The process of charge generation in organic photovoltaics proceeds through the formation of a charge transfer complex. In this species, a hole on the donor molecule is found near an electron on the acceptor molecule. The energy of the charge transfer complex plays the role of the effective bandgap in organic photovoltaics and is strongly correlated to the open-circuit voltage of the solar cell. Furthermore, because current is generated by the dissociation of charge transfer complexes, And yet, for all their importance, we still know little about the structure-property relationships of these species. One reason for this situation is the inherent disorder of donor-acceptor interfaces, which impedes systematic studies. In this talk I will present results on well-defined donor-acceptor interfaces obtained in bimolecular crystals, which show that the amount of active donor-acceptor interface can be controlled by processing and measured. Furthermore, dilute ternary blends with an excess of fullerenes will be used to address the issue of delocalization of charge-transfer states and its effect on open-circuit voltage. Finally, we will also address the effect of energetic disorder at the interface. These simple model systems demonstrate that there is much to be learned about the structure-property relationships of charge transfer states.

Photodiodes made from neat donor polymer or neat fullerene acceptor photo-junctions have strong energy dependence in their Internal Quantum Efficiencies (IQEs), where higher energy photons deliver higher efficiencies for exciton dissociation and / or charge collection. Classic papers by Archipov and Bässler [1] contend that charge generation is mediated by excitation excess energy and / or field assisted effects as predicted by the Onsager-Braun model. Blends of acceptor and donor that form bulk heterojunctions (BHJs - the most common architecture in organic solar cells) generally do not show such energy dependence in their IQEs, although there is much debate as to whether high excitation energies generate “hot excitons”, which in turn deliver higher charge generation efficiencies [2-5].
It has recently been shown that the IQE in high efficiency BHJs is often flat [3,6]. We have however, identified a polymer-fullerene combination that does have an energy dependent IQE [7]. We attribute this phenomenon not to hot excitons but to unequal photoinduced hole and electron transfer efficiencies (so called Channel I and Channel II pathways). We have studied these processes in several model donor systems namely: PCPDTBT, PCDTBT and DPP-DTT, all with fullerene acceptors, and rationalise our findings in terms of system energetics and charge generation kinetics.
In our paper we will describe these findings and also show how optimizing the relative Channel I and II efficiencies can deliver improved solar cell performance. The work also demonstrates the important role fullerenes play in light harvesting in bulk heterojunctions.
References:
[1] Archipov et al. Phys. Stat. Sol. (a) 201, 1152 (2004).
[2] Lee et al. JACS 132, 11878 (2010).
[3] Vandewal et al. Nature Mater. 13, 63 (2014).
[4] Grancini et al. Nature Mater. 12, 29 (2013).
[5] Armin et al. Nature Mater. 12, 593 (2013).
[6] Armin et al. ACS Photonics 1, 173 (2014).
[7] Armin et al. JACS (2014) dx.doi.org/10.1021/ja50533

Charge separation and recombination via charge transfer (CT) states at the donor-acceptor (DA) interface are key in determining both the practical and thermodynamic limiting efficiencies of organic solar cells. While the nature of the charge separation process continues to be debated, the CT density of states (DOS) distribution at the DA interface and its occupation function under solar illumination have received less attention. Most device modeling to date assumes quasi-equilibrium Fermi-Dirac statistics in determining charge distributions and thus recombination rates at the DA interface; however, recent work¹ has called this into question, suggesting that occupation of the molecular site DOS is in fact far from equilibrium, particularly near the DA interface.
Here, we explore the density of occupied states at the DA interface for small molecule solar cells consisting of the donor N,Nprime;-bis(1-naphthyl)-N,Nprime;-diphenyl-1,1prime;-biphenyl-4,4prime;-diamine (NPD) and the acceptor C₆₀ using pump-probe photocurrent spectroscopy in conjunction with CT electroluminescence and photoluminescence. Spectrally resolving the change in external quantum efficiency induced by steady-state background illumination, we observe characteristic changes in the low energy region dominated by CT absorption that differ strongly in the case of planar and bulk heterojunctions. We propose a model in which photocurrent generating CT absorptive transitions are bleached by charge carriers residing on molecules at the DA interface and show that it enables the interface DOS occupation function to be quantified directly. Taken together with bias and temperature dependent measurements, these results indicate that the occupation function depends strongly on DA interface morphology and is much farther from equilibrium in bulk heterojunction devices than in their planar heterojunction counterparts.
Independent assessment of this conclusion is provided by CT electroluminescence (EL) and photoluminescence (PL), where we observe a blue-shift in the PL spectrum relative to EL that ranges from 0 to 80 meV for planar and bulk heterojunction devices, respectively. Interpreting this shift within the framework of the generalized Planck equation for luminescent radiation, we show that the ratio of PL to EL intensities enables the difference between their effective temperatures, T_PL and T_EL, to be quantified. Consistent with our pump-probe measurements, these data indicate that T_PL can exceed T_EL by nearly 100 K in bulk heterojunction devices. The non-equilibrium nature of charge distributions at the DA interface is expected to be a general phenomenon for disordered organic solar cells and has important implications for recombination and device modeling that will be discussed.
1. A. Melianas, V, Pranculis, A. Devi#382;is, V. Gulbinas, O. Inganäs, M. Kemerink, Adv. Funct. Mater.2014, 24, 4507.

The mechanism of charge carrier generation and recombination in polymer-based bulk
heterojunction solar cells remains a topic of great interest and debate. In addition to the
difficulties in deciding where in the complex polymer-fullerene blend carriers are
generated, an agreed understanding of the actual exciton-to-carrier process remains
equally elusive. At the heart of this misunderstanding is the role played by a chargetransfer
(CT) state, which is assumed to reside at the polymer-fullerene interface. Most
models locate this species at the interface of a pure polymer phase with a pure fullerene
phase, and cite carrier delocalization as a means of explaining how the strong
coulombic binding energy is overcome to yield the separate carriers. And yet there is
also growing opinion that a well-mixed phase of polymer and fullerene is responsible for
the majority of free carriers. So how do we rationalize these two scenarios?
This presentation will focus on some recent studies of polyfluorene (PFO or F8) doped
lightly with a number of fullerenes of differing electron affinities (reduction potentials).
Using time-resolved microwave conductivity (TRMC) and photoluminesence
spectroscopy, we demonstrate the production of both free carriers and strong emission,
which is attributed to an exciplex or possibly an interfacial CT state. The presentation
will discuss the role that this species does or does not play in the role of free carrier
generation.

The investigation of the photophysics of organic-organic heterostructures composed by narrow band-gap polymers has revealed in the last years a complex excitation scenario where charge transfer excitons play a very important role [1]. I will report about the investigation performed on two sets of narrow band gap polymers both composed by a silol and a C-bridged polymer [2]. The similarities of these two sets of polymers in blends with the fullerene derivative PCBM and the effect of the morphology (additives) on the charge transfer excitons population will be discussed. Furthermore, with the help of confocal microscope micrographs I will evidence that the highest charge transfer exciton population is localized in regions where a more intimate mixture of donors and acceptors molecules is present.[3]
References
[1] C. Piliego and M. A. Loi, J. Mater. Chem., 22, 4141 (2012).
[2] T. M. Clarke, J. Peet, C. Lungenschmied, N. Drolet, X. Lu, B. Ocko, A. J. Mozer, and M. A. Loi, J. Mat. Chem. A, 2, 12583 (2014); M. C. Scharber, C. Lungenschmied, H-J. Egelhaaf, G. Matt, M. Bednorz, T. Fromherz, J. Gao, D. Jarzab and M. A. Loi, Energy Environ. Sci. 4, 5077 (2011).
[3] M. Manca, C. Piliego, E. Wang, M. R Andersson, A. Mura and M. A. Loi, J. Mater. Chem. A, 1, 7321 (2013).

Organic solar cell materials are often unintentionally doped, which can be verified easily using capacitance voltage measurements. Here we explain and demonstrate the influence of this unintentional doping on charge carrier collection and on the interpretation of various optoelectronic measurements. Doping leads to a redistribution of the electric field in the device that depends strongly on the doping density, the permittivity and the thickness of the absorber layer. Doping has a particularly strong influence on the electric field if the doping density is high, the permittivity is low and the thickness is large enough. We show that doping densities in polymer:fullerene blends are often in the 10¹⁷ cm^-3 range and therefore high enough to redistribute the electric field substantially in devices with absorber thicknesses below 100 nm. At such high doping densities, the space charge region remains constant with absorber thickness and a thickness series is essentially governed by the overlap of the optical generation profile with the space charge region. At lower doping densities, even thicker absorbers will be fully depleted leading to substantially different thickness dependent properties. In particular, we find that in doped devices charge collection losses at short circuit are nearly linear with light intensity leading to a short circuit current which is linear with light intensity despite the fact that non-geminate recombination is strong [1]. We also observe reaction orders to change strongly with thickness in case of low doping and to stay constant in case of high doping, which can be explained by the strong influence of the depletion width on reaction orders [1,2]. Finally, we discuss why charge collection in sufficiently thick normal and inverted geometry solar cells can be affected by doping. This is due to the increased charge carrier collection in the space charge region, which is facing the sun in case of the inverted geometry or facing the back contact in case of a normal geometry [3].
[1] F. Deledalle, T. Kirchartz, et al. submitted
[2] F. Deledalle, P. S. Tuladhar, J. Nelson, J. R. Durrant, T. Kirchartz, J. Phys. Chem. C 118, 8837 (2014)
[3] G. F. A. Dibb, et al. Sci. Rep. 3, 3335 (2013)

Several donor-acceptor polymers have achieved power conversion efficiency (PCE) greater than 9% and internal quantum efficiency (IQE) greater than 90% in optimized polymer-fullerene bulk heterojunction solar cells. However, most of these devices optimize with active layers less than 200 nm thick and the devices suffer from poor fill factors when the active layer thickness is increased to improve light absorption. To maximize the PCE of bulk heterojunction solar cells, 300 nm thick devices with fill factor greater than 0.75 and 90% external quantum efficiency (EQE) are needed. We use a combination of experimental results and a numerical device simulator to determine the charge-carrier mobility needed to achieve a high fill factor in a 300 nm thick bulk heterojunction solar cell. We fabricate devices with hole mobility varying from 2x10^-7 to 4x10^-4 cm² Vs^-1 and determine how the fill factor varies with device thickness and light intensity for each mobility. Using these experimental results, we calibrate our device simulator and show that the device simulator captures the essential physics involved in charge transport in these devices. We then simulate the performance of solar cells with various charge-carrier mobility and find that the electron and hole mobility must both be greater than 5x10^-3 cm² Vs^-1 in order to attain a fill factor greater than 0.75 in a 300 nm thick device. When the polymer hole mobility is decreased below 10^-3 cm² Vs^-1, the solar cells suffer from space charge build-up and the device fill factor suffers. This result is promising because the electron mobility of the most commonly used fullerene-derivative, PCBM, is already ~5x10^-3 cm² Vs^-1. Thus, the hole mobility of donor polymers only needs to be increased from the typical values of 10^-4 - 10^-3 cm² Vs^-1 to 5x10^-3 cm² Vs^-1 in order to achieve a single junction bulk heterojunction solar cell with EQE greater than 90% and PCE greater than 10%.

The potentially low cost of printed organic PV (OPV) is a major attraction of this technology. With falling costs of inorganic PV, there is a less reason to believe this will be a sufficient advantage. The reward of better optical incoupling for OPVs help to improve the energy budget of the solar module, thus somewhat compensating for a lower power efficiency. This is of importance for building integration of OPV, where walls are more attractive than roofs. For building integration of OPV, the aesthetic aspect is crucial, and the advantage of different colours of OPVs may compensate for lower efficiency. In building envelopes, very large areas may be accessible and the energy budget more attractive. We develop semitransparent polymer/fullerene solar modules printed on plastic, and use the semitransparent high conductivity conjugated polymer PEDOT both as anode and cathode. Semitransparent cells can be stacked in tandem and triple junctions, reducing the transmission but generating more electrical power. With dielectric reflectors in the form of light scattering layers, light may be returned to the active layer for reabsorption and electrical power generation. They may also be used to balance the electrical power generation between several materials in tandem devices, creating a palette of weak and strong colours, transmissive or opaque.

As a fundamental electrostatic limit, the space charge limit (SCL) for photocurrent is a universal feature and of paramount importance in organic semiconductors with unbalanced electron/hole mobility and high exciton generation. Here, we propose a new concept of plasmonic-electrical effect to manipulate the electrical properties (photocarrier generation, recombination, transport, and collection) of semiconductor devices with the help of plasmonically induced light redistribution. As a proof-of-concept, organic solar cells (OSCs) incorporating metallic planar and grating anodes are systematically investigated for normal and inverted device structures. Interestingly, although strong plasmonic resonances induce abnormally dense photocarriers around a grating anode, the grating-inverted OSC is exempt from space charge accumulation (limit) and degradation of electrical properties in contrast to the planar-inverted and planar-normal ones. It is because abnormally redistributed holes by the plasmonic-electrical effect, despite of the typically low mobility of holes, shorten hopping path of low mobility holes to reach the grating anode [1]. Our results show that together with the improved light absorption benefiting from the plasmonic-optical effect, the power conversion efficiency of the grating-anode OSC can be almost four times larger than that of the planar-anode OSC. Consequently, the work contributes to the evolution of device architecture to break the SCL with detailed multiphysics explanations. Moreover, the proposed plasmonic-electrical concept will open up a novel way to manipulate both optical and electrical properties of organic semiconductor devices for optoelectronic applications.
[1] W.E.I. Sha, X. Li, W.C.H. Choy*, Scientific Reports, DOI: 10.1038/srep06236.

In many organic semiconductors the electron transport is strongly limited by traps which are intrinsically located in the material. Trap-limited electron transport also decreases the efficiency of organic solar cells. Although the mostly used fullerene acceptors in polymer:fullerene solar cells feature trap-free electron transport, low optical absorption of fullerene derivatives limits maximum attainable efficiency. A way to higher efficiencies is the use of strongly absorbing acceptors that exhibit trap-free electron transport. Acceptors with LUMO deeper than 3.7 eV are especially interesting since electron trapping in organic semiconductors is universal and dominated by a trap located at ~3.6 eV. In this study the electron transport in solar cells based on perylene diimide (PDI) acceptors is investigated. An electron mobility of 3E-9 m²/Vs is found in (1:4) MEH-PPV:PDI-1 blends, whereas a mobility of 1.8 E-11 m²/Vs is measured in (2:1) P3HT:PDI-1 blends. The performance and loss processes of polymer:PDI derivaves solar cells are discussed.

Organic photovoltaics (OPVs) have the potential to be a competitive alternative to inorganic solar cells. Unlike inorganic solar cells, OPVs use cheaper starting materials, less extreme processing conditions, and are compatible with flexible substrates. Yet, they are still too inefficient and too unstable to contend with other solar cells on the market. Here we use a lateral bulk hetrojunction (LBH) device scheme in conjunction with scanning photocurrent microscopy (SPCM) to help understand the origin of the low efficacies in OPVs.
Most polymer systems are said to follow Langevin bimolecular recombination, leading to a decrease in efficiency as device thickness increase. P3HT and KP115 are two polymers that do not follow Langevin recombination statistics and their efficiencies do not depend on film thickness. It was previously thought that crystalline regions of annealed P3HT were the reason for this non-Langevin behavior, but KP115 does not show any crystalline domains or dependence on annealing. These systems will be compared to the very efficient PTB7 polymer, where the main recombination pathway has been demonstrated to be bimolecular recombination. In addition the effects of additives such as DIO can be probed using this device structure and microscopy.
Scanning microscopy will be used on of both Langevin and non-Langevin polymers to find the inherent differences that could account for these different recombination dynamics. Using LBHs will provide a new way to easily change the distance that charge carriers travel , where measurements in vertical devices are limited by the thickness of spun cast layers. In addition, by maintaining a uniform thickness between measurements, the absorption characteristics of the OPV cells do not change. Lastly, the transit length for charge carriers can easily be varied by tens of nanometers using e-beam lithography or hundreds of microns through conventional lithography, screen printing, and other methods. Such studies stand to increase the efficiencies of organic photovoltaics, which are characteristically cheaper and more flexible than traditional inorganic solar cells.

The overall power conversion efficiency in an organic solar cell depends on the balance between the rates of exciton dissociation, recombination and separation at the donor acceptor interface. Inability to design, control and engineer these interfaces remains a key bottleneck in their widespread use for the next generation organic electronic devices. Here, we show that we can control the ordering at the P3HT/C60 interface in bilayer device geometry by inserting a monolayer of oligothiophenes, which leads to a complete suppression in the exciplex (or charge transfer state) recombination. We observe that the photocurrent increases by 300 times, which in turn results in an increase in the overall power conversion efficiency by an order of magnitude. Moreover, we find that the oligothiophene with an odd number of rings (ter and penta oligothiophene) exhibit a much higher increase in the photocurrent in comparison to the oligothiophene with an even number of rings (tetra oligothiphene). STM measurements reveal that the oligothiophene with odd and even number of rings differ in their ordering respectively, that has a big effect on the overall device performance. We also find that this ordering is highly dependent on the side functional groups in the oligothiophenes. The mechanism of photocurrent generation will be discussed and a simple transport model will be used to explain the change in the charge transfer and recombination rates and predict current-voltage curves.

Solar thermal fuels (STFs) are an unconventional paradigm for solar energy conversion and storage which is attracting renewed attention. A material absorbs sunlight and stores the energy chemically via an induced structural change, which can later be reversed to release the energy as heat. An example is the azobenzene molecule which has a cis-trans photoisomerization with these properties, and can be tuned by chemical substitution and attachment to templates such as carbon nanotubes, forming energy-harvesting nanostructures [A. M. Kolpak et al., Nano Lett. 11, 3156 (2011); T. Kucharski et al., Nat. Chem. 6, 441 (2014)]. By analogy to the Shockley-Queisser limit for photovoltaics, we analyze thermodynamic constraints on STF efficiency, stored energy density, and storage lifetime, underscoring the key role of the free energy. In particular, we derive limits to the quantum yield for photoisomerization and the degree of conversion in the photostationary state. We show optimal values of material parameters and compare to ab initio calculations and experimental measurements.

Molecular doping is a key technique on the roadmap towards flexible and cheap organic CMOS with reliable parameter control, requiring both efficient p- and n-type doping. However, in contrast to molecular p-dopant compounds, the highly efficient n-dopant tetrakis(1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]pyrimidinato)ditungsten(II) (W₂(hpp)₄) is sensitive to immediate degradation in air due to its exceptional low ionization potential of just 2.4eV.
Here, we study the air stability of various host:W₂(hpp)₄ combinations by conductivity measurements and photoemission spectroscopy (UPS/XPS). A partial passivation of W₂(hpp)₄ molecules against oxidation is found, however, this effect depends on the energy levels of the used host material. This finding can be explained either by host-dopant hybrid formation or by a down-shift of the W₂(hpp)₄ energy levels upon charge transfer to a host material with deeper lying energy levels. Our results show the feasibility of temporarily handling efficiently n-doped organic thin-films in air, e.g., during structuring of OFETs by lithography. Furthermore, they provide new conclusions on the relativ energy level alignment of dopant and host species in mixed organic films in general.

Small molecule (SM) bulk heterojunction (BHJ) solar cells, comprising blends of SM donor with fullerene acceptor have recently demonstrated power conversion efficiencies nearing 10%. Of the parameters determining device performance in both polymer and SM solar cells, the open-circuit voltage (V_OC) contains untapped potential for improvement. There is universally a large loss in cell voltage, compared to the difference between donor HOMO and acceptor LUMO energies, the effective upper limit to V_OC. When this loss is measured against the charge transfer state energy, or E_CT, most OPV blends are found to lose approximately 0.6 ± .1 V, which is attributed to a significant density of states within the photovoltaic bandgap. A complete picture of the energetics in solar cells, and therefore a clear picture of V_OC loss, is currently lacking in the literature.
In this work, we determine the full energetics—from the energy of the charge transfer state, through the range of densities under illumination at V_OC, and down to states occupied at equilibrium—of six different solar cell blends based on a small-molecule donor and either fullerene or polymer acceptor. For all of the systems studied, analysis of charge density at open-circuit voltage over a range of illumination intensities reveals an exponential dependence of V_OC on charge density. These fits can be extrapolated over the range of 10¹⁴-10²⁰ cm^-3 to obtain excellent agreement with experimental estimates of the occupied DOS at low and high state densities, suggesting a consistent shape of the DOS over this range. We find that the role of energetic disorder in determining V_OC is defined by a trade-off between the benefit of improved free carrier generation and the cost of lower-lying states. Therefore, understanding V_OC requires knowledge not only of the energetics, but also of the charge density, which we find to vary over an order of magnitude in the systems studied here.

Thin films of polythiophene (PT) have been extensively investigated as active layers in electronic and optoelectronic devices, especially as active layers in Organic Photovoltaics (OPVs) and Field Effect Transistors (OFETs). However, to reach the market PT-based devices still have to overcome a few obstacles: 1) to better understand the interaction between oxygen and the polymer molecule, and; 2) the importance of space-charged depletion formed in the interface electrode/semiconductor due to the Fermi Level alignment. In order to better evaluate those processes, dark CELIV measurements in pristine and oxygen exposed ITO/rr-P3HT/Ag device structures are performed. A partial photovoltaic device structure is used to demonstrate the effect of oxygen doping on the transport mechanism in rr-P3HT thin films. It will be shown that unexposed devices display a single CELIV peak located in the middle of the extracting voltage pulse. As the device is exposed to air, the original peak declines gradually in intensity, while another peak rises at the beginning of the extraction period at low voltages and increases in amplitude with prolonged oxygen exposure. This phenomenon of two interacting CELIV current peaks is discussed in terms of extracted charges from the space charge regime at the ITO/rr-P3HT interface and the p-doping mechanism in rr-P3HT due to oxygen exposure.

A milestone in the development of organic photovoltaics (OPVs) has been the introduction of processing additives (a small amount of high boiling point solvent, typically diiodooctane, DIO), which may significantly increase the device performance in many photovoltaic blends. One of the major ways by which additives improve OPV device performance is by helping to optimise the active layer morphology. In addition to morphology, another parameter which is vitally important to OPV performance is the energetic disorder. However, little is known concerning the effect of additives on the energetic disorder in OPVs.
We investigate how additives affect the energetic disorder in a benzodithiophene-based copolymer (PBDTTT-C-T), a model system because of the widespread use of the benzodithiophene unit in highly efficient devices. Based on temperature-dependent mobility measurements on both diode and transistor configurations, we demonstrate that the additive (DIO) lowers energetic disorder in the blend. We show that the reduction in energetic disorder occurs primarily for electrons in acceptor domains, while the disorder for holes is relatively unaffected by DIO. The ability of DIO to decrease the energetic disorder is confirmed by highly sensitive measurements of the weak charge-transfer state emission. Wide-angle (WAXS) and small-angle X-ray scattering (SAXS) measurements suggest the origin of this reduced energetic disorder is due to increased aggregation and a larger average fullerene domain size upon addition of DIO.

Organic semiconductors often have much lower dielectric constants than their inorganic counterparts. This leads to a high exciton binding energy. Thermal energy near room temperature is not enough to break the exciton into free carriers. With the interest of increasing the exciton dissociation efficiency, we have added both neat and functionalized insulating inorganic nanoparticles (5 nm Al2O3 and 50 nm BaTiO3) into the active layer of P3HT:PCBM and PTB7:PC70BM based solar cells. Not only do the nanoparticles increase the dielectric constant of the composite film, but they also change the distribution of the built in electric field in the device. Simulation shows that a single particle inside a P3HT domain can greatly increase the exciton dissociation at the P3HT - nanoparticle interface leading to an increase in short circuit current. SEM analysis shows that the neat nanoparticles, as expected, form large aggregates ranging in size from a few hundred nanometers to tens of microns. The functionalized BaTiO3 nanoparticles show a much smaller range of aggregate size ranging from individual particles to a few micron. However, the majority of these aggregates are rod-like with two dimensions less than 100 nm and third dimension up to a micron. Since these aggregates are large compared to the film thickness, the film is no longer smooth and uniform, and the composite film thickness can vary from a few hundred nanometer to a few micron within one sample. Since the majority of the aggregates have one dimension in the optical wavelength regime, and the film thickness is not uniform, we have the added benefit of light scattering and having a longer path length inside the active layer, increasing the absorption efficiency. We show spectral and global light transmission curves verifying the scattering effects. Light I-V curves show that the solar cell performance with nanoparticles is roughly the same or better than the control device. An improvement in short circuit current up to 10% has been observed with similar or better fill factor than the control device. It is interesting that these large, nonabsorbing, insulating aggregates, which can greatly increase the thickness of the film, do not hinder the solar cell performance, but in some cases increases it. This could have future implications for other inorganic additives and shows that these OPV systems are fairly robust to inorganic additives.

The molecular weight (MW) of poly(3-hexylthiophene) is an important factor influencing the photovoltaic properties of bulk heterojunction organic solar cells based on this material. However, since different synthetic processes or repetitive soxhlet extractions - generally applied to obtain the different MW batches under study - result in samples with simultaneously varying regioregularity (RR) and polydispersity index (PDI), it has not been possible yet to find an unambiguous correlation between MW and the photovoltaic performance. We introduce recycling gel permeation chromatography as a versatile technique to fractionate the donor polymer and thereby obtain a systematic variation of the number average molecular weight (M_n =11 - 91 kg mol^-1) with an almost constant PDI and RR. Polymer crystallinity and conjugation length are evaluated by UV-Vis spectroscopy, rapid heat-cool calorimetry and selected area electron diffraction, and are found to be deeply affected by MW. This in turn influences the behavior of the charge transfer state energy, measured via Fourier transform photocurrent spectroscopy, and therefore the open-circuit voltage. The short circuit current is also affected by the MW, but mainly due to a change in absorption coefficient. The apparent recombination order is determined using transient photovoltage and photocurrent techniques and is shown to be linked to the morphology of the polymer:fullerene blend. Finally, a correlation between recombination and fill factor is suggested.

Bulk heterojunction polymer-polymer blends are actively pursued as excellent candidates for various types of optoelectronic devices such as organic solar cells (OSCs), ambipolar field effect transistors (FETs), and organic light emitting diodes (OLEDs). For the synthesis of conjugated polymers, the rational control of molecular weight for ideal polymer-polymer blend morphology is challenging owning to the lack of fundamental knowledge on intermolecular, polymer-polymer, and polymer-solvent interactions. In a systematic study of a polymer-polymer blend where the molecular weights of both the electron-donating and the electron-accepting polymers are simultaneously tuned, the resulting blend film morphology is strongly influenced by polymer intermolecular aggregation and polymer-polymer miscibility. The study of several physical properties of conjugated polymers such as radius of gyration, solubility and miscibility provide invaluable insights into obtaining optimal blend film morphology with small domain sizes and percolative polymer networks. Using this strategy, we synthesized conjugated polymers with proper molecular weights for photoactive layers having improved exciton dissociation and charge transport characteristics, yielding all-polymer solar cells with high power conversion efficiency (PCE) up to 3.5%.

In the last years a dramatic increase in the efficiency of solution processed polymer/fullerene solar cells has been reported. However, the fundamental processes involved in the conversion of absorbed photons to free charges are still not fully understood. In this work, we use time delayed collection field (TDCF) [1] experiments with exceptionally high time resolution to investigate the charge carrier dynamics in the highly efficient PCDTBT:PC₇₀BM system. Although this system has a high fill factor of around 70% and an internal quantum efficiency approaching unity under steady state illumination conditions [2], TDCF experiments reveals non-geminate recombination on the 10 ns time-scale, even for charge carrier densities comparable to one sun illumination. This loss becomes significantly accelerated at higher pulse fluence. To identify the reason for this rapid loss, the recombination dynamics was further investigated with constant white light background illumination, which introduces a tunable steady state carrier density, and on thicker devices. We observe that the short term decay dynamics is not affected by the background carrier density. It is concluded that the main reason for the nongeminate loss observed at the 10 ns time scale is recombination of hot charges close to the contacts. As recombination occurs mainly between hot electrons and holes, this loss channel seems to be insignificant under steady state illumination. Our result imply that transient experiments, considering the dynamics of “freshly-generated” charges should be considered with great care when aiming at the understanding of device function under steady-state illumination conditions.
1 S. Albrecht, W. Schindler, J. Kurpiers, J. Kniepert, J. C. Blakesley, et al., Journal of Physical Chemistry Letters 2012 / 3, 640.
2 S. H. Park, A. Roy, S. Beaupre, S. Cho, N. Coates, et al., Nature Photonics 2009 / 3, 297.

Perovskite-based photovoltaics have thoroughly triggered the interest of many research groups, and have become a field of their own on an unprecedented time scale. Energy conversion efficiencies approaching 20% as a result of just a few years of research demonstrate the enormous potential of organometal halide perovskites for photovoltaic applications.^[1] In particular, several perovskites built from ammonium salts and lead salts have been shown to exhibit many excellent features for application as active layer materials in solar cells: high absorption coefficient, low exciton binding energy and high charge carrier mobility are a few of their desirable properties that lead to device efficiencies that easily trump organic and dye-sensitized solar cells. However, the downsides of the efficient perovskites currently being studied are still very much in the shade of the remarkable progress and high performance achieved so far. Much like organic solar cells, perovskite solar cells are not intrinsically stable and, given their different nature, stabilization strategies for organic solar cells cannot be easily transferred to perovskites. In addition, lead (Pb), being the key metal to incorporate in the perovskite crystal to obtain high efficiencies, is criticized due to its toxicity and is deemed a hurdle for the commercialization of the technology. Besides these issues, there are many aspects in which an objective comparison between both technologies is appropriate. This contribution therefore provides a critical examination of the assets and drawbacks of both organic and perovskite solar cells. Different facets, covering physical and chemical properties, processability and operational stability are highlighted, leading to a nuanced SWOT analysis for an objective evaluation of the current state-of-the-art.
[1] M. A. Green, A. Ho-Baillie, H. Snaith, Nat. Photon.8, 506-514 (2014).

The molecular weight (MW) of polythieno[3,4-b]-thiophene-co-benzodithiophene (PTB7) has a remarkable influence on the operation of corresponding organic solar cells. A clear understanding of the influence of MW on molecular structure and performance of PTB7-based solar cells is still lacking. Moreover when MW and polydispersity index (PDI) are defined in an article, the distribution itself is often not shown while this can play a role in the formation of the microstructure. Therefore we investigate several commercially available batches of PTB7 with different MWs of which some even exhibit a bimodal distribution. We find that when the distribution is dominated by the presence of low MW components, the photovoltaic conversion efficiency drops significantly (from 7% to 3%), mainly due to lowering of the short-circuit current density (J_sc) and the fill factor (FF). The behavior of these photovoltaic parameters is explained by mapping the microstructure of the active layer at different length scales with Atomic Force Microscopy (AFM) and X-Ray Diffraction spectroscopy (XRD). Furthermore field-effect-transistor (FET) and photo induced charge extraction by linearly increasing voltage (photo-CELIV) measurements were performed to probe the charge carrier mobility in the direction parallel to the substrate as well as orthogonal to the substrate, yielding a significant difference in both directions for the high MW batches. Transient analysis of photovoltage (TPV) and photocurrent (TPC) measurements revealed a significantly higher apparent recombination order for the low MW batches, indicating that the amount of traps is significantly higher in these systems. These measurements are supplemented by matrix assisted laser desorption/ionisation time-of-flight (MALDI-TOF) experiments on the low MW batches, revealing a large amount of homocouplings between the donor and acceptor groups of the polymer. We conclude that the presence of low MW acts as a plasticizer preventing the formation of an interconnected network of polymer aggregates which governs charge transport in PTB7 systems.

