Attachment of Protoporphyrin Dyes to ... - ACS Publications

ARTICLE pubs.acs.org/JPCC

Attachment of Protoporphyrin Dyes to Nanostructured ZnO Surfaces: Characterization by Near Edge X-ray Absorption Fine Structure Spectroscopy Ruben Gonzalez-Moreno,†,‡ Peter L. Cook,§ Ioannis Zegkinoglou,§,|| Xiaosong Liu,|| Phillip S. Johnson,§ Wanli Yang,|| Rose E. Ruther,^ Robert J. Hamers,^ Ramon Tena-Zaera,# F. J. Himpsel,§ J. Enrique Ortega,†,z,∞ and Celia Rogero*,† †

Centro de Física de Materiales (CSIC-UPV/EHU), Material Physics Center (MCP), 20018, San Sebastian, Spain Instituto de Ciencia de Materiales de Madrid (ICMM/CSIC), 28049, Madrid, Spain § Department of Physics, University of Wisconsin, Madison, Wisconsin 53706-1390, United States Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States ^ Department of Chemistry, University of Wisconsin, Madison, Wisconsin 53706-1390, United States # Energy Department, CIDETEC-IK4, 20009, San Sebastian, Spain z Donostia International Physics Center (DIPC), 20018, San Sebastian, Spain ∞ Universidad del Pais Vasco, Departamento de Fisica Aplicada I, 20018, San Sebastian, Spain

)

‡

ABSTRACT: The attachment of H2- and metal (Co- and Zn-) protoporphyrin IX molecules to ZnO nanorods and single-crystal surfaces is investigated by Near Edge X-ray Absorption Fine Structure (NEXAFS) spectroscopy. The carboxyl groups of the protoporphyrin are found to be essential for anchoring the molecules to ZnO surfaces. The crystallographic orientation of the exposed ZnO face has an influence on the dye immobilization, with the highest uptake observed for the oxygen-terminated ZnO (000-1) surface. The preparation conditions are crucial for the dye immobilization. Under certain preparation conditions, there is a Zn atom exchange between the H2-protoporphyrin and the ZnO surface, i.e., a metalation of H2protoporphyrin IX to form Zn-protoporphyrin. Moreover, in the presence of chenodeoxycholic acid as coabsorber, the ZnO single-crystal surfaces are etched, as indicated by the loss of the orientation-dependent spectral features. These results help to pinpoint the chemical reactions that are responsible for the poor efficiency of ZnO-based dyesensitized solar cells, especially those built from ZnO nanorod arrays.

’ INTRODUCTION Dye-sensitized solar cells (DSSCs) are becoming serious contenders for low-cost photovoltaics. Their basic building blocks are a metal oxide as an electron acceptor, a dye as a light sensitizer, and a redox electrolyte as an electron donor. Sunlight creates excitons in the dye sensitizer which are separated into electrons and holes by transfer to the acceptor and donor, respectively. Since the pioneering report by O’Regan and Gr€atzel in 1991,1 very impressive results have been obtained, reaching an energy conversion efficiency up to 11%.2 However, further improvements in both efficiency and stability by introducing new materials and engineering their interfaces are required to make DSSCs commercially viable. For improving the efficiency, it is necessary to find the optimal combination of energy levels in the dye, acceptor, and donor. This depends not only on the choice of the materials but also on their interactions. For example, the anchoring of the dye molecules to the metal oxide is crucial for obtaining high coverage and efficient charge transfer. The oxidation potential of the excited dye needs to be sufficiently r 2011 American Chemical Society

negative for efficient electron injection into the conduction band of the acceptor. In other words, the LUMO of the dye molecule has to lie sufficiently far above the conduction band minimum of the metal oxide. If the LUMO lies too low, the output current decreases, and if it lies too high, the output voltage is reduced. Likewise, the HOMO of the dye has to lie sufficiently far below the valence band maximum of the donor, i.e., the redox potential of the electrolyte. The best performance has been achieved with a randomly packed TiO2 nanoparticle network sensitized with rutheniumbased organic dyes.1,2 This design achieves high quantum efficiency due to the efficient electron injection into the energymatched conduction band of TiO2. At the same time, electron hole recombination is avoided by rapid filling of the hole via the electrolyte. However, this is accompanied by a substantial voltage drop. Furthermore, the use of the rare metal, Ru, in the best dyes Received: April 18, 2011 Revised: July 18, 2011 Published: July 22, 2011 18195

