Langmuir 2007, 23, 12337-12343
12337
Structural, Photophysical, and Photochemical Characterization of 9-Anthracenecarboxylate-Hydrotalcite Nanocomposites: Evidence of a Reversible Light-Driven Reaction Loredana Latterini,*,† Morena Nocchetti,† Gian Gaetano Aloisi,† Umberto Costantino,† Frans C. De Schryver,‡ and Fausto Elisei† CEMIN-Centro Eccellenza Materiali InnoVatiVi Nanostrutturati, Dipartimento di Chimica, UniVersita` di Perugia, Via Elce di Sotto, 8, 06123 Perugia, Italy, and Laboratory of Photochemistry and Spectroscopy, Katholieke UniVersiteit LeuVen, Celestijnenlaan 200F, B-3001 HeVerlee-LeuVen, Belgium
Downloaded via UNIV OF WINNIPEG on January 23, 2019 at 14:09:52 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
ReceiVed May 22, 2007. In Final Form: August 13, 2007
ZnAl hydrotalcites containing increasing amounts of 9-anthracenecarboxylate anion (9AC) have been obtained via an anion-exchange procedure. In particular, intercalated and/or surface-exchanged samples were prepared to study the effect of the chromophore packing on their photophysical and photochemical behavior. Surface-exchanged samples were obtained by equilibrating the carbonate form of the ZnAl hydrotalcite with dilute solutions of 9AC. The nitrate form of the ZnAl hydrotalcite was instead chosen for the preparation of intercalation compounds. The maximum loading of 9AC was found to be 44% of the anion-exchange capacity. The obtained nanostructured materials were characterized by chemical and thermal analysis and X-ray powder diffractometry and studied for their photophysical and photochemical properties. The absorption and emission spectra of the materials revealed the formation of 9AC aggregates. The time-resolved fluorescence properties of the hybrid materials were investigated in bulk and under space-resolved conditions. The fluorescence decays appeared to be quite complex and were affected by the microenvironment and the experimental conditions. Generally, a shortening of the main fluorescence decay component was observed with increasing matrix loading, thus suggesting the occurrence of nonradiative processes in competition with fluorescence at high chromophore concentrations. Indeed, the occurrence of an electron-transfer process to water molecules, which led to the formation of 9AC radical, was observed spectrophotometrically in the sample with high 9AC loading. The electron-transfer process was completely reversible under air-equilibrated conditions.
1. Introduction Layered solids are materials with interesting physical properties related to their structural anisotropy, and many of them can be functionalized by intercalation of species having specific properties.1 For example, intercalation of guest species containing chromophore groups into different layered host compounds gives rise to nanostructurated materials with photochemical and photophysical properties such that they have potential applications in various fields.2 It must be pointed out that these properties generally differ from those of the pure guest species, being affected by the organization of the guest molecules in the interlayer region as well as by the host-guest interactions. Particularly interesting in this context are layered solids belonging to the family of hydrotalcite-like compounds (HTlc’s), also called layered double hydroxides, for their structural and compositional features and simple and inexpensive preparation methods and because they have positively charged layers and allow the intercalation of * Corresponding author. Tel.: ++39-75-5855636. Fax: ++39-755855598. E-mail:
[email protected]. † Universita ` di Perugia. ‡ Katholieke Universiteit Leuven. (1) (a) Mu¨ller-Warmuth, W., Scho¨llhorn, R., Eds. Progress in Intercalation Research; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1994. (b) Ferey, G. J. Solid State Chem. 2000, 152, 37. (c) Schulz-Ekloff, G.; Wo¨hrleb, D.; van Duffelc, B.; Schoonheydt, R. A. Microporous Mesoporous Mater. 2002, 51, 91. (d) Alberti G.; Costantino, U. In ComprehensiVe Supramolecular Chemistry; Solid-State Supramolecular Chemistry: Two and Three-dimensional Inorganic Networks; Alberti, G., Bein, T., Eds.; Pergamon Press: Oxford, U.K., 1996; Vol. 7, Chapter 1. (2) (a) Ogawa, M.; Kuroda, K.; Chem. ReV. 1995, 95, 399. (b) Costantino, U.; Nocchetti, M. In Layered Double Hydroxides: Present and Future; Nova Science Publishers, Inc.: New York, 2001; p 383. (c) Latterini, L.; Nocchetti, M.; Aloisi, G. G.; Costantino, U.; Elisei, F. Inorg. Chim. Acta 2007, 360, 728.
