Materials Science and Engineering C 50 (2015) 64–73
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Biodegradation-tunable mesoporous silica nanorods for controlled drug delivery Sung Bum Park a, Young-Ho Joo a, Hyunryung Kim b, WonHyoung Ryu b,⁎, Yong-il Park a,⁎ a b
School of Advanced Materials & System Engineering, Kumoh National Institute of Technology, 1 Yangho-dong, Gumi, Gyeongbuk 730-701, Republic of Korea School of Mechanical Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Republic of Korea
a r t i c l e
i n f o
Article history: Received 5 August 2014 Received in revised form 26 December 2014 Accepted 23 January 2015 Available online 24 January 2015 Keywords: Mesoporous Biodegradation Silica nanorods Anodic aluminum oxide Drug delivery
a b s t r a c t Mesoporous silica in the forms of micro- or nanoparticles showed great potentials in the field of controlled drug delivery. However, for precision control of drug release from mesoporous silica-based delivery systems, it is critical to control the rate of biodegradation. Thus, in this study, we demonstrate a simple and robust method to fabricate “biodegradation-tunable” mesoporous silica nanorods based on capillary wetting of anodic aluminum oxide (AAO) template with an aqueous alkoxide precursor solution. The porosity and nanostructure of silica nanorods were conveniently controlled by adjusting the water/alkoxide molar ratio of precursor solutions, heattreatment temperature, and Na addition. The porosity and biodegradation kinetics of the fabricated mesoporous nanorods were analyzed using N2 adsorption/desorption isotherm, TGA, DTA, and XRD. Finally, the performance of the mesoporous silica nanorods as drug delivery carrier was demonstrated with initial burst and subsequent “zero-order” release of anti-cancer drug, doxorubicin. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Mesoporous silica is an attractive biomaterial because of their stable pore structure based on their high regularity, high surface to volume ratio, biocompatibility, biodegradation, and control of pore size at nanoscale [1–3]. Recently, mesoporous silica nanoparticles have drawn much attention as promising drug carriers in controlled drug delivery since it biodegrades in a controlled manner. In particular, biodegradable mesoporous silica has been used as drug carriers to store unstable proteins or poorly-soluble drugs for sustained and controlled delivery to target tissues [4–8]. The biodegradation and drug loading capacity of mesoporous silica were reported to vary according to their porosity and pore size [4,9–12]. Thus, modulation of the porosity and pore size of mesoporous silica can enable robust control of drug release kinetics from drug delivery system based on mesoporous silica. Existing fabrication processes for mesoporous silica nanoparticles include particle growth techniques using laser ablation and atomic layer deposition, anodizing, growth in liquid phase, electrolytic plating, and chemical vapor deposition (CVD) using supports [9–11]. More recently, a silicon wafer etching process has been developed to fabricate porous silicon nanowires with biodegradability for controlled drug delivery [4]. However, those approaches are expensive and time consuming. It is also difficult to control the porous structure that determines the biodegradation rate of
⁎ Corresponding authors. E-mail addresses: [email protected] (W. Ryu), [email protected] (Y. Park).
http://dx.doi.org/10.1016/j.msec.2015.01.073 0928-4931/© 2015 Elsevier B.V. All rights reserved.
mesoporous silica carriers. As another way to control the silica polymer network or pore structure of silica ceramics, sol–gel processes using tetrarethylorthosilicate (Si(OC 2H5 ) 4) (TEOS) have been explored. The TEOS-based sol–gel processes were focused on controlling pore morphology of silica structure by change of water quantity [1,13–15], change of the rates of hydrolysis and condensation reaction [16], stepwise addition of ingredients with time intervals [17], and addition of acidic and basic catalysts [1,17,18]. In this study, we propose an AAO template-based sol–gel process that allows for diverse control of the compositions and processes to “conveniently tune” the porosity and biodegradability of mesoporous silica nanorods (NRs) (Fig. 1). In particular, we report significant change in biodegradation rates of fabricated NRs by controlling their porosity and heat-treatment temperature as well as Na addition. Porosity apparently increases the corrosion surfaces of amorphous silica pore walls inside NRs, which are susceptible to water-based corrosion attack, and, therefore, is a very effective variable to control biodegradability. We increased the porosity of the fabricated NRs up to about 47% by controlling water amount of the precursor solution and annealing temperature. Effect of both water content in precursor solutions and heat-treatment on the microstructure, porosity, and crystallinity of NRs was also monitored to understand the biodegradability of mesoporous NRs. Furthermore, sodium (Na) at the maximum amount up to 5 mol% was added to precursor solutions to fabricate Na-doped mesoporous NRs (NaNRs) with accelerated biodegradation. Such incorporation of Na into amorphous silica network was expected to disrupt siloxane bond (Si–O–Si) by forming Na2 O ending groups. This can lead to
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
65
Fig. 1. Schematic diagram of silica sol-based nanorod fabrication using AAO template.
seriously reduced chemical stability of NRs and faster breakdown of silica structure. Fabrication of NaNRs was demonstrated and their accelerated biodegradation dependent on Na content was also confirmed. Finally, drug release performance of NRs and NaNRs using anti-cancer drug, doxorubicin, was demonstrated and discussed.
2. Materials and methods 2.1. Synthesis of mesoporous NRs Nanorod samples with different water contents in precursor solutions were prepared to control the pore size and structure of NRs
Fig. 2. FE-SEM images of (a) NR1 and (b) NR2 formed in AAO templates.
66
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
(b)
Fig. 3. FE-SEM images of fabricated (a) NR1 and (b) NR2 heat-treated at 150, 350, and 650 °C after removal of AAO templates.
