, 'A' in your case. It often occurs when the HPLC pump has: (A) been sitting around unused for long periods; (B) When the system is left in water, buffer or other inappropriate liquids over time or (C) when the MCGV is just worn out ('A' is the most used channel).
BTW: One of the first things to check are the four (4) solvent pickup frits. You are replacing them at regular intervals, right? You are keeping them clean too? If one of them develops a restriction, then it can create a vacuum on the valve resulting in internal leakage and poor performance. Be sure to flush down the entire HPLC with fully degassed liquid before doing any troubleshooting.
OK, Options:
(1) Replace the MCGV. Is has a lifetime and it may be that it is over (your pump is 13 years old). You could run a formal gradient mixing test to check the operation of each channel to be sure [https://hplctips.blogspot.com/2014/05/gradient-mixing-test-for-your-hplc-pump.html ].
(2) Flush/Clean the MCGV. Fill a beaker with HPLC grade IPA/Water (50/50). Place ALL FOUR of the clean solvent pickups in it. Open the prime purge valve. Flush, one-at-a-time, each channel in 100% IPA. Make sure you account for the delay volume in the low pressure tubing and degasser too (that could be 30+ mls per line if you have a G1322A!). Do this for 'A', 'B' and so on. Flush 50-100 mls of IPA at 2 mL/min to waste, per channel. Do this for ALL channels. Stop the flow, set 'A' back to 100% again, purge with 100% IPA and set the flow at 1 mL/min, let some IPA run through it for > 10 minutes, then set flow to 0. Let it sit in IPA for ~1 hour. Return and flush all channels out with your own mobile phases and test the valve again. *Repeat this for each "sticking" valve channel. What happens and why this sometime helps is the seal inside of the valve dry out, then re-hydrates from extended contact with the alcohol. Sometimes it reforms back into shape after the soak in liquid. Plus the IPA rinses and washes away any particulate matter. This may work to restore some MCGVs that have been ignored, abused or contaminated. If it fails, then it is time to buy a new MCGV valve.
Hi William, thank you so much for your help. I'll definitely try that.
Hi Victoria Calatrava
We tried to physically block the channel to prevent it from taking volume and did not succeed.
That is because the valve is still open inside so solvent can mix between channels. Blocking the port will not help. As noted above, you need to either "fix" or replace the valve unit.
If you are running acetonitrile ensure your bottle contains is a 90% solution rather than running it neat. Running ACN neat can cause the valve to gum up. At that point no amount of rinsing or sonicating will help.
As Michael has already said, pure acetonitrile leads to the valves sticking together in the long run. Acetonitrile contains small amounts of acrylonitrile (acetonitrile is a by-product of acrylonitrile production) and this then tends to polymerize with the described consequences.
Regards
Markus
Thanks Leticia Gomes Pontes, I won't try that then. I'll go for William Letter 's advice. Thank you Michael Toner and Markus Christ for the note. We are using an acetonitrile:methanol:water mixture (45:45:10, v:v:v), so I think that shouldn't be a problem.
- "Running ACN neat can cause the valve to gum up".
The above correct statement by Michael applies to many types of HPLC inlet and outlet valves, but is not applicable in this specific case because the problematic part is an electronic (magnetic) solenoid valve (MCGV), not a gravity or spring check valve. The seals in these solenoid valves are generally not affected by ACN as they have plenty of power to overcome most types of problems that plague other valves. The internal solenoid plunger seals wear out, dry out and sometimes just get stuck. They do not last forever (~ 10 years is the average lifetime we see for the MCGV and also the electronic AIV too. Both are very reliable and rugged parts which take a lot of abuse), but I have personally revived many of the with a good IPA flush so that I what I recommend first.
Hi William Letter, I could finally test your suggestion and I observed a good improvement but channel A is still taking up volume when it is not active.
Before, flushing 100%B for 5 min at 2 mL/min, channel A took 5 mL and channel B took 5 mL. Now, after IPA, for the same conditions I observe that B takes 8 mL and channel A just 2 mL. I tried your protocol twice on channel A without further improvement. I assume now it is not an electronic problem, but related to the seals you mentioned. In your experience, do you think we could fix it completely? Would you have any further recommendation to do so?
Thank you so much again, it was very helpful.
I do not recall suggesting that you only try it for 5 minutes. In my previous note, the total time each channel should be exposed to IPA should be 1-2 hours (for all of the steps). If you followed my earlier advice and found little improvement, then it is time to replace the MCGV.
As I mentioned before, this "trick" helps fix the problem when the seals are dried out or have a small amount of contamination on them. If they are simply worn out, it will not repair them as new seals are needed (and each solenoid is sold as part of one assembly). Purchase and install one new MCGV and the new part should last another 10+ years too.
Yes, I tried for 1 hour with IPA, the 5 min test was only to check if the flow was restored after that. I guess it is time to change the MCGV...
Thank you so much for your advice, it was truly helpful.
Thank you very much for bring up this problem. I found it and the discussion very useful and would like to ask a few things about flushing/cleaning the MCGV. Did you use 100% IPA or IPA/water (50/50)? When setting flow at 0 mg/mL, is the pump turned on to check if it is taking up solvent?
Should the protocol of flushing the MCGV be applied to all channels to be able to restore the MCGV although only one channel is blocked?
Is it possible to change the seal if the MCGV is failed to restore after the flushing step or the only option is to install the new one?
Hoang-Anh: the answers to your questions are above in my earlier post.
As noted earlier, the initial flush is done at 50/50 (IPA/Water) and each channel is allowed to soak in 100% IPA.
The HPLC "pump" is always turned 'On'. When flow is set to 0 mL/min, there is no flow through the system.
All channels should be flushed as stated in the post.
The MCGV is not user-serviceable and should be replaced if damaged or defective. It wears out over time. If one channel is bad, the others are also worn too. It is more cost effective to replace the assembly.
- Badges
- Science method