Fluctuating baseline on HPLC-RI

[Alps_hiker]

Hi all,

We have recently observed an intermittent fluctuating baseline on our HPLC-RI system when running sugars.

It's the same column, and we don't run anything other than a sugars determination method on it with extracted centrifuged PTFE-filtered food extracts. The mobile phase is ACN:H2O, 81:19, and we have a temperature gradient upto 40C on the detector.

Has anyone else experienced this before?

Many Thanks,

James.

[Consumer Products Guy]

we have a temperature gradient up to 40C on the detector

1. I don't really understand "temperature gradient up to 40C on the detector". We never used column temperature gradient or detector gradient with RID (our detectors couldn't do that, and not sure why an RID detector would need to do that).

2. We'd run the system with mobile phase recycling overnight before use with RID.

3. We insulated any lines as well as possible.

4. We'd pre-mix mobile phase (81:19 works differently than 80:20? 80:20 is within range of instrument or human error), and bypass the mixing valve and pump from one pump only when using RID.

[DR]

Include a picture showing the fluctuations and include time on your x axis...

[Alps_hiker]

Hi, thanks for the reply ...

we have a temperature gradient up to 40C on the detector

1. I don't really understand "temperature gradient up to 40C on the detector". We never used column temperature gradient or detector gradient with RID (our detectors couldn't do that, and not sure why an RID detector would need to do that).

I should have been more clear on the gradient. I meant to say that the temperature increases from column to detector, but each unit is set at a specific temperature. The detector is at 40C.

2. We'd run the system with mobile phase recycling overnight before use with RID.

We've had a brief look at recycling our mobile phase. With the cost of Acetonitrile, it would be a good idea. How do you recycle it?

3. We insulated any lines as well as possible.

I know temperature fluctuations cause variations in RI response, but the wave is repeating every few seconds, so it's unlikely to be the cause of the problem

4. We'd pre-mix mobile phase (81:19 works differently than 80:20? 80:20 is within range of instrument or human error), and bypass the mixing valve and pump from one pump only when using RID.

We also prepare the mobile phase, and have found the 81:19 ratio the best for our needs.

I have attached a photo of the chromatography, didn't realise it had to be a URL instead of the typical JPEG format.

I think the fluctuation on the baseline may be due to the pump, as one pump stroke is roughly the same time as one wave cycle. The pressure is varying by approx 20psi during a run, with a starting pressure of 1400psi.

Any further suggestions welcome,

Many Thanks.

[Alps_hiker]

[TylerSmith123]

Hi Alps,

Unfortunately, your image doesn't work (unless you want to sign into microsoft).

Now to the issue: is this change in pressure (+- 20 psi) a constant issue with the pump's strokes, or just an overall trend for the day? I could reasonably believe, in any traditional system, that the psi could shift 20+- throughout the day due to out-side, uncontrollable, phenomenon.

I'd bring up some of your chromatograms and, assuming you can measure and track pressure, to look at the fluctuations on those and see if this is something that is continuous, or just constantly rising. As your pump completes a stroke, or as your pump is running, it may slip and cause that sudden spike in pressure you're seeing. Recently, the seals on my pump were getting old and displayed a regular jump of about 15+- psi, and I was able to fix that by replacing the seals. This was very easy to notice though, as I'm constantly staring at my instrument haha.

Another way you could test the pump hypothesis (ie, if your flow is off perhaps and your pump isn't pumping the selected amount) simply take your flow-path to that goes to waste and redirect it into a flask and measure the amount of liquid over a certain period of time. The volume should match your flow-rate over the allotted time. However, you would also see this issue manifest in your separation and, since you haven't indicated any changes, this is likely not the issue.

I hope some of this helps,
TS.

[uzman]

You may try to connect a 1meter of 0.17mm ID PEEK tubing ( as a pulse damper ) between pump and injector.

Most of the time this will help to reduce the noise of RI detectors.

Pump pressure will increase slightly.

[Alps_hiker]

Hi there,

Thanks for the suggestions, I'll try them out this week.

Incidentally, the piston and support seals (or rings) were only replaced a few months ago (February), so it would be disappointing if those were the issue with this.

Best Regards.

[Alps_hiker]

[Consumer Products Guy]

Hi, thanks for the reply ...

we have a temperature gradient up to 40C on the detector

1. I don't really understand "temperature gradient up to 40C on the detector". We never used column temperature gradient or detector gradient with RID (our detectors couldn't do that, and not sure why an RID detector would need to do that).

I should have been more clear on the gradient. I meant to say that the temperature increases from column to detector, but each unit is set at a specific temperature. The detector is at 40C.
We always had both at same temperature, had the line between them short and insulated.


2. We'd run the system with mobile phase recycling overnight before use with RID.

We've had a brief look at recycling our mobile phase. With the cost of Acetonitrile, it would be a good idea. How do you recycle it?
At night, before such instrument use, we'd pump mobile phase back into the solvent reservoir, all night. Instrument at "use" flow rate and temperatures. We did not recycle during actual assays.

3. We insulated any lines as well as possible.

I know temperature fluctuations cause variations in RI response, but the wave is repeating every few seconds, so it's unlikely to be the cause of the problem

4. We'd pre-mix mobile phase (81:19 works differently than 80:20? 80:20 is within range of instrument or human error), and bypass the mixing valve and pump from one pump only when using RID.

We also prepare the mobile phase, and have found the 81:19 ratio the best for our needs.

I have attached a photo of the chromatography, didn't realise it had to be a URL instead of the typical JPEG format.

I think the fluctuation on the baseline may be due to the pump, as one pump stroke is roughly the same time as one wave cycle. The pressure is varying by approx 20psi during a run, with a starting pressure of 1400psi.

Any further suggestions welcome,

Many Thanks.

[Multidimensional]

Some quick comments and questions:.
(1) Alps, I do not think you understood "Consumer Products Guy's" comment about :"recyling". I think he meant they used to run the system, recycling the mobile phase when NOT USING the system (as in, BEFORE running samples to allow the column and system plenty of time to stabilize before analysis). You should NOT recycle the mobile phase when running samples.
(2) You report a +/- 20 psi change for 1400 psi system pressure as a possible "problem"? This is NOT a problem or indication of pressure instability.
(3) Is there someone in your group with many years of industrial chromatography experience who is familiar with your specific HPLC system and software? If not, please consider hiring someone to help you.
(4) The temp of the column and detector should be the same. This is really important ! RID is super sensitive to temperature.
(5) Are you using the RID flush valve to purge the flow cell before auto-zeroing the detector to start a run?
(6) As requested before, if you could post actual chromatograms with scales shown and the the method conditions (column, dimensions, flow etc), then someone might be able to provide a constructive tip.

Here is a free link to an authoritative article written about issues using a RID. There are lots of tips in there they may or may not assist you (as we do not have enough information to know which of many possible issues you may be experiencing.

"HPLC Baseline Stabilization Tips for Refractive Index Detectors (RI or RID)"; https://hplctips.blogspot.com/2018/02/h ... s-for.html

[DR]

https://www.chromforum.org/viewtopic.php?f=1&t=2617