Encapsulation efficiency of solid lipid hybrid particles prepared using the PGSS® technique and loaded with different polarity active agents

J. of Supercritical Fluids 54 (2010) 342–347

Contents lists available at ScienceDirect

The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu

Encapsulation efficiency of solid lipid hybrid particles prepared using the PGSS® technique and loaded with different polarity active agents C.A. García-González a,∗ , A. Argemí b , A.R. Sampaio de Sousa c , C.M.M. Duarte c,d , J. Saurina b , C. Domingo a,∗ a

Instituto de Ciencia de Materiales de Barcelona (CSIC), Campus de la UAB s/n, E-08193 Bellaterra, Spain Department of Analytical Chemistry, University of Barcelona, Diagonal 647, E-08028 Barcelona, Spain Instituto de Biologia Experimental e Tecnológica (IBET), Aptd. 12, 2781-901 Oeiras, Portugal d Instituto de Tecnologia Química e Biológica (ITQB), Av. República (EAN), 2780-157 Oeiras, Portugal b c

a r t i c l e

i n f o

Article history: Received 23 December 2009 Received in revised form 4 May 2010 Accepted 15 May 2010 Keywords: Solid lipid particles PGSS® process Encapsulation Hybrid materials Dissolution tests

a b s t r a c t The manufacture of particulate hybrid carriers containing a glyceryl monostearate (Lumulse® GMS-K), a waxy triglyceride (Cutina® HR), silanized TiO2 and different active agents (caffeine, glutathione or ketoprofen) was investigated with the aim of producing controlled drug delivery systems based on solid lipid particles. Particles were obtained using the supercritical PGSS® (particles from gas saturated solutions) technique. Experiments were performed at 13 MPa and 345 K, according to previous measurements of lipid melting points. Solid lipid particles were loaded with silanized TiO2 and caffeine, glutathione or ketoprofen in percentages of 6–7 wt% for the mineral filler and 4.2, 5.6 and 16.1 wt% for the respective drugs. The particles obtained were analyzed in the solid state by thermogravimetric and X-ray diffraction analysis and scanning electron microscopy. Drug contents in the precipitated lipid samples and their elution profiles were studied by HPLC. Hydrophobic drugs, such as ketoprofen, were more efficiently encapsulated in the lipophilic lipidic matrix than hydrophilic drugs, such as caffeine and glutathione. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Pharmaceutical and cosmetic industries are working on the development of new strategies for the formulation and processing of matrixes containing active compounds for controlled release. Particulate and colloidal carrier systems have attracted growing interest concerning drug delivery. Based on the carrier material, the conventional vehicles can generally be divided into polymeric and lipidic systems. To avoid potential toxicological problems associated with the degradation of synthetic and semi-synthetic polymeric systems, a great deal of interest is currently being focused on lipid-based carrier systems, inter-alia liposomes and lipid oil-in-water emulsions [1,2]. These vehicles are composed of physiological lipids, such as phospholipids, cholesterol or triglycerides, and, thus, their toxicological risk is null. However, the storage stability of liposomes is limited and the large scale production and sterilization of these carriers is complicated. Many drawbacks associated with conventional liquid-like lipid drug carriers can be overcome using solid lipid particles (SLPs) that combine the superior biodegradability and

∗ Corresponding authors. Tel.: +34 935801853; fax: +34 935805729. E-mail addresses: [email protected] (C.A. García-González), [email protected] (C. Domingo). 0896-8446/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.supflu.2010.05.011

biocompatibility as well as ease of manufacture of lipids with the advantages of the solid-like state. In SLPs, the solid matrix provides enhanced physical and chemical stability, facilitates surface modification for targeting, and drug release is controlled by degradation of matrix constituents rather than by diffusion [3–8]. Conventional methods for the production of SLPs, such as solvent-emulsification-evaporation, double emulsion, ultrasonication and spray chilling or drying, are either multi-step processes or operate with organic solvents [1,9]. Conversely, technology based on the use of supercritical carbon dioxide (scCO2 ) has been described as a one-pot strategy capable of encapsulating drugs into organic solvent-free lipid particles [10–21]. In this work, hybrid SLPs composed of a lipid matrix entrapping mineral filler nanoparticles and loaded with an active compound were obtained using the particles from gas saturated solutions (PGSS® ) supercritical method [22]. The technique consists of dissolving scCO2 in the bulk of a melted lipid mixture with dispersed titanium dioxide (TiO2 ) nanoparticles and dissolved/dispersed drugs, and the subsequent quick expansion through a nozzle, causing the atomization of the melt, the complete evaporation of the gas and the precipitation of the SLPs. A mixture of two C18 triglycerides (Lumulse® GMS-K and Cutina® HR) was used as a matrix. Both lipids have applications in cosmetic and pharmaceutical industries [23,24]. The mixture

