Axially Bound Ruthenium Phthalocyanine Monolayers on Indium Tin

Aug 10, 2017 - *E-mail: [email protected]., *E-mail: [email protected]. ... Using PM-ATR, ks,opt values of 2.2 × 103 and 2.4 × 103 s–...
2 downloads 0 Views 2MB Size
Research Article www.acsami.org

Axially Bound Ruthenium Phthalocyanine Monolayers on Indium Tin Oxide: Structure, Energetics, and Charge Transfer Properties Ramanan Ehamparam,† Luis E. Oquendo,† Michael W. Liao, Ambjorn K. Brynnel, Kai-Lin Ou, Neal R. Armstrong, Dominic V. McGrath,* and S. Scott Saavedra* Department of Chemistry and Biochemistry, University of Arizona, Tucson, Arizona 85721, United States S Supporting Information *

ABSTRACT: The efficiency of charge collection at the organic/transparent conducting oxide (TCO) interface in organic photovoltaic (OPV) devices affects overall device efficiency. Modifying the TCO with an electrochemically active molecule may enhance OPV efficiency by providing a charge-transfer pathway between the electrode and the organic active layer, and may also mitigate surface recombination. The synthesis and characterization of phosphonic acid-ruthenium phthalocyanine (RuPcPA) monolayer films on indium tin oxide (ITO), designed to facilitate charge harvesting at ITO electrodes, is presented in this work. The PA group was installed axially relative to the Pc plane so that upon deposition, RuPcPA molecules were preferentially aligned with the ITO surface plane. The tilt angle of 22° between the normal axes to the Pc plane and the ITO surface plane, measured by attenuated total reflectance (ATR) spectroscopy, is consistent with a predominately inplane orientation. The effect of surface roughness on RuPcPA orientation was modeled, and a correlation was obtained between experimental and theoretical mean tilt angles. Based on electrochemical and spectroelectrochemical studies, RuPcPA monolayers are composed predominately of monomers. Electrochemical impedance spectroscopy (EIS) and potential modulated-ATR (PMATR) spectroscopy were used to characterize the electron-transfer (ET) kinetics of these monolayers. A rate constant of 4.0 × 103 s−1 was measured using EIS, consistent with a short tunneling distance between the chromophore and the electrode surface. Using PM-ATR, ks,opt values of 2.2 × 103 and 2.4 × 103 s−1 were measured using TE and TM polarized light, respectively; the similarity of these values is consistent with a narrow molecular orientation distribution and narrow range of tunneling distances. The ionization potential of RuPcPA-modified ITO was measured using ultraviolet photoelectron spectroscopy and the results indicate favorable energetics for hole collection at the RuPcPA/ITO interface, indicating that this type of TCO modification may be useful for enhancing charge collection efficiency in OPV devices. KEYWORDS: ruthenium phthalocyanine, phosphonic acid, molecular orientation, indium tin oxide, ATR spectroelectrochemistry, electron-transfer kinetics, organic photovoltaics



INTRODUCTION Organic photovoltaics (OPVs) are continuing to evolve as a promising renewable energy source for the future.1−3 As internal quantum efficiencies approach 100% and overall device efficiencies continue to advance, it is becoming increasingly important to understand the remaining sources of inefficiency in OPVs, especially inefficiencies in charge extraction from the active layers, such that further improvements can be made.4−6 Indium tin oxide (ITO) is a commonly used transparent conducting oxide (TCO) for hole extraction in OPVs, but it has some significant disadvantages: (a) the hydrophilic surface of clean ITO is poorly wetted by the hydrophobic organic semiconductors commonly used as donor layers in OPVs;7−11 (b) the surface of ITO is electrically heterogeneous with areas of low conductivity that may present a barrier to interfacial charge transport;12,13 and (c) the energy alignment between the Fermi level (EF) of ITO and the ionization potential of the donor layer controls the hole extraction rate in an OPV, and the EF may not be well aligned with the frontier energy levels of © 2017 American Chemical Society

many commonly used donor materials, which can result in inefficient charge carrier extraction.12,14−18 ITO can be derivatized with small-molecule modifiers to modulate its surface properties, such as work function, surface energy, surface conductivity, and the spatial heterogeneity in these properties, which affects its performance as an electrode in an OPV. For example, charge extraction rates can be enhanced by tuning the work function of ITO, using smallmolecule modifiers, to match the ionization potential of the donor material.8,19−23 A phosphonic acid (PA) is frequently used as the anchoring group because PAs bind strongly to most TCOs, including ITO, and are compatible with numerous types of tail groups that provide for tunability of TCO surface properties.10 Received: May 24, 2017 Accepted: August 10, 2017 Published: August 10, 2017 29213

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces

resonance for 1H (7.26 ppm for CDCl3 and 7.22 ppm for pyridine-d5). Chemical shifts for 31P NMR were referenced to H3PO4 (0 ppm). UV−vis measurements were performed using a Shimadzu UV-2401PC spectrophotometer and a 0.1 cm quartz cuvette. Molar absorptivity (ε) values were calculated from Beer’s Law plots using four data points. Electrospray ionization (ESI) mass spectra were recorded on a 9.4 T Bruker Apex-Qh Fourier transfer ion-cyclotron resonance instrument. The samples were dissolved in MeOH:MeCN 1:1 in the concentration range of 1−5 μM and this solution was sprayed using an Apollo II ESI source. ESI conditions and mass spectrometer tuning parameters were conventional except that the ESI source temperature (drying gas temperature) was kept at 100 °C which was necessary to avoid thermal dissociation of RuPc complexes. Substrate Cleaning Procedure. Glass slides coated with indium tin oxide (ITO; thickness = ca. 100 nm) were purchased from Colorado Concept Coatings (total thickness of ca. 1 mm or ca. 0.4 mm; denoted CC ITO) and Thin Film Devices (total thickness of ca. 1 mm; denoted TFD ITO). CC ITO was used for all experiments excepting those for which TFD ITO is specified. ITO slides were cleaned by first scrubbing with 1% Triton X-100, followed by successive sonication in 1% Triton X-100, water (Barnstead nanopure, 18.3 MΩ·cm) and ethanol, for 15 min for each step. Cleaned ITO slides were stored in ethanol and then subjected to a final cleaning in a low temperature air plasma (Harrick PDC-3XG) at medium RF level for 15 min immediately prior to any experiments. UV−vis Attenuated Total Reflectance (ATR) Spectroscopy. A detailed description of the custom built, UV−vis ATR spectrometer can be found elsewhere.4,43,44 Briefly a collimated broadband source (Xe lamp) was coupled into and out of the waveguide (an ITO-coated slide) using two BK7 prisms (n = 1.51). The outcoupled light from the waveguide was directed into a monochromator (Newport MS260i) and detected using a CCD (Andor iDus420A). The two prisms were placed 44.5 mm apart producing 10 total internal reflections at the ITO/solution interface. The reflection angle was 67−72°. ATR measurements were acquired using either transverse electric (TE) or transverse magnetic (TM) polarized light which was selected using a half wave Fresnel rhomb. Potential-Controlled ATR Spectroscopy. Potential-controlled ATR spectroscopy4 on RuPcPA films on ITO was performed using an ATR flow cell modified with a conventional three-electrode system. An ITO-coated slide was both the waveguide and the working electrode (active area = 0.8 cm2), a Pt wire was used as the counter electrode, and Ag/AgNO3 (0.01 M TBAP in acetonitrile) served as the pseudoreference electrode. The supporting electrolyte was a degassed solution of 0.1 M TBAP in acetonitrile. After polarized ATR spectra were measured at open circuit potential, spectra were acquired at fixed potentials from −0.2 to 0.6 V, at 50 mV intervals, vs Ag/AgNO3. Potential Modulated Attenuated Total Reflectance (PMATR) Spectroscopy. A detailed description of PM-ATR spectroscopy can be found elsewhere.4,44,45 Briefly, experiments were performed with an ATR flow cell having a conventional three-electrode system, using the counter and pseudoreference electrodes described above. ITO slides (total thickness = ca. 0.4 mm) served as both the ATR element and the working electrode (active area = 0.8 cm2). A degassed solution of 0.1 M TBAP in acetonitrile was used the background electrolyte solution. Additional experimental details are given in SI. Cyclic Voltammetry. Cyclic voltammetry (CV) experiments on RuPcPA films on ITO were performed using a CH420A potentiostat (CH Instruments) and a conventional three-electrode cell, using the counter and pseudoreference electrodes described above. The active area of the ITO working electrode was either 0.66 cm2 or 0.071 cm2. Procedures for diffusion controlled voltammetry on dissolved RuPcPA and precursor molecules are given in SI. Electrochemical Impedance Spectroscopy (EIS). Electrochemical impedance spectroscopy (EIS) was performed using an EG&G Model 263A potentiostat/galvanostat with a Model 1025 frequency response detector. Impedance data were collected using PowerSuite 2.00.5 software (Princeton Applied Research). The ATR flow cell and supporting electrolyte described in the PM-ATR section

