Letter pubs.acs.org/JPCL
Cite This: J. Phys. Chem. Lett. 2018, 9, 6864−6870
Efficiency of Charge-Transfer Doping in Organic Semiconductors Probed with Quantitative Microwave and Direct-Current Conductance Andrew J. Ferguson,#,† Obadiah G. Reid,#,†,‡ Sanjini U. Nanayakkara,† Rachelle Ihly,† and Jeffrey L. Blackburn*,† †
National Renewable Energy Laboratory, Golden, Colorado 80401, United States Renewable and Sustainable Energy Institute, University of Colorado Boulder, Boulder, Colorado 80303, United States
J. Phys. Chem. Lett. Downloaded from pubs.acs.org by UNIV OF NORTH DAKOTA on 11/20/18. For personal use only.
‡
S Supporting Information *
ABSTRACT: Although molecular charge-transfer doping is widely used to manipulate carrier density in organic semiconductors, only a small fraction of charge carriers typically escape the Coulomb potential of dopant counterions to contribute to electrical conductivity. Here, we utilize microwave and direct-current (DC) measurements of electrical conductivity to demonstrate that a high percentage of charge carriers in redoxdoped semiconducting single-walled carbon nanotube (s-SWCNT) networks is delocalized as a free carrier density in the π-electron system (estimated as >46% at high doping densities). The microwave and four-point probe conductivities of hole-doped s-SWCNT films quantitatively match over almost 4 orders of magnitude in conductance, indicating that both measurements are dominated by the same population of delocalized carriers. We address the relevance of this surprising one-to-one correspondence by discussing the degree to which local environmental parameters (e.g., tube−tube junctions, Coulombic stabilization, and local bonding environment) may impact the relative magnitudes of each transport measurement. “free” carrier density that can contribute to long-range transport.18 For example, several studies on semiconducting polymers (SPs) indicate that only 1−5% of charges injected by molecular dopants are mobile,19,20 consistent with expectations based on poor dielectric screening.17,21 Both electrical (direct-current, DC) and noncontact quasioptical (ν = 8 GHz to 10 THz) methods can be used to probe the conductance of organic and inorganic semiconductors. Quasi-optical methods using electromagnetic radiation in the gigahertz (GHz) and terahertz (THz) frequency range are powerful techniques and have been employed for measuring the dark conductance of a broad variety of semiconductors.22−24 As an example, we recently utilized an X-band microwave cavity combined with quantitative numerical modeling to measure the ca. 9 GHz dark conductivity of methylammonium lead iodide (MAPbI3) single crystals and related mixed composition organic-metal-halide perovskite thin films,25 as well as several metal−organic26 and covalent− organic27 frameworks. Because long-range transport to electrical contacts is not required for the dark microwave conductivity measurement, it is free from the influence of contact resistance and is less sensitive to the presence of grain boundaries, provided charge carriers do not diffuse across the
O
rganic semiconductors (OSCs), broadly defined as carbon-based semiconductors where conduction occurs predominantly through the π orbitals of a sp2-hybridized carbon network, are an important class of electronic materials for optical and electronic applications such as photovoltaics,1−3 photodetectors,2 field-effect transistors (FETs),4−6 and energyharvesting devices.7−9 Many of these applications require, or can be improved by, the control over a well-defined majority carrier concentration of either electrons or holes. Many strategies exist for doping OSCs, such as small molecules, semiconducting polymers (SPs), and single-walled carbon nanotubes (SWCNTs): electrostatic gating, electrochemically modulating the Fermi energy, or adsorption of redox molecules.8,10−12 Thin films of semiconducting single-walled carbon nanotubes (s-SWCNTs) have demonstrated strong performance in applications ranging from photovoltaic3,13 and thermoelectric energy harvesting7−10,14 to digital logic.15,16 Small-molecule redox dopants are particularly useful for doping s-SWCNT thin films because of the high surface area of individual s-SWCNTs and their bundles, as well as the intrinsic porosity of disordered s-SWCNT thin-film networks.9−11,14 Because of the low dielectric constant of OSCs, a significant fraction of the charges injected by physisorbed redox molecules is expected to remain predominantly localized near the dopant site as a Coulomb-bound “mirror charge” or “charge puddle”.12,17 The remainder escape the Coulomb potentials of the counterions, contributing to a delocalized © XXXX American Chemical Society
Received: October 5, 2018 Accepted: November 14, 2018
6864
DOI: 10.1021/acs.jpclett.8b03074 J. Phys. Chem. Lett. 2018, 9, 6864−6870
Letter
The Journal of Physical Chemistry Letters grains within the cycle time of the probing field.23,28 Similarly, it is not necessarily expected a priori that the microwave and DC conductivity measurements should match for doped OSCs, because the microwave measurement can probe charge carriers with appreciable mobility which do not contribute to DC conductivity. For example, the photoconductivity of individualized SWCNTs can be readily measured in solution using the same microwave probe frequency as in this report.29 In this study, we utilize a specially fabricated X-band microwave cavity to demonstrate a one-to-one correspondence of microwave and DC conductivity over nearly 4 orders of magnitude in conductivity for redox-doped p-type s-SWCNT thin films.9,10,14 We estimate that the population of “free” mobile charge carriers measured by both techniques represents >46% of the total charge carrier population at high doping densities, roughly an order of magnitude larger than typically observed for SPs, despite similar expectations for dielectric constants and Coulomb binding energies in both systems. Figure 1 shows the dependence of optical absorbance on the injected hole density [h+] for ca. 20−25 nm thick (6,5) s-
our previous studies,10 the volumetric density of injected holes scales with the amount of OA adsorbed on the SWCNT surfaces within the film, leading to the inverse correlation between OA concentration and S11 intensity/area.9,10,14 Quenching of the S11 is accompanied by the rise of a new absorption at ca. 1.06 eV (1170 nm), which has been attributed to the optical transition of a charged exciton or trion (Χ+).30,31 Figure 2a shows the dependence of the microwave cavity resonance curve on the doping density of a (6,5) film mounted
Figure 2. (a) Microwave resonance curves for (6,5):SMP s-SWCNT thin-film network doped p-type with progressively increasing hole concentration, [h+]. Arrows are a guide to the eye for increasing [h+]. (b) Simulated resonance curves (solid black lines) for five representative experimental resonance curves (colored circles) obtained at different hole concentrations. Extracted conductivity values are provided in the legend.
