easy way for increasing resolutions

[arjanschoonen]

The theory
Instead of using a longer column, resolutions in GC can be increased substantially by providing backpressure on the column and slowing down the linear velocity to its new optimum. The longitudinal diffusion (B in Van Deemter equation) will be less and separations improve. This may be as well applicable to LC separations.

Backpressure in GC can be provided by mounting a thinner capillary -no stationary phase needed here- between column and detector.
In literature about GCxGC, where two columns are coupled in this manner, it is often stated that a peak gets 'focused'. Focussing means the same as decreasing longitudinal diffusion.

The experiment
Isopropylbenzene and n-propylbenzene were injected at different flows on a 30 m x 0.25 mm x 0.25 um DB-1 column connected to a BPX-50 column of i m x 0.1 mm x 0.1 um. Resolutions were calculated at 10% peak height. Linear velocities were calculated as the quotient of column length and retention time of methane. The optimum linear velocities (U opt.) were assigned to the highest resolutions. Pressures were calculated with Hagen-Poisseuille and the universal gas law (see http://www.chem.hope.edu/~polik/Chem345 ... osity.html)

i (m) 0; 1.5; 3.0
U opt. (cm/s) 30 (lit.); 15; 12.4
inlet p (bar) 1.95; 3.01; 4.05
back p (bar) 1.01; 2.52; 3.64
Resolution n.a.; 4.9; 6.1

You can see backpressure and resolution increase and the optimum velocity drop. The attached BPX50 column has little selectivity between iso- and n-propylbenzene, but it decreases the pressuredrop over the DB-1 column and hence increases the resolution. Unfortunately there is no resolution data without the BPX50 column attached.

The questions
Are you familiar with using backpressure for improving separations?
If so, how do you restrict your column outlet?
If not, why don't you try it?

[chromatographer1]

Yes, I am familiar with the technique. It has been used in industry for a couple of decades, just like industry has used multidimensional chromatography since the late 1960s. I even used the technique in my paper published in Analytical Chemistry in 1995, although this technique was used, not to increase resolution, but was used to sharpen volatile peaks that would normally be broadened by the large injection volume.

Thanks for posting the information.

Rodney George

[aldehyde]

Thats interesting, I'd love to see more data. Anyone else have experience with this? Are there down sides?

[chromatographer1]

The raising of the pressure under which the chromatography proceeds increases the temperature or flow rate that is required for any analyte to elute from the column. If one is trying to chromatograph analytes that require high temperatures, then this technique may be detrimental.

This technique allows a linear flow rate over the length of the column to be set at a level closer to the optimal, rather than having the linear flow change over the pressure drop of the column, ie slower at the head and faster at the tail.

Another downside is, of course, maintaining leak-proof connections.

best wishes,

Rodney George

[Peter Apps]

Increasing the back pressure will surely increase viscosity - which will decrease optimum linear velocity and increase the time taken to get the separation. This is equivalent to the marginally higher maximum resolution that you can get with nitrogen vs hydrogen as carrier, at a cost of runs that are about four times as long.

The focussing in GCxGC is achieved by cryofocussing at the end of the first dimension column, not by the back pressure created by the narrow bore second dimension column.

Peter

[lmb]

The theory
Instead of using a longer column, resolutions in GC can be increased substantially by providing backpressure on the column and slowing down the linear velocity to its new optimum.

I respectfully disagree.

If backpressure means the same as outlet pressure then, according to theory and experimental evidence,

1) backpressure only affects analysis time – the higher is the backpressure the longer is the analysis
2) backpressure has practically NO EFFECT on resolution that can be obtained in analysis of a given sample by a given column

CLARIFICATION for one-dimensional GC:

Develop a method with any backpressure you wish. Using GC Method Translation (available free of charge from http://www.chem.agilent.com/en-US/Suppo ... ation.aspx), this method can be translated into another method (with atmospheric outlet pressure for FID, with vacuum at the column outlet in GC-MS, etc.) which will yield almost exactly the same resolution for ALL peak pairs in a chromatogram. A choice of carrier gas also has practically no effect on the resolution that can be obtained in analysis of a given sample by a given column.

Another way to put it: If a desired resolution cannot be obtained at the column outlet pressure required by a specific detector (atmospheric for FID, vacuum for MS, etc.), no further improvement can be obtained at any other outlet pressure (or with other carrier gas). Changing stationary phase, using a longer and/or narrower column are the ONLY viable alternatives for improving the resolution.

This theoretically justified outcome has been verified in numerous experiments.

CLARIFICATION for two-dimensional GC:

Peak resolution can be significantly increased (without increasing analysis time) by using two-dimensional GC instead of one-dimensional GC. However, this improvement has nothing to do with the backpressure. It has everything to do with using two columns of different polarities and refocusing the sample slices during their transfer from the primary column into the secondary one.

[chromatographer1]

I agree with Peter's statements. The separation improvement is not achieved without a cost in time.

The improvement of separation is generally at the 'front' end of the chromatogram and is similar to the data that arjanschoonen has presented. It also requires a careful setting of carrier gas flow.

This technique will not transform the world of chromatography. It has been practiced for 40 years and hasn't changed anything yet.

Rodney George

[thohry]

I think the increase of pressure will not affect the resolution (or at lease at the pressures used in normal GC). If you analyze a mix by GC-FID and GC-MS with the same column and the same conditions (temperature program, flowrate...), I can say that the 2 resolution levels are much the same. even though the pressure in GC-MS is lower.