Step gradients - column shock?

[Camisotro]

My group is using a rather old (1994) gradient LC pump with a 1 mL delay volume, borrowed from our collaborators. There is a slim to nil chance of getting our hands on anything else for this project.

I know that LC/MS if the ion source is ESI, optimal flow rates are in the 200-300 µL/min range. But this also means that my gradient is delayed by 3-5 minutes plus the column's own void volume - so more like 4-6 minutes. Even if I set up a gradient to run starting at t=0.01, the sample will still experience 4-6 minutes of the initial isocratic phase. And after the gradient is over I have to wait several more minutes to let the back pressure reach an equilibrium state.

We do have access to 2 other isocratic pumps, however. And from what I can tell, an optimal method would involve eluting a short low-organic segment to waste (to wash out the matrix components) followed by a higher-organic segment to elute the analytes into the MS (monitoring by MS/MS, paramters already known).

So I was thinking what I might do is use the switching valve of our MS to program (for example) 2 minutes at 15% methanol and then step up to 5 minutes at 60% methanol, no gradient. This would have much better throughput than programming it all on the single gradient pump.

Would the sudden solvent change have a negative effect on the column? Would it help to provide a small mixing volume (not as large as the problematic 1 mL of course)?

What if I'm just using a small guard column - is that going to be more sensitive to the sudden change?

[Uwe Neue]

There are potentially many ways around the problem, depending on the equipment that you have. Here are some thoughts:

Can you tell your LC equipment to start the gradient, and make the injection only at a time just before the gradient gets to the column inlet?

Are you using automatic injection or manual injection? If you do a manual injection, you can buy a separate injector for a few dollars and do the injection with a timed delay.

Is your MS capable of tolerating higher flow rates? Some MS systems do 2 mL/min with little difficulty. So you may not need to be hung up on 200 - 300 microL/min.

If you can't fix the system, you could potentially start the gradient with a very polar solvent, say 5% organic. When you inject, the sample might just sit at the column inlet and do nothing, until you get to a solvent composition that makes the peaks move. This is an automated washing step for the matrix components, which is not a bad thing anyway. If you have a switching valve, you can run all the LC sample washing at high flow, and only switch to a low flow into the MS once the elution solvent reaches the column.

Anyway, I don't know what you have, but you could do a lot of things with a little thought.

[Loekie]

You can split the flow before the injector. you can have the pump at a nice flow without diluting your sample

[Camisotro]

Can you tell your LC equipment to start the gradient, and make the injection only at a time just before the gradient gets to the column inlet?

No. I thought of that but the autosampler and liquid pump only communicate with the MS via contact closures. The MS sends out a "ready" signal; the autosampler then prepares the injection and sends the "inject out" signal, which starts the clock on the pump and the MS program. There is no way to do a consistent delayed injection protocol because the time elapsed between "ready" and "inject out" is variable. Only the MS talks directly to the computer; the autosmampler and LC pump must be programmed from the front panel.

Are you using automatic injection or manual injection? If you do a manual injection, you can buy a separate injector for a few dollars and do the injection with a timed delay.

We are using an autosampler and we will need to do hundreds of runs with whatever method I come up with.

Is your MS capable of tolerating higher flow rates? Some MS systems do 2 mL/min with little difficulty. So you may not need to be hung up on 200 - 300 microL/min.

Tolerate, yes; but at such high flow rates, ESI-MS suffers incredible degradation of peak height. So even though the peaks would become narrower, my detection limit would get worse by a factor of 5 or more.

If you can't fix the system, you could potentially start the gradient with a very polar solvent, say 5% organic. When you inject, the sample might just sit at the column inlet and do nothing, until you get to a solvent composition that makes the peaks move. This is an automated washing step for the matrix components, which is not a bad thing anyway. If you have a switching valve, you can run all the LC sample washing at high flow, and only switch to a low flow into the MS once the elution solvent reaches the column.

This is what I'd been hoping to do, but the time it takes to re-equilibrate after starting a gradient on our pump is quite long at flow rates that work well on ESI.

Anyway I'm still wondering how bad or not-so-harmful an abrupt switch might be for the column.

[Camisotro]

You can split the flow before the injector. you can have the pump at a nice flow without diluting your sample

This thought also did occur to me... lots of potential wasted liquid though depending on the flow rate.

[lmh]

(1) Whether you can do sudden changes depends on your column. If you are using a modern silica-based reverse phase column, you can change solvent as quickly as you want, change flow rate as quickly as you want, or if you like, drop it on the floor and stamp on it (but not recommended!). If in doubt, ask the column supplier.

(2) Splitting flow may waste a bit of solvent, but if you only have a 1mL/min pump, you haven't got much choice. Solvent (even acetonitrile) is cheap compared to an LC-MS system. And look on the bright side: if you need to collect fractions for other analyses, you already have a flow available! Splitting flow will cost you a T-piece and a bit of tubing.

(3) Whoever is in charge of your MS facility needs to think carefully about the cash investment in an MS, and the balance of wasted MS time versus the cost of a modern pump. You can get a simple gradient pump (second hand?) for very little cash nowadays, and you have correctly identified that not having such a pump is going to slow you down phenomenally.

