In spectroscopy we have unknown and refrence, if u have some peak in your refrence and u haven't it on your unknown, u have negative peak, u can test it by a sample with color and water as refrence and a sample of water as unknown or any other type.

It simply means that sample is more transmitting than the reference. i.e., reference cuvette is dirty ( e.g., fingerprint spot). If you are measuring absorbance after performing the baseline, add your sample without disturbing the cuvette. Sometimes we tend to clean the cuvette after measuring the baseline. If this is not the reason a spectra will be really helpful here. Another common error could be "autozero" command. If your sample has real absorbance at the point of "autozero" wavelength then the absorbance will go to negative where the real absorbance is low or zero.

Many thanks to all for your comments. To Peter Kapusta, attached is one of the UV-vis spectra we have collected. I can only tell you that the compound is a pyrimidine. As I mention above the spectrum appears to be quite consistent regardless of the instrument or cells we are using. We have not yet measured the compounds emission.

Peter Kapusta, it's not the parent pyrimidine system, it is a pyrimidine derivative, quite a simple one but not known in the literature. We typically fully characterise all compounds we publish and that includes collecting the UV-vis data. The concentration is 10-5 M and its in DCM. The cells are quartz. I cannot give you more details of the chemical structure. We have three pyrimidines out of over 10 analogues which show this behaviour. All were measured using the same equipment, solvents, cells and by the same pair of hands. The run down from 800 nm is just a coincidence, many of our other compounds have low energy absorptions. It's not related however to this problem.

Pete, many thanks for wasting your time, I didn't realise your time was so precious, you must be an extremely busy and important individual. My apologies.

We will look at the compounds fluorescence.

I am really curious to know the reason behind the peculiar shape of the spectrum (with such huge negative absorbance). Please do let us know if you find the answer. Really sorry to see the discussion went wrong. Though fluorescence can contribute to the detected intensity I never had such experience even for highly emissive molecules like quinine sulfate, rhodamine 6G. We use Jasco- V530 spectrometer.

Tuhin Khan, thanks for your e-mail. It is unexpected and while human error is the most probable cause we haven't found out yet what that error could be. I will for sure update you with our findings. It must be something obvious that is right in front of our eyes and we can't see it.

 This phenomenon could be seen in a laser where the intensity of emission exceeds that of absorption ,though I think that is not the case here.

Dear Tuhin, and others. The problem related to the unusual negative absorption has now been resolved. The controls we had carried out insured the problem was not related to instrumentation, the cells or the compound itself. We reevaluated the controls and realised that we had not checked the batch of solvent we were using. It appears that the solvent contained an unknown impurity which led to the effect, not sure how though as it seemed to be rather compound specific, but with fresh DCM the absorption spectrum was rectified. Thanks again to everyone, including Peter, for their comments.

many factor affected the absorption of the sample, one of them the impurities in the solvent or sample you analyzed, to solve this problem check well the solvent purity.