The position of the respective energy levels at the donor/acceptor interfaces of organic solar cells is a crucial factor for the macroscopic solar cell performance. For instance, the open circuit voltage (V_{#119900;#119888;}) is related to the quasi-Fermi level splitting, the latter being limited by the energy difference between the highest occupied molecular orbital (HOMO) of the donor and the lowest unoccupied molecular orbital (LUMO) of the acceptor, the so-called effective band gap. To gain detailed insights into this correlation and into the impact of heterojunction interface energetics on the overall device characteristics, tailoring of energy levels by chemical modification offers a promising approach.
In this contribution we investigate the impact of gradual fluorination on the photophysical properties of F_nZnPc thin films and F_nZnPc/C₆₀ (with n = 0, 4, 8, 16) bilayer cells. As confirmed by ultraviolet photoelectron spectroscopy measurements, the HOMO and LUMO levels of the phthalocyanines are shifted towards lower energies upon increasing degree of fluorination. Due to the resulting increase of the effective band gap at the donor/acceptor interface, an enhancement of V_oc by 27 % and 50 % is detected for F₄ZnPc/C₆₀ and F₈ZnPc/C₆₀ bilayer devices, respectively. For the F₁₆ZnPc/C₆₀ solar cells however, a decrease in V_oc is observed. Complementary external quantum efficiency and photoluminescence measurements reveal a different working principle, which is ascribed to the special energy level alignment of the photoactive materials and comprises the transfer of excitons as a whole from the fullerene acceptor phase to the phthalocyanine layer and the subsequent dissociation at the MoO₃/F₁₆ZnPc interface. By their matching HOMO and LUMO level positions at the common heterojunction interface in combination with complementary absorption characteristics, we consider F₁₆ZnPc/C₆₀ bilayers to be a promising candidate for application in cascade and ternary structure based thin film photovoltaic devices.

Among the parameters that characterize the current-voltage curve of a solar cell, the fill-factor (FF) is the least well understood, making targeted improvements difficult. Here, we show that the FF is determined by the competition between charge extraction and recombination, according to a relationship that holds for a wide variety of organic solar cells across the whole range of 0.26-0.74. We quantify this competition with a dimensionless parameter theta, proportional to the ratio between recombination and extraction rates.
Our results are achieved by experimental measurements as well as by means of drift-diffusion simulations of organic solar cells. We simulate a large number of devices, varying charge carrier mobilities, recombination rate, light intensity, energy levels, and active layer thickness over a wide range to reproduce typical experimental conditions. Experimentally, we measure charge carrier mobilities and bimolecular recombination rates using a combination of steady-state and transient extraction techniques. The parameter theta is varied by producing devices based on different blends, by varying the processing conditions, or by measuring the same device in different conditions of temperature and incident light. Several polymer:fullerene, polymer:polymer and small molecules blends are considered.
For both simulated and experimental devices, when FF is plotted against theta the data collapse onto one universal curve, showing that all the fill-factors follow from the competition between recombination and extraction. Additionally, we show a decrease of FF with reducing light intensity for low bandgap blends, due to their low value of open-circuit voltage.
Our findings explain why FFs change significantly with thickness, light intensity and material properties. In particular, we provide a guideline for the targeted improvement of the FF of organic solar cells with given thickness, generation rate of free charges and internal voltage. The possibility for targeted improvement of FF is a promising achievement towards the enhancement of the power conversion efficiency of organic solar cells.

Fullerene cages doped with atoms, molecules or clusters in his interior are known as "endohedral fullerenes”. These compounds are actually studied [1] by their potential applications in molecular electronics, pharmacy and as contrast agents in medicine. We have studied the fullerene C60 doped with endohedral atoms from H to Ne, also the dimers Li₂, N₂ and F₂ were studied trapped inside C₆₀ and C₈₀. The potential energy surface of these system are studied by scans of a rigid geometry of the corresponding fullerene with the semi- empirical method PM6 and by density functional theory (DFT) using the hybrid functional B3LYP and basis 6-31G for all atoms. Once the atomic positions of the endohedral species with minimal energy along various relevant directions from the fullerene&’s center were calculated, a full geometry optimization at DFT level was done using the B3LYP/6-31G(d,p) level of theory. Analysis of the optimized geometries was performed in terms of absorption energies, charge distributions and other properties. The bonding properties between cage and the endohedral species were studied using Bader&’s quantum theory of atoms in molecules (QTAIM). The results show that: the HOMO-LUMO gap are tunable with the endohedral species; the interaction analysis shows a behavior going from Van der Waals to covalent bonding; and the unstable atomic N in its quadruplet state can be stabilized by the C₆₀. The C₆₀ can transfer charge from and to the endohedral species, opposite to its most known acceptor behavior. Comparison with experimental results is presented.
Keywords: Fullerene, Endohedral, C60
[1] Popov A. A., Yang S and Dunsch L., “Endohedral Fullerenes”, Chem. Rev., 2013, 113 (8), 5989.
E- mail: [email protected]

Photovoltaic active layers comprised of semiconducting single-walled carbon nanotubes (s-SWCNTs) in conjunction with buckminsterfullerene (C₆₀) have realized power conversion efficiencies of around 1%, demonstrating their potential for low cost solar energy conversion applications. We employ a non-contact transient photoconductivity technique, time-resolved microwave conductivity (TRMC), to probe photoinduced free carrier generation, transport and decay at the s-SWCNT-fullerene “bilayer” interface. The observations allow us to rationalize photovoltaic (PV) device performance based on this material combination and may point to potential avenues to improve SWCNT-C₆₀ PV devices.
Measurements of the mono-chiral (7,5) s-SWCNT-C₆₀ interface demonstrate long-lived charge separation in bilayers following excitation of (7,5) s-SWCNTs at both the fundamental excitonic transition as well as excitation of a phonon sideband coupled to this excitonic transition. The carrier generation rate is dependent on the thickness of the (7,5) s-SWCNT layer due to the requirement for excitons in the SWCNTs to reach the interface with C₆₀. Following charge separation, the recombination of charges across the SWCNT-C₆₀ interface is significantly influenced by the volumetric hole density within the thin SWCNT layer. We are able to successfully reproduce the photoconductance decay transients by utilizing a model that considers a trap-limited recombination of free C₆₀ electrons and free SWCNT holes.
We then turn our attention to the thermodynamics of free carrier generation at the s-SWCNT-fullerene interface. A study of a multi-chiral s-SWCNT/C₆₀ interface shows that the efficiency of free carrier generation is determined by the diameter-dependent s-SWCNT bandgap. This alters the energetic offset between the electron affinities of the s-SWCNTs and C₆₀ and hence the driving force for exciton dissociation to free carriers. We show that the presence of the C₆₀ interface has little influence on the exciton dissociation for large diameter (8,7) and (9,7) s-SWCNTs. We subsequently remove uncertainties due to the presence of multiple s-SWCNT chiralities by measuring (7,5) s-SWCNTs in conjunction with various perfluoroalkylfullerene derivatives that possess tunable electronic properties. Here we probe the dependence of free carrier generation on the electron affinity of the fullerene derivative and demonstrate the appearance of a Marcus “inverted region” where an increase in the driving force for electron transfer results in a reduction of free carrier generation. These two studies allow us to make further refinements to estimates of diameter-dependent electronic properties of s-SWCNTs.

It is widely believed that the open circuit voltage V_oc of organic photovoltaic cell (OPV) is related to the difference between the ionization energy of donor, I_D, and the electron affinity of acceptor, A_A, by the following empirical formula,
e V_oc = (I_D - A_A) - E (1)
where e is the elementary charge and E accounts for the energy loss (about 0.3 eV). The A_A values are often discussed based on reduction potentials in solution measured by cyclic voltammetry (CV) because the electron affinities of organic materials were not able to be determined precisely in solid.
Recently we have demonstrated a new technique called low-energy inverse photoemission spectroscopy (LEIPS) which determine the electron affinity of solid materials with the precision of 0.1 eV [1]. Using LEIPS, the electron affinities of acceptor materials most frequently used for organic photovoltaic cells, C₆₀, C₇₀, PC61BM, PC71BM and ICBA are determined in solid. The determined electron affinities, A_A, are in the range of between 3.4 and 4.0 eV [2].
Using A_A and I_D measured in the solid state using LEPS and photoemission spectroscopy, respectively, the relation between V_oc and (I_D - A_A) is examined. The I_D and V_oc values are taken from the literature: I_D of the donor polymers including P3HT, PBTTT, MDMO-PPV, MDMO-PPV, PCDTBT, and PBDTTPD, and V_oc of the 30 combinations of the donor and acceptor materials. The relation disagrees with Eq 1, and is well described by the following formula,
e V_oc = a (I_D - A_A) (2)
with the slope of a=0.62. This relation can be understood if the loss term E in Eq 1 depends linearly on V_oc; assuming E = b eV_oc in Eq 1 yields the coefficient b = 1/a - 1. The result strongly suggests that the relation Eq 1 should be reconsidered and that the loss term E is function of V_oc or (I_D - A_A).
There are three possible explanations for the discrepancy between the previous and present studies. First, I_D and A_A in the previous studies are estimated from the oxidation and reduction potentials, respectively, using CV. The oxidation and reduction potentials increase weakly with the increase of ionization energy and electron affinity, respectively, with the slopes ranging between 0.7 and 0.8, which explains the slope a in Eq 2 is smaller than unity. Second, I_D and A_A of the individual materials are used in the above discussion. The interface dipole between the donor and acceptor in an actual OPV should affect the difference of energy levels across the interface (I_D-A_A). Third, the magnitude of the loss term E depends on the recombination process. The recombination rate can be related to the interface energy difference (I_D-A_A) making E a function of V_oc in Eq 2.
References:
1) H. Yoshida, Chem. Phys. Lett, 539-540, 180-185 (2012); H. Yoshida, Anal. Bioanal. Chem. 406, 2231-2237 (2014).
2) H. Yoshida, J. Phys. Chem. C, DOI: 10.1021/jp509141y.

In order to clarify the role of excess photon energy above the optical absorption edge of an organic solar cell comprising a polymeric donor and an acceptor we measured the field dependence of the photocurrent in a bilayer assembly as a function of the photon quantum energy. Upon optical excitation of the donor an electron is transferred to the acceptor forming a coulombic bound electron-hole pair. Since the subsequent escape is a field assisted process photogeneration saturates at higher electric fields, the saturation field being a measure of the separation of the eh-pair. Using the low-bandgap polymers PCDTBT or PCPDTBT as a donor and C₆₀ as well as TNF as acceptors in a bilayer assembly we find that the saturation field decreases when the photon energy is roughly 0.5 eV above the S₁-S₀ 0-0 transition of the donor. This translates into an increase of the size of the eh-pair up to about 13 nm which is close to the coulomb capture radius. It is controlled by the delocalization of the hole of the donor. This increase correlates with the onset of higher electronic states with partial charge transfer character. It demonstrates that accessing higher electronic states does favor photogeneration but excess vibrational energy plays no role. Experiments on intrinsic photogeneration in donor photodiodes without acceptors support this reasoning.

We have recently demonstrated that following photo-excitation and electron transfer at a donor-acceptor interface, the process of final electron-hole separation against their mutual coulomb potential can be described by a so-called “effective mass model”. [1,2]
We were able to show with our simulation for two types of polymers that an increase in conjugation length represented via the effective mass improves the solar cell efficiency. The first example is a poly-para-phenylene type polymer series where the stiffness of the polymer backbone is increased. [2] Further we have investigated Donor-Acceptor polymers. One polymer with alternating Donor and Acceptor blocks in the backbone and the other polymer has a donor backbone with acceptor units in the side chains. [3] In this system we tried to investigate the interplay of chemical structures and its influence on the conjugation of the polymer.
In this work we have improved our numerical method to quantitatively assess how the interplay of electronic on-chain coupling (parameterized through the effective mass), excited state lifetime and hole mobility affect the exciton dissociation yield.
We can show that mobilities obtained by this new method are consistent with on-chain mobilities described elsewhere.
[1] C. Schwarz, H.Bässler, I. Bauer, J.-M. Koenen, E. Preis, U. Scherf and A. Köhler, Adv. Mat. 2012, 24, 922-925
[2] C. Schwarz, S. Tscheuschner, J. Frisch, S. Winkler, N. Koch, H. Bässler and A. Köhler, Phys. Rev. B 2013, 87, 155205
[3] K. Neumann, C. Schwarz, A. Köhler and M. Thelakkat, J. Phys. Chem. C 2014, 118(1), 27-36

Molecular p-doping of organic semiconductors (OSCs) is done by the admixture of strong molecular acceptors as dopants. In line with the common perceptions of molecular doping, as inferred from inorganic semiconductor physics, the prototypical conjugated polymer poly(3-hexylthiophene) (P3HT) doped with tetrafluoro-tetracyanoquinodimethane (F4-TCNQ) shows complete integer charge transfer (ICT) [1]. Furthermore, a related study doped thiophene-tetrafluorobenzene copolymers of different length revealed the localization of this process to one quaterthiophene (4T) unit [2]. For a number of small molecular OSCs, however, we found molecular p-doping to be due to the formation of intermolecular ground-state charge transfer complexes (CPXs) of OSC and dopant instead of ICT, which leads to an energy-level splitting between a doubly occupied bonding and an empty antibonding supramolecular hybrid orbital being detrimental to the doping efficiency. Its magnitude depends on the individual energy levels of dopant and host, on the nodal structure of their frontier molecular orbitals, and on their relative orientation [3,4].
Here, we aim at exploring inherent differences in the molecular doping of conjugated oligo- and polymers by comparing the F4-TCNQ doping of P3HT, where ICT is localized at one 4T unit [2], to that of molecular 4T [5]. Doping 4T films increases conductivity by up to three orders of magnitude, which is comparable to P3HT [1]. However, surprisingly small shifts of the cyano-vibrational bands of F4-TCNQ in infrared absorption spectra, which are characteristic for the degree of charge transfer (δ), reveal that only partial charge transfer occurs (δ = 0.21) instead of ICT. By further employing the weaker dopants TCNQ, F1-, and F2-TCNQ the situation is found to be qualitatively the same while no clear trend is found between δ and the dopant electron affinity. The finding of well-defined infrared bands points towards an equally well-defined mutual orientation between OSC and dopant moieties. By combining grazing-incidence x-ray diffraction reciprocal space mapping and morphological investigations, we observe phase separation between pure 4T and 1:1 co-crystalline regions of dopant and host already at low doping concentrations (< 1.5%). In marked contrast to P3HT, optical absorption spectroscopy reveals evidence for CPX formation by the occurrence of new sub-gap absorptions that scale with the dopant EA [4], which is corroborated by theoretical modelling on the density functional theory level. Overall, our study highlights inherent differences between the fundamental doping mechanisms at work in small molecular and polymeric semiconductors of identical chemical composition.
[1] P. Pingel et al., Phys. Rev. B 87, 115209 (2013); [2] P. Pingel et al., J Phys Chem Lett 1, 2037 (2010); [3] I. Salzmann et al., Phys. Rev. Lett. 108, 035502 (2012); [4] H. Méndez et al., Angew. Chem. Int. Ed. 52, 7751 (2013); [5] H. Méndez et al., submitted (2014).

Due to the nature of low dielectric constant in organic solar cells, excitons requires additional energy for dissociation. Not only did it affect J_sc, but Voc as well by affecting loss of photovoltage due to dielectric effect, the dielectric effect on the exciton dynamic as well as charge transfer (CT) state formation under the influence of dielectric nature is also important to the solar cell performance. In this report, the dielectric effect on charge transfer (CT) states formation and delocalization of CT excitons will be discussed in this isoindigo polymer system.
We found that two isoindigo polymers with identical energetics and similar chemical structures as well as pristine dielectric constant, there is a drastic difference in dielectric after blending with PC₇₁BM have been discovered. Charge modulated electroabsorption spectroscopy (CMEAS) shows that the difference in permittivity affects the hot CT states position while both have identical HOMO/LUMO levels. Additionally, with photocurrent spectral response we found that the CT states in higher dielectric constant polymer domains results in a higher output of photo-generated carriers and the transient PL result shows that this is a result of a more delocalized CT states that is not observable in either pristine nor lower dielectric polymer matrix, indicating the existence of a more delocalized CT excitons as a result of reduced columbic interaction.

Organic photovoltaics (OPV) represent a promising route toward lightweight, flexible, and low-cost renewable energy generation. An empirical reaction order higher than 2 has been experimentally observed in bulk-heterojunction (BHJ) OPV devices.¹ Trap-assisted recombination and increased surface recombination were proposed theoretically as the major reasons to explain the high reaction order.² However, understanding of how different active materials as well as contact layers affect the recombination dynamics and thus device performance is still lacking. We have studied BHJ solar cells made with different donor polymers (P3HT, PCDTBT, PBDTTPD), acceptors (PC₆₁BM, PC₇₁BM, ICBA), and hole transport layer materials (PEDOT:PSS, P-type NiO_x or CoO_x, and N-type MoO₃ or WO₃) using current density-voltage (J-V) measurement and impedance spectroscopy (IS). An equivalent circuit incorporating a recombination resistance and chemical capacitance is used to analyze IS data acquired at open circuit bias, and recombination parameters including carrier density and carrier lifetime at steady state are extracted. A significant difference in reaction order is revealed between devices made with different donor polymers. The IS results are corroborated by morphology difference revealed by atomic force microscopy. Recombination at the contact layer interface can be probed in devices with very thin active layer where surface recombination dominates. Utilizing drift-diffusion simulation that incorporates trap-assisted and surface recombination models, we will discuss the effects of doping, structure and energetic disorder, and hole transport layer on device recombination dynamics.
Reference:
1. Shuttle, C. G. et al. Experimental determination of the rate law for charge carrier decay in a polythiophene: Fullerene solar cell. Appl. Phys. Lett.92, 093311 (2008).
2. Kirchartz, T. & Nelson, J. Meaning of reaction orders in polymer:fullerene solar cells. Phys. Rev. B86, 165201 (2012).
This project is sponsored by National Science Foundation DMR-1305893

Buckminsterfullerene (C₆₀) and Copper(II) Phthalocyanine (CuPc) have been widely used as a donor-acceptor pair in Organic Photovoltaic (OPV) devices with good efficiencies of up to 5%. To date, crystallinity and molecular orientation have not been optimized for OPV performance. We will describe an approach to this optimization that takes advantage of our recent work creating crystalline flat-lying CuPc film structures on graphite[1] to promote highly crystalline layer-by-layer growth of C60 films. Using grazing incidence wide-angle x-ray scattering and atomic force microscopy, we observe a nearly ideal, highly ordered bilayer film of C₆₀ / CuPc / HOPG. We hypothesize that such bilayers will improve the Voc, Jsc, and fill factor in actual solar cells devices. Flat lying CuPc allows a strong co-facial interface with C₆₀, which induces layer-by-layer C₆₀ growth in highly ordered FCC crystals, covering the flat lying CuPc single crystal domains. Very similar morphologies are observed when varying the film thickness of either material from 5 nm up to more than 100 nm, suggesting that this advantageous morphology is not only robust, but may actually be preferred by these materials. The large, ordered domains observed in both materials and the co-facial interface should greatly increase charge carrier mobility leading to increased fill factor. We envision that bilayer films such as described here can use very high quality graphene as a transparent conducting electrode to simultaneously optimize crystallinity and molecular orientation for solar cell performance.
This work was supported by the U. S. Department of Energy (DE-FG02-98ER45737) and NSF CAREER award DMR-1056861. TM partially supported by GAANN Fellowship.
References:
[1] T. McAfee, E. Gann, T. Guan, S. C. Stuart, J. Rowe, D. B. Dougherty, H. Ade, Cryst. Growth Des. 14, 4394 (2014).

Molecular orientation of phthalocyanine has significant influences on characteristics of organic solar cells (OSCs), such as absorption, charge transfer (CT), and exciton diffusion (ED). CuI is a well-known templating layer for controlling molecular orientation of phthalocyanines, such as ZnPc and PbPc [1,2]. Here, we report that deposition of PbPc on a templating layer consisting of ZnPc on CuI leads to PbPc molecular orientation that is more favorable to photocurrent generation in the near infrared spectral region. We fabricate three types of OSCs whose structures are glass / ITO / templating layer / 25nm PbPc / 35nm C₆₀ / 8nm BCP / 100nm Ag, where the templating layer is: none (device A), 1nm CuI (device B), and 1nm CuI / 6nm ZnPc. The external quantum efficiency (EQE) and power conversion efficiency (PCE) of Device B are higher than those of device A, which is in good agreement with previous study [1]. Notable is that EQE in the near infrared (650 nm to 950 nm) region of device C are more enhanced than device B. EQEs at 880 nm are 16.3% (device A), 30.5% (device B), and 39.1% (device C) and PCEs are 1.54% (device A), 2.64% (device B), and 3.07% (device C).
By conducting total absorption (TA) measurement, we found that TA efficiency of device C (= 82.4%) at 880 nm is lower than that of device B (= 89.2%). This means that the increase in internal quantum efficiency (IQE) of device C at 880 nm is 38.7%, which is sufficiently large to overcompensate the decrease in optical absorption. Atomic force microscope measurement shows that the surface roughness and folding ratio of the PbPc layer in each sample is similar to one another, suggesting that the difference in donor/acceptor interfacial area is ruled out as a possible reason for the IQE increase of device C. By performing grazing incidence wide angle X-ray scattering (GIWAXS) measurements, we compared crystallinity, crystal phase, and molecular orientation of PbPc layers in the three types of devices. The PbPc layers in devices B and C are similar in crystallinity, both being more crystalline than that in device A. The crystal phase is also similar for PbPc layers in devices B and C, and compared to that in device A, they show more pronounced triclinic signals at q_{(theta;=59°)}=0.53#8491;^-1, q_(theta;=0°)=0.9#8491;^-1. In terms of molecular orientation, however, there is a notable difference: device C has a diffraction peak at q_(theta;=0°)=0.53 #8491;^-1, where no peak is observed at that point for device B. This indicates that a fraction of PbPc molecules that are oriented standing up (edge-on) is larger in device C than in device B [1]. We discuss further studies on how this change in molecular orientation affect ED and CT efficiency, leading to the enhanced IQE of PbPc molecules on the templating bilayer.
References:
[1] K. Vasseur, et al., Chem. Mater. 2011, 23, 886-895
[2] B. rand, et al., Adv. Funct. Mater. 2012, 22, 2987-2995

Organic bulk-heterojunction (BHJ) solar cells have shown a bright future because of their potentially low production cost. However, the performance of BHJ solar cells is known to be limited by insufficient light harvesting and low exciton generation rates. Recently, the localized surface plasmon resonance (LSPR) effects of metallic nanostructures, which enhance light absorption in the visible range, have proven useful in improving performance for organic BHJ devices. These metallic nanostructures are mainly gold or silver based expensive materials, which could increase the fabrication cost. In this work, we embedded the plasmonic copper nanoparticles (NPs) into poly(3,4-thylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) layers of organic BHJ solar cells to replace the use of expensive Au or Ag nano-materials. We found the power conversion efficiency of poly(3-hexylthiophene) (P3HT) based solar cells increased from 3.58 to 3.96%, and that of the device based on poly[[4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b']dithiophene-2,6-diyl][3-fluoro-2-[(2-ethylhexyl)carbonyl]thieno[3,4-b]thiophenediyl]] (PTB7) increased from 6.79 to 7.43%. The copper NPs enhanced light harvesting efficiency because of their LSPR effects, and thus improved short circuit current densities, which was the dominant reason for the performance improvements. We also found the accompanied increase in maximum exciton generation rate through analyzing the photocurrent densities. Our results indicate that, compared with gold and silver nano-structures, copper NP is an inexpensive and effective plasmonic material for organic BHJ solar cells.

Solar cells which have several layers chosen to absorb different parts of the solar spectrum offer the best hope to convert solar energy to electricity at high efficiency, though high cost is an issue. Recently combinations of organic films with inorganic ones are attracting attention because organic films can be implemented on top of inorganic ones at low cost [1]. In this research, a variation of this idea which combines two inorganic layers with an intermediate organic layer is investigated. Organic layer consists of a charge transfer complex such as tetrathiafulvalene (TTF) and 7,7,8,8-tetracyanoquinodimethane (TCNQ) pair. The inorganic layer which faces the sunlight is a wide bandgap semiconductor, for example zinc oxide (ZnO) and the other one is a narrow bandgap semiconductor, for example germanium (Ge). The organic layer can be made very thin, of the order of nm, since the TTF/TCNQ can function even as one monolayer. The thin organic film facilitates charge transfer from the top layer to the bottom one and compensates for the inferior charge transport properties of organic materials. Such a device structure has the potential also to improve long term durability of the organic film, since it becomes encapsulated between two inorganic ones. In order for such a device to function as intended, organic materials must adhere to the inorganic surfaces well. Both TTF and TCNQ are known to adhere to ZnO, as reported in a prior paper [2]. In this research, their interactions with Ge and Si surfaces have been studied using quantum mechanical calculations. Optimal positioning of the organic molecule with respect to the inorganic surface is calculated using the DFT method with B3LYP functional and Pople type basis sets augmented with polarization functions. Proximity of the molecule to the surface and its interaction energy with the inorganic material are estimated from the resulting geometry. TTF which is the electron donating molecule that faces the narrow bandgap semiconductor has an interaction energy of 0.1 eV at a proximity of 0.37 nm to the Ge surface. Corresponding values for the Si surface are an interaction energy of less than 0.05 eV and a proximity of 0.40 nm. TCNQ which is the electron accepting molecule that faces the wide bandgap semiconductor has an interaction energy of 0.3 eV at a proximity of 0.32 nm to the ZnO surface [2]. Other organic/inorganic material combinations suitable for the suggested device structure are being analyzed and additional results will be reported during the presentation. [1] M. Reinhard, et al., “CIGS/Organic single junction and tandem hybrid solar cells,” MRS Spring Meeting, April 2014 [2] H. Gokturk, “Matching oxide semiconductors with organic materials for solar applications,” MRS Spring Meeting, April 2014

In spite of the successful enhancement of the power-conversion efficiency (PCE) in organic bulk heterojunction solar cells by surface plasmon resonance (SPR), the incorporation of several tens of nm-sized (25-50 nm) metal nanoparticles (NPs) has some limitations to further enhancing the PCE due to concerns related to possibly transferring non-radiative energy and disturbing the interface morphology. Instead of tens of nm-sized metal NPs, here, we have incorporated dodecanethiol stabilized Au nanoclusters (Au:SR, R=the tail of thiolate) with sub-nm-sized Au₃₈ cores on inverted bulk heterojunction (BHJ) solar cells. Although metal NPs less than 5 nm in size do not show any scattering or electric field enhancement of incident light by SPR effects, the incorporation of emissive Au:SR nanoclusters provided effects that were quite similar to those of tens of nm-sized plasmonic metal NPs. Due to effective energy transfer, based on the protoplasmonic fluorescence of Au:SR, the highest performing solar cells fabricated with Au:SR clusters yielded a PCE of 9.15%; this value represents an ~20% increase in the efficiency compared to solar cells without Au:SR nanoclusters.

Agent-based modelling (ABM) is a research area that has been developed in the artificial intelligence community. The purpose of ABM is to get a converged point of intelligent behaviour from an emergent process occurring between multiple objects or agents interacting one another. Nowadays agent-based simulations with over two million agents are realised by development of hardwares, especially supercomputers and GPUs.
Kinetic Monte Carlo (kMC) simulation as a computational methodology is commonly used to measure performance of quantum efficiency on organic solar cells. kMC does literally Monte Carlo process with kinetics for the serial processes of photon absorption, exciton generation, exciton diffusion, charge (electron and hole) generation, charge transport and recombination of charges.
We attempt to move our viewpoint for the simulation from process-oriented one which kMC describes to agent-oriented one which declares particles and their interaction rules. This movement initialise the agent based modelling approach. Gains from agent-based simulation are as follows. First, we can immediately reuse well prepared software packages and hardwares for massive agent simulations. Second, we are allowed to concentrate on just what rules have to be defined on agents or particles. Rules are particle-level limited, orthogonal and declarative. The rules are defined as physical forces and logical relations. Last, model modification becomes simple and easy when simulation results are different from experimental results because the modification is achieved by just modifying rules for agents. New knowledge might be revealed through this benefit.
We implement massive agent based kinetic Monte Carlo on organic solar cells using a well-known agent-based modelling package, FLAME (Flexible Large-scale Agent Modelling Environment), supercomputers and GPUs. In this talk, we explain the details of how to implement it and show the simulation results from it.

Organic photovoltaic cells (OPVCs) have recently surmounted the 10% mark in power conversion efficiency and are close to the edge of being commercialized. This goal was reached by introducing new donor and acceptor materials with increased absorption strength and a more favorable energetic alignment. To further increase thin film solar cell efficiency - organic as well as inorganic PVCs - research directs its focus more and more on techniques for third generation photovoltaics. Therefore light management and energy transfer from surface plasmons (SPs) to molecules for exciton generation plays an important role.
Due to very thin films and short exciton diffusion lengths, absorption is normally below unity in organic photovoltaics. An increase in absorption over a broad spectral range can be achieved by the use of SP active structures like corrugated metal electrodes and metallic nanoparticles. To gain from this additional absorption, it is necessary to transfer energy from the SP to molecules by generating excitons, which can contribute to the short-circuit current of the OPVC.
To prove the principle of coupling between SPs and excitons, we investigated semi-transparent organic solar cells, in which SPs are excited at interfaces of thin metal films and a dielectric medium by using a Kretschmann configuration setup. Therefore it is essential, that the dielectric medium has a smaller refractive index than glass, e.g. lithium fluoride (LiF) or air.
To compare the SP coupling to different orientations of the transition dipole moment - as it is known from OLEDs - two donor materials were used, diindenoperylene (DIP) and dibenzo-tetraphenyl-periflanthen (DBP). Both molecules have the transition dipole moment along the long axis. While DIP crystallizes with nearly upright standing molecules on the underlying PCBM film, DBP grows amorphous with predominantly lying molecules.^1,2
To locate the angular position of the SP resonance the reflectance of the OPVC is measured angle dependent. A simultaneously measured photo current reveals the impact of SPs in these OPVCs. The use of different donors shows, that coupling from SPs to excitons only leads to a positive effect for upright-standing transition dipole moment orientation.
¹ Wagner et al., Adv. Func. Mater. 2010, 20, 4295.
² Grob et al., Appl. Phys. Lett. 2014, 104, 213304.