inside the cavity. The Supporting Information contains additional details on the cavity design, its dynamic range for conductance, and the numerical simulations used to extract conductance values. The microwave cavity is, to a good approximation, a damped-driven harmonic oscillator that exhibits a Lorentzian resonance line shape characterized by three parameters: resonant frequency (f 0, ca. 9.91 GHz for this cavity), power reflection coefficient on resonance (R0 = Pr/Pi where Pr is the reflected power and Pi is the incident power), and full width at half-maximum (fwhm). The quality factor (Q) of the cavity is given by Q= f 0/fwhm. As the hole density increases, the cavity resonance first deepens (solid arrow, Figure 2) before reaching zero (critical coupling, see the Supporting Information), after which the power reflection coefficient increases again and the fwhm begins to increase rapidly with increasing hole density (dashed arrow, Figure 2). We calculate the sample conductance by fitting numerical
Figure 1. Absorption spectra of OA-doped (6,5):PFO-BPy (a) and (6,5):SMP (b) thin-film networks where the concentration of the OA dopant is increased to inject progressively larger densities of holes into the film.
SWCNT films progressively doped p-type with triethyloxonium hexachloroantimonate (OA). Two types of (6,5) sSWCNT samples are studied here, with the sample details provided in Experimental Methods. As the concentration of OA is increased in the dopant solution, the oscillator strength of the S11 exciton transition is increasingly quenched in the doped thin films because of phase-space filling. As shown in 6865
DOI: 10.1021/acs.jpclett.8b03074 J. Phys. Chem. Lett. 2018, 9, 6864−6870
Letter
The Journal of Physical Chemistry Letters
Figure 3. 9.9 GHz microwave conductivity extracted from the resonance curves as a function of (a) the fractional bleaching of the first exciton, S11, transition and (b) the DC conductivity measured by linear four-point probe resistivity measurements, for two different (6,5) s-SWCNT films progressively doped p-type with OA.
nonzero: 0% < γfree < 100%.12,17 Excitonic optical transitions should be quenched by both free and localized charges, whereas the conductivity measurements are sensitive to the population of free charge carriers. Because the 9.9 GHz conductivity increases with the same dependence as the DC conductivity, with respect to progressively increasing absorption quenching (and hence total injected carrier density), this tells us that the bifurcation between free and localized charge carriers is fairly binary, in that there does not appear to be a “continuum” of charge carrier mobility values for the free carriers measured via microwave and DC conductivity. As we discuss below, the microwave measurement can probe charge carriers with significantly shorter confinement lengths than those probed in the current samples, i.e. carriers with appreciable local mobility but that do not contribute to DC conductivity. Interestingly, Figure 3 suggests that these types of carriers either do not exist or are an extreme minority within doped s-SWCNT thin films over many orders of magnitude for carrier density and conductivity. If a range of confinement lengths were available for carriers at a given carrier density, one would expect the microwave conductivity to exceed the DC conductivity because the microwave measurement would be sensitive to a large subset of charge carriers with appreciable local mobility and the DC measurement would be sensitive only to carriers delocalized enough to contribute to millimeterscale transport. To better understand the balance between free and localized holes in the films studied here, we can estimate the value of γfree. Currently, we restrict this analysis to the highest doping density, and the detailed calculation of γfree is given in the Supporting Information. Briefly, we use a value of 1.7 × 10−17 cm2/atom for the S11 optical cross section (σS11)32 and an exciton correlation length (ξe) of 2.0 ± 0.7 nm33 (i.e., complete quenching of S11 at a hole density of ca. 0.5 holes/ nm) to estimate a total volumetric hole density (nfree + nlocalized) of 1.7 × 1020 holes/cm3. In turn, we can estimate nfree from the peak sample conductivity (1.6 × 104 S m−1), but we must utilize an appropriate hole mobility. As an estimate for the hole mobility, we consider the typical range found in FET measurements from the literature for (6,5) s-SWCNT thin films, μh = 1−8 cm2 V−1 s−1.34−37 Using the lower limit of this range (μh = 1 cm2 V−1 s−1) produces unphysical values for γfree (370%). A mobility value of μh = 3.7 cm2 V−1 s−1 generates γfree = 100%, whereas μh = 8 cm2 V−1 s−1 generates γfree = 46%.
