In lab you improved the signaltonoise ratio SN for the Raman

In lab you improved the signal-to-noise ratio (S/N) for the Raman experiment by averaging spectra. If you want to improve the S/N by a factor of 4 (beyond what you observe for a single spectrum), how many spectra should you average? You are setting the sampling parameters on a high field FT-MS instrument. If you want to detect signals up to a frequency of 25,000 Hz with a resolution (minimum distinguishable step in frequency between signals) of 0.5 Hz, then what is the minimum sampling rate at which you should record the data? what is an appropriate observation time over which to record the data?

Solution

(4)

The S/N ratio for a coadded spectrum is given by

Sx/Nx =(n)^0.5 (S/N)

4*S/N=(n)^0.5(S/N)

n=16

the number of spectra should average is 16