Impact of Molecular Dipole Moments on Fermi Level Pinning in Thin

Article pubs.acs.org/JPCC

Impact of Molecular Dipole Moments on Fermi Level Pinning in Thin Films Stefanie Winkler,†,‡ Johannes Frisch,†,‡ Patrick Amsalem,‡ Stefan Krause,† Melanie Timpel,‡ Matthias Stolte,§ Frank Würthner,§ and Norbert Koch*,†,‡ †

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH, Albert-Einstein-Str. 15, 12489 Berlin, Germany Humboldt-Universität zu Berlin, Institut für Physik, 12489 Berlin, Germany § Institut für Organische Chemie and Center for Nanosystems Chemistry, Universität Würzburg, Am Hubland, 97074 Würzburg, Germany ‡

S Supporting Information *

ABSTRACT: A series of merocyanine dyes with a wide range of molecular dipole moments were deposited on metal oxides covering a wide work function (Φ) range (2.3 to 7.0 eV), and the energy level alignment at these interfaces was studied with photoelectron spectroscopy. We find that a preferential orientation of the merocyanines and their dipoles in the monolayer systematically lowers Φ of the oxides such that Fermi-level (EF) pinning at the highest occupied molecular level of the merocyanines only occurs for very high Φ oxides (≥6 eV). Correspondingly, pinning at the electron affinity level can readily be achieved also with moderate oxide Φ, e.g., for indium tin oxide, and electron transfer between the merocyanines to these oxides can proceed readily. Noteworthy, the EF-pinning behavior and the associated Φ values seem independent of the molecular dipole moment magnitude, most likely due to the self-limiting effect of Φ as soon as the pinning regime is reached.

■

INTRODUCTION Merocyanine dyes are molecules consisting of a donor (D) and an acceptor (A) part providing a strong dipolar character and high polarizability to these D−A molecules, enabling high absorption coefficients and tunable electronic character from polyene- to polymethine-type chromophores.1−3 These properties make merocyanines naturally attractive for application in organic photovoltaics and particularly dye-sensitized solar cells (DSSCs).4−9 Despite its importance for efficient charge collection, the energy level alignment at the D−A dye/ electrode interfaces has not yet been studied extensively,10−16 and the mechanisms (particularly including the effects due to the molecular dipoles) that determine the interface electronic properties remain to be established, which will then allow correlation with device performances.17−20 To achieve low resistance electrical contacts in organic (opto-) electronic devices, a proper matching of the electrode Fermi level (EF) to the charge transport levels of the organic semiconductor is necessary.21−23 Within the Schottky−Mott limit, i.e., assuming vacuum level alignment, one might try estimating the relative position of the energy levels at the interface simply from the values of the electrode work function (Φ) as well as the ionization energy (IE) and electron affinity (EA) of the organic semiconductor. However, for interfaces between organic semiconductors and metals, the invalidity of vacuum level alignment is established meanwhile.23,24 For chemically passivated metal substrates or inert nonmetallic electrodes, it was found that vacuum level alignment does hold as long as Φ is within the limits set by the critical © 2014 American Chemical Society

work function values for Fermi level pinning at the lowest unoccupied and highest occupied energy levels (Φpin‑ and Φpin+, respectively), which are in principle specific to every organic semiconductor but in practice defined by the sample structure and composition dependent density of states distribution near the frontier energy levels, i.e., the gap state distribution.25,26 However, if the sample structure does not vary too much, the spread of critical Φ values is less than 0.2 eV. For an electrode exceeding these limits, vacuum level alignment would place the electrode EF below the highest occupied molecular orbital (HOMO) level or above the lowest unoccupied molecular orbital (LUMO) level of the organic semiconductor, corresponding to an electronic nonequilibrium situation. To establish equilibrium, charges are transferred across the interface, giving rise to interface dipoles that shift the vacuum level; this phenomenon is commonly referred to as Fermi level pinning.22,27 The above is well established for conventional organic semiconductors, which typically have no or very small molecular dipole moments, or at least no preferential dipole orientation at interfaces or in thin films. In contrast, merocyanines have substantial molecular dipoles and their impact on Φpin‑ and Φpin+ at interfaces with electrodes is yet unexplored. Received: March 3, 2014 Revised: May 5, 2014 Published: May 6, 2014 11731

