Influence of Electrode Surface Composition and Energetics on Small

Dec 19, 2012 - The potential applied to the device was swept by use of a source meter (Keithley 2400) and in-house software (National Instruments Labv...
0 downloads 0 Views 894KB Size
Subscriber access provided by UNIV OF ARIZONA

Article

Influence of Electrode Surface Composition and Energetics on Small-Molecule Organic Solar Cell Performance: Polar Versus Non-Polar Donors on ITO Contacts Jeremy Gantz, Diogenes Placencia, Anthony J Giordano, Seth R. Marder, and Neal R. Armstrong J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/jp307546v • Publication Date (Web): 19 Dec 2012 Downloaded from http://pubs.acs.org on January 4, 2013

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

Influence of Electrode Surface Composition and Energetics on Small-Molecule Organic Solar Cell Performance: Polar Versus Non-Polar Donors on ITO Contacts

Jeremy Gantz,1 Diogenes Placencia,1 Anthony Giordano,2 Seth R. Marder,2 and Neal R. Armstrong1*

1 = Department of Chemistry & Biochemistry, University of Arizona, Tucson, Arizona 85721 2 = School of Chemistry & Biochemistry, and the Center for Organic Photonics and Electronics, Georgia Institute of Technology, Atlanta Georgia 30332-0400 * - [email protected]

Keywords: organic solar cell, phthalocyanine, ITO contacts, UV-photoemission spectroscopy, organic heterojunctions

     

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

ABSTRACT We present a systematic study of the band-edge energy offsets and shifts in local vacuum levels for indium tin oxide (ITO)/donor heterojunctions, using vacuum-deposited chloro-indium phthalocyanine (ClInPc), titanyl phthalocyanine (TiOPc), pentacene (PEN), and copper phthalocyanine (CuPc) donor layers. We include a comparison of the performance of ITO/donor/C60 planar heterojunction (PHJ) organic solar cells (OPVs) as a function of activation and modification of the ITO contact. UV-photoemission spectroscopy (UPS) was used to characterize the interfacial region between these donors and ITO to infer the ordering of these first deposited 1-2 monolayers as a function of ITO activation and modification. For the polar donors ClInPc and TiOPc, deposited on air-plasma (AP) treated, high work function ITO (φeff ≈ 5.1-5.2 eV), shifts in local vacuum level observed during deposition of the first two monolayers of donor molecules suggest that the halo-metal or oxo-metal bond points (on average) toward the oxide surface in the first monolayer. Inversion of this orientation is inferred during formation of the second monolayer. Thin film absorbance data for the Q-band spectra of ClInPc or TiOPc 1-2 monolayer on properly AP-ITO confirmed the organization of the first deposited layers. Using non-activated ITO contacts, or contacts modified with pentafluoro benzyl phosphonic acid (F5BPA), disrupts this order in the first two monolayers as revealed in the UPS and absorbance data, leading to decreases in OPV open-circuit photopotential (VOC), fill factor (FF) and efficiency (η). For nonpolar donors like PEN and CuPc, shifts in local vacuum level as a function of donor coverage were less sensitive to ITO surface composition, and PEN-based OPVs show relatively invariant VOC, JSC, FF and η, regardless of ITO pretreatment or F5BPA modification. The interaction of ClInPc donor layers with AP-ITO is strong enough to generate photovoltaic

     

ACS Paragon Plus Environment

2

Page 3 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

activity in the absence of a C60 acceptor layer (ITO/ClInPc/Al device configurations), where photocurrent appears to be generated at the ITO/ClInPc interface. Such an effect is not seen for ITO/ClInPc/Al devices on non-activated ITO substrates, highlighting the importance of the interactions between the donor layer and the hole-harvesting interface in PHJ OPVs. These results suggest that the organization of a dipolar donor layer at a hole-harvesting oxide contact can affect OPV performance where charge harvesting efficiency is sensitive to molecular organization in the interfacial region near the contact. Table of Contents Figure

     

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

INTRODUCTION Organic photovoltaic solar cells (OPVs) have shown steady improvements in efficiency, both for blended heterojunction (BHJ) polymer/small molecule and for planar heterojunction (PHJ) small molecule OPVs, with research cells now routinely exceeding 4-8%.1-15 Interface composition and energetic offsets, between the hole- and electron-harvesting contacts and the active layers, are increasingly seen as critical to the performance of OPVs.16-24 The best OPV performance results when the efficiencies of hole- and electron-harvesting at opposing contacts are high and balanced, combined with low surface recombination rates.16,17,20 While interlayer films can be added at both contacts to enhance efficiency for collection of one type of charge carrier,6,18-20 molecular-level interactions between certain donor or acceptor molecules, and the contact or the interlayer, are predicted to have an impact upon the efficiency of charge harvesting, interface recombination, and overall OPV performance.16 Previous studies from this group have explored the use of strongly dipolar trivalent metal and tetravalent metal phthalocyanines (Pcs), such as titanyl phthalocyanine (TiOPc) and chloroindium phthalocyanine

(ClInPc),

as

donor materials in both planar (Pc/C60

heterojunctions) and textured heterojunction format OPVs. Solvent vapor annealing of the asdeposited Pc layer has been used to texture the donor layer and to enhance the near-IR response, producing up to 5 enhancements in both short-circuit photocurrent (JSC) and overall OPV device performance.13,25-27 16,17,20,28 Rand and coworkers, using combinations of X-ray diffraction and vibrational spectroscopies, have also recently noted the dependence of OPV performance on molecular structure for this same type of donor material. 14,15

     

ACS Paragon Plus Environment

4

Page 5 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

Other groups have been interested in the molecular electronic properties of these Pcs.  29‐32 Field effect transistors have been developed using TiOPc as the channel material in organic field effect transistors, to study the effects of molecular packing on field-effect hole mobility, where differences in Pc packing architecture were achieved using surface modified Si/SiO2 substrates in the channel region.30 When deposited on octadecyltrichlorosilane (OTS) treated Si/SiO2 substrates at elevated temperatures, the triclinic Phase II polymorph of TiOPc is preferentially formed, leading to hole mobilities in the range of 2-4 cm2V-1S-1, 3 orders of magnitude higher than seen in transistors using the Phase I polymorph. 30,31 It is clear that the interactions of these dipolar Pc layers with substrates and contacts can influence both molecular packing and ultimately device performance. In our studies we have noted a significant sensitivity of OPV performance to the composition and surface activation of the hole-harvesting contact, indium tin oxide (ITO).  18,19,33,34 As shown here, this sensitivity to contact surface composition is not strictly dependent upon the effective work function of the contact. Recent experimental and modeling studies of ITO contacts have also shown the importance of compositional and energetic heterogeneity in determining the efficiency of hole-harvesting and OPV performance.

18,19,34,35

It is clear that even for freshly cleaned and activated ITO

contacts and interlayers: (i) only a small fraction of the ITO surface may be electrically active; and (ii) charge mobilities in the donor layer, or in an interlayer immediately adjacent to the contact, play a significant role in determining the probability that charges will be harvested before undergoing recombination.

