Synthesis, Characterization, and in Vitro pH-Controllable Drug

Oct 12, 2010 - To accomplish pH-controllable drug release on mesoporous carrier, one of the ..... International Journal of Polymeric Materials and Pol...
0 downloads 0 Views 2MB Size
pubs.acs.org/Langmuir © 2010 American Chemical Society

Synthesis, Characterization, and in Vitro pH-Controllable Drug Release from Mesoporous Silica Spheres with Switchable Gates Qiang Gao,†,‡ Yao Xu,*,† Dong Wu,† Wanling Shen,§ and Feng Deng§ †

State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, P. R. China, ‡Department of Faculty of Material Science & Chemistry Engineering, China University of Geosciences, Wuhan 430074, P. R. China, and §State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071, P. R. China Received July 25, 2010. Revised Manuscript Received September 16, 2010 To accomplish pH-controllable drug release on mesoporous carrier, one of the best ways is to graft stimuli-responsive organic molecules around mesopore outlets. In this work, the pH-responsive propyldiethylenetriamine groups (abbreviative phrase: multiamine chains) were grafted around mesopore outlets of mesoporous silica spheres (MSS) and expected to act as pH-responsive gates. To this end, three multiamine-grafted MSS (i.e., NM1, NM2, and NM3) were synthesized under different reaction temperatures and reaction times. The reaction temperature and time for multiamine grafting were 25 °C and 12 h for NM1, 100 °C and 1 h for NM2, and 100 °C and 12 h for NM3, respectively. Through systematic investigations of TEM, SEM, N2 adsorption/desorption, TG, and 29Si MAS NMR, it was found that NM3 had the highest grafting amount of multiamine chains. It was further confirmed that the multiamine chains around the pore outlets of NM3 played the role of “molecular switch” that could well control the transport of guest drug molecules. In contrast, the multiamine chains around the pore outlets of NM2 and NM3 did not show gate effect. The difference should be decided by the fact whether the grafting amount of multiamine chains around mesopore outlets were sufficient under determined reaction temperature and time. In the tests of in vitro drug release, multiamine-gated MSS (i.e., NM3) showed highly sensitive response to the solution pH. At high pH (pH 7.5), ibuprofen (IBU) in this carrier released rapidly and completely within 2 h; at low pH (pH 4.0 or 5.0), only a small part of the IBU (13 wt %) was slowly released from this carrier and the most of IBU was effectively confined in mesopores.

1. Introduction Since the discovery of M41S mesoporous silica by Mobil scientists in 1992,1 these kinds of nanoporous materials have shown great potential for the applications in catalysis,2 adsorption,3 separation,4 and sensor.5 Recently, the application of mesoporous silica has been extended to the field of drug delivery.6-8 As a new drug carrier, mesoporous silica has the following remarkable characteristics:9,10 (1) high surface area and large pore volume create preconditions for large drug loding capacity; (2) highly ordered mesopores provide a possibility for drugs to achieve uniform distribution; (3) easy surface modification and adjustable pore diameter can easliy delay the drug release; and (4) its nontoxicity and biocompatibility meet medical requirements. Therefore, as a novel drug carrier mesoporous silica has attracted numerous studies.1-10 However, normal mesoporous silica (e.g., MCM-41 and SBA-15) are unintelligent materials for the controllable release that precisely matches the actual physiological needs at proper time and/or proper site.11 *Corresponding author. Tel: þ86-351-4049859. Fax: þ86-351-4041153. E-mail: [email protected]. (1) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Nature 1992, 359, 710. (2) Martin-Aranda, R. M.; Cejka, J. Top. Catal. 2010, 53, 141. (3) Gao, Q.; Xu, W. J.; Xu, Y.; et al. J. Phys. Chem. B 2008, 112, 2261. (4) Bao, X. Y.; Zhao, X. Y. Langmuir 2009, 25, 1807. (5) Asefa, T.; Duncan, C. T.; Sharma, K. K. Analyst 2009, 134, 1980. (6) Vallet-Regi, M.; Ramila, A.; del Real, R. P.; Perez-Pariente, J. Chem. Mater. 2001, 13, 308. (7) Vallet-Regi, M.; Balas, F.; Arcos, D. Angew. Chem., Int. Ed. 2007, 46, 7548. (8) Wang, S. B. Mciroporous Mesoporous Mater. 2009, 117, 1. (9) Wan, Y.; Zhao, D. Y. Chem. Rev. 2007, 107, 2821. (10) Vallet-Regi, M.; Gonzalez, L. R.; et al. J. Mater. Chem. 2006, 16, 26. (11) Qiu, Y.; Park, K. Adv. Drug Delivery Rev. 2001, 53, 321.

