Management would like a lot of things, but we are scientists and have practical and physical limits governed by scientific laws.
In order to use UV (DAD) to determine the purity of a sample you would need to know the identity and characteristics (Ext coef) of every compound in a sample. This is true of most detectors. There is no true universal detector for HPLC. Alone, MS is no better in this regard as different compounds ionize to different degrees depending on many parameters. Anyone that thinks that ELSD is a universal detector has not actually used one (calibrate different compounds in the same sample with one some day over more than one order of magnitude and see what you get) before or is reading the marketing/advertising literature put out by their distributors or factory. ELSD's respond differently to different compounds based on many things (solvent flow, particle size, atomization, drift tube temperature, stability of the sample...). If you want to know the analytical purity of a sample, then you should use different types of systems to arrive at a consensus. For example, in our laboratory, we use HPLC (UV), LC-MS and NMR all together to get a qualitative idea of overall purity. Even this is not perfect as their are compounds that will just not get picked up sometimes using all three of these techniques. Just because you do not detect something does not mean that it is not there. However, to be reasonable and not spend five years on each sample we compromise and use these three techniques together to develop a purity value. You must select techniques that are appropriate for your types of samples (pharma drugs in our case).
Now, many of the HPLC manufacturer's provide a DAD Purity software application. This application has many settings and parameters that must be tuned to each sample/run. With these systems you can achieve any "Purity" result you like by manipulating the parameters. In all cases, the system assumes that all peaks/compounds have the same coextinction eff and absorb the same at the same concentration. *If you perform a multi-point calibration for different KNOWN compounds, then they can be accurate in determining overall 'UV' purity, BUT again, only for known, calibrated samples. Beware, this is another technique that can get you into serious trouble if you do not understand the science behind it AND how to correctly set up the software parameters (not to mention have obtained an excellent, clean chromatogram under correct conditions BEFORE you use the program). ** I have seen many scientists over the years mis-use these programs in major Pharma companies. Please get help if you do not understand them.
