s/n: RMS - interpretation

[NoraStern]

I (student) posted a chromatogram of my TQ Quattro premier below (Quercetin) where I tried to meaure S/N ratio. Unfortunately I have a result of 394.91 (see above peak). How should I interprete this result? In litertaure s/n is normally 3 or something similar. I'm using Masslynx 4.1

[Rndirk]

By definition it means that you have about 400 times more signal than you have noise, in the regions where your indicated both signal and noise to be measured (indicated by the red lines i presume).

In literature often a S/N of 3 is used to indicate the minimal S/N you need to be talking about an actual peak. This is not a golden rule for everything though.

It's not part of your question but I can't resist: you have way too many data points leading to a 'noisy' peak. How broad is it (in retention time X-axis)?

[NoraStern]

Hi,

thanks for your answer! And you are totally right, that my peak looks weird. I have a very strong tailing and my peak including the tailing is about 1.5 min broad. The peak in the hplc chromatogram ist just 0.5 min long so I assume that I have lots of noise in the TQ? But from where does it come? I'm just measuring quercetin dissolved in methanol. My mobile phase is acetonitrile and water with 0.03 % acetic acid. I dissolved the quercetin in the mobile phase but it didn't change the peak shape.

Unfortunalety I have no one who could help me at the lab. I'm trying to improve the set up of the TQ and the HPLC but nothing helped yet. So, if you have any ideas, i am happy to hear them!

[JMB]

The Broad tailing peak that you see by mass spec. is NOT due to noise; it is signal from the analyte.
The noise should be measured where there is no signal, somewhere in the range 25-75 scans.
I would bet $$$ that your peak in the HPLC chromatogram is virtually symmetrical and that the tailing observed in the mass chromatograms is due to interactions of the analyte with the steel components of the ESI probe/source.
I have experienced a similar, Broad tailing peak in mass chromatograms when working with sodium dodecyl sulphonate (SDS).

[bunnahabhain]

JMB: Were you able to solve this tailing issue? I have seen similar problem with analytes containing phosphate functional group and quarternary ammonium compounds. :/

PS: Sorry for hijacking this thread...

[NoraStern]

That sounds very interesting, JMB! So what can I do to improve my signal in the TQ? Have your read about this in literature or can you recommend literature containing this problem?

I am cleaning the source and the T-wave every two months because I'm not using it that often and just measuring my standard solution. My HPLC peak (DAD) looks quite symmetrical but has a small tailing. But still it looks way better than the MS peak....

[Rndirk]

Is the HPLC-DAD chromatogram measured on the same instrument, under the same chromatographic conditions? I'm not convinced that the extensive width of your peak is due to broadening in the probe.

JMB: Were you able to solve this tailing issue? I have seen similar problem with analytes containing phosphate functional group and quarternary ammonium compounds. :/

PS: Sorry for hijacking this thread...

I can recommend the following article for interaction between phosphate functional groups and stainless steel:

Y. Asakawa et al. / J. Chromatogr. A 1198–1199 (2008) 80–86 "Suppression effects of carbonate on the interaction between stainless steel and phosphate groups of phosphate compounds in high-performance liquid chromatography and electrospray ionization mass spectrometry".

I would also like to note that the ESI probe is not the only part in most LCMS systems where the analyte is in contact with stainless steel !

[NoraStern]

I have two different machines, the HPLC with the DAD detector is from Agilent (software chemstation) and the TQ from Waters quattro premier (software Masslynx). Could it be that the communication is not working correctly between the two systems?

I thought that my broading comes from the manual set-up...therfore I wanted to tune the ESI and the three quadrupoles again.

[Rndirk]

Peak broadening in retention does not happen (or very rarely) in the detector itself. Also, i don't see how the communication between the instruments is relevant.

I was implying in my previous post that, in my opinion, your peak is broader on instrument 1 compared to instrument 2 because of differences in the LC part of the system (column, mobile phases, gradient, ....)

[NoraStern]

Maybe we misunderstood each other. I have two machines standig next to each other and they are connected. So the sample goes from the HPLC to the TQ. That's why I mentioned the communication. Therefore I do not understand why my peak is broader in the MS chromatogram than in the HPLC chromatogram. There are no differences in the LC system because there is just one. Thats why I was interested to hear the theory about interactions with steel components because I cannot explain myself the discrepancy between HPLC and TQ.

Actually I should use the TQ because I can detect smaller amounts but unfortunately my method is not very sensitive and I have this broad peak with a strong tailing

[JMB]

JMB: Were you able to solve this tailing issue? I have seen similar problem with analytes containing phosphate functional group and quarternary ammonium compounds. :/

PS: Sorry for hijacking this thread...

I was not.

[JMB]

Run any NEUTRAL, non-ionic, non-hydroxy compound through the HPLC-ESI-MS system and compare peak widths.
This will show whether,
1) there is significant broadening in the ESI source. There should not be!
2) your communication/acquisition parameters are set OK.
3) You should observe a delay of a few seconds due to transit of the analyte from UV detector to MS detector. The peak width will also increase very slightly.

In my opinion, Your observed peak broadening is consistent with multiple adsorption/desorption cycles of Quercetin in the ESI interface.

Until the manufacturers come up with a passivated steel interface............

[JMB]

I cannot access the Asakawa paper mentioned above, but you may find that a mobile phase of MeCN/ aqueous carbonate (delete the acetic acid) will give acceptable HPLC results and non-broadening MS peak.
Before doing that, however, make up the Quercetin in MeCN/aq. carbonate and inject directly into MS; this will let you know if you can reduce peak broadening.

[BZK]

You can't perform an accurate measure of S/N with your peak. The top of peak is very noisy. In that point the S/N could be 15/20. Software is stupid because make a baseline measure, before or after or both of peak, and take as right that the noise is the same during the peak eluition. This is not your case. You need to do a job for discovering why your top of peak is so noisy. After that all other consideration can be done.

[mattmullaney]

Hi NoraStern,

I am curious, what is the Quercetin peak width in minutes from the PDA-generated chromatogram as compared to that of the total ion chromatogram? There must be a way to show both chromatograms using time as the domain (x-axis). Also, knowing the data rates for the PDA as well as the MS may be helpful.

I was taught to use PEEKSil for the tubing connecting the PDA to the MS, perhaps that would help here.