exchangeable anions.3 Anionic chromophores can indeed be intercalated within HTlc’s, and their arrangement can lead to a modulation of the chromophore-chromophore interactions.4 In this context, the control of the relative position of the chromophores can affect their photophysical and photochemical properties. Anthracene derivatives, similarly to many other aromatic compounds, present photophysical and photochemical behaviors that are strongly dependent on the microenvironment5 and on the experimental conditions, which can lead to excimer or dimer photoinduced formation,6 photooxidation,7 or thermal8 or photoinduced9 electron transfer. (3) (a) Trifiro`, F.; Vaccari, A. In ComprehensiVe Supramolecular Chemistry; Solid-State Supramolecular Chemistry: Two and Three-dimensional Inorganic Networks; Alberti, G., Bein, T., Eds.; Pergamon Press: Oxford, U.K., 1996; Vol. 7, p 251. (b) Jones, W.; Newman, S. P. New J. Chem. 1998, 105. (c) Rives, V.; Ulibarri, M. A. Coord Chem. ReV. 1999, 181, 61. (d) Khan, A. I.; O’Hare, D. J. Mater. Chem. 2002, 12, 1. (e) Leroux, F.; Taviot-Gueho, C. J. Mater. Chem. 2005, 15, 3628. (4) (a) Aloisi, G. G.; Coletti, N.; Costantino, U.; Elisei, F.; Nocchetti, M. Langmuir 1999, 15, 4454. (b) Aloisi, G. G.; Costantino, U.; Elisei, F.; Latterini, L.; Nocchetti, M. J. Mater. Chem. 2002, 12, 3316. (c) Ge´raud, E.; Bouhent, M.; Derriche, Z.; Leroux, F.; Pre´vot, V.; Forano, C. J. Phys. Chem. Solids 2007, 68, 818. (5) (a) Mizobe, Y.; Miyata, M.; Hidaki, I.; Hasegawa, Y.; Tohnai, N. Org. Lett. 2006, 8, 4295. (b) Kaanumalle, L. S.; Gibb, C. L. D.; Gibb, B. C.; Ramamurthy, V. J. Am. Chem. Soc. 2005, 127, 3674. (6) (a) Hashimoto, S.; Ikuta, S.; Asashi, T.; Masuhara, H. Langmuir 1998, 14, 4284. (b) Kwon, O-H.; Yu, H.; Jang, D-J. J. Phys. Chem. B 2004, 108, 3970. (c) Bouas-Laurent, H.; Castellan, A.; Desvergn, J.-P.; Lapouyade, R. Chem. Soc. ReV. 2000, 29, 43. (7) Fudickar, W.; Fery, A.; Linker, T. J. Am. Chem. Soc. 2005, 127, 9386. (8) Marquis, S.; Moisette, A.; Vezin, H.; Bre´mard, C. J. Phys. Chem. B 2005, 109, 3723 and references therein. (9) (a) Kiwi, J.; Dhananjeyan, M. R.; Nadtochenko, V. J. Phys. Chem. A 2002, 106, 7138. (b) Worral, D. R.; Williams, S. L.; Wilkinson, F. J. Phys. Chem. B 1997, 101, 4709.
10.1021/la7014989 CCC: $37.00 © 2007 American Chemical Society Published on Web 10/26/2007
12338 Langmuir, Vol. 23, No. 24, 2007
9-Anthracenecarboxylic acid (9ACA) has been widely studied in homogeneous and heterogeneous media to explore its complex photophysical and photochemical behavior. An interesting discussion has arisen concerning the assignment of the fluorescence spectra of 9ACA observed upon changing the experimental conditions (pH, ground-state concentration, medium proticity); in particular, in addition to a spectrum with a sharp structure due to the anthracene moiety, a broad structureless band appears at lower energy, and time-resolved measurements have shown that this emission band decays with a rate constant different from that of the structured band.10 The fluorescence decay is even more complex when the compound is dispersed in heterogeneous media such as surfactant solutions or modified clays.11 This dual fluorescence has been assigned either to acidbase equilibria11 or to dimer/excimer formation.12 Furthermore, 9ACA is known to undergo efficient photodimerization reactions both in solution13 and in the solid state,14 but it also undergoes electron-transfer reactions when absorbed on TiO2 particles.15 In the present contribution, samples of 9AC intercalated in HTlc’s and/or surface-exchanged have been prepared, and the effect of a constrained environment on the photochemical and photophysical behavior of 9AC has been explored. In particular, detailed absorption and emission measurements resolved in time and space allowed for a better understanding of the deactivation paths of the anthracene derivative when accommodated within the layered matrix. Furthermore, evidence for a reversible photoinduced ionization process was obtained. The spectral changes related to the photochemical reaction allow for the prediction of potential applications of such a hybrid material (9AC-HTlc) in which the charge separation can be photochemically controlled. 