Modification of Mesoporous Silica Surface by Immobilization of Functional Groups for Controlled Drug Release

&is paper introduces the synthesis of mesoporous silica nanoparticles (MSNs) with three different groups such as amine, thiol, and sulfonic acid, along the internal surface. Trimethyl[3-(trimethoxysilyl)propyl]ammonium chloride was used to modify the external surface of the nanomaterials. Such materials allow control of the drug release from MSN pores. Multifunctional MSNs were loaded with doxycycline (Doxy) to study their capacities and uploading time. &e loading profile indicates that sulfonic groups in the internal surface were the most efficient surfaces with a loading capacity of ca. 35% in 90min in acidic media.

3-Aminopropyltriethoxysilane (APTES) and 3-mercaptopropyltrimethoxysilane (MPTMS) have been widely utilized to modify silica surface with amine and mercapto groups, respectively, and used for several applications [17,19,20]. Huang et al. have prepared a catalyst based on modified mesoporous silica coated with MPTMS. iol groups on the surface were converted to sulfonic acid and used as a catalyst for the esterification reaction [21]. Mosquera et al. reported the synthesis of gold nanoparticles stabilized with silica monolayers. e silica monolayer was formed after MPTMS immobilization on the surface of gold nanoparticles and hydrolysis in basic media [22]. Shen et al. have synthesized PEGylated capped MSNs and covered with MPTMS, to be used as a gatekeeper [23]. Cheng et al. reported the functionalization of MSNs with pH-sensitive trimethylammonium as pH-controllable drug release [24]. Xia et al. reported the preparation of MSNs coated with amine and encapsulated doxorubicin inside their pores and then sealed the external surface with cellulose through esterification [25]. Liu and coworkers reported the synthesis of MSNs coated with amine and used to load with doxorubicin (Dox) [26]. Chen et al. reported the preparation of mesoporous silica modified with amine for loading and releasing Dox using a capping concept. e drug can be released with control minor by detachment of the capped nanoparticles that block the MSN pores [27]. Shao et al. reported the preparation of Janus silvermesoporous silica nanoparticles (Ag-MSNs) as nanocarriers [28]. It was found that Dox-loaded Ag-MSNs were pHresponsive drug release, leading to inhibit the growth of cancer cell, rather than human normal cells. Wang et al. engineered Janus-structured gold-mesoporous silica nanoparticles and modified with folic acid for enhancing HCC targeting and internalization [29]. Dox release from loaded nanosystem was studied from the mesopores in a pH-responsive manner. Dong and coworkers have reported the synthesis of Janus magnetic head with Dox-loaded mesoporous silica as "nanobullets" [30,31]. Such nanoparticles were used to inhibit the growth of cancer cell, due to pHpromoted Dox release in the interior of cancer cells. MSNs with large internal pores have been prepared for the controlled delivery of biomacromolecules (protein and nucleic acid) with low toxicity [32].
Cheng et al. developed a multifunctional nanosystem based on Dox-gated MSNs for combined tumor chemotherapy and photothermal and gene therapies [33].
is nanoplatform could specifically target cancer cells, minimizing damage to the normal cells. A new drug delivery system based on d-a-tocopheryl polyethylene glycol 1000 succinate (TPGS)-functionalized polydopamine-coated MSNs was developed by Cheng et al. [34]. Polydopamine-coated MSNs could degrade at low pH, leading to release Dox to damage cell lung cancer. Dual pH and glucose sensitivity nanocarrier were designed to control the drug release via the detection of glucose and pH levels [35]. Oxaliplatin prodrug (Oxa(IV)) encapsulated in TPGS functionalized MSNs was synthesized to release the drug in acidic media in cancerous tissue and the reductants in cancer cells [36]. A facile drug-selfgated MSN nanocarrier via a pH-sensitive benzoic-imine covalent bond was demonstrated by Zeng et al. [37]. At weak acidic cancerous cells, loaded Dox could be released due to the hydrolyzation of benzoic-imine bonds. Zhang et al. fabricated bioresponsive controlled delivery nanoplatform based on the assembly and disassembly of gold nanoparticles from the MSN surface [38]. Drug could be trapped inside the nanopores by AuNPs functionalized with aptamer, and the drug can be released in the presence of adenosine molecules.
Li and coworkers reported the synthesis of magnetic MSNs with high drug-loading amounts and pH-responsive control drug release. Dox-loaded magnetic MSNs could selectively damage the cancer cells in vitro [39]. Yue et al. have prepared MSNs and mesoporous organosilica nanoparticles (ss-MONs) as nanocarriers for loading Dox in liver cancer cell treatment [40]. e ss-MONs-Dox showed better anticancer efficacy than MSNs-Dox, due to that ss-MONs have good degradability with reductive glutathione, leading to the enhancement of Dox release. Chen et al. have recently reported the synthesis of pH/glutathione (GSH) dual-sensitive nanoplatform based on silica nanoparticle for targeting cancer cells [41]. e novel nanocarrier was pH-sensitive in the weakly acidic cancerous cells.
In this work, mesoporous silica nanoparticles (MSNs) were synthesized with amine, thiol, and sulfonic groups along the inner surface and with high surface area and average pore size of ca. 5 nm.
ereafter, MSN external surfaces were modified with trimethyl[propyl]ammonium chloride groups. All materials were loaded with doxycycline (Doxy) to study their capacities and uploading time. Amine, thiol, and sulfonic groups in the internal surfaces may allow precise control of the drug release from MSN pores. is hypothesis is examined using BET, FT-IR, and UV-vis.

