Molecular electrical doping of organic semiconductors has emerged as key strategy to improve the performance of organic (opto-)electronic devices and to extend their range of functionality. However, despite impressive success in practical applications, the fundamental processes underlying molecular electrical doping are still not fully understood to date. Notably, this is reflected in the observation that the two important classes of organic semiconductors, small conjugated molecules and conjugated polymers, exhibit a vastly different phenomenology upon doping with no satisfactory explanation put forward to date. The present contribution provides an overview on pertinent literature and recent own work focusing on the role of ground-state charge transfer complex formation upon doping prototypical organic semiconductors. In particular, in the prospect of providing a unifying picture of molecular electrical doping, profound differences between the fundamental mechanisms at work in small molecular and polymeric compounds of identical chemical composition and similar microstructure are discussed.

We present multi-photon OLEDs where enhanced light emission was achieved by stacking two OLEDs utilizing a regular device architecture (top cathode) and an intermediate charge carrier generation layer (CGL) for monolithic device interconnection. With respect to future printing processes for organic optoelectronic devices, all functional layers were deposited from solution. The CGL comprises a low-work function zinc oxide layer that was applied from solution under ambient conditions and at moderate processing temperatures and a high-work function interlayer that was realized from various solution processable precursor-based metal oxides, like molybdenum-, vanadium- and tungsten-oxide. Since every injected electron-hole pair generates two photons, the luminance and the current efficiency of the tandem OLED at a given device current are doubled while the power efficiency remains constant. At a given luminance, the lower operating current in the tandem device reduces electrical stress and improves the device life-time. ToF-SIMS, TEM/FIB and EDX analyses provided evidence of a distinct layer sequence without intermixing upon solution deposition.

In this work we focus on the formation of different kinds of polarons upon doping of thin films of poly(3-hexylthiophene) (P3HT). We elucidate profoundly the cyclic voltammogram to fit the number of oxidation peaks and, hence, to calculate the number of charge carriers per thiophene unit. These values are correlated with in-situ spectroelectrochemical measurements ranging from UV-VIS to mid-IR. In-situ UV-VIS measurements show a gradual decrease of the HOMO-LUMO transition. At the same time a new broad absorption band arises, that continuously increases up to a certain potential, above which it disappears. With in-situ spectroelectrochemical mid-IR experiments conducted in the attenuated total reflection (ATR) mode, we observe a broad absorption band, which shifts its maximum as oxidation proceeds. Interestingly, this shifting occurs at the same applied potential as the disappearing of this arised absorption band in the UV-VIS. Moreover, new doping induced infrared active vibration (IRAV) modes appear which hardly change in the different oxidation levels. Electron paramagnetic resonance (EPR) measurements are performed, indicating the persistent formation of radical cations up to a certain potential, above which the EPR signal changes its shape. These results emphasize the formation of polarons which are converted into another species as oxidation proceeds. We present a possible model which supports our spectroscopic data. Additionally, all these measurements have been confirmed by in-situ chemical doping experiments using iodine as an oxidation agent.

It has been discovered that polythiophene molecules (P3AT) aggregate into 1D nanostructures by solution-induced crystallization. In the case of p-doped P3AT, the doped polymer cations could have steric arrangements that are different from their neutral forms. Herein the doping-induced molecular packing processes are studied by absorption spectroscopy and atomic force microscopy (AFM). The doping-induced conformation change and Coulomb interactions influence the molecular packing of the P3AT nanostructures. On the other hand, the P3AT π-π stacked structures are more efficient at delocalizing charges, which could also enhance the charge transfer from dopants. The absorption spectra show distinctive bands for molecular packing and doping products, respectively. By systematically changing the dopant concentration, quantitative kinetic studies are carried out to correlate the correlated growth dynamics of the two absorption bands. p-doping are shown to substantially facilitate the π-π stacking of the conjugated polymers into 1D aggregates even at marginally low p-dopant concentrations. The doped 1D polymer nanostructures are also compared with the non-doped ones by AFM to reveal their morphological differences. This investigation will greatly strengthen our understanding on the chemical doping process for conjugated polymers.

Molecular doping is an important strategy to optimize organic electronic devices. Transport layers in OLEDs and solar cells are typically doped at less than few percent weight with molecular dopants to finely tune the carrier density and thus modulate carrier transport. The microscopic mechanism is believed to be the formation of charge transfer complexes between the dopant and the semiconductor. The formation of this hybrid states, as well as the complete charge transfer from dopant to semiconductor, strongly relies on the wavefunction overlap between the molecules. For example, it is known that in efficient doping of polymers it is important to have a cofacial arrangement between the dopant and the π orbital system of the polymer backbone. This scenario may be more unpredictable when the conjugated polymer is made of donor acceptor moieties in the repeat unit, which recently is one of the most common chemical architectures pursued for high performance polymer devices. The central question is how does doping work in these alternating copolymers?
In this communication, following our recent studies on doping of conjugated copolymers (1,2), we present an explanation for the low doping efficiency in donor-acceptor copolymers. By performing a combined experimental and theoretical study, focussed on infrared active molecular vibrations, we demonstrate how the geometrical position and orientation of the dopant F4-TCNQ (Tetrafluoro-tetracyano-quinodimethane) influences the doping efficiency in the copolymer PCPDTBT. We first analyze the remarkable changes in the conjugated copolymer vibrational modes upon doping with low molar ratio concentrations of F4-TCNQ, ranging from 1 to 7% , where 1% correspond to one dopant molecule every 100 copolymer repeat units. Appreciable changes in the vibrational spectrum due to transfer of charge from the polymer to the dopant acceptor only occur above 4%. We contrast these results with the same experiments performed on the homopolymer PCPDT, i.e. the polymer based on the donor moiety of PCPDTBT only. Surprisingly, although the homopolymer has the same ionization potential as PCPDTBT, signatures of infrared active vibrational modes from charges are recorded for a doping concentration as low as 1%. We have based our interpretation on density-functional-theory calculations and modelled the vibrational modes of complexes between the polymers and F4-TCNQ. Theory shows how the experimental spectra are a sum of different geometrical configurations in the dopant/polymer complexes. Remarkably, complete charge transfer between the copolymer PCPDTBT and the dopant occurs only when the dopant is docking to the donor moiety of the chain, unravelling one of the reasons for low doping efficiency (3). We further show experiments using the dopant F6-TCNNQ and discuss the role of molecular size. (1) Deschler, et al. PRL 107, 127402 (2011). (2) Tunc et al. Org. Elect. 13, 290 (2012). (3) Di Nuzzo et al. Nature Comm. 6, 6440 (2015).

The ability to precisely control the equilibrium carrier concentration in organic semiconducting devices is of great interest. As early as 1977, it was shown that the conductivity of polyacetylene could be systematically controlled over 11 orders of magnitude by doping using a range of halogens. Today, thermally evaporated organic light-emitting diodes (OLEDs) benefit from the use of doped transport layers; lowering operation voltages, reducing the device&’s sensitivity to electrode work functions, and enhancing device lifetime. However, in solution-processed organic optoelectronic devices the choice and accessibility of doped injection or transport layers is more limited. The ability to solution process doped layers is of extreme importance for high throughput production of organic electronic devices via roll-to-roll or ink-jet printing. In this talk, I will discuss the approach of using Lewis acids to modify the absorption and charge transport properties of π-conjugated systems with an available lone pair of electrons. By modulating the stoichiometry and strength of the added Lewis acid, a wide-range of optical properties were accessible without the need for rigorous synthesis. By formation of the Lewis acid-base adduct, electron density could be withdrawn from the π-system, narrowing the band gap. Using the Lewis acid tris(pentafluorophenyl)borane (BCF), we demonstrate that the absorption, the photoluminescence and electroluminescence, and charge transport of conjugated polymers can be modulated. The adduct formation leads to lower energy absorption and emission transitions, extended PL lifetimes, and increased solid state quantum yields. These properties allowed the strategy to be successfully demonstrated in PLED devices to modify the electroluminescence (EL) characteristics while keeping the luminance efficiency constant. Furthermore, addition of the Lewis acid effectively p-dopes the hole transport in the parent polymer, leading to increases in the free hole density and thus the charge-carrier mobility. This methodology is advantageous since the polymer, BCF, and the adduct have excellent solubility in organic solvents, negating the need for polar co-solvents that result in substandard and thin polymer layers.

The strong electron(hole)-phonon coupling in semiconducting polymers lends itself to the application of vibrational spectroscopies to investigate both the precise effect of electrons or holes on the polymer chain and to determine the degree of chemical, electrochemical or electrostatic doping. Raman microscopy is particularly useful to monitor local doping and spatially resolve doping profiles, e.g., within a transistor channel, as has been shown for operating electrolyte-gated single-walled carbon nanotube network transistors (Adv. Mater. 26 7986-92, 2014).
Here we demonstrate how Raman microscopy can be used to monitor the doping level in electrochemically and chemically doped semiconducting polymers ranging from regio-regular poly(3-hexylthiophene), which shows a significant broadening of the symmetric C=C stretching mode around 1460 cm^-1 with increasing p-doping, to high-mobility donor-acceptor polymers such as the diketopyrrolo-pyrrole thiophene copolymer DPPT-TT. Using the specific spectral features of the charged semiconducting polymers and the mapping capabilities of the Raman microscope it is possible to investigate the efficiency of chemical dopants and to image carrier density and doping profiles in-situ, for example, in electrochemical transistors. The effect of doping on the Raman modes of low and high mobility polymers may also provide further insight into the origin of the different charge transport regimes in these semiconductors.

The low ionization potentials of highly reducing organic n-dopants and host radical anions makes n-doping a much greater challenge than p-doping. However, many modern electronics such as transistors, complementary circuits, light-emitting diodes, photovoltaics, and thermoelectrics either require or benefit from both n- and p-type conduction. It is a long-standing and partially realized goal of the organic electronics community to mass produce devices by exploiting modern printing processes. To accomplish this will require new materials, device architectures, and processing methods. The goal of this work was to advance the state of solution-processable conductive n-type organic materials for use in printed organic solar cells, transistors, and thermoelectric devices.
A bottom-up mechanistic approach was used to design new organic n-dopants and n-dopable host semiconductors. In previous work, the n-doping mechanism of 1,3-dimethylbenzimidazole (DMBI-H) derived dopants was studied in solution, and it was discovered that DMBI-H dopants react with fullerenes by hydride transfer. Following this study, a new class of dimeric dopants ((DMBI)2) were developed to eliminate the dependence of the doping reaction on the hydrogenation thermodynamics of the host. The (DMBI)2 dopants were employed in a thorough synthetic, spectroscopic, and electrical study of new n-dopable conjugated polymers. Ultimately, several polymers with higher conductivities when n-doped than current state-of-the-art materials were identified. The results of this study indicate that the polaron delocalization length is the most relevant parameter to optimize to achieve high conductivity n-doped polymer films. Building upon this work, a class of self-n-doped polymers with remarkable stability in air were developed. Finally, as a demonstration of the utility of the new materials reported in this work, solution-processed organic solar cells with n-doped layers were fabricated.

Although open shell organic molecules (free radicals) are of fundamental interest, with few exceptions, they are generally reactive and unstable. We report solution-processed stable diradicaloids with temperature tunable electrical conductivity via a mechanism of self-doping, a result that is promising for advanced device applications. Electrical measurements show a remarkable electrical self-doping for the diradicaloids at room temperature (RT) with the doping strength highly tunable and reversible with temperature (T), attributed to the formation of radical ion species within the solid state. The self-doping in diradicaloids is confirmed by intentional doping with an external dopant and T-dependent X-ray photoelectron spectroscopy that shows an increase of the nitrogen cations accompanied with a stoichiometric change of the nitrogen in the triazinyl rings at higher temperatures.

We report the status of our research program to develop solution-processable air-stable p-doped polymers with ultrahigh workfunctions larger than 5.4 eV and up to 6.0 eV in collaboration with our industry partner. Previously it was thought that ultrahigh workfunctions beyond 5.3 eV are not possible in ambient because of oxidation of the water couple. We report that these problems can be alleviated to produce novel hole-injection layers (HILs) are able to provide ohmic injection into solution-processed organic semiconductors with deep ionization potentials up to 6.0 eV, and have sufficient air and thermal stability to afford a useful processing window. We also demonstrate that suitable members of this family of HILs also provide higher open-circuit voltages, and surprisingly also higher fill factors, than conventional PEDT:PSSH for organic solar cells fabricated with photoactive layers that have larger ionization potentials than 5.0 eV. In one example with PCDTTBT: PCBM as photoactive layer, power conversion efficiencies are improved by 40%from 5% to 7% simply by using these HILs in place of PEDT:PSSH. We will also report the first organic CMOS circuit elements built by electrode differentiation leveraging on our hole- and electron-injection interlayer technologies.

Semiconducting polymers have inherent trap, or gap, states due to structural/morphological imperfections and irregularities, or to impurities. These states can lead to impeded charge transport and, in the case of organic photovoltaic systems, result in an increased probability of charge recombination. It has been shown that low molecular doping levels (<1%) engender the passivation of gap states in organic semiconductor films. When co-deposited during vacuum deposition of C₆₀, an increasing ratio of n-dopants causes an exponential increase in the conductivity until the dopant density approaches the trap density of C₆₀, after which contribution of carriers to the conduction band begins. Contrary to the relatively even spatial distributions of dopant molecules that one would expect to be afforded by evaporation, processing of films from solution which are doped at low levels could result in less homogenous films, which might in turn impact their electronic properties. The details of the solution doping and deposition process can then significantly affect the distribution of the dopants in the film, depending on miscibility of the ion pairs with solvents and host materials.
This presentation reports doping of the widely-studied semi-crystalline donor polymers P3HT, PTB7, & PCDTBT with soluble molecular p-dopants derived from molybdenum tris[dithiolene] exhibiting high electron affinities ~5.5 eV, which resulted in passivation of trap states, but not without marked effects on material order, even at the lowest levels examined (10^-4 wt%). In highly ordered P3HT, the incorporation of the dopant is found to affect phase formation and consistently lower photovoltaic PCE. However with PTB7, and PCDTBT, where charge transport is less dependent on short-range order, doping results in an improvement in current density, and PCE, at dopant ratios of ca. 0.1-0.2 wt%.

The lifetime of the OLED device has been one of the critical hurdles for commercialization, and it is especially true for blue OLED. Notably, the intrinsic chemical stability of host materials for emission layer of blue OLED becomes more relevant due to the high emission energy of the blue guest materials, which may cause the degradation of surrounding host materials and eventually the failure of the entire device. Therefore, understanding the nature of the bonds and its dissociation phenomena of the blue OLED host materials are utmost importance for the better lifetime as well as the performance. Here, we present a systematic computational study on C-N bond dissociation of a series of carbazole(Cz)-dibenzothiophene(DBT) derivatives (Cz(x)DBTs, x=1, 2, 3, and 4), which is essentially a simpler variations of one of the commercialized high triplet host materials for blue OLED: DCzDBT (2,8-di(9H-carbazol-9-yl)dibenzo[b,d]thiophene). Our calculations show that the CzDBTs also have high triplet energies and compatible HOMO/LUMO levels. We found that the extra electron of CzDBT anion was redistributed towards the C-N bond upon bond stretching, forming a partially cleaved bond, and that is why the anions are more susceptible to C-N bond dissociation than cations or neutral species. Based on this, the substitution effect on the C-N bond stability was examined, and the results confirmed that an electron-withdrawing substituent with a strong inductive effect can improve the C-N bond stability of the CzDBTs without compromising their excited states energies.

Energy level alignment at organic heterojunctions plays a crucial role not only in determining the performance of organic electronic devices, but also in correlating the electronic interaction of organic semiconductors. Here photoemission spectroscopy is used to investigate the energy level alignment of organic bulk heterojunction (BHJ) which formed via solution mixing of conjugated polymer (electron donor) and fullerene derivatives (electron acceptor). Owing to the preferential vertical phase segregation with polymer dominated on the surface, we find that an abnormally high [6,6]-phenyl-C₇₁-butyric acid methyl ester (PCBM) to polymer ratio is required to be able to acquire the signals from the both components. By using this approach, we have successfully differentiated the interface dipole and energy level bending at the BHJs by photoemission spectroscopy. In addition, the effective band gap extracted from the energy difference between the highest-occupied molecular-orbital (HOMO_D) of the polymer and the lowest-unoccupied molecular-orbital (LUMO_A) of PCBM have excellent agreement with the values obtained from temperature dependent open-circuit voltage (V_OC) measurement in photovoltaic cells. The results demonstrate a facile approach to determine the energy gap in BHJ thin films, and sight light into the fundamental correlation between the energetic alignment and photovoltage in organic solar cells.

The design and performance of organic photovoltaic cells is dictated in part by the magnitude of the exciton diffusion length (L_D). Despite the importance of this parameter, there have been few investigations connecting L_D and materials purity. Here, we investigate L_D for the organic small molecule N,Nprime;-bis(naphthalen-1-yl)-N,Nprime;-bis(phenyl)-benzidine (α-NPD) as native impurities are systematically removed from the material. Thin films deposited from the as-synthesized material yield an L_D, as measured by photoluminescence quenching, of (3.9 ± 0.5) nm with a corresponding photoluminescence efficiency (eta;_PL) of (25 ± 1)%. After purification by thermal gradient sublimation, the value of L_D is increased to (4.7 ± 0.5) nm with a corresponding eta;_PL of (33 ± 1)%. Using a model of diffusion by Förster energy transfer, the variation of L_D with purity is predicted as a function of eta;_PL and is in good agreement with measurements. The observed increase in eta;_PL and concomitant increase in the exciton lifetime suggest a reduction in the concentration of exciton quenching impurities with purification. Interestingly, a similar behavior is also observed as a function of the thin film deposition rate. Films grown from the purified material at a high deposition rate give L_D = (5.3 ± 0.8) nm with eta;_PL= (37 ± 1)%. The results of this work highlight the role of impurities in determining L_D, while also providing design rules for materials purity and device processing.