dx.doi.org/10.1021/jp203590p | J. Phys. Chem. C 2011, 115, 18195–18201

The Journal of Physical Chemistry C makes such DSSCs inappropriate for large-scale applications, which is opposite to the motivation of using DSSCs as inexpensive alternatives to standard solar cells based on semiconductors. Thus, ruthenium-free light sensitizers have attracted great interest because of their modest cost.3 12 Among them, porphyrins are especially attractive.13 In fact, Gr€atzel's group reported recently a DSSC with 11% efficiency using a YD-2 (a porphyrin dye) as a light sensitizer.12 In the context of finding new materials, ZnO has been a very popular alternative to TiO2 electron acceptor material. Both have a very similar position of the conduction band minimum and a similar band gap. An appealing property of ZnO is its anisotropy, which allows the growth of single-crystal nanorods that are oriented along the c-axis and exhibit well-defined (10-10) side faces.14 19 Such nanorod arrays exhibit ideal architecture, providing a direct pathway for the electrons, optimizing the conductivity along a single-crystal wire, and minimizing the charge carrier path. However, DSSCs based on ZnO nanoparticles show relatively low efficiencies (98.0%), 2 M KCl (>99.5%) ultrapure aqueous solution, saturated with bubbling oxygen. The charge density was 5 C/cm2. Further details are given elsewhere.19,24b The nanorod samples used in the present study present well-defined straight columnar structures perpendicular to the substrate that indicate high crystallinity. Scanning Electron Microscopy (SEM) micrographs (Figure 1) show the side cross section and top views of the ZnO nanorod array samples. The mean nanorod diameter and length are around 200 and 700 nm, respectively. The observed rods exhibit a clear hexagonal columnar structure morphology that, together with X-ray diffraction patterns, pointed out that nanorods grow with the c-axis mainly perpendicular to the substrate and the (10-10) faces on the side. Sensitization with Dyes. We used four different protoporphyrin IX molecules, purchased by Sigma Aldrich: Protoporphyrin IX (H2PPIX), Protoporphyrin IX zinc(II) (ZnPPIX), Protoporphyrin IX cobalt chloride (Cl-CoPPIX), and Protoporphyrin IX dimethyl ester (H2PPIXester). Molecules were used as purchased, without any further purification. Sensitization of the ZnO surfaces was performed using two slightly different procedures: Procedure 1 for protoporphyrin sensitization: 20 mL of solution consisting of 2.25 mg of protoporphyrins and 3.14 mg of chenodeoxycholic acid (CDCA) is prepared in ethanol. The ZnO samples are immersed in the solution for 20 min. Eventually the samples are thoroughly rinsed with ethanol to remove the excess of physisorbed molecules. This procedure is similar to the one used for the immobilization of other dyes, such as indoline D149.24 18196

dx.doi.org/10.1021/jp203590p |J. Phys. Chem. C 2011, 115, 18195–18201

The Journal of Physical Chemistry C

ARTICLE

determining the orientation of the molecules, the angle of incidence was varied from normal incidence (s-polarization) to 60 from normal (predominantly p-polarization).

’ RESULTS

Figure 2. N 1s absorption edge of Cl Co protoporphyrin (Cl-CoPPIX), free-protoporphyrin (H2PPIX), and H2 protoporphyrin dimethyl ester (esterH2PPIX), attached to ZnO nanorods using Procedure 1 (full lines) compared to the molecular powders (dotted lines, demagnified). H2 protoporphyrin dimethyl ester (esterH2PPIX) on ZnO is almost undetectable. This demonstrates the need of carboxylic groups for attaching molecules to ZnO efficiently.

Procedure 2 for protoporphyrin sensitization: 20 mL of solution of 2.25 mg of PPIXs in ethanol. The ZnO samples are immersed in the solution for 90 min. During the immersion, the solution is sonicated to prevent the formation of aggregates. After the sensitization, the samples are rinsed with ethanol to remove the physisorbed molecules. Near Edge X-ray Absorption Fine Structure Spectroscopy (NEXAFS). NEXAFS spectra were measured with total electron yield (TEY) detection at two different beamlines, i.e., the VLSPGM beamline at the Synchrotron Radiation Center (SRC) in Madison and Beamline 8.0 at the Advanced Light Source in Berkeley. This technique has been used previously to measure unoccupied orbitals of porphyrin dyes.30a c The photon energy was calibrated at the Zn 2p edge using the binding energy of the 2p3/2 level of freshly scraped zinc metal at 1021.8 eV and at near the N 1s edge using the sharp 2p-to-3d transition in TiO2 (rutile) at 458.0 eV as the secondary standard. This value was established by measuring TiO2 powder side-by-side with gas-phase N2 trapped inside an irradiated imide.30d The first vibrational line of N2 at 400.9 eV31 served as the primary standard. All spectra are normalized to the incident photon flux. To remove the effects of beam fluctuations and decay, the sample current is divided by the current from a mesh coated in situ with Au. After this division, a linear background is subtracted using an extrapolation of the pre-edge signal. This normalization produces a signal proportional to the density of N atoms. The dipole matrix element associated with the N 1s to 2p π* transitions produces a cos2(θ) intensity distribution with respect to the angle θ between the polarization vector and the direction of the 2p valence orbital. Thus, the π* resonance intensity is maximized if the polarization vector of the light is perpendicular to the plane of the aromatic rings in the porphyrin. For

