Communication via Electron and Energy Transfer between Zinc Oxide

Feb 23, 2009 - Steady-state absorption spectroscopy as well as steady-state and time-resolved emission studies confirmed the electronic communication ...
0 downloads 0 Views 1MB Size
Download PDF
J. Phys. Chem. C 2009, 113, 4669–4678

4669

Communication via Electron and Energy Transfer between Zinc Oxide Nanoparticles and Organic Adsorbates Renata Marczak,† Fabian Werner,‡ Jan-Frederik Gnichwitz,§ Andreas Hirsch,*,§ Dirk M. Guldi,*,‡ and Wolfgang Peukert*,† Institute of Particle Technology, Friedrich-Alexander-UniVersity Erlangen-Nuremberg, Cauerstrasse 4, 91058 Erlangen, Germany, Department of Chemistry and Pharmacy, Friedrich-Alexander-UniVersity Erlangen-Nuremberg, Egerlandstrasse 3, 91058 Erlangen, Germany, and Institute of Organic Chemistry, Friedrich-Alexander-UniVersity Erlangen-Nuremberg, Henkestrasse 42, 91054 Erlangen, Germany ReceiVed: December 5, 2008; ReVised Manuscript ReceiVed: January 13, 2009

Stable ZnO nanoparticles suitable for further surface functionalization were synthesized in the liquid phase from homogeneous ethanolic solutions of the precursors lithium hydroxide and zinc acetate. It was found that the growth of the particles was governed by temperature as well as the presence of the reaction byproduct lithium acetate during the aging process. In particular, the reaction could be almost completely arrested by removal of this byproduct. The “washing” consisted of repeated precipitation of the ZnO particles by addition of alkanes such as heptane, removal of the supernatant, and redispersion in ethanol. Furthermore, the surface of the colloidal ZnO nanoparticles was successfully modified by catechol-anchoring group containing dye molecules, i.e., 5-(N-(3,4-dihydroxyphenethyl)-2-phenoxyacetamide)-10,15,20-(p-tert-butyltriphenyl)porphyrinatozinc (DOPAZ) and 5-(3,4-dihydroxy-N-phenylbenzamide)-10,15,20-tris(4-tert-butylphenyl)porphyrinatozinc (CAMIZ), for the study of photochemical properties. Thermogravimetric analysis proved the stability of the catechol anchor groups. Steady-state absorption spectroscopy as well as steady-state and time-resolved emission studies confirmed the electronic communication between the ZnO nanoparticles in their excited state and both of the porphyrins. More than 96% emission quenching of ZnO can be achieved by addition of the porphyrins, proving that the visible emission of the ZnO is caused by surface states, since only the surface of the particles was altered by the grafting experiments. Moreover, with increasing porphyrin concentrations the lifetimes changed from 46.0 to 15.3 ns. The shortened lifetimes prompt a new deactivation pathway, namely, through the electronic coupling of the porphyrins to the ZnO nanoparticle. Assuming that the decrease in lifetime is entirely due to electron transfer to the porphyrins, a rate constant of 0.35 × 108 s-1 could be determined for this process. When testing the excited state of the porphyrin in comparative assays between ZnO and Al2O3, we conclude a similar electron transfer deactivation. Introduction Zinc oxide is an attractive material for a broad range of electronic, optical, and piezoelectric applications due to its direct band gap and excellent thermal, chemical, and structural properties.1 For example, ZnO has been suggested for use in application such as solar cells,2-5 light emitting diodes,6 transparent electrodes,7 sensors,8,9 and many other devices.10,11 The various applications of ZnO nanoparticles are due to sensitivity to surrounding environments and superior luminescence and photoelectric properties. Different synthesis routes have been developed, including solid-vapor phase thermal sublimation,12 spray pyrolysis,13,14 RF plasma synthesis,15 sonochemical or microwave-assisted synthesis,16,17 and hydrothermal processing.18,19 However, wetchemical synthesis of ZnO is an area of particular interest, since * Corresponding authors. Phone +49 9131 8529400; Fax +49 9131 85 29402; E-mail [email protected] (W.P.). Phone +49 9131 8527341; Fax +49 9131 85-28307; E-mail [email protected] (D.M.G.). Phone +49 9131 85-22537; Fax +49 9131 85-26864; E-mail [email protected] (A.H.). † Institute of Particle Technology, Friedrich-Alexander-University Erlangen-Nuremberg. ‡ Department of Chemistry and Pharmacy, Friedrich-Alexander-University Erlangen-Nuremberg. § Institute of Organic Chemistry, Friedrich-Alexander-University Erlangen-Nuremberg.

it provides a low-temperature, economical way to produce various ZnO nanostructures.20-24 Growth from solution typically starts with nucleation and is followed by growth of the nuclei until the metal cation concentration reaches the solubility of the oxide. Then particle aging proceeds in the mother liquors. Synthesis conditions such as temperature,25 overall concentration of the precursors,21,23 water concentration,26 ions present in solution,27,28 and solvents29 have been shown to influence particle nucleation and growth. Therefore, the investigation of ZnO nanoparticles and the controlled synthesis thereof is of great interest due to their wide range of potential applications. A very interesting example of an application are solar cells based on nanocrystalline metal oxides, i.e., TiO2 and ZnO, developed by Gra¨tzel and co-workers.2,4 This devicessdyesensitized solar cells (DSSCs)sare based on photoelectrochemical dye-sensitized mesoporous metal-oxide electrodes and have been the subject of intense interest for years. This is due to the fact that they combine potentially low cost with medium performance as an alternative to traditional photovoltaic devices. Interfacial electron transfer reactions are reported on TiO230-34 and ZnO5,35-37 in the fast and ultrafast time regimes and are investigated to understand the mechanism which ultimately will assist in improving the solar light conversion efficiency. Despite the considerable interest in these materials, many questions are

10.1021/jp810696h CCC: $40.75  2009 American Chemical Society Published on Web 02/23/2009

4670 J. Phys. Chem. C, Vol. 113, No. 11, 2009

Marczak et al.

