In Situ FTIR Investigation of the Kinetics of Silica Polycondensation in

Jul 25, 2014 - Here the kinetics of silica polycondensation are investigated in situ in the presence of a set of representative surfactant templates f...
5 downloads 14 Views 3MB Size
Download PDF
Article pubs.acs.org/JPCC

In Situ FTIR Investigation of the Kinetics of Silica Polycondensation in Surfactant Templated, Mesostructured Thin Films Venkat R. Koganti,† Saikat Das, and Stephen E. Rankin* Department of Chemical and Materials Engineering, University of Kentucky, 177 F.P. Anderson Tower, Lexington, Kentucky 40506-0046, United States S Supporting Information *

ABSTRACT: A detailed Fourier transform infrared spectroscopy (FTIR) study of the kinetics of polycondensation of surfactant-templated mesostructured silica thin films was carried out with the goal of understanding how to manipulate the synthesis and processing of these films to achieve a desired architecture. The evolution of silica condensation was followed both during the sol preparation process and in situ for a time period on the order of minutes to hours after initial film deposition. The kinetics were measured in the presence of three commonly used classes of surfactant templates: P123 [poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) triblock copolymer), Brij-56 (poly(ethylene oxide] [n ∼ 10] hexadecyl ether), and CTAB (cetyltrimethylammonium bromide). The induction period for polycondensation (an initial period during which the rate of change of silanol content is slow) was different in each case, with P123 giving the longest induction time followed by Brij-56 and finally CTAB. Humidity was found to increase the induction time for polycondensation in general, and the initial period of up to 30 min after film deposition was identified as a critical “tunable steady state” interval during which conditions can be adjusted to tune the film properties for different applications, for instance, to alter the nature of the mesophase or its orientation by imposing external forces by confining interfaces or by electrical, magnetic, or flow fields.

■

INTRODUCTION Since the initial reports of preparing surfactant-templated ordered mesostructured silica films by spin1−3 and dip coating,4 the evaporation induced self assembly (EISA) process5 has been widely adopted as an approach to engineered nanostructured materials.6−10 Because it is a rapid, simple approach to preparing thin films with well-defined organic domains or pores of sizes on the order of several nanometers, EISA has been used to prepare films for a wide range of applications, including protective films,11−13 low-k dielectrics,14−16 sensors,17−21 nanomaterial templates,22−29 and membranes.30−36 Despite the advantages of these materials, being able to consistently tune the architecture of the films for a desired application requires a thorough understanding of the physical and chemical changes that occur in the film during sol preparation, deposition, and aging.10 To understand structural evolution in EISA thin films, several groups have used a combination of X-ray scattering and transmission electron microscopy (TEM) techniques.37−46 These techniques have provided a detailed picture of the evolution of long-range order and thus help to understand the effects of process parameters, including pH, temperature, and surfactant type on the mesostructure. As recently reviewed by Faustini et al.,10 the picture that emerges during dip coating involves micelle formation and mesophase formation at the film/air interfaces within the first few seconds of the coating © 2014 American Chemical Society

process, followed by further mesophase formation throughout the film and order−order phase transitions between mesophases (depending on the system and its phase diagram). Along with mesostructure formation and transitions, condensation of the precursor occurs while at the same time species can exchange between the film and vapor phase. The period where structure has emerged but the components of the film are still mobile enough to be altered has been called the “tunable steady state” (TSS) by Grosso et al.43,47 The TSS helps to explain why the structure of EISA-derived films can be tuned by adjusting variables such as the humidity of the vapor,48−50 the sol aging time (which affects the timescale for condensation after coating),50,51 the pH of the film,38,52,53 the presence of solvent vapors,54 and confining the coating using materials of different surface energy.55−59 While X-ray and electron microscopy studies have shown how long-range order evolves in EISA films, spectroscopy is needed to follow the corresponding chemical changes occurring during the process. Understanding the molecular-level changes in the silica network is essential to understand the synthesis mechanisms of these materials. Nuclear magnetic resonance (NMR) provides the most detailed information regarding the Received: June 6, 2014 Revised: July 25, 2014 Published: July 25, 2014 19450

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

bonding of silicon-based materials,60−69 and, in fact, provided much of the insight needed to develop the process for EISA. Under acid-catalyzed conditions typically used for EISA, the hydrolysis of silicon alkoxides such as tetraethoxysilane (TEOS) is fast and reversible, which allows this process to reach equilibrium.70−73 The polycondensation process exhibits negative first-shell substitution effects and cyclization, which tends to lead to cagelike intermediate oligomers74−80 and slow development of homogeneous network made up of chains of cycle-containing species.81,82 The rapid hydrolysis and slow polycondensation at low pH provides desirable conditions for mesoporous silica film formation because the as-deposited films contain active silicates that react slowly enough to allow an ordered mesostructure to form by coassembly with surfactants.5,10,43 Despite the high level of detail it provides, NMR is not applicable to the EISA process in thin films due to the small volumes of material involved and the high viscosity of the deposited films. Instead, Fourier transformed infrared (FTIR) spectroscopy can be used to follow the evolution of functional groups and some structural details as a coated solution evolves.44 Several prior studies of the sol−gel process of alkoxysilane precursors to form powders and gels83−87 have identified and used key bands, including the Si−OH stretching band at ∼960 cm−1, siloxane stretching bands near 1100−1200 cm−1, and bands at 550−600 nm−1 originating from the skeletal vibration of the 4-fold siloxane rings.88 Innocenzi has provided an excellent overview of the application of infrared spectroscopy to sol−gel derived silica films.89 That article reviews the correlation between IR spectra and microstructure of sol−gel silica films and explains the various bands occurring in the IR spectra of sol−gel silica thin films. Deconvolution techniques have been used to detect changes in skeletal vibrations of the Si−O−Si network related to cyclization and progress of polycondensation.90−92 For example, Innocenzi et al.93 were able to successfully resolve the LO3-TO3 pair (longitudinal optic−transverse optic splitting of the asymmetric Si−O−Si stretch), cyclic species absorption bands, and also disorder− order transition bands (LO4-TO4) in surfactant-templated silica films during thermal treatment. Recently, Innocenzi et al.44,94−97 have shown that advanced synchrotron techniques can provide highly resolved kinetic information about the EISA process immediately following the deposition of surfactant-containing sol−gel films.44,94−97 Using this approach, IR spectroscopy and X-ray scattering have been coupled to gain substantial insight into the evaporation and selfassembly process occurring during the first few seconds of the dip-coating process during EISA.44,98 However, the focus of these studies has mainly been on the drying process itself99,100 and timescales up to the first minute or two of the EISA coating process.44,95,97 The tunable steady state noted by several researchers9,14,33,101 involves timescales on the order of several minutes following coating, and so there still exists a need to study the kinetics of the polycondensation of silica precursors in surfactant-containing EISA films for intervals on the order of minutes to hours, comparable to the recent study of 3glycidoxypropyltrimethoxysilane film formation by Innocenzi et al.102 Here the kinetics of silica polycondensation are investigated in situ in the presence of a set of representative surfactant templates for thin films both before and after coating deposition. Prior to deposition, the changes in various bands at different stages during the preparation of the coating sol are

monitored using ATR-FTIR. In situ FTIR studies after coating are accomplished by depositing the surfactant-templated silica films onto thin Si wafers. By monitoring the wafers at the exact same spot before and after coating, it is possible to follow the changes of the sol−gel film. As an indicator of the stage of the coating process and the extent of polycondensation, particular attention is paid to the Si−OH stretching band at 960 cm−1. The in situ FTIR experiments are used to better understand the duration and nature of the tunable steady state (TSS) in mesostructured silica films, and its dependence on the type of surfactant and humidity of the vapor in contact with the film.