C.A. García-González et al. / J. of Supercritical Fluids 54 (2010) 342–347

343

Fig. 1. Process flow diagram of the equipment used for PGSS® SLPs formation. T: CO2 reservoir, EX1: CO2 cooling unit, P1: pump, V1–V5: valves, Ve1: high pressure mixing chamber, Ve2: collector vessel, T1: heater, N1: nozzle.

of two lipids was preferred as a matrix to individual compounds, since together they form crystals with many imperfections offering space to accommodate the additives [13,17,25]. Supercritically silanized nanoparticulate titanium dioxide (TiO2 ) [26–29] was incorporated to the mixture as an anticaking additive. This material is also an effective UV-blocker [13,30,31]. The hybrid SLPs were loaded with three different active agents: caffeine, glutathione and ketoprofen. Caffeine and glutathione are hydrophilic active compounds with photoaging prevention capacity [32,33]. Ketoprofen is a lipophilic non-steroidal antirheumatic drug available in the form of coated tablets for oral administration and gel for topical application. To counteract the short elimination half-life of ketoprofen, encapsulation of the drug in lipidic formulations is the method usually chosen for sustained delivery [34–36]. These three active agents were chosen as model compounds for encapsulation in the lipidic matrix because of their different lipophilic behaviours. Solid state characterization of the prepared samples together with preliminary drug dissolution tests in water were used to analyze the encapsulation ability of the chosen lipidic mixture.

2. Experimental 2.1. Materials Lumulse® GMS-K (GMS, glyceryl monostearate) and Cutina® HR (HCO, hydrogenated castor oil) were kindly provided by Lambent Technologies and José M. Vaz Pereira S.A., respectively. TiO2 nanometric particles (∼20 nm in diameter) were supplied by Degussa (TiO2 P25) and silanized with octyltriethoxysilane following a scCO2 procedure published elsewhere [37]. Ketoprofen (Kt), caffeine (Cf) and glutathione (Gl) were purchased from Sigma–Aldrich. Carbon dioxide (CO2 , 99.998 mol% purity) was supplied by Air Liquide. Chemicals utilized for high pressure liquid chromatography (HPLC) characterization were sodium hydroxide, phosphoric acid (85%), ammoniumformate and formic acid, all of them from Sigma–Aldrich (analytical reagents). Ultrapure water (Millipore, Milford), methanol and ethanol (both from Merck, HPLC grade) were used for the preparation of mobile phases and standard solutions.

2.2. Equipment and procedure The process flow diagram of the PGSS® equipment used to produce the SLPs is shown in Fig. 1. CO2 was fed by a high-pressure piston pump (P1, Haskel model MCPV-71) to a 0.5 L high-pressure stirred vessel (Ve1, Parr Instruments) until the desired working pressure was reached. The autoclave contained a mixture constituted by GMS:HCO in a ratio 50:50 wt%, 5 wt% of TiO2 nanoparticles and different amounts (from 9 to 17 wt%) of active compound. In some of the performed experiments, a certain amount of water was also added to the vessel. The autoclave was heated using a thin band heater (T1, Watlow STB3J2J1). After 1 h of stirring, necessary for mixture equilibration, the system was depressurized by opening valve V3 (Parker 4M4Z-B2LJ) and atomized through a 600 ␮m cone nozzle (N1, Spraying Systems Co.) into a 10 L atmospheric collector (Ve2) where particles were recovered. 2.3. Analytical methods 2.3.1. Solid state characterization The thermal stability of the obtained samples was determined using thermogravimetric analysis (TGA, Perkin Elmer 7) under Ar atmosphere and raising the temperature at a rate of 5 K min−1 . Photographs of the samples were taken using a JEOL JSM 6300 scanning electron microscope (SEM). Samples were also analyzed by X-ray diffraction (XRD) from a 2-value of 5–30◦ with a Rigaku Rotaflex RU200 B instrument, using a step of 0.02◦ and the CuK␣1 radiation. 2.3.2. HPLC characterization The chromatographic system consisted of an Agilent 1100 Series Instrument (Agilent Technologies, Waldbronn, Germany) equipped with a G1311A quaternary pump, a G1379A degasser, a G1315B diode-array detector furnished with a 13-␮L flow cell, a G1329B automatic injector and a chemstation for data acquisition and analysis. A CyberScan model 2500 potentiometer (precision of ±0.1 mV) with a combined pH electrode ORION 9103SC was used for pH measurements. The analytical column was a Synergi Hydro-RP C18 column (150 mm × 4.6 mm i.d., particle size 4 ␮m, 80 Å) equipped with a guard column (4 mm × 3 mm i.d.), both from Phenomenex (Torrance, CA, USA). An isocratic elution at the constant flow rate of 1 mL min−1 was utilized. For caffeine analysis, the chro-