2. Experimental Section 2.1. Chemicals. 9-Anthracenecarboxylic acid was supplied by Fluka and used without any further purification. All other chemicals were C. Erba RP-ACS products. 2.2. Preparation of Hydrotalcite-like Compounds. Hydrotalcite ZnAl having the formula [Zn0.63Al0.37(OH)2](CO3)0.185‚0.5H2O (ZnAl-CO3) and an ion-exchange capacity (IEC) of 3.51 mequiv/g was prepared by the urea method.16 The carbonate form was converted into the chloride form (ZnAl-Cl) by titration of the sample dispersed in a 1 mol/dm3 NaCl solution (1 g/100 cm3) with a 0.1 mol/dm3 HCl aqueous solution by means of a radiometer automatic titrator operating in pH stat mode and at pH equal to 5. The solid obtained was washed with deionized, CO2-free water and dried in a desiccator containing P4O10. The latter form was exchanged with nitrate anions by equilibrating 1 g of the ZnAl-Cl form with 50 mL of 1 mol/dm3 NaNO3 CO2-free aqueous solution, under nitrogen atmosphere at room temperature, for 1 day. Finally, the sample was filtered, washed with deionized, CO2-free water and dried in a desiccator containing P4O10. The formula of the nitrate form was [Zn0.63Al0.37(OH)2](NO3)0.37‚0.6H2O (subsequently denoted ZnAl-NO3), and it had an IEC of 3.12 mequiv/g. 2.3. Preparation of Anthracene Hydrotalcite Derivatives. Weighed amounts of ZnAl-NO3 were equilibrated with volumes of 0.25 mol/dm3 sodium 9-anthracenecarboxylate in a water/acetone mixture (1/1 v/v) to have molar ratios of 9AC in solution to NO3in the solid of 0.5 (sample I) and 2 (sample II). The solution was (10) Werner, T. C.; Hercules, D. M. J. Phys. Chem. 1969, 75, 2005. (11) Momiji, I.; Yoza, C.; Matsui, K. J. Phys. Chem. B 2000, 104, 1552. (12) Ghoneim, N.; Scherrer, D.; Suppan, P. J. Lumin. 1993, 9, 17133. (13) Cowan, D. O.; Schmiegel, W. W. J. Am. Chem. Soc. 1972, 94, 6779. (14) Ito, Y.; Fujita, H. J. Org. Chem. 1996, 61, 5677. (15) Martini, I.; Hodak, J.; Hartland, G. V.; Kamat, P. V. J. Chem. Phys. 1997, 19, 8064. (16) Costantino, U.; Marmottini, F.; Nocchetti, M.; Vivani, R. Eur. J. Inorg. Chem. 1998, 1439.
Latterini et al. kept at room temperature for 72 h. To obtain a sample with 9AC exchanged only on the surface (sample SE), the previous conditions were used, but the starting hydrotalcite was the ZnAl-CO3 form, and the 9AC/CO32- molar ratio was 0.1. The recovered solids, obtained after being washed with a water/acetone mixture, were dried at room temperature over saturated NaCl solution (75% relative humidity). 2.4. Analysis. The metal content of the HTlc’s was obtained by EDTA titrations after about 100 mg of solid had been dissolved in a few drops of concentrated HCl and diluted with water to 50 mL. Anions such as chlorides and nitrates were measured by means of ion chromatography after 0.1 g of the samples had been equilibrated with 1 mol/dm3 Na2CO3 solution for 12 h. The 9AC content of the solids was determined by UV-vis spectroscopy at 360 nm after a weighed amount of the samples (∼50 mg) had been dissolved in a few drops of concentrated HCl and properly diluted with 5 × 10-2 mol/dm3 acetate buffer. The amounts of water and carbonate were obtained by thermogravimetry. The weight loss curves as a function of temperature also gave information on the carbonate, nitrate, and 9AC contents. 