Surprisingly High Conductivity and Efficient Exciton Blocking in

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Communication

Surprisingly High Conductivity and Efficient Exciton Blocking in Fullerene:Wide-Energy-Gap Small Molecule Mixtures Kevin Bergemann, Jojo Amonoo, Byeongseop Song, Peter F Green, and Stephen R. Forrest Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.5b00908 • Publication Date (Web): 05 May 2015 Downloaded from http://pubs.acs.org on May 11, 2015

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.

Nano Letters 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 27

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

Nano Letters

Surprisingly High Conductivity and Efficient Exciton Blocking in Fullerene:Wide-Energy-Gap Small Molecule Mixtures Kevin J. Bergemann1, Jojo A. Amonoo2, Byeongseop Song3, Peter F. Green2,4, Stephen R. Forrest*,1,3,4 1

2

3

Department of Physics, University of Michigan, Ann Arbor, MI, 48109 USA

Applied Physics Program, University of Michigan, Ann Arbor, MI 48109, USA

Department of Electrical Engineering and Computer Science, Ann Arbor, MI, 48109 USA

4

Materials Science and Engineering, University of Michigan, Ann Arbor, MI, 48109 USA *Corresponding Author E-Mail: [email protected]

Abstract We find that mixtures of C60 with the wide energy gap, small molecular weight semiconductor bathophenanthroline (BPhen) exhibit a combination of surprisingly high electron conductivity and efficient exciton blocking when employed as buffer layers in organic photovoltaic cells. Photoluminescence quenching measurements show that a 1:1 BPhen:C60 mixed layer has an exciton blocking efficiency of 84±5%, compared to that of 100% for a neat BPhen layer. This high blocking efficiency is accompanied by a 100-fold increase in electron conductivity compared with neat BPhen. Transient photocurrent measurements show that charge transport through a neat BPhen buffer is dispersive, in contrast to non-dispersive transport in the compound buffer. Interestingly, although the conductivity is high, there is no clearly defined insulating-to-conducting phase transition with increased insulating BPhen fraction. Thus, we

1

ACS Paragon Plus Environment

Nano Letters

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

infer that C60 undergoes nanoscale (80%) BPhen fractions. Keywords: Organic photovoltaics; organic buffer layer; diffusion; charge transport; percolation; charge transfer

2


Page 2 of 27

Page 3 of 27

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

Nano Letters

Exciton confinement in the active regions of organic photovoltaics (OPVs) provided by a blocking layer between the cathode and electron acceptor is a necessary element for achieving high solar power conversion efficiency. The blocker serves four functions: (i) confinement of excitons within the active layer to prevent quenching at the organic/cathode interface1,2, (ii) prevention of damage to the acceptor layer during the deposition of the cathode2,3, (iii) provision of a transparent spacer to optimize the optical field distribution within the active layer4, and (iv) promotion of efficient charge extraction following exciton dissociation in the active region. Multiple types of blockers have been developed, with the major differences being the mechanism for charge conduction. The most effective means for achieving confinement is the double heterostructure5, where a cathode-side organic buffer layer forms a blocking, nested (Type I) heterojunction with the acceptor. A wide, highest occupied molecular orbital (HOMO)-lowest unoccupied MO (LUMO) energy gap material such as bathocuproine1,2 (BCP) or bathophenanthroline6,7 (BPhen) with a LUMO energy significantly closer to the vacuum level than that of the acceptor layer can transfer charge through defect states induced during metal deposition5. Materials such as 3,4,9,10 perylenetetracarboxylic bisbenzimidazole (PTCBI) and 1,4,5,8-napthalene-tetracarboxylic-dianhydride (NTCDA) conduct electrons along LUMOs that align with those of the acceptor8. In contrast, tris-(acetylacetonato) ruthenium(III) [Ru(acac)3] and related compounds with very shallow HOMO levels conduct holes from the cathode that recombine with electrons from the acceptor layer at the buffer/acceptor interface9,10. The recently developed mixed buffer11,12 that is comprised of an electron conducting fullerene (e.g. C60) blended into a wide HOMO-LUMO energy gap material such as BPhen uses a fourth mechanism. The fullerene provides conductive pathways through the wide gap, insulating material which, in turn, blocks (or “filters”) excitons. The use of the mixed buffer

3


Nano Letters

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

results in an increase in the power conversion efficiency and fill factor in both planar and mixed heterojunction OPVs11,12. In the planar device, the improvement is due to reduced excitonpolaron quenching arising from the significantly enhanced charge extraction compared to a neat layer of BPhen.