344

C.A. García-González et al. / J. of Supercritical Fluids 54 (2010) 342–347

Table 1 Amount of additives added to the lipidic mixture for PGSS® processing and obtained results for the different prepared samples. Sample

GMS:HCO:TiO2 :9Cf GMS:HCO:TiO2 :17Cf GMS:HCO:TiO2 :10Cf/w GMS:HCO:TiO2 :9Gl GMS:HCO:TiO2 :9Kt GMS:HCO:TiO2 :16Kt a b

Processed mixture [wt%]

Precipitated SLPs [wt%]

TiO2

Drug

Water

TiO2

Drug

5 5 3 5 5 5

9 17 10 9 9 16

– – 40 – – –

5.7 5.7 6.8 7.9 5.1 5.3

4.2 3.6 12.5b 5.6 8.6 16.1

Eluted druga [wt%] (t2h )

61 – 80 100 – 20

Weight percentage of eluted drugs after 2 h of dissolution tests. Caffeine hydrate (C8 H10 N4 O2 ·0.8H2 O).

matographic eluent consisted of water/methanol (30:70, v/v) and chromatograms were recorded at 272 nm. For glutathione analysis, the eluent consisted of 75 mM phosphoric acid aqueous solution (pH 2.6)/acetonitrile (97:3, v/v) and chromatograms were recorded at 210 nm. For ketoprofen analysis, the mobile phase consisted of 10 mM formic acid/formate aqueous solution (pH 3.2)/methanol (30:70, v/v) and chromatograms were recorded at 266 nm. These HPLC conditions were used in the determination of the percentage of entrapment as well as in the monitoring of the drug release. 2.3.3. Determination of the percentage of drug entrapment In order to measure the total drug content in the precipitated particles, each drug was first extracted from the lipidic matrix and, then, determined by HPLC. Between 20 and 40 mg of the sample containing ketoprofen was treated with 10 mL of a methanol:water mixture (1:1, v/v). In the same way, 50–100 mg of samples containing caffeine were treated with 20 mL of an ethanol:water mixture (1:1, v/v) at 313 K, in accordance to the method described in the literature [17]. Lastly, around 10–20 mg of the sample containing glutathione was treated with 10 mL of MilliQ water. Subsequently, the samples were vortexed (OVAN® Vibra Mix, Lovango S.L.), centrifuged for 5 min at 3500 rpm (Rotanta 460 RS Hettich Zentrifugen) and filtered through a 0.45 ␮m Nylon filter (Scharlab). A volume of 20 ␮L of each sample solution was injected into the HPLC system for quantification. Standard solutions of ketoprofen and caffeine to be used in the calibration were prepared in methanol. Glutathione standard solutions were prepared in water. Detection limits were estimated for a signal-to-noise ratio of 3. Average repeatabilities for the retention time and peak area were calculated from 8 replicates. 2.3.4. Drug elution tests in water The drug encapsulation was preliminary investigated via dissolution tests in water. For ketoprofen and caffeine, 10 mg of samples containing these drugs were placed in vessels containing 200 mL of distilled water. For glutathione, 20 mg of sample were added to 50 mL of distilled water. All the elution tests were carried out at 310 ± 0.1 K fixing the stirring rate at 70 rpm (magnetic stirrer IKA® RCT Basic). Processes were chromatographically monitored to obtain the corresponding elution profiles. For this, aliquots of 500 ␮L, withdrawn over the period under study, were filtered and analyzed by HPLC. In parallel, identical volumes of fresh water were added to maintain the level of liquid. 3. Results and discussion Operating conditions, set at 13 MPa and 345 K, were chosen according to fundamental previous studies that consisted of measuring the melting point variation of the individual and mixed lipids at different pressures of scCO2 [13,27]. The selected working temperature was lower than the melting point of the studied drugs, which have values of 500–501 K for caffeine, 465–468 K for glutathione and 367–370 K for ketoprofen. The three studied drugs have a significantly different lipophilic character. For instance, the

