Substituent-Dominated Structure Evolution during Sol−Gel Synthesis

May 5, 2010 - †School of Chemistry and Materials Science, Shaanxi Normal University, Xi'an 710062, People's Republic of. China, and ‡Department of...
0 downloads 0 Views 2MB Size
pubs.acs.org/Langmuir © 2010 American Chemical Society

Substituent-Dominated Structure Evolution during Sol-Gel Synthesis: A Comparative Study of Sol-Gel Processing of 3-Glycidoxypropyltrimethoxysilane and Methacryloxypropyltrimethoxysilane Shukun Shen,† Peipei Sun,† Wei Li,† Atul N. Parikh,*,‡ and Daodao Hu*,† †

School of Chemistry and Materials Science, Shaanxi Normal University, Xi’an 710062, People’s Republic of China, and ‡Department of Applied Science, University of California, Davis, California 95616 Received October 24, 2009. Revised Manuscript Received April 8, 2010

The sol-gel processes of 3-glycidoxypropyltrimethoxysilane (GPTMS) and methacryloxypropyltrimethoxysilane (MAPTMS) have been followed by fluorescence spectroscopy with pyranine as a photophysical probe. The experimental results showed that this probe is sensitive to the structural evolution and microenvironment polarity. The specific comparison of the structural evolution in two substituted organotrialkoxysilanes, namely, MAPTMS and GPTMS, illustrates the ability of the substituents to interact with the microenvironment via electrostatic interactions. Interestingly, these interactions determine the kinds of intermediate supramolecular structures that form during the sol-gel process and hence control the structure of the ensuing sol-gel end product. In particular, the amphiphile-like character of the MAPTMS intermediates contrasts with the biamphiphilic character of their GPTMS counterparts, driving distinctly different transient and local molecular organizations, which in turn modulate the hydrolysis and condensation reactions during the sol-gel process.

1. Introduction Substituted trialkoxysilanes have proven to be versatile building blocks for designing silica-based organic-inorganic hybrid materials1-3 via sol-gel methods. The hydrolytic polycondensation of the trialkoxysilyl residue produces an inorganic silica network, and the covalently linked organic substituents allow the incorporation of a host of useful functionalities.3 Within this class of organic-inorganic materials, (3-glycidoxypropyl) trimethoxysilane4 (GPTMS) and methacryloxypropyl trimethoxysilane5 (MAPTMS) assume special importance for emerging technological applications of sol-gel-derived materials. This is because their organic substituents (viz. epoxide or methacryloxy groups) sensitively respond to optical, thermal, and chemical stimulii by undergoing intermolecular polymerization and thus a dramatic property change.6 Indeed, organic-inorganic hybrid materials prepared using these GPTMS and MAPTMS precursors are finding niche applications in the design of many technologically useful materials including antiscratch coatings, contact lens materials, passivation *To whom correspondence should be addressed. Tel: þ86 29 85307584. Email: [email protected] (D.H.) Tel: +1 (530) 304-7523. Email: anparikh@ ucdavis.edu (A.N.P). (1) Baney, R. H.; Itoh, M.; Sakakibara, A.; Suzuki, T. Chem. Rev. 1995, 95, 1409–1430. (2) Loy, D. A.; Shea, K. J. Chem. Rev. 1995, 95, 1431–1442. (3) Sanchez, C.; Ribot, F.; Lebeau, B. J. Mater. Chem. 1999, 9, 35–44. (4) Innocenzi, P.; Brusatin, G.; Guglielmi, M.; Bertani, R. Chem. Mater. 1999, 11, 1672–1679. (5) Kahraman, M. V.; Kugu, M.; Menceloglu, Y.; Kayaman-Apohan, N.; Gungor, A. J. Non-Cryst. Solids 2006, 352, 2143–2151. (6) Philipp, G.; Schmidt, H. J. Non-Cryst. Solids 1984, 63, 283–292. (7) Brusatin, G.; Della Giustina, G.; Romanato, F.; Guglielmi, M. Nanotechnology 2008, 19. (8) Lukowiak, A.; Strek, W. J. Sol-Gel Sci. Technol. 2009, 50, 201–215. (9) Soppera, O.; Croutxe-Barghorn, C.; Carre, C.; Blanc, D. Appl. Surf. Sci. 2002, 186, 91–94. (10) Zhang, J. T.; Huang, S. W.; Zhuo, R. X. Colloid Polym. Sci. 2005, 284, 209– 213. (11) Innocenzi, P.; Brusatin, G.; Babonneau, F. Chem. Mater. 2000, 12, 3726– 3732.

