Reference Deconvolution

Top  Previous  Next

Reference Deconvolution is a signal processing method to remove experimental effects that broaden lines. Reference deconvolution is a technique which uses the shape of a single resonance line to measure the actual frequency distribution produced by the local magnetic field in homogeneity and then deconvolves that distribution from the whole spectrum. It is a simple linear process which requires no prior knowledge of the number of lines, their intensities, or their relaxation characteristics. No fitting procedures are used.

 

This ability to generate sharp line spectra is very important for DOSY measurements, improving t1 noise in 2D spectra, qNMR (through reducing truncation effects), and we can expect it to also add robustness to Verification, metabolomics, and in many more real-life situations.

 

Here you can see an example of the NMR spectrum which has been folded several times to optimize digitization. To correct the severe lineshape problems and spinning sidebands, we will apply 'reference deconvolution' by following the menu 'Processing/More Processing/Reference Deconvolution'. After that you will notice that the mouse cursor will change to an integral curve; the next step will be to click&drag over the region of interest (which must be always a singlet) to display the Reference Deconvolution dialog box, from where additional settings can be applied:

 

Reference Deconvolution

 

After having clicked on the OK button, you will get the result:

 

Reference Deconvolution2

 

Checking the 'Advanced' box will display additional tabs with settings for the 13C and 28Si satellites; to minimize the artifacts and to select the peaks of interest for the reference deconvolution model:

 

reference deconvolution_advanced

 

You can choose the ideal line shape (Gaussian or Lorentzian) from the dialog above.

 

The model settings can be saved to be used with other datasets by clicking on the 'Load/Save' buttons.Saving the model will store:

 

1. The experimental shape between specified limits.

2. The peaks found by peak picking in the same region, including info if the peaks are selected or deselected by the user.

3. The parameter which shows if user want to use experimental shape or the synthetic shape from the peaks.

 

Loading a settings file and clicking on the "Show Reference" button, will display all the GSD peaks in the reference region (the non-selected ones will appear in grey). You can switch to the experimental shape (by unchecking the option "Create shape from peaks"):

 

RefDecAdv

 

See also this blog post