Synthesis of Mesoporous Silica Nanoparticles (MSNs).
In a 250 ml flask, CTAB (1 g) was dissolved in 155 mL deionized water and stirred for 10 min, and then, NH 4 OH (7 mL) was added to the mixture. A mixture of n-hexane (20 mL) and TEOS (5 mL) was added to the reaction under stirring and left overnight at 35°C. Prepared nanoparticles were separated and washed with deionized water (3 times) and ethanol (3 times) and then dried in an oven at 110°C.

Synthesis of Mesoporous Silica Nanoparticles with
Amine in the Pores (MSNs-NH 2 ). In a 250 ml flask, CTAB (1 g) was dissolved in 155 mL deionized water and stirred for 10 min, and then, NH 4 OH (7 mL) was added to the mixture. A solution contained n-hexane (20 mL), APTES (0.2 mL), and TEOS (5 mL) was added to the reaction under stirring and left overnight at 35°C. Prepared nanoparticles were separated and washed with deionized water (3 times) and ethanol (3 times) and then dried in an oven at 110°C.

Synthesis of Mesoporous Silica Nanoparticles with iol in the Pores (MSNs-SH).
In a 250 ml flask, CTAB (1 g) was dissolved in 155 mL deionized water and stirred for 10 min, and then, NH 4 OH (7 mL) was added to the mixture. A solution contained n-hexane (20 mL), MPTMS (0.2 mL), and TEOS (5 mL) was added to the reaction under stirring and left overnight at 35°C. Prepared nanoparticles were separated and washed with deionized water (3 times) and ethanol (3 times), and then, dried in an oven at 110°C.

Functionalization of the External Surface of MSNs with Trimethyl[propyl]ammonium Chloride Group (TMPAC-MSNs).
MSNs (1 g) were suspended in 40 mL methanol containing TMSPAC (2 mmol). e reaction mixture was refluxed overnight (Scheme 1(a)). e product was separated by centrifugation, washed 5 times with methanol, and dried under vacuum. e template was removed using a solution of ammonium nitrate in ethanol (10 mg/mL) at 85°C for 12 h under stirring. e nanoparticles were centrifuged, washed 5 times with ethanol, and dried in an oven.

Functionalization of the External Surface of MSNs-NH 2 with Trimethyl[propyl]ammonium Chloride Group
(TMPAC-MSNs-NH 2 ). 1 g of MSNs-NH 2 was suspended in 40 mL methanol containing TMSPAC (2 mmol). e reaction mixture was refluxed overnight (Scheme 1(b)). e product was separated by centrifugation, washed 5 times with methanol, and dried under vacuum. e template was removed using a solution of ammonium nitrate in ethanol (10 mg/mL) at 85°C for 12 h under stirring. e nanoparticles were centrifuged, washed 5 times with ethanol, and dried in an oven.

Functionalization of the External Surface of MSNs-SH with
Trimethyl [propyl]ammonium Chloride Group (TMPAC-MSNs-SH). 1 g of MSNs-SH was suspended in 40 mL methanol containing TMSPAC (2 mmol). e reaction mixture was refluxed overnight (Scheme 1(c)). e product was separated by centrifugation, washed 5 times with methanol, and dried under vacuum. e template was removed using a solution of ammonium nitrate in ethanol (10 mg/mL) at 85°C for 12 h under stirring. e nanoparticles were centrifuged, washed 5 times with ethanol, and dried in an oven.

