Annoying problem keeps popping up before starting runs!

[EmpowersBane]

I run a 40 minute gradient starting at 18% Acetonitrile and 82% Buffer on UPLC and I almost have to pray that my check valves stay put for the entire run of 20 injections. I start up my system as usual, priming out the lines and needle cycles etc and I want to equilibrate to 100% ACN for a half hour or so before I hit go to prime the column before the run and 8 times out of 10 my check valves will go nuts (pressure up and down) about 15mins into this and I have to replace it and start the run again.
I'm only doing this to wash off any retained material at the head of the column ( I also have a ghost trap fitted downstream) before I hit the column with 82% Buffer. Are check valves really that sensitive??

[HPLC chemist]

What are you injecting? ACN and buffer are not unusual. What is your buffer and what is your wavelength?

[Consumer Products Guy]

Annoying problem keeps popping up before starting runs!

In my case it was my pointy-haired boss showing up!

[EmpowersBane]

Im injecting blank solvent at the start of the run followed by related substances samples. Its dual wavelength one active at 220nm the other at 240nm, buffer is 20mM potassium phosphate, all standard stuff but I wonder am I doing my prep before the run right. A workmate told me to always blast the column with close to 100%ACN for a bit (10 or 20mins) before you start your run, but by doing that it seems to clog the check valves and that takes up time.

There is so much advice out there about check valves and the best way to activate and prime for gradient buffer runs but none of it is consistent. ANy advice?!

[HPLC chemist]

It seems your injector may be clogged by what the formulation you're injecting. I would develop a needle wash that guarantees your washing off the crap from your injector.

In my previous life I killed a column in a week because I was injecting a 'high' concentration of an excipient (in order to detect the RS) used in the manufacture of a USP protein standard (excipient = gelatin). Eventually I had to develop a secondary standard that only contained the RS in a diluent (no gelatin).

[EmpowersBane]

Thanks for the reply. My method has a weak and strong wash, weak is 60% Water and 20% each of Methanol and Acetonitrile. Strong wash is 30% each of Methanol and Acetonitrile and 40% Water. I must have left the ins method as default for strong and weak wash volumes, are there any regulations on this?

Would a clogged injector make the check valves stop working? I'm always wary of putting too much organic through the system as Acetonitrile is known for sticking to check valves and causing pressure issues.

[HPLC chemist]

Damn!

Take your column offline and wash the system with ~500 mL of 'hot' water and then Methanol at a flow of 5 mL/min. This should clean any residual buffer from the system and clog your check values. It seems someone did not clean the system after use.

[HPLC chemist]

What are you injecting on your column? What is your injection volume? What is your sample concentration?

[HPLC chemist]

High concentrations of ACN (>50%) are known to precipitate Phosphate buffer. What is your gradient timetable? Can you adjust it?

[HPLC chemist]

Can you use Methanol instead of ACN (which is non-protic)?

[DR]

100% ACN is well known for causing "float" problems w/ ruby/sapphire check valve cartridges - where the ball floats a bit over the seat, allowing intermittent leakage.

Fortunately for you, ceramic alternatives are available and immune to "floating" as a result of frequent 100% ACN exposure. Try some!

[EmpowersBane]

Thanks for the replies, to answer a few qs:

Injection vol = 6 microliters, I cant adjust gradient as its a validated method and the gradient starts at 80% Buffered mobile phase and 20% of Acetonitrile and cycles back to this over 42 minutes.
I always perform a washdown after my run which flushes water through the system for 2 hours followed by a low flow of acetonitrile for 2 hours then shut off.
It could be that my weak and strong wash amounts aren't sufficient. I'm running blend samples which contain about 2-4% active in a lactose excipient.

Why is methanol considered superior to acetonitrile for cleaning down UPLC? I have heard that a few times but don't understand the science behind it. I will stop blasting the column with 100% ACN before the run I don't think that's helping.

[HPLC chemist]

Methanol is protic, that is it has a proton. Most molecules are subject to proton interactions like hydrogen bonding.

It is likely that Lactose has contaminated your system (even though it is a binder). Try using the enzyme Lactase (in water) as one of your washes.

[EmpowersBane]

Thanks for that, I should add that I filter the samples through an acrodisk 0.45 micron filter, but its certainly possible that some lactose has contaminated the system.

[Hollow]

Don't let your check-valves rest in ACN. Do a wet-prime with MeOH before shut down (only pump. Column rest in high ACN)
Doing so reduced our issues with CVs.

Maybe also compare the quality of your acetonitrile. Some may contain more impurities that may polymerize on the CV. Even if gradient quality.

And I would filter my samples and buffer with 0.2μm for UPLC. Pure solvents and solvent mixtures we don't filter, but if salts were involved we do.

By the way, what system do you have? A low- or high-pressure mixing?