Yes, fluorescein can self-quench due to overlapping excitation and emission spectra. Excitation can be transfered by FRET to another fluorescein molecule. FRET is strongly dependent on the distance, therefore dilution can lead to stronger fluorescence.

Apart from FRET, primary inner filter effect will lead to a similar observation. What is the absorbance at excitation wavelength? Is it above 0.5? High absorbance will lead to decreased fluorescence.

Thanks every one. For my experiment I am using fluorescein free acid as a model drug. I am using 1 mg per ml conc. of florescein to determine the encapsulation efficiency of some nanoparticles. For that, I have to measure the concentration of fluorescein remained in the supernatant after the encapsulation processss (Absorbance and fluorescence). In this point I am facing the above problem. Every time I am getting higher fluorescence and absorbance intensity of supernatant then the initial concentration. Kindly suggest regarding this. Should I dilute each sample before absorbance and fluorescence measurement?

1mg/ml is a very high concentration. In fact it si about 3 mmol/l in concentration, corresponding to an absorbance value of 276. For such a dense solution, all the light gets absorbed long before it reaches the part of the cuvette, where the fluorescence is detected (at least for a usual 90° detection setup in a 10mm fluorescence cuvette). All the light reaching the detector is result of a series of re-absorption and re-emission processes.

You need to dilute by about 300x - 1000x to get into the area of linear relation relationships...

Thank you so much for the valuable information.