through sol–gel processes. Nanorods with low water content (NR1) were fabricated by preparing a precursor solution at the ratio of H2O/ TEOS = 3.5 and impregnating porous templates with the precursor solution. Another group of nanorods with higher water content (NR2) was prepared using a precursor solution with the ratio of H2O/TEOS = 14. TEOS and ethanol were mixed at a ratio of 1:3.9 for 60 min and a predetermined amount of water for hydrolysis and hydrochloric acid (HCl) as catalyst was added to the solutions. Anodic aluminum oxide (AAO) filters (Anodisc®, Whatman Inc., NJ, USA, thickness: 50 μm, pore size (inner diameter): 200 nm, filter diameter: 47 mm) was selected as a porous template because of its thermal and mechanical stabilities suitable for nanorod fabrication. The AAO template was immersed and stored for 2 h in a starting solution for the impregnation of AAO templates (Fig. 1). Upon the completion of the sol–gel process, the AAO template filled with synthesized nanorods was dried for 2 h at 80 °C in order to minimize cracks and shrinkage of the gel nanorods. Heat treatment was performed for 120 h at 150, 350,
and 650 °C. After drying and heat treatments, the AAO template was dissolved and removed in an aqueous 3 M NaOH solution to obtain free-standing nanorods. The resulting free-standing NRs were rinsed in ethanol and then dried again at 80 °C. Sodium-incorporated silica nanorods (NaNRs) were also fabricated using a precursor solution synthesized by adding a solution of sodium nitrate (NaNO3, 99.0%, DC Chemical Co. Seoul, Korea): ethanol = 1:5 vol.% into the precursor solution of NR1. The NaNRs were fabricated by the following processes identical to NR1 fabrication. Targeted compositions NaNR-2% and NaNR-5% were 2:98 and 5:95 mol% of Na2O:SiO2, respectively. 2.2. Characterization of mesoporous NRs The pore size and distribution of the fabricated NRs were verified through SEM/TEM analysis. The specific surface area, pore volume, and porosity of NR1s and NR2s were measured based on N2
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
67
(b)
Fig. 4. TEM images of (a) NR1 and (b) NR2 heat-treated at 150, 350, and 650 °C.
adsorption/desorption isotherm using a pore size analyzer (BELSORPmax, BEL Inc., Osaka, Japan). The measurement was conducted under a nitrogen atmosphere with the weight of each nanorod set at 2 g. The average lengths of NRs were estimated from cross-sectioned SEM images of the NRs. The thermal response of NRs was measured using TG-DTA (SDT Q600, TA Instrument Inc., New Castle, DE, USA) under temperature elevation of 10 °C per minute up to 900 °C. To observe phase changes in NR1 and NR2 according to heat-treatment temperature, the XRD patterns of the samples were analyzed (SWXD (X-MAX/ 2000-PC), Rigaku, Tokyo, Japan) within the range of 2θ = 5°–60° and full width at half maximum (FWHM) peak intensities measured from the XRD patterns at 2θ = 23° were compared.
2.3. Biodegradation test of mesoporous NRs In order to verify the biodegradability of silica nanorods that were obtained by the composition control of precursor solutions, silica
nanorods were placed in a PBS solution stored at 36.5 °C for three days. For prevention of nanorod dispersion in a PBS solution after removal of an AAO template, the AAO template was fixed on a metal substrate using an epoxy layer such that nanorods were immobilized on the epoxy layer on the metal substrate even after removal of the AAO template.
2.4. Drug release test Nanorod samples of NR1, NaNR-2%, NaNR-5% (20 mg each) were dispersed in 6 ml of aqueous doxorubicin (DOX, Sigma-Aldrich, St. Loius, MO, USA) solution (0.5 mg/ml) and sonicated for 5 min. The nanorod dispersion was light-sealed and stirred using a magnetic bar at 200 rpm for 24 h in order to load DOX on nanorods. The DOXloaded nanorods were rinsed thoroughly with 6 ml of distilled water until the rinsed liquid was free of DOX residue. The washed nanorods were collected using a centrifuge filter and dried for 24 h at room
68
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
When silica nanorods were fabricated without any acidic catalyst, small particles attached to external surface of nanorods were observed. However, with the use of HCl as an acid catalyst in our experiments, no particle formation was observed as shown in Figs. 2 & 3. Acid catalysts helped the formation of extended, weakly branched chains [21] to form extremely fine microstructures without formation of unwanted particles such as NaOH [21–23]. On the other hand, high water content in the precursor solution increases cylinder-shaped, homogeneously distributed fine pores [24] or small, dense spherical-shaped particles [25]. NR2 with higher water content (H2O/TEOS = 14) produced NRs with homogeneously distributed nanopores (Fig. 4). In the previous studies by Lee et al. [26], porous silica was obtained using precursor solutions with the water content range of H2O/TEOS = 7.8–18 and HCl catalyst. This conforms well to our results in this experiment. The distribution of nanopores on the surface of the nanorods was observed by TEM (Fig. 4). The nanopores were homogeneously distributed on the surface and inside of the fabricated NRs. The measured porosity and the size of pores of the NR2s were also larger compared to the NR1s with low water content. This is mainly ascribed to the formation of overlapped polymer clusters consisting of weak and loose silica network by fast polymerization reaction. It is also consistent with the results from previous works that increase of water content accelerated the hydrolysis rate in both acid- and base-catalyzed solutions [19,20].
3.2. Porosity and pore size
(b) Fig. 5. N2 adsorption/desorption isotherm of (a) NR1 and NR2 heat-treated at 150 °C, (b) NR2 heat-treated at 350 and 650 °C.
temperature in a vacuum oven. DOX release tests were conducted by immersing the DOX-loaded nanorods in pH 7.4 phosphate buffer saline (PBS) in an incubator and the media were refreshed every designated time point for 8 days. UV/VIS spectrophotometer (V-650, JASCO Corporation, Tokyo, Japan) was used to measure the quantities of DOX released at the wavelength of 480 nm. The DOX loading amount was calculated by subtracting the remaining amount of DOX in the rinsed water from the total amount in the stock solution. 3. Results and discussion
N2 adsorption/desorption isotherm of NRs (heat-treated at 150 °C) is illustrated and summarized in Fig. 5a and Table 1. According to the six classified types of adsorption isotherm of porous powder [27], the NR2 shows a typical Langmuir-type isotherm curve for “mesoporous” materials with nanopores having diameter of 2–50 nm. The large hysteresis is due to capillary condensation through bottle-neck typed mesopores. On the other hand, the curve of NR1 is similar to the typical N2 adsorption/ desorption isotherm for “non-porous” materials, although it shows slight hysteresis. These results match well with the acquired values of specific surface area, pore volume, pore size, and porosity as shown in Table 1. The porosity of NR2 (heat-treated at 150 °C) was 47.4%, which was about twice higher than the porosity (23.3%) of NR1. Such increase in the porosity and the specific surface area for NRs fabricated with higher water content is ascribed to fine nanopores formed between delicate SiO2 frames consisting of small polymer chains which are rapidly created by a quick gel reaction [25–27]. On the other hand, with increasing heat-treatment temperature of NR2 from 150 °C up to 650 °C, the pore size and the porosity were reduced from sintering effect. This also resulted in the increase of the density of NR2 due to pore shrinkage by the high temperature treatment. The observed adsorption/desorption isotherm of the NR2 at these elevated temperatures (Fig. 5b) also strongly suggests the collapse of pore structures during heattreatment showing similar patterns with the curve of NR1 having low porosity. Therefore, in addition to water/TEOS ratio, heat-treatment temperature can also be a useful variable to control the porosity and biodegradability of mesoporous NRs.