Multiple exciton generation by singlet fission is a fascinating approach to double internal quantum efficiency of organic photovoltaics by producing two electrons from one photon. To be able to split one single exciton into two triplet excitons, energy level of T₁ state of molecules has to be lower than half of S₁ state. However, in general, the T₁ level of π-conjugated molecules tends to be higher than the half of the S₁. Only some compounds such as diphenylbenzofurans, carotenoids, and oligoacenes are considered to be able to satisfy such an energy level requirement, and has been investigated theoretically and in terms of photochemistry. Especially in the applications to actual OPV devices, only a few oligoacences such pentacene, tetracence and rubrene (5,6,11,12-tetraphenyltetracene) have been reported to exhibit the performance based on the singlet fission mechanism.
In this study, we report the non-acence compounds having the T₁ state close to the half level of S₁ state, and their singlet fission characteristics in the organic photovoltaics. The compounds are based on 9,9&’-bifluoreneylidene (BF) structure, in which two fluorene planes are twisted each other through the central C = C double bond, due to steric hindrance of hydrogens at 1,8-positons. This distortion of the central double bond gives biradical nature to the compounds, and consequently their T₁ levels can be reduced. In addition to the control of the T₁ level, absorption of visible light and charge transporting abilities are necessary to the compounds. We synthesized three BF derivatives having thiophene, thieno[3,2-b]thiophene, and bithiophene groups between the two fluorene units: 2,5-bis(fluorene-9-ylidene)-2,5-dihydrothiophene (ThBF), 2,5-bis(fluorene-9-ylidene)-2,5-dihydrothieno[3,2-b]thiophene (TThBF), and 5,5'-bis(fluorene-9-ylidene)-5,5'-dihydro-2,2'-bithiophene (BThBF). The bulk heterojunction type OPV devices, with the BF compounds as p-type materials and C₆₀ or PDIF-CN2 as n-type materials, were fabricated. When C₆₀ was selectively photoexcited under monochromatic light, photocurrent intensity was not affected by magnetic field in the devices, showing that any triplet excitons were not involved in photocurrent generation. On the other hand, when BF compounds were selectively photoexcited, the photocurrent of the device with C₆₀ increased with the increasing of magnetic filed, whereas the photocurrent of the device with PDIF-CN2 decreased. We suppose the mechanisms are the followings. The large ΔE(CT) of BF and C₆₀ suppress charge separation of triplet excitons produced by singlet fission, and suppressed singlet fission by magnetic filed results in less population of triplet excitons and increased photocurrent. On the other hand, the small ΔE(CT) of BF and PDIF-CN2 enables charge separation of triplet excitons, and suppressed singlet fission by magnetic filed results in less population of triplet excitons and decreased photocurrent

Singlet fission has attracted a great deal of attention to provide an alternative pathway to enhance efficiencies in organic photovoltaic (OPV) devices. To convert one singlet exciton into two triplet excitons, the triplet energy has to be lower than half of the singlet energy. Oligoacenes such as pentacene and tetracene are known to exhibit singlet fission phenomenon, since they satisfy the requirement of the energy levels. However, there are few reports on singlet fission compounds other than oligoacene compounds, in OPV devices.
We demonstrate singlet fission phenomena in a series of thienoquinoidal compounds by magnetic field dependence of photocurrent in OPVs. In the three thienoquinoidal compounds, thiophene (Th), thieno[3,2-b]thiophene (TTh), and bithiophene (BTh) are incorporated as a core between two fluorene (BF) units with double bonds, respectively: ThBF, TThBF, and BThBF. The distortion between the two fluorene planes through the thienoquinoid core leads a biradical character, and consequently their E(T₁) decreased. TD-DFT calculations using the single-crystal structure of the compounds indicated that the E(T₁) of TThBF and BThBF were lower than half of the E(S₁). We fabricated OPV devices using the thienoquinoidal compounds as a donor, C₆₀ and PDIF-CN₂ as an acceptor, respectively. With PDIF-CN₂ having enough deep LUMO level for charge separation of triplet excitons in the donor, the photocurrents obtained by the excitation of the donor decreased with increasing magnetic field. This result indicates that the magnetic field suppressed singlet fission rate because of Zeeman splitting of triplet state, resulting in the reduction of the triplet excitons and the photocurrent from their charge separation. On the other hand, in devices with C₆₀, the photocurrents under the excitation of the donor increased with increasing magnetic field. This result is because suppression of the singlet fission by magnetic field reduced the triplet excitons that cannot be separated to charges because of shallow LUMO level of C₆₀.

Understanding of defects or trap states is crucial to further improve the performance of Organic Photovoltaic (OPV) devices. A previously unknown defect distribution has been reported recently for Regio-regular P3HT based OPVs. This distribution was identified from a deeper defect band in the trap Density of States (tDOS) Energy spectra. To characterize the physical origin responsible for this distribution, we probed defects in both regioregular and regiorandom P3HT based devices. For the regio-random devices, we found higher tDOS for the new defect distribution while lower tDOS for the lower energy band attributed to oxygen doping. Coupling the defect measurements with impedance spectroscopy, we also show that in regiorandom devices mobility is more than one order of magnitude lower than the regioregular devices. This lower mobility, as well as current density, in regiorandom devices validate the presence of higher amount of defect distribution in these devices. Using impedance spectroscopy we also show that transit time for electron diffusion and minority carrier lifetime are both one order of magnitude higher in the regiorandom case, whereas lifetime mobility product is still lower. These results provide more insight into the understanding of defect distributions in OPV and how they affect device performance.

Organic solar cells based on oligomer (small molecule) materials have achieved major progress in the last few years and are currently delivering the highest efficiencies of all-solid state organic solar cells with 12% reported for triple-junction devices by Heliatek. However, for a broad application of organic solar cells, module efficiencies well beyond 10% are needed, which requires for laboratory solar cells in small area efficiencies beyond 15%. One key aspect for improvement is to raise the open-circuit voltage: For the best available organic solar cells, the voltages are typically 0.6-0.7 eV below the optical gap , i.e. losses are much higher than e.g. GaAs or even perovskite cells which have losses around 0.3-0.5V. In this talk, we discuss our recent experimental and theoretical work on energy losses in small-molecule organic solar cells. The electrical characterization of solar cells is compared to DFT calculations and spectroscopic data for the charge transfer excitations. Temperature-dependent electrical measurements are used to determine the effective energy gap of the cell and point towards the energy of the lowest charge transfer state at the donor-acceptor interface as the ultimate limit for the open circuit voltage.

Photocurrent spectroscopy is used to measure both the charge transfer and exciton optical absorption spectra of various bulk heterojunction organic solar cells. The energy difference between the polymer HOMO energy and the fullerene LUMO energy is obtained from the spectra, along with the disorder energy. Combining information from cells with several different polymers and fullerenes allows measurements of the energy differences between HOMO or LUMO energies for many important polymers and fullerenes for OPV applications with an estimated uncertainty of only ~50 meV. Heterojunction band offsets are obtained for the various cells, distinguishing between the excitonic and the single-carrier band offsets. The cell open-circuit voltage is shown to be closely correlated with the interface band gap. The exciton disorder energy is directly correlated to the band-tail disorder and we also consider the effects of exciton thermalization on the charge generation mechanism. Our data indicates that an energy offset between the polymer exciton and the charge transfer ground state below about 0.25 eV adversely affects the cell performance, while a HOMO band offset below about 0.2minus;0.3 eV also degrades cell performance but by a different mechanism.

A major source of loss in the power conversion efficiency of organic photovoltaic cells is low open-circuit voltages (V_oc), being significantly smaller than the optical gap of the absorber material. Typical strategies to improve the cell voltage involve tuning the interfacial energy gap between donor and acceptor by choosing suitable material combinations. However, the energy loss between the resulting charge transfer state and the open-circuit voltage is almost invariantly between 0.5 and 0.6 eV. Thus, one has to explore new strategies to further reduce energy losses in OPVs.
Using sexithiophene (6T) as donor material in combination with three different acceptors (including fullerene (C₆₀), diindenoperylene (DIP), and a zinc chlorodipyrrin (ZCl)) we demonstrate a tunability of V_oc from less than 0.5 V in the C₆₀ case to almost 1.5 V for DIP. From temperature dependent electrical characteristics in combination with photocurrent and electroluminescence spectroscopy we are able to estimate the energy and coupling strength of the involved charge transfer states and their implications for V_oc. From these experiments, two different contributions are identified: For the rod-like molecules 6T and DIP there is a strong dependence of intermolecular coupling on the mutual orientation of donor and acceptor molecules at the interface, with an edge-on configuration being favourable for reducing recombination losses [1]. By contrast, in the case of ZCl as acceptor an intramolecular symmetry breaking charge transfer seems to be the dominant effect for efficient charge separation.
[1] U. Hörmann et al. (submitted)
[2] A. Bartynski et al. (submitted)

Efficient organic solar cells are based on (electron) donor-acceptor heterojunctions. An optically generated excited molecular state (exciton) is dissociated at this junction, forming a charge-transfer (CT) state in an intermediate step before the electron and hole are completely separated. The observed highly efficient dissociation of this Coulombically bound state raises the question on the dissociation mechanism.
In this work, we show that the observed high quantum yields of charge carrier generation and CT state dissociation are due to a highly delocalized and consequently weakly bound CT state. In our study we employ ternary blends which contain two small-molecule donors combined with the acceptor C60. Instead of showing distinct absorption and emission features of a molecular donor-acceptor CT state, these blends exhibit spectra which are shifted in energy dependent on the composition of the blend. Identifying a new geminate-pair loss mechanism via donor excimers, we find that the hole on the small-molecule donor is delocalized in addition to the electron on the fullerene acceptor, and effective charge separation depends on the energetic offset between excimer and CT states. Thus, the charges upon interface charge transfer and even in the case of back transfer and recombination are much more delocalized than commonly assumed. Besides the importance for solar cells, our observation is fundamentally important since it might open the path for “bandgap engineering” with continuously energy tunable organic semiconductors, similar to inorganic semiconductor heterostructures.

Organic photovoltaics (OPVs) depend on the molecular interface between electron donating molecules (D) and electron accepting molecules (A) for exciton dissociation and charge separation. The energetic landscape at and around this interface is thus expected to play a significant role in determining charge separation probabilities and OPV device performance. For example, if D or A molecules at the interface are higher energy states for holes or electrons, respectively, then this will provide an energetic driving force for charges to move away from the interface. Furthermore, the energetic landscape at the interface will set the approximate limit of the open circuit voltage (V_OC) of the OPV material system. In this work we explore how the material state (crystalline vs. amorphous) and film composition affect the energy levels of small molecule donors and electron accepting C₆₀ molecules. This is accomplished through ultraviolet photoelectron spectroscopy and external quantum efficiency measurements to probe ionization potentials (IP) and charge-transfer (CT) state energies, respectively, where CT states are intermolecular states formed between D and A molecules. These measurements show that the IPs can vary by up to 0.5 eV between molecules in pure and blend states. These shifts in IP correspond with shifts in the CT state energy, with variations of up to 0.5 eV depending on whether the device is a bilayer or if a small percentage of donor material is blended with C₆₀. These large energetic shifts between pure and mixed phases will provide a significant driving force for charge separation in bulk-heterojunction devices, where a three-phase morphology consisting of pure phases and a mixed phase has been shown to exist. In some donor material/C₆₀ bilayer devices intermixing occurs to a large extent, which results in a CT state energy that reflects the IP of the donor material in a blend film and not in the pure state. Consequently, the V_OC of these PV devices is significantly higher than what is predicted based on the IP of the pure donor material.

Advances in controlled synthesis, processing, and tuning of the properties of organic conjugated polymers and peroskites have enabled significantly enhanced performance of organic and hybrid electronic devices. Our laboratory employs a molecular engineering approach to develop processible low band-gap polymers with high charge carrier mobility for enhancing power conversion efficiency of single junction solar cells to ~11%. We have also developed several innovative strategies to modify the interface of bulk-heterojunction devices and create new device architectures to fully explore their potential for solar applications.
The performance of polymer and hybrid solar cells is strongly dependent on their efficiency in harvesting light, exciton dissociation, charge transport, and charge collection at the metal/organic/metal oxide or the metal/perovskite/metal oxide interfaces. In this talk, the integrated approach of combining material design, interface, and device engineering to significantly improve the performance of polymer and hybrid perovskite photovoltaic cells (PCE of >17%) will be discussed. Specific emphasis will be placed on the development of low band-gap polymers with reduced reorganizational energy and proper energy levels, formation of optimized morphology of bulk-heterojunction layer, and minimized interfacial energy barriers using functional surfactants and graphene oxide. At the end, several new device architectures and optical engineering strategies to make tandem cells and semitransparent solar cells will be discussed to explore the full promise of polymer and perovskite hybrid solar cells.

With to the development of new donor and acceptor materials, OPV blends are becoming capable of generating high currents. This places the understanding of charge extraction in competition to non-geminate recombination at the centre current OPV research. We recently applied different transient photocurrent techniques to study charge extraction in a variety of photovoltaic organic systems, including polymer-polymer and polymer-fullerene blend. Also, a novel approach was employed to quantify the effective mobility of charge extraction under application-relevant conditions.
We find that fill factors approaching 70 % can be realized in thick active layers provided that both types of carriers exhibit mobilities exceeding 10^-3 cm²/Vs. For a particular series of polymers blended with fullerene, the device performance is shown to be exclusively determined by the hole mobility. Here, the continuous increase of mobility upon modification of the chemical structure can be related to a preferential face-on orientation and increased pi-pi stacking intensity of the hole-conducting polymer backbones. The result shows the potential of properly-designed polymers to enable high fill factors in thick devices, as required by mass production technologies. Notably, even in these high mobility blends, the photocurrent is markedly affected by the resistivity of the layer, leaving room for further significant improvements of device performance.
[1] S. Albrecht, J.R. Tumbleston, S. Janietz, I. Dumsch, S. Allard, U. Scherf, H. Ade, D. Neher, JPCL 2014, 5, 1131
[2] W. Li, S. Albrecht, L. Yang, S. Roland, J.R. Tumbleston, T. McAfee, L. Yan, M.A. Kelly, H. Ade, D. Neher, W. You, J. Am. Chem. Soc. 2014, DOI: 10.1021/ja5067724

Polymer solar cell (PSC) technology has attracted much attention due to its promise as low-cost conversion of solar energy. Despite recent progress, several limitations are holding back PSC development. For instance, current high-efficiency (>9.0%) PSCs{Liao, 2013 #103} are restricted to materials combinations that are based on limited donor polymers and only one specific fullerene acceptor, PC₇₁BM. Furthermore, the PSC field lacks an effective approach to control the polymer:fullerene blend morphology that is critical in achieving high PSC performance. Here we show multiple cases of high-performance thick-film PSCs (efficiencies up to 10.8%, fill factors up to 77%) via the formation of a near-optimal polymer:fullerene morphology that contains highly crystalline yet reasonably small polymer domains. This morphology is controlled by the temperature-dependent aggregation behavior of the donor polymers during casting and is insensitive to the choice of fullerenes. Our comparative study show that the choice of alkyl chains is critically important in enabling optimal aggregation and morphology control. The uncovered aggregation and design rules yieldthree high-efficiency (>10%) donor polymersand will allow further synthetic advances, process optimizations, and matching of both the polymer and fullerene materials, potentially leading to significantly improved performance and increased design flexibility.

Merging inorganic and organic solar cells in a series connected hybrid multi-junction is shown to be an elegant approach to prepare efficient solar cells with a total active layer thickness well below 1 µm. Complementary absorption, high absorption coefficients, and the ease of fabrication make organic low band-gap (LBG) materials suitable for the use in multi-junction solar cells in combination with amorphous silicon (a-Si:H). Since a-Si:H is insoluble in all organic solvents, the organic top cell comprising the LBG polymer mixed with PC₆₁BM could be easily deposited from solution. Several material combinations were tested as recombination layer, with Al:ZnO or ITO combined with MoO₃ or PEDOT:PSS giving the best results. Transfer matrix based optical modeling was employed to predict the optimum layer thicknesses of each layer in the full stack and to verify the results. External quantum efficiency measurements show that all planar hybrid multijunctions are current limited by the a-Si:H middle junction, but that the red shifted absorption of the LBG donor materials increases the short circuit current by having a higher transparency on the visible part of the solar spectrum. Furthermore, light scattering by transparent, rough front contacts is used to increase the absorption and, therefore, current in the amorphous silicon sub-cells. The presented multi-junction solar cells are highly efficient, showing high open circuit voltages, suitable for water splitting, and high fill factors up to 80 %. Therefore, merging inorganic/organic sub-cells in multi-junction devices bears great potential as efficient, truly thin film solar cells.

Charge transfer (CT) states are bound combinations of an electron and a hole located on separate molecules. They play a crucial role in organic optoelectronic devices, mediating the generation of charge in organic photovoltaics (OPVs) and light emission in organic light emitting devices (OLEDs). Here, we report the first direct nanoscale imaging of charge transfer states in donor-acceptor blends. After generation of a CT state, we demonstrate that energy moves geminately over distances of 5-10nm, driven in part by energetic disorder and diffusion to lower energy sites. Our direct imaging is complemented by magnetic field studies. Exchange splitting in the CT states is a sensitive function of the electron-hole separation, demonstrating that CT states ‘breathe&’ during their diffusion, expanding before ultimately recombining. In addition, measurements of spin relaxation confirm the donor-acceptor composition dependence of nanoscale transport of CT states. Transport of CT states thus provides an alternative method for mixing singlet and triplet states in OLEDs, and drift of CT states under an electric field may explain the high charge generation yields in some OPVs.

The nanoscale crystallinity at organic interfaces influences all stages of charge carrier generation of organic photovoltaic devices and is therefore a key property for maximizing the efficiency. While x-ray diffraction is an efficient tool to identify crystalline phases in organic films, it has not yet been possible to image the local distribution of crystalline phases at the nanoscale. X-ray nanobeams, however, start to change this situation and we show that it is now possible to image organic crystalline phases with submicron resolution in real devices, e.g. below metallic contacts. Images with even higher resolution are obtained by a combination of infrared spectroscopy (IR) and scattering-type scanning near field optical microscopy (s-SNOM). Once the IR laser frequency is tuned to resonance with a molecular vibration the film, the s-SNOM technique allows to image recrystallization phenomena in highly ordered thin films with a typical resolution of 20nm, equal to the diameter of a metallic AFM tip. The recrystallization pattern revealed by the NeaSNOM setup in pentacene films is quite surprising; we observe that the pattern is rather well defined with elongated bulk phase domains nucleating within a thin film matrix [1]. We argue that thermal expansion upon sample cooling is the trigger for this regular recrystallization phenomenon, which we observe continues for months of shelf storage. Right now, we also explore molecule/fullerene films to access the nanoscale structure organic heterojunctions of highly efficient photovoltaic cells.
[1] C. Westermeier, A. Cernescu, S. Amarie, C. Liewald, F. Keilmann, B. Nickel
Sub-micron phase coexistence in small-molecule organic thin films revealed by infrared nano-imaging, Nature Communications 5, Article Number 4101, DOI: 10.1038/ncomms5101(2014)

From the starting point of organic electronics in the 1990's, steadily increasing effort is put into the improvement of organic solar cell (OSC) performance. This led to a tremendous growth rate of their photo conversion efficiency. The progress is based on new materials tailored for OSC applications as well as on advanced skills in device engineering. But many of the models and characterization techniques applied in the field of organic electronics are restricted to small clusters of molecules or model systems as individual interfaces only, thus lacking prediction when it comes to full devices. With scanning Kelvin probe microscopy (SKPM) we access the physics of entire OSC devices and bridge the gap between the molecular and the macroscopic understanding. On this road, the nature of SKPM as a surface characterization method and of OSCs as horizontally layered devices poses an experimental challenge. In our measurements we gain access to the OSC cross sections by milling trenches into the devices with a focused ion beam (FIB).
Our scanning probe station is placed within the vacuum of a scanning electron microscopy (SEM)/FIB cross beam system. We expose the OSC cross section with the FIB and place the cantilever under SEM observation right at the FIB spot. Through defined contacting and illumination we are able to investigate the potential distribution of the solar cell in vivo with SKPM.
We vary the device architecture of vacuum processed F₄ZnPc/C₆₀ OSCs in a defined manner. By means of characterization with SKPM, we link the induced changes in solar cell performance to modifications of their nanoscale electric potential distribution. We thus identify loss mechanisms and their localization within the cell. Based on our results with different preparation methods for cross section exposure, we give an overview of challenges and prospects for cross-sectional SKPM studies.

Small molecule organic solar cells often require optimization of processing conditions to induce phase separation and crystallinity in order to achieve high solar cell performance. The charge transfer state (CTS) is believed to be the state through which vast majority of charge generation and charge recombination take place in organic solar cells. It has been shown that the energy of radiative bimolecular recombination through the CTS strongly correlates with the open circuit voltage (V_OC) under illumination, marking this recombination process as more significant than just defining photocurrent losses in a solar cell. In most reported polymer:fullerene blends the effect of processing conditions on the CTS is an energetic change often observed by CTS peak emission shifts. We demonstrate that in three high-performing solution-processed small molecule organic solar cell blends, phase separation and crystallinity correlate with singlet emission in the blend EL. The as-cast devices, with poor solar cell performance and negligible phase separation, show only CTS emission. By processing with the solvent additive diiodooctane (DIO), the solar cells achieve high photoconversion efficiencies of 7%, and also show additional emission from donor exciton singlets. It has been suggested that the presence of singlet emission in blend luminescence is correlated to insufficient energetic offset to drive charge separation. In the blends studied here, however, we show singlet emission is correlated with greatly enhanced fill factors (FF) and short circuit currents (J_SC).
Because the CTS is an interfacial state that only exists by blending of two materials, its properties are often probed by means of optical transitions with the hope of understanding the processes that take place at the donor-acceptor interface.

Our group recently developed the use of high angle annular dark field (HAADF) scanning transmission electron tomography (STET) in combination with discrete area reconstruction technique (DART) for the measurement of polymer fullerene morphology. The resulting images can resolve multiple gray levels and have a voxel size of 1.4x1.4x1.4 nm, which makes them by far the most highly resolved and chemically detailed images of bulk heterojunction (BHJ) layers ever published. We apply this technique along with energy filtered STEM to the study of morphology in both PTB7 based OPV device layers and to polymer/PbS PV layers. These measurements are compared against reflectometry techniques and thin sectioned layers to verify that vertical information is accurately reconstructed in the 3D images.

In organic solar cells, excitons are dissociated by charge separation between an electron donor (e.g. conjugated polymer) and an electron acceptor (e.g. fullerene derivative). For solid-state blends of the two components, their precise arrangement into a bulk heterojunction (BHJ) plays an important role, and this microstructure was recently revealed to be very complex. Not only can the polymer and fullerene arrange into either amorphous or crystalline neat domains of variable size, but an additional intimately mixed polymer-fullerene phase has been identified. We show here that microstructure and photophysical properties in polymer:fullerene blends go hand-in-hand to determe the efficiency of organic solar cells. Ultrafast transient absorption, electro-absorption and fluorescence up-conversion spectroscopy were used to investigate the dynamics of charge separation in samples with well-characterized microstructure. Blends of the pBTTT and PBDTTPD polymers with PCBM gave us access to different scenarios, such as a single intimately mixed polymer:fullerene phase, an intermixed phase with additional pure PCBM clusters, or a three-phase microstructure of pure polymer aggregates, pure fullerene clusters and intermixed regions. Moreover, we have selectively excited either the fullerene or the polymer in order to evaluate the effect of microstructure on both the electron and hole transfer pathways. We discuss how an ideal microstructure can promote generation of free charge carriers and relate the high photovoltaic performance of the PBDTTPD blend (up to 8.5 %) to its favorable properties in terms of microstructure and exciton delocalization. Overall, the fundamental insights of our study allow precise control of the optical and electronic properties in OPV materials by targeting optimized microstructures.

The efficiency of bulk-heterojunction polymer:fullerene solar cells critically depends on the dominant length scale of the morphology formed during drying of the photoactive film. Time-resolved studies that allow studying the kinetics of film and morphology formation have been performed on a blend of a diketopyrrolopyrrole-based polymer with a fullerene derivative, deposited from solvents with and without co-solvent. Using in-situ time-resolved optical techniques, combined with quantitative analysis of ex-situ transmission electron microscopy, we are able to determine the parameters that influence the dominant length scales in the phase separated films. Without co-solvents, large fullerene domains are formed by liquid-liquid phase separation, whose size is governed by the drying rate when normalized to the final thickness. With co-solvents, on other hand, the dominant length scale is not determined by liquid-liquid demixing, but by polymer aggregation. Dominant length scales of the polymer fibers that are formed by the aggregation can be controlled by adjusting the nature of the co-solvent mixture.

The practice of combining electron rich (donor, D) and electron deficient (acceptor, A) moieties to make low-band gap co-polymers has led to dramatic efficiency gains in organic photovoltaics (OPV). The flexibility afforded by combining different D and A building blocks to make distinct co-polymers allows for the tuning of the optoelectronic properties of the active layer to improve device performance. This tuning is currently predicted to a reasonable extent using electronic structure methods based on vacuum phase calculations. However, device properties such as spectral broadening and hole mobility depend critically on morphology. Such considerations need to be addressed by methods beyond calculations on isolated molecules. One can predict the morphology of a polymer film using classical, atomistic molecular dynamics (MD) simulations and appropriate force fields. Here, we report results from large-scale (volumes greater than 1000 nm³) MD simulations for a variety of high power conversion efficiency DA co-polymers. The internal packing and relative positions of D and A moieties were established via the analysis of radial distribution functions, inter-ring dihedral distributions and computed GIWAXS spectra. The effect of polymer packing on electronic structure was also analyzed using quantum chemical methods such as time dependent density function theory in order to determine the spectral broadening that occurs once a polymer is placed within a film. In addition, generalized inter-site electronic coupling distribution functions were computed using Corresponding Orbital Transformation for each pair of sites within the film. The relationship of these calculations to measured hole mobilities in neat polymers is described. Finally, we discuss the implications of these results for understanding structure-property relations in OPV.

The addition of small quantities of polystyrene (PS), a typical insulator, was recently shown to increase the power conversion efficiencies of solution deposited molecular organic solar cells. The performance is further improved by the presence of diiodooctane (DIO), a more conventional solvent additive. How the two additives function in tandem to provide optimal enhancement was not initially understood mechanistically. Thus, we probed how the addition of PS and DIO affects film formation kinetics of the p-DTS(FBTTh₂)₂ and (6,6)-phenyl C₇₁-butyric acid methyl ester (PC₇₁BM) system, a bulk heterojunction blend film with high-performance organic photovoltaic (OPV) applications. This thin film characterization was accomplished through in-situ monitoring of absorbance, thickness and crystallinity during spin-casting - using ellipsometry, UV-Vis spectrometry and grazing-incidence wide-angle X-ray diffraction, respectively. These techniques probe the complex “wet” film environment and allow for high temporal resolution during the solution-solid transition.
In-situ measurement shows that the PS and DIO additives promote donor crystallite formation on different time scales and through different mechanisms. PS-rich films retain chlorobenzene solvent, slightly extending evaporation time and promoting phase separation earlier in the casting process. However, this extended time and chlorobenzene alone is not sufficient to attain the right morphology for optimal PCE results before dryness. Here is where the solvent additive, DIO, comes into play: its very low vapor pressure extends the time scale of film evolution for a considerably longer time and allows for crystalline rearrangement of the donor phase for more than a minute after casting. This behavior is not observed for chlorobenzene, retained by PS-rich films.

Recent improvements in small molecular-weight organic photovoltaics (OPVs) have been realized by controlling the morphology of thin-films down to the nanometer scale. Organic layer morphology affects device efficiencies, operational lifetimes, and failure mechanisms. One method to effectively control the morphology of organic layers is via growth by organic vapor phase deposition (OVPD)¹. By using an inert carrier gas, OVPD provides extra energy for organic molecules to find an equilibrium morphological configuration as they adsorb onto the substrate. Compared to conventional vacuum thermal evaporation (VTE), the use of a carrier gas in OVPD dramatically changes many aspects of the film deposition kinetics.
In this work, we demonstrate organic photovoltaic (OPV) cells based on a nanocrystalline mixed tetraphenyldibenzoperiflanthen (DBP):C₇₀ heterojunction grown by OVPD with a power conversion efficiency, PCE = 6.7%, compared to 6.2% for analogous, optimized devices grown by VTE. Due to the lower electrical resistance of the nanocrystalline layers formed via OVPD, the active region thickness of the OVPD-grown device can be significantly thicker than the one grown by VTE². We also show that morphological changes over time in the bathophenanthroline (Bphen) cathode buffer layer in these organic solar cells strongly impact device reliability, and that these changes are reduced when the underlying OPV active region is grown by OVPD as opposed to VTE. The enhanced layer stability leads to significantly improved device operational lifetime.
When Bphen is deposited onto a tetraphenyldibenzoperiflathen (DBP):C₇₀ mixed active region grown by VTE, the morphological transformation in Bphen is found to reduce the open-circuit voltage (V_OC) from 0.91 V to 0.52 V after aging at 250 hr under simulated AM 1.5G solar illumination, resulting in a 50% decrease in power conversion efficiency from PCE = 6.0% to 3.1%. The morphological degradation also results in electrical shorts across the devices—the initial yield of 2 mm² VTE-grown devices is 93% and decreases to 66% after aging. Here, the rough surface of the OVPD active layer is found to pin the morphology of Bphen, resulting in a longer device lifetime and higher yield compared with analogous VTE-grown devices. Devices with OVPD-grown active layers experience little change in preserving 80% of their initial PCE of 6.7% after 250 hr of operation while maintaining a yield of 93%.
¹ F. Yang, M. Shtein, S. R. Forrest, Nature Materials, 2004, 4, 37.
² B. Song, C. Rolin, J. D. Zimmerman, S. R. Forrest, Advanced Materials, 2014, 26, 2914.

Semitransparent organic photovoltaic cells (OPV) have attracted growing interest in recent years due to their specific applications for various (semi-)transparent architectures like windows, car roofs, greenhouses and other building integration photovoltaic (BIPV) architectures. Like all kinds of photovoltaics, a critical step towards commercialization of the OPV technology is the manufacturing of large-area modules with minimal efficiency losses compared to the small-size devices.
Another important issue that has to be addressed is the realization of solution-processability by which the cost potential of this promising solar technology can be realized. We choose silver nanowires (AgNWs) as material for transparent electrodes, due to their excellent optoelectronic properties and their compatibility with large-area coating techniques such as slot-die coating and spray coating. Highly transparent lab-scale solar cells with AgNWs as both bottom and top electrode have been fabricated entirely from solution. In order to give a rationale for electrode design, we will also discuss quantitatively the interdependence between transparency and efficiency for semi-transparent OPV, with respect to the choice of electrode materials and the absorption spectrum of the active layer.
The right combination of innovative electrodes and smart processing technologies demonstrates a viable path towards high efficient module technology. We will show highly efficient solution-processed semitransparent polymer solar cell modules (ST-modules) based on ultra-fast laser patterning. The use of AgNWs as top electrodes enables the whole module stack to be processed in ambient air. The efficient ohmic contact of the top AgNWs with the bottom ITO electrodes and high accuracy of the laser positioning enable the as-fabricated modules to show high electrical fill factor (FF) of 63% and extremely high geometric FF of >95%, respectively. The resulting semitransparent modules with active areas of 6400 mm² showed a PCE value of ~3.0% which is ~85% of the single cell reference (10 mm²).

