- Previous Article
- Next Article
- Table of Contents
Journal of Supercritical Fluids, Vol.24, No.3, 251-263, 2002
The effect of initial drop size on particle size in the supercritical antisolvent precipitation (SAS) technique
The objective of this study was to find the effect of the initial droplet size on the final particle size in the semi-continuous SAS technique. In this technique the solute of interest is first dissolved in a conventional solvent. Particle formation is accomplished by spraying this solution continuously through a nozzle into a chamber containing subcritical or supercritical carbon dioxide. The droplet sizes were calculated from four alternative equations. The theoretical particle sizes were calculated from the calculated droplet sizes assuming that one agglomerated microparticle is produced from each droplet. The theoretical mean particle sizes and the measured particle sizes were compared. The particle formation model was tested with poly(L-lactid acid) (L-PLA) in dichloromethane (DCM) and carbon dioxide. The smallest L-PLA particles were formed by spraying L-PLA/DCM solution into liquid CO, below 304 K and 8.0 MPa. Under these conditions, the mean agglomerate particle diameter was less than 4 pm. When spraying the solution into supercritical CO2 at 308-323 K and 10-14 MPa, the mean particle size was 8-14 pm. Higher temperatures and pressures increased the mean particle size. The particles were strongly agglomerated at the highest temperature and pressure (333 K and 17 MPa). Recrystallization of L-PLA into small particles is possible below 308 K and 10 NlPa using the SAS technique. According to the droplet formation models, the effects of temperature, pressure, CO2 velocity and density, nozzle diameter, interfacial tension between liquid and CO2 and Reynolds and Weber numbers on the droplet size is small. The calculated theoretical particle size agreed reasonably well with the final particle size at the near critical temperature and pressure of CO2. However, the droplet and particle formation models did not explain the observed changes of particle sizes with changes in operating conditions. It is suggested that the initial droplet sizes, formed at the nozzle, do not have an effect on the final particle size. (C) 2002 Elsevier Science B.V. All rights reserved.