Study of Acetic Acid-Catalyzed Nanocomposite ... - ACS Publications

---

Langmuir 2003, 19, 7587-7591

7587

Study of Acetic Acid-Catalyzed Nanocomposite Organic/ Inorganic Ureasil Sol-Gel Ionic Conductors E. Stathatos and P. Lianos* Engineering Science Department, University of Patras, 26500 Patras, Greece

B. Orel, A. Surca Vuk, and R. Jese National Institute of Chemistry, Hajdrihova 19, SI-1000 Ljubljana, Slovenia Received February 10, 2003. In Final Form: May 27, 2003 Sol-gel ionic conductors based on nanocomposite ureasil have been studied by attenuated total reflection infrared spectroscopy, time-resolved fluorescence probing, and direct current conductivity. Gels were obtained by acetic acid-catalyzed solvolysis under ambient conditions. IR measurements have shown that acetic acid by itself cannot lead to gelation. Gelation is attained with help from ambient humidity. However, acetic acid solvolysis regulates gel evolution by formation of intermediate products, such as silicon ester and Si-O-Si oligomers. Fluorescence probing and conductivity measurements helped determine the right concentration of acetic acid in the sol to obtain gels with an optimal ionic conductivity. Fluorescence probing has also revealed that ionic conductivity is favored when gels are made of short-polyether-chain ureasil.

Introduction The sol-gel method for the synthesis of inorganic or nanocomposite organic/inorganic gels has become one of the most popular chemical procedures. This popularity stems from the fact that sol-gel synthesis is easy and it is carried out at ambient or a slightly elevated temperature so that it allows nondestructive organic doping. Indeed, the sol-gel method has led to the synthesis of a great variety of materials, the range of which is continuously expanding. Thus, the simple incorporation of organic dopants as well as the formation of organic/inorganic nanocomposites offer the possibility of efficient dispersion of functional compounds in gels; it allows modification of the mechanical properties of the gels and provides materials with very interesting optical properties. A typical sol-gel route for making oxide matrices and thin films is followed by the hydrolysis of alkoxides, for example, alkoxysilanes, alkoxytitanates, and so forth.1 However, a review of the recent literature reveals an increasing interest in another sol-gel route based on organic acid solvolysis of alkoxides.2-6 This second method seems to offer substantial advantages in several cases, and it is becoming the method of choice in the synthesis of organic/inorganic nanocomposite gels.3,4 As it has been earlier found by Pope and Mackenzie7 and later verified by others,4-6 organic (for example, acetic) acid solvolysis proceeds by a two-step mechanism that involves intermediate ester formation. Simplified reaction schemes * Author to whom correspondence should be addressed. Telephone 30-2610-997587, fax 30-2610-997803, e-mail lianos@ upatras.gr. (1) Hench, L. L.; West, K. J. Chem. Rev. 1990, 90, 33. (2) Green, H. W.; Le, P. K.; Grey, J.; Au, T. T.; Sailor, M. J. Science 1997, 276, 1826. (3) Bekiari, V.; Lianos, P. Chem. Mater. 1998, 10 (12), 3777. (4) Stathatos, E.; Lianos, P.; Lavrencic-Stangar, U.; Orel, B. Adv. Mater. 2002, 14, 354. (5) Ivanda, M.; Music, S.; Popovic, S.; Gotic, M. J. Mol. Struct. 1999, 480-481, 645. (6) Birnie, D. P. J. Mater. Sci. 2000, 3, 367. (7) Pope, E. J. A.; Mackenzie, J. D. J. Non-Cryst. Solids 1986, 87, 185.

showing gel formation either by hydrolysis or by organic acid solvolysis are presented by the following reactions. (Note that in these reactions only one metal-bound ligand is taken into account,6 while acetic acid (AcOH) is chosen to represent organic acids in organic acid solvolysis.)

Hydrolysis M-OR + H2O f M-OH + ROH

(1)

Acetic acid solvolysis M-OR + AcOH f M-OAc + ROH

(2a)

ROH + AcOH f ROAc + H2O

(2b)

M-OAc + ROH f ROAc + M-OH

(2c)