Recently all-carbon based solar cells have attracted attention as potential candidates for innovative photovoltaic devices. Carbon-based materials such as graphene, carbon nanotubes (CNT) and amorphous carbon (a-C) have the potential to present physical properties comparable to those of silicon-based materials with advantages such as low cost, solution processing, air stability, and higher thermal stability. In particular a-C structures are promising systems in which both sp² and sp³ hybridization coordination are present in different proportions depending on the specific density, providing the possibility of tuning their optoelectronic properties and achieving comparable sunlight absorption to amorphous silicon.
In this work we employ accurate computational approaches to design suitable device architectures, such as bulk heterojunctions (BHJ) or p-n junctions, consisting of a-C as the active layer material. These structures must enable successful electron and hole extraction as well as reduced sources for carrier recombination in order to achieve large currents and voltages. Regarding a BHJ construction, we carry out ab initio molecular dynamics and density functional theory calculations for a large statistical set of interfaces between a-C structures, with different densities, and C nanostructures (such as CNT and fullerene) to relate the optoelectronic properties of the interface to the stoichiometry of a-C. We demonstrate that the energy alignment between the a-C mobility edges and the occupied and unoccupied states of the CNT or C₆₀ can be widely tuned by varying the a-C density to obtain a type II interface, a fundamental prerequisite for charge transfer mechanism. In order to employ a-C materials in p-n junctions we analyze with the same level of accuracy the p-type and n-type doping of a-C focusing mainly on an evaluation of the Fermi level and work function dependence on doping. Our results highlight promising features of a-C as the active layer material of thin-film solar cells.

Hybrid organic/silicon heterostructures have become of great interest for photovoltaic application due to their promising features (e.g. easy fabrication in a low-temperature process) for cost-effective photovoltaics. These hybrid devices aim at combining the advantages of the well-developed crystalline silicon solar technology and the low-cost processable organic semiconductors.
This work is focussed on solar cells with a hybrid heterojunction between the polymer poly(3-hexylthiophene-2,5-diyl) (P3HT) and n-doped monocrystalline silicon. P3HT was spin-cast from a toluene solution at ambient environment onto a prestructured and cleaned silicon wafer. As semi-transparent top contact, a thin (15 nm) Au layer was deposited on top of the P3HT by vacuum evaporation and for the backside contact, we used a thick vacuum-deposited Al layer. The processed devices had an active area of 5 x 5 mm².
Devices with different P3HT thicknesses were processed by varying the solution concentration and the rotation speed. As a reference, we further prepared an Au/n-Si Schottky diode solar cell without P3HT. The devices were characterized by AFM, wavelength-dependent photocurrent (J_SC) measurements and illuminated current density-voltage (J-V) measurements. In addition, P3HT layers with different thicknesses were deposited onto glass substrates for transmission measurements.
The J-V measurements of the hybrid devices show a significant increase in open-circuit voltage (V_OC) compared to the Schottky diode reference from 0.29 V up to 0.53 V for the best performing hybrid devices. The increased V_OC indicates that P3HT functions as efficient electron blocking layer reducing the reverse electron current into the anode and thereby increasing V_OC.
In contrast, by inserting the P3HT layer the short-circuit current density J_SC did not significantly change. This correlates with the wavelength-dependent J_SC measurements which do not show any contribution of P3HT to the J_SC of the hybrid solar cells. For devices with a thick P3HT layer, we could even see a decreased photocurrent in the wavelength range of P3HT absorption at 400 - 650 nm. Therefore, the hybrid P3HT/Si interface seems to be not effective in separating excitons generated in P3HT, although the energy difference ΔE asymp; 1 eV between the LUMO of P3HT and the conduction band of silicon is larger than the exciton binding energy E_Ex of 0.6 - 0.7 eV in P3HT. Different possible mechanisms for the hindered charge generation in P3HT (e.g. interface dipole, hole trapping in P3HT) are discussed.
For the best performing hybrid P3HT/silicon solar cells, we could achieve power conversion efficiencies (PCE) (AM1.5 illumination) up to 6.5% with a V_OC of 0.52 V, a J_SC of 18.6 mA/cm² and a fill factor (FF) of 67%. This is more than twice the efficiency of the reference Schottky diode with a PCE of 3% and opens up the way for even higher values with the P3HT film contributing to the photocurrent of the hybrid solar cell.

Highly efficient organic devices based on the blend of an electron-donating and an electron-accepting material require fine control of the phase separation in order to improve the charge transport by creating electron acceptor-donor interfaces in the nano-scale. Aiming for the control of the film nanomorphology across large areas devices and toxic solvent free inks, we present organic photovoltaic devices (PVDs) based on polymer nanoparticles synthesized in aqueous media through the miniemulsion process [1]. A new bulk heterojunction blend composed of organic semiconducting nanoparticles containing the polymer poly[{2,6-(4,8-didodecylbenzo[1,2-b:4,5-b&’]dithiophene)}-alt-{5,5-(2,5-bis(2-butyloctyl)-3,6-dithiophen-2-yl- 2,5dihydropyrrolo[3,4-c] 57pyrrole-1,4-dione)}] (P(TBT-DPP)) and the fullerene indene-C60-bisadduct (ICBA) was successfully miniemulsified in water. We have prepared dispersions with several concentrations of the P(TBT-DPP):ICBA blend ratios (1:1, 3:7, 1:3) and tested as printed active layer in the PVDs. The nanoparticle aqueous dispersions were doctor blade coated in air onto PEN/PEDOT:PSS substrates whereas the PEDOT:PSS layer was also printed using the blade coating technique. Then a layer of C₆₀ (40 nm) and Al (100 nm) were thermally evaporated. The nanoparticle devices have shown improved performance when compared to the conventional P(TBT-DPP):ICBA blends blade coated in the same conditions. The nanoparticle devices exhibited higher short current density, higher fill factor and lower series resistance. The best blend ratio was found to be 3:7 and optimized devices exhibit power conversion efficiencies up to 2.63% (AM1.5). As indicated by the film morphology probed by Atomic Force Microscopy, this method has shown to be an efficient approach to impose a fine fixed length scale of phase separation and improved charge transport. In the conventional BHJ blends the phase separation is in the order of 500 nm, whereas in the nanoparticle films it is in the range of 80 - 90 nm which is the average size of the nanoparticles in the dispersions. The electrical transport properties of the nanoparticle films are fully described by a phenomenological model that relates intrinsic film morphology to the photovoltaic response.
[1] K. Landfester, R. Montenegro, U. Scherf, R. Güntner, U. Asawapirom, S. Patil, D. Neher, T. Kietzke, Adv. Mater. 2002, 14, 651.

Nowadays, the growing demand for renewable and sustainable resources has triggered the researches in advanced materials for the solar cell application.[1] Among them, metal oxide nanomaterials (such as ZnO, TiO_X) are widely utilized for the fabrication of inverted organic solar cells (OSCs) which have the merits of good stability and high efficiencies. In the inverted OSCs, buffer layers between the organic active layer and electrodes usually use metal oxide nanomaterials to improve the device performance.[2] The metal oxides influence device stability as an oxygen scavenger that can protect organic active layer against a degradation in air. To date, most buffer layers have to be fabricated by vacuum-evaporation or high-temperature treatment (>200 °C), and their properties can hardly be tuned to match the change in active materials for better performance OSCs.
Herein, we design and fabricate high efficiency and stable inverted OSCs by employing solution processed Zn_1-xMg_xO (ZMO) films with high transparency and tunable bandgaps as a novel class of cathode buffer layers (CBLs).[3] By incorporating ternary ZMO films as the CBLs, and combining PC₇₁BM with a low-bandgap polymer, PTB7 in the active layer, inverted OSCs exhibit PCEs up to 7.83% (8.31-8.35% by a further optimization), much better than those of control devices without a CBL (PCE=3.48%) or with a ZnO CBL (PCE=7.11%) under the same conditions. A high efficiency of 9.39% was also achieved for the inverted PSCs based on another low-bandgap polymer PTB7-Th. These new ZMO CBLs are highly transparent, and their bandgaps and energy levels can be refinedly tailored by the Mg doping, thus allowing us to improve the cell performance by optimizing the interfacial properties. This new type of CBLs can also be extended to other solar cell systems. The results open an efficient and wet-chemical strategy for the design and fabrication of high-performance BHJ solar cells by incorporating multielement semiconductors with variable bandgaps as novel CBLs.
References
1. Z.G. Yin, Q.D. Zheng, Adv. Energy Mater. 2012, 2, 179.
2. J. Huang, Z.G. Yin, Q.D. Zheng, Energy Environ. Sci. 2011, 4, 3861.
3. Z.G. Yin, Q.D. Zheng, S-C. Chen, D.D. Cai, L.Y. Zhou, J. Zhang, Adv. Energy Mater. 2014 ,4, 1301404.

A tandem solar cell based on a combination of an amorphous silicon (a-Si) and polymer solar cell (PSC) has been demonstrated for the first time. When connected in series, the open-circuit voltages of these two sub-cells add perfectly and a 20 % higher efficiency is achieved when compared to a single a-Si cell. Of greater interest, however, is the fact that connecting in parallel provides a 60 % greater efficiency, with near perfect addition of the current densities. Since these tandem devices can be readily fabricated by low-cost methods, they require only a minor increase in the total manufacturing cost. Furthermore, the absorption spectrum of the device can be easily tuned by altering the conjugated polymer used for the PSC back cell. A combination of a-Si and PSC therefore clearly provides a compelling solution to reducing the cost of electricity produced by photovoltaics.

Based on our experimental and numerical simulation results, we present and discuss the influence of gold, silica coated and silver enhanced gold nanostars on the optical properties and on the performance of organic devices, such as organic solar cells and organic light-emitting diodes (OLEDs).
Despite the advances in the field of organic light emitting devices and solar cells, the realization of high performance organic devices presents a major challenge. Here, we present promising possibilities to increase the spontaneous emission rate in organic semiconductors. For this propose, we designed and synthesized plasmonic nanostars coated with a very thin silica layer, so that the hot-spots at the tips of the nanostars remain still easily accessible. In addition, we can fine-tune the spectral position of the plasmon resonances of the nanostars in specific spectral regions for the optimal overlap with the emission of different organic semiconductors. An optimal overlap in the green spectral region is achieved by silver-enhancement of the nanostars, while for the near-infrared spectral region the overlap is achieved by controlling the size and sharpness of the gold nanostar tips.
We demonstrate that the emission of organic semiconductors solutions and thin films can be considerable enhanced upon addition of the spectrally matching plasmonic nanostars. Furthermore, we successfully incorporated the nanostars in fully working OLEDs. Compared to the conventional OLED, the plasmonic OLEDs show both: significantly enhanced electroluminescence as well as substantially enhanced photoluminescence upon external optical excitation.
Our experimental and numerical simulation results allow us to discuss beneficial and limiting factors for the optical properties and for the performance of the plasmonic OLEDs and organic solar cells.

Plasmonic metal nanostructures are widely used to improve efficiency of organic photovoltaics by optical field enhancement in thin polymer films [1].
In this work we demonstrate for the first time a novel coupling mechanism between photoinduced polaronic transitions in conjugated polymer P3HT and particle-plasmon nanoantennas resonant in the infrared spectral region. This method is completely different from commonly used surface plasmon nanostructures resonantly enhancing excitonic transitions in the spectral region above the energy gap of the polymer [2].
Experimentally, large area nanoantenna samples were fabricated by hole-mask colloidal nanolithography using a tilted-angle-rotation evaporation technique covered by a thin P3HT film. Polarized steady-state photoinduced absorption measurements (PIA) of the hybrid polymer-nanoantenna system clearly indicate near-field interaction between charged polarons in the conducting polymer and localized particle-plasmons. Moreover, coupling between these two quasi-particles is demonstrated by enhancement of both polaron transitions and Infra-Red Active Vibrational Modes (IRAV) of the polymer, which directly gauge the density of photogenerated charge carriers. This new concept may be used to enhance power conversion efficiency of conventional polymer photovoltaic cells by recovering thermal photon energy, facilitating polaron formation and exciton dissociation.
[1] Atwater, H. A. & Polman, A. Plasmonics for improved photovoltaic devices. Nat. Mater. 9, 205-213 (2010).
[2] Gan, Q., Bartoli, F. J. & Kafafi, Z. H. Plasmonic-Enhanced Organic Photovoltaics: Breaking the 10% Efficiency Barrier. Adv. Mater. 25, 2385-2396 (2013).

Scarcity of the fossil fuels and repetitive price spikes of them have driven the communities to seek alternative and rapidly renewable energy sources. Photovoltaic technology, which utilizes sunlight to generate clean energy, is one of the promising solutions to the above challenge. Polymer based organic solar cells (OSC) has become a competitive material over silicon based solar cell. The ability to OSC as a surface coating form having thickness less than 1mm over any substrate regardless the shape to cover the large surface area is an example of its novelty. Such materials have ability to convert any physical object into an active light harvesting solar cell devise. However, the challenges are remain in expanding OSC technology due to its low efficiency compared to silicon base solar panels and comparably high cost of materials. Even though the cost of the OSCs is comparably lower than the Si based devises, it can be further reduced by preparing polythiophenes from cheap raw materials. The current study reports potential synthetic and technological approaches to generate thiophene polymers from the plant extract of Tageteserecta for the first time. Thiophenes and their derivatives are also a family of compounds in the defense system of Tagetes plants to repel herbivores. The concentrations of thiophene level in plant tissues increase as the plants get older, reaching a maximum during the reproductive stages. A number of thiophenes derivatives were reported and identified in the roots of Tagetes species including 5-(3-buten-1-ynyl)-2,2-bithienyl, 5-(4-hydroxy-1-butynyl)-2,2-bithienyl, 5-(4-acetoxy-1-butynyl)-2,2-bithienyl and 2,2:5,2-terthienyl. Out of the above thiophene derivatives 2,2:5,2-terthieny, which is also known as α-T, consist of three fused rings of thiophene monomers in regioregular manner. During the current study these thiophene based compounds were extracted from the dry roots of the Tagetes plants and fully characterized using spectroscopic and chromatographic techniques. Then the extracted dimer and trimer units of thiophenes were partially purified. These thiophenes mixture with dimmers and trimers of thiophenes were then selectively polymerized using cationic polymerization technique. The final products were mainly a polymerized product of α-T. The final polythiophene obtained from α-T were characterized for their ability for the light harvesting properties. The degree of polymerization as well as the length of the conjugated system was able to control by controlling the polymerization reaction conditions.

Thermomechanical degradation of inverted polymer solar cells is of particular interest for operational reliability, particularly in light of weak internal interfaces and mechanically fragile layers. We have recently shown how annealing temperature and time increases the adhesive properties at the weak interface between P3HT:PCBM and PEDOT:PSS. However, the role of thermal cycling, mechanical fatigue and their synergistic effect remains unclear, which is important to better understand real-world exposure conditions.
We subjected representative polymer solar cells to temperature cycles of - 40C to 85C. We monitored the adhesion energy using a thin-film delamination technique. Surprisingly, we found that the solar cells are not weakened after thermal cycling. However, surface analysis on the delaminated fractured samples using XPS and AFM revealed that the P3HT:PCBM and PEDOT:PSS layers became more susceptible to cohesive as opposed to adhesive failure observed in as processed devices. We found that for a low number of thermal cycles the fracture path meandered cohesively through the P3HT:PCBM and PEDOT:PSS layers, but then failed cohesively within the P3HT:PCBM layer after ~50 thermal cycles. We used a kinetic analysis technique to establish a debond model for the effect of cyclic temperature variation on the delamination of polymer cells. In addition, we performed mechanical fatigue measurements on the thermal cycled samples to explore the synergistic effects of mechanical and thermal degradation. Understanding the degradation mechanisms in operational conditions is essential to improve reliability and mechanical integrity in future designs.

Hybrid organic inorganic photovoltaics have proven to be a promising setting for metal nanowire network transparent electrodes due to facile solution processing already necessary to deposit the organic layer[1]. In particular, silver nanowires integrated in PEDOT:PSS/n-Si heterojunction solar cells have produced published power conversion efficiency of over 10%[2]. Optical processing of these nanowire network electrodes increases the contact at narrow nanowire-nanowire junctions, thereby lowing the in-plane or sheet resistance of the electrode [3]. We will demonstrate that this optical processing can be performed on top of sensitive organic thin films by using a pulsed laser near a plasmonic resonance in the silver nanowires. This resonant process allows for very low doses of energy to effectively interact only with the nanowire network layer of the device. The laser processed devices show improvements to short circuit current and to fill factor, leading to increased power conversion efficiency. We find that the benefits of such a laser processing technique, while dramatic (as much as a 50% increase in power conversion efficiency) scales inversely with the initial nanowire density in the electrode. This laser process is energy tunable, as such the same laser can be used to scribe or pattern the devices. Additionally, considerations must be made in deposition technique, and in the silver nanowire ink preparation to ensure the device integrity. We will report devices made with power conversion efficiency up to 13% with a solution processed heterojunction and top electrode, eliminating the need to vacuum deposit metal finger electrodes.
[1] Nagamatsu, K. A., Member, S., Avasthi, S., Jhaveri, J., Member, S., & Sturm, J. C. (2014). A 12 % Efficient Silicon / PEDOT#8239;: PSS Heterojunction Solar Cell Fabricated at < 100 #9702; C, 4(1), 260-264.
[2] Chen, T.-G., Huang, B.-Y., Liu, H.-W., Huang, Y.-Y., Pan, H.-T., Meng, H.-F., & Yu, P. (2012). Flexible silver nanowire meshes for high-efficiency microtextured organic-silicon hybrid photovoltaics. ACS Applied Materials & Interfaces, 4(12), 6857-64.
[2] Spechler, J. A., & Arnold, C. B. (2012). Direct-write pulsed laser processed silver nanowire networks for transparent conducting electrodes. Applied Physics A, 108(1), 25-28.

In this work, the DC sputter deposition of Molybdenum oxide thin-films at room temperature has been investigated by systematically varying oxygen partial pressures (1.2x10^-3 mbar, 2.3x10^-3 mbar and 3 x10^-3 mbar) and sputtering powers (100W, 150W, 200W and 250W). The films were deposited on BK7 glass for analyzing the optical transmittance and surface roughness as a function of deposition parameters. The surface roughness does not change substantially with oxygen partial pressure, indicating a morphology that is independent of this parameter. The surface roughness was also investigated on commercial ITO substrates, which served as basis for resistivity measurements using silver top contacts. The films as-deposited with 1.2x10^-3 mbar oxygen partial pressure at 250W sputtering power were the ones resulting in the lowest resistivity of around 4x10⁴ ohm^.cm, indicating a semiconducting behaviour of the films, although the O/Mo ratio is low. As the oxygen concentration increases up to 3 x10^-3 mbar, the resistivity of the films enhances by about 4 orders of magnitude.
In order to extract the stoichiometry of the films, the samples were prepared on Si substrates followed by deposition of a gold layer, which was used as a capping layer against surface oxidation and as a calibration film for Rutherford backscattering spectroscopy. Although the changes in the oxygen content between the samples deposited with a low amount of oxygen in the chamber and the samples deposited with a higher oxygen partial pressure is only minor, the transmittance spectra show large differences in optical transparency (up to around 80% in difference between 300-900 nm). Since the microstructure of the different films, investigated here via high resolution transmission electron microscopy (HRTEM), in all cases is consistent with amorphous films, the dramatic changes in transmittance and conductivity observed here are not ascribed to different crystal phases in the films, but rather to intrinsic defects, which dominate the optical and electrical properties of these amorphous films. This work therefore points on an interesting possibility to modify the optical and electrical properties of MoOx films without employing the normally used post-annealing processes. It therefore presents a viable approach for using sputtered Molybdenum oxide thin-films in organic electronic and optoelectronic devices, where Molybdenum oxide is widely used as an interfacial layer.

Organic solar cells (OSC&’s) have attracted much attention in the past years due to their low costs, light weight and mechanical flexibility. A promising method for improving the power conversion efficiencies of the devices is by incorporating structured electrodes in the solar cell architecture. That way light absorption in the active layers of the devices can be improved. A cheap and large-scale production compatible method for structuring the electrodes in OSC&’s is by the use of Anodic Alumina Oxide (AAO) membranes. Here, nano-scale pores of controlled dimensions are formed through anodic oxidation of sputter deposited high purity Al films. The Al deposition conditions are controlled in order to modify the roughness and the grain size of the Al layers, as those parameters critically affect the subsequent pore formation during the anodization process. The anodization of the Al layers occurs in an electrochemical cell in H₂SO₄, H₂C₂O₄ and H₃PO₄ solutions, in order to tune the AAO pore diameter and interpore distance.
Following anodization, the fabricated AAO is selectively etched away in H₂CrO₄/H₃PO₄ mixtures, in order to reveal the underlying Al nanoscale dimples, which are present at the bottom of the pores. The light-trapping properties of these dimples are investigated as a function of their dimensions and ordering. The optical properties of the dimples are characterized mainly via reflection measurements, supported by laser ablation based measurements of the field enhancement [1]. The experiments are compared to FDTD calculations to further explain the mechanisms of light-trapping in these structures. The nanoscale dimples as light-trapping nanostructure are integrated into inverted P3HT/PCBM organic solar cell devices. The impact from the different dimple structures on the resulting power conversion efficiency of the devices is investigated and optimized for obtaining the highest device performance.
[1] Fiutowski J., Maibohm C., Kjelstrup-Hansen J., and Rubahn H.-G. Appl. Phys. Lett.98, 193117 (2011)

ABSTRACT: Organic photovoltaic (OPV) devices have attracted worldwide attention over the past few decades due to their inherent advantages including flexibility, lightweight, large-scale printability using roll-to-roll processing techniques, and potential utility in a wide variety of products.[1] Progress in developing fully printed solar cells using blends of commercially available active materials, such as poly(3-hexylthiophene) and phenyl-C₆₁-butyric acid, deposited on patterned indium/tin oxide-coated polyethylene terephthalate (ITO-PET) substrates has already been reported.[2] Our research has focused on the utilisation of various deposition techniques such as gravure, reverse gravure, slot die and screen-printing methods for the continuous fabrication of fully printed OPV modules. Module performance is highly susceptible to degradation on exposure to atmospheric moisture and oxygen and so a major challenge is the preparation of modules having adequately long operating times for their intended end-use. In order to realize this objective we are investigating various pre-commercial barrier encapsulation films, adhesives and edge-sealing materials to develop new encapsulation protocols which inhibit the ingress of moisture and oxygen. In our work we have found that pre-entrained moisture in the encapsulation materials, and post-encapsulation ingress of moisture/oxygen through adhesive layers and around electrical contacts are significant lifetime-limiting factors. By exploring a range of encapsulation materials and using new encapsulation techniques we have prepared 100 cm² fully printed OPV modules that exhibit a shelf-life of more than 1 year under ambient conditions and which display good durability under the outdoor conditions. Progress in this research will be reported.
Key words: Flexible photovoltaic modules, Organic photovoltaics, Lifetime, Encapsulation, Barrier materials, Stability
References:
[1] R. Po, A. Bernardi, A. Calabrese, C. Carbonera, G. Corso, A. Pellegrino, Energy & Environmental Science 2014, 7, 925.
[2] J. Yang, D. Vak, N. Clark, J. Subbiah, W. W. H. Wong, D. J. Jones, S. E. Watkins, G. Wilson, Solar Energy Materials and Solar Cells 2013, 109, 47.

Organic light-emitting diodes (OLEDs) for lighting and display, and organic photovoltaic (OPV) devices are all extremely sensitive to ambient moisture, and therefore have to be encapsulated. Large-area organic light-emitting diodes (OLEDs) for general lighting are the most sensitive to ambient degradation. Water ingress into the OLED stack leads to a local oxidation of the cathode, resulting in the formation non-emissive regions called black spots. A single black spot of sup3; 100 µm in diameter is visible by the naked eye, and the OLED is considered to be a reject, even when the black spot area is insignificant to the total device area. Logically, when a similar amount of degradation would occur in an organic solar cell, where the degradation is not determined by visibility but solely by a decreasing efficiency, the decreasing efficiency would be below detection level. Hence, a thin film encapsulation suitable for OLEDs would likely be suitable for organic solar cells too.
Here, we will show the progress in the thin film encapsulation of OLEDs on foil, which are black spot free for at least 2500h at 60 °C and 90 % relative humidity (accelerated climate conditions). We have applied the same encapsulation method to flexible organic solar cells, leading to stable organic solar cells with less than 5 % relative decrease of initial efficiency (T95) for more than 7000h at 85 °C/85 % rel. hum.
However, depending on the application in mind, different thin film encapsulation routes can be chosen. Each encapsulation process has different obstacles that need to be overcome before a production of flexible devices can be realized. For example, for large-area OPV, encapsulation by lamination of roll-to-roll (R2R) produced barrier foils is often preferred. Here, special attention to the edges and side leakage is required when the R2R encapsulated panels are cut to discrete products. We will demonstrate that side leakage can be reduced such that laminated barriers on top of OPV panels will have a sufficient life time of more than 10 years. Furthermore, we will show that the current state of the art R2R produced barrier foils of Holst Centre have a water vapour transmission rate (WVTR) of 10^-6 g/m²middot;day with the most basic barrier present, and the performance can subsequently be increased when required by specific applications.

Ongoing advances in stretchable electronics necessitate the development of electronic materials that not only meet the performance expectations of conventional materials for rigid substrates, but also are capable of maintaining stable performance under high mechanical deformation. Transparent electrodes (TEs) as essential constituents of solar cells and light-emitting diodes need to meet stringent performance metrics in terms of sheet resistance and optical transmittance. Despite their exemplary high transmittance and low sheet resistance, commercially used indium tin oxide (ITO) TEs fall short of these expectations, due to increasing costs of materials and fabrication processes as well as brittleness and lack of tolerance for mechanical deformations. Alternative materials introduced to replace ITO, such as nanotubes and metallic nanowire meshes, provide limited stretchability for meeting the sheet resistance and transparency requirements.
This work presents fibrous based transparent electrodes that provide superior performance to ITO and provide a high level of flexibility and stretchability. Nanofiber structures are known for their flexible nature due to the small diameter of the fibers. In this work composites of nanofibers and metal oxide materials are presented as platform for transparent electrodes. Integration of these composite TE as bottom and top transparent electrodes for organic solar cells are demonstrated. Both spin-coated and sprayed organic solar cells are shown based on these nanofiber TEs to demonstrate scalability of the proposed TEs.

Organic solar cells (OSCs) have attracted considerable interest in the last years due to their potential application in low-cost and large-area production. The typical structure of a BHJ-SC consists on an active layer that is a blend of p and n materials, sandwiched between two electrodes, a transparent conductive oxide (TCO) and an evaporated metal. The TCO mainly used is indium tin oxide (ITO) which unfortunately presents serious issues related to i) the release of oxygen and indium into the organic layer, ii) the poor transparency in the blue region, iii) its stiffness, which prevents its use in flexible solar cells, and iv) the large cost due to the limited supply of indium. Several candidates have been reported to replace ITO, such as conductive films based on carbon nanotubes (CNT) or graphene. The limited thermal and chemical stability and the high surface roughness of CNT films, make this option less desirable. In this context the use of graphene as semitransparent electrode has been proposed and photovoltaic devices have been already demonstrated also for flexible substrates. In this presentation we propose an approach in which the graphene electrode is 3D structured and acts not only as a contact, but also as photonic crystal. We simulated several solar cell architectures considering different materials, such as PEDOT:PSS or MoO₃ as hole transporting layer (HTL) and P3HT:PCBM or PTB7:PC70BM as active layer.
The study has been performed using an electromagnetic simulator permitting the calculation of absorbed photons in the active layer for any geometrical configuration. We optimized the grating structure demonstrating an enhancement of the optical absorption up to 27% respect to a flat solar cell with same active layer thickness.
We report also the fabrication of organic solar cells with a multi-layer graphene as semitransparent electrode. The multilayer graphene was grown by rapid thermal annealing on nickel films using a solid carbon source and transferred to glass substrates using a wet chemical transfer method. The obtained devices have been characterized in terms of optical and electrical characteristics reaching a preliminary maximum efficiency of 1,5% with two different HTL (PEDOT:PSS and MoO3) and P3HT:PCBM as active layer. This efficiency, which is strongly affected by the quality of the graphene contact that reached a minimum sheet resistance of 800Omega;/#9633;, is about half of the efficiency achieved with the standard TCO based structure (~3%).

Solution-processed thin film solar cells, such as organic photovoltaic solar cells (OPV), are made by several methods such as spin coating, spray coating, etc. The processing parameters greatly affect the stability, wetting/dewetting, morphology, shunt and series resistance and field factor and therefore the device performance. Behavior of thin films and ultrathin films used in PV solar cells is different from thick films. By definition, if the thickness of a liquid film is well below the capillary length, where the capillary (surface tension) and intermolecular forces (van der Walls and electrostatic forces) are much larger than the gravitational force, and the film is called a thin film. A micron-sized or nanometer-sized liquid film satisfies the definition of a thin film. An important issue in films is the ability of a liquid to wet its substrate. Once a liquid film is formed by a dynamic and non-equilibrium process the liquid film may remain stable, or may be unstable or metastable leading to dewetting or film rupture. The stability of a thin film depends on the film thickness, substrate thickness and surface energy. Therefore, in order to ensure that stable coating is formed, the film and surface properties must be fine-tuned and controlled. Formation of solar cell films with minimum defects is a critical step in fabrication of thin film solar cells. Defects reduce the shunt resistance and the field factor.
Three modes of instability in a thin and ultrathin film may be encountered: spinodal dewetting, which occurs in unstable films, due to disturbances in the film and growth of waves, heterogeneous nucleation and dewetting that may occur in a metastable film if nucleation sites, defects and contamination exist so as to initiate dewetting, and dewetting due to thermal nucleation, which is a special case that may occur if the interface effective potential is zero. At above a critical or transition film thickness, which is in the range of 5 nm to 50 nm, films are usually metastable and film instability occurs due to heterogeneous dewetting. Given that solar cell films have a thickness in or above this range, the films may be deteriorated by this instability mechanism. Heterogeneous dewetting can be avoided by eliminating sources of nucleation in the film. This work is applied and is dedicated to the characteristics of spin-coated and spray-coated PEDOT:PSS hole blocking polymer thin films on glass substrates. Film wetting/dewetting and rupture and film roughness are analyzed using AFM, SEM and Confocal Laser Microscopy. Conditions leading to defects are identified and discussed. Also, due to the emergence of paper solar cells, wetting characteristics of a dye solution sprayed on regular and glossy papers are performed. Optical microscopy images reveal interesting results and patterns regarding the effect of various parameters such as substrate speed, temperature, and porosity as well as spray passes on wetting and coating quality.

A novel method of precisely constructing stable and controllable conjugated polymer (CP)/fullerene nanostructures is presented. By building in non-covalent interactions between CP nanofibers (NFs) and fullerene derivatives, supramolecular polymer/fullerene composite NFs are obtained in solution for the first time. Specifically, a conjugated block copolymer having poly(3-hexylthiophene) (P3HT) backbone selectively functionalized with polar isoorotic acid (IOA) moieties, P1, is used as the building block. Self-assembly of P1 in mixed solvents leads to well-defined NFs decorated with IOA groups on the periphery, onto which phenyl-C61-butyric acid methyl ester (PCBM) and modified PCBM molecules are subsequently attached non-covalently via complementary hydrogen bonding interactions. Formation of such complex structures are studied in detail and confirmed by NMR, UV-Vis absorption spectroscopy, transmission electron microscopy (TEM), atomic force microscopy (AFM), and X-ray scattering measurements. Thin films of such composite NFs displayed controlled morphologies at both macroscopic (long-range ordering) and microscopic (polymer packing orientation) levels. Application of these composite NFs in organic photovoltaic (OPV) devices leads to not only superior performance but also much improved thermal stability, when compared with conventional bulk heterojunction (BHJ) devices.
References
(1) “Stable and Controllable Polymer/Fullerene Composite Nanofibers through Cooperative Noncovalent Interactions for Organic Photovoltaics.” Li, F.; Yager, K. G.; Dawson, N. M.; Jiang, Y.-B.; Malloy, K. J.; Qin, Y. Chem. Mater. 2014, 26, 3747.
(2) “Complementary Hydrogen Bonding and Block Copolymer Self-Assembly in Cooperation toward Stable Solar Cells with Tunable Morphologies.” Li, F.; Yager, K. G.; Dawson, N. M.; Yang, J.; Malloy, K. J.; Qin, Y. Macromolecules 2013, 46, 9021.

