International Journal of Pharmaceutics 473 (2014) 87–94
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Polymorphism and kinetic behavior of binary mixtures of triglycerides Franco Pattarino a, *, Ruggero Bettini b , Andrea Foglio Bonda a , Andrea Della Bella b , Lorella Giovannelli a a b
Dipartimento di Scienze del Farmaco, Università degli Studi del Piemonte Orientale “A. Avogadro”, Largo Donegani, 2/3, 28100 Novara, Italy Dipartimento di Farmacia, Università degli Studi di Parma, Parco Area delle Scienze, 27/A, 43124 Parma, Italy
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 March 2014 Received in revised form 19 June 2014 Accepted 22 June 2014 Available online 24 June 2014
The work is aimed at investigating the polymorphism and the phase transition kinetics of binary lipid mixtures with potential application in controlled drug delivery. The lipid systems, constituted of glyceryl tristearate (GTS) added with different amounts (1.0–7.5% w/w) of a medium-chain liquid triglyceride (C10–C12 acyl derivative – MCT), were studied by differential scanning calorimetry, by X-ray diffraction and hot-stage microscopy. The liquid lipid, although present in small amount, modified the thermal profile and the diffraction pattern of the systems, indicating that it promoted the formation of the GTS stable polymorph, b, during the re-solidification of the melted mixture. This promotion effect of MCT was concentration-dependent and evident for systems containing MCT > 2.5%. Also the kinetics of transformation of GTS polymorphs was affected by the percentage of the liquid component. The a ! b-transition was a biphasic process which for GTS–MCT mixture (99:1) superimposed that of pure GTS, while followed a different trend for systems containing percentages of MCT higher than 2.5. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Lipids Triacyl glycerides Polymorphism Solid/liquid lipid mixture Kinetics phase behavior
1. Introduction Many liquid, semi-solid and solid lipids are involved in the formulation of drug dosage forms because of their versatile properties. Nowadays, many natural, semisynthetic and synthetic products are employed as matrices for drug formulation or as coating materials, and they can be of a single chemical structure or mixtures of definite and tunable composition. Among the lipids for pharmaceutical applications, triacyl glycerides (TAGs) represent the most employed classes. TAGs allow formulating active pharmaceutical ingredients, often possessing very poor solubility and/or permeability, thus enhancing their bioavailability; they make possible taste masking or protection of sensitive drugs from environmental or biological accidents; they can control modulate the delivery of drugs from modified release dosage forms. Moreover, most of them are GRAS listed (FDA, 2014) biodegradable and nontoxic (Traul et al., 2000). TAGs as lipid excipients offer undoubted advantages from the technological point of view as they can be processed by several methods, such as melt extrusion (Liu et al., 2001), melt granulation (Evrard et al., 1999; Zhang and Schwartz, 2003), spray cooling (Cavallari et al., 2005), spray drying
* Corresponding author. Tel.: +39 321 375863; fax: +39 321 375621. E-mail address: [email protected] (F. Pattarino). http://dx.doi.org/10.1016/j.ijpharm.2014.06.042 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
(Chauhan et al., 2005), supercritical fluid techniques (Salmaso et al., 2009), which generally employ mild process conditions. Most of these techniques involve the melting of lipids; the crystallization of metastable polymorphic forms during resolidification step represents an issue that should be addressed. TAGs in general exhibit three polymorphic forms (a, b0 and b) which have been identified by different techniques, such as X-ray diffraction (Lavigne et al., 1993; Norton et al., 1985), differential scanning calorimetry (DSC) (Hagemann and Rothfus, 1983; Bouzidi and Narine, 2012), Raman spectroscopy (Simpson and Hagemann, 1982), calorimetry (Norton et al., 1985; Charbonnet and Singleton, 1947) and microscopy (Okada, 1964). Many factors influence how a TAG crystallizes from the melt, including composition, tempering regime, presence of other lipids or additives and mechanical treatment (shear, stirring, etc.). A systematic examination of the properties of TAG-containing materials, with particular emphasis on polymorphism, crystallization and microstructure of individual, binary and ternary systems has been conducted (Boodhoo et al., 2008, 2009a,b; Bouzidi et al., 2010). The knowledge of the structure and properties of lipids used in pharmaceutical application is of extreme importance for their influence on the performance of the dosage form. In particular, the conversion from metastable to stable form of a lipid excipient, that may occur during the product shelf life, would alter the formulation and compromise its physical integrity, modify the distribution of the
88
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
Fig. 1. DSC profiles (a) and XRPD spectra (b) of GTS and GTS–MCT freshly re-solidified systems (*GTS as received; **GTS melted and re-solidified).
drug in the dosage form and change the solid state properties of the active substance; these changes are often referred as aging in the literature (Sutananta et al., 1994). Therefore, the behavior of the dosage form and the release and bioavailability of the carried drug could be unexpected and unpredictable. Mixtures of lipids containing a significant proportion of a liquid lipid are often employed as carriers of drugs, and these liquid components are medium-chain or unsaturated fatty acid glycerides. These substances are included in the mass of the lipid carrier in order to enhance the solubility of lipophilic drugs and to prevent the drug segregation or leakage during the shelf-life of the product; when present in high percentage they can form liquid domain in the mass of the formulation (Xia et al., 2014). The importance of these liquid lipids lies in their interactions with the drug, but also with the other components of a formulation: it was demonstrated that mediumchain glycerides, when associated with long-chained lipids, significantly affect the crystalline structure and stability of the system, and their effect is concentration-dependent (Windbergs et al., 2009). The present work is aimed at investigating the phase transition behavior of binary TAGs mixtures constituted of glyceryl tristearate (GTS) as solid component, and caprylic/capric triglyceride (MCT) as liquid lipid. In particular, systems containing low amounts of MCT were studied with the aim of elucidating the role played by the medium-chain triglycerides on the GTS crystal phase formation as well as on its polymorphic conversion during and after the mixture production.
Fig. 2. Difference between onset and maximum peak temperatures as a function of peak temperature. Circle = a-form of pure GTS and GTS–MCT mixtures with MCT 2.5%; square = a-form of GTS–MCT mixtures with MCT > 2.5%; triangle = bform of pure GTS and GTS–MCT mixtures.
2. Experimental methods 2.1. Materials Glyceryl tristearate (Dynasan 118 Mikrofein – 97.7%) was a CREMER product (CREMER OLEO GmbH, Witten, D), caprylic/capric triglyceride (LabrafacTM Lipophile WL1349) was a gift from Gattefossé (Milano, I). The materials were used as received, without any further treatment. 2.2. Preparation of mixtures The mixtures constituted of GTS and MCT (liquid content from 0.0 to 7.5% w/w) were prepared by melting the solid lipid in a water bath (85.0 C) and incorporating the liquid triglyceride under stirring. The melt was then solidified by immersion in an ice bath (0.0 C). 2.3. Differential scanning calorimetry DSC analyses were performed with a Pyris 1 (PerkinElmer, USA). The samples were prepared introducing an aliquot of the melted mixture equivalent to 5 mg of GTS in an aluminum pan that was then rapidly cooled in an ice bath (0.0 C) and sealed. Thermal scans were recorded at a heating rate of 5 C min1 under dry nitrogen purge (20 mL min1). Two sets of analysis were carried out: the scanning range was from 25 to 85 C in the first and from 55.0 to 25.0 C in the second one. The sample was maintained at
Fig. 3. DSC of freshly re-solidified systems in the temperature range from 55.0 to 25.0 C: the profiles from 20.0 to 20.0 C are shown.
