(APTES) Silanization of Superparamagnetic Iron Oxide Nanoparticles

52 downloads 0 Views 384KB Size Report
Nov 20, 2013 - ABSTRACT: Silanization of magnetic ironoxide nanoparticles with ... including silica,1,2 ceramics,3 titania,4 and magnetic iron oxide.
Article pubs.acs.org/Langmuir

Kinetics of (3-Aminopropyl)triethoxylsilane (APTES) Silanization of Superparamagnetic Iron Oxide Nanoparticles Yue Liu,† Yueming Li,‡ Xue-Mei Li,*,† and Tao He*,† †

Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai, China 201210 State Key Laboratory of Metastable Materials Science and Technology, College of Materials Science and Engineering, Yanshan University, Qinhuangdao 066004, China



S Supporting Information *

ABSTRACT: Silanization of magnetic ironoxide nanoparticles with (3-aminopropyl)triethoxylsilane (APTES) is reported. The kinetics of silanization toward saturation was investigated using different solvents including water, water/ethanol (1/1), and toluene/methanol (1/1) at different reaction temperature with different APTES loading. The nanoparticles were characterized by Fourier transform infrared spectroscopy, vibrating sample magnetometry, transmission electron microscopy, and thermal gravimetric analysis (TGA). Grafting density data based on TGA were used for the kinetic modeling. It is shown that initial silanization takes place very fast but the progress toward saturation is very slow, and the mechanism may involve adsorption, chemical sorption, and chemical diffusion processes. The highest equilibrium grafting density of 301 mg/g was yielded when using toluene/methanol mixture as the solvent at a reaction temperature of 70 °C.



INTRODUCTION The (3-aminopropyl)triethoxylsilane (APTES) compound is an important silane coupling agent and is a widely used grafting agent to promote interfacial behavior of inorganic oxides including silica,1,2 ceramics,3 titania,4 and magnetic iron oxide nanoparticles (MNPs).5 APTES surface-functionalized MNPs in particular found applications in many biological applications including cell separation6 and enzyme separation, diagnosis as MRI contrast agent,7−9 as magnetically controlled drug carriers,10 and hyperthermia treatment media as illustrated in many literature reports.2,11−19 The mechanism of surface functionalization is the formation of Fe−O−Si bond between the nanoparticles and silane ligand, very similar to the siloxane layer formation on silica. However, accurate control of the layer quality has been a difficult task.8,11,20−22 Silane attachment may exist in physisorption, hydrogen bonding, or electrostatic attachment besides covalent linkage.20 Even for covalently linked ligands, complications may be present in monolayer and multilayer formation.23 The general silanization sequence follows by the hydrolysis of triethoxyl groups into trihydroxyl groups and then polycondensation of the hydroxyl groups with the surface © 2013 American Chemical Society

hydroxyls groups from the MNP surface. However, several reaction routes may be involved, which makes the grafting density of the silane ligands highly dependent on the reaction conditions as illustrated in Scheme 1. For example, after hydrolysis the silanol groups may react with surface hydroxyl groups leading to surface silanization. The surface-immobilized ligands then condense with vicinal silanol groups forming either a monolayer or highly branched polycondensed structure. Overall, the kinetics of the silanization is critical for controlling the layer formation process.11 Layer formation kinetics of octadecyltrichlorosilane (ODS)24 and octyltriethoxyl silane (OTS) has been studied widely.25−28 It is generally believed that the silanization is fast with that of the reaction kinetics following the Langmuir model in the short time range leading to a self-assembled monolayer and the process toward saturation is of slow kinetics.27,29 For APTES silanization process in particular, there have been many studies on the molecular structure,20 type of bonding,21 and surface Received: August 23, 2013 Revised: November 20, 2013 Published: November 20, 2013 15275

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282

Langmuir

Article

Scheme 1. Schematic Illustration of Possible Reaction Routes for APTES Silanization of Magnetic Nanoparticles

Table 1. The Silanization Operating Conditions for MNPs entrya

solvent

T (°C)

initial APTES conc. (g/L)

MNPw30−2% MNPw30−0.2% MNPw70−2% MNPw70−0.2% MNPM30−2% MNPM30−0.2% MNPM70−2% MNPM70−0.2% MNPE30−2% MNPE30−0.2% MNPE70−2% MNPE70−0.2%

H2O H2O H2O H2O MeOH/toluene MeOH/toluene MeOH/toluene MeOH/toluene EtOH/H2O EtOH/H2O EtOH/H2O EtOH/H2O

30 30 70 70 30 30 70 70 30 30 70 70

20 2 20 2 20 2 20 2 20 2 20 2

reaction time (min) 60, 120, 300, 600, 60, 120, 300, 600, 60, 120, 300, 600, 60, 600, 60, 600, 60, 600, 60, 600, 60, 600, 60, 600, 60, 600, 60, 600, 60, 600,