Attachment of Dye Molecules to ZnO Nanorods. One of the important factors in the design of DSSCs is the selection of the anchoring group for the dye molecules to the ZnO surface. This determines the dye coverage and can be used to enhance the electron injection efficiency into the ZnO. Carboxylic or phosphoric acids were found to be the best anchoring groups on TiO2.32 35 To test whether carboxylic acid is also favorable on ZnO surfaces, we have immobilized three slightly different protoporphyrin IX (PPIX) molecules on the ZnO nanorod: H2PPIX and Cl-CoPPIX, each containing two propionic acid groups, and H2PPIXester, where the two carboxylic acid groups are replaced by two ester groups. The N 1s NEXAFS spectra of such dye-coated ZnO surfaces are used to determine the amount of attached dye and its chemical bonding to the surface. Since N atoms occur only in the dye, this is a highly specific measurement. Figure 2 shows the N 1s spectra of ZnO nanorod surfaces obtained after dye sensitization with H2PPIX, Cl-CoPPIX, and H2PPIXester by using Procedure 1 (see the Experimental Details). The three solid-line spectra in Figure 2 correspond to the three molecules after the dye sensitization normalized to the spectra from the clean ZnO nanorod surface (dotted lines correspond to the bulk powder N 1s NEXAFS spectra). For H2PPIX and Cl-CoPPIX, one can see strong π* peaks of the dye molecules below 400 eV, whereas for the H2PPIXester there is hardly any significant signal. This striking difference demonstrates that a successful dye sensitization can be achieved with the H2PPIX and Cl-CoPPIX but not for the H2PPIX dimethyl ester. Since the only difference between the H2PPIX and H2PPIXester is the substitution of the two carboxylic acid groups by two dimethyl ester groups, the NEXAFS results demonstrate that the carboxylic acid groups are essential for anchoring the dye molecules. The carboxylic groups can form either an ester-like linkage (CdO) or a carboxylate linkage (COO ): either only the deprotonated oxygen is bonded to the surface while the other remains in its CdO configuration (ester-like linkage) or the two O ions become equivalent and are both bonded to the surface (carboxylate linkage). The former favors the efficiency of the electron injection.36 For molecules such as N3 or N719, it has been demonstrated that carboxylate occurs,34,35,37 although the ester linkage can be induced, improving the efficiency of the cell.36 For metal protoporphyrin IX molecules, it has been shown that the linkage to TiO2 rutile surfaces is mediated by the deprotonation of the carboxylic acid and the formation of the carboxylate group38,39 which is also the behavior on metal surfaces.40 Unfortunately, because of the presence of O atoms in ZnO, the analysis of the NEXAFS O 1s edge provides little information about the linkage (XPS measurements were also taken, but again the oxygen atoms in the ZnO hide the contributions coming from the molecule). The spectra for adsorbed molecules in Figure 2 are very similar to the analogous spectra for molecular bulk powders (dotted lines). The main difference is a rapidly increasing background at higher photon energies which is likely associated with the uncertainties in the background subtraction caused by the low signal-to-background ratio for a monolayer of molecules on ZnO. The π* peaks are virtually identical, indicating that the 18197


The Journal of Physical Chemistry C

Figure 3. N 1s absorption edge of H2 , Zn , and Cl Co Protoporphyrin IX attached to ZnO nanorods using Procedure 2 (full lines) compared to the molecular bulk powders (dotted lines). Procedure 2 causes a strong change in the spectrum of the H2 Protoporphyrin IX after attachment to ZnO. This can be explained by a displacement of H2 by a Zn atom from the ZnO surface, i.e., a metalation (see the vertical lines). The opposite reaction, i.e., a removal of the Zn atom from the ZnO surface, can be inferred from two weak additional features in the spectrum of attached Zn Protoporphyrin.

macrocycle is not involved in the dye-to-surface bond when using Procedure 1 for the sensitization. However, when changing the sensitization procedure, we have been able to detect substantial changes in the NEXAFS spectra of the molecules during adsorption, as shown in Figure 3. In this case, Procedure 2 was used for the attachment of three different protoporphyrin molecules, i.e., H2PPIX, ZnPPIX, and Cl-CoPPIX. As in Figure 2, the full lines are for molecules adsorbed on ZnO nanorods and the dotted lines for molecular bulk powders. The largest difference in the spectra between adsorbed molecules and the molecular powder is observed for H2PPIX, attached using Procedure 2. The π* peaks near 400 eV are shifted dramatically, indicating a major chemical change in the H2PPIX molecules during adsorption. Upon closer inspection, one finds that the spectrum has become similar to the spectra of the other two protoporphyrins, which contain metal atoms (see the vertical lines). Therefore, we conclude that H2PPIX acquires Zn atoms from the ZnO surface and thereby becomes metalated. Such an effect has recently been found on Cu single-crystal surfaces where H2PPIX becomes metalated by Cu surface atoms at room temperature.40 On ZnO only a small fraction of the H2PPIX molecules remain intact, while the majority becomes metalated. For the ZnPPIX dye, the overall shape of the spectra is similar after adsorption. Nevertheless, some significant differences can be detected at 397.5 and 399.7 eV, where two new π* features appear. Their positions coincide with the π* peaks of the metalfree protoporphyrin. That indicates that some of the ZnPPIX molecules in direct contact with the oxide surface lose their metal atom. Thus, we have the inverse of the metalation observed in ZnPPIX. Such a reaction has been found in previous work on TiO2,39 where a combination of XPS and NEXAFS was used to

ARTICLE

Figure 4. Polarization-dependent N 1s absorption edges of H2 Protoporphyrin IX on the three low-index ZnO single-crystal surfaces and for ZnO nanorods. Red lines are for a polarization parallel to the surface and blue lines for mixed polarization. The (nominally) oxygenterminated ZnO(000-1) surface exhibits the strongest polarization dependence and the highest coverage. The inset illustrates the configuration of the molecule standing upright on the ZnO (000-1) surface.