logarithmic partition coefficient in octanol–water was 2.77 for the lipophilic ketoprofen [38], and −0.50 and −0.87 for the hydrophilic caffeine [39] and glutathione [40] drugs, respectively. The chemical formulation of the mixtures initially added to the autoclave for the formation of the hybrid SLPs are shown in Table 1. The weight of mineral filler in the PGSS® particles obtained was quantified using TGA. Values of ∼5–7 wt% were found from sample pyrolysis assays at temperatures higher than 725 K [41]. Therefore, under working experimental conditions, the percentage of mineral filler in the obtained particles was maintained or slightly increased with respect to the initially added amount to the mixing chamber for PGSS® processing. The enrichment in the inorganic part of the precipitated composite was due to the permanence of part of the organic matter in the mixing vessel, which was not transported to the collection chamber during expansion. This fact was particularly relevant in the low yield experiments, performed with hydrophilic drugs where nozzle blockage was frequent. The physical state of the different phases present in the precipitated composite was studied by XRD analysis. For both lipids, the only polymorphic phase observed after precipitation was the ␤ phase, with three strong reflections at 2 ∼ 5.4◦ , 19.7◦ and 22.1◦ [30]. The presence of TiO2 in the SLPs was confirmed by the anatase peak at 25.3◦ . Experiments with the mixture of chemicals GMS:HCO:TiO2 :xCf were performed with two initial caffeine contents: x = 9 or 17 wt%. For the sample processed with 9 wt% of caffeine, the HPLC analysis indicated that the drug loading in the precipitated GMS:HCO:TiO2 :9Cf sample was 4.2 wt% (Table 1), which corresponded to ∼50 wt% of the initially added amount of caffeine. The caffeine loss was related to the relatively high solubility of this solute in scCO2 under working conditions [42], and further precipitation inside the mixing vessel during decompression [13,27]. In the XRD pattern, this sample had a signal at 2 = 12.0◦ , corresponding to the major peak from caffeine crystals (Fig. 2). SEM images of the processed GMS:HCO:TiO2 :9Cf microparticles showed the presence of needle-like caffeine crystals deposited on the SLPs surface (Fig. 3(a)). For sample GMS:HCO:TiO2 :17Cf with a 17 wt% of caffeine initially added, blockage of the nozzle happened during expansion, likely caused by undissolved caffeine crystals. Caffeine is a hydrophilic compound presenting low solubility in the used lipophilic lipids. Hence, most probably the caffeine crystals added to the mixing vessel were not completely dissolved in the melted matrix, but dispersed. The increase of caffeine percentage in the initial mixture was not reflected in a higher drug loading in the precipitated SLPs (Table 1). In an attempt to increase the loaded amount of caffeine, a certain amount of water was added to the initial mixture (sample GMS:HCO:TiO2 :10 Cf/w). The caffeine dissolved in the water phase, and this solution was emulsified with the lipids during the stirring step (Lumulse® GMS-K is used as an emulsifier) prior to expansion. Free water was not occluded in the precipitated SLPs. However, the presence of water during precipitation led to the formation of the non-stoichiometric caffeine hydrate C8 H10 N4 O2 ·0.8H2 O, as shown

C.A. García-González et al. / J. of Supercritical Fluids 54 (2010) 342–347

Fig. 2. XRD characterization of PGSS® processed SLPs.