The improvement in the mixed device results from enhanced electron

extraction, again due to higher conduction through the buffer. The mixed buffer is analogous to the concept used in doped transport layers but differs in several important respects, rendering it qualitatively different. The dopant in doped transport layers is generally a metal13 or an organic with energy levels offset from the acceptor material14, which results in quenching of excitons at the acceptor/transport layer interface. This is often treated by incorporating a thin layer of the host material between the doped transport layer and the acceptor. As the conductive fullerene component of the mixed buffer possesses the same HOMO and LUMO as the fullerene acceptor, such precautions are unnecessary. The mixing fraction of fullerene in the mixed buffer is also significantly larger than that of doped transport layers13–15, with constant device efficiencies for 20%-80% C60 fraction. Here we quantitatively investigate and model the physical mechanisms of conduction and exciton blocking by the compound blocker using a combination of photoluminescence (PL) quenching16,17, transient photocurrent18–20, resistivity, conductive atomic force microscopy (cAFM) and x-ray diffraction (XRD) measurements. We find that the compound buffer is greater than the sum of its parts, with the C60 and BPhen contributing to the current transport and exciton blocking characteristics of the buffer to a degree that is strikingly greater than their mixing fraction would suggest. A 1:1 (by volume) BPhen:C60 mixed buffer blocks 84±5% of the excitons incident from the active layer, which is only a slight decrease from that of a neat BPhen layer. The resistance of the compound buffer remains more than an order of magnitude lower

4


Page 4 of 27

Page 5 of 27

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

Nano Letters

than pure BPhen even at a very high (>80%) BPhen fractions, implying that nanoscale phase segregation in the C60 allows it to conduct charges over the entire range of mixing fractions21–25. Electron transport through the compound buffer is non-dispersive, which is qualitatively different from conduction through neat BPhen. The improved charge transport properties of the compound buffer combined with efficient blocking characteristics leads to the increased efficiency for both planar and mixed active layer OPVs. The exciton population in an organic semiconductor is modeled using the steady-state diffusion equation5,

L2D

∂2 n − n + τ G(x) = 0 , ∂x 2

(1)

where n is the exciton density, LD is the diffusion length, τ is the exciton natural decay lifetime, and G(x) is the exciton generation profile as a function of distance (x) from the blocking/active region interface. Equation 1 is the basis of spectrally-resolved photoluminescence quenching (SR-PLQ)16,17, a method to measure the LD of a photoluminescent material. This technique uses two identical films of the material, one capped with a blocking layer (with boundary condition

∂n(0) / ∂x = 0 ) and the other with a quenching layer (with n(0) = 0).

For an exponentially

decaying G(x) in the absorbing layer, the ratio, η, of the photoluminescence of the optically pumped sample capped with a blocking (B) layer to that with a quenching (Q) layer is: 16,17

η=

PLB = α ' LD + 1 PLQ

5


(2)

Nano Letters

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

where α ' = α

cos(θ r )

Page 6 of 27

is the absorption coefficient of the active region at wavelength, λ, corrected

for the propagation angle, θr, in the layer as measured from normal incidence. The slope of η vs.

α’ in Eq. (2) gives the exciton diffusion length. This method has been extended to optically thin films through the use of transfer matrices to calculate G(x)17. For a perfectly quenching interface, n (0 ) = n Q = 0 , whereas for a perfect blocker,

n(0) = n B . Intermediate between these extremes is the non-ideal interface characterized by its blocking efficiency, ϕ, such that nNI = φ nB , where nNI is the exciton density at the non-ideal interface. In this case, Eq. (2) becomes:

η (λ ) =

PLB = φα (λ )LD + 1 . PLQ

(3)