simulations of the microwave cavity characteristics to the data in Figure 2a, as described in detail in ref 25 and in the Experimental Methods and Supporting Information. Representative fits to experimental data are presented in Figure 2b for five different doping levels of the (6,5):SMP s-SWCNT film. The 9.9 GHz conductivity values extracted from these simulations are presented in the legend for each doping level. Figure 3a demonstrates that as carrier density increases and S11 is quenched (cf. Figure 1), the sample conductance measured at 9.9 GHz increases. Figure 3b compares the 9.9 GHz conductivity values extracted for two films from the dark microwave conductivity measurement to the DC conductivity measurements performed on the same two films. The 9.9 GHz and DC conductivity values coincide remarkably well over nearly 4 orders of magnitude change in conductivity for the (6,5) s-SWCNT films, as highlighted by the proximity of all data points to the black line (y = x) that serves as a guide to the eye in Figure 3b. The one-to-one match in Figure 3b indicates that, at least for redox-doped s-SWCNT thin films, the two measurements are sensitive to the same population of charge carriers, regardless of the total charge carrier density injected by the adsorbed redox molecules. The above discussion demonstrates the surprising nature of the near-perfect agreement between DC and GHz conductivity demonstrated in Figure 3 for doped s-SWCNTs. This correlation has several important implications for the impact of molecular dopants on carrier transport within highly enriched s-SWCNT thin films. First, when a redox molecule is used to dope s-SWCNTs with charge carriers via physisorption on the SWCNT surface, one can assume some fraction of the injected charge carriers remain localized near the dopant site as a “mirror charge”; the remainder escape the Coulomb potential of the counterion, contributing to a delocalized “free” carrier density that can be measured by microwave absorption and/or DC conductivity measurements. Here, we define the “branching ratio” between free and localized carriers (γfree) as the percentage of delocalized (nfree) charge carriers relative to the total charge carriers (nfree + nlocalized) at a given doping density. γfree = 100 ×
(n free
n free + nlocalized)
(1)
The branching ratio, γfree, cannot exceed 100% and, for doped OSCs with measurable DC conductivity, must be 6866
DOI: 10.1021/acs.jpclett.8b03074 J. Phys. Chem. Lett. 2018, 9, 6864−6870
Letter
The Journal of Physical Chemistry Letters
Figure 4. (a) Representative atomic force microscopy (AFM) image of a (6,5):SMP thin-film network, where the gray curves schematically illustrate the probability of locating a carrier positioned at the left-hand boundary of the dashed rectangle during the time scale of the microwave probe cycle, for Ld = 100 nm with the electric field of the microwave probe polarized in the direction indicated by the double-headed arrow. (Note: This illustration is not meant to signify the actual electric field orientation in the microwave measurement.) (b) Histograms of the junction− junction distances for (6,5):PFO-BPy and (6,5):SMP networks extracted from the AFM images, for a minimum of 100 bundle segments, indicating that the average junction−junction distance is less than the estimated carrier diffusion length.
