In vitro release of l -phenylalanine from ordered mesoporous materials

Microporous and Mesoporous Materials 177 (2013) 32–36

Contents lists available at SciVerse ScienceDirect

Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso

In vitro release of L-phenylalanine from ordered mesoporous materials Joanna Goscianska, Anna Olejnik, Robert Pietrzak ⇑ ´ , Umultowska 89b, 61-614 Poznan ´ , Poland Faculty of Chemistry, Adam Mickiewicz University in Poznan

a r t i c l e

i n f o

Article history: Received 9 March 2013 Received in revised form 18 April 2013 Accepted 19 April 2013 Available online 28 April 2013 Keywords: Mesoporous silicas Release studies L-phenylalanine Amino acid Active compounds

a b s t r a c t The applicability of different types of mesoporous materials as pharmaceutical carriers for L-phenylalanine was evaluated. Silicas, such as SBA-15, SBA-16 and KIT-6 were synthesised by hydrothermal method with tetraethyl orthosilicate as the silica source and triblock copolymer P123 as a template. XRD and TEM studies confirmed an ordered hexagonal structure of SBA-15 and a cubic structure of SBA-16 and KIT-6. The materials are characterised by well-developed specific surface areas and large pore volumes. Adsorption of L-phenylalanine over various mesoporous silicas was studied from solution of different pH (5.6– 9.4). The greatest sorption capacity was observed at pH 5.6, which is close to the isoelectric point of L-phenylalanine (pI = 5.48). The amount of L-phenylalanine adsorbed on the mesoporous materials decreases at pH 5.6 in the following sequence: KIT-6 > SBA-15 > SBA-16 that was strongly related to the average pore diameter of the samples. The structural and textural features of the silicas seem to be responsible for the different L-phenylalanine release rate. Additionally, it was found that L-Phe release rate exhibited the pH sensitivity. These phenomena allow control of the experiment according to the needs. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction Mesoporous silicas have attracted the attention of many scientists all over the world as delivery vehicles of biologically active substances [1,2]. Their applications as carriers for various molecules has been under intensive in vitro and recently also in vivo research. Mesoporous materials are interesting to be applied in delivery systems because of their outstanding features such as high surface area, great pore volumes, well–ordered, tunable pores and nontoxicity [3–10]. It has been proved that these materials are capable of carrying high dosages of various drugs in mesopores [11–13]. Several studies have proved that different biologically active molecules can be loaded and successfully released from mesoporous materials [14–17]. It has been suggested that the pore diameter of the carrier should be adapted according to the released molecule size [18]. Therefore, in this study we try to access the applicability of various mesoporous materials for loading and release of L-phenylalanine. L-phenylalanine is the form of amino acid that occurs naturally in proteins in human body [19]. It plays essential role as a precursor in the biosynthesis of L-tyrosine and is crucial in biochemical processes regarding the synthesis of several neurotransmitters (such as L-dopa, dopamine, epinephrine, thyroxine and melanin). Furthermore, it can be converted through several pathways to phenylethylamine that is suggested to elevate mood and have an influence on the synthesis of brain neuropep⇑ Corresponding author. Tel.: +48 61 8291581; fax: +48 61 8291555. E-mail address: [email protected] (R. Pietrzak). 1387-1811/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.micromeso.2013.04.021