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
89
the starting temperature for 15 min before the analysis. Each measurement was conducted in triplicate.
Hot-stage observations were carried out at 5 C min1 from 25 to 75 C under a flux of dry nitrogen (50 mL min1).
2.4. X-ray powder diffraction (XRPD)
3. Results and discussion
Diffraction patterns were recorded with an APD 2000 diffractometer (G.N.R. s.r.l., Italy). Measurements were performed in symmetrical reflection mode with CuK a radiation (l = 1.54 Å) with Göbel mirror bent multilayer optics in the angular range of 3–40 (2u) and at a scan speed of 0.2 s1.
The method used for the preparation of the systems afforded materials that were macroscopically homogeneous: moreover, before re-solidification, they were maintained for 10 min above the melting temperature, in order to remove possible crystallization grains (the “memory”) of the initial crystal structure of the solid component (Jenning et al., 2000). The protocol selected for the preparation of the samples for DSC analysis allowed an optimal distribution of the material into the crucible, avoiding the interferences due to the incomplete and variable contact between the sample and the bottom of the pan. The scanning rate for DSC analysis was chosen on the basis of the results from some preliminary thermal runs carried out at different rates (ranging for 0.1 and 10 C min1): the selected scanning rate (5 C min1)
2.5. Hot-stage microscopy Hot-stage microscopy analyses were conducted on a HSF 91 plate (Linkam Scientific Instrument, U.K.) at 20 magnification with an Optiphot2-Pol microscope (Nikon, Japan) fitted with a video camera (3CCD Color Vision Camera Module, Sony, Japan) equipped with a DAC-200 (Data video, Korea) digital interface.
Fig. 4. Pictures of pure GTS (left column) and binary mixture GTS plus 7.5% MCT (right column), taken at three different temperatures. Magnification 20.
90
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
Fig. 5. a-form enthalpy values as a function of MCT proportion in the mixtures immediately after re-solidification.
allowed obtaining thermal profiles characterized by acceptable and reproducible “resolution” of the peaks. The DSC analysis of GTS, as received from the manufacturer, gave the thermal profile of Fig. 1a in which a single endothermic peak with onset at 70.1 0.8 C was observed: it corresponds to the melting of the stable b-form of the lipid. A freshly re-solidified sample of glyceryl tristearate (GTS m&r) exhibited a melting endotherm referable to the metastable a-form (53.1 0.6 C), a crystallization exotherm (59.5 1.3 C) and a melting endotherm (67.2 0.9 C) that can be associated to the stable b-form of this lipid. It has to be underlined that the crystallization peak of b-form started after completion of the a-form melting: the peaks of the two events were well resolved. On the other hand, the crystallization and the melting peaks of the stable form were partially superimposed to each other, the endothermic peak beginning before the end of the exothermic one. The results of XRPD analysis, reported in Fig. 1b confirmed that pure GTS from the manufacturer was in the stable form: its pattern is typical of the b-polymorph, the peaks at 19.4, 23.1 and 24.1 2u being characteristic of the triclinic parallel structure (T//subcell) of the triglyceride (Bouzidi and Narine, 2012). At the opposite, freshly re-solidified GTS exhibited the single peak at 21.4 2u: this signal is typical of the H subcell of a-form, where chain–chain interactions are non-specific, and the lattice is loosely packed. The addition of low amounts (from 1.0 to 7.5%) of MCT to GTS formed mixtures that, immediately after re-solidification, showed different thermal and diffraction patterns.
For systems containing MCT 2.5%, the onsets of the a-form melting event are identical to that of GTS alone (53.1 C), while higher proportions of MCT determined a shift toward progressively lower temperatures. The introduction of short- or medium-chain triglycerides generally accelerated the rate of polymorphic transition of solid glycerides, and this change has been associated to the interactions between the two substances (Jenning et al., 2000). The difference between the onset and maximum peak temperatures of a thermal event for a mixture of substances can be taken as a measure of the crystal order disturbance induced by the minor component of the system. From Fig. 2 it can be seen that a low amount of MCT (2.5% – cluster 1) in the mixture did not significantly modify the onset-peak difference, while its value increased for higher content of the liquid lipid (cluster 2). These findings suggest that MCT 2.5% did not significantly alter the crystal structure of the metastable polymorph, being likely located among the crystal lamella of GTS. For medium-chain triglycerides in percentages >2.5%, higher values of onset-peak temperature difference could be observed. This finding indicates that MCT interacts with the solid lipid, probably at subcell unit level, and its presence forms a-crystals with slightly altered structure (Jenning et al., 2000). The presence of MCT has a significant effect on the crystallization of GTS stable form (Fig. 1). The exothermic peak appeared at progressively lower temperatures as the MCT percentage increased, and the shift of the peak determines the loss of resolution between the a-melting and b-crystallization signals and suggests the presence of defects in the crystals of the formed b-phase or the formation of small GTS crystals that showed lower onset melting temperature (Bettini et al., 2003). In fact, for all the re-solidified systems, the melting peak of b-form occurred at a temperature of 70.9 C, which was slightly lower than that of untreated GTS (72.6 C). Again, this difference can be explained taking into account that the b-phase is rapidly formed during the thermal scanning run and that it may be constituted of small crystals due to the relatively high rate of solid phase formation. The presence of MCT in the system seems not to perturb the crystal structure of the stable form, as the similarity of the onset-peak difference values indicates (Fig. 2 – cluster 3). These results allow hypothesizing that the liquid lipid is excluded from the lattice of b-crystals. DSC analyses of pure GTS, pure MCT and freshly re-solidified binary mixtures, carried out over a lower scanning temperatures interval (from 55.0 to 25.0 C) (Fig. 3), evidenced, almost for systems at the higher percentage levels, a melting event with about 14.5 C onset temperature that corresponded to the melting of MCT alone whose intensity was proportional to the amount of MTC in the binary mixture. This result indicates that MTC solidify as such in the binaries, thus supporting the above-mentioned hypothesis.
Fig. 6. DSC traces (a) and XRPD patterns (b) of pure GTS m&r during storage.
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
91
Fig. 7. DSC traces (a) and XRPD patterns (b) of GTS:MCT (99.0:1.0) system during storage.
With respect to the above reported point, the determination of the solubility of MCT in GTS, particularly under ambient temperature would be diriment. However, this is not an easy task from an experimental point of view. A recent work (Takeuchi et al., 2003) reported that for mixtures of monoacid TAGs, when the acids bound to glycerol differed by 4 or 6 carbon atoms, both TAGs were not miscible in all polymorphic forms. Nevertheless, we have carried out hot-stage microscopy measurements with physical mixtures of MCT and GTS in 1:1 weight ratio in the 25.0–75.0 C temperature range (data not shown). These measurements
evidenced a certain degree of miscibility between MCT and GTS
a-form well above the melting temperature of the latter (53.0 C), while for the GTS b-phase, miscibility could be observed only upon
GTS melting, namely after 70 C. This is not surprising, as it is well known that the solubility of a metastable form is always higher than that of the relevant stable form in whatever solvent (Pranzo et al., 2010). Moreover, as already stated, the melting peak of the a-form and the recrystallization peak of the b-form are partially superimposed in the trace of GTS–MCT mixtures (Fig. 1), due to the fact that the
Fig. 8. DSC traces (a) and XRPD patterns (b) of GTS:MCT (97.5:2.5) system during storage.