36-39

For contacts such as ITO, and transparent conducting

oxides and interlayer materials in general, the molecular organization of the donor or acceptor

     

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

layer at the metal oxide/active layer interface may be one of several critical determinants of OPV efficiency. Excessive heterogeneity in composition and energetics is likely to enhance surface recombination, limiting OPV efficiency.  20,40   As we demonstrated recently in modeling studies, donor layers immediately adjacent to the hole-harvesting contact that lead to high charge mobilities, can mitigate many of the effects of electrical heterogeneity in the contact. 18 Charges that arrive at an electrically inactive region can build in local concentration and diffuse laterally in these adjacent layers until reaching an electrically active region. This process competes with recombination to an extent dictated by the arrival rate of these charges at the contact, the separation between electrically active regions, and the charge mobility. Based on these observations we hypothesize that ordered donor layers, adjacent to the hole-harvesting contact, will lead to higher charge harvesting efficiencies than disordered layers, and for certain Pc donor layers, where charge mobilities may be intrinsically low, this may be an important consideration in OPV performance. This paper focuses on a detailed study of the composition and energetics of the first few monolayers of ClInPc, TiOPc, CuPc and pentacene (PEN) deposited on ITO electrodes, which have undergone different degrees of cleaning, pre-treatment and chemical modification. We use UV-photoemission spectroscopies (UPS), visible wavelength absorbance, and the performance of PHJ (donor/C60) OPVs to probe the nature of the interaction between ClInPc or TiOPc donor layers and ITO contacts. These results are compared with studies of the non-polar donors copper phthalocyanine (CuPc) and PEN. OPVs based on the two polar Pc donors are sensitive to the near-surface composition of ITO. UPS studies suggest that the first monolayer of ClInPc or TiOPc molecules deposited on high work function (ca. 5.1 eV) air-plasma (AP) activated ITO,

     

ACS Paragon Plus Environment

6

Page 7 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

adopt orientations with the oxo-metal or halo-metal bond pointed toward the activated ITO surface (Figure 1). For the second monolayer of ClInPc or TiOPc molecules, the bond dipoles appear to reverse orientation with concomitant changes in local vacuum level. The trivalent and tetravalent metal Pcs have internal dipole moments as high as 3 Debye. UPS studies by Ueno and coworkers have shown that ordered TiOPc monolayers and bilayers, and ordered monolayers and bilayers of related asymmetric Pcs such as chloro-aluminum phthalocyanine (ClAlPc), adopt orientations on highly ordered pyrolytic graphite (HOPG) that produce opposed oxo-metal or halo-metal bond dipoles in first-deposited ordered bilayers, and that these architectures lead to changes in local vacuum level reflecting changes in the internal dipole orientation.41-47 When deposited on HOPG, however, these Pcs typically adopt a flat lying orientation in the first monolayer, with the oxo-metal or halo-metal bond perpendicular to the substrate plane, while the second layer positions this bond in the opposite orientation. We show here that when ordering occurs in the first ClInPc layer, as indicated by sharp reversals in effective work function between the first and second monolayers of Pc, good OPV performance results. If this ordering is not achieved, OPV performance is compromised with decreased VOC and lowered fill-factors (FF), , and increased series resistance (RS), leading to significant decreases in device efficiency (η).16,20,24,28 Modification of AP-ITO electrodes with a dipolar

small

molecule

phosphonic

acid,

pentafluoro-benzylphosphonic2,3,4,5,6-

pentaflourobenzylphosphonic acid (F5BPA) produces an ITO contact with lowered surface free energy and with an effective work function ca. 5.1 eV,48,49 close to the work function achieved on AP-ITO and the ionization potential (IP) of the tetravalent and trivalent metal Pcs. Despite this high work function and wettability by non-polar donor layers, the F5BPA modification leads

     

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

to less organized Pc layers, as indicated in both the UPS results, and in visible spectroscopic characterization of these thin Pc films, and poorer ClInPc/C60 or TiOPc/C60 OPV performance, i.e., on F5BPA-modified ITO the Pc layers cannot form the same kind of organized layers achieved on the OP or AP activated ITO surface. Non-polar donors, such as PEN, appear to be less sensitive to composition and activation of the ITO surface. Monotonic changes in local vacuum level are observed as the donor coverage increases on all ITO contacts, and only small differences are measured in PEN/C60 OPV device performance as the contact surface composition is altered. Recent studies of PEN/C60 OPVs by Sharma et al., using ITO contacts modified with a range of phosphonic acids including F5BPA, showed that OPV performance was insensitive to composition of the phosphonic acid surface modifier as long as the effective work function was above the IP of the PEN donor.50 These results suggest that the interactions between the donor and the ITO contact are weak, and the energetic barrier to charge extraction from the PEN donor to the contact is small.18 We also note that the intrinsic charge mobilities in PEN layers is among the highest observed for most molecular semiconductors,  51,52 which also favors charge collection even at heterogeneous contacts.  18 For polar donors such as ClInPc and TiOPc, however, it appears that efficiencies of hole-harvesting and overall OPV efficiency are much more sensitive to molecular orientation and organization at the metal oxide/donor interface. We also show that the strength of interaction of ClInPc thin films with AP-ITO is sufficient to generate photovoltaic activity in simple ITO/ClInPc/Al devices, where no acceptor layer (C60) is present. These results are reminiscent of Schottky-diode-like OPVs generated from single dye layers prior to the introduction of donor/acceptor PHJ OPVs, in which exciton dissociation and

     

ACS Paragon Plus Environment

8

Page 9 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

photocurrent generation was believed to occur at the dye/contact interface.53-55 For these simpler devices based on less polar donor layers (e.g., ITO/CuPc/Al), negligible photocurrent generation is seen at the ITO/CuPc interface. For ITO/ClInPc devices based on non-activated ITO surfaces the photocurrent generation at the oxide/donor interface is greatly suppressed. These observations are likely to be important in OPV platforms where either donor or acceptor phases have large internal dipoles which induce specific interactions with the contact electrodes, which must be controlled to achieve optimum device performance. EXPERIMENTAL UV-photoemission spectroscopy (UPS, He I excitation at 21.2 eV) was used to probe the interface between ITO and the donor materials (Kratos Axis-Ultra spectrophotometer). The sample was biased at -9.0 volts to enhance the yield of low kinetic energy electrons in the UPS data.56 The spot size was 1 mm to 3 mm. We plot the UPS data on the binding energy scale.13,25,56,57 Work functions on clean Au were first established for each day of spectrometer use, calculated according to φ = 21.2 eV – (HKE – LKE), where HKE is the high kinetic energy edge (photoemission from the Fermi level for clean Au), and LKE is the low kinetic energy emission edge.  44,58 The work function of atomically clean Au foils (Alfa Aesar) was verified to be ca. 5.1 eV, and the absolute kinetic energy (KE) of the Au Fermi edge was frequently measured to ensure instrument calibration was sustained. We assume ITO substrates are in electronic equilibrium with the spectrometer, i.e. that the Fermi energy for the ITO substrate is at the same absolute kinetic energy as Fermi edge emission for the Au sample. Work functions are then computed for all subsequent heterojunctions on ITO

     

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

using this Fermi energy as the HKE.48,49 For determination of ionization potential (IP) we use the high kinetic energy edge associated with the first recognizable photoemission feature of the organic layer, typically ionization of the C 2pz orbital in the molecular semiconductor, and the difference between this energy and the LKE, and the source energy, to compute IP. This protocol has been documented in several publications.13,25,56-59 Donor molecules were deposited on ITO-coated glass (Colorado Concept Coating LLC) in a high-vacuum deposition system attached to the photoelectron spectrometer, the deposition rate was monitored using a quartz crystal monitor (Agilent 53131A) and custom-designed quartz oscillator (QCM-Newark). Deposition rates of less than ca. 1 Å/sec were achievable on either Au or ITO substrates. The ITO coating was 150 nm thick with a sheet resistance of 15 Ω/sq. (fourpoint probe). 1-in. by 1-in. substrates were cleaned by: (1.) washing with deionized water and scrubbing with 10% Triton X-100 aqueous solution; (2.) sonicating in 10% Triton X-100 aqueous solution for 15 min., ; (3.) rinsing and scrubbing with Nanopure (18 MΩ) water; (4.) sonicating in Nanopure water for 15 minutes; (5.) rinsing and scrubbing in 100% ethanol; and (6.) sonicating in absolute-proof ethanol for 15 minutes. This six-step treatment protocol leads to what we term DSC-ITO surfaces.34,35 For ITO substrates used for OPVs, 1 in. by 1 in. sections were rinsed with absolute ethanol and then patterned with S1813 positive photoresist (Rohm and Haas), followed by exposure, development, etching, and removal of the photoresist to create the necessary patterns to prepare OPVs with an active area of 0.125 cm2. Each patterned ITO substrate was etched using aqua regia (3:1 ratio) at 120 °C for approximately 30 s. Upon removal of the photoresist the substrates were cleaned using the six-step procedure described above, and stored in absolute ethanol. Prior