3.1. Morphology study of NRs The lengths and shapes of the nanorods were affected by impregnation time. Long impregnation time resulted in elongated and centerfilled nanorod forms, while decrease in impregnation time produced relatively shorter and hollow tube form. This is because gel polymerization started from the surfaces of top and the column walls of AAO template. The gel networks spread from these surfaces toward the inside of the nanopores of AAO templates. Therefore, long gelation time increased nanorod lengths and also effectively filled the open columnar nanopores.
Table 1 BET analysis results. H2O/TEOS Heat-treatment BET surface Pore volume Pore size Porosity (mol ratio) temp. (°C) area (m2/g) (cm3/g) (Å) (%) NR1 3.5 NR2 14 14 14
150 150 350 650
67.80 78.78 68.78 44.67
0.094 0.483 0.152 0.081
55.41 65.23 58.12 46.15
23.3 47.4 29.2 21.9
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
69
(b)
Fig. 6. FE-SEM images of biodegradation process of (a) NR1 and (b) NR2 heat-treated at 150 °C after incubation in PBS solution.
3.3. Biodegradation of NRs The SEM images in Fig. 6 show the morphology of NR1 and NR2 (both heat-treated at 150 °C) observed for three days to monitor their biodegradation process. After three days, NR1 shows relatively slow biodegradation and aggregation of each NR by a capillary force during their drying for SEM analysis. NR2 degraded at faster speed than NR1. Most NR2 nanorods collapsed and dissolved after 48 h. This difference in biodegradation rate between NR1 and NR2 is explained by the higher porosity and surface area of NR2 than NR1. This also confirms that the differences in processing conditions and the resulting microarchitectures between NR1 and NR2 have profound influence on their biodegradability.
Fig. 7 shows DT/TGA analysis results for NR1 and NR2 (both heattreated at 150 °C). Under 100 °C, both nanorods showed the weight loss of about 10% and 19%, respectively. This seems to be caused by evaporation of H2O absorbed onto the pores. The weight loss of NR2 was about twice higher than NR1. This demonstrates a higher degree of water absorption in NR2 that has higher porosity than NR1. Additional weight losses of 8% and 4%, respectively in NR1 and NR2 were observed within the range of 100 °C–600 °C. This difference is due to the evaporation/decomposition of residual organic matters. From the retarded weight loss and relatively large exothermic combustion that continues up to 600 °C, the main cause of the reduced weight loss of NR2 seems to be relatively small amount of residual non-reactive alkoxy group (− OR) in its gel state. The large
70
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
(a)
(b) (b) Fig. 7. DT-TGA analysis of (a) NR1 and (b) NR2. Fig. 8. XRD peak intensity of (a) NR1 and (b) NR2 at 2θ = 23°.
amount of H2O drastically increases the hydrolysis reaction of TEOS that changes alkoxy groups to alcohol which easily evaporates below 100 °C. As shown in Fig. 8, the change in the XRD peak intensity at 2θ = 23° of NR1 and NR2 with different heat-treatment temperatures was measured over two days of biodegradation. NR1 and NR2 heat-treated at 150, 350, and 650 °C were immersed in PBS solutions for two days for the analysis with 24-hour intervals. In both NR1 and NR2, broad peaks by amorphous silica was observed near 2θ = 23° (data not shown). On the other hand, the relative intensity observed at 2θ b 10° is larger in NR2 compared to NR1. This seems to be caused by the higher porosity of NR2. The peak intensity at 2θ = 23° decreases as biodegradation proceeds and, as seen in both cases, the change of peak intensity is dramatic in the samples heat-treated at 150 °C while negligible change is detected in those heat-treated at 650 °C. In Fig. 9, the full width at half maximum (FWHM) XRD intensities of NR1 and NR2 at each biodegradation time were measured. In PBS solutions, the FWHM values increased over time for all samples. This indicates gradual decrease of their crystallinity because the silica glass network collapsed through biodegradation. Regardless of heattreatment temperatures, the FWHM of NR2 was higher than that of NR1. This confirms that NR2 has lower crystallinity and higher porosity compared to NR1. This also explains the faster biodegradation of NR2 than NR1. Fig. 10 shows the change in the average lengths of NR2 when immersed in a PBS solution. The length decrease of NR2 heat-
treated at 150 °C was more significant compared to NR2 samples treated at 350 and 650 °C. This indicates that heat-treatment of NRs at higher temperature sintered NRs, resulting in decrease of biodegradation speed. In some samples, severe aggregation of the NR2s heat-treated at 650 °C was observed without their degradation after 3 days being immersed in a PBS solution. 3.4. Biodegradation of NaNRs The SEM images in Fig. 11 show the morphology of NaNR-2% and NaNR-5% (heat-treated at 150 °C) observed for three days monitoring the biodegradation process. Compared to NRs, NaNRs showed faster degradation and increase of Na content further accelerated the decomposition. After a day, a significant portion of NaNRs already collapsed. In particular, NaNR-5% showed the fastest biodegradation and was perfectly removed after 3 days. As shown in Fig. 12, after 3 days immersed in PBS solution at room temperature, NaNR2% and NaNR-5% lost about 28% and 37% of their weight, respectively. On the other hand, NR1 samples (without any Na inclusion) lost about 10% of its original weight. The accelerated biodegradation speed of the NaNRs is attributed to the seriously reduced chemical stability of the silica networks due to ‘scissoring’ of Si–O–Si oxide network by introducing Na2 O as an ending group, which is welldefined in the studies on silicate glasses. Furthermore, Na-
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
71
Fig. 10. Change of average length of NR2 heat-treated at 150, 350, and 650 °C when incubated in PBS solution.
incorporated silica is vulnerable to an acidic corrosion attack compared to basic or alkaline circumstances because the Na 2O ending group in silica network affords a preferential attacking point to acidic corrosion. The partially dissolved silica surface results in continuous mechanical collapse of entire silica oxide body by accelerated corrosion. 3.5. Drug release performance of NRs
(b)
DOX loading amount in NR1 and NaNR-5% was calculated by subtracting the remaining amount of DOX in the rinsed water from the total amount in the stock solution. Drug loading efficiency (loaded DOX/nanorods, w/w) was determined as 10.1% and 14.8% for NR1 and NaNR-5%, respectively. In order to understand the performance of the NRs as drug delivery carriers, drug-loaded NRs were immersed in a PBS solution and incubated at 36.5 °C. As a model drug, DOX, an anti-cancer drug, was used in this experiment. DOX, positivelycharged under physiological pH [28], was loaded based on ionic bonding with the negatively-charged surface of NRs. When biodegradation of NRs occurs, the break-down of the silica materials releases the DOX molecules attached on their surface. As shown in Fig. 13, initial burst was observed within the first 24 h and subsequently DOX was linearly released until the 8th day of incubation. It is also noteworthy that NaNR-5% released DOX faster than NR1 within the first 24 h. This conforms well to the difference in biodegradation speeds between NR1 and NaNRs, as shown in Fig. 11. Such initial burst and subsequent “zero-order” style release are desired for cancer treatments. The initial burst of anti-cancer drug can damage most of cancer cells in target tumor, while the following constant delivery of the drug at lower delivery rate can prevent any reoccurrence of cancer at the delivery site. 4. Conclusions
(c) Fig. 9. FWHM analysis of NR1 and 2 heat-treated at (a) 150, (b) 350, and (c) 650 °C.
