Although I've reviewed some theories, I hope to know effects of flow rate on the GC resolution (for both packed and capillary column GC).
Thanks a lot!
By Dan Vassilaros on Wednesday, July 14, 1999 - 10:37 am:
The HETP vs average linear velocity plots (known as "van Deemter" plots) go through a minimum where the linear velocity of the carrier gas is optimized in terms of its impact on resolution. At that optimum point, the carrier gas linear velocity in a sense "balances" resolution loss due to longitudinal diffusion with that due to ineffective radial mass transfer. These plots are generated from isothermal studies.
The basic rule is that a carrier gas linear velocity lower than the optimum will lead to decreased resolution (because of greater band broadening due to longitudinal diffusion), and a linear velocity higher than the optimum will lead to decreased resolution (because of less effective radial mass transfer). This is straight-forward, not a lot of magic here, just a matter of finding the best compromise among several independent forces affecting the band width.
The matter starts getting tricky if you are doing temperature programmed analyses and the carrier gas is pressure-controlled. As you know, the linear velocity decreases with temperature increase because of increasing gas viscosity, given a constant head pressure. YOu can figure out which side of the optimum point the linear velocity of the carrier gas will be, at the higher end of the temperature program, if you set it at the optimum value at the lower end of the temperature ramp. You have to account for the 20-50% drop in carrier gas linear velocity with a 100-200 C temperature rise.
Even EPC with so-called flow control does not necessarily effectively ramp the linear velocity to account for temperature-induced viscosity increase.
By D.Z.Zou on Wednesday, July 14, 1999 - 02:53 pm:
Thank you very much!
By John Hinshaw on Thursday, July 15, 1999 - 11:37 am:
A plot of average linear carrier-gas velocity vs. column effiency (HETP, or height equivalent to a theoretical plate) shows that in most situations the loss of efficiency caused by moving off the optimum at any single temperature is less when moving to higher velocities than to lower velocities; the curve is flatter at higher speeds. The exception to this is for very thick film columns (df > 2 um) where the losses may be similar.
In general there are two situations that arise concerning the carrier gas velocity when temperature programming the column. First, classical (non-EPC) instruments with fixed-pressure inlets should be set 20-30% above the pressure for optimum flow at the initial oven temperature. Then, when the oven temperature increases, the effect of the resulting drop in the average linear velocity will be minimized because the column is operating on the flatter portion of the HETP curve. This is especially true for Hydrogen carrier gas.
For instruments with computer-controlled inlet pressure, use either the constant flow or constant average linear velocity modes. Both will increase the carrier pressure with increasing oven temperature, and will keep the column operating on the flat part of the curve. Even in this situation, it is still a good idea to set the initial velocity somewhat above optimum. A bonus side-effect will be shorter analysis times due to the increased pressure (relative to constant pressure operation) at higher temperatures.
Having said all that, the optimum velocity shifts with increasing temperature, due to changes in gas-gas and gas-liquid diffusion coefficients. These relationships have been quantitatively measured for only a few analyte/carrier/stationary phase combinations, but in general the shift in optimum is small compared to the effects of the changing carrier-gas viscosity.
By michael dunn (host-209-214-119-128.bna.bellsouth.net - 18.104.22.168) on Monday, July 19, 1999 - 06:16 pm:
practically speaking, inject smallest volume possible for your application, run carrier gas velocity as fast as you can to resolve your most critical compounds
Posting is currently disabled in this topic. Contact your discussion moderator for more information.