by the signals appearing at 2 = 8.1◦ , 10.6◦ , 12.2◦ and 13.2◦ in the XRD spectrum of the obtained sample (Fig. 2) [43]. The presence of crystals was also noticed in the SEM images (Fig. 3(b)). In this experiment, near all the added amount of caffeine to the processing vessel was present in the precipitated particles, since the solubility of the caffeine hydrate in scCO2 was reduced with respect to anhydrous caffeine. Similar experiments were performed using the mixture of chemicals GMS:HCO:TiO2 :9Gl (9 wt% in glutathione). The resulting sample contained a 5.6 wt% of drug (Table 1). In the XRD analysis, the presence of an intense signal at 2 = 22.4◦ indicated the formation of glutathione crystals in the precipitated SLPs. The process yield was relatively low due to nozzle blockage, likely caused by glutathione crystals. The approach of emulsifying the lipids with water to increase the loaded amount could not be used for glutathione, due to its quick oxidation in aqueous solutions [44]. Indeed, glutathione should be kept refrigerated and protected from light until given to the patient. A preliminary stability test was performed in the PGSS® coprecipitated particles with glutathione, in which the sample GMS:HCO:TiO2 :9Gl was kept at room temperature in a container tightly closed for 2 months. After this time, the HPLC analysis revealed only a 1% increase in oxidized glutathione when compared with original sample.

345

Finally, ketoprofen was processed in the mixture GMS:HCO:TiO2 :xKt (x = 9 or 16 wt%). HPLC analysis indicated that the ketoprofen content in the precipitated samples was 8.6 and 16.1 wt% (Table 1), respectively. In the XRD spectrum, raw ketoprofen had narrow well-defined peaks at 2 = 6.4◦ , 18.2◦ and 23.2◦ . For the sample processed with 9 wt% ketoprofen, only an incipient peak at 2 = 23.2◦ could be observed. On the contrary, the main peaks of pure ketoprofen appeared in the system processed with 16 wt% of the drug (sample GMS:HCO:TiO2 :16Kt in Fig. 2), indicating the formation of a significant amount of crystals. In this system, the chemical compatibility of the hydrophobic ketoprofen with the lipidic matrix favored the dissolution of the drug and the homogenization of the mixture. However, ketoprofen is relatively soluble in scCO2 [45] and, thus, it was partially solubilized in the supercritical phase during lipid melting and mixing step. During expansion, the quick and sudden depressurization taking place through the nozzle led to ketoprofen supersaturation and crystals precipitation on the surface of the formed lipid particles. An in vitro dissolution test in water was performed to investigate the encapsulation efficiency for the three studied drugs. Encapsulated systems were expected to be eluted to the medium in a prolonged way, while active agents with a low degree of entrapment should be dissolved very fast as the raw drugs did. The diffusion of the different active compounds from solid lipid particles was investigated over time by HPLC. The weight percentage of eluted drug at 2 h (t2h ) was used as a practical measurement for comparison of the behaviour of the different studied systems (Table 1). High dissolution rates (t2h = 100% in all the cases) were obtained when raw drugs were tested under similar experimental conditions. Different dissolution profiles were obtained when caffeine loaded SLPs were processed in the absence or presence of water (Fig. 4). Quick and sudden dissolution of all the caffeine (t2h = 100%) was obtained for sample GMS:HCO:TiO2 :10Cf/w processed with water, indicating a lack of hydrated caffeine encapsulation. In contrast, a more sustained release (t2h = 64%) with an initial intense burst of 40–50 wt% was observed for encapsulated anhydrous caffeine (sample GMS:HCO:TiO2 :9Cf in Fig. 4). This profile may be explained by the fast dissolution of the needle-like crystals of caffeine precipitated on the surface of the particles (Fig. 3(a)), followed by a prolonged dissolution of the caffeine contained within the lipid core. Fig. 4 also shows the dissolution profile of glutathione loaded SLPs (sample GMS:HCO:TiO2 :9Gl). For this sample, the drug was released very quickly from the lipidic matrix (t2h = 100%), in the form of a burst. Ketoprofen release profile differed significantly from the previous ones. For sample GMS:HCO:TiO2 :16Kt, after an initial burst of 15%, the rest of the drug was eluted in a prolonged

Fig. 3. SEM images of PGSS® precipitated hybrid SLPs: (a) sample GMS:HCO:TiO2 :9Cf, and (b) GMS:HCO:TiO2 :10Cf/w.