Here, φ = 1 for a perfectly blocking interface and φ = 0 for a perfect quencher. If LD is known, ϕ can be directly measured. If the PL from three different samples capped with a blocker, a quencher, and a non-ideal blocking/quenching layer is measured, then both ϕ for the non-ideal blocker and LD for the luminescent material can be independently determined. Traditional descriptions of charge transport in semiconductors assume that the hopping time of a carrier is small compared to its transit time through the material. A narrow charge packet injected into one side of a film will then travel at a constant velocity characterized by the carrier mobility with a time-dependent Gaussian spread in the packet as predicted by the central limit theorem. In contrast, dispersive transport occurs when the hopping time is on the order of the transit time so a small number of hops can drastically change the transit time for a given charge26,27. The central limit theorem is therefore no longer applicable, as it requires a large

6


Page 7 of 27

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

Nano Letters

number of independent random events, and the spread in the charge packet becomes nonGaussian. Dispersive transport is characterized by a time-dependent mobility that is generally lower than for the case of non-dispersive transport. The current response in the dispersive limit where charges are injected with a delta- or step-function shape is: t − (1− β ) , t < t c I (t ) ∝  −(1+ β ) t , t ≥ tc

,

(4)

where tc is the transit time across the layer, and β is an empirical factor in the domain {0,1}. This behavior is in contrast to the exponential decay of the current as expected for non-dispersive transport. In the mixed conductor (C60)/insulator (BPhen) blends, it is useful to determine whether or not there is phase separation that will affect the conductive properties of the material. For this purpose, we examine the distribution of the current based on its autocorrelation function, a method that is useful in identifying spatial order. For a 2-D current map of current density, j(r), the autocorrelation function is defined as28–30: C (r ) = ∫ j (r ') j (r '+ r )dr '

(5)

Here, r is the separation between any two points in the image. In a completely random distribution, C(r) is a constant independent of separation. The autocorrelation function may, therefore, be used to look for phase segregation in a current map of a conductor-insulator blend such as used in the compound buffer.

7


Nano Letters

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

All materials were deposited by thermal evaporation in high vacuum (base pressure 9:1 BPhen:C60. The decrease in PCE is due to a reduction in fill factor (FF) from 61% to 57%. As shown in Fig. 3, the resistivity of the compound buffer gradually increases with BPhen fraction, with a sharp jump seen at 100% BPhen as the film becomes insulating. The resistivity of 100% BPhen of 6 ± 1 x 1011 Ω cm matches literature values35. Just as BPhen has a disproportionate effect on the exciton blocking characteristics of the buffer, C60 has a similarly pronounced effect on its charge transport properties. Even at 9:1 BPhen:C60, the resistance of the buffer is an order of magnitude less than neat BPhen, and the conduction is non-dispersive. This indicates that C60 undergoes nanoscale phase segregation that allows it to form conductive paths even at very high BPhen concentrations. This is further supported by the constant efficiency of OPV devices under 1 sun AM1.5G illumination employing the buffers with mixtures ranging from 1:3 BPhen:C60 to 4:1 BPhen:C60, as shown in the inset of Fig. 3. 2-D current maps of the compound buffer with varying blend ratios obtained by cAFM at an applied bias of -1.0 V are shown in Fig. 4. For the 1:2 C60:BPhen buffer the current density data are analyzed using Eq. (5). Autocorrelation of the current density vs distance was constant, indicative of a random distribution of conductor and insulator as shown in Fig. 5, inset. The transition from conducting C60 to mostly insulating BPhen is apparent in Fig. 4. The neat C60 sample shows bulk conductivity, with only isolated patches of slightly (20%-40%) decreased

11


Nano Letters

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 27

current. Layers with larger BPhen concentrations show increasingly larger insulating areas, with the 1:2 C60:BPhen mixture showing both a maximum current of 100 times less than pure C60, and conduction only through isolated regions. Layers with lower concentrations of C60 were too insulating to be imaged. Phase segregation, if it exists therefore occurs on scales below the cAFM instrument resolution of 10 nm. We further examined the nanostructure of the mixed buffer using XRD, with scans of three BPhen:C60 mixtures shown in Fig. 5. Neat C60 exhibits peaks corresponding to diffraction from the (002), and (112) crystal planes, with a domain size of 20 nm calculated from the peak Scherrer broadening36. For 1:1 C60:BPhen, a small peak corresponding to the (112) crystal plane is still visible. The broadened peak indicates that the C60 domains are only 3 ± 1 nm, which is too small to be imaged by cAFM. When the C60 was further diluted in the 1:9 BPhen:C60 sample, no XRD peaks are observed. We attribute this to the small quantity of C60 that cannot result in observable peaks at our instrumental resolution, although their absence does not rule out the possibility of small-scale phase segregation. Improved charge extraction through the compound buffer explains the improvement in both mixed and planar heterojunction devices.