     

ACS Paragon Plus Environment

10

Page 11 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

to introduction to the vacuum systems, DSC-ITO substrates were removed from the ethanol solution and dried under a stream of nitrogen. DSC-ITO substrates were activated in a plasma generator (Harrick PDC-32G) in either dry oxygen or dry air at approximately 0.4 Torr, with etching for 10.5 minutes in oxygen or 8.5 minutes in air to produce OP- or AP-ITO substrates respectively. TiOPc, C60, CuPc, ClInPc, and PEN (Mer Corp, Aldrich) were all triply sublimed and bathocuproine (BCP, Aldrich) was sublimed twice. All molecules were thoroughly degassed in the vacuum system prior to use. Aluminum top contacts (Alfa Aesar) were deposited with 99.999% rated material. A custom high vacuum deposition system was used to prepare the various OPV test cells used in this study, as described previously.13,25,26 All molecules used for vacuum deposition were loaded into a Knudsen-type cell and deposited at equal to or less than 9  10-7 Torr. Thin films were deposited at a rate of approximately 1 Å/sec, measured with a 10 MHz quartz crystal microbalance (QCM-Newark) and a frequency monitor (Agilent, model 53131A). Solvent annealing of the TiOPc and ClInPc 1-to-2 monolayer thin films were carried out by placing the films face up in closed containers (Nalgene 4 oz, 125 mL polypropylene) containing saturated chloroform vapor, a 3 hour exposure was used for TiOPc films and a 5 hour exposure was used for ClInPc films. Top contacts were deposited using thermal evaporation, and measured with a 6MHz quartz crystal microbalance (Tangydine) and a frequency monitor (Inficon, model 758-500-G1). F5BPA was prepared as discussed previously, and used to modify ITO surfaces using small variations of the previously described procedures.48,49 F5BPA solutions were prepared in

     

ACS Paragon Plus Environment

11

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

absolute-proof ethanol to a final concentration of 10 mM. Cleaned ITO and AP-activated substrates were submersed in this F5BPA solution and heated to 75 °C for 1 hr. The substrates were then removed and submersed in a 5% v/v solution of triethylamine (Sigma Aldrich) and sonicated for 10 minutes to remove weakly sorbed F5BPA. For test devices, a custom made current-voltage (J/V) system was used to measure six devices per substrate. The device area used for all J/V curves was 0.125 cm2. J/V measurements were performed in a nitrogen filled glovebox (Mbraun Labmaster) where water and oxygen levels were below 0.1 ppm. A current controlled, 300 W Xenon arc lamp (Newport) was used as the light source. The beam path from the source to the sample was filtered using an AM 1.5 filter (Melles Griot) to simulate the solar spectrum. The filtered light was then optically diffused using an engineered diffuser with an output of 40 degrees (Newport). The power density at the surface of the devices was tuned with a flat-response thermopile (Newport) and cross checked with a calibrated silicon photodiode (Newport – Model 818-SL with OD3 Attenuator) to achieve 100 mW/cm2. The potential applied to the device was swept using a source meter (Keithley 2400) and in-house software (National Instruments Labview 8.2). The potential was swept from -1.00 V to 1.50 V using 5.00 mV steps, starting from negative potential. Absorbance measurements were carried out with a UV/Visible spectrophotometer (Agilent Technologies - Model 8453) at 1 and 4 nm intervals, with an integration time of 25 seconds and a range of 400 to 1100 nm.

     

ACS Paragon Plus Environment

12

Page 13 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

RESULTS AND DISCUSSION UPS and absorbance spectroscopies for ClInPc and PEN films on ITO Figures 2, 3 and 4 summarize the UPS characterization of ITO/ClInPc and ITO/PEN heterojunctions, as a function of Pc or PEN coverage on both AP-ITO and DSC-ITO – these two donor materials provided the most striking differences in UPS data and OPV performance and are the primary focus of this presentation. UPS data for ITO/TiOPc and ITO/CuPc heterojunctions are shown in Supporting Information (Figures S1 and S2). The UPS data for deposition of ClInPc on F5BPA-modified ITO is shown in Figure 5 and discussed further below. The effective work function (φeff) was ca. 5.1 – 5.2 eV for AP-ITO in these experiments, and was the starting point to compute changes in φeff as ClInPc coverage increased (Figure 4A). For ClInPc on AP-ITO we see significant decreases in local vacuum level (Δφeff ≈ -0.6 eV) for deposition of approximately one monolayer (as estimated from our thickness monitor, and from the thin film absorbance data discussed below), and an immediate increase in φeff and IP for Pc coverages exceeding one monolayer. The high KE ionization feature corresponding to the highest occupied molecular orbital (HOMO) levels in the Pc are also broadened with increasing Pc coverage suggesting the possibility of more than one environment for Pcs in this film, with slightly different IPs. This broadened ionization peak in the second monolayer can be fit with two components separated in energy by 0.2 eV (see discussion below).44,47,60 Changes in φeff and IP are much less pronounced for deposition of ClInPc on DSC-ITO at low Pc coverages (Figures 2C,D and 4B).

     

ACS Paragon Plus Environment

13

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

These changes in local vacuum level for deposition of the first monolayer of ClInPc are consistent with those expected for a dipolar adsorbate where the positive end of the dipole is, on average, pointed away from the substrate plane (Figure 1); while in the second monolayer, the negative end of the molecular dipole is pointed away from the substrate plane and compensates for some of the change in φeff created by deposition of the first monolayer. 48,49,56,57,61 In Figure 1 we show schematically the first monolayer with the halo-metal bond along the normal axis, using sections of published Phase I and Phase II single crystal structures to suggest orientation and the interrelationship between the two polymorphs in these thin films,62 although we acknowledge that configurations with tilted halo-metal or oxo-metal are quite possible. Thin films of TiOPc, VOPc, and ClAlPc on HOPG also show a coverage dependent shift in both φeff and IP due to changes in the oxo-metal or halo-metal bond dipole orientation, and a splitting of the highest kinetic energy ionization peaks.41,42,44,47,60 These studies, however, show the plane of the Pc lying parallel to the substrate plane, dictated by van der Waals interactions between the non-polar HOPG surface and the Pc, with the polar oxo-metal or halo-metal bond pointed away from the surface in the first monolayer, and an opposed dipole orientation in the second monolayer. Formation of the first ordered monolayer and addition of the second monolayer led to shifts in φeff of ca. +/- 0.35 eV, accompanied by shifts in IP of up to 0.2 eV between first and second monolayers for each Pc, dependent upon the apparent dipole moment of the oxo-metal or halo-metal bond, and the precise alignment of adjacent molecules. Similar effects have recently been documented by Monti and coworkers for sub-monolayer to monolayer coverages of oxo-vanadium naphthalocyanine on HOPG.63

     

ACS Paragon Plus Environment

14

Page 15 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

Our UPS data suggest that the halo-metal bond dipole is, on average, pointed toward the polar oxide surface in the first Pc monolayer. The UPS data by itself, however, is not sufficient to suggest whether the orientation of the Pc macrocycle is parallel to the substrate plane or tilted. Other studies of the orientation of asymmetric Pcs on surfaces more polar than HOPG have suggested a tilted orientation for the first deposited Pc monolayer.64,65 Figure 5 shows the UPS data for ClInPc deposited on F5BPA-modified ITO.

