Communication DOI: 10.1002/cvde.201004289 CVD Co3O4 Nanopyramids: a Nano-Platform for Photo-Assisted H2 Production** Dedicated to Prof. Mauro Graziani, an exemplary scientific guide, on the occasion of his retirement By Davide Barreca,* Paolo Fornasiero,* Alberto Gasparotto, Valentina Gombac, Chiara Maccato, Andrea Pozza, and Eugenio Tondello
Hydrogen is considered a promising environmentally friendly energy source since it can be used as a post-fossil fuel energy vector to feed fuel cells in a clean and efficient way.[1,2] Unfortunately, its effective and sustainable production is still an open challenge for the transition to a real H2based economy.[3,4] In this regard, a strategically appealing goal is the use of electromagnetic radiation to produce H2 by means of photo-assisted processes, starting from widely available source materials.[1] Yet, at variance with water photo-splitting, photo-reforming is a less explored method for deriving H2 from biomass extracts such as glucose, and oxygenates such as alcohols.[5,6] The development of such processes relies on the availability of suitable active systems endowed with the required catalytic activity. In this regard, materials based on cobalt oxides and, in particular, spinel-type Co3O4, have been the object of considerable attention for applications in Fischer-Tropsch syntheses,[7] hydrocracking of crude fuels,[8] oxidation reactions,[9,10] and ethanol steam reforming.[3,4] Recently, their photocatalytic activity has been investigated in the degradation of dyes,[11,12] and in oxygen evolution from water under mild conditions.[13] Nevertheless, only very few [*] Dr. D. Barreca CNR-ISTM and INSTM Department of Chemistry - Padova University, via F. Marzolo, 1 - 35131 Padova, Italy E-mail:
[email protected] Prof. P. Fornasiero, Dr. V. Gombac Department of Chemistry - ICCOM-CNR Trieste Research Unit INSTM Trieste Research Unit - Trieste University, via L. Giorgieri, 1 34127 Trieste, Italy E-mail:
[email protected] Dr. A. Gasparotto, Dr. C. Maccato, Dr. A. Pozza, Prof. E. Tondello Department of Chemistry, Padova University and INSTM, via F. Marzolo, 1 - 35131 Padova, Italy [**] This work was financially supported by CNR-INSTM PROMO, Fondazione CR Trieste, COFIN-PRIN 2007 Project ‘‘2nd generation sustainable processes for the hydrogen production from renewable sources’’ and Regione Lombardia-INSTM program ‘‘Produzione e uso di Idrogeno in Campo energetico: Sviluppo di nanoarchitetture innovative a Base di Ossidi metallici (PICASSO)’’. Thanks are also due to Mr. A. Ravazzolo (CNR-ISTM, Padova, Italy) for the valuable technical help.