Molecular modifiers used to achieve these changes may be strictly dipolar in character,8,10 or may feature redox-activity. A redox-active modifier with a reduction potential that is matched to the frontier energy levels of the donor material may serve as a charge transfer “bridging” layer at the ITO/donor interface in an OPV.1,4−6,10 In addition to the electroactive moiety, a number of structural variables are available in the design of a redox-active modifier, such as the type of anchoring group, the composition and length of the linker, and the presence of peripheral solubilizing groups.10 These variables in turn control molecular orientation, degree of aggregation, interface dipole magnitude and orientation, electronic coupling between the redox center and the electrode, tunneling distance, and so on; all of these are predicted to mediate charge collection kinetics and efficiency at the ITO/donor interface.4,9,10,24−26 Phthalocyanines have been investigated extensively as donor chromophores due to a number of favorable properties such as broad absorption in the near IR, photoconductivity, and thermal stability.27,28 Recently, we have studied the structure and charge transfer behavior of zinc phthalocyanines (ZnPcs) immobilized on ITO using PA anchoring groups attached to the macrocycle periphery with alkyl linkers.4,9,29 The molecular orientation distribution and degree of aggregation in ZnPc monolayers was found to have a significant influence on the electron-transfer (ET) kinetics, with the fastest ET exhibited by monomeric ZnPc molecules having the macrocycle plane largely parallel to the ITO surface plane. Pcs having ruthenium as the central metal of the macrocycle have also been examined for energy conversion applications.30−35 The higher coordination number of Ru allows attachment of pyridine-based axial ligands that provide for tethering of RuPcs to metal and metal oxide surfaces via ligand exchange,36−39 and also minimizes aggregation of the Pc macrocycle.34 For example, immobilization of RuPc on nanocrystalline TiO2 using a 4-pyridine carboxylic acid axial ligand has been extensively studied.30,31,33,34,39,40 In dyesensitized solar cells, these types of molecules have produced photocurrent yields of over 60% in the near-IR region.39 RuPc molecules also have been utilized in OPV platforms.41,42 In this work, we report the synthesis of a RuPc derivative containing 4-pyridine phosphonic acid as an axial ligand that serves as the anchoring group to tether the molecule to ITO. Bulky t-butyl groups at the periphery of the macrocycle are present to increase solubility in common organic solvents.33 Upon attachment to ITO, this design is expected to produce a RuPc monolayer composed of predominantly monomeric species with an in-plane orientation and a relatively short tunneling distance between the macrocycle and the electrode, which should minimize dispersity in the ET kinetics.4,9 Herein we report detailed characterization of RuPcPA monolayers on ITO using electrochemical, spectroelectrochemical, and photoelectron spectroscopy methods to assess relationships between monolayer structure, redox thermodynamics, ET kinetics, and energy level alignment.



EXPERIMENTAL SECTION

Materials, Synthesis, and Characterization Methods. All chemicals were purchased from commercial suppliers and used as received unless otherwise noted. RuPc complexes with axial ligands were prepared according to modified literature methods.32,34,35 A detailed description along with NMR characterization is given in Supporting Information (SI). NMR data were collected on a Bruker 500 MHz. Chemical shifts were referenced to the deuterated solvent 29214

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces Scheme 1

were used. A sinusoidal AC voltage of 10 mVRMS in the frequency range of 1 Hz - 100 kHz was applied at DC potentials of −0.2 or 0.3 V. Photoelectron Spectroscopy. X-ray photoelectron spectroscopy (XPS) was performed on a Kratos Axis-Ultra spectrometer. The X-ray source was a monochromatic Al Kα photon source (1486.6 eV, 300 W) using normal takeoff angles. Each sample was loaded into the spectrometer and first evaluated to determine sample composition and stability. Survey scans of the samples were obtained using the following settings: 1 eV step size, 100 ms dwell time, and 160 eV pass energy. High-resolution spectra were collected at 0.1 eV step size, 250 ms dwell time, and 20 eV pass energy for at least 4 sweeps for each element region. Samples were electrically coupled to the spectrometer and no sample bias was applied. Additional details are given in SI. Ultraviolet photoelectron spectroscopy (UPS) was performed using a Kratos Axis-Ultra spectrometer with a Specs UVS-20 He (I) (21.2 eV) excitation source. The analyzer was set to 5 eV pass energy, 250 ms dwell time, and 0.01 eV step size. A sample bias of 10.00 V was applied between the sample stage and the detector to ensure that the lowest kinetic energy photoelectrons were collected. The instrument was operated ca. 1.0 × 10−7 Torr. The Fermi energy of the spectrometer was ascertained by evaluating freshly cleaved HOPG and atomically clean Au foils. At least three different spots were interrogated on each sample. Additional details are given in SI.



dichlorobenzene to yield a mixture of two products, Pc 3 (RuPcPO3Et2) and Pc 4. The mixture was separated by column chromatography, leveraging the difference in polarities between the asymmetric and symmetric complexes. The minimum temperature required to promote asymmetric ligand exchange was found to be ca. 130 °C, to yield Pc 3 in 27%. Under these conditions, the yield for the symmetric product, Pc 4, was 4%. We converted the phosphonate group of Pc 3 to a phosphonic acid derivative to provide a functional group capable of binding strongly to ITO. This was achieved by reacting Pc 3 with TMSBr in CH2Cl2 to yield Pc 5 (RuPcPA), as shown in Scheme 2, which is the first reported example of a phosphonic acid RuPc complex. We investigated the electronic properties of the RuPc complexes by UV−vis spectroscopy (Figure 1). They exhibited

RESULTS AND DISCUSSION

Synthesis and Spectral Properties of RuPc Complexes. We prepared the RuPc complexes with axial ligands as outlined in Schemes 1 and 2. Briefly, 4-tert-butyl phthalonitrile was condensed with RuCl3, 1,8diazabicyclo[5.4.0]undec-7-ene (DBU) and pyridine as the axial ligand in pentanol to form Pc 1 (RuPcPy) as a blue complex. Pc 1 was subjected to an axial ligand exchange using diethyl(4-pyridinyl)phosphonate (2) at 130 °C for 6 h in

Figure 1. UV−vis absorbance spectra of RuPcPy (1), RuPcPO3Et2 (3), and RuPcPA (5) dissolved in CH2Cl2 at 100 μM. Spectra are normalized to the wavelength of the Q-band absorbance maxima (λmax = ca. 630 nm) to compare the spectral profiles.