Fig. 9. DSC traces (a) and XRPD patterns (b) of GTS:MCT (95.0:5.0) system during storage.
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
Fig. 10. DSC traces (a) and XRPD patterns (b) of GTS:MCT (92.5:7.5) system during storage.
92
latter started at a lower temperature, while DSC profile of GTS m&r showed resolved peaks for both these endothermic and exothermic events. This indicates that MCT, although present in a very small amount, promotes the crystallization of b-polymorph. This assumption is supported also by the observation made with the HS microscopy (Fig. 4). In the case of the sample not containing MCT a demolition of the starting crystal followed by the formation of a new one was observed which showed a melting onset at about 70 C. On the contrary, with the binary containing 7.5% MCT, the crystallization of the b-polymorph occurred progressively, without a clear evidence of melting and recrystallization (only very small chances in crystal birefringence could be observed), over a large temperature interval starting at temperatures lower than 55.0 C. Final melting onset occurred around 62.0 C. Also, XRPD profile of MCT-containing systems showed an increased complexity (Fig. 1b): the intensity of the signal at 21.4 2u (associated to a-form of the stearyl glyceride) decreased with the increase of the percentage of MCT and, at the same time, the peaks at 19.4, 23.1 and 24.1 2u became more evident. It is worth noting that the changes of peak intensity seem directly related to the amount of MCT in the mixture. The complexity of XPRD patterns for the binary systems can be explained taking into account that MCT, if present as a separate component with respect to solid GTS (as previously hypothesized), can enhance the noise of the XRPD spectrum, giving reduced evidence for the small peaks of the primary lipid (GTS). MCT has also significant effects on the enthalpy of GTS peaks (Fig. 1a). The enthalpy of peaks were calculated to gain deeper insight into the melting and re-crystallization behavior of GTS–MCT mixtures: in order to obtain homogeneous and comparable values of this parameter, DH values were calculated by the tangential (backward and forward) skim method. As for the onset of b-crystallization a shift toward lower temperature was observed, DH values associated to this event were considered not fully comparable each other, so they was excluded from the calculation. In addition, the enthalpy values of b-melting were not considered because a large portion of this polymorph formed during the analysis. In Fig. 5, the calculated DH values for the a-melting are reported as a function of MCT concentration, and the regression analysis carried out on these data indicated a highly significant and inverse linear correlation between the two variables (R2 = 0.9772, p < 0.001). In other words, the amount of the a-form decreases with the MCT content in the mixture. This finding strongly suggested that MCT hinders the formation of the metastable GTS form during the re-solidification from the melted mixture. The kinetics of a ! b-conversion was also evaluated: samples of each system were stored at 25.0 C for 100 days and analyzed at predetermined times during this period. The obtained results are reported in Figs. 6–10. As previously reported (Windbergs et al., 2009), the thermal profiles of pure GTS, melted/re-solidified and stored for 60 days at room temperature, showed the presence of the three characteristic events. In our work, the thermal profiles of GTS (Fig. 6a), stored in the same conditions of temperature, have been acquired over a longer period of time (about 100 days). Immediately after resolidification, a-melting, b-crystallization and b-melting peaks were present, and the same signals can be observed after 100 days of storage: at this time, a significant signal relevant to the melting of the a-form of GTS can be appreciated, and the b-crystallization peak was clearly evident. A loss of resolution between the crystallization peak of the b-form and the melting peak of a-form, stemming from the progressive shift of the onset of b-form crystallization peak toward lower temperatures, was observed. At the same time, the resolution between the crystallization and
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
93
Fig. 11. Kinetics of a ! b-conversion for GTS and GTS–MCT systems: DH of a-polymorph as a function of storage time.
melting signals of b-polymorph increased. Matovic et al. (2005) have previously demonstrated that, during the thermal scan of triglycerides, a b0 -phase crystallized from the melted a-form; in our case, the presence of this polymorph cannot be detected because of its rapid conversion to the stable b-form. From the XRPD patterns of Fig. 6b, it can be seen that the samples analyzed after 35 and 100 days of storage showed the 21.4 2u peak as main signal, although decreasing over time, while the peaks relevant to the b-form appeared (19.4 and 24.1 2u) with intensity increased over time. For GTS–MCT systems, substantially different findings were obtained: the signals associated to a-melting and to b-crystallization and -melting were present in the DSC profiles of freshly re-solidified samples (Figs. 7–10a), but both a-melting and b-crystallization signals significantly decreased in intensity as a function of storage time. These findings indicated that a progressively higher amount of b-polymorph was present in the mixture over time, as a result of the promotion of the b-phase formation by MCT. This promotion effect is MCT-concentration dependent and makes more rapid the conversion to the stable form. Again, XRPD analysis results (Figs. 7–10b) support this explanation: the proportion of signals at 19.4 and 24.1 2u increased over the time with respect to the peak associated to the a-form. The diffractograms of the binary systems were more complex and not so sharp and neat as for pure GTS b-form because of the presence of liquid MCT that altered the background of XRPD pattern. Finally, the DH values of melting of the unstable form were plotted as a function of the storage time (Fig. 11). It can be noticed that the polymorphic transformation of GTS freshly re-solidified was characterized by a biphasic kinetics: the enthalpy of a-form dropped quickly in about 4 days; thereafter its decrease pursues more slowly. The significant decrease of enthalpy observed for the first period allows hypothesizing that, immediately after resolidification, a surplus of energy acquired during the melting and stored in the crystals, dissipates quite rapidly and GTS consolidates its crystalline structure. The successive phase of the process is characterized by a slow, constant decrease of the a-melting enthalpy due to the progressive a ! b-conversion.