2.5. Instrumentation. X-ray powder diffraction (XRPD) patterns were recorded on a computer-controlled Philips 1710 diffractometer (40 kV, 20 mA), using a graphite monochromator. Thermoanalytical characterizations were performed with a TG-DTA STA 449 C Jupiter thermoanalyzer operating at a 5 °C min-1 heating rate under a 30 mL/min flow of air. The morphology of the intercalation compounds was analyzed by scanning electron microscopy (SEM) on an electron microscope (Philips XL30) equipped with an electron gun of LaB6. Absorption spectra of the solid samples were recorded with a homemade spectrophotometer that uses a deuterium-halogen lamp (DH-2000-FHS) as a light source and a CCD as a detector (2001100 nm range, 2048 pixels, 86 photons/count) and is equipped with an integration sphere for recording reflectance spectra. A bar of barium sulfate was used as the reference to calibrate the spectrophotometer. The recorded spectra were analyzed with the KubelkaMunk equation to enable comparisons among different samples. Fluorescence spectra of all samples were recorded by a fluorimeter (Spex Fluorolog2) in front face configuration between the excitation and the emission light. The spectra were corrected for the response of instrument components at each wavelength. Time-resolved fluorescence decay profiles were measured by the time-correlated single-photon-counting (TCSPC) technique using the 488 nm output (8.18 MHz, 1.2-ps fwhm) of a mode-locked regeneratively amplified Ti:sapphire laser (Spectra Physics).17 For space-resolved fluorescence decay measurements, the same laser source directed into an inverted microscope (Olympus IX 70) and focused onto a sample through a 1.4-NA oil-immersion objective (Olympus) was used. Fluorescence was collected through the same objective and detected with an avalanche photodiode. To avoid sample degradation, all measurements were carried out at low laser intensity, which was spectrophotometrically controlled. The fluorescence images and space-resolved spectra were recorded with a laser scanning confocal microscope (Nikon, PCM2000), already described,18 using an Ar-ion laser (λexc ) 488 nm) as the light source. The images were obtained with a 60×, 1.4-NA oilimmersion objective (512 × 512 pixels) with the pinhole set at 10 µm. The fluorescence spectra were recorded by coupling the microscope to the previously described fluorimeter. In particular, the light emitted from the sample was collected by the objective and sent to the acquisition compartment (monochromator and photomultiplier tube) of the fluorimeter. Nanosecond laser flash photolysis experiments were performed using the third harmonic of a Nd:Yag laser (λ ) 355 nm, pulse width ≈ 7 ns, energy ≈ 150 mJ/pulse) to pump an optical parametric oscillator (OPO) and to generate the visible wavelengths. The OPO (17) Maus, M.; Rousseau, E.; Cotlet, M.; Schweitzer, G.; Hofkens, J.; van der Auweraer, M.; De Schryver, F. C.; Krueger, A. ReV. Sci. Instrum. 2001, 72, 36. (18) Latterini, L.; Elisei, F.; Aloisi, G. G.; Costantino, U.; Nocchetti, M. Phys. Chem. Chem. Phys. 2002, 4, 2792.
Characterization of 9AC-HTlc Nanocomposites
Langmuir, Vol. 23, No. 24, 2007 12339
Table 1. 9AC Added and Taken Up and Solid Composition sample
9AC added (mmol/IEC)a
9AC taken up (mmol/IEC)a
solid composition
Ib IIb SEc
0.5 2 0.05
0.23 0.43 0.01
[Zn0.63Al0.37(OH)2](9AC)0.085(NO3)0.22(CO3)0.033‚1.4H2O [Zn0.63Al0.37(OH)2](9AC)0.16(NO3)0.21‚0.79H2O [Zn0.63Al0.37(OH)2](9AC)0.004(CO3)0.366‚0.5H2O
a IEC for ZnAl-NO3 ) 3.12 mequiv/g; IEC for ZnAl-CO3 ) 3.51 mequiv/g. b Sample obtained from ZnAl-NO3. c Sample obtained from ZnAl-CO3.
was adjusted to excite the sample at 490 nm (pulse width ≈ 4 ns, energy ≈ 1.5 mJ/pulse). Diffuse-reflectance laser flash photolysis spectra and decay traces were measured using a homemade accessory for reflectance measurements.