dx.doi.org/10.1021/jp5021615 | J. Phys. Chem. C 2014, 118, 11731−11737

The Journal of Physical Chemistry C

Article

sputtering (500 eV, 1 μA) and annealing at 400 °C (final annealing step in 10−6 mbar O2 for 10 min). The sputtered/ annealed MoO3 films are denoted as MoO3_cl, whereas those used without treatment after evaporated are denoted as MoO3_e. Absence of sample contaminations were confirmed by X-ray photoelectron spectroscopy (XPS) prior to evaporation of the merocyanines (for the synthesis of the molecules, see ref 28). Thin film preparation of merocyanines 1−4 was performed via sublimation from resistively heated quartz crucibles using a rate of about 0.2−2 Å/min. Nominal film mass−thickness of 1−4 (density:1.35 g/cm3), MoO3 (4.7 g/ cm3), and Sm (7.5 g/cm3) was further monitored using a quartz crystal microbalance. To prepare Sm2O3 we exposed freshly evaporated metal Sm films to 10−5 mbar oxygen for 20 min. Ultraviolet photoelectron spectroscopy (UPS) was performed using a helium gas discharge lamp (21.22 eV) and a photon energy of 35 eV (BESSY) with a very low photon flux (ca. 100 times attenuated compared to standard commercial sources) in order to minimize sample irradiation damage. XPS was performed using Al Kα radiation (1486.58 eV) and a photon energy of 620 eV (BESSY) to check for sample contamination. All spectra were recorded at room temperature and normal emission using a Specs Phoibos 100 (at HU-Berlin) and a Scienta SES 100 (at BESSY) hemispherical energy analyzer, with 120 meV energy resolution for UPS. To determine the sample work function the secondary electron cutoff spectra were measured with the sample biased at −10 V to overcome the analyzer work function.

Therefore, we investigated a series of merocyanines (see Figure 1) with various molecular dipole moment adsorbed on

■

Figure 1. Chemical structures of the four investigated merocyanine dyes 1−4. All consist of the same donor part with varying acceptor moiety. Their ground state dipole moments, as determined by electrooptical absorption spectroscopy,28 are given next to the structure.

RESULTS AND DISCUSSION Figure 2 shows the nominal film thickness (θ) dependent evolution of the work function (Φ), the HOMO-onset (Eonset) position with respect to the electrode Fermi level (EF) [i.e., hole injection barrier (HIB)], and the ionization potential (IP) when depositing 1 on Sm2O3 (Φox = 2.3 eV), ITO (Φox = 4.3 eV), TiO2 (Φox = 5.25 eV), MoO3_cl (Φox = 6.0 eV), and MoO3_e (Φox = 7.0 eV) (see the Experimental Details section above for oxide preparation conditions). 1 is discussed first because it is the most planar molecule within the studied merocyanine series and steric hindrance should therefore play only a minor role for the arrangement of the molecules with respect to each other and the substrate (see Supporting Information for all valence region spectra). Figure 2a shows that the IP of 1 for thick (∼5 nm) films is 5.50 eV on all substrates. By subtracting the optical gap of 2.25 eV,28 we provide an upper limit for the EA of 3.25 eV; the exciton binding energy, yet not determined experimentally for 1, must be taken into account for a more accurate EA value. However, it is also important to note that the orientation of the molecular dipole moment with respect to the surface normal has a significant effect on the IP of molecular films.33 Consequently, the constant IP on all substrates (see Figure 2a) indicates the same structure and molecular orientation for films with θ > 30 Å. Beyond this film thickness no preferred net dipole moment orientation persists, otherwise Φ would not be constant from there on. Thus, in thick films of 1 the molecular dipoles are either oriented parallel to the surface or the molecules arrange such that neighboring dipoles cancel each other macroscopically (i.e., antiparallel arrangement, as discussed in reference 3). In the low θ regime the deposition of 1 significantly changes Φ, but a saturation close to θ = 8 Å occurs. It is, therefore, reasonable to assume that the molecular monolayer (ML) is closed at this nominal coverage.34 Only for

five different metal oxide substrates by photoemission spectroscopy and studied the film thickness dependent energy level alignment, particularly to unravel an influence of the molecular dipoles on EF-pinning. We used metal oxide substrates covering a wide range of substrate-Φ (Φox), from 2.3 eV (Sm2O3) to 7.0 eV (MoO3), in order to reach EF-pinning at occupied and unoccupied levels. We find that the molecular dipoles lower the effective work function of the oxide substrates by ca. 1 eV because preferential orientation of the molecules in the monolayer persists. This, in turn, leads to the observation of Fermi level pinning at 1 eV higher pristine substrate Φox than expected from the ionization energy and electron affinity of the molecules alone. This molecular dipole moment induces a substantial work function change, which must be taken into account when designing interfaces for either electron or hole extraction in photovoltaic cells.