Scheme 2

an intense Soret band with an absorbance maximum at ca. 318 nm which is a characteristic feature of RuPc complexes.30,33,46 The absorbance maximum of the Q-band was at 630−632 nm, accompanied by a high energy shoulder that can be attributed to a charge transfer band, as previously reported for RuPc molecules with pyridine and polypyridine axial ligands.46 The Q-band of Pc 5 in CH2Cl2 was slightly broadened compared to Pcs 1 and 3, suggesting that Pc 5 forms aggregates and that the PA group plays a role. Dissolving Pc 5 in a protic solvent (ethanol) decreased the broadening of this peak (see 29215

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces

Molecular Orientation. We determined the molecular orientation of Pc 5 adsorbed on CC ITO from the dichroic ratio (ρfilm = ATE,film/ATM,film) measured at 632 nm using ATR spectroscopy, following procedures described in previous publications.4,9,47 TE- and TM-polarized ATR spectra of a Pc 5 film on CC ITO are shown in Figure S14A. To correct for the difference in interfacial electric field intensities in TE and TM polarizations, we used the method described by Mendes et al.,47 as described in SI, and circularly polarized absorbance in the plane of the Pc macrocycle was assumed. The mean tilt angle (θμ) between the axis normal to the macrocycle plane and the axis normal to the CC ITO substrate plane was determined to be 37°(±1) (n = 3). This angle is larger than expected; binding via the axial PA ligand was expected to produce a more in-plane orientation and a smaller θμ. Numerous studies of molecular orientation of PA modifiers on ITO have been published (see recent review10); however, none of these included pyridyl phosphonic acid. The most relevant of the modifiers studied to date is phenyl phosphonic acid (PPA) for which Gliboff et al.48 reported a tilt angle of 19° (±4) between the plane of the phenyl ring and the ITO surface plane. This result agrees well with density functional theory calculations which predict a tilt angle of 15°−22° when the PA-ITO binding is bidentate.7 The surface of TFD ITO is considerably smoother than that of CC ITO, as shown by the atomic force microscopy (AFM) images presented in Figure S16. Notably, the studies of Gliboff et al.48 were performed on TFD ITO. This information suggests that the larger than expected tilt angle of Pc 5 on CC ITO may be due to its higher surface roughness. To assess this possibility, the molecular orientation of Pc 5 adsorbed on TFD ITO was determined using ATR spectroscopy. The result, θμ = 22° (±1) (n = 3), agrees well with both the experimental and theoretical studies of PPA orientation on ITO. To further address the discrepancy between the Pc 5 tilt angles measured on CC and TFD ITO, we performed a theoretical study using the approach described by Simpson and Rowlen.49 The details of the method are presented in SI and outlined here. AFM images of air plasma-cleaned CC and TFD ITO were collected using a scan size of 500 nm × 500 nm (see examples in Figure S16). With reference to the schematic of a surface roughness feature with a coordinate frame shown in Figure S17, the distribution of local surface normal tilt angles, defined with respect to the macroscopic ITO surface normal (the Z-axis), was then calculated from the AFM data. Figure S18 shows the local surface normal tilt angle distributions for CC ITO and TFD ITO. The shapes agree well with those calculated for fused silica by Simpson and Rowlen.49 The CC ITO distribution is broader and more asymmetric than the TFD ITO distribution. The most probable local surface normal tilt angle (ϕmp) is 7° for TFD and 13° for CC ITO. The average local surface normal tilt angle (ϕave) is 9° for TFD and 21° for CC ITO. We also modeled the molecular tilt angle distribution of adsorbed Pc 5. A molecular axis was placed randomly (with respect to the local azimuthal plane) on the ITO surface and tilted at 22° relative to the local surface normal obtained from AFM data. This angle was chosen to match the θμ measured for Pc 5 on TFD ITO. The local tilt angle was then calculated for each vector with respect to the Z-axis. Figure 3 shows the theoretical tilt angle distributions on CC ITO and TFD ITO. Similar to Figure S18, the CC curve is broader and asymmetrically weighted toward higher angles, with a most probable tilt angle (θmp) of 25° and an average tilt angle (θave) of 29°. For the TFD curve, both θmp and θave are 23°. Overall,

comparison of spectra in ethanol and CH2Cl2 in Figure S10A). This decrease is likely due to hydrogen bonding between the phosphonic acid of Pc 5 and ethanol, which competes with aggregation via PA−PA interactions. The ratio of the absorbance values at the Soret and Q-band maxima in CH2Cl2 was 0.50−0.55 and was independent of concentration in the range of 5−100 μM, which indicates that higher-order aggregate formation was minimal in this concentration range (Figure S10B). The molar extinction coefficient for Pc 5 at 630 nm measured in CH2Cl2 was 60 000 M−1cm−1. Adsorption Kinetics and Spectroscopic Characterization of Pc 5 Adsorbed on ITO. We monitored chemisorption of Pc 5 to ITO slides using ATR spectroscopy in TE polarization. The detailed experimental procedure is presented in SI. Spectra collected as a function of time are shown in Figure S11A. The absorbance at 632 nm reached an asymptote after 20 min, indicating saturation of the ITO surface (Figure S11B). After the flow cell was flushed with pure ethanol, substantial absorbance was still present, indicating that Pc 5 molecules were strongly adsorbed to ITO. We performed control experiments with Pcs 1 and 3 using the same experimental conditions, and observed no detectable absorbance after flushing the ATR flow cell with pure solvent. This indicates that the PA functional group acts as an anchor to chemically bind Pc 5 to ITO and that negligible physisorption of molecules lacking a PA group occurs on ITO.4 There is a small red shift, ca. 7 nm, in the Q-band in the ATR spectrum of Pc 5 on ITO relative to the UV−vis absorbance spectrum of Pc 5 dissolved in ethanol (Figure 2). We attribute

Figure 2. Absorbance spectrum of Pc 5 dissolved in ethanol (100 μM) and TE-polarized ATR spectrum of a Pc 5 film on ITO. Absorbance values are normalized to the wavelength of the Q-band maxima (λmax = ca. 630 nm) to compare the spectral profiles.

this shift to minor changes in the frontier orbital energies,4,9 and PA deprotonation may be a contributing factor. More importantly, the ATR spectrum did not show evidence for the presence of aggregated species in the Pc 5 film; this is in contrast to ZnPcPA adsorbed to ITO which exhibits a broadened, blue-shifted Q-band due to cofacial aggregation.4 We attribute the absence of significant aggregation in the Pc 5 film to the molecular design that binds via the axial ligand and is expected to align the plane of the macrocycle largely parallel to the ITO surface plane, preventing cofacial intermolecular interactions. 29216

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces

Figure 3. Theoretical tilt angle distributions of Pc 5 molecules placed at 22° with respect to local surface normal on TFD ITO (blue) and CC ITO (red).

the results of these modeling studies show that a higher surface roughness shifts the theoretical distribution toward higher angles, and thus provide an explanation for why the θμ for Pc 5 on CC ITO is greater than that on TFD ITO. Finally, these results also illustrate the importance of considering the contribution of molecular-scale surface roughness to an ensemble measurement of molecular orientation. Electrochemical and Spectroelectrochemical Studies of RuPc Derivatives. We performed diffusion controlled cyclic voltammetry of Pcs 1, 3, and 5 dissolved in CH2Cl2 to assess the effect of structure on redox thermodynamics. Experimental details are given in SI, example CVs are presented in Figure S12, and midpoint and peak potentials vs Ag/AgNO3 are listed in Table S1. Pc 1 exhibited two successive, reversible one-electron oxidations of the Pc macrocycle in the potential range of 0 to 1.2 V, consistent with the literature.50 Pcs 3 and 5 also underwent two oxidations in the same potential window (see data in Table S1), although the second oxidation was not fully reversible.31,32 We assessed the voltammetric behavior of the three RuPcs after each was incubated with ITO followed by rinsing, and the results are shown in Figure 4A. Only Pc 5 showed redox activity, confirming that the PA group is necessary for chemisorption to ITO. The midpoint potential of first oxidation of adsorbed Pc 5 was ca. 0.30 V vs Ag/AgNO3, equivalent to that measured for dissolved Pc 5; thus, immobilization on ITO did not measurably affect Pc 5 redox thermodynamics. The anodic−cathodic peak separation was less than 20 mV and was independent of the scan rate (see Figure S13A), and the current density scaled linearly with scan rate (Figure S13B); these results confirm that the first oxidation was surface confined and reversible. The CV of Pc 5 exhibited only one peak in the potential range where the first oxidation of Pcs is typically observed. In contrast, in an earlier study of ZnPcPA adsorbed to ITO, the CV showed two oxidation peaks in this range that were assigned to monomeric and aggregated subpopulations in the film.4 The presence of only one peak for Pc 5 on ITO is consistent with aggregation being minimal or eliminated due to binding via the axial ligand, and the absence of spectral evidence for aggregation in the ATR spectrum (Figure 2), as discussed above, provides further support for this interpretation. A similar lack of aggregation was observed for ZnPc(PA)4 films on ITO15 and likewise this molecule was shown to adsorb in a largely in-plane orientation. However, we note that the

Figure 4. (A) Representative CVs measured after ITO was incubated with Pc 1 (green), Pc 3 (blue), or Pc 5 (red) solution, then subjected to extensive rinsing with ethanol. The background electrolyte was 0.1 M TBAP in acetonitrile and the scan rate was 100 mV/s. (B) Representative potential-dependent UV−vis ATR spectra of a Pc 5 film on CC ITO measured using TE polarization over an applied potential range of 0 (Voc) to 0.6 V. (C) Normalized TE absorbance at 632 nm (red circles) vs applied potential. The blue line is a sigmoidal fit to the experimental data and the red dashed line is the first derivative of the sigmoidal fit.