Langmuir 2010, 26(22), 17133–17138

To overcome the shortcoming, one strategy is to graft stimuliresponsive organic molecules around the mesopore outlets. In the carrier system, mesoprous silica is used as drug containers and the grafted organic molecules are employed as switchable gates. The “open” or “closed” state of the gates can be controlled by specific stimuli signals (e.g., pH, temperature or light) and determine whether the drug should be released from the mesopores or not. To date, several research groups have taken great efforts to explore such drug carrier systems.12-26 Fujiwara and co-workers15-17 grafted coumarin molecules around the mesopore outlets, and realized a light-controllable release of guest molecules from mesoporous hosts at the first time. Very recently, Zink et al.19-24 explored a series of smart mesoporous carriers with switchable (12) Aznar, E.; Martinez-Manez, R.; Sancenon, F. Expert Opin. Drug Delivery 2009, 6, 643. (13) Aznar, E.; Coll, C.; Marcos, M. D.; et al. Chem.;Eur. J. 2009, 15, 6877. (14) Slowing, I. I.; Vivero-Escoto, J. L.; Wu, C. W.; et al. Adv. Drug Delivery Rev. 2008, 60, 1278. (15) Mal, N. K.; Fujiwara, M.; Tanaka, Y. Nature 2003, 421, 350. (16) Mal, N. K.; Fujiwara, M.; Tanaka, Y.; et al. Chem. Mater. 2003, 15, 3385. (17) Fujiwara, M.; Terashima, S.; et al. Chem. Commun. 2006, 4635. (18) Casasus, R.; Marcos, M. D.; Manez, R. M.; et al. J. Am. Chem. Soc. 2004, 126, 8612. (19) Hernandez, R.; Tseng, H. R.; Wong, J. W.; et al. J. Am. Chem. Soc. 2004, 126, 3370. (20) Nguyen, T. D.; Tseng, H. R.; Celestre, P. C.; et al. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 10029. (21) Nguyen, T. D.; Liu, Y.; Saha, S.; et al. J. Am. Chem. Soc. 2007, 129, 626. (22) Park, C.; Oh, K.; et al. Angew. Chem., Int. Ed. 2007, 46, 1455. (23) Leung, K. C. F.; Nguyen, T. D.; Stoddart, J. F.; Zink, J. I. Chem. Mater. 2006, 18, 5919. (24) Patel, K.; Angelos, S.; Dichtel, W. R.; et al. J. Am. Chem. Soc. 2008, 130, 2382. (25) Giri, S.; Trewyn, B. G.; et al. Angew. Chem., Int. Ed. 2005, 44, 5038. (26) Lai, C. Y.; Trewyn, B. G.; et al. J. Am. Chem. Soc. 2003, 125, 4451.

Published on Web 10/12/2010

DOI: 10.1021/la102952n

17133

Article

Gao et al.

Figure 1. Schematic illustration of the synthesis strategy of multiamine-gated MSS and its working principle as pH-responsive drug carrier.

