Peijian Wangâ and Michael D. Barnesâ â¡. â Department of Physics and â¡Department of Chemistry, University of Massachusetts, Amherst, Amherst, M...
Oct 19, 2017 - We report on spatially correlated wavelength-resolved photoluminescence and Kelvin probe force microscopy to probe ground state ...
Jun 23, 2014 - Mapping Bright and Dark Modes in Gold Nanoparticle Chains using. Electron Energy Loss Spectroscopy. Steven J. Barrow,. â . David Rossouw,.
Jul 30, 2012 - Using novel bottom-up fabrication methods to construct such plasmonic ... poly(sodium 4-styrenesulfonate) in a 0.1 M solution of sodium chloride in water. ..... *Fax +41(0)22 379 61 03; Tel +41(0)22 379 65 52; e-mail.
Dec 6, 2010 - In this paper, we therefore study the electromagnetic responses of a cluster ... For a fixed polarization, the extinction Ïext and scattering Ïsca cross sections .... are proportional to the volume of a fictitious sphere with its radi
Dec 6, 2010 - Applied Physics, National Chengchi UniVersity, Taipei 116, Taiwan, and Department of ... National Taiwan UniVersity, Taipei 106, Taiwan.
Sep 5, 2017 - The circular polarization arises from the Zeeman splitting of exciton levels, whereas lifetime reduction results from a polarization-preserving field-induced mixing of exciton levels. We analyze these experimental findings in terms of a
Feb 21, 2018 - (44) Although the major features of valence and conduction bands originate from the inorganic sub lattice (PbX3â), the fast motion of dipolar organic cations may strongly affect the dynamics of charge carriers.(43, 45) Of particular
... Transition Between Bright and Dark Excitons in. Donor/Acceptor Conjugated Polyelectrolyte Complexes. William R. Hollingsworth,. â a. Timothy J. Magnanelli,.
Sep 18, 2011 - Clapp , A. R.; Medintz , I. L.; Mattoussi , H. Forster Resonance Energy ..... Park , S. J.; Link , S.; Miller , W. L.; Gesquiere , A.; Barbara , P. F. Effect ...
Subscriber access provided by Universitaetsbibliothek | Johann Christian Senckenberg
Communication
Disentangling ‘bright’ and ‘dark’ interactions in ordered assemblies of organic semiconductors Peijian Wang, and Michael D. Barnes Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.7b03394 • Publication Date (Web): 19 Oct 2017 Downloaded from http://pubs.acs.org on October 20, 2017
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.
Disentangling ‘bright’ and ‘dark’ interactions in ordered assemblies of organic semiconductors Peijian Wang,1 and Michael D. Barnes*1,2 1
Department of Physics Department of Chemistry University of Massachusetts, Amherst Amherst, Massachusetts 0100, United States 2
Abstract:We report on spatially correlated wavelength-resolved photoluminescence and Kelvin Probe Force Microscopy (KPFM) to probe ground state charge-transfer coupling and its correlation with pi-stacking order in nanoscale assemblies of a small molecule ntype organic semiconductor, tetraazaterrylene (TAT). We find a distinct upshift in surface potential contrast (SPC) corresponding to a decrease in work function in TAT in the transition from disordered spun-cast films to ordered crystalline nanowire assemblies, accompanied with a nanowire size dependence in the SPC shift suggesting that the shift depends on both ground state charge transfer interaction and a size(volume) dependent intrinsic doping associated with the nitrogen substitutions. For the smallest nanowires studied (surface height ≈ 10-15 nm), the SPC shift with respect to disordered films is + 110 meV, in close agreement with recent theoretical calculations. These results illustrate how ‘dark’ (ground-state) interactions in organic semiconductors can be distinguished from ‘bright’ (excited-state) exciton coupling typically assessed by spectral measurements alone. Keywords: work function, disorder, charge transfer coupling, kelvin probe force microscopy, photoluminescence spectra, molecular packing