296
wileyonlinelibrary.com
studies are available on the use of powdered binary (Co3O4)[14,15] and ternary (CoX2O4, X ¼ Al, Ga, In)[2] spinels for light-driven hydrogen production, whereas the photo-assisted H2 generation by supported Co3O4 nanosystems has never been reported. In this regard, a key step is the development of preparation routes joining the advantages of the elimination of the filtration processes required by powdered catalysts and the performances offered by nanostructured materials with tailored properties. In this context, CVD and related methods represent preferred choices for their inherent flexibility and the possibility of controlling the nanosystem morphology by a proper choice of the process parameters, a key step to the achievement of unprecedented system performances in photo-assisted H2 production.[16] Recently, we have reported on the synthesis of cobalt oxide nanomaterials with phase composition tuneable from CoO to Co3O4.[17] In particular, the use of suitable growth conditions enabled the preparation of Co3O4 nanopyramid assemblies, very attractive for use in hydrogen generation because of the possibility of exploiting the synergy between the peculiar system nano-organization and the Co3O4 catalytic activity. Herein, we focus on the CVD of Si(100)-supported Co3O4 nanopyramids, with particular attention to their morphological characterization and the study of their performances for H2 production by oxygen-assisted photo-reforming of methanol solutions. To the best of our knowledge, no similar studies have ever been performed. Field emission scanning electron microscopy (FE-SEM) images of systems obtained under O2 þ H2O atmospheres (Fig. 1) showed the formation of pyramidal assemblies uniformly covering the Si(100) substrates. The formation of such peculiar aggregates could be associated with the preferential exposure of the (111) planes, the most stable ones in face-centered cubic structures like spinel-type Co3O4,[17] thereby decreasing the overall system surface energy. An increase in the growth temperature from 400 to 500 8C produced an increase of the aggregate mean lateral size from 270 to 550 nm (Figs. 1a-b). In addition, higher magnification images (Fig. 1b, inset) revealed the presence of a spiral-like surface texture, suggesting that the growth occurred according to a spiral dislocation mechanism.[18] The corresponding cross-sectional pictures were characterized by a well-ordered structure, with an increase of the mean nanodeposit thickness from 460 20 nm (400 8C) to 630 20 nm (500 8C). The high purity of the synthesized Co3O4 systems was confirmed by energy dispersive X-ray (EDX) spectroscopy analyses (see Fig. 1). Irrespective of the growth conditions, the patterns were characterized by the presence of the sole cobalt [Co La: 0.78 keV; Co Ka1: 6.92 keV] and oxygen [O Ka1: 0.52 keV] signals, along with the Si Ka substrate
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Vap. Deposition 2010, 16, 296–300
Communication
Fig. 1. Plane-view and cross-sectional FE-SEM images of Co3O4 nanopyramids synthesized at a) 400 8C, and b) 500 8C. In the insets of the left figures, higher magnification micrographs are also displayed. The EDXS scans for O Ka1 and Co Ka1 X-ray signals were recorded along the marked line for the 400 8C-grown specimen.
peak at 1.74 keV, without any detectable carbon contamination. The homogeneous formation of Co3O4 through the whole nanodeposit thickness was investigated by means of local line-scans at several sample points. As a representative example, typical spectra are displayed in Figure 1, showing a uniform chemical composition from the external surface up to the substrate interface. The intensity ratio between the O Ka1 and Co Ka1 X-ray signals, mediated along the whole line scan, provided values very close to those expected for Co3O4. The formation of this phase was further confirmed by X-ray photoelectron spectroscopy (XPS) measurements, yielding the typical Co 2p profile expected for Co3O4. The main Co 2p component was centered at a binding energy (BE) of 780.7 eV, with a mean Auger parameter of 1552.2 eV, irrespective of the adopted deposition temperature.[17] Chem. Vap. Deposition 2010, 16, 296–300
Further insight into the system morphological features was obtained by atomic force microscopy (AFM). In line with the FE-SEM results, AFM analyses confirmed the homogeneous growth of Co3O4 nanoaggregates on the Si(100) substrates, along with the above discussed size increase, on going from 400 to 500 8C. Figure 2 shows the 2 mm 2 mm AFM images for the same specimens as in Figure 1, along with the corresponding line profiles. The latter suggested an increase of the surface roughness with the deposition temperature. Indeed, from the 2 mm 2 mm areas, root mean square (rms) roughness values could be calculated, yielding 28 nm (400 8C) and 62 nm (500 8C). As can be seen, an increase in the deposition temperature resulted in a rougher texture of Co3O4 nanosystems, as expected on the basis of FE-SEM characterization (compare Fig. 1).
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.cvd-journal.de
297
Communication
Fig. 2. Representative AFM surface images (2 mm 2 mm) of Co3O4 nanopyramids synthesized at a) 400 8C, and b) 500 8C. Height profiles along the marked lines are also reported.