supermolecular nanovalves around the pore outlets of mesoporous silica, and investigated the stimuli-controllable release behaviors. All these pioneering works showed that the key of this graft approach on pore outlets was to select suitable organic molecules as “switchable gates”. As revealed by Martinez-Manez,27 the propyldiethylenetriamine groups (multiamine chains) are highly sensitive to the solution pH. At high pH, multiamine chains come close to each other through the hydrogen bonding interaction; at low pH, multiamine chains will be protonated and tend to stay as far away from each other as possible. Therefore, it is possible that the multiamine chains grafted on mesopore outlets endow the mesoporous silica carrier with a capability of ‘molecular switch’.27 Martinez-Manez’s findings are of great significance, nevertheless, no further attempts are made for the application of multiamine chains in drug delivery. On the other hand, it is documented that mesoporous silica spheres (MSS) is a potential candidate of drug carrier because of its perfect geometrical symmetry and thus uniform physicochemical properties.28-30 A highly effective pHresponsive carrier system may be obtained because polyamnie chains around the pore outlets of MSS can construct excellent ‘gates’ for drug molecules. Inspired by the advantages of MSS and multiamine chains, in this work we grafted multiamine chains on MSS particles to prepare above-mentioned pH-responsive drug carrier system. The synthesis strategy and possible working principle of this system are described in Figure 1. The multiamine-gated MSS may be an ideal carrier for targeting drug delivery through a pH-responsive mechanism. At low pH (e.g., in the stomach), the protonated multiamine chains block the mesopore outlets, and thus the drug release is minimized. At high pH (e.g., in the intestine), the neutral multiamine chains attract each other through hydrogen bonding and open the pathway for drug release (see Figure 1).

2. Experimental Section 2.1. Preparation of Multiamine-Gated MSS. The preparation of multiamine-gated MSS could be roughly divided into (27) Casasus, R.; Marcos, M. D.; Martinez-Manez, R.; et al. J. Am. Chem. Soc. 2004, 126, 8612. (28) Xu, W. J.; Gao, Q.; Xu, Y.; et al. Powder Technol. 2009, 191, 13. (29) Manzano, M.; Alina, V.; Arean, C. O.; et al. Chem. Eng. J. 2008, 137, 30. (30) Collilla, M.; Manzano, M.; Lzquierdo-Barba, I.; Vallet-Regi, M.; et al. Chem. Mater. 2010, 22, 1821.

17134 DOI: 10.1021/la102952n

two steps. At the first step, MSS was synthesized via a modified St€ ober method.28 The structure directing agent and the silica source are hexadecyltrimethylammonium bromide (CTAB) and tetraethyl orthosilicate (TEOS), respectively. Typically, 3 g of CTAB was dissolved in 828 mL of deionized water and 1438 mL of absolute ethanol at room temperature. Then 72 mL of aqueous ammonia solution (25 wt %) was added to this clear solution and stirred for 15 min. After addition of 10 mL of TEOS, the resultant mixture was stirred for 2 h to form a suspension. The solid products were collected by filtration, washed several times with H2O/C2H5OH, and finally dried at 70 °C. After it dried completely, the solid product containing templates was labeled as MSS-T. At the second step, the multiamine groups were grafted on MSS-T through a post-treatment method. Briefly, 1 g of MSS-T solid was dried in a vacuum environment and then added into 50 mL of absolute toluene with magnetic stirring. After mixing, 4 mL of (trimethoxysilyl)propyldiethylenetriamine (denoted as multiamine-TMS) was added into the above suspension. Then, the grafting reaction proceeded with different reaction temperature and reaction time. These detailed grafting conditions can be found in Table 1. The solid products were filtrated, washed by toluene, dried at 70 °C, and finally denoted as NMx-T, where x = 1, 2, or 3 and refers to the different grafting condition of multiamine. 2.2. Removal of Template. The employed methods for the template removal of NMx-T are summarized in Table 1. The templates inside MSS-T, NM1-T, and NM2-T can be removed by ion-exchange with NH4NO3/C2H5OH solution or NaCl/ (CH3OH þ H2O) solution. The solution of NH4NO3/C2H5OH (pH 4.2) was prepared by dissolving 0.3 g of NH4NO3 into 150 mL of ethanol. The solution of NaCl/(CH3OH þ H2O) (pH 6.9) was prepared by dissolving 3.28 g of NaCl into 280 mL of CH3OH and 56 mL of H2O. Typically, 1 g of dried MSS-T powder was dispersed into the above NH4NO3/C2H5OH solution (or NaCl/ (CH3OH þ H2O) solution), and the mixture was stirred at 70 °C for 1 h. The solid product was recovered by filtration and dried at 70 °C for 1 h. The above treatment was repeated twice. The finally obtained solid product was further dried at 70 °C for 24 h and denoted as MSS-R. Similar to MSS-R, the other two products with removal of templates were denoted as NM1-R and NM2-R. The templates inside NM3-T was removed by ion-exchange with NaCl/(CH3OH þ H2O), and the resultant product was denoted as NM3-R. To primarily test the pH-sensitivity of multiamine gates, NM3T was ion-exchanged three times by NH4NO3/C2H5OH solution, and the resultant product was denoted as NM3-N. It is noteworthy that the NH4NO3/C2H5OH solution is acidic (low pH), Langmuir 2010, 26(22), 17133–17138

Gao et al.