3. Results and Discussion 3.1. Preparation and Chemicalphysical Characterization of the Nanocomposites. Intercalation compounds containing 9-anthracenecarboxylate were prepared via an anion-exchange reaction starting from ZnAl-NO3. The 9-anthracenecarboxylic acid was deprotonated by adding sodium hydroxide to the solution to reach a pH of about 9.5. All operations were performed under nitrogen to avoid CO2 contamination. Samples having different contents of 9AC in the interlayer region were obtained by equilibrating the ZnAl-NO3 with appropriate volumes of 9AC solution give molar ratios between the amount of 9AC in solution and the IEC of 0.5 and 2. In each intercalation reaction, the total content of the guest is given by the sum of the species present in the interlayer region and those exchanged on the surface. To study the contribution of the 9AC exchanged on the surface of the microcrystals, a sample containing 9AC only on the surface was prepared. For this purpose, the carbonate form of the hydrotalcite was employed. Indeed, it is well-known that the carbonate anions can be removed from the interlayer region only by treatment with acidic solution;19 therefore, the amount of 9AC present on the obtained solid is ascribed to surface carbonate/ 9AC exchange. The composition of the materials obtained is reported in Table 1. It can be noted that the maximum amount of 9AC intercalated did not exceed 44% of the IEC, even when the hydrotalcite was equilibrated with solutions containing a large excess of 9AC. Charge balance requires the presence of other anions in the interlayer region. The ion chromatography of samples previously dissolved in hydrochloric acid confirmed the presence of nonexchanged nitrate anions. Only for sample I were small amounts of carbonate anions also found to be present. The X-ray powder diffraction (XRPD) patterns of some samples in comparison to that of the starting ZnAl-NO3 material are reported in Figure 1. The XRPD pattern of sample II (Figure 1c) shows an X-ray reflection at 20.4 Å assigned to the 9AC phase, with a broad second reflection, probably as a result of the convolution of the second reflection of the 9AC phase and the first reflection of the nitrate phase (8.9 Å). The XRPD patterns of sample I (Figure 1b), in addition to the phases discussed above, presents a diffraction line at 7.6 Å characteristic of the carbonate phase, as confirmed by chemical analysis. As expected, sample SE exhibits the typical XRPD pattern (not shown) of the carbonate form. Figure 2 shows the weight loss (TG) and the differential thermal analysis (DTA) of sample II as a function of temperature. The first step, from room temperature to 240 °C, corresponds to the loss of 0.79 mol/mol of hydration water. The second step has been attributed to the thermal decomposition of nitrate anions and exchanged 9AC and overlaps with the dehydroxylation of (19) Miyata, S. Clays Clay Miner. 1983, 31, 305. (b) Hines, D. R.; Solin, S. A.; Costantino, U.; Nocchetti, M. Phys. ReV. B 2000, 61, 11348.
Figure 1. XRPD patterns of (a) ZnAl-NO3 starting material, (b) sample I, and (c) sample II intercalation compounds dried at room temperature (75% relative humidity).
Figure 2. TG and DTA curves of sample II intercalation compound. Heating rate, 5 °C/min; air flow, 30 mL/min.
the lamellae. The formulas reported in Table 1 were obtained by taking into account the fact that the solids at 1000 °C consist of a ZnO and ZnAl2O4 mixture in a stoichiometric ratio. The hydrotalcite microcrystals were also examined by scanning electron microscopy (SEM). Figure 3a shows a micrograph of sample SE, in which packets of lamellae with a hexagonal form having a diameter of 5-8 µm and a thickness of about 0.8 µm are evident. Figure 3b,c presents micrographs of samples I and II showing that, notwithstanding the intercalation process, the morphology is unchanged but the thickness of the microcrystals is increased to about 1.3 µm for the former and to 1.4 µm for the latter. This increase is qualitatively in agreement with the increase in the interlayer distance and hence the swelling of the microcrystals detected by XRPD upon 9AC intercalation; the resolution of the SEM micrographs prevents a more accurate comparison. As already observed, sample II consists of two co-existing phases, one containing essentially 9AC anions (interlayer distance of 20.4 Å) and the other containing the original nitrate anions (interlayer distance of 8.9 Å). It was of interest to obtain a schematic picture of the probable disposition of the 9AC in the interlayer region, which was done with the Hyperchem program
12340 Langmuir, Vol. 23, No. 24, 2007
Latterini et al.