SCHEME 1: (a) Synthesis of 5-(N-(3,4-Dihydroxyphenethyl)-2-phenoxyacetamide)-10,15,20-(p-tert-butyltriphenyl)porphyrinatozinc, DOPAZ, and (b) Synthesis of 5-(3,4-Dihydroxy-N-phenylbenzamide)-10,15,20-tris(4-tert-butylphenyl)porphyrinatozinc, CAMIZ

to be answered with regard to their fundamental properties, such as charge carrier dynamics, electronic transport, and energy level distribution.5,36,38 There are various ways to anchor organic and/or inorganic dyes onto metal oxides’ host surfaces: (I) covalent attachment by anchoring groups, (II) electrostatic interactions, via ion exchange, ion-pairing, or donor-acceptor interactions, (III) hydrophobic interactions leading to self-assembly of long alkyl chains, (IV) hydrogen bonding, (V) van der Waals, and (VI) physical entrapment inside the pores or cavities of hosts such as cyclodextrins, micelles, etc. However, the covalent bonding is the most stable form of attachment. The covalent attachment is realized by a variety of anchoring groups with different affinities to the several metal-oxide surfaces.39,40 Typically the best anchoring groups for metal oxides are phosphonic acids (P(O)(OH)2), followed by carboxylic acids (COOH) and their derivatives. The main disadvantages of these anchoring groups are, however, their poor solubility in common organic solvents and their ability to dissolve some of the metal oxides, especially ZnO. To overcome these limitations, silanes, ethers, acetylacetonate, and salicylates have been explored, but their instability against water, acids, and bases hampers their applications. In this paper, we report the introduction of catechol-anchoring groups (3,4-dihydroxybenzo compounds) for grafting porphyrins onto ZnO nanoparticles. There are only a few examples41-43 where catechol functionalities are used for grafting functional molecules onto metal-oxide surfaces although it is known that catechol forms very stable metal complexes and that the anchoring group is relatively stable and soluble. Porphyrins are known to facilitate a good control over the reactions taking place in a DSSC.44 Based on a simple colloidal method to synthesize ZnO nanoparticles, we present a suitable new approach for further derivatization of their surface. The particles were prepared from zinc acetate dihydrate in ethanolic solution under basic conditions. To learn more about the growth mechanism, we employed in situ UV-vis absorption spectros-

copy and dynamic light scattering (DLS) to quantify both particle size and energy band gap as they vary during the course of time in order to study the influence of aging conditions on the growth of ZnO nanocrystals. The synthesis of the catecholanchoring group containing molecules, i.e. 5-(N-(3,4-dihydroxyphenethyl)-2-phenoxyacetamide)-10,15,20-(p-tert-butyltriphenyl)porphyrinatozinc, DOPAZ, and 5-(3,4-dihydroxy-Nphenylbenzamide)-10,15,20-tris(4-tert-butylphenyl)porphyrinatozinc, CAMIZ, is reported as well. In particular, we used dopamine in the case of DOPAZ as the effective anchor coupled via a modified Steglich coupling reaction with a porphyrin core followed by the metalation of the porphyrin with zinc (Scheme 1a). The second molecule, i.e., CAMIZ, was synthesized with a 3,4-dihydroxybenzoic acid derivative to shorten the distance between porphyrin and anchoring group (Scheme 1b). Both molecules are very well soluble in the most common organic solvents. Because their synthesis includes an amide coupling, purification on silica gel is needed. This is, however, a crucial point of the synthesis, since they may anchor to silica, which leads to a certain loss in yield in product. Nevertheless, the yields are quite good despite this enduring problem. Moreover, the surface functionalization of the ZnO nanocrystals with organic molecules, namely, DOPAZ and CAMIZ (Scheme 2), enabled examining the interactions between ZnO nanoparticles and electroactive porphyrins by using steady-state absorption and emission spectroscopies as well as time-resolved emission spectroscopy. Experimental Section Synthesis of ZnO Nanoparticles. All chemicals were analytical grade reagents purchased from commercial sources and used without further purification. Colloidal ZnO nanoparticles were prepared by hydrolyzing zinc acetate dihydrate in basic ethanol solution. The overall preparation procedure was adapted from Spanhel, Anderson,22 and Meulenkamp20 with a

Synthesis, Characterization of ZnO Nanoparticles SCHEME 2: (a) 5-(N-(3,4-Dihydroxyphenethyl)-2-phenoxyacetamide)-10,15,20-(p-tert-butyltriphenyl)porphyrinatozinc, DOPAZ, and (b) 5-(3,4-Dihydroxy-Nphenylbenzamide)-10,15,20-tris(4-tertbutylphenyl)porphyrinatozinc, CAMIZ, Molecules onto the ZnO Nanoparticle Surface

few modifications. A 2.19 g (0.01 mol) sample of zinc acetate dihydrate was dissolved in 100 mL of boiling ethanol at atmospheric pressure and cooled down to the synthesis temperature. A white powder of anhydrous zinc acetate precipitated close to room temperature.20 A 0.34 g (0.014 mol) sample of lithium hydroxide was dissolved in 100 mL of boiling ethanol and cooled to the synthesis temperature. Then, the lithium hydroxide solution was added dropwise to the zinc acetate solution under vigorous stirring. The reaction mixture became transparent after addition of about 1/3 volume of the lithium hydroxide solution. The ZnO colloid was stored at e0 °C to prevent rapid particle growth. In order to remove the reaction byproduct, lithium acetate, the ZnO suspension was washed by repeated flocculation of ZnO affected by addition of n-heptane. The supernatant was separated from the ZnO white precipitate by centrifugation and decantation. For colloidal characterization,