Article

Table 1. Multiamine-Grafting Conditions, Employed Methods for Template Removal, Texture Parameters, and IBU-Loaded Amounts of Samples IBU-loaded amount sample MSS-T MSS-R NM1-T NM1-R NM2-T NM2-R NM3-T NM3-R NM3-N

grafting condition

25 °C, 12 h 100 °C, 1 h 100 °C, 12 h

2

template removal method

NH4NO3/C2H5OH or NaCl/(CH3OHþH2O) NH4NO3/C2H5OH or NaCl/(CH3OHþH2O) NaCl/(CH3OHþH2O) NH4NO3/C2H5OH

but NaCl/(CH3OH þ H2O) solution is neutral (relatively high pH). Therefore, the removal of templates (i.e., CTAB) from NM3-T should be limited by the protonated multiamine groups in NH4NO3/C2H5OH solution rather than in NaCl/(CH3OH þ H2O). If it is that case, the surface area and pore volume of NM3-N should be less than that of NM3-R. 2.3. Drug Storage. IBU was dissolved in hexane at a concentration of 33 mg/mL. Then, 1 g of solid adsorbent was dispersed in 30 mL of this solution at room temperature. After stirring for 72 h, IBU-loaded solid sorbents was separated by centrifugation and dried at 50 °C. To measure the loading amount of IBU, 2.0 mL of filtrate was diluted to 50 mL and then analyzed using UV-vis spectroscopy at 264.5 nm wavelength. The amount of IBU loaded in mesoporous adsorbent was calculated by the following equation: ð1Þ

where m1 and m2 correspond respectively to the initial mass of IBU and mesoporous adsorbent added into hexane solution, C is IBU concentration of the solution prepared by diluting 2 mL of filtrates to 50 mL in volumetric flask, and V is the volume of hexane solution for drug loading. 2.4. In Vitro Drug Release. In our work, each in vitro drug release was done in triplicate. The release fluid with pH 4.0 was prepared by dissolving 4.425 g of KH2PO4 and 32.25 g of citric acid monohydrate into 2500 mL of water. The release fluid with pH 5.0 was prepared by dissolving 12.9 mL of acetic acid and 60.0 g of anhydrous sodium acetate into 2500 mL of water. The release fluid with pH 7.5 was prepared by dissolving 20.25 g of KH2PO4 and 4.53 g of NaOH into 2500 mL of water. In our experiments, all used water is deionized. IBU-loaded solid sorbents was immersed in release fluids with different pH values (i.e., pH 4.0, 5.0, or 7.5) at 37 °C with stirring at a rate of 100 rpm. The release fluid (2.0 mL) was withdrawn at a give time interval, and supplied with the same volume of fresh release fluid. The drug concentration in the fluid was measured with UV-vis spectroscopy at a wavelength of 264 nm. Calculation of the corrected concentration of released IBU is based on the following equation: t-1 vX Cc ¼ Ct þ Ct ð2Þ V 0 where Cc is the corrected concentration at time t, Ct is the apparent concentration at time t, v is the volume of sample taken, and V is the total volume of release fluid. 2.5. Instrumentation. The powder X-ray diffraction (XRD) patterns were recorded on a Bruker diffractometer using Cu KR radiation. N2 adsorption-desorption isotherm was obtained on a Micromeritics Tristar 3000 pore analyzer at 77 K. Transmission electronic microscopes (TEM) of samples were observed on Hitachi H-600. Field emission scanning electronic microscopes (FE-SEM) were obtained on LEO 1530VP. 29Si MAS NMR experiments were performed on a Varian Infinityplus-300 spectrometer using 7.5 mm probe under magic-angle spinning: the resonance frequency was 79.5 MHz, the 90° pulse width was Langmuir 2010, 26(22), 17133–17138