7708 DOI: 10.1021/la904040c

layers for microelectronics, multifunctional coatings, chemical and biological sensors, and optical devices.7-12 The typical route for synthesizing organic-inorganic materials using MAPTMS is to employ acidic conditions to catalyze the inorganic sol-gel polymerization and to use a free-radical initiator or thermal curing for the methacryloxy functional groups’ polymerization.4,13 For GPTMS, a low-pH environment provides a chemical stimulus for both the inorganic sol-gel polymerization and the organic polymerization of the epoxide (forming poly(ethylene)oxide (PEO) linkages). From a material structure point of view, the organic and inorganic networks in the composite materials prepared through the two precursors can form, which offers an easy way to synthesize organic-inorganic composite materials with molecular-level homogeneity. It is worthwhile to note that there is an effect of a former network on a later one’s formation, and the resulting synergy directly influences the structure of the final sol-gel organic-inorganic hybrid material. Indeed, on the basis of the results from gel permeation chromatography, Piana and Schubert14 found that in the absence of catalyst the hydrolysis of MAPTMS was slower than that of GPTMS. Moreover, the hydrolysis of GPTMS resulted in rather small oligomers instead of high polymers and that the oligomers were relatively inert toward further condensation; however, contrary to GPTMS hydrolysis, the mean molecular mass did not approach a constant value even after a long time. Innocenzi and co-workers11 have recently shown that synthesis conditions that promote rapid silica condensation (e.g., high catalyst concentrations) result in a lower degree of polymerization of the epoxy group in GPTMS. They ascribe this observation to increased steric repulsion, which limits the access of epoxy groups for cross-linking and polymerization. Pankow and Schmidt-Naake (12) Schottner, G. Chem. Mater. 2001, 13, 3422–3435. (13) Loy, D. A.; Baugher, B. M.; Baugher, C. R.; Schneider, D. A.; Rahimian, K. Chem. Mater. 2000, 12, 3624–3632. (14) Piana, K.; Schubert, U. Chem. Mater. 1994, 6, 1504–1508.

Published on Web 05/05/2010

Langmuir 2010, 26(11), 7708–7716

Shen et al.

