S/N vs column length, GCMS

[Georgsv]

Hi,

In my opinion, column length is inversely related to the S/N due to "column bleeding".

Could anyone comment on the S/N dependency on column length for a GCMS?

Thanks,
George

[Peter Apps]

Welcome.

Two factors come into play - column bleed you have already identified, and peak width. Longer columns elute wider peaks, wider peaks are lower for a given area, hence less detectable.

Peter

[Dimitri]

Can we really gauge the relationship of column length to S/N without considering theoretical plates? And that being said, won't that inevitably lead us back to the Van Deemter equation?

[Amirav]

George
The most important factor that affects the column bleed as a noise source is the elution temperature.
If you use twice shorter column and the same temperature program rate the elution time will be about similar but the elution temperature will be about 20C lower hence the column bleed would be much lower (exponentially lower with the temperature) hence your S/N will improve IF the column bleed is the most important noise source. This effect is far greater than column bleed rate that lineraly depends on the column length.
We work with the Aviv Analytical 5975-SMB GC-MS with Cold EI with short columns and high column flow rates and use this effect to increase the range of compounds amenable for aalysis
For references please write me at amirav@tau.ac.il
Amirav

[James_Ball]

Column bleed would be affected by both column length and film thickness. Signal to noise would be affected by column bleed versus peak height. Peak height would be affected by column length, film thickness, column diameter, carrier flow rate and temperature program. Faster temperature ramps produce sharper peaks and slower ramps produce wider peaks.

At maximum temperature for the phase I would doubt that cutting the length in half would reduce the column bleed signal by half, but at a lower temperature it might.