fwhm of the Pc 5 oxidation and reduction peaks was ca. 150 mV (in the scan rate range of 0.1−0.5 V/s) which is greater than the Nernstian value of 91 mV for a one-electron transfer. This suggests that there is some heterogeneity in the local environment of adsorbed Pc 5 molecules that gives rise to a nonideal distribution of redox potentials,51,52 which is not unexpected given that the surface properties of ITO are known to be spatially heterogeneous.12,13,53 Similar broadening was observed for ZnPc(PA)4 films on ITO.9 The electroactive surface coverage of Pc 5 adsorbed on CC ITO was estimated by integrating the Faradaic current (average of the oxidation and reduction peaks) in Figure 4A).54 The result was 3.1(±0.4) × 10−11 mol·cm−2 (n = 5) which is 29217

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces

which the imaginary part of the capacitance is plotted against frequency (Figure 5A).58,59 The imaginary capacitance was

approximately 75% of full monolayer coverage, assuming that 4 nm2 is the projected area of an adsorbed RuPc molecule.31 This projected area assumes that the Pc macrocycle is parallel to the ITO surface plane, which is a reasonable approximation based on the orientation results presented above, as well as other studies with similar molecules.31,55 Rawling et al.31 studied the adsorption of RuPc complexes with thioacetate axial ligands on Au and showed that the different peripheral and axial ligand substituents on RuPcs significantly affected the surface coverage and molecular order of the film. Complexes with four peripheral tertiary butyl groups formed lower surface coverage films that were relatively disordered, which was attributed to the existence of four possible positional isomers for each molecule. Their results agree with and provide a likely explanation for the submonolayer coverage observed here. Potential-step ATR spectroelectrochemistry4,43,56 of Pc 5 films was performed as a complement to conventional electrochemistry and to assess conditions for PM-ATR spectroscopy. Spectra acquired in TE polarization as a function of applied potential are shown in Figure 4B. Increasing the potential (i.e., decreasing the Fermi level) from −0.2 V to +0.8 V with respect to the Ag/AgNO3 pseudoreference electrode caused bleaching of the Q-band at ca. 630 nm and the appearance of a new band at ca. 550 nm, which is consistent with the results reported by Rawling et al.33 for RuPc complexes. We obtained similar results when the experiment was performed using TM polarization (Figure S19A). In Figure 4C and Figure S19B, the change in absorbance at 632 nm is plotted vs potential. A sigmoidal function was fit to the data and the first derivative of the sigmoidal curve was then used to determine the midpoint potential. We obtained a value of 0.3 V in both TE and TM polarizations which is consistent with the CV data and is assigned to macrocyclic oxidation of adsorbed Pc 5.33 Electron-Transfer Kinetics of Pc 5 Films on ITO. The ET kinetics of Pc 5 films were evaluated using two complementary techniques, EIS and PM-ATR. In EIS, the impedance is measured at a fixed DC potential with a superimposed AC voltage of variable frequency. Figure S20A shows the Nyquist plot of a Pc 5 film on ITO measured at a DC potential of −0.2 V vs Ag/AgNO3, a potential at which Pc 5 is not redox-active. A vertical line indicating capacitive-like behavior was observed as expected.57 The equivalent circuit at a potential where the film is redox-inactive can be modeled as a double-layer capacitance (Cdl) and a solution resistance (Rs). The capacitive behavior is also observed when the phase angle approaches 90° at low frequencies as shown in the Bode-phase plot in Figure S20B. The EIS experiment was repeated at 0.3 V where the Pc 5 film is redox-active, and a capacitive-like behavior (straight line) was also observed in the Nyquist plot (Figure S20A). A semicircle was expected, however, in the high frequency region due to the ET reaction. The absence of a prominent semicircle is likely due to a fast ET rate (i.e., a small charge transfer resistance). The magnitude of the impedance at low frequency measured at 0.3 V is less than that measured at −0.2 V (see Figure S20B for the Bode-magnitude plot), which indicates that impedance contribution is indeed due to redox activity.57 A similar observation has been reported for ITO modified with PA-modified perylenediimide molecules.57 Due to the absence of a semicircle feature in the Nyquist plot, we did not fit the impedance data to an equivalent circuit model. To obtain the apparent heterogeneous ET rate constant from these data, we constructed a Bode capacitance plot in

Figure 5. (A) Plot of imaginary capacitance vs AC frequency for a Pc 5 film on ITO at DC potentials of 0.3 V (red circles) and −0.2 V (green squares) vs Ag/AgNO3. The sinusoidal AC voltage and the frequency used in these experiments were 10 mVRMS and 1−100 kHz, respectively. The difference imaginary capacitance curve (blue diamonds) was obtained by subtracting the −0.2 V curve from the 0.3 V curve. The peak frequency in the difference curve, f 0, was used to calculate the heterogeneous ET rate constant (k s,EIS ). (B) Representative normalized complex plane plot of the electroreflectance signal of a Pc 5 film on ITO measured in TE polarization. The DC potential and the amplitude of the sinusoidal AC voltage were 0.3 V vs Ag/AgNO3 and 30−40 mVRMS, respectively. Measurements were performed over a frequency range of 0.1−3000 Hz. The optical switching frequency (ω) at which real reflectance (Re(Rac)) is zero, indicated by the dashed ellipse, was used to obtain the heterogeneous ET rate constant (ks,opt).

calculated from Cim = −Im[(jωZ)−1] where Z is the complex impedance and ω is the frequency in rads−1. In Figure 5A, a peak is observed at frequencies >1 kHz in both the −0.2 and 0.3 V imaginary capacitance curves (denoted by green squares and red circles, respectively). The peak in the −0.2 V curve is due to the RC time constant of electrochemical cell, whereas in the 0.3 V curve, it is due to both the redox process and the RC time constant. To remove the latter contribution, we performed a linear subtraction of the two curves. The ET rate constant (ks,EIS) was then calculated from ks,EIS = πf0 where f 0 is the frequency of the peak in the difference imaginary capacitance curve (blue diamonds in Figure 6A).59 For the trial shown in Figure 5A, f 0 was 1151 Hz. The ks,EIS, value obtained from this analysis was 4.0 (±0.5) × 103 s−1 (n = 3). This result is discussed below after presentation of the PM-ATR data. 29218

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces

Figure 6. (A) Ultraviolet photoelectron spectra of (I) air plasma-cleaned ITO, (II) air plasma-cleaned ITO followed by an ethanol soak (control ITO), and (III) Pc 5-modified ITO. The high binding energy (HBE) region, the full spectra, and the low binding energy (LBE) region are shown in the left, middle, and right panels, respectively. EF is shown as the vertical dashed line in the full and LBE spectra, assuming electronic equilibrium between the sample and spectrometer. The energy axis in the spectra is relative to EF. The HBE and LBE edges and BEHOMO, which are used to define the spectral widths w, w′, and w″, are also shown where appropriate. The widths are used to estimate local work function, IP, and EHOMO, respectively, as shown in the middle panel. Refer to the text for the relevant equations. (B) Energy level diagram obtained from UPS data showing estimates of IP and IP′ of Pc 5 films on ITO with consideration of changes in the local vacuum level.