We report a novel fabrication method to precisely control the geometry, porosity, and biodegradation rate of mesoporous silica nanorods for controlled drug delivery. Mesoporous NRs with varying porosities were fabricated by controlling H2O/TEOS ratio and heattreatment temperatures through a sol–gel process. AAO templatebased formation of mesoporous silica nanorods successfully controlled the length of NRs by adjusting the impregnation time of the precursor solution of NRs in AAO templates. Uniform mesoporous
72
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
(b)
Fig. 11. FE-SEM images of biodegradation process of (a) NaNR-2% and (b) NaNR-5% incubated in PBS solution.
nanorods were obtained with reduced cracks and shrinkage of nanorods by drying and heat-treatment processes that affected pore formation. The distribution and size of pores inside nanorods were also controlled by water quantity used in the hydrolysis step of the sol–gel process. Higher water content in the precursor solution leads to higher porosity of the nanorods and also increased biodegradability. Heat-treatment temperature also significantly affected the biodegradability of mesoporous NRs by increasing their crystallinity which attributed to fortify silica network and thus to increase resistant toward water-based corrosion attack. As a drug delivery carrier, the mesoporous NRs showed initial burst and “zero-order”
release kinetics of anti-cancer drug, which is ideal for cancer treatment. In addition, NaNRs were fabricated using AAO template method and they showed accelerated biodegradation compared to NRs. It was demonstrated that the biodegradation speed was increased with the increase of Na amount.
Acknowledgments This paper was supported by Research Fund, Kumoh National Institute of Technology.
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
73
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.01.073. References
Fig. 12. Weight change comparison of NR1, NaNR-2%, and NaNR-5% in PBS solution (pH 7).
Fig. 13. DOX release test in vitro on NR1 and NaNR-5% for 8 days.
[1] J.C. Ro, I.J. Chung, J. Non-Cryst. Solids 130 (1991) 8. [2] R.K. Iler, The Chemistry of Silica: Solubility, Polymerization, Colloid and Surface Properties and Biochemistry of Silica, John Wiley & Sons, New York, 1979. [3] H. Böttcher, P. Slowik, W. Suss, J. Sol-Gel Sci. Technol. 13 (1998) 277. [4] C. Chiappini, X. Liu, J.R. Fakhoury, M. Ferrari, Adv. Funct. Mater. 20 (2010) 2231. [5] L. Vaccari, D. Canton, N. Zaffaroni, R. Villa, M. Tormen, E. di Fabrizio, Microsc. Electron. Eng. 83 (2006) 1598–1601. [6] I.I. Slowing, J.L. Vivero-Escoto, C.-W. Wu, V.S.-Y. Lin, Adv. Drug Deliv. Rev. 60 (2008) 1278–1288. [7] K.M.L. Taylor, J.S. Kim, W.J. Rieter, H. An, W. Lin, W. Lin, J. Am. Chem. Soc. 130 (2008) 2154–2155. [8] K.-W. Hu, K.-C. Hsu, C.-S. Yeh, Biomaterials 31 (2010) 6843–6848. [9] C.J. Brinker, G.W. Scherer, Sol–Gel Science, Academic Press, 1990. [10] H. Böttcher, C. Jagota, J. Trepte, K.-H. Kallies, H.J. Haufe, Control. Release 60 (1999) 57. [11] M. Ahola, P. Kortesuo, I. Kangasniemi, J. Kiesvaara, A. Yli-Urpo, Int. J. Pharm. 195 (2000) 219. [12] N.K. Mal, M. Fujiwara, Y. Tanaka, Nature 421 (2003) 350. [13] S. Sakka, K. Kamiya, J. Non-Cryst. Solids 48 (1982) 31. [14] I. Strawbridge, A.F. Craievich, P.F. James, J. Non-Cryst. Solids 72 (1985) 139. [15] B.E. Yoldas, J. Polym. Sci. 24 (3475) (1986) 3487. [16] C.J. Brinker, K.D. Keefer, D.W. Shaefer, C.S. Ashley, J. Non-Cryst. Solids 48 (1982) 47. [17] C.J. Brinker, K.D. Keefer, D.W. Shaefer, C.S. Ashley, J. Non-Cryst. Solids 63 (1984) 50. [18] J.Y. Ying, J.B. Benziger, J. Non-Cryst. Solids 147/148 (1992) 222. [19] I. Strawbridge, A.F. Craievich, P.F. James, J. Non-Cryst. Solids 72 (1985) 139. [20] L.C. Klein, G.J. Garvey, in: C.J. Brinker, D.E. Clark, D.R. Ulrich (Eds.),Better Ceramics Through Chemistry III, Mater. Res. Soc. Symp. Proc., 32, 1984, p. 33. [21] C.J. Brinker, K.D. Keefer, D.W. Shaefer, R.A. Assink, B.D. Kay, C.S. Ashley, J. Non-Cryst. Solids 63 (1984) 50. [22] C.J. Brinker, K.D. Keefer, D.W. Shaefer, C.S. Ashely, J. Non-Cryst. Solids 48 (1982) 47. [23] N. Nogami, Y. Moriya, J. Non-Cryst. Solids 37 (1980) 191. [24] L.C. Klein, Sol–Gel Technology for Thin Films, Fibers, Preforms, Electronics and Specialty Shapes, 389, Noyes Publication, NJ, U.S.A., 1988 [25] I. Strawbridge, A.F. Craievich, P.F. James, J. Non-Cryst. Solids 72 (1985) 139. [26] J.-H. Lee, W.J. Kim, J. Lee, J. Korean Ceram. Soc. 34 (2) (1997) 216. [27] S. Brunauer, L.S. Deming, W.S. Deming, E. Teller, J. Am. Ceram. Soc. 62 (1940) 1723. [28] H. Meng, M. Liong, T. Xia, Z. Li, Z. Ji, J.I. Zink, A.E. Nel, ACS Nano 4 (8) (2010) 4539–4550.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Biodegradation-tunable mesoporous silica nanorods for controlled drug delivery Sung Bum Park a, Young-Ho Joo a, Hyunryung Kim b, WonHyoung Ryu b,⁎, Yong-il Park a,⁎ a b
School of Advanced Materials & System Engineering, Kumoh National Institute of Technology, 1 Yangho-dong, Gumi, Gyeongbuk 730-701, Republic of Korea School of Mechanical Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Republic of Korea
a r t i c l e
i n f o
Article history: Received 5 August 2014 Received in revised form 26 December 2014 Accepted 23 January 2015 Available online 24 January 2015 Keywords: Mesoporous Biodegradation Silica nanorods Anodic aluminum oxide Drug delivery
a b s t r a c t Mesoporous silica in the forms of micro- or nanoparticles showed great potentials in the field of controlled drug delivery. However, for precision control of drug release from mesoporous silica-based delivery systems, it is critical to control the rate of biodegradation. Thus, in this study, we demonstrate a simple and robust method to fabricate “biodegradation-tunable” mesoporous silica nanorods based on capillary wetting of anodic aluminum oxide (AAO) template with an aqueous alkoxide precursor solution. The porosity and nanostructure of silica nanorods were conveniently controlled by adjusting the water/alkoxide molar ratio of precursor solutions, heattreatment temperature, and Na addition. The porosity and biodegradation kinetics of the fabricated mesoporous nanorods were analyzed using N2 adsorption/desorption isotherm, TGA, DTA, and XRD. Finally, the performance of the mesoporous silica nanorods as drug delivery carrier was demonstrated with initial burst and subsequent “zero-order” release of anti-cancer drug, doxorubicin. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Mesoporous silica is an attractive biomaterial because of their stable pore structure based on their high regularity, high surface to volume ratio, biocompatibility, biodegradation, and control of pore size at nanoscale [1–3]. Recently, mesoporous silica nanoparticles have drawn much attention as promising drug carriers in controlled drug delivery since it biodegrades in a controlled manner. In particular, biodegradable mesoporous silica has been used as drug carriers to store unstable proteins or poorly-soluble drugs for sustained and controlled delivery to target tissues [4–8]. The biodegradation and drug loading capacity of mesoporous silica were reported to vary according to their porosity and pore size [4,9–12]. Thus, modulation of the porosity and pore size of mesoporous silica can enable robust control of drug release kinetics from drug delivery system based on mesoporous silica. Existing fabrication processes for mesoporous silica nanoparticles include particle growth techniques using laser ablation and atomic layer deposition, anodizing, growth in liquid phase, electrolytic plating, and chemical vapor deposition (CVD) using supports [9–11]. More recently, a silicon wafer etching process has been developed to fabricate porous silicon nanowires with biodegradability for controlled drug delivery [4]. However, those approaches are expensive and time consuming. It is also difficult to control the porous structure that determines the biodegradation rate of
⁎ Corresponding authors. E-mail addresses: [email protected] (W. Ryu), [email protected] (Y. Park).
http://dx.doi.org/10.1016/j.msec.2015.01.073 0928-4931/© 2015 Elsevier B.V. All rights reserved.
mesoporous silica carriers. As another way to control the silica polymer network or pore structure of silica ceramics, sol–gel processes using tetrarethylorthosilicate (Si(OC 2H5 ) 4) (TEOS) have been explored. The TEOS-based sol–gel processes were focused on controlling pore morphology of silica structure by change of water quantity [1,13–15], change of the rates of hydrolysis and condensation reaction [16], stepwise addition of ingredients with time intervals [17], and addition of acidic and basic catalysts [1,17,18]. In this study, we propose an AAO template-based sol–gel process that allows for diverse control of the compositions and processes to “conveniently tune” the porosity and biodegradability of mesoporous silica nanorods (NRs) (Fig. 1). In particular, we report significant change in biodegradation rates of fabricated NRs by controlling their porosity and heat-treatment temperature as well as Na addition. Porosity apparently increases the corrosion surfaces of amorphous silica pore walls inside NRs, which are susceptible to water-based corrosion attack, and, therefore, is a very effective variable to control biodegradability. We increased the porosity of the fabricated NRs up to about 47% by controlling water amount of the precursor solution and annealing temperature. Effect of both water content in precursor solutions and heat-treatment on the microstructure, porosity, and crystallinity of NRs was also monitored to understand the biodegradability of mesoporous NRs. Furthermore, sodium (Na) at the maximum amount up to 5 mol% was added to precursor solutions to fabricate Na-doped mesoporous NRs (NaNRs) with accelerated biodegradation. Such incorporation of Na into amorphous silica network was expected to disrupt siloxane bond (Si–O–Si) by forming Na2 O ending groups. This can lead to
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
65
Fig. 1. Schematic diagram of silica sol-based nanorod fabrication using AAO template.
seriously reduced chemical stability of NRs and faster breakdown of silica structure. Fabrication of NaNRs was demonstrated and their accelerated biodegradation dependent on Na content was also confirmed. Finally, drug release performance of NRs and NaNRs using anti-cancer drug, doxorubicin, was demonstrated and discussed.
2. Materials and methods 2.1. Synthesis of mesoporous NRs Nanorod samples with different water contents in precursor solutions were prepared to control the pore size and structure of NRs
Fig. 2. FE-SEM images of (a) NR1 and (b) NR2 formed in AAO templates.
66
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
(b)
Fig. 3. FE-SEM images of fabricated (a) NR1 and (b) NR2 heat-treated at 150, 350, and 650 °C after removal of AAO templates.
through sol–gel processes. Nanorods with low water content (NR1) were fabricated by preparing a precursor solution at the ratio of H2O/ TEOS = 3.5 and impregnating porous templates with the precursor solution. Another group of nanorods with higher water content (NR2) was prepared using a precursor solution with the ratio of H2O/TEOS = 14. TEOS and ethanol were mixed at a ratio of 1:3.9 for 60 min and a predetermined amount of water for hydrolysis and hydrochloric acid (HCl) as catalyst was added to the solutions. Anodic aluminum oxide (AAO) filters (Anodisc®, Whatman Inc., NJ, USA, thickness: 50 μm, pore size (inner diameter): 200 nm, filter diameter: 47 mm) was selected as a porous template because of its thermal and mechanical stabilities suitable for nanorod fabrication. The AAO template was immersed and stored for 2 h in a starting solution for the impregnation of AAO templates (Fig. 1). Upon the completion of the sol–gel process, the AAO template filled with synthesized nanorods was dried for 2 h at 80 °C in order to minimize cracks and shrinkage of the gel nanorods. Heat treatment was performed for 120 h at 150, 350,
and 650 °C. After drying and heat treatments, the AAO template was dissolved and removed in an aqueous 3 M NaOH solution to obtain free-standing nanorods. The resulting free-standing NRs were rinsed in ethanol and then dried again at 80 °C. Sodium-incorporated silica nanorods (NaNRs) were also fabricated using a precursor solution synthesized by adding a solution of sodium nitrate (NaNO3, 99.0%, DC Chemical Co. Seoul, Korea): ethanol = 1:5 vol.% into the precursor solution of NR1. The NaNRs were fabricated by the following processes identical to NR1 fabrication. Targeted compositions NaNR-2% and NaNR-5% were 2:98 and 5:95 mol% of Na2O:SiO2, respectively. 2.2. Characterization of mesoporous NRs The pore size and distribution of the fabricated NRs were verified through SEM/TEM analysis. The specific surface area, pore volume, and porosity of NR1s and NR2s were measured based on N2
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
67
(b)
Fig. 4. TEM images of (a) NR1 and (b) NR2 heat-treated at 150, 350, and 650 °C.