346

C.A. García-González et al. / J. of Supercritical Fluids 54 (2010) 342–347

Fig. 4. Drug release profiles in water at 310 K with a stirring rate of 70 rpm, obtained for particles loaded with anhydrous caffeine (sample GMS:HCO:TiO2 :9Cf), hydrated caffeine (sample GMS:HCO:TiO2 :10Cf/w), glutathione (sample GMS:HCO:TiO2 :9Gl), and ketoprofen (sample GMS:HCO:TiO2 :16Kt).

way (t2h = 20%, Fig. 4). Ketoprofen elution can be described with the solid-solution model [4], with the drug molecularly dispersed in the lipid matrix. 4. Conclusions A lipidic matrix of GMS and HCO (50 wt%) was used to form composite powders loaded with active agents (caffeine, glutathione or ketoprofen) and a mineral filler (silanized TiO2 ) using the PGSS® process. The matrix was coprecipitated with 6–7 wt% of silanized TiO2 , and with a 4–5 wt% of hydrophilic caffeine or glutathione and ca. 16 wt% of hydrophobic ketoprofen. The three studied systems had different elution behaviours depending on the lipophilic character of the active agent. Hydrophilic drugs, such as caffeine and glutathione, had a limited interaction with the lipophilic matrix components and thus a limited capacity to be encapsulated. As a consequence, for both drugs, most of the active agent was eluted to the medium as a burst in the first few hours of the dissolution test. The matrix used was more efficient for the encapsulation of caffeine (t2h = 64%) than glutathione (t2h = 100%). However, even when the encapsulation was not highly effective for glutathione, the obtained results indicated that the lipidic matrix kept this compound mainly in the reduced state. On the other hand, hydrophobic ketoprofen was entrapped in the lipidic particles as a molecular dispersion. As a result, the obtained systems could be used for sustained release (t2h = 20%), particularly of interest in topical applications, taking also advantages of the UV-shielding ability of the dispersed TiO2 . Acknowledgements The financial support of the Spanish projects MAT-2007-63355E, Ingenio 2010 CEN-20081027 (CDTI) and CTQ2008-05370/PPQ is gratefully acknowledged. Additional support for this work has been provided by the Generalitat of Catalonia under project 2009SGR666. C.A. García-González gives acknowledgment to CSIC for its funding support through I3P program. References [1] R.H. Müller, M. Radtke, S.A. Wissing, Solid lipid nanoparticles (SLN) and nanostructured lipid carriers (NLC) in cosmetic and dermatological preparations, Adv. Drug Deliv. Rev. 54 (2002) S131–S155. [2] S.S. Davis, Coming of age of lipid-based drug delivery systems, Adv. Drug Deliv. Rev. 56 (2004) 1241–1249. [3] A. zur Mühlen, C. Schwarz, W. Mehnert, Solid lipid nanoparticles (SLN) for controlled drug delivery—drug release and release mechanism, Eur. J. Pharm. Biopharm. 45 (1998) 149–155.