(Figure 4). It was assumed that the layers of the host did not change in structure as a consequence of intercalation and that all of the nitrate anions were exchanged. The experimental interlayer distance was found to be in agreement with a disposition of the anions as a bilayer film of oriented species interacting by π-π overlap. 3.2. Absorption and Emission Properties of the Nanocomposites. Absorption and emission spectra of 9ACA recorded under different experimental conditions are shown in Figure 5. The spectra obtained from 9ACA in solution (ethanol and water as solvents, pH 12) presented a structured absorption with five bands in the 300-400 nm region typical of the anthracene chromophore.20 On the other hand, the solid samples presented a broad and structureless absorption spectrum; in particular, the
spectrum of sample II was slightly blue-shifted in the main band compared to that of pure 9ACA and showed a second band at 480 nm, probably due to 9AC aggregates formed in the interlayer region. The formation of anthracene derivative aggregates in confined environments has already been observed and described.4b,5,6 The effect of 9AC packing and/or loading on the spectral features was further investigated in the present work; Figure 6 shows the absorption spectra of the samples at different loadings. The spectrum obtained from the surface-exchanged sample has a structure similar to the solution spectra, indicating that, under these conditions, the chromophore-chromophore interactions are weak although not negligible. In the cases of intercalated samples, the spectra lost the characteristic structure even at low loading, and an absorption tail at wavelengths above 400 nm appeared and became more evident with increasing concentration of 9AC in the interlayer region. This spectral behavior is in agreement with the formation of anthracene aggregates in a constrained environment whose concentration is dependent on the 9AC loading in the inorganic matrix. Fluorescence spectra of 9AC samples are shown in Figure 7. Similarly to the absorption spectra of the compound in solution, the fluorescence spectra presented a structured shape (Figure 7a) as has been reported in the literature;21 the emission of the solid samples showed spectra with shapes and positions strongly related to the nature of the sample and the excitation wavelength. Upon excitation in the main absorption band (λexc ) 366 nm), the sample obtained by surface exchange of 9AC on the inorganic matrix presented an emission spectrum with maxima at 415 and 435 nm and a shoulder at 465 nm, which resembles the spectrum of 9ACA in solution (λmax ) 387, 410, 435, and 464 nm), thus confirming the presence of free chromophores not interacting each other. Instead, the emission spectra of the intercalated samples were structureless and presented a maximum at 500 nm, similarly to what was observed for the pure 9ACA solid sample. When the ZnAl-9AC samples were excited on the absorption shoulder (λexc ) 460 nm), the emission spectra in all cases appeared markedly red-shifted (λmax ) 600 nm) and structureless. This contribution can be attributed to emission from the aggregate species that are likely formed even in the sample where the chromophore was externally exchanged on the matrix. Similar spectral behavior was previously observed when anthracene or other aromatic molecules were confined in space.5b,11 The timedependent fluorescence intensity of 9AC-HTlc complexes afforded complex decays that changed with the excitation and emission wavelengths. In particular, upon excitation at 366 nm and collection of the emission below 500 nm, the decay was essentially fitted by a monoexponential function (98%) independently of the position of the chromophore in the matrix and the loading with a decay time of 0.11 ns. This value is in good agreement with the τF value of 0.1 ns reported in the literature for 9ACA inserted in zeolites6b at least for high loadings of the chromophore, confirming the occurrence of specific interactions leading to the formation of complex structures (aggregates for which it is not possible to establish the number of interacting molecules and their relative orientations) that shorten the τF values obtained from homogeneous solutions (1.3-4.5 ns depending on the medium).13 On the other hand, when the hybrid materials investigated were excited at 460 nm, the decays appeared to be more complex and could be satisfactorily reproduced only by use of triexponential functions with fitting parameters, decay times (on the order of 2 × 102, 1 × 103, and 3-4 × 103 ps), and relative amplitudes affected by the emission wavelength and the
(20) Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M. T. In Handbook of Photochemistry, 3rd ed.; CRC Press: Boca Raton, FL, 2006.
(21) Chandross, E. A.; Fergusson, J.; McRae, E. G. J. Chem. Phys. 1966, 45, 3546.
Figure 3. SEM images of 9AC (a) exchanged on the surface of HTlc, sample SE and intercalated in HTlc, samples (b) I and (c) II.
Characterization of 9AC-HTlc Nanocomposites
Langmuir, Vol. 23, No. 24, 2007 12341
Figure 4. Two different views of computer-generated models showing the most probable disposition of 9AC anions in the interlayer region of ZnAl.
Figure 5. Absorption spectra of (a) sample II and of 9ACA (b) as the pure solid, (c) in water at pH 12 (1 × 10-5 M), and (d) in ethanol (1 × 10-5 M).
Figure 7. Emission spectra of 9ACA (1a) in ethanol (2 × 10-6 M), (1b) in water at pH 12 (2 × 10-6 M), and (1c) as a pure solid; of 9AC (2a) exchanged on the surface of HTlc sample SE and intercalated in HTlc samples (2b) I and (2c) II (λexc ) 366 nm); and (3a) exchanged on the surface of HTlc sample SE and intercalated in HTlc samples (3b) I and (3c) II (λexc ) 460 nm).