The ability to accurately capture the colour quality of images independent of the illuminant is critical for many applications including robotics and machine vision. It has been theoretically shown that illuminant independent detection is possible if the photodetector is able to achieve spectral responses with full width-half-maxima (FWHM) of less than 100 nm.[1] Current photodetectors use a broadband photodetector in combination with colour filters, which generally do not meet the aforementioned criteria and furthermore the use of filters introduces additional complexity and cost.
Organic photodiodes typically comprise a thin (asymp;100 nm) photo-active layer with strong optical absorption over the appropriate spectral range, sandwiched between two electrodes with electron and hole injection layers. Such devices are low finesse electro-optical cavities, and hence their photoresponse is not only related to the absorption of the photo-active layer, but also strongly governed by the optical field distribution within the cavity.
In this presentation we will discuss how we combined a narrow-absorbing chromophore and optical cavity tailoring to achieve very narrow FWHM responses.[2] In particular we demonstrate how the photoresponse of a particular device design can be predicted through optical modeling prior to fabrication, thus enabling rapid optimization of the photodiode structure.
[1] R.D. Jansen-van Vuuren, A. Pivrikas, A.K. Pandey, P.L. Burn, Colour selective organic photodetectors utilizing ketocyanine-cored dendrimers, Journal of Materials Chemistry C, 1 (2013) 3532-3543.
[2] D.M. Lyons, A. Armin, M. Stolterfoht, R.C.R. Nagiri, R.D. Jansen-van Vuuren, B.N. Pal, P.L. Burn, S.-C. Lo, P. Meredith, Narrow band green organic photodiodes for imaging, Organic Electronics, 15 (2014) 2903-2911.

Silicon photo-junctions are the mainstay of current visible light imaging and photodetector technologies. Silicon photodiodes possess large linear dynamic ranges and detectivities and high speeds. However, a band gap of 1.1 eV means that they exhibit an unwanted infrared response, which must be optically supressed, thus reducing image quality and increasing architectural complexity. This has generated significant efforts to find IR-blind alternatives [1] with junction materials and architectures that can be simply and cheaply processed. An ideal substitute for silicon as the active junction in visible light photodiodes has not to date been identified.
In our work, wavelength selective light detection has been achieved with thick junction organic photodiodes (OPDs). By drawing upon detailed knowledge [2] of the electro-optical properties of the photoactive materials and structure of the junction, we can tune the spectral response of the photodiodes. IR-blind broadband visible OPDs have been created with performance metrics similar to silicon in small and large (2500 mm²) area devices on flexible substrates. Such devices can be pixelated in a CMOS configuration for low-noise image sensing [3]. Based upon the same strategy, we have also demonstrated the first sub-100 nm Full-Width-at-Half-Maximum visible-blind red and NIR photodetectors with state-of-the-art performance across the relevant figures of merit. Paradoxically, we used broadband absorbing organic semiconductors and utilize the electro-optical properties of the junction to create the narrowest band photoresponses to date [4]. The design concept allows for response tuning and is generic for other spectral windows.
In our paper we will present the strategies used to develop and performance of our narrow and broadband OPDs. We will also discuss how these strategies are material-agnostic and applicable to other disordered semiconductors.
References:
[1] Konstantatos et al. Nature photonics1, 531 (2007).
[2] Armin et al. ACS Photon. 1, 173 (2014).
[3] Armin et al. Laser Photon. Rev. DOI 10.1002/lpor.201400081 (2014),
[4] Armin et al. Nature Comm. Submitted (2014).

Organic solar cells (OSCs), due to its fundamental limitations, have resorted to the use of buffer layers to achieve good device performance. There have been several materials that were reported to aid high efficiency solar cell devices such as conducting organic molecules or metal oxides. In this study, we investigated the effects of oxidation state in copper oxide based buffer layer in relation to its role in device performance. We demonstrated the use of evaporated-Cu_xO films as an effective hole transporting buffer layer in improving the device performance as well as its stability. A hole transporting layer with a fully-oxidized CuO film with PCE of 4.06% was achieved using a P3HT:PC₆₁BM-based solar cells with device degradation of only 25% after 40 days. In the study, the variation in the oxidation state affects the band position of buffer layer and built-in voltage of the device, therefore, leading to variation in device performances. Lastly, we examine cause of device deterioration and proposed. We have proposed a strategy in recovering the device efficiency.

Tandem and triple junction polymer solar cell can overcome the limitations of a single-layer bulk heterojunction photovoltaic cell which has narrow absorption maximum and width. In tandem or multi-junction solar cell, two or more cells with different absorption spectra can be stacked together in order to harvest solar energy more efficiently. Lossless and transparent interfacial layer plays the most important role in achieving high open-circuit voltage (Voc) in tandem or multi-junction polymer solar cells. Here, we demonstrate a low temperature solution processed interfacial layer (PEDOT:PSS/AZO/PEIE) for double and triple junction polymer solar cell. Solution processed and chemically stable PEDOT:PSS/AZO/PEIE layer can efficiently double the Voc of tandem polymer solar cell and it can be considered as useful candidate for triple junction polymer solar cell without any need of UV light illumination.

We demonstrate the use of TiO₂ nanorods with well controlled lengths as excellent electron extraction materials in significantly improving the performance of inverted polymer solar cells (PSCs). The cells containing long nanorods outperform the devices using amorphous TiO₂ particles as the electron extraction layer, mainly by a two-fold increase in short-circuit current and fill factor. The enhanced charge extraction is attributed to the high electron mobility in crystalline nanorods and their preferential alignment during film formation. Furthermore, transient photocurrent studies suggest the presence of fewer interfacial defects in the nanorods interlayers, which can effectively decrease carrier recombination and suppress electron trapping.

Organic solar cells have attracted great attention due to their potential to enable lightweight, flexible, large-area, and cost-effective photovoltaic technology. In order to optimize their performance, a variety of hole transporting materials has been applied to them. Among the materials, poly(3,4-ethylenedioxy-thiophene):polystyrene sulfonate (PEDOT:PSS) is one of the most promising candidates because its earth-abundant element composition (e.g., C, H, O, and S) facilitates future cost-effective, large-scale manufacturing. However, because its strong acidity corrodes organic donor and anode materials, new neutral hole-transporting polymers are needed to expand the applicability of organic photovoltaics further.
In this sense, poly(3,4-dimethoxythiophene) (PDMT) can be a neutral alternative to PEDOT:PSS due to the same earth-abundant element composition, but typical PDMT does not have a high conductivity compared to PEDOT:PSS. We have modified PDMT via oxidant chemical vapor deposition (oCVD), which has a remarkably improved conductivity relative to the typical PDMT. Specifically, the nature of oxidative polymerization achieved in oCVD reactors generates doped PDMT with anion dopant ions (Cl^-), which is advantageous for increasing conductivity. Consequently, oCVD-processed PDMT can have a high conductivity without acidity, and thus it becomes one of the most suitable candidates to replace PEDOT:PSS.
In this study, PDMT hole transporting layer (HTL) is successfully integrated into organic photovoltaic devices for the first time. Though PDMT is insoluble and infusible, and thus typically difficult to process, patterned thin films of this regioregular polymer were easily prepared using a vacuum-based vapor-printing technique (i.e., oCVD combined with in-situ shadow masking). The advantages of vapor-printed PDMT over spin-coated PEDOT:PSS were systematically uncovered using photo-conductive atomic force microscopy and organic field-effect transistors. In particular, vapor-printed PDMT HTL functions better than spin-coated PEDOT:PSS HTL by enhancing short-circuit current and fill factor in DBP:C₆₀ photovoltaic devices. The maximum power conversion efficiency (PCE) was 4.1% for employing vapor-printed PDMT HTL, and 3.5% for using spin-coated PEDOT:PSS HTL. Furthermore, vapor-printed PDMT HTL demonstrates much longer-term stability in terms of PCE because PDMT is a neutral material, unlike acidic PEDOT:PSS. The photovoltaic device with vapor-printed PDMT HTL maintained 83% of its optimum efficiency after 17 days in a N₂-filled glove box, while one with spin-coated PEDOT:PSS HTL retained only 12% of its best efficiency under the same conditions. The advances of this work can also be applied to different-type solar cells and any other organic electronics because oCVD is a powerful platform technology, independent of material solubility and substrate properties.

Solution processible cadmium sulfide (CdS) with a high electron mobility and a conjugated polyelectrolyte (CPE) were utilized as cathode interlayer on ITO to construct inverted solar cells based on a poly({4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b&’]dithiophene-2,6-diyl}{3-fluoro-2-[(2-ethylhexyl)carbonyl]thieno[3,4-b]-thiophenediyl}) (PTB7) as the polymer donor and [6,6]-phenyl-C₇₁-butyric acid methyl ester (PC₇₁BM) as the acceptor. The power conversion efficiency (PCE) of the inverted solar cells without the cathode interlayer (only ITO) was 2.03% due to poor V_OC of 0.306 V and FF of 43.8%. Using a CdS layer as a cathode interlayer significantly improved V_OC of 0.589 V and FF of 55.7%, giving the PCE of 4.90%. Combined solution processed CdS and CPE layers deposited on ITO substrate enhanced the PCE of 6.81% with V_OC of 0.747 V, and FF of 59.3%. Characteristics of devices with a ZnO cathode interlayer were also compared. Our results suggest that hybrid inorganic/organic materials are promising candidates as a cathode interlayer in high efficiency inverted solar cells through the modification of interface contacts.

We demonstrate a highly efficient inverted bulk heterojunction polymer solar cell using a wet-chemically prepared ZnO with a self-organized ripple nanostructure as an electron extraction selective layer and the blend of poly[[4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-bprime;]dithiophene-2,6-diyl][3-fluoro-2-[(2-ethylhexyl)-carbonyl]-thieno-[3,4-b]thiophenediyl]] (PTB7) and [6,6]-phenyl C71 butyric acid methyl ester (PC70BM) as an active light absorbing layer. In order to enhance the electron extraction efficiency, the ZnO ripple surface was modified with various metal carbonate materials such as Li₂CO₃, K₂CO₃, Na₂CO₃, Cs₂CO₃, and (NH₄)₂CO₃. Inclusion of an additional metal carbonate layer led to improve the electron extraction properties by modifying the energy level and electron transport properties without destruction of ZnO ripple structures. The highest performing solar cells fabricated with sodium carbonate yielded a PCE of 9.32%; this value represents an ~10% increase in the efficiency compared to solar cells without metal carbonate treatment.

Photovoltaic (PV) cells, which convert sunlight into electricity, have been greatly developed in particular including solution processed solar cells such as organic, hybrid organic-inorganic and dye-sensitized solar cells. The progress in the performance of solution based solar cells is achieved by newly developed materials having low optical band-gap and based on lead halide perovskite. To derive the potential of materials, the optimization of device is an important work. The optimization of device for high performance requires to probe several parameters including device thickness, ratio of donor and acceptor materials, and the quantity of additive. However, the spin-coating that is generally used to optimize the device involves many sampling processes with regard to donor-acceptor ratio and thickness, which means a significant loss of time and materials. In this study, we utilized the spray coating as roll-to-roll compatible process to rapidly screen and optimize the solution based solar cells. With controlling flow rate gradually and introducing dual feed nozzle, we could conduct single deposition parameter screening for optimization of thickness or one parameter in state of fixing the other parameter and composition screening for optimization of ratio of two materials. To access this technique, we fabricated polymer solar cells using poly(3-hezylthiophene) (P3HT):(6,6)-phenyl C61 butyric acid methyl ester (PCB61M) and poly(N-9&’-hepta-decanyl-2,7-carbazole-alt-5,5-(4&’,7&’-di-2-thienyl-2&’,1&’,1&’-benzothiadiazole)) (PCDTBT):(6,6)-phenyl-C71-butyric methyl ester (PC71BM). Despite different coating process, the optimum ratios of materials obtained from this technique have similar values compared with spin coating process as 1:1 and 1:3.5. Also, we attempted optimization of hybrid organic-inorganic based solar cell (CH₃NH₃PbI₃) with controlling the quantity of methylammonium iodid (MAI) added on PbI₂ film in sequential process, which showed peak efficiency of 7.48 %. We anticipate that the introduction of this system in spray process has great potential to save time and materials for optimization of various solution based devices.

Recent advances in organic photovoltaics (OPV) have resulted in power conversion efficiencies (PCE) of >10%. Most high efficiency OPVs use poly(3,4-ethylenedioxythiophene):polystyrenesulfonate (PEDOT:PSS) as the hole transport layer (HTL). However, the stability of these cells can be limited by PEDOT:PSS reacting with the active layer [1] or the anode electrode [2]. Hence, further progress in this field requires the development of alternative HTL materials that can increase device lifetime and PCE. Copper(I) thiocyanate (CuSCN) is an inexpensive inorganic, molecular, p-type semiconductor that exhibits excellent hole transport properties and high optical transparency due to its large band-gap (>3.5 eV) [3]. Its wide availability combined with processing versatility makes it ideal for large area opto/electronic applications. This unique combination of attractive characteristics has led to CuSCN being utilised in dye-sensitised solar cells [4] and more recently in hybrid perovskite-based photovoltaic cells [5]. Despite its potential, however, reports on OPV devices utilising CuSCN are extremely limited.
Here we report on the use of CuSCN as a HTL material in high efficiency solution-processed OPV cells. Field-effect transistor measurements have been used to assess the charge transport properties of CuSCN films processed from solution at 100 °C. Combining transistor measurements with other thin-film analysis techniques allowed us to optimise fabrication parameters and improve the device performance. The identification of inexpensive sulphur and non-sulphur based solvents facilitated the production of high concentration CuSCN solutions (10-40 mg mL^-1), which enabled deposition of HTL with controlled thickness. Using these novel formulations, we were able to solution-deposit CuSCN films with high uniformity over large area substrates at 100 °C. Average optical transparencies approaching 90% (400-1300 nm) were measured for CuSCN-based HTL - a value significantly higher than that measured for standard PEDOT:PSS interlayers. Furthermore, we were able to modify the conductivity and work function of CuSCN films with the addition of suitable chemical dopants; hole mobilities exceeding twice that of pure CuSCN films were demonstrated for optimised doped systems. Finally, bulk heterojunction OPV cells made using optimised CuSCN-based HTL, demonstrated PCE in the range 6-8%, which are superior to those measured in reference cells with PEDOT:PSS. The dramatically enhanced OPV performance was attributed to the extreme transparency and excellent electron blocking properties of CuSCN, demonstrating its tremendous potential for use in next generation low-cost photovoltaic technologies.
[1]A. Garcia, et al., Adv. Mater., 24, 5368, 2012
[2]M.P. de Jong, et al., Appl. Phys. Lett., 77, 2255, 2000
[3]P. Pattanasattayavong, Adv. Mater., 25, 1504, 2013
[4]G.R.R.A. Kumara, et al., Sol. Energy Mater. Sol. Cells, 69, 195, 2001
[5]P. Qin, et al., Nat. Commun., 5, 3834, 2014

We present a room temperature method for fabrication of metal oxide films using metalorganic precursors compatible with organic photovoltaic layers, without the use of toxic solvents. With potential for use as electrode material, the films may also be incorporated into hybrid devices.
Manganese (II) and copper (II) oxide films have been produced by UV irradiation of metal phthalocyanines deposited on silicon. The original morphology of the crystalline phthalocyanine film creates a templating effect on the oxide formed, offering a method of patterning (1).
Here, we have applied the method to zinc analogues of the precursors, treating films with UV light in a partial atmosphere of oxygen to yield oxygen deficient zinc oxide and removal of organic material, as confirmed by secondary ion mass spectrometry and x-ray photoelectron spectroscopy. Comparison of the analogues has enabled determination of the mechanism by which the oxide forms, capitalisation of which will increase the yield.
Precursors may be grown to form various morphologies such as films and nanowires by modification of deposition parameters, allowing optimisation of device efficiency and alternative geometries. Co-deposition of a selection of metal phthalocyanines may be used to form doped zinc oxides with improved properties.
Potential uses, including charge injection layers for hybrid devices, antireflection coatings on solar cells and electrodes, will be discussed.
(1) J. A. Gardener, I. Liaw, G. Aeppli, I. W. Boyd, S. Fiddy, G. Hyett, T. S. Jones, S. Lauzurica, R. G. Palgrave, I. P. Parkin, G. Sankar, M. Sikora, A. M. Stoneham, G. Thornton, and S. Heutz, The Journal of Physical Chemistry C 2011115 (27), 13151-13157

Networks of silver nanowires (AgNWs) are promising candidates for transparent conducting electrodes in organic photovoltaics (OPV). These networks exhibit comparable opto-electrical performances to the commonly used indium tin oxide (ITO) while being cheaper, less energy consuming, and more flexible under repeated bending cycles. To decrease the initially high sheet resistance of as-prepared AgNWs, a post-annealing step (90 min@200 °C) is commonly used, being detrimental for processing on polymeric substrates. Beside these high temperatures involved in bottom-electrode fabrication, a fast and scalable technique to enable the roll-to-roll processing of AgNW top-electrodes for small molecule-based OPV is not yet known. Currently, only processing on polymer-based solar cells or direct lamination is reported.
We present novel low temperature-based methods to integrate silver nanowires in organic small molecule-based photovoltaics, either as transparent and highly conductive bottom or top-electrode. The bottom-electrodes are prepared by organic matrix assisted low-temperature fusing. In this solvent-mediated process, polymers with both hydrophobic and hydrophilic side chains are coated below AgNWs. Hereby, the hydrophilic part of the polymer directs into the polar solvent, while the hydrophobic part couples with the hydrophobic NW shell. During drying of the solvent, the junctions strongly bend over each other and the electrical contact of the nanowire junctions is significantly improved. In comparison to networks without these polymeric sublayers, the sheet resistance of the whole network is decreased by several orders of magnitude.
AgNW top-electrodes are realized by modifying high-quality wires with a perfluorinated stabilizer. Hereby, the corresponding AgNWs can be nicely dispersed in highly fluorinated solvents. These solvents are used to pattern organic light emitting diodes in our photolithography lab since they do not damage the small molecule layers. Accordingly, our modified AgNW dispersion can be spray-coated onto all kind of OPV devices. Both bottom- and top-electrodes show sheet resistances of <11 Omega;/#9633; at >87 % transparency directly after spray-coating at very low substrate temperatures of <80 °C. We also demonstrate the implementation of our novel AgNW electrodes in organic solar cells. The corresponding devices show almost identical performance compared to organic solar cells exploiting ITO as bottom or thermally evaporated thin metal as top-electrode.

The brittleness and fragility of many current high performing organic solar cells (OSCs) might limit their applicability in applications demanding mechanical compliance. The design of organic semiconductors that can be significantly deformed would facilitate roll-to-roll production, mechanical robustness for portable applications, conformal bonding to curved surfaces, and would enable large-scale solar farms based on ultra-thin organic modules that can survive outdoor environment. This paper describes our efforts to understand and control the structural parameters that influence the mechanical properties of modern conjugated polymers. We identify several key determinants of the mechanical properties including molecular structures, polymorphisms, and resulting microstructures. Our studies on the elasticity and ductility of polymer:fullerene bulk heterojunction blends reveal the strong influence of size and purity of the fullerene, the effect of processing additives as plasticizers, and the details of molecular mixing. We also describe the applications of especially deformable materials in stretchable and mechanically robust devices. Our principal conclusion is that while the field of plastic electronics has achieved impressive gains in the last several years in terms of electronic performance, all semiconducting polymers are not equally deformable, and thus materials tested on glass substrates may fail in real-world applications and may not be amenable to stretchable—or even modestly flexible—systems. Our results should inform the engineering of new materials to maximize both mechanical resilience and photovoltaic performance.

We have discovered that when a relatively pure fullerene layer (e.g., PCBM, ICBA, bis-PCBM, C₆₀, etc.) is exposed to evaporated metal (e.g., Ca, Al, Mg, Ag, Au, etc.), the metal penetrates substantially into the fullerene film, altering its dielectric properties and inducing an appreciable number of equilibrium free carriers by chemical doping. With cross-sectional TEM, we find that metal incorporation is so extensive that the final result is a film with ~3-20 nm diameter metal nanoparticles embedded in a fullerene matrix. The consequences of this unexpected discovery are vast, opening possibilities for plasmonic-scattering enhanced Ohmic contacts for OPVs, controlled fullerene-layer doping, and ultra-high dielectric constant fullerene films. This finding also partly explains the universal leakage current observed in such diodes and forces us to re-examine the properties of fullerene-based bilayer devices. Interestingly, we find that conjugated polymers act as effective getters for the metal penetrant, efficiently stopping most metals from penetrating further into the organic film. Here we present details regarding the effect of different metals, fullerene derivatives, and evaporation conditions on the penetration/doping of fullerene layers in order to explore the underlying effects behind this unanticipated discovery.

In this contribution new solution-processable Poly(ethlyendioxythiophene) (PEDOT) and electrode materials for organic photovoltaic (OPV) applications will be introduced. This includes materials and concepts for inverted architecture and ITO-free OPV devices as well as solution processed metallic silver and a new material range of completely water-free organic solvent based PEDOT. ¹ This highly complementary set of new developments allows the realization of an all solution-processable approach to OPV addressing important points for large scale production and compatibility.
Different PEDOT:PSS dispersions have been developed and tailored towards their use as hole transport layers and transparent conductive electrodes in OPV devices. For inverted structure top contacts a highly conductive grade with very good film formation and wettability on photoactive layers was developed to allow semi-transparent grid based top contacts. A second type is based on organic solvent PEDOT:PSS sytems with enhanced adhesion to the photo-active layer increasing the stack robustness. For the bottom contact a highly transparent PEDOT smoothening layer is presented which demonstrates a high transparency even with a few hundred nanometer film thickness. This material was designed to be combined with silver grids or silver nanowire mesh for ITO-free devices.
A highlight will be the introduction of different metallic silver materials applicable to a variety of substrates and architectures. For bottom grids different low-temperature (<250 °C) silver resinates can be printed to form uniform and structured metallic silver grids with clear advantages over metal inks including high conductivity and smooth metallic surface.
Our organic solvent based PEDOT range highlights a new class of organic electronics materials which are based on non-polar solvents and are completely water-free. Different solvent systems are possible allowing a versatile orthogonal approach for numerous applications in OPV.
¹ W. Lövenich, Polymer Science, Ser. C, 2014, 56, 1, 136-144.

Considering bulk-heterojunction (BHJ) organic solar cells, which seem to be very prospective in the future, one of the limiting factors is the development of the transparent electrodes. Due to the limited amount of indium in nature and thus its high price, replacement for widely used indium tin oxide (ITO) is highly desired. One of the possibilities is to substitute ITO with carbon based electrodes such as heavily boron-doped conductive diamond (BDD) [1]. Diamond which is generally recognized as an insulating material, once successfully doped becomes a wide-bandgap semiconductor material with excellent potential due to the unique combination of its physical and electronic properties. The addition of boron atoms has a strong influence on diamond layers electrical conductivity. High boron concentrations typically result in conductive system with electrical properties comparable to metallic materials.
Present research demonstrates the possibility of the ITO replacement using nanocrystalline BDD. Diamond layers deposited by means of chemical vapor deposition of several thicknesses were prepared containing various boron concentrations in a gas phase, optimization of growth conditions at high boron/carbon ratio can lead to low sheet resistance comparable or exceeding the ITO samples. Dependence of the above-mentioned parameters on the optical and electrical properties of the BDD was studied, to achieve the optimal conditions for the effective application of the diamond electrodes in organic electronics.
BHJ organic solar cells were fabricated to test the potency of the BDD application in the photovoltaic devices. Obtained results proved the possibility of the aforesaid application. Achieved values of the efficiency approaching 50% of the efficiency of regular ITO based solar cells and can be compared with ones obtained using graphene electrodes in the organic BHJ solar cells, however applied material is much cheaper and easier to produce.
[1] Candy Haley et al., Adv. Funct. Mater. 2010, 20, 1313-1318

Conventional solar cells generate one electron for each absorbed photon, wasting a photon&’s excess energy above the bandgap. Exciton fission in organic molecules splits a high-energy singlet exciton into a pair of low-energy triplets. In solar cells, it promises to double the photocurrent for the blue/green solar spectrum and overcome the fundamental photovoltaic efficiency limit. In this talk, we will describe an organic photovoltaic cell based on exciton fission, which for the first time produced more than one electron per incident photon in the visible spectrum. Also, we will discuss the fundamental mechanism governing singlet fission and present a first principles expression that successfully predicts the fission rate in materials with vastly different structures.

Singlet fission (SF), a molecular process involving the conversion of one singlet exciton into two triplets, may be used to help design new efficient solar cells. We study the microscopic mechanisms in singlet fission by using the recently developed symmetrical quasi-classical (SQC) nonadiabatic molecular dynamics simulation approach [Cotton and Miller, J. Phys. Chem. A, 117, 7190, 2013; Meyer and Miller, J. Chem. Phys. 70, 3214, 1979]. The robust SQC method allows the effects of energy levels, electronic couplings and electronic phonon couplings on SF to be thoroughly investigated. Some interesting results related to bath modeling, such as low frequency bath and low temperature, and pathway (quantum) interference may inspire new design principles for SF materials.

Current solar cells are unable to harness the energy from photons that are in excess of the semiconductor bandgap. This is known as the Shockley-Queisser (SQ) limit, which restricts for example, the efficiency of silicon solar cells to a maximum of 33%. The ability of certain organic semiconductors to undergo efficient singlet fission (SF) may allow the SQ limit to be circumvented. SF is a process that converts a singlet-excited state into a two triplet electron hole pairs. If an appropriate acceptor collects these multiple electron-hole pairs, the SQ limit can be exceeded.
Here, we study the yield and mechanism of such an energy transfer process using covalently bound tetracene and PbSe nanocrystals (NCs). Tetracene is chosen because it has been shown to have SF yields close to 100%. PbSe NCs are an appropriate acceptor material because the triplet energy of tetracene is relatively close to the its first excited state. Previous work has shown that the yield of SF depends on the crystallinity of molecules. Our study includes the organic synthesis of functionalized derivatives of tetracene to control/ increase the crystallinity of this molecule bound to PbSe. The relationship between molecular structure and energy transfer will be investigated.

Organic materials that undergo singlet exciton fission have received a great deal of recent focus due to their potential to serve as key components in low-cost, multijunciton photovoltaic cells. During singlet fission, an excited electron in a spin singlet configuration relaxes by using some, but not all, of its energy to excite a second electron on a neighboring molecule, creating two spin triplet excitations. While this process can be exceedingly fast in some systems (~100 fs - 100 ps) as it does not violate spin conservation since the generated triplet pair is spin correlated and net spin singlet, the ability for a particular molecular pair to undergo singlet fission depends on their degree of electronic coupling that in turn reflects their relative intermolecular arrangement. In disordered systems, such as polycrystalline or amorphous thin films, this can cause singlet fission to occur over a multitude of timescales, as singlet excitons must first diffuse to regions that possess adequate intermolecular coupling to allow fission. During this diffusive phase, processes that can compete with singlet fission can quench excitations and negatively impact triplet yields. Here, we present the results of transient absorption and steady-state and time-resolved emission experiments that indicate that excimer formation can compete with singlet fission in disordered acene films. As amorphous films of 5,12-diphenyltetracene (DPT) are cooled, a strong, featureless red-shifted emission band is observed that is indicative of excimer-like emission. Likewise, transient absorption experiments indicate that the triplet yield drops from 122% at room temperature to ~60% as the system is cooled to 77K. We have constructed a kinetic model that can adequately model competition between the diffusion of photogenerated singlet excitations to excimer forming sites as well as singlet fission sites in DPT and related acene films as a function of temperature. Our results suggest that excimer formation to some extent can compete with singlet fission at room temperature in disordered films and may reduce triplet yields. This underscores the importance of controlling the intermolecular arrangement of molecules in thin films designed to undergo singlet fission.

We demonstrate the successful incorporation of a solution-processable singlet exciton fission material, 6,13-bis(triisopropylsilylethynyl)pentacene (TIPS-pentacene), into infrared photovoltaic devices. TIPS-pentacene rapidly converts high-energy singlet excitons into pairs of triplet excitons via singlet fission, thereby potentially doubling the photocurrent from high-energy photons. At the same time, the low-energy photons are captured by small bandgap lead chalcogenide (PbSe or PbS) nanocrystals, which also serve as electron acceptors. This is the first singlet fission system since the pentacene-nanocrystals devices that performs with substantial efficiency, with a maximum power conversion efficiency above 4.8%, and external quantum efficiencies up to 60% in the TIPS-pentacene absorption range (~550-700 nm). With nanocrystals of suitable bandgap, the TIPS-pentacene internal quantum efficiency reaches 160±40%, confirming that singlet fission is operational in these devices.

Very recently, organic solar cells (OSC) with very low concentrations of donor molecules (~1-10%) in a fullerene matrix have been shown to work remarkably well with efficiencies of more than 5%. Given the diluted nature of the donor molecules this kind of OSC is sometimes called "homoeopathic" OSC, or "Organic Schottky Junction Diodes" following one hypothesis for their working mechanism. These OSC represent an exciting new device architecture. On the one hand, their working mechanisms are not yet well understood, on the other hand, they provide an excellent model system to learn more about the semiconducting organic materials, in particular about the interface between donor and acceptor molecules. As one example, by varying the concentration of transparent donor molecules in C₆₀, we were able to show that the recombination scales with the amount of available interface between donor and C₆₀, which in turn allowed us to vary the open circuit voltage V_oc without changing the energy of the charge transfer state E_CT. The energy loss E_CT-eV_oc for the best performing TAPC-C₆₀ combination (5%vol donor) was 0.52eV, i.e. much lower than the typical energy loss of about 0.6±0.05eV in OSC. These findings give new design criteria for better organic materials and eventually higher efficiencies.

Utilization of 2,3,4-tri(3-(dimethylamino)propoxy)fulleropyrrolidine (C₆₀-N) as an electrode modification interlayer in organic solar cells improves the device power conversion efficiency (PCE), enables operation of air-stable high work functional metals (Ag, Cu, Au) as cathodes, potentially eliminating the vacuum deposition steps and increasing the device lifetimes, and eases the tolerance to the interlayer film thickness variations that is critical for reliability of solution-based device fabrication processes. The functionality of C₆₀-N stems from the interaction of dipolar amine side groups with a metal surface, leading to a reduced work function of the electrode. Here, we discuss an approach of utilizing the dipolar C₆₀-N as a bulk additive in the bulk heterojunction active layer which is composed of [6,6]-phenyl C₇₁-butyric acid methyl ester (PC₇₁BM) acceptor and either 7,7&’-(4,4-bis(2-ethylhexyl)-4H-silolo[3,2-b:4,5-b&’]dithiophene)-2,6-diyl)bis(6-fluoro-4-(5&’-hexyl-[2,2&’-bithiophen]-5yl)benzo[c][1,2,5] thiadiazole) (p-DTS(FBTTh₂)₂) or thieno[3,4-b]thiophene-a-benzodithiophene with 2-(ethylhexyl)thienyl side chains (PCE-10) as the small-molecule or polymer donor, respectively. We demonstrate that electric field poling of the devices with C₆₀-N additive results in significantly improved PCE, from 5.2% to 8.0% in p-DTS(FBTTh₂)₂:PC₇₁BM devices and from 6.8% to 10.1% in PCE-10:PC₇₁BM devices, which is higher than in the optimized devices that do not contain the additive, 7.2% and 9.2%, respectively. We discuss the origin of C₆₀-N functionality and the details of device operation, including stability after electric field poling. A mechanism of the additive operation that includes an electric-field induced alignment of dipolar side groups and thermal annealing under poling of diode-like structures in forward direction will be discussed. Incorporation of electrotropic molecules, i.e. with molecular conformations being responsive to electric field, as the bulk additives promises to become an effective universal approach for improving the efficiency of bulk heterojunction organic solar cells.