It has been previously shown that planar

heterojunctions with a neat BPhen layer exhibit charge pile-up at the buffer-acceptor interface, resulting in a high steady-state charge density in the active layers, and hence a high rate of exciton-polaron quenching37. Exciton-polaron quenching is not present in mixed donor-acceptor heterojunctions due to the drastically reduced exciton concentration, but inefficient charge extraction in such devices leads to increased bimolecular recombination and a similar loss in device efficiency12.

12


Page 13 of 27

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

Nano Letters

Percolation theory has been extensively applied to understanding the conductivity of disordered conductor-insulator composites21–25. The transition point from insulator to conductor is dependent on the crystal structure, with the close-packed BPhen:C60 structures expected to have a conducting threshold of ~20% C60 concentration38–41. Surprisingly, we observe no such threshold in either the transient photocurrent or resistance measurements. This suggests that C60 undergoes nanoscale phase segregation when mixed in BPhen, which allows it to form conductive pathways even at extremely low concentrations. As discussed above, autocorrelation analysis of the cAFM scans of the compound buffer shows no ordering, suggesting that phase segregation occurs on a smaller scale than the instrument resolution of 10 nm, which is consistent with the XRD measurements in Fig. 5. For a layer of finite thickness, the threshold concentration for conduction follows:

pc − pc (L ) = AL−1.14

(6)

where pc is the critical conductor fraction for an infinitely thick layer, pc(L) is the critical fraction for a layer thickness of L, and lattice constant (assuming a cubic molecular arrangement), a. = 1.1 nm for34 C60. Also, A is an empirical constant. For the 10 nm thick compound buffer used in transient photocurrent measurements, L ~ 9a, and L ~ 91a for the 100 nm thick layer used in resistivity measurements. The lack of a transition point at high BPhen concentration is therefore attributed to either nanoscale phase segregation of C60, or large A. For thin layers used in our OPVs, either explanation suggests a high conductivity of the mixed buffer across almost the entire range (≤80% BPhen) of mixing fractions. In conclusion, we have examined charge and exciton transport through a compound fullerene:BPhen buffer.

We find that BPhen has a disproportionate effect on the exciton

13


Nano Letters

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 27

blocking characteristics of the buffer for a given mixing ratio. For example, 1:1 BPhen:C60 buffer was found to block 84±5% of incident excitons instead of the 75% expected from a linear interpolation between 0% and 100% BPhen, and other mixing fractions showed a similarly pronounced blocking effect. Indeed, C60 was also found to have a disproportionate effect on charge conduction, with non-dispersive transport observed even in 1:9 C60:BPhen layers, which is qualitatively different from the dispersive transport observed for neat BPhen. Furthermore, the lack of a percolative phase transition suggests small-scale phase segregation of C60 creates conductive pathways even in the 1:9 compound buffer. In combination, these characteristics indicate that the compound buffer can improve efficiency for a range of OPV architectures, as has been demonstrated in both planar and mixed heterojunction devices. The authors gratefully acknowledge Dr. Ardavan Oskooi, Professor Jeramy Zimmerman, Dr. Olga Griffith, and Mr. Quinn Burlingame for useful discussions. This work was partially supported by the Center for Solar and Thermal Energy Conversion at the University of Michigan (Department of Energy, Energy Frontier Research Center, Award No. DE-SC0000957, KJB, measurements, analysis, JAA, measurements, PFG), the Air Force Office of Scientific Research (SRF, analysis, project direction), and the SunShot Program of the Department of Energy, and NanoFlex Power Corp. (BS, measurements). One of the authors of this work (SRF) has an equity interest in one of its sponsors (NFP Corp.)