F5BPA

modifies the initial effective surface work function of ITO to φeff = ca. 5.1 eV, nearly the same effective work function seen for AP-ITO, owing to the strong internal molecular dipole in this molecule, and the bond dipole of the phosphonic acid group attaching the modifier to the oxide surface.48,49 φeff decreases with deposition of the first sub-monolayer of ClInPc and then increases, as measured for DSC-ITO; but does not increase with Pc coverage at the rate for deposition on AP-ITO, suggesting that inversion of the Pc dipole orientation on F5BPA-modified ITO does not take place to the same extent as for AP-ITO. OPV behavior on such modified surfaces is negatively impacted as discussed further below. For PEN films on both AP-ITO and DSC-ITO (Figures 3 and 4) we see only monotonic decreases in φeff with increasing PEN film thickness, and no abrupt changes in φeff as the first equivalent monolayer is completed and the second one begins. Similar large shifts in the local vacuum level have been reported previously for the deposition of PEN on conductive substrates, which were attributed to changes in the polarizability of the environment, and the distribution in orientation and packing of adjacent PEN molecules, as coverage increases.51,66-68 For perfluoropentacene thin films, however, Koch et al. did note a significant difference in local

     

ACS Paragon Plus Environment

15

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

vacuum levels (and hole injection barriers) for thin films of this material, owing to the nonplanar configuration of the fluorinated PEN derivative, and in subsequent reports expanded this observation to note differences in IP and local vacuum level that might arise in PEN films in general when their orientation in a thin film is changed.  69    Continuous shifts in vacuum level with increasing CuPc coverage on AP-ITO are shown in the Supporting Information (as is seen for PEN films) where no abrupt changes in φeff versus coverage are seen at low coverages of CuPc. For molecules like PEN and CuPc, gradients in intermolecular electronic interactions along the surface normal have been invoked to explain these vacuum level shifts.51,67,68,70-72 Optical characterization of thin ClInPc/ITO films Significant differences are seen in the absorbance spectra of these ultrathin ClInPc films deposited on AP-ITO versus DSC-ITO substrates. The visible absorbance spectra for ca. 2 monolayers of ClInPc on AP-ITO are shown in Figure 1 (upper right insert) and Figure 6, for both as-deposited and the solvent annealed ClInPc films.  13,14,26 The complete conversion of the Phase I absorbance spectrum to the Phase II polymorph Q-band spectrum with its characteristically split and red-shifted profile, is only achieved for ClInPc films deposited on APITO. Structural differences between the monoclinic Phase I polymorph and the triclinic Phase II polymorph involve primarily the lateral translation of Pcs in one layer relative to the adjacent layer, shortening the c-axis dimension, and bringing adjacent macrocycles into closer proximity, producing new excited state energy levels.  73    As we show in Figure 1, ordering of the first monolayer of deposited Pc is likely to facilitate that transition. On ITO surfaces not supporting Pc growth that leads to the work function changes noted in our UPS data (i.e. ClInPc or TiOPc

     

ACS Paragon Plus Environment

16

Page 17 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

films deposited on either DSC-ITO or AP-ITO modified with F5BPA), we see only a slight narrowing of the Q-band spectrum, with no indication of this polymorphic transition. These absorbance spectra are thus an additional indicator of the proposed ordering in the first 1-2 monolayers of ClInPc and TiOPc on AP-ITO. In the Supporting Information section we include absorbance spectra for ClInPc on AP-ITO ranging from ca. 1. nm up to 4.7 nm film thickness, showing a systematic red-shift in the Q-band maximum. These shifts are similar to earlier observations of shifts in Q-band absorbance during layer-by-layer deposition of well-ordered epitaxial ClInPc and TiOPc ultra-thin films on layered semiconductors, where the formation of dimer-like aggregates, with opposed halo-metal or oxometal bond orientations, leads to a lowering of excitonic transition energy for thicknesses up to ca. 4-to-5 monolayers.29,74-76 Studies to be communicated elsewhere also show that this type of ordering and Phase I to Phase II polymorphic transitions in the first deposited Pc layers can affect the types of morphologies and OPV activity seen in thicker Pc films, in bulkheterojunction OPV platforms. For PEN films on all versions of ITO (Figures 6C versus 6D), we see no significant differences in absorbance spectra, after correction for background absorbance, suggesting only minimal differences in aggregate architecture or degree-of-ordering on the different ITO surfaces. We have previously shown that as-deposited films of both TiOPc and ClInPc, on either OPor AP-ITO, at thicknesses up to 20 nm, are sufficiently conformal that the ITO sub-grain structure persists in AFM and SEM images even at these Pc thicknesses.13,25,26 For thinner

     

ACS Paragon Plus Environment

17

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

ClInPc or TiOPc films, e.g. those at ca. 2ML coverages, regardless of ITO pretreatment, we have not been able to distinguish differences in topography or rms surface roughness, when using ITO films with the roughness of those used in these studies (ca. 1.5 – 3.0 nm rms roughness),26 which is not surprising since the critical interactions occur at just the Pc/ITO interface. OPV performance for ClInPc/C60 and PEN/C60 heterojunctions on activated and chemically modified ITO Device performance for ClInPc/C60 and PEN/C60 OPVs, on AP-ITO and DSC-ITO, is shown in Figure 7, and on F5BPA-modified ITO in Figure 8, with pertinent device parameters summarized in Table 1. For these studies we chose to examine only OPV performance for the as-deposited Pcs, i.e., the Phase I polymorph. Device performance for TiOPc/C60 and CuPc/C60 OPVs are summarized in Supporting Information. In previous studies we have shown that solvent annealing both TiOPc and ClInPc films increase the Pc/C60 interfacial contact area, and significantly increases photocurrent response, but also complicates some of the interactions between the Pc donor and the ITO contact.13,25-27 Work to be communicated elsewhere will demonstrate how these contact/Pc interactions affect the performance of bulk-heterojunction OPVs based on the donors studied here. For ClInPc/C60 devices under AM1.5G illumination, disorder in the as-deposted Pc films leads to poor device performance, with low FF, and in some cases an S-shaped J/V response, consistent with enhanced charge recombination near VOC.16,17,28,40 As predicted from our recent modeling studies,18,19 recent investigations of the electrical properties of partically blocked ITO electrodes,77 and from several past studies of partially-blocked or contaminated ITO electrodes,78

     

ACS Paragon Plus Environment

18

Page 19 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

both the contribution to recombination near VOC, and contributions to RS for ClInPc/C60 devices increases for DSC- versus AP-ITO contacts by almost a factor of ten. The state of the ITO surface is critical in controlling charge harvesting efficiencies near VOC, and RS, owing to the overall coverage and separation distance between blocked regions or regions with energetically distinct (high) barriers to charge harvesting, especially for active layer materials with low charge mobilities.18,36,37,39 Blocked regions, or regions with high barriers to charge harvesting, that are small and well separated, or active layers with high charge mobilities, lead to J/V behavior suggesting high injection efficiencies, transitioning at low applied voltages to space-charge limited currents. 77 If charge mobility parallel to the plane of the substrate, and close to the holeharvesting contact, is sensitive to the ordering in the donor layer as has been proposed for tetravalent metal Pcs, such as TiOPc,30,32 we can expect that less ordered thin-films of ClInPc and TiOPc will be quite sensitive to the surface coverage and size of the blocked regions on the electrode. For PEN/C60 devices we observed only small differences in device performance for DSCITO, AP-ITO and F5BPA-modified ITO substrates, despite the difference in work function for DSC-ITO versus either AP-ITO or F5BPA-ITO substrates. Device performance is almost independent of ITO surface composition provided that the work function is greater than the IP of this donor layer, as shown in earlier studies of phosphonic acid-modified ITO contacts, where F5BPA-modified ITO was used to create a contact with a work function comparable to that seen for oxygen plasma treated, or air-plasma treated ITO.50 Similar conclusions can be drawn from OPV data for TiOPc/C60 with TiOPc donor layers showing extreme sensitivity to ITO

     