Excited triplets states are ubiquitous in Organic Electronic (OE) devices and are often considered an undesired species.
In solar cells triplet excitons lead to a loss of energy due to their energetic position in relation to the acceptor, leading to decreased efficiencies. Triplet states also play a major role in the aging process of OE polymers and in device deterioration, affecting OLEDs, OFETs and OPV cells alike.
In the field of OLEDs a lot of research has been conducted in order to utilise triplets excitons, immensely boosting their efficiency. In the younger fields of OFET and OPV applications the interactions of triplet excitons with the bulk materials are less well understood. Additionally, triplet states still remain highly elusive and difficult to characterise.
Our goal is therefore to better understand triplet excitons by creating new tools for more in-depth analyses. At the same time we aim at improving the efficiencies of semi-conducting polymers for OPV and OFET applications. One way to achieve this is by making use of the increased life-time of triplet states which can - under the right conditions - lead to increased exciton diffusion lengths.
We designed a Donor-Acceptor system with the energy levels such that they allow for the extraction of triplet excitons in an OPV device. We therefore incorporated platinum into the polymer backbone and used the heavy metal&’s spin-orbit coupling effect to generate triplet excitons directly on the polymer chain. This also enabled us to analyse the full generation pathway spectroscopically.
The added platinum moiety leads to improved polymer properties and to an increase in extracted charges. The knowledge gained gives clear indications on how to further increase the stability of devices and can potentially turn one of the major loss mechanisms in OPV, the random formation and decay of triplets, into a secondary extraction pathway.

Harnessing the energy of light in low-intensity environments, such as indoor, would bolster the development of stand-alone low-power electronics, relevant for new applications such as wearables or the internet of things. Organic solar cells are well-adapted for such applications as they can be printed at low-cost on flexible substrates and they retain good performances at low-light intensities. However, given the low power available, the energy thus generated cannot be used readily and needs to be stored in order to be useful.
In this work, we investigate the integration of printable organic solar cells with fully printed super-capacitors for energy generation and storage in low-light intensity environments. The interest of adding a rectifying diode to prevent self-discharge is also discussed. A solution-processed bulk-heterojunction organic solar cell is electrically coupled with a fully printed super-capacitor. The amounts of power generated and of energy stored are then studied under different light intensities. Under 1-sun (100 mW/cm²), the cell produces 4.5mW/cm² and the super-capacitor stores up to 19.6 mJ/cm². When the light intensity is decreased to 17 mW/cm², the system is still able to store up to 14.5 mJ/cm² of energy, 67 % of its 1-sun value. This work suggests that a fully-printed monolithic device comprising an organic solar cell and a super-capacitor would provide a viable solution for energy generation and storage in low-light intensity environments.

Small conjugated molecules (SM) are rapidly gaining momentum as a valid alternative to semiconducting polymers for the production of solution-processed bulk heterojunction (BHJ) solar cells, as they allow to overcome current limitations imposed by the intrinsic polydispersity of long conjugated chains and low batch-to-batch reproducibility. The major issue with SM-BHJ solar cells is the low carrier mobility due to the scarce control on the phase segregation process and consequent lack of preferential percolative pathways for free carriers to the extraction electrodes. In this talk, I will demonstrate a new paradigm for fine tuning the phase segregation in SM-BHJs based on the post-deposition exploitation of latent hydrogen bonding in binary blends of functionalized electron-donor moieties mixed with PCBM. The strategy consist in the chemical protection of the hydrogen bond forming sites of the donor species with a thermo-labile functionality whose controlled thermal cleavage leads to the formation of highly crystalline, stable, phase-separated molecular aggregates. This approach allows for fine tuning of the nanoscale film connectivity and thereby to simultaneously optimize the generation of geminate carriers at the donor-acceptor interfaces and the extraction of free charges via ordered phase-separated domains. As a result, the PV efficiency undergoes an over twenty-fold increase with respect to control devices. The structural, morphological and photophysical implications of the hydrogen bond networking are thoroughly investigated by combining X-ray diffraction, scanning probe techniques and ultrafast spectroscopy experiments on both H-networked and control devices. Importantly, results show that the strong cohesive forces introduced by H-networking stabilize the film morphology and optical activity both in term of suppressed solubility in organic solvents and for shelf times as long as six months.
This strategy, demonstrated here with a binary mixture of diketopyrrolopyrrole derivatives with PCBM can in principle be extended to other molecular systems for achieving highly efficient, stable, small-molecule BHJ solar cells.

The open circuit voltage of organic solar cells is very far from its potential upper limit due to substantial charge recombination of various types, whose origin still remains to be accurately determined. Herein, we summarize what we have learned about these limitations by studying the radiative part of recombination in OPVs. We study the limiting mechanisms by first employing steady state measurements of recombination as a function of charge carrier density by accurately evaluating the diode ideality factor. The diode ideality is related to the order of recombination and we first assess it via light intensity dependent open circuit voltage characterization under the influence of a varying temperature. We focus our study on the ratio between radiative and non-radiative recombination via the interfacial charge transfer state (CT) as determined by absolute CT electroluminescence efficiency measurements, also as a function of temperature. The charge transfer state governs the radiative recombination in OPV bulk heterojunctions and is therefore crucial to evaluate in this context. Substantially improving the radiative efficiency of OPVs to increase the open circuit voltage is the only way to put these promising photovoltaic converters in efficiency parity with their inorganic or perovskite counterparts.

The theory typically used to correlate photon emission to photovoltaic performance in solar cells is based on detailed balance theory and has been derived and discussed extensively for the case of crystalline inorganic semiconductors with linear recombination mechanisms. One feature of this opto-electronic reciprocity is that the electroluminescence (EL) intensity increases exponentially with voltage, however, the spectral shape of the EL does not change with applied voltage. Therefore, if the shape of the EL spectrum depends on voltage, as observed for some organic solar cells, the validity range of this relation between the external quantum efficiency and EL emission, based on detailed balance theory, has been left behind and a more sophisticated theory is required.
In order to investigate the influence of the charge transfer (CT) states on the EL spectrum, we used a simple rate model approach. This model includes the singlet state of the polymer, the CT state and the charge separated state. For every state there are rates for thermal and radiative generation and recombination as well as for injection, and extraction from other states. In order to allow for any variations of the EL spectrum as a function of voltage, we need a non-linear model that takes into account a finite density of CT states per unit energy. In our model, we approximate the density of CT states by an exponential tail with the characteristic energy E_ch.
The introduced rate model describes three cases that we observe in our experiments: no voltage-dependence of the EL spectrum, a peak shift towards higher energies, and the increase of a second peak with increasing voltage. Our model shows that the ratio of the recombination rates (R+S) to injection rate n₀p₀H (n₀p₀ - carrier concentration in thermal equilibrium) determines whether the peak of the EL emission shifts in the experimentally accessible voltage range or not. The shift is due to saturation effects of the CT state. The shoulder is explained by radiative recombination in the donor, which increases exponentially with exp(qV/kT), whereas the emission of the CT state is proportional to exp(qV/E_ch). Consequently, a shoulder can be observed if the polymer emission is close in energy to the CT emission and if E_ch>kT.
Thus, while making a quantitative model to describe the EL emission would be quite difficult in such disordered systems as organic solar cells, our simple model shows under which circumstances the EL starts to deviate from the simple situation described by the opto-electronic reciprocity.

We study recombination in polymer:fullerene bulk-heterojunction (BHJ) solar cells by analyzing the forward-bias dark current as a function of temperature in three common high-performance BHJ materials systems. Our interpretation of the data is centered around applying the Shockley-Read-Hall (SRH) formalism to the case of recombination between mobile carriers and an exponential distribution of localized band-tail states. Remarkably, we find that the analysis gives a simple proportional relation between the dark-current ideality factor and the localized band-tail slope. Moreover, our model predicts a nearly temperature independent exponential slope of the dark J-V-T curves, which is in excellent agreement with experimental data. In general, diode current-voltage measurements give good agreement with the analytical model, confirming that the band-tail recombination mechanism applies to several important organic solar cell materials systems. We further demonstrate that the assumptions of the model are reasonable with detailed numerical drift-diffusion modeling. The implications of our findings are that SRH recombination through an exponential distribution of localized states is likely the dominant loss process in many important OPV systems, and that this process can be easily monitored by consideration of the dark-current ideality factor.

Organic solar cells (OPV) have developed rapidly over the past decade but still lag behind their inorganic counterparts in efficiency due to a combination of low open-circuit voltages and the difficulty of making optically thick cells with high fill factors. In this work we develop a comprehensive framework for understanding and improving the open-circuit voltage (V_oc) of organic solar cells based on quasi-equilibrium between Charge Transfer (CT) states and free carriers. We first show that the ubiquitous reduced Langevin recombination observed in organic solar cells implies quasi-equilibrium and then use statistical mechanics to calculate how many CT states will be populated in the solar cell at each voltage. This general result allows us to quantitatively assign the open-circuit voltage losses to a combination of interfacial energetic disorder, high CT state binding energies, large degrees of mixing and fast electron transfer at the donor/acceptor interface. Our work reconciles Langevin recombination with detailed balance calculations of V_oc based on the CT state absorption spectrum and gives an intuitive physical explanation for why the open-circuit voltage of organic solar cells is almost always 500 to 700 meV below the measured energy of the CT state.
To quantify the impact of energetic disorder on V_oc we develop a new temperature dependent Charge Transfer state absorption measurement. By looking at how the apparent CT state energy (E_ct) varies with temperature we can directly extract the degree of interfacial energetic disorder since it causes a characteristic 1/T temperature dependence in E_ct. We study five different OPV systems and find between 60 and 105 meV of disorder, contributing an additional 75-225 mV of V_oc loss not included in the 500-700 mV recombination loss mentioned above. Our theoretical result shows that disorder happens to affect E_ct in precisely the same manner as V_oc, which is why it does not affect their difference. Finally, we consider various strategies by which the open-circuit voltage could be improved, quantify their impact and show that the best gains could come from slightly raising the bulk dielectric constant of the active layer and reducing energetic disorder at the donor/acceptor interface.

Despite vast research on organic semiconducting materials and devices over the last two decades, significant gaps in fundamental understanding exist in key areas. A detailed understanding of charge carrier recombination is particularly important to promote efficient radiative recombination in light emitting diodes, minimize power conversion efficiency losses in photovoltaics, and enhance sensitivity in photodiodes. Bimolecular charge recombination in these devices is commonly described using the Langevin model, which assumes an encounter-limited process. However, blends for organic photovoltaics have often been measured to have major deviations from the Langevin model, most significantly a recombination rate that is several orders of magnitude less than expected. Here, transient experimental and computational simulation results are presented to help elucidate the origins of the unexpected nongeminate recombination dynamics[1,2], focusing on the role of morphology, charge carrier mobility, density of states, and charge-transfer state properties. We demonstrate that greatly reduced recombination rates are not an inherent property of phase separated systems and show how charge-transfer state properties can have a critical impact on the recombination dynamics.
1. J. Gorenflot, M. C. Heiber et al. J. Appl. Phys. 115, 144502 (2014).
2. M. C. Heiber, V. Dyakonov, and C. Deibel (under review)

Further improvements of the efficiency of organic solar cells are expected e.g. from advances in chemistry and the optimization of charge carrier transport. To investigate the recombination and transport processes in organic solar cells we use the technique of pulsed electrically detected magnetic resonance (pEDMR) where we measure the change of the photocurrent caused by resonant X-band microwave pulses in the presence of an external magnetic field. This has already proven to be a highly valuable tool in the investigation of inorganic photovoltaic cells. As test devices for this technique, we use bulk heterojunction P3HT/PCBM (poly(3-hexylthiophene-2,5-diyl)/[6,6]-phenyl C₆₁ butyric acid methyl ester) solar cells. Under illumination at a temperature of 10 K we are able to observe both positively and negatively charged polarons and can identify them as partners in a spin-dependent pair process by experiments using two microwave frequencies, thus providing evidence for a bipolar polaron pair recombination process. Using the time resolution and sensitivity of pulsed EDMR we are able to quantify the spin dynamics of the system and measure the lifetime of parallel spin pairs, the lifetime of antiparallel spin pairs, the spin decoherence time and the coupling strength between the spin partners. Experiments comprising a nanosecond laser pulse excitation enable a time-resolved investigation of the charge carrier generation and separation process. All of these microscopic parameters provide valuable information for an optimization of overall solar cell efficiencies.

One of the most critical problems in organic solar cells is their degradation during operation. Extrinsic degradation is caused by oxidization in the ambient atmosphere, whereas intrinsic degradation is due to the incident solar radiation. Even when the extrinsic degradation is minimized by operation in a controlled environment, organic solar cells (OSCs) show intrinsic degradation in sunlight alone, where the short circuit current (Jsc) and fill factors can decrease by more than 20% over a period of several days.
We seek to understand the fundamental atomistic mechanisms underlying the degradation process. We use the PTB7:PCBM71 donor-acceptor system as a prototype since this has led to the highest efficiency OSCs. Since light absorption and exciton formation occurs primarily within the low band gap PTB7 polymer, we investigate conformational changes that can be caused by light. Simulations utilize ab-initio density functional theory utilizing the well-established SIESTA local orbital approach.
We find that PTB7 polymer can undergo very low energy structural rearrangements by migration and re-bonding of either O or H atoms. O-induced rearrangement involves local motion of the bridging O atom to the thiophene group with formation of new hydroxl (OH) groups within the thiophene chain, and less than a 0.5 eV increase in energy. Activation energy barriers for O motion are < 1 eV, indicating feasibility of deformation under light soaking. We additionally find H induced motion leading to new C=O bond formation and rupture of PTB7, producing low energy final states < 0.5 eV. These conformational changes degrade the ability of the PTB7 donors to produce excitons and may degrade photo-currents.
In the category of extrinsic changes, we also have simulated ready oxidation of the PTB7 polymer to oxygen and water molecules, where new C-O and C-OH bonds are formed within the thiophene group. Other organic absorber materials will be discussed. We compare these atomic conformational changes with the observed degradation phenomena, to develop lifetime models and ways to improve device durability.

With efficiencies over 10% and potential commercial applications like building integrated photovoltaics in sight, increasing device stability has become a major challenge for organic solar cells. We have studied the degradation of a variety of organic photovoltaic (OPV) materials and observe an increase of energetic disorder in all investigated materials. This increased energetic disorder lowers the open-circuit voltage in amorphous materials by broadening the density of states. An increased recombination rate constant is not necessary to explain the open-circuit losses and also not observed. Crystalline materials are found to be less sensitive to increased energetic disorder.
Using charge extraction and photocurrent spectroscopy, we show an increased energetic disorder after the burn-in period for amorphous materials. Temperature dependent measurements of the open-circuit voltage over a wide range of light intensities for fresh and aged devices confirm this result. The changes in open-circuit voltage after ageing are temperature dependent and well described with a simple model that predicts the influence of increased energetic disorder on the open-circuit voltage. Transient photo-voltage measurements show no evidence of decreased carrier lifetime. This confirms that not a change in recombination dynamics, but rather a shift in the density of states leads to a lower open-circuit voltage.
Crystalline materials also show an increase of energetic disorder, but no significant losses in open-circuit voltage. With charge extraction measurements we show that the carrier densities at open-circuit voltage are several times higher in crystalline materials than in amorphous materials. This is most likely facilitated by the separation of holes and electrons in aggregated domains of polymer and fullerene, respectively. For the large carrier density in crystalline materials, an increase of energetic disorder has a relatively smaller effect than in amorphous materials where the steady state carrier concentration is much lower. Moreover, the use of crystalline materials in organic solar cells seems to be beneficial also for other reasons, like reduced photo-oxidation rates and triplet concentrations.

In polymer solar cell (PSC) fabrication, high boiling point solvent additives such as diiodooctane (DIO) are often used to optimize the active layer morphology and achieve high power conversion efficiency. We have successfully fabricated single-junction PSCs with efficiencies exceeding 9% using this approach; however, we find that residual amounts of the solvent additive remain in the film after deposition and that the presence of the additive is deleterious to the long-term photo-stability of the active layer. We have developed a protocol to identify solvent additive remaining in the film and have utilized several spectroscopic measurements including FTIR to characterize additive-induced degradation. Consequently, we were then able to design and synthesize novel polymers that exhibit better photochemical stability. Additionally, we have explored alternative solvent additives that are less reactive under solar illumination even if they remain in the film post deposition. With this approach, we demonstrate stable, high-performance PSCs.

We explore the early aging (i.e. burn-in) characteristics of archetype organic photovoltaic (OPV) cells with fullerene C₆₀ acceptor layers. Initially, the short-circuit current density is found to decrease rapidly within the first several hours, after which the rate of change with time decreases substantially.^[1] Using time-resolved Fourier transform infrared spectroscopy, we find that several C₆₀ absorption features decrease over a similar time-scale, which we attribute to its photo-induced oligomerization.^[2] An analytical model, developed based on the decreased exciton lifetime and diffusion length of C₆₀ oligomers, is used to describe the burn-in process. The model predicts the lack of burn-in in blended heterojunction devices—since in mixed junctions the exciton diffusion length is always much larger than the average distance between dissociating donor/acceptor interfaces. Also, our results suggest that the increased stability of C₇₀-based OPVs is due to their reduced symmetry compared with C₆₀, which sterically inhibits their oligomerization. Our analytical model, in combination with direct chemical verification of the sources of material degradation over time, provides powerful insights into the failure modes of OPVs, which can be used to understand and inform the design of long operational lifetime devices.
[1] X. Tong, N. Wang, M. Slootsky, J. Yu, and S. R. Forrest, Sol. Energy Mater. Sol. Cells, vol. 118, pp. 116-123, Nov. 2013.
[2] P. C. Eklund, Y. Wang, J. M. Holden, A. M. Rao, and P. Zhou, Thin Solid Films, vol. 257, pp. 185-203, 1995.

We have studied the degradation of polymer solar cells (PSCs) and find that operational lifetimes can be improved when the PSCs are illuminated in an oxygen free environment. In PSCs made with PCDTBT and PC₇₁BM, we observe an average lifetime that exceeds 15 years with both the standard and inverted architectures.
To minimize the uncertainty of solar cell packaging and test PSCs in a controlled environment, we constructed an environmental chamber that keeps oxygen content below 0.1 part per million (ppm) and water content at 0.4 ppm. In PSCs made with PCDTBT and PC₇₁BM and aged for over 7500 hours at their maximum power point under such atmospheric conditions, we observe a period of burn-in loss followed by a period of linear degradation. While inverted architecture devices burn-in more slowly than standard architecture PSCs, all solar cells lose nearly 40% of their starting efficiency over the first 3500 hours of monitoring. The burn-in loss is dominated by losses in open-circuit voltage (Voc) and fill factor (FF). Over an additional 4000+ hours of monitoring, the solar cell degradation is minimal. The PSC lifetimes are extrapolated to exceed 15 years, and some individual solar cells have extrapolated lifetimes that exceed 25 years.
Ways to reduce burn-in loss and minimize long-term degradation under more realistic packaging conditions are explored. In both respects, we find that materials with a higher crystallinity are more stable. Under constant illumination and inert atmospheres, we find that PSCs made from more amorphous polymers like PCDTBT and regiorandom P3HT show a severe photo-induced efficiency loss of 25% during the first 60 hours, while more crystalline materials like ZZ115 or regioregular P3HT show burn-in losses of less than 10%. By performing photobleaching experiments on thin films in air, we also find that increasing a material&’s crystallinity increases its photochemical stability. For example, amorphous films of rubrene photobleach three orders of magnitude faster than crystalline films. Higher crystallinity could improve a material&’s stability for a variety of reasons. The increased density of more crystalline materials could reduce the solubility of photochemical reactants (O₂, H₂O, synthetic catalysts, organic impurities) and products, and we do find that the photobleaching rate correlates well with a material&’s density. Interchain stabilization in crystallites could reduce the number of available conformations, effectively reducing the number of possible reaction products. Crystallinity can also increase backbone planarity, which could both reduce intersystem crossing rate (and reduce triplet concentrations) and increase the energy required to break conjugated bonds.

Side-chain engineering has a significant impact to develop high performance semiconducting π-conjugated polymers for electronic and optoelectronic applications. The side-chains affect the molecular energy levels, internal morphology, and charge transport property in the solid states. In particular, it affects the miscibility with fullerene derivatives and modulates the device performance in organic solar cells. This study demonstrates that incorporation of branched side-chains where the branching position is far from the polymer backbone revealed the advantages of improved solubility with effective π-π intermolecular interactions between conjugated polymer chains. The specific effect of the alkyl-chain branching position is thought to be critical according to related work on organic electronics. The conjugated polymers bearing isoindigo-based monomer were successfully synthesized with the side-chain branching point. Their physical, photophysical, and electrochemical properties were investigated in detail. Eventually, thin film transistors and photovoltaic cells with PCBM were fabricated to study the structure-property relationship.

Diketopyrrolopyrrole (DPP)-based organic semiconductors have considered highly promising candidates for organic thin-film transistor and organic photovoltaic (OPV) cell owing to their outstanding charge transport properties. Although a number of research have been carried out using DPP-based organic materials, there is a few reports about systematic study from small molecules to high molecular weight polymers. In this work, we designed and synthesized DPP-based small molecules (i.e. dimer, and tetramer) and polymer composed of same repeating unit to compare their physical properties, electronic properties, and photovoltaic performances. In particular, conjugated tetramer has been investigated intensely after fabricating thin-film transistors and photovoltaic cells with PCBM. In brief, the DPP-based tetramer could exhibit the unique molecular aggregation behavior and internal morphology after blending with PCBM. In addition, these solution-processed bulk heterojunction cells led us to study donor structure-PV property relationships.

The use of the donor-acceptor (D-A) concept to create an alternating semiconducting copolymer structure has been recently demonstrated as a highly efficient approach to improve the semiconductor performance. In particular, diketopyrrolopyrrole(DPP)-based D-A conjugated polymers have been intensively employed for use in solution processed polymer solar cells. In this study, alternating conjugated terpolymers have been synthesized by using DPP and various donor monomers.The advantageous effect of A-D-A-D&’ terpolymer was investigated and compared with D-A polymers bearing only single donors. For example, a new alternating terpolymer (-(A-D₁-A-D₂)_n-) is composed of electron accepting DPP units (A) between two different electron donors such as pyrene (D₁) and thiophene (D₂). The molecular energy levels, intermolecular interaction, light harvesting property, and charge transport property could be tuned with the kind of donor monomer. Eventually, we fabricated thin-film transistors and photovoltaic cells with PCBM and investigated the effect of terpolymer structure on the device performances.

By adding a small amount of a high boiling point solvent such as diiodooctane (DIO) to solutions of regioregular poly-(3-hexylthiophene-2,5-diyl) (P3HT), films with different amounts of aggregated P3HT can be obtained by spin-coating. Here we use different spectroscopic methods and techniques of surface analysis to show that the use of DIO results in films containing a smaller fraction of aggregated phase, albeit with a higher degree of order within the aggregated phase. From spatially resolved spectroscopy it appears that material orientation during the drying process leads to the formation of boundaries containing highly ordered aggregates, with a high fraction of remaining amorphous material.

Solution-processed small molecules have attracted a great deal of attention in the field of bulk heterojunction organic photovoltaics (OPVs), owing to their advantages of well-defined structures, easy purification and low batch-to-batch variations. The packing of these small molecules into crystal structures in the active layer involving fullerenes is an important factor for improving the power conversion efficiency of its devices since the molecular packing in the active layer critically affects the charge transport. Because the extent of molecular packing can be very sensitive to its chemical structure, synthesizing small molecules with well-defined architectures that provide good packing can lead to high efficiency devices.
Here, we synthesized a series of acceptor-donor-acceptor (A-D-A) organic molecules, comprising oligo-thiophene as central donor and thiobarbituric acid (TB) units as end caps. The oligo-thiophene consists of two units: an electro-donating core and a π-conjugated bridging unit of terthiophene which linked the core and end caps. By changing the core structure and their alkyl chain length, we found that the molecular symmetry and coplanarity would significantly affect the melting/crystallization behaviour and the formation of crystalline domains in the blend film with fullerene. Molecules with symmetrical core display high crystallinity in their pristine state, but they have different driving forces in crystallization, presumably because of different degrees of coplanarity. Additionally, the alkyl chain structure was also an important factor of the controlling molecular symmetry and their crystallinity, whereas the molecule featuring unequal length side chain tends to be an amorphous material.

Since the synthesis of the fullerene derivative [6,6]-Phenyl C₆₁ butyric acid methyl ester (PC61BM), bulk heterojunction (BHJ) devices have shown the most promising structure for organic photovoltaic. Combined with low band gap copolymer, the power conversion efficiency (PCE) exceeds 6%. In particular, copolymers using Si bridging atoms, like the poly[2,7-(9,9-bis(2-ethylhexyl)-dibenzosilole)-alt-4,7-bis(thiophen-2-yl)benzo-2,1,3-thiadiazole] (PSiF-DBT), are promising materials due to their broad absorption and their charge transport properties. BHJ devices using PSiF-DBT:PC61BM as active layer reached PCE of 5.4% for devices without thermal annealing [1]. The main reason for these improved optoelectronic properties is that large Si atom, and consequently large C-Si bonds, modify the entire geometry, leading to a more crystalline structure, allowing a better stacking between the chains. The efficiency of BHJ devices is strongly dependent on the blend morphology. A nanoscale phase separation is desired to ensure the best donor acceptor contact leading to high exciton dissociation efficiency. Here we present a study of BHJ device using PSiF-DBT:PC71BM as active layer. XPS (X-Ray Photoelectron Spectroscopy) technique was used to probe both PSiF-DBT and PSiF:PC71BM films. XPS analysis of PSiF:PC71BM film shows that the chemical homogeneity of the film surface is improved with the temperature increase in thermal annealing. By analyzing the ratio of the areas between these atomic contributions (A_{C-N + C-S}/A_{C-O + C=O}) it is possible to indicate whether the film surface is rich in the polymer or PC71BM phases. The results of this analysis show that a more polymeric richer surface is achieved by thermal annealing. Photovoltaic characterization showed that as cast device has its PCE limited by its low value of short circuit density of current (J_{sc =} 4.06 mA/cm²). Associated with a low value of open circuit voltage (V_oc = 0.48 V) and a fill factor (FF) of 42%, lead to a PCE of 0.82%. This occurs mainly because, in the as cast film, a desired morphology was not achieved. With a non favorable interface for exciton dissociation a limited value for both J_sc, and V_oc is expected. As the annealing temperature increases the phase separations decreases, leading to a highly homogeneous film after 200^oC. In agreement with XPS results discussed before, thermal annealing makes the film surface more chemically homogeneous impacting directly in devices performance. After the thermal treatment at 200°C a noticeable improvement is observed in J_sc and in V_oc (9.36 mA/cm² and 0.57 V, respectively) rising the PCE to 2.5%. Due to the more homogeneous surface a more suitable donor/acceptor surface was achieved making the exciton dissociation more efficient.
[1] E. Wang, L. Wang, L. Lan, C. Luo, W. Zhuang, J. Peng, Y. Cao, Applied Physics Letters, 92 (3) 2008.

The fine tuning of the energy level and band gap of donor polymer is critically important for high power conversion efficiency (PCE) in polymer solar cells (PSCs). In this work, we systematically studied energy level control by using inductive and resonance effects for donor polymers. Three monomer units have been designed and synthesized by introducing the cyano (CN) and alkoxy (OR) groups onto 4,4'-didodecyl-2,2'-bithiophene (BT) unit step by step. The BT moiety was used as the reference structure. The synthesized monomers were polymerized with benzo[1,2-b:4,5-b&’]dithiophene (BDT) as a counter unit to afford three polymers (PBDT-BT, PBDT-BTC, and PBDT-BTCox). We investigated the effects of CN and OR groups on the energy level, band gap, and device performance in detail. As a result, the HOMO energy level was dramatically lowered by introducing CN group from -5.37 to -5.58 eV and the band gap became narrow significantly by introducing CN and OR groups from 2.1 to 1.74 eV via inductive and resonance effects. The resultant photovoltaic performances showed high open-circuit voltages (V_oc) ranging from 0.82 V to 0.98 V. Among the series, the polymer solar cell based on the blend of PBDT-BTCox and PC₇₁BM gave the best photovoltaic performance, with a V_oc of 0.86 V, and a PCE of 5.06%. We demonstrated that the inductive and resonance effects could be powerful synthetic strategy for fine tuning of energy level and band gap of conjugated polymers.

As computational speed and resources continue to improve, the design of molecules for particular applications has increasingly involved the computational prediction of molecular properties prior to undertaking synthesis. Moreover, high throughput approaches that create thousands of candidates and compute their properties have become common, with a variety of molecular and solid-state efforts embodied within the materials genome approach. In the work described here we discuss our approach to discover novel low-band gap materials for organic photovoltaic (OPV) applications by automated combinatorial generation of candidate co-polymers and small molecules and subsequent calculation of their electronic properties. The auto-generation and computation was performed using our Simulation Toolkit for Renewable Energy Atomistic Molecular Modeling (STREAMM) toolkit. The molecules consist of electron rich (donor, D) and electron deficient (acceptor, A) moieties, treated as modular building blocks, to make low-bandgap materials. Such materials have led to dramatic OPV efficiency gains. We have computed the electronic and optical properties of more than 200,000 molecules or polymers, with the results imported into a database with a searchable web-based interface. The computed energy levels and absorption spectra were calculated using various electronic structure methods and basis sets and the effectiveness of various methods in comparison to experiment was determined. A variety of data mining and informatics methods were applied to the database and the design rules inferred from these analyses for low band gap materials will be discussed.