tides [19]. Phenylalanine has potential antidepressant and analgesic effects and it seems to be useful in the vitiligo treatment. On the other hand, it is thought that L-phenylalanine can exacerbate symptoms of phenylketonuria [20] and dyskinesia in some schizophrenic patients. Clinical trials have resulted in mixed results and more research is required. In some cases L-phenylalanine is essential for human body, therefore we have developed a carrier for this active compound that enables its slow-rate delivery. To the best of our knowledge, there has been no article regarding the potential use of ordered mesoporous silicas for amino acids release. Furthermore, we have chosen L-phenylalanine as a model for protein release because it has both aromatic hydrophobic region as well as hydrophilic functional groups (Fig. 1). The main aim of this study was to analyse the effect of ordered mesoporous materials such as SBA-15, SBA-16 and KIT-6 on the loading and release of L-phenylalanine. 2. Experimental procedure 2.1. Sample preparation A highly ordered SBA-15 sample was synthesized using the triblock copolymer, EO20PO70EO20 (Pluronic P123, BASF) as template and tetraethyl orthosilicate (TEOS, Aldrich) as the silica source, following the synthesis procedure reported by Zhao et al. [21]. The starting composition was 0.0017 mol of P123: 0.10 mol TEOS: 0.60 mol HCl: 20 mol H2O. In a typical synthesis, 1.1 g TEOS was added dropwise to 19.0 ml of 1.6 M HCl containing 0.5 g of P123

33

J. Goscianska et al. / Microporous and Mesoporous Materials 177 (2013) 32–36

O

Total pore volume and average pore diameter were determined as well. 2.4. Transmission electron microscopy (TEM)

OH

For TEM measurements, powdered samples were deposited on a grid with a perforated carbon film and transferred to a JEOL 2000 electron microscope operating at 80 kV.

NH2

9.7 × 4.7Å Fig. 1. Chemical approximations.

structure

of

phenylalanine

with

its

cylindrical

size

at 35 °C. The mixture was stirred with a magnetic stirrer until TEOS was completely dissolved. Then, the mixture was placed in an oven for 24 h at 35 °C and subsequently for 6 h at 100 °C. The white solid product was filtered without washing and dried at 100 °C for 24 h in air oven. Finally, the product was calcined at 550 °C in air to remove the template. SBA-16 mesoporous silica was prepared exactly as reported recently by Ryoo et al. [22] using triblock copolymer in a ternary copolymer-butanol-water system and low-acid concentration. The aqueous mixture of Pluronic F127 copolymer (EO106PO70EO106, Aldrich) with 1-butanol (Aldrich, 99%) was applied to create a mesotructure to achieve an ordered self-assembly of the silica source tetraethyl orthosilicate (TEOS, Aldrich). Typically, 3 g of the copolymer F127 was dissolved in a solution of 144 g water and 5.94 g of hydrochloric acid (Chempur, 35%). After 30 min 9 g of the co-surfactant 1-butanol was added. After 1 h stirring 14.2 g TEOS was added to the solution. At a constant temperature of 45 °C the mixture was further stirred for 24 h. The mixture was then placed in an oven for 24 h at 100 °C. The molar gel composition of the synthesis mixture was 0.00024 F127: 0.12 C4H10O: 0.068 TEOS: 0.057 HCl: 8 H2O. The white solid product was filtered without washing and dried at 100 °C for 24 h in air oven. At last, the product was calcined at 550 °C in air to remove the template. KIT-6 sample was synthesized as follows: 4.0 g of Pluronic P123 (BASF) was dissolved in 144 g of distilled water and 7.9 g of hydrochloric acid (Chempur, 35%) solution upon stirring at 35 °C [23]. After complete dissolution, 4.0 g of 1-butanol was added immediately. After 1 h stirring, 8.6 g of TEOS was added to the homogeneous clear solution. The mixture was kept under vigorous and continuous stirring at 35 °C for 24 h. Subsequently, the reaction mixture was aged at 100 °C for 24 h under static condition. The molar gel composition of the synthesis mixture was 0.0007 P123: 0.054 C4H10O: 0.041 TEOS: 0.076 HCl: 8 H2O. The product was filtered without washing and dried at 100 °C for 24 h in air oven. Finally, the sample was calcined at 550 °C in air to remove the template.