3 Introduction In organic electronics, fundamental processes of exciton and charge transport are mediated by inter-molecular interactions which are highly sensitive to the molecular packing geometry.1-3 This structural information in aggregated organic semiconductors is typically inferred from vibronic structure in absorption or emission, as a result of the sensitivity of the vibration-less (0-0) transition intensity (relative to that of the 0-1, and higher vibronic side bands) to the sign and magnitude of a point-dipole interaction between neighboring chromophores.4, 5 This well-known excited-state (‘bright’) interaction’ gives rise to so-called H or J-aggregate spectral features corresponding to relative diminution or enhancement of 0-0/0-1 intensity ratios which can be used to directly extract exciton coupling strength. More recent theoretical and experimental work point to a more complex picture of exciton coupling which is mediated by both point-dipole coupling and charge-transfer (CT) interactions, leading to so-called H(J) J(H) aggregate model in which the first label (H or J) denotes the sign of the dipoledipole coupling, and the second label denotes the sign of the CT interaction.6-8 The mixing of these different coupling modes is predicted to profoundly impact spectral properties and have important consequences on exciton and charge transport in ordered organic semiconductor assemblies.9, 10 However, because these two coupling modes are entangled in absorption or photoluminescence measurements, very little is known experimentally regarding the relative strength of these interactions and the impact on ground state electronic structure of the aggregate. In this paper, Figure 1. Energy level schematic for the TAT system in the we show with spatially transition from isolated (uncoupled) molecule to crystalline registered aggregate. The ground states are coupled by a hole-transfer photoluminescence and interaction, th, which raises the valence band maximum by ≈ 2 KPFM measurements, |
|. Note that the surface potential contrast (SPC) is measured relative to the Pt reference, thus larger (positive) clear connections SPC corresponds to a decrease in energy separation between between pi-stacking order vacuum and Fermi level. as signaled by changes in vibronic intensity ratios, and surface potential contrast (SPC) that derive from energy shifts in the Fermi level associated with ground state coupling. In the HJ aggregate model developed by Spano and coworkers, the usual Holstein Hamiltonian describing Frenkel exciton (FE) coupling via dipole-dipole interaction is augmented with short-range CT interactions as well as FE-CT mixing terms.9, 10The CT
and electron transfer integrals that describe the energy associated with transfer of hole/electron between adjacent molecules, and they are linked with the CT coupling energy by the second order perturbation theory.8, 11, 12 Because this interaction requires wave-function overlap, both and
are short-range interactions (acting over a distance of ≈ 3.5 A) and are exquisitely sensitive (in sign and magnitude) to sub-Å atomic registration displacements between molecules.6,7,8,9 In the ground state, the coupled electronic wavefunction interaction is described by
with an energy bandwidth of ≈ 4 |
|; if each individual molecule is a charge-neutral singlet, then each state in the ground state band is filled and the valence band maximum is ≈ 2 |
|. In the general case of organic semiconductors where the coupling is strongest along the pi-stacking direction, it becomes difficult – if not impossible- to parse the different coupling modes (FE vs CT) by photoluminescence (PL) methods alone. In this work, we show that combining (bright) spectrally resolved PL with (dark) electrostatic scanning probes allows quantitative energy location of the Fermi level and direct Figure 2. (Left) Experimental schematic of combined PL correlation with pi-stacking spectral and KPFM imaging setup. The scanning probe order, and direct interrogation measurements are performed above the sample, while of ground state couplings. optical interrogation utilizes epi-illumination and highnumerical aperture (100x, 1.4 N.A.) objective coupled with