Thus, from this selection of literature values for σS11, ξe, and μh, we estimate a lower limit of γfree = 46% for the percentage of free charge carriers injected by the redox dopants that are probed by both DC and microwave conductivity measurements. It is important to note here that the coincidence of the two conductance measurements does not indicate that the branching ratio γfree remains constant over all doping levels. However, the matched values of microwave and DC conductance indicate that if this branching ratio changes as a function of doping level, the two conductance measurements are dominated by the same population of charge carriers at all doping levels. Second, as alluded to above, the different length scales probed by the four-point probe and microwave conductivity experiments imply that they may not generally measure the same electrical conductivity on nanostructured samples. Because the probe pins utilized for the four-point probe measurement are spaced 1 mm apart and the average length of s-SWCNTs in the SWCNT dispersions (and presumably bundles within these films) is less than one micrometer, the DC conductivity measurement requires the carriers to traverse across many bundle−bundle junctions. In contrast, the microwave conductivity measurement does not require longrange (i.e., intertube) transport, only that the carriers are free to diffuse many times their natural (lattice and/or phonon) scattering length on the time scale of the probe cycle (100 ps). This means that for a mobility of 4−8 cm2 V−1 s−1, the transport length scale probed by the microwave measurement is ∼80−115 nm (Ld = (6Dt)1/2), much shorter than an individual nanotube. However, analysis of high-resolution atomic force microscopy (AFM) images (Figures 4 and S6) demonstrate that the average distance between bundle−bundle junctions (dj) is on the order of 50−65 nm, smaller than Ld if μ = 4−8 cm2 V−1 s−1. Because dj < Ld, the microwave conductance measurement probes carriers that encounter several junctions over the cycle time of the microwave probe (illustrated by the gray curves in Figure 4a). In contrast, the 1 mm probe spacing of the fourpoint probe measurement implies that this measurement samples many thousands of junctions. Thus, the close correspondence of both measurements indicates that probing
diffusive carriers that traverse a small number of junctions (i.e., ≤10) in the microwave measurement is a representative sampling of the drift transport statistics that the DC conductivity measurement samples as carriers traverse many thousands of junctions. Interestingly, the microwave measurement is not dominated by the unimpeded “intratube” transport of holes that would presumably have mobility values in the range of several hundred to ca. 1000 cm2 V−1 s−1.16 Thus, while these majority carriers are quite delocalized along the segments of SWCNT bundles between junctions, it appears that the bundle−bundle junctions may impose barriers to transport both between crossing bundles (“interbundle”) and along an individual bundle (“intrabundle”). In turn, the transport process probed in both measurements is likely limited by the hopping or tunneling of these delocalized carriers from site to site, over or through these transport barriers (bundle−bundle junctions). This conclusion is consistent with both the high estimate for γfree above and with the “heterogeneous” model popularized by Kaiser38,39 for transport within semiconducting polymer and SWCNT thin films. It is informative to compare the value for γfree estimated here for doped s-SWCNTs to estimates obtained for doped SPs from the literature. In a detailed recent study, Pingel and Neher found integer charge transfer from the molecular acceptor 2,3,5,6-tetrafluoro-7,7,8,8-tetracyanoquinodimethane (F4TCNQ) to poly(3-hexylthiophene) (P3HT) but also observed that only ca. 5% of the injected holes escaped the Coulomb potential of the negatively charged F4TCNQ counterions.19 This finding is in line with observations on other polymer systems, where the mobile charge injection efficiency of molecular dopants is typically found to be in the 1−5% range,20,40 an order of magnitude lower than the γfree value estimated here for doped s-SWCNTs. The equivalently low dielectric constant (ε ≈ 4) and high Coulomb binding energies (100s of meV) expected for both systems, doped sSWCNTs and doped SPs, suggest that permittivity effects alone are not sufficient to predict the value of γfree.17 In an attempt to rationalize the high γfree value observed here, we consider the potential impacts of the local bonding environment and density of states on the nature of injected 6867
DOI: 10.1021/acs.jpclett.8b03074 J. Phys. Chem. Lett. 2018, 9, 6864−6870
Letter
The Journal of Physical Chemistry Letters
ment and can be used to discern doping and transport mechanisms in a wide variety of both organic and inorganic semiconductors.
