J. of Supercritical Fluids 51 (2009) 50–56 Contents lists available at ScienceDirect The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu The supercritical micronization of solid dispersions by Particles from Gas Saturated Solutions using experimental design Michael Brion a,∗ , Séverine Jaspart a , Leonardo Perrone b , Géraldine Piel a , Brigitte Evrard a a Laboratory of Pharmaceutical Technology, Pharmacy Department, CIRM, University of Liège, 4000 Liège, Belgium b Laboratory of Pharmaceutical Technology, Pharmacy Department, University of Camerino, 62032 Camerino (MC), Italy article info Article history: Received 17 April 2009 Received in revised form 29 June 2009 Accepted 30 June 2009 Keywords: Experimental design Plackett–Burman Central composite design Supercritical carbon dioxide Solid dispersion Particles from Gas Saturated Solutions abstract The Particle from Gas Saturated Solutions process was successfully used to micronize solid dispersions containing hydrophilic carriers and a new chemical entity, YNS3107. By means of experimental design, the effects of several experimental parameters on micronization were investigated. Within the chosen exper- imental conditions, the results showed that the autoclave temperature, autoclave pressure, drug loading, flow rate of carbon dioxide and air pressure were significant parameters. During the optimization step, the most relevant parameters of the screening were optimized using a central composite design meanwhile other factors were kept constant. Optimal conditions were used to produce microparticles with a volume weighted mean diameter of 30.4 m. The closeness between the measured and predicted response, eval- uated at 28.1 m, demonstrated the validity of the statistical analyses. Finally, an enhancement of the rate of dissolution of YNS3107 in the solid dispersion microparticles was measured using USP II dissolution test apparatus. © 2009 Elsevier B.V. All rights reserved. 1. Introduction In most cases, oral administration of new chemical entities (NCE) results in poor bioavailability because of their poor water-solubility and low dissolution rate. Solid dispersions have been widely used to improve these specific properties [1–3]. Generally, a solid dis- persion (SD) is a dispersion of one or more hydrophobic drug(s) in a polymeric hydrophilic matrix at a solid state. These drugs can be dispersed molecularly, in clusters or in a combination of both. Such dosage forms are conventionally prepared by the fusion, solvent or fusion–solvent method. Supercritical micronization processes, such as rapid expansion of a supercritical solution [4,5] (RESS), supercritical anti-solvent [6,7] (SAS) or Particles from Gas Saturated Solutions [8–14] (PGSS TM ), have gained increasing attention and may be considered as interesting alternatives for the micronization of solid dispersions. The Particle from Gas Saturated Solutions process, in particular, exhibits some important advantages such as the fact that drug(s) and carrier(s) do not require a high solubility in supercritical fluid (SCF), the fact that it is a one-step process and the fact that it provides the possibility of avoiding the use of organic solvents. However, the SCF, generally supercritical carbon dioxide (scCO 2 ), ∗ Corresponding author at: Avenue de l’hopital, 1, B36, Tour 4, +2, 4000 Liège, Belgium. Tel.: +32 4 366 4306; fax: +32 4 366 4301. E-mail address: mbrion@ulg.ac.be (M. Brion). must be soluble [15] in the solid matrix in order to obtain a molten system for the micronization. The present study intends to investigate, by using design of experiments, the effects of several experimental parameters on the supercritical micronization of solid dispersions containing polymeric excipients (PEG4000, PEG400 and poloxamer 407) and YNS3107. During the optimization, the appropriate mathematical model describing the Particle from Gas Saturated Solutions process was elaborated and used in order to find the optimal conditions of micronization producing the smallest SD microparticles. Finally, the dissolution profiles of the produced microparticles, the correspond- ing physical mixture and pure active ingredient were compared. 2. Materials and methods 2.1. Materials CO 2 (99.998%) was supplied by Air Liquide (Liège, Belgium) and YNS3107 by UCB Pharma S.A. (Braine-l’Alleud, Belgium). Methanol (LiChrosolv ® ), acetonitrile (LiChrosolv ® ), polyethylene glycol M w ∼ 4000 (PEG4000), potassium dihydrogen phosphate (pro-analysis quality), dipotassium hydrogen phosphate (pro- analysis quality) and polysorbate 80 were purchased from Merck (Darmstadt, Germany). Polyethylene glycol M w ∼ 400 (PEG400), sodium hydroxide pellets (Ph. Eur.) and poloxamer 407 were pro- vided by Fagron (Waregem, Belgium), VWR (Leuven, Belgium) and BASF (Burgbernheim, Germany), respectively. All products were 0896-8446/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.supflu.2009.06.021