■

EXPERIMENTAL DETAILS Photoemission experiments were performed at HU-Berlin with a multitechnique ultrahigh vacuum (UHV) apparatus consisting of interconnected sample preparation (base pressure 5 × 10−10 mbar) and analysis (base pressure 1 × 10−10 mbar) chambers and at the end station SurICat (beamline PM4) at the synchrotron light source BESSY II, consisting also of interconnected sample preparation (base pressure 1 × 10−8 mbar) and analysis (base pressure 1 × 10−10 mbar) chambers. Indium-tin-oxide (ITO) covered glass substrates (sheet resistance 15−30 Ω), TiO2(100) single crystals, and MoO3 thin films were in situ cleaned via repeated cycles of Ar-ion 11732



Article

Figure 2. (a) Ionization Potential (IP), work function (Φ) and HOMO-onset position (Eonset − EF) depending on the nominal film thickness (θ) of merocyanine 1 on Sm2O3 (blue triangles down, band gap: 4.33 eV29), ITO (black squares, band gap: 4.0 eV30), TiO2(111) (green circles, band gap: 3.4 eV31), MoO3_cl (orange triangle up) and MoO3_e (wine diamonds, band gap: 3.0 eV32). (b) Corresponding energy level schemes (using the optical gap 2.25 eV28 to estimate the LUMO level of 1 and the band gaps of the oxides to estimate the CB-onset position). The critical work function values for EF-pinning are highlighted by a square/circle surrounding the corresponding numbers.

MoO3_e a fairly higher θ of ca. 30 Å is required for reaching the saturation of value of Φ (Φsat). This effect is attributed to the heterogeneity of the as-deposited MoO3_e film that leads to a less homogeneous growth of 1 and/or to additional energy level bending due to strong EF-pinning (vide infra).35 Upon monolayer formation, large interface dipoles, i.e., work function changes ΔΦ, of about 1 eV and more are observed in all cases, ruling out vacuum level alignment. For the two extreme Φox substrate cases (see Figure 2a), corresponding to MoO3_cl/MoO3_e (Φox= 6.0 eV/7.0 eV, respectively) and Sm2O3 (Φox= 2.3 eV), Φox is either higher than the IP or lower than the EA of 1. Consequently, in the case of Sm2O3 we find the LUMO level pinned at EF, which results in a Φ increase of 1.05 eV, giving a final Φsat of 3.35 eV. Additional valence spectral features due to (partial) filling of the LUMO upon charge transfer might be expected to arise close to EF. However, the density of charged molecules is typically very low and is likely here below the detection limit of our experimental setup. Likewise, the HOMO level gets EF-pinned for MoO3 substrates, resulting in the observed Φ decrease of 1.3 and 2.3 eV, respectively, giving a final Φsat of 4.7 eV. We assign the two Φsat values of 3.35 and 4.7 eV to Φpin‑ and Φpin+, which would typically be synonymous with the substrate Φox values at which pinning sets in. However, we show in the following that this is not the case for the dipolar molecules investigated here. Turning to ITO (Φox= 4.3 eV) and TiO2 (Φox= 5.25 eV) as substrates, one would expect vacuum level alignment for the 1/ ITO interface, as the work function of ITO falls into the interval given by Φpin‑ and Φpin+, which is highlighted as gray

shaded area in Figure 3a. For the 1/TiO2 interface Φox falls outside this interval and a relatively small interface dipole of ca.

Figure 3. (a) Φsat measured for thick films of 1 as a function of pristine substrate Φox. The region where critical substrate Φox values for pinning and Φsat values are expected to coincide (and where vacuum level alignment is typically observed) is shaded in gray. For 1 this region is shifted (arrow) by ca. 1 eV to higher Φox values. This is attributed to the preferential orientation of the molecular dipole moments that cause the shift ΔΦdip. (b) Effect of preferential orientation of molecular dipole moments on ΔΦdip.

