Peculiar Peaks in HPLC Blank runs_multiple products

[Pharm_hplc]

Hello Everyone.

I am from a pharma industry, and am developing methods for various products.
To give you a little brief, the methods are being developed in a R&D block and validated in a qc block. Both the blocks are in the same premises, but different buildings at a distance of about 100 meters from each other. R&D Block as Split AC installed, QC has AHUs installed with filtered air. (AHU of QC and microbiology lab is common)

We have developed methods for a product and all of the chromatographs (Raw material Assay & RS, Finished Product Assay and RS) were smooth and with the peaks at the right RTs. We transferred the method to the QC lab, but have been observing really peculiar behaviour in the BLANK runs in all HPLCs.

There are unknown peaks in blank runs in all HPLCs ( there are about 4 HPLC) at 5 and 26 minute and the peak area increases with every run. There are no peaks initially for 4-5 runs but start appearing from the 6th run and keep increasing. We tried changing the mobile phase A, B and even the diluent but the peak still remains. We changed the column to a one with higher pH spec, but the peak still remains though the RT shifts a little to 7 and 30 minutes. We changed the vial and mobile phase bottles. The peculiarity is that there is no peak in R&D, but two ghost peaks are always there in all runs, and they keep increasing.

Mobile phase is acetonitrile based. Diluent is ACN + Naoh.

As the peak increases over time, we thought it to be a leeching issue, so have tried changing everything but nothing seems to be working. Any support would be greatly appreciated

[Consumer Products Guy]

I'd extend the run time a bunch if isocratic and see if more peaks come out.

If gradient, I'd increase the time at high organic, and ALSO lengthen the equilibration time before next injection.

[Pharm_hplc]

I'd extend the run time a bunch if isocratic and see if more peaks come out.

If gradient, I'd increase the time at high organic, and ALSO lengthen the equilibration time before next injection.

Hello,

It is a gradient method.

If i increase the organic ratio or time the peak of interest will also be shifted to the left.

Another observation is that this peak of 5 minute is now appearing in blanks of ALL the products that are being run now. Please note we have tried HPLC of different makes also (Waters, Shimadzu) but the peak remains and increases on every run.

Can it be due to environment?

[scottythree]

Assuming water is used what kind of water is used in your mobile phase. Is there differences between what was used in the R&D vs. QC? If there is try making some with the R&D water and running it in the QC lab.

[Pharm_hplc]

Hi,

We have found the reason, it was the vial septa. We soaked the septa in the diluent for 12 hours and then we ran the sequence.

We got the peaks exactly at the RTs of the unknown peaks. Confirmed that it was the septa. We were using waters vials, and on using shimadzu vials the chromatograph was very smooth.

[Pharm_hplc]

Hi,

We have found the reason, it was the vial septa. We soaked the septa in the diluent for 12 hours and then we ran the sequence.

We got the peaks exactly at the RTs of the unknown peaks. Confirmed that it was the septa. We were using waters vials, and on using shimadzu vials the chromatograph was very smooth.

Problem again. The shimadzu vials gave no peaks for 24 hours, but have started giving out peaks, although small, after 24 hours.