Fluorescent and phosphorescent organic light-emitting diodes (OLEDs) were fabricated using a charge-transfer-featured compound, 6-{3,5-bis-[9-(4-t-butylphenyl)-9H-carbazol-3-yl]-phenoxy}-2-(4-t-butylphenyl)-benzo[de]isoquinoline-1,3-dione (CzPhONI), as a host. CzPhONI exhibits a twisted intramolecular CT character, and the density functional theory calculations have shown that the overlap between highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of CzPhONI is zero. The HOMO of CzPhONI is merely located on the dicarbazylphenyl moiety, while the LUMO is only distributed on the naphthalimide unit. Therefore, the excited state of CzPhONI should possess small exchange energy, resulting in the small DE_ST, which is beneficial for triplet energy up-conversion.
Fluorescent dyes of rubrene and 4-(dicyanomethylene)-2-t-butyl-6-(1,1,7,7-tetramethyljulolidyl-9-enyl)-4H-pyran (DCJTB) and phosphorescent dyes of bis[2-(4-tertbutylphenyl)benzothiazolato-N,C^2prime;]iridium (acetylacetonate) [(t-bt)₂Ir(acac)] and tris(1-phenylisoquinoline)iridium(III) [Ir(piq)₃] were doped into CzPhONI to form the emissive layers of OLEDs. The results showed that the external quantum efficiencies of both fluorescent and phosphorescent OLEDs were exceeding their theoretical limits. Based on the analysis of EL characteristics, the high device performance of fluorescent and phosphorescent OLEDs was attributed to both efficient energy transfer and triplet energy up-conversion, and direct exciton formation was also involved in phosphorescent OLEDs. In addition, the host film possessed high thermal and morphological stabilities due to the attachment of steric bulks on host molecule, resulting in the high doping concentration for both fluorescent and phosphorescent dyes. The use of a intramolecular charge-transfer-featured compound as the host for both fluorescent and phosphorescent emitters is very promising, and this work provides an applicable new route for developing high performance OLEDs.

Organic photovoltaic (OPV) cells employing bulk-heterojunctions (BHJ) have been intensely investigated in the past decade. One of the important hurdles for achieving high efficiencies OPV cells is the limitation imposed by the low hole mobility of the light absorbing polymer. In this study, we highlight two strategies of enhancing hole mobilities and power conversion efficiencies (PCEs) in PTB7:PC₇₁BM based OPV cells. First, we developed a family of fluorenone-based acceptors. They were used as dopants in PTB7:PC₇₁BM BHJs. Small concentrations (~0.5 % by weight) of such dopants are found to boost the PCEs of PTB7:PC₇₁BM from 7.6 % to about 8.3%. The hole transport behaviors of the undoped and doped BHJs were investigated in details by dark-injection space-charge limited current technique and admittance spectroscopy. Additional impacts arising from doping was further examined by measuring the subgap optical absorptions of the BHJs with photothermal deflection spectroscopy. The results suggest that improved hole transport occur after doping. Besides doping, we also investigate a polymer rich BHJs using PTB7:PC₇₁BM as a model system. Such a BHJ was obtained by adjusting the concentration of solvent additive in the BHJ solution. We discovered that the hole mobility such a BHJ was also improved by a factor of 2-3 while OPV employing thick layers of PTB7:PC₇₁BM exceeding a PCE of 7% can be demonstrated.

Large area electronics are currently witnessing a boost with the introduction of new materials, device architectures and methods of integration for various circuit functions. Nevertheless, energy efficiency, reliability and uniformity of performance are still important research topics.
The source-gated transistor (SGT) is a three-terminal thin-film device which relies on a potential barrier deliberately introduced at the source as its current control mechanism. This type of device can produce very high gain with low saturation voltage, and is very tolerant to large variations in geometry, making it ideal for large-area, cost-efficient analog and digital circuits.
The most convenient way of realizing the source barrier is by creating a Schottky contact, but depending on the material system and fabrication process, this approach is not always successful or controllable. In polysilicon devices, we have previously shown that the profile of the Schottky barrier can be tuned by ion implantation at the metal-semiconductor interface, leading to changes in transconductance, drain current activation energy, and saturation characteristics.
Here, we use results from solution-processed devices together with numerical simulations to investigate means of fabricating transistors which behave like SGTs without the need of rigorous control of the properties of the Schottky source contact. Potential routes to achieving this type of devices include heterostructures, contact area (as opposed to electrode) modification, doping, and interface engineering. We generically call transistors with barriers realized in this fashion bulk unipolar SGTs (BUSGTs).
BUSGTs may have superior dynamic range of the on current, larger on/off ratio and lower temperature dependence of the drain current than Schottky barrier SGTs (SBSGTs). Moreover, their architecture makes them suited to both staggered electrode configurations (top or bottom contact; bottom or top gate), which could provide some integration advantages.
SGT properties make them suited for a variety of sensing and control applications of printed, solution processed and flexible electronics: from noise-tolerant digital to high-gain low-power analog.

We demonstrate that it is possible to tune charge carrier balance in a bulk-heterojunction (BHJ) solar cell. To achieve this, we investigate the impacts of a solvent additive, 1,8-diiodooctane (DIO) on both hole and electron transports in a state of the art bulk-heterojunction (BHJ) system, namely PTB7:PC₇₁BM. For a polymer:fullerene weight ratio of 1:1.5, besides changes in the BHJ film morphology, the electron mobility in the blend film increases by two orders of magnitude from around 10^-5 to 10^-3 cm² V^-1 s^-1 with the DIO concentration while almost no change is found in the hole mobility (~10^-4 cm² V^-1 s^-1). For lower DIO concentrations, the electron mobility is suppressed because of large, but poorly connected PC₇₁BM domains. For higher concentrations of DIO, the electron mobility is improved progressively and the hole mobility becomes the limiting factor. Between 1 - 5 vol%, the electron and hole mobilities are balanced, and under these conditions, an optimized power conversion efficiencies (PCE) between 7-7.5% can be obtained. Using the Gaussian disorder model (GDM), we found that the DIO concentration modifies fundamentally the average hopping distances of electrons. At DIO concentrations much smaller than 3 vol%, the BHJ film possesses low electron mobilities corresponding to longer hopping distances of 2.4 - 3.1 nm. On the other hand, at DIO concentrations much larger than 3 vol%, the BHJ film possesses high electron mobilities corresponding to smaller hopping distances of 0.7 - 1.1 nm. Our work suggest that hole mobility is the bottleneck on the PCE, and the amount of fullerene is in excess. By increasing the DIO concentration in the processing solutions, we demonstrate that the fullerene content of the BHJ film can be significantly reduced from 1:1.5 to 1:1 while the optimized performance can still be preserved.

Charge traps in polymer gate dielectrics determine the electrical stability of organic field-effect transistors (OFETs), and polar alkoxy groups are wellknown extrinsic charge traps. However, the actual location of intrinsic charge traps in nonpolar polymer gate dielectrics has been poorly understood yet. Here, we demonstrate that the skeletal structure of polymer chain plays an important role in determining the electrical stability. To verify it, we prepared linear and branched polystyrene (l-PS and b-PS) and blended them, in which branched segments provide much larger free volume than the other segments. The current-insulating performance and field-effect mobility increased with decease of b-PS portion. In particular, the bias-stress stability was remarkably varied according to the change of b-PS portion even though all measurements excluded reactive components such as oxygen and water; the increase of b-PS resulted in time-dependent decay of mobility and threshold voltage under bias stress. This indicates that the branched segments in b-PS provide intrinsic and metastable charge trap sites. Our result suggests that the skeletal structure of polymeric chains in gate dielectric is one of the important factors affecting intrinsic long-term operational stability of OFET devices.

Organic light emitting diodes using phosphorescent dyes (PHOLEDs) have excellent performance and an internal quantum efficiency approaching 100%. To maximize performance, PHOLED devices use a conductive organic host material with a phosphorescent guest that is sufficiently dispersed to avoid concentration quenching. One of the most widely used organic compounds is green phosphorescent fac-tris (2-phenylpyridine) iridium, [Ir(ppy)₃]. In this work, we used the effect of the resonance vibration of the substrate during thermal deposition in high vacuum environment of the co-deposition of [Ir(ppy)₃] into host organic material, for reducing the clusters growth formed in the co-deposited film. It is found that the distribution of the [Ir(ppy)₃] concentration in the host material is more homogeneous in the case of the films co-deposited on vibrating substrate, as confirmed by means of scanning transmission electron microscopy (STEM) equipped with HAADF (High-Angle Annular Dark-Field) and EDS (Energy Dispersive X-Ray Spectroscopy) detectors. This analysis technique, employed for the first time in co-deposited organic thin films, permits to obtain simultaneously an image and its respective chemical information, allowing to undoubtedly characterizing their distribution and morphology.

The fabrication of OLEDs by high throughput printing techniques requires the development of solution processable electron injection layers. In this context, two classes of organic materials, aliphatic amines such as polyethylenimine (PEI) and amino-functionalized polyfluorenes such as PFN, have raised interest. However, processing of PEI poses a big challenge as films need to be very thin (<5 nm) in order to reach a high device performance [1,2]. In contrast, processing of PFN is easier but OLEDs that use it as electron injection layer exhibit limited power efficiencies due to high operational voltages [3,4].
In this work, we introduce a novel amino-functionalized polyfluorene, that consists of multiple PEI-like tertiary amine side-chains connected to the polyfluorene backbone via an amide, as an electron injection material. As a result of its molecular structure, layer thicknesses of up to 20 nm can be used in OLEDs while high power efficiencies and low operational voltages are maintained.
We solution process OLEDs that use a PPV derivative as emitting layer and either PEI, PFN or our new material in combination with silver as cathode layer. OLEDs that use our polyfluorene exhibit a current efficiency of ~ 7.5 cd/A compared to ~ 6 cd/A for PFN and ~ 7 cd/A for PEI. At the same time, due to the chemical structure of our material, operational voltages are lowered by more than 1 V compared to PFN. These results can be correlated to kelvin probe measurements that show that the new polyfluorene reduces the work-function of silver substrates by ~ 0.9 eV, exceeding the reduction observed for PFN and PEI by ~ 0.5 and 0.2 eV, respectively. AFM measurements furthermore confirm that film formation of our new material is similar to PFN and thus larger thicknesses can be used in devices.
These results show that our new polyfluorene combines the advantages of PEI and PFN, namely simple processing and a good OLED performance.
REFERENCES
[1] Zhou et al., Science 2012, 336:327-332, 2012.
[2] Stolz et al., ACS Applied Materials and Interfaces 2014, 6:6616-6622, 2014.
[3] Zeng et al., Advanced Materials 2007, 19: 810-814
[4] Zheng et al., Nature Communications2013,4:1971

Ammonia (NH₃) gas sensors based on organic field-effect transistor (OFET) using poly(methyl methacry) (PMMA) blending with zinc oxide (ZnO) nanoparticles as a gate dielectric layer were fabricated. Compared to those with the pure PMMA dielectric layer, the sensing properties of these devices using ZnO/PMMA blend as the gate dielectric layer were significantly improved when the sensors exposed to various concentrations of NH₃, and the percentage response under 75 ppm NH₃ was nearly 10 folds higher than that using pure PMMA. The results showed that there was a remarkable shift in the threshold-voltage as well as a change in the field-effect mobility after exposed to NH₃ gas. By analyzing the morphologies of the dielectrics and pentacene (which acted as the organic semiconductor) films and the electrical characteristics of OFETs, it was found that ZnO/PMMA blend gate dielectric layer was responsible for the enhanced sensing properties. As the interaction force between polar ZnO surface and NH₃ (which is polar molecule) is relatively strong, the introduction of ZnO nanoparticles to the surface of dielectric can increase the number of NH₃ molecules absorbed on the dielectric/semiconductor interface. As a consequence, more hole-trapping sites are obtained on blend dielectric surface than on pure PMMA dielectric. Besides, the decreased grain size of pentacene was formed on the ZnO/PMMA blend dielectric, facilitating NH₃ to diffuse into the conducting channel and then interact with the ZnO nanoparticles. Moreover, the environmental stability of the OFET sensors with ZnO/PMMA blend dielectric was measured, and the sensing property maintained after stored in atmosphere for 40 days. In order to confirm the functionality of the ZnO/PMMA blend dielectric in OFET based sensors, copper phthalocyanine (CuPc) was employed as the organic semiconducting layer in OFET for NO₂ detection. Compared to those with the pure PMMA dielectric, significant enhancement of NO₂ sensing property was also observed. Encouraged by the above results, a simple blending method to improve the sensing properties of OFET based gas sensors was proposed. These performance improved OFET-sensors based on ZnO/PMMA blend dielectric pave a novel way to the formation and modulation of OFETs based gas sensor as well as potential for low-cost, fast, and portable electronic nose.

This work describes the construction and characterization of self-assembled thin films of N,N&’-(2-phosphonoethyl)-3,4,9,10-perylenediimide (PPDI) on indium-tin oxide (ITO) substrates using the zirconium phosphonate technique (ZP) . Films with up to 20 layers were grown by deposition of alternating layers of zirconium cations and the imide¹ PPDI.
The films were immersed in a deaerated solution of 1,4-diazabicyclo[2.2.2]octane (DABCO) (10mM) in acetonitrile and irradiated with a high pressure mercury lamp . In situ reduction of the dye was observed, generating a light purple film (initially red) containing PPDI radical anions. This radical formation can be attributed by the electron-transfer from DABCOsup3; to PPDI. The formation of the corresponding anion radical was observed, with the absorption maxima at 719 nm, 812 nm and 969 nm, corresponding to the closest values reported by Marcon and Brochsztain² for a chemically generated PDI^#9679;- using sodium dithionite.
The stability observed for the PDI radical anions immobilized in zirconium phosphonate films can be mainly attributed to two factors, namely ring stacking, resulting in spin pairing, and the highly organized zirconium phosphonate framework.
Cyclic voltammetry (CV) data showed the typical two step redox process usually observed with PDI derivatives, giving first an anion radical (PDI^#9679;-, E_1/2 = -491 mV) followed by a dianion species (PDI^2-, E_1/2 = -900 mV). The peak currents were proportional to the number of layers in the films. The surface coverage calculated by integrating the area under the CV was 2.8 x 10¹⁴ molecules/cm² per layer.
These results suggest that regular PPDI/ZP films were formed, with the same amount of dye incorporated at each deposition cycle, making these films potential candidates for the construction of new materials with technological applications such as corrosion protection coatings, biosensors, solar cells or OLEDs.

Since its advent decades ago, the field of molecular electronics has come a long way, evolving into a vast interdisciplinary and applied area of research. Possessing an amazingly diverse range of structures and properties, single molecules are now well expected to be the functional building blocks in future computing technologies, as well as bio and ferroelectronics. The utility of molecular bridging with organic groups in an active or passive role between two solid-state contacts has been extended into the field of electrochemistry with one of the attached contacts being replaced by an electroactive organic redox group. This single contact electrochemical avenue has widened the reach of molecular electronics towards organic light emitting diodes and molecular photovoltaics, organic radical batteries, and biosensor. To simulate electron transport in such systems, we propose an electrochemical charge transfer model utilizing the non-equilibrium Green&’s function (NEGF) formulation of the Landauer quantum transport picture; with the electrochemically active redox group described by a Marcus-Gerischer density of states distribution. The formulation is implemented in the study of ultra/fast linear sweep voltammetry, applied to capture electron transfer between a metal substrate and a redox group bridged by a molecular chain. Based on this formulation, the electronic coupling independent voltammetric spectra is predicted from which the redox reorganization energy may be extracted. Moreover, two methods based on voltammetric peak potentials and the degree of reaction completion are examined as possible techniques to measure the electronic coupling between the redox group and substrate. In general, the results are expected to aid in the bridging of the molecular electronics and electrochemistry communities.

This work reports the mobility enhancement of poly(3-hexylthiophene) (P3HT) based organic thin film transistors (OTFTs) by incorporating gold nanorods (Au NRs). Through varying the doping concentration and surface modifier of the Au NRs in P3HT matrix, the P3HT/Au composite with 0.5 mg mL^-1 pyridine capped Au NRs exhibits a hole mobility of 0.059 cm² V^-1 s^-1, which is 8 times higher than that of pristine P3HT. This remarkable improvement of mobility is originated from the enhanced crystallinity and optimized orientation of P3HT after doping with Au NRs. In addition, the appropriate surface modification can produce more efficient hole conduction of Au NRs.

Carbon nanotube thin film transistors (CNT-TFTs) have been developed as promising devices toward achieving applications for printed electronics. We have already succeeded in fabricating high performance CNT-TFTs with mobilitiesy of 3-5 cm²/Vs or more. However, Organic TFTs often indicate large contact resistance (R_c), which limits the supply current and degrades the performance of the TFTs. Although we suspected that CNT-TFTs also show large R_c, there are have been few reports on this issue. Therefore, in this study, we evaluated the R_c of CNT-TFTs.
Generally, the transfer length method (TLM) is a widely accepted method to evaluate R_c. This conventional method derives the R_c without considering the transfer length (L_t), which is a function of the sheet resistance of the semiconducting layer over the contact area and the contact resistivity at the interface between the channel and the electrodes. It is well known that the R_c drastically change when the width of the electrode is comparable to L_t. Hence, the width of the electrode and the L_t must be considered when evaluating the R_c. Our approach is to fabricate several CNT-TFTs with electrodes of various widths and evaluate the contact resistance considering the relative dimensions between electrode width and L_t. Moreover, the additional electrodes was were added to the conventional TLM structure to obtain the potential distribution in the semiconducting layer over the contact area. We derived a one -dimensional model about for the relationship ofbetween current and voltage around the contact interface and compared it towith the result of the measurement results obtained by the improved TLM. Their details of them will be shown provided in this meeting.