J. Phys. Chem. C, Vol. 113, No. 11, 2009 4671 the ZnO flocculates were redispersed in ethanol. In order to obtain the powder, the ZnO flocculates were dried under nitrogen for about 5 min. Particle Characterization. Particle size distribution of ZnO nanoparticle suspensions were determined via dynamic light scattering (DLS) by using a Malvern Nano ZS Instrument with a 633 nm “red” laser. For each sample, 10 measurements were performed and the average value per sample was calculated. Optical properties of the nanoparticles were determined from UV-vis absorption spectra recorded using a Cary 100 Scan Spectrometer (Varian) with a 10 mm path length cuvette. Structural analysis of the ZnO nanoparticles was performed in a D8 Advance (Bruker AXS) X-ray diffractometer (XRD) using Cu KR radiation (0.154 nm). The measurement was in the range of 20° e 2θ e 70°. HRTEM images were obtained using a Philips CM 300 UltraTwin microscope with the particles deposited on a standard copper grid supported carbon film. FTIR spectra were recorded on a Varian Excalibur Spectrometer FTS 3100 with a resolution of 2 cm-1 using the Easy Diffuse reflectance accessory. The samples were packed into a small sample cup. A reproducible sample surface was achieved by smoothing with a razor blade. Synthesis of the Dye Molecules. 5-(N-(3,4-Dihydroxyphenethyl)-2-phenoxyacetamide)-10,15,20-(p-tert-butyltriphenyl)porphyrin. A solution of 500 mg of porphyrin 1 (Scheme 1)45 (0.55 mmol) and 200 mL formic acid was stirred for 4 h to obtain the deprotected acid. Progress of the reaction was followed via TLC. The solvent was removed on a rotary evaporator, transferred into toluene, and evaporated twice to remove any residual formic acid. The product was finally dried under reduced pressure. The dried product was dissolved in DMF at 0 °C, and after addition of 315 mg of EDC (1.65 mmol), 222 mg of HOBT (1.65 mmol), and 200 mg of DMAP (1.65 mmol), the solution was allowed to stir for 1 h at 0 °C. Subsequently 155 mg of 3-hydroxytyramine hydrochloride (dopamine) (0.825 mmol) was added to the solution, and the mixture was stirred at room temperature for 48 h. The solvent was removed on a rotary evaporator, and the residue was purified with column chromatography on silica gel with a mixture of dichloromethane/ methanol (19:1) as eluent. The product was obtained by crystallization in pentane. Yield: 410 mg (75%). 1H NMR (400 MHz, 25 °C, THF-d8): δ ) -2.66 (s, 2H), 1.60 (s, 27H), 2.76 (t, 3J ) 7.3 Hz, 2H), 3.53 (dd, 3J ) 6.5 Hz, 3J ) 14.1 Hz, 2H), 4.70 (s, 2H), 6.55 (dd, 4J ) 2.1 Hz, 3J ) 7.9 Hz, 1H), 6.66 (d, 3 J ) 7.9 Hz, 1H), 6.71 (d, 4J ) 2.0 Hz, 1H), 7.36 (d, 3J ) 8.7 Hz, 2H), 7.52 (t, 3J ) 5.8 Hz, 1H), 7.71 (s, 1H), 7.82 (m, 7H), 8.14 (m, 8H), 8.82 (d, 3J ) 4.0 Hz, 8H) ppm. 13C NMR (100.5 MHz, 25 °C, THF-d8): δ ) 31.87, 35.45, 36.22, 41.52, 68.88, 114.07, 115.92, 116.52, 120.44, 120.52, 120.96, 124.51, 131.61, 131.76, 135.26, 136.27, 136.48, 140.42, 144.40, 146.48, 151.41, 159.18, 168.10 ppm. UV/vis (ethanol): λmax (log ) 417 (5.26), 519 (4.58), 551 (4.51), 593 (4.42), 646 nm (4.36). MS (FAB): m/z (%) 993 [M]+. 5-(N-(3,4-Dihydroxyphenethyl)-2-phenoxyacetamide)-10,15,20(p-tert-butyltriphenyl)porphyrinatozinc (DOPAZ). An amount of 110 mg of 5-(N-(3,4-dihydroxyphenethyl)-2-phenoxyacetamide)-10,15,20-(p-tert-butyltriphenyl)porphyrin (0.11 mmol) was dissolved in 100 mL of THF, and 80 mg of zinc acetate (0.44 mmol) was added. The mixture was heated to reflux over a period of 4 h. The reaction was monitored by TLC by the disappearance of the free base porphyrin. The solution was concentrated on a rotary evaporator, precipitated with water, and finally filtered. The dried product was obtained as a purple red solid in very good yield (113 mg, 98%). 1H NMR (400

4672 J. Phys. Chem. C, Vol. 113, No. 11, 2009 MHz, 25 °C, THF-d8): δ ) 1.62 (s, 27H), 2.76 (t, 3J ) 7.3 Hz, 2H), 3.54 (dd, 3J ) 6.5 Hz, 3J ) 14.1 Hz, 2H), 4.71 (s, 2H), 6.56 (dd, 4J ) 2.1 Hz, 3J ) 7.9 Hz, 1H), 6.67 (d, 3J ) 7.9 Hz, 1H), 6.71 (d, 4J ) 2.0 Hz, 1H), 7.34 (d, 3J ) 8.7 Hz, 2H), 7.54 (t, 3J ) 5.8 Hz, 1H), 7.72 (s, 1H), 7.80 (d, 3J ) 8.3 Hz, 6H), 7.84 (s, 1H), 8.12 (m, 8H), 8.85 (m, 8H) ppm. 13C NMR (100.5 MHz, 25 °C, THF-d8): δ ) 31.93, 35.41, 36.25, 41.52, 68.91, 113.67, 116.51, 120.54, 120.86, 121.45, 124.09, 131.99, 132.19, 132.27, 135.34, 136.39, 137.82, 141.69, 144.38, 146.48, 150.83, 151.13, 151.24, 158.78, 168.17 ppm. UV/vis (ethanol): λmax (log ) 424 (5.23), 557 (3.95), 598 nm (3.67). MS (FAB): m/z (%) 1053 [M]+. 5-(N,2,2-Triphenylbenzo[d][1,3]dioxole-5-carboxamide)10,15,20-tris(4-tert-butylphenyl)porphyrin. An amount of 100 mgof3,4-diphenylmethylenedioxyprotocatechuicacid46 (0.31mmol) was dissolved in 200 mL of dichloromethane at 0 °C. Then 130 mg of DCC (0.62 mmol) and 85 mg of HOBT (0.62 mmol) were added, and the solution was stirred for 1 h at 0 °C. After that period 250 mg of 5-(4-aminophenyl)-10,15,20-tris(4-tertbutylphenyl)porphyrin 2 (Scheme 1)47 (0.62 mmol) was added to the solution, and the mixture was stirred at room temperature for 96 h. The solvent was removed on a rotary evaporator, and the residue was purified by column chromatography on silica gel with dichloromethane as eluent. The yield was 120 mg (35%). 1H NMR (400 MHz, 25 °C, CDCl3): δ ) -2.76 (s, 2H), 1.61 (s, 27H), 7.04 (d, 3J ) 7.9 Hz, 1H) 7.42-7.65 (m, 12H), 7.76 (d, 3J ) 8.3 Hz, 6H), 8.00 (d, 3J ) 8.5 Hz, 2H), 8.03 (s, 1H), 8.14 (m, 6H), 8.22 (d, 3J ) 8.5 Hz, 2H), 8.87 (m, 8H) ppm. 13C NMR (100.5 MHz, 25 °C, CDCl3): δ ) 31.63, 34.84, 108.05, 108.51, 118.31, 120.30, 121.86, 123.66, 126.34, 128.47, 129.49, 129.34, 131.11, 134.54, 135.33, 137.75, 138.51, 139.25, 139.73, 145.91, 150.52, 150.56, 165.47 ppm. UV/vis (ethanol): λmax (log ) 417 (5.07), 520 (4.31), 557 (4.28), 596 (4.18), 648 nm (4.17). MS (FAB): m/z (%) 1099 [M]+. 5-(3,4-Dihydroxy-N-phenylbenzamide)-10,15,20-tris(4-tertbutylphenyl)porphyrin. An amount of 100 mg of 5-(N,2,2triphenylbenzo[d][1,3]dioxole-5-carboxamide)-10,15,20-tris(4tert-butylphenyl)porphyrin was dissolved in 100 mL of formic acid. The solution was stirred further for 5 h. The progress of the reaction was monitored via TLC. After completion of the reaction, the solvent was removed by rotary evaporation. The mixture was transferred into toluene and evaporated twice to remove any residual formic acid. The final product is obtained by recrystallization in toluene overnight at 0 °C with a yield of 83 mg (98%). 1H NMR (400 MHz, 25 °C, THF-d8): δ ) 2.66 (s, 2H), 1.61 (s, 27H), 6.86 (d, 3J ) 8.2 Hz, 1H), 7.49 (dd, 4 J ) 2.1 Hz, 3J ) 8.2 Hz, 1H), 7.56 (d, 4J ) 2.1 Hz, 1H), 7.82 (d, 3J ) 8.2 Hz, 6H), 8.15 (m, 8H,), 8.25 (d, 3J ) 8.5 Hz, 2H), 8.51 (s, 1H), 8.67 (s, 1H), 8.83 (m, 6H), 8.91 (d, 3J ) 4.6 Hz, 2H), 9.59 (s, 1H) ppm. 13C NMR (100.5 MHz, 25 °C, THFd8): δ ) 31.94, 35.51, 115.42, 115.98, 118.78, 120.40, 120.99, 124.58, 128.18, 131.89, 135.36, 135.72, 137.78, 140.52, 141.14, 146.29, 150.07, 151.43, 166.33 ppm. UV/vis (ethanol): λmax (log ) 417 (5.03), 518 (4.50), 551 (4.49), 596 (4.43), 647 nm (4.36). MS (MALDI-TOF): m/z (%) 934 [M]+. 5-(3,4-Dihydroxy-N-phenylbenzamide)-10,15,20-tris (4-tertbutylphenyl)porphyrinatozinc (CAMIZ). An amount of 80 mg of 5-(3,4-dihydroxy-N-phenylbenzamide)-10,15,20-tris(4-tertbutylphenyl)porphyrin (0.09 mmol) was dissolved in 50 mL of THF, and 65 mg of zinc acetate (0.36 mmol) was added. The mixture was heated to reflux over a period of 4 h. The reaction was monitored by TLC by the disappearance of the free base porphyrin. The solution was concentrated on a rotary evaporator, precipitated with water, and finally filtered. The dried product

