Filtering blank peaks

Hello,

You could create a CC that looks to see if there is a peak in the blank within specific RT limits. The biggest issue with this is you cannot keep the same peak in the blank from being used by more than one peak in the sample. See the attached pdf for a CCF file to do this calculation. The attached CC finds the closest peak in the blank within the RT range and returns the RT and Area. It then uses the RT and 10 * Area to set a flag on the peak if it was found in the blank. It will not use any identified peaks from the blank. In the tables, I use the CC flag as a filter. You can see the peaks at 0.591 and 0.654 both were in the +-0.1 window with the 0.65 peak in the blank. The peak area of the 0.591 was more than 10x the blank peak area so it was not flagged, but the 0.654 peak was within the 10x area range and was flagged. If these had both been in the area window, then both would have been flagged against the same blank peak. 

Marty

/cfs-file/__key/communityserver-discussions-components-files/26/Blank-Subtract.pdf

Thank you Martin 

I got the idea,  i will try using this on my sequence and see if i can get some good results

I could also think of a couple other options.

First, you can adopt a blank subtraction policy, which should remove any peaks in nonblanks that are present in blanks. This also makes chromatograms look great, but if the blanks are a different solvent than the samples, then it won't be applicable and it is a subtraction so having variable sizes of those peaks will cause some artifacts in the baseline. However, I would argue that if they still exist after blank subtraction that their concentrations are likely just higher in the samples and they should be measured.

Second, you can adopt a naming convention for peaks in blanks. We do this for our unknowns. For example, we have the compounds named starting with "Unknown", then we filter them out of specific tables and filter them into other ones. You could do something like starting with "Blank_Impurity" and do the same at least for the tables.

Hello,

My concern with chromatographic blank subtraction has to do more with the chromatography shifting in RT. When that happens even if the blank peak is smaller than the peak in the standard or sample it can shift enough to cause artifacts on your peaks. I typically only look to blank subtraction for baseline drift correction or some very specific applications. As Cole said if you willing to name your impurities the subtraction becomes much easier and the issue with using the same blank peak multiple times would be resolved. The CC would need to change to identified peaks and a filter clause for the current compound name. 

Marty