Supporting Information Available: Calculations of the optical field within a C60/C70 sample used in modeling. This material is available free of charge via the Internet at http://pubs.acs.org.

14


Page 15 of 27

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

Nano Letters

References (1)

O’Brien, D. F.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 1999, 74, 442–444.

(2)

Parthasarathy, G.; Adachi, C.; Burrows, P. E.; Forrest, S. R. Appl. Phys. Lett. 2000, 76, 2128–2130.

(3)

Bulović, V.; Tian, P.; Burrows, P. E.; Gokhale, M. R.; Forrest, S. R.; Thompson, M. E. Appl. Phys. Lett. 1997, 70, 2954–2954.

(4)

Burns, S. E.; Pfeffer, N.; Grüner, J.; Remmers, M.; Javoreck, T.; Neher, D.; Friend, R. H. Adv. Mater. 1997, 9, 395–398.

(5)

Peumans, P.; Yakimov, A.; Forrest, S. R. J. Appl. Phys. 2003, 93 (7), 3693–3723.

(6)

Chan, M. Y.; Lee, C. S.; Lai, S. L.; Fung, M. K.; Wong, F. L.; Sun, H. Y.; Lau, K. M.; Lee, S. T. J. Appl. Phys. 2006, 100, 094506.

(7)

Wang, N.; Yu, J.; Zang, Y.; Huang, J.; Jiang, Y. Sol. Energy Mater. Sol. Cells 2010, 94, 263–266.

(8)

Lassiter, B. E.; Wei, G.; Wang, S.; Zimmerman, J. D.; Diev, V. V.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 2011, 98, 243307.

(9)

Rand, B. P.; Li, J.; Xue, J.; Holmes, R. J.; Thompson, M. E.; Forrest, S. R. Adv. Mater.

2005, 17, 2714–2718. (10) Renshaw, C. K.; Schlenker, C. W.; Thompson, M. E.; Forrest, S. R. Phys. Rev. B 2011, 84, 045315. (11) Bartynski, A. N.; Trinh, C.; Panda, A.; Bergemann, K.; Lassiter, B. E.; Zimmerman, J. D.; Forrest, S. R.; Thompson, M. E. Nano Lett. 2013, 13, 3315–3320.

15


Nano Letters

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

(12) Xiao, X.; Bergemann, K. J.; Zimmerman, J. D.; Lee, K.; Forrest, S. R. Adv. Energy Mater.

2013, 4, 1301557. (13) Parthasarathy, G.; Shen, C.; Kahn, A.; Forrest, S. R. J. Appl. Phys. 2001, 89, 4986–4992. (14) Uhrich, C.; Wynands, D.; Olthof, S.; Riede, M. K.; Leo, K.; Sonntag, S.; Maennig, B.; Pfeiffer, M. J. Appl. Phys. 2008, 104, 043107. (15) Drechsel, J.; Männig, B.; Kozlowski, F.; Pfeiffer, M.; Leo, K.; Hoppe, H. Appl. Phys. Lett.

2005, 86, 244102. (16) Lunt, R. R.; Giebink, N. C.; Belak, A. A.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys.

2009, 105, 053711. (17) Bergemann, K. J.; Forrest, S. R. Appl. Phys. Lett. 2011, 99, 243303. (18) Tress, W.; Corvers, S.; Leo, K.; Riede, M. Adv. Energy Mater. 2013, 3, 873–880. (19) Street, R. A. Phys. Rev. B 2011, 84, 075208. (20) McNeill, C. R.; Hwang, I.; Greenham, N. C. J. Appl. Phys. 2009, 106, 024507. (21) Roldughin, V. I.; Vysotskii, V. V. Prog. Org. Coat. 2000, 39, 81–100. (22) Xie, N.; Shao, W.; Feng, L.; Lv, L.; Zhen, L. J. Phys. Chem. C 2012, 116, 19517–19525. (23) Rubin, Z.; Sunshine, S. A.; Heaney, M. B.; Bloom, I.; Balberg, I. Phys. Rev. B 1999, 59, 12196–12199. (24) Heaney, M. B. Phys. Rev. B 1995, 52, 12477–12480. (25) Viswanathan, R.; Heaney, M. B. Phys. Rev. Lett. 1995, 75, 4433–4436. (26) Pfister, G.; Scher, H. Adv. Phys. 1978, 27, 747–798. (27) Scher, H.; Montroll, E. W. Phys. Rev. B 1975, 12, 2455–2477. (28) Forrest, S. R.; Jr, Witten, T. A., J. Phys. Math. Gen. 1979, 12, L109. (29) Weitz, D. A.; Lin, M. Y.; Sandroff, C. J. Surf. Sci. 1985, 158, 147–164.