Understanding the nature of charge carriers in molecular semiconductors, typically termed polarons, is indispensable for rational material design that targets superior (opto-) electronic device functionality. The traditionally conceived picture of the corresponding single particle energy levels invokes singly occupied molecular states within the energy gap of the semiconductor, that should be observable both by direct and inverse photoemission (IPES). However, despite intense experimental effort, clear experimental evidence for the HOMO-derived, singly occupied state of the cation is still missing employing ultraviolet photoelectron spectroscopy (UPS), even though, in reverse, LUMO-derived, singly occupied states have been clearly observed for the anion.
Here, we seek to provide such evidence by employing a sample concept, which allows both, deliberately generating C₆₀ cations and assessing their energetics by complementary photoemission techniques X-ray photoelectron spectroscopy (XPS), UPS and IPES. The sample concept is based on the fact that C₆₀ molecules, exhibiting an ionisation energy (IE=6.4 eV) lower than the work function (Phi;) of the supporting metal, show Fermi-level (E_F) pinning. There, as the substrate E_F is moved into the occupied density of states of the molecular adsorbate, electron transfer occurs from the molecules into the metal.
Concomitantly, strong electronic coupling between C₆₀-molecules and clean metal surfaces, that might mask or at least significantly alter the spectral information of cations is inhibited by inserting a thin passivating MoO₃-interlayer (Phi; = 6.8 eV) between an atomically clean Au(111) single-crystal surface and the molecular adsorbate.
This approach allows observing the superimposed spectral contributions of neutral molecules and cations, that can be clearly deconvoluted over the entire energy region. Supported by DFT calculations, the energy-level shift in neutral molecules due to inter-site Coulomb interaction with nearby cations is estimated to be 0.5 eV. Naturally, the on-site Coulomb interaction exceeds that value and is experimentally determined here to amount to 1.4 eV for the valence hole in the C₆₀-film. Importantly, on-site Coulomb interaction splits the highest occupied molecular orbital in an upper unoccupied sub-level and a lower occupied sub-level.
Consequently, we suggest replacing the widely established picture of the single-particle energy levels associated to charge carriers in molecular semiconductors, which might lead to new insights already from a re-interpretation of previous experimental results. More importantly, however, it might inspire new experiments addressing the fundamental properties of charge carriers in weakly interacting molecular systems, in particular, those induced by electrical doping of molecular semiconductors through admixing strong electron donors or acceptors.

Conjugated polymers (CPs), a kind of excellent conducting materials, have been widely applied in optical devices, such as light-emitting diodes (LEDs), organic photovoltaic cells (OPVs) and organic thin film transistors (OTFTs), due to their large-conjugated backbones which cause delocalized electronic structure,. The excellent light-harvesting and light-amplifying properties of CPs also create them a class of encouraging fluorescence probes in material science fields, such as biosensor. However, device fabrication process using CPs for water based material science fields has a difficulty, because that is not water-soluble. So, conjugated polymer nanoparticles (Pdots) have received considerable attention as potential application, such as biosensor in aqueous solution, which are fabricated by reprecipitation or micro-emulsification method.
The size-dependent tunable electronical, optical and magnetical properties of nanomaterials are essential to the current excitement and developing applications of nanomaterials. Among such nanomaterials, semiconductor quantum dots (QDs) are generally used material for active component in device technology and biomedical technology, though research in biomedical technology using QDs has declined owing to their lousy properties, such as toxicity and harmful fabrication process for QDs.
In the work, we studied about size-dependent tunable fluorescence color of Pdots from single conjugated polymer as like as QDs. We prepared our Pdots with various particle sizes through two different method, reprecipitation and micro-emulsification method. Our CPs are composed of phenylene backbone with benzoselenadiazole as electron accepting moiety. Benzoselenadiazole in CPs is included small amount with regard to phenylene parts, in order to obtain different fluorescence color in solution and solid state. So, control of particle size can induce change in amount of electron transfer from phenylene as electron donor to benzoselenadiazole as electro acceptor. For instance, small sized Pdot emitted blue light excited at 348 nm, which is influenced by only phenylene backbone, but large sized Pdot emitted yellowish green light which is influenced by benzoselenadiazole as long wavelength emitter. Finally, we demonstrated size-dependent tunable fluorescence color of Pdots not only by reprecipitation method, but also by micro-emulsification method. And this result of our work suggests that prepared Pdots with various particle sizes in this work are able to use in diverse research fields, instead of QDs. These size-controlled Pdots from single conjugated polymer can emit diverse fluorescence light excited at only single wavelength light source.

Today a variety of functions of organic molecular solids such as ferroelectric, ferromagnetism, superconductivity, switching and so on are realized. However, these functions are generally anisotropic. In order to use their functions in organic electronics effectively, the assembly control of polycrystallites in the thin film is required. Recently, we demonstrated the formation of highly ordered thin films of molecular conductors on organic semiconductor crystals by contacting the vapor of electron donor [1].
In this work, we fabricate a molecular ferroelectric thin film on a molecular semiconductor crystal. It was already known that phenazine-chloranilic acid (Phz-H₂ca) co-crystals show large spontaneous ferroelectricity at 253 K [2]. We contacted Phz vapor to crystals including H₂ca, with expecting Phz-H₂ca formation on crystal surfaces.
Single crystal of H₂ca and co-crystal of quinoxaline-chloranilic acid (QX-H₂ca) were used as substrate crystals. The crystals were grown by the vacuum sublimation method. Vaporized Phz was contacted and absorbed on the surfaces of the substrates under ambient conditions. As a result, thin films of Phz-H₂ca co-crystals were obtained on surfaces of both of crystals. The obtained thin films were confirmed as Phz-H₂ca from IR spectra. Arrangement of Phz-H₂ca polycrystals on H₂ca was found to be random. On the other hand, on QX-H₂ca, a highly ordered alignment of Phz-H₂ca co-crystals was achieved. The obtained thin film on QX-H₂ca showed ferroelectric behavior around 253 K. In addition, the anisotropy of the ferroelectric function was observed in the thin film. In the presentation, the mechanism of the crystal growth and the ferroelectric property of the thin film will be discussed.
[1] T. Mikasa, et. al., Thin Solid Films. 579, 38-43 (2015)
[2] S. Horiuchi, et. al., Nature Mater. 4, 163-166 (2005)

Recently, several research groups reported that there is an increase in the conductivity of Poly (3,4 Ethyldioxythiophene) Polystyrene Sulfonate (PEDOT:PSS) thin films if exposed to sulfuric acid. However, there is no information on the electrical behavior of acid-treated PEDOT:PSS films as a function of frequency. This work presents an impedance spectroscopy characterization of the of PEDOT:PSS thin films as a function of acid exposure time. PEDOT:PSS thin films were drop casted onto silicon and fused silica substrates. Prior to the drop casting, the substrates were cleaned with acetone/isopropanol/methanol for 5 minute and then rinsed with DI water. The substrates were also treated with oxygen and oxygen/nitrogen plasmas to achieve the best PEDOT:PSS wettability. Once the PEDOT:PSS films were drop casted, they were annealed at 120⁰C for 10 minutes to remove water, and then the films were exposed to sulfuric acid for 5, 10, 20, 40, 60, and 80 minutes. Our preliminary results show that exposure to sulfuric acid for extended periods of time, the electrical impedance decreases, and thus the electrical conductivity increases.

Fullerene derivatives are the most commonly used electron acceptors in solution-processed bulk heterojunction (BHJ) organic solar cells due to their high electron affinity, good electron transport and capability of forming favorable nanoscale phase separation with most electron donors. However, fullerene derivatives have their own physical and structural insufficiency, such as narrow absorption in the visible region and difficulty in tuning the energy levels by chemical modification. Perylenediimide derivatives (PDIs) have attracted significant attention because of their excellent thermal, chemical and photochemical stability, strong electron-accepting ability, and high electron mobility. Despite these promising advantages, PDIs based BHJ organic solar cells usually show low power conversion efficiencies (PCEs), mainly due to their strong aggregation tendency to form highly-crystalline thin films when drop cast from solution. Recent works have shown that solar cell efficiency can be significantly increased by tailoring PDIs chemical structures to control their tendency to aggregate in the solid state. Much effort has been devoted to developing twisted PDI dimmers, in which two PDI units are linked at bay positions (1,6,7,12-positions) or imide positions either directly or through various functional groups, such as thiophene, and indacenodithiophene linkers. Another efficient approach to suppressing aggregation tendency of PDIs is functionalizing PDI monomer at the imide positions, bay positions or 2,5,8,11-positions. However, compared to high PCEs of twisted PDI dimmers, PCEs of solar cells fabricated with monomeric PDIs are relatively low. The reported PCEs in the literature are typically less than 3%. In order to get suitable PDI monomer materials with low aggregation, in this work, we synthesized a PDI monomer (N,N&’-dicyclohexyl-1,6,7,12-tetraphenyl-3,4,9,10-perylenediimide, TP-PDI), in which four phenyl groups are appended to the bay positions of PDI unit. Optimized molecular structure of TP-PDI by using density functional theory (DFT) method exhibited core-twisting by 15° and the dihedral angle between the bay substituents and the PDI core is about 42°. By combining a high efficiency polymer donor, namely, poly[4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b:4,5-b&’]dithiophene-co-3-fluorothieno[3,4-b]thiophene-2-carboxylate] (PTB7-Th), TP-PDI generates a high PCE of ~4.1%, indicating that TP-PDI is a promising electron acceptor.

Conventionally, doping to organic semiconductors has been performed only in the concentration around % order. In this study, extremely low concentration doping reaching 1 ppm was performed for the first time and detailed mechanism on the doping effects in organic photovoltaic cell was clarified.
As the simplest cell, we adopted the n⁺p-homojunction having one-sided abrupt junction. Acceptor doping concentration in p-layer was changed with 0, 1, 10, 100, 1,000 ppm. Once acceptor doping is performed, holes and electrons inevitably act as majority carrier and minority carrier, respectively. From 0 to 10 ppm-doping, fill factor increased due to the appearance of majority carrier. From 10 to 100 ppm-doping, photocurrent density increased due to the built-in potential increase, namely, the formation of n⁺p-homojunction. 1.3 times enhanced photocurrent was obtained by 100 ppm accepter doping directly to the photoactive organic co-deposited layer. The present technique is generally applicable to the organic solar cell.

Tailoring the surface of dielectric layer in organic thin film transistors (OTFTs) is highly important to form a good interface with the following channel layer. Especially, since the thickness of the dielectric layer must be as low as possible to secure the low-power operation of devices, the surface treatment procedure must be free of damage to the ultrathin dielectric layer. However, the mechanical/chemical stability of organic dielectrics are generally not as great as that of inorganic ones. Moreover, the polymer dielectrics with ultrathin thickness are known to exhibit substantially depressed as compared to the corresponding bulk materials, resulting in viscoelastic behavior when they heated. All these factors make tailoring the surface of ultrathin polymer dielectrics highly challenging.
Here, we applied a simple surface treatment method onto an ultrathin (<15nm) organosilicon-based dielectric layer deposited via initiated chemical vapor deposition (iCVD) to tailor its surface to make it compatible with various kinds of organic semiconductors without degrading its insulating property. The dielectric layer used in this work is poly(1,3,5-trimetyl-1,3,5-trivinyl cyclotrisiloxane) (pV3D3), which can meet a wide range of requirements for next-generation electronic devices, such as a large energy gap (>8 eV), extremely low leakage current (less than 10^-9 A/cm² at 3 MV/cm) even with ultralow thickness (~ 6 nm), and resistance to a tensile strain of up to 4%
On the 15nm-thick pV3D3 dielectric layer, a simple oxygen plasma treatment was applied to form a molecular thin SiO_x capping layer on the pV3D3. The capping layer on ultrathin pV3D3 greatly enhanced the thermal stability of the dielectrics: the insulating property of plasma-treated pV3D3 was maintained even after the thermal exposure of 280 °C up to 3.5 h in ambient air, confirming the improved thermal and environmental stability of ultrathin pV3D3. Furthermore, the plasma-treated pV3D3 thin film exhibited high flexibility up to tensile strain of 3.3%, which is far superior to previously reported SAM/AlO_x-based dielectric layer, whose yield strain was 1.5%.
Thanks to the improved stability of pV3D3, additional silane-based self-assembled monolayer (SAM) treatment could also be utilized for the purpose of surface modification of the ultrathin pV3D3 dielectric layer. With the surface-modified pV3D3, low-voltage operating (<5 V) pentacene OTFTs with improved device performance could be achieved. TFTs on flexible substrates were also demonstrated, which indicates the approach is suitable for developing flexible/soft electronic devices. The simple, but versatile method to form oxide capping layer could significantly improve the device performance by controlling the surface functionality, as well as increasing the thermal stability of the ultrathin pV3D3 dielectric layer.

The technique of doping organic semiconductor layers has greatly increased the performance of optoelectronic devices such as solar cells and light emitting diodes. Hereby, doping is achieved by implementing either strong acceptor (p-doping) or strong donor (n-doping) molecules into the matrix of an organic semiconductor. By applying molecular doping it is possible to increase the film conductivity by several orders of magnitude and to control the Fermi level position in the organic semiconductor to facilitate carrier injection via barrier lowering or carrier tunneling.
In the last years, the number of fundamental investigations regarding the doping process has increased, and today an increasing variety of n- and p-dopants are available for these studies. In this talk, I will give an overview over the various aspects related to doping, including effects on the interface alignment, conductivity, and the passivation of trap state. The current status of research will be examined and the open question, mainly regarding the principle behind the doping process and the generally observed low doping efficiency, will be discussed.

Chemical doping is an important technique to alter the charge-transport properties or conductivity of both small molecule and polymer organic semiconductors that find widespread applications in organic electronic devices. However, research into doping organic semiconductors, in particular polymer semiconductors, is still challenging. Controlled doping of semiconductors is vital to create ohmic contacts thereby enhancing the performance of electronic devices made from such semiconductors.
In previous studies, it was found that organic semiconductor conductivity could be increased upon doping with metal atoms or gases. However, the resultant semiconductor electronic devices are unstable as the dopant diffuses under operation. Better stability is achieved through the use of molecular dopants, due to their larger size, but the attained doping effect is smaller: this is caused by strong Coulomb interactions between electrons and holes in organic semiconductors, which significantly reduce the number of free charge carriers.^[1,2] Organic semiconductors do not generally possess the high refractive indices of inorganic semiconductors—the higher the refractive index the greater the screening of the Coulomb interaction.
In this work, we investigate the enhancement of polymer conductivity using poly(3-hexyl)thiophene (P3HT) doped with the small molecule tetrafluoro-tetracyanoquinodimethane (F₄TCNQ) by adding inorganic nanoclay platelets. Previous studies have found that almost all F₄TCNQ molecules undergo integer charge transfer with P3HT sites. However, the doped polymer conductivity is still limited.^[2] To achieve a much high conductivity of a polymer semiconductor, a high refractive index material is needed. Here, we selected a synthetic nanoclay additive (LaponiteRD) as an additive, because dispersions of such nanoparticles have been shown to change the refractive indices of polymers^[3] and decreased Coulomb interactions are expected with increased refractive indices. Additionally, the edges of LaponiteRD crystals have localized positive charges, which anions may become associated with; Coulomb interactions between the electrons and holes may be decreased if the dopant anions become associated with the LaponiteRD particles, physically removing them from the vicinity of the P3HT chains. Detailed research on how the additive affects the conductivity, optical properties and film quality of the doped polymer-nanoclay blend will be presented. The method describe here could be widely applied in polymer semiconductor/small molecular dopant systems. Furthermore, organic electronic devices could benefit from this work greatly.
[1] B. Lüssem, M. Riede, K. Leo, Phys. Status Solidi2013, 210, 9.
[2] P. Pingel, D. Neher, Phys. Rev. B2013, 87, 115209.
[3] J. Böhm, J. Hauszlig;elt, P. Henzi, K. Litfin, T. Hanemann, Adv. Eng. Mater.2004, 6, 52.

We report the development of unprecedented n-doped polymers with sufficient stability to be processed from solution and in air, and which can provide shallow effective workfunctions down to 3.7 eV for useful ohmic electron injection into a number of important organic semiconductors with electron affinity larger than 3.6 eV, including the DPP family, PNDI and PCBM. We demonstrate for the first time the fabrication of inverted organic solar cells based on the coating of the electron-extraction layer from solution onto a first electrode, followed by the photoactive layer and the hole-extraction layer followed by the second electrode. With P3HT:PCBM as photoactive layer, similar power conversion efficiencies as those obtained in conventional device structures were obtained, without the use of low-workfunction metal electrodes. The discovery of a new polymer concept that can now provide polymers to be processed into viable n-doped polymer layers opens new device architectures and processing schemes that are not previously possible.

We measure dramatic increases in the composite conductivity (σ, from 0.0001 to 0.01 S/cm) and a simultaneous rise in the Seebeck coefficient (S, from ~100 to >1000 µV/K) of p-type polymers with the addition of weakly oxidizing metal acetylacetonate dopant complexes. The in situ growth of secondary phase particles within polymer matrices can be observed during film formation, and a charge transfer complex (or internal dipole) develops at the polymer-particle interface, resulting in a solid-state weakly-doped composite with altered density of states. S and σ are measured and systematically investigated for blended composites fabricated by one-pot solution-processing methods utilizing conjugated polymers (PQT12, P3HT, PBTTT-C14) and 10-90 wt% additives (Co(acac)₃, Fe(acac)₃, Mn(acac)₃). We characterized the terms in the thermoelectric power factor (S²σ) for these materials, and determined fundamental behavior regarding the enhancement of power factor in these hybrid composite systems as a function of additive concentration (at room temperature in air). We obtain peak power factors around 10-100 µW/mK² for composites with additive concentrations of around 60-80 wt%, which we can improve by more precisely guiding the growth and distribution of secondary particles.
We investigated the effects of utilizing processing solvents with different volatility (THF, chlorobenzene, and dichlorobenzene). Similarly we investigated possible effects of film thickness, or volume of drop-cast solution, by drop-casting incremental amounts of solution within 2D wells of the same surface area. Particle morphologies depend on solvent type, and also on the amount of solution drop-cast per area, due to varying processing rates and diffusion times. Kelvin Probe Microscopy and Ultraviolet Photoelectron Spectroscopy are also used to interrogate the altered energy landscape of polymer-particle composites.
The inorganic material work and composite electrical measurements were supported by the National Science Foundation, Division of Materials Research, Grant number 1005398. Polymer synthesis was supported by the Department of Energy, Office of Basic Energy Sciences, Grant Number DE-FG02-07ER46465.