1440 1440 1440 1440 1440 1440 1440 1440 1440 1440 1440 1440

equilibrium grafting density (De, mg/g)b 243.9 96.6 275.5 248.1 259.1 198.8 301.2 278.6 243.9 209.6 281.7 263.2

a

The subscripts are denoted as the reaction conditions. For example, MNPw70−0.2% refers to aminated MNP prepared using water as the solvent at 70 °C reaction temperature and 0.2% (v/v) APTES loading. bResults from data fitting for the equilibrium grafting density (De) by the pseudo-second order kinetics.

derivatization conditions30 but the layer formation kinetics has not been reported to the authors’ knowledge. Moreover, the surface coverage of APTES for MNP has not been reported, nor the complete plot of reaction kinetics. The understanding of the kinetics of APTES silanization would help to clarify whether a monolayer is formed and eventually the quality control of the layer.11,21

In this report, surface functionalization of MNP with APTES is investigated under different reaction conditions in order to reveal the layer formation kinetics, especially the kinetics in the process toward saturation. The maximum grafting density and reaction constants are estimated based on the kinetic model. The use of different solvents was carried out in order to assess solvent effects on the reaction kinetics. The results may help 15276

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282

Langmuir

Article

Figure 1. TEM images of MNPs (left) and MNP-NH2 (right). kOe. Specific surface area was calculated via the Brunauer−Emmett− Teller (BET) method according to nitrogen adsorption−desorption measurements performed at 77 K on a Micromeritics ASAP 2020 adsorption analyzer.

the quality control of the silane layer in terms of reaction time, temperature, and solvent selection.



EXPERIMENTAL SECTION



Materials. FeCl3·6H2O and FeCl2·4H2O were both purchased from Sigma-Aldrich. Sodium hydroxide was purchased from Longxi Chemicals (Shantou), China. The (3-Aminopropyl)-triethoxysilane (APTES) compound was purchased from Alfa Aesar. All the organic solvents were of reagent grade and used as received unless stated otherwise. Doubly distilled water was used throughout the experiments. Synthesis of MNP Iron Oxide Nanoparticles. Water was deoxygenated by bubbling N2 for 30 min. A stock solution of NaOH was prepared by dissolution of NaOH (23.08 g, 0.57 mol) in water (1 L). FeCl3·6H2O (5.4 g, 20 mmol) and FeCl2·4H2O (4.0 g, 20 mmol) were dissolved in 80 and 120 mL distilled water, respectively. The molar ratio of Fe3+/Fe2+ was 1:1. The two solutions were mixed together and added into the alkaline solution prepared above slowly under vigorous stirring. The mixture was heated to 80 °C and stirred for 30 min under N2. The black precipitates were washed 3 times with distilled water followed by magnetic decantation and dried under vacuum. Silanization of MNPs by APTES. For optimization of the surface coating process of the silanization, different solvent and reaction temperature were attempted as shown in Table 1. When water-based solvent was used, the reaction procedure is as follows. Briefly, the nanoparticles obtained from coprecipitation (0.5 g) were dispersed in the solvent (100 mL) and then APTES (0.2 or 2 mL) was added into the mixture under mechanical stirring. The reaction mixture was heated up to 30 or 70 °C for a specified time then cooled down to room temperature followed by magnetic decantation and thorough washing with ethanol. The samples were then dried under rotary evaporation and subsequently in an oven (60 °C) for 24 h. When a mixture of methanol/toluene (volume ratio, 1:1) was used as the solvent, the procedure is as follows. The nanoparticles were dispersed in 100 mL of the solvent, sonicated for 30 min, and heated up to 95 °C till 50 mL of the solvent was evaporated. Thereafter, methanol (50 mL) was added. The operation was repeated three times to ensure the solution was completely anhydrous. Subsequently, the volume of reaction mixture was fixed to 100 mL by the addition of methanol followed by APTES addition and stirred for a certain time at 30 or 70 °C under N2. The product was washed several times with anhydrous ethanol and distilled water by magnetic decantation and dried under vacuum. Characterization. Infrared spectra of the nanoparticles were recorded on an AVA TAR 360 Fourier transform infrared spectroscopy (Thermo Nicolet) by pressing the MNPs into a KBr pellet. The surface grafting density of the dry nanoparticles was determined by thermal gravimetric analysis (TGA) (DSC204, NETSCH) under a N2 atmosphere at a heating rate of 10 K/min. The heating temperature range was from room temperature to 800 °C. Morphology of the nanoparticles was characterized by field emission scanning electron microscopy (FESEM) (S4800, Hitachi) and transmission electron microscopy (TEM) (JEM-2010UHR, Joel). Magnetic properties of the MNP were recorded on a vibrating sample magnetometry (VSM, EV7, ADE) and carried out at room temperature in magnetic fields up to 10