In the case of systems containing MCT in low amount (2.5%), a kinetics similar to that of pure GTS could be observed. With respect to the first phase of the kinetics process, it has to be recalled that DH values at t = 0 were different and proportional to the concentration of MCT (see Fig. 6), as the liquid lipid promoted the crystallization of the b-form from the melt during the system re-solidification. For these systems, in the second phase of the process, an almost linear relationship between DH and time was observed. The linear regression analysis on the data relevant to the systems prepared with MCT at 0.0, 1.0 and 2.5% indicated that the conversion from metastable to stable form of lipid occurred with a quite similar rate (0.177, 0.177 and 0.165 J g1 d1, respectively). For systems containing MCT in concentration 5.0%, only the first step of the transformation process could be appreciated: the amount of a-polymorph crystallized from the melted mixtures was very low, and a short time was sufficient for its complete conversion to the stable b-polymorph. When MCT concentration in the binary mixture was 7.5%, the transition to the more stable form was completed in 20 min, while for the system containing 5% MCT, after about 1 day, only a very small amount of the metastable form (DH a-form accounted for 0.565 J g1) was present. 4. Conclusions Binary systems constituted of glyceryl tristearate and mediumchain triglyceride show thermodynamic and kinetics behaviors very different from that of the pure GTS. The results presented here demonstrate that even a limited amount of MTC (lower than 7.5% of the total mass) significantly affects the structure of the lipid mixture and affords, within a very short time, systems containing only the more stable crystalline form of GTS. Although present in very low percentages, MCT has significant influence on the evolution of the phase behavior of the GTS polymorphs: MCT partially hinders the formation of the metastable form of the glyceryl stearate during the re-solidification of the melted mixture, while it accelerates the formation of the b-form during both DSC scan and storage at ambient temperature. The investigation and the findings reported here are of primary importance from the pharmaceutical point of view. In fact, these
94
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
phenomena are expected to impact on the quality attributes of a finished dosage form, compromising the physical stability and/or modifying the pharmaceutical performances (drug release) of the formulation. The data reported in this work represent a tool for better understanding the phase behavior of the lipid excipient used in dosage form production, not only to avoid the above reported risks, but also for designing drug delivery systems with desired drug release kinetics. Acknowledgments The authors thank Dr. G. Ventimiglia and Dr. S. Bellomi for performing XRPD analysis. References Bettini, R., Rossi, A., Lavezzini, E., Pasquali, I., Giordano, F., 2003. Thermal and morphological characterization of micronized acetylsalicylic acid powders prepared by rapid expansion of a supercritical solution. J. Therm. Anal. Calorim. 73, 487–497. Boodhoo, M.V., Kutek, T., Filip, V., Narine, S.S., 2008. The binary phase behavior of 1,3-dimyristoyl-2-stearoyl-sn-glycerol and 1,2-dimyristoyl-3-stearoyl-sn-glycerol. Chem. Phys. Lipids 154, 7–18. Boodhoo, M.V., Bouzidi, L., Narine, S.S., 2009a. The binary phase behavior of 1,3dicaproyl-2-stearoyl-sn-glycerol and 1,2-dicaproyl-3-stearoyl-sn-glycerol. Chem. Phys. Lipids 157, 21–39. Boodhoo, M.V., Bouzidi, L., Narine, S.S., 2009b. The binary phase behavior of 1,3dipalmitoyl-2-stearoyl-sn-glycerol and 1,2-dipalmitoyl-3-stearoyl-sn-glycerol. Chem. Phys. Lipids 160, 11–32. Bouzidi, L., Narine, S.S., 2012. Relationships between molecular structure and kinetic and thermodynamic controls in lipid systems. Part II: phase behavior and transformation paths of SSS, PSS and PPS saturated triacylglycerols – effect of chain length mismatch. Chem. Phys. Lipids 165, 77–88. Bouzidi, L., Boodhoo, M.V., Kutek, T., Filip, V., Narine, S.S., 2010. The binary phase behavior of 1,3-dilauroyl-2-stearoyl-sn-glycerol and 1,2-dilauroyl-3-stearoylsn-glycerol. Chem. Phys. Lipids 163, 607–629. Cavallari, C., Rodriguez, L., Albertini, B., Passerini, N., Rosetti, F., Fini, A., 2005. Thermal and fractal analysis of diclofenac/Gelucire 50/13 microparticles obtained by ultrasound-assisted atomization. J. Pharm. Sci. 94, 1124–1134. Charbonnet, G.H., Singleton, W.S., 1947. Thermal properties of fats and oils. J. Am. Oil Chem. Soc. 24, 140–142.
Chauhan, B., Shimpi, S., Paradkar, A., 2005. Preparation and characterization of etoricoxib solid dispersions using lipid carriers by spray drying technique. AAPS PharmSciTech 6, E405–E412. Evrard, B., Amighi, K., Beten, D., Delattre, L., Moës, A.J., 1999. Influence of melting and rheological properties of fatty binders on the melt granulation process in a highshear mixer. Drug Dev. Ind. Pharm. 25, 1177–1184. FDA, 2014. Generally Recognized as Safe (GRAS). 21CFR Part 182. Available at: http:// www.fda.gov/Food/IngredientsPackagingLabeling/GRAS/ (accessed 20.03.14). Hagemann, J.W., Rothfus, J.A., 1983. Polymorphism and transformation energetics of saturated monoacid triglycerides from differential scanning calorimetry and theoretical modeling. J. Am. Oil Chem. Soc. 60, 1123–1131. Jenning, V., Thünemann, A.F., Gohla, S.H., 2000. Characterisation of a novel solid lipid nanoparticle carrier system based on binary mixtures of liquid and solid lipids. Int. J. Pharm. 199, 167–177. Lavigne, F., Bourgaux, C., Ollivon, M., 1993. Phase transitions of saturated triglycerides. J. Phys. IV Proc. 3, 137–140. Liu, J., Zhang, F., Ginity, Mc, J.W, 2001. Properties of lipophilic matrix tablets containing phenylpropanololamine hydrochloride prepared by hot-melt extrusion. Eur. J. Pharm. Biopharm. 52, 181–190. Matovic, M., van Miltenburg, J.C., Los, J., Gandolfo, F.G., Floter, E., 2005. Thermal properties of tristearin by adiabatic and differential scanning calorimetry. J. Chem. Eng. Data 50, 1624–1630. Norton, I.T., Lee-Tuffnell, C.D., Ablett, S., Bociek, S.M., 1985. A calorimetric, NMR and X-ray diffraction study of the melting behavior of tripalmitin and tristearin and their mixing behavior with triolein. J. Am. Oil Chem. Soc. 62, 1237–1244. Okada, M., 1964. The transformation of SSS crystal from a form to b form by melt grown method. J. Electron. Microsc. 180–181. Pranzo, M.B., Cruickshank, D., Coruzzi, M., Caira, M.R., Bettini, R., 2010. Enantiotropically related albendazole polymorphs. J. Pharm. Sci. 99, 3731–3742. Salmaso, S., Elvassore, N., Bertucco, A., Caliceti, P., 2009. Production of solid lipid submicron particles for protein delivery using a novel supercritical gas-assisted melting atomization process. J. Pharm. Sci. 98, 640–650. Simpson, T.D., Hagemann, J.W., 1982. Evidence of two b' phases in tristearin. J. Am. Oil Chem. Soc. 59, 169–171. Sutananta, W., Craig, D.Q.M., Newton, J.M., 1994. The effects of ageing on the thermal behaviour and mechanical properties of pharmaceutical glycerides. Int. J. Pharm. 111, 51–62. Takeuchi, M., Ueno, S., Sato, K., 2003. Synchrotron radiation SAXS/WAXS study of polymorph-dependent phase behavior of binary mixtures of saturated monoacid triacylglycerols. Cryst. Growth Des. 3, 369–374. Traul, K.A., Driedger, A., Ingle, D.L., Nakhasi, D., 2000. Review of the toxicologic properties of medium-chain triglycerides. Food Chem. Toxicol. 38, 79–98. Windbergs, M., Strachan, C.J., Kleinebudde, P., 2009. Investigating principles of recrystallization from glyceride melts. AAPS PharmSciTech 10, 1224–1233. Xia, D., Cui, F., Gan, Y., Mu, H., Yang, M., 2014. Design of lipid matrix particles for fenofibrate: effect of polymorphism of glycerol monostearate on drug incorporation and release. J. Pharm. Sci. 103, 697–705. Zhang, Y.E., Schwartz, J.B., 2003. Melt granulation and heat treatment for wax matrix-controlled drug release. Drug Dev. Ind. Pharm. 29, 131–138.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Polymorphism and kinetic behavior of binary mixtures of triglycerides Franco Pattarino a, *, Ruggero Bettini b , Andrea Foglio Bonda a , Andrea Della Bella b , Lorella Giovannelli a a b