Molecular n- and p-dopants can be used to increase the conductivity of, and facilitate carrier injection into, organic semiconductors. They can also be used as surface dopants to modify the work functions of inorganic electrode materials, and to tune both work function and conductivity in two-dimensional materials, such as graphene. The simplest strategy to designing n-dopants is to use molecules, D, that function as simple one-electron reductants, i.e., react with semiconductor molecules, A, to form D⁺ and A^- if the ionization energy of D is sufficiently low. However, the low dopant ionization energies required to n-dope all but the most easily reducible organic semiconductors generally render dopants highly air and water sensitive. This problem can potentially be circumvented by identifying or developing reductants in which electron transfer is coupled to a chemical reaction. Several such approaches to kinetically stabilizing dopants have been reported and will compared in this presentation. In particular, the presentation will focus on dimeric dopants, D₂, that react with A to give D⁺ and A^-, and dopants of the form DH that react to form D⁺, A^-, and hydrogenated derivatives of A. D₂ and DH dopants will be discussed for both the cases where D is either a 19-electron organometallic sandwich compound, such as rhodocene or ruthenium pentamethylcyclopentadienyl mesitylene, and where D is a substituted benzoimidazoline radical. The dopant stability, the mechanisms by which they react with acceptors and the side products formed, and estimates of their strength as dopants will be discussed.

In the early days of organic semiconductors, doping and Fermi level control were largely ingnored. The discovery of efficient doping (for a review, see e.g. [K. Walzer et al., Chem. Rev. 107, 1233 (2007); B Lüssem et al., physica status solidi (a) 210, 9 (2013)]) brought the technology also to organics. In the past years, the doping technology has achieved also a breakthrough in applications and is now contained in almost all organic devices on the market. In this talk, I will discuss some basic aspects of the doping process which are little understood. Furthermore, I will address device applications such as OLED [S. Reineke et al., Nature 459, 234 (2009)], solar cells [R. Meerheim et al., Appl. Phys. Lett. 105, 063306 (2014)] and organic transistors [B. Lüssem et al., Nature Comm. 4, 2775 (2013); X.H. Liu et al., Appl. Phys. Lett. 106, 103301 (2015)] where the doping of bulk layers and interfaces leads to novel and improved devices.

Understanding the working mechanism of electrical doping in organic semiconductors is essential for optimizing organic semiconductor devices such as organic light emitting diodes, solar cells or transistors. In contrast to molecular p-doping of organic semiconductors, n-doping has the additional problem of air sensitivity. The successful transfer of an electron to the lowest unoccupied molecular orbital (LUMO) of typical matrix materials requires n-dopants exhibiting shallow highest molecular orbitals (HOMO), rendering them prone to reactions with, e.g. oxygen or water. Hence, air stable precursor compounds, enclosing the actual n-dopant species, have been suggested to circumvent the challenging preparation and storage of the materials under inert atmosphere.
We here study the stability of n-doped organic semiconductors under air exposure by ultraviolet photoelectron spectroscopy (UPS) and electrical measurements to determine the Fermi level position in the gap and the electrical conductivity after different air exposure times. Highly efficient and air stable cationic DMBIs [1] and DMBI dimers [2] are compared to the established air sensitive n-dopants W₂(hpp)₄ and Cr₂(hpp)₄ [3]. The investigated matrix materials are C₆₀, ZnPc and Bis-HF1-NTCDI. At high doping concentrations, new filled states below the expected LUMO level of the matrix arise in the UPS spectra. Contrary to expectations, the air stability of the n-doped organic semiconductors is not exclusively determined by the stability of the pure dopant. It is connected to the energetic position of these new states at high doping concentrations. Finally, it will be discussed, whether these states are related to the LUMO level of the matrix, the HOMO level of the dopant [4] or to hybridization states of matrix and dopant [5].
[1] P. Wei et al., J. Am. Chem. Soc. 134, 3999-4002, 2012
[2] B. Naab et al., Advanced Materials 26, 4268-4272, 2014
[3] T. Menke et al., Applied Physics Letters 100, 093304, 2012
[4] M. L. Tietze et al., Phys. Status Solidi A 210, 2188-2198, 2013
[5] I. Salzmann et al., PRL 108, 035502, 2012

Using very low concentrations of dopants to eliminate the effect of gap states in organic semiconductors is a promising way to improve carrier mobility in organic thin films, adjust energy level alignments at interfaces, and achieve overall better device performance. We investigate here the prototypical hole-transport material copper phthalocyanine (CuPc) p-doped with very small concentrations of molybdenum tris[1,2-bis(trifluoromethyl)ethane-1,2-dithiolene] (Mo(tfd)₃) [1], using a combination of ultraviolet and X-ray photoemission spectroscopy (UPS, XPS), inverse photoemission spectroscopy (IPES), and variable temperature current voltage (VTIV) measurement. Density of gap states, Fermi level position, conductivity, and hole-hopping activation energy are measured for doping molar ratios (MR) ranging between 10^-4 and 2×10^-2. The undoped CuPc film exhibits a pronounced exponential tail of states extending above the edge of the highest occupied molecular orbital (HOMO). As expected, doping induces a Fermi level shift (by ~ 0.45 eV) toward the CuPc HOMO and a considerable increase in film conductivity. The latter is attributed to the filling (and de-activation) of the deepest traps and tail states upon the progressive introduction of dopants. On the other hand, the standard deviation of the Gaussian distribution of the CuPc HOMO is seen to increase with the doping concentration, denoting a significant broadening of the density of states, an effect predicted by Arkhipov et al. [2]. Whereas the decay length of the exponential tail of states decreases with increasing doping concentration, doping also seems to introduce more shallow gap states as seen from the enhanced tail of HOMO feature. The VTIV measurements show a dramatic increase in conductivity with decreasing activation energy as the doping concentration increases, confirming the conclusion from UPS data that ultra-lowing doping is filling the deep gap (trap) states. Finally, a series of theoretical simulations, considering dopant-induced electrostatic and structural disorder, were performed to support our conclusions.
[1] Y. Qi et al., J. Am. Chem. Soc. 131, 12530 (2009).
[2] V. I. Arkhipov et al., Phys. Rev. B 71, 045214 (2005).

Tandem or triple junctions are considered as the most promising avenue to unlock the 15% efficiency level for organic solar cells (OSCs).[1] Vertically stacked inverted organic tandem cells require an interconnect between the two sub-cells, that allows for an efficient recombination of photo-generated electrons from the upper cell with holes from the lower sub-cell.[2] Typically, thin evaporated metal layers (~ 1 nm of Ag or Au) are used to optimize these interconnects, which complicates fabrication. Alternatively, PEDOT:PSS is frequently employed. Unfortunately, the acidity of PEDOT:PSS limits its applicability. Moreover, reliability issues related to the use of PEDOT:PSS have been reported.
Here we present the first all-oxide interconnect for organic tandem cells, which is based on a bilayer of a high work-function (WF) metal-oxide (e.g. MoO_x or VO_x[3]) and low WF tin-oxide (SnO_x). We have recently shown that SnO_x as an electron extraction layer affords excellent inverted single junction OSCs, which do not require activation by UV light.[4] Using photo-electron spectroscopy and Kelvin probe analysis, we are able to unveil the electronic line-up at the interface of MoO_x and SnO_x. Remarkably, both oxides are n-type semiconductors with a WF of 5.2 eV and 4.1 eV, respectively. A large interface dipole (~ 0.5 eV) is found between the two oxides, which affords almost ideal alignment of their conduction bands, with a negligible offset of only 50 meV. As a result, electrons extracted by the SnO_x from the upper sub-cell are efficiently handed over from the SnO_x to the MoO_x and recombine with holes at the interface of organic/MoO_x of the lower sub-cell. This working mechanism is in contrast to the established picture where charge recombination is claimed to take place in the center of the interconnect. In tandem devices based on PCDTBT:PC70BM (high-band gap) and a PDPP3T:PC60BM (low band-gap)[5] we evidence ideal, loss-free addition of the V_OC and an efficiency (~ 8%) which is more than 60% higher than that of the individual sub-cells.
[1] A. R. M. Yusoff, D. Kim, H. P. Kim, F. K. Schneider, W. J. da Silva and J. Jang, Energy Environ. Sci. 8, 303-316 (2015)
[2] T. Ameri, N. Li and C. J. Brabec, Energy Environ. Sci. 6, 2390-2413 (2013)
[3] K.Zilberberg, S. Trost, J. Meyer, A. Kahn, A. Behrendt, D. Lützenkirchen-Hecht, R. Frahm, T. Riedl, Adv. Funct. Mater. 21, 4776 (2011).
[4] S. Trost, A. Behrendt, T. Becker, A. Polywka, P. Görrn and T. Riedl, Adv. Energy Mat. DOI: 10.1002/aenm.201500277 (in press).
[5] J. C. Bijleveld, A. P. Zoombelt, S. G. J. Mathijssen, M. M. Wienk, M. Turbiez, D. M. de Leeuw and R. A. J. Janssen, J. Am. Chem. Soc. 131, 16616-16617 (2009)

Advancements in n-type (electron transporting) organic materials have been limited by a scarcity of stable n-type doping strategies compatible with facile processing. Tethering dopants to a semiconducting core offers a promising route for designing next generation n-type organic materials. We recently demonstrated that trimethylammonium functionalization with hydroxide counterions, tethered to a perylene diimide core by alkyl spacers, facilitated solution-processing and resulted in extremely high carrier concentrations (~10²⁰ carriers/cm³) and best-in-class performance in thermoelectric thin films^[1]. In this presentation, we show that a chemical transformation in the charged end groups upon thin film drying is critical to the underlying mechanism that enables charge carrier generation in these self-doping materials in the solid-state. We draw these conclusions by complementing thermoelectric characterization of these variants with insight on the electronic and structural property changes from optical spectroscopy, EPR, XPS, and GIWAXS experiments. Furthermore, we demonstrate that the self-doping is broadly tunable via counteranion-mediated control of dopant activation and that the use of tethered functionality is highly generalizable to other n-type molecular cores of interest, including naphthalene diimides, diketopyrrolopyrroles, and PCBM. Our findings help shape future strategies for enhancing n-type organic electronic material performance, especially for applications demanding high carrier concentrations, such as thermoelectrics. Coupling such advances in n-type material system design with complementary p-type organic materials helps bring the development of high-performing, fully solution-processable, organic electronic devices closer to reality.
[1] Russ et al. Adv. Materials (2014), DOI: 10.1002/adma.201306116

The electronic structures of the hybrid interfaces between conducting oxides and electron-accepting or electron-donating organic semiconductors are detailed via a combination of DFT calculations and UPS/XPS spectroscopies [1-3]. We have considered representative systems such as ZnO interacting with PTCDI, PTCDA, C60, or sexithienyl. The interfacial electronic interactions are found to depend critically on the nature of surface defects / dopants. In particular, the large interface dipole coming from substantial charge transfer from ZnO to PTCDI or PTCDA can be properly described only when accounting for the surface defects that confer ZnO its n#8209;type properties.
The presentation will thus focus on a discussion of the significant influence that defect sites and dopants have on the specifics of interfacial interactions and, as a consequence, on carrier injection or extraction.
[1] P. Winget et al., Advanced Materials 26, 4711 (2014).
[2] M. Gruenewald et al., Journal of Physical Chemistry C 119, 4865 (2015).
[3] H. Li and J.L. Bredas, submitted for publication (2015).

The field of hybrid inorganic-organic semiconductor systems (HIOS) that potentially allow combining the advantages of both material classes steadily grows. Important processes in HIOS-based devices, like energy or charge transfer, are controlled by the energy level alignment (ELA) at the HIOS interface, which therefore critically influences device efficiency. However, an unfavorable interfacial ELA is often encountered for a given HIOS. This necessitates modifying the ELA, which, e.g., can be achieved with the help of ultrathin interlayers of organic donor / acceptor molecules, as was recently shown for ZnO applied in a hybrid light emitting diode (LED) [1]. The adsorption of organic acceptor molecules was found to increase the ZnO work function due to two contributions, i.e., an interface dipole at the organic-inorganic interface and band bending modification inside the ZnO [2]. A drawback of ZnO for many envisioned devices, however, is that it is naturally n-doped and hard to p-dope. In contrast, GaN is another promising wide band gap inorganic semiconductor with both types of doping available. It is applied in LEDs, high electron mobility transistors (HEMTs), and other electronic devices.
To investigate how the above findings for ZnO transfer to GaN and to other doping regimes, the two molecular organic acceptors 1,4,5,8,9,12-hexaazatriphenylenehexacarbonitrile (HATCN) and 2,2'-(perfluoronaphthalene-2,6-diylidene)dimalononitrile (F6-TCNNQ) were vacuum-deposited on non-intentionally doped GaN (0001) and, for comparison, also on ZnO (0001). By means of ultraviolet photoelectron spectroscopy, a huge work function increase (up to 1.5 eV and 2.1 eV for HATCN and 1.7 eV and 2.8 eV for F6-TCNNQ on GaN and on ZnO, respectively) was observed for monolayer coverage, which stems mostly from an interface dipole between substrate and molecules. The contribution of band bending within GaN (ca. 0.2 eV, as determined from X-ray photoelectron spectroscopy) was found to be significantly smaller than in ZnO (0.9 eV), contrary to what is expected from theoretical calculations that predict the contribution of band bending to be more pronounced for a lower doping concentration [3].
It should be noted, that the way of preparing the GaN surface does not affect the obtained results. The ELA for ordered GaN surfaces (as evidenced by a well-developed LEED pattern) is the same as when depositing on sputtered GaN surfaces, where order is lost. Moreover, the absence of chemical shifts in the substrate core levels indicates physisorption of the molecules rather than chemisorption.
Our investigations show that the energy level tuning scheme via organic interlayers, already successfully employed for ZnO, holds great promise for GaN as well.
[1] Schlesinger et al., Nat. Commun. 6, 6754 (2015)
[2] Schlesinger et al., Phys. Rev. B 87, 155311 (2013)
[3] Xu et al., Phys. Rev. Lett. 111, 226802 (2013)

The emergence of two-dimensional (2D) layered materials has gained intense attention in the recent past due to their wide application in diverse areas such as transistors, sensors, piezoelectric devices, thermoelectrics, and solar cells. Controlled doping of these 2D materials such as graphene and transition-metal dichalcogenides (TMDC) can provide a powerful tool for modifying their electrical and optical properties, and for controlling device performance. This presentation will focus on the doping studies of these 2D materials. By applying the dopants, the work function of CVD graphene can be tuned from ca. 3 to 5 eV, and the sheet resistance of monolayer graphene can be reduced by more than 90%. Organic field-effect transistors and solar cells devices with doped graphene electrodes were fabricated, the performance of which is comparable to, or even better than, that of similar devices with metal or metal- oxide electrodes. Work- function engineering of graphene electrode via doping was proved to be important in reducing the carrier injection barriers. In this presentation, doping studies of two TMDCs, molybdenum disulfide (MoS₂) and tungsten diselenide (WSe₂), will also be discussed; these materials were characterized by electrical measurements, ultraviolet- and X-ray-photoelectron spectroscopy (UPS and XPS), and Raman spectroscopy. The doping effects can be controlled through the choice of dopant, treatment time, and the concentration of dopant in solution.

In general, molecular compounds have closed-shell electron strucutures. Therefore, in order to generate a functionality in molecular solids, it is neccesary that the closed-shell turns into opened-shell strucuture. Typically, highly conductive functions in molecular materials are obtained by formation of charge-transfer complexes between electron donor and acceptor molecules. Tetrathiafulvalene (TTF) and tetracyanoquinodimethane (TCNQ) are well-known electron donor and acceptor molecules, respectively, and TTF-TCNQ complex composed of them is also a well-known molecular conductor. However, it was reported that good electric conduction was observed at the interface where TTF and TCNQ crystals were conjugated [1]. We have confirmed reproduction of the results and attempted to clarify the mechanism of carrier-doping into the interface through various measurements. As a result, we found that the origin of metallic feature is simple charge-transfer at the interface and generation of nano-size TTF-TCNQ crystals on the surface [2]. Then, we attempted to demonstrate highly electric conduction by simple carrier injection at the interface by investigating the transport properties of the contact interfaces between various donors and acceptors.
As a result, the contact interface between (phthalocyaninato)nickel(II) (Ni(Pc)) and 2,5-difluoro-7,7,8,8-tetracyanoquinodimethane (F₂TCNQ) has been found to show metal-like transport properties. Although Ni(Pc) and F₂TCNQ are an electron donor and an acceptor, respectively, the combination of these two components does not yield any charge transfer (CT) complex crystals. Infrared spectra show that the highly conductive feature originates from charge injection at the contact interface. The thermoelectric power of the mixed powder reveals that the transport at the contact interface is dominated by the holes in the Ni(Pc) crystal [3]. The details of the transport and some other physical properties will be presented.
[1] H. Alves, et al., Nature Mater. , 7, 574-580 (2008).
[2] Y. Takahashi, et al., J. Phys. Chem. C 116, 700-703 (2012).
[3] Y. Takahashi, et al., Chem. Mater. 26, 993-998 (2014).

As a p-type dopant, tetrafluoro-tetracyanoquinodimethane (F₄TCNQ) is believed to form charge-transfer complexes (CTC) with electron-donor molecules such as thiophenyl or carbazole derivatives. In our earlier work,¹ the impact of F₄TCNQ on the indium-tin-oxide (ITO) surface modified by a self-assembled monolayer (SAM) of t-butyl-carbazole-substituted phosphonic acids (t-BCPA) was investigated theoretically at the density functional theory (DFT) level; it was assumed that F₄TCNQ can either form a CTC with the carbazole fragment of t-BCPA, or adsorb directly onto the ITO surface on the side of b-BCPA. Here, we discuss a joint experimental and theoretical study where we combine DFT calculations on multiple t-BCPA/F₄TCNQ/ITO configurations, with experimental characterizations involving scanning force microscopy (SFM), X-ray and ultraviolet photoelectron spectroscopy (XPS and UPS), and optical absorption (UV/Vis) spectroscopy. Our results offer an improved understanding of the charge transfer among the ITO surface, the t-BCPA SAM, and the F₄TCNQ dopant. They are consistent with the recent model proposed by Oehzelt et al.,² that charge transfer can occur directly through an insulating layer without the formation of a CTC and can be attributed to charge equilibration across interfaces.
References:
1. Li, H.; Winget, P.; Bredas, J.-L., Surface Modification of Indium-Tin-Oxide via Self-Assembly of a Donor-Acceptor Complex: A Density Functional Theory Study. Adv. Mater.2012, 24, 687.
2. Oehzelt, M.; Koch, N.; Heimel, G., Organic Semiconductor Density of States Controls the Energy Level Alignment at Electrode Interfaces. Nature Commun., 2014, 5, 4174.