0.55 eV is expected to establish the pinning work function of 4.7 eV. Clearly, the observed results differ from these expectations (see Figure 2). For 1/ITO an interface dipole decreases Φox to Φpin‑ (3.35 eV), and for 1/TiO2 Φox is decreased to 3.6 eV, which is much lower than Φpin+ (4.7 eV). 11733



Article

Figure 4. Work function (Φ) in dependence of the film coverage on ITO (black squares, band gap: 4.0 eV30) and TiO2(111)-surface (green circles, band gap: 3.4 eV31) for the merocyanine dyes (a) 1, (b) 2, (c) 3, and (d) 4. In conjunction the energy level schemes are provided, where we used the optical gap 2.25, 1.90, 2.10, and 1.75 eV28 to provide the LUMO-positions. (For more details find the evolution of IP and Eonset − EF in the Supporting Information.)

To visualize their interdependence,22,27 Φsat of 1 is plotted versus the corresponding pristine substrate Φox in Figure 3a. In contrast to common observations (gray shaded area, Figure 3a) here the saturated work function values Φsat of 1 in both pinning situations (Sm2O3 and MoO3) cannot be assigned to the critical substrate work function Φox values (Φpin+ and Φpin‑), at which pinning sets in. To elucidate this behavior, the various possible contributions to the observed interface dipole are discussed for ITO and TiO2 more explicitly in the following. One contribution may be the “push-back” effect, which was found to be ΔΦPB ≈ −0.3 eV for organic materials on oxides.36,37 Apparently, this value is much smaller than the 1 eV difference seen in Figure 3 and therefore cannot be the sole cause. In addition, we need to consider a preferential orientation of the molecular dipoles of 1 in the monolayer (ML), leading to a work function change ΔΦdip. A possible dipole moment orientation within the ML is that the electron poor (rich) part faces the vacuum side, inducing a Φ decrease (increase) [see Figure 3b, (i) and (ii)]. If all molecules are lying on the surface, or if they form dimers with antiparallel orientation of dipoles (as known to be often the case for merocyanines38) the contribution ΔΦdip of the molecular dipole moment is zero [see Figure 3b, (iii)]. It is now important to recall that the electron withdrawing carbonyl and/ or cyano groups can act as anchoring groups as they are known

to form weak coordination bonds to metal atoms.39−41 This anchoring can induce the vertical orientation of the intermolecular dipole moment in the monolayer, with the negative pole pointing toward the metal oxide surface [situation (i) in Figure 3b], which reduces Φ. To estimate the contribution of the oriented dipole moments to the total interface dipole, we use the Helmholtz equation assuming a compact layer of vertically aligned 1 molecules enDμ⊥ ΔΦdip = εrεD (eq 1) where e is the elementary charge, ε0 is the vacuum permittivity, we use a packing density nD = 2 × 1014 cm−2,17 a relative permittivity of εr = 3 of the ML, and the molecular dipole moment μ⊥= 3.8 D.28 With this we find ΔΦdip ≈ −1 eV. Taking into account both already mentioned contributions, i.e., ΔΦPB and ΔΦdip, Φox of pristine TiO2 (5.25 eV) can be reduced to 3.95 eV due to the deposition of a ML of 1. This value is in the range of−but still higher than−the measured Φsat of 3.60 eV. However, obviously a further contribution is involved, which is most likely resulting from the chemical coordination39−41 of 1 to the TiO2 surface. This leads to an overall ΔΦ of −1.65 eV, which is assumed to be equal for all investigated substrates from first approximation as long as pinning-induced charge transfer does not occur in addition. 11734