3

-1

surface area (m g ) pore volume (cm g )

NH4NO3/C2H5OH or NaCl/(CH3OHþH2O)

wt% ¼ ðm1 - ð50=2ÞCVÞ=ðm2 þ ðm1 - ð50=2ÞCVÞÞ

-1

42 710 43 640 44 580 46 493 46

0.046 0.396 0.038 0.351 0.041 0.311 0.040 0.238 0.041

wt %

mg m-2

17.4

0.297

16.7

0.313

16.0

0.328

15.5

0.372

Figure 2. XRD patterns of (a) MSS-R, (b) NM1-R, (c) NM2-R, and (d) NM3-R. measured to be 4.8 μs, and a repetition time of 100 s for singlepulse experiments was used. The UV-vis absorbance spectra were measured with a Shimadzu UV-3150 spectrophotometer. Thermogravimetric analysis (TGA) was carried out on a PerkinElmer TGA-7 thermal analyzer. The samples were heated from 25 to 1000 °C at 10 °C/min under air.

3. Results and Discussion 3.1. Structural Characteristics of Samples. XRD patterns of MSS-R, NM1-R, NM2-R, and NM3-R are shown in Figure 2. An obvious peak (100) can be observed for all these samples, indicating their ordered mesoporous structures.28 Compared with MSS-R, the (100) peaks of NM1-R, NM2-R, and NM3-R samples appear at the relatively higher angle positions and the order was NM3-R > NM2-R > NM1-R, implying that some multiamine chains have been implanted into mesopore channels of NMx-R, and the increase of reaction temperature and/or reaction time can benefit this implantation. In theory, these multiamine chains implanted into mesopores should be divided into two parts with one part around pore outlets and the other part onto inner pore wall. We will discuss later which part was dominant. Figure 3 shows the TEM and SEM images of the four samples. It can be found that all of these samples exhibit monodisperse, uniform, and spherical morphologies with an average diameter about 660 nm. As above-mentioned, MSS with an ideal sphere structure should be beneficial for the uniform distribution of multiamine chains around pore outlets. In addition, it should be pointed out that the length of multiamine chain can be ignored as compared with the diameter of MSS, so the four MSS-type samples have almost equal particle size. N2 adsorption/desorption isotherms and BJH pore size distributions of MSS-R, NM1-R, NM2-R, and NM3-R are shown in Figure 4. It can be found that MSS-R, NM1-R, and NM2-R possess typical type IV isotherms and H1 hysteresis loops that are the characteristics of mesoporous solid (see Figure 4a-c). Moreover, it can be observed that the pore sizes of these three samples DOI: 10.1021/la102952n

17135

Article

Figure 3. TEM images of (a) MSS-R, (b) NM1-R, (c) NM2-R, and (d) NM3-R. SEM images of (e) MSS-R and (f) NM3-R.