Figure 6. Absorption spectra of 9AC (a) exchanged on the surface of HTlc sample SE and intercalated in HTlc samples (b) I and (c) II.
loading. This behavior suggests that the 9AC distribution in the inorganic matrix (in terms of chemical structure and/or microenvironment) is not homogeneously, similarly to what was previously observed for anthracene in zeolite nanocavities.6b This behavior prevents a more precise description of the 9AC aggregates. 3.3. Confocal Fluorescence Microscopy. Confocal fluorescence imaging of the ZnAl-9AC samples was carried out to check the fluorescence distribution on the matrix. In particular, the aggregate species were mainly excited using 488 nm as the excitation source, and the fluorescence images recorded their distribution. In the sample having only surface-exchanged 9AC, the detected emission presented a very low intensity localized in micrometer-sized spots of the matrix grains. This observation indicates that, in sample SE the aggregates are few and localized in a specific region of the inorganic clay. In the case of intercalated samples, the images appear to be more intense and the bright spots more frequent, as shown, for example, in Figure 8b. In this sample, aggregate formation appears to be much more favorable upon intercalation of the chromophores and increasing of the
anthracene concentration, as already observed by the spectrophotometric analysis. However, the fluorescence spectra of the sample recorded through the microscope (Figure 8c) indicate that the arrangements and/or locations of the fluorophore in the matrix are not unique. This conclusion is further confirmed by space-resolved fluorescence decay time measurements carried out on sample II through a confocal microscope upon excitation at 488 nm. In general, the fluorescence revealed complex decays that could be satisfactorily fitted by triexponential functions with decay times in agreement with those obtained in bulk, thus confirming the presence of a distribution of decay times even at a small scale that reflects fluorophore arrangements. In particular, an effect of the heterogeneity in interaction and environment on the fitting parameters was observed. The relative contribution of the shortest component (ca. 2 × 102 ps) was higher when the decay traces were collected in areas with higher intensity. Furthermore, a comparison of the fluorescence traces recorded in the areas with the highest intensities for the different samples (SE, I, and II) suggests that the decays are affected by the loading, as shown by the fitting parameters reported in Table 2. The data show a significant reduction of the short-component decay time with increasing 9AC loading in the matrix; this behavior can be explained through the occurrence of nonradiative processes in competition with fluorescence, which appears more efficient when 9AC is more concentrated.
12342 Langmuir, Vol. 23, No. 24, 2007
Latterini et al.
Figure 9. Absorption spectra of 9AC intercalated in HTlc sample II recorded after 0, 15, 60, and 320 min of irradiation at (a) 460 and (b) 350 nm. Inset: Absorbance at 685 nm as a function of time upon irradiation (circles) and storage in the dark (squares).
Figure 8. (a,b) Fluorescence confocal images of 9AC intercalated in HTlc sample II (λexc ) 488 nm). (c) Fluorescence spectra recorded in the regions marked with a blue circle (solid line) and yellow square (dashed line) in the image of Figure 8b. Table 2. Space-Resolved Fluorescence Decay Times of 9AC-HTlc Materials sample
τF1 (ns)
τF2 (ns)
τF3 (ns)
SE I II
0.20 ( 0.07 0.16 ( 0.05 0.13 ( 0.03
1.9 ( 0.3 1.6 ( 0.3 1.3 ( 0.3
4.5 ( 1.0 6.2 ( 1.8 5.0 ( 1.0
3.4. Evidence of a Light-Driven Reaction. When sample II was excited at 460 or 490 nm, more interesting spectral changes were observed (see Figure 9a). In particular, a decrease of the absorbance in the 200-390 nm range was detected with increasing irradiation time, and a new band whose intensity increased with the irradiation time appeared at 685 nm. It has to be noted that such a modification was not observed when the sample was kept in the dark, thus indicating that the changes are due to a photoinduced process. Furthermore, the photoinduced process was thermally reversible up to the complete recovery of the original spectrum of the starting material. In fact, when the irradiated sample was stored in the dark, the absorbance at 685 nm decreased with a half-time of about 5 h, and after about 13 h, the signal completely vanished, as shown by the absorbance trend at 685 nm with time upon irradiation and storage in the dark (inset of Figure 9a). In contrast, the absorption in the UV region increased until the original values were reached. The reaction could be cycled several times by alternating irradiation and dark storage periods (inset Figure 9a) with a reproducible optical response. This behavior indicates the absence of degradation processes under the experimental conditions used. It has to be noted that, upon irradiation at wavelengths below 400 nm (see Figure 9b, λexc ) 350 nm, as an example), no significant spectral changes were observed. Moreover, irradiation of samples at lower 9AC loadings did not lead to such marked spectral changes. These observations suggest that the aggregate species are responsible for the photoreaction. To collect information on the species involved in the reaction, the effect of irradiation time on the fluorescence intensity was controlled under different experimental conditions. In fact, a comparison of the fluorescence spectra recorded at increasing irradiation times revealed a substantial decrease of the emission
Figure 10. Emission kinetics of 9AC intercalated in an HTlc recorded under irradiation of sample II when (a) desiccated, (b) equilibrated with THF, and (c) hydrated (λexc ) 460 nm, λem ) 610 nm).