charges in SPs and s-SWCNTs. First, recent studies demonstrate that molecular doping of OSCs can be considered a two-step process, consisting of single-electron transfer from donor to acceptor and subsequent dissociation of the resulting ground-state integer charge-transfer complex (carrier release). Leo et al. suggest that carrier release is thermally activated, with activation energies typically below 25 meV (full range of ca. 9−50 meV) for several small molecule host−dopant combinations.18,41 Energetic disorder within the host OSC reduces the activation energies for carrier release such that they are nearly an order of magnitude lower than Coulomb binding energies.18 Second, charge transfer in OSCs is often associated with the formation of polarons, whereby bond angles and lengths are locally modified in the vicinity of the charge.12,22 Such polaronic stabilization tends to localize charges in place at the site of the dopant counterion42 (on top of, and distinct from, the expected Coulombic stabilization). Rigid structures (e.g., SWCNTs) tend to favor smaller reorganization energy and better delocalization of charges,43,44 whereas less rigid structures (e.g., SPs with torsional degrees of freedom about bonds and large bond-length rearrangements) lead to larger reorganization energies and more localized charges.45 The considerations discussed above imply that energetic disorder within redox-doped OSCs should contribute to appreciable carrier release even at room temperature and that the extent to which these carriers contribute to conductivity may be tied to structural properties that facilitate carrier delocalization. Thus, we speculate that the high value of γfree observed here for the relatively crystalline (at least locally) s-SWCNT system results, at least in part, from their particularly low charge-transfer reorganization energies. Injected charges in s-SWCNTs should have less polaronic character than those in SPs (i.e., not leading to large lattice distortions in the vicinity of the charge) and should be delocalized over many bonds within the SWCNT π-electron system. We note that similar conclusions have been reached for the doping efficiency of polymers that can adopt either crystalline or amorphous phases. As an example, the higher doping efficiency for F4TCNQ in regioregular P3HT, relative to regiorandom P3HT, has been attributed to lower torsional reorganization (and a correspondingly higher degree of charge delocalization) in the rigid crystalline domains of the former.46,47 In summary, the methods developed here establish a robust strategy for the rigorous quantification of quasi-optical conductivity and a path toward quantifying the branching ratio between free and localized charge carriers that can be applied to a broad range of both organic and inorganic semiconductors. While the estimate obtained here for γfree contains appreciable uncertainty, the ultimate conclusion is clear: a significant fraction of charge carriers injected by physisorbed redox molecules into s-SWCNT networks (>50%) are delocalized in the π electron system as a free carrier density. We suggest that the high value of γfree observed here is related to the inherently low charge-transfer reorganization energy within s-SWCNTs, which in turn leads to the efficient delocalization of charges away from the oppositely charged molecular counterion. This design rule can be generally applied to the realization of higher conductivity and higher-performing devices based on broad classes of redox-doped OSCs. The rigorous quantitative comparison of DC and dark microwave conductivity measurements demonstrated here provides critical benchmarking for the microwave conductivity measure-
■
EXPERIMENTAL METHODS Two separate (6,5) s-SWCNT inks were prepared for this study. One ink consisted of (6,5) s-SWCNTs wrapped by poly[(9,9-dioctylfluorenyl-2,7-diyl)-alt-co-(6,6′-[2,2′-bipyridine])] (PFO-BPy) and is labeled as (6,5):PFO-BPy. The other sample is prepared by exchanging the initial PFO-BPy wrapping polymer with a supramolecular cleavable polymer 1,1′-(((1E,1′E)-(9,9-didodecyl-9H-fluorene-2,7-diyl)bis(ethene-2,1-diyl))bis(6-methyl-4-oxo-1,4-dihydropyrimidine5,2-diyl))bis(3-dodecylurea) (SMP), and that sample is labeled as (6,5):SMP.14 The preparation of individualized sSWCNT inks is described in the Supporting Information, and the characterization of the (6,5):SMP sample will be described in more detail in a forthcoming publication. The (6,5) inks were spray-coated onto clean glass substrates, as described in the Supporting Information. The (6,5) s-SWCNT films were doped p-type with different hole concentrations by soaking films in solutions of triethyloxonium hexachloroantimonate (OA) in dichloroethane with varying OA concentrations at 78 °C.9,10,14 Following the soak in OA, each film was subsequently dipped in acetone (room temperature) to remove excess unbound OA from the surface. The absorption of the films was then measured using a Cary 5000 optical spectrophotometer in transmission mode. The absorption is not corrected for reflection. Nanotube bundle network topography images and film thicknesses were measured using an atomic force microscope (Park Systems XE/70 Atomic Force Microscope) in noncontact (tapping) mode using either Olympus AC160TS or Nanosensors Super Sharp SiliconTM − Non-Contact/Tapping mode (SSS-NCHR) AFM probes. The (6,5):PFO-bpy film was measured as 19 ± 3 nm thick, and the (6,5):SMP was measured as 24 ± 5 nm thick. The four-point probe method was used to measure DC conductivity of the (6,5) films, as described previously.9,10,14 Dark microwave conductivity was performed in an X-band microwave cavity, as described previously.23,25 To calculate the conductance of each sample for the microwave measurement, we used the commercially available COMSOL Multiphsics (v4.3) finite element package to solve Maxwell’s equations for the electromagnetic field distribution within the cavity, as described in additional detail in the Supporting Information. Replicates of each measurement were performed to get statistically meaningful average values and standard deviations (see the Supporting Information). For each film, the DC and 9.9 GHz measurements were performed back-to-back (in no particular order) for the same film at a given carrier concentration (as determined by the OA concentration used to dope the film). The calculations of DC and quasi-optical conductivities were performed in a “blind” fashion, in that the calculations were done independently and no correction factors were subsequently applied to conductance values obtained by either method.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.8b03074. 6868