RESULTS AND DISCUSSION MNP Preparation and Silanization. Fe3O4 MNPs were prepared via coprecipitation of Fe2+ and Fe3+ ions (molar ratio 1:1) with an average diameter of 9 nm (Figure 1) as determined by TEM. The silanization of MNP with APTES is schematically shown in Scheme 1 with ideal conditions being that a monolayer is formed (route 1). The resulting MNP is denoted as MNP-NH2. The reaction routes depicted in Scheme 1 include APTES hydrolysis and surface attachment by the formation of Fe−O−Si bonds, leading to monolayer (route 1) or multilayer formation (Route 2). Ionic bonding may also be present which is not illustrated. Several (mixture) solvents were investigated with respect to reaction temperature and APTES loading as illustrated in the following paragraphs. All the reaction conditions explored are shown in Table 1. The subscripts in letters, numbers and percentage are referred as the solvents (with W, M, and E representing water, methanol/toluene mixture, and ethanol/water mixture, respectively), reaction temperature, and initial APTES charging percentage with respect to the solvent (v/v). Water as the Solvent. Water was used as the solvent because the MNPs are well dispersible in water. FT-IR was used to verify the successful surface modification, where MNPs after surface modification showed a similar spectrum to that of APTES (Supporting Information Figure S1). Vibrating sample magnetometry was used for the characterization of the magnetization of the MNPs (Supporting Information Figure S2). The saturation magnetization (Ms) of MNP was 50 emu/g for uncoated MNPs and 40 emu/g for MNPw70−2% with no remnant magnetization indicating that the both MNPS are superparamagnetic. X-ray diffraction patterns of the MNPs showed typical Fe3O4 diffraction peaks at 35.8° corresponding to the (311) planes (Supporting Information Figure S3) before and after surface modification indicating surface functionalization did not change the crystal structure of nanoparticles. Thermal gravimetric analysis was used to quantify and compare the degree of silanization at different reaction time, temperature, and APTES charging. Thermal gravimetric data were converted to silane grafting densities (D g ) by using independent calculation for molecular weight (Mw) of the removed component (NH2C3H6) during TGA treatments. It should be noted that TGA does not allow quantifying the composition of the layer growing on the nanoparticles but rather the amount of material grown on the particles. Because no other silane ligands were present, TGA was used to quantify the amount APTES grown on the MNPs. It is assumed that all 15277

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282

Langmuir

Article

the silane ligands were fully hydrolyzed because it would not be possible to estimate the contribution of partially hydrolyzed ones bearing ethoxyl (OEt) moieties. It should be noted that the weight loss below 200 °C was not taken into consideration for grafting density determination in order to minimize effects of solvent and ligand incorporation or physisorption (Supporting Information Figure S4 shows a typical TGA curve). APTES silanization of MNPs were carried out at 30 and 70 °C, each with two different APTES loadings (0.2 and 2%), respectively at different reaction time. The Dg analyzed by thermal gravimetric analysis was plotted against the reaction time as illustrated in Figure 2. For all the reactions, Dg increased

Figure 2. Grafting density change of the MNP-NH2 nanoparticles under different reaction conditions using water as the solvent. (□) 70 °C, 2% APTES; (○) 70 °C, 0.2% APTES; (△) 30 °C, 2% APTES; (▽) 30 °C, 0.2% APTES.

Figure 3. The kinetics of MNP nanoparticles silanization with APTES using water as the solvent as fitted by pseudo- second order model (a) and Elovich model (b).

kinetics describes a chemical adsorption process. Elovich model describes the second order kinetics by assuming the actual surfaces are heterogeneous for chemical sorption, which gives correlation coefficients ranged from 0.69 to 0.98. The intraparticle diffusion model with Dt versus t1/2 resulted in fittings with r2 lower than 0.95 (shown in Supporting Information Figure S6). If intraparticle diffusion occurs, then Dt versus t1/2 will be linear and the line will pass through the origin when intraparticle diffusion is the only rate limiting process. According to Supporting Information Figure S5, it appears that the fitting indeed is linear but does not go through the origin. This indicates that the silanization mechanism is complex and that adsorption, chemical sorption, and intraparticle diffusion models all contribute to the rate determining steps. The pseudo-second order and Elovich kinetic equations are shown in Table 2, where Dt and De, are grafting density at reaction time (t) and at equilibrium respectively. In every regression, the sum of error squared (SSE) between the predicted values (Dcalc) and the experimental results (Dexp) was calculated according to