Article

Noteworthy, in the case of ITO this overall ΔΦ is too large to explain the observed −0.95 eV interface dipole. ΔΦ for ITO may well differ because of different molecular tilting (thus changes of μ⊥), less dense packing, or different chemical coordination. As in this case Φsat coincides exactly with Φpin‑, oxide-to-organic charge transfer is most likely already involved, limiting the magnitude of the interface dipole.42 Therefore, the critical work function value at which LUMOpinning sets in (Φpin‑) is at least 0.95 eV higher than the Φsat value in a LUMO-pinned situation. On the other hand, we have ruled out EF-pinning of the HOMO for Φox = 5.25 eV (TiO2), but for Φox = 6.0 eV (MoO3_cl) HOMO-pinning was observed. Therefore, the critical value, at which HOMOpinning sets in (Φpin+) is at most 1.3 eV higher than the Φsat value found for the clear HOMO-pinned situation. We can thus determine the observed delay of EF-pinning to be between 0.95 and 1.3 eV. The cause for this delayed EF-pinning is the preferential orientation of molecules on the oxide surfaces, which reduces the effective work function by ca. 1.0 eV compared to the pristine surfaces. This situation is somewhat reminiscent of molecules deposited on atomically clean metal surfaces, where the push-back effect (in the absence of chemical bonding of the molecules) lowers the effective work function “seen” by the molecules, and also leads to a “delayed” EF-pinning.43 Note that the physical mechanisms in these two cases differ substantially. To investigate the influence of the dipole moment magnitude on the energy level alignment the series of merocyanines was extended by 2−4 with the molecular dipole moment now covering the range from 3.8 to 13.3 D. As substrates the most application relevant substrates ITO and TiO2 were chosen. The thickness-dependent evolution of Φ of 1−4 deposited on TiO2 and ITO are shown in Figure 4, along with the energy level diagram. Deposition induced changes of Φ in Figure 4 saturate at different film thicknesses, most likely due to different growth modes of the different merocyanines. It is striking that the Φsat values are similar for all molecules on ITO (Φsat= 3.35−3.50 eV) and slightly higher on TiO2 (Φsat = 3.60−3.85 eV); no apparent correlation of Φsat to the magnitude of the molecular dipole moment can be found. The corresponding energy level diagrams strongly suggest that EF-pinning at the molecules’ LUMO occurs on ITO for 1−4, and no pinning occurs on TiO2. The fact that we find no correlation between the molecular dipole moment and the final work function in the case of EF-pinning (ITO) can be understood from the fact that electronic equilibrium needs to be established and the impact of ΔΦdip (and accordingly the impact of the molecular dipole moment) becomes negligible.42 Similarly to what was observed for 1, the interface dipole is limited by the occurrence of EFpinning, and therefore, we do not expect any correlation between Φsat and dipole strength, particularly as it is not even existent in the nonpinning case (TiO2). Here, the absence of correlation may be attributed to an increasing dimerization tendency3,17,38 of the molecules with increasing dipole moment already within the ML, which counteracts to the effect of preferred dipolar orientation. Also, molecules with stronger molecular dipoles may have an increasing tendency to lie flat on the oxides, which as well would reduce the magnitude of ΔΦdip. From a device perspective, these findings have a profound impact as it turns out that the HIB (Eonset − EF in the figures, see also SI) is rather high when employing ITO or TiO2 as electrodes, in contrast to a priori expectations, which have

initially motivated the use of merocyanines with ITO contact as donors in planar heterojunctions in combination with C60 as acceptor material.17 The better performance of photovoltaic cells when substituting ITO by a MoO3 interlayer can readily be understood from our data and can be assigned to EF-pinning at the HOMO of the merocyanine only for very high substrate Φox values. For TiO2 we find large values for the hole injection barrier as well, which means, in turn, low electron injection barriers. This should be energetically favorable for the application in dye-sensitized solar cells, where electrons are to be collected in the TiO2 conduction band.44−46

■

CONCLUSION In this photoemission study of merocyanine/metal oxide interfaces we revealed the importance of the molecular dipole moments for the energy level alignment. A coordination bond formation of the merocyanines with the oxide surfaces leads to a preferred orientation of molecular dipoles in the monolayer, which decreases the work function, while dimerization in the multilayer regime does not further impact Φ due to the antiparallel alignment of neighboring molecular dipole moments. As a consequence, EF-pinning of the molecular frontier energy levels does not set in at pristine oxide Φ values that would be expected from the Φsat values found for thick molecular films in a pinned situation. The substantial Φ lowering effect of the merocyanines themselves causes a “delay” of EF-pinning by ca. 1 eV. Another important finding is that ITO and TiO2 in conjunction with the merocyanines 1−4 result in very high hole injection barriers, in contrast to general expectations; thus these oxides are ideally suited for electron extraction from the merocyanines. To achieve good electrical contact for injection/extraction of holes, very high Φ (≥6 eV) oxides are required, e.g., MoO3.

■

ASSOCIATED CONTENT

S Supporting Information *

Valence electronic structure of all employed materials and interfaces (Sm2O3, ITO, TiO2, MoO3_cl, MoO3_e and 1−4 on TiO2) can be found. The thickness-dependent evolution of 1 valence spectra in the vicinity of the EF on all substrates is provided additionally to proof the absence of any LUMOderived state. Following Figure 4 (main text) the evolution of IP and Eonset − EF is provided for molecules 1−4. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +49 30 2093 7819. E-mail: norbert.koch@physik. hu-berlin.de. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Ruslan Oysanikov for support at BESSY and Jürgen P. Rabe for providing access to the in-house UPS setup. F.W. and M.S. acknowledge financial support by the Bavarian State Ministry of Science, Research, and the Arts within the Collaborative Research Network “Solar Technologies go Hybrid”. S.W., J.F., P.A, S.K., M.T., and N.K acknowledge support by the DFG (SPP1355, SFB951) and the HelmholtzEnergie-Allianz “Hybrid-Photovoltaik”. 11735