■

EXPERIMENTAL SECTION Materials. P123 (BASF); cetyltrimethylammonium bromide (CTAB, Aldrich, ≥99%); Brij-56 (Fluka) and tetraethoxysilane (TEOS, >99%, Fluka); deionized ultrafiltered water, concentrated sulfuric acid, and 30% aqueous H2O2 (all from Fisher Scientific); and anhydrous ethanol (Aaper Alcohol & Chemical) were all used as received. (100) Cut single-side polished silicon wafers were obtained from University Wafer. Preparation of Coatings. Three surfactants were employed in this work: P123, a triblock copolymer with average block composition (ethylene oxide)20(propylene oxide)70(ethylene oxide)20; cetyltrimethylammonium bromide (CTAB); and Brij-56, a hexadecyl ether of poly(ethylene oxide) with the primary component being decaethylene oxide hexadecyl ether. Tetraethoxysilane was used as a silica precursor. A stock solution of prehydrolyzed silane was prepared using the acid-catalyzed procedure described by Lu et al.4,5 First, TEOS, ethanol, water, and HCl (mole ratio 1:3.8:1:5 × 10−5) were refluxed at 70 °C for 90 min. Then, the remaining water and HCl were added, increasing the concentration of HCl to 7.34 mM. After stirring this mixture at 25 °C for 15 min, the sols were aged at 50 °C for 15 min. The required amount of surfactant was dissolved in ethanol and this solution was added to the above aged silica sol with constant stirring. The final mole ratios were 1 TEOS:22 C2H5OH:5 H2O:0.004 HCl:0.01 P123 or 0.055 Brij-56 or 0.098 CTAB. The surfactant amounts were selected based on reports shown to give two-dimensional (2D) hexagonal mesoporous silica films for P123,58 Brij-56 (see below), and CTAB.4 Silicon wafer sections were dip coated with each sol at a withdrawal speed of 7.6 cm/min. FTIR Sampling. During the sol preparation, a small amount of the coating solution was withdrawn at regular intervals and the FTIR spectrum of this solution was collected using a ZnSe ATR trough accessory (Pike Technologies, Madison, WI). To obtain the FTIR spectra of the coated films, a 1 × 1 cm section of thin (100) Si wafer (200 μm thick) was first cleaned using deionized water, acetone, and isopropanol in that sequence. This cleaned Si wafer was held between two nickel-coated axially aligned ring-shaped neodymium magnets and Teflon spacers with an outer diameter just small enough to fit inside of the cell and an inner diameter larger than the FTIR beam, and this system (Si wafer between two ring magnets) was mounted inside of a 5 cm path length gas cell accessory for the FTIR spectrometer to collect the background (see Figure S1 of the Supporting Information). The gas cell (also from Pike Technologies) consisted of a closed cylindrical glass cell capped at both ends with IR-transparent ZnSe windows through which the IR beam could enter and then leave. The cell had side inlet and outlet fittings to introduce any gas, thus allowing environmental control without contaminating the IR 19451

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

compartment. In this work, vapors with different relative humidity were passed through the cell (from completely dry N2 to water-saturated air). Humidity was measured in all cases using a humidity/temperature monitor (Fisher Scientific). After collecting the background, the Si wafer was removed, coated with the silica sol synthesized as described above, and then transferred back into the gas cell immediately to collect the spectra as a function of time. Successful removal of the background from the oxide layer on the wafer required carefully repositioning the wafer in the gas cell exactly as it had been positioned during background collection. In addition, it was found to be necessary to use single-side polished wafers to avoid interference from partial reflection of the IR beam on both sides of the wafer. It typically took 50 s to transfer a wafer into the gas cell after coating. Therefore, the earliest spectra of the silica films presented in this work were collected ∼50 s after coating. FTIR spectra were collected using a ThermoNicolet Nexus 470 FTIR spectrometer with sealed and desiccated optics. The resolution of the spectra was 2 cm−1, and 64 scans were acquired per spectrum. Characterization of Films. The thickness of each film after complete curing was estimated by measuring the thickness of films coated from the same sols onto silicon wafers, using a Gaertner 7109-C-338G ellipsometer. For Brij-56, the formation of a 2D hexagonal mesoporous material was confirmed by dip coating the sol onto glass slides that were cleaned with a 7:3 (by volume) mixture of concentrated sulfuric acid and 30% H2O2. After dip coating at a withdrawal speed of 7.6 cm/min, the films were aged at room temperature for 24 h, then 70 °C for 24 h, and finally at 120 °C for 24 h. The films were then calcined in air at 500 °C for 4 h. Samples were prepared for transmission electron microscopy (TEM) by scraping the film from the glass slide, suspending the powder in an epoxy resin, and slicing the resin with a microtome. TEM was performed using a JEOL 2000FX instrument operated at 200 kV.

Figure 1. Typical ATR FTIR spectrum of the silica sol just before dip coating Si wafers with a P123-containig sol. Numbers indicate the peak wavenumbers (in cm−1) of the major bands.

Table 1. Major Infrared Bands Observed and Their Assignments wavenumber (cm−1) ∼3350

O−H stretch87,103

∼3000−2800

C−H stretch87

∼1640 ∼1440−1320 ∼1250−1020

H−O−H bending103 various CH2 and CH3 vibrations Si−O−Si antisymmetric stretching modes89−92 CH3 rock C−O antisymmetric/C−C stretch87 C−O stretch, C−C Si−OH5,103 stretch C−C/C−O stretch87,106 Si−O−Si symmetric stretch89/ CH2, CH3 rock104,106,107 C−O symmetric stretch105 cyclic siloxane species89,90,93 transverse optical Si−O−Si rocking89,90

∼1163 ∼1082 ∼1045 ∼955 ∼877 ∼798−805

■

RESULTS AND DISCUSSION For all three types of surfactants, uniform, macroscopically defect free SiO2 films were deposited on Si wafers using the procedure described above. The thickness of the films after curing was measured by ellipsometry (data not shown) to be approximately 250 nm regardless of the surfactant used. Before discussing the kinetics of curing, the reactions occurring during sol aging are discussed. The FTIR history of the precursor sol was obtained before depositing the sol on Si wafers by withdrawing a small amount of sol at different time periods and obtaining the FTIR spectrum using a ZnSe ATR trough accessory. Figure 1 shows a typical ATR-FTIR spectrum of the final sol after the addition of surfactant P123. Spectra of sols with other surfactants (CTAB and Brij-56) showed similar bands, with the obvious exception of the bands coming from the surfactants. The bands in the spectra were identified based on literature assignments and comparison to the ATR-FTIR spectra of pure TEOS and EtOH (not shown but available in standard databases such as the NIST Chemistry Web Book, http:// webbook.nist.gov), and are summarized in Table 1. Figure 2 shows the FTIR spectra of the sol at different stages starting from the first stage during which the initial amounts of TEOS, H2O, EtOH, and HCl are added and aged at 70 °C, to the final stage after adding the surfactant (P123 in this case). For clarity, the spectra are shown in the wavenumber range from 1800 to

band assignment

∼783 ∼620 ∼450

chemical species responsible ethanol, water, Si− OH of silanols ethanol, TEOS and surfactant molecular water ethanol, TEOS siloxane skeleton TEOS87,104 TEOS,84 ethanol87 ethanol87 silanols ethanol siloxane bonds, ethanol TEOS105 siloxane bonds siloxane bonds

650 cm−1. No significant qualitative changes were seen in the high-wavenumber portions of the spectra. At the far left end of the spectrum in Figure 1, a very strong, broad band centered near 3320 cm−1 was observed. This is the O−H stretching band and is broad because the FTIR sol contains several O−H containing compounds in various states of hydrogen bonding that contribute to this band.87,103 Ethanol, H2O, and Si−OH all contribute to the appearance of this band. Also at high wavenumber, bands near 2980−2870 cm−1 were found due to CH2 stretches from ethanol, ethoxide groups of TEOS and partially hydrolyzed silanes, and the surfactant.87 Because of the large number of contributors to the highwavenumber bands, little could be discerned about the evolution of the sol from these bands. At lower wavenumbers, the spectra in Figure 2 reveal more information about the evolution of the sol. The band due to bending of molecular water occurring at a wavenumber of around 1640 cm−1 decreased over the initial 90 min aging period at 70 °C because of its consumption during the hydrolysis of TEOS.103 The intensity of this band increased again after adding the remaining H2O and HCl in the second 19452