tunneling distances between the macrocycle and ITO. A similar hypothesis was advanced in an earlier report to explain the polarization-independent ET kinetics of ZnPc(PA)4 adsorbed on ITO, which was also shown to adopt a largely in-plane orientation.9 If the Pc 5 film had a broad tilt angle distribution, ks,opt(TE) would reflect molecules with smaller tilt angles and ks,opt(TM) polarization would reflect molecules with larger tilt angles, and these rate constants would not be polarizationindependent. In previous studies, we compared the structure and ET kinetics of ZnPcPA and ZnPc(PA)4 monolayers on ITO.4,9 The ks,opt values for Pc 5 are greater than those observed for ZnPcPA, which was adsorbed via a 10-carbon methylene linker and formed a relative disordered film composed of monomeric and aggregated subpopulations with a broad tilt angle distribution. The ks,opt values for the monomeric subpopulation were ca. 200 s−1. In contrast, the ks,opt values for ZnPc(PA)4 were about 8-fold greater than those for Pc 5. These differences can be rationalized by considering estimated macrocycleelectrode tunneling distances (d): For monomeric ZnPcPA tilted at 33° and Pc 5 tilted at 22°, d is estimated to be 8.3 and 5.8 Å, respectively. ZnPc(PA)4 bound to ITO via 3−4 PA groups should be largely parallel to the ITO surface which dictates a shorter d; e.g., for a conformation with a single gauche defect in each alkyl linker, d is predicted to be 2.8 Å.9 Energy Level Alignment. To evaluate the thermodynamic feasibility for hole extraction across the RuPc/ITO interface, energy level alignment was assessed using energy levels estimated from both the electrochemical and UPS data. We calculated the first ionization potential (IP) from the onset potential of the first oxidation peak, Eox, of the Pc 5 film on ITO which was 0.18 V vs Ag/AgNO3. To correct this value to the vacuum scale, a value of −4.48 eV versus vacuum was used for the potential of the normal hydrogen electrode (NHE).63 The Ag/AgNO3 electrode was calibrated against ferrocene for which the midpoint potential was measured to be 0.16 V and assumed to be 0.64 V vs NHE.64 The IP was estimated from IP = |−(Eox + 4.96)| eV = 5.14 eV. This result is comparable to

PM-ATR is a waveguide-based form of electroreflectance (ER) spectroscopy60,61 that can be used to measure ET rate constants of electrochemically active films of chromophores. Measurements made using TE- and TM-polarized light allow the kinetic information to be correlated with molecular orientation, which is not possible with EIS.4,9,45 The working principle of PM-ATR is described elsewhere,4,9,44,45,62 and the experimental procedures used in this study are given in SI. Briefly, a DC potential with a superimposed sinusoidal potential modulation of variable frequency is applied to the ITO working electrode on which the redox-active film is immobilized. The DC potential and wavelength are selected such that the potential modulation produces a significant change in the ATR signal; for Pc 5 films, these values were 0.3 V and 632 nm, respectively (see SI). The real and imaginary parts of the ATR signal are measured as a function of modulation frequency. The resulting complex plane plot is fit with a polynomial function to determine the optical switching frequency (ω) at which the real reflectance is zero. Example plots in TE and TM polarization are shown in Figures 5B and S22, respectively. The apparent heterogeneous ET rate constant (ks,opt) is calculated from ks,opt = 0.5ω2RsCdl using Rs and Cdl values measured by EIS. Using Rs = 5(±1) Ω·cm2 and Cdl = 12(±1) μF/cm2, the ks,opt obtained for Pc 5 films on ITO in TE and TM polarizations were 2.2(±0.3) × 103 s−1 (n = 5) and 2.4(±0.6) × 103 s−1 (n = 5), respectively. The rate constants obtained by EIS and PM-ATR agree well, differing by a factor of less than two. This is the first direct comparison of kinetic data obtained by these techniques. The slight difference between ks,EIS and ks,opt may be due to the assumption in the PM-ATR measurement that all circuit elements corresponding to electrochemical processes are ideal. In reality, they are not ideal. For example, the double-layer capacitance exhibits a slight deviation from 90° at low frequency as can be seen from the Nyquist plot (Figure S18B). We interpret the equivalent ks,opt values measured in TE and TM polarizations as evidence of a narrow distribution of Pc 5 tilt angles in the film, which produces a small distribution of 29219

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces

ITO (Figure 2). The resulting vacuum level-adjusted energy band diagram is shown in Figure 6B. The Φ of AP-ITO was found to be 5.0 eV, which is consistent with published values.9,29 Shifts in the local vacuum level of 0.3 and 1.1 eV were observed for control ITO and Pc 5modified ITO, respectively. Soaking the ITO in EtOH likely passivated some of the high energy surface sites exposed by air plasma treatment, and thus a vacuum level shift is expected, which reduced Φ to 4.7 eV. Deposition of Pc 5 further decreased Φ by ca. 0.8 eV relative to the control ITO, a shift indicative of a change in the local surface dipole at the ITO/Pc 5 interface, as observed for similar systems.9,29 The IP and IP′ of Pc 5/ITO were found to be 4.7 and 6.2 eV, respectively. The IP value differs from the electrochemically determined IP by ca. 0.5 eV. Differences in energy levels measured by electrochemistry and photoelectron spectroscopy also have been reported for ZnPcs on ITO9,29 and RuPcs on TiO265 and are attributed to the changes in the local surface dipole at the Pc/ TCO interface, solvation effects, and differences in the dielectric medium at the interface, as discussed above. EHOMO and EHOMO−1 of the Pc 5 film (Figure S24) were 5.4 and 7.0 eV, respectively. The HOMO feature lies 1.4 eV below the EF of ITO which indicates that favorable energetic alignment for hole extraction exists at the Pc 5/ITO interface. This in turn suggests that from a thermodynamic perspective, charge transfer from a subsequently deposited Pc active layer to Pc 5-modified ITO may not be contact limited, assuming that proper energy level alignment of the active layer is achieved.

previously reported IPs of RuPcs determined electrochemically.42,65,66 It is well-established that chemisorption of a PA modifier on ITO can shift the local vacuum level due to a change in the magnitude and/or orientation of the local surface dipole.10 An electrochemical estimate of IP does not account for interface dipole changes and associated local vacuum level shifts, which may play an important role in OPV performance. Furthermore, an electrolyte-solvated Pc monolayer may not reflect the local environment at a Pc active layer/ITO interface in a solid-state OPV. We therefore examined the energetics of Pc 5-modified ITO using UPS. Prior to UPS experiments, XPS was performed to evaluate the substrate surface composition. Spectra of the N 1s and Ru 3d regions, shown in Figure S23, confirmed the adsorption and stability in high vacuum of Pc 5 films on ITO. Ultraviolet photoelectron spectra of air plasma-cleaned (AP) ITO, control ITO (AP-ITO soaked in ethanol, the deposition solvent), and Pc 5-modified ITO are shown in Figure 6A. The binding energies (BE) of the photoelectrons are displayed in the spectra with respect to the spectrometer/sample Fermi energy, EF, (assuming electronic equilibration between the sample and the spectrometer) defined as zero. The left panel shows an expanded view of the high binding energy (HBE) or low kinetic energy (LKE) edge, the center panel shows the full spectra, and the right panel is an expanded view of the low binding energy (LBE) or high kinetic energy (HKE) edge, with respect to EF. Each panel also features hash marks for HBE (LKE), LBE (HKE), BEHOMO (KEHOMO), and EF, where applicable. BEHOMO and KEHOMO are the energy of the center of the HOMO photoemission feature in the BE and KE scales, respectively. The center panel additionally has lines marked as w, w′, and w″ defined by the spectral width between HBE (LKE) and BEHOMO (KEHOMO), LBE (HKE), and EF, respectively. The right panel was treated with systematic background corrections to enhance the Pc 5 photoemission features by reducing the photoelectron contributions from the substrate, He (I) satellites, and secondary electron emission. The details of the correction protocol are reported in the literature and also are given in SI (Figure S24).67,68 A peak in the spectrum observed near EF (ca. 1.4 eV) is due to photoemission from the highest occupied molecular orbital (HOMO) of Pc 5. Similar observations of HOMO features were reported for RuPc with bis(4-carboxypyridine) in the axial position and adsorbed on TiO2.65 We attribute a second photoemission peak at ca. 3.0 eV to the HOMO−1 of the molecule. Despite the low surface coverage of Pc 5, both the HOMO and HOMO−1 features are resolved, although the HOMO−1 feature is detectable only with implementation of the correction protocol. Using the approach described in previous papers,9,10,29,69 IP, IP′, effective local work function (Φ), and EHOMO and EHOMO−1 were calculated for each sample. Briefly, the Φ is determined from Φ = hν − w. IP and IP′ are determined from IP = hν − w′, where w′ is defined by the photoemission onset of the HOMO and HOMO−1, respectively. We also calculate EHOMO and EHOMO−1 in the same way as IP, but using the center of the HOMO and HOMO−1 photoemission features, respectively, rather than the photoemission onsets; i.e., EHOMO = hν − w″. The binding energies used to derive IP, IP′, EHOMO, and EHOMO−1 are tabulated in SI (Table S2). The electron affinity (EA) was obtained by adding the optical band gap (Eopt) to the IP values. Eopt was estimated from the low energy onset of the Q-band absorbance in the TE absorbance spectrum of Pc 5 on