16


Page 16 of 27

Page 17 of 27

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

Nano Letters

(30) Cleary, T.; Samson, R.; Gentry, J. W. Aerosol Sci. Technol. 1990, 12, 518–525. (31) Xiao, X.; Zimmerman, J. D.; Lassiter, B. E.; Bergemann, K. J.; Forrest, S. R. Appl. Phys. Lett. 2013, 102, 073302. (32) Kazaoui, S.; Minami, N.; Tanabe, Y.; Byrne, H. J.; Eilmes, A.; Petelenz, P. Phys. Rev. B

1998, 58, 7689–7700. (33) Orlandi, G.; Negri, F. Photochem. Photobiol. Sci. 2002, 1, 289–308. (34) Arbogast, J. W.; Darmanyan, A. P.; Foote, C. S.; Diederich, F. N.; Whetten, R. L.; Rubin, Y.; Alvarez, M. M.; Anz, S. J. J. Phys. Chem. 1991, 95, 11–12. (35) Earmme, T.; Jenekhe, S. A. Appl. Phys. Lett. 2013, 102, 233305. (36) Krätschmer, W.; Lamb, L. D.; Fostiropoulos, K.; Huffman, D. R. Nature 1990, 347, 354– 358. (37) Verreet, B.; Malinowski, P. E.; Niesen, B.; Cheyns, D.; Heremans, P.; Stesmans, A.; Rand, B. P. Appl. Phys. Lett. 2013, 102, 043301. (38) Xu, X.; Wang, J.; Lv, J.-P.; Deng, Y. Front. Phys. 2014, 9, 113–119. (39) Powell, M. J. Phys. Rev. B 1979, 20, 4194–4198. (40) Van der Marck, S. C. Int. J. Mod. Phys. C 1998, 9, 529–540. (41) Lorenz, C. D.; Ziff, R. M. J. Phys. Math. Gen. 1998, 31, 8147–8157.

17


Nano Letters

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 27

Table 1. Comparison of simulated and measured exciton blocking characteristics of the mixed buffer. Pure BPhen and NPD are assumed to be a perfect blocker and quencher, respectively.

% Bphen

Predicted Blocking

Blocking Efficiency ϕ

Blocking Efficiency ϕ

Efficiency (%)

Relative to Blocker (%)

Relative to Quencher (%)

0

50

50 ± 2

46 ± 2

25

63

76 ± 3

69 ± 11

50

81

84 ± 5

82 ± 5

66

95

87 ± 4

87 ± 4

100

100

--

--

NPD

0

--

--

18


Page 19 of 27

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

Nano Letters

Figure 1. Photoluminescence (PL) intensity from an 80 nm thick C70 layer capped with an 8 nm thick layer of BPhen (exciton blocking), 8 nm thick NPD (quenching), or a 20 nm thick 1:1 BPhen:C60 compound buffer. The blocking efficiency, ϕ, is calculated from the PL intensity relative to the blocking and quenching caps. Although the compound buffer sample PL overlaps that of the blocking sample, it corresponds to ϕ = 84±5% instead of 100% due differing optical fields in the two samples.

Inset: Sample structure used for PL measurements. Samples were

excited at a wavelength of 540 nm, at an incidence angle θ=30° and measured at θ=60°.

Figure 2. Photocurrent vs. time (t) of an ITO / 10 nm MoOx / 54 nm DBP:C60 1:8 / buffer / 100 nm Ag organic photovoltaic (OPV) cell following the end of a 530 nm wavelength optical pulse. The buffer was either an 8 nm thick BPhen, or a 10 nm thick compound buffer capped with 5 nm BPhen to protect the underlying layer from damage during metal deposition. Current for the neat BPhen buffer (black squares) follows t-2, indicative of dispersive transport. All compound buffer BPhen fractions (25%, blue triangles; 50%, green diamonds; 90%, red circles) show an exponential decay transient (indicated by the dashed line as a guide to the eye), suggesting nondispersive transport. Inset: Time constants of exponential transient photocurrent decay versus BPhen fraction.