Interfacial barriers at electrode-semiconductor contacts limit charge collection efficiency and hamper performance of organic electronic devices. Doping of the semiconductor near the interface can mitigate charge extraction or injection problems by allowing charge tunneling through barriers with reduced width. We have demonstrated that polymer acids can act as p-type dopants for polymer donors, such as poly(3-hexylthiophene-2,5-diyl) (P3HT) and poly[[9-(1-octylnonyl)-9H-carbazole-2,7-diyl]-2,5-thiophenediyl-2,1,3-benzothiadiazole-4,7-diyl-2,5-thiophenediyl] (PCDTBT). The performance of contact-doped organic photovoltaics nearly matches the performance of devices comprised of traditional hole transport layers such as PEDOT:PSS. By varying the pendant acidic groups of the polyelectrolyte between aromatic sulfonic acid, trifluoromethane sulfonimide, and perfluorosulfonic acid, we find the effectiveness of doping the conjugated polymer at the interface depends on the strength of the pendant acid group with stronger acid moieties being capable of creating more carriers in the doped system. Deposition of acidic polymeric dopants at the anode allows high carrier densities, of order 10²⁰ cm^-3, to be obtained in polymer semiconductors near the electrode interface. The charge carrier density also depends on the miscibility between polymeric dopants and conjugated polymers. The overall doping efficacy near electrodes therefore depends on the interplay between the strength of pendant acid groups and miscibility between polymeric dopants and conjugated polymers.

It is usually believed that high work function hole transport layers (HTLs) improve the performance of organic photovoltaic (OPV) devices by increasing the built-in field. Here we show the existence of an optimal work function for a given donor polymer; HTLs with work function higher than the optimal value produce devices with lower short-circuit current (J_sc) while open-circuit voltage (V_oc) and fill factor (FF) are unaffected. By controlling the atmosphere and solvent exposure of freshly thermally evaporated MoO_x films, we are able to vary its work function from 4.9 eV to 5.8 eV. OPV devices fabricated on these MoO_x HTLs, including P3HT:PCBM, PTB7:PC₇₁BM and PCDTBT:PC₇₁BM, show a decreased J_sc as the MoO_x work function increases beyond an optimal value, which is donor materials specific and is originated from the Fermi level pining of contact layer to the integer charge transfer state of donor material. Capacitance-voltage measurements indicate that the dopant density in active layer increases as the MoO_x work function increases, consistent with the increased injection current in hole-only device, suggesting a contact-induced doping mechanism. The decrease in J_sc with increasing p-type dopant density is found to be linear and attributed to: (1) a reduced carrier generation at the p-doped region due to increased polaron-exciton quenching, as shown by shorter photoluminescence lifetime, and (2) an increased recombination rate both at the anode contact and in the flat band region of active layer, as confirmed by drift-diffusion modeling using SCAPS software and external quantum efficiency measurement. This study shows that this contact-induced unintentional doping in active layer is deleterious for OPV devices. Thus, there exists an optimal work function for MoO_x HTL, at which it dopes the active layer minimally but still maintains a sufficiently high built-in field. The relation between such an optimal work function and the donor material energy levels will be reported. This is the first study that clearly demonstrates higher work-function HTLs can be detrimental to device performance.
This project is sponsored by National Science Foundation DMR-1305893.

The ability to tune the properties of the dielectric material is at special interest for organic thin-film transistor (OTFT) performance. ^[1] Therefor it is essential to control the physical and chemical composition of the dielectric material, especially at the dielectric semiconductor interface. ^[2,3]
Devices using a hybrid metal oxide/ethylene glycol based self-assembled monolayer (SAM) dielectric offer the possibility to combine insulation with the ability to dope the layer. The incorporation of specific ions into such SAMs was shown by Cho et al. ^[4] This offers us new possibilities to dope the dielectric layer in a device with non-covalently bond species, which directly influence device performance by changing the dipole moment of the SAM layer.
Here we present an ethylene glycol based phosphonic acid SAM, (2-(2-(2-methoxyethoxy) ethoxy)ethyl) phosphonic acid (CH₃-(OC₂H₄)₃-PA), acting as cage for alkali metal cations in a hybrid aluminum oxide (AlO_x) stack dielectric. The ions were added by a simple immersion step of the previously deposited SAM. Applied as dielectric in a 2#8209;tridecyl-benzothieno-benzothiophene (C₁₃#8209;BTBT) based organic thin-film transistor the presence of the ions lead to improvements in mobility and a shift of the threshold voltage could be detected.
Furthermore molecular dynamics simulation shows the structure of the SAM, the positioning of these ions in the layer and the mechanism of this doping by the incorporation of an aligned dipole into the layer.
This doping of SAMs provides a large potential in low voltage electronics, by influencing the device performance without changing the initial device setup. This makes this system suitable for a large variety of devices.
[1] Halik, M.; Hirsch, A. Adv. Mater.2011, 23, 2689-2695.
[2] Novak, M.; Schmaltz, T.; Faber, H.; Halik, M. Appl. Phys. Lett.2011, 98, 093302.
[3] Lenz, T.; Schmaltz, T.; Novak, M.; Halik, M. Langmuir2012, 28, 13900-13904.
[4] Cho, E. S.; Kim, J.; Tejerina, B.; Hermans, T. M.; Jiang, H.; Nakanishi, H.; Yu, M.; Patashinski, A. Z.; Glotzer, S. C.; Stellacci, F.; Grzybowski, B. a. Nat. Mater.2012, 11, 978-985.

Bilayer organic solar cells rely on very thin coated films that allow for effective light absorption and charge carrier transport away from the heterojunction at the same time. However, thin films are difficult to coat on rough substrates or over large areas, resulting in adverse shorting and low device fabrication yield. Chemical p-type doping of organic semiconductors can reduce Ohmic losses in transport layers through increased conductivity, potentially allowing the use of thicker active films. Recently, the performance of cyanine dye / C₆₀ bilayer cells could be improved by using oxygen or NOBF₄ as chemical dopants. However, drawbacks because of an inherently unstable dopant, side reactions with the solvent and/or detrimental film formation properties made both dopants inappropriate for further investigations. In this study a stable Co(III)-complex was used to p-type dope a heptamethine cyanine. The effect of dopant addition was studied using conductivity measurements and cell performance trends for various active layer thicknesses. For dye films thicker than 50 nm, doping increased the power conversion efficiency by more than 30%, primarily because of a pronounced increase in the fill factor. The stability of the oxidized radical dication was studied in film and solution. Doped species are not only formed by chemical doping, but also after light absorption and electron transfer directly at the heterojunction in OSC, or over a thicker layer during operation of light-emitting electrochemical cells (LECs). We found that the stability of cyanine dye-LECs is closely related to the degradation mechanisms observed in chemically doped films.
References:
1. S. Jenatsch, T. Geiger, J. Heier, C. Kirsch, F. Nüesch, A. Paracchino, D. Rentsch, B. Ruhstaller, A. C. Véron, R. Hany_,Sci. Technol. Adv. Mater. 16 (2015) 035003
2. A. Pertegas, D. Tordera, J. J. Serrano-Perez, E. Orti, H. J. Bolink, J. Am. Chem. Soc.2013, 135, 18008-18011

It is widely accepted that injection of carriers from the metal contacts into the semiconductor is crucially determining the performance of organic devices, particularly that of organic light-emitting diodes and transistors. This fuels the on-going quest to identify the relevant microscopic mechanisms underlying injection into typically disordered semiconductors and to cast these insights into a comprehensive model. Currently, analytical models are not universally valid due the vast range of values that crucial parameters such as injection barriers, carrier mobilities, molar doping ratios, and energetic disorders can adopt. To bypass the short-comings state-of-the-art macroscopic expressions, we directly study the injected current across the interface between a metal and an undoped semiconductor as a function of the energetic disorder and the superimposed electric field by using Kinetic Monte Carlo. This model inherently considers the hopping motion of charges in presence of energetic disorder and all carrier-carrier interactions. However, the need to handle charge concentrations as high as 10²⁶m^-3 (be that due to efficient injection in the presence of large fields or doping) and large volume sizes poses a serious challenge to simulations, as then the exact dynamic Monte Carlo method (DMC) becomes inherently slow and exceeds reasonable computational resources. Thus, we developed a fast update method being as accurate as DMC while being only slightly more expensive than the First Reaction Method (FRM) commonly used to study charge migration in organic donor acceptor blends.
We demonstrate that, independent of a given barrier, reasonable field strength, and degree of energetic disorder, there is an accumulation of charges next to the metal surface due to the attraction to their image charges in the metal. Remarkably, the such accumulated charges are found to play a decisive role for ultimately injecting further carriers into the bulk: For small injection barriers, the resulting space charge causes a band bending such that resonant, quasi-ohmic injection is enabled. For large barriers, it is essentially the Coulomb interaction of a carrier with close-by charges that promotes the carrier from the interface into the bulk rather than returning instantaneously back into the contact.

The intermolecular interactions between an electron donor (conjugated polymer with alternating electron-rich and electron-poor moieties) and an electron acceptor (fullerene or fullerene derivative) in bulk heterojunctions are known to critically influence the efficiency of an organic solar cell. A better understanding of the nature of these interactions and of the electronic structure at the donor:acceptor interface can help in designing more efficient solar cells. Recently, Graham et al. have shown that the intermolecular packing between fullerene and polymer can be controlled by the side chains appended to the conjugated polymer backbone [1]. In general, higher power conversion efficiency is obtained in devices where the fullerenes preferentially dock on the electron-poor moieties. Here, using density functional theory (DFT) and symmetry-adapted perturbation theory (SAPT) methodologies, we examine the nature of the intermolecular interactions between fullerenes (C₆₀ and PC₆₀BM) and poly-benzo[1,2-b:4,5-bprime;]dithiophene-thieno[3,4-c]pyrrole-4,6-dione (PBDTTPD). Our calculations allow us to rationalize the differences in binding when the acceptor docks on the electron-rich vs. electron-poor moieties of the polymer and their implications in terms of interfacial electronic structure.
[1] Graham, K. R.; Cabanetos, C., Jahnke, J. P.; Idso, M. N.; Labban, A. E.; Ndjawa, G. O. N.; Heumueller, T.; Vandewal, K.; Salleo, A.; Chmelka, B. F.; Amassian, A.; Beaujuge, P.; McGehee, M. D. J. Am. Chem. Soc.,2014, 136, 9608.

The promise of low-cost processing and flexible circuitry has driven intense research in carbon-based electronics. The performance of carbon-based materials can be effectively tuned by the application of redox reagents as dopants to increase conductivity and decrease injection barriers. For n-type dopants, the ability to reduce a variety of materials, formation of stable relatively immobile ions, and air stability are desirable properties. We have designed and synthesized a series of new benzimidazole-based dimeric dopants, which are moderately air-stable in the solid state, that react with a variety of electron-transport materials, including fullerenes, perylene diimides, and bis(triisopropylsilyl)pentacene to form the corresponding radical anions and monomeric benzimidazolium cations. The doping behavior has been extensively characterized in both solution and films, through UV-vis, ultraviolet photoelectron spectroscopy, photothermal deflection spectroscopy, and electron paramagnetic resonance. Different doping mechanisms were founded for different dimeric dopants. These dimers prove to be effective dopants for a variety of vapor- and solution-processed materials, and react much faster than the corresponding monomeric hydrobenzoimidazole reductants. The conductivities of doped films are several orders of magnitude higher than those of the undoped materials. Notably, the conductivity of doped C₆₀ reached a record average value for molecular doped C₆₀ films of 12.0 S/cm. All of these suggest that these new benzimidazole dimeric dopants could be useful for carbon-based electronic applications.

The dark current density in donor/acceptor organic planar heterostructure devices at a given forward voltage bias can either increase or decrease when an insulating spacer layer is added between the donor and acceptor layers. The dominant current flow process in these systems involves the formation and subsequent recombination of interfacial exciplex states. If the exciplex recombination rate limits current flow, an insulating interface layer decreases the dark current. However, if the exciplex formation rate limits the current, an insulating interface layer may increase the dark current. We present a device model to describe this behavior, and we discuss relevant experimental data.
Collaborators: Sun Yin (Shandong University), Wanyi Nie (LANL), Aditya D. Mohite (LANL), Darryl L. Smith (LANL), and P. Paul Ruden (University of Minnesota)

Organic photovoltaics offer a flexible, low-cost solution to the energy problem, but they are held back by low efficiencies. One of the reasons the efficiency is limited is that the devices do not reach their theoretical maximum open-circuit voltage. Modeling predicts that the loss in V_OC is caused by energetic disorder, material mixing, and short charge-transfer (CT) lifetimes [1], all of which depend on the bulk and interface morphology. At the same time, short-circuit current and fill factor depend on recombination in the device, which has been shown to primarily occur at the interfaces in planar devices [2]. A thorough understanding of the morphology and interfacial structure is necessary to link the processing conditions to energetic disorder, mixing, CT state lifetimes, and recombination kinetics, enabling a systematic improvement of device performance.
This work focuses on a bilayer model system consisting of an evaporated fullerene layer (C60) on top of a spin-cast small molecule (H1). Bilayers have a well-defined interface, allowing an in-depth study of the interaction between the donor and acceptor materials. Previous work has shown that there may be interdiffusion at the interfaces of bilayers [3] so a careful characterization of the interface is necessary. GIWAXS, molecular modeling, NEXAFS, AFM, and SEM are used to study the film morphology under a variety of processing conditions. These measurements reveal that H1 assumes a complex interwoven structure that flips in orientation with the addition of a solvent additive. It is also shown that C60 forms a sharp interface with H1 with little mixing between the two materials independent of H1 orientation. The controllable interface makes this bilayer an ideal system for studying the link between structural and opto-electronic properties. Understanding this link will be crucial toward improving device V_OC, J_SC, and fill factor in a systematic way.
[1] T. M. Burke, S. Sweetnam, K. Vandewal, and M. D. McGehee, Adv. Energy Mater. 2015, 5, 11.
[2] A. Foertig, A. Wagenpfahl, T. Gerbich, D. Cheyns, V. Dyakonov, and C. Deibel, Adv. Energy Mater. 2012, 2, 12.
[3] N. D. Treat, M. A. Brady, G. Smith, M. F. Toney, E. J. Kramer, C. J. Hawker, M. L Chabinyc, Adv. Energy Mater. 2011, 1, 82

Printed electronics is a rapidly developing field of research which covers any electronic devices or circuits that can be processed using direct printing techniques. Among those printing techniques, inkjet printing is a technique of increasing interest for organic field-effect transistors (FETs) due to its fully data driven and direct patterning#12290; In this work, the effect of dopant concentration and substrate temperature on the micro- and nano-structure and function of p-type polymer PCDTPT field-effect transistors has been investigated. The crystallinity and packing of conjugated polymer has been examined by synchrotron x-ray diffraction. The detailed information about dopant/polymer interface and domains were investigated by small angle neutron scattering. It is found that the polymer ordering, domains, and dopant/polymer interface were significant influenced by the dopant concentration and substrate temperature. Finally, the relationship among dopant concentration, morphology, and function has been established for polymer FET from inkjet printing.

Chemical doping can be utilized to improve the conductivity of many organic semi-conductor materials. Solution-based chemical doping is favored in the device fabrication process due to its low-cost, low energy demand, and increased compatibility with large-scale printing techniques. P-doping of poly(3-hexyltiophene), a conjugated polymer, with the strong oxidizing agent7,7,8,8-Tetracyano-2,3,5,6-tetrafluoroquinodimethane, (F4TCNQ) in most organic solvents is a well-known example of chemical doping. This work presents the doping kinetics studies of P3HT with F4TCNQ via in situ UV-Vis spectroscopy. Our results indicate that temperature will impact the progress of chemical doping of P3HT in the solution phase. It was found that an optimal temperature range will result in increased doping rates for P3HT/F4TCNQ, likely due to the competition between thermodynamic and kinetic factors. The reversibility of this doping reaction is also studied. Polythiophenes comprised of different alkyl chain lengths will also be investigated to study their influence on the charge transfer reaction between the host/dopant molecules.

Organic thin film transistors still suffer from an insufficient stability and cannot compete with the
performance of their inorganic counterparts. Nevertheless, highly promising advances in switching
speeds were achieved upon aggressive scaling of device dimensions such as the channel length
and the gate dielectric thickness [1]. The benefits of reducing the channel length, i.e., the
separation of source and drain are, however, counteracted by non-desired short channel effects
and increasingly dominant contact resistances. Introducing local doping near the metal electrode -
semiconductor interface has been demonstrated to efficiently suppress the latter.[1] Theoretical
work indicates that, without the benefit of doping, the contact resistance is strongly related to the
efficiency of injection at the metal-semiconductor interfaces and depends not only on the carrier
mobilities and injection barriers, but also on the device dimensions, the orientation of the injecting
surface with respect to the semiconductor-insulator interface, and the point of operation. [2]
To provide a solid basis for a deeper understanding of the doping-induced effects at the contact
interface, we investigate the interplay between short channel effects and injection for undoped
devices. We utilize two-dimensional drift-diffusion-based simulations including the self-consistent
consideration of thermionic and tunneling injection, interface recombination, and back drift, to
determine the contact resistance and short channel effects directly from the simulation of the
device at a given point of operation. The considered channel lengths vary by three orders of
magnitudes, i.e., range between several micrometers to 300 nanometers and below. We
particularly focus on how the onset and the extent of short channel effects for given material
properties depend on the injection barrier and the actual device architecture, i.e., the staggered
(top-contact bottom gate) or the coplanar (bottom-contact bottom-gate) device configuration.
[1] F. Ante et al. Small 7 (2011) 1186-1191, DOI: 10.1002/smll.201002254
[2] M. Gruber et al., Adv. Funct. Mater. 23 (2013) 2941-2952, DOI: 10.1002/adfm.201203250.