significantly within the first hour of reaction time. For all reactions, the initial 1 h of reaction contributed the most of the grafting density increase, especially for the reactions carried out at higher reaction temperatures (70 °C). This finding agrees with literature in that silanization is a fast process and over 90% of saturation is achieved within the first hour of reaction.29 However, it is observed that at lower reaction temperature, the initial grafting density is highly dependent on the initial silane concentration. For example, at 30 °C the initial APTES grafting density was very low, only 64 mg/g at 0.2% of APTES loading; even after 24 h of reaction, it was only 95 mg/g. However, at 2% APTES loading the initial grafting density was 224 mg/g and reached to 247 mg/g indicating that initial silane concentration is a key factor for the silanization reaction. On the other hand, the reaction temperature plays also important roles. At 0.2% APTES loading, 70 °C initial 1 h of reaction yielded a Dg of 236 mg/g, much higher than that at 30 °C. It is interesting though to compare the Dg change with time of MNPw30−2% and MNPw70−0.2%. Although the initial Dg of MNPw70−0.2% was higher than MNPw30−2%, the final Dg was comparable to each other, indicating that temperature is overall a more critical factor in the silanization processes. The kinetics of the silanization reaction in water was assessed on the basis of the amount of deposited silane (Dt, mg/g) as a function of reaction time (t, min). Several kinetic models were used to fit our experimental data including Lagergren pseudofirst order (Supporting Information Figure S5), pseudo-second order (Figure 3a), Elovich (Figure 3b) and intraparticle diffusion (Supporting Information Figure S6) models. Lagergren pseudo-first order model did not yield a linear fitting and therefore was not considered further. Pseudo-second order fitting (Figure 3a) gave results with correlation coefficient (r2) in the range of 0.9973−0.9999. Pseudo second order

N

SSE =

∑i = 1 (Dexp − Dcal )2

(1) N where N is the number of experiments. The pseudo second order kinetic constant was calculated from the intercept and slope of the plots of t/Dt versus t. The De was determined as the inverse of the slope of the regression. The rate constant is determined by slope2/intercept. The rate constants were very low as seen in Table 2. For MNPw‑70−2%, the equilibrium grafting density was 275 mg/g as the highest and 96.6 mg/g for MNPw‑30−0.2% as the lowest, following the Pseudo-second order model. Comparison of the experimental results with the data 15278

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282

Langmuir

Article

Table 2. Kinetic Parameters of Different Models for APTES Silanization pseudo-second ordera

Elovichb

t/(Dt) = [(1/k2De) + t/Dt] MNP MNPw30−2% MNPw30−0.2% MNPw70−2% MNPw70−0.2% MNPM30−2% MNPM30−0.2% MNPM70−2% MNPM70−0.2% MNPE30−2% MNPE30−0.2% MNPE70−2% MNPE70−0.2%

De (mg/g) 244.0 97.2 274.9 247.9 259.1 198.8 301.2 278.6 243.9 209.6 281.7 263.2

−1

k2 × 10 (g mg 4

13.22 2.8 4.16 8.54 1.62 4.67 3.19 10.70 14.0 12.1 3.76 6.8

−1

min )

Dt = (1/β)ln(αβ) + (1/β)ln(t) R

2

0.9974 0.9999 0.9999 0.9999 0.9990 0.9996 0.9992 0.9951 0.9997 0.9999 0.9999 0.9998

SSE 2.81 1.18 2.21 1.19 10.64 2.27 4.83 0.94 2.92 0.98 6.78 1.53

α (mg g

−1

6.35 140.51 3.24 1.86 5.454 5.90 3.41 2.070 1.5107 1.189 1.3102 5.661

min

× 10 × × × × × × × × × ×

−2)

39

1014 1012 105 105 1010 1017 1044 1023 1021 1022

β (g mg−1 min−1)

R2

SSE

0.4018 0.1041 0.1401 0.1347 0.07092 0.091308 0.097982 0.1626 0.4451 0.2826 0.1956 0.2225

0.9832 0.9975 0.9424 0.9747 0.8582 0.9968 0.9543 0.9934 0.937 0.996 0.6931 0.937

0.1351 0.2004 0.7293 0.6554 4.4358 0.4811 2.4813 1.2885 0.4507 0.1742 2.632745 0.9015