Photo-assisted properties of the nanodeposits relating to H2 production were evaluated using 1:1 methanol/water solutions under UV-vis illumination, both in the presence and in the absence of O2 (see the Experimental Section). No appreciable gas-phase product evolution (H2, CH4, CO, CO2, or formaldehyde) was observed under the present experimental conditions when UV-vis illumination and photocatalyst were not present simultaneously. As a prototype for the observed behavior, Figure 3 shows the gas-phase product evolution as a function of irradiation time. In the absence of O2 (Fig. 3a), a relatively modest and transient H2 evolution was observed, with the concomitant formation of relatively small amounts of CO, CO2, CH4, and traces of formaldehyde. As can be seen, hydrogen evolution underwent a rapid decrease within the first two hours. Conversely, when O2 was present in the reaction medium (Fig. 3b), appreciable differences were observed in the Co3O4 nanodeposit reactivity. In fact, under these 298
www.cvd-journal.de
Fig. 3. H2 (&), CO2 (), CO (^), and CH4 (~) production rate per unit area vs. irradiation time for a Co3O4 nanodeposit a) in the absence of O2, and b) in the presence of O2.
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Vap. Deposition 2010, 16, 296–300
Communication
conditions, significant and constant hydrogen evolution was observed for at least 30 h, indicating a promising stability of the investigated nanomaterials in view of practical applications. The evolution of CO, CO2, and CH4 was observed even in this case. In addition, HCHO and CH3OCH3 were detected in the gas phase, while traces of formic acid and methylformate were found in the liquid phase. It is worth noting that the addition of O2 considerably increased the evolution rate of H2 (compare Fig. 3a), while the evolution of carbon-containing by-products was significantly less affected. Traces of dimethoxymethane were also observed in this case. In order to explain the observed reactivity, it is necessary to consider not only the general photocatalysis mechanism, but also the well known capability of cobalt oxides to promote oxidation reactions and the accessibility of cobalt (II) and (III) oxidation states. Whereas the formation of methoxy species by gas-phase interaction of methanol with cobalt oxide surfaces is well documented,[1-3] a thorough understanding of the subsequent degradation steps is a more difficult task, especially in a liquid phase containing a large amount of both water and methanol, as in the present case. Bulk Co3O4 is a p-type semiconductor with a band gap < 2.0 eV,[2,17] indicating the possibility of an efficient UV-vis radiation absorption. As a consequence, Co3O4 illumination leads to electron excitation to the conduction band, along with the simultaneous formation of holes in the valence band.[11] Thanks to the capability of Co3O4 to promote the activation of O2 and oxygen-containing species, photo-generated electrons can react with adsorbed oxygen and water molecules to produce highly oxidizing species, such as O2, H2O2, and OH. The latter, along with photoproduced holes, can react with methoxy species adsorbed onto the surface of Co3O4. The photo-assisted activity improvement observed upon oxygen introduction in the reaction environment can be mainly explained taking into account that O2 present on the surface of the photocatalyst acts as a scavenger for the photo-produced electrons, inhibiting recombination phenomena of charge carriers and thus extending their lifetime. Nevertheless, under the present conditions, additional factors have to be considered in the description of O2 influence on the system
reactivity. In fact, previous in-situ studies pointed to the occurrence of an equilibrium between CoO and Co3O4 depending on the surrounding atmosphere, Co3O4 being observed as pure phase only in the presence of oxygen. In addition, the total oxidation of methanol to CO2, and its partial oxidation to formaldehyde, were suggested as preferential reaction pathways over Co3O4 and CoO surfaces, respectively.[3] On this basis, when photo-assisted experiments are performed in the absence of oxygen, it is reasonable to suppose that H2 evolution results in a partial reduction of Co3O4 to CoO. In this case, the nanosystem surface is likely to be mainly covered by methoxy species (Scheme 1), acting as a poison for the catalyst. Dehydrogenation of these species might occur only at temperatures higher than 80 8C, leading to formaldehyde and surface hydroxyls, with subsequent degradation to CO/CO2.