pointed out that the conversion of the double bonds in MAPTMS is quite low because of the tight packing of the methacryloxy groups in the organosilicate, leading to obstruction through the growing polymer chains.15 The results from NMR and FT-IR also indicated that the substituents in organic siloxanes could affect the final sol-gel products.11,16-20 Although the effect of organic moieties in organically substituted trialkoxysilanes on the structure of corresponding organosilicate has been found and some explanations have been given, the effect of the molecular organization that occurs during the sol-gel processes has scarcely been considered. It is well known that the luminescence properties of fluorescent dyes are sensitively dependent on changes in the local chemical microenvironment in heterogeneous systems and real-time monitoring of fluorescence emission offers an excellent tool for characterizing many aspects of sol-gel chemistry including reaction kinetics, aging, and drying. As first reported by Avnir and co-workers21 more than two decades ago, the use of fluorescent probes to characterize sol-gel reactions has gained considerable popularity.22 Many previous studies mainly focused on the change in the microenvironment of products during the transition from sol to gel or/and gel to xerogel. The evolution, especially from organosiloxanes to hydrolyzed intermediates, hydrolyzed species to condensed oligomer, and oligomer to gel, however, is rarely considered. In two recent studies, we used steady-state fluorescence emission due to 2-napthanol23 and pyrene24 to monitor the hydrolysis and condensation processes of MAPTMS and GPTMS organosiloxanes in real time. Both probes are excellent indicators of the microenvironment polarity. The temporal dynamics of the fluorescence emission in those studies led us to postulate a large-scale microstructural evolution via supramolecular assembly and the dynamic reorganization of partially hydrolyzed/polymerized intermediates during sol-gel conversion. To develop a molecular-level understanding of the difference in hydrolysis and condensation between MAPTMS and GPTMS, in the work reported here we employ pyranine (8-hydroxy-1,3,6pyrene trisulfonic acid trisodium salt) as a photoprobe to directly compare the microstructural evolution and determine the role of environment polarity during the sol-gel transition in MAPTMS and GPTMS. Pyranine is particularly well suited for our study because of its stability and high sensitivity to the proton donor-acceptor abilities of the surrounding medium (or large changes in the pH) during sol-gel conversion.25,26 Specifically, as the proton-transfer property of the sol-gel reaction medium changes, pyranine exhibits markedly different luminescence behavior.27,28 For instance, in strongly polar, proton-accepting (15) Pankow, O.; Schmidt-Naake, G. Macromol. Mater. Eng. 2004, 289, 990996. (16) Delattre, L.; Dupuy, C.; Babonneau, F. J. Sol-Gel Sci. Technol. 1994, 2, 185–188. (17) Lavrencic Stangar, U.; Sassi, A.; Venzo, A.; Zattin, A.; Japelj, B.; Orel, B.; Gross, S. J. Sol-Gel Sci. Technol. 2009, 49, 329–335. (18) Lazghab, M.; Saleh, K.; Guigon, P. Chem. Eng. Res. Des. 2009, DOI: 10.1016/j.cherd.2009.11.005. (19) Innocenzi, P.; Brusatin, G.; Guglielmi, M.; Bertani, R. Chem. Mater. 1999, 11, 1672–1679. (20) Innocenzi, P.; Esposto, M.; Maddalena, A. J. Sol-Gel Sci. Technol. 2001, 20, 293–301. (21) Avnir, D.; Levy, D.; Reisfeld, R. J. Phys. Chem. 1984, 88, 5956–5959. (22) Dunn, B.; Zink, J. I. Chem. Mater. 1997, 9, 2280–2291. (23) Hu, D. D.; Croutxe-Barghorn, C.; Feuillade, M.; Carre, C. J. Phys. Chem. B 2005, 109, 15214–15220. (24) Shen, S. K.; Hu, D. D. J. Phys. Chem. B 2007, 111, 7963–7971. (25) Dunn, B.; Zink, J. I. J. Mater. Chem. 1991, 1, 903–913. (26) Pouxviel, J. C.; Dunn, B.; Zink, J. I. J. Phys. Chem. 1989, 93, 2134–2139. (27) Agmon, N.; Huppert, D.; Masad, A.; Pines, E. J. Phys. Chem. 1991, 95, 10407–10413. (28) Suwaiyan, A.; Aladel, F.; Hamdan, A.; Klein, U. K. A. J. Phys. Chem. 1990, 94, 7423–7429.

Langmuir 2010, 26(11), 7708–7716

Article Chart 1. Chemical Structures of Pyranine, MAPTMS, and GPTMS

media (e.g., water at pH >3), the fluorescence from pyranine is dominated by the green emission at 515 nm due to the preponderance of its deprotonated excited state (PyO-*). In contrast, ambient media that do not readily accept protons (e.g., methanol) stabilize the protonated form of pyranine (PyOH*), which exhibits emission in the blue at 430 nm. In previous studies using sol-gel materials, pyranine has been successfully used to (1) measure the water/methanol content of aluminosilicate bulk gels during the sol-gel-xerogel transformations; (2) determine the effect of the acidity of the medium on water consumption during gelation; (3) determine the chemical evolution during thin sol-gel film formation via dip coating; and (4) characterize the changes in the pore environment during drying and aging of the sol-gel material.25,26,29 Our results demonstrate that a complex interplay among changes in (1) the solvent chemistry, (2) a substituentinduced supramolecular organization (or lack thereof), and (3) local and transient fluctuations in environmental polarity determines both the kinetics of the sol-gel process and the final structure of the condensed phase. These results would offer new insights into understanding the difference between the sol-gel processes of MAPTMS and GPTMS.