DOI: 10.1021/acs.jpclett.8b03074 J. Phys. Chem. Lett. 2018, 9, 6864−6870
Letter
The Journal of Physical Chemistry Letters
■
(9) MacLeod, B. A.; Stanton, N. J.; Gould, I. E.; Wesenberg, D.; Ihly, R.; Owczarczyk, Z. R.; Hurst, K. E.; Fewox, C. S.; Folmar, C. N.; Holman Hughes, K.; et al. Large N- and P-Type Thermoelectric Power Factors from Doped Semiconducting Single-Walled Carbon Nanotube Thin Films. Energy Environ. Sci. 2017, 10, 2168−2179. (10) Avery, A. D.; Zhou, B. H.; Lee, J.; Lee, E.-S.; Miller, E. M.; Ihly, R.; Wesenberg, D.; Mistry, K. S.; Guillot, S. L.; Zink, B. L.; et al. Tailored Semiconducting Carbon Nanotube Networks with Enhanced Thermoelectric Properties. Nat. Energy 2016, 1, 16033. (11) Mistry, K. S.; Larsen, B. A.; Bergeson, J. D.; Barnes, T. M.; Teeter, G.; Engtrakul, C.; Blackburn, J. L. N-Type Transparent Conducting Films of Small Molecule and Polymer Amine Doped Single-Walled Carbon Nanotubes. ACS Nano 2011, 5, 3714−3723. (12) Salzmann, I.; Heimel, G.; Oehzelt, M.; Winkler, S.; Koch, N. Molecular Electrical Doping of Organic Semiconductors: Fundamental Mechanisms and Emerging Dopant Design Rules. Acc. Chem. Res. 2016, 49, 370−378. (13) Guillot, S. L.; Mistry, K. S.; Avery, A. D.; Richard, J.; Dowgiallo, A.-M.; Ndione, P. F.; van de Lagemaat, J.; Reese, M. O.; Blackburn, J. L. Precision Printing and Optical Modeling of Ultrathin Swcnt/C60 Heterojunction Solar Cells. Nanoscale 2015, 7, 6556−6566. (14) Norton-Baker, B.; Ihly, R.; Gould, I. E.; Avery, A. D.; Owczarczyk, Z. R.; Ferguson, A. J.; Blackburn, J. L. Polymer-Free Carbon Nanotube Thermoelectrics with Improved Charge Carrier Transport and Power Factor. ACS Energy Lett. 2016, 1, 1212−1220. (15) Brady, G. J.; Joo, Y.; Wu, M.-Y.; Shea, M. J.; Gopalan, P.; Arnold, M. S. Polyfluorene-Sorted, Carbon Nanotube Array FieldEffect Transistors with Increased Current Density and High on/Off Ratio. ACS Nano 2014, 8, 11614−11621. (16) Brady, G. J.; Way, A. J.; Safron, N. S.; Evensen, H. T.; Gopalan, P.; Arnold, M. S. Quasi-Ballistic Carbon Nanotube Array Transistors with Current Density Exceeding Si and GaAs. Sci. Adv. 2016, 2, e1601240. (17) Gregg, B. A.; Chen, S.-G.; Cormier, R. A. Coulomb Forces and Doping in Organic Semiconductors. Chem. Mater. 2004, 16, 4586− 4599. (18) Tietze, M. L.; Benduhn, J.; Pahner, P.; Nell, B.; Schwarze, M.; Kleemann, H.; Krammer, M.; Zojer, K.; Vandewal, K.; Leo, K. Elementary Steps in Electrical Doping of Organic Semiconductors. Nat. Commun. 2018, 9, 1182. (19) Pingel, P.; Neher, D. Comprehensive Picture of P-Type Doping of P3HT with the Molecular Acceptor F4TCNQ. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 87, 115209. (20) Zhang, Y.; de Boer, B.; Blom, P. W. M. Controllable Molecular Doping and Charge Transport in Solution-Processed Polymer Semiconducting Layers. Adv. Funct. Mater. 2009, 19, 1901−1905. (21) Ghosh, R.; Chew, A. R.; Onorato, J.; Pakhnyuk, V.; Luscombe, C. K.; Salleo, A.; Spano, F. C. Spectral Signatures and Spatial Coherence of Bound and Unbound Polarons in P3HT Films: Theory Versus Experiment. J. Phys. Chem. C 2018, 122, 18048. (22) Tsutsui, Y.; Okamoto, H.; Sakamaki, D.; Sugiyasu, K.; Takeuchi, M.; Seki, S. Landscape of Charge Carrier Transport in Doped Poly(3-Hexylthiophene): Noncontact Approach Using Ternary Combined Dielectric, Paramagnetic, and Optical Spectroscopies. J. Phys. Chem. Lett. 2018, 9, 3639−3645. (23) Reid, O. G.; Yang, M.; Kopidakis, N.; Zhu, K.; Rumbles, G. Grain-Size-Limited Mobility in Methylammonium Lead Iodide Perovskite Thin Films. ACS Energy Lett. 2016, 1, 561−565. (24) Dicker, G.; de Haas, M. P.; Warman, J. M.; de Leeuw, D. M.; Siebbeles, L. D. A. The Disperse Charge-Carrier Kinetics in Regioregular Poly(3-Hexylthiophene). J. Phys. Chem. B 2004, 108, 17818−17824. (25) Reid, O. G.; Moore, D. T.; Li, Z.; Zhao, D.; Yan, Y.; Zhu, K.; Rumbles, G. Quantitative Analysis of Time-Resolved Microwave Conductivity Data. J. Phys. D: Appl. Phys. 2017, 50, 493002. (26) Dolgopolova, E. A.; Brandt, A. J.; Ejegbavwo, O. A.; Duke, A. S.; Maddumapatabandi, T. D.; Galhenage, R. P.; Larson, B. W.; Reid, O. G.; Ammal, S. C.; Heyden, A.; et al. Electronic Properties of Bimetallic Metal−Organic Frameworks (MOFS): Tailoring the
Detailed experimental methods; cavity design and numerical simulations for dynamic range and conductivity calculation; comparison of optical absorption for (6,5) s-SWCNT dispersion and film; (6,5) sSWCNT network morphology; estimation of free carrier “branching ratio” (γfree) (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: jeff
[email protected]. ORCID
Andrew J. Ferguson: 0000-0003-2544-1753 Obadiah G. Reid: 0000-0003-0646-3981 Jeffrey L. Blackburn: 0000-0002-9237-5891 Author Contributions #
A.J.F. and O.G.R. contributed equally to this work.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was authored by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by the Solar Photochemistry Program, Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences, U.S. Department of Energy (DOE). The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paidup, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.