Dipartimento di Scienze del Farmaco, Università degli Studi del Piemonte Orientale “A. Avogadro”, Largo Donegani, 2/3, 28100 Novara, Italy Dipartimento di Farmacia, Università degli Studi di Parma, Parco Area delle Scienze, 27/A, 43124 Parma, Italy
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 March 2014 Received in revised form 19 June 2014 Accepted 22 June 2014 Available online 24 June 2014
The work is aimed at investigating the polymorphism and the phase transition kinetics of binary lipid mixtures with potential application in controlled drug delivery. The lipid systems, constituted of glyceryl tristearate (GTS) added with different amounts (1.0–7.5% w/w) of a medium-chain liquid triglyceride (C10–C12 acyl derivative – MCT), were studied by differential scanning calorimetry, by X-ray diffraction and hot-stage microscopy. The liquid lipid, although present in small amount, modified the thermal profile and the diffraction pattern of the systems, indicating that it promoted the formation of the GTS stable polymorph, b, during the re-solidification of the melted mixture. This promotion effect of MCT was concentration-dependent and evident for systems containing MCT > 2.5%. Also the kinetics of transformation of GTS polymorphs was affected by the percentage of the liquid component. The a ! b-transition was a biphasic process which for GTS–MCT mixture (99:1) superimposed that of pure GTS, while followed a different trend for systems containing percentages of MCT higher than 2.5. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Lipids Triacyl glycerides Polymorphism Solid/liquid lipid mixture Kinetics phase behavior
1. Introduction Many liquid, semi-solid and solid lipids are involved in the formulation of drug dosage forms because of their versatile properties. Nowadays, many natural, semisynthetic and synthetic products are employed as matrices for drug formulation or as coating materials, and they can be of a single chemical structure or mixtures of definite and tunable composition. Among the lipids for pharmaceutical applications, triacyl glycerides (TAGs) represent the most employed classes. TAGs allow formulating active pharmaceutical ingredients, often possessing very poor solubility and/or permeability, thus enhancing their bioavailability; they make possible taste masking or protection of sensitive drugs from environmental or biological accidents; they can control modulate the delivery of drugs from modified release dosage forms. Moreover, most of them are GRAS listed (FDA, 2014) biodegradable and nontoxic (Traul et al., 2000). TAGs as lipid excipients offer undoubted advantages from the technological point of view as they can be processed by several methods, such as melt extrusion (Liu et al., 2001), melt granulation (Evrard et al., 1999; Zhang and Schwartz, 2003), spray cooling (Cavallari et al., 2005), spray drying
* Corresponding author. Tel.: +39 321 375863; fax: +39 321 375621. E-mail address: [email protected] (F. Pattarino). http://dx.doi.org/10.1016/j.ijpharm.2014.06.042 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
(Chauhan et al., 2005), supercritical fluid techniques (Salmaso et al., 2009), which generally employ mild process conditions. Most of these techniques involve the melting of lipids; the crystallization of metastable polymorphic forms during resolidification step represents an issue that should be addressed. TAGs in general exhibit three polymorphic forms (a, b0 and b) which have been identified by different techniques, such as X-ray diffraction (Lavigne et al., 1993; Norton et al., 1985), differential scanning calorimetry (DSC) (Hagemann and Rothfus, 1983; Bouzidi and Narine, 2012), Raman spectroscopy (Simpson and Hagemann, 1982), calorimetry (Norton et al., 1985; Charbonnet and Singleton, 1947) and microscopy (Okada, 1964). Many factors influence how a TAG crystallizes from the melt, including composition, tempering regime, presence of other lipids or additives and mechanical treatment (shear, stirring, etc.). A systematic examination of the properties of TAG-containing materials, with particular emphasis on polymorphism, crystallization and microstructure of individual, binary and ternary systems has been conducted (Boodhoo et al., 2008, 2009a,b; Bouzidi et al., 2010). The knowledge of the structure and properties of lipids used in pharmaceutical application is of extreme importance for their influence on the performance of the dosage form. In particular, the conversion from metastable to stable form of a lipid excipient, that may occur during the product shelf life, would alter the formulation and compromise its physical integrity, modify the distribution of the
88
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
Fig. 1. DSC profiles (a) and XRPD spectra (b) of GTS and GTS–MCT freshly re-solidified systems (*GTS as received; **GTS melted and re-solidified).
drug in the dosage form and change the solid state properties of the active substance; these changes are often referred as aging in the literature (Sutananta et al., 1994). Therefore, the behavior of the dosage form and the release and bioavailability of the carried drug could be unexpected and unpredictable. Mixtures of lipids containing a significant proportion of a liquid lipid are often employed as carriers of drugs, and these liquid components are medium-chain or unsaturated fatty acid glycerides. These substances are included in the mass of the lipid carrier in order to enhance the solubility of lipophilic drugs and to prevent the drug segregation or leakage during the shelf-life of the product; when present in high percentage they can form liquid domain in the mass of the formulation (Xia et al., 2014). The importance of these liquid lipids lies in their interactions with the drug, but also with the other components of a formulation: it was demonstrated that mediumchain glycerides, when associated with long-chained lipids, significantly affect the crystalline structure and stability of the system, and their effect is concentration-dependent (Windbergs et al., 2009). The present work is aimed at investigating the phase transition behavior of binary TAGs mixtures constituted of glyceryl tristearate (GTS) as solid component, and caprylic/capric triglyceride (MCT) as liquid lipid. In particular, systems containing low amounts of MCT were studied with the aim of elucidating the role played by the medium-chain triglycerides on the GTS crystal phase formation as well as on its polymorphic conversion during and after the mixture production.
Fig. 2. Difference between onset and maximum peak temperatures as a function of peak temperature. Circle = a-form of pure GTS and GTS–MCT mixtures with MCT 2.5%; square = a-form of GTS–MCT mixtures with MCT > 2.5%; triangle = bform of pure GTS and GTS–MCT mixtures.
2. Experimental methods 2.1. Materials Glyceryl tristearate (Dynasan 118 Mikrofein – 97.7%) was a CREMER product (CREMER OLEO GmbH, Witten, D), caprylic/capric triglyceride (LabrafacTM Lipophile WL1349) was a gift from Gattefossé (Milano, I). The materials were used as received, without any further treatment. 2.2. Preparation of mixtures The mixtures constituted of GTS and MCT (liquid content from 0.0 to 7.5% w/w) were prepared by melting the solid lipid in a water bath (85.0 C) and incorporating the liquid triglyceride under stirring. The melt was then solidified by immersion in an ice bath (0.0 C). 2.3. Differential scanning calorimetry DSC analyses were performed with a Pyris 1 (PerkinElmer, USA). The samples were prepared introducing an aliquot of the melted mixture equivalent to 5 mg of GTS in an aluminum pan that was then rapidly cooled in an ice bath (0.0 C) and sealed. Thermal scans were recorded at a heating rate of 5 C min1 under dry nitrogen purge (20 mL min1). Two sets of analysis were carried out: the scanning range was from 25 to 85 C in the first and from 55.0 to 25.0 C in the second one. The sample was maintained at
Fig. 3. DSC of freshly re-solidified systems in the temperature range from 55.0 to 25.0 C: the profiles from 20.0 to 20.0 C are shown.