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

during the initial 90 min period due to partial hydrolysis, and it effectively disappeared by the second stage of sol preparation (when all water and HCl have been added). The FTIR bands between wave numbers 1350 and 1500 cm−1 occur due to various deformations in CH2 and CH3 groups that are explained at greater detail elsewhere.87,104 No distinct quantitative trend was identified from these bands. Another important feature of the spectra is a band near 955 cm−1 coming from the stretching of the Si−OH bond.103 Unfortunately, this region (∼960 cm−1) also contains bands from C−C stretching and CH3 bending deformations of ethanol.87,104 Because of this overlap, it was difficult to deconvolute the contributions of Si−OH from the ethanol bands in the sol; as discussed below, this problem was alleviated after the films were deposited and the ethanol evaporated. Still, the appearance of this band signifies the presence of silanols after hydrolysis of the ethoxysilanes. The region between 1120 and 1020 cm −1 contains bands originating from the antisymmetric stretching of the Si−O−Si89 and also from the C−O and C−C stretches of TEOS84 and ethanol.87 At the point when the sol contained hydrolyzing TEOS species, the bands were probably dominated by TEOS and the solvent ethanol. This region also has bands originating from the bending of CH3 and CH2 bonds of the organic species present in the solution. The overlapping bands in this region are difficult to interpret clearly due to the high concentration of ethanol, although Tejedor−Tejedor et al.87 and Mondragon et al.105 have discussed the interpretation of this region of the spectrum. As discussed below, when ethanol was removed, the siloxane bands could be clearly resolved. Figure 1 also shows a small peak at 798 cm−1. This could have been caused by Si− O−Si symmetric stretches,89 but it appeared close to bands due to CH2 and CH3 rocking motions.104,106,107 Another band of significant intensity appeared around 877 cm−1 and was due to the stretching of C−C/C−O bonds in the ethanol present in the system.87 The set of spectra if Figure 2 suggest that the silica sol was completely hydrolyzed before coating under the conditions used, and that there was no significant concentration of hydrolyzable ethoxy groups present in the system after adding the surfactant (no peaks at 1168 and 790 cm−1 were present in the sol after the entire quantities of H2O and HCl were added). Upon dip coating, drying of ethanol drives the coassembly of surfactants and silicate species into an ordered mesophase. The compositions used here have already been shown to give 2D hexagonal mesoporous materials for CTAB4 and P123.58 Figure 4 shows a representative TEM image of the Brij-56 templated film after coating onto a glass slide and calcining the film to remove the surfactant. The image contains both strip patterns consistent with viewing the 2D hexagonal structure perpendicular to the columnar pore channels (indicated with black arrows) and hexagonally close-packed spots consistent with viewing the channels directly from the edge (indicated with a dotted box). The (100) d-spacing from the image is about 5 nm, which is consistent with prior studies reporting 2D hexagonal silica templated using Brij-56.108,109 Thus, all three films that were synthesized in this study are expected to have comparable mesostructures. Figure 5 shows a typical FTIR spectrum of a P123-containing silica film on a Si wafer, 20 h after deposition. As noted in Experimental Section, the contribution of the native oxide layer of the silicon wafer (presumably with a thickness on the order of 10 Å based on prior studies of room temperature oxide

Figure 2. FTIR history of P123-based sol at different times during the synthesis. The age of the sol increases from bottom to top, and spectra are included (a) at the beginning of the first stage of sol preparation, before heating; after the first stage of heating at 70 °C for (b) 15, (c) 26, (d) 32, (e) 48, and (f) 90 min; after the second stage of adding additional HCl and H2O at 25 °C for (g) 1 and (h) 15 min; after the third stage at 50 °C for (i) 15 min; and (j) after adding P123 and ethanol to the fully aged sol. Numbers indicate the peak wavenumbers (in cm−1) of the major bands.

stage of sol preparation and remained high because the final amount of water was greater than the stoichiometric amount for TEOS hydrolysis. In the final spectrum (after adding surfactant), dilution with ethanol caused the intensity of this band to drop significantly. These changes in intensity are depicted in Figure 3, which shows the height of the H2O band

Figure 3. Heights from FTIR spectra of the water bending band (1642 cm−1) and C−O asymmetric stretch of TEOS (783 cm−1) as a function of the total time elapsed during sol aging. The temperatures during each stage of the process are indicated.

near 1642 cm−1 versus time. The gradual decrease of intensity of the water bending band over the first 90 min of aging was also accompanied by decreases in the intensities of the bands associated with CH3 rocking motion87,104 and C−O symmetric stretching105 in TEOS around wave numbers 1163 and 783 cm−1, respectively. The intensities of these bands dropped to almost zero after the addition of the remaining HCl and H2O by the end of the second aging stage, due to further ethoxide hydrolysis. Figure 3 also shows the evolution of the height of the band near 783 cm−1. The intensity of this band decreased 19453

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

3000−2800 cm−1 wavenumbers were caused by the CH2 stretching vibrations of the surfactant. The weak peaks between wave numbers of 1500−1330 cm−1 were caused by the CH2 and CH3 vibrations of the surfactant species. The strongest band in the spectrum occurred at 1080 cm−1 and had a highwavenumber shoulder. This is caused by the antisymmetric stretching motion of the O atom in Si−O−Si bonds parallel to the Si−Si line.89 The intense shoulder on the high frequency end of this band is caused by the longitudinal optic−transverse optic (LO−TO) splitting of the vibrational modes. The band that was observed at wavenumber ∼950 cm−1 is due to the Si− OH stretching of the silanols. The band at wavenumber ∼805 cm−1 is due to the symmetric stretching of the O atom along a line bisecting the Si−O−Si angle.36 A broad band around wavenumber 620 cm−1 can be attributed to cyclic species.89 Such cyclic species are well-known in NMR studies of reacting TEOS sols.60,68,70,71,75,81 Several researchers have found IR evidence for ring structures in sol−gel silica films and have attributed the broad peak between wave numbers 550−640 cm−1 to their presence.89,90,93 The band at ∼450 cm−1 comes from the transverse-optical (TO) rocking motions perpendicular to the Si−O−Si plane of the oxygen atoms bridging two adjacent Si atoms.89,90 While complex changes in Si−O−Si network components can be discerned from FTIR by using deconvolution,46,89 the kinetics of the silica condensation in surfactant templated films after deposition were followed directly by monitoring the changes in the intensity of the silanol stretching band near 955 cm−1. Figure 6 shows the FTIR spectra as a function of time elapsed after coating of silica films with different surfactants cast onto silicon wafers. The spectra shown in Figure 6 were collected in a highly humid environment (relative humidity ∼95%) achieved by passing laboratory air saturated with water through the FTIR gas cell. For clarity, the spectra are shown only in the 600−1400 cm−1 wavenumber range, which include the band originating from the stretching of silanols (∼960 cm−1) and also the band from Si−O−Si antisymmetric stretches (∼1070 cm−1). To obtain a clear view of the silanol band, the data are presented in the reverse of the usual spectrum orientation (with wavenumbers increasing from left to right). The intensity of the silanol band clearly decreases as a function of time for all samples. This is consistent with the fact that the hydrolyzed silica precursors were condensing and thus decreasing the concentration of silanols. However, Figure 6 shows that the kinetics of this decrease in intensity of silanols depends on the type of surfactant present in the system. When the surfactant was P123, the silanol intensity remained constant for almost 20−30 min after coating and then decreased to reach a steady state value (Figure 6a). This induction period is defined as the time Ti during which the silanol band intensity does not decrease appreciably, which suggests that condensation has not begun. Induction periods have also been reported during in situ studies of surfactanttemplated particle formation with P123, but this period is more related to the emergence of ordered mesoporous precipitates from dispersed, weakly bound silica/surfactant aggregates.111−113 Here, the induction period is attributed to the related initial weak interactions between silicates and PEO chains, which slows down the polycondensation process until significant separation between the surfactant and silicate phases has been established. When the surfactant was changed to Brij56 (Figure 6b), delayed condensation was observed as in Figure 6a, but Ti was reduced to ∼10−20 min. When the surfactant

Figure 4. TEM micrograph of a calcined Brij-56 templated mesoporous film which was scraped from the substrate, suspended in epoxy resin, and microtomed. Black arrows indicate stripe patterns resulting from viewing the 2D hexagonal phase perpendicular to the columnar pores, and the dotted box shows a region where the pores are visible directly from their ends.

Figure 5. FTIR spectrum of a fully cured P123-templated silica film (20 h after deposition) coated onto a Si wafer. Numbers indicate the peak wavenumbers of the major bands, in cm−1.

growth110) was removed by collecting the background for the FTIR spectrum with the wafer in place (see Figure S2 of the Supporting Information for background spectrum). For consistency, it was important to use the same section of the wafer when measuring spectra after coating. No further subtraction of the wafer background was required if this was done. In addition, thin Si wafers (200 μm) were used in order to be able to obtain reasonable signal-to-noise in the spectrum of the coating. Ethanol evaporates quickly during coating, and water evaporates until it reaches equilibrium with the film during the initial minutes of curing.10 The only components in the film 20 h after deposition were condensed silica precursors and surfactant. In comparison with Figure 1a, all bands associated with ethanol and ethoxy groups centered near 1400 and 877 cm−1 were absent and new bands associated with condensed silicate species were present. Starting from the left end of the spectrum, the first band observed was the O−H stretching band at around 3400 cm−1 due to dangling Si−OH bonds and possibly adsorbed water. The strong bands between 19454

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

the silanol peak versus time after film deposition for each of the surfactants at RH ∼95%. Note that there may be differences in the absolute concentration of silanol species at the start of FTIR data collection in the three films, so the absolute intensities should not be compared, although the sample-tosample variability in the initial peak heights for each type of film was measured to be only ±3.6%. This plot clearly shows that Ti was maximum when P123 was used as a surfactant, followed by Brij-56. When CTAB was used as a surfactant, no induction time was observed and the silanol band intensity dropped immediately from the start of FTIR observation. These changes in the silanol band were also accompanied by corresponding changes in the antisymmetric stretching of Si− O−Si bonds. To show clearly the types of changes observed, Figure 8 shows several representative FTIR spectra of the

Figure 8. FTIR spectra of the P123-templated silica films aged at 95% RH (a) as a function of time elapsed after deposition where from bottom to top spectra are for 1, 22, 27, 49, and 75 min of aging and (b) after calcination. Figure 6. Time-dependent spectra of silica films on Si wafers collected between 1 min after coating and 150 min, aged in a 95% RH environment and with (a) P123, (b) Brij-56, or (c) CTAB as the surfactant. The positions of the Si−OH and antisymmetric Si−O−Si stretching bands are indicated.

