(279h) Prediction of Drug Solubility in Supercritical Carbon Dioxide Using Equation of State Based On Hole Theory With Molecular Information
AIChE Annual Meeting
2013
2013 AIChE Annual Meeting
Engineering Sciences and Fundamentals
Thermophysical Properties and Phase Behavior IV
Tuesday, November 5, 2013 - 10:15am to 10:30am
There are great interests in preparation of nanometer size materials because of their unique properties in the filed of pharmaceutical, coating, environmental, chemical processing, electronic, and sensing applications. In the pharmaceutical industry, their particle sizes are important for the bioavailability. The bioavailability can be improved by the reduction of the drug particle size. The development of methods for the preparation of nanoparticles for drug has received a great deal of attention. Various crystallization techniques using supercritical fluids have recently been proposed as preparation method of drug nanoparticles. Rapid expansion from supercritical solution method (RESS) is the most well-known crystallization technique using supercritical fluid. We proposed the strategies for nanoparticle design of drugs using RESS technique in our previous work. The mean sizes of the particles micronized by RESS process can be correlated with the supersaturation defined as the difference between the solubilities in supercritical carbon dioxide and the sublimation pressure on the particle collection part. To develop and design the RESS process, a knowledge of drug solubility in supercritical carbon dioxide is necessary and important. In this work, the solubility of several pharmaceutical compounds in supercritical carbon dioxide were predicted using an equation of state. The coordination numbers around the molecules in mixture were estimated from the molecular shape factor which is given by the molecular volume and surface area calculated an quantum calculation based of conductor-like screening model. The prediction performances were evaluated by the comparison with the experimental data It is found that the prediction model proposed in this work is able to represent the experimental data within the same order of the magnitude of the solubility without data fitting.