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
89
the starting temperature for 15 min before the analysis. Each measurement was conducted in triplicate.
Hot-stage observations were carried out at 5 C min1 from 25 to 75 C under a flux of dry nitrogen (50 mL min1).
2.4. X-ray powder diffraction (XRPD)
3. Results and discussion
Diffraction patterns were recorded with an APD 2000 diffractometer (G.N.R. s.r.l., Italy). Measurements were performed in symmetrical reflection mode with CuK a radiation (l = 1.54 Å) with Göbel mirror bent multilayer optics in the angular range of 3–40 (2u) and at a scan speed of 0.2 s1.
The method used for the preparation of the systems afforded materials that were macroscopically homogeneous: moreover, before re-solidification, they were maintained for 10 min above the melting temperature, in order to remove possible crystallization grains (the “memory”) of the initial crystal structure of the solid component (Jenning et al., 2000). The protocol selected for the preparation of the samples for DSC analysis allowed an optimal distribution of the material into the crucible, avoiding the interferences due to the incomplete and variable contact between the sample and the bottom of the pan. The scanning rate for DSC analysis was chosen on the basis of the results from some preliminary thermal runs carried out at different rates (ranging for 0.1 and 10 C min1): the selected scanning rate (5 C min1)
2.5. Hot-stage microscopy Hot-stage microscopy analyses were conducted on a HSF 91 plate (Linkam Scientific Instrument, U.K.) at 20 magnification with an Optiphot2-Pol microscope (Nikon, Japan) fitted with a video camera (3CCD Color Vision Camera Module, Sony, Japan) equipped with a DAC-200 (Data video, Korea) digital interface.
Fig. 4. Pictures of pure GTS (left column) and binary mixture GTS plus 7.5% MCT (right column), taken at three different temperatures. Magnification 20.
90
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
Fig. 5. a-form enthalpy values as a function of MCT proportion in the mixtures immediately after re-solidification.
allowed obtaining thermal profiles characterized by acceptable and reproducible “resolution” of the peaks. The DSC analysis of GTS, as received from the manufacturer, gave the thermal profile of Fig. 1a in which a single endothermic peak with onset at 70.1 0.8 C was observed: it corresponds to the melting of the stable b-form of the lipid. A freshly re-solidified sample of glyceryl tristearate (GTS m&r) exhibited a melting endotherm referable to the metastable a-form (53.1 0.6 C), a crystallization exotherm (59.5 1.3 C) and a melting endotherm (67.2 0.9 C) that can be associated to the stable b-form of this lipid. It has to be underlined that the crystallization peak of b-form started after completion of the a-form melting: the peaks of the two events were well resolved. On the other hand, the crystallization and the melting peaks of the stable form were partially superimposed to each other, the endothermic peak beginning before the end of the exothermic one. The results of XRPD analysis, reported in Fig. 1b confirmed that pure GTS from the manufacturer was in the stable form: its pattern is typical of the b-polymorph, the peaks at 19.4, 23.1 and 24.1 2u being characteristic of the triclinic parallel structure (T//subcell) of the triglyceride (Bouzidi and Narine, 2012). At the opposite, freshly re-solidified GTS exhibited the single peak at 21.4 2u: this signal is typical of the H subcell of a-form, where chain–chain interactions are non-specific, and the lattice is loosely packed. The addition of low amounts (from 1.0 to 7.5%) of MCT to GTS formed mixtures that, immediately after re-solidification, showed different thermal and diffraction patterns.
For systems containing MCT 2.5%, the onsets of the a-form melting event are identical to that of GTS alone (53.1 C), while higher proportions of MCT determined a shift toward progressively lower temperatures. The introduction of short- or medium-chain triglycerides generally accelerated the rate of polymorphic transition of solid glycerides, and this change has been associated to the interactions between the two substances (Jenning et al., 2000). The difference between the onset and maximum peak temperatures of a thermal event for a mixture of substances can be taken as a measure of the crystal order disturbance induced by the minor component of the system. From Fig. 2 it can be seen that a low amount of MCT (2.5% – cluster 1) in the mixture did not significantly modify the onset-peak difference, while its value increased for higher content of the liquid lipid (cluster 2). These findings suggest that MCT 2.5% did not significantly alter the crystal structure of the metastable polymorph, being likely located among the crystal lamella of GTS. For medium-chain triglycerides in percentages >2.5%, higher values of onset-peak temperature difference could be observed. This finding indicates that MCT interacts with the solid lipid, probably at subcell unit level, and its presence forms a-crystals with slightly altered structure (Jenning et al., 2000). The presence of MCT has a significant effect on the crystallization of GTS stable form (Fig. 1). The exothermic peak appeared at progressively lower temperatures as the MCT percentage increased, and the shift of the peak determines the loss of resolution between the a-melting and b-crystallization signals and suggests the presence of defects in the crystals of the formed b-phase or the formation of small GTS crystals that showed lower onset melting temperature (Bettini et al., 2003). In fact, for all the re-solidified systems, the melting peak of b-form occurred at a temperature of 70.9 C, which was slightly lower than that of untreated GTS (72.6 C). Again, this difference can be explained taking into account that the b-phase is rapidly formed during the thermal scanning run and that it may be constituted of small crystals due to the relatively high rate of solid phase formation. The presence of MCT in the system seems not to perturb the crystal structure of the stable form, as the similarity of the onset-peak difference values indicates (Fig. 2 – cluster 3). These results allow hypothesizing that the liquid lipid is excluded from the lattice of b-crystals. DSC analyses of pure GTS, pure MCT and freshly re-solidified binary mixtures, carried out over a lower scanning temperatures interval (from 55.0 to 25.0 C) (Fig. 3), evidenced, almost for systems at the higher percentage levels, a melting event with about 14.5 C onset temperature that corresponded to the melting of MCT alone whose intensity was proportional to the amount of MTC in the binary mixture. This result indicates that MTC solidify as such in the binaries, thus supporting the above-mentioned hypothesis.
Fig. 6. DSC traces (a) and XRPD patterns (b) of pure GTS m&r during storage.
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
91
Fig. 7. DSC traces (a) and XRPD patterns (b) of GTS:MCT (99.0:1.0) system during storage.
With respect to the above reported point, the determination of the solubility of MCT in GTS, particularly under ambient temperature would be diriment. However, this is not an easy task from an experimental point of view. A recent work (Takeuchi et al., 2003) reported that for mixtures of monoacid TAGs, when the acids bound to glycerol differed by 4 or 6 carbon atoms, both TAGs were not miscible in all polymorphic forms. Nevertheless, we have carried out hot-stage microscopy measurements with physical mixtures of MCT and GTS in 1:1 weight ratio in the 25.0–75.0 C temperature range (data not shown). These measurements
evidenced a certain degree of miscibility between MCT and GTS
a-form well above the melting temperature of the latter (53.0 C), while for the GTS b-phase, miscibility could be observed only upon
GTS melting, namely after 70 C. This is not surprising, as it is well known that the solubility of a metastable form is always higher than that of the relevant stable form in whatever solvent (Pranzo et al., 2010). Moreover, as already stated, the melting peak of the a-form and the recrystallization peak of the b-form are partially superimposed in the trace of GTS–MCT mixtures (Fig. 1), due to the fact that the
Fig. 8. DSC traces (a) and XRPD patterns (b) of GTS:MCT (97.5:2.5) system during storage.