Kelvin Probe Force Microscopy an imaging spectrograph. (a) shows a surface height image of TAT nanowires, and (b) the corresponding PL image (KPFM) is a scanning probe technique that has proven to be along with (c) PL spectrum obtained in the highlighted region. The insets in the spectrum shows the molecular especially powerful in structure of TAT. elucidating connections between molecular packing morphology and electronic properties,13-17 and tailored material properties.18, 19 This technique exploits a capacitive interaction between a metal-coated tip (typically Pt), and the material under investigation; the surface potential contrast (SPC) in units of volts represents the work function difference between the tip and sample with a lateral resolution defined by the nominal tip radius, and energy resolution of the order of a few meV. Previous work in our laboratory combined spectrally resolved photoluminescence and Kelvin Probe Force Microscopy (KPFM) in separate measurements has indicated effects of morphology on electronic properties of nanowire and nanocluster systems. 20 21, 22 Interestingly, spectrally similar low-molecular weight (MW) H-type aggregates show work functions that can vary by as much as 200 meV, while high-MW J-type aggregates may be completely dissimilar spectrally, yet have almost identical SPC, thus highlighting the
5 need for spatially registered PL and surface potential measurements.22 In a similar spirit, Frisbie and coworkers have illustrated the influence of perturbations of packing order caused by mechanical strain on work function of rubrene single crystals.23, 24 In the present study, we used 7,8,15,16 tetraazaterrylene (TAT), a nitrogen-substituted terrylene system that functions as an n-type semiconductor.25 TAT readily forms crystalline aggregates from either physical vapor transport,25, 26 or solution-based selfassembly.12 TAT has a number of interesting and important properties useful for the interrogation of (packing) structure correlated electronic properties: First and foremost, TAT aggregates are highly photostable and its vibronic structure is preserved upon aggregation where the intensity profile encodes information on packing structure and pistacking order. Second, TAT crystallizes in a monoclinic unit cell with no known polymorphs.25,26 Finally, TAT aggregates have been studied extensively theoretically,8, 9 where both FE and CT coupling matrix elements have been computed thus providing a quantitative basis for experimental comparison with theory.
6 Figure 1 shows an energy level schematic and illustration of the experimental design. The ground state electronic coupling of TAT monomers generates a HOMO band with an energy bandwidth of ≈ 4 |
|, where
is the holetransfer integral. Semi-empirical and density functional calculations of
by Spano and co-workers gave 545 and 502 cm-1 (62 meV).8 Because each TAT Figure 3. (A, B) TEM images and diffraction patterns (insets) of TAT nanowire monomer is a bundles and isolated wires prepared by self-assembly from toluene solution. neutral singlet in The wires show characteristic reflections (q1, q2) of a monoclinic unit cell with the ground state, real space distances of 3.9 Å, and 9.6 Å associated with crystallographic a and each level in the b-c, directions respectively (see Ref. 25). (C, D) Surface height and SPC image ground state of isolated TAT nanowire, along with (E)axially resolved PL spectra obtained from the either end and middle. The spectra clearly indicate different packing band is filled, motifs at the nanowire ends which induces spectra similar to disordered TAT. which should The inset words reveal the spectral properties type. produce an upshift in the work function of ≈ 2 |
|. This shift in the VBM should depend only on the magnitude of |