2. Experimental Section 2.1. Materials. Pyranine, glycidoxypropyltrimethoxysilane (GPTMS), and methacryloxypropyltrimethoxysilane (MAPTMS) were purchased from Aldrich and used without further purification. Their chemical structures are shown in Chart 1. Water was deionized. Methanol was spectroscopic grade. Hydrochloric acid was reagent grade. The ionic exchange resin (ESINE Naþ) was obtained from CHAUNY. 2.2. Sample Preparation. In this research, sample preparation includes the following aspects.

2.2.1. Sample Preparation for Preliminary Measurements. 1. For Pyranine Fluorescence in Water/Methanol Mixtures. Methanol-containing pyranine (10-3 M) as the stock solution was first prepared. The stock solution (2.5 mL) was added to a 25 mL volumetric flask, and then acidified water (pH 2) in the desired volume percent was added and the complete volume was attained with methanol. In the research, 7, 10, 18, 22, 26, 30, and 32% water by volume percent were used.

2. For Fluorescence Spectroscopy of Pyranine in a Mixture of Methanol and Precursors. The MAPTMS or GPTMS solu-

tion containing pyranine (10-4 M) was first prepared. Three (29) Nishida, F.; McKiernan, J. M.; Dunn, B.; Zink, J. I.; Brinker, C. J.; Hurd, A. J. J. Am. Ceram. Soc. 1995, 78, 1640–1648.

DOI: 10.1021/la904040c

7709

Article

Shen et al.

milliliters of the solution was added to the cell (48  12.5  12.5 mm3; volume, 3.5 mL), and then 100 μL of methanol at constant intervals was added to the cell. As a result, solutions with different amounts of methanol were gradually obtained. It is worth mentioning that the concentration of pyranine in solution is basically maintained at 10-4 M in our experiment.

2.2.2. Sample Preparation for a Comparison of the Fluorescence Behavior of Pyranine during the Sol-Gel Processing of MAPTMS and GPTMS. Partial hydrolysis and the condensation of precursors were carried out by adding 0.75 (for MAPTMS, rw = 0.75) and 1.5 (for GPTMS, rw = 1.5) molar equivalents of acidified water (0.01 M HCl) to the siloxane precursors in a fluorescence cell (48  12.5  12.5 mm3; volume, 3.5 mL). Pyranine at a concentration of 10-4 M was introduced just after the acidified water addition. Then the cell was closed and shaken until the solution became clear for the fluorescence measurements.

2.2.3. Sample Preparation for Determining the Transfer of a Proton from the Acidified Water Phase to the GPTMS Phase. One milliliter of water (pH 5.5 or 3 containing 10-4 M pyranine) was carefully added to the cell containing 2 mL of GPTMS to ensure the formation of a clear boundary layer between bulk GPTMS and water. Then the fluorescence spectrum for both phases was recorded. We note that this experiment was allowed to proceed for about 30 min because the boundary became blurred after that time.

2.2.4. Sample Preparation for Investigating the Effect of Solvent Removal on the GPTMS Sol-Gel Reaction. The mixture in the cell as mentioned for GPTMS was vacuumed for a certain time after the reaction was allowed to proceed for a given time, and the fluorescence of the resulting mixture was measured after the end of vacuum. 2.3. Measurements. Fluorescence spectra were recorded using a fluorescence spectrophotometer (Fluoromax-2). The emission spectrum was obtained using an excitation wavelength of 345 nm. The fluorescence cell was closed for all measurements except as indicated.