2.5. L-phenylalanine adsorption L-phenylalanine solution of concentration 30 mmol/l was prepared by dissolving amino acid in potassium phosphate buffer solutions (pH 5.6–9.4). In each adsorption experiment, 0.2 g of a mesoporous silica (SBA-15, SBA-16, KIT-6) was suspended in 50 ml of L-phenylalanine solution. The resulting mixture was continuously stirred in a closed batch at room temperature (48 h). The amount of amino acid adsorbed was calculated by subtracting the amount found in the supernatant liquid after adsorption from the amount of amino acid present before addition of the adsorbent by UV absorption at the kmax of L-phenylalanine, 257 nm.

2.6. L-phenylalanine release experiment The appropriate mesoporous material containing L-phenylalanine was placed in the Enhancer cell. In order to maintain appropriate experiment conditions and steady surface area, the mesoporous materials with adsorbed amino acid were sandwiched between two porous synthetic nets. In vitro release studies were performed with the use of an USP Apparatus 2 (Varian Vankel 7010). Potassium phosphate buffer at pH 5.6, pH 7.2 and pH 9.4 was used as the receiving medium. The medium (200 ml) was maintained at 32.0 ± 0.5 °C and stirred at 100 rpm. The samples were filtered through 35 lm HDPE FullFlow filters and the concentration of the dissolved L-phenylalanine was monitored by UV–Vis spectrophotometer at 257 nm. The absorbance of the sample aliquots was used to assess the amount of compound release at each time point. In addition, the reference standard solutions of L-phenylalanine was prepared in the appropriate receptor fluid in order to generate the standard curve of absorbance versus concentration. 2.7. Kinetics calculations In order to obtain the release rate of L-phenylalanine from mesoporous materials the amount of the substance which was released per time unit was taken into account. The release of amino acid in

SBA-16

SBA-15

KIT-6

The materials obtained were characterised by X-ray diffraction (XRD) using a D8 Advance diffractometer (Bruker) (Cu Ka radiation, k = 0.154 nm), with a step size 0.02° in the low-angle range.

Intensity, a.u.

2.2. Powder X-ray diffraction (XRD)

2.3. Nitrogen sorption Characterization of the pore structure of samples obtained was performed on the basis of low-temperature nitrogen adsorption– desorption isotherms measured on a sorptometer Quantachrome Autosorb iQ. Prior to adsorption measurements, the samples were degassed in vacuum at 300 °C for 2 h. Surface area and pore size distribution were calculated by BET and BJH methods, respectively.

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

0

2 θ,

Fig. 2. Low-angle X-ray diffraction patterns of ordered mesoporous silica.

34

J. Goscianska et al. / Microporous and Mesoporous Materials 177 (2013) 32–36

the first 4 h of the experiment was linear when the amount released per square centimetre was plotted as a function of the square root of time, in accordance with Higuchi’s model [24,25]. On the basis of these plots the release rate was calculated.