Marczak et al. was obtained as a purple red solid in nearly quantitative yield (81 mg, 95%). 1H NMR (400 MHz, 25 °C, THF-d8): δ ) 1.62 (s, 27H), 6.86 (d, 3J ) 8.2 Hz, 1H), 7.48 (dd, 4J ) 2.1 Hz, 3J ) 8.2 Hz, 1H), 7.55 (d, 4J ) 2.1 Hz, 1H), 7.81 (d, 3J ) 8.2 Hz, 6H), 8.14 (m, 8H), 8.22 (d, 3J ) 8.7 Hz, 2H), 8.44 (s, 1H), 8.60 (s, 1H), 8.86 (m, 6H), 8.92 (d, 3J ) 4.7 Hz, 2H), 9.52 (s, 1H) ppm. 13C NMR (100.5 MHz, 25 °C, THF-d8): δ ) 30.65, 35.52, 115.42, 115.98, 118.78, 120.40, 120.99, 124.58, 128.18, 131.78, 131.99, 135.36, 135.72, 137.78, 140.52, 141.14, 146.29, 150.07, 150.74, 151.15, 151.43, 166.33 ppm. UV/vis (ethanol): λmax (log ) 424 (5.08), 559 (4.06), 601 nm (3.98). MS (FAB): m/z (%) 995 [M]+. Spectroscopic Characterization. In order to investigate the interaction between ZnO nanoparticles and dye molecules, the ethanolic solutions of ZnO nanoparticles and different concentration of porphyrins were mixed and sonicated for 15 min. Thermogravimetric measurements were carried out on a Netzsch STA 409 CD. The powder of the samples was obtained by drying the concentrated colloid at 30 °C in vacuum for ca. 5 h. A small amount of the powders (about 15-20 mg) was heated from room temperature to 1000 °C under a helium flow of 80 mL min-1. The UV-vis spectra were recorded with a PerkinElmer Lambda 2 spectrophotometer. Steady-state fluorescence studies were carried out with a Fluoromax 3 (Horiba) instrument. Fluorescence lifetimes were measured with a Laser Strobe fluorescence lifetime spectrometer (Photon Technology International) with 337-nm laser pulses from a nitrogen laser equipped with a stroboscopic detector. Results and Discussion ZnO Nanoparticles Synthesis. ZnO nanoparticles were formed via precipitation in ethanolic solution. The zinc acetate and lithium hydroxide solutions in ethanol were mixed at different temperatures under vigorous stirring. In order to follow the progress of oxide formation, the suspensions of the nanoparticles were aged in their mother liquors and the crystal growth was monitored by UV-vis spectroscopy and DLS. The influence of nanocrystal size on the electronic structure of semiconducting material is represented by the band gap increasing with decreasing of the particle sizes, which is attributed to the quantum confinement effect. ZnO shows this effect for particles smaller than 8 nm.25,48,49 Hence, the measurement of absorption spectra provides a convenient way to investigate particle growth. Figure 1 shows the absorption spectra of the ZnO suspension recorded immediately after the preparation (3 min) and performed in 60 min intervals for a total of 14 h. Just after the zinc acetate was mixed with the lithium hydroxide, the absorbance spectrum shows a well-defined exciton peak at 300 nm, which suggests a very fast nucleation. In addition, a marked red shift in the absorption edge was observed during the early formation stagesan indication of fast crystal growth. These absorption measurements also allow gathering insight into the crystal size distribution. In fact, the sharp excitonic peak in the absorption spectra, as seen in the case of small nanocrystals (at shorter times), is indeed indicative of a narrow size distribution of the corresponding nanoparticles in the sample. For the larger particles, on the other hand, sharp excitonic features are absent in the absorption spectra (at longer times). Instead only a broad and featureless absorption edge is registered. The latter is due to the fact that a number of exciton peaks appear at different energies corresponding to different sized nanocrystals, which are superimposed onto each other. As a matter of fact, we may expect that the nanocrystals reveal a rather broadened size distribution as the aging time increases. Independent confirma-

Synthesis, Characterization of ZnO Nanoparticles

Figure 1. Room temperature absorption spectra recorded at different aging times of ZnO nanocrystals in ethanolic suspension. Red-shift in the absorption onset was observed between interval time 0 and 60 min.