Polymeric semiconductor materials are the promising candidates for flexible electronic devices with low-cost and easy processing. However, In spite of the significant improvement of organic semiconducting materials, almost of polymeric semiconductors show relatively low mobility and high hysteresis than silicon or metal oxide based devices. Especially, n-type organic semiconductors have been demonstrated still inferior performance compared with p-type materials. Various studies of n-type OFET devices have been conducted to improve their performances. In this study, we fabricated the n-type organic field effect transistors (OFET) with TiO₂ doping layer as dielectric, in order to improve the electrical characteristics such as threshold voltage and the charge injection properties. In consideration of the channel stability, structure with top gate bottom contact was selected. Au source/drain electrodes were thermally deposited on si-wafer substrate and poly{[N,N9-bis(2-octyldodecyl)-naphtha-lene-1,4,5,8-bis(dicarboximide)-2,6-diyl]-alt-5,59-(2,29-bithiophene)};[(P(NDI2OD-T2))](polyera, activink n2200) n-type polymer semiconductor was spun coated or ink-jet printed as a semiconductor layer. TiO₂ nanoparticles (average diameter-21nm, sigma-aldrich) was used as a doping agent and dissolved in n-butylacetate with triton-X100 surfactant. Poly-methylmethacrylate (PMMA, sigma-aldrich, Mw ~120,000) dielectric layer was also spun coated, and thermally evaporated Au gate electrode and inkjet-printed Ag electrodes were compared. In case of device with TiO₂-doped dielectric layer, the threshold voltage was shifted significantly to positive bias (+10V) from negative bias (-20V) of device with PMMA-only dielectric layer. The mobility of device with TiO₂ doped dielectirc layer was shown to be 2x10^-3cm²/Vs, which is increased by three times compared to with PMMA-only dielectirc layer device (0.7x10^-3cm²/Vs). In addition, device with the printed Ag gate electrode showed nearly identical performance with a device having the thermally deposited electrode. Improvement of device properties of n-type OFETs using modified dielectric layers were beneficial for a design of NMOS inverter device (load TFT with doped dielectric) and PMMA-only dielectric devices (drive TFT).

The family of hydrogen-bonded organic semiconductors has attracted quite some attention in organic electronic applications during recent years, due to their extraordinary stability. In this work we went beyond present device characterizations and focus on in situ spectroscopic measurements. We study the redox behavior of quinacridone, a hydrogen-bonded organic semiconductor, and pentacene, which is analogous in structure (five fused rings) and size. Cyclic voltammetric studies of quinacridone show two well-defined oxidation peaks which are correlated with in situ transmittance UV-VIS spectroelectrochemistry. While in the first oxidation step hardly any changes occur, during the second oxidation step, several new pronounced absorption bands appear and the HOMO-LUMO transition decreases significantly. These results are combined with in situ attenuated total reflection (ATR) FTIR spectroelectrochemistry. Here, a broad absorption band arises during the first oxidation peak that decreases as oxidation proceeds. Moreover, electron paramagnetic resonance (EPR) measurements are performed, showing the formation of different radical cations during these oxidation steps. Analogous results are obtained by in situ chemical doping with iodine as an oxidation agent. Interestingly, chemical oxidation reaches only the first oxidation peak. As compared to quinacridone, pentacene behaves spectroscopically differently. It shows only one oxidation peak in its cyclic voltammogram. Several electronic transitions appear in UV-VIS and a broad absorption band arises in the mid-IR range, whose maximum shifts by applying higher oxidation potentials. Further, significant changes in vibrational absorption bands are observed after reaching a high oxidation level. In addition, the EPR-signal increases at low oxidation levels, indicating the formation of radicals; it decreases at high oxidation levels. Based on these data we present a model which considers structural changes on a molecular level for both molecules. The various radical cation formation in quinacridone and pentacene, respectively, are suggested due to heteroatomic structure of quinacridone, and, hence, the hydrogen-bonding in the solid state. Furthermore, we analyze the reduction processes of both molecules using the same spectroscopic techniques.

Thermoelectic materials have been of interest in scavenging waste heat; however many efficient inorganic materials are rare, brittle, and contain toxic elements. Organic thermoelectrics offer a potential alternative because they are abundant, mechanically flexible, solution processable and operative at low temperatures (<200°C). Thermoelectric devices require complementary n- and p-type legs, which ideally will have similar thermoelectric properties. A route to achieve this goal is to use an ambipolar polymer, which can be used in both legs. Here we investigate the electronic doping mechanism of an ambipolar polymer based on a co-polymer of Pol{((E)-3-(5-([8,8'-biindeno[2,1-b]thiophenylidene]-2-yl)thiophen-2-yl)-2,5-bis(2-octyldodecyl)-6-(thiophen-2-yl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione} (P(BTP-DPP)) doped with organometallic small molecule dopants that lead to p- or n-type conduction. The molecular design of P(BTP-DPP) allows for efficient packing of dopants, leading to charge transfer and relatively good electrical conductivity. Transmission electron microscopy provided the dopant distribution and morphology within the doped blends. Electron paramagnetic resonance was used to determine the efficiency of the dopant in generation of charge carriers and to provide insight into the charge transport mechanism. Temperature dependent Seebeck and conductivity measurements were used to determine the peak operating temperature of these materials. Maximum n-type conductivities of ~10^-1 S/cm were achieved; these values are amongst the best known for n-type semiconducting polymers.

We successfully fabricate high performance bottom-contact organic field-effect transistors (OFETs) using AgNW networks electrodes by spray deposition. The synthesized AgNWs have the dimension of 40-80 nm in diameter and 30-80 µm in length and are randomly distributed and interconnected to form a 3D hollow framework. The AgNWs networks, deposited by spray coating, yield an average optical transmittance of up to 88 % and a sheet resistance as low as 10 ohm/sq. For using AgNWs as source/drain electrodes of OFETs with bottom-contact configuration, the large contact resistance at AgNWs/organic channel remains critical issue for charge injection. To enhance charge injection, we fabricate semiconductor crystals on the AgNW using adsorbed residual poly(N-vinylpyrrolidone) layer. The resulting bottom-contact OFETs exhibit high mobility up to 1.02 cm²/Vs, and are similar to that of the top-contact Au electrodes OFETs with low contact resistance. A morphological study shows that the pentacene crystals coalesced to form continuous morphology on the nanowires, and are highly interconnected with those on the channel. These features contribute to efficient charge injection and encourage the improvement of the bottom-contact device performance. Furthermore, large contact area of individual AgNWs spreading out to the channel at the edge of the electrode also improves device performance.

Recent researches have demonstrated MoO3 to be an effective hole transporting layer (HTL) in various OLED devices and OPV devices. MoO3 is not only known for multifunctions of hole transporting, electron blocking, and exciton blocking, but also a successful replacement for PEDOT:PSS, resulting fill factor increment. Meanwhile, the conduction band edge of MoO3 with a band gap of ~3.1 eV varies from 2.3 to 6.7 eV depending on its thickness and process condition, which tunes the band bending of adjacent materials due to the Fermi energy difference.
In this work, we studied the role of MoO3 as HTL in fullerene-based Schottky junction solar cells and effects on the band bending of adjacent layers. The photovoltaic devices were fabricated with a structure composed of ITO/HTL/C70/BCP/Al. Here MoO3 or MoO3/TPTPA (tris[4-(5-phenylthiophen-2-yl)phenyl]amine) layer was used as the HTL to observe the difference in their role in C70-based Schottky solar cells. The current density-voltage measurements show the increased open circuit voltage (Voc) and the maintained short-circuit current density (JSC) of the device with MoO3-C70 Schottky junction, compared to the VOC and JSC of the device without HTL. The high VOC of the device is attributed to the Fermi energy-difference induced band bending in the C70 layer and energy level tuning in the ITO electrode, while a smaller energy offset between the conduction band edge of MoO3 and the LUMO
level of C70 is not effective in electron blocking to the ITO electrode, showing low JSC. By adopting double HTL layer of MoO3/TPTPA, the lower work function of TPTPA than that of MoO3 induces effective electron blocking, leading to the enhancement in both VOC and JSC of the device. Therefore, our study suggests that MoO3 is not an effective electron blocking layer in fullerene-based Schottky junction solar cells, but does increases Voc and has great potential as an electrode dopant to enhance overall performance.

Generally, organic compounds have closed-shell electronic structure, and they are insulating and non-magnetic. In 2008, an epoch-making doping technique for organic crystals was reported by Morpurgo, et al [1]. Conductivity enhancement and metallic functionality were realized at the contact interface between a TTF (tetrathiafuluvalene) single crystal and a TCNQ (7,7,8,8-tetracyanoquinodimethane) single crystal. These characteristics were derived from carrier doping from electron donors to acceptors. Such a contact doping technique has great potential values, and we utilized it for the combinations that do not give any charge-transfer (CT) crystals, because conductive TTF- TCNQ CT crystals were found to form at the interface in the case of TTF and TCNQ contact [2].
In this work, we attempted to demonstrate metal-like behavior with using contact doping between electron acceptor, F₂TCNQ, and various donor organic crystals. The charge conduction properties and the chemical conditions of the interfaces were investigated. By the current-voltage measurements, it was confirmed that the electrical conductivities of all interfaces were enhanced compared to as-grown F₂TCNQ crystal. In addition, the results of the temperature dependence of the carrier transport properties revealed that several interfaces with F₂TCNQ showed metal-like behavior. Furthermore, temperature dependence of the thermoelectric power (TEP) was measured to investigate the majority of the charge carriers at the contact interfaces. In this presentation, we will show charge conduction properties of interfaces formed by F₂TCNQ and several donor crystals, and discuss about the relationship between the charge transport and ionization potential of donors and π-π interactions between the donors, as well as the majority of the charge carriers at the contact interfaces.
[1] H. Alves, A. S. Molinari, H. Xie and A. F. Morpurgo, Nature Mater., 7, 574-580 (2008)
[2] Y. Takahashi, K. Hayakawa, T. Naito, and T. Inabe, J. Phys. Chem. C., 116, 700-703 (2012)

With the aim of developing compact, highly sensitive, and high-resolution color video cameras, we have proposed a color image sensor with three stacked organic photoconductive films (OPFs) that are sensitive to only one of the primary color components (red (R), green (G), and blue (B)) and have a transparent signal readout circuit [1-2]. The need for a color separation optical system, such as a dichroic prism or color filter array, is eliminated when using this stack-type organic image sensor because color separation can be achieved in depth direction with only the image sensor. In addition, this type of sensor uses incident light efficiently because the lights, which are not absorbed by a certain OPF, are absorbed by the respective underlying OPF intended for the specific color component. Because of this stacked structure, the light receiving areas of all the RGB pixels can be maximized, even on a single sensor chip. However, it is essential to use three OPFs that are selectively sensitive to R, G, or B light, and the incident photon-to-current conversion efficiencies of available films are insufficient, especially for R-sensitive films based on phthalocyanine (Pc) derivatives, such as ZnPc and TiOPc, which have relatively low efficiencies (20%-30%). In addition, the dark current of the organic film must be suppressed as much as possible, and preferably to a level comparable to that of Si photodiodes.
In this study, boron sub-2, 3-naphthalocyanine chloride (SubNc) was investigated as a red-sensitive OPF. A photoconductive cell was fabricated, and its current-voltage characteristics with and without light irradiation and external quantum efficiency (EQE) were evaluated. The structure of the photoconductive cell was as follows: glass substrate/In-Zn-O/spiro-2CBP (30 nm thick)/SubNc (50 nm)/Alq3 (30 nm)/Al (50 nm) (spiro-2CBP = 2, 7-bis(9-carbazolyl)-9, 9-spirofluorene; Alq3 = tris (8-hydroxyquinolinato) aluminum). The spiro-2CBP and Alq3 were inserted between the SubNc and electrodes as dark current injection blocking layers. These three layers were successively evaporated in a vacuum on the indium-zinc-oxide-patterned substrate. The SubNc film absorbed red region light well with an absorption peak of 695 nm. The EQE of the cell reached 80% when 690 nm light was irradiated at a power of 50 µW/cm² and the applied bias was 15 V. In addition, the blocking layers effectively suppressed the dark current of the OPF, which was 20 nA/cm² at an applied voltage of 15 V. These results indicate that SubNc is a promising candidate as a red-sensitive OPF for RGB stacked sensors.
REFERENCES
[1] H. Seo et al., Jpn. J. Appl. Phys., 23, 024103 (2011)
[2] T. Sakai et al., Proc. SPIE, 9022, 90220J-1 (2014)

Thin films of organic semiconductor Poly (3,4 ethyldioxythiophene) Polystyrene Sulfonate (PEDOT:PSS) were produced by drop casting method onto silicon and fused silica substrates. These films were then treated with sulfuric acid for various amounts of time (i.e., 10, 20, 40, and 60 minutes). The acid treatment acts as a charge doping process. In order to understand this bulk doping mechanism, FT-IR spectroscopy and Spectroscopic Ellipsometry were used to analyze the difference between untreated and acid-treated PEDOT:PSS films. Resistivity measurements using the Van Der Pauw method were performed to determine the electrical conductivity of the PEDOT:PSS. Our preliminary results show that when PEDOT:PSS films were treated with sulfuric acid the PSS is protonated and precipitates out leaving only the PEDOT film left on the substrate surface. In conjunction with this, a decrease in thickness and an increase in electrical conductivity were observed.

The threshold voltage of organic field-effect transistors (OFETs) has a great dependence on several parameters: voltage stress conditions, presence of oxygen, incidence of light and others. Recently, we have observed that the combination of short exposure to oxygen from ambient environment and the incidence of light in the visible range during the polarization of P3HT OFETs in the depletion regime is responsible to a shift, in direction to more positive values of the threshold voltage, which is dependent on the amplitude of the maximum voltage in the depletion direction, accompanied by an increase of the active layer intrinsic conductivity (and, consequently, to a decrease of the on/off ratio). This behavior is attributed to the formation of shallow traps levels due to the complexation of oxygen with the thiophene ring. A dynamic study of the evolution of the electrical properties of the transistors show clearly the appearance of two distinct processes: one with a relaxation time in the order of few minutes and a second one about two orders of magnitude longer. In the present work, we present the results of the influence on the characteristic relaxation times with the oxygen content of the environment used during the electrical measurements. It was observed that the higher the oxygen content in the measurement environment, the faster is the relaxation time of the electrical parameters. This study can be used to find the optimum conditions in order to stabilize the electrical response of organic P3HT-based devices and obtain maximum device performance. (Authors acknowledge FAPESP grant # 2013/24461-7).

An aqueous solution processable molybdenum oxide (MoO_x) layer is developed in this study as an alternative to PEDOT:PSS as a hole transport layer (HTL) for organic solar cells. The MoO_x layer has increased optical transparency compared to the reference PEDOT:PSS. This increases light absorption leading to higher short circuit current and power conversion efficiency (PCE).
Polymer solar cells are fabricated using PCDTBT:PC[70]BM (poly[N - 9prime;-hepta-decanyl-2,7-carbazole-alt-5,5-(4prime;,7prime;-di-2-thienyl-2prime;,1prime;,3prime;-benzothiadiazole)] (PCDTBT):[6,6]-phenyl C70-butyric acid methyl ester (PC[70]BM) (1:4 weight ratio) by using ITO/HTL/PCDTBT:PC[70]BM/TiO_x/Al device structure. A thin layer of Molybdenum oxide (MoO_x) is prepared by spin coating MoO_x precursor solutions (MoO₃ dissolved in ammonium hydroxide and water). The device using the MoO_x layer (current density - voltage curves, EQE, IQE) were compared to either no hole extraction layer or PEDOT-PSS layer. The optimized conditions of MoO_x hole transport layer outperforms the reference PEDOT:PSS layer (3.5 %PCE) used in our labs reaching 4.4 % PCE.
Furthermore, charge extraction using a nanosecond switch technique (TRCE) combined with photovoltage decay measurements are introduced to specifically study the effect of interfacial layers on charge recombination kinetics in the active layer of organic solar cells. The results show increased lifetime of charge carriers using optimized charge extraction layers at the same charge density which can reduce charge recombination at the electrode surface resulting in an increasing in PCE.

Charge transport properties of phenylimidazole type emitter doped organic materials were studied by designing organic materials derived from a pyridoindole moiety. The pyridoindole derivatives could exhibit both good hole and electron transport properties due to hole transport character of carbazole and weak electron transport character of pyridoindole. The doping of phenylimidazole type emitter retarded hole transport and decreased hole density. However, good hole transport character of the pyridoindole derivatives compensated the hole trapping of the phenylimidazole type dopant, and could balance hole and electron density. On the other hand, the doping of electron trapping doping could not keep the similar hole and electron density due to relatively strong hole transport character of the pyridoindole derivatives. Therefore the doping of phenylimidazole type emitter was an effective way of balancing holes and electrons in the pyridoindole type organic materials.

In spite of glorious improvement in soft electronics, interfacial issues between each functional layer have been pointed out as drawback. Especially, tandem structured transistors are easily exposed to interface issues such as fracture, delamination, and slip, which provokes bad stability of device in bending condition. In this manner, one can use a multifunction elastomer electrolyte, which simultaneously works as substrate and ionic dielectric in the same layer. Through this device architecture, the reduced number of layers in the device could decrease versatile interface failures. Further, such elastomer electrolyte gives tremendous benefit like high transparency, low power operation, and even stretchability in the devices.
In this talk, we present highly flexible and ultralow power freestanding transistor gated by elastomer electrolyte. This device operates at low voltage (V_ds = -0.5V, V_gs = -1V) and show efficient bending tensile strain up to 7%. Especially, carrier mobility and on current of flexible transistor were sustainable in moderate bending (bending tensile strain ~3.5%). Moreover, the device showed the exceptional resistance to fatigue in cyclic bending test. We believe that our device design will provide rational guide to flexible electronics.

