decreasing retention time problems

[kyle432]

hi all

currently i am struggling with some hplc chromotography showing a decreasing retention time on a uplc. the main peaks retention time is normally 18 minutes and decreases to 12 minutes during a long period run. this shouldn't be happening. however during a new test i did i used the same buffer that is normally used except this time using two lines from two different bottles. this presents new problems as i expected the retention time decrease from thye first line, however what i did not expect was the second line to then take the retention time back to how it had originally started.

the gpv is brand new and so is the pump so these are not the issue. the column is also not the problem.

does anyone have any ideas as to what is happening at all?

thanks

[kyle432]

also i have been on the web looking for solutions and all have no relevance that i have found.

Decreasing Retention Times
Active sites on column packing - Use mobil-phase modifier, competing base (basic compounds), or increase buffer strength; use higher coverage column packing.
Column overloaded with sample - Decrease sample amount or use larger-diameter column.
Increasing flow rate - Check and reset pump flow rate.
Loss of bonded stationary phase or base silica - Use mobile-phase pH between pH 2 and pH 8
Varying column temperature - Thermostat or insulate column; ensure laboratory temperature is constant


all stated above are not applicable to my problem as i have checked these and none are an issue

[Gerhard Kratz]

Did you check pH value of your buffer? For me it Looks like in line 1 and 2 a different buffer pH!

[kyle432]

hi gerhard

yes we have a specific method to keep to which is recorded with a 0.1­­­ plus or minus boundary so the pH isnt the problem. its very strange in my experience iv never seen this before. all guidelines have been met so it is definitely not that sort of issue which makes me believe it is our system, however its only been used for two months now and all parts are fine. flow rate is stable temperature stable its just odd.

[Peter Apps]

Can you exclude the possibility that something is evaporating from the mobile phase reservoir while it is in use ? - for example, do you use continuous helium purging to degas the mobile phase ?

Peter

[kyle432]

hi peter

thanks for the ideas.

the bottles being used are kept closed. my supervisor has informed me that the method can exclude evaporation and that they have tried the purging you described in the past and this made no difference to this method run.

any more ideas please pass them on because we've hit a wall on what this could be.

thanks

[Tomtotom]

Perhaps it is a stupid thought, but your bottle may be shut to tight. So the pressure decreases in the bottle and the solvent is held back. It only makes sense if you are running a gradient.

[kyle432]

hi tom

our bottle toppers have holes within, not big holes though. however it may be worth trying but my supervisor has discounted this as a possibility.

again any new ideas will be a big help.


thanks

[Gerhard Kratz]

You mentioned that your System is new. Anyhow, please check the magnitic valves on the low pressure side if flow of each chanel/line is correct.

[aceto_81]

Maybe phase dewetting/collapse?

Any chance that you can provide some more details on the method used (eg column, mobile phase, temperature, sample solvent, ....)?

Ace

[HPLCaddict]

As already asked, some more details regarding the method would be fine.
Is it isocratic or gradient? If isocratic, premixed or dial-a-mixed by the machine?
If gradient, what's the gradient programme? Flow-rate? Column dimensions?
Was the retention time change abrupt or developing slowly over the course of the sequence? Are we talking about reversed-phase or some other mode of chromatography?

Without more details, only guessing could be done here...

If I interpret correctly, you're running an isocratic method. Using the very same (!) eluent on line 1 gives you different retention times as on line 2, correct? Sounds like system failure...

[kyle432]

hi all

sorry about this.

its a gradient programme with two primary buffers at a 1mil/min flow rate with 40oC temp. the change is developing slowly over the sequence. a full sequence can take nearly three days requiring around 4 litres of solvent A and 800 ml of the B.

my investigation i described earlier was still using buffer A except i seperated the 5L buffer into two bottles of smaller size and swapped the lines between sample brackets. this in turn would show retention times starting from the point required however still decreasing over time.

thanks

[tom jupille]

There *is* a way to determine if column deterioration is due to the mobile phase or to the sample, but it's time-consuming to do.

What it involves is running a series of samples over a long enough period to measure the change, then running the same number of blanks, then running the series of samples again. You then plot the performance (in this case, your retention time) as a function of total elapsed time, and again as a function of sample number (ignoring the blanks).

If the problem is due to the mobile phase, then the plot vs. elapsed time will be fairly linear, but the plot vs. injection number will show a "cliff edge" (because the deterioration continued even though you weren't injecting samples).

If the problem is due to the sample, then the plot vs. injection number will be fairly linear, but the plot vs. elapsed time will show a plateau corresponding to the period when samples were *not* being injected.

There is a bit more detail on our web site:
http://www.lcresources.com/resources/TSWiz/hs390.htm

[aceto_81]

No organic, just buffers?

[kyle432]

thanks for that tom ill try this and see what happens.

and aceto its just buffers no organics involved.

thanks again for all the help