Figure 2. Number size distributions recorded by DLS at different aging times of the ZnO crystals in ethanolic suspension at 25 °C. The size distribution broadening was observed between interval time 0 and 60 min.

tion for this hypothesis was deduced from measurements that focused on the number-weighted size distributions recorded using DLS at different times during the aging of the particles (Figure 2). Here, the longer the aging proceeds, the larger particles with broader size distributions were detected. The influence of the aging temperature, i.e., 10, 25, and 35 °C, on the particle growth was also examined by using UV-vis absorption spectroscopy and DLS. For each of the absorption spectra, the average band gap energies were calculated from the absorption onset (Figure 3) and correlated with the different aging times. The higher the aging temperature, the smaller the energy band gap observed at the same time. Thus, larger ZnO nanocrystals were formed. Hydrodynamic diameter changes that were monitored during the aging by DLS (Figure 4) further confirm this. These results imply that the particle size kept on evolving, even when their suspensions were stored at low temperatures. Moreover, higher aging temperature leads to lower colloidal stability. At a temperature of 35 °C, about 7 h after the start of the synthesis, the transparent ZnO suspension turned white with the simultaneous increase in hydrodynamic diameter (Figure 4). Such behavior is indicative of particle coagulation. Additionally, a marked decrease in the corresponding band gap

J. Phys. Chem. C, Vol. 113, No. 11, 2009 4673 energy is observed (Figure 3). Important in this context is that the energy band gap gradually shifts toward the value for the macrocrystalline ZnO (∼3.34 eV)50 as the size of the particles grows. The ability to obtain various stable particle sizes is based on the phenomena that the growth of the particles is governed not only by aging temperature and time but also by the presence of the reaction byproduct, lithium acetate, during the aging process. The reaction could be almost completely stopped by removing this byproduct and, in turn, providing unique control over the exact particle size. This process, so-called “washing”, consisted of reversible flocculation of the ZnO nanoparticles by addition of n-heptane.20 After centrifugation and removal of the supernatant, the white ZnO flocculates were either redispersed in ethanol to obtain a transparent colloid or dried under nitrogen for about 5 min to get a white powder. The powder could be stored and redispersed even months after its synthesis. To verify this restriction on crystal growth, the ZnO flocculation was enforced after 4 h of aging, and the optical, compositional, and structural characterizations were carried out on the washed ZnO nanocrystal solution as well as the redispersed powder. No significant difference was registered.

Figure 3. Calculated band gap energies as a function of aging time for the ZnO nanoparticles in ethanolic suspension at 10 °C (triangles), 25 °C (circles), and 35 °C (squares). The band gap energy can be tuned continuously between 3.3 and 3.9 eV by choosing the reaction time and temperature.

Figure 4. Particle aging in the ZnO ethanolic suspension at 10 °C (triangles), 25 °C (circles), and 35 °C (squares) monitored by DLS.

4674 J. Phys. Chem. C, Vol. 113, No. 11, 2009 ZnO Nanoparticles’ Optical, Compositional, and Structural Characterizations. Figure 5 shows the steady-state absorption and emission spectra recorded for the suspension containing the washed ZnO nanoparticles. The absorption onset of 354 nm corresponds to a band gap energy of 3.5 eV. The emission spectrum consists of a relatively narrow emission band in the UV region at 355 nm and an intense broad emission band, which is observed in the visible region of the spectrum and that maximizes at 530 nm. The UV emission is due to direct recombination of photogenerated charge carriers, i.e., electrons in the conduction band and holes in the valence band (exciton emission).22,51-55 The width of the exciton emission corresponds to an inhomogeneous broadening of the particle size distribution. The visible emission, on the other hand, relates to the defects in ZnO and has been extensively examined.51-60 Peaks observed in the range from 494 to 582 nm are attributed to oxygen vacancies.55,58-60 Thus, in our study the emission at 530 nm (2.34 eV) originates from oxygen vacanciessa fact that could be rationalized by insufficient oxidation conditions during the nanoparticles synthesis. The experimental number size distribution of the washed ZnO nanoparticles remained in a narrow range, as shown in Figure 6, with a mean hydrodynamic diameter of 4.90 and 4.0 nm obtained by DLS and from HRTEM image analysis, respectively. These results are in accordance with the size distribution constructed from conversion of the absorption spectrum (Figure 5 solid line) by an algorithm developed by Peukert et al.61 This algorithm converts the measured absorption spectra into the single particle contributions using the bulk absorption coefficient determined by Bergstro¨m62 and the tight binding model (TBM)63 to correlate the measured wavelengths with distinct particle sizes. Structural analysis of the sample was performed by using X-ray diffraction. In the XRD pattern of the ZnO nanoparticles presented in Figure 7, no impurity peaks are observed. All the diffraction peaks are well assigned to the standard hexagonal phase of ZnO with a wurtzite structure reported in JCPDS card no. 36-1451 (a ) 3.249 Å, c ) 5.206 Å). The broadness of the XRD peaks reveals the nanocrystalline nature and size of the ZnO nanocrystals. An average crystallite size of 4.7 nm was calculated by using the Debye-Scherrer’s equation64 coinciding with the previously described mean particles size. The crystalline nature of the ZnO nanoparticles is also evidenced in the HRTEM images (Figure 8) that gives rise to lattice features. Moreover, it is seen that the nanocrystals are highly monodisperse.

Figure 5. Room temperature steady-state absorption (solid line) and emission (dotted line) spectra recorded for the washed ZnO nanocrystals in ethanolic suspension. Excitation wavelength was 330 nm.

Marczak et al.

Figure 6. Number size distributions of the washed ZnO nanoparticles’ ethanolic suspension obtained by DLS (black line), image analysis of HRTEM (gray columns), and inversion of the UV-vis spectrum (Figure 5a) based on the TBM (blue line).