are 2.6 nm, 2.5 and 2.3 nm, respectively, and all these distributions are very narrow. In contrast, NM3-R shows an approximate Type I isotherm and a pore size of about 2.0 nm (see Figure 4d). This finding demonstrates again that the high reaction temperature (100 °C) and long reaction time (12 h) can promote the implantation of multiamine chains into mesopore channels. 29 Si MAS NMR is employed to learn the information about the graft degree of multiamine chains on NM1-R, NM2-R, and NM3-R. The 29Si MAS NMR spectra of the tested samples are depicted in Figure 5. Table 2 summarizes these observed resonances in 29Si MAS NMR spectra, and the corresponding relative peak areas. The resonance peaks at -110, -100.6, and -91.1 ppm can be attributed to Q4, Q3, and Q2 [Qn =Si(OSi)n(OH)4-n, n= 2∼4] silicon species. In addition, the broad peak approximately at -62.7 ppm should be assigned to T [T = multiamine-Si(OSi)3] species. By calculating the values of T/(Tþ2Q2þQ3), the percentages of multiamine groups in total surface groups can be determined as 4.1%, 7.2%, and 15% for NM1-R, NM2-R, and NM3R. On the other hand, the surface areas of MSS-T and MSS-R are 42 and 710 m2/g (see Table 1), which correspond to the external surface area (Se) and the total surface area (St) of MSS, respectively. By calculating the values of Se/St, the percentage of external surface groups in total surface groups can be determined as 5.9% for MSS (also for NMx, x = 1, 2, and 3). Through comparing these calculated results, it can be also concluded that a considerable part of multiamine chains have been implanted into pores of NM3-R. As some studies demonstrated,15,16 this part of multiamine chains should be absolutely located at mesopore outlets where a few templates leach out during the multiaminegraft reaction. Multiamine chains would not be grafted on pore wall far away from pore outlets because the dense filling of templates in the mesopores depress the permeation of multiamine-TMS.15,16 Therefore, the multiamine chains on NM3-T are distributed around outlets of mesopores as well as the external surface. These multiamine chains around mesopore outlets are expected to act as the pH-responsive gates. In the cases of NM1-R and NM2-R, the amounts of multiamine group (4.1% and 7.2%, respectively) are close to that of external surface groups (5.9%). Are some multiamine chains also grafted around mesopore outlets of NM1-R and NM2-R? According to the data of pore size distributions (see Figure 4b,c), it is convincing that some multiamine chains have been grafted around outlets of NM1-R and NM2-R samples, despite the corresponding graft amounts should be lower than that of NM3-R. These differences in graft amounts around outlets should originate from the graft reaction conditions, and might affect the pH-controllable performance significantly. TGA curves of the materials under study are shown in Figure 6. It can be seen that MSS exhibits a major weight loss at a temperature up to ca. 100 °C, which can be attributed to the 17136 DOI: 10.1021/la102952n

Gao et al.

desorption of physically adsorbed water.31 An additional weight loss occurs at higher temperatures due to the further condensation of the silicate mesopore walls.32 In the case of multiamine-grafted samples, the TGA curves exhibit a rapid decline in the temperature range from ca. 200 to 650 °C, which should be ascribed to the removal of multiamine groups, through which the existence of multiamine groups on the surface of NM1-R, NM2-R, and NMR-3 can be further confirmed .Moreover, the largest weight loss of NM3-R demonstrates that the grafting amount of multiamine groups on NM3-R is much higher than those of NM2-R and NM1-R (see Figure 6). 3.2. Removal of Templates and pH-Sensitivity of Multiamine Chains. NH4NO3/C2H5OH ion-exchange is a common approach for the removal of templates within mesoporous silica. Table 1 summarizes the texture parameters of MSS, NM1, NM2, and NM3 before and after NH4NO3/C2H5OH ion-exchanges. It can be found that through NH4NO3/C2H5OH ion-exchange the surface area and pore volume of MSS increase from 42 and 0.046 cm3/g up to 710 and 0.396 cm3/g, respectively, indicating the templates have been effectively removed. This is consistent with the reported results.33 Similarly, the surface areas and pore volumes of NM1 and NM2 increase significantly after NH4NO3/ C2H5OH ion-exchange, also showing the validity of this method (see Table 1). Moreover, it is found that there are almost no differences between NH4NO3/C2H5OH and NaCl/(CH3OH þ H2O) methods used for the template removal of MSS, NM1, and NM2. The interesting result is that, after NH4NO3/C2H5OH ionexchange, the surface area and the pore volume of NM3 are almost unchanged (see Table 1). The reason should be that multiamine chains around the pore outlets have played a switch role efficiently. Because multiamine is stronger alkaline than ammonia, Hþ from NH4þ would readily pass to multiamine chains. This causes the multiamine chains to be positively charged, and thus multiamine chains act as “gates” and are in a “closed” state. However, why are the multiamine chains of NM1 and NM2 also protonated but they do not act as the role of gate? The possible reason might be that the graft amount of multiamine chains around the pore outlets of the two samples is too low to achieve the gate effect. In contrast, the ion-exchange with NaCl/(CH3OH þ H2O) can remove the template of NM3-T efficiently. As shown in Table 1, the surface area and pore volume of NM3-R reach 493 and 0.238 cm3/g, respectively, after ion-exchange three times. This should be explained by the reason that the multiamine gates are in an “open” state in relatively high pH environment and thus allow the leaching out of templates. In addition, we find that further ion-exchange does not bring the increase of surface areas and pore volumes of NM3 (not shown), indicating that the templates of NM3 have been removed completely through three time ionexchange. Through comparing the behaviors of multiamine chains in acidic NH4NO3/C2H5OH solution and neutral NaCl/ (CH3OH þ H2O) solution, we can primarily infer that multiamine chains have played the role of the “molecular switch” that can control the transport of molecules via a pH-dependent “openclose” mechanism. 3.3. Drug Loading. Table 1 also lists the drug storage data on four mesoporous carriers. It can be found that the IBU adsorption capacity increases with the surface area of samples. MSS-R has the largest surface area of 710 m2/g, and thus the adsorption amount of IBU reachs 17.4 wt %. NM3-R has the minimum (31) Jaroniec, C. P.; Kruk, M.; Jaroniec, M. J. Phys. Chem. B 1998, 102, 5503. (32) Lim, M. H.; Stein, A. Chem. Mater. 1999, 11, 3285. (33) Gao, Q.; Xu, Y.; Wu, D.; Li, X. A. J. Phys. Chem. C 2009, 113, 12753.