CONCLUSIONS

Pc 5, a TCO modifier bearing an axial ligand with a PA anchor, binds strongly to ITO, forming a film with near monolayer surface coverage. The axial ligand binding was expected to produce a planar orientation that minimizes Pc-Pc aggregation. We confirmed both effects: (a) CV and spectroelectrochemical studies revealed the presence of only monomers in Pc 5 films (i.e., dimerization was insignificant); (b) orientation studies using ATR spectroscopy show that in the Pc 5 film, the plane of the macrocycle is largely aligned with the ITO surface plane. We modeled the effect of ITO surface roughness on the tilt angle measurement and the results agree well with the experimental results on CC (rougher) and TFD (smoother) ITO − the rougher surface broadens the orientation distribution and shifts the mean tilt angle to a higher value. ET rate constants at the Pc 5/ITO interface were measured using EIS and PM-ATR, and these approaches produced similar results, with rate constants in the range of 2−4 × 103 s−1. Comparing the rate constants measured here with those obtained in prior studies of ZnPcs on ITO suggest that the rapid kinetics of Pc 5/ITO can be attributed to a short tunneling distance between the Pc macrocycle and the electrode surface.4,9,29 Using PM-ATR, equivalent ks,opt values were measured using TE and TM polarized light. This result is consistent with a narrow tilt angle distribution in Pc 5 films and a consequently narrow range of tunneling distances. Ionization energies and vacuum level shifts of Pc 5/ITO were measured using CV and UPS. The results show that the Pc 5 film does not present a barrier for hole extraction at the ITO/organic monolayer interface, indicating that Pc 5 is a viable surface modifier for an OPV based on a Pc donor layer. 29220

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces



(6) Ratcliff, E. L.; Zacher, B.; Armstrong, N. R. Selective Interlayers and Contacts in Organic Photovoltaic Cells. J. Phys. Chem. Lett. 2011, 2, 1337−1350. (7) Gliboff, M.; Sang, L.; Knesting, K. M.; Schalnat, M. C.; Mudalige, A.; Ratcliff, E. L.; Li, H.; Sigdel, A. K.; Giordano, A. J.; Berry, J. J.; Nordlund, D.; Seidler, G. T.; Brédas, J. L.; Marder, S. R.; Pemberton, J. E.; Ginger, D. S. Orientation of Phenylphosphonic Acid SelfAssembled Monolayers on A Transparent Conductive Oxide: A Combined NEXAFS, PM-IRRAS, and DFT Study. Langmuir 2013, 29, 2166−2174. (8) Hotchkiss, P. J.; Li, H.; Paramonov, P. B.; Paniagua, S. A.; Jones, S. C.; Armstrong, N. R.; Brédas, J.-L.; Marder, S. R. Modification of the Surface Properties of Indium Tin Oxide with Benzylphosphonic Acids: A Joint Experimental and Theoretical Study. Adv. Mater. 2009, 21, 4496−4501. (9) Lin, H.-C.; MacDonald, G. A.; Shi, Y.; Polaske, N. W.; McGrath, D. V.; Marder, S. R.; Armstrong, N. R.; Ratcliff, E. L.; Saavedra, S. S. Influence of Molecular Orientation on Charge-Transfer Processes at Phthalocyanine/Metal Oxide Interfaces and Relationship to Organic Photovoltaic Performance. J. Phys. Chem. C 2015, 119, 10304−10313. (10) Paniagua, S. A.; Giordano, A. J.; Smith, O. L.; Barlow, S.; Li, H.; Armstrong, N. R.; Pemberton, J. E.; Bredas, J. L.; Ginger, D.; Marder, S. R. Phosphonic Acids for Interfacial Engineering of Transparent Conductive Oxides. Chem. Rev. 2016, 116, 7117−7158. (11) Paniagua, S. A.; Li, E. L.; Marder, S. R. Adsorption Studies of a Phosphonic Acid on ITO: Film Coverage, Purity, and Induced Electronic Structure Changes. Phys. Chem. Chem. Phys. 2014, 16, 2874−2881. (12) Brumbach, M.; Veneman, P. A.; Marrikar, F. S.; Schulmeyer, T.; Simmonds, A.; Xia, W.; Lee, P.; Armstrong, N. R. Surface Composition and Electrical and Electrochemical Properties of Freshly Deposited and Acid-Etched Indium Tin Oxide Electrodes. Langmuir 2007, 23, 11089−11099. (13) MacDonald, G. A.; Veneman, P. A.; Placencia, D.; Armstrong, N. R. Electrical Property Heterogeneity at Transparent Conductive Oxide/Organic Semiconductor Interfaces: Mapping Contact Ohmicity Using Conducting-Tip Atomic Force Microscopy. ACS Nano 2012, 6, 9623−9636. (14) Bel Hadj Tahar, R.; Ban, T.; Ohya, Y.; Takahashi, Y. Tin Doped Indium Oxide Thin Films: Electrical Properties. J. Appl. Phys. 1998, 83, 2631−2645. (15) Dhere, R. G.; Gessert, T. A.; Schilling, L. L.; Nelson, A. J.; Jones, K. M.; Aharoni, H.; Coutts, T. J. Electro-optical Properties of Thin Indium Tin Oxide Films: Limitations on Performance. Sol. Cells 1987, 21, 281−290. (16) Ishibashi, S.; Higuchi, Y.; Ota, Y.; Nakamura, K. Low Resistivity Indium−Tin Oxide Transparent Conductive Films. II. Effect of Sputtering Voltage on Electrical Property of Films. J. Vac. Sci. Technol., A 1990, 8, 1403. (17) Shallcross, R. C.; Stubhan, T.; Ratcliff, E. L.; Kahn, A.; Brabec, C. J.; Armstrong, N. R. Quantifying the Extent of Contact Doping at the Interface between High Work Function Electrical Contacts and Poly(3-hexylthiophene) (P3HT). J. Phys. Chem. Lett. 2015, 6, 1303− 1309. (18) Zhang, H.; Shallcross, R. C.; Li, N.; Stubhan, T.; Hou, Y.; Chen, W.; Ameri, T.; Turbiez, M.; Armstrong, N. R.; Brabec, C. J. Overcoming Electrode-Induced Losses in Organic Solar Cells by Tailoring a Quasi-Ohmic Contact to Fullerenes via Solution-Processed Alkali Hydroxide Layers. Adv. Energy Mater. 2016, 6, 1502195. (19) Koh, S. E.; McDonald, K. D.; Holt, D. H.; Dulcey, C. S.; Chaney, J. A.; Pehrsson, P. E. Phenylphosphonic Acid Functionalization of Indium Tin Oxide: Surface Chemistry and Work Functions. Langmuir 2006, 22, 6249−6255. (20) Li, H.; Paramonov, P.; Bredas, J.-L. Theoretical Study of the Surface Modification of Indium Tin Oxide with Trifluorophenyl Phosphonic Acid Molecules: Impact of Coverage Density and Binding Geometry. J. Mater. Chem. 2010, 20, 2630−2637. (21) Sharma, A.; Kippelen, B.; Hotchkiss, P. J.; Marder, S. R. Stabilization of the Work Function of Indium Tin Oxide Using

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b07394. Detailed synthesis procedures; NMR characterization; spectral and electrochemical characterization of dissolved RuPcs; Pc 5 film preparation and adsorption kinetics; electrochemical, spectroelectrochemical, ATR, EIS, PMATR, XPS, and UPS characterization of Pc 5 films; and methods to determine the mean tilt angle in Pc 5 films and model the influence of ITO surface roughness (PDF)



AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

S. Scott Saavedra: 0000-0002-9946-2664 Author Contributions †

(R.E., L.E.O.) These authors contributed equally.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was supported as part of the Center for Interface Science: Solar-Electric Materials (CIS:SEM), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0001084. Authors thank Dr. Arpad Somogyi of the University of Arizona, Chemistry and Biochemistry Department Mass Spectrometry Facility, for his assistance. All Multimode AFM images and data were collected in the W.M. Keck Center for Nano-Scale Imaging in the Department of Chemistry and Biochemistry at the University of Arizona. This instrument purchase was supported by Arizona Technology and Research Initiative Fund (A.R.S.§15-1648). Portions of this research have been published as part of the Ph.D. dissertation of Ramanan Ehamparam.70