P123-templated silica films as a function of the time elapsed after deposition in a 95% relative humidity environment from t = 1 min to t = 75 min. Also included is the FTIR spectrum of the silica film after calcination at 500 °C for 4 h in air. Only the region from 1300 to 850 cm−1 is shown to highlight the bands originating from Si−O−Si antisymmetric stretching and Si− OH stretching. The broad intense peak at ∼1100 cm−1 shifted toward lower wavenumbers as time elapsed. This peak shift was accompanied by the development of a shoulder on the high wavenumber side. This shoulder can be attributed to LO splitting of the TO3 stretching mode.89 The observation of this shoulder has particular importance in sol−gel materials, especially in thin films, because it has been observed that the LO3 modes are not detected or weakly detected in films when the structure is only weakly cross-linked.89,90 The TO3 and LO3 mode intensities were clearly enhanced as the film evolves in time. This was also accompanied by a shift in the TO3 mode to lower wave numbers (∼1070 cm−1), and the LO3 mode to higher wavenumbers (up to a final value of ∼1250 cm−1) upon calcination. Calcination was also accompanied by the complete disappearance of the peak around 955 cm−1, originating from Si−OH. This signified the complete condensation of the silica network upon calcination of the film. Perhaps the most important observation in Figures 6 and 8 is that the increase in intensity and onset of drift of the TO3 peak

used was CTAB (Figure 6c), condensation started immediately after depositing the film, resulting in a continuous drop in the silanol band intensity immediately after film deposition (Ti ∼ 0). To quantify these observations, Figure 7 plots the height of

Figure 7. Height of the silanol Si−OH stretching band near ∼955 cm−1 vs time for films containing (a) P123, (b) Brij-56, or (c) CTAB and aged at 95% RH. 19455

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

to lower wave numbers coincided with the onset of the decrease in the Si−OH band intensity. Once the Si−OH band reached a steady-state value, there was almost no change in the location of the TO3 and LO3 bands. Significant changes occurred after this stage only upon calcination of the silica film, which induced complete condensation of the silanols. This demonstrates that the induction period in the P123 and Brij-56 templated films was in fact due to a delay in the onset of condensation. Brij-56 templated films also showed the same features as P123 templated films with respect to the changes in the TO3 and LO3 band positions during the aging of the film. When CTAB was used as a surfactant, the splitting of the TO3 and LO3 was seen in the very first spectrum of the film (Figure 6c), indicating that significant condensation occurred during film deposition. No qualitative changes were seen after that. The other variable investigated in this work was the dependence of Ti on the humidity of the environment in which the coated silica film was cured. Figure 9 plots the peak

Figure 10. Height of the silanol Si−OH stretching band near ∼955 cm−1 vs time for P123-templated silica films cured in an environment for which the relative humidity is changed. After 9 min, the stream passing through the gas cell was switched from humid air (95% RH) to dry nitrogen (0% RH).

the amount of water increases the water content of the film and, therefore, slows the condensation reaction. This effect is prominent for P123 surfactants because of the hydrophilicity and length of the poly(EO) (where EO = ethylene oxide) headgroups, which are able to interact with silicates through hydrogen bonding to mediate the effects of RH on curing kinetics. For comparison, Figure 11 shows plots of the peak heights of the Si−OH stretching band vs time for Brij-56 and CTABtemplated films as a function of the gas-phase humidity during curing. Brij-56 templated films responded to humidity similarly to P123 templated films, except that the induction time periods were always shorter. The induction time period decreased as the RH of the vapor decreases, and when the film was cured in a dry N2 atmosphere, no changes in the Si−OH stretching band intensity could be seen. Similar to P123-templated films, this phenomenon can be attributed to fast condensation of the silica network in a dry N2 atmosphere, before the first spectrum can be measured. CTAB-templated films did not show any dependence of the evolution of the silanol band intensity on the gas phase humidity except for the case when the gas phase was dry N2. Similar to P123 and Brij-56 templated films, no changes in the silanol band were observed in a dry N2 atmosphere because of rapid condensation of the silica network. In all experiments conducted here, Ti (P123) > Ti (Brij-56) > Ti (CTAB) ∼ 0 min. This is attributed to differences in the structure of the surfactant. P123 has a large polymeric headgroup. In sol−gel-derived silica materials, there is evidence in the literature for headgroup penetration into the silica walls in P123-templated materials, resulting in microporosity in the silica network.114−116 Sundblom et al.117 concluded that attractive interactions between PEO chains and silicates are entropically driven, although hydrogen bonding may also play a significant role.118,119 Both factors are expected to contribute to the induction period. A consistent observation of decreasing polycondensation kinetics due to the addition of Pluornic F127 to a sol−gel film was reported by Innocenzi et al.97 When the surfactant used was Brij-56 [CH3(CH2)15(EO)10], which also has a polymeric headgroup, delayed condensation of silica was observed, but it was reduced because of the smaller average molecular weight of the poly(ethylene oxide) headgroups. In contrast, CTAB [CH3(CH2)15N(Br)(CH3)3] has an small ionic

Figure 9. Height of the silanol Si−OH stretching band near ∼955 cm−1 versus time for P123 templated silica films cured in the presence of air with a relative humidity of (a) 95% RH (saturated), (b) 47% RH, (c) 25% RH, or (d) dry nitrogen with 0% RH.

heights of the Si−OH stretching band near 960 cm−1 versus time for P123-templated films as a function of the gas-phase relative humidity during curing. Note that the RH during the coating process itself (which occurred in ambient lab air) was the same for all samples and less than 50%. Spectra were obtained in conditions ranging from completely saturated with water vapor (∼95% relative humidity) to dry N2 environment (∼0% relative humidity). The induction time was found to increase as the humidity increased for P123-containing films. In the dry N2 environment, no changes in the silanol band intensity could be resolved with time, which is attributed to fast condensation of the silica network before the first FTIR spectrum was measured. To confirm that the condensation of silica in a dry N2 environment was rapid, the changes in the FTIR spectra induced by suddenly switching the environment from humid air to dry N2 were measured. Figure 10 shows the height under the silanol band for a P123 templated film as a function of aging time after coating the film. The film was initially kept in an environment saturated with water vapor, but 9 min after deposition the environment was changed to dry N2. The intensity of the silanol band dropped immediately at this point, suggesting that the dry N2 environment induces fast condensation of the silica network. The change in induction time can therefore be explained by the change in the gas phase water content. A small amount of water in the vapor phase allows condensation to proceed immediately, while increasing 19456

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

the existence of an induction time after the deposition of the silica films, when the silica condensation is slowest. The induction period is equivalent to the TSS and is longest for surfactants with large nonionic poly(EO) headgroups at high relative humidity. This observation is important for two reasons. First, it shows that controlling the coating environment after curing is important to obtaining reproducible results when making films by EISA. The humidity (and by inference also the partial pressure of the solvent) must be controlled to ensure that the mesostructure does not change (for instance due to swelling) while it is still in the induction period. Second, it provides the opportunity to utilize the induction time to engineer the film properties, for instance, to bring the films into contact with functional surfaces to alter the alignment of the mesophases,58 obtain different mesophases,49,120 imprint functionality into the films, or to manipulate the mesophase with physical or optical patterning or fields.