M-OR + M-OAc f ROAc + M-O-M

(2d)

where M is a metal (for example, Si or Ti) and R is a short alkyl chain (for example, methyl, ethyl, or isopropyl). Hydrolysis (1) produces a highly reactive hydroxide species M-OH, which, by inorganic polymerization, produces oxide, that is, M-O-M, that is the end product of the sol-gel process. More complicated is acetic acid solvolysis (2), where several different possibilities may define different intermediate routes to obtain oxide. Reaction 2a is a prerequisite of the remaining three reactions. The occurrence of reaction 2b would mean that water may be formed, which may lead to hydrolysis. Reaction 2c would create reactive M-OH, which would form oxide, while reaction 2d directly leads to oxide formation. The above possibilities have been demonstrated by various researchers by spectroscopic techniques.4-7 However, there still exists a lot of uncertainty, and there is no concrete model to describe a well-established procedure leading to oxide formation by organic acid solvolysis. For this reason, more work needs to be carried out on these systems. Reactions 2 reveal one certain fact. The quantity of acetic acid in solution will be crucial in affecting intermediate routes. Thus, reaction 2b is possible only if an excess of acetic acid is present. Also, the quantity of acetic acid will define

10.1021/la0300535 CCC: $25.00 © 2003 American Chemical Society Published on Web 07/19/2003

7588

Langmuir, Vol. 19, No. 18, 2003

Stathatos et al.

Figure 1. Chemical structure of the ureasil precursors (PPG4000, n ∼ 68; PPG230, n ∼ 3).

whether the solvolysis steps will simultaneously affect all available alkoxide ligands or will leave some of them intact and subject to hydrolysis reactions. For this reason and by realizing the importance of AcOH availability, we have decided to isolate and study this particular effect. The obtained results are presented in the present work. The sol-gel precursor used in the present work is not a simple alkoxide but a complex compound that consists of a polyether chain with two triethoxysilane end groups covalently bound by urea bridges (ureasil), as can be seen in Figure 1. This material is studied because it is used to make gel electrolytes applied to the construction of dyesensitized photoelectrochemical solar cells and electrochromic devices.4,8 For this reason, we have studied the effect of acetic acid solvolysis on the quality of the gels as related to their capacity to act as ionic conductors. Consequently, we have characterized the materials by conductivity measurements. In addition, as we have previously shown, ionic conductivity in a nanocomposite gel depends on the capacity of the ion-conducting organic subphase to percolate.9 Our experience teaches us that one, almost ideal, method to study percolation conditions in a complex material is to analyze time-resolved fluorescence quenching reactions between mobile species exclusively solubilized in the ion-conducting organic subphase. Subsequently, a large portion of the present work is devoted to time-resolved fluorescence quenching analysis. The data presented in the following begin with attenuated total reflection infrared (ATR IR) spectroscopy, which has been employed in an effort to define the route followed by the acetic acid solvolysis of ureasil precursors, in relation to reactions 1 and 2. Experimental Section Materials. All the chemicals used in the present work have been purchased from Aldrich. Transparent conductive oxide electrodes were cut from a SnO2:F coated glass (8 Ω/square) purchased from Hartford Glass Co., U.S.A. Synthesis of Hybrid Precursors. Two different unhydrolyzed alkoxysilane-polyether precursors were prepared as in previous publications.10 Poly(propylene glycol)bis(2-aminopropyl ether) of molecular weight 4000 or 230 and 3-isocyanatopropyltriethoxysilane (ICS; molar ratio ICS/diamine ) 2) were mixed in tetrahydrofuran (THF) under reflux (64 °C) for 6 h. The isocyanate group of ICS reacts with the amino groups of poly(propylene glycol)bis(2-aminopropyl ether) (acylation reaction), producing urea connecting groups between the polymer units and the inorganic part. After evaporation of THF under a vacuum, a viscous precursor was obtained, which is stable at room temperature for several months. The abbreviated names of the precursors used in the present work are PPG4000 and PPG230, referring to molecular weights 4000 and 230, respectively (cf. Figure 1). Sol-Gel Synthesis and Film Deposition. A total of 2 g of the precursor was mixed with 2 g of alcohol. After stirring for 5 min, AcOH was added and the mixture was stirred for about 5 h. Finally, a glass slide, previously cleaned in sulfochromic solution, was dipped into the sol and withdrawn at a speed of 31 mm/min. A thin, transparent, and optically uniform film was (8) Dahmouche, K.; Atik, M.; Mello, N. C.; Bonagamba, T. J.; Panepucci, H.; Judeinstein, P.; Aegerter, M. A. Sol. Energy Mater. Sol. Cells 1998, 54, 1. (9) Stathatos, E.; Lianos, P.; Krontiras, C. J. Phys. Chem. 2001, 105, 3486. (10) Stathatos, E.; Lianos, P.; Lavrencic-Stangar, U.; Orel, B.; Judeinstein, P.; Langmuir 2000, 16, 8672.

