Chromatography Forum

Hi everybody - Happy Monday.

I've been able to fix an old Shimadzu UPLC and C18 column with the forum's help. You all are awesome. Now I'm trying to sort out an issue with another Shimadzu UPLC but using a phenomenex rezex RHM-monosaccharide column for C6 and C5 sugar and acid-sugar detection. I washed it with 0.025 M H2SO4 at a flow of 0.2 mL/min for about 24 hours at 85-degrees C. Swapped it over to nano pure filtered water until it equilibrated. Let it run for another 4 hours. The baseline looked pretty solid. I loaded a few known samples (Glucose, Xylose, Erythrose) turned the flow up to 0.5 mL/min and dropped the temp to 65-degrees. RID is at 65 as well. NO UV detection in the method. Samples run until the first peak, then - I'll attach a screen shot here... What could be making the know standard jump like this? Pressure is all good.





Thoughts appreciated!!

Best,
Carbon

Good monday

Maybe the column dimensions would be a helpful information too.

But...

...washed it with 0.025 M H2SO4 at a flow of 0.2 mL/min for about 24 hours at 85-degrees C.
Swapped it over to nano pure filtered water until it equilibrated. Let it run for another 4 hours. The baseline looked pretty solid.
I loaded a few known samples (Glucose, Xylose, Erythrose) turned the flow up to 0.5 mL/min and dropped the temp to 65-degrees...

...is a bit confusing, especially the last sentence and no clear reference to the pictures.

1) if you changed the temperature and flow just shortly before your injection, then your system is in non-equlibrated state. The flow may be a small issue but the temperature will take a while to reach 65°C.
> the pressure trace may be ok but has increased by about 50% between the two pictures, if looked closely.

>> let the system run at the same temperature as your separation method. Low flow is ok, then ramp up to working flow and wait until the pressure is stable again. Then inject your samples.
I would even keep the temperature for (overnight) standby after the sequence, as thermal equilibration is slow.

2) do you have a backpressure regulator at the RID outlet? This may help for steady baseline and signal.
If not, add a dedicated one (there are posts with links on other topics in this forum) or just connect a piece of smallbore tubing - but make sure it stays within the limits of the RID cell under all working and flushing conditions (flow, temperature, solvent)! You may also want to put a big, red warning label with the maximum flow on the PC screen.
Maybe calculate the length of the tubing with the viscosity of MeOH 50% and only about half of the limit of the cell, to stay on the safe side).
The pressure drop can be calculated by the Hagen-Poisseuille equation or use some online calculators e.g. the one on the Waters website: https://www.waters.com/webassets/other/ ... itters.htm

Sorry. Column dims are: 300 x 7.8mm with 8 micron particle size

I will have to investigate the pressure regulator question and post back.

After the temp adjustment, the mobile phase flowed for about 90 more minutes, but I let it run overnight (last night) so I will look further into your other questions and try the injection again after the overnight equilibriation.

another question:

How long is your runtime per injection?

Maybe you see some artefacts and baseline shift from previous injections?

From the column dimension and flow rate, one could expect a t0 of about 17-19 min.
This would correspond to the first main peak at 19.5 min in the first picture (or maybe it's the signal increase at 22.5 min, but then maybe check the flowrate accuracy of your system).
But in the second picture, there is a real peak already at 15.5 min, maybe even two peaks at 10-12.5 min.

Maybe you can check with an application note from Phenomenex to prepare a simple refernce mixture (like you already did, but I don't know if there are reference chromatograms available).
Maybe prolong your runtime until you're sure to get all of the substances from the mixture out and the signal returns back to baseline before the next injection.

Do several injections to check repeatability and maybe change the concentrations of the test compounds to see if the detector is working as excpected too.

I've had similar experience with this column in the past. Some of the things which had contributed to the unstable baseline were
- air coming from air conditioners (apparently)
- bubbles being formed due to near-empty level of liquid used for purging lines
- dissolved gases in mobile phase
- sudden changes in flow rate & cell temperature not being constant/stable.

I'm not sure which one of these issues has the largest effect, but since I've resolved them the baseline has been relatively stable.