[2] In a different way, oxygen introduction into the reaction atmosphere is believed to have further positive effects: (i) the suppression of Co3O4 reduction to CoO; (ii) the reaction with adsorbed methoxy species to form adsorbed formates and hydrogen, resulting in a less extensive catalyst poisoning. Formates, in turn, can be further decomposed to hydrogen and CO/ CO2[19] (Scheme 1). In conclusion, we have presented a convenient CVD route to Co3O4 systems on Si(100) starting from Co(hfa)2 TMEDA, resulting in homogeneous assemblies of faceted nanopyramids. The obtained materials were tested for the first time in the photo-assisted H2 production from methanol/ water media. A remarkably stable hydrogen evolution rate was observed over significant periods of time, provided that oxygen was present in the reaction environment. The present findings pave the way to the development of mixed cobalt oxide-containing nanocomposites for further advancements in photo-assisted hydrogen generation, with particular attention to the selectivity towards H2 obtainment at the expense of carbon-containing gaseous by-products. The engineering of the present nanosystems as multifunctional catalytic platforms for both hydrogen generation and oxygen activation will also be the object of future studies, with particular attention to the use of sustainable oxygenates, such as glycerol.
Scheme 1. Simplified sketch of the main proposed reaction pathways over Co3O4 nanosystems in CH3OH/H2O media under the present photo-assisted conditions.
Chem. Vap. Deposition 2010, 16, 296–300
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.cvd-journal.de
299
Communication
Experimental Co3O4 nanopyramids were prepared using a cold-wall CVD quartz reactor equipped with a resistively heated susceptor and an external precursor reservoir. Co(hfa)2TMEDA, synthesized as recently reported [20], was vaporized at 60 8C and transported into the reaction chamber by an electronic grade O2 flow (100 sccm), whereas an additional 100 sccm O2 flow, bubbled through a water reservoir heated to 50 8C, was supplied independently. Depositions were performed for 2 h under a total pressure of 10.0 mbar on previously cleaned 1 cm 1 cm 1 mm Si(100) substrates, maintained at temperatures between 400 and 500 8C. Under these conditions, water partial pressure was estimated to be 1.5 mbar. FE-SEM measurements were carried out by means of a Field Emission Zeiss SUPRA 40VP apparatus, equipped with an Oxford INCA x-sight X-ray detector for EDX analyses. XPS analyses were performed using a Perkin Elmer F 5600ci spectrometer with a standard Al Ka source (1486.6 eV), under a working pressure smaller than 109 mbar. The reported BEs were corrected for charging effects by assigning a value of 284.8 eV to the adventitious C 1s signal. AFM images were obtained by a NT-MDT SPM Solver P47H-PRO instrument operating in tapping mode and in air. Images were recorded on different sample areas in order to check surface homogeneity. rms roughness values were calculated after background subtraction through plane fitting. Photo-assisted experiments were carried out on 1:1 water/methanol mixtures in a discontinuous batch reactor directly connected to an Agilent HP 6890 Gas Cromatographic (GC) apparatus for the on-line analysis of gaseous products. A 125 W low-pressure mercury lamp was used for UV-vis irradiation, and no activity was observed in the absence of illumination. Blank tests were performed on the bare Si(100) substrates. Experiments, in the absence of O2, were carried out after de-aeration of the reaction mixture by Ar bubbling and by maintaining a controlled Ar flow throughout the irradiation time. For the experiments in the presence of O2, a controlled flow of O2 (5%)/Ar was used. GC analysis of H2, O2, N2, CH4, and CO was performed using a Molsieve 5A column, with Ar as the carrier, and a thermal conductivity detector (TCD). A PoraPLOT Q column, with He as the carrier, was connected in series to a methanator and to a flame ionization detector (FID) to analyze the carbon-containing compounds [16]. Analysis of the liquid phase was performed using an Agilent 7890A GC system (column: J&W DB-225ms, 60 m, 0.25 mm, 0.25 mm) coupled with a 5975C VL MSD with triple-Axis Detector.
