is this sepperation accepted ?

[WESAM]

Hi there ,



1-is these three compounds well seperated ( i mean is it acceptable in term of resolution ?. espeacially between 1 & 2 in two cases

a- 72 % me
b- 75% what is the best strength because i want the unknown peak in between go far .

2- how can i calculate the resolution in 75 % between the unknown and number 2 .I know the equation of the resolution but i don't know how do i calcualte w1/2 or wide at a half high because it only written here in the paper ( Area , Heigh , Area % , Height% ).

Any help will be much aprreciated

[tom jupille]

The resolution is not very good, but it may be good enough depending on what you are trying to accomplish (what kind of accuracy and precision do you need?). If:
- this is not a "regulatory" (e.g., pharmaceutical / environmental / food), and
- the peaks in your actual sample will be comparable in size (as they are here), and
- your accuracy requirement is plus/minus a few percent,
then what you are showing here will probably work.

Resolution in your 75% chromatogram is probably best calculated by hand-estimating the baseline widths: print out the chromatogram and draw the best straight-line fits to the front and back slopes of the peaks and measure the separations where they cross the baseline. Remember that resolution is only an approximation in any case.

As far as "optimizing" the separation, merely playing with the % methanol is not going to do a lot for you (that said, something in between what you are showing would probably be the best compromise).

Right now, you are getting about 5,000 plates for the last peak (by my calibrated eyeball!). If you don't have time to experiment, you could always couple two columns in series, which should improve resolution by a factor of about 1.4 . A better approach would be to get a more efficient column with the same packing chemistry (e.g., go from a 5 micron to a 3 micron packing) for the same kind of improvement.

Even better (but more time-consuming) would be to change the selectivity (move the peaks with respect to one another) by changing to a different solvent (e.g. ACN) or a different temperature, or a different pH, or a different stationary phase.

[WESAM]

Tom thanks mate for you .In the above graph 75 % , the concentration was 1 ppm but because i want to at least down to 0.01 ppm , i found that when injecting 0.1 ppm peak number two ( IS ) comes at the sholdure of the unknown peak . 78% doesn't help also ( as you said ) down than 72 will co-eluting .The only choice for me as i don't have any column ( poor laboratory ) and i don't have oven is to try to change PH ? but to what level ? what do thing

[Uwe Neue]

I would first go for a still larger retention, since the separation does not look too bad. Maybe 70% methanol will do.

For a possibly larger change in selectivity, go to acetonitrile as the organic modifier. Somewhere between 60% and 65% is a good start (62% maybe?).

[HW Mueller]

It would be interesting to see what flow rate optimization would bring here.

[danko]

The peak between 1 and 2 (one and a half?) looks bad - suffers from terrible tailing. I'd go for pH or/and ion strength optimization.
It's obvious that longer retention makes things worse. So, less organing sounds like a less rewarding path.

Best Regards

[Uwe Neue]

1. Hans, flow rate manipuation usually does close to nothing in an optimization. It is commonly just a waste of time.
2. Danko, the separation with the larger retention appears to show additional features of the separation not seen before, such as a smalkl peak between the tailing peak and peak 2, and potentially another shoulder on the fornt of peak 2. Admittedly, such features could also be caused by the injection solvent, but it should be followed up.
3. WESAM, what do you do to control the pH? And do your analytes have ionic functional groups? Danko may have a point there...

[WESAM]

Well , i don't control the PH i just prepare it simply as methanol water and the analytes ( 1 = anailine groube ) 2 &3 = carbamate group

Tomorrow i will try 70 % it may help , not sure if decreasing flow rate from 1 to 0.7 would help !

[HW Mueller]

Uwe, exactly because of the "usually" and "commonly" I wondered how it goes here.
"Usually" an aniline tails without pH/ion control, but not here.

Is it known what sort of compound(s) comprises the peak between 1 and 2?

[WESAM]

Oh mate this compound between 1& 2 is nightmare , right now i spend one year with my supervisor and we don't know what is it , any way we assume it is impurity from methanol and 15 different batches give the same peak area .So, no way to get red of it and that's why it causes much trouble to my seperation .

I think i need to control aniline group at PH=9 no sure

[Uwe Neue]

You don't need pH control for the carbamates, but it would be useful for the analine, and especially for the unknown tailing peak between 1 and 2.

For pH control, use a phosphate buffer at pH 3. This may do wonders already.

Don't waste your time with changing the flow rate.

[HW Mueller]

What a waste: Two extra runs to see a possible effect of flow rate.
What are you trying to say, Uwe? That one should just use 1 mL/min for all times and every substance, and never consider that different diffusion properties, and whatever, influences the van Deemter curve?

[tom jupille]

What I think Uwe is saying (or, what I would say if I were in his place ) is that the effect on resolution of changing the flow rate is relatively small compared to the effect of changing selectivity. Small-molecule Van Deemter curves are fairly flat, even for 5-micron particles, and 1 mL/min in a 4.6 mm id column is not *that* far off from the optimum anyway, so that I would expect the improvement in plates to be at most around 20% -- and that translates to less than a 10% improvement in resolution (the square root bites again!).

I agree with Hans that you have to take that generalization in context; for larger molecules, the Van Deemter C-term is higher and flow *is* more important, but in this case, I really doubt that a flow rate change would come anywhere near giving the required improvement in resolution.

[Uwe Neue]

Here is the list of things that change a separation (for ionizable compounds):

1. Drastic change in pH (low to high)
2a. Organic solvent (methanol versue acetonitrile)
2b. Column type (no, not from C18 to C18, but from C18 to Phenyl, or EPG)
3. Temperature
4. Gradient slope
5. Ionic strength
...
...
...
10. Flow rate

[HW Mueller]

Maybe I should have stated the reason for asking about flow rate change. I have the info in the back of my head that I saw some improvement, due to flow rate change, in badly chromatographing matrix elements. Unfortunatly I didn´t document this anywhere, I am not going through thousands of chroms to find it. I don´t even remember whether the broad peaks could have been due to macromolecules.
So I just wonder whether WESAM may see something like that.

Besides, two extra runs doesn´t kill anybody.