3. Results and Discussion 3.1. Preliminary Measurements. During sol-gel processing, methanol-to-water and precursor-to-methanol ratios both undergo continuous changes, and the fluorescence of pyranine is sensitive to both changes. Therefore, it is necessary to get some basic information about the effects of the ratios on the fluorescence of pyranine, so the following two sets of preliminary experiments were performed. 3.1.1. Pyranine Fluorescence in Water/Methanol Mixtures. An important medium change during sol-gel conversion involves the formation of methanol. As methoxy (or ethoxy) silanes hydrolyze in the aqueous phase, methanol (or ethanol) is quantitatively released into the reaction mixture, thus altering the water-to-methanol ratio. As methanol enters the aqueous phase, its proton-acceptor property changes. Thus, the time-dependent ratio of the luminescence due to protonated and deprotonated pyranine during the sol-gel reaction may reflect the extent of the hydrolysis reaction during the sol-gel process due to the generation of methanol. To obtain the change in the fluorescence of pyranine with respect to the methanol concentration, preliminary fluorescence spectroscopy was carried out for pyranine in water/ methanol mixtures with different concentrations of methanol. The values of water/methanol ratios were chosen to represent the limiting range of conditions achieved during sol-gel processing. Representative fluorescence spectra are shown in Figure 1. For comparison with the sol-gel process, the acidified water (pH 2) containing 10-4 M pyranine was used to mix with methanol. The amount of methanol added reduces the water concentration in the 7710 DOI: 10.1021/la904040c

Figure 1. Fluorescence spectra of pyranine (10-4 M) in water (pH 2)/ methanol mixtures. The percentages shown are the volume percentages of water (7, 10, 18, 22, 26, 30, and 32%).

range of 32 and 7 vol % water. The spectrum for the highestpolarity medium investigated, namely, 32% water, reveals two overlapping peaks with emission maxima centered at 430 and 515 nm due to protonated and deprotonated pyranine, respectively. As the amount of methanol in the pyranine environment is raised (such as would occur because of the progression of the sol-gel reaction), the peak at ∼515 nm assigned to the deprotonated pyranine decreases and the one at ∼430 nm correspondingly increases. It is also notable that as the polarity of the medium decreases as a result of the increase in the amount of methanol, the weighted mean of the peak envelope due to PyOH* exhibits a small red shift. These observations are in excellent correspondence with those reported previously.26 This simple experiment confirms not only the ability of low-concentration pyranine to provide a reliable fluorescence measure of the changes in the polarity of the medium because of the addition (or generation) of methanol but also the stability of pyranine in the sol-gel-like environment. 3.1.2. Fluorescence Spectroscopy of Pyranine in a Mixture of Methanol and Precursors. In addition to the generation of methanol during the sol-gel reaction, the presence of sol-gel precursors, namely, MAPTMS or GPTMS, can also influence pyranine’s photophysical properties in the sol-gel medium. To establish the role of sol-gel precursors in modulating pyranine fluorescence, we carried out additional control experiments. Figure 2a,b shows a series of fluorescence spectra of pyranine in mixtures of methanol in MAPTMS and GPTMS as a function of the relative ratio of precursor to methanol. Note that these mixtures are anhydrous and do not contain any appreciable amount of water. Several features of these data are noteworthy. First, the spectra are characterized by broad envelopes with maximum emission intensity at ∼430 nm due to PyOH* with little or no emission for the PyO-* form at ∼515 nm. This behavior is consistent with the preferential partitioning of pyranine in the methanol phase (compared to sol-gel precursors) and additionally suggests that the addition of sol-gel precursors does not significantly alter the proton-transfer property of the methanol medium to any measurable extent. Second, the intensity of the envelopes at ∼430 nm for both MAPTMS/methanol and GPTMS/methanol mixtures decreases monotonically as the precursor concentration is raised. Note, however, that this decrease in emission due to PyOH* is not accompanied by an equivalent increase in emission due to PyO-*. This lends additional support to the foregoing inference that the fluorescence intensity changes observed are not due to any substantial differences in the polarity Langmuir 2010, 26(11), 7708–7716

Shen et al.