| (which in turn depends on packing arrangement and order), but independent of sign. In the experiment, isolated nanowire structures are illuminated using an inverted microscope and the photoluminescence is imaged with a CCD camera or imaging spectrograph; with suitable nanowire orientation along the direction of the entrance slit of the spectrograph, spectra along the nanowire axis was readily imaged. Figure 2 shows a schematic of the experimental setup. TAT samples were drop-cast from dilute toluene solution onto plasma-cleaned glass coverslips. A Digital Instruments BioScope AFM (with AppNano ANSCM-PA Pt-coated Si cantilever probes), was used to probe surface height and surface potential contrast of nanostructures in the field of view of the microscope (see Supporting Information for experimental details on registration of the PL and AFM coordinates). PL imaging and spectral measurements were performed in an epi-illumination geometry with 440 nm laser excitation. The typical experimental protocol involved first measuring the surface topology and surface potential contrast (relative to Pt) in a two-pass interleaved scan with a 40nm lift height. Then the AFM probe was retracted to perform the PL imaging on the same region of interest. Results and Discussion
7 Figure 3 shows representative surface height, SPC, and axially-resolved spectra from a single TAT nanowire, along with a TEM micrograph of a TAT nanowire solution-grown from toluene. The center portion of the TAT nanowire shows a spectrum that is essentially indistinguishable from physical vapor transport (PVT) grown single crystals, and the (same) crystal structure is supported by the TEM measurement. The only deviation from crystal PL spectra is found near the ends of the nanowire structure which have a vibronic intensity envelope significantly different from that of the crystal. As seen from the surface height measurements, the nanowire structures grown from solution have a ‘ribbon’-like quality, and are typically 50-100 nm in width, and range in height from 10 to 50 nm. Similar to the spectra, the surface potential contrast (SPC) has relatively little axial variation along the nanowire, consistent with a uniform packing structure, with some significant deviations at the nanowire ends. We also observe some significant bundling of nanowire structures resulting in much larger – presumably polycrystalline – structures. Figure 4 shows contour plots of SPC vs sample height from different TAT samples: spun-cast TAT thin-films, isolated nanowire, and different nanowire bundles. The contour plots were generated from correlated height and SPC values from different nanowire (or film) samples from which density plots were made by 2D binning. Figure 4. (A) Contour plots of SPC vs. sample height comparing spun-cast The contours shown in TAT thin film, isolated nanowire, and different nanowire bundles. (B) Figure 4 (A) represent Representative PL spectra of spun-cast TAT, nanowire, and single-crystal 2D Gaussian fits to the TAT (the spectrum for single crystal TAT is adapted with permission from density plots; the raw Ref 26. Copyright 2014 Royal Society of Chemistry). height/SPC data, scatter plots and density plots are given in Supporting Information. Also included in Figure 4 are representative PL spectra of the different samples, along with the PL spectra of physical vapor transport grown TAT single crystal.26 The spun-cast film shows a PL spectrum with vibronic intensity ratios (00/01) of 1.4; virtually identical to dilute TAT solution, and those of single TAT molecules assayed in polymer-supported thin films. (The preparation of spun cast film is in Supporting Information) The TAT nanowire and nanowire bundles show PL spectra very similar to single-crystal TAT with only small variations in origin and higher vibronic side-band intensities. Measured SPC from the different nanowire and nanowire bundles, however, vary significantly: The mean SPC for the thin isolated wire (presumably similar to that shown in Fig. 3B) is 1.72 V, while the different representative
