Dionex Ultimate 3000 HPLC enquiry: Low peak area

[aduragbemi]

I am experiencing low peak areas for the samples I run on the Dionex Ultimate 3000 HPLC machine.

To troubleshoot, I ran my samples for quantifying ascorbic acids and carotenoids using the HPLC (Agilent) of a different lab and compared the final concentration with the result I got using our Dionex ultimate 3000 HPLC, and the difference in final concentration was huge.

I further try to troubleshoot the problem myself by doing the following:
Sample verification, column verification, method parameter verification, and checks for leakages. In essence, I did all as advised in the link by Thermofisher ( https://knowledge1.thermofisher.com/Chromatography/Liquid_Chromatography/LC_Knowledge/Low_Peak_Areas) except number 9. Also, the UV lamp looks good but can't say much about the flow cell.

I would need suggestions on how to tackle this.

Thanks,
Amo Aduragbemi

[TylerSmith123]

Hello Amo,

What are you using to quantify final concentration? If you are using the other lab's values then you are certainly not going to get the same results. Each instrument is different, and the exact same results (especially something like area) can vary widely due to a variety of different parameters. That is why you need to develop the standard curve yourself for your parameters/instrument.

Unless I am mistaken and this issue is a recent/sudden manifestation that deviates from expected results.

TS

[aduragbemi]

Hi TylerSmith,

Thanks so much for your response.

To explain better, I verified the final concentration of my sample for ascorbic acid and made a calibration curve using another HPLC.

After obtaining the result, I ran the same sample. I made another calibration curve using our lab Dionex Ultimate 3000 HPLC to get the final concentration of my samples for ascorbic acid.

Then, I compared the two results from the two HPLC, I observed a huge difference in the final concentration of ascorbate acid of more than 100 µg/g FW.

Ideally, the difference shouldn't be significant.

I bought a new column and also checked that there are no air bubbles, the autosampler seems to be working fine, optimized my method and the UV lamp also has operating hours of less than 2000. What I am not sure about is the flow cell.

So I am thinking of washing the flow cell with nitric oxide.

What would you suggest in my case?

I need to get the machine working asap but with limited experience, since I just started learning chromatography more than 6 months ago.

The information for my ascorbic acid analysis is as follows:

the mobile phase used are 0.03M orthophosphoric acid (A), and methanol (B). The final gradient used is 100% (A) at 0min; 90% (A) at 3min; 10% (A) at 7min; 100% (A) at 8min; 100% (A) at 9min with a flow rate of 0.5ml/min.

[TylerSmith123]

I wouldn't go as far as washing the flow-cell just yet. Especially if I am understanding your issue correctly. So, just to reiterate, you have run two different sets of standards, one on the "good" LC, and then one on the Dionex 3000. These standards should be the exact same and are used to make the calibration curve for both instruments which sounds great. Then, you're injecting your unknown concentration ascorbic acid sample and this is where the problem lies. On the good instrument, you receive a normal concentration but on the Dionex you do not. To me, if your standards are working successfully on both instruments, I would take a look at the sample prep of the unknown. What is the column you're running these samples on, and are you diluting them in mobile phase or something else?
I assume all samples are injected through the autosampler, standards included, have you attempted to try manual injection?
Another potential issue may be the autosampler itself, or maybe the vials. Is there enough volume for the autosampler to suck up and inject? I believe many autosamplers require an extra amount of volume of sample per injection and this may be impacting your results?
I completely agree with you the results should be the same as the calibration curves are made for each respective instrument and should be accurately predictive if everything went to plan correctly. I hope someone else chimes in as well with possible solutions-- but good luck in the meantime!

[aduragbemi]

Hi TylerSmith,

I am grateful for your input so far.

The standard dilution was made with the re-agent I used for extraction which was 3% metaphosphoric acid. I also ensured that the columns I used in both equipment were of the same dimension 250x4.6 pore size 5nm.

Yes, I injected them all through the autosampler and I haven't tried manual injection yet.

In terms of the volume, I will check that and rerun but I believe the volume of 1ml is great enough.

Thanks for your intake so far.

I also appreciate it if anyone could chime their input also because I have tried all I know and it's getting frustrating by the day.

Thanks