The goal of this research is to synthesize the linear and hyperbranched polythiophene derivatives containing diketopyrrolopyrrole (DPP) as linking groups, and to investigate the thermal, optical, electrochemical, and photovoltaic properties of those derivatives. Poly(3-hexylthiophene) (P3HT) was also synthesized for comparison in this study. All polymers were synthesized via the Universal Grignard metathesis with high regioregularity >96%. The thermal characteristics of polymers were investigated by differential scanning calorimeter and thermogravimetric analyzer, showing a first-stage weight loss at about 300 ^oC; in addition, polymers containing DPP groups possess less weight loss than P3HT after heating, indicative of enhanced thermal stabilities. The UV-vis absorption spectra of the DPP-containing polymers are similar to that of P3HT in film state. Besides, polymers containing DPP groups show distinct attenuation in fluorensencent emission, indicating that excitons are not easy to recombine to emit light in those materials; in other words, there are more opportunities for carriers to transport to both electrodes to increase current. The electrochemical properties of polymers were analyzed by cyclic voltammetry, revealing that introduction of DPP groups may result in decreasing HOMO and LUMO levels of polymers. All synthesized polymers were used as active layers for fabrication of inverted solar devices with the configuration of ITO/ZnO nanorods/ionic PF/polymer:PC₆₁BM/ PEDOT/WO₃/Au, using ZnO nanorod arrays as electron transporting layer, ionic PF as wetting layer, PEDOT as hole transporting layer, and WO₃ as hole extraction layer. Photovoltaic devices based on linear polythiophene series as active layers showed the open-circuit voltage (V_OC) of 0.55-0.58 V, short-circuit current (J_SC) of 8.62-16.21 mA/cm², fill factor (FF) of 37-41%, and power conversion efficiency (PCE) of 1.73-3.74%. Besides, solar devices based on hyperbranched polythiophene series as active layers showed V_OC, J_SC, FF, and PCE values of 0.55-0.58 V, 9.49-11.87 mA/cm², 36-38%, and 2.01-2.49%, respectively.

Bulk heterojunction films, which typically comprise a polymer donor and fullerene acceptor, are considerably stiffer than films of the neat polymer. The increase in stiffness upon blending is dependent on the miscibility of the polymer and the fullerene, and potentially on the details of molecular mixing and microstructure. This paper describes the effects of molecular mixing and microstructure on the tensile modulus of polythiophenes in blends with [6,6]-phenyl C₆₁ butyric acid methyl ester (PC₆₁BM). We also describe the evolution of molecular mixing and morphology—specifically the demixing of the amorphous mixed phase—under strain and subsequent relaxation of the strain utilizing a combination of spectroscopic techniques. Our conclusions include the role of molecular mixing on the tensile modulus of films, the correlation between the mechanical properties of neat polymers and polymer:fullerene blends, and the extent of demixing of mixed phases under strain and its effect on photovoltaic performance. A greater understanding of the ways in which molecular mixing and phase separation dictate the mechanical properties of photoactive blends is of critical importance for the rational design of functional devices and for their stability in outdoor and portable environments.

High short circuit current generating single junction wide band gap polymer solar cells were fabricated using alternating copolymer poly{2-octyldodecyloxy-benzo[1,2-b;3,4-b]dithiophene-alt-5,6-bis(dodecyloxy)-4,7-bis(2,2'-bithiophene-5-yl)-benzo[c][1,2,5]-thiadiazole} (PBDT-AOBT) blended with fullerene derivatives in different weight ratios. The effects of increasing fullerene weight ratios on ordering, morphological changes and charge transport mechanisms were investigated using UV-Vis spectroscopy, atomic force microscopy (AFM) and Photo-charge extraction by linearly increasing voltage (Photo-CELIV) measurements. The UV-Vis absorption spectra showed that increasing PCBM mixing ratio lowers peak absorption wavelength and absorption intensity in visible region. For 1:1 and 1:2 PBDT-AOBT:PCBM ratios, the absorption peak and intensity in visible region remains same while there is a dip around 400nm for 1:1 ratio may be attributed to lack of PCBM content in the blend. When the ratio was increased to 1:3 and 1:4, maximum absorption intensity significantly reduced in visible region from 400 nm to 630 nm. Higher fullerene mixing ratio reduces the interaction between polymer chain which increases bonding and anti-bonding band gap distance and causes less π- π^* transition. From UV-Vis spectra 1:2 ratio was found to be optimal among all other ratios which also reflected in photovoltaic performance in terms of moderate short circuit current density (Jsc) and fill factor (FF). Adding 1,8-diiodooctane (DIO) additive with parent solvent chlorobenzene (CB) into 1:2 blend films of PBDT-AOBT:PC₆₀BM further improved both short circuit current density and fill factor, leading to an increased efficiency to 4.75%. Further changing solvent from CB to chloroform (CF) with DIO additive increased the Jsc from 9.80 mA/cm2 to 12.20 mA/cm2 and reduced the FF from 71.5% to 62% and eventually limit the efficiency to 5.15%. Photo-charge extraction by linearly increasing voltage (Photo-CELIV) measurements showed highest charge carrier mobility in the 1:2 film among all the ratios, which was further enhanced by introducing DIO additive.

Traditionally, photocurrent generation in a typical OPV has been thought to occur predominantly by photoexcitation of the electron donor and subsequent electron transfer to an acceptor material with a higher electron affinity than the donor (the Channel I mechanism). More recently it has been acknowledged that photoexcitation of the electron acceptor can also play a significant role in photocurrent generation, especially in low donor concentration devices. Photoinduced hole transfer or the so-called “Channel II” process occurs when the electron acceptor is photoexcited followed by oxidation of the electron donor. In order to probe the dynamics of the less-studied Channel II process it is desirable to have an electron donor with an absorption profile outside the visible spectrum. To this end, we have developed a series of triphenylamine (TPA)-based dendrimers and poly(dendrimers) possessing good hole mobility to act as hole acceptors and transporters. For the dendrimers we have shown that subtle variations to the surface substituents of these materials lead to changes in the basic physical properties as well as intermolecular interactions and charge mobility, which impacts the device performance. The poly(dendrimers) were prepared using ring opening metathesis polymerization. It was found that the ionization potential and the optical gap of each dendrimer-poly(dendrimer) combination were similar while the hole mobility of the latter was improved by an order of magnitude. In this presentation we will discuss the preparation and properties of the materials and show how they can be used for the unambiguous identification of the Channel II charge generation process in bulk heterojunction blends with PC₇₀BM.

The impact of controlled solvent vapor exposure on the morphology, structural evolution and function of small molecule based bulk-heterojunction solar cells is examined, focusing on the effect of solubility of the fullerene and small molecule in the annealing vapor on morphological evolution and device performance. The results indicate that exposure of this active layer to the solvent vapor controls the ordering of the small molecule and PCBM phase separation very effectively, presumably by inducing component mobility as the solvent plasticizes the mixture. The detailed morphlogy is obtained by small angle nuetron sacttering and neutron reflectometry, which shows that the size of small molecule crystals and PCBM phase separation is strongly associated with the solubility of small molecule and PCBM in the solvent vapor. The judicious choice of solvent vapor, therefore, provides a unique method to exquisitely control and optimize the morphology of small molecule :fullerene mixtures.

Organic photovoltaics (OPV) have seen a strong growth over the past two decades as promising candidates for renewable energy production. Organic solar cells show a number of interesting features, amongst which low cost, (semi-)transparency, light weight, flexibility, improved low-light performance and compatibility to large scale roll-to-roll production. Moreover, the technology allows for a fine-tuning of the color and deposition in different forms and shapes, appealing from an aesthetical point of view. At present, power conversion efficiencies (PCE&’s) exceeding 9% have been achieved for single-junction solution-processed OPV devices through simultaneous photoactive material and device optimization. However, growth toward commercial applications requires devices with long lifetimes, which is one of the remaining challenges for OPV.[1]
As the photoactive layer in polymer:fullerene bulk heterojunction solar cells is comprised of a solution-deposited blend of an organic donor and acceptor material, leading to the formation of a specific peak-performing (nano)morphology, it is susceptible to several degradation mechanisms, either photo-induced or thermally triggered, leading to phase separation and/or degradation of the bulk materials.[2] In previous work, we have focused on two particular methods toward enhanced durability of polymer:fullerene OPV blends to prolonged thermal stress, either by increasing the glass transition temperature of the conjugated polymer or by structural engineering of the side chain pattern.[3,4] In the present contribution, we have combined both approaches by the design and synthesis of a small series of prototype (PCPDTBT) low bandgap copolymers with functionalized side chains and high Tg&’s (160minus;215 °C). Accelerated aging tests (@85 °C) showed that the thermal stability of the donor-acceptor blends can be improved by the insertion of ester or alcohol moieties on the polymer (CPDT) side chains. Efficiencies over 60% of the initial value could be maintained over 600 hours.
[1] Joslash;rgensen, M.; Norrman, K.; Gevorgyan, S.A.; Tromholt, T.; Andreasen, B.; Krebs, F.C. Adv. Mater.2012, 24, 580.
[2] Cardinaletti, I.; Kesters, J.; Bertho, S.; Conings, B.; Piersimoni, F.; D&’Haen, J.; Lutsen, L.; Nesladek, M.; Van Mele, B.; Van Assche, G.; Vandewal, K.; Salleo, A.; Vanderzande, D.; Maes, W.; Manca, J.V., J. Photon. Energy 2014, 4, 040997.
[3] Vandenbergh, J.; Conings, B.; Bertho, S.; Kesters, J.; Spoltore, D.; Esiner, S.; Zhao, J.; Van Assche, G.; Wienk, M.M.; Maes, W.; Lutsen, L.; Van Mele, B.; Janssen, R.A.J.; Manca, J.; Vanderzande, D.J.M., Macromolecules2011, 44, 8470.
[4] (a) Bertho, S.; Campo, B.; Piersimoni, F.; Spoltore, D.; D&’Haen, J.; Lutsen, L.; Maes, W.; Vanderzande, D.; Manca, J., Sol. Energy Mater. Sol. Cells2013, 110, 69; (b) Kesters, J.; Kudret, S.; Bertho, S.; Van den Brande, N.; Defour, M.; Van Mele, B.; Penxten, H.; Lutsen, L.; Manca, J.; Vanderzande, D.J.M.; Maes, W., Org. Electron.2014, 15, 549.

Commercial photovoltaic devices are still primarily based on inorganic materials (for example, silicon, CdTe, CIGS, or III-V group semiconductors).[1] But, despite their high efficiency, these devices show limitations as high production costs and environmental impact. On the other side, devices based on organic semiconducting polymers may enable compatibility with flexible substrates, large-area and low-cost production. During the last years it has been demonstrated that a potential replacement of inorganic components seems possible while achieving reasonably high power conversion efficiencies.[2]
Althought the solar spectrum covers a wavelength range from ultraviolet to near-infrared (NIR), most of the reported components of bulk heterojunction-type organic solar cells show a HOMO-LUMO gap of ge; 2.0 eV. For this reason, the resulting cells only able to harvest visible light, limiting the performance of polymer-based solar cells (PSCs).[3]
We present two novel low band gap semiconducting donor-acceptor polymers showing a broad absorption band that is extended up to 1200 nm. The synthetic strategy uses suitable acceptor and donor building blocks as N-alkyl-4,7-di(thien-2-yl)-2,1,3-benzothiadiazole-5,6-dicarboxylic imide, and 4-(2-ethylhexyl)-4H-dithieno[3,2-b:2',3'-d]pyrrole or (4,8-bis(octyloxy)benzo[1,2-b:4,5-b']dithiophene thus extending the absorption bands into the NIR. In addition, the incorporation of long alkyl side into the polymer backbone provides sufficiently high solubility. Therefore, they can be processed at room temperature.
[1] Dou, L.; J. You; Hong, Z.; Xu, Z.; Li, G.; Street, R. A.; Yang, Y. Adv.Mater. 2013, 25, 6642-6671.
[2] Mayer, A. C.; S. R. Scully; Hardin, B. E.; Rowell, M. W.; McGehee, M. D. Materials Today 2007, Vol. 10, N° 11, 28-33.
[3] Yue, W.; Zhao, Y.; Shao, Shuyan, Hongkun, T.; Zhiyuan, X.; Geng, Y.; Wang, F. J. Mater. Chem., 2009, 19, 2199-2206.

Organic electronic devices have attracted much attention as lightweight, low-cost, and easy-to-process replacements for inorganic devices. In organic devices, polymer morphology such as chain orientation, crystallinity, crystal size, and domain composition critically influences on electronic and optical properties which will ultimately determine the device performance. In the present study, we suggest the use of mass producible nanoimprint lithography as a method to change molecular orientation as well as domain composition and its applications to organic solar cells. We fabricated nanorods or nanowells of conducting polymers with different chemical structure, dimension, and crystallinity based on patterning with soft PFPE templates. GIWAXS measurements showed different chain orientation and crystallinity which are quite different from the bulk crystallinity of conducting polymers. We adopted P3HT nanorods with face-on orientation or low bandgap polymers (LBPs) for the ordered heterojunction organic solar cell architecture. We were able to control the chain orientation as well as to fine-tune the domain composition based on nanoimprinted P3HT or LBP nanorods coupled with thermal annealing to control the mutual diffusion between P3HT (LBP) and PCBM. Finally, we could establish a clear relationship between nano-morphology, micro-electrical property (which was characterized by conductive AFM) and macro-device performance. Based on the structure-device relationship, we optimized the morphology of ordered organic solar cells and obtained more than 3 times higher power conversion efficiency than bilayer organic solar cells.

Benzo[1,2-b:4,5-b&’]dithiophene (BDT) derivatives are among the most widely used electron-donating units in polymer backbone for polymer solar cells (PSCs). The most commonly used derivative in BDT family is the 4,8-dialkoxy-substituted BDT due to earlier usage and easier synthesis than others and some polymers of dialkoxyBDT show excellent PCEs. To improve the photovoltaic performance, structural modifications of BDT by introducing novel substituents were performed by many research groups. By replacing dialkoxy group to dialkyl, dialkylthiol or dithiophene, HOMO levels of related polymers become deeper and Voc of the related PSCs are increased. But as a trade-off of increased Voc, fill factors of the dialkyl, dialkylthiol or dithiophene incorporated BDT polymer based PSCs are decreased. Here we present newly designed and synthesized novel BDT based polymers and their application to PCSs.

Thin film polymer photovoltaics based on blend system of electron donor and electron acceptor for photoactive layer are receiving extensive academic and commercial interest due to their unique advantages such as low processing cost, solution processability, light weight, and flexibility on the plastic substrate. For the polymer photovoltaics to become a viable energy source in near future, charge carrier mobility and light absorption should be improved. To improve mobility of donor polymers, rigid and coplanar building blocks have been usually introduced into the polymer backbone, because their highly planar π-conjugated system leads to efficient π-orbital delocalization and finally, enhance charge carrier transport property. Due to rigid and coplanar fused rings, thiazolothiazole (TzTz) unit shows extended π-electron system and strong π-π stacking. Thienylenevinylene (TV) also has good planarity and an extended π-conjugation because of the presence of a vienylene spacer between two thiophene uints. In addition, for increasing absorption of solar spectrum, one way is to design low-band-gap polymer which consists of alternative donor and acceptor units in its conjugated backbone. Thiazole is a widely used electron-deficient fused ring due to the electron-withdrawing property of imine group (C=N), which makes the TzTz-TV copolymers acts as a donor-acceptor system. In this presentation, we report synthesis of TzTz-TV copolymer and two other comparable materials to check the variation of planarity in polymer structures and characterized all three polymers by measuring thermal analysis, UV-Vis absorption, cyclic voltammetry, x-ray diffraction measurement. Additionally, device performance of organic thin film transistor was measured in order to explain the relationship between backbone planarity and charge carrier mobility of the polymer. Finally, thin film polymer photovoltaic cells were fabricated using three polymer:PCBM systems and their performance was investigated.

Porphyrin and its derivatives have been intensively studied because of their importance in the fields of photochemical, photobiological, and photovoltaic applications. For the development of inexpensive and renewable energy sources, the functionalized porphyrins are promising candidates for the material source in solar cells, in terms of a natural light-harvesting structure with a large π-conjugation on the carbon-nitrogen macrocyclic body.
Here we demonstrate the photovoltaic effect based on the variation of π-conjugation length of π-conjugated porphyrin derivatives containing Zn-porphyrins with thiophene units at the meso positions. For the realization of effective conjugation of system, the porphyrins were further modified by introducing ethynyl linkers between the porphyrin framework and the thiophenes. The light-absorption characteristics were gradually changed by extending the thiophene units and enhanced Q-band intensities were observed. After fabricating OPVCs, the current density-voltage (J-V) curves and the spectral responses of photocurrent were measured. In order to study the transport of charge carriers in active materials, the exciton generation rate and dissociation probability were further analyzed. The crystallinity of the π-conjugated donor material plays an important role in charge generation, and the planarity of the porphyrin derivatives can improve the power conversion efficiency (PCE) of OPVCs.

In spite of the cost of silicon solar cells decreasing in recent years, there is considerable interest in solar cells with lightweight and flexible forms, where organic semiconductor-based photovoltaic cells (OPVs) have the potential to offer low cost, lightweight and flexible devices. At this stage there is still much to be improved for OPVs as the efficiencies are still lower than their inorganic counterparts.
Polymeric donor materials have been widely studied in comparison to non-polymeric compounds in OPVs. The main drive in polymer materials development has been to create narrow optical gap polymers, which are usually mixed with fullerene acceptors to form bulk heterojunction devices. In recent years, polymeric donors have emerged with efficiencies reaching 10%¹ with quite a range of materials now providing efficiencies of around 7%. However, the problem with semiconducting polymers is that control of the regioregularity, polydispersity, and molecular weight from batch-to-batch is not a simple process.
Non-polymeric chromophores possess certain advantages over polymeric materials such as the relatively ease in synthesis and purification. The optical properties of well defined chromophores can also be fine-tuned more precisely, compared to polymers, where even a small structural change can alter either the optical or physical properties to a great extent. In this context, diketopyrrolopyrrole (DPP) based small molecules have shown promising results with devices efficiencies reaching around 5#8210;6%.^2,3 Key properties of the DPP core unit are strong π-π interactions, a relatively high electron affinity, and relatively stable Highest Occupied Molecular Orbital (HOMO) and Lowest Unoccupied Molecular Orbital (LUMO) levels. These properties have been the main drivers in their use in OPVs. To facilitate favorable π-π interactions, recent work on solution processable non-polymeric DPP-bithiophene units has involved incorporation of different low ionization potential and high electron affinity end groups, allowing for enhanced intermolecular interactions between adjacent molecules for improved charge transport.
In this presentation our new approaches to engineering non-polymeric DPP-bithiophene donors by incorporating different electron accepting end groups will be presented. Their structure, photophysical, thermal and electrochemical properties of the new DPPs, and device performance (>4% power conversion efficiencies) will be discussed.
References
1. Green, M. A., Emery, K., Hishikawa, Y., Warta, W. and Dunlop, E. D., Prog. Photovolt: Res. Appl. 22, 701-710 (2014).
2. Liu, J., Sun, Y., Moonsin, P., Kuik, M., Proctor, C. M., Lin, J., Hsu, B. B., Promarak, V., Heeger, A. J. and Nguyen, T.-Q., Adv. Mater. 25, 5898-5903 (2013).
3. Lin, Y., Ma, L., Li, Y., Liu, Y., Zhu, D. and Zhan, X., Adv. Energy Mater. 3, 1166-1170 (2013).

A series of two-dimensional (2D) conjugated polymers containing benzodithiophene (BDT) and difluorobenzothiadiazole (2FBT) units, namely PBDT2FBT-T₁, PBDT2FBT-T₂, PBDT2FBT-T₃, and PBDT2FBT-T₄, were designed and synthesized with four different chromophoric side chains; π-conjugation lengths of the 2D-conjugated polymers were systematically extended along side-chains functionalized with mono-, bi-, ter-, and quarterthienyl. We found that two-dimensionally extended π-conjugation of conjugated polymers improves the light absorption property both in low- and high-energy absorption bands. In addition, these π-conjugated side chains efficiently increase the electronic density and interchain aggregation of conjugated polymers, leading to an enhanced charge transport ability by developing an interpenetrating nano-fibrillar morphology with interdigitated polymer packing. Our results revealed a noticeable enhancement of power conversion efficiency of polymer solar cells with dimensional and topological optimum of π-conjugated side chains; the PCE values increased up to 6.48 % from 2.17 % without any post-treatments, processing additives, or optical spacers.

In order to realize the full potential of organic photovoltaics (OPV) as cheap, light, and flexible energy harvesting devices challenges with respect to limited conversion efficiency and lifetime have to be overcome. Since thermal energy under operating conditions is not sufficient for spontaneous dissociation of the initially formed excitons, additional energy must be provided through a electronic band transition at donor:acceptor interfaces, the bulk heterojunction morphology needs to provide sufficient interfaces between donor or acceptor domains of ca. 10 nm domain size as well as pathways for the dissociated electrons and holes to travel to the respective cathode and anode electrodes during charge extracshy;tion.
To deepen the understanding of morphology evolution in bulk heterojunction (BHJ) P3HT:PCBM organic photovoltaics (OPVs) system by thermal treatment, we have analyzed domain-size-dependent interfacial energy values from coarse-grained Molecular Dynamics (MD) modelling and then employed the functional relationship between interfacial energy and domain size in Monte Carlo simulations (MC) of the morphology evolution. Thereby initial conditions associated with optimal interfacial surface area, continuous volume, as well as domain sizes and spatial distributions of the phase separated domains were identified, confirming earlier suggestions that a 1:1 P3HT:PCBM blend ratio has the potential to exhibit the most efficient morphology for exciton dissociation and charge transport. Our simulations then reveal how pre-seeding of P3HT crystal at the anode side prior to the annealing process can be instrumental to pin the formation of P3HT at the favourable electrode especially when seeding exceeds a threshold of 10% surface coverage, while denser seeding patterns beyond the threshold did not improve the active layer morphology further. Seeding also favourably increases the local dimensionality of the intradomain pathways for dissociated charge carriers in the BHJ systems from 2.5 to 2.7, thus enhancing charge transport efficiency, while the accompanying increase in domain size from 14 nm to ca. 20 nm somewhat deteriorates charge dissociation.
It should be noted that a one-sided donor seeding does not yield the desired dominance of the acceptor phase at the opposite side of the active layer. The observed trilayer depth profile implies that the commonly used thickness 100 nm of the active layer is not ideal for ensuring that donor and acceptor phases dominate at opposite ends of the active layer. Further simulations on crystal seeding, such as including PCBM seeding near the cathode, could thus further improve charge collection efficiency. The combination of MD and Monte Carlo method used in this work can equally be applied to novel donor:acceptor materials that have potential for improved V_oc through tuned LUMO energy levels or extended lifespan.

Microphase-separated block copolymers composed of electron donor and acceptor blocks may provide morphology control to address many of the challenges in organic photovoltaics. Fully conjugated block copolymers, with careful design of the donor and acceptor blocks, can self-assemble into periodic nanostructures beneficial for charge photogeneration. Previously, we have demonstrated that poly(3-hexylthiophene)minus;blockminus;poly- ((9,9-dioctylfluorene)-2,7-diyl-alt-[4,7-bis(thiophen-5-yl)-2,1,3-benzothiadiazole]-2prime;,2Prime;-diyl) (P3HT-b-PFTBT) can self-assemble into 10 nm lamellae with alternating electron donor and acceptor domains. The resulting solar cell performance is approximately 3% which clearly outperforms blends of the same component obtained due to the self-assembly of the block copolymers. Nevertheless, one of the challenges in controlling the self-assembly of fully conjugated block copolymers is controlling the interplay between crystallization of the P3HT block and microphase separation between the donor and acceptor. To this end, we have examined the kinetics of the morphological evolution during two processes: solution casting and thermal annealing. By using in-situ wide angle and small angle grazing incidence X-ray scattering to monitor crystallization of the P3HT blocks and microphase separation between the two blocks, we identify the crucial parameters that are the driving force for the formation of the lamellar structure. We find that during film drying, P3HT crystallization happens on a much faster time scale than phase separation of the two blocks but the crystallization is significantly suppressed with respect to neat materials, enabling the microphase separation to proceed at time scales after crystallization of P3HT takes place. This enables the mesoscale structure to develop during processes such as thermal annealing, because self-assembly of the lamellar structure takes place before the crystallization of P3HT is complete.

Fully conjugated block copolymers have the potential to overcome many of the limitations of mixtures and blends as photoactive layers in solar cells; furthermore, they may serve as model systems to study fundamental questions regarding optoelectric properties and charge transfer. However, the synthesis of fully conjugated block copolymers remains a challenging issue in the field. We have optimized the two-step synthesis of poly(3-hexylthiophene)minus;blockminus;poly-((9,9-dioctylfluorene)-2,7-diyl-alt-[4,7-bis(thiophen-5-yl)-2,1,3-benzothiadiazole]-2prime;,2Prime;-diyl) (P3HT-b-PFTBT), which is composed of Grignard metathesis for polymerization of P3HT followed by chain extension through a Suzuki-Miyaura polycondenstation. We find that the concentration of the Grignard reagent is critical for end-group control such that P3HT is terminated by H at one end and Br at the other. Furthermore, we can utilize an asymmetric feed ratio of monomers for the Suzuki-Miyaura reaction to minimize the amount of uncoupled homopolymers and to control the molecular weight of the second block. We demonstrate that we can utilize these strategies for the synthesis of block copolymers beyond P3HT-b-PFTBT.

Interface property is an important issue in developing high efficiency polymer solar cells. The interface property could have significant impact on charge extract efficiency, governing the device performance. For this sake, we systematically select a series of thioaromatic molecules of proper lengths to tune the surface characters of the metal-oxide array in inverted ITO/ZnO-nanorod/poly(3-hexythiophene):(6,6)-phenyl C₆₁ butyric acid methyl ester (P3HT:PCBM)/Ag devices. In addition to physically improving the compatibility between ZnO-nanorod and polymer blend contact junctions, those conjugated thioaromatic molecules modulate the nanostructured donor and acceptor percolated networks without significantly affecting the photon absorption efficiency and the exciton dissociation rate. It is found that there is a trade-off between the surface energy of the selected molecules and the surface roughness of the subsequent deposited film. The balanced charge transport architecture can be achieved by using thioaromatic molecules of lower surface energy along with a high surface roughness induced in the successive deposited film; i.e., longer, linearly arranged aromatic rings. As high as 80% enhancement in the power conversion efficiency of the engineered device has been achieved. The obtained result can be generalized to other hybrid polymer-nanocrystal systems as well.

Recent requirements for low cost energy solutions have led to increased research in the area of organic photovoltaics (OPVs), on account of their potential flexibility, low cost and large scale manufacture. The success of these devices, however, has been hampered by low carrier mobilities, resulting in poorly performing devices. Whilst bulk properties are regularly tested; the conduction properties of local morphology, known to have a large impact on device performance, are rarely considered. These have seldom been tested at temperatures beyond room temperature, even though this is a factor that is known to influence device performance [1]. Nanoscale studies, thus far, have not employed the use of in-situ heating, leaving the question of nanoscale morphology and carrier transport dependence on temperature an open question.
Here, local charge carrier mobilities are quantitatively measured in hole only devices; since poly-3-hexyl-thiophene (P3HT) is a quintessential component of the ‘workhorse&’ OPV cell of the polymer community, P3HT:PCBM (Phenyl-C61-butyric acid methyl ester). Using conductive atomic force microscopy (c-AFM) we demonstrate the correlation of localised space charge limited current (SCLC) measurements, and those of material properties such as Young&’s modulus and surface adhesion. We use traditional tip-surface contact mechanics, namely the JKR model, to measure the tip-surface contact area therefore defining a quantitative current density. Based on the work of Reid et al. [2], modified Mott-Gurney models are fitted to the measured current density as a function of bias (J-V) measurements. The same contact mechanics yield surface values of Young&’s modulus and adhesion [3]. These properties are probed across a range of temperatures in-situ, under a continual nitrogen flow. This not only enables device capability to be realistically tested in various operating environments, but allows for the investigation of electrical and physical properties under annealing conditions. We demonstrate that annealing not only increases the hole mobility, as previously shown by Li et al. [4], but changes the physical morphology of the film. The correlation of these properties may lead to a deeper understanding of the nature of charge transport in conjugated polymers. Such methods can be readily applied to other systems.
[1] W. Bagienski and M. C. Gupta, Sol. Energy Mater. Sol. Cells, 2011, 95, 933-941.
[2] O. G. Reid, K. Munechika, and D. S. Ginger, Nano Lett., 2008, 8, 1602-1609.
[3] V. L. Popov, Contact Mechanics and Friction, Springer Berlin Heidelberg, Berlin, Heidelberg, 2010.
[4] G. Li, V. Shrotriya, Y. Yao, and Y. Yang, J. Appl. Phys., 2005, 98.

Benzo[1,2-b:4,5-b&’]dithiophene (BDT)¹ and analogous asymmetric molecules (e.g. 9,9-dialkylfluorene² and 4,4-dialkyl-4H-silolo[3,2-b:4,5-b&’]dithien-2-yl³) have been used extensively in polymer and small molecule materials for organic photovoltaic (OPV) cells. The solubilizing groups of the asymmetric analogs have directionality and generate molecular geometries that often contribute to columnar packing in the solid state, which can have significant impact on charge transport in the material⁴. In order to impart this directionality to molecular organic semiconductors (OSC&’s) containing BDT units, we have devised a path to asymmetrically substituted BDT. For photovoltaic studies, both donor and acceptor molecules containing this unit were synthesized as a comparison to literature semiconductors^2,3a. To demonstrate the connection between symmetry of molecules and device functionality, two large molecular OSC&’s, similar to literature donor molecules⁴, were synthesized employing a central unit of either symmetrically or asymmetrically substituted BDT. The use of a symmetrically substituted core unit should restrict the solid state ordering to a slip-stack or 2-D brickwork motif, while asymmetric substitution may lead to columnar packing motifs. Key synthetic steps, physical properties and device characteristics of these molecules are illustrated.
1. (a.) Huo, L. and Hou, J. Polym. Chem. 2011, 2, 2453-2461. (b.) Li, Y.; Pan, Z.; Miao, L.; Xing, Y.; Li, C.; Chen, Y. Polym. Chem. 2014, 5, 330-334.
2. Gui, K.; Mutkins, K.; Schwenn, P.E.; Krueger, K.B.; Pivrikas, A.; Wolfer, P.; Stutzmann, N.S.; Burn, P.L.; Meredith, P. J. Mater. Chem. 2012, 22, 1800-1806.
3. (a.) Fang, Y.; Pandey, A.K.; Nardes, A.M.; Kopidakis, N.; Burn, P.L.; Meredith, P. Adv. Energy Mater. 2013, 3, 54-59. (b.) Zhou, H.; Zhang, Y.; Mai, C-K.; Collins, S.D.; Nguyen, T-Q.; Bazan, G.C.; Heeger, A.J. Adv. Mater. 2014, 26, 780-785.
4. Zhugayevych, A.; Postupna, O.; Bakus, R.C., II; Welch, G.C.; Bazan, G.C.; Tretiak, S. J. Phys. Chem. C2013, 117, 4920-4930.

The recent surge of enthusiasm in bulk-heterojunction (BHJ) organic solar cells (OSCs) has been driven by the goal of fabricating flexible and light-weight solar cells via facile, low-cost solution processing techniques. The power conversion efficiencies (PCEs) of OSCs have improved rapidly, mainly due to the development of a variety of high-performance donor polymers with near-ideal optical and electronic properties. Small molecule organic solar cells (SMOSCs) have also received significant attention because they are easily synthesized and purified, yield well-defined structures, offer good batch-to batch reproducibility, and provide reproducible solar cell performances. Several strategies for designing organic donors that may be used in high-performance OSCs have been established. These strategies involve (1) forming a conjugated donor-acceptor (D-A) molecular backbone to achieve broad and strong light absorption bands in the visible and near-infrared regions; or (2) incorporating planar molecular structures that pack closely to form crystalline structures, enhance the charge carrier mobility, and yield a large short-circuit current (J_SC). The DTBDT unit is a linearly fused derivative of benzodithiophene, with an enlarged planar area that can enhance the structural ordering of the materials, thereby introducing superior intermolecular charge transport properties. In this presentation, we designed and synthesized benzo[1,2-b:4,5-b&’]dithieno[3,2-b] thiophene (DTBDT)-based small molecules for the fabrication of solution-processable organic solar cells (OSCs).