intensity in the 500-700 nm range. These findings confirm the observation that the aggregate species (emitting at 600 nm) is responsible for the reaction and that the effect of irradiation times on the absorption spectra can be related to the contribution of the reactive species to the absorption in the UV region. A comparison of the absorption spectra obtained after irradiation with literature data6a,8,9 suggests that the photoproduct is the 9AC radical formed upon photoionization of the aggregate chromophores assisted most likely by the hydration water molecules present in the interlayer region, as shown by the formula of the compounds. To confirm this assignment, irradiation experiments were carried out on the samples under different experimental conditions: hydrated, dehydrated at 100 °C, and equilibrated with THF. The effect of sample conditions on the photoreaction was quantified by following the fluorescence intensity as a function of irradiation time (Figure 10) and by diffuse-reflectance laser flash photolysis (Figure 11). In particular, the fluorescence intensity of the sample containing the structural hydration water decreased by about 75% in 2 h of irradiation; when sample II was dehydrated at 100 °C, the photoprocess was much slower, as a 10% decrease of the fluorescence intensity was observed in 2 h of irradiation. This behavior is in agreement with the occurrence of a photoionization process assisted by the water crystallized in the interlamellar region; in particular, the excited 9AC aggregates lose an electron that is stabilized in the inorganic matrix by the crystallized water, leading to the formation of the 9AC radical. This conclusion is further confirmed by the results recorded when the dried sample was equilibrated with THF, a solvent of medium polarity with electron-donating properties. In this case, the irradiation induced a 20% fluorescence reduction in 2 h, which means that the photoreaction channel between 9AC and the inorganic matrix is less favorable. Diffusereflectance laser flash photolysis measurements were performed
Characterization of 9AC-HTlc Nanocomposites
Langmuir, Vol. 23, No. 24, 2007 12343
optimize the process to make the material useful for practical applications.
Conclusions
Figure 11. Diffuse-reflectance absorption spectra of sample II recorded after the laser pulse (λexc ) 490 nm).
upon excitation at 490 nm. When hydrated sample II was excited, a transient signal in the region of 500-800 nm (Figure 11) having a maximum at 680 nm was observed. The transient spectrum recorded is in good agreement with the spectrum observed under stationary irradiation in the same spectral region (500-800 nm) and with those reported in the literature for 9-anthracene carboxylic acid on silica gel22 and assigned to the radical cation. The transient signal did not decay in the time window observed (80 µs), and its formation was hidden by the instrumental time response, preventing further kinetic information from being obtained. When diffuse-reflectance laser flash experiments were carried out on dehydrated and/or THF-equilibrated sample II, no transient signal was detected in the 500-800 nm region (data not shown), probably because of the low efficiency of the electrontransfer process, in agreement with the fluorescence results. These results are particularly interesting because the photoreaction could be initiated and its efficiency controlled using a visible light source even though the back-reaction is thermally controlled, although further experiments would be necessary to (22) Worral, D. R.; Williams, S. L.; Wilkinson, F.; Crossley, J. E.; BouasLaurent, H.; Desvergne, J.-P. J. Phys. Chem. B 1999, 103, 9255.
Nanostructured materials containing 9AC only on the microcrystal surface or intercalated were prepared by performing anionexchange reactions and taking advantage of the selectivity of ZnAl hydrotalcite toward different anions. The arrangement of the chromophores at the surface or between the sheets of the lamellar solid affected their photophysical and photochemical behavior. The fluorescence properties of the materials were investigated, and it was found that the deactivation paths of the chromophore are determined by the microenvironment and the experimental conditions. In particular, the constraints imposed by the inorganic matrix led to the formation of 9AC aggregates that are characterized by quite complex decay mechanisms of their excited states, as indicated by single-photon timing measurements carried out in bulk and spatially resolved. Furthermore, when the loading of the chromophore in the matrix was increased, corresponding to an increase in the concentration of aggregate species, the occurrence of a photoinduced electrontransfer process from 9AC to the hydration water molecules was observed to occur with increasing efficiency. The accumulation of a photoproduct was observed only when the sample at higher chromophore loading was excited in the absorption band of the aggregates. The formation of the 9AC radical was spectrophotometrically observed under different experimental conditions and appeared to be completely reversible. To better characterize the electron-transfer process, in situ ESR and photocurrent measurements are currently under way. However, for the controlled production of charge-separated species, the constrained environment of hydrotalcites appears to be suitable and will be further tested with other aromatic compounds. Acknowledgment. The authors thank the Ministero per l’Universita` e la Ricerca Scientifica e Tecnologica (Rome, Italy) for financial support though Project FIRB (RBNE017MB5). LA7014989