ACS Paragon Plus Environment

19

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

composition. CuPc donor layers are less sensitive to ITO composition than ClInPc or TiOPc,77  but appear to be more sensitive than for OPVs based on PEN donor layers. OPV performance for ITO/ClInPc single-junction devices lacking an electron acceptor As a further indication of the importance of the interaction between the ITO surface and the ClInPc donor layer, Figure 9 summarizes the OPV behavior for devices built without a C60 acceptor layer. Although it has been widely confirmed that the majority of photocurrent in PHJ OPVs is formed at a molecular donor/acceptor interface,53,79-81 it can be seen that appreciable photocurrent is created in ITO/ClInPc/Al devices (Figure 9C,D), while almost negligible photocurrent response is seen for ITO/CuPc/Al devices (Figure 9A,B). For ClInPc single junction devices on AP-ITO the VOC is 0.75 V, with a JSC of 3.2 mA/cm2, about half that seen in ITO/ClInPc/C60/Al devices. Photocurrent action spectra to be reported elsewhere (IPCE plots) confirm photocurrent production mainly at the ITO/ClInPc interface, with a maximum IPCE value of ca. 10% for photocurrent production and harvesting at zero bias (short-circuit condition). Unlike devices with a C60 acceptor layer, the FF for these single component OPVs is less than 0.2, consistent with recombination limited performance, i.e. high internal fields are required to sweep out charges created at the ITO/ClInPc interface, to the collection electrode. In this case it is anticipated that efficiency is limited by transport of electrons across the Pc layer to the top contact, assuming that electron mobility in the Pc layer is significantly lower than the hole mobility.53,80,82,83 For ITO/ClInPc/Al devices built on DSC-ITO (SI section) the JSC is reduced to 0.83mA/cm2, and the VOC is also reduced to 0.72 V, consistent with the hypotheses that: (i) photocurrent is mainly produced at the ITO/Pc interface and, (ii) the interaction of the Pc with that interface affects the probability for photocurrent production, and is strongly dependent

     

ACS Paragon Plus Environment

20

Page 21 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

on the composition of the oxide and the structure of the first-deposited monolayers of Pc donor. This observation has interesting implications for OPV behavior for any donor layer which produces photocurrent at the contact/donor interface, since this may actually enhance recombination probabilities for charges (in this case holes) created at the donor/acceptor interface some 15-20 nm away. Forrest and coworkers have recently noted that charge carriers may be spontaneously formed in the bulk of donor layers (direct carrier generation), not at a D/A interface, and that the sweep out of these charge carriers can show the same type of field dependence noted for our single junction devices.84 We cannot also fully exclude some bulk generation of charge carriers as well, especially in light of the strong internal dipoles in trivalent or tetravalent metal Pcs such as ClInPc. Studies to be communicated elsewhere focus on photocurrent generation in active layers where ClInPc is fully mixed with C60 in bulkheterojunction OPVs, where photocurrent yields are higher relative to photocurrent that may arise at the Pc/ITO interface. Conclusions We have shown the effect of various ITO surface pre-treatments on the device performance for PHJ organic photovoltaics using widely variable polar and non-polar electron donor materials. This work suggests that orientation of polar electron donor materials, such as ClInPc and TiOPc, has a significant impact on hole-collection and OPV efficiencies. Activating the ITO surface increased the performance of ClInPc-based OPVs by approximately twofold. Previous studies have demonstrated that AP- or OP-ITO surfaces are quite polar with high surface free energies. Orientation of the polar end of the halo-metal or oxo-metal bond in these Pcs, towards this polar oxide surface, is a reasonable expectation.48,49 This sensitivity to surface energy is also

     

ACS Paragon Plus Environment

21

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

seen in the UPS vacuum level shifts for deposition of the first two monolayers of the polar Pcs, and suggests that these effects may be important in determining the performance of OPVs based on polar donors and/or acceptors, or where one or both components of the active layer has a strong internal dipole. Using a fluorinated benzyl phosphonic acid to lower the surface free energy while maintaining the high work function of air-plasma treated ITO,48,49 we find that OPV performance is still negatively impacted for these polar donor layers. For a non-polar donor like PEN, which does not demonstrate specific interactions with the ITO surface, OPV performance is not changed provided the effective work function of the contact exceeds the IP and transport HOMO energies of the donor layer. We hypothesize that when active layer materials are brought into contact with electrically and compositionally heterogeneous hole- or electron-collection electrodes, for systems in which dipolar interactions control molecular structure at the electrode/active layer interface, substantial attention should be paid to the design of these interfaces in order to optimize OPV performance. ACKNOWLEDGEMENTS This work was supported as part of the Center for Interface Science: Solar Electric Materials, 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. The Authors would like to thank Ed Autz and Lee Macomber from the Department of Chemistry and Biochemistry Machine shop, Mike Read and Kevin Bao from CHIEF (CBC) and Associate Staff

     

ACS Paragon Plus Environment

22

Page 23 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

Scientist Paul Lee for their constant support and direction in the fabrication and implementation of critical components used in this project. Supporting Information Available: Additional UPS data for CuPc and TiOPc heterojunctions, and OPV data for these same heterojunctions, along with the absorbance data for a larger range of ClInPc coverages on AP-ITO, can be found free of charge via the internet at http://pubs.acs.org

References   (1)  Sun, Y.; Welch, G. C.; Leong, W. L.; Takacs, C. J.; Bazan, G. C.; Heeger, A. J. Nat. Mater.  2012, 11, 44‐48.    (2)  Gong, X.; Tong, M. H.; Brunetti, F. G.; Seo, J.; Sun, Y. M.; Moses, D.; Wudl, F.; Heeger, A.  J. Adv. Mater. 2011, 23, 2272‐+.    (3)  Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.; Leclerc, M.;  Lee, K.; Heeger, A. J. Nature Photonics 2009, 3, 297‐U295.    (4)  Scharber, M. C., Wuhlbacher, D., Koppe, M., Denk, P., Waldauf, C., Heeger, A. J., Brabec,  C. L. Adv. Mater. 2006, 18, 789‐+.    (5)  Steim, R.; Ameri, T.; Schilinsky, P.; Waldauf, C.; Dennler, G.; Scharber, M.; Brabec, C. J.  Sol. Energy Mater. Sol. Cells 2011, 95, 3256‐3261.    (6)  Steim, R.; Kogler, F. R.; Brabec, C. J. J. Mater. Chem. 2010, 20, 2499‐2512.    (7)  Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S. J.; Williams, S. P. Adv. Mater.  2010, 22, 3839‐3856.    (8)  Zhong, S.; Zhong, J. Q.; Wang, X. Z.; Huang, M. Y.; Qi, D. C.; Chen, Z. K.; Chen, W. J. Phys.  Chem. C 2012, 116, 2521‐2526.    (9)  Verreet, B.; Muller, R.; Rand, B. P.; Vasseur, K.; Heremans, P. Organic Electronics 2011,  12, 2131‐2139.    (10)  Beaumont,  N.;  Hancox,  I.;  Sullivan,  P.;  Hatton,  R.  A.;  Jones,  T.  S.  Energy  &  Envr.  Sco.  2011, 4, 1708‐1711.    (11)  Yang, J. L.; Schumann, S.; Hatton, R. A.; Jones, T. S. Organic Electronics 2010, 11, 1399‐ 1402.    (12)  Sullivan, P.; Jones, T. S.; Ferguson, A. J.; Heutz, S. Appl. Phys. Lett. 2007, 91.    (13)  Placencia,  D.;  Wang,  W.  N.;  Gantz,  J.;  Jenkins,  J.  L.;  Armstrong,  N.  R.  J.  Phys.  Chem.  C  2011, 115, 18873‐18884.    (14)  Vasseur, K.; Rand, B. P.; Cheyns, D.; Temst, K.; Froyen, L.; Heremans, P. J. Phys. Chem.  Lett 2012, 3, 2395‐2400. 