[Camisotro]

(2) Splitting flow may waste a bit of solvent, but if you only have a 1mL/min pump, you haven't got much choice. Solvent (even acetonitrile) is cheap compared to an LC-MS system. And look on the bright side: if you need to collect fractions for other analyses, you already have a flow available! Splitting flow will cost you a T-piece and a bit of tubing.

(3) Whoever is in charge of your MS facility needs to think carefully about the cash investment in an MS, and the balance of wasted MS time versus the cost of a modern pump. You can get a simple gradient pump (second hand?) for very little cash nowadays, and you have correctly identified that not having such a pump is going to slow you down phenomenally.

2) To be clear, it's not a 1 mL/min pump... it's a gradient pump with a void volume of 1 mL, but the pump's flow rate can go as low as 10 µL/min. The flow rate I'm looking at is 200-300 µL/min.

3) We are not an MS facility. We are an MS research group in a university and normally we do zero chromatography. We have entered into a collaboration with a group that had LC equipment to lend us (which they use for UV and GC-MS; they don't have LC-MS in their own lab). And when this project is over 6-12 months from now, they will take their equipment back and my group will probably never do LC again after I graduate.

How little is "very little cash"? Expenses are tight and my higher-ups probably can't justify investing in equipment they won't need again.

[Uwe Neue]

A decent column will not have a problem with abrupt solvent changes.

there may still be a way in which you can combine column requilibration with the injection in a way that is beneficial. Think of the solvent generation train that happens at the pump as being offset from what is happening at the column outlet. So you can program a step back to the initial conditions before your last peak has eluted. So the reequilibration step of the next run and potentially the start of your gradient for the next run is programmed while your peaks are still eluting. It looks weird for the programming of the method, but it should work.

[lmh]

could you consider hiring a pump, or even talk to the MS manufacturer, and see if you can borrow a pump from them? If they think there's a sale in it, they might lend you one. The advantage of hiring/borrowing from the MS manufacturer is that they'll probably be able to sort out all software and installation issues, and give you the benefit of a working LC-MS system. The way to swing it with them is to point out that if the MS lab love it, and are considering further collaborations with labs that need chromatography, then this could easily be a future sale.

Worst case scenario, they say no.

Good luck!

[Camisotro]

A decent column will not have a problem with abrupt solvent changes.

there may still be a way in which you can combine column requilibration with the injection in a way that is beneficial. Think of the solvent generation train that happens at the pump as being offset from what is happening at the column outlet. So you can program a step back to the initial conditions before your last peak has eluted. So the reequilibration step of the next run and potentially the start of your gradient for the next run is programmed while your peaks are still eluting. It looks weird for the programming of the method, but it should work.

That's close to what I was imagining too as a possibility. While it's easy to program the pump to start re-equilibration before peaks have eluted (because of delay time), it's a little harder to start the next run's gradient during the previous run.

The reason is, it would make the next run's gradient depend on the time gap between injections. Usually this gap is about 30 seconds but it's not 100% constant. It's shortest when repeating an injection from the same vial, longer when switching to a new vial with a wash program, and longest when the autosampler pauses to rinse the buffer tubing between runs.

[Camisotro]

could you consider hiring a pump, or even talk to the MS manufacturer, and see if you can borrow a pump from them? If they think there's a sale in it, they might lend you one. The advantage of hiring/borrowing from the MS manufacturer is that they'll probably be able to sort out all software and installation issues, and give you the benefit of a working LC-MS system. The way to swing it with them is to point out that if the MS lab love it, and are considering further collaborations with labs that need chromatography, then this could easily be a future sale.

Worst case scenario, they say no.

Good luck!

I don't really have the connections or negotiation skills to make something like that happen. But it's a neat thought

Renting a pump, I hadn't considered that idea. I have no concept of how much it'd cost, but it'd really have to be worth it for that to be even under consideration.

[Uwe Neue]

Sh.... Well, tell the autosampler that he is gotta get his act together or he will get fired...

[Camisotro]

Sh.... Well, tell the autosampler that he is gotta get his act together or he will get fired...

Haha, I'll try that but he knows it's not a serious threat. We only have the one

[Uwe Neue]

Ok, I reread your info. A delay volume of 1 mL is not bad at all, and we used to run very fast LC/MS on systems like this. You objected, based on the idea that your loose a lot of sensitivity when you do that. Unless you have a MS system from the stone age, this should not be the case. So let me ask you the question why you think that your sensitivity goes down when you run at 1 mL/min. There could be problem in the thought, rather than a real problem.

Peak area decreases 5-fold when you change the flow rate 5-fold, because the peaks elute 5 times faster. But this is not a loss in sensitivity, unless the detector noise increases drastically with flow. With the MS systems that I know, the noise goes up a little bit with increased flow, bot not 5-fold for a flow rate change from 0.2 to 1.0 mL/min. Only maybe 50% or so. So to run at 1 mL/min is not a real problem.

[Tomasz]

You wrote you have 2 isocratic pumps. Try combining them together and build your own gradient pump. You need a low-volume mixer for your mobile phase. You will mix the 2 mobile phases from the pumps on the high pressure side. If your pumps have problems with controling low flows, add splitters (with some restriction capillaries) before the mixer. You can re-use the phase from each pump as you will splits the phase before the mixer.
I know this is a little crazy idea but maybe it could work