3. Results and discussion All obtained mesoporous materials were characterised to establish their structural, textural and morphological properties. Fig. 2 shows the X-ray diffraction patterns at a low-angle range characteristic of the mesostructured silicas with highly ordered arrangement. The profile of SBA-15 indicates three well-developed reflections corresponding to the planes (1 0 0), (1 1 0) and (2 0 0), evidencing the hexagonal structure (space group p6 mm). The XRD pattern of SBA-16 shows a narrow single peak centred at 2H  0.9° corresponding to the plane (1 1 0) which confirms the cubic structure of the space group Im3m. Two low-intensity peaks correspond to the planes (2 0 0) and (2 1 1). The XRD profile of KIT-6 silica reveals a well-developed and highly intense peak at 2H  1° corresponding to the plane (2 1 1) and reflections in the range 2H  1.5–1.9°, testifying to a highly ordered cubic structure of the space group Ia3d. Transmission electron microscopy (TEM) images presented in Fig. 3 confirm that all silica materials synthesised have ordered structures with uniform mesopores. The textural parameters of mesoporous materials presented in Table 1 influence the sorption capacity and the amount of L-phenylalanine released. All silicas obtained have high surface areas and large pore volumes. KIT-6 is characterised by the largest pore volume (0.97 cm3/g) and size (5.83 nm), which can suggest its greatest sorption capacity towards large biomolecules such as amino acids. From among the samples synthesised, the smallest pore volume and size of 0.66 cm3/g and 3.29 nm, respectively, had SBA-16. Fig. 4 shows the nitrogen adsorption isotherms of materials obtained. The isotherms of SBA-15 and KIT-6 are of type IV according to the IUPAC classification and exhibit H1 type hysteresis loops, which are typical of materials with a constant cross-section. SBA16 exhibits a type IV isotherm with hysteresis loop of H2 type, which is difficult for interpretation. It has an adsorption branch of a low slope and a steep desorption branch. It is usually obtained as a consequence of the difference in the mechanisms of condensation and evaporation taking place in pores of narrow necks and wide main parts, often described as ink bottle pores. All ordered mesoporous silicas obtained were tested in the process of L-phenylalanine adsorption and release. It was found that the amount of L-Phe adsorbed was strongly related to the pH value of the adsorbing solution [26]. As reported by Vlasova [27], at pH 8–11, the carboxyl groups of L-phenylalanine tend to be dissociated giving the compound a more negative charge (Fig. 5). At low pH (2– 4) the amino group, as well as the overall molecule, becomes positively charged. At an interim pH known as the isoelectric point,

amino acid in solution has, on average, a net charge of zero. At this point, a solution contains positively and negatively charged L-phenylalanine ions in equal quantities. A solution at the isoelectric point, pI, also has a large concentration of ionic species that are comprised of both charged amino and carboxyl groups. Therefore, as shown in Table 2, the amount of L-Phe adsorbed at pH 9.4 was lower, because of electrostatic interactions between mesoporous materials and amino acid. The maximum sorption capacity of ordered mesoporous materials was at the buffer solution pH of 5.6, which is relatively close to the isoelectric point of L-Phe (pI = 5.48). At this pH the electrostatic interactions between host and guest are negligible, which enables more intensive adsorption of L-Phe than in any other pH range. At pH 5.6, the amount of L-Phe adsorbed on the mesoporous materials (Table 2) decreases in the following sequence: KIT-6 > SBA-15 > SBA-16, which is strongly related to their average pore diameter. Large pore size of KIT-6 and SBA-15 were found to favour the adsorption of the amino acid molecules at pH 5.6. In vitro release of L-phenylalanine from ordered mesoporous materials was carried out for 48 h in a phosphate buffer solution at three different pH values: at pH 5.6 close to the isoelectric point of L-Phe, pH 7.2 close to the physiological pH and pH 9.4. The experimental results of amino acid release are shown in Table 2 and Fig. 6. Moreover the release rate data are presented in the Table 3. A different behaviour between SBA-15, SBA-16 and KIT-6, depending on pH value was detected. Electrostatic, hydrophobic and steric interactions are all likely to be important in amino acid release from ordered mesoporous materials. The fact that at pH 5.6 more than 50% of adsorbed L-phenylalanine was not released is related to the strength of the interactions between amino acid and material surface. In addition, at pH 5.6 the interactions between L-Phe molecules and mesoporous materials are strong and can be responsible for close packing of L-Phe on mesoporous adsorbents. Furthermore, we suggest that L-Phe adsorbed previously at pH 5.6 could be packed inside the pores of the material, therefore amino acid could not be easily released from these materials and the percentage release of L-Phe from mesoporous materials was as follows – 30% from SBA-16, 25.8% from KIT-6 and 25% from SBA-15. The higher release was observed for L-Phe in potassium phosphate buffer at pH 7.2 compared to the results observed at pH 5.6. The reason could be the fact that amino acid was located mostly on the external surface of the ordered mesoporous material, which may have influenced the release process. The same explanation could be applied to the release study carried out at pH 9.4, however at this pH a slow release pattern was observed and we suggest that the experiment should be prolonged. The pore diameter of the material is another important factor affecting the release of the loaded substance. The material with small pore diameter (SBA-16) was characterised by poor adsorption capacity (only 58.9 mg of L-Phe adsorbed), which also

Fig. 3. TEM micrographs of SBA-15 (A), SBA-16 (B) and KIT-6 (C).