REFERENCES

(1) Kippelen, B.; Brédas, J. L. Organic Photovoltaics. Energy Environ. Sci. 2009, 2, 251−261. (2) Collins, S. D.; Ran, N. A.; Heiber, M. C.; Nguyen, T.-C. Small is Powerful: Recent Progress in Solution-Processed Small Molecule Solar Cells. Adv. Energy Mater. 2017, 7, 1602242. (3) Heremans, P.; Cheyns, D.; Rand, B. P. Strategies for Increasing the Efficiency of Heterojunction Organic Solar Cells: Material Selection and Device Architecture. Acc. Chem. Res. 2009, 42, 1740− 1747. (4) Lin, H.-C.; Polaske, N. W.; Oquendo, L. E.; Gliboff, M.; Knesting, K. M.; Nordlund, D.; Ginger, D. S.; Ratcliff, E. L.; Beam, B. M.; Armstrong, N. R.; McGrath, D. V.; Saavedra, S. S. Electron-Transfer Processes in Zinc Phthalocyanine−Phosphonic Acid Monolayers on ITO: Characterization of Orientation and Charge-Transfer Kinetics by Waveguide Spectroelectrochemistry. J. Phys. Chem. Lett. 2012, 3, 1154−1158. (5) Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.; Leclerc, M.; Lee, K.; Heeger, A. J. Bulk Heterojunction Solar Cells With Internal Quantum Efficiency Approaching 100%. Nat. Photonics 2009, 3, 297−302. 29221

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces Organic Surface Modifiers in Organic Light-Emitting Diodes. Appl. Phys. Lett. 2008, 93, 163308. (22) Wang, H.; Gomez, E. D.; Guan, Z.; Jaye, C.; Toney, M. F.; Fischer, D. A.; Kahn, A.; Loo, Y.-L. Tuning Contact Recombination and Open-Circuit Voltage in Polymer Solar Cells via Self-Assembled Monolayer Adsorption at the Organic−Metal Oxide Interface. J. Phys. Chem. C 2013, 117, 20474−20484. (23) Zhang, Y.; Galoppini, E. Organic Polyaromatic Hydrocarbons as Sensitizing Model Dyes for Semiconductor Nanoparticles. ChemSusChem 2010, 3, 410−428. (24) Myahkostupov, M.; Piotrowiak, P.; Wang, D.; Galoppini, E. Ru(II)-Bpy Complexes Bound to Nanocrystalline TiO2 Films through Phenyleneethynylene (OPE) Linkers: Effect of the Linkers Length on Electron Injection Rates. J. Phys. Chem. C 2007, 111, 2827−2829. (25) Zheng, Y.; Jradi, F. M.; Parker, T. C.; Barlow, S.; Marder, S. R.; Saavedra, S. S. Influence of Molecular Aggregation on Electron Transfer at the Perylene Diimide/Indium-Tin Oxide Interface. ACS Appl. Mater. Interfaces 2016, 8, 34089−34097. (26) Zheng, Y. L.; Giordano, A. J.; Shallcross, R. C.; Fleming, S. R.; Barlow, S.; Armstrong, N. R.; Marder, S. R.; Saavedra, S. S. Surface Modification of Indium Tin Oxide with Functionalized Perylene Diimides: Characterization of Orientation, Electron-Transfer Kinetics and Electronic Structure. J. Phys. Chem. C 2016, 120, 20040−20048. (27) Claessens, C. G.; Hahn, U.; Torres, T. Phthalocyanines: From Outstanding Electronic Properties to Emerging Applications. Chem. Rec. 2008, 8, 75−97. (28) Singh, V. K.; Kanaparthi, R. K.; Giribabu, L. Emerging Molecular Design Strategies of Unsymmetrical Phthalocyanines for Dye-Sensitized Solar Cell Applications. RSC Adv. 2014, 4, 6970−6984. (29) Polaske, N. W.; Lin, H.-C.; Tang, A.; Mayukh, M.; Oquendo, L. E.; Green, J. T.; Ratcliff, E. L.; Armstrong, N. R.; Saavedra, S. S.; McGrath, D. V. Phosphonic Acid Functionalized Asymmetric Phthalocyanines: Synthesis, Modification of Indium Tin Oxide, and Charge Transfer. Langmuir 2011, 27, 14900−14909. (30) Rawling, T.; Austin, C.; Buchholz, F.; Colbran, S. B.; McDonagh, A. M. Ruthenium Phthalocyanine-Bipyridyl Dyads as Sensitizers for Dye-Sensitized Solar Cells: Dye Coverage versus Molecular Efficiency. Inorg. Chem. 2009, 48, 3215−3227. (31) Rawling, T.; Austin, C.; Hare, D.; Doble, P.; Zareie, H.; McDonagh, A. Thin Films of Ruthenium Phthalocyanine Complexes. Nano Res. 2009, 2, 678−687. (32) Rawling, T.; McDonagh, A. Ruthenium Phthalocyanine and Naphthalocyanine Complexes: Synthesis, Properties and Applications. Coord. Chem. Rev. 2007, 251, 1128−1157. (33) Rawling, T.; Xiao, H.; Lee, S.-T.; Colbran, S. B.; McDonagh, A. M. Optical and Redox Properties of Ruthenium Phthalocyanine Complexes Tuned with Axial Ligand Substituents. Inorg. Chem. 2007, 46, 2805−2813. (34) Yanagisawa, M.; Korodi, F.; He, J.; Sun, L.; Sundström, V.; Åkermark, B. Ruthenium Phthalocyanines with Axial Carboxylate Ligands: Synthesis and Function in Solar Cells Based on Nanocrystalline TiO2. J. Porphyrins Phthalocyanines 2002, 06, 217−224. (35) Yang, X.; Kritikos, M.; Åkermark, B.; Sun, L. Axial Ligand Exchange Reaction on Ruthenium Phthalocyanines. J. Porphyrins Phthalocyanines 2005, 09, 248−255. (36) Huc, V.; Armand, F.; Bourgoin, J. P.; Palacin, S. Covalent Anchoring of Phthalocyanines on Silicon Dioxide Surfaces: Building up Mono- and Multilayers. Langmuir 2001, 17, 1928−1935. (37) Huc, V.; Bourgoin, J.-P.; Bureau, C.; Valin, F.; Zalczer, G.; Palacin, S. Self-Assembled Mono- and Multilayers on Gold from 1,4Diisocyanobenzene and Ruthenium Phthalocyanine. J. Phys. Chem. B 1999, 103, 10489−10495. (38) Li, X.; Xu, W.; Wang, X.; Jia, H.; Zhao, B.; Li, B.; Ozaki, Y. Ultraviolet-visible and Surface-enhanced Raman Scattering Spectroscopy Studies on Self-Assembled Films of Ruthenium Phthalocyanine on Organic Monolayer-Modified Silver Substrates. Thin Solid Films 2004, 457, 372−380. (39) Nazeeruddin, M. K.; Humphry-Baker, R.; Grätzel, M.; Nazeeruddin, M. K.; Humphry-Baker, R.; Grätzel, M.; Murrer, B. A.