show that a fraction of the ZnPPIX molecules adsorbed on the (110) surface of rutile TiO2 lose their central metal atom. For Cl-CoPPIX, we do not observe any significant changes. Both the peak positions and their relative intensities remain the same after adsorption by Procedure 2. In this molecule, the Co metal core is protected by a Cl atom to prevent the oxidation. This protection can inhibit also the interaction of the metal core with the substrate and prevent the loss of the metal core observed for the ZnPPIX molecules. On TiO2, it has been established that the chenodeoxycholic acid coadsorbed with the dye, preventing the formation of molecular aggregates by H bonds.41 In fact, this coadsorption has been reported to improve the photocurrent and the photovoltage of the DSSCs. By complementary XPS experiments, we have determined that there is also coadsorption for the ZnO surfaces. Comparing the relative intensities of C 1s and N 1s core levels on the solutions of both procedures and after immobilization on the ZnO surface, we determine that there is a fraction of CDCA coadsorbed together with the H2PPIX molecules. The presence of the CDCA could inhibit the reactivity of the central core with the Zn ions, preventing the metalation observed in Procedure 2. Thus, using this sensitization method, the molecules are highly modified when interacting with the substrate. It is important to have control of all of these changes since they affect the dye properties and therefore the properties for the DSSC applications. Attachment to ZnO Single-Crystal Surfaces. The formation of well-organized molecular layers on the metal oxide surface is 18198


The Journal of Physical Chemistry C important for the efficiency of DSSCs. Disordered layers or molecular aggregates inhibit the electron transfer from the molecular LUMO into the metal oxide conduction band. Also, a higher density of sensitizer molecules improves the harvesting of light. Furthermore, a solid connection of the molecules to the electron acceptor improves charge injection. For protoporphyrin molecules, the ideal situation would be the formation of organized layers where the macrocycles are parallel to each other and perpendicular to the oxide surface, anchored only by the two carboxylic groups. If the molecules lie flat on the surface, the coverage would be reduced and consequently the light absorption. To investigate the organization of these molecules on the surface, we have performed polarization-dependent NEXAFS measurements of H2PPIX attached using Procedure 1 to the three low-index ZnO single-crystal surfaces, i.e., ZnO(10-10), ZnO(0001), and ZnO(000-1). The ZnO(10-10) surface is nonpolar and thus has the lowest surface energy. It forms the side faces of the crystalline ZnO nanorods which dominate the surface area. The polar ZnO(0001) and ZnO(000-1) surfaces are unstable and reconstruct into complex structures to reduce their electrostatic energy.42 45 Nominally, the ZnO(0001) surface is Zn-terminated, and ZnO(000-1) is O-terminated; however, in view of their complex reconstructions, this terminology has little meaning. Figure 4 shows polarization-dependent N 1s spectra for the three low-index ZnO surfaces and for a ZnO nanorod substrate. Procedure 1 was used for the dye immobilization in all four cases since, in this way, only the carboxylic groups chemically interact with the surfaces. Red lines correspond to normal incidence (with the electric field vector parallel to the surface) and blue lines to 60 from normal (where a substantial component of the electric field vector is perpendicular to the surface). The black dotted spectrum at the bottom of the figure corresponds to a molecular powder. On the polar ZnO(000-1) surface, the spectra are clearly polarization dependent. When going from normal incidence to 60, the π* peaks below 404 eV become less intense relative to the σ* resonance at 407 eV. Since the dipole matrix element associated with π* excitations has its maximum for the electric field vector perpendicular to the plane of the central tetrapyrrole ring of H2PPIX, this observation shows that the molecules are oriented nearly perpendicular to the surface, as indicated by the inset. This orientation is favorable for high efficiency solar cells, as discussed above. For ZnO(0001), the other polar face, the polarization dependence is much smaller but still detectable and in the same direction as for ZnO(000-1). This indicates partial orientation of the molecules with a small preference toward being upright. The position of the π* peaks does not indicate any metalation, which is consistent with the conclusion that the molecules tend to be upright and therefore do not touch the surface at their center. The nonpolar ZnO(10-10) face exhibits weak polarization dependence with a tendency toward upright orientation, similar to ZnO(0001). For the ZnO nanorod sample, we hardly observe any polarization dependence. This is consistent with their structure, which combines side walls consisting of ZnO(10-10) faces and a top surface dominated, in principle, by ZnO(0001).19 From the intensity of the molecular π* peaks at the N 1s edge relative to the ZnO background, it is possible to infer the relative coverage at different surfaces. The surface with the highest density of molecules is ZnO(000-1), which also has the strongest

ARTICLE

Figure 5. Polarization-dependent Zn 2p absorption edges for H2 Protoporphyrin IX attached to (A) single-crystal ZnO(1010) and (B) ZnO nanorods. Procedure 1 was used in (A), and Procedure 2 in (B). Solid lines correspond to the bare surfaces and dotted lines to the dye-sensitized surfaces. Red lines are for a polarization parallel to the surface (normal incidence) and blue lines for mixed polarization.

