The DAD is a Diode Array Detector, which is a type of UV-Visible detector for HPLC. So, it’s not exactly clear what you’re comparing it to or based on what measure.
However, I will say that the best way is usually to compare the results obtained from running the same calibrated methods on both instruments for the same amounts of the compound(s) of interest (which would have to contain a chromophore in order to be detectable by this technique). In your case, the caffeine concentrations of standards should be the same between the two instruments. If you can provide more information about what instruments you’re looking to compare, we might be able to offer more specific advice.
Thank you for the reply!
Well, the problem is that I just know my instrument (HPLC-DAD Agilent mod.1260), I don't know the instrument used by the external laboratory. I have a sample (food supplement in powder) that contains a known title of caffeine (0.32%) and I've already done the analysis. I'd like to confirm my results with the ones of another lab, but as you said, it's better to compare the same type of instrument..
Hello i am not a chemist but a software developer, from my understanding:
UV-Detectors (like a diode array detector) or a VWD (variable wavelength detector), typically output an "absorbance" signal which is a physical property and should not be detector dependent ( see: Absorbance - Wikipedia ). But their are many more variables to this, like wavelength bandwidth and reference wavelength / bandwidth.. Some detectors have a fixed bandwidth with the DAD you can choose it, but the results might still be a little bit different.
To measure concentration you would typically compare the absorbance signal of a standard sample with a know caffeine concentration to your "unknown" sample. Using a wavelength where caffeine absorbs the most light ~270nm.
In the end you have to develop a method which works best for your specific type of sample which only measures caffeine and is not influenced by any other compounds in your sample or external parameters.