Figure 2. (a) Fluorescence spectra of pyranine (10-4 M) in a MAPTMS/methanol mixture. (b) Fluorescence spectra of pyranine (10-4 M) in a GPTMS/methanol mixture.

of the medium. Rather, this decreased quantum yield for pyranine in MAPTMS- and GPTMS-enriched methanolic media is most likely due to aggregation-induced fluorescence self-quenching. Such quenching is also consistent with the observed red shift in the fluorescence emission most clearly visible for GPTMS (inset of Figure 2). Such self-quenching is not surprising because increasing the relative fraction of hydrophobic MAPTMS or GPTMS in the mixtures makes the polarity of the medium decrease. As a result, the hydrophilic pyranine molecules are favorable to aggregation, which necessarily elevates the local concentration of pyranine in the mixture. In this case, the concentration quenching of fluorescence often occurs.23 These observations imply that the fluorescence intensity of pyranine is sensitive to the decrease in the amount of precursor during the sol-gel process. 3.2. Comparison of the Fluorescence Behavior of Pyranine During the Sol-Gel Processing of MAPTMS and GPTMS. The fluorescence spectra of pyranine during the entire sol maturation period for MAPTMS are dominated by the pyranine emission at ∼430 nm due to PyOH*. Panels a and b in Figure 3 summarize the evolution of wavelength at maximum intensity and peak intensity as a function of the sol maturation time (rw = 0.75). The plots immediately reveal four regimes of distinctly different fluorescence behavior. Regime I during the early phase of sol maturation is seen from the onset until about 60 h. This regime is characterized by oscillations in both wavelengths (between 447 and 450 nm) and the intensity of fluorescence emission. Between 60 and ∼150 h, regime I is replaced by regime II, characterized by a plateau in the fluorescence properties. In this regime, the wavelengths remain essentially fixed, displaying little or no time dependence, and the intensity exhibits a small decrease. In regime III, between 150 and ∼250 h, the fluorescence emission intensity continuously declines and the wavelength exhibits a small monotonic blue shift from roughly 446 to 444 nm. Langmuir 2010, 26(11), 7708–7716

Article

The replacement of this regime by regime IV is signaled by a second plateau that extends from ∼250 h until the maximum monitoring time of 600 h during sol maturation and aging. Figure 3c,d shows the changes in the emission wavelength and intensity of pyranine over time during GPTMS sol maturation. Comparing the fluorescence behavior of the GPTMS system with that of the MAPTMS system, a dramatic difference is immediately obvious: unlike the MAPTMS system, both the emission intensity and wavelength during the sol-gel reaction of the GPTMS system are essentially constant. This conspicuous absence of fluorescence emission fluctuations in the GPTMS system suggests fundamental differences in the microenvironment surrounding pyranine during its sol maturation. The implications of this difference in the temporal evolution of pyranine fluorescence for MAPTMS and GPTMS systems are addressed below. For MAPTMS, the dramatic oscillations in pyranine fluorescence observed in regime I are analogous to those observed in our previous work when 2-naphthol was used as a fluorescent probe.23 In that study, a detailed analysis of fluorescence fluctuations during the sol maturation of MAPTMS, in conjunction with a systematic study of the photophysical properties of 2-napthanol, allowed us to deduce a detailed molecular picture in terms of simultaneous changes in solvent chemistry and the structure of the incipient sol-gel material. Our current observations using a pyranine photoprobe lend further support to this picture and offer further structural insights because of its unique microenvironment dependence. Regime I captures changes during the early phase of the sol-gel process. At the onset of sol maturation, headgroup hydrolysis of MAPTMS precursor produces methanol and silanol bearing hydrolyzed MAPTMS species. Athough the introduction of additional methanol into the sol must change pyranine fluorescence reflecting a changing water/methanol ratio, the observed changes in pyranine emission here can not be accounted for by the changes in the water/methanol ratio itself because both amounts of water and methanol in the reaction mixture during the early stage are very low (