obtained, and it was used for the fluorescence studies. Fluorescent probes and quenchers {tris(2,2′-bipyridine)ruthenium dichloride hexahydrate [Ru(bpy)32+], pyrene, coumarine-153 [C-153}, and methylviologen [MV2+]} were introduced by previous solubilization in the alcohol used to prepare the precursor solutions. For the conductivity measurements, 0.5 M KI and 0.05 M I2 were added in the sol, giving an I3-/I- redox couple.4,9 The choice of this electrolyte was made because it is used in dye-sensitized photoelectrochemical solar cells, which the studied material is destined for.4 For the IR measurements, we used a sol made by a simple mixture of PPG230 with AcOH not containing ethanol, to avoid interference from the absorption bands of the latter. Experimental Methods. ATR IR spectra have been recorded by placing one drop of the sol on the top of a horizontal ATR crystal. Spectra have been recorded either by providing a dry N2 atmosphere or by exposing the sample to ambient humidity. The sol used for the IR measurements was a simple mixture of AcOH and PPG230 without any additive. Time-resolved fluorescence decay profiles were recorded with the single-photon-counting technique using an IBH hydrogen flash lamp as the excitation source. An Oxford Instruments PC multichannel analyzer card was used for data storage over 1024 channels. The direct current (DC) conductivity measurements were made by placing one drop of the gel containing electrolyte between two SnO2:F coated glass electrodes and squeezing them to give a solid film of a thickness of 6 µm, defined by appropriate spacers. The applied tension was 0.5 V, while the current was measured by a multimeter providing 1-µA resolution. Both the fluorescence probing and the conductivity measurements were performed at ambient conditions.

Results and Discussion Study of ATR IR Spectra. To get an idea of which of the reactions of the previous scheme 2 are valid in our case, we have studied gels made of PPG230 by ATR IR spectroscopy. PPG230 has been chosen because, as it will be seen in the following paragraphs, it is the material of choice for increased ionic conductivity. ATR IR spectra have been recorded in the absence of ethanol. This was necessary to follow the ethanol production during gelation. The sample was one drop of the sol specially prepared for IR measurements, deposited on the horizontal ATR cell. Spectra were recorded either in a dry N2 atmosphere or by exposing the sample to ambient humidity. In the first case, the evolution of the IR absorption was monitored in a time of several hours, as is seen in Figure 2A. In the inset of Figure 2A, the left side of the spectrum has been purposely expanded to discern information on the evolution of the sol-gel reactions, while the right side of the spectrum is expanded and presented in Figure 2B. The 1780-1200 cm-1 spectral range contains carbonyl stretching modes11 of AcOH at 1714 and 1754 cm-1(C-O stretching) and 1412, 1360, and 1291 cm-1 (C-O stretching and OH deformation). In the course of time, the intensity of all the peaks extensively decreased. This is due to the evaporation of AcOH from the sol due to its exposure to the gaseous environment. However, the decrease of the intensity of these modes revealed the appearance of new modes, one scarcely discernible at 1734 cm-1 and two much stronger bands at 1373 and 1167 cm-1, assigned to EtOAc ester,12 and one new band, appearing at 1724 cm-1, assigned to Si-OAc.2,13 The formation of silicon ester is (11) Arkles, B.; Silicon-esters-alkoxy and acyloxysilanes; Gelest Inc.: Tully Fown, PA, 1995 (product catalog). (12) Kline, A. A.; Mullins, M. E.; Cornilsen, B. C. J. Am. Ceram. Soc. 1991, 74, 2559. (13) Smith, A. L.; Anderson, D. R. Appl. Spectrosc. 1984, 38, 822.

Ureasil Sol-Gel Ionic Conductors

Langmuir, Vol. 19, No. 18, 2003 7589

Figure 3. Temporal ATR IR spectra of PPG230 catalyzed with AcOH after exposure to ambient humidity.

simultaneous with the decrease of the alkoxide Si-OEt absorption bands at 1166 and 956 cm-1 (Figure 2A,B). Furthermore, inspection of Figure 2B revealed that EtOH (bands at 1055 and 880 cm-1) formed immediately after mixing PPG230 with AcOH. In the course of measurements, EtOH evaporated and the intensities of the corresponding bands diminished. The formation of SiOAc and EtOH justifies the validity of reaction 2a, while the formation of EtOAc supports validity of any of the reactions 2b-d. However, no Si-OH groups have been detected in the spectra of Figure 2B. Si-OH exhibits absorption bands14 in the same spectral range (950-925 cm-1) as PPG230 (956-933 cm-1) or simple alkyloxysilane groups, as it has been found by separately studying the IR spectra of pure precursors. This complicates the study of Si-OH absorption. Nevertheless, if Si-OH were formed, the decrease of the absorption of the Si-OEt band at 956 cm-1 should be accompanied by a simultaneous increase of the absorption in the range 930-924 cm-1. Such an increase was not detected; therefore, no support is provided for the validity of reaction 2c. The validity of reaction 2d is derived from a small but distinct blue frequency shift and an intensity increase of the 1015 cm-1 band attributed to the Si-O-Si bonds (Figure 2B). Such limited amounts of M-O-M bonds lead to the formation of oligomers,13,15 which do not give gelling. Indeed, samples kept in a dry atmosphere never gelled. This is also true for other precursors studied in the past.3 On the contrary, when the sol was exposed to ambient humidity, it gelled and the IR spectrum dramatically changed. Figure 3 shows the evolution of the spectrum after exposure to ambient humidity of the sample previously kept in a dry N2

