Thanks for the reply.

The matrix is not soluble in water. However to determine the EE% the matrix loaded drug was dispersed in a solvent in which the drug was soluble. The drug is soluble in ethanol,acetonitrile and DMSO. After sonication and mechanical stirring overnight, high speed centrifugation is done to separate out the insoluble matrix, and the supernatant was used for analysis using UV-VIS spectrophotometer. After centrifugation the inner part of the pellet appears to take color of the drug even after repeated wash. Therefore, feel that 100 % of the drug is not coming out of the matrix..

Color is an indication of the presence of drug in the medium unless it is observed as a reflection at the interphase. Is drug fluorescent? You may have some of the color or, strong color illusion for minute concentartion of the drug. Make sure drug is ALL extracted. If drug is coming in the organic phase, try enriching the extraction by saturating the inorganic medium. A solo medium extraction may work but not probably to the fullest to extract all. The partitioning phase may be needed to pick all drug molecules - which depends on the drug's structure, the conditions and the medium/phases you are using.

Thanks for the reply.

The material is fluorescent, and it is very well known natural product curcumin. I have tried to extract it from ceramic matrix by ethanol and acetonitrile.

Thanks for confirming! There are a lot of questions on curcumin color problem. Check on the questions-answers from David E. Bautista-Erazo profile on the RG. Will take up from there! It would have been helpful if you had disclosed the full info in the beginning!

Thanks for the reply. I thought the problem is with the experiment and not concerned with the nature of the compound.

Sorry for that.

Did you check the questions-answers from David E. Bautista-Erazo profile on the RG about the Curcumin?

 I checked the discussions from David E. Bautista-Erazo profile on RG. I think it is not relevant to my question and doesn't address any solution of my problem.

The problem is in the matrix your drug is in! The fluorescence part of the drug because of its presence may be a small contributor of the ratio of the drug not recovered back which shows the color. Just check what percentage of the drug you are recovering from the matrix on a model test system in a simulated condition. Is the matrix a solo system or a mix of materials, then use separate matrix for simulated experiment and run a calibrated & standardized GC or HPLC for confirmations. This could also give a clue to your EE query. See also for porosity based retention of your material!

The matrix is a solo system and the affinity of curcumin towards the solvent is much much more than towards the matrix.

Please tell how to calculate the percentage of the drug recovery from the matrix on a model test system in a simulated condition? 

Start with known quantity-concentration of the drug and similar ratioed quantity-similar ratioed weight of the matrix (material need to be of same - structure and density also). If you used fresh curcumin - use fresh here too. Treat both -drug and matrix in the way it simulates or nearly simulates (especially pH care needed, solvents acidity-basicity at trace levels, temperature, shaking intensity and time etc.) your original experimental conditions. Run quantitative analysis for the simulated experiment recovered drug and compare to see how much drug is returned after retaining by the matrix. Both UV-Vis (not just UV) and HPLC can be employed to see concentration.