polarization dependence. Such an increase of the molecular orientation is often observed with self-assembled monolayers when the coverage approaches its maximum value. This effect can be rationalized as a crowding effect, where the interaction between the molecules increases when their distance approaches the van der Waals radius. These results help explain the low efficiencies observed for DSSCs using ZnO nanorods instead of randomly oriented nanoparticle networks.24 Since the indoline molecules use the same immobilization procedure and follow the same adsorption mechanism, the results presented here can be extrapolated to explain the low efficiency of nanorods versus nanowires found in ref 24. The majority of the surface area of ZnO nanorods consists of the nonpolar, nonreactive ZnO(10-10) side faces which are less attractive for anchoring perpendicularly oriented dye molecules. Consequently, it is critical to find an efficient attachment chemistry for the ZnO(10-10) surface to take advantage of the ideal topology of crystalline ZnO nanorods. As we discuss next, there is another explanation for the nonpolarization dependence and for the low coverage of the ZnO surfaces, which is the etching of the surfaces induced by the acid solution using Procedure 1. Etching Reaction at the Interface. Another potential hurdle for ZnO as acceptor material for DSSCs is its unstable character and its tendency to dissolve in acidic solutions.26,28,46 50 For example, when ZnO is sensitized with N3, it has been shown that N3/Zn2+ aggregates form and are adsorbed onto the ZnO surface, which reduces the charge-collection efficiency and the lightharvesting efficiency of the cell.46 Even with only two COOH groups per molecule, it has been recently demonstrated that the ZnO nanoflower structures can be etched when N719 is immobilized and that the longer the time, the higher the etching.26 To obtain better insight into this problem, we have used NEXAFS at the Zn 2p edge with ZnO single crystals and nanorods. 18199


The Journal of Physical Chemistry C Since ZnO grows in the wurtzite crystal structure, there is a difference of the NEXAFS spectra depending on whether the electric field, E, is parallel or perpendicular to the c-axis. Figure 5 illustrates this effect. On the left panel (Figure 5(A)), the red and blue solid lines correspond to the measurements of Zn 2p edge on the single crystal at both polarizations with respect to the caxis: the red spectrum corresponds to the light polarization parallel to the c-axis of the crystal, i.e., perpendicular to the ZnO(10-10) face, and the blue spectrum corresponds to the light polarization vector being perpendicular to the c-axis. The colors in the right panel (Figure 5(B)) are opposite since the orientation of the c-axis is rotated by 90. The most significant difference in both cases is the pronounced peak at about 1032 eV that appears for light polarization parallel to the c-axis of the crystals and the small shoulder at around 1034 eV detected with the light polarization perpendicular to the c-axis. Detrimental etching reactions are detected in NEXAFS by a loss in this polarization dependence of the spectra which is taken as a sign of reduced crystallinity in the surface region (typically 5 10 nm with total electron yield detection). Protoporphyrin IX dyes, as happens with the N719, have only two carboxylic acid groups and could be considered as a gentle sensitizer.34 However, in Procedure 1 there is an extra source of carboxylic acid groups since the solution contains chenodeoxycholic acid,24,51 which also absorbs on the surface. In Figure 5(A), the dotted lines correspond to the two spectra measured at both polarizations after the sensitization using Procedure 1. In the spectra measured at normal incidence (c ^ E), the peak at 1032 eV completely disappears, and both spectra become equivalent, indicating the loss in the surface crystallinity. However, when the molecules are immobilized using Procedure 2 (dotted spectra in Figure 5(B)), there is no damage of the surface. The spectra measured before and after the sensitization (solid and dotted lines in Figure 5(B), respectively) exhibit exactly the same line shape, and the polarization dependence is still visible. Even though the sensitization time in Procedure 2 was longer than in Procedure 1 (90 min for the former against 20 min for Procedure 1), the etching of the surface is almost undetectable.

’ CONCLUSIONS In summary, element-specific spectroscopy is used to investigate the chemistry and electronic structure at the interface between dye molecules and the ZnO acceptor electrode for dye-sensitized solar cells. For ZnO single-crystal substrates, the uptake of protoporphyrin dyes depends strongly on the crystallographic orientation, with the largest coverage observed at the nominally oxygen-terminated ZnO(000-1) surface. This has an impact on the dye coverage on ZnO nanorods which are desirable for devices because of their crystallinity. They are oriented in the [0001] direction and thus exhibit mainly the nonpolar ZnO(10-10) surfaces where the uptake is small. This finding calls for the development of new attachment methods that are optimized for ZnO(10-10). The interface chemistry depends on the sensitizer solution since ZnO can be etched. In that case, the orientation dependence of the interface chemistry is lost. Moreover, under certain preparation conditions (using an ethanol solution with the dye molecules), an interface reaction takes place where the hydrogen in H2 Protoporphyrin is substituted by a Zn atom from the substrate. Such interface reactions call for careful control of the solution chemistry to immobilize dye molecules on ZnO (work in progess). It might be advantageous to use gas-phase chemistry, such as atomic layer deposition (ALD).

ARTICLE

’ AUTHOR INFORMATION Corresponding Author

*Tel.: (+34) 943015804. Fax: (+34) 943015800. E-mail: [email protected].