Received: March 30, 2010 Revised: May 7, 2010
300
www.cvd-journal.de
[1] F. Ga¨rtner, B. Sundararaju, A.-E. Surkus, A. Boddien, B. Loges, H. Junge, P. H. Dixneuf, M. Beller, Angew. Chem. Int. Ed. 2009, 48, 9962. [2] A. Walsh, K.-S. Ahn, S. Shet, M. N. Huda, T. G. Deutsh, H. Wang, J. A. Turner, S.-H. Wei, Y. Yan, M. M. Al-Jassim, Energy Environ. Sci. 2009, 2, 774. [3] J. Llorca, P. R. de la Piscina, J.-A. Dalmon, J. Sales, N. Homs, Appl. Catal, B 2003, 43, 355. [4] M. Ni, D. Y. C. Leung, M. K. H. Leung, Int. J. Hydrogen Energy 2007, 32, 3238. [5] R. M. Navarro, M. C. Sa´nchez Sa´nchez, M. C. Alvarez-Galvan, F. del Valle, J. L. G. Fierro, Energy Environ. Sci. 2009, 2, 35. [6] X. L. Fu, J. L. Long, X. X. Wang, D. Y. C. Leung, Z. X. Ding, L. Wu, Z. Zhang, Z. H. Li, X. Z. Fu, Int. J. Hydrogen Energy 2008, 33, 6484. [7] A. Y. Khodakov, W. Chu, P. Fongarland, Chem. Rev. 2007, 107, 1692. [8] D. G. Castner, Ph. R. Watson, I. Y. Chan, J. Phys. Chem. 1989, 93, 3188. [9] S. Zafeiratos, T. Dintzer, D. Teschner, R. Blume, M. Ha¨vecker, A. Knop-Gericke, R. Schlo¨gl, J. Catal. 2010, 269, 309. [10] J. Kirchnerova, M. Alifanti, B. Delmon, Appl. Catal, A 2002, 231, 65. [11] X. Lou, J. Han, W. Chu, X. Wang, Q. Cheng, Mater. Sci. Eng, B 2007, 137, 268. [12] Y. Chen, L. Hu, M. Wang, Y. Min, Y. Zhang, Colloids Surf, A 2009, 336, 64. [13] F. Jiao, H. Frei, Angew. Chem. Int. Ed. 2009, 48, 1841. [14] S. Ikeda, T. Takata, T. Kondo, G. Hitoki, M. Hara, J. N. Kondo, K. Domen, H. Hosono, H. Kawazoe, A. Tanaka, Chem. Commun. 1998, 2185. [15] S. Ikeda, T. Takata, M. Komoda, M. Hara, J. N. Kondo, K. Domen, A. Tanaka, H. Hosono, H. Kawazoe, Phys. Chem. Chem. Phys. 1999, 1, 4485. [16] D. Barreca, P. Fornasiero, A. Gasparotto, V. Gombac, C. Maccato, T. Montini, E. Tondello, ChemSusChem 2009, 2, 230. [17] D. Barreca, A. Gasparotto, O. I. Lebedev, C. Maccato, A. Pozza, E. Tondello, S. Turner, G. Van Tendeloo, CrystEngComm 2010, 12, 2185. [18] R. A. Laudise, in: Treatise on Solid State Chemistry (Ed. N. B. Hannay) Plenum Press, New York-London 1975, vol. 5, p. 413. [19] M. M. Natile, A. Glisenti, Chem. Mater. 2002, 14, 3090. [20] G. Bandoli, D. Barreca, A. Gasparotto, C. Maccato, R. Seraglia, E. Tondello, A. Devi, R. A. Fischer, M. Winter, Inorg. Chem. 2009, 48, 82.
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Vap. Deposition 2010, 16, 296–300