8 nanowire bundles show much higher average SPCs (increased Fermi level) that are weakly correlated with surface height. The nitrogen atom substitutions, located at “shoulder” positions, intrinsically dope carriers into the ensemble. This “intrinsic doping”, when size-volume and carrier delocalization increases, further raises the Fermi level in addition to the ground state interaction, and diminishes the energy gap between Fermi and vacuum level, thus reducing the work function (increasing SPC relative to Pt). Considering the oligoacene structural similarity of TAT and pentacene and fully established physical parameters for the model organic semiconductor pentacene, an estimate of the energy shift, ∆E, from the VBM to the Fermi level (see Fig. 1) can be made from the carrier density formula from semiconductor physics, p = NV Exp[-∆E / kB T], where Nv is the effective density of states, given by 2(2 π m* kB T/h2)3/2, with m* as the effective carrier mass, kB is Boltzmann’s constant, h is Planck’s constant, and T is the absolute temperature.27 Using values adopted from literature, in which m* ≈ 4 me, NV is ≈ 2 x 1020 cm-3 at T = 300 K.28 and number of holes, p, by Kobayashi and coworkers,29 p = 3.25 x 1016 cm-3, we estimate an energy shift, ∆E ≈ 230 meV, which is in the same order of magnitude and reasonable agreement with the SPC upshift between the thin nanowire, and the nanowire bundles. Figure 5 shows the correlation between average SPC and 00-01 vibronic intensity ratio for several different TAT nanowire and nanocluster samples. Because the 00— 01 intensity ratio encodes information on both pistacking order and packing geometry, variations in SPC amongst the different TAT samples reveal the connection with CT-modified ground state (dark interactions), and PL spectral properties (bright interactions). The superficially H-like (00-01 intensity ratios < Figure 5. Correlation of average SPC with vibronic intensity ratio for spun-cast TAT, nanowires, and both H-like and J-like 1) nanowire samples show nanoclusters. The color shading indicates H-like (blue), and SPC ≈ 300 meV higher J-like (red) spectral signatures. The similar SPC for both (increase in Fermi level) than cluster types suggests a similar magnitude of
for the Jthe uncoupled (disordered) and H-like clusters (nominally 85 meV). spun-cast TAT film. Interestingly, we find that both small clusters (with diffraction limited PL images) as well as some of the wire ends, show PL spectral characteristic of J-like aggregates (00-01 intensity ratio > 1) that also show elevated SPC relative to disordered TAT implying fairly strong ground state
9 coupling in these systems as well. We believe these J-like nanoclusters have a qualitatively different packing geometry (illustrated by the structural cartoon) which may have a CT (ground-state) interaction – nominally 85 meV) similar in magnitude as the crystalline TAT but with possibly a different sign. As suggested in recent work by Spano and coworkers, a slip of ≈ 3 Å along the long (x-) molecular axis – approximately 30% of molecular length – could invert the sign of the CT interaction,9 while also promoting a more ‘head-to-tail’ dipole-dipole orientation thus contributing to the J-like spectral signatures of these species. While more work is required to specifically identify the packing motif underlying the spectra, realization of such J-J aggregate (both CT and FE interactions are negative) could provide a route to significantly enhanced exciton mobility. Summary and Conclusions We have demonstrated how ‘dark’ ground-state CT inter-chromophore interactions and the correlation with pi-stacking order can be probed quantitatively in isolated nanowire structures using spatially registered PL imaging and Kelvin Probe Force Microscopy. The measured upshift in Fermi level in thin TAT nanowires relative to disordered spuncast films of ≈ 110 meV is in good agreement with the theoretically computed value of 2|
| ≈ 120 meV. We observed a more or less monotonic decrease in Fermi level with decreasing pi-stacking order as signaled by increasing vibronic intensity ratio. Interestingly, we also observed a nanowire size dependence in the Fermi level which we attribute to a volume-dependent intrinsic n-type doping which further increases the energy difference between the Fermi level and valence band maximum(VBM). More generally, these studies point to a route towards a more detailed understanding of intermolecular interactions in organic semiconductors and realization of tailored packing motifs that enhance charge and/or energy transport in a directionally specific way.