     

ACS Paragon Plus Environment

23

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL     (15)  Rand, B. P.; Cheyns, D.; Vasseur, K.; Giebink, N. C.; Mothy, S.; Yi, Y. P.; Coropceanu, V.;  Beljonne, D.; Cornil, J.; Bredas, J. L.; Genoe, J. Adv. Funct. Mater. 2012, 22, 2987‐2995.    (16)  Wagenpfahl,  A.;  Rauh,  D.;  Binder,  M.;  Deibel,  C.;  Dyakonov,  V.  Phys.  Rev.  B  2010,  82,  115306.    (17)  Wagenpfahl, A.; Deibel, C.; Dyakonov, V. IEEE Selected Top. Quan. Elect. 2010, 16, 1759‐ 1763.    (18)  Zacher, B.; Armstrong, N. R. J. Phys. Chem. C 2011, 115, 25496‐25507.    (19)  Ratcliff, E. L.; Zacher, B.; Armstrong, N. R. J. Phys. Chem. Lett. 2011, 1337‐1350.    (20)  Tress, W.; Leo, K.; Riede, M. Phys. Rev. B 2012, 85.    (21)  Petersen, A.; Kirchartz, T.; Wagner, T. A. Phys. Rev. B 2012, 85.    (22)  Cowan, S. R.; Banerji, N.; Leong, W. L.; Heeger, A. J. Adv. Funct. Mater. 2012, 22, 1116‐ 1128.    (23)  Kirchartz, T.; Pieters, B. E.; Kirkpatrick, J.; Rau, U.; Nelson, J. Phys. Rev. B 2011, 83.    (24)  Street, R. A.; Song, K. W.; Cowan, S. Organic Electronics 2011, 12, 244‐248.    (25)  Wang, W.; Placencia, D.; Armstrong, N. R. Organic Electronics 2011, 12, 383‐393.    (26)  Placencia, D.; Wang, W. N.; Shallcross, R. C.; Nebesny, K. W.; Brumbach, M.; Armstrong,  N. R. Adv. Funct. Mater. 2009, 19, 1913‐1921.    (27)  Brumbach, M.; Placencia, D.; Armstrong, N. R. J. Phys. Chem. C 2008, 112, 3142‐3151.    (28)  Street, R. A. Phys. Rev. B 2011, 84.    (29)  Coppede, N.; Castriota, M.; Cazzanelli, E.; Forti, S.; Tarabella, G.; Toccoli, T.; Walzer, K.;  Iannotta, S. J. Phys. Chem. C 2010, 114, 7038‐7044.    (30)  Li, L. Q.; Tang, Q. X.; Li, H. X.; Yang, X. D.; Hu, W. P.; Song, Y. B.; Shuai, Z. G.; Xu, W.; Liu,  Y. Q.; Zhu, D. B. Adv. Mater. 2007, 19, 2613‐+.    (31)  Tada, H.; Touda, H.; Takada, M.; Matsushige, K. Appl. Phys. Lett. 2000, 76, 873‐875.    (32)  Norton, J. E.; Bredas, J. L. J. Chem. Phys. 2008, 128, 7.    (33)  Marrikar, F. S.; Brumbach, M.; Evans, D. H.; Lebron‐Paler, A.; Pemberton, J. E.; Wysocki,  R. J.; Armstrong, N. R. Langmuir 2007, 23, 1530‐1542.    (34)  Brumbach,  M.; Veneman, P. A.;  Marrikar, F. S.; Schulmeyer, T.; Simmonds, A.; Xia, W.;  Lee, P.; Armstrong, N. R. Langmuir 2007, 23, 11089‐11099.    (35)  Armstrong,  N.  R.;  Veneman,  P.  A.;  Ratcliff,  E.;  Placencia,  D.;  Brumbach,  M.  Acc.  Chem.  Res. 2009, 42, 1748‐1757.    (36)  Shen, Y. L.; Hosseini, A. R.; Wong, M. H.; Malliaras, G. G. ChemPhysChem 2004, 5, 16‐25.    (37)  Scott, J. C.; Malliaras, G. G. Chem. Phys. Lett. 1999, 299, 115‐119.    (38)  Malliaras, G. G.; Scott, J. C. J. Appl. Phys. 1999, 85, 7426‐7432.    (39)  Shen, Y.; Klein, M. W.; Jacobs, D. B.; Campbell Scott, J.; Malliaras, G. G. Phys. Rev. Lett.  2001, 86, 3867‐3870.    (40)  Street, R. A.; Schoendorf, M. Phys. Rev. B 2010, 81.    (41)  Kera,  S.,  Yamane,  H.,  Fukagawa,  H.,  Hanatani,  T.,  Okudaira,  K.  K.,  Seki,  K.,  Ueno,  N.  J.  Electron Spectrosc. Relat. Phenom. 2007, 156, 135‐138.    (42)  Fukagawa,  H.,  Kataoka,  T.,  Hosourni,  S.,  Kera,  S.,  Ueno,  N.  J.  Electron  Spectrosc.  Relat.  Phenom. 2007, 156, 37.    (43)  Fukagawa, H., Yamane, H., Kera, S.,. Okudaira, K. K., Ueno, N. Phys. Rev. B 2006, 73.    (44)  Kera,  S.;  Yabuuchi,  Y.;  Yamane,  H.;  Setoyama,  H.;  Okudaira,  K.  K.;  Kahn,  A.;  Ueno,  N.  Phys. Rev. B 2004, 70. 

     

ACS Paragon Plus Environment

24

Page 25 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL     (45)  Kera, S., Fukagawa, H., Kataoka, T., Hosoumi, S., Yamane, H., Ueno, N. Phys. Rev. B 2007,  75.    (46)  Fukagawa, H.; Yamane, H.; Kera, S.; Okudaira, K. K.; Ueno, N. Phys. Rev. B 2006, 73.    (47)  Fukagawa, H.; Hosoumi, S.; Yamane, H.; Kera, S.; Ueno, N. Phys. Rev. B 2011, 83.    (48)  Hotchkiss, P. J.; Jones, S. C.; Paniagua, S. A.; Sharma, A.; Kippelen, B.; Armstrong, N. R.;  Marder, S. R. Acc. Chem. Res. 2012, 45, 337‐346.    (49)  Paniagua, S. A.; Hotchkiss, P. J.; Jones, S. C.; Marder, S. R.; Mudalige, A.; Marrikar, F. S.;  Pemberton, J. E.; Armstrong, N. R. J. Phys. Chem. C 2008, 112, 7809‐7817.    (50)  Sharma,  A.;  Haldi,  A.;  Potscavage,  W.  J.;  Hotchkiss,  P.  J.;  Marder,  S.  R.;  Kippelen,  B.  J.  Mater. Chem. 2009, 19, 5298‐5302.    (51)  Griffith, O. L.; Anthony, J. E.; Jones, A. G.; Lichtenberger, D. L. J. Am. Chem. Soc. 2010,  132, 580‐586.    (52)  Muntwiler, M.; Yang, Q.; Tisdale, W. A.; Zhu, X. Y. Phys. Rev. Lett. 2008, 101, 4.    (53)  Tang, C. W. Appl. Phys. Lett. 1986, 48, 183‐185.    (54)  Chamberlain, G. A. Solar Cells 1983, 8, 47‐83.    (55)  Martin, M.; Andre, J. J.; Simon, J. Nouv. J. Chim. New J. Chem. 1981, 5, 485‐490.    (56)  Alloway,  D.  M.;  Graham,  A.  L.;  Yang,  X.;  Mudalige,  A.;  Colorado,  R.;  Wysocki,  V.  H.;  Pemberton,  J.  E.;  Randall  Lee,  T.;  Wysocki,  R.  J.;  Armstrong,  N.  R.  J.  Phys.  Chem.  C.  2009,  113,  20328‐ 20334.    (57)  Alloway, D. M.; Armstrong, N. R. Appl. Phys. A Mater. Sci. Processing 2009, 95, 209‐218.    (58)  Cahen, D.; Kahn, A. Adv. Mater. 2003, 15, 271‐277.    (59)  Hwang, J.; Wan, A.; Kahn, A. Mat. Sci. Engr. R 2009, 64, 1‐31.    (60)  Duhm,  S.;  Xin,  Q.;  Hosoumi,  S.;  Fukagawa,  H.;  Sato,  K.;  Ueno,  N.;  Kera,  S.  Adv.  Mater.  2012, 24, 901‐+.    (61)  Alloway,  D.  M.;  Hofmann,  M.;  Smith,  D.  L.;  Gruhn,  N.  E.;  Graham,  A.  L.;  Colorado,  R.;  Wysocki, V. H.; Lee, T. R.; Lee, P. A.; Armstrong, N. R. J. Phys. Chem. B 2003, 107, 11690‐11699.    (62)  Mizuguchi, J.; Rihs, G.; Karfunkel, H. R. J. Phys. Chem. 1995, 99, 16217 ‐ 16227.    (63)  Monti, O. L. A. J. Phys. Chem. Lett. 2012, 3, 2342‐2351.    (64)  Kera, S.; Yamane, H.; Honda, H.; Fukagawa, H.; Okudaira, K. K.; Ueno, N. Surf. Sci. 2004,  566, 571‐578.    (65)  Basova, T. V.; Kiselev, V. G.; Plyashkevich, V. A.; Cheblakov, P. B.; Latteyer, F.; Peisert, H.;  Chasse, T. Chem. Phys. 2011, 380, 40‐47.    (66)  Niederhausen,  J.;  Amsalem,  P.;  Frisch,  J.;  Wilke,  A.;  Vollmer,  A.;  Rieger,  R.;  Mullen,  K.;  Rabe, J. P.; Koch, N. Phys. Rev. B 2011, 84.    (67)  Salzmann, I.; Duhm, S.; Heimel, G.; Oehzelt, M.; Kniprath, R.; Johnson, R. L.; Rabe, J. P.;  Koch, N. J. Am. Chem. Soc. 2008, 130, 12870‐+.    (68)  Koch,  N.;  Gerlach,  A.;  Duhm,  S.;  Glowatzki,  H.;  Heimel,  G.;  Vollmer,  A.;  Sakamoto,  Y.;  Suzuki, T.; Zegenhagen, J.; Rabe, J. P.; Schreiber, F. J. Am. Chem. Soc. 2008, 130, 7300‐7304.    (69)  Duhm, S.; Heimel, G.; Salzmann, I.; Glowatzki, H.; Johnson, R. L.; Vollmer, A.; Rabe, J. P.;  Koch, N. Nat. Mater. 2008, 7, 326‐332.    (70)  Watkins, N. J.; Yan, L.; Gao, Y. L. Appl. Phys. Lett. 2002, 80, 4384‐4386.    (71)  Wheeler, W. D.; Parkinson, B. A.; Dahnovsky, Y. J. Chem. Phys. 2011, 135.    (72)  Schroeder, P. G.; France, C. B.; Park, J. B.; Parkinson, B. A. J. Phys. Chem. B 2003, 107,  2253‐2261. 