’ ACKNOWLEDGMENT We acknowledge funding through Spanish research projects PET2008-109, BIO 2007-67523, intramural project 200960I159, and HOPE CSD2007-0007 (Consolider-Ingenio 2010). This work was supported by the Spanish MICINN (PIB2010US 00652, MAT2010 21156 C03 01, and MAT2010 21156 C03 03) and the Basque Government (IT 257 07). This work was supported in the U.S. by the NSF under the awards CHE-1026245 and DMR-0537588 (SRC) and by the DOE under the contracts DEFG02-01ER45917 (end station), DEAC03-76SF00098 (ALS), and ZnO Surface Functionalization was supported by DOE Basic Energy Sciences Grant DE-FG02-09ER16122 (RJH and RER). Dr. Doug Taube at the ALS is gratefully acknowledged for his help in the dye-solution preparation. R.T.-Z. acknowledges the support of the Program “Ramon y Cajal” of the MICINN. ’ REFERENCES (1) O’Regan, B.; Gr€atzel, M. Nature 1991, 353, 737. (2) (a) Gr€atzel, M. J. Photochem. Photobiol. A: Chem. 2004, 164, 3. (b) Nazeeruddin, M. K.; De Angelis, F.; Fantacci, S.; Selloni, A.; Viscardi, G.; Liska, P.; Ito, S.; Takeru, B.; Gr€atzel, M. J. Am. Chem. Soc. 2005, 127, 16835. (c) Chiba Y.; Islam A.; Watanabe Y.; Komiya R.; Koide N.; Han L. Jpn. J. Appl. Phys., Part 2, 45, 2006, L638. (d) Gao, F; Wang, Y.; Zhang, J.; Shi, D.; Wang, M.; Humphry-Baker, R.; Wang, P.; Zakeeruddin, S. M.; Gr€atzel, M. Chem. Commun. 2008, 23, 2635. (3) (a) Horiuchi, T.; Miura, H.; Uchida, S. Chem. Commun. 2003, 3036. (b) Horiuchi, T.; Miura, H.; Sumioka, K.; Uchida, S. J. Am. Chem. Soc. 2004, 126, 12218. (c) Ito, S.; Zakeeruddin, S. M.; Humphry-Baker, R.; Liska, P.; Charvet, R.; Comte, P.; Nazeeruddin, M. K.; Pchy, P.; Takata, M.; Miura, H.; Uchida, S.; Gr€atzel, M. Adv. Mater. 2006, 18, 1202. (4) (a) Kitamura, T.; Ikeda, M.; Shigaki, K.; Inoue, T.; Anderson, N. A.; Ai, X.; Lian, T.; Yanagida, S. Chem. Mater. 2004, 16, 1806. (b) Hara, K.; Sato, T.; Katoh, R.; Furabe, A.; Yoshihara, T.; Murai, M.; Kurashige, M.; Ito, S.; Shinpo, A.; Suga, S.; Arakawa, H. Adv. Funct. Mater. 2005, 15, 246. (5) (a) Kim, S.; Choi, H.; Kim, D.; Song, K.; Kang, S. O.; Ko, J. Tetrahedron 2007, 63, 9206. (b) Kim, S.; Choi, H.; Baik, C.; Song, K.; Kang, S. O.; Ko, J. Tetrahedron 2007, 63, 11436. (c) Jung, I.; Lee, J. K.; Song, K. H.; Song, K.; Kang, S. O.; Ko, J. J. Org. Chem. 2007, 72, 3652. (6) (a) Velusamy, M.; Thomas, K. R. J.; Lin, J. T.; Hsu, Y.; Ho, K. Org. Lett. 2005, 7, 1899. (b) Hagberg, D. P.; Edvinsson, T.; Marinado, T.; Boschloo, G.; Hagfeldt, A.; Sun, L. Chem. Commun. 2006, 2245. (c) Liang, M.; Xu, W.; Cai, F.; Chen, P.; Peng, B.; Chen, J.; Li, Z. J. Phys. Chem. C 2007, 111, 4465. (7) (a) Ferrere, S.; Zaban, A.; Greg, B. A. J. Phys. Chem. B 1997, 101, 4490. (b) Ferrere, S.; Greg, B. A. New J. Chem. 2002, 26, 1155. (c) Shibano, Y.; Umeyama, T.; Matano, Y.; Imahori, H. Org. Lett. 2007, 9, 1971. (8) (a) Ehret, A.; Stuhl, L.; Spitler, M. T. J. Phys. Chem. B 2001, 105, 9960. (b) Ushiroda, S.; Ruzycki, N.; Lu, Y.; Spitler, M. T.; Parkinson, B. A. J. Am. Chem. Soc. 2005, 127, 5158. (9) Tatay, S.; Haque, S. A.; O’Regan, B.; Durrant, J. R.; Verhees, W. J. H.; Kroon, J. M.; Vidal-Ferran, A.; Gavia, P.; Palomares, E. J. Mater. Chem. 2007, 17, 3037. (10) (a) Yao, Q.-H.; Shan, L.; Li, F.-Y.; Yin, D.-D.; Huang, C.-H. New J. Chem. 2003, 27, 1277. (b) Chen, Y.-S.; Li, C.; Zeng, Z.-H.; Wang, W.-B.; Wang, X.-S.; Zhang, B.-W. J. Mater. Chem. 2005, 15, 1654. (11) (a) Hara, K.; Sato, T.; Katoh, R.; Furube, A.; Ohga, Y.; Shinpo, A.; Suga, S.; Sayama, K.; Sugihaa, H.; Arakawa, H. J. Phys. Chem. B 2003, 18200