Langmuir 2010, 26(22), 17133–17138

Gao et al.

Article

Figure 4. Nitrogen adsorption/desorption isotherms and pore size distributions of samples: (a) MSS-R, (b) NM1-R, (c) NM2-R, and (d) NM3-R.

Figure 5.

29

Si MAS NMR spectra of NM1-R, NM2-R, and

NM3-R. Table 2. Relative Peak Area (%) in 29Si MAS NMR Spectra of Multiamine-Grafted Samplesa 29

Si MAS NMR, relative peak area (%)

samples

Q2

Q3

Q4

T

T/(T þ 2Q2 þ Q3)

NM1-R 1.84% 43.1% 57.1% 2.0% 4.1% NM2-R 2.65% 39.8% 58.3% 3.5% 7.2% NM3-R 2.70% 33.2% 56.9% 7.2% 15% a 2 Q represents the silicon species of (OH)2Si(-OSi-)2; Q3 represents the silicon species of (OH)Si(-OSi-)3; Q4 represents the silicon species of Si(-OSi-)4; T represents the silicon species of multiamineSi(-OSi-)3.

surface area of 493 m2/g, and its IBU adsorption amount is reduced to 15.5 wt %. With moderate surface areas of 640 and 580 m2/g, NM1-R and NM2-R also have moderate IBU adsorption capacities of 16.7 and 16.0 wt %, respectively. However, if the adsorption capacity per unit mass is converted into the corresponding adsorption capacity per unit area, it can be found that the adsorption capacity of IBU increase with the grafting amount of multiamine chains. In detail, the samples MSS-R, NM1-R, NM2-R, and NM3-R have the adsorption capacities of 0.297, 0.313, 0.328, and 0.372 mg/m2, respectively (see Table 1). The result should be reasonable because the increasing amount of Langmuir 2010, 26(22), 17133–17138

Figure 6. TGA curves of MSS-R, NM1-R, NM2-R, and NM3-R.

multiamine chains on sample provides more surface basic sites to enhance the adsorption of the acidic drug IBU. 3.4. In Vitro Drug Release. Figure 7 shows the cumulative IBU release curves from MSS-R, NM1-R, NM2-R, and NM3-R in different release fluids with different pH values (pH 4.0, 5.0 and 7.5). For the MSS-R, it can be found that IBU release is very fast in the release fluid of pH 7.5, and the release amount reachs about 85% within 1 h and almost 100% within 2 h (see Figure 7a). In the case of pH 5.0 or 4.0, the release of IBU from MSS-R become slow slightly (see Figure 7a). According to our previous study, this relatively slow release should be caused by relatively low solubility of IBU in acidic solution.33 For NM1-R and NM2-R, at pH 7.5 IBU releases are almost the same as in the case of MSS-R (see Figure 7b,c). Considering the two samples possessed multiamine chains (i.e., basic adsorption sites), it seems to be that both the releases should be slower than that of MSS-R, and IBU release from NM1-R is slower than NM2-R. The actual result might be explained by the common “leveling effect”;34 that is, the difference between different individuals can be eliminated under specified condition. Specifically, IBU can quickly dissolve into (34) Ji, H. F.; Xu, H.; Xu, X. H. Chem. Phys. Lett. 2001, 343, 325.