Figure 7. X-ray powder pattern of ZnO nanoparticles. All peaks correspond to the ZnO wurtzite structure. The broadness of the XRD peaks reveals the nanocrystalline nature of the ZnO powders. The average crystallite size was calculated to be 4.7 nm using Debye-Scherrer’s equation.

As it was reported before,56 the ZnO particle surface is not bare but is stabilized with acetate groups, which originate from the precursor materials and are adsorbed on the surface of the crystals. Acetate ions that are bound to the ZnO surface are effectively preventing coagulation of colloidal ZnO nanoparticles. The presence of acetate groups on the ZnO particle surface is confirmed by infrared spectroscopy (Figure 9) showing bands at about 1585 and 1415 cm-1 and at around 1343 cm-1.56 These bands correspond to CdO stretching (1585 cm-1) and C-O stretching (1415 cm-1). The smaller band at 1343 cm-1 is due to the weakly bound acetic acid molecules. The intensive band below 500 cm-1 is attributed to the vibrational modes of Zn-O.65,66 ZnO Nanoparticles’ Surface Functionalization. To examine the capability for the surface functionalization of the ZnO surface, the washed ZnO particles were mixed with solutions of dye molecules, i.e., DOPAZ and CAMIZ. The bond stability of the bond between dye and nanoparticles was probed with TGA-MS. The results are summarized in Figure 10. A weight loss of about 13% is discernible in the range of 300-350 °C in the case of not-functionalized ZnO as well as for the dyesensitized nanoparticles due to the loss of acetate groups

Synthesis, Characterization of ZnO Nanoparticles

J. Phys. Chem. C, Vol. 113, No. 11, 2009 4675

Figure 10. TGA curves of pure ZnO nanoparticles (solid line), ZnO nanoparticles grafted with CAMIZ (dashed line), and ZnO nanoparticles grafted with DOPAZ (dotted line) in the range of 25-1000 °C over 5 h.

Figure 8. High resolution transmission electron microscopy image of ZnO nanocrystals.

spectra of ethanolic suspensions of the ZnO in the absence and in the presence of the dye molecules. The peak at 330 nm corresponds to the absorption of ZnO nanoparticles, whereas the strong feature at 424 nm and weaker features at 550 and 600 nm are attributed to the porphyrins’ Soret and Q-bands, respectively. A broad absorption of the dye molecules is also observed in the UV range.

Figure 9. Infrared spectrum of ZnO nanoparticles covered with acetate groups.

adsorbed on the surface. This trend documents that carboxylic acids are relatively labile, what renders their use as anchoring group to metal oxides rather limiting. In the range of 600-780 °C, an additional weight loss takes place when the dyes are functionalized onto the nanoparticles. In particular, an additional 15% and 30% were noted for CAMIZ and DOPAZ grafted onto ZnO, respectively. Implicit is that the relatively flexible anchoring group of DOPAZ containing an ethyl spacer allows a higher dye coverage onto the surface, when compared to the rigid anchoring group of CAMIZ. However, the catechol group is bound very strongly to ZnO surfaces and is stable up to 600 °C, exceeding the stability of carboxylic anchor groups by far. Electronic Communication. Next, the interactions of ZnO nanoparticles with the dye molecules, i.e., DOPAZ and CAMIZ, were tested in titration experiments by using steady-state absorption and emission spectroscopy. In these cases the ZnO nanoparticles concentrations were kept constant (1.5 × 10-4 M, based on the Zn2+ concentration), while those of the dye molecules were varied incrementally between 0 and 2.25 × 10-6 M and 0 and 4.49 × 10-6 M for DOPAZ and CAMIZ, respectively. Figures 11a and b show the UV-vis absorption

Figure 11. Room temperature absorption spectra of ZnO in ethanol in the presence of different concentrations of (a) DOPAZ (0-2.25 µM) and (b) CAMIZ (0-4.49 µM).

4676 J. Phys. Chem. C, Vol. 113, No. 11, 2009

Marczak et al.

Figure 13. Normalized fluorescence lifetime measurements for the visible band of ZnO nanoparticles in the presence of different CAMIZ concentrations: (a) 0, (b) 0.57, and (c) 1.60 µM.

Figure 12. Room temperature fluorescence spectra of ZnO nanoparticles in ethanol, exhibiting an optical absorption of 0.1 at the wavelength of photoexcitation of 330 nm, in the presence of different concentrations of (a) DOPAZ (0-2.25 µM) and (b) CAMIZ (0-4.49 µM).

As we described before, the visible emission from ZnO colloids arises from oxygen vacancies. In turn, the 530 nm emissionsupon photoexcitation at 330 nmsemerged as a useful probe to monitor charge transfer processes at the ZnO interface, thus to investigate the nanoparticles’ interaction with dye molecules. To see whether the ZnO emission is sensitive to the presence of the dye molecules, we monitored the ZnO emission at different concentrations of DOPAZ and CAMIZ (Figures 12a and b). For both of the porphyrins, the ZnO emission intensity decreases with increasing porphyrin concentrations. More than 97% emission quenching of ZnO was achieved when adding 2.25 µM of DOPAZ, whereas the quenching is more than 96% when adding 4.49 µM of CAMIZ. When the porphyrin molecules are adsorbed onto ZnO nanoparticles with perpendicular or parallel orientation relative to the surface of the particle, they occupy an area of about 0.25 or 1.96 nm2, respectively. Thus, to form a monolayer, the maximum number of the adsorbed molecules per nanoparticles is estimated to be about 120 and 15, respectively. In order to deactivate the excited state of ZnO nanoparticles quantitatively, the number of needed DOPAZ and CAMIZ molecules per nanoparticle is 35 and 67, respectively. This corresponds to a porphyrin density of about 1 molecule nm-2 and 2 molecule nm-2. Aggregates have been ruled out since the corresponding shifts in the Soret band have not been observed (Figures 11a