■
REFERENCES
(1) Cheng, P.; Li, G.; Zhan, X.; Yang, Y. Next-Generation Organic Photovoltaics Based on Non-Fullerene Acceptors. Nat. Photonics 2018, 12, 131−142. (2) Arnold, M. S.; Zimmerman, J. D.; Renshaw, C. K.; Xu, X.; Lunt, R. R.; Austin, C. M.; Forrest, S. R. Broad Spectral Response Using Carbon Nanotube/Organic Semiconductor/C60 Photodetectors. Nano Lett. 2009, 9, 3354−3358. (3) Blackburn, J. L. Semiconducting Single-Walled Carbon Nanotubes in Solar Energy Harvesting. ACS Energy Lett. 2017, 2, 1598− 1613. (4) Sirringhaus, H. 25th Anniversary Article: Organic Field-Effect Transistors: The Path Beyond Amorphous Silicon. Adv. Mater. 2014, 26, 1319−1335. (5) Franklin, A. D. Nanomaterials in Transistors: From HighPerformance to Thin-Film Applications. Science 2015, 349, aab2750. (6) Schneider, S.; Brohmann, M.; Lorenz, R.; Hofstetter, Y. J.; Rother, M.; Sauter, E.; Zharnikov, M.; Vaynzof, Y.; Himmel, H.-J.; Zaumseil, J. Efficient N-Doping and Hole Blocking in Single-Walled Carbon Nanotube Transistors with 1,2,4,5-Tetrakis(Tetramethylguanidino)ben-zene. ACS Nano 2018, 12, 5895−5902. (7) Blackburn, J. L.; Ferguson, A. J.; Cho, C.; Grunlan, J. C. CarbonNanotube-Based Thermoelectric Materials and Devices. Adv. Mater. 2018, 30, 1704386. (8) Oshima, Y.; Kitamura, Y.; Maniwa, Y.; Yanagi, K. Fabrication of Thermoelectric Devices Using Precisely Fermi Level-Tuned Semiconducting Single-Wall Carbon Nanotubes. Appl. Phys. Lett. 2015, 107, 043106. 6869
DOI: 10.1021/acs.jpclett.8b03074 J. Phys. Chem. Lett. 2018, 9, 6864−6870
Letter
The Journal of Physical Chemistry Letters Density of Electronic States through MOF Modularity. J. Am. Chem. Soc. 2017, 139, 5201−5209. (27) Hughes, B. K.; Braunecker, W. A.; Bobela, D. C.; Nanayakkara, S. U.; Reid, O. G.; Johnson, J. C. Covalently Bound Nitroxyl Radicals in an Organic Framework. J. Phys. Chem. Lett. 2016, 7, 3660−3665. (28) O’Connor, B. T.; Reid, O. G.; Zhang, X.; Kline, R. J.; Richter, L. J.; Gundlach, D. J.; DeLongchamp, D. M.; Toney, M. F.; Kopidakis, N.; Rumbles, G. Morphological Origin of Charge Transport Anisotropy in Aligned Polythiophene Thin Films. Adv. Funct. Mater. 2014, 24, 3422−3431. (29) Park, J.; Reid, O. G.; Blackburn, J. L.; Rumbles, G. Photoinduced Spontaneous Free-Carrier Generation in Semiconducting Single-Walled Carbon Nanotubes. Nat. Commun. 2015, 6, 8809. (30) Dowgiallo, A.-M.; Mistry, K. S.; Johnson, J. C.; Blackburn, J. L. Ultrafast Spectroscopic Signature of Charge Transfer between SingleWalled Carbon Nanotubes and C60. ACS Nano 2014, 8, 8573−8581. (31) Matsunaga, R.; Matsuda, K.; Kanemitsu, Y. Observation of Charged Excitons in Hole-Doped Carbon Nanotubes Using Photoluminescence and Absorption Spectroscopy. Phys. Rev. Lett. 2011, 106, 037404. (32) Schöppler, F.; Mann, C.; Hain, T. C.; Neubauer, F. M.; Privitera, G.; Bonaccorso, F.; Chu, D.; Ferrari, A. C.; Hertel, T. Molar Extinction Coefficient of Single-Wall Carbon Nanotubes. J. Phys. Chem. C 2011, 115, 14682−14686. (33) Lüer, L.; Hoseinkhani, S.; Polli, D.; Crochet, J.; Hertel, T.; Lanzani, G. Size and Mobility of Excitons in (6, 5) Carbon nanotubes. Nat. Phys. 2009, 5, 54. (34) Rother, M.; Brohmann, M.; Yang, S.; Grimm, S. B.; Schießl, S. P.; Graf, A.; Zaumseil, J. Aerosol-Jet Printing of Polymer-Sorted (6,5) Carbon Nanotubes for Field-Effect Transistors with High Reproducibility. Adv. Electron. Mater. 