Dt, De, t, and k2 represent grafting density at reaction time (t), grating density at equilibrium, reaction time, and the reaction rate constant. bα and β represent the initial sorption rate (mg·g−1 min−2) and desorption constant (mg·g−1 min−1), respectively. a

fitting showed surprisingly that the equilibrium grafting density was very close to the experimental results at 24 h of reaction time, indicating that the equilibrium grafting density may have been reached. On the other hand, for MNPw‑70−2% it could be seen that within the first hour of reaction time the grafting density had reached 90% of De, but for MNPw‑30−0.2%, the grafting density had only reached two-thirds of De. Moreover, due to kinetic reasons even at 24 h of reaction time the Dg of MNPw‑30−0.2% was still much lower than that of MNPw‑70−2%, a little more than one-third. Therefore, it appeared that although both reaction temperature and initial silane concentration were important parameters in determining the equilibrium grafting density of the nanoparticles, reaction temperature appeared more important for silanization than the initial silane concentration. The kinetics of silanization has been widely studied and reports mainly focused on the growth of monolayer on flat substrate including oxidized silicon, titania.23,31−36 Literature results showed that the silanization follows a first order Langmuir adsorption kinetics. For example, Balgar et al.28 used atomic force microscopy to characterize the monolayer growth and showed that the growth of octadecyltrichlorosilane proceeds via the formation of islands with a sometimes branched shape indicative for a DLA-type growth mechanism. Garcia et al.24 characterized the surface silanization kinetics of the TiO2 nanoparticles with octytriethoxysilane under compressed CO2 and reported that the self-assembled monolayer was formed in a short time (within 15 min) and the kinetic equation resembled the Langmuir adsorption model for monolayer formation. However, the kinetic data were based on a reaction time up to 120 min. There are few reports exploring the entire layer formation mechanism for APTES to the authors’ knowledge. It is shown according to our results that when water is used as the solvent the silanization follows a two-stage reaction kinetics with the first stage a fast increase in grafting density and thereafter a process of slow saturation in silanization. Moreover, the overall reaction kinetics appears to be a complex one involving adsorption, chemical sorption, and intraparticle diffusion processes. Other Solvents. In order to verify the silanization mechanism, water/ethanol (1/1) mixture (denoted with

subscript E) and methanol/toluene mixture (denoted with subscript M) were also used as solvents for the silanization. The grafting density change with time is plotted in Figure 4. The

Figure 4. Grafting density of magnetic MNP-NH2 nanoparticles prepared using water/ethanol (a) and toluene/methanol (b) as the solvent. (□) MNP70−2%; (○) MNP70−0.2%; (△) MNP30−2%; (▽) MNP30−0.2% .

initial grafting density increased rapidly for both solvents system, similar to water as the solvent. Again, for both mixed solvents within the first hour of reaction the growth of Dg is fast, agreeing with literature results that silanization takes place rapidly and the process toward saturation is a slow process. The difference in the Dg with different solvents is most probably caused by the complicated solvent effects. Solvent effects might be ascribed to the competition effects, differences in polarity 15279

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282

Langmuir

Article

contrast to the initial silane concentration. Nevertheless, it should be pointed out that at a low reaction temperature and a low silane concentration, the progress toward saturation may be longer than 24 h with a low surface coverage. Surface Coverage. If a monolayer of silane is formed on the MNP surface, the maximum grafting density can be calculated, according to the following equation

and diffusivity of the solvent molecules, agreeing with literature results.27,37,38 On the other hand, it appeared that the initial Dg for water/ethanol mixture solvent is higher than the methanol/ toluene mixture, which may indicate the methanol/toluene mixture is not suitable for silanization at ambient temperature (around 30 °C) with a low initial silane concentration. Overall, it can be seen that when methanol/toluene (1/1) mixture is used, the highest grafting density is yielded at 70 °C using 2% of APTES loading. The kinetic fittings of silanization over longer time by the pseudo-second order are shown in Figure 5a,b, respectively, for

⎛ S ⎞⎛ M ⎞ Dg = ⎜ BET ⎟⎜⎜ W ⎟⎟ ⎝ Ssi ⎠⎝ Na ⎠

(2)

where SBET is the specific surface area of the bare nanoparticle (m2/g), Ssi is the occupied surface area of an silane ligand (nm2), Mw is the molecular weight of APTES (221 g/mol), and Na is the Avogadro constant (6.02 × 1023 /mol). The surface area of the MNP was calculated via the BET method according to nitrogen adsorption−desorption measurements with an average value of 162 m2/g. Assuming that the occupied surface area of a silane ligand is 0.2 nm2,39 a grafting density of 298 mg/g is expected according to eq 2. Interestingly, the experimental result of Dg at 70 °C (301 mg/g) using toluene/methanol mixture solvent is very close to the theoretical prediction by BET calculation, assuming a monolayer exists on the surface. Moreover, based on the Pseudo-second order model the equilibrium Dg of 301.2 mg/g was expected, which was also close to the data based on BET calculation. Therefore, it can be seen that the pseudo-second order kinetics can be used rationally to describe the kinetics of APTES silanization of MNP. Moreover, these results indicate that a monolayer (route 1) is most likely resulted during the surface grafting process. Also, to prepare an APTES silane layer with a good quality the reaction should be carried out at higher temperature (70 °C) using methanol/toluene mixed solvents at a large excess of APTES. Figure 5. The kinetics fitting of MNP nanoparticles silanization with APTES using ethanol/water mixture (a) and methanol/toluene (b) as the solvent as fitted by Pseudo- second order model. (▽) MNP30−0.2%; (○) MNP30−2%; (△) MNP70−0.2%; (□) MNP70−2%.



