Fractal analysis of SEM images and mercury intrusion porosimetry data for the microstructural characterization of microcrystalline cellulose-based pellets A. Go ´ mez-Carracedo, C. Alvarez-Lorenzo, R. Coca, R. Martı ´nez-Pacheco, A. Concheiro, J.L. Go ´ mez-Amoza * Departamento de Farmacia y Tecnologı ´a Farmace ´utica, Universidad de Santiago de Compostela, Santiago de Compostela 15782, Spain Received 7 May 2008; received in revised form 2 September 2008; accepted 4 September 2008 Available online 17 October 2008 Abstract The microstructure of theophylline pellets prepared from microcrystalline cellulose, carbopol and dicalcium phosphate dihydrate, according to a mixture design, was characterized using textural analysis of gray-level scanning electron microscopy (SEM) images and thermodynamic analysis of the cumulative pore volume distribution obtained by mercury intrusion porosimetry. Surface roughness evaluated in terms of gray-level non-uniformity and fractal dimension of pellet surface depended on agglomeration phenomena during extrusion/spheronization. Pores at the surface, mainly 1–15 lm in diameter, determined both the mechanism and the rate of theophylline release, and a strong negative correlation between the fractal geometry and the b parameter of the Weibull function was found for pellets containing >60% carbopol. Theophylline mean dissolution time from these pellets was about two to four times greater. Textural analysis of SEM micrographs and fractal analysis of mercury intrusion data are complementary techniques that enable complete characterization of multiparticulate drug dosage forms. Ó 2008 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Extrusion; Image analysis; SEM; Texture; Microstructure 1. Introduction The surface texture and the inner structure of the pellets strongly depend on drug proportion [1], composition and volume of the wetting liquid [1,2–4], additives [5,6] and spheronization and drying conditions [1,7–11], and criti- cally determine the relevant properties such as friability, flowability, wettability, adhesion to various substrates and drug delivery behavior [1,12–15]. Several methods are suitable for quantifying the surface roughness of solid materials. Contact or stylus profilometry measures the ver- tical movements of a sharp tip mounted on a cantilever resulting from the irregularities of the surface [16]. The main drawbacks of this approach are related to the dam- age/deformation of the surface of soft materials caused by the movement of the tip [17]. If the material has ade- quate light reflection properties, these problems can be overcome using non-contact or laser profilometry [12]. Image analysis of scanning electron microscopy (SEM) micrographs is also used to extract quantitative informa- tion of fractal geometry and surface texture [18,19]. Fractal analysis was originally developed as a quantitative tool for measuring the complexity of structures that show a pattern of self-similarity (i.e., they look roughly the same at any scale) [20]. Mathematically, the self-similarity implies a power relationship between size and the measurement scale, which enables the application of the fractal geometry principles to the characterization of widely diverse materi- als [21,22]. Image texture informs about the structural order of the surfaces as a function of the intensity, position 1359-6454/$34.00 Ó 2008 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actamat.2008.09.009 * Corresponding author. Tel.: +34 981563100x14883; fax: +34 981547148. E-mail address: joseluis.gomez.amoza@usc.es (J.L. Go ´ mez-Amoza). www.elsevier.com/locate/actamat Available online at www.sciencedirect.com Acta Materialia 57 (2009) 295–303