and b). Thus, it is safe to assume a “standing” orientation of the porphyrins relative to the nanoparticle surface and a grafting through the catechol group. In summary, our emission quenching results prove the electronic communication between the ZnO nanoparticles in their excited state and both of the porphyrins. Furthermore, it suggests that the visible emission of the ZnO is caused by surface states, since only the surface of the particles is altered by these grafting experiments. If indeed the observed emission quenching of ZnO arises from the charge transfer interaction with the dye molecules, enhanced decay rates of the visible emission with increased concentration of the porphyrins should be observed. In order to confirm this assumption, the samples were analyzed by time-resolved fluorescence spectroscopy. The experiments on the lifetime measurements were carried out by exciting ZnO nanoparticles at 355 nm and monitoring the visible emission at 530 nm. Figure 13 displays the emission decay profiles of ZnO nanoparticles in the presence of different quencher concentrations. All the emission decays could be best fitted by a biexponential fitting function in the studied time range, and the decay components were used to determine the average lifetimes.67 With increasing porphyrin concentrations, the lifetimes change from 46.0 ns down to 15.3 ns. The shortened lifetimes prompt to a new deactivation pathway, which is now available for the deactivation of the excited state. In the absence of the dye molecules, the ZnO emission decay reflects the charge carrier recombination via radiative and nonradiative processes.51-60,67 In the presence of the porphyrins, an additional pathway is introduced, namely, through the electronic coupling of the porphyrins to the ZnO nanoparticles. If the decrease in lifetime is entirely due to electron transfer to the porphyrins, a rate constant of 0.35 × 108 s-1 could be determined for this process.67 Finally, in the context of excited-state interactions, we have added fluorescence experiments that focus on the DOPAZ emission rather than on that of ZnO. In particular, porphyrins were grafted to ZnO and Al2O3, and the porphyrin fluorescence was tested as a function of conduction band energy, that is, ZnO versus Al2O3. Notable, higher conduction band energies in Al2O3srelative to that of ZnOsrender an electron transfer deactivation of the porphyrin singlet excited-state thermodynamically unfeasible. The corresponding fluorescence spectra

Synthesis, Characterization of ZnO Nanoparticles

Figure 14. Room temperature fluorescence spectra of ZnO (solid line) and Al2O3 (dotted line) DOPAZ-sensitized films, exhibiting the same optical absorption at the wavelength of photoexcitation of 424 nm.

of the ZnO and Al2O3 films are gathered in Figure 14. The porphyrin emission when grafted to Al2O3 is nearly unchanged relative to what is seen in reference experiments. In stark contrast, the porphyrin emissionsexciting into the Soret band at 424 nm sis nearly quantitatively quenched in the ZnOsensitized films. Taking the aforementioned into account, we conclude charge injection that evolves from the photoexcited porphrins with an injection efficiency of about 90%. Conclusions The synthesis of ZnO nanoparticles in ethanol by using a simple colloidal method was presented. It was shown that the growth of the particles was governed by temperature as well as the presence of the reaction byproduct lithium acetate during the aging process. This process could be almost completely stopped by removal of this byproduct by repeated flocculation of the ZnO particles by addition of n-heptane. Those ZnO nanoparticles were shown to be suitable for further derivatization of their surface. The catechol-functionalized zinc porphyrins could be used as support architectures to anchor the ZnO nanoparticles. Steady-state absorption spectroscopy as well as steady-state and time-resolved emission studies confirmed an interaction between the ZnO and dye molecules and electron injection from the dye molecules into the ZnO. The results presented in this study open the way toward the design of ordered ZnO-based nanostructures that can harvest light energy efficiently. Acknowledgment. The authors gratefully acknowledge the funding of the German Research Council (DFG), which, within the framework of its “Excellence Initiative”, supports the Cluster of Excellence “Engineering of Advanced Materials” (www. eam.uni-erlangen.de) at the University of Erlangen-Nuremberg. The authors want to thank Christoph Dotzer for the TGA-MS measurements and Dr. Robin Klupp Taylor for the HRTEM measurements. References and Notes (1) Spahnel, L. J. Sol-Gel Sci. Technol. 2006, 39, 7. (2) O’Regan, B.; Gra¨tzel, M. Nature 1991, 353, 737. (3) Quintana, M.; Edvinsson, T.; Hagfeldt, A.; Boschloo, G. J. Phys. Chem. C 2007, 111, 1035. (4) Redmond, G.; Fitzmaurice, D.; Gra¨tzel, M. Chem. Mater. 1994, 6, 686.