Unbalanced hole and electron mobilities in organic photovoltaic cells (OPVCs) can be limiting factors for both key device parameters the short-circuit current density (J_SC) and the open-circuit voltage (V_OC). In particular, conjugated polymer/fullerene-based OPVCs suffer from the low hole mobility in the donor polymer as compared to the typically higher electron mobility in the fullerene which promotes, e.g., parasitic charge recombination at the polymer/fullerene interface. In this context, molecular electrical p-doping emerged as valuable approach for enhancing both the hole mobility and the conductivity through increased carrier concentration in the donor material. There, typically strong electron acceptors are introduced into the conjugated polymer matrix for enhancing the hole mobility (in the low doping concentration regime) as well as the conductivity by increasing the mobile-carrier concentration in the donor polymer. In order to explore the impact of molecular electrical doping in photovoltaic applications, the electronic, optical, and morphological properties of p-doped polyfuran (PF) films were investigated over a wide range of doping ratios employing tetrafluoro-tetracyanoquinodimethane (F4TCNQ) as prototypical molecular p-dopant. From the presence of the characteristic F4TCNQ anion transitions together with the spectral signature of the positive polaron of the donor polymer in optical absorption spectroscopy data we deduce integer-charge transfer between PF and F4TCNQ in the doped films. Already at a dopant loading as low as 2%, we observe a conductivity increase of one order of magnitude which, consequently, doubles J_SC of fullerene-based bilayer OPVCs built thereof. For higher doping ratios, however, F4TCNQ is found to precipitate at the heterojunction between the doped donor polymer and the fullerene acceptor, as evidenced by scanning force microscopy. Ultraviolet photoelectron spectroscopy reveals that the presence of F4TCNQ at the PF/fullerene interface acts beneficial to the energy-level alignment for OPVC applications: the open-circuit voltage is almost doubled to 0.4 V, as the energy separation between the onset of the PF valence-band and the lowest unoccupied molecular orbital onset of the fullerene was increased from 0.5 eV to 0.9 eV.
Overall, our results show that molecular electrical doping not only favorably acts on conductivity but also enables tuning the energy-level alignment in OPVCs, thus paving the way for achieving enhanced functionality in organic electronic devices.
S. Yu, J. Frisch, A. Opitz, E. Cohen, M. Bendikov, N. Koch, and I. Salzmann, "Effect of molecular electrical doping on polyfuran based photovoltaic cells", Appl. Phys. Lett. 106 (2015) 203301.

Organic light-emitting devices are promising elements for display and artificial lighting technologies. In the fabrication of full color displays, one of the challenges is complex device fabrication to get full-color emission containing red, green, blue emissions in small areas [1]. Furthermore, three color displays can be used to get white light emission via complicated methods [2]. In this perspective, color-tuning have always drawn significant attention of the researchers in the field either to achieve white light, multi- or full-color devices. A simple light-emitting device which can emit a wide range of colors is always desirable to simplify the fabrication techniques.
In our work, color-tunable polymer organic light emitting devices were fabricated using a hybrid device architecture combining organic light emitting diode (OLED) and light emitting electrochemical cell (LEC) characteristics in one structure. A novel alcohol-soluble ionically functionalized polyfluorene (PFN) was utilized as an emissive polyelectrolyte on top of a conventional OLED stack. Depending on the thickness of the PFN layer, we defined a fine tuning in the color emission ranging from yellow-green (x=0.39, y=0.47) towards the edge of white emission region (x=0.29, y=0.4) according to the Commission Internationale de I&’Eclairage (CIE) 1931 chromaticity diagram. The luminance-current density-voltage, the lifetime and the electroluminescence characteristic of the organic light emitting devices were obtained. The solution processed bilayer hybrid device comprising an air-stable Ag cathode exhibited a maximum luminance of ~ 500 cd/m². We highlight this new method as a strategy for solution processed white or multi-color polymer organic light emitting devices.
References
[1] H. Kobayashi, S. Kanbe, S. Seki, H. Kigchi, M. Kimura, I. Yudasaka, S. Miyashita, T. Shimoda, C. R. Towns, J. H. Burroughes, and R. H. Friend, Synth. Met. 111-112, 125 (2000).
[2] Brian B, D'Andrade W., Thompson M. E., and Forrest S. R., Controlling Exciton Diffusion in Multilayer White Phosphorescent Organic Light Emitting Devices, AdvancedMaterials, 14, 2, (January 2002), 147-151.

For years, it has been very difficult to answer the question, “what is key difference between organic semiconductor and inorganic counterpart?” The difficulty of giving the answer exists in scientific points. The below summarizes characteristics of organic semiconductors (molecular crystals), which can be origins of peculiar scientific features that cannot be expected in inorganic counterparts.¹
(1) Intermolecular interaction in molecular crystals is very weak.
(2) Molecular size is very large.
(3) Symmetry of the molecular structure is very low (the structure is planar or complicated 3-dimensional structure etc.).
(4) Each molecule consists of light elements (mainly H, C, O, N), but the molecular weight is very large [i.e. pentacene (C₂₂H₁₄ ) MW= 278; C₆₀ MW= 720; metal-free phthalocyanine (C₃₂ H₁₈ N₈) MW= 514]. All of these molecules are much heavier than Si (atomic weight= 28) and available heaviest elements (atomic weight ~100)].
It is thus convinced that mysterious properties of organic semiconductors may be brought from each or interplay of these characteristics. One of good examples of peculiar properties from these characteristics is band gap states in molecular crystals, which may interestingly originate from all of these (1~4). Small structural imperfections can be easily introduced by weak perturbation (1), such as weak contact to other materials², gas exposure³ etc., resulting in more density of states (due to 1-3) than the density expected from “introduced defects”(2,3). Such gap states have broad energy distribution and give tailing states from HOMO or LUMO into the band gap (1-3). At an electron-transferred interface there are fully relaxed polarons with different binding energies according to energies of molecular vibrations and crystal phonons (4). These polarons thus exist near the band edges, overlap with the tailing gap states and therefore cannot be separated from the tailing states. All of these states in the gap control the Fermi level location in the gap and charge trapping, namely the energy level alignment at electrode-organic^2,3 and organic-organic interfaces⁴ and charge mobility⁵. We will introduce typical experimental results on these gap states.
¹ N. Ueno, in "Electronic Processes in Organic Electronics" (Springer Series in Materials Science Vol. 209), Eds. H. Ishii, K. Kudo, T. Nakayama, and N. Ueno (Springer, Tokyo 2015), Chap.1, pp.3-9.
² T. Sueyoshi et al., Appl. Phys. Lett. 95, 183303 (2009), T. Sawabe et al., Appl. Phys. A 95, 225 (2009).
³ T. Sueyoshi et al., Appl. Phys. Lett. 96, 093303 (2010), F. Bussolotti et al., Phys. Rev. Lett. 110, 267602 (2013).
⁴ H.-Y. Mao et al., Org. Electronics 12, 534 (2011), F. Bussolotti et al., Phys. Rev. B 89, 115319 (2014).
⁵ J.-P. Yang et al., Org. Electronics 15, 2749 (2014).

The performance of organic electronic devices is steadily improving thanks to the advancement in understanding and controlling the organic material molecular structure and electron transport properties. Yet, a few major issues still need to be addressed, such as the achievement of an efficient charge carrier injection. The implementation of low contact resistance devices usually relies on interface physics (matching the metal electrode work function to the molecular energy levels of the semiconductors), or on dedicated device architectures. The electrical doping of organic films can be a very attractive way to further improve the efficiency of organic devices and ion implantation, the process often used to this aim in the fabrication of inorganic semiconductor devices, has not been yet applied to small molecule organic semiconductors.
We report on the effects of low energy (30keV) ion implantation on thin films of Pentacene, carried out to investigate the efficacy of this process in the fabrication of organic electronic devices. Two different ions, Ne and N, have been implanted and compared, to assess the effects of a different reactivity within the hydrocarbon matrix. A strong modification of the electrical conductivity, stable in time, is observed following ion implantation. This effect is significantly larger for N implants (up to 8 orders of magnitude), that are here shown to introduce stable charged species within the hydrocarbon matrix, not only damage as is the case for Ne implants. Fully operational Pentacene thin film transistors have also been implanted and we show how a controlled N ion implantation process can induce stable modifications in the threshold voltage, without affecting the device performance. We have monitored the effectiveness of low-energy ion implantation in controlling and stabilizing the organic thin film resistivity over a long time (over 2000 hours), proving how ion implantation can be safely carried out on fully operational OTFTs. The electrical properties of the Pentacene layer and of the OTFT have been investigated by means of current-voltage and photocurrent spectroscopy analyses, while the structural modification induced by ion implantation have been characterized by depth resolved X-ray Photoemission Spectroscopy analyses.
[1] B.Fraboni, P.Cosseddu, Y.Q. Wang, R.K. Schulze, Z.F. Di, A.Cavallini, M. Nastasi and A.Bonfiglio Organic Electronics, 12, 1522 (2011)
[2] B.Fraboni, A.Scidagrave;, P.Cosseddu, Y.Q. Wang, M. Nastasi, S.Milita and A.Bonfiglio Semicond.Sci Technol. 2015 in press
[3] P.Cosseddu, B.Fraboni, A.Scidagrave;, Y.Q. Wang, M. Nastasi and A.Bonfiglio Synthetic Metals 2015 in press

Being able to control charge density in a semiconductor by doping is a key step towards enabling technologically relevant electronic devices. In spite of progress in many areas of development in conjugated polymers, controllable and efficient charge control via doping remains elusive. In particular doping efficiencies are low and the doping process is not always reproducible. We study the P3HT/F4TCNQ model system. Using spectroscopic characterization we are able to determine that in the solid state the charged state is a polaron at all dopant concentrations, in contrast to solutions where bipolarons are observed. By using a closed system with known F4TCNQ concentrations we measure the absorption cross-section of polarons in P3HT. This measurement provides a simple method to determine the amount of charge in doped P3HT. We also measure the mobility as a function of doping and find that it has a non-monotonic behavior.
In order to provide a better understanding of what controls doping efficiency, we study the precursor species in solution prior to casting the film and find that association of P3HT and F4TCNQ in solution is necessary for high efficiency of the doping process. Such association can be simply controlled by controlling the temperature of the solution. Finally, conductive-tip AFM is used to study the location of the dopants in the film with the goal of elucidating how the spatial distribution of dopants is related to the film microstructure.
Our studies on a model system provide insights into the doping mechanisms in conjugated polymers and explain some of the limitations observed with current dopants.

This talk reviews two key aspects of doping in organic molecular and polymer semiconductors. The first pertains to the fine control of very low densities of carriers added in the host, with dopant-to-host molar ratios (MR) varying in the 10^-4-10^-3 range. Filling electronic trap and tail states without adding “free” carriers to the system dominates in this doping regime, leading to the de-activation of deep gap states and a super-linear increase in carrier mobility and film conductivity. Examples are given for three molecular (n-doped C₆₀, and p-doped CuPc and TIPS-pentacene) and one polymer (n-doped P(NDI₂OD-T₂)) semiconductors [1-3]. We also demonstrate for this very low doping regime the ability to control the threshold voltage of C₆₀, TIPS-pentacene and P(NDI₂OD-T₂) field-effect transistors without significantly affecting the field-effect mobility and ON/OFF ratio of the device [2,4]. The second part of the talk focuses on dopant diffusion. This aspect of doping is crucial for the stability of devices that include spatially-confined doped regions, e.g. doped interfaces for injection enhancement, or p-i-n-like structures. Using a combination of electron spectroscopy, secondary ion-mass spectrometry (SIMS), and current-voltage measurements on diode-like structure, we show strikingly different rates of diffusion of the p-dopant Mo(tfd-CO₂Me)₃ in three different polymers, i.e., P3HT, PBDTTT-C and poly-TPD. While very rapid room temperature diffusion occurs in pure rr-P3HT, presumably along disordered areas, no measurable diffusion in observed in P3HT-PCBM blends [5]. Considerably slower diffusion is seen in PBDTTT-C, which is more amorphous than P3HT. Finally, the dopant is entirely stable even at elevated temperature in amorphous poly-TPD. Consequences for doped devices are discussed.
[1] S. Olthof et al, Phys. Rev. Lett. 109, 176601 (2012)
[2] J. Belasco et al., Appl. Phys. Lett. 105, 063301 (2014)
[3] A. Higgins et al., Appl. Phys. Lett. 106, 163301 (2015)
[4] S. Olthof et al., Appl. Phys. Lett. 101, 253303 (2012)
[5] A. Dai et al., Org. Electr. 23, 151 (2015)

In recent years, non-fullerene acceptor based organic solar cells have attracted considerable attention since they have a broad optical absorption feature and a flexibility of energy level control. Subphthalocyanine (SubPc) acceptor, in particularly, exhibited superior device efficiency of over 8% [1]; being comparable to fullerene acceptor based devices. To clarify the energy loss mechanisms in the SubPc acceptor based solar cells in detail, we investigated the electronic structures at SubPc/buffer/Ag heterointerfaces by means of synchrotron based in-situ ultraviolet photoelectron spectroscopy (UPS). The SubPc/buffer/Ag heterostructures were formed by depositing buffer materials on Ag and subsequently depositing SubPc onto buffer/Ag stack structure in a step-by-step way in a vacuum deposition chamber. A series of pyridine based acceptor molecules (BCP, TPBi and TAZ) were applied as the buffer layers. For all buffer/Ag stack structures, metal induced gap states within the HOMO-LUMO gap were observed. These states were located near the LUMO levels and their density of states reached to the Fermi level. In general, energy position and electron occupation of gap states in organic semiconductor layers affects their carrier conductivity. Thus, the gap states act as shallow donors, which enhance the electron conductivity of the buffer layers. The work function of the buffer/Ag stack structures showed fairly small values (3.3-3.6 eV), so the buffer layers have a role of promoting a smooth carrier injection from cathode to acceptor layer. Nevertheless, the energy difference between LUMO of SubPc and Fermi level of the buffer is estimated to be ~0.8 eV. The carrier injection barrier of the system was quite huge as compared with that of C60/buffer/Ag heterostructure (~0.2 eV). The difference in the barrier height might be caused by the density of disorder originated gap states in the acceptor layers and/or acceptor/buffer heterointerfaces [2] since structural disordering is easily introduced in non-planar aromatic systems (SubPc). We consider the disorder originated gap states, which were located at the deep energy position, caused the Fermi level pinning and the carrier injection barrier at the SubPc/buffer/Ag heterostructures.
[1] K.Cnops et al., Nat. Comm. 5, 3406 (2014). [2] T. Sueyoshi et al., APL 95, 183303 (2009).

Interface energetics strongly affects the performance of organic devices, such as organic photovoltaic cells (OPVCs). A full understanding of the donor/acceptor (DA) interface is required to optimize charge carrier generation and propose design criteria for more efficient materials. Especially the evolution of the relative energies of the frontier orbitals of the donor and acceptor materials are of particular interest [1], because a small change of the energy offsets of these frontier orbitals at the very interface impacts the free carrier lifetime, and consequently overall OPVC performance [2]. The electronic structure of the prototypical DA pair metal-phthalocyanine/fullerene, has been widely investigated, employing ultraviolet and X-ray photoelectron spectroscopy measurements (UPS and XPS, respectively). Previous studies showed that the energy shift at their interface is not always monotonic [3-5]: The vacuum level and occupied states at first shift downward at the direct interface, followed by a successive upward shift in the C₆₀ layer. However, a comprehensive understanding of this complex energy level evolution has not been reached.
Experimentally, we determined a clear ‘kink&’ of the work function shift at the zinc phthalocyanine (ZnPc, bottom layer)/C₆₀ (top layer) interface prepared on 2 nm molybdenum trioxide (MoO₃)-covered Au(111). The analysis of the UPS spectra revealed that the vacuum level shifts downward by 0.3 eV upon depositing 1 nm C₆₀ on the ZnPc layer. In addition to the concurrent downward shifts of the occupied states, the highest occupied molecular orbital (HOMO) level was broadened [6]. This implies an increase in the energetic disorder caused by structural imperfection and/or intermixing between the donor and acceptor layer upon interface formation [6]. Further deposition of C₆₀ pins its lowest unoccupied molecular orbital (LUMO) to the Fermi level and then results in an upward energy shift. Corresponding electrostatic calculations [7] quantitatively reproduced these complex results and clearly show that the broadened ZnPc HOMO, due to interface disorder, leads to the ‘kink&’ of the electrostatic potential [6].
Complementary results on the perfluorinated copper phthalocyanine/sexithiophene interface, where the energetic landscape is reversed [6] suggest these interface shifts to be a general phenomenon and that they should be taken into account to fully understand the energetics at organic heterojunctions.
[1] S. Sweetnam et al., J. Am. Chem. Soc. 136, 14078 (2014).
[2] S. Izawa et al., Adv. Mater. DOI: 10.1002/adma.201500840 (2015).
[3] A. Wilke et al., IEEE J. Sel. Topics Quantum Electron. 16, 1732 (2010).
[4] K. Akaike et al., Org. Electron. 14, 1 (2013).
[5] Y. Nakayama et al., Adv. Energy Mater.4, DOI: 10.1002/aenm.201301354 (2014).
[6] K. Akaike et al., submitted.
[7] M. Oehzelt et al., Nat. Commun.5, 4174 (2014).

Single component small-molecule organic semiconductors (OSC) have made substantial progress in commercial applications such as in displays, but heavily rely on the intrinsic properties of the OSC, which severely limit the type of applications where these materials can be adopted. Organic binary charge transfer (CT) compounds are regaining considerable attention as a source for new engineering (i.e., doping) and design opportunities. CT complexes consist of two types of OSC: an electron donor (D) and an electron acceptor (A). The degree of CT between the D and A molecules has profound impact on the electrical conductivity properties of the complex ranging from superconducting,^1,2 metallic conducting,^2,3 semiconducting,^2,4 or insulating.² In addition, polymorphism in CT compounds can impact the conductivity, which can vary up to many orders of magnitude for the different stacking configurations.^{5, 6} CT OSC offer the potential of functional tunability which make them markedly different from other semiconductors.
The CT semiconductor DBTTF-TCNQ (dibenzotetrathiafulvalene-7,7,8,8-tetracyanoquinodimethane) exists in two polymorphs (rectangular and elliptical) as single crystals, and each has distinct electronic properties. The two polymorphs undergo different degree of CT as measured by X-ray photoelectron spectroscopy (XPS). The degree of CT impacts the electronic properties, as observed by the positions of the highest occupied molecular orbital (HOMO) by UV photoelectron spectroscopy (UPS) and electrical properties in a field-effect transistor (FET) devices. The rectangular polymorph undergoes significantly more CT between the D and A, and this manifests itself through a favorably aligned HOMO position and greater electron mobility (when compared to the elliptical polymorph).
Aside from tuning the electronic properties of CT semiconductors itself, the interfaces between them and metals can impact transfer characteristics in a FET.^4,7 While it has been proposed that the tunable electrical properties are due to the ideal electronic energy alignment,⁷ there has not been a comprehensive physical and electronic investigation of these interfaces. Here, we perform detailed studies of the interface formation between organic metals and DBTTF-TCNQ, to link the interface structure to the electrical results. These spectroscopic results combined with FET electrical measurements provide a thorough physical and electronic understanding of interfaces between CT organic metals and semiconductors and will be presented.
1. D. Jérome et al, J. Physique Lett. 41 (4), 95-98 (1980).
2. K. P. Goetz et al, J. Mater Chem C 2 (17), 3065-3076 (2014).
3. H. Alves et al, Nat Mater 7 (7), 574-580 (2008).
4. Y. Takahashi et al, Appl. Phys. Lett. 86 (6), 063504 (2005).
5. T. Mori and H. Inokuchi, Solid State Commun. 59 (6), 355-359 (1986).
6. T. Mori and H. Inokuchi, Bull. Chem. Soc. Jpn. 60 (1), 3 (1987).
7. Y. Takahashi et al, Appl. Phys. Lett. 88 (7), 073504 (2006).