■

CONCLUSION Here, the synthesis of mesoporous silica films using CTAB, P123, and Brij-56 surfactants was monitored in situ using FTIR spectroscopy. The kinetics of the sol−gel process were studied both before and after depositing the silica films. Prior to film deposition, FTIR spectra of the sol were collected at regular intervals using ATR-FTIR. Monitoring various bands in the FTIR spectra of the sol showed that with the widely used sol preparation technique published by Lu et al.,4 TEOS is completely hydrolyzed prior to deposition of the films. This hydrolysis was accompanied by the growth of a Si−OH stretching band. After depositing the silica sol on Si wafers, transmission FTIR spectra of the silica films were collected at regular intervals of time. The evolution of the bands from the silanols and the siloxane species showed that the condensation of the silanols in the film continues to occur over a significant amount of time after coating deposition, and that the rate of condensation depends on the kind of surfactant used and the relative humidity of the curing environment. In some cases, evidence was found for the presence of an induction time after the film deposition (Ti) during which the condensation rate of the silica network was at its minimum and (presumably) the silica network is still flexible. Monitoring the silanol band intensity versus time showed that Ti is greatest when the surfactant used is P123, less but still observable with Brij-56, and not observed with CTAB. These differences are attributed to penetration of the poly(ethylene oxide) chains of P123 and Brij-56 into the silica walls, perhaps also aided by interactions between the terminal OH groups of these surfactants with silica. The interaction of CTAB with silica is physical, and occurs only at the micelle−silica interface, which apparently does not cause a large induction time for silanol condensation. The induction time was also found to be enhanced by a high humidity environment, especially for P123. All of the changes in the silanol band intensity were found to be accompanied by corresponding changes in the siloxane band intensities consistent with greater solidifaction of the siloxane network upon loss of the Si−OH stretching intensity. This is an important observation, suggesting that the silica network in surfactant-templated mesoporous films is flexible for an interval on the order of several minutes, during which the film environment can be manipulated by postcoating modifications.58 Thus, it is not necessary to assume that changes in the mesostructure are complete within the time frame that the final ordered mesophase has developed, which is on the order of a

Figure 11. Height of the silanol Si−OH stretching band near ∼955 cm−1 vs time for (a) Brij-56-templated silica films and (b) CTABtemplated silica films cured in environments of different relative humidity.

headgroup, which is not expected to penetrate into the silica network.114 Therefore, condensation was not delayed by interactions between silanols and the headgroup, and thus, no induction time was observed by FTIR with CTAB. Prior studies have also reported effects of the relative humidity and solvent vapor pressure of the mesostructure of the dip-coated EISA-derived films. As discussed in the Introduction, Cagnol et al.49 found that relative humidity affects the mesophase formed in silica films and, therefore, proposed the idea of a TSS during which the mesostructure of the silica films can be changed by manipulating the environment. They found a change in the mesophase by changing the relative humidity of the coating environment 10 min after film deposition, unlike many previous studies which assume that in the EISA process, all changes happen during the initial seconds after deposition. Babonneau et al.120 used in situ X-ray scattering experiments to understand the effect of pH of the coating sol and the ethanol vapor pressure of the coating environment on the final mesophase obtained in surfactanttemplated silica films. They observed that high ethanol concentrations in the vapor phase maintains high ethanol concentration in the films, which lowers the viscosity and thus promotes phase transitions.120 The present study explains the reasons for such an observation in terms of the silica condensation process, which is influenced by both the relative humidity of the environment and the kind of surfactant present in the system. The FTIR results presented here demonstrate 19457

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

few seconds or minutes after depositing the film. The current results suggest the existence of the additional time frame on the order of at least up to 20−30 min for postdeposition modifications in PEO-based surfactant templated films.

■

(12) Li, X.; Song, L. Y.; Vogt, B. D. Tuning Mechanical Properties of Mesoporous Silicas Using Associating Homopolymers/Block Copolymer Blends as Templates. J. Phys. Chem. C 2008, 112, 53−60. (13) Wang, J. F.; Cheng, Q. F.; Tang, Z. Y. Layered Nanocomposites Inspired by the Structure and Mechanical Properties of Nacre. Chem. Soc. Rev. 2012, 41, 1111−1129. (14) Soler-Illia, G.; Innocenzi, P. Mesoporous Hybrid Thin Films: The Physics and Chemistry Beneath. Chem.Eur. J. 2006, 12, 4478− 4494. (15) Plawsky, J. L.; Kim, J. K.; Schubert, E. F. Engineered Nanoporous and Nanostructured Films. Mater. Today 2009, 12, 36− 45. (16) Volksen, W.; Miller, R. D.; Dubois, G. Low Dielectric Constant Materials. Chem. Rev. 2010, 110, 56−110. (17) Coll, C.; Bernardos, A.; Martinez-Manez, R.; Sancenon, F. Gated Silica Mesoporous Supports for Controlled Release and Signaling Applications. Acc. Chem. Res. 2013, 46, 339−349. (18) Urek, S. K.; Francic, N.; Turel, M.; Lobnik, A. Sensing Heavy Metals Using Mesoporous-Based Optical Chemical Sensors. J. Nanomater. 2013, 2013, 1−13. (19) Wagner, T.; Haffer, S.; Weinberger, C.; Klaus, D.; Tiemann, M. Mesoporous Materials as Gas Sensors. Chem. Soc. Rev. 2013, 42, 4036−4053. (20) Walcarius, A. Mesoporous Materials and Electrochemistry. Chem. Soc. Rev. 2013, 42, 4098−4140. (21) Melde, B. J.; Johnson, B. J. Mesoporous Materials in Sensing: Morphology and Functionality at the Meso-Interface. Anal. Bioanal. Chem. 2010, 398, 1565−1573. (22) Urbanova, V.; Walcarius, A. Silica-Based Nanoporous Materials. Z. Anorg. Allg. Chem. 2014, 640, 537−546. (23) Holmes, J. D.; Spalding, T. R.; Ryan, K. M.; Lyons, D.; Crowley, T.; Morris, M. A. The Use of Templated Mesoporous Materials as Templates for the Development of Ordered Arrangements of Nanowire and Nanorods of Electronically Important Materials. In Nanoporous Materials III; Sayari, A., Jaroniec, M., Eds. 2002; Vol. 141, pp 337−344. (24) Ryan, K. M.; Erts, D.; Olin, H.; Morris, M. A.; Holmes, J. D. Three Dimensional Architectures of Ultra-High Density Semiconducting Nanowires Deposited on Chip. J. Am. Chem. Soc. 2003, 125, 6284−6288. (25) Wang, D. H.; et al. Electrodeposition of Metallic Nanowire Thin Films Using Mesoporous Silica Templates. Adv. Mater. 2003, 15, 130− 133. (26) Guo, D. J.; Ding, Y. Porous Nanostructured Metals for Electrocatalysis. Electroanalysis 2012, 24, 2035−2043. (27) Lee, K. R.; Kwon, Y. U. Hard Templates for Fabrication of Nanostructured Films. Nano 2010, 5, 75−87. (28) Tian, B. Z.; et al. Facile Synthesis and Characterization of Novel Mesoporous and Mesorelief Oxides with Gyroidal Structures. J. Am. Chem. Soc. 2004, 126, 865−875. (29) Yang, H. F.; Zhao, D. Y. Synthesis of Replica Mesostructures by the Nanocasting Strategy. J. Mater. Chem. 2005, 15, 1217−1231. (30) Kumar, P.; Guliants, V. V. Periodic Mesoporous OrganicInorganic Hybrid Materials: Applications in Membrane Separations and Adsorption. Microporous Mesoporous Mater. 2010, 132, 1−14. (31) Fujita, S.; Koiwai, A.; Kawasumi, M.; Inagaki, S. Enhancement of Proton Transport by High Densification of Sulfonic Acid Groups in Highly Ordered Mesoporous Silica. Chem. Mater. 2013, 25, 1584− 1591. (32) Elbert, J.; Krohm, F.; Ruttiger, C.; Kienle, S.; Didzoleit, H.; Balzer, B. N.; Hugel, T.; Stuhn, B.; Gallei, M.; Brunsen, A. PolymerModified Mesoporous Silica Thin Films for Redox-Mediated Selective Membrane Gating. Adv. Funct. Mater. 2014, 24, 1591−1601. (33) Alonso, B.; Fayon, F.; Massiot, D.; Amenitsch, H.; Malfatti, L.; Kidchob, T.; Costacurta, S.; Innocenzi, P. Hybrid Organic-Inorganic Mesostructured Membranes: Interfaces and Organization at Different Length Scales. J. Phys. Chem. C 2010, 114, 11730−11740. (34) Chen, Z.; et al. DNA Translocation through an Array of Kinked Nanopores. Nat. Mater. 2010, 9, 667−675.

ASSOCIATED CONTENT

S Supporting Information *

FTIR gas cell with ZnSe windows and background single beam FTIR spectrum. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address †

Celsion, Inc., 997 Lenox Dr. Suite 100, Lawrenceville, NJ 08648. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This material is based upon initial FTIR analysis which was conducted as part of a U.S. Department of Energy EPSCoR University-National Laboratory Partnership Award under Grant DE-FG02-03ER46033 and analysis and interpretation of the data performed as part of a U.S. Department of Energy EPSCoR Implementation Award under Grant DE-FG0207ER46375.

