yes. You can use LC-MS, provided that impurities peaks are not isomers ie same MW. The strategy that gives the broadest coverage is to use DAD scanning in conjunction with LC-MS and if you suspect the peak is not homogenous, you coud also use an orthogonal separation method eg CE, IC. This approach is generally accepted by regulators.
other detectors like UV detectors, electrochemical detectors and fluorescent detectors are also used to detect the peak purity in chromatogram. the above detectors including PDA are very effective advantages like high sensitivity,and suitability for gradient elution. other detectors like refractive index detectors are also available but these are not sensitive compare to above detectors. the precaution is mobile phase used should be of hplc grade and free from air bubbles, this may give errors in data.
You can vary the mobile phase and column temperature conditions slightly to be sure that a second peak does not resolve from the main peak. You will need to do this anyway if you are evaluating robustness. Also, if you are optimizing the method through a DOE approach you can look for resolution of underlying peaks. This condition modification combined with use of DAD should answer the questions.
LC-MS is the surest way of differentiating co-eluting peaks. The only proviso is you must use volatile buffers. If you are limited to using DAD then I suggest varying the pH which may have some selectivity if their are ionisable groups present.
And in case of doubt, you could put the fraction in a NMR tube. This can also rule out the presence of any compounds that have no chromophore and a low ionization efficiency in the MS.
The old-fashioned way was to use wavelength ratioing. Some of the older UV detectors offered that capability. If your data system allows area slices, you could do it even without such a detector, although it would be a tedious exercise.
Collect the peak fraction and look at it by TLC in a variety of solvent systems. If the compound is stable it will work if not may get breakdown products. You may need to get rid of the solvent in a gentle way to get sensitivity.
In RP-HPLC, the finest way in analyzing peak purity is mass spectrometry. However, if your molecules are not easily ionizable, you may perform your runs with very long gradients and high temperature (50 to 70 °C). In my experience, slow gradients, higher than 50 °C, and using higher theoretical plate number columns as well, constitute a good way to separate overlying chromatographic signals.
Another (old school way) is to measure the melting point of the eluted peak (compound). In this case you can simply collect the peak kof interest, remove the solvent and determine the melting point of the solid residue by a microscope. If the compound is pure the melting interval will be narrow, but if there are some impurities present the melting interval will be somewhat wider.
The peak purity of non volatile sample can be done by RI Detector, Fluorescence Detectors. If you want to know the mass fragmentation of that peak then you have to run the sample on LC-MS-MS (Ion trap system) using either ESI probe or APCI or APPI mode depending on the nature of the sample. If the the sample is volatile then you can use GC with FID, TCD, NPD or ECD Detectors. For mass fragmentation you can use GC-MS either single Qudrupole or iontrap mass system.