Fig. 9. DSC traces (a) and XRPD patterns (b) of GTS:MCT (95.0:5.0) system during storage.
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
Fig. 10. DSC traces (a) and XRPD patterns (b) of GTS:MCT (92.5:7.5) system during storage.
92
latter started at a lower temperature, while DSC profile of GTS m&r showed resolved peaks for both these endothermic and exothermic events. This indicates that MCT, although present in a very small amount, promotes the crystallization of b-polymorph. This assumption is supported also by the observation made with the HS microscopy (Fig. 4). In the case of the sample not containing MCT a demolition of the starting crystal followed by the formation of a new one was observed which showed a melting onset at about 70 C. On the contrary, with the binary containing 7.5% MCT, the crystallization of the b-polymorph occurred progressively, without a clear evidence of melting and recrystallization (only very small chances in crystal birefringence could be observed), over a large temperature interval starting at temperatures lower than 55.0 C. Final melting onset occurred around 62.0 C. Also, XRPD profile of MCT-containing systems showed an increased complexity (Fig. 1b): the intensity of the signal at 21.4 2u (associated to a-form of the stearyl glyceride) decreased with the increase of the percentage of MCT and, at the same time, the peaks at 19.4, 23.1 and 24.1 2u became more evident. It is worth noting that the changes of peak intensity seem directly related to the amount of MCT in the mixture. The complexity of XPRD patterns for the binary systems can be explained taking into account that MCT, if present as a separate component with respect to solid GTS (as previously hypothesized), can enhance the noise of the XRPD spectrum, giving reduced evidence for the small peaks of the primary lipid (GTS). MCT has also significant effects on the enthalpy of GTS peaks (Fig. 1a). The enthalpy of peaks were calculated to gain deeper insight into the melting and re-crystallization behavior of GTS–MCT mixtures: in order to obtain homogeneous and comparable values of this parameter, DH values were calculated by the tangential (backward and forward) skim method. As for the onset of b-crystallization a shift toward lower temperature was observed, DH values associated to this event were considered not fully comparable each other, so they was excluded from the calculation. In addition, the enthalpy values of b-melting were not considered because a large portion of this polymorph formed during the analysis. In Fig. 5, the calculated DH values for the a-melting are reported as a function of MCT concentration, and the regression analysis carried out on these data indicated a highly significant and inverse linear correlation between the two variables (R2 = 0.9772, p < 0.001). In other words, the amount of the a-form decreases with the MCT content in the mixture. This finding strongly suggested that MCT hinders the formation of the metastable GTS form during the re-solidification from the melted mixture. The kinetics of a ! b-conversion was also evaluated: samples of each system were stored at 25.0 C for 100 days and analyzed at predetermined times during this period. The obtained results are reported in Figs. 6–10. As previously reported (Windbergs et al., 2009), the thermal profiles of pure GTS, melted/re-solidified and stored for 60 days at room temperature, showed the presence of the three characteristic events. In our work, the thermal profiles of GTS (Fig. 6a), stored in the same conditions of temperature, have been acquired over a longer period of time (about 100 days). Immediately after resolidification, a-melting, b-crystallization and b-melting peaks were present, and the same signals can be observed after 100 days of storage: at this time, a significant signal relevant to the melting of the a-form of GTS can be appreciated, and the b-crystallization peak was clearly evident. A loss of resolution between the crystallization peak of the b-form and the melting peak of a-form, stemming from the progressive shift of the onset of b-form crystallization peak toward lower temperatures, was observed. At the same time, the resolution between the crystallization and
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
93
Fig. 11. Kinetics of a ! b-conversion for GTS and GTS–MCT systems: DH of a-polymorph as a function of storage time.
melting signals of b-polymorph increased. Matovic et al. (2005) have previously demonstrated that, during the thermal scan of triglycerides, a b0 -phase crystallized from the melted a-form; in our case, the presence of this polymorph cannot be detected because of its rapid conversion to the stable b-form. From the XRPD patterns of Fig. 6b, it can be seen that the samples analyzed after 35 and 100 days of storage showed the 21.4 2u peak as main signal, although decreasing over time, while the peaks relevant to the b-form appeared (19.4 and 24.1 2u) with intensity increased over time. For GTS–MCT systems, substantially different findings were obtained: the signals associated to a-melting and to b-crystallization and -melting were present in the DSC profiles of freshly re-solidified samples (Figs. 7–10a), but both a-melting and b-crystallization signals significantly decreased in intensity as a function of storage time. These findings indicated that a progressively higher amount of b-polymorph was present in the mixture over time, as a result of the promotion of the b-phase formation by MCT. This promotion effect is MCT-concentration dependent and makes more rapid the conversion to the stable form. Again, XRPD analysis results (Figs. 7–10b) support this explanation: the proportion of signals at 19.4 and 24.1 2u increased over the time with respect to the peak associated to the a-form. The diffractograms of the binary systems were more complex and not so sharp and neat as for pure GTS b-form because of the presence of liquid MCT that altered the background of XRPD pattern. Finally, the DH values of melting of the unstable form were plotted as a function of the storage time (Fig. 11). It can be noticed that the polymorphic transformation of GTS freshly re-solidified was characterized by a biphasic kinetics: the enthalpy of a-form dropped quickly in about 4 days; thereafter its decrease pursues more slowly. The significant decrease of enthalpy observed for the first period allows hypothesizing that, immediately after resolidification, a surplus of energy acquired during the melting and stored in the crystals, dissipates quite rapidly and GTS consolidates its crystalline structure. The successive phase of the process is characterized by a slow, constant decrease of the a-melting enthalpy due to the progressive a ! b-conversion.