■

REFERENCES

(1) Ogawa, M. Formation of Novel Oriented Transparent Films of Layered Silica-Surfactant Nanocomposites. J. Am. Chem. Soc. 1994, 116, 7941−7942. (2) Ogawa, M. A Simple Set-Gel Route for the Preparation of SilicaSurfactant Mesostructured Materials. Chem. Commun. 1996, 32, 1149−1150. (3) Ogawa, M. Preparation of Transparent Thin Films of SilicaSurfactant Mesostructured Materials. Supramol. Sci. 1998, 5, 247−251. (4) Lu, Y. G. R.; Drewien, C. A.; Anderson, M. T.; Brinker, C. J.; Gong, W.; Guo, Y.; Soyez, H.; Dunn, B.; Huang, M. H.; Zink, J. I. Continuous Formation of Supported Cubic and Hexagonal Mesoporous Films by Sol−Gel Dip-Coating. Nature 1997, 389, 364−368. (5) Brinker, C. J.; Lu, Y.; Sellinger, A.; Fan, H. Evaporation-Induced Self-Assembly. Nanostructures Made Easy. Adv. Mater. 1999, 11, 579− 585. (6) Guliants, V. V.; Carreon, M. A.; Lin, Y. S. Ordered Mesoporous and Macroporous Inorganic Films and Membranes. J. Membr. Sci. 2004, 235, 53−72. (7) Brinker, C. J.; Dunphy, D. R. Morphological Control of Surfactant-Templated Metal Oxide Films. Curr. Opin. Colloid Interface Sci. 2006, 11, 126−132. (8) Yamauchi, Y.; Suzuki, N.; Radhakrishnan, L.; Wang, L. Breakthrough and Future: Nanoscale Controls of Compositions, Morphologies, and Mesochannel Orientations toward Advanced Mesoporous Materials. The Chemical Record 2009, 9, 321−339. (9) Innocenzi, P.; Malfatti, L. Mesoporous Thin Films: Properties and Applications. Chem. Soc. Rev. 2013, 42, 4198−4216. (10) Faustini, M.; Boissiere, C.; Nicole, L.; Grosso, D. From Chemical Solutions to Inorganic Nanostructured Materials: A Journey into Evaporation-Driven Processes. Chem. Mater. 2014, 26, 709−723. (11) Sellinger, A.; Weiss, P. M.; Nguyen, A.; Lu, Y. F.; Assink, R. A.; Gong, W. L.; Brinker, C. J. Continuous Self-Assembly of OrganicInorganic Nanocomposite Coatings That Mimic Nacre. Nature 1998, 394, 256−260. 19458

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

(35) Brinker, C. J. Evaporation-Induced Self-Assembly: Functional Nanostructures Made Easy. MRS Bull. 2004, 29, 631−640. (36) Liu, N. G.; Dunphy, D. R.; Atanassov, P.; Bunge, S. D.; Chen, Z.; Lopez, G. P.; Boyle, T. J.; Brinker, C. J. Photoregulation of Mass Transport through a Photoresponsive Azobenzene-Modified Nanoporous Membrane. Nano Lett. 2004, 4, 551−554. (37) Koltz, M.; A, P. A.; Ayral, A.; Menager, C.; Grosso, D.; Van der Lee, A.; Cabuil, V.; Babonneau, F.; Guizard, C. The True Structure of Hexagonal Mesophase-Templated Silica Films as Revealed by X-Ray Scattering Effects of Thermal Treatments and of Nanoparticle Seeding. Chem. Mater. 2000, 12, 1721−1728. (38) Grosso, D.; Balkenende, A. R.; Albouy, P. A.; Ayral, A.; Amentisch, H.; Babonneau, F. Two-Dimensional Hexagonal Mesoporous Sillica Thin Films Prepared from Block Copolymers: Detailed Characterization and Formation Mechanism. Chem. Mater. 2001, 13, 1848−1856. (39) Grosso, D.; Babonneau, F.; Albouy, P.; Amenitsch, H.; Balkenende, A. R.; Brunet-Bruneau, A.; Rivory, J. An in Situ Study of Mesostructured CTAB-Silica Film Formation During Dip Coating Using Time-Resolved SAXS and Interferometry Measurements. Chem. Mater. 2002, 14, 931−939. (40) Doshi, D. A.; Gibaud, A.; Goletto, V.; Lu, M.; Gerung, H.; Ocko, B.; Han, S. M.; Brinker, C. J. Peering into the Self-Assembly of Surfactant Templated Thin-Film Silica Mesophases. J. Am. Chem. Soc. 2003, 125, 11646−11655. (41) Gibaud, A.; Grosso, D.; Smarsly, B.; Baptiste, A.; Bardeau, J. F.; Babonneau, F.; Doshi, D. A.; Chen, Z.; Brinker, C. J.; Sanchez, C. Evaporation-Controlled Self-Assembly of Silica Surfactant Mesophases. J. Phys. Chem. B 2003, 107, 6114−6118. (42) Grosso, D.; Babonneau, F.; Sanchez, C.; Soler-Illia, G. J.; Crepaldi, E. L.; Albouy, P.; Amenitsch, H.; Balkenende, A. R.; BrunetBruneau, A. A First Insight in the Mechanism Involved in the SelfAssembly of 2D-Hexagonal Templated SiO2 and TiO2 Mesostructured Films During Dip-Coating. J. Sol-Gel Sci. Technol. 2003, 26, 561−565. (43) Grosso, D.; Cagnol, F.; Soler-Illia, G.; Crepaldi, E. L.; Amenitsch, H.; Brunet-Bruneau, A.; Bourgeois, A.; Sanchez, C. Fundamentals of Mesostructuring through Evaporation-Induced SelfAssembly. Adv. Funct. Mater. 2004, 14, 309−322. (44) Innocenzi, P.; Malfatti, L.; Kidchob, T.; Costacurta, S.; Falcaro, P.; Piccinini, M.; Marcelli, A.; Morini, P.; Sali, D.; Amenitsch, H. TimeResolved Simultaneous Detection of Structural and Chemical Changes During Self-Assembly of Mesostructured Films. J. Phys. Chem. C 2007, 111, 5345−5350. (45) Grosso, D.; Ribot, F.; Boissiere, C.; Sanchez, C. Molecular and Supramolecular Dynamics of Hybrid Organic-Inorganic Interfaces for the Rational Construction of Advanced Hybrid Nanomaterials. Chem. Soc. Rev. 2011, 40, 829−848. (46) Innocenzi, P.; Malfatti, L.; Kidchob, T.; Falcaro, P.; Costacurta, S.; Guglielmi, M.; Mattei, G.; Bello, V.; Amenitsch, H. ThermalInduced Phase Transitions in Self-Assembled Mesostructured Films Studied by Small-Angle X-Ray Scattering. J. Synchrotron Radiat. 2005, 12, 734−738. (47) Grosso, D.; Boissière, C.; Nicole, L.; Sanchez, C. Preparation, Treatment and Characterisation of Nanocrystalline Mesoporous Ordered Layers. J. Sol-Gel Sci. Technol. 2006, 40, 141−154. (48) Gibaud, A.; Dourdain, S.; Gang, O.; Ocko, B. M. In Situ Grazing Incidence Small-Angle X-Ray Scattering Real-Time Monitoring of the Role of Humidity During the Structural Formation of Templated Silica Thin Films. Phys. Rev. B 2004, 70, 4. (49) Cagnol, F.; G, D.; Soler-Illia, G. J.; Crepaldi, E. L.; Babonneau, F.; Amenitsch, H.; Sanchez, C. Humidity-Controlled Mesostructuration in CTAB-Templated Silica Thin Film Processing. The Existence of a Modulable Steady State. J. Mater. Chem. 2003, 13, 61−66. (50) Urade, V. N.; Bollmann, L.; Kowalski, J. D.; Tate, M. P.; Hillhouse, H. W. Controlling Interfacial Curvature in Nanoporous Silica Films Formed by Evaporation-Induced Self-Assembly from Nonionic Surfactants. II. Effect of Processing Parameters on Film Structure. Langmuir 2007, 23, 4268−4278.