[4] R.H. Müller, K. Mäder, S. Gohla, Solid lipid nanoparticles (SLN) for controlled drug delivery-review of the state of the art, Eur. J. Pharm. Biopharm. 50 (2000) 161–177. [5] S.A. Wissing, R.H. Müller, Cosmetic applications for solid lipid nanoparticles (SLN), Int. J. Pharm. 254 (1) (2003) 65–68. [6] C.S. Maia, W. Mehnert, M. Schäfer-Korting, Solid lipid nanoparticles as drug carriers for topical glucocorticoids, Int. J. Pharm. 196 (2000) 165–167. [7] J. Pardeike, A. Hommoss, R.H. Müller, Lipid nanoparticles (SLN, NLC) in cosmetic and pharmaceutical dermal products, Int. J. Pharm. 366 (1–2) (2009) 170– 184. [8] V. Jenning, A. Gysler, M. Schäfer-Korting, S.H. Gohla, Vitamin A loaded solid lipid nanoparticles for topical use: occlusive properties and drug targeting to the upper skin, Eur. J. Pharm. Biopharm. 49 (2000) 211–218. [9] W. Mehnert, K. Mäder, Solid lipid nanoparticles: production, characterization and applications, Adv. Drug Deliv. Rev. 47 (2–3) (2001) 165–196. [10] J. Jung, M. Perrut, Particle design using supercritical fluids: literature and patent survey, J. Supercrit. Fluids 20 (2001) 179–189. [11] N. Elvassore, M. Flaibani, A. Bertucco, P. Caliceti, Thermodynamic analysis of micronization processes from gas-saturated solution, Ind. Eng. Chem. Res. 42 (2003) 5924–5929. [12] M. Rodrigues, N. Peiric¸o, H. Matos, E. Gomes de Azevedo, M.R. Lobato, A.J. Almeida, Microcomposites theophylline/hydrogenated palm oil from a PGSS process for controlled drug delivery systems, J. Supercrit. Fluids 29 (2004) 175–184. [13] C.A. García-González, A.R. Sampaio da Sousa, A. Argemí, A. López Periago, J. Saurina, C.M.M. Duarte, C. Domingo, Production of hybrid lipid-based particles loaded with inorganic nanoparticles and active compounds for prolonged topical release, Int. J. Pharm. 382 (2009) 296–304. [14] M.J. Whitaker, J. Hao, O.R. Davies, G. Serhatkulu, S. Stolnik-Trenkic, S.M. Howdle, K.M. Shakesheff, The production of protein-loaded microparticles by supercritical fluid enhanced mixing and spraying, J. Control. Release 101 (2005) 85–91. [15] A.R. Sampaio de Sousa, M. Calderone, E. Rodier, J. Fages, C.M.M. Duarte, Solubility of carbon dioxide in three lipid-based biocarriers, J. Supercrit. Fluids 39 (1) (2006) 13–19. [16] M. Calderone, E. Rodier, J.J. Letorneau, J. Fages, Solidification of Precirol® by the expansion of a supercritical fluid saturated melt: from the thermodynamic balance towards the crystallization aspect, J. Supercrit. Fluids 42 (2007) 189–199. [17] A.R. Sampaio de Sousa, A.L. Simplício, H.C. de Sousa, C.M.M. Duarte, Preparation of glyceryl monostearate-based particles by PGSS® : application to caffeine, J. Supercrit. Fluids 43 (1) (2007) 120–125. [18] A.R. Sampaio de Sousa, R. Silva, F.H. Tay, A.L. Simplício, S.G. Kazarian, C.M.M. Duarte, Solubility enhancement of trans-chalcone using lipid carriers and supercritical CO2 processing, J. Supercrit. Fluids 48 (2) (2009) 120–125. [19] S.G. Kazarian, Polymer processing with supercritical fluids, Polym. Sci. C 42 (2000) 78–101. [20] S. Spilimbergo, G. Luca, N. Elvassore, A. Bertucco, Effect of high-pressure gases on phase behaviour of solid lipids, J. Supercrit. Fluids 38 (3) (2006) 289–294. [21] F. Temelli, Perspectives on supercritical fluid processing of fats and oils, J. Supercrit. Fluids 47 (3) (2009) 583–590. [22] E. Weidner, Z. Knez, Z. Novak, Process for preparing particles or powders, Patent WO95/21688 (1995). [23] A.G. Gopala Krishna, Influence of viscosity on wax settling and refining loss in rice bran oil, J. Am. Oil Chem. Soc. 70 (9) (1993) 895–898. [24] D.S. Ogunniyi, Castor oil: a vital industrial raw material, Bioresour. Technol. 97 (9) (2006) 1086–1091. [25] K. Westesen, H. Bunjes, M.H.J. Koch, Physicochemical characterization of lipid nanoparticles and evaluation of their drug loading capacity and sustained release potential, J. Control. Release 48 (1997) 223–236. [26] C. García-González, J. Fraile, A.M. López-Periago, C. Domingo, Preparation of silane-coated TiO2 nanoparticles in supercritical CO2 , J. Colloid Interface Sci. 338 (2009) 491–499. [27] C. García-González, J. Saurina, J.A. Ayllón, C. Domingo, Preparation and characterization of surface silanized TiO2 nanoparticles under compressed CO2 : reaction kinetics, J. Phys. Chem. C 113 (2009) 13780–13786. [28] E. Loste, A. Fanovich, J. Fraile, C. Domingo, Anhydrous supercritical carbon dioxide method for the controlled silanization of inorganic nanoparticles, Adv. Mater. 16 (2004) 739–744. [29] C. Domingo, E. Loste, J. Fraile, Grafting of trialkoxysilane on the surface of nanoparticles by conventional wet alcoholic and supercritical carbon dioxide deposition methods, J. Supercrit. Fluids 37 (2006) 72–86. [30] C.L. Hexsel, S.D. Bangert, A.A. Hebert, H.W. Lim, Current sunscreen issues: 2007 Food and Drug Administration sunscreen labelling recommendations and combination sunscreen/insect repellent products, J. Am. Acad. Dermatol. 59 (2008) 316–323. [31] N.J. Lowe, N.A. Shaath, M.A. Pathak, Sunscreens: development, evaluation, and regulatory aspects Cosmetic Science and Technology Series, vol. 15, 2nd ed., Marcel Dekker, New York, 1997. [32] C. Bertin, H. Zunino, J.C. Pittet, P. Beau, P. Pineau, M. Massonneau, C. Robert, J. Hopkins, A double-bind evaluation of the activity of an anti-cellulite product containing retinol, caffeine, and ruscogenine by a combination of several noninvasive methods, J. Cosmet. Sci. 52–4 (2001) 19. [33] L. Montenegro, F. Bonina, L. Rigano, S. Giogilli, S. Sirigu, Protective effect evaluation of free radical scavengers on UVB induced human cutaneous erythema by skin reflectance spectrophotometry, Int. J. Cosmet. Sci. 17 (3) (1995) 91–103. [34] S. Ciceri, H.J. Hamann, I. Hurner, P. Kurka, J. Maasz, Ketoprofen liposomes, US Patent 5,741,515 (1998).

