Choice of compound for dead time measurement in LCMS/LC-RI.

[haagiboy]

Hi,

We have two different detectors coudled to a HPLC. We do not run them in series, but have to swap between the two detectors. One is a MSD, and the other is a RID.

The retention time is ofcourse different between the two. So in an effort to find which peaks correspond to which between the two detectors I am looking into finding the "dead time" of our two systems.

I want to inject a compound which will be unretained in our column and show up in our detector at time t0. That t0 will then be defined as the time it takes for the mobile phase to flow through the column and lines/tubings.

The relative retention times of compounds will then be calculated based on this t0.

So here is my question:

What type of compound will not be retained in a Agilent Hi-Plex Ca (duo) column? The column works mainly by ligand interaction and size. The mobile phase we use is pure distilled water.

From Agilent website:

"Agilent Hi-Plex Ca (Duo) uses a duo-calcium counter-ion to provide selectivity differences through its ability to interact with the hydroxyl group. Hi-Plex Ca (Duo) typically uses isocratic conditions with water as the eluent and temperature as the main variable for resolution control."



We are also looking into options on how to improve our resolution/separation of the peaks by changing our mobile phase, but we are not sure what mobile phases will work with our column and ESI-MSD! Any help in this topic will also be appreciated.

Image of RID (top) and MSD (bottom) chromatogram: https://i.imgur.com/JkbxtD1.png

Overlayed on each other:

https://i.imgur.com/UmgOzxw.png

Method description:

Agilent Technologies 1260 Infinity HPLC, with a Hi-Plex Ca (Duo) 300 x 6.5mm column, and a differential
refractive index detector (RID). Water was used as mobile phase, with a flow rate at 0.300
mL/min, injection volume was 5.0 µl, RID temperature 35 °C, and the column temperature
used was 80 °C.

[tom jupille]

Even *defining* t0 gets tricky, and measuring it is even worse; check these threads from the "old" Chromatogaphy Forum:
- http://www.lcresources.com/discus/messa ... 19991210pm
- http://www.lcresources.com/discus/messa ... 20041222pm
I suspect that *any* relatively early eluting standard compound would work for your purposes. I also suspect that the retention time differences you are seeing have more to do with column chemistry or temperature than they do with plumbing.

The retention time is of course different between the two.

I'm not sure about the "of course". Those are big columns and, unless you have a *horribly* plumbed system, there should be a negligible difference in the tubing volume. Your overlay chromatograms are showing retention differences on the order of a minute, so at 0.3 mL/min, you would have to have a volume difference on the order of 300 microliters. 0.007" id tubing has about 0.25 microliters of volume per centimeter of length; it would take around a 75 cm *difference* in tubing length to account for the retention time difference.

If this were my problem, the first thing I would do would be to swap the columns between the two systems and see if the retention times go with the columns.

[lmh]

Why do you need an unretained standard to determine the difference? Provided your chromatography is reproducible, it doesn't matter what you use; it should be possible to use any peak that you can recognise in both detectors. Run a standard using each, and see what the difference is?

[haagiboy]

The retention time is of course different between the two.

I'm not sure about the "of course". Those are big columns and, unless you have a *horribly* plumbed system, there should be a negligible difference in the tubing volume. Your overlay chromatograms are showing retention differences on the order of a minute, so at 0.3 mL/min, you would have to have a volume difference on the order of 300 microliters. 0.007" id tubing has about 0.25 microliters of volume per centimeter of length; it would take around a 75 cm *difference* in tubing length to account for the retention time difference.

If this were my problem, the first thing I would do would be to swap the columns between the two systems and see if the retention times go with the columns.

Thanks for the reply!

I might not have been clear enough, but it's the same system and column. The msd and rid are right next to eachother, but not connected in series. We literally change tubing connectors after the column to guide the flow into either the msd or the rid.

In other words, we first separate the compounds in the column, which is the same for both detectors since it's the same setup, and then we either send it to msd where the peak for a known compound is for example 18.8 minutes. That same compound, when the lc column is connected to the rid will have another retention time. Which most likely is due to horrible tubing.

I am a chemical engineer PhD student, and working with other chemical engineers. We got the msd half a year ago or so, in order to better identify several unknown peaks. I work with catalysis of biomass to transportation fuels. As of know we get cellulose into polyols, mostly diols and triols.

As for better separation, how would adding acetonitrile help? We can not run a gradient as we only have one pump.

In a perfect world I would be able to identify most of my compounds with the msd, and then identify them again in the rid based on the retention time. Can I add an internal standard and define relative retention time relative to that compound? And that relative retention time would be the same for both the msd and rid?

Some of the peaks in the rid show some shoulders and coelution problems. Which is bad, since we need to calculate the selectivity and yield of our products.

If I were to run standards of everything that we might have (based on chemistry and msd), it would cost an additional 2000 usd on top of the 50 standards we have done already.

[haagiboy]

Why do you need an unretained standard to determine the difference? Provided your chromatography is reproducible, it doesn't matter what you use; it should be possible to use any peak that you can recognise in both detectors. Run a standard using each, and see what the difference is?

Yeah this was more or less what I was hoping I could do
I will try to find an internal standard that we can use and define the relative retention time based on that peak. That would work fine for identifying which peaks are which, msd vs rid?

[lmh]

I think it might be the natural desire to calculate relative retention times that is making your life unnecessarily complicated, because you're right, it would force you to know the system dead volume from injection to detector for both detectors.

But actually you can think about this another way. Analytes appear out the end of the column at some time T. They then take an extra A seconds to arrive at the RID, or an extra B seconds to arrive at the MS. A and B don't depend on chromatography and are completely constant. They're simply a matter of flow-rate and the quantity of tubing between the end of the column and the detector (and some MS detectors have a substantial volume in their spray needle, which people tend to forget, but which delays and broadens MS peaks).
You can measure A-B using any peak, and you don't need to know the dead volume for this. If any known peak arrives at the MS 0.283min later than it would have arrived at the RID, then every other peak should arrive at the MS 0.283min after it arrived at the RID, so if you want the mass spectrum of an unknown peak with RT 4.0min by RID, look in the MS at 4.283min (and trust that your chromatography and plumbing are reproducible!).
Hope this helps...

[haagiboy]

Thank you! Yes, this is what I wanted to confirm. That delta RT is constant between rid and msd.