Associated Content The Supporting Information is available free of charge on the ACS Publications website: http://pubs.acs.org. Raw scatter plots and density plots by 2D binning; The raw height/SPC data of TAT nanowire bundles, thin nanowire, nanoclusters; PL image of Jlike TAT and H-like TAT; Registration of the PL and AFM coordinates. Author Information Corresponding Author: * E-mail: [email protected]. Acknowledgements We thank Ned Burnett and Alejandro Briseno of the Polymer Science and Engineering Dept. at UMass Amherst for the TEM imaging of the TAT nanowires. We also thank Connor Boyle and Dhandapani Venkataraman of the Chemistry Department at UMass Amherst for the synthesis of the TAT used in this study. Funding for this work was provided by the US Department of Energy. The single crystal TAT spectrum is adapted with permission from Ref. 26. Copyright 2014 Royal Society of Chemistry
References 1. Haken, H.; Reineker, P. Z Phys 1972, 249, 253-268. 2. Beljonne, D.; Yamagata, H.; Bredas, J. L.; Spano, F. C.; Olivier, Y. Phys Rev Lett 2013, 110, 226402(5). 3. Ryno, S. M.; Risko, C.; Bredas, J. L. Acs Appl Mater Inter 2016, 8, 14053-14062. 4. Spano, F. C. Accounts Chem Res 2010, 43, 429-439. 5. Clark, J.; Silva, C.; Friend, R. H.; Spano, F. C. Phys Rev Lett 2007, 98, 206406(10). 6. Yamagata, H.; Spano, F. C. J Chem Phys 2012, 137, 184901, 1-14. 7. Yamagata, H.; Pochas, C. M.; Spano, F. C. J Phys Chem B 2012, 116, 14494-14503. 8. Yamagata, H.; Maxwell, D. S.; Fan, J.; Kittilstved, K. R.; Briseno, A. L.; Barnes, M. D.; Spano, F. C. J Phys Chem C 2014, 118, 28842-28854. 9. Hestand, N. J.; Tempelaar, R.; Knoester, J.; Jansen, T. L. C.; Spano, F. C. Phys Rev B 2015, 91,195315, 1-7. 10. Hestand, N. J.; Spano, F. C. J Chem Phys 2015, 143, 244707, 1-16. 11. Hestand, N. J.; Yamagata, H.; Xu, B. L.; Sun, D. Z.; Zhong, Y.; Harutyunyan, A. R.; Chen, G. G.; Dai, H. L.; Rao, Y.; Spano, F. C. J Phys Chem C 2015, 119, 22137-22147. 12. Labastide, J. A.; Thompson, H. B.; Marques, S. R.; Colella, N. S.; Briseno, A. L.; Barnes, M. D. Nat Commun 2016, 7, 10629, 1-7. 13. Shao, G. Z.; Rayermann, G. E.; Smith, E. M.; Ginger, D. S. J Phys Chem B 2013, 117, 46544660. 14. Giridharagopal, R.; Shao, G. Z.; Groves, C.; Ginger, D. S. Mater Today 2010, 13, 50-56. 15. Pingree, L. S. C.; Rodovsky, D. B.; Coffey, D. C.; Bartholomew, G. P.; Ginger, D. S. J Am Chem Soc 2007, 129, 15903-15910. 16. Umeda, K.; Kobayashi, K.; Ishida, K.; Hotta, S.; Yamada, H.; Matsushige, K. Jpn J Appl Phys 1 2001, 40, 4381-4383. 17. Ostrowski, D. P.; Lytwak, L. A.; Mejia, M. L.; Stevenson, K. J.; Holliday, B. J.; Vanden Bout, D. A. Acs Nano 2012, 6, 5507-5513. 18. Hutchison, J. A.; Liscio, A.; Schwartz, T.; Canaguier-Durand, A.; Genet, C.; Palermo, V.; Samori, P.; Ebbesen, T. W. Adv Mater 2013, 25, 2481-2485. 19. Liscio, A.; Palermo, V.; Samori, P. Accounts Chem Res 2010, 43, 541-550.
11 20. Baghgar, M.; Barnes, A. M.; Pentzer, E.; Wise, A. J.; Hammer, B. A. G.; Emrick, T.; Dinsmore, A. D.; Barnes, M. D. Acs Nano 2014, 8, 8344-8349. 21. Baghgar, M.; Labastide, J. A.; Bokel, F.; Hayward, R. C.; Barnes, M. D. J Phys Chem C 2014, 118, 2229-2235. 22. Baghgar, M.; Barnes, M. D. Acs Nano 2015, 9, 7105-7112. 23. Wu, Y. F.; Chew, A. R.; Rojas, G. A.; Sini, G.; Haugstad, G.; Belianinov, A.; Kalinin, S. V.; Li, H.; Risko, C.; Bredas, J. L.; Salleo, A.; Frisbie, C. D. Nat Commun 2016, 7, 10270, 1-9. 24. Ellison, D. J.; Lee, B.; Podzorov, V.; Frisbie, C. D. Adv Mater 2011, 23, 502-507. 25. Fan, J.; Zhang, L.; Briseno, A. L.; Wudl, F. Org Lett 2012, 14, 1024-1026. 26. Wise, A. J.; Zhang, Y.; Fan, J.; Wudl, F.; Briseno, A. L.; Barnes, M. D. Phys Chem Chem Phys 2014, 16, 15825-15830. 27. Hatch, R. C.; Huber, D. L.; Hochst, H. Phys Rev B 2009, 80,081411(4). 28. Yi, H. T.; Gartstein, N.; Podzorov, V. Sci Rep-Uk 2016, 6, 23650, 1-11. 29. Kobayashi, S.; Nishikawa, T.; Takenobu, T.; Mori, S.; Shimoda, T.; Mitani, T.; Shimotani, H.; Yoshimoto, N.; Ogawa, S.; Iwasa, Y. Nat Mater 2004, 3, 317-322.