adsorption/desorption isotherm using a pore size analyzer (BELSORPmax, BEL Inc., Osaka, Japan). The measurement was conducted under a nitrogen atmosphere with the weight of each nanorod set at 2 g. The average lengths of NRs were estimated from cross-sectioned SEM images of the NRs. The thermal response of NRs was measured using TG-DTA (SDT Q600, TA Instrument Inc., New Castle, DE, USA) under temperature elevation of 10 °C per minute up to 900 °C. To observe phase changes in NR1 and NR2 according to heat-treatment temperature, the XRD patterns of the samples were analyzed (SWXD (X-MAX/ 2000-PC), Rigaku, Tokyo, Japan) within the range of 2θ = 5°–60° and full width at half maximum (FWHM) peak intensities measured from the XRD patterns at 2θ = 23° were compared.
2.3. Biodegradation test of mesoporous NRs In order to verify the biodegradability of silica nanorods that were obtained by the composition control of precursor solutions, silica
nanorods were placed in a PBS solution stored at 36.5 °C for three days. For prevention of nanorod dispersion in a PBS solution after removal of an AAO template, the AAO template was fixed on a metal substrate using an epoxy layer such that nanorods were immobilized on the epoxy layer on the metal substrate even after removal of the AAO template.
2.4. Drug release test Nanorod samples of NR1, NaNR-2%, NaNR-5% (20 mg each) were dispersed in 6 ml of aqueous doxorubicin (DOX, Sigma-Aldrich, St. Loius, MO, USA) solution (0.5 mg/ml) and sonicated for 5 min. The nanorod dispersion was light-sealed and stirred using a magnetic bar at 200 rpm for 24 h in order to load DOX on nanorods. The DOXloaded nanorods were rinsed thoroughly with 6 ml of distilled water until the rinsed liquid was free of DOX residue. The washed nanorods were collected using a centrifuge filter and dried for 24 h at room
68
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
When silica nanorods were fabricated without any acidic catalyst, small particles attached to external surface of nanorods were observed. However, with the use of HCl as an acid catalyst in our experiments, no particle formation was observed as shown in Figs. 2 & 3. Acid catalysts helped the formation of extended, weakly branched chains [21] to form extremely fine microstructures without formation of unwanted particles such as NaOH [21–23]. On the other hand, high water content in the precursor solution increases cylinder-shaped, homogeneously distributed fine pores [24] or small, dense spherical-shaped particles [25]. NR2 with higher water content (H2O/TEOS = 14) produced NRs with homogeneously distributed nanopores (Fig. 4). In the previous studies by Lee et al. [26], porous silica was obtained using precursor solutions with the water content range of H2O/TEOS = 7.8–18 and HCl catalyst. This conforms well to our results in this experiment. The distribution of nanopores on the surface of the nanorods was observed by TEM (Fig. 4). The nanopores were homogeneously distributed on the surface and inside of the fabricated NRs. The measured porosity and the size of pores of the NR2s were also larger compared to the NR1s with low water content. This is mainly ascribed to the formation of overlapped polymer clusters consisting of weak and loose silica network by fast polymerization reaction. It is also consistent with the results from previous works that increase of water content accelerated the hydrolysis rate in both acid- and base-catalyzed solutions [19,20].
3.2. Porosity and pore size
(b) Fig. 5. N2 adsorption/desorption isotherm of (a) NR1 and NR2 heat-treated at 150 °C, (b) NR2 heat-treated at 350 and 650 °C.
temperature in a vacuum oven. DOX release tests were conducted by immersing the DOX-loaded nanorods in pH 7.4 phosphate buffer saline (PBS) in an incubator and the media were refreshed every designated time point for 8 days. UV/VIS spectrophotometer (V-650, JASCO Corporation, Tokyo, Japan) was used to measure the quantities of DOX released at the wavelength of 480 nm. The DOX loading amount was calculated by subtracting the remaining amount of DOX in the rinsed water from the total amount in the stock solution. 3. Results and discussion
N2 adsorption/desorption isotherm of NRs (heat-treated at 150 °C) is illustrated and summarized in Fig. 5a and Table 1. According to the six classified types of adsorption isotherm of porous powder [27], the NR2 shows a typical Langmuir-type isotherm curve for “mesoporous” materials with nanopores having diameter of 2–50 nm. The large hysteresis is due to capillary condensation through bottle-neck typed mesopores. On the other hand, the curve of NR1 is similar to the typical N2 adsorption/ desorption isotherm for “non-porous” materials, although it shows slight hysteresis. These results match well with the acquired values of specific surface area, pore volume, pore size, and porosity as shown in Table 1. The porosity of NR2 (heat-treated at 150 °C) was 47.4%, which was about twice higher than the porosity (23.3%) of NR1. Such increase in the porosity and the specific surface area for NRs fabricated with higher water content is ascribed to fine nanopores formed between delicate SiO2 frames consisting of small polymer chains which are rapidly created by a quick gel reaction [25–27]. On the other hand, with increasing heat-treatment temperature of NR2 from 150 °C up to 650 °C, the pore size and the porosity were reduced from sintering effect. This also resulted in the increase of the density of NR2 due to pore shrinkage by the high temperature treatment. The observed adsorption/desorption isotherm of the NR2 at these elevated temperatures (Fig. 5b) also strongly suggests the collapse of pore structures during heattreatment showing similar patterns with the curve of NR1 having low porosity. Therefore, in addition to water/TEOS ratio, heat-treatment temperature can also be a useful variable to control the porosity and biodegradability of mesoporous NRs.