CONCLUSIONS



ASSOCIATED CONTENT

Kinetics of APTES silanization of supermagnetic iron oxide nanoparticles was investigated with the use of different solvents and under different reaction temperatures. The results show that initial silanization takes place very fast and can account for over 80% of the degree of grafting. The process toward saturation is very slow and may take 24 h before reaching equilibrium. Various kinetic models were used for the data fitting, and it was found that the reaction mechanism may involve pseudo-second order, Elovich, and intraparticle diffusions models. Overall, this report clarified the kinetics of APTES silanization toward saturation and may serve as a guideline for the preparation of aminated MNPs with controlled surface grafting density.

ethanol/water mixture and methanol/toluene mixture as the solvents. The data fitting with Elovich model is not shown, but the reaction rate constants, r2 and SSE, are shown in Table 2. All the reaction data fit well with the pseudo-second order kinetics and Elovich models indicating that the solvent does not change the reaction kinetics for the slow saturation silanization process. The reaction constants are shown in Table 2. The calculated equilibrium adsorption capacity is 301.2 mg/g using toluene/methanol mixture as the solvent at a reaction temperature of 70 °C and a reaction time of 24 h based on pseudo-second order model. It should be noted that the data fitted by the pseudo-second order kinetics model were surprisingly close to the values obtained at 24 h of reaction time. These results indicate that the silanization might have reached an equilibrium state and longer reaction time may not increase the degree of silanization. In summary, the use of different solvents does not change the kinetics of the silanization process. However, the degree of silanization depends greatly on the reaction conditions. It may indicate that to prepare a silane layer with suitable surface coverage, the silanization temperature is a key parameter in

S Supporting Information *

The FTIR spectra, VSM loops, XRD patterns, and a typical TGA curve of the MNPs are shown in Figures S1−S4, respectively. The kinetics data fitting by Lagergren pseudo-first order and intraparticle diffusion models are shown in Figures S5 and S6, respectively. Particle size distribution of the MNPs is shown in Figure S7. This material is available free of charge via the Internet at http://pubs.acs.org. 15280