The Journal of Physical Chemistry C 107, 597. (b) Morandeira, A.; Boschloo, G.; Hagfeldt, A.; Hammarstrm, L. J. Phys. Chem. B 2005, 109, 19403. (c) Koumura, N.; Wang, Z.-S.; Mori, S.; Miyashita, M.; Suzuki, E.; Hara, K. J. Am. Chem. Soc. 2006, 128, 14256. (12) Bessho, T.; Zakeeruddin, S. M.; Yeh, C. Y.; Wei-Guang Diau, E.; Gr€atzel, M. Angew. Chem., Int. Ed. 2010, 49, 6646. (13) Campbell, W. M.; Burrell, A. K.; Officer, D. L.; Jolley, K. W. Coord. Chem. Rev. 2004, 248, 1363. (14) Vayssieres, L Adv. Mater. 2003, 15, 464. (15) (a) Baxter, J. B.; Walker, A. M.; van Ommering, K.; Aydil, E. S. Nanotechnology 2006, 17, S304. (b) Baxter, J. B.; Aydil, E. S. Appl. Phys. Lett. 2005, 86, 053114. (16) Leschkies, K. S.; Divakar, R.; Basu, J.; Enache-Pommer, E.; Boercker, J. E.; Carter, C. B.; Kortshagen, U. R.; Norris, D. J.; Aydil, E. S. Nano Lett. 2007, 7, 1793. (17) Olson, D. C.; Lee, Y.-J.; White, M. S.; Kopidakis, N.; Shaheen, S. E.; Ginley, D. S.; Voigt, J. A.; Hsu, J. W. P. J. Phys. Chem. C 2007, 111, 16640. (18) Uhlrich, J. J.; Franking, R.; Hamers, R. J.; Kuech, T. F. J. Phys. Chem. C 2009, 113, 21147. (19) Tena-Zaera, R.; Elias, J.; Levy-Clement, C.; Mora-Sero, I.; Luo, Y.; Bisquert, J. Phys. Status Solidi B 2008, 205, 2345–2350. (20) Zhang, Q. F.; Chou, T. P.; Russo, B.; Jenekhe, S. A.; Cao, G. Z. Angew. Chem., Int. Ed. 2008, 47, 2402. (21) Yoshida, T.; Zhang, J.; Komatsu, D.; Sawatani, S.; Minoura, H.; Pauporte, T.; Lincot, D.; Oekermann, T.; Schlettwein, D.; Tada, H.; Wohrle, D.; Funabiki, K.; Matsui, M.; Miura, H.; Yanagi, H. Adv. Funct. Mater. 2009, 19, 17. (22) Xu, F.; Dai, M.; Lu, Y.; Sun, L. J. Phys. Chem. C 2010, 114, 2776. (23) Saito, M.; Fujihara, S. Energy Environ. Sci. 2008, 1, 280. (24) (a) Guillen, E.; Idigoras, J.; Berger, T.; Anta, J. A.; FernandezLorenzo, C.; Alcantara Navas, R. J.; Martin-Calleja, J. Phys. Chem. Chem. Phys. 2011, 13, 207. (b) Guillen, E.; Azaceta, E.; Peter, L. M.; Zukal, A.; Tena-Zaera, R.; Anta, J. A. Energy Environ. Sci. 2011, 4, 245. (25) Law, M.; Greene, L. E.; Johnson, J. C.; Saykally, R.; Yang, P. Nat. Mater. 2005, 4, 455. (26) Yan, F.; Huang, L.; Zheng, J.; Huang, J.; Lin, Z.; Huang, F.; Wei, M. Langmuir 2010, 26, 7153. (27) Guillen, E.; Casanueva, F.; Anta, J. A.; Vega-Poot, A.; Oskam, G.; Alcantara, R.; Fernandez-Lorenzo, C.; Martin-Calleja, J. J. Photochem. Photobiol., A 2008, 200, 364. (28) Keis, K.; Lindgren, J.; Lindquist, S. E.; Hagfeldt, A. Langmuir 2000, 16, 4688. (29) Peulon, S.; Lincot, D. Adv. Mater. 1996, 8, 166. (30) (a) Cook, P. L.; Liu, X.; Yang, W.; Himpsel, F. J.; J. Chem. Phys. 2009, 131, 194701. (b) García-Lastra, J. M.; Cook, P. L.; Himpsel, F. J.; Rubio, A.; J. Chem. Phys. 2010, 133, 151103. (c) Cook, P. L.; Yang, W.; Liu, X.; García-Lastra, J. M.; Rubio, A.; Himpsel, F. J.; J. Chem. Phys. 2011, 134, 204707. (d) Johnson, P. S.; Cook, P. L.; Liu, X.; Yang, W.; Bai, Y.; Abbott, N. L.; Himpsel, F. J. J. Chem. Phys. 2011, 135, 044702. (31) Sodhi, R. N. S.; Brion, C. E. J. Electron Spectrosc. Relat. Phenom. 1984, 34, 363. (32) Chen, G.; Zheng, K.; Mo, X.; Sun, D.; Meng, Q.; Chen., G. Mater. Lett. 2010, 64, 1336. (33) Wegner, G.; Baum, P.; Muller, M.; Norwig, J.; Landfester, K. Macromol. Symp. 2001, 175, 349. (34) Thavasi, V.; Renugopalakrishnan, V.; Jose, R.; Ramakrishna, S. Mater. Sci. Eng. R 2009, 63, 81. (35) Nazeeruddin, M. K.; Humphry-Baker, R.; Liska, P.; Gr€atzel, M. J. Phys. Chem. B 2003, 107, 8981. (36) Murakoshi, K.; Kano, G.; Wada, Y.; Yanagida, S.; Miyazaki, H.; Matsumoto, M.; Murasawa, S. J. Electroanal. Chem. 1995, 396, 27–34. (37) De Angelis, F.; Fantacci, S.; Selloni, A.; Nazeeruddin, M. K.; Gr€atzel, M. J. Phys. Chem. C 2010, 114, 6054. (38) De Angelis, F.; Fantacci, S.; Sgamellotti, A.; Pizzotti, M.; Tessore, F.; Biroli, A. O. Chem. Phys. Lett. 2007, 447, 10. (39) Rienzo, A.; Mayor, L. C.; Magnano, G.; Satterley, C. J.; O’Shea, J. N.; Ataman, E.; Schnadt, J.; Schulte, K. J. Chem. Phys. 2010, 132, 084703.