In the case of systems containing MCT in low amount (2.5%), a kinetics similar to that of pure GTS could be observed. With respect to the first phase of the kinetics process, it has to be recalled that DH values at t = 0 were different and proportional to the concentration of MCT (see Fig. 6), as the liquid lipid promoted the crystallization of the b-form from the melt during the system re-solidification. For these systems, in the second phase of the process, an almost linear relationship between DH and time was observed. The linear regression analysis on the data relevant to the systems prepared with MCT at 0.0, 1.0 and 2.5% indicated that the conversion from metastable to stable form of lipid occurred with a quite similar rate (0.177, 0.177 and 0.165 J g1 d1, respectively). For systems containing MCT in concentration 5.0%, only the first step of the transformation process could be appreciated: the amount of a-polymorph crystallized from the melted mixtures was very low, and a short time was sufficient for its complete conversion to the stable b-polymorph. When MCT concentration in the binary mixture was 7.5%, the transition to the more stable form was completed in 20 min, while for the system containing 5% MCT, after about 1 day, only a very small amount of the metastable form (DH a-form accounted for 0.565 J g1) was present. 4. Conclusions Binary systems constituted of glyceryl tristearate and mediumchain triglyceride show thermodynamic and kinetics behaviors very different from that of the pure GTS. The results presented here demonstrate that even a limited amount of MTC (lower than 7.5% of the total mass) significantly affects the structure of the lipid mixture and affords, within a very short time, systems containing only the more stable crystalline form of GTS. Although present in very low percentages, MCT has significant influence on the evolution of the phase behavior of the GTS polymorphs: MCT partially hinders the formation of the metastable form of the glyceryl stearate during the re-solidification of the melted mixture, while it accelerates the formation of the b-form during both DSC scan and storage at ambient temperature. The investigation and the findings reported here are of primary importance from the pharmaceutical point of view. In fact, these
94
F. Pattarino et al. / International Journal of Pharmaceutics 473 (2014) 87–94
phenomena are expected to impact on the quality attributes of a finished dosage form, compromising the physical stability and/or modifying the pharmaceutical performances (drug release) of the formulation. The data reported in this work represent a tool for better understanding the phase behavior of the lipid excipient used in dosage form production, not only to avoid the above reported risks, but also for designing drug delivery systems with desired drug release kinetics. Acknowledgments The authors thank Dr. G. Ventimiglia and Dr. S. Bellomi for performing XRPD analysis. References Bettini, R., Rossi, A., Lavezzini, E., Pasquali, I., Giordano, F., 2003. Thermal and morphological characterization of micronized acetylsalicylic acid powders prepared by rapid expansion of a supercritical solution. J. Therm. Anal. Calorim. 73, 487–497. Boodhoo, M.V., Kutek, T., Filip, V., Narine, S.S., 2008. The binary phase behavior of 1,3-dimyristoyl-2-stearoyl-sn-glycerol and 1,2-dimyristoyl-3-stearoyl-sn-glycerol. Chem. Phys. Lipids 154, 7–18. Boodhoo, M.V., Bouzidi, L., Narine, S.S., 2009a. The binary phase behavior of 1,3dicaproyl-2-stearoyl-sn-glycerol and 1,2-dicaproyl-3-stearoyl-sn-glycerol. Chem. Phys. Lipids 157, 21–39. Boodhoo, M.V., Bouzidi, L., Narine, S.S., 2009b. The binary phase behavior of 1,3dipalmitoyl-2-stearoyl-sn-glycerol and 1,2-dipalmitoyl-3-stearoyl-sn-glycerol. Chem. Phys. Lipids 160, 11–32. Bouzidi, L., Narine, S.S., 2012. Relationships between molecular structure and kinetic and thermodynamic controls in lipid systems. Part II: phase behavior and transformation paths of SSS, PSS and PPS saturated triacylglycerols – effect of chain length mismatch. Chem. Phys. Lipids 165, 77–88. Bouzidi, L., Boodhoo, M.V., Kutek, T., Filip, V., Narine, S.S., 2010. The binary phase behavior of 1,3-dilauroyl-2-stearoyl-sn-glycerol and 1,2-dilauroyl-3-stearoylsn-glycerol. Chem. Phys. Lipids 163, 607–629. Cavallari, C., Rodriguez, L., Albertini, B., Passerini, N., Rosetti, F., Fini, A., 2005. Thermal and fractal analysis of diclofenac/Gelucire 50/13 microparticles obtained by ultrasound-assisted atomization. J. Pharm. Sci. 94, 1124–1134. Charbonnet, G.H., Singleton, W.S., 1947. Thermal properties of fats and oils. J. Am. Oil Chem. Soc. 24, 140–142.
Chauhan, B., Shimpi, S., Paradkar, A., 2005. Preparation and characterization of etoricoxib solid dispersions using lipid carriers by spray drying technique. AAPS PharmSciTech 6, E405–E412. Evrard, B., Amighi, K., Beten, D., Delattre, L., Moës, A.J., 1999. Influence of melting and rheological properties of fatty binders on the melt granulation process in a highshear mixer. Drug Dev. Ind. Pharm. 25, 1177–1184. FDA, 2014. Generally Recognized as Safe (GRAS). 21CFR Part 182. Available at: http:// www.fda.gov/Food/IngredientsPackagingLabeling/GRAS/ (accessed 20.03.14). Hagemann, J.W., Rothfus, J.A., 1983. Polymorphism and transformation energetics of saturated monoacid triglycerides from differential scanning calorimetry and theoretical modeling. J. Am. Oil Chem. Soc. 60, 1123–1131. Jenning, V., Thünemann, A.F., Gohla, S.H., 2000. Characterisation of a novel solid lipid nanoparticle carrier system based on binary mixtures of liquid and solid lipids. Int. J. Pharm. 199, 167–177. Lavigne, F., Bourgaux, C., Ollivon, M., 1993. Phase transitions of saturated triglycerides. J. Phys. IV Proc. 3, 137–140. Liu, J., Zhang, F., Ginity, Mc, J.W, 2001. Properties of lipophilic matrix tablets containing phenylpropanololamine hydrochloride prepared by hot-melt extrusion. Eur. J. Pharm. Biopharm. 52, 181–190. Matovic, M., van Miltenburg, J.C., Los, J., Gandolfo, F.G., Floter, E., 2005. Thermal properties of tristearin by adiabatic and differential scanning calorimetry. J. Chem. Eng. Data 50, 1624–1630. Norton, I.T., Lee-Tuffnell, C.D., Ablett, S., Bociek, S.M., 1985. A calorimetric, NMR and X-ray diffraction study of the melting behavior of tripalmitin and tristearin and their mixing behavior with triolein. J. Am. Oil Chem. Soc. 62, 1237–1244. Okada, M., 1964. The transformation of SSS crystal from a form to b form by melt grown method. J. Electron. Microsc. 180–181. Pranzo, M.B., Cruickshank, D., Coruzzi, M., Caira, M.R., Bettini, R., 2010. Enantiotropically related albendazole polymorphs. J. Pharm. Sci. 99, 3731–3742. Salmaso, S., Elvassore, N., Bertucco, A., Caliceti, P., 2009. Production of solid lipid submicron particles for protein delivery using a novel supercritical gas-assisted melting atomization process. J. Pharm. Sci. 98, 640–650. Simpson, T.D., Hagemann, J.W., 1982. Evidence of two b' phases in tristearin. J. Am. Oil Chem. Soc. 59, 169–171. Sutananta, W., Craig, D.Q.M., Newton, J.M., 1994. The effects of ageing on the thermal behaviour and mechanical properties of pharmaceutical glycerides. Int. J. Pharm. 111, 51–62. Takeuchi, M., Ueno, S., Sato, K., 2003. Synchrotron radiation SAXS/WAXS study of polymorph-dependent phase behavior of binary mixtures of saturated monoacid triacylglycerols. Cryst. Growth Des. 3, 369–374. Traul, K.A., Driedger, A., Ingle, D.L., Nakhasi, D., 2000. Review of the toxicologic properties of medium-chain triglycerides. Food Chem. Toxicol. 38, 79–98. Windbergs, M., Strachan, C.J., Kleinebudde, P., 2009. Investigating principles of recrystallization from glyceride melts. AAPS PharmSciTech 10, 1224–1233. Xia, D., Cui, F., Gan, Y., Mu, H., Yang, M., 2014. Design of lipid matrix particles for fenofibrate: effect of polymorphism of glycerol monostearate on drug incorporation and release. J. Pharm. Sci. 103, 697–705. Zhang, Y.E., Schwartz, J.B., 2003. Melt granulation and heat treatment for wax matrix-controlled drug release. Drug Dev. Ind. Pharm. 29, 131–138.