3.1. Morphology study of NRs The lengths and shapes of the nanorods were affected by impregnation time. Long impregnation time resulted in elongated and centerfilled nanorod forms, while decrease in impregnation time produced relatively shorter and hollow tube form. This is because gel polymerization started from the surfaces of top and the column walls of AAO template. The gel networks spread from these surfaces toward the inside of the nanopores of AAO templates. Therefore, long gelation time increased nanorod lengths and also effectively filled the open columnar nanopores.
Table 1 BET analysis results. H2O/TEOS Heat-treatment BET surface Pore volume Pore size Porosity (mol ratio) temp. (°C) area (m2/g) (cm3/g) (Å) (%) NR1 3.5 NR2 14 14 14
150 150 350 650
67.80 78.78 68.78 44.67
0.094 0.483 0.152 0.081
55.41 65.23 58.12 46.15
23.3 47.4 29.2 21.9
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
69
(b)
Fig. 6. FE-SEM images of biodegradation process of (a) NR1 and (b) NR2 heat-treated at 150 °C after incubation in PBS solution.
3.3. Biodegradation of NRs The SEM images in Fig. 6 show the morphology of NR1 and NR2 (both heat-treated at 150 °C) observed for three days to monitor their biodegradation process. After three days, NR1 shows relatively slow biodegradation and aggregation of each NR by a capillary force during their drying for SEM analysis. NR2 degraded at faster speed than NR1. Most NR2 nanorods collapsed and dissolved after 48 h. This difference in biodegradation rate between NR1 and NR2 is explained by the higher porosity and surface area of NR2 than NR1. This also confirms that the differences in processing conditions and the resulting microarchitectures between NR1 and NR2 have profound influence on their biodegradability.
Fig. 7 shows DT/TGA analysis results for NR1 and NR2 (both heattreated at 150 °C). Under 100 °C, both nanorods showed the weight loss of about 10% and 19%, respectively. This seems to be caused by evaporation of H2O absorbed onto the pores. The weight loss of NR2 was about twice higher than NR1. This demonstrates a higher degree of water absorption in NR2 that has higher porosity than NR1. Additional weight losses of 8% and 4%, respectively in NR1 and NR2 were observed within the range of 100 °C–600 °C. This difference is due to the evaporation/decomposition of residual organic matters. From the retarded weight loss and relatively large exothermic combustion that continues up to 600 °C, the main cause of the reduced weight loss of NR2 seems to be relatively small amount of residual non-reactive alkoxy group (− OR) in its gel state. The large
70
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
(a)
(b) (b) Fig. 7. DT-TGA analysis of (a) NR1 and (b) NR2. Fig. 8. XRD peak intensity of (a) NR1 and (b) NR2 at 2θ = 23°.
amount of H2O drastically increases the hydrolysis reaction of TEOS that changes alkoxy groups to alcohol which easily evaporates below 100 °C. As shown in Fig. 8, the change in the XRD peak intensity at 2θ = 23° of NR1 and NR2 with different heat-treatment temperatures was measured over two days of biodegradation. NR1 and NR2 heat-treated at 150, 350, and 650 °C were immersed in PBS solutions for two days for the analysis with 24-hour intervals. In both NR1 and NR2, broad peaks by amorphous silica was observed near 2θ = 23° (data not shown). On the other hand, the relative intensity observed at 2θ b 10° is larger in NR2 compared to NR1. This seems to be caused by the higher porosity of NR2. The peak intensity at 2θ = 23° decreases as biodegradation proceeds and, as seen in both cases, the change of peak intensity is dramatic in the samples heat-treated at 150 °C while negligible change is detected in those heat-treated at 650 °C. In Fig. 9, the full width at half maximum (FWHM) XRD intensities of NR1 and NR2 at each biodegradation time were measured. In PBS solutions, the FWHM values increased over time for all samples. This indicates gradual decrease of their crystallinity because the silica glass network collapsed through biodegradation. Regardless of heattreatment temperatures, the FWHM of NR2 was higher than that of NR1. This confirms that NR2 has lower crystallinity and higher porosity compared to NR1. This also explains the faster biodegradation of NR2 than NR1. Fig. 10 shows the change in the average lengths of NR2 when immersed in a PBS solution. The length decrease of NR2 heat-
treated at 150 °C was more significant compared to NR2 samples treated at 350 and 650 °C. This indicates that heat-treatment of NRs at higher temperature sintered NRs, resulting in decrease of biodegradation speed. In some samples, severe aggregation of the NR2s heat-treated at 650 °C was observed without their degradation after 3 days being immersed in a PBS solution. 3.4. Biodegradation of NaNRs The SEM images in Fig. 11 show the morphology of NaNR-2% and NaNR-5% (heat-treated at 150 °C) observed for three days monitoring the biodegradation process. Compared to NRs, NaNRs showed faster degradation and increase of Na content further accelerated the decomposition. After a day, a significant portion of NaNRs already collapsed. In particular, NaNR-5% showed the fastest biodegradation and was perfectly removed after 3 days. As shown in Fig. 12, after 3 days immersed in PBS solution at room temperature, NaNR2% and NaNR-5% lost about 28% and 37% of their weight, respectively. On the other hand, NR1 samples (without any Na inclusion) lost about 10% of its original weight. The accelerated biodegradation speed of the NaNRs is attributed to the seriously reduced chemical stability of the silica networks due to ‘scissoring’ of Si–O–Si oxide network by introducing Na2 O as an ending group, which is welldefined in the studies on silicate glasses. Furthermore, Na-
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
71
Fig. 10. Change of average length of NR2 heat-treated at 150, 350, and 650 °C when incubated in PBS solution.
incorporated silica is vulnerable to an acidic corrosion attack compared to basic or alkaline circumstances because the Na 2O ending group in silica network affords a preferential attacking point to acidic corrosion. The partially dissolved silica surface results in continuous mechanical collapse of entire silica oxide body by accelerated corrosion. 3.5. Drug release performance of NRs
(b)
DOX loading amount in NR1 and NaNR-5% was calculated by subtracting the remaining amount of DOX in the rinsed water from the total amount in the stock solution. Drug loading efficiency (loaded DOX/nanorods, w/w) was determined as 10.1% and 14.8% for NR1 and NaNR-5%, respectively. In order to understand the performance of the NRs as drug delivery carriers, drug-loaded NRs were immersed in a PBS solution and incubated at 36.5 °C. As a model drug, DOX, an anti-cancer drug, was used in this experiment. DOX, positivelycharged under physiological pH [28], was loaded based on ionic bonding with the negatively-charged surface of NRs. When biodegradation of NRs occurs, the break-down of the silica materials releases the DOX molecules attached on their surface. As shown in Fig. 13, initial burst was observed within the first 24 h and subsequently DOX was linearly released until the 8th day of incubation. It is also noteworthy that NaNR-5% released DOX faster than NR1 within the first 24 h. This conforms well to the difference in biodegradation speeds between NR1 and NaNRs, as shown in Fig. 11. Such initial burst and subsequent “zero-order” style release are desired for cancer treatments. The initial burst of anti-cancer drug can damage most of cancer cells in target tumor, while the following constant delivery of the drug at lower delivery rate can prevent any reoccurrence of cancer at the delivery site. 4. Conclusions
(c) Fig. 9. FWHM analysis of NR1 and 2 heat-treated at (a) 150, (b) 350, and (c) 650 °C.