     

ACS Paragon Plus Environment

25

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL     (73)  Nakai, K.; Ishii, K.; Kobayashi, N.; Yonehara, H.; Pac, C. J. Phys. Chem. B 2003, 107, 9749‐ 9755.    (74)  Chau, L. K.; England, C. D.; Chen, S. Y.; Armstrong, N. R. J. Phys. Chem. 1993, 97, 2699‐ 2706.    (75)  Dienel, T., Forker, R., Leo, K., Fritz, T. J. Phys. Chem. C 2007, 111, 14593‐14596.    (76)  Walzer,  K.,  Toccoli,  T.,  Pallaoro,  A.,  Iannotta,  S.,  Wagner,  C.,  Fritz,  T.,  Leo,  K.  Surf.  Sci.  2006, 600, 2064‐2069.    (77)  MacDonald, G. A.; Veneman, P. A.; Placencia, D.; Armstrong, N. R. ACS Nano 2012.    (78)  Xue, J. G.; Uchida, S.; Rand, B. P.; Forrest, S. R. Appl. Phys. Lett. 2004, 84, 3013‐3015.    (79)  Perez,  M.  D.;  Borek,  C.;  Forrest,  S.  R.;  Thompson,  M.  E.  J.  Am.  Chem.  Soc.  2009,  131,  9281‐9286.    (80)  Rand, B. P.; Burk, D. P.; Forrest, S. R. Phys. Rev. B 2007, 75.    (81)  Forrest, S. R. MRS Bull. 2005, 30, 28‐32.    (82)  Gregg, B. A.; Hanna, M. C. J. Appl. Phys. 2003, 93, 3605‐3614.    (83)  Gregg, B. A. In Molecules as Components of Electronic Devices 2003; Vol. 844, p 243‐257.    (84)  Renshaw, C. K.; Zimmerman, J. D.; Lassiter, B. E.; Forrest, S. R. Phys. Rev. B 2012, 86.   

     

ACS Paragon Plus Environment

26

Page 27 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

 

Table 1 – Summary of relevant device parameters measured for various pre-treatments of ITO using ClInPc and PEN as donor materials in the device architecture. a Open circuit photopotential . b Short-circuit current. c Fill factor. d Predicted efficiency under AM1.5G illumination. e Series resistance in the dark. f Shunt resistance in the dark. g Reverse saturation current (estimated from lowest dark current at 0 V). For all devices reported the statistical variations are less than: Voc ±0.01, Jsc ±0.20, FF ±0.02, η ±0.14.

     

ACS Paragon Plus Environment

27 

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

 

Figure 1 – (upper left) Schematic view of ClInPc; (upper right) The absorbance spectra for ca. 2 monolayers of ClInPc as-deposited on AP-ITO (black curve, “Phase I”) and the same Pc film after solvent-annealing (red curve, “Phase II”); such clear optical transitions are seen only for 1-to-2 monolayer films on AP-ITO, where some degree of ordering of the Pc film is possible (see below);  (lower) A schematic edge on view of two monolayers of the monoclinic Phase 1 polymorph of ClInPc, which, upon solvent annealing converts to the triclinic Phase II polymorph – these figures adapted from the single crystal structures for TiOPc, as summarized in Reference 62. For simplicity we show the halometal or oxo-metal bond normal to the substrate plane, demonstrating that, with sufficient ordering in the first monolayer, the Phase I to Phase II transition might be achieved by a simple displacement of the second layer with respect to the first. This configuration does not preclude a more tilted orientation for the first deposited monolayers of Pc.

     

ACS Paragon Plus Environment

28

Page 29 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

Intensity (arb. units)

(A)

(B)

ClInPc (AP ITO)

19.1 nm

19.1 nm

4.7 nm 2.3 nm 1.1 nm

4.7 nm 2.3 nm 1.1 nm

0.5 nm

0.5 nm

0.2 nm

0.2 nm

0.0 nm

0.0 nm

17.4

17.1

16.8

16.5

16.2

15.9

4

3

Binding Energy (eV)

(C) Intensity (arb. units)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

2

1

0

-1

Binding Energy (eV)

(D)

ClInPc (DSC ITO)

19.1 nm

19.1 nm

4.7 nm 2.3 nm

4.7 nm 2.3 nm

1.1 nm

1.1 nm

0.5 nm

0.5 nm

0.2 nm

0.2 nm 0.0 nm

0.0 nm

18.0

17.7

17.4

17.1

16.8

Binding Energy (eV)

16.5 4

3

2

1

0

-1

Binding Energy (eV)

Figure 2 – UPS spectra for the deposition of ClInPc films (0-19 nm thickness) on (panel A,B) AP-ITO and (panel C,D) DSC-ITO. Work functions and ionization potentials are calculated as discussed in the text, using energy differences between high and low kinetic energy (HKE and LKE) edges. For deposition on AP-ITO the shift in the LKE edge reverses direction above coverages corresponding to completion of the first Pc monolayer, and the high KE ionization feature broadens and can be fit with two peaks, as per previous studies of similar dipolar Pcs.32,35,36    

     

ACS Paragon Plus Environment

29

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

Intensity (arb. units)

(A)

(B)

Pentacene (AP ITO)

13.6 nm 6.8 nm 3.3 nm 1.6 nm 0.8 nm

13.6 nm 6.8 nm 3.3 nm

0.3 nm

0.3 nm 0.1 nm

1.6 nm 0.8 nm

0.1 nm 0.0 nm

0.0 nm

18.4 18.0 17.6 17.2 16.8 16.4 16.0

4

3

(C)

2

1

0

-1

Binding Energy (eV)

Binding Energy (eV)

Intensity (arb. units)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 36

Pentacene (DSC ITO)

(D)

13.6 nm

13.6 nm 6.8 nm

6.8 nm 3.3 nm 1.6 nm 0.8 nm 0.3 nm

3.3 nm 1.6 nm 0.8 nm 0.3 nm 0.1 nm 0.0 nm

0.1 nm 0.0 nm

18.3

18.0

17.7

17.4

Binding Energy (eV)

17.1

4

3

2

1

0

-1

Binding Energy (eV)

Figure 3 – UPS spectra for the deposition of PEN films (0-14 nm thickness) on (panel A,B) AP-ITO and (panel C,D) DSC-ITO. UPS data plotted on an absolute KE scale, as discussed in Figure 2. In contrast to ClInPc films on ITO (Figure 2), we see only monotonic changes to the local work function and small changes to the effective IP as a function of surface coverage, regardless of ITO pretreatment.      