Functionalization of the External Surface of MSNs-SO
). 1 g of MSNs-SH was suspended in 40 mL methanol containing TMSPAC (2 mmol). e reaction mixture was refluxed overnight (Scheme 1(d)). e product was separated by centrifugation, washed 5 times with methanol, and dried under vacuum. e template was removed using a mixture of acetic acid and hydrogen peroxide under refluxing for 12 h under stirring. e final product was centrifuged, washed with ethanol, and dried in an oven.

Doxycycline (Doxy) Loading and pH-Triggered Release.
To determine the loading capacity and efficiency of encapsulation at different times, each prepared nanoparticle was dispersed in deionized water with a concentration of 1 mg/mL. Doxy was added to the suspension with a concentration of 0.5 mg Doxy per mg of fabricated nanoparticles, and the pH of the resulting mixture was adjusted to 6.5 or 7.4. Each mixture was shaken at room temperature for 5, 10, 20, 30, 40, 60, 120, 180, 240, 360, or 480 min. After each time period, the mixtures were centrifuged and washed two times with 1 cm 3 buffer solutions. All fragments of each  experiment were combined to determine the Doxy concentration in the final solution by the colorimetric method. e unloaded Doxy concentration was calculated on the basis of a calibration curve of Doxy in buffer.
Doxy-loaded nanoparticles of each type were suspended in (pH � 3, pH � 5, pH � 6.5, pH � 7.4, and pH � 8.5) solution at a concentration of 0.25 mg NPs/mL. At a certain time, the nanoparticles were separated, and aliquot supernatant was removed and replaced with an equal volume of buffer solution. e amount of Doxy released was determined based on the calibration curve. e efficiency of encapsulation (EE) and loading capacity (LC) was calculated as follows: weight of loaded drug weight of nanoparticles × 100, EE(%) � weight of loaded drug weight of drug in feed × 100. (1)

