Chromatography Forum

Hello all,

I'm completely new to the forum and I mostly registered because I realised that the client support and lab technicians I work with are not real experts concerning my system.
My goal is to separate and quantify different monosaccharides and for that I use the following setup:

- All hardware except for the tubing and the column is from the Agilent 1260 Infinity series
- Iso quartenary pump G1310B (includes degasser)
- ALS (autosampler) G1329B
- TCC (column oven) G1316A
- RID (refractive index detector) G1362A

I tried three columns:
- Shodex SP0810 + guard column
Optimised the separation at 86/14 V% Water/Acetonitrile (premixed, isocratic), flow = 1ml/min, Tcolumn= 80 C, Trid = 55 C, injection 10 ul, 20 min. run
- Bio-rad HPX-87H
Optimised the separation at 5 mM H2SO4 in Water (premixed, isocratic), flow = 0.6 ml/min, Tcolumn= 65 C, Trid = 55 C, injection 10 ul, 20 min. run
- Zorbax NH2
Tried 80/20 V% Acetonitrile/water (premixed, isocratic), flow = 1ml/min, Tcolumn= 40 C, Trid = 40 C, injection 10 ul, 20 min. run

The Shodex gave the best results in terms of separation, but the solvent peak interferes with the glucose peak, the baseline is really shaky in the low concentration range.
The Bio-rad could not separate the sugars completely, only 5 of the 8 sugars were clean peaks (but then it is not a specialized sugar column)
The Zorbax NH2 I could not check because of a bad baseline.

In all cases the pump pressure does not have a bigger deviation than +/- 0.4 bar. Everything without acetonitrile gives a slick, flat, clean baseline, but as soon as I enter acetonitrile in the system (premixed, isocratic!) the baseline starts jumping like crazy! I cannot distinghuish concentrations of 0.5 mg/ml from the baseline. This is after leaving the system flush for > 2h and flushign the purge cell of the RID > 30 minutes.

I know that the difference in viscosity and refractive index can give problems, but I thought that if the solvent is properly mixed and degassed that should not happen anymore.

If that helps I can post some spectra with and without injection for the Zorbax (since I use the highest acetonitrile concentration there).

Hopefully I am just inexperienced and this matter can be solved very easily, I'm looking forward to your comments!

Best regards,

Adriaan

When we want sensitivity with RID, we bypass the multi-channel gradient valve (MCGV) which smooths the RID baseline. If using a quaternary HPLC pump system, it is recommended to bypass the multi-channel mixing valve using an adapter (Agilent Technologies #0100-1847) that is routed directly into the HPLC pump active inlet valve.

A possible cause , the column is not clean and acetonitrile eluting some junk.

Also, route the waste tubing of RID nearly half a meter up , then send to the waste , this will help to reduce noise , and fill the recycle tubing , even it is not used.

I cannot help with the noise issue, but if you are having resolution problems you might want to look at this web-site for carbohydrate separations.

http://www.nacalai.co.jp/global/update/ ... _hplc.html

MeCN/H2O mobile phase.....ELSD.....

I have no connection with the company, just came across the website.

Hello

What is you optical unit temperature?
If optical unit temperature is less than your column temperature your mobile phase will be rapidly cooled down (even if it is few degrees difference) - it could be a reason for unstable baseline.

Regards

Tomasz Kubowicz

Wow, such a fast response and all the answers seem worth trying or considerate, I love it!

@Consumer Products Guy: We have plenty of spare products, both brandless or Agilent, I will try the bypass as soon as I found/ordered the adapter. I guess you mean bypassing from the exit of the degasser to the entrance of the pumphead?

@Uzman: The current column was operated under the same conditions with only standards on a different system with an ELSD detector for three weeks. I also tried running without any column (running with a clean constraint from pump to detector) with water, perfect baseline, and with with acetonitrile/water, noisy baseline again.

I will try routing the waste up and filling the recycle tubing, thanks for the tip.

@JMB: The separation for higher concentrations is fine with the Zorbax, my main concern now is the baseline. I bookmarked the webpage and will look into it when I solved the baseline problem, thank you.

@Tomasz: In my first description I posted Trid, with which I meant optical unit temperature (Temperature RID). For the Zorbax column, the oven and RID temperatures are set at the same values. I usually monitor pump pressure, column temperature and detector unit temperature. The column temperature is not more off than 0.5 degrees throughout a run and for the optical unit temperature it is even 0.1 degrees Celcius maximum.

As soon as I have improvements (or not) I will post again!

Best,

Adriaan

@Consumer Products Guy: We have plenty of spare products, both brandless or Agilent, I will try the bypass as soon as I found/ordered the adapter. I guess you mean bypassing from the exit of the degasser to the entrance of the pumphead?

Yes. Here's what I do.

1. I screw the adapter directly into the pumphead (and use a union and column plug to plug the line fitting there).

2. I unscrew the tube (from the degasser outlet) into the MCGV and screw that into the Agilent adapter at the pump.

3. I use a "low pressure plug" to plug off the open hole in the MCGV.

4. I use a sandwich bag to hold those parts when not bypassing the MCGV, to keep them all together.

Ok, but the problem is that the adapter is different than the normal connection that comes from the MCGV, so I cannot put it directly in the pumphead.
You can see this clearly at page 150 of the manual:
http://www.chem.agilent.com/Library/use ... USR_EN.pdf

Can I not just use the tubing that comes out of the degasser and normally goes into the MCGV to connect the degasser to the pump inlet?

Best,

Adriaan

Ok, but the problem is that the adapter is different than the normal connection that comes from the MCGV, so I cannot put it directly in the pumphead.
You can see this clearly at page 150 of the manual:
http://www.chem.agilent.com/Library/use ... USR_EN.pdf

Can I not just use the tubing that comes out of the degasser and normally goes into the MCGV to connect the degasser to the pump inlet?

Best,

Adriaan

I don't see anything on page 150 of that link that is relevant.

"When we want sensitivity with RID, we bypass the multi-channel gradient valve (MCGV) which smooths the RID baseline. If using a quaternary HPLC pump system, it is recommended to bypass the multi-channel mixing valve using an adapter (Agilent Technologies #0100-1847) that is routed directly into the HPLC pump active inlet valve."

The female thread of the active inlet valve is typical Swagelok HPLC for 1/16 inch tubing, 1/4 inch-28. The male thread of tubing to and from the degasser is 10-32. That's why the adapter is needed, has male 1/4-28 and female 10-32 threads. This is what #0100-1847 looks like, perhaps yours already has this?



This has been discussed before, see http://46.51.172.102/viewtopic.php?t=16478&p=80154

Maybe it is because of the note at page 166:

By default, 1260 Infinity pumps do not have an active inlet valve. If an AIV shall be installed,
please contact your Agilent service representative

Anyway, my pumphead looks like this and the connection at the bottom (which I considered to be the pump head inlet) is not directly connected to the degasser:


I found the adapter in our inventory, but no place to connect it to the pump.

Best,

Adriaan

From your photo: connect diectly from degasser like shown in your photo, no adapter needed.

Sorry, I'm not familiar with anything higher in model number than Agilent 1200.