■

Article

(20) Steinmann, V.; Kronenberg, N. M.; Lenze, M. R.; Graf, S. M.; Hertel, D.; Meerholz, K.; Bürckstümmer, H.; Tulyakova, E. V.; Würthner, F. Simple, Highly Efficient Vacuum-Processed Bulk Heterojunction Solar Cells Based on Merocyanine Dyes. Adv. Energy Mater. 2011, 1, 888−893. (21) Koch, N. Organic Electronic Devices and Their Functional Interfaces. ChemPhysChem 2007, 8, 1438−1455. (22) Braun, S.; Salaneck, W. R.; Fahlman, M. Energy-Level Alignment at Organic/Metal and Organic/Organic Interfaces. Adv. Mater. 2009, 21, 1450−1472. (23) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. Energy Level Alignment and Interfacial Electronic Structures at Organic Metal and Organic Organic Interfaces. Adv. Mater. 1999, 11, 605−625. (24) Hill, I. G.; Rajagopal, A.; Kahn, A.; Hu, Y. Molecular Level Alignment at Organic Semiconductor-Metal Interfaces. Appl. Phys. Lett. 1998, 73, 662−664. (25) Sueyoshi, T.; Fukagawa, H.; Ono, M.; Kera, S.; Ueno, N. LowDensity Band-Gap States in Pentacene Thin Films Probed with Ultrahigh-Sensitivity Ultraviolet Photoelectron Spectroscopy. Appl. Phys. Lett. 2009, 95, 183303. (26) Bussolotti, F.; Kera, S.; Kudo, K.; Kahn, A.; Ueno, N. Gap states in Pentacene Thin Film Induced by Inert Gas Exposure. Phys. Rev. Lett. 2013, 110, 267602. (27) Osikowicz, W.; de Jong, M. P.; Salaneck, W. R. Formation of the Interfacial Dipole at Organic-Organic Interfaces: C60/Polymer Interfaces. Adv. Mater. 2007, 19, 4213−4217. (28) Krause, S.; Stolte, M.; Würthner, F.; Koch, N. Influence of Merocyanine Molecular Dipole Moments on the Valence Levels in Thin Films and the Interface Energy Level Alignment with Au(111). J. Phys. Chem. C 2013, 117, 19031−19037. (29) Dakhel, A. A. Dielectric and Optical Properties of Samarium Oxide Thin Films. J. Alloys Compd. 2004, 365, 233−239. (30) Hamberg, I.; Granqvist, C. G. Evaporated Sn-Doped In2O3 Films - Basic Optical-Properties and Applications to Energy-Efficient Windows. J. Appl. Phys. 1986, 60, R123−R159. (31) Uozumi, T.; Okada, K.; Kotani, A. Electronic-Structures of Ti and V Oxides - Calculation of Valence Photoemission and Bremsstrahlung Isochromat Spectra. J. Phys. Soc. Jpn. 1993, 62, 2595−2599. (32) Kroger, M.; Hamwi, S.; Meyer, J.; Riedl, T.; Kowalsky, W.; Kahn, A. P-type Doping of Organic Wide Band Gap Materials by Transition Metal Oxides: A Case-Study on Molybdenum Trioxide. Org. Electron 2009, 10, 932−938. (33) Duhm, S.; Heimel, G.; Salzmann, I.; Glowatzki, H.; Johnson, R. L.; Vollmer, A.; Rabe, J. P.; Koch, N. Orientation-Dependent Ionization Energies and Interface Dipoles in Ordered Molecular Assemblies. Nat. Mater. 2008, 7, 326−332. (34) Koch, N.; Duhm, S.; Rabe, J. P.; Vollmer, A.; Johnson, R. L. Optimized Hole Injection with Strong Electron Acceptors at OrganicMetal Interfaces. Phys. Rev. Lett. 2005, 95, 237601. (35) Lange, I.; Blakesley, J. C.; Frisch, J.; Vollmer, A.; Koch, N.; Neher, D. Band Bending in Conjugated Polymer Layers. Phys. Rev. Lett. 2011, 106, 216402. (36) Winkler, S.; Frisch, J.; Schlesinger, R.; Oehzelt, M.; Rieger, R.; Räder, J.; Rabe, J. P.; Müllen, K.; Koch, N. The Impact of Local Work Function Variations on Fermi Level Pinning of Organic Semiconductors. J. Phys. Chem. C 2013, 117, 22285−22289. (37) Greiner, M. T.; Helander, M. G.; Tang, W. M.; Wang, Z. B.; Qiu, J.; Lu, Z. H. Universal Energy-level Alignment of Molecules on Metal Oxides. Nat. Mater. 2012, 11, 76−81. (38) Würthner, F.; Yao, S.; Debaerdemaeker, T.; Wortmann, R. Dimerization of Merocyanine Dyes. Structural and Energetic Characterization of Dipolar Dye Aggregates and Implications for Nonlinear Optical Materials. J. Am. Chem. Soc. 2002, 124, 9431−9447. (39) Barteau, M. A. Organic Reactions at Well-Defined Oxide Surfaces. Chem. Rev. 1996, 96, 1413−1430. (40) Diebold, U. The Surface Science of Titanium Dioxide. Surf. Sci. Rep 2003, 48, 53−229.