DOI: 10.1021/la102952n

17137

Article

Gao et al.

Figure 7. Cumulative release of IBU from (a) MSS-R, (b) NM1-R, (c) NM2-R, and (d) NM3-R.

basic solution (pH 7.5) owing to the strong affinity of basic solution with IBU, so multiamine chains and Si-OH groups become indistinguishable in adsorption force for retention of IBU. At low pH, the IBU release rates from MSS-R, NM1-R, and NM2-R show a trend of gradual slowdown. For example, at pH 5.0 for 8 h the cumulative release amounts of IBU from MSS-R, NM1-R, and NM2-R are 72%, 66%, and 59%, respectively (see Figure 7a-c). This result indicates that in acidic solution the graft of multiamine chains shows a “distinguishing effect” on drug release. The grafted multiamine chains are helpful to delay the release of IBU to a certain degree. Nevertheless, as previously mentioned, NM2-R and NM3-R do not have the performance of “molecular switch”. Figure 7d shows the cumulative IBU release curve from NM3R. At pH 7.5, the release profile is nearly same as that of MSS-R. Therefore, it can be concluded that the multiamine chains are in the “open” state at pH 7.5 and do not hinder the transportation of IBU. However, in the case of pH 5.0 or 4.0, the release of IBU from NM3-R is evidently slow (see Figure 7d), and both release amounts are less than 13% even after 24 h. Obviously, the multiamine chains have played the role of pore switch in the release of IBU from NM3-R by a pH-dependent “open-close” mechanism. We hope that the release of all of the loaded IBU can be controlled perfectly by these multiamine chains. However, it is inevitable for the drug to be adsorbed on the external surface of NM3-R. Therefore, though the multiamine chains are in “closed” state in the acidic environment, a certain amount of IBU is still desorbed. However, in acidic solution, does IBU release only from the external surface of NM3-R? In order to clarify this problem, we make a rough analysis. It has been known that the external surface of NM3-R accounts for 5.9% of its total surface, so the release of IBU should not exceed 5.9% if these multiamine gates are closed ideally. On the other hand, from our previous study31 it has been learned that IBU on the external surface of NM3-R should release completely within 8 h at low pH. However,

17138 DOI: 10.1021/la102952n

in present work the IBU release from NM3-R is sustained during the entire investigated period and reached about 13% at pH 5.0 after 24 h. This reveals that the multiamine chains can block the mesopores of NM3-R efficiently, but still leave gaps for IBU transport. By optimizing the molecular structure, we learn that multiamine chain in diameter is about 1.3 nm, whereas the pore diameter of MSS is about 2.6 nm. The multiamine chains seem to be able to fully close pores when multiamine chains are perpendicular to pore wall. In fact, in acidic solution the terminal of multiamine chain is a positively charged amino group, thus the multiamine chains can not link head to head, but leave some gaps. Therefore, some IBU go through these gaps to release out. Of course, as our results show, this release is quite difficult (see Figure 7d).

4. Conclusion Through grafting a pH-responsive organic groups (multiamine chains) around the mesopore outlets of MSS by a long reaction time (12 h) at high temperature (100 °C), the gated MSS (namely NM3-R) has been successfully prepared. Systematic investigations were adopted to probe the pH sensitivity of NM3-R. The multiamine chains played the role of a “molecular switch” that can control the release of guest molecules. NM3-R showed a pHresponsive control for drug release. At high pH (pH 7.5), IBU drug loaded in NM3-R releases rapidly and completely (within 2 h); at low pH (pH 4.0 or 5.0), only a small part of the IBU (13 wt %) was slowly released from the NM3-R and the most of IBU is effectively confined in the mesopores. Acknowledgment. The financial support from the National Native Science Foundation (No. 20573128), State Key Laboratory of Coal Conversion Foundation (No. 10-11-610), and Special Fund for Basic Scientific Research of Central Colleges, China University of Geosciences (Wuhan) (No. CUGL090307) are acknowledged.

Langmuir 2010, 26(22), 17133–17138