Baseline Absorption in Gradient Elution

[TNA]

Dear all,

I've developed some HPLC methods for an OTC drug which contains Acetaminophen and Ibuprofen. I've used gradient elution in its determination since those two components have different polarity which is quite difficult when using an isocratic. The condition is as shown below:

Column: C8 encapped
Mobile Phase: ACN (A) : Monobasic Potassium Phosphate 0.025 M, adjusted with phosporic acid ad pH 3.0 (B)
Gradient program : 0 min: 2% A (hold for 9 min); 10 min: 50% A (hold for 10 min); 21 min: 2% A (hold for 7 min --- reequilibrium)
Flow rate : 1.5 ml / minutes
Wavelength : 250 nm and 223 nm

I've got such a good separation with the condition above but there was a problem with the baseline. When I run baseline, mostly before the analysis was started, there was a ghost peak at the retention time of Ibuprofen. I've changed the composition of mobile phase while it was running at high organic content (after 10 minutes run) as well as filtering the buffer through column before its using but all of them seemed not quite useful. The ghost peak remained to be appear at the same retention time of Ibuprofen.

I need some suggestions and considerations to obtain a better determination method.

Thank you

[danko]

Hi TNA,

In my terminology (but there could be other opinions out there) a ghost peak is a peak that didn’t elute in the previous run/injection (also called late eluter) which elutes in the following run/injection. That kind of peaks, are easily acknowledged, because they are quite broad/flat compared to other peaks at the approximate elution time. Also, this would be quite a coincident if a ghost peak eluted exactly at the analyte’s specific retention time.
So, you need to examine the observed peak with regards to its shape. If it is well-shaped (i.e. slim) and the retention time coincides with the analyte’s retention time, the probability is quite high that you’re experiencing a “carry-overâ€

[TNA]

Dear danko,

Firstly thank you for your kind reply.

I forgot to mention that the retention time of Acetaminophen is 7.8 minutes and Ibuprofen is 18.64 minutes when I run the gradient program above. Therefore I tried to change the gradient composition while it was running at higher organic strength.

The ghost peak looks good and quite slim so as other peaks. I used to run baseline (0 uL of volume injection) before running samples in order to anticipate samples interferences. The ghost peak appeared even at the first injection.

Moreover I've tried to change the buffer with 0.2% trietilamin in water, adjusted with phosporic acid ad pH 3.0. I got the baseline profile wasn't so different and there still was a ghost peak appeared exactly at the retention time of Ibuprofen. Is it make sense if that ghost peak come from phosporic acid I used to adjust? How could I prove it?

What does "carry-over" peak actually mean? And how should I run simple test you've mentioned above?

Thank you

[TNA]

Dear danko,

For addition information, I've already tried to use water instead of phosphate buffer with the same gradient program. The result was the baseline relatively less of peaks and there was no ghost peak at the retention time of analyte. But unfortunately using water instead of phosphate buffer isn't specific for the analytes. That's why I suspect the ghost peak is caused by phosphate.

[danko]

Hi again,

The simplest test is to inject different volumes of the blank (i.e. phosphate buffer).
You didn’t mention what your injection volume was, so I’ll just refer to a randomly chosen figure.
Let’s say you inject 10 μL. Then you could set up the test as follows:
Inject 5 μL phosphate buffer twice. Then 10 μL twice and finally 20 μL twice.
If the replicates are consistent (i.e. approx. the same peak area for both injections at the same level) whilst the peak area increases proportionally to the injection volume, then your phosphate buffer is the source of the peak (i.e. contaminated). If the peak area on the other hand decreases with every consecutive injection, then the peak is caused by “carry-overâ€

[TNA]

Dear all,

I've changed the pH of phosphate buffer into 6.6. Therefore I used monobasic potassium phosphate, adjusted ad pH 6.6 with potassium hydroxide. I used the same gradient program as I mentioned on my first post.

The result: there was a ghost peak, appeared exactly at the retention time of analyte. Then my last assumption that I suspected this ghost peak is caused by phosphoric acid was wrong. The ghost peak still appeared even when I didn't use phosphoric acid in adjusting.

Does anyone have some better ideas or suggestion to solve my problem?

Thank You

[TNA]

Recently I've tried to change gradient slope when I used buffer pH 6.6 as shown below :

0 min : 2% A
10 min : 30% A
20 min : 30% A
21 min : 2% A
28 min : 2% A (re-equilibrium)

But the ghost peak still appeared at the retention time of analyte.

[tom jupille]

It looks like your mobile phase is contaminated with ibuprofen.

You can confirm by the classic test of running three blank runs with increasing equilibration times in between. If the size of the ghost peak increases with increasing equilibration time, then you have contamination in the "A" solvent.