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282

Langmuir



Article

superparamagnetic iron oxide nanoparticles for application as magnetic resonance imaging contrast agents. J. Mater. Res. 2012, 27, 1846−1852. (14) Rouhana, L. L.; Schlenoff, J. B. Aggregation resistant zwitterated superparamagnetic nanoparticles. J. Nanopart. Res. 2012, 14. (15) Deligoz, H.; Baykal, A.; Senel, M.; Sozeri, H.; Karaoglu, E.; Toprak, M. S. Synthesis and characterization of poly(1-vinyltriazole)grafted superparamagnetic iron oxide nanoparticles. Synth. Met. 2012, 162, 590−597. (16) Hong, S. C.; Lee, J. H.; Lee, J.; Kim, H. Y.; Park, J. Y.; Cho, J.; Lee, J.; Han, D. W. Subtle cytotoxicity and genotoxicity differences in superparamagnetic iron oxide nanoparticles coated with various functional groups. Int. J. Nanomed. 2011, 6, 3219−3231. (17) Tang, T.; Fan, H.; Ai, S.; Han, R.; Qiu, Y. Hemoglobin (Hb) immobilized on amino-modified magnetic nanoparticles for the catalytic removal of bisphenol A. Chemosphere 2011, 83, 255−264. (18) Rother, D.; Sen, T.; East, D.; Bruce, I. J. Silicon, silica and its surface patterning/activation with alkoxy- and amino-silanes for nanomedical applications. Nanomedicine 2011, 6, 281−300. (19) Aissaoui, N.; Bergaoui, L.; Landoulsi, J.; Lambert, J.-F.; Boujday, S. Silane Layers on Silicon Surfaces: Mechanism of Interaction, Stability, and Influence on Protein Adsorption. Langmuir 2011, 28, 656−665. (20) Acres, R. G.; Ellis, A. V.; Alvino, J.; Lenahan, C. E.; Khodakov, D. A.; Metha, G. F.; Andersson, G. G. Molecular Structure of 3Aminopropyltriethoxysilane Layers Formed on Silanol-Terminated Silicon Surfaces. J. Phys. Chem. C 2012, 116, 6289−6297. (21) Asenath Smith, E.; Chen, W. How To Prevent the Loss of Surface Functionality Derived from Aminosilanes. Langmuir 2008, 24, 12405−12409. (22) Thakurta, S. G.; Subramanian, A. Fabrication of dense, uniform aminosilane monolayers: A platform for protein or ligand immobilization. Colloids Surf., A 2012, 414, 384−392. (23) Fadeev, A. Y.; McCarthy, T. J. Self-Assembly Is Not the Only Reaction Possible between Alkyltrichlorosilanes and Surfaces: Monomolecular and Oligomeric Covalently Attached Layers of Dichloro- and Trichloroalkylsilanes on Silicon. Langmuir 2000, 16, 7268−7274. (24) Garcia-Gonzalez, C. A.; Saurina, J.; Ayllon, J. A.; Domingo, C. Preparation and Characterization of Surface Silanized TiO2 Nanoparticles under Compressed CO2: Reaction Kinetics. J. Phys. Chem. C 2009, 113, 13780−13786. (25) Wang, Y.; Lieberman, M. Growth of Ultrasmooth Octadecyltrichlorosilane Self-Assembled Monolayers on SiO2. Langmuir 2003, 19, 1159−1167. (26) Harada, Y.; Girolami, G. S.; Nuzzo, R. G. Growth Kinetics and Morphology of Self-Assembled Monolayers Formed by Contact Printing 7-Octenyltrichlorosilane and Octadecyltrichlorosilane on Si(100) Wafers. Langmuir 2004, 20, 10878−10888. (27) Rozlosnik, N.; Gerstenberg, M. C.; Larsen, N. B. Effect of Solvents and Concentration on the Formation of a Self-Assembled Monolayer of Octadecylsiloxane on Silicon (001). Langmuir 2003, 19, 1182−1188. (28) Balgar, T.; Bautista, R.; Hartmann, N.; Hasselbrink, E. An AFM study of the growth kinetics of the self-assembled octadecylsiloxane monolayer on oxidized silicon. Surf. Sci. 2003, 532, 963−969. (29) Fadeev, A. Y.; Helmy, R.; Marcinko, S. Self-Assembled Monolayers of Organosilicon Hydrides Supported on Titanium, Zirconium, and Hafnium Dioxides. Langmuir 2002, 18, 7521−7529. (30) Howarter, J. A.; Youngblood, J. P. Optimization of Silica Silanization by 3-Aminopropyltriethoxysilane. Langmuir 2006, 22, 11142−11147. (31) Carraro, C.; Yauw, O. W.; Sung, M. M.; Maboudian, R. Observation of Three Growth Mechanisms in Self-Assembled Monolayers. J. Phys. Chem. B 1998, 102, 4441−4445. (32) Fadeev, A. Y.; McCarthy, T. J. Trialkylsilane Monolayers Covalently Attached to Silicon Surfaces: Wettability Studies Indicating That Molecular Topography Contributes to Contact Angle Hysteresis. Langmuir 1999, 15, 3759−3766.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Tel: +86-21-20350881. Fax: +86-2120325112. *E-mail: [email protected]. Tel: +86-21-20325162. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors would like to thank the partial financial support from National Natural Science Fund China (Project nos. 21176119, 51202212), the National Key Basic Research Program of China (973 Program with Project no. 2012CB932800), and Shanghai Ministry of Science and Technology (Project no. 13DZ1205100).