We report the development of a novel dopant-interlayer technology to make general ohmic contacts to bottom contact organic semiconductor field-effect transistors that overcome the limitations of previous methods based on surface modification using self-assembly monolayers, intermediate semiconductor layers, and electrochemistry. The method is able to impose ohmic contact at will by interfacial doping of the adjacent semiconductor. Using this approach, we demonstrate for a state-of-the-art DPP polymer, the electrode contact resistance decreased from 60 kOhm cm to less than 20 kOhm cm, corresponding to a contact resistivity of 4 Ohm cm², the lowest so far. Apparent linear field-effect mobility of the polymer increases from 0.25 to 0.45 cm²/ V s. Detailed characterization reveals the method is self-aligning to the metal electrode, which crucially avoids doping of the channel, and UPS reveals the contact is indeed appropriately p-doped.

In 1962 Mark and Helfrich demonstrated that the current in a semiconductor containing traps is reduced by N/N_t^r, with N the amount of transport sites, N_t the amount of traps and r a number that depends on the energy distribution of the traps. For r>1 the possibility opens that trapping effects can be nearly eliminated when N and N_t are simultaneously reduced. Solution-processed conjugated polymers are an excellent model system to test this hypothesis, because they can be easily diluted by blending them with a large band gap semiconductor. We demonstrate that in conjugated polymer blends with 10% active semiconductor and 90% large band gap host the typical strong electron trapping can be effectively eliminated. As a result we were able to fabricate polymer light-emitting diodes with balanced electron and hole transport and reduced non-radiative trap-assisted recombination, leading to a doubling of their efficiency at nearly ten times lower material costs.

While the power conversion efficiency of polymer solar cells has greatly increased over the last years by improved donor-acceptor systems and new device geometries, some mechanisms that are at the heart of organic photovoltaics are still not fully understood. In this regard, processes limiting the device performance like recombination are of special interest.
Recombination events that macroscopically influence the sample conductivity alter the spin-dependent current, which is the observable in an experiment not frequently applied to organic solar cells so far - namely electrically detected magnetic resonance (EDMR). Spin pairs that give rise to an EDMR signal can be found at deep levels due to inherent disorder, at dopant sites that also act as recombination centers, or at contact interfaces. The EDMR method also is extremely sensitive and can detect signals from only 1,000 charge carriers at room temperature.^[1]
In this contribution, we apply EDMR as a spectroscopic tool to elucidate the origin of recombination in P3HT:[60]PCBM bulk-heterojunction solar cells, chosen as a well-known model system. We show that the combination of EDMR with light-dependent current-voltage analysis allows forming a picture of spin-dependent processes that affect the lifetimes and transport of elementary excitations inside these devices under realistic operating conditions. While the latter technique is well known and commonly used for the determination of key solar cell parameters like, e.g. the fill factor, the recombination spectroscopy technique is still young and much less widespread in the community. Only recently a compact setup to measure EDMR on complete and functional devices based on a commercial benchtop spectrometer was presented, making the EDMR method more easily accessible.^[2] Here, we show that in combination, these two experimental techniques can yield valuable information and new insights into carrier transport processes like recombination which may limit the performance of solar cell devices. Furthermore, we show EDMR results on paramagnetic doping of organic semiconductor micro-crystals, which lead to a fingerprint for spin-dependent recombination events near a dopant molecule. A combination of EDMR and site-specific doping can thus help to pinpoint recombination events inside devices by distinguishing, e.g., surfaces from the bulk of organic semiconductors. The presented EDMR technique can thus be a valuable tool in the emerging field of doped intrinsic organic semiconductors.
[1] W. Harneit, C. Boehme, S. Schäfer, K. Huebener, K. Fostiropoulos, K. Lips, Phys. Rev. Lett.2007, 98, 216601.
[2] M. Eckardt, J. Behrends, D. Münter, W. Harneit, AIP Adv.2015, 5, 047139.

A low work function contact is desirable for the cathode interface in organic electronic devices. Low work function metals however, are prone to rapid oxidation and corrosion. Therefore, Al is used along with interlayers inserted at the Al/organic interface. These interlayers improve the function of Al as an electron extracting/injecting contact and impede Al-organic bonding. Amongst these interlayers, LiF has proved its significant role in providing better electron injection in organic light emitting diodes and enhanced electron extraction in organic solar cells. Nevertheless, it is still important to disentangle the extrinsic effects of LiF that potentially influence the properties of the active organic compound when deposited as the cathode interlayer.
In this work, we study the role of LiF interlayers as a doping agent in fullerene-based devices. We perform current-voltage (J-V) and capacitance-voltage characterization of capacitors out of fullerene derivatives capped with LiF/Al versus Ba/Al and Al top contacts. The effective dielectric constants extracted from fullerene/LiF/Al capacitors were significantly higher than the pristine values while other capacitors retrieved the pristine values indicating that LiF dopes the fullerene layer. In addition, high doping densities as determined from Mott-Schottky analysis and increased current densities close to Ohmic regime were observed as a signature of doping effect of LiF. In bulk-heterojunction solar cells with LiF/Al cathodes, devices with 4:1 ratios of fullerene:polymer showed distinct doping changes with respect to 1:1 ratios. While in the case of the pristine polymers, LiF did not impose any doping effect. Our J-V measurements in macroscopic level and local conductivity measurements microscopic level with Atomic Force Microscopy in proved that LiF should not necessarily be capped with Al in order to be able to dope the fullerenes films.
Our investigations suggest that the intercalation of LiF into films of fullerene derivatives upon evaporation gives rise to alkali metal-doping of these molecular semiconductors. Therefore the more openly packed fullerene derivatives are more influenced by the doping. Such doping effect cannot be explained merely with formation of a thin n-doped layer adjacent to the cathode to facilitate charge injection/extraction, as accepted within the organic electronic community. In conclusion, even though many scenarios promote LiF as a modifying interlayer for organic electronic devices, one can hardly disentangle the doping effect of LiF in fullerene based devices. Therefore, properties such as mobility, conductivity, built-in voltage and the dielectric constant, extracted from the devices in which monolayers of LiF have been used, will be dubious in conveying the correct values. LiF doped fullerene derivatives, however, can be applied as electron extracting buffer layers in organic or perovskite solar cells or find suitable applications in organic thermoelectrics.

Light-emitting electrochemical cells (LECs) comprise a blend of an emitting conjugated species and mobile ions (a dissolved salt) as the active material positioned between two charge-injecting electrodes. During operation the mobile ions redistribute, so that a light-emitting p-n junction doping structure can form in-situ. This unique operational mode allows for the utilization of thick and uneven active materials and air-stabile electrodes, which in turn paves the way for an attractive fault-tolerant and scalable fabrication in the ambient.
A straightforward analysis reveals that the LEC performance has been far from optimized, and that it is critically important to better control the in-situ doping process and identify improved electrolytes. A “too-high” doping level and a non-optimum active material thickness are common LEC features, which are concomitant with significant exciton quenching and doping-induced self-absorption. We show that these setbacks can be mitigated via a rational selection of electrolyte concentration and the thickness of the active material. The currently employed electrolytes suffer from an inadequate electrochemical stability window, and it is particularly important to improve their cathodic stability. By following such well-motivated guidelines and employing improved materials, we are today able to deliver LECs with an ambient operational lifetime (at a luminance >100 cd/m²) of > 6 months and which feature an efficiency of 16 lm/W at significant brightness.
We also report a new easily adoptable and scalable ambient-air fabrication technique, which effectively resolves problems with non-homogenous light-emission from large-area devices. The merit of the technique is demonstrated in the form of uniform light emission from large-area (>100 cm²) solution-processed devices and via the realization of light-emission from highly complex-shaped objects, such as a kitchenware fork! We further report that it is possible to attain high-quality white emission from single-emitter LECs, and that bilayer LECs, fabricated with a mask-free and additive process, can deliver static messages with a high resolution of 170 ppi.

White polymer light-emitting electrochemical cells (PLECs), which are based on fluorescent π-conjugated polymers (FCPs) and polymeric solid electrolytes (PSEs) composed of salts and polar polymer matrices, are seen as one of the most promising technologies for future lighting devices. PLECs have p-i-n junction structures which are formed by in-situ electrochemical doping of the FCPs after device fabrication. The FCPs are p-doped on the anode side and n-doped on the cathode side, and light emission originates from the i region formed between the p- and n-doped regions. We present two different approaches to realize color-stable white PLECs as shown below.
1. Using poly(9,9-di-n-dodecylfluorenyl-2,7-diyl) (PFD) as the blue exciton emitter and an excited complex consisting of PFD segments and π-conjugated triphenylamine molecules as the long-wavelength exciplex emitter.
2. Light-mixing of blue exciton emission from PFD and orange exciton emission from poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-diphenylphenylenevinylene](MEH-PPV).

Poly-3-hexylthiophene (P3HT) is one of the most widely studied and characterized organic semiconductors. The neutral, polaron, and bipolaron forms have unique chemical signatures that can be spectroscopically resolved and thus P3HT is a “standard” to better understand how doping processes in organic systems are different than for traditional inorganic semiconductors. The central focus of this talk will be elucidation of the interconnected properties of structure, energetics, and electron transfer at doped interfaces. Points of emphasis will include simultaneous nano-texturing and doping of P3HT films and the critical differences for electron transfer due to structure and energetics for amorphous versus nano-crystalline regions. Simple planar type II heterojunctions will serve as a test platform to understand the interfacial charge transfer with fullerenes, with evidence for both electron donation and electron acceptance of the P3HT, as a function of doping density. In addition, it will be shown that changes in doping density can be used to alter the injection barrier in simple organic light emitting diode platforms.

Recently, conversion efficiency of organic thin-film solar cell exceeded 10%. In 1991, I proposed pin junction incorporating co-deposited i-interlayer consisting of two kinds of organic semiconductors (so-called bulkheterojunction), which is an indispensable for present organic solar cells [1,2].
In this paper, bandgap science for organic thin-film solar cells [3], including (i) ‘seven-nines&’ purification of organic semiconductors [3], (ii) pn-control of organic semiconductors by impurity doping [4-6], (iii) Doping mechanism investigated by Kelvin band-mapping [7], (vi) pn-control of the photovoltaic co-deposited films [8,9], (v) Ionization sensitization of doping [10], (vi) ppm-doping effects in the simplest n⁺p-homojunction organic photovoltaic cells, will be presented.
1. M. Hiramoto et al., Appl. Phys. Lett., 58, 1062 (1991).
2. M. Hiramoto et al., J. Appl. Phys., 72, 3781 (1992).
3. M. Hiramoto, M. Kubo, Y. Shinmura et al., Electronics, 3, 351 (2014).
4. M. Kubo et al., Appl. Phys. Lett., 98, 073311 (2011).
5. M. Kubo et al., AIP Advances, 1, 032177 (2011).
6. Y. Shinmura et al., AIP Advances, 2, 032145 (2012).
7. Y. Shinmura et al., Appl. Phys. Express, 7, 071601 (2014).
8. N. Ishiyama et al., Appl. Phys. Lett., 99, 133301 (2011).
9. N. Ishiyama et al., Org. Electron, 14, 1793 (2013).
10. Y. Shinmura et al., Appl. Phys. Lett., 105, 183306 (2014).

This paper discusses the mechanism of generation of free charges in organic photovoltaic cells (OPV) from electrostatically bound electron-hole pairs. We explain why this process is efficient also when interfacial charge-transfer (CT) states are generated by direct optical excitation. We used semiclassical quantum dynamics at short timescale and Redfield theory at long timescale to cover both the process of dissocation and the relaxation to the lowest energy state. Our calculations suggest that a CT state with an intermediate electron-hole separation can evolve into a charge-separated (CS) state on ultrafast timescales (~100 fs) as a result of quantum diffusion. On long timescales, however, the CS states ultimately relax to the low-energy CT states due to the interaction with the thermal bath, indicating that the yield of free charge carrier generation is determined by the interplay between ultrafast charge separation, due to quantum diffusion, and the much slower quantum relaxation process.
M. Lee, J. Arago, A. Troisi, Charge Dynamics in Organic Photovoltaic Materials: Interplay between Quantum Diffusion and Quantum Relaxation, J. Phys. Chem. C, DOI: 10.1021/acs.jpcc.5b03989

In this work, we also show that the performances of inverted organic solar cells (OSCs) can be significantly improved through making quasi-ohmic cathode contacts via solution-processed alkali hydroxide (LiOH, NaOH and KOH) to treat MeO_X (ZnO and AZO) layer. We used a blend of low bandgap diketopyrrolopyrrole-quinquethiophene alternating copolymer pDPP5T-2 and Phenyl-C61-butyric acid methyl ester (PC₆₁BM) as our model for photovoltaic devices. The insertion of the alkali hydroxide layer switches cathode surfaces, from barrier-limited to quasi-ohmic electron extraction due to the lower work function, band bending induced by electron-doping and the passivation of defect of ETL. This effective quasi-ohmic cathode contacts can suppress the bimolecular recombination at the interface, leading to increased charge collection, and remarkable enhancement of PCE from 3.26% to approximately 5.90% for AZO devices, from 4.86% to approximately 6.00% for ZnO devices.
In electron-only devices (ITO/MeO_X/with or without alkali hydroxide interlayer/pDPP5T-2:PC₆₁BM/Ca/Ag), the devices without the alkali hydroxide interlayer present j-V characteristics resembling those of a Schottky contact because of a Schottky barrier for electron injection from MeO_X to active layer. Insertion of an alkali hydroxide interlayer produces quasi-ohmic properties evidenced by the linear character of their j-V characteristics. The injection current from the alkali hydroxide / MeOx device follows a space-charge limited current (SCLC) relationship, and is completely symmetrical with the injection current from Ca/Ag. From thickness-dependent valence band photoemission spectra for C60 on AZO or AZO/KOH, it&’s clear that KOH modification of AZO provides a powerful interface states, which in agreement with ground-state interfacial charge transfer from AZO to C60. This situation is similar to the K doping of C60. Band bending induced by electron-doping from the cathode with alkali hydroxide layer enables quasi-ohmic contacts for charge-carrier extraction.[1] Moreover, the devices with alkali hydroxide modified AZO or ZnO exhibit much better long-term stability, especially for the one with the KOH interlayer. Furthermore, in the devices with alkali hydroxides, the alkali ions can fill electron traps on the surface of metal oxides. It is in good agreement with that alkali hydroxides resulted in a larger number of reduced surface defect sites of AZO or ZnO via the analysis of photoluminescence. Therefore, the strong “light-soaking” effect is not observed in the device by solution-processed alkali hydroxides treatment.
This strategy of inserting a alkali hydroxide layer exemplify the importance of quasi-ohmic contact as a tool for improved the efficiency of OSCs on an industrial fabrication level, and open a promising and practical route to cathode interfacial modification via solution-processed techniques.
[1] H. B. Wang, P. Amsalem, G. Heimel, I. Salzmann, N. Koch, M. Oehzelt, Adv Mater 2014, 26, 925.

A wide range of high-performance organic solar cells (OSCs) rely on interlayers that enable efficient charge extraction. A common phenomenon of OSCs incorporating metal-oxide based electron extraction layers is the requirement to expose the devices to UV light in order to achieve reasonable characteristics - commonly referred to as “light-soaking” issue. [1,2]. This behaviour appears to be of general validity for various metal-oxide layers regardless of the active donor/acceptor system and it applies to single junction devices as well as to multi junction cells. The requirement of UV activation of OSCs may, however, impose severe problems if substrates with limited UV transmission, UV blocking filters, multi-junction devices or indoor applications are considered. In these cases, OSCs with interlayers of TiO_x or ZnO that require UV activation would remain inefficient. Thus, the development of charge extraction materials that do not rely on UV activation has been identified as a critical challenge to achieve highly efficient and long term stable devices.[3]
In this communication, we demonstrate that the necessity of UV activation that comes with the use of ZnO and TiO_x based electron extraction layers (EELs) can be avoided by using pristine tin oxide (SnO_x). We will demonstrate the preparation of SnOx by low-temperature atomic layer deposition (ALD) using tetrakis(dimethylamino)tin(IV) (TDMASn) and water as precusors.[4] Alternatively, we have very recently developed a facile solution based route using TDMASn dissolved in isopropoxyethanol which will be shown to afford SnO_x with essentially the same functionality as that prepared by ALD. The required processing temperatures remain below 120°C and are therefore compatible with flexible polymer substrates.
While for ZnO and TiO_x UV light (hnu; > E_g) is required to lower their work function, the WF of SnO_x is not changed by UV illumination. In single carrier devices, SnO_x is evidenced to allow for efficient electron extraction even without UV activation. As a result, OSCs comprising SnO_x as EEL show well-behaved solar cell characteristics with a high FF and high PCE even without the UV spectral range of the AM1.5 solar spectrum, which is in strong contrast to devices based on TiO_x and ZnO, which need to be activated by UV illumination. The general applicability is demonstrated for OSCs based on the most common fullerene electron acceptors, e.g. PC₇₀BM or PC₆₀BM. Our results are expected to be of general relevance for organic solar cells where UV light activation is undesired or not even possible.
[1] H. Schmidt et al., Appl. Phys. Lett. 2010 , 96 , 243305
[2] S. Trost, et al., Adv. Energy Mater, 2013, 3, 1437-1444
[3] J. Adams et al., Energy & Environmental Science 2015, 8, 169.
[4] S. Trost et al., Adv. Energy Mat. DOI: 10.1002/aenm.201500277 (in press).