Efficient Near IR Sensitization of Nanocrystalline TiO2 Films by Ruthenium Phthalocyanines. Chem. Commun. 1998, 719−720. (40) Morandeira, A.; Lopez-Duarte, I.; O’Regan, B.; Martinez-Diaz, M. V.; Forneli, A.; Palomares, E.; Torres, T.; Durrant, J. R. Ru(II)Phthalocyanine Sensitized Solar Cells: The Influence of CoAdsorbents Upon Interfacial Electron Transfer Kinetics. J. Mater. Chem. 2009, 19, 5016−5026. (41) Capobianchi, A.; Tucci, M. Ruthenium Phthalocyanine Thin Films for Photovoltaic Applications. Thin Solid Films 2004, 451−452, 33−36. (42) Fischer, M. K. R.; López-Duarte, I.; Wienk, M. M.; MartínezDíaz, M. V.; Janssen, R. A. J.; Bäuerle, P.; Torres, T. Functionalized Dendritic Oligothiophenes: Ruthenium Phthalocyanine Complexes and Their Application in Bulk Heterojunction Solar Cells. J. Am. Chem. Soc. 2009, 131, 8669−8676. (43) Doherty, W. J.; Donley, C. L.; Armstrong, N. R.; Saavedra, S. S. A Broadband Spectroelectrochemical ATR Instrument for Molecular Adlayer Studies. Appl. Spectrosc. 2002, 56, 920−927. (44) Doherty, W. J.; Wysocki, R. J.; Armstrong, N. R.; Saavedra, S. S. Potential-Modulated, Attenuated Total Reflectance Spectroscopy of Poly(3,4-ethylenedioxythiophene) and Poly(3,4-ethylenedioxythiophene Methanol) Copolymer Films on Indium−Tin Oxide. J. Phys. Chem. B 2006, 110, 4900−4907. (45) Araci, Z. O.; Runge, A. F.; Doherty, W. J.; Saavedra, S. S. Correlating Molecular Orientation Distributions and Electrochemical Kinetics in Subpopulations of an Immobilized Protein Film. J. Am. Chem. Soc. 2008, 130, 1572−1573. (46) Guez, D.; Markovitsi, D.; Sommerauer, M.; Hanack, M. Photophysical Properties of a Ruthenium(II) Phthalocyanine. Chem. Phys. Lett. 1996, 249, 309−313. (47) Mendes, S. B.; Bradshaw, J. T.; Saavedra, S. S. Technique for Determining the Angular Orientation of Molecules Bound to the Surface of an Arbitrary Planar Optical Waveguide. Appl. Opt. 2004, 43, 70−78. (48) Gliboff, M.; Li, H.; Knesting, K. M.; Giordano, A. J.; Nordlund, D.; Seidler, G. T.; Brédas, J.-L.; Marder, S. R.; Ginger, D. S. Competing Effects of Fluorination on the Orientation of Aromatic and Aliphatic Phosphonic Acid Monolayers on Indium Tin Oxide. J. Phys. Chem. C 2013, 117, 15139−15147. (49) Simpson, G. J.; Rowlen, K. L. Quantification of “Local” Surface Orientation: Theory and Experiment. J. Phys. Chem. B 1999, 103, 1525−1531. (50) Dolphin, D.; James, B. R.; Murray, A. J.; Thornback, J. R. Synthetic and Oxidation Studies of Ruthenium(II) Phthalocyanine Complexes. Can. J. Chem. 1980, 58, 1125−1132. (51) Clark, R. A.; Bowden, E. F. Voltammetric Peak Broadening for Cytochrome c/Alkanethiolate Monolayer Structures: Dispersion of Formal Potentials. Langmuir 1997, 13, 559−565. (52) Finklea, H. O.; Liu, L.; Ravenscroft, M. S.; Punturi, S. Multiple Electron Tunneling Paths Across Self-Assembled Monolayers of Alkanethiols with Attached Ruthenium(II/III) Redox Centers. J. Phys. Chem. 1996, 100, 18852−18858. (53) Armstrong, N. R.; Veneman, P. A.; Ratcliff, E.; Placencia, D.; Brumbach, M. Oxide Contacts in Organic Photovoltaics: Characterization and Control of Near-Surface Composition in Indium-Tin Oxide (ITO) Electrodes. Acc. Chem. Res. 2009, 42, 1748−1757. (54) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; John Wiley & Sons: New York, 2001; Chapter 14. (55) Li, Z.; Lieberman, M.; Hill, W. XPS and SERS Study of Silicon Phthalocyanine Monolayers: Umbrella vs Octopus Design Strategies for Formation of Oriented SAMs. Langmuir 2001, 17, 4887−4894. (56) Bradshaw, J. T.; Mendes, S. B.; Armstrong, N. R.; Saavedra, S. S. Broadband Coupling Into a Single-Mode, Electroactive, Planar Integrated Optical Waveguide for Spectroelectrochemical Analysis of Surface Confined Redox Couples. Anal. Chem. 2003, 75, 1080−1088. (57) Silva, B. P. G.; de Florio, D. Z.; Brochsztain, S. Characterization of a Perylenediimide Self-Assembled Monolayer on Indium Tin Oxide 29222

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223

Research Article

ACS Applied Materials & Interfaces Electrodes Using Electrochemical Impedance Spectroscopy. J. Phys. Chem. C 2014, 118, 4103−4112. (58) Bueno, P. R.; Mizzon, G.; Davis, J. J. Capacitance Spectroscopy: A Versatile Approach To Resolving the Redox Density of States and Kinetics in Redox-Active Self-Assembled Monolayers. J. Phys. Chem. B 2012, 116, 8822−8829. (59) Nahir, T. M.; Bowden, E. F. Measurement of the Rate of Adsorption of Electroactive Cytochrome c to Modified Gold Electrodes by Electrochemical Impedance Spectroscopy. Langmuir 2002, 18, 5283−5286. (60) Feng, Z. Q.; Imabayashi, S.; Kakiuchi, T.; Niki, K. Determination of the Electrode Kinetic Parameters of a Species Immobilized on Electrodes Using the Electroreflectance (ER) Voltammogram. J. Electroanal. Chem. 1996, 408, 15−20. (61) Feng, Z. Q.; Sagara, T.; Niki, K. Application of PotentialModulated UV-Visible Reflectance Spectroscopy to Electron Transfer Rate Measurements for Adsorbed Species on Electrode Surfaces. Anal. Chem. 1995, 67, 3564−3570. (62) Zheng, Y.; Saavedra, S. S. Characterization of Charge-Transfer Kinetics at Organic/Electrode Interfaces Using Potential-modulated Attenuated Total Reflectance (PM-ATR) Spectroscopy. Anal. Sci. 2017, 33, 427−433. (63) Kelly, C. P.; Cramer, C. J.; Truhlar, D. G. Single-Ion Solvation Free Energies and the Normal Hydrogen Electrode Potential in Methanol, Acetonitrile, and Dimethyl Sulfoxide. J. Phys. Chem. B 2007, 111, 408−422. (64) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C. Electrochemical Considerations for Determining Absolute Frontier Orbital Energy Levels of Conjugated Polymers for Solar Cell Applications. Adv. Mater. 2011, 23, 2367−2371. (65) Liu, G.; Klein, A.; Thissen, A.; Jaegermann, W. Electronic Properties and Interface Characterization of Phthalocyanine and RuPolypyridine Dyes On TiO2 Surface. Surf. Sci. 2003, 539, 37−48. (66) Tong, B.; Yang, H.; Xiong, W.; Xie, F.; Shi, J.; Zhi, J.; Chan, W. K.; Dong, Y. Controlled Fabrication and Optoelectrical Properties of Metallosupramolecular Films Based on Ruthenium(II) Phthalocyanines and 4,4′-Bipyridine Covalently Anchored on Inorganic Substrates. J. Phys. Chem. B 2013, 117, 5338−5344. (67) Ehamparam, R.; Pavlopoulos, N. G.; Liao, M. W.; Hill, L. J.; Armstrong, N. R.; Pyun, J.; Saavedra, S. S. Band Edge Energetics of Heterostructured Nanorods: Photoemission Spectroscopy and Waveguide Spectroelectrochemistry of Au-Tipped CdSe Nanorod Monolayers. ACS Nano 2015, 9, 8786−8800. (68) Li, X.; Zhang, Z.; Henrich, V. E. Inelastic Electron Background Function for Ultraviolet Photoelectron Spectra. J. Electron Spectrosc. Relat. Phenom. 1993, 63, 253−265. (69) Schlettwein, D.; Hesse, K.; Gruhn, N. E.; Lee, P. A.; Nebesny, K. W.; Armstrong, N. R. Electronic Energy Levels in Individual Molecules, Thin Films, and Organic Heterojunctions of Substituted Phthalocyanines. J. Phys. Chem. B 2001, 105, 4791−4800. (70) Ehamparam, R. Band Edge Energetics and Charge Transfer Processes in Semiconductor-Metal Heterostructured Nanorods as Photocatalysts and Metal Oxide Electrode-Organic Semiconductor Interfaces in Organic Photovoltaics. Ph.D. Dissertation, University of Arizona, 2015. http://hdl.handle.net/10150/566213.

29223

DOI: 10.1021/acsami.7b07394 ACS Appl. Mater. Interfaces 2017, 9, 29213−29223