2017, 3, 1700080. (35) Brohmann, M.; Rother, M.; Schießl, S. P.; Preis, E.; Allard, S.; Scherf, U.; Zaumseil, J. Temperature-Dependent Charge Transport in Polymer-Sorted Semiconducting Carbon Nanotube Networks with Different Diameter Distributions. J. Phys. Chem. C 2018, 122, 19886. (36) Bottacchi, F.; Petti, L.; Späth, F.; Namal, I.; Tröster, G.; Hertel, T.; Anthopoulos, T. D. Polymer-Sorted (6,5) Single-Walled Carbon Nanotubes for Solution-Processed Low-Voltage Flexible Microelectronics. Appl. Phys. Lett. 2015, 106, 193302. (37) Schießl, S. P.; Rother, M.; Luttgens, J.; Zaumseil, J. Extracting the Field-Effect Mobilities of Random Semiconducting Single-Walled Carbon Nanotube Networks: A Critical Comparison of Methods. Appl. Phys. Lett. 2017, 111, 193301. (38) Kaiser, A. B. Electronic Transport Properties of Conducting Polymers and Carbon Nanotubes. Rep. Prog. Phys. 2001, 64, 1. (39) Kaiser, A. B. Systematic Conductivity Behavior in Conducting Polymers: Effects of Heterogeneous Disorder. Adv. Mater. 2001, 13, 927−941. (40) Lee, J.-H.; Kim, H.-M.; Kim, K.-B.; Kabe, R., Jr.; Anzenbacher, P.; Kim, J.-J. Homogeneous Dispersion of Organic P-Dopants in an Organic Semiconductor as an Origin of High Charge Generation Efficiency. Appl. Phys. Lett. 2011, 98, 173303. (41) Higgins, A.; Mohapatra, S. K.; Barlow, S.; Marder, S. R.; Kahn, A. Dopant Controlled Trap-Filling and Conductivity Enhancement in an Electron-Transport Polymer. Appl. Phys. Lett. 2015, 106, 163301. (42) Bakulin, A. A.; Rao, A.; Pavelyev, V. G.; van Loosdrecht, P. H. M.; Pshenichnikov, M. S.; Niedzialek, D.; Cornil, J.; Beljonne, D.; Friend, R. H. The Role of Driving Energy and Delocalized States for Charge Separation in Organic Semiconductors. Science 2012, 335, 1340−1344. (43) Ihly, R.; Mistry, K. S.; Ferguson, A. J.; Clikeman, T. T.; Larson, B. W.; Reid, O. G.; Boltalina, O. V.; Strauss, S. H.; Rumbles, G.; Blackburn, J. L. Tuning the Driving Force for Exciton Dissociation in Single-Walled Carbon Nanotube Heterojunctions. Nat. Chem. 2016, 8, 603−609. (44) Kunai, Y.; Liu, A. T.; Cottrill, A. L.; Koman, V. B.; Liu, P.; Kozawa, D.; Gong, X.; Strano, M. S. Observation of the Marcus Inverted Region of Electron Transfer from Asymmetric Chemical
Doping of Pristine (N,M) Single-Walled Carbon Nanotubes. J. Am. Chem. Soc. 2017, 139, 15328−15336. (45) Coffey, D. C.; Larson, B. W.; Hains, A. W.; Whitaker, J. B.; Kopidakis, N.; Boltalina, O. V.; Strauss, S. H.; Rumbles, G. An Optimal Driving Force for Converting Excitons into Free Carriers in Excitonic Solar Cells. J. Phys. Chem. C 2012, 116, 8916−8923. (46) Gao, J.; Stein, B. W.; Thomas, A. K.; Garcia, J. A.; Yang, J.; Kirk, M. L.; Grey, J. K. Enhanced Charge Transfer Doping Efficiency in JAggregate Poly(3-Hexylthiophene) Nanofibers. J. Phys. Chem. C 2015, 119, 16396−16402. (47) Gao, J.; Niles, E. T.; Grey, J. K. Aggregates Promote Efficient Charge Transfer Doping of Poly(3-Hexylthiophene). J. Phys. Chem. Lett. 2013, 4, 2953−2957.
6870
DOI: 10.1021/acs.jpclett.8b03074 J. Phys. Chem. Lett. 2018, 9, 6864−6870