35

J. Goscianska et al. / Microporous and Mesoporous Materials 177 (2013) 32–36 Table 1 Textural parameters of the ordered mesoporous silica. Sample

Total surface area [m2/g]

Total pore volume [cm3/g]

Average pore diameter [nm]

SBA-15 SBA-16 KIT-6

790 804 668

0.82 0.66 0.97

5.24 3.29 5.83

Table 2 Data related to L-Phe adsorbed and released from the mesoporous materials (after 48 h).

pH 5.6

Amount of L-Phe adsorbed [mg] Amount of L-Phe released [mg] % of L-Phe released Amount of L-Phe adsorbed [mg] Amount of L-Phe released [mg] % of L-Phe released Amount of L-Phe adsorbed [mg] Amount of L-Phe released [mg] % of L-Phe released

pH 7.2

SBA-16

KIT-6

600

KIT-6

SBA-16

64.2 15.7 25.0 31.2 19.7 63.1 11.8 5.8 49.1

69.3 15.8 25.9 36.3 24.9 68.5 14.3 3.9 27.2

58.9 17.8 30.9 38.5 17.8 46.4 19.7 6.1 33.9

3

-1

Volume adsorbed, cm g (STP)

pH 9.4 SBA-15

SBA-15

Amount of L-Phe released [mg]

(A)

400

200

0 0,0

0,3

0,6

0,9

0,0

0,3

0,6

0,9

0,0

0,3

0,6

0

Relative pressure, p/p 0 Fig. 4. Nitrogen adsorption isotherms of mesoporous materials.

max. 64.2 mg

10

SBA-15 KIT-6 SBA-16

5

10

20

30

40

50

t [h]

(B) 100 SBA-15 KIT-6 SBA-16

80

60

40

20

0 0

10

20

30

40

50

t [h] Fig. 6. Amount (A) and percentage (B) of L-phenylalanine released from SBA-15, KIT-6 and SBA-16 in potassium phosphate buffer at pH 5.6.

4. Conclusions The work presented has been focused on the adsorption and release of L-phenylalanine from three mesoporous materials such as SBA-15, SBA-16, KIT-6 having different surface area, pore volume and pore diameter. All silicas synthesised are characterised by

O OH

Acidic

max. 69.3 mg 15

0

O

NH3

max. 58.9 mg

0

% released

influences the rate of release of this amino acid from this material. Moreover, it has been shown by Horcajada et al. that even small changes in the pore diameter can affect the release rate of active compound from ordered mesoporous materials [28]. This phenomenon was also observed in our studies at pH 5.6, in SBA-16 with the smallest pore diameter, the released amount of L-Phe was higher when compared to two other materials. It can be related to the fact that in this material a smaller amount of L-Phe has been adsorbed, so the process of release of the active substance will be faster. Furthermore, we also suggest that the highest pore diameter of KIT-6, could be responsible for L-Phe crystallisation inside the pores, thus leading to a slower release of this substance, as proposed by Shen [29]. On the other hand, when the experiment was performed at pH 7.2 the greatest amount of the substance was released from KIT-6 of the greatest pore diameter. For this reason, the release of bioactive L-Phe essentially depends on the experimental conditions. The effect of pore diameter on the release of active compound has been analysed in other studies and it has been found that the drug release generally increases with increasing pore diameter [30–32]. Additionally, it should be mentioned that the large number of hydroxyl groups in mesoporous materials could be involved in the formation of hydrogen bonds with L-phenylalanine. On the other hand, these materials could also form hydrogen bonds with molecules of water from the buffer solution. Different adsorption capacity and diverse release rate of L-Phe could be also influenced by various structures of mesoporous materials. Hexagonal structure was observed in SBA-15 and cubic arrangement of mesopores was in both KIT-6 and SBA-16.