 

     

ACS Paragon Plus Environment

30

Page 31 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

0.4

(A)

0.0 -0.1

5.2

0.2

5.1

0.1

5.1

0.0

5.0

5.0

-0.1

4.9

-0.5

ClInPc (DSC ITO) 5.3

5.2

-0.3 -0.4

(B)

0.3

4.9

-0.2

-0.6

4.8

4.8

-0.3

-0.7 0

5

10

15

0

20

5

(C)

15

20

Thickness (nm)

Thickness (nm) 0.0

10

Pentacene (AP ITO)

-0.2 -0.4

5.3

0.0

5.2

-0.1

5.0

eff

-0.6

IP

5.1

(D)

Pentacene (DSC ITO)

4.65

4.60

-0.2

4.55

IP

eff

-0.2

5.3

IP eff

ClInPc (AP ITO)

IP

0.1

eff

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

-0.3

-0.8

4.50

4.9 -0.4

-1.0

4.45

4.8 -1.2

-0.5 0

5

10

15

0

Thickness (nm)

5

10

15

Thickness (nm)

Figure 4 – Change in effective work function (Δφeff – black lines) and IP (blue lines) as a function of of ClInPc (panel A,B) and PEN (panel C,D) film thickness for AP-ITO (panel A,C) and DSC-ITO (panel B,D) – on DSC-ITO smaller and more variable changes in both Δφeff and IP are observed as a function of PEN coverage.  

     

ACS Paragon Plus Environment

31

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL    

Intensity (arb. units)

 

(A)

ClInPc (F5BPA ITO)

 

9.6 nm 4.8 nm 2.4 nm 1.2 nm 0.6 nm 0.3 nm 0.0 nm

17.2

17.0

16.8

16.6

16.4

Binding Energy (eV)

(B)

Figure 5 – UPS data (A = high KE; B = low KE) for deposition of low coverages of ClInPc on F5BPA-modified ITO; (C) the change in Δφeff (black line) and IP (blue line) as a function of ClInPc coverage on F5BPAmodified ITO.

9.6 nm 4.8 nm 2.4 nm 1.2 nm 0.6 nm 0.3 nm 0.0 nm

4

3

2

1

0

Binding Energy (eV) ClInPc (F5BPA ITO)

0.00

5.40

-0.05 5.35

-0.15

5.30

IP

-0.10

eff

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 36

-0.20 5.25 -0.25

(C)

-0.30

 

0

2

4

6

8

5.20

10

Thickness (nm)

   

ACS Paragon Plus Environment

32

Page 33 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

 

Figure 6 – Absorbance spectra for ClInPc films on (A) DSC-ITO and (B) AP-ITO both before (black line) and after (red line) solvent annealing to achieve the Phase II polymorph; and absorbance spectra for comparable coverages of PEN on (C) DSC-ITO and (D) AP-ITO showing only minor changes in the absorbance.

     

ACS Paragon Plus Environment

33

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

ClInPc

100

DSC Light AP Light DSC Dark AP Dark

10

Photocurrent (mA/cm2)

Photocurrent (mA/cm2)

20

0

(A) -10 -1.0

-0.5

0.0

0.5

1.0

10 1 0.1 0.01 1E-3 1E-4 1E-5

1E-7 -1.0

1.5

Pentacene

10

(C) -0.5

0.0

0.5

1.0

1.5

100

DSC Light AP Light DSC Dark AP Dark

0

-10 -1.0

-0.5

Applied Potential (V)

Photocurrent (mA/cm2)

20

(B)

1E-6

Applied Potential (V)

Photocurrent (mA/cm2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 36

0.0

0.5

1.0

Applied Potential (V)

1.5

10 1 0.1 0.01 1E-3 1E-4 1E-5

(D)

1E-6 1E-7 -1.0

-0.5

0.0

0.5

1.0

Applied Potential (V)

1.5

  

Figure 7 – Linear and log J/V plots (light = solid lines, dark = dashed lines) for ClInPc/C60 (panel A,B) and PEN/C60 (panel C,D) OPVs on AP-ITO (blue) and DSC-ITO (red). PEN/C60 OPVs are remarkably insensitive to the degree of activation of the ITO substrate. ClInPc/C60 OPVs are quite sensitive to the type of activation for the ITO substrate, and on substrates with a high fraction of blocked sites the series resistance is higher, the fill factor reduced, and overall device efficiency reduced.

     

ACS Paragon Plus Environment

34

Page 35 of 36

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL  

ClInPc on F5BPA ITO

15

Photocurrent (mA/cm2)

Photocurrent (mA/cm2)

20

Dark Light

10 5 0 -5

(A) -10 -1.0

-0.5

0.0

0.5

1.0

10 1 0.1 0.01 1E-3 1E-4 1E-5 1E-6 -1.0

1.5

(B) -0.5

Applied Potential (V) PEN on F5BPA ITO

15

Dark Light

5 0 -5

(C) -0.5

0.0

0.5

1.0

1.5

100

10

-10 -1.0

0.0

Applied Potential (V)

Photocurrent (mA/cm 2)

20

Photocurrent (mA/cm 2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

0.5

1.0

1.5

10 1 0.1 0.01 1E-3 1E-4

(D)

1E-5 1E-6 -1.0

-0.5

Applied Potential (V)

0.0

0.5

1.0

1.5

Applied Potential (V)

Figure 8 – Linear and log J/V plots (light = solid lines, dark = dashed lines) for ClInPc/C60 (panel A,B) and PEN/C60 (panel C,D) OPVs on F5BPA-modified ITO. As with deposition on DSC-modified ITO, the OPV performance is decreased significantly for ClInPc/C60 heterojunctions and not significantly changed for PEN/C60 heterojunctions when deposited on the F5BPA-modified ITO substrates.  

     

ACS Paragon Plus Environment

35

The Journal of Physical Chemistry

jp-2012-07546v Gantz et al. J.Phys.Chem. C Revised Dec 13 2012 FINAL      

Photocurrent (mA/cm2)

20 15 10

100

ITO(AP)/CuPc/Al Average Dark Average Light

10 1 0.1

5

0.01 0 1E-3 -5

1E-4

-10

(A)

-15 -1.5

-1.0

-0.5

0.0

0.5

1.0

(B)

1E-5 1.5

1E-6 -1.5

-1.0

Applied Potential (V) 20

Photocurrent (mA/cm2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 36

15 10

-0.5

0.0

0.5

1.0

1.5

Applied Potential (V) 100

ITO(AP)/ClInPc/Al Average Dark Average Light

10 1 0.1

5

0.01 0

1E-3 -5

1E-4

-10 -15 -1.5

(C) -1.0

-0.5

0.0

0.5

Applied Potential (V)

1.0

1E-5 1E-6 -1.5 1.5

(D) -1.0

-0.5

0.0

0.5

1.0

1.5

Applied Potential (V)

Figure 9 – Log and linear J/V plots (red = illuminated, black = dark) for ITO(AP)/CuPc/Al and ITO(AP)/ClInPc/Al heterojunctions. The ITO(AP)/CuPc heterojunctions (panel A,B) generated virtually no discernible photoresponse, while the ITO(AP)/ClInPc heterojunctions (panel C,D) generate a clear photovoltaic response, albeit one with very low FF (0.34) suggestive of a charge transport limited, or recombination limited, photocurrent response near VOC (0.75V).  

     

ACS Paragon Plus Environment

36