J. Phys. Chem. C, Vol. 113, No. 11, 2009 4677 (5) Zeng, L.; Dai, S.; Xu, W.; Wang, K. Plasma Sci. Technol. 2006, 8, 172. (6) Colvin, V. L.; Schlamp, M. C.; Alivisatos, A. P. Nature 1994, 370, 354. (7) Duran, P.; Capel, F.; Tartaj, J.; Moure, C. AdV. Mater. 2002, 14, 137. (8) Rout, C. S.; Raju, A. R.; Govindaraj, A.; Rao, C. N. R. Solid State Commun. 2006, 136. (9) Yang, M.; Wang, D.; Peng, L.; Zhao, Q.; Lin, Y.; Wei, X. Sens. Actuators, B 2006, 117, 20. (10) Johnson, J. C.; Yan, H.; Schaller, R. D.; Haber, L. H.; Saykally, R. J.; Yang, P. J. Phys. Chem. B 2001, 105, 11387. (11) Zhang, X. T.; Liu, Y. C.; Zhang, L. G.; Zhang, J. Y.; Lu, Y. M.; Shen, D. Z.; Xu, W.; Zhong, G. Z.; Fan, X. W.; Kong, X. G. J. Appl. Phys. 2002, 92, 3293. (12) Wang, Z. L. J. Phys.: Condens. Matter 2004, 16, R829. (13) Okuyama, K.; Lenggoro, I. W. Chem. Eng. Sci. 2003, 58, 537. (14) Park, S. B.; Kang, Y. C. J. Aerosol. Sci. 1997, 28, S473. (15) Sato, T.; Tanigaki, T.; Suzuki, H.; Saito, Y.; Kido, O.; Kimura, Y.; Kaito, C.; Takeda, A.; Keneko, S. J. Cryst. Growth 2003, 255, 313. (16) Hong, R. Y.; Shen, Z. H.; Li, H. Z. J. Process Eng. 2005, 5, 693. (17) Hu, X. L.; Zhu, Y. J.; Wang, S. W. Mater. Chem. Phys. 2004, 88, 421. (18) Liu, B.; Zeng, H. C. J. Am. Chem. Soc. 2003, 125, 4430. (19) Zhang, H.; Yang, D.; Ji, Y.; Ma, X. Y.; Xu, J.; Que, D. L. J. Phys. Chem. B 2004, 108, 3955. (20) Meulenkamp, E. A. J. Phys. Chem. B 1998, 102, 5566. (21) Pacholski, C.; Kornowski, A.; Weller, H. Angew. Chem., Int. Ed. Engl. 2002, 41, 1188. (22) Spanhel, L.; Anderson, M. A. J. Am. Chem. Soc. 1991, 113, 2826. (23) Verges, M. A.; Mifsud, A.; Serna, C. J. J. Chem. Soc., Faraday Trans. 1990, 86, 959. (24) Harnack, O.; Pacholski, C.; Weller, H.; Yasuda, A.; M. J.; Wessels, J. M. Nano Lett. 2003, 3, 1097. (25) Wong, E. M.; Bonevich, J. E.; Searson, P. C. J. Phys. Chem. B 1998, 102, 7770. (26) Hu, Z.; Escamilla Ramirez, D. J.; Heredia Cervera, B. E.; Oskam, G.; Searson, P. C. J. Phys. Chem. B 2005, 109, 11209. (27) Hu, Z.; Herrera Santos, J. F.; Oskam, G.; Searson, P. C. J. Colloid Interface Sci. 2005, 288, 313. (28) Hu, Z.; Oskam, G.; Penn, R. L.; Pesika, N.; Searson, P. C. J. Phys. Chem. B 2003, 107, 3124. (29) Hu, Z.; Oskam, G.; Searson, P. C. J. Colloid Interface Sci. 2003, 263, 454. (30) Miller, R. D. J.; McLendon, G. L.; Nozik, A. J.; Schmickler, W.; Willig, F. Surface electron-transfer processes; VCH: New York, 1995. (31) Hagfeldt, A.; Gra¨tzel, M. Chem. ReV 1995, 95, 49. (32) Hannapel, T.; Burfeindt, B.; Storck, W.; Willig, F. J. Phys. Chem. B 1997, 101, 6799. (33) Ghosh, H. N.; Asbury, J. B.; Lian, T. J. Phys. Chem. B 1998, 102, 6482. (34) Ghosh, H. N. J. Phys. Chem. B 1999, 103, 10382. (35) Spanhel, L.; Weller, H.; Henglein, A. J. Am. Chem. Soc. 1987, 109, 6632. (36) Asbury, J. B.; Wang, Y.; Lian, T. J. Phys. Chem. B 1999, 103, 6643. (37) Bedja, I.; Kamat, P. V.; Hua, X.; Lappin, A. G.; Hotchandani, S. Langmuir 1997, 13, 2398. (38) de Jongh, P. E.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Kelly, J. J. J. Phys. Chem. B 2000, 104, 7686. (39) Kalyanasundaram, K.; Gra¨tzel, M. Coord. Chem. ReV. 1998, 177, 347. (40) Gallopini, E. Coord. Chem. ReV. 2004, 248, 1283. (41) Verma, S.; Kar, P.; Das, A.; Palit, D. K.; Ghosh, H. G. J. Phys. Chem. C 2008, 112, 2918. (42) Ramakrishna, G.; Verma, S.; Jose, D. A.; Krishna Kumar, D.; Das, A.; Palit, D. K.; Ghosh, H. N. J. Phys. Chem. B 2006, 110, 9012. (43) Ramakrishna, G.; Jose, D. A.; Krishna Kumar, D.; Das, A.; Palit, D. K.; Ghosh, H. N. J. Phys. Chem. B 2005, 109, 15445. (44) Forneli, A.; Planells, M.; Sarmentero, M. A.; Martinez-Ferrero, E.; O’Regan, B. C.; Ballester, P.; Palomares, E. J. Mater. Chem. 2008, 18, 1652. (45) Wessendorf, F.; Gnichwitz, J.-F.; Sarova, G. H.; Hager, K.; Hartnagel, U.; Guldi, D. M.; Hirsch, A. J. Am. Chem. Soc. 2007, 129, 16057. (46) Iacazio, G.; Perissol, C.; Faure, B. J. Microbiol. Methods 2000, 42, 209. (47) Liu, X.; Liu, J.; Jin, K.; Yang, X.; Peng, Q.; Sun, L. Tetrahedron 2005, 61, 5655. (48) Brus, L. E. J. Phys. Chem. B 1986, 90, 2555. (49) Koch, U.; Fojtik, A.; Weller, H.; Henglein, A. Chem. Phys. Lett. 1985, 122, 507. (50) Sadao, A. Handbook on Physical Properties of Semiconductors; Springer-Verlag: New York, 2004.

4678 J. Phys. Chem. C, Vol. 113, No. 11, 2009 (51) van Dijken, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink, A. J. Lumin. 2000, 90, 123. (52) Bahnemann, D. W.; Kormann, C.; Hoffmann, M. R. J. Phys. Chem. 1987, 91, 3789. (53) Sakahora, S.; Ishida, M.; Anderson, M. A. J. Phys. Chem. B 1998, 102, 10169. (54) van Dijken, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink, A. J. Phys. Chem. B 2000, 104, 1715. (55) van Dijken, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink, A. J. Lumin. 2000, 87-89, 454. (56) Sakohara, S.; Tickanen, L. D.; Anderson, M. A. J. Phys. Chem. 1992, 96, 11086. (57) Monticone, S.; Tufeu, R.; Kanaev, A. V. J. Phys. Chem. B 1998, 102, 2854. (58) Vanheusden, K.; Warren, W. L.; Seager, C. H.; Tallant, D. R.; Voigt, J. A. J. Appl. Phys. 1996, 79, 7983. (59) Fujihara, S.; Ogawa, Y.; Kasai, A. Chem. Mater. 2004, 16, 2965.

Marczak et al. (60) Vanheusden, K.; Seager, C. H.; Warren, W. L.; Tallant, D. R.; Voigt, J. A. Appl. Phys. Lett. 1996, 68, 403. (61) Gradl, J.; Segets, D.; Peukert, W. Manuscript in preparation. (62) Bergstroem, L.; Meurk, A.; Arwin, H.; Rowcliffe, D. J. J. Am. Ceram. Soc. 1996, 79, 339. (63) Viswanatha, R.; Sapra, S.; Satpati, B.; Satyam, P. V.; Dev, B. N.; Sarma, D. D. J. Mater. Chem. 2004, 14, 661. (64) West, A. R. Solid State Chemistry and Its Applications; John Wiley & Sons: New York, 1984; p 174-175. (65) Hong, R. Y.; Qian, J. Z.; Cao, J. X. Powder Technol. 2006, 163, 160. (66) Kwon, Y. J.; Kim, K. H.; Lim, C. S.; Shim, K. B. J. Ceram. Process. Res. 2002, 3, 146. (67) Vietmeyer, F.; Seger, B.; Kamat, P. V. AdV. Mater. 2007, 19, 2935.

JP810696H