We report a novel fabrication method to precisely control the geometry, porosity, and biodegradation rate of mesoporous silica nanorods for controlled drug delivery. Mesoporous NRs with varying porosities were fabricated by controlling H2O/TEOS ratio and heattreatment temperatures through a sol–gel process. AAO templatebased formation of mesoporous silica nanorods successfully controlled the length of NRs by adjusting the impregnation time of the precursor solution of NRs in AAO templates. Uniform mesoporous
72
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
(a)
(b)
Fig. 11. FE-SEM images of biodegradation process of (a) NaNR-2% and (b) NaNR-5% incubated in PBS solution.
nanorods were obtained with reduced cracks and shrinkage of nanorods by drying and heat-treatment processes that affected pore formation. The distribution and size of pores inside nanorods were also controlled by water quantity used in the hydrolysis step of the sol–gel process. Higher water content in the precursor solution leads to higher porosity of the nanorods and also increased biodegradability. Heat-treatment temperature also significantly affected the biodegradability of mesoporous NRs by increasing their crystallinity which attributed to fortify silica network and thus to increase resistant toward water-based corrosion attack. As a drug delivery carrier, the mesoporous NRs showed initial burst and “zero-order”
release kinetics of anti-cancer drug, which is ideal for cancer treatment. In addition, NaNRs were fabricated using AAO template method and they showed accelerated biodegradation compared to NRs. It was demonstrated that the biodegradation speed was increased with the increase of Na amount.
Acknowledgments This paper was supported by Research Fund, Kumoh National Institute of Technology.
S.B. Park et al. / Materials Science and Engineering C 50 (2015) 64–73
73
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.01.073. References
Fig. 12. Weight change comparison of NR1, NaNR-2%, and NaNR-5% in PBS solution (pH 7).
Fig. 13. DOX release test in vitro on NR1 and NaNR-5% for 8 days.
[1] J.C. Ro, I.J. Chung, J. Non-Cryst. Solids 130 (1991) 8. [2] R.K. Iler, The Chemistry of Silica: Solubility, Polymerization, Colloid and Surface Properties and Biochemistry of Silica, John Wiley & Sons, New York, 1979. [3] H. Böttcher, P. Slowik, W. Suss, J. Sol-Gel Sci. Technol. 13 (1998) 277. [4] C. Chiappini, X. Liu, J.R. Fakhoury, M. Ferrari, Adv. Funct. Mater. 20 (2010) 2231. [5] L. Vaccari, D. Canton, N. Zaffaroni, R. Villa, M. Tormen, E. di Fabrizio, Microsc. Electron. Eng. 83 (2006) 1598–1601. [6] I.I. Slowing, J.L. Vivero-Escoto, C.-W. Wu, V.S.-Y. Lin, Adv. Drug Deliv. Rev. 60 (2008) 1278–1288. [7] K.M.L. Taylor, J.S. Kim, W.J. Rieter, H. An, W. Lin, W. Lin, J. Am. Chem. Soc. 130 (2008) 2154–2155. [8] K.-W. Hu, K.-C. Hsu, C.-S. Yeh, Biomaterials 31 (2010) 6843–6848. [9] C.J. Brinker, G.W. Scherer, Sol–Gel Science, Academic Press, 1990. [10] H. Böttcher, C. Jagota, J. Trepte, K.-H. Kallies, H.J. Haufe, Control. Release 60 (1999) 57. [11] M. Ahola, P. Kortesuo, I. Kangasniemi, J. Kiesvaara, A. Yli-Urpo, Int. J. Pharm. 195 (2000) 219. [12] N.K. Mal, M. Fujiwara, Y. Tanaka, Nature 421 (2003) 350. [13] S. Sakka, K. Kamiya, J. Non-Cryst. Solids 48 (1982) 31. [14] I. Strawbridge, A.F. Craievich, P.F. James, J. Non-Cryst. Solids 72 (1985) 139. [15] B.E. Yoldas, J. Polym. Sci. 24 (3475) (1986) 3487. [16] C.J. Brinker, K.D. Keefer, D.W. Shaefer, C.S. Ashley, J. Non-Cryst. Solids 48 (1982) 47. [17] C.J. Brinker, K.D. Keefer, D.W. Shaefer, C.S. Ashley, J. Non-Cryst. Solids 63 (1984) 50. [18] J.Y. Ying, J.B. Benziger, J. Non-Cryst. Solids 147/148 (1992) 222. [19] I. Strawbridge, A.F. Craievich, P.F. James, J. Non-Cryst. Solids 72 (1985) 139. [20] L.C. Klein, G.J. Garvey, in: C.J. Brinker, D.E. Clark, D.R. Ulrich (Eds.),Better Ceramics Through Chemistry III, Mater. Res. Soc. Symp. Proc., 32, 1984, p. 33. [21] C.J. Brinker, K.D. Keefer, D.W. Shaefer, R.A. Assink, B.D. Kay, C.S. Ashley, J. Non-Cryst. Solids 63 (1984) 50. [22] C.J. Brinker, K.D. Keefer, D.W. Shaefer, C.S. Ashely, J. Non-Cryst. Solids 48 (1982) 47. [23] N. Nogami, Y. Moriya, J. Non-Cryst. Solids 37 (1980) 191. [24] L.C. Klein, Sol–Gel Technology for Thin Films, Fibers, Preforms, Electronics and Specialty Shapes, 389, Noyes Publication, NJ, U.S.A., 1988 [25] I. Strawbridge, A.F. Craievich, P.F. James, J. Non-Cryst. Solids 72 (1985) 139. [26] J.-H. Lee, W.J. Kim, J. Lee, J. Korean Ceram. Soc. 34 (2) (1997) 216. [27] S. Brunauer, L.S. Deming, W.S. Deming, E. Teller, J. Am. Ceram. Soc. 62 (1940) 1723. [28] H. Meng, M. Liong, T. Xia, Z. Li, Z. Ji, J.I. Zink, A.E. Nel, ACS Nano 4 (8) (2010) 4539–4550.