Results and Discussion
Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were utilized to study the morphology of the prepared samples. SEM image shows that mesoporous silica is nearly spherical nanoparticle, with an average particle size of ca. 200 nm, as shown in Figure 1(a). TEM image reveals that prepared samples are dispersed without aggregation, with a diameter ranging from 150 to 250 nm, which shows a good agreement with the SEM image (Figure 1(b)). Highly ordered array pores were observed from the TEM image, with a pore size of ca. 5 nm. No significant differences in SEM and TEM images were observed between nonmodified and modified MSNs, which is similar to the observation reported previously [42] ( Figures 1S and 2S). Furthermore, the particle size distribution was investigated using dynamic light scattering (DLS). Although some particles had sizes around 400 nm, the majority ranged between 100 and 300 nm. e average particle size was 220 nm for all fabricated nanomaterials ( Figure 3S).
is shows a good agreement with the size estimated by SEM and TEM.
e physicochemical properties of all the samples are presented in Table 1. Generally, there were reductions in surface area, and pore volumes were noted for functionalized MSNs. e average pore size was between 5 and 6 nm for all prepared samples, which agrees with the pore size that observed in the TEM image.
FT-IR was utilized to characterize the bare MSNs, TMPAC-MSNs, TMPAC-MSNs-NH 2 , TMPAC-MSNs-SH, and TMPAC-MSNs-SO − 3 , as shown in Figure 3. It can be seen that wide bands at 1250-1050 cm −1 were ascribed to siloxane groups (Si-O-Si) bands, and peak at 800 cm −1 was ascribed to Si-O stretching vibration. Peak was noticed in all prepared samples at 1650 cm −1 , which was ascribed to water bending vibration. e absorption peaks of surface-modified nanoparticles were noted at 1490 cm −1 , which were ascribed to C-H as bending vibration in -CH 2 -groups, and peaks around 2930 cm −1 were assigned to C-H as stretch vibration. ese characters provide an evidence of the successful functionalization of surface MSNs.
Elemental analysis was used to estimate the molecules attached to the surfaces of bare MSNs, TMPAC-MSNs, MSNs-NH 2 , TMPAC-MSNs-NH 2 , MSNs-SH, TMPAC-MSNs-SH, MSNs-SO − 3 , and TMPAC-MSNs-SO − 3 using the measured percentages of carbon, hydrogen, nitrogen, and sulfur (Table 2). e results clearly show an increase in the percentage of N, C, S, and H elements on the modified MSNs compared to bare MSNs, which confirms the presence of the functional groups in internal and external surfaces of MSNs.
Doxy is a tetracycline antibiotic that is widely used to treat serious bacterial infections, such as cholera, typhus, and syphilis. To the best of our knowledge, little work focuses on the interaction between Doxy and the internal surface of mesoporous silica nanoparticles. e main goal of this study is to investigate the effect of selected functional groups in the inner surface on the drug release. To minimize the external surface effect, the outer surface of all fabricated nanoparticles was functionalized with TMPAC. Loading and 4 Journal of Chemistry    Journal of Chemistry release of Doxy were investigated as drug delivery vectors for TMPAC-MSNs, TMPAC-MSNs-NH 2 , TMPAC-MSNs-SH, and TMPAC-MSNs-SO − 3 . Doxy was loaded in these different MSNs samples at pH of 6.5 and 7.4 and at different loading time to allow Doxy to interact with the inner surfaces of the MSNs (Figure 4). For TMPAC-MSNs, TMPAC-MSNs-NH 2 , and TMPAC-MSNs-SH, the loading capacity dramatically increased to ca. 20%, in 60 min, and then remains at 25%, after 90 min for TMPAC-MSNs and TMPAC-MSNs-NH 2 at both pH 6.5 and 74. However, in 60 min, the loading capacity of TMPAC-MSNs-SO − 3 increased to ca. 20% at pH 7.4 and 30% at pH 6.5 and then remains at 35%, after 90 min. e enhanced encapsulation efficiency of TMPAC-MSNs-SO − 3 compared to other fabricated materials indicates that Doxy (positive) could interact physically to the interior surfaces of TMPAC-MSNs-SO − 3 . e encapsulation efficiency for all fabricated materials was ca. 40% at different pH, except TMPAC-MSNs-SO − 3 . e encapsulation efficiency of MSNs with sulfonic groups in the internal surface was ca. 60% in acidic media. e surface charge of all prepared nanoparticles was found to be around 40 mV at pH 7.4, due to the surface functionalization with TMPAC (permanent positively charged) on the external surface of MSNs. It was noted that Doxy-loaded MSNs had similar surface zeta potential of unloaded MSNs ( Figure 4S).
e Doxy release profile of the fabricated nanoparticles was investigated in mild acidic (3, 5, and 6.5) and basic (7.4 and 8.5) pH conditions at room temperature ( Figure 5). e unmodified inner surfaces of MSNs material present clear drug release behaviour ( Figure 5(a)). At pH higher than 5, about 20% release was noticed after an incubation of 72 hours. e drug release was higher at pH 3, to ca. 25%, which may be assigned to the physical adsorption between the drug and the inner surfaces of the samples. e release profile of  Journal of Chemistry TMPAC-MSNs-NH 2 exhibited a constant release of ca. 35% at pH 8.5, after a 72-hour incubation period ( Figure 5(b)). Upon decreasing the pH to 5, 6.5, and 7.4, approximately 40% of Doxy was released after a 72-hour incubation period. However, almost 50% of the drug was released when the Doxy-loaded TMPAC-MSNs-NH 2 exposed to pH 3. e observation may be assigned to the attractive and repelling forces between the amino-modified internal surface and Doxy.
e Doxy release profile of TMPAC-MSNs-SH is presented in Figure 5(c). No difference in the release behaviour was observed when the pH was changed. is results that the interaction behaviour between the thiol group and Doxy is not affected by the pH. Yet, when the thiol group was converted to sulfonic groups, Doxy release was ca. 55% after 72 hours at pH 7.4 and 8.5. In acidic media, there was a reduction in Doxy release due to the physical interaction between the surface and the drug.

Conclusions
In summary, we reported the preparation of MSNs different functional groups in the internal surface, and high surface area of ∼1000 m 2 /g, large pore volume of 1.2 cm 3 /g, and average pore size of ∼5.5 nm. Such materials were covered with cationic groups by attaching trimethyl [3-(trimethoxysilyl)propyl]ammonium chloride in the external surface. e characterizations via TEM, FT-IR, BET, and elemental analysis confirmed that the MSNs were prepared Journal of Chemistry 7 successfully.
e loading capacity of TMPAC-MSNs-SO − 3 was found to be the highest with 30% at pH 6.5. Furthermore, the drug release was ca. 55% in a relatively controlled manner.

Data Availability
All data used to support the findings of the study are available within the article and from the corresponding author.

Conflicts of Interest
e author declares no conflicts of interest.
Acknowledgments e author extends this appreciation to the Deanship of Scientific Research at King Saud University for funding this work through the research project.