REFERENCES

(1) Würthner, F.; Wortmann, R.; Meerholz, K. Chromophore Design for Photorefractive Organic Materials. ChemPhysChem 2002, 3, 17−31. (2) Marder, S. R.; Kippelen, B.; Jen, A. K. Y.; Peyghambarian, N. Design and Synthesis of Chromophores and Polymers for ElectroOptic and Photorefractive Applications. Nature 1997, 388, 845−851. (3) Würthner, F.; Meerholz, K. Systems Chemistry Approach in Organic Photovoltaics. Chem.Eur. J. 2010, 16, 9366−9373. (4) Grätzel, M. Dye-Sensitized Solar Cells. J. Photochem. Photobiol. C 2003, 4, 145−153. (5) Hagfeldt, A.; Boschloo, G.; Sun, L. C.; Kloo, L.; Pettersson, H. Dye-Sensitized Solar Cells. Chem. Rev. 2010, 110, 6595−6663. (6) Meyer, T.; Ogermann, D.; Pankrath, A.; Kleinermanns, K.; Müller, T. J. J. Phenothiazinyl Rhodanylidene Merocyanines for DyeSensitized Solar Cells. J. Org. Chem. 2012, 77, 3704−3715. (7) Sayama, K.; Hara, K.; Mori, N.; Satsuki, M.; Suga, S.; Tsukagoshi, S.; Abe, Y.; Sugihara, H.; Arakawa, H. Photosensitization of a Porous TiO2 Electrode with Merocyanine Dyes Containing a Carboxyl Group and a Long Alkyl Chain. Chem. Commun. 2000, 1173−1174. (8) Wu, W. J.; Hua, J. L.; Jin, Y. H.; Zhan, W. H.; Tian, H. Photovoltaic Properties of Three New Cyanine Dyes for DyeSensitized Solar Cells. Photochem. Photobiol. Sci. 2008, 7, 63−68. (9) Tang, J.; Wu, W. J.; Hua, J. L.; Li, J.; Li, X.; Tian, H. Starburst Triphenylamine-Based Cyanine Dye for Efficient Quasi-Solid-State Dye-Sensitized Solar Cells. Energ Environ. Sci. 2009, 2, 982−990. (10) Edvinsson, T.; Li, C.; Pschirer, N.; Schöneboom, J.; Eickemeyer, F.; Sens, R.; Boschloo, G.; Herrmann, A.; Müllen, K.; Hagfeldt, A. Intramolecular Charge-Transfer Tuning of Perylenes: Spectroscopic Features and Performance in Dye-Sensitized Solar Cells. J. Phys. Chem. C 2007, 111, 15137−15140. (11) Cao, L.; Wang, Y. Z.; Zhong, J. Q.; Han, Y. Y.; Zhang, W. H.; Yu, X. J.; Xu, F. Q.; Qi, D. C.; Wee, A. T. S. Electronic Structure, Chemical Interactions and Molecular Orientations of 3,4,9,10Perylene-tetracarboxylic-dianhydride on TiO2(110). J. Phys. Chem. C 2011, 115, 24880−24887. (12) Bach, U.; Krüger, J.; Grätzel, M. The Concept of Dye-Sensitized Mesoporous Solid-State Heterojunction Solar Cells. P Soc. Photo-Opt Ins 2000, 4108, 1−7. (13) Goh, C.; Scully, S. R.; McGehee, M. D. Effects of Molecular Interface Modification in Hybrid Organic-Inorganic Photovoltaic Cells. J. Appl. Phys. 2007, 101, 114503. (14) Rühle, S.; Greenshtein, M.; Chen, S. G.; Merson, A.; Pizem, H.; Sukenik, C. S.; Cahen, D.; Zaban, A. Molecular Adjustment of the Electronic Properties of Nanoporous Electrodes in Dye-Sensitized Solar Cells. J. Phys. Chem. B 2005, 109, 18907−18913. (15) Johansson, E. M. J.; Scholin, R.; Siegbahn, H.; Hagfeldt, A.; Rensmo, H. Energy Level Alignment in TiO2/Dipole-Molecule/P3HT Interfaces. Chem. Phys. Lett. 2011, 515, 146−150. (16) Berhe, S. A.; Zhou, J. Y.; Haynes, K. M.; Rodriguez, M. T.; Youngblood, W. J. Electron Transport in Acceptor-Sensitized Polymer-Oxide Solar Cells: The Importance of Surface Dipoles and Electron Cascade Effects. ACS Appl. Mater. Interfaces 2012, 4, 2955− 2963. (17) Ojala, A.; Bürckstümmer, H.; Hwang, J.; Graf, K.; von Vacano, B.; Meerholz, K.; Erk, P.; Würthner, F. Planar, Bulk and Hybrid Merocyanine/C60 Heterojunction Devices: A Case Study on Thin Film Morphology and Photovoltaic Performance. J. Mater. Chem. 2012, 22, 4473−4482. (18) Ojala, A.; Petersen, A.; Fuchs, A.; Lovrincic, R.; Polking, C.; Trollmann, J.; Hwang, J.; Lennartz, C.; Reichelt, H.; Hoffken, H. W.; Pucci, A.; Erk, P.; Kirchartz, T.; Würthner, F. Merocyanine/C60 Planar Heterojunction Solar Cells: Effect of Dye Orientation on Exciton Dissociation and Solar Cell Performance. Adv. Funct Mater. 2012, 22, 86−96. (19) Steinmann, V.; Kronenberg, N. M.; Lenze, M. R.; Graf, S. M.; Hertel, D.; Bürckstümmer, H.; Würthner, F.; Meerholz, K. A Simple Merocyanine Tandem Solar Cell with Extraordinarily High OpenCircuit Voltage. Appl. Phys. Lett. 2011, 99, 193306. 11736



