different sepration with same column flow

[mr3573]

I analyzed a sample containing 4 components with isothermal and programming method. In isothermal method two last components was fairly separated while with programming two components merge together and totally three peaks appeared.
Temperature of isothermal was 130 C and retention time of desired peaks was about 7.4. in that case linear velocity shown by GC was 23 and constant flow was 1.3
In programming manner oven temperature has increased so that desired peaks (converted to one) appears at 130 C, linear velocity and column flow was exactly as isothermal but retention time changed to 12.2
Question is:
While linear velocity and column oven and column flow are the same for specific peak(or peaks) and their appears in different retention time, what is reason of different separation(no separation versus fairly separated)?
The column is CP-wax 52 CB 60m 0.32, 0.5 micro

[chromatographer1]

Over a range of temperatures the retention of a peak changes under a constant flow. This has to do with the attraction of the liquid phase for the analyte over a range of temperatures.

Another analyte may have a different retention pattern, perhaps with less variation over the same range of temperatures.

If A is more strongly retained at low temperatures than B but is retained less at higher temperatures than is B , when a range of temperatures is used the average retention may be the same, thus the same retention time. But at a temperature X or Y they may separate, but when programmed between X and Y they may coelute.

Run the analytes at a high temperature and at a low temperature and see what happens.

best wishes,

Rod

[mr3573]

After analyzing same sample by isothermal method in two different temperatures (80 and 110) I noticed two desired peak separated in high temperature and in low temperature only one peak appeared.
Let me more explanation about my question
Suppose we have two components A and B. during isothermal test (for example 110 as above) A and B will be separated at specified retention time (i.e. 7.5) .i will have a defined linear velocity and column flow for A and B at their retention time. Also I try to change method to programming so that A and B appears at same linear velocity and temperature (constant flow method same as isothermal) but different retention time (i.e. 12). I though if I kept same column flow, linear velocity and column temperature, A and B have same behavior and separation. But actually wasn’t that. When retention time changed, separation between A and B will change regardless column flow and temperature. If my conclusion is right, how to predict best condition for separation of A and B

[Don_Hilton]

If you are just trying to separate two or three components, you have a couple of options.
The first is: make injections at a few sets of conditions and settle on a set of conditions that gets the job done. It may not be the exact optimum - but if it does the job, you're there. Best outcome depends on what you are trying to do. For some the question is analysis cycle time, for others it is maximum peak resolution, and for others is would be maximum signal to noise. For many of us, we pick a compromise between these – and perhaps other factors. Depending on your constraints, you can make a few injections changing one parameter and plot the outcome, and then change another and plot the outcome. Or, you can use a simplex technique and change multiple parameters at a time.

If you need a more polished optimization the second option is: software designed to model the physical chemistry going on in the GC column. Years ago I had a copy of DryLab -- and it may still be around for GC. You can save a lot of time - but my recollection was that the software was not cheap.

And, I suppose there is a third option, you can read the papers upon which software like DryLab is based and use it to do the math. I read the papers I could find, thought about doing the math - then bought the software...