ARTICLE

(40) Gonzalez-Moreno, R.; Sanchez-Sanchez, C; Trelka, M.; Otero, R.; Cossaro, A.; Verdini, A.; Floreano, L; Ruiz-Bermejo, M; Garcia-Lekue, A.; Martin-Gago, J. A; Rogero, C. J. Phys. Chem. C 2011, 115, 6849. (41) (a) Yum, J. H.; Jang, S. R.; Baker, R. H.; Gr€atzel, M.; Cid, J. J.; Torres, T.; Nazeeruddin, M. K. Langmuir 2008, 24, 5636. (b) Yum, J. H.; Walter, P.; Huber, S.; Rentsch, D.; Geiger, T.; N€uesch, F.; Angelis, F. D.; Gr€atzel, M.; Nazeeruddin, M. K. J. Am. Chem. Soc. 2007, 129, 10320. (c) Kay, A.; Gr€atzel, M. J. Phys. Chem. B 1993, 97, 6272. (d) Schlichth€orl, G.; Huang, S. Y.; Spargue, J.; Frank, A. J. J. Phys. Chem. B 1997, 101, 8141. (e) Zaban, A.; Ferrere, S.; Gregg, B. A. J. Phys. Chem. B 1998, 102, 452. (f) Kusama, H.; Arakawa, H. Sol. Energy Mater. Sol. Cells 2004, 81, 8. (42) Torbr€ugge, S.; Ostendorf, F.; Reichling, M. J. Phys. Chem. C 2009, 113, 4909. (43) Jedrecy, N.; Sauvage-Simkin, M.; Pinchaux, R. Appl. Surf. Sci. 2000, 162, 69. (44) Dulub, O.; Diebold, U.; Fresse, G. Phys. Rev. Lett. 2003, 90, 016102. (45) King, S. T.; Parihar, S. S.; Pradhan, K.; Johnson-Steigelman, H. T.; Lyman, P. F. Surf. Sci 2008, 602, L131. (46) Horiuchi, H.; Katoh, R.; Hara, K.; Yanagida, M.; Murata, S.; Arakawa, H.; Tachiya, M. J. Phys. Chem. B 2003, 107, 2570. (47) Seow, Z. L. S.; Wong, A. S. W.; Thavasi, V.; Jose, R.; Ramakrishna, S.; Ho, G. W. Nanotechnology 2009, 20, 045604. (48) Westermark, K.; Rensmo, H.; H.; Siegbahn, H.; Keis, K.; Hagfeldt, A.; Ojam€ae, L.; Persson, P. J. Phys. Chem. B 2002, 106, 10102. (49) Taratula, O.; Galoppini, E.; Mendelsohn, R. Langmuir 2009, 25, 2107. (50) Chou, T. P.; Zhang, Q. F.; Cao, G. Z. J. Phys. Chem. C 2007, 111, 18804. (51) Lee, C. W.; Lu, H. P.; Lan, C. M.; Huang, Y. L.; Liang, Y. R.; Yen, W. N.; Liu, Y. C.; Lin, Y. S.; Diau, E. W. G.; Yeh, C. Y. Chem.—Eur. J. 2009, 15, 1403.

’ NOTE ADDED AFTER ASAP PUBLICATION This manuscript was originally published on the web on August 18, 2011, with an error to the author affiliations and Acknowledgment Section. The corrected version was reposted on August 26, 2011.

18201