REFERENCES

(1) Libertino, S.; Giannazzo, F.; Aiello, V.; Scandurra, A.; Sinatra, F.; Renis, M.; Fichera, M. XPS and AFM Characterization of the Enzyme Glucose Oxidase Immobilized on SiO2 Surfaces. Langmuir 2008, 24, 1965−1972. (2) Mäkilä, E.; Bimbo, L. M.; Kaasalainen, M.; Herranz, B.; Airaksinen, A. J.; Heinonen, M.; Kukk, E.; Hirvonen, J.; Santos, H. A.; Salonen, J. Amine Modification of Thermally Carbonized Porous Silicon with Silane Coupling Chemistry. Langmuir 2012, 28, 14045− 14054. (3) Chen, R.; Jiang, Y.; Xing, W.; Jin, W. Fabrication and Catalytic Properties of Palladium Nanoparticles Deposited on a Silanized Asymmetric Ceramic Support. Ind. Eng. Chem. Res. 2011, 50, 4405− 4411. (4) Mahmoudi, M.; Sahraian, M. A.; Shokrgozar, M. A.; Laurent, S. Superparamagnetic Iron Oxide Nanoparticles: Promises for Diagnosis and Treatment of Multiple Sclerosis. ACS Chem. Neurosci. 2011, 2, 118−140. (5) Can, K.; Ozmen, M.; Ersoz, M. Immobilization of albumin on aminosilane modified superparamagnetic magnetite nanoparticles and its characterization. Colloids Surf., B 2009, 71, 154−159. (6) Sweetman, M. J.; Shearer, C. J.; Shapter, J. G.; Voelcker, N. H. Dual Silane Surface Functionalization for the Selective Attachment of Human Neuronal Cells to Porous Silicon. Langmuir 2011, 27, 9497− 9503. (7) Feng, B.; Hong, R. Y.; Wang, L. S.; Guo, L.; Li, H. Z.; Ding, J.; Zheng, Y.; Wei, D. G. Synthesis of Fe3O4/APTES/PEG diacid functionalized magnetic nanoparticles for MR imaging. Colloids Surf., A 2008, 328, 52−59. (8) Yamaura, M.; Camilo, R. L.; Sampaio, L. C.; Macedo, M. A.; Nakamura, M.; Toma, H. E. Preparation and characterization of (3aminopropyl) triethoxysilane-coated magnetite nanoparticles. J. Magn. Magn. Mater. 2004, 279, 210−217. (9) Arsalani, N.; Fattahi, H.; Laurent, S.; Burtea, C.; Vander Elst, L.; Muller, R. N. Polyglycerol-grafted superparamagnetic iron oxide nanoparticles: highly efficient MRI contrast agent for liver and kidney imaging and potential scaffold for cellular and molecular imaging. Contrast Media Mol Imaging 2012, 7, 185−94. (10) Neuberger, T.; Schopf, B.; Hofmann, H.; Hofmann, M.; von Rechenberg, B. Superparamagnetic nanoparticles for biomedical applications: Possibilities and limitations of a new drug delivery system. J. Magn. Magn. Mater. 2005, 293, 483−496. (11) Zhu, M.; Lerum, M. Z.; Chen, W. How To Prepare Reproducible, Homogeneous, and Hydrolytically Stable AminosilaneDerived Layers on Silica. Langmuir 2011, 28, 416−423. (12) Van, T. N.; Lee, Y. K.; Lee, J.; Park, J. Y. Tuning Hydrophobicity of TiO2 Layers with Silanization and Self-Assembled Nanopatterning. Langmuir 2013, 29, 3054−3060. (13) Larsen, B. A.; Hurst, K. M.; Ashurst, W. R.; Serkova, N. J.; Stoldt, C. R. Mono and dialkoxysilane surface modification of 15281

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282

Langmuir

Article

(33) Bunker, B. C.; Carpick, R. W.; Assink, R. A.; Thomas, M. L.; Hankins, M. G.; Voigt, J. A.; Sipola, D.; de Boer, M. P.; Gulley, G. L. The Impact of Solution Agglomeration on the Deposition of SelfAssembled Monolayers. Langmuir 2000, 16, 7742−7751. (34) Bautista, R.; Hartmann, N.; Hasselbrink, E. Two-Dimensional Aggregation of Species with Weak and Strong Bonding Interactions: Modeling the Growth of Self-Assembled Alkylsiloxane Monolayers. Langmuir 2003, 19, 6590−6593. (35) Chauhan, A. K.; Aswal, D. K.; Koiry, S. P.; Gupta, S. K.; Yakhmi, J. V.; Suergers, C.; Guerin, D.; Lenfant, S.; Vuillaume, D. Self-assembly of the 3-aminopropyltrimethoxysilane multilayers on Si and hysteretic current-voltage characteristics. Appl. Phys. A: Mater. Sci. Process. 2008, 90, 581−589. (36) Yang, Y.; Bittner, A. M.; Baldelli, S.; Kern, K. Study of selfassembled triethoxysilane thin films made by casting neat reagents in ambient atmosphere. Thin Solid Films 2008, 516, 3948−3956. (37) Liao, Z.; Pemberton, J. E. Structure−Function Relationships in High-Density Docosylsilane Bonded Stationary Phases by Raman Spectroscopy and Comparison to Octadecylsilane Bonded Stationary Phases: Effects of Common Solvents. Anal. Chem. 2008, 80, 2911− 2920. (38) Duchet, J.; Chabert, B.; Chapel, J. P.; Gérard, J. F.; Chovelon, J. M.; Jaffrezic-Renault, N. Influence of the Deposition Process on the Structure of Grafted Alkylsilane Layers. Langmuir 1997, 13, 2271− 2278. (39) Kojio, K.; Takahara, A.; Omote, K.; Kajiyama, T. Molecular Aggregation State of n-Octadecyltrichlorosilane Monolayers Prepared by the Langmuir and Chemisorption Methods. Langmuir 2000, 16, 3932−3936.

15282

dx.doi.org/10.1021/la403269u | Langmuir 2013, 29, 15275−15282