(51) Besson, S.; Gacoin, T.; Ricolleau, C.; Jacquiod, C.; Boilot, J. P. Phase Diagram for Mesoporous CTAB-Silica Films Prepared under Dynamic Conditions. J. Mater. Chem. 2003, 13, 404−409. (52) Boudot, M.; Gaud, V.; Louarn, M.; Selmane, M.; Grosso, D. SolGel Based Hydrophobic Antireflective Coatings on Organic Substrates: A Detailed Investigation of Ammonia Vapor Treatment (AVT). Chem. Mater. 2014, 26, 1822−1833. (53) Collard, X.; Van der Schueren, B.; Rooke, J. C.; Aprile, C.; Su, B. L. A Comprehensive Study of the Reaction Parameters Involved in the Synthesis of Silica Thin Films with Well-Ordered Uni-Directional Mesopores. J. Colloid Interface Sci. 2013, 401, 23−33. (54) Lee, Y.-F.; Chang, K.-H.; Chu, C.-Y.; Chen, H.-L.; Hu, C.-C. Microstructure Tuning of Mesoporous Silica Prepared by EvaporationInduced Self-Assembly Processes: Interactions among Solvent Evaporation, Micelle Formation/Packing and Sol Condensation. RSC Adv. 2011, 1, 401−407. (55) Keller, A.; Segal-Peretz, T.; Kauffmann, Y.; Frey, G. L. Control over in-Channel Mesostructure Orientation through Aam Surface Modification. Phys. Chem. Chem. Phys. 2013, 15, 13637−13645. (56) Platschek, B.; Keilbach, A.; Bein, T. Mesoporous Structures Confined in Anodic Alumina Membranes. Adv. Mater. 2011, 23, 2395−2412. (57) Koganti, V. R.; Dunphy, D.; Gowrishankar, V.; McGehee, M. D.; Li, X. F.; Wang, J.; Rankin, S. E. Generalized Coating Route to Silica and Titania Films with Orthogonally Tilted Cylindrical Nanopore Arrays. Nano Lett. 2006, 6, 2567−2570. (58) Koganti, V. R.; Rankin, S. E. Synthesis of Surfactant-Templated Silica Films with Orthogonally Aligned Hexagonal Mesophase. J. Phys. Chem. B 2005, 109, 3279−3283. (59) Stein, A.; Rudisill, S. G.; Petkovich, N. D. Perspective on the Influence of Interactions between Hard and Soft Templates and Precursors on Morphology of Hierarchically Structured Porous Materials. Chem. Mater. 2014, 26, 259−276. (60) van Beek, J. J.; Seykens, D.; Jansen, J. B. H.; Schuiling, R. D. Incipient Polymerization of SiO2 in Acid Catalyzed Tmos Sol-Gel Systems with Molar Water/Alkoxide Ratio between 0.5 and 32. J. NonCryst. Solids 1991, 134, 14−22. (61) Assink, R. A.; Kay, B. D. Sol-Gel Kinetics by NMR. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 1991, 32, 506−7. (62) Assink, R. A.; Kay, B. D. Study of Sol-Gel Chemical Reaction Kinetics by NMR. Annu. Rev. Mater. Sci. 1991, 21, 491−513. (63) Doughty, D. H.; Assink, R. A.; Kay, B. D. Hydrolysis and Condensation Kinetics of Dimeric Sol-Gel Species by Silicon-29 NMR Spectroscopy. Adv. Chem. Ser. 1990, 224, 241−50. (64) Assink, R. A.; Kay, B. D. Sol-Gel Kinetics: Silicon-29 NMR and a Statistical Reaction Model. Mater. Res. Soc. Symp. Proc. 1988, 121, 25−35. (65) Kay, B. D.; Assink, R. A. High Field 1H NMR Studies of Sol-Gel Kinetics. Mater. Res. Soc. Symp. Proc. 1986, 73, 157−64. (66) Assink, R. A.; Kay, B. D. Proton NMR Studies of the Sol-Gel Transition. Mater. Res. Soc. Symp. Proc. 1984, 32, 301−6. (67) Pouxviel, J. C.; Boilot, J. P. Kinetics Study of the Acidic Catalyzed Polymerization of Tetraethoxysilane by Silicon-29 NMR. Mater. Res. Soc. Symp. Proc. 1988, 121, 37−42. (68) Pouxviel, J. C.; Boilot, J. P.; Beloeil, J. C.; Lallemand, J. Y. NMR Study of the Sol/Gel Polymerization. J. Non-Cryst. Solids 1987, 89, 345−60. (69) Bonhomme, C.; Coelho, C.; Baccile, N.; Gervais, C.; Azaïs, T.; Babonneau, F. Advanced Solid State NMR Techniques for the Characterization of Sol−Gel-Derived Materials. Acc. Chem. Res. 2007, 40, 738−746. (70) Sanchez, J.; Rankin, S. E.; McCormick, A. V. 29Si NMR Kinetic Study of Tetraethoxysilane and Ethyl-Substituted Ethoxysilane Polymerization in Acidic Conditions. Ind. Eng. Chem. Res. 1996, 35, 117−129. (71) Rankin, S. E.; Macosko, C. W.; McCormick, A. V. Sol-Gel Polycondensation Kinetic Modeling: Methylethoxysilanes. AIChE J. 1998, 44, 1141−1156. 19459

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

(72) Rankin, S. E.; Šefčík, J.; McCormick, A. V. Similarities in the Hydrolysis Pseudoequilibrium Behavior of Methyl-Substituted Ethoxysilanes. Ind. Eng. Chem. Res. 1999, 38, 3191−3198. (73) Rankin, S. E.; McCormick, A. V. Hydrolysis Pseudoequilibrium: Challenges and Opportunities to Sol-Gel Silicate Kinetics. Chem. Eng. Sci. 2000, 55, 1955−1967. (74) Kasehagen, L. J.; Rankin, S. E.; McCormick, A. V.; Macosko, C. W. Modeling of First Shell Substitution Effects and Preferred Cyclization in Sol-Gel Polymerization. Macromolecules 1997, 30, 3921−3929. (75) Rankin, S. E.; Macosko, C. W.; McCormick, A. V. Importance of Cyclization During the Condensation of Hydrolyzed Alkoxysilanes. Chem. Mater. 1998, 10, 2037−2040. (76) Rankin, S. E.; Kasehagen, L. J.; McCormick, A. V.; Macosko, C. W. Dynamic Monte Carlo Simulation of Gelation with Extensive Cyclization. Macromolecules 2000, 33, 7639−7648. (77) Šefčík, J.; Rankin, S. E. Monte Carlo Simulations of Size and Structure of Gel Precursors in Silica Polycondensation. J. Phys. Chem. B 2003, 107, 52−60. (78) Li, X.; Rankin, S. E. Multiscale Dynamic Monte Carlo/ Continuum Model of Drying and Nonideal Polycondensation in SolGel Silica Films. AIChE J. 2010, 56, 2946−2956. (79) Li, X.; Rankin, S. E. Influence of Unlimited 3-Membered Ring Cyclization on a Multiscale Dynamic Monte Carlo/Continuum Model of Drying and Curing in Sol-Gel Silica Films. Chem. Eng. Sci. 2011, 66, 1015−1026. (80) Li, X.; Rankin, S. E. Multiscale Modeling of the Effects of Unlimited Four-Membered Ring Formation on Sol-Gel Silica Film Molecular Structure. AIChE J. 2013, 59, 707−718. (81) Devreux, F.; Boilot, J. P.; Chaput, F.; Lecomte, A. Sol-Gel Condensation of Rapidly Hydrolyzed Silicon Alkoxides: A Joint Si-29 NMR and Small-Angle X-Ray-Scattering Study. Phys. Rev. A 1990, 41, 6901−6909. (82) Devreux, F.; Boilot, J. P.; Chaput, F.; Sapoval, B. NMR Determination of the Fractal Dimension in Silica Aerogels. Phys. Rev. Lett. 1990, 65, 614−617. (83) Wood, D. L.; Rabinovich, E. M. Study of Alkoxide Silica Gels by Infrared Spectroscopy. Appl. Spectrosc. 1989, 43. (84) Matos, M. C.; Ilharco, L. M.; Almeida, R. M. The Evolution of TEOS to Silica Gel and Glass by Vibrational Spectroscopy. J. NonCryst. Solids 1992, 147-148, 232−237. (85) Calabro, D. C.; Valyocsik, E. W.; Ryan, F. X. In Situ ATR/FTIR Study of Mesoporous Silicate Synthesis. Micropor. Mater. 1996, 7, 243−259. (86) Holmes, S. M.; Zholobenko, V. L.; Thursfield, A.; Plaisted, R. J.; Cundy, C. S.; Dwyer, J. In Situ FTIR Study of the Formation of MCM-41. J. Chem. Soc., Faraday Trans. 1998, 94, 2025−2032. (87) Tejedor-Tejedor, M. I.; Paredes, L.; Anderson, M. A. Evaluation of ATR-FTIR Spectroscpy as an “in Situ” Tool for Following the Hydrolysis and Condensation of Alkoxysilanes under Rich H2O Conditions. Chem. Mater. 1998, 10, 3410−3421. (88) Yoshino, H.; Kamiya, K.; Nasu, H. IR Study on the Structural Evolution of Sol-Gel Derived SiO2 Gels in the Early Stage of Conversion to Glasses. J. Non-Cryst. Solids 1990, 126, 68−78. (89) Innocenzi, P. Infrared Spectroscopy of Sol-Gel Derived SilicaBased Films: A Spectra-Microstructure Overview. J. Non-Cryst. Solids 2003, 316, 309−319. (90) Primeau, N.; Vautey, C.; Langlet, M. The Effect of Thermal Annealing on Aerosol-Gel Deposited SiO 2 Films: A FTIR Deconvolution Study. Thin Solid FIlms 1997, 310, 47−56. (91) Parrill, T. M. Transmission Infrared Study of Acid-Catalyzed Sol-Gel Silica Coatings During Room Ambient Drying. J. Mater. Res. 1992, 7, 2230−2239. (92) Muroya, M. Correlation between the Formation of Silica Skeleton Structure and Fourier Transform Relection Infrared Absorption Spectroscopy Spectra. Colloids Surf. 1999, 157, 147−155. (93) Innocenzi, P.; Falcaro, P.; Grosso, D.; Babonneau, F. OrderDisorder Transitions and Evolution of Silica Structure in Self-