C.A. García-González et al. / J. of Supercritical Fluids 54 (2010) 342–347 [35] A.I. Arida, M.M. Al-Tabakha, Encapsulation of ketoprofen for controlled drug release, Eur. J. Pharm. Biopharm. 66 (1) (2007) 48–54. [36] J.A. Cordero, M. Camacho, R. Obach, J. Domenech, L. Vila, In vitro based index of topical anti-inflammatory activity to compare a series of NSAIDs, Eur. J. Pharm. Biopharm. 51 (2) (2001) 135–142. [37] C. García-González, J. Fraile, A.M. López-Periago, J. Saurina, C. Domingo, Measurements and correlation of octltriethoxysilane solubility in supercritical CO2 and assembly of functional silane monolayers on the surface of nanometric particles, Ind. Eng. Chem. Res. 48 (2009) 9952–9960. [38] X. Liu, H. Hefesha, G. Scriba, A. Fahr, Retention behavior of neutral and positively and negatively charged solutes on an immobilized-artificial-membrane (IAM) stationary phase, Helv. Chim. Acta 91 (8) (2008) 1505–1512. [39] R.S. Plumb, W.B. Potts III, P.D. Rainville, P.G. Alden, D.H. Shave, G. Baynham, J.R. Mazzeo, Addressing the analytical throughput challenges in ADME screening using rapid ultra-performance liquid chromatography/tandem mass spectrometry methodologies, Rapid Commun. Mass Spectrom. 22 (14) (2008) 2139–2152. [40] Calculated value using Advanced Chemistry Development (ACD/Labs) Software v9.04, presented in SciFinder database.

347

[41] C.A. García-González, J.M. Andanson, S.G. Kazarian, C. Domingo, J. Saurina, Application of principal component analysis to the thermal characterization of silanized nanoparticles obtained at supercritical carbon dioxide conditions, Anal. Chim. Acta 635 (2009) 227–234. ˜ R.S. Mohamed, M.G. Baer, P. Mazzafera, Extraction of purine [42] M.D.A. Saldana, alkaloids from Maté (Ilex paraguariensis) using supercritical CO2 , J. Agric. Food Chem. 47 (9) (1999) 3804–3808. [43] H.G.M. Edwards, E. Lawson, M. de Matas, L. Shields, P. York, Perkin Trans. 2 (10) (1997) 1985–1990. [44] A. Trapani, V. Laquintana, N. Denora, A. Lopedota, A. Cutrignelli, M. Franco, G. Trapani, G. Liso, Eudragit RS 100 microparticles containing 2-hydroxypropyl-␤cyclodextrin and glutathione: physicochemical characterization, drug release and transport studies, Eur. J. Pharm. Sci. 30 (2007) 64–74. [45] C.A. García-González, A. Vega-González, A.M. López-Periago, P. SubraPaternault, C. Domingo, Composite fibrous biomaterials for tissue engineering obtained using a supercritical CO2 antisolvent process, Acta Biomater. 5 (4) (2009) 1094–1103.