Figure 3. Resistivity as a function of BPhen fraction in a 100 nm thick C60:BPhen mixed layer. The resistitivity only increases sharply when all C60 is eliminated from the layer, suggesting that the C60 forms conductive paths even at 90% BPhen. Inset: Power conversion efficiency of OPVs

19


Nano Letters

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 27

under 1 sun AM1.5G illumination vs. BPhen fraction in a C60:BPhen buffer. Device efficiency was roughly constant from 20% to 80% BPhen, showing the buffer had efficient exciton blocking and charge extraction over this entire range. Device efficiency decreases at > 80% BPhen, primarily due to a decrease in fill factor from 61±1% to 57±1% due to reduced charge extraction efficiency.

Figure 4. Conducting atomic force microscope images (cAFM) of C60:BPhen blends, with (a)100 vol.%, (b) 75 vol.%, (c) 50 vol.%, and (d) 33 vol.% C60, taken at -1.0 V tip bias. The 100% C60 samples (a) show only small areas of reduced conduction. As the BPhen concentration is increased, patches of the film become increasingly insulating, until at 33 vol.% C60, conduction occurs only in isolated regions. This change from widespread to isolated conduction is accompanied by a 100X decrease in current.

Figure 5. X-ray diffraction pattern of 200 nm thick C60 (black line, data divided by 20), 1:1 BPhen:C60 (red dashed line), and 9:1 BPhen:C60 (blue dotted line) films on sapphire substrate taken in the Bragg-Brentano configuration using the Cu-Kα line. Peaks from the (002) and (112) crystal planes are visible for the neat C60 film, with peak widths corresponding to a domain size of 20 ± 1 nm. The 1:1 BPhen:C60 compound buffer shows a smaller and wider (112) peak, corresponding to a 3 ± 1 nm domain size. No crystallinity was observed in the 9:1 BPhen:C60 buffer. Inset: Averaged values of the current autocorrelation function C(r) vs. distance (r) of three, 2 µm x 1 µm 2:1 BPhen:C60 conductive-tip AFM images taken at 3 V bias. The magnitude

20


Page 21 of 27

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

Nano Letters

of C(r) is constant over two orders of magnitude of distance, from which we infer a random distribution of BPhen and C60 at all length scales. The cAFM spatial resolution is 10 nm.

21


Nano Letters

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

Page 22 of 27

Blocking θ Mixed 1:1

Cap C70 Quartz


Quenching

Nano Letters

100

Time constant (µs)

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

Photocurrent (Normalized)

Page 23 of 27

-1

10

2.0 1.8 1.6 1.4 1.2 1.0 0.8 0 10 20 30 40 50 60 70 80 Bphen BPhen Fraction Frac>on (%) (%)

-2

10

2

4

Time (µs)


6

Nano Letters

12 7.5

10

11

Efficiency (%)

7.0 6.5 6.0 5.5 5.0 20

40

60

80

100

BPhen Fraction Frac>on (%) (%) Bphen

10

Conducting Insulating

Resistivity (Ω cm)

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

10

Page 24 of 27

10

9

10

60

65

70

75

80

85

90

BPhen Fraction Frac>on (%) (%) Bphen ACS Paragon Plus Environment

95 100

Page 25 of 27

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

(a)

(b)

Nano Letters

100% C60

75% C60

(c)

50% C60

(d)

33% C60


Nano Letters

3

(002)

3

4.0x10

Counts

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

6.0x10

Page 26 of 27

2.0x103

Neat C60, divided by 20 1:1 C60:BPhen

(112)

1:9 C60:BPhen

0.0 8

10

12

14

16

18

2θ (degree)


20

22

24

Page 27 of 27

Nano Letters

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 ACS Paragon Plus Environment

Surprisingly High Conductivity and Efficient Exciton Blocking in

Recommend Documents