Assembled Mesostructured Silica Films Studied through FTIR Spectroscopy. J. Phys. Chem. B 2003, 107, 4711−4717. (94) Innocenzi, P.; Kidchob, T.; Bertolo, J. M.; Piccinini, M.; Guidi, M. C.; Marcelli, C. Time-Resolved Infrared Spectroscopy as an in Situ Tool to Study the Kinetics During Self-Assembly of Mesostructured Films. J. Phys. Chem. B 2006, 110, 10837−10841. (95) Innocenzi, P.; Kidchob, T.; Malfatti, L.; Costacurta, S.; Takahashi, M.; Piccinini, M.; Marcelli, A. In-Situ Study of Sol-Gel Processing by Time-Resolved Infrared Spectroscopy. J. Sol-Gel Sci. Technol. 2008, 48, 253−259. (96) Innocenzi, P.; Malfatti, L.; Kidchob, T.; Grosso, D. Controlling the Processing of Mesoporous Titania Films by in Situ FTIR Spectroscopy: Getting Crystalline Micelles into the Mesopores. J. Phys. Chem. C 2010, 114, 10806−10811. (97) Innocenzi, P.; Malfatti, L.; Kidchob, T.; Falcaro, P.; Guidi, M. C.; Piccinini, M.; Marcelli, A. Kinetics of Polycondensation Reactions During Self-Assembly of Mesostructured Films Studied by in Situ Infrared Spectroscopy. Chem. Commun. 2005, 2384−2386. (98) Marcelli, A.; Innocenzi, P.; Malfatti, L.; Newton, M. A.; Rau, J. V.; Ritter, E.; Schade, U.; Xu, W. IR and X-Ray Time-Resolved Simultaneous Experiments: An Opportunity to Investigate the Dynamics of Complex Systems and Non-Equilibrium Phenomena Using Third-Generation Synchrotron Radiation Sources. J. Synchrot. Radiat. 2012, 19, 892−904. (99) Innocenzi, P.; Malfatti, L.; Piccinini, M.; Marcelli, A.; Groso, D. Water Evaporation Studied by in Situ Time-Resolved Infrared Spectroscopy. J. Phys. Chem. A 2009, 113, 2745−2749. (100) Innocenzi, P.; Malfatti, L.; Piccinini, M.; Marcelli, A. Evaporation-Induced Crystallization of Pluronic F127 Studied in Situ by Time-Resolved Infrared Spectroscopy. J. Phys. Chem. A 2010, 114, 304−308. (101) Malfatti, L.; Innocenzi, P. Sol-Gel Chemistry: From SelfAssembly to Complex Materials. J. Sol-Gel Sci. Technol. 2011, 60, 226− 235. (102) Innocenzi, P.; Figus, C.; Takahashi, M.; Piccinini, M.; Malfatti, L. Structural Evolution During Evaporation of a 3-Glycidoxypropyltrimethoxysilane Film Studied in Situ by Time Resolved Infrared Spectroscopy. J. Phys. Chem. A 2011, 115, 10438−10444. (103) Fidalgo, A.; Ilharco, L. M. Correlation between Physical Properties and Structure of Silica Xerogels. J. Non-Cryst. Solids 2004, 347, 128−137. (104) van der Vis, M.; Konings, R.; Oskam; Snoeck, T. The Vibrational Spectra of Gaseous and Liquid Tetraethoxysilane. J. Mol. Struct. 1992, 274, 47−57. (105) Mondragon, M. A.; Costano, V. M.; Garcia, J. M.; Tellez, S. C. Vibrational Analysis of Si(OC2H5)4 and Spectroscopic Studies on the Formation of Glasses via Silica Gels. Vibr. Spectrosc. 1995, 9, 293. (106) Allinger, N. L.; Rahman, M.; Lii, J. H. A Molecular Mechanics Force Field (MM3) for Alcohols and Ethers. J. Am. Chem. Soc. 1990, 112, 8293−8307. (107) Téllez, L.; Rubio, J.; Rubio, F.; Morales, E.; Oteo, J. L. FT-IR Study of the Hydrolysis and Polymerization of Tetraethyl Orthosilicate and Polydimethyl Siloxane in the Presence of Tetrabutyl Orthotitanate. Spectrosc. Lett. 2004, 37, 11−31. (108) Lei, J. H.; Liu, D.; Guo, L. P.; Yan, X. M.; Tong, H. Fabrication and Characterization of Hexagonal Mesoporous Silica Monolith Via Post-Synthesized Hydrothermal Process. J. Sol-Gel Sci. Technol. 2006, 39, 169−174. (109) Gonçalves, M. C.; Attard, G. S. Nanostructured Mesoporous Silica Films. Rev. Adv. Mater. Sci. 2003, 4, 147−164. (110) Morita, M.; Ohmi, T.; Hasegawa, E.; Kawakami, M.; Ohwada, M. Growth of Native Oxide on a Silicon Surface. J. Appl. Phys. 1990, 68, 1272−1281. (111) Flodstrom, K.; Teixeira, C. V.; Amenitsch, H.; Alfredsson, V.; Linden, M. In Situ Synchrotron Small-Angle X-Ray Scattering/X-Ray Diffraction Study of the Formation of SBA-15 Mesoporous Silica. Langmuir 2004, 20, 4885−4891. (112) Khodakov, A. Y.; Zholobenko, V. L.; Imperor-Clerc, M.; Durand, D. Characterization of the Initial Stages of SBA-15 Synthesis 19460

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461

The Journal of Physical Chemistry C

Article

by in Situ Time-Resolved Small-Angle X-Ray Scattering. J. Phys. Chem. B 2005, 109, 22780−22790. (113) Ruthstein, S.; Schmidt, J.; Kesselman, E.; Talmon, Y.; Goldfarb, D. Resolving Intermediate Solution Structures During the Formation of Mesoporous SBA-15. J. Am. Chem. Soc. 2006, 128, 3366−3374. (114) Ryoo, R.; Ko, C. H.; Kruk, M.; Antochshuk, V.; Jaroniec, M. Block-Copolymer-Templated Ordered Mesoporous Silica: Array of Uniform Mesopores or Mesopore-Micropore Network? J. Phys. Chem. B 2000, 104, 11464−11471. (115) Li, D.; Guan, X.; Song, J.; Di, Y.; Zhang, D.; Ge, X.; Zhao, L.; Xiao, F. Highly Efficient Synthesis of Ordered Mesoporous Silica Materials with Controllable Microporosity Using Surfactant Mixtures as Templates. Colloids Surf., A 2006, 272, 194−202. (116) Van Der Voort, P.; Benjelloun, M.; Vansant, E. F. Rationalization of the Synthesis of SBA-16: Controlling the Microand Mesoporosity. J. Phys. Chem. B 2002, 106, 9027−9032. (117) Sundblom, A.; Palmqvist, A. E. C.; Holmberg, K. Study of the Pluronic-Silica Interaction in Synthesis of Mesoporous Silica under Mild Acidic Conditions. Langmuir 2010, 26, 1983−1990. (118) Gage, R. A.; Currie, E. P. K.; Stuart, M. A. C. Adsorption of Nanocolloidal SiO2 Particles on PEO Brushes. Macromolecules 2001, 34, 5078−5080. (119) Zhao, D. Y.; Huo, Q. S.; Feng, J. L.; Chmelka, B. F.; Stucky, G. D. Nonionic Triblock and Star Diblock Copolymer and Oligomeric Surfactant Syntheses of Highly Ordered, Hydrothermally Stable, Mesoporous Silica Structures. J. Am. Chem. Soc. 1998, 120, 6024− 6036. (120) Alonso, B.; Balkenende, R.; Albouy, P.; Durand, D.; Babonneau, F. Directing Role of pH and Ethanol Vapour on the Formation of 2D or 3D Mesostructured Silica and Hybrid OrganoSilica Thin Films. New J. Chem. 2002, 26, 1270−1272.

19461

dx.doi.org/10.1021/jp505651j | J. Phys. Chem. C 2014, 118, 19450−19461