20

H3O +

O -

O

OH

NH3

Isoelectric Point Fig. 5. Phenylalanine dissociation equilibrium.

O NH2

Basic

36

J. Goscianska et al. / Microporous and Mesoporous Materials 177 (2013) 32–36

Table 3 Release rate of L-Phe from mesoporous material in buffer solution at different pH (calculated till 4 h of the experiment). Material SBA-15 KIT-6 SBA-16

pH 5.6 k [mg cm 2 h 0.29 ± 0.01 0.45 ± 0.02 0.21 ± 0.01

1/2

]

pH 7.2 k [mg cm 2 h 0.28 ± 0.09 0.12 ± 0.02 0.34 ± 0.08

1/2

]

pH 9.4 k [mg cm 2 h

1/2

]

0.32 ± 0.01 0.18 ± 0.02 0.27 ± 0.02

highly ordered mesoporous structure. Biologically active molecules i.e. amino acids can be loaded and successfully released from these materials. Small changes in the pore size of the materials influenced the adsorption and release of L-phenylalanine. A very significant parameter that affects both processes is the pH of the buffer solution applied. The mesoporous silica materials studied show the largest sorption capacity at pH 5.6, close to the isoelectric point of L-phenylalanine (pI = 5.48). Above this pH value, the amount of adsorbed amino acid decreases. At pH 5.6 the electrostatic interactions between host and guest are negligible, which permits more intensive adsorption of L-phenylalanine than in any other pH range. In release experiment, at pH 5.6 interactions between L-Phe molecules and mesoporus materials are high and can be responsible for close packing of L-Phe on mesoporous adsorbents. We suggest that the amino acid adsorbed at pH 5.6 could be packed inside the pores of the material, therefore the amino acid could not be easily released from these materials and the percentage release of L-Phe from mesoporous materials was as follows – 30% from SBA-16, 25.8% from KIT-6 and 25% from SBA-15. L-Phe adsorption and release rate were sensitive to pH. This phenomena allows to control the experiment. The amount of this amino acid adsorbed on the mesoporous materials and released from them is strongly related to the average pore diameter of the materials and the type of their structure. References [1] M. Vallet-Regí, A. Rámila, R.P. del Real, J. Pérez-Pariente, Chem. Mater. 13 (2001) 308–311. [2] A.B. Foraker, R.J. Walczak, M.H. Cohen, T.A. Boiarski, C.F. Grove, P.W. Swaan, Pharm. Res. 20 (2003) 110–116.