Article

(41) Matsuura, A. Y.; Obayashi, T.; Kondoh, H.; Ohta, T.; Oji, H.; Kosugi, N.; Sayama, K.; Arakawa, H. Adsorption of Merocyanine Dye on Rutile TiO2(110). Chem. Phys. Lett. 2002, 360, 133−138. (42) Egger, D. A.; Zojer, E. Anticorrelation between the Evolution of Molecular Dipole Moments and Induced Work Function Modifications. J. Phys. Chem. Lett. 2013, 4, 3521−3526. (43) Koch, N. Energy Levels at Interfaces between Metals and Conjugated Organic Molecules. J. Phys.: Condens. Matter 2008, 20, 184008. (44) Mishra, A.; Fischer, M. K. R.; Bäuerle, P. Metal-Free Organic Dyes for Dye-Sensitized Solar Cells: From Structure: Property Relationships to Design Rules. Angew. Chem. Int. Ed. 2009, 48, 2474−2499. (45) Ooyama, Y.; Shimada, Y.; Ishii, A.; Ito, G.; Kagawa, Y.; Imae, I.; Komaguchi, K.; Harima, Y. Photovoltaic Performance of DyeSensitized Solar Cells Based on a Series of New-Type DonorAcceptor-π-Conjugated Sensitizer, Benzofuro[2,3-c]oxazolo[4,5-a]carbazole Fluorescent Dyes. J. Photochem. Photobiol. A 2009, 203, 177−185. (46) Hao, Y.; Yang, X. C.; Cong, J. Y.; Tian, H. N.; Hagfeldt, A.; Sun, L. C. Efficient Near Infrared D-π-A Sensitizers with Lateral Anchoring Group for Dye-Sensitized Solar Cells. Chem. Commun. 2009, 4031− 4033.

11737


Impact of Molecular Dipole Moments on Fermi Level Pinning in Thin

Recommend Documents