atmosphere. As is seen in Figure 3, AcOH is at the end completely consumed, save for a small contribution from AcOH dimers absorbing at 1714 cm-1. Meanwhile, the ethanol band at 1052 cm-1 increased while the Si-OEt modes at 1166 and 956 cm-1 and the Si-OAc at 1724 cm-1 completely disappeared at the end. The obtained gels contained an appreciable amount of Si-OH, as was detected by an increase of the band at 915 (and 3500, not shown) cm-1. These results reveal that hydrolysis by means of scheme 1, prevails by exposure to ambient humidity. We then conclude that pure acetic acid solvolysis of ureasil precursors is an extremely slow process not leading to gelation. Gels can be formed by exposure to humidity. This fact is, unfortunately, not sufficiently stressed in the literature.16 We believe that organic acid regulates the gelling process, affecting the structure and the quality of the final gels by two major means: (1) by slowing down chemical reactions leading to gelling and, thus, allowing the evaporation of volatile liquid components, such as ethanol, and (2) by forming intermediate species such as Si-OAc. Very recent data indicate that carboxylic acids with lower pK values17 than AcOH (pK ) 4.75) seem to lead to more extensive polymerization and gelling. This question is further studied in our laboratories. The data of the present paragraph show that it is hard to make usable gels by only mixing acetic acid and ureasil precursors, especially if they are protected from humidity. For this reason, the precursor solutions actually used to study ionic mobility were different. Ethanol was employed as a solvent while no measures have been taken to prevent or even control humidity, including small quantities of water always attached to the reagents used. In addition, mixtures of ethanol and acetic acid lead to esterification and simultaneous water release, according to the reaction EtOH + AcOH f Et-OAc + H2O. This additional water also participates in the slow gelling process affecting the structure of the end product. Even though such samples cannot be studied by IR spectroscopy, the catalytic presence of AcOH is obvious, while the importance of its molar fraction is assessed in the following paragraphs. Time-Resolved Luminescence Quenching Analysis. Luminescence quenching analysis has been carried out by using two luminophore/quencher couples probing domains of different polarities. Both the hydrophilic Ru(bpy)32+/MV2+ and the hydrophobic pyrene/C-15310,18 luminophore/quencher couples can be incorporated in the

(14) Fujiwara, M.; Wessel, H.; Park, H. S.; Roesky, H. W. Chem. Mater. 2002, 14, 4975. (15) Sharp, K. G. J. Sol.-Gel Sci. Technol. 1994, 2, 35.

(16) Maurie, M.; Novak, A. J. Chem. Phys. 1965, 62, 10. (17) Colthup, N. B.; Doly, L. H.; Wiberley, S. E. Introduction to Infrared and Raman Spectroscopy; Academic Press: New York, 1964.

Figure 2. Temporal ATR IR spectra of PPG230 catalyzed by AcOH kept in a dry N2 atmosphere in the spectral range 1800800 cm-1 (A). The inset shows an expansion of the left side of the spectrum and part B shows an expansion of the right side of the spectrum.

7590

Langmuir, Vol. 19, No. 18, 2003

Stathatos et al.

Figure 5. f values obtained with the pyrene/C-153 luminophore/quencher couple. The concentrations (measured in the original sol) were 1 mM for pyrene and 1.1 mM for C-153.

Figure 4. (a) Experimental decay profile of Ru(bpy)32+ in the presence of MV2+ together with a fitted curve, excitation pulse, and residual distribution. (b) f values obtained with the Ru(bpy)32+/MV2+ luminophore/quencher couple. The concentrations (measured in the original sol) were 1 mM for Ru(bpy)32+ and 20 mM for MV2+.

nanocomposite material, but they are expected to probe domains of substantially different polarities. For this reason, analyses were carried out with both couples. The luminescence decay profiles of a luminophore in the presence of a quencher in a restricted medium can be expressed as a sum of the stretched exponentials according to the following equation:9,10,18,19

I(t) ) I0 exp(-t/τo) exp[-C1(t/τo)f + C2(t/τo)2f] (3) 0