[3] Z.G. Feng, Y.S. Li, D.C. Niu, L. Li, W. Zhao, H. Chen, L. Li, J. Gao, M. Ruan, J. Shi, Chem. Commun. 23 (2008) 2629–2631. [4] A. Sayari, S. Hamoudi, Chem. Mater. 13 (2001) 3151–3168. [5] D. Brunel, A.C. Blanc, A. Galarneau, F. Fajula, Catal. Today 73 (2002) 139–152. [6] J. Liu, X.D. Feng, G.E. Fryxell, L.Q. Wang, A.Y. Kim, M.L. Gong, Adv. Mater. 10 (1998) 161–165. [7] A. Stein, B.J. Melde, R.C. Schroden, Adv. Mater. 12 (2000) 1403–1419. [8] R. Mellaerts, J.A.G. Jammaer, M. Van Speybroeck, H. Chen, J. Van Humbeeck, P. Augustijns, G. Vanden Mooter, J.A. Martens, Langmuir 24 (2008) 8651–8659. [9] S.P. Hudson, R.F. Padera, R. Langer, D.S. Kohane, Biomaterials 29 (2008) 4045– 4055. [10] S.R. Blumen, K. Cheng, M.E. Ramos-Nino, D.J. Taatjes, D.J. Weiss, C.C. Landry, B.T. Mossman, Am. J. Respir. Cell Mol. 36 (2007) 333–342. [11] M. Vallet-Regi, J.C. Doadrio, A.L. Doadrio, I. Izquierdo-Barba, J. Perez Pariente, Solid State Ionics 172 (2004) 435–439. [12] Y.F. Zhu, J.L. Shi, Y.S. Li, H.R. Chen, W.H. Shen, X.P. Dong, J. Mater. Res. 20 (2005) 54–61. [13] V. Ambrogi, L. Perioli, F. Marmottini, S. Giovagnoli, M. Esposito, C. Rossi, Eur. J. Pharm. Biopharm. 32 (2007) 216–222. [14] C. Charnay, S. Bégu, C. Tourné-Péteilh, L. Nicole, D.A. Lerner, J.M. Eur, J. Pharm. Biopharm. 57 (2004) 533–540. [15] M. Van Speybroeck, V. Barillaro, T.D. Thi, R. Mellaerts, J. Martens, J. Van Humbeeck, J. Vermant, P. Annaert, G. Van den Mooter, P. Augustijns, J. Pharm. Sci. 98 (2009) 2648–2658. [16] M.J.K. Thomas, I. Slipper, A. Walunj, A. Jain, M.E. Favretto, P. Kallinteri, D. Douroumis, Int. J. Pharm. 387 (2010) 272–277. [17] Y. Zhang, Z. Zhi, T. Jiang, J. Zhang, Z. Wang, S. Wang, J. Controlled Release 145 (2010) 257–263. [18] I. Izquierdo-Barba, A. Martinez, A.L. Doadrio, J. Pérez-Pariente, M. Vallet-Regí, Eur. J. Pharm. Sci. 26 (2005) 365–373. [19] M.B. Stargrove, J. Treasure, L.D. McKee, Herb, Nutrient, and Drug Interactions: Clinical Implications and Therapeutic Strategies, Elsevier, Missouri, 2008. [20] H.M. Clark, C.C.C. Alves, A.S. Franca, L.S. Oliveira, LWT –Food Sci. Technol. 49 (2012) 155–161. [21] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrikson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548–552. [22] F. Kleitz, L.A. Solvyov, G.M. Anilkumar, S.H. Choi, R. Ryoo, Chem. Commun. 13 (2004) 1536–1537. [23] T.W. Kim, F. Kleitz, B. Paul, R. Ryoo, J. Am. Chem. Soc. 127 (2005) 7601–7610. [24] T. Higuchi, J. Pharm. Sci. 50 (1961) 874–875. [25] T. Higuchi, J. Pharm. Sci. 51 (1962) 802–804. [26] J. Goscianska, A. Olejnik, R. Pietrzak, Adsorption 19 (2013) 581–588. [27] N.N. Vlasova, L.P. Golovkova, Colloid J. 66 (2004) 657–662. [28] P. Horcajada, A. Rámila, J. Pérez-Pariente, M. Vallet-Regi, Microporous Mesoporous Mater. 68 (2004) 105–109. [29] S.C. Shen, W.K. Ng, L. Chia, J. Hu, R.B.H. Tan, Int. J. Pharm. 410 (2011) 188–195. [30] J. Andersson, J. Rosenholm, S. Areva, M. Linden, Chem. Mater. 16 (2004) 4160– 4167. [31] F. Qu, G. Zhu, S. Huang, S. Li, J. Sun, D. Zhang, S. Qiu, Microporous Mesoporous Mater. 92 (2006) 1–9. [32] V. Cauda, L. Mühlstein, B. Onida, T. Bein, Microporous Mesoporous Mater. 118 (2009) 435–442.