A screening design of experiments has been applied to supercritical antisolvent precipitation of amoxicillin (AMC) using carbon dioxide (CO2) and N-methylpyrrolidone (NMP) as antisolvent and solvent, respectively. The design of experiment (DOE) proposed is useful for identifying the key factors involved in the SAS process in just a few runs at an early stage of experimentation. Seven factors have been studied, and two levels were assigned to each. A fractional factorial design with 2(7-4) experiments has been used. Mean particle size (PS) and particle size distribution (PSD) of the processed amoxicillin were chosen as responses to evaluate the process performance. Within the range of the operating conditions investigated, concentration and liquid flow rate proved to be the key factors having most effect on both PS and PSD and thus, the most important factors for controlling the formation of sub-micro particles of amoxicillin by the SAS technique. Moreover, the success of the precipitation process was influenced by the location of the operating conditions within the P-x-y equilibrium diagram of the CO2-NMP system. (c) 2007 Elsevier B.V. All rights reserved.