Water-based polymer nanoparticle dispersions (solar paint) offer the prospect of addressing two of the main challenges associated with printing large area organic photovoltaic devices; namely how to control the nanoscale architecture of the active layer and eliminate the need for hazardous organic solvents during device fabrication. In this paper we review progress in the field of nanoparticulate organic photovoltaic (NPOPV) devices and future prospects for large scale manufacture of solar cells based on this technology.

The performance of organic solar cells consisting of a donor/acceptor bulk heterojunction has rapidly improved over the past few years. Major efforts have been focused on developing a variety of donor materials to gain access to different regions of the solar spectrum as well as to improve carrier transport properties. On the other hand, the most utilized acceptors are still restricted to the fullerene family, which includes PC₆₁BM, PC₇₁BM and ICBA. In comparison to the existing high performing donor materials, the molecule pool of acceptors is still extremely limited, thus inhibiting the further development of organic solar cell in fundamental research and practical application.
Despite intensive research focused on creating an efficient all polymeric solar cells for the past decade, only three different recipes with PCE only reaching 2% have been reported. In recent one year, there have been further improvements to the PCE which were up to 5.8%. In all these reports, the best performance devices are still fabricated via halogenated solvents, such as chloroform, chlorobenzene or dichlorobenzene. Such toxic solvents are expensive and difficult to handle in large scale printing process for organic solar cell fabrication. We report an all polymer solar cell with PCE exceeding 5% fabricated via non-halogenated solvent with small amount of additive. The solubility of the polymer in toluene was enhanced by our novel polymer side chain engineering approach. The phase separation size of the two components in the blend are fine tuned by the concentration of additives. The thermal stability of such solar cell is tested. The devices can be heated up to 250 ^oC and the PCE still keeps up to 4.8%. To the best of our knowledge, this is the highest performance all polymer solar cell with PCE up to 5% fabricated with non-halogenated solvents.

Conjugated polymer nanoparticles (NPs) recently gained large interest as a new class of functional nanomaterials offering novel possibilities from fundamental investigations to eco-friendly processing. For bulk heterojunction organic photovoltaics, the formulation of NPs as confined packages of active layer blend materials provides the possibility to study the morphology-dependent electro-optical properties in depth on the nanoscale. Here, the NPs can be regarded as intermediate systems between bulk films and single molecules, as they offer the functionality of bulk materials but without significant heterogeneity.^[1,2]
For the first time, eco-friendly water-based blend NP dispersions of the low band gap donor polymer poly[[9-(1-octylnonyl)-9H-carbazole-2,7-diyl]-2,5-thiophenediyl-2,1,3-benzothiadiazole-4,7-diyl-2,5-thiophene-diyl] (PCDTBT)^[3] and the fullerene acceptor phenyl-C₇₁-butyric acid methyl ester (PCBM[70]) were prepared using the miniemulsion technique^[4,5,6] and characterized for their relevance in the field of opto-electronic devices. The optical properties of the NP dispersions with different PCDTBT:PCBM[70] ratios were studied using steady-state absorption and emission as well as time-resolved fluorescence spectroscopy in a time range from nanoseconds to hundred femtoseconds in order to achieve insight into the polymer conformation and the interaction with PCBM[70]. These experiments revealed the quenching of PCDTBT by the fullerene, possibly by photo-induced charge separation. The morphology of the different NPs was studied using advanced TEM characterization. For the first time, analytical electron microscopy was applied to obtain structural information down to 2 nm resolution employing STEM in combination with electron energy-loss spectroscopy (EELS). This combination referred to as STEM spectral imaging (STEMSI) was used to acquire spatially resolved low energy-loss spectra across single particles.
Finally, it has to be emphasized that the presented blended nanoparticles based on PCDTBT and PCBM[70], materials which are usually processed with hazardous solvents, have the clear advantage of being water-based dispersions. This opens possibilities for a truly eco-friendly processing of state-of-the-art organic bulk heterojunction solar cells.
[1] Z. Hu, D. Tenery, M.S. Bonner, A.J. Gesquiere, J. Lumin.2010, 130, 771.
[2] T. Kietzke, D. Neher, K. Landfester, R. Montenergo, R. Güntner, U. Scherf, Nat. Mater.2003, 2, 408.
[3] S. Wakim, S. Beaupré, N. Blouin, B. Aich, S. Rodman, R. Gaudiana, M. Leclerc J. Mater. Chem.2009, 19, 5351.
[4] K. Landfester, Angew. Chem. Int Ed.2009, 48, 4488.
[5] A. Ethirajan, K. Landfester, Chem. Eur. J.2010, 16, 9398.
[6] S. Ulum, N. Holmes, D. Darwis, K. Burke, A.L. D. Kilcoyne, X. Zhou, W. Belcher, P. Dastoor, Sol. Energ. Mat. Sol. C. 2013, 110, 43.

The industrial fabrication of organic solar cells is often hampered by toxic solvents that require strong safety precautions. Whereas the use of chlorinated solvents or toxic hydrocarbons is feasible in the lab, their use in large scale printing processes would lead to enormous operational costs, which conflicts with the goal of cost effective production. In this work, we fabricate organic solar cells from nanoparticle dispersions in alcohol. For the absorber layer, we disperse, investigate and use P3HT:ICBA nanoparticles in environmentally friendly dispersion agents such as ethanol. For the preparation of the dispersions, we intentionally omitted any stabilizers that otherwise remain in the active layer, negatively affecting the device performance. In an inverted solar cell architecture comprising a nanoparticulate P3HT:ICBA layer, the power conversion efficiency approaches 5% thereby matching the performance of reference devices that were fabricated from dichlorobenzene. Atomic force microscope images of the P3HT:ICBA surface visualize the merging of the nanoparticles upon thermal annealing. Irradiation intensity dependent current measurements show a gradual change from bimolecular recombination towards monomolecular recombination at higher annealing temperatures within the bulk-heterojunction. The universality of this approach allows the use of other common organic photo-active materials for the fabrication of solution processable organic solar cells from non-hazardous solvents. [Adv. Mater., doi:10.1002/adma.201402360 (2014)]

We present a new photoactive polymer BBTI based on benzo[1,2-b:3,4-b&’:5,6-d&’]trithiophene (BTT) and 2,1,3-benzothiadiazole-5,6-dicarboxylic imide (BTI). The synthetic design allows for alkyl chains to be introduced on both the electron-rich and electron-deficient components, which in turn allows for rapid optimization of the polymer&’s alkyl chain substitution pattern. As a consequence, the optimized polymer shows high performance with a maximum efficiency of 8.3% in organic photovoltaic devices processed in a commercially viable fashion without the need for solvent additives, annealing or device engineering.
To the best of our knowledge, the PCE of 8.3% for BBTI achieved by simple solution processing from a single solvent is the best OPV device performance reported to date in a standard device configuration (conventional or inverted) without the use of any solvent additives or annealing steps. We believe that the aspect of simple processing conditions and thus a robust and reproducible protocol for device fabrication is an important consideration when designing novel photoactive materials and obviously of great significance for the commercialization of organic solar cells.

Solution processed organic solar cells are a promising next generation energy source and have the potential to be manufactured by the cost-competitive roll-to-roll printing process. Therefore, intensive research activities have been made recently and state-of-the-art device efficiency has been improving rapidly. However, most of these devices have been made by processes that are not scalable, typically spin coating, and there has been a considerable gap between the record efficiency of lab scale devices and large scale devices produced by scalable methods. Here, we present two scalable coating methods. We have used the spray coating method as a rapid material screening tool. By having automated position control, various deposition conditions including blend ratio, thickness and feed ratio can be screened in single experiments. We have also used a 3D printer based slot die coater as a lab-to-fab translation tool. The 3D printing platform provide automated control of x,y,z-positioning, coating speed, acceleration and temperature. The controllability allows us to mimic printing conditions in typical roll-to-roll manufacturing processes. The 3D printer based slot die coater was used to produce small scale printed organic solar cells with over 6% efficiency. The process was translated to produce a 47.3 cm² module with 4.56% efficiency. The processes developed in a batch process have been transferring to roll-to-roll printer. Recent progress on the translation process will be presented. The techniques developed for organic solar cells have been adopted to perovskite solar cells, which is another promising solution processed solar cell technology with rapidly increasing record efficiencies. Printing processes for all layers except electrodes have been developed and over 12% efficiency for a printed perovskite is achieved via roll-to-roll compatible processes. Progress toward fully roll-to-roll produced perovskite solar cells will be also presented.

The current efforts to improve the OPV device performance almost exclusively employ spin-coating of the active materials in a protected atmosphere, with efficiencies now reaching 12% for small area devices on the order of millimeter squares. The fabrication conditions by which these very high electrical efficiency cells are currently produced generally cannot be simply transferred to a large-scale roll-to-roll industrial production scheme. The efficiency discrepancy between organic solar cells produced in the laboratory and large-scale devices manufactured in a roll-to-roll printing process need to be addressed before the commercialization of large area printed OPV devices. Correlating the process conditions, the morphology and electrical performance of the OPV devices are critical for solving the challenge of mass-producing high efficient large area organic solar cells.
To solve the above mentioned grand challenge, we developed a mini roll-to-roll compatible printing setup for organic solar cells with the capability to follow the film formation during solvent dry in situ with small and wide angle X-ray scattering at Stanford Synchrotron radiation facility. By using this set-up, time resolution down to 10ms was achieved to probe the drying kinetic and crystallization process of the organic semiconductor materials. This set-up also allows to use multiple inks that being delivered to the roll-to-roll printer head with different composition of the active layer between the donor and acceptor materials. We printed a variety of different BHJ OPV materials (both polymer fluorine systems and all polymer systems) on the flexible ITO/PET substrates and characterized the print process using in situ small/wide angle X-ray scattering (SAXS, WAXS). Wide range process parameters were investigated in details during R2R coating process, including solvent quality, drying temperature, shearing speed, blend ratio of donor and acceptor materials, additives, in order to obtain the optimized OPV morphology. By extracting the peak intensity, full widths at half maximum (FWHM) and peak position, rich information about the drying process as well as dried film are obtained. Finally the morphology of the OPV devices are correlated to electrical performance of the device (e.g. V_OC , Jsc , FF, PCE), thus shed light on the best protocol for drying process to obtain the highest possible efficiency for the R2R coating of OPV materials on flexible substrates.

The organic solar cells have reached such a level of efficiency that one can start thinking of commercialization. Transforming cells into modules in the consequent next step. In this abstract we present a combination of subtractive and additive laser processes to introduce freedom of form interconnection in semi-finished solar cells. A complete OPV stack excluding the top electrode is deposited full area on a substrate followed by subtractive laser process to define the cells and by an additive laser process to make interconnects.
The subtractive laser process involves layer selective ablation of OPV layers from the substrate. On a full area covered solar cell, a P1 and P2 scribe are performed. General laser ablation processes from the top are sensitive to particle generation and corresponding shorts. The novelty in this approach is to come from the back where higher power and gravity enable cleaner surfaces with lesser particle generation.
Subsequently the laser induced forward transfer (LIFT) is applied to deposit dielectric and conductive pastes. LIFT is an additive process which has shown a lot of potential in multilateral deposition. The main advantage of this technique is that higher solid content materials can be deposited which is normally not possible with ink jet printing. The dielectric paste provides the necessary electrical isolation required during the interconnection process. The conductive paste ensures the electrical interconnection between the cells and the external contacts.
By integrating both subtractive and additive laser processes with the 2D freedom of laser, this approach enables complete flexibility and freedom of design for modules on large area.
This process has been shown with P3HT:PCBM based modules on glass/ITO. It is however not limited to a specific substrate, bottom electrode or photoactive layer. ZnO, P3HT:PCBM and PEDOT were subsequently spin coated on a glass slide with full area ITO. As a first step, the LIFT of the Ag top electrode was evaluated. Optimal annealing temperature and time of the deposited silver were studied with small area solar cells. Efficiencies of 3% which is comparable to an evaporated Ag electrode, were reached.
Next, the laser ablation process as well as the LIFT of the dielectric paste were investigated. By laser ablation with a frequency tripled ( 355nm) Nd:YAG, Coherent Talisker picosecond laser, and illuminated from the back, the P1 and P2 scribe were selectively opened. Closed dielectric paste lines could be deposited within the precision of 20 µm and a thickness of around 150 µm. This corresponds to an interconnection width of 500 µm with these initial settings.
Finally we combined the subtractive laser ablation with the additive LIFT to create functional modules. As proof of principle, we built a 4-cell module of 3 cm² with expected voltage and current, and low leakage current. The fill factor was still a bit low due to non-optimal cell design.

Organic photovoltaic cells (OPV) have made tremendous progress in power conversion efficiency (PCE) over the past few years. While the state-of-the-art OPV cells based on multi-junction structures have achieved a PCE of >11%^[1], the scaling of OPV cell performance with area, which is essential element in their practical use, has not been systematically studied, particularly for multi-junction OPV cells. Here we investigate the performance of multi-junction OPV cells as a function of device area. Single, tandem, triple- and four-junction OPV cells were fabricated via vacuum thermal evaporation with active areas ranging from 0.01 to 1 cm². The single junction OPV cell with a tetraphenyldibenzoperiflanthene (DBP):C₇₀ planar-mixed heterojunction structure shows a PCE loss of 15% for 1 cm² cell compared to 0.01 cm² cell while the tandem, triple- and four-junction cells exhibit significantly reduced PCE loss with area. The efficiency loss for single junction and multi-junction OPV cells was analyzed using an equivalent circuit model. We find that the sheet resistance of indium tin oxide (ITO) which increases with device area is the primary current limitation in large area cells. The lower photocurrent but higher voltage in multi-junction cells leads to the reduced resistive losses compared to single junction cells with their higher photocurrent but lower voltage. We find that metal sub-electrodes reduce the series resistance by 50% for 1cm² cells. Our results suggest that the multi-junction OPV cells can efficiently reduce the loss in PCE with area compared to an analogous single junction cell, making multi-junction cells particularly promising for large area devices. With our understanding on the scaling of OPV cells, we further designed and fabricated 5x5 cm² modules with 1cm² sub-cells. We have achieved a high yield of 100% for sub-cells and efficiency variation within 10% across the module.
References
[1] X. Che, X. Xiao, J. D. Zimmerman, D. Fan, S. R. Forrest, Advanced Energy Materials, in press.

The transfer from hero solar cells, produced in the lab, to entirely printed, high efficiency modules has so far been proven to be a difficult challenge. Many obstacles are hidden in the up-scaling process, all contributing to a prominent loss in power conversion efficiency.
In order to tackle the challenge, which will decide whether or not printed photovoltaics will be commercially successful, we investigate both the optical (through a transfer matrix formalism method) and the electrical (through analytical simulations) losses taking place in the up-scaling process. These analyses have to be complemented by a proper choice of high quality electrode materials. These are, e.g., low-temperature solution processable silver inks with high reflectivity (up to 98% in the visible range) and low sheet resistance (0.1 ohm/sq) for opaque modules. For semitransparent electrodes, silver nanowires (transmission higher than 90% at 550 nm and sheet resistance lower than 10 ohm/sq) and high performance, ITO-free, sputtered dielectric/ metal/ dielectric electrodes (T around 90% at 550 nm and sheet resistance < 8 ohm/sq) are employed. This allows us, together with ultra-fast (pulse duration of about 300 fs), high resolution (higher than 20 µm) laser structuring to open the way to a zero-loss transition from lab cells towards large area modules. New, commercially available and efficient materials, such as pDPP5T-2, DPPTT-T, PTB7 and a gen-2 donor from Merck Chemical, will be introduced. Based on these materials, large area modules (> 35 cm²) were R2R processed and an overall geometric fill factor higher than 95% could be achieved, with total area efficiencies of around 70% of the reference cells.
At last, our approach towards the proper formulation of green inks for active layers will be shown: film formation control and relatively fast drying at low (< 100 °C) temperatures which guarantee the formation of a controlled composite microstructure in the active layer.

Transparent electrodes are an important part of future low cost photovoltaic cells. In the last years a number of different approaches has been investigated to offer roll-to-roll processable alternatives to the commonly used TCOs at competitive costs.
We have studied a number of materials and technologies for application in small molecule OPV cells, which will be presented here. The performance, growth[1] and stability of thin metal films (Ag, Cu) including different seed layers and metal alloys like Ag:Ca (Rs=27.3 Omega; /sq. and T= 84.3% at 580 nm including the substrate) were investigated [2,3]. Additionally, metal nanowire percolation type electrodes of silver (10.5 Omega; /sq. and T= 84.8% at 550 nm)[4,5] and copper nanowires (14.2 Omega; /sq. and T= 79.1% at 550 nm)[6] have been manufactured by spray and dip coating and integrated into efficient organic PV cells. Finally, we will show the use of carbon nanotubes and doped small molecules (n-C60)[7] as transparent electrodes and discuss the applicability of network type electrodes for OPV.
If time permits, the influence of the electrode on the device stability and its barrier properties [8] depending on the growth conditions will be presented as well.
[1] F. Nehm, S. Schubert, L. Müller-Meskamp, K. Leo, Thin Solid Films 556 (2014) 381.
[2] S. Schubert, J. Meiss, L. Müller-Meskamp, K. Leo, Adv. Energy Mater. 3 (2013) 438.
[3] S. Schubert, L. Müller-Meskamp, K. Leo, Adv. Funct. Mater. (2014), available online
[4] C. Sachse, L. Müller-Meskamp, L. Bormann, Y.H. Kim, F. Lehnert, A. Philipp, B. Beyer, K. Leo, Org. Electron. 14 (2012) 143.
[5] F. Selzer, N. Weiszlig;, L. Bormann, C. Sachse, N. Gaponik, L. Müller-Meskamp, A. Eychmüller, K. Leo, Org. Electron. (2014) , available online
[6] C. Sachse, N. Weiszlig;, N. Gaponik, L. Müller-Meskamp, A. Eychmüller, K. Leo, Adv. Energy Mater. 4 (2014).
[7] S. Schubert, Y.H. Kim, T. Menke, A. Fischer, R. Timmreck, L. Müller-Meskamp, K. Leo, Sol. Energy Mater. Sol. Cells 118 (2013) 165.
[8] H. Klumbies, M. Karl, M. Hermenau, R. Rösch, M. Seeland, H. Hoppe, L. Müller-Meskamp, K. Leo, Sol. Energy Mater. Sol. Cells 120 (2014) 685.

Silver nanowires (AgNWs) have been considered as one of the most promising materials as transparent electrodes for various optoelectronic devices because of their outstanding optoelectronic properties. AgNWs as window electrodes for single junction polymer solar cells (PSCs) have routinely achieved comparable efficiencies to the ITO based counterparts. We present here the exploration of AgNWs as intermediate electrode for the construction of multi-junction PSCs.
Series-connected tandem PSCs have advanced rapidly in recent years. The performance of the tandem devices is substantially dependent on the photovoltaic performance of the subcells. Despite its importance, a valid method that can directly detect the photovoltaic performance of the subcells is so far lacking. We will show firstly that AgNWs can be used as middle electrode for facile characterization of the photovoltaic behavior of the subcells. We introduce a thin layer of AgNWs between the solution-processed charge recombination layers without compromising the photovoltaic performance of the tandem devices. Due to the accessibility of the AgNWs, the J-V and EQE characteristics of the subcells can be independently measured. This work provide a facile tool to investigate the correlation between the performance of tandem devices and the constituting subcells which is of high importance for further optimization of series-connected tandem devices.
Secondly, we will demonstrate completely solution processed parallel tandem PSCs using silver nanowires (AgNWs) as intermediate charge collecting electrode in combination with a rational interface design. Our monolithically fabricated parallel tandem devices showed high FFs of ~60% and the cumulatively added J_sc extracted from the subcells suggesting no resistance losses. The as-designed intermediate layers are mechanically robust and compatible with most of the polymers we used. In addition, our calculations indicate that compared to the series-connected counterparts, parallel tandem solar cells can achieve comparable high efficiencies but with a more flexibility in selecting material combinations.
Triple-junction solar cells are promising device architectures for further improving the efficiency of photovoltaic devices. Due to the presence of the stringent current-matching criteria, recent demonstrations on series-connected triple-junction PSCs showed very limited efficiency gain compared to their double-junction counterparts. We will finally propose and demonstrate a novel triple-junction architecture which combines a series/parallel interconnection. Series-connected bottom two subcells are connected in parallel with a third subcell using AgNWs as intermediate electrode. The entire triple-junction device is processed from solution in ambient air. Due to the excellent functionality of the intermediate layers, the triple junction shows high FF of ~60% and Voc of ~1 V.

Thin film solar cells suffer from reduced absorption across the solar spectrum as compared to state-of-the-art silicon based solar devices. In order to increase the efficiency of thin film photovoltaic devices it is therefore of paramount importance to increase the absorption cross section of the active layer. A large range of various techniques have therefore been established over the last years, where colloidal structures and entities play an important role. In particular colloidal structures based on noble metals such as Au or Ag, have attracted a lot of attention due to the field enhancement associated with their plasmonic resonances. Another important step for the future development of thin film solar cells is the replacement of the transparent ITO electrode with other more abundant and potentially flexible interfaces. Quite interestingly, both properties (plasmonic light management and transparent electrode) can be combined by the fabrication of metallic nanomesh structures, which are in the focus of this contribution.
We demonstrate the facile, large-area fabrication of metallic nanomesh structures based on the self-assembly of colloidal monolayers. We assess the rich optical properties of such highly ordered and periodic structures, where we aim to give a full understanding of different modes, which can be local and propagating surface plasmons, scattering, or grating diffraction. For this we employ angle- and polarization-dependent UV/Vis spectroscopy.
We tune the periodicity of these nanomesh arrays from a few 100 nm up to 1.5 µm and thereby address the entire solar spectrum. Based on the detailed understanding of the optical properties of such nanomesh structures, we integrate them into photovoltaic cells. We investigate, which optical mode and structural entity is most promising for the realization of an ITO-free photovoltaic device.

The high conductivity of metals has stimulated interest in metal nanowire (NW) networks and meshes as alternative transparent electrodes to replace indium-tin-oxide (ITO). We have recently shown that two-dimensional nanowire networks can outperform ITO both in terms of transparency and sheet resistance (van de Groep et al., Nano Letters 12, 3138-3144 (2012)). Here, we employ soft-imprint nanolithography to transfer this small-area concept into large-area applications of NW networks. We demonstrate the first centimeter-scale NW network based functional devices and show the unique combination of both mode-matched light trapping and charge collection in a single multifunctional layer.
First, we use the facile fabrication of large area NW networks to systematically vary NW width (55 - 130 nm, 30 nm high) and pitch (300 - 1000 nm in 100 nm steps) and study the influence on the optical transmittance and sheet resistance. We demonstrate strongly reduced sheet resistances (8.7 - 27.5 Omega;/sq) compared to ITO (34.3 - 63.6 Omega; /sq) while maintaining high optical transmittance (45 - 92 % averaged transmission, weighted for AM1.5 solar spectrum photon density). Tuning the pitch and width allows the selection of a desired resistance/transmittance combination that can be optimized for a specific application.
Next, we demonstrate the first centimeter-scale printed NW network based functional devices. We use NW networks on glass substrates to fabricate P3HT-PCBM polymer solar cells in a superstrate configuration. While this material is not chosen for its high efficiency, it serves as a well characterized model system. The result of this work is generic, and applicable to all thin film devices. First, 30 nm of PEDOT:PSS conductive polymer are spin coated onto the NW networks, followed by 230 nm P3HT-PCBM. Thermal evaporation of 1 nm LiF, 200 nm Al and 500 nm Ag finishes the devices.
We show that the application of NW networks as transparent conductor induces no current leakage, as the open-circuit voltage is ~550 mV and the fill-factor ~0.66, equal to that of the ITO reference cell. The short-circuit current is 7.35 - 7.65 mA/cm², compared to 9.37 mA/cm² for ITO. This difference can readily be eliminated by using thinner wires. Finally, we observe clear peaks in the external quantum efficiency (EQE) in the weakly absorbing spectral range if the NW network is present.
We use angle-resolved EQE measurements to demonstrate that these peaks originate from the NW network scattering of light into guided modes in the absorber layer, increasing the absorption efficiency. The modal dispersion shows good correspondence with calculated guided modes. By tuning the NW pitch and dimensions, the wavelength and coupling efficiency can be optimized. This demonstrates how engineered 2D NW networks can combine polarization-independent mode-matched light trapping and charge collection in a single multifunctional layer.

Semi-transparent (ST) organic solar cells with potential application as power generating windows are studied. The main challenge is to find proper transparent electrodes with desired electrical and optical properties. In this work, this is addressed by employing an amphiphilic conjugated polymer PFPA-1 modified ITO coated glass substrate as the ohmic electron-collecting cathode and PEDOT:PSS PH1000 as the hole-collecting anode.
Different light trapping techniques are developed for the ST solar cells for enhanced absorption. By stacking two or more ST solar cells on top of each other, and connect them in series or parallel, we show that a stacked ST solar cell can perform more efficiently than a standard solar cell based on the same active materials. By employing an echelle grating structure in combination with the ST solar cells, we demonstrate an increased photocurrent generation up to 24%, compared to the ST solar cells with a highly reflective silver planar mirror. We have also developed efficient dielectric scatterers based on a mixture of TiO2 nanoparticles and polydimethylsiloxane (PDMS). The dielectric scatterer is shown to act as an efficient light trapper for the ST solar cells and an excellent photocurrent balancer for tandem solar cells. The characteristics of polydimethylsiloxane, such as flexibility and the ability to stick conformably to surfaces, also remain in the dielectric scatterers, which makes the demonstrated light trapping configuration highly suitable for large scale module manufacturing of roll-to-roll printed organic single- or tandem-junction solar cells.

Organic photovoltaic (OPV) devices are attractive as they are amenable to low-cost and low-temperature roll-to-roll solution processes suitable for flexible substrates. The dominant choice for transparent electrodes of OPVs is indium-tin oxide (ITO) due to its high transparency (> 90%) and low sheet resistance (< 100 Omega;/square). However, ITO films are brittle and have high costs associated to manufacturing, raw material, as well as, material waste control, which negates the cost savings associated to use of flexible substrates. Among all the potential alternatives of ITO, graphene is considered as one of the most promising candidates. A monolayer graphene, a 2-dimentional material, has notable transparency (97.8%), as well as, good electrical function under bending and stretching. However, to replace ITO with graphene, a key issue is to reduce the sheet resistance of graphene. The lowest reported sheet resistance of about 30 Ohm/square is only for multiple layers of graphene [1]. In most of the literature, sheet resistance values of monolayer graphene are still around several hundred or thousand Ohms [2-4].
This work demonstrates centimeter scale monolayer graphene grown by chemical vapor deposition and transferred to glass and flexible substrates for fabrication of OPV devices. Subsequent deposition of nanocrystalline MoO₃, ZnO, and Al-doped ZnO layers on the monolayer graphene significantly improves the sheet resistance of the hybrid transparent electrode. In addition, chemical doping is applied to further enhance the conductance of the SLGs. Electrical and optical properties of the combined transparent electrodes are investigated. OPV devices on both rigid and highly flexible substrates using the hybrid electrodes are fabricated, showing OPV performances, such as power conversion efficiency, comparable to the control devices made on ITO-coated glass. The stability of OPV devices under different bending conditions is also demonstrated.
[1] Nair, R.R., et al. Science320: 1308 (2008)
[2] Suk, J.W., et al. ACS Nano, 5: 6916-6924 (2011)
[3] Li, X., et al. Nano Lett., 9: 4359-4363 (2009)
[4] Li, X., et al. Nat. Nanotechnol.3: 538-542 (2008)

Polymer solar cells (PSCs) have attracted considerable attention over the past several years due to their unique advantages of low cost, light weight, and great potential for realization of flexible and large-area devices. Great efforts have been paid to achieve PSCs with high efficiency, easy processing, and good stability. Toward these goals, inverted PSCs based on ITO cathode have become an optional device configuration. It has been demonstrated that inserting a cathode interlayer between the ITO cathode and the active layer is quite necessary to realize high efficiency inverted PSCs. In comparison to inorganic ZnO interlayer, using a hydrophilic polymer interlayer not only affords easy processing without high temperature thermal annealing, but also supplies inverted PSCs with high efficiency and good stability.
In this presentation, we would demonstrate three strategies to develop hydrophilic polymers as efficient cathode interlayer for inverted PSCs. The first one is using ultra-high molecular weight polymer interlayer. Water-soluble polyacrylamide with a molecular weight of 3 million as cathode interlayer could decrease the work-function of ITO and hence lead to suitable energy level alignments for efficient electron collection. Obvious enhancements of efficiency as well as air-stability over 1 year were found. The interlayer polymer with giant molecular size and numerous hydrophilic side groups would facilitate the formations of large and intimate single-molecular binding area with the ITO substrate, and afford more reliable interfacial contacts to resist possible moisture- and/or stress-induced delamination. For the second strategy, organic/organic combined cathode interlayers based on a water-soluble non-conjugated polymer PDMC and an alcohol-soluble conjugated polymer PFN were introduced. The water-soluble high molecular PDMC layer facilitated to form strong adsorption on ITO cathode while the alcohol-soluble PFN could provide both compatibly interfacial contacts with the bottom PDMC interlayer and the upper organic active layer. With PTB7 as the polymer donor, inverted PSCs based on the combined interlayers showed the highest efficiency over 9% and the most excellent air-stability during 2 months, if compared a single PDMC or PFN cathode interlayer. For the third strategy, alcohol-soluble conjugated polymers with large band-gaps (~3.5 eV), such as poly(1,3-phenylenes), poly(3,6-fluorenes), and twisted poly(2,7-fluorenes), were synthesized for the cathode interlayer of inverted PSCs. The large band-gap characteristics supplied very good solar light incidence, and better long-term stability of an interlayer due to “self-protection” under solar irridiation may be established. Inverted PSCs based on the cathode interlayers could show efficiencies around 9%.

Carrier selective interfaces are crucial for high performance bulk heterojunction (BHJ) organic photovoltaic (OPV) devices. The conducting polymer blend PEDOT:PSS [poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate)] is the most used hole transport layer in OPV devices. However, its selectivity for holes has been attributed to the presence of a PSS-rich film that forms on the top surface of the PEDOT:PSS film. In standard architecture devices this PSS-rich layer can react with polymers in the BHJ forming a doped p-type interlayer, which enhances the selectivity of the PEDOT:PSS - active layer interface. In inverted devices this doped interlayer does not form due to a lower concentration of PSS at the interface of the PEDOT:PSS and the active layer. As a result, the PEDOT:PSS contact may be less selective for holes in inverted devices than it is in standard devices. To improve the selectivity of PEDOT:PSS in inverted devices, polymer interlayers are inserted to create a doped interlayer between the BHJ and PEDOT:PSS. Optical measurements show that the interlayer polymers dope the BHJ polymer. The dark and light current-voltage characteristics show increased compensation voltage in devices with the polymer interlayers, indicative of increased selectivity. This translates to increased power conversion efficiency due to improvements in open-circuit voltage and fill factor.

Molecule-doped conjugated polymers have been proved to be one of powerful methods to improve the performance of polymer-based electronics compared to the pristine state: increasing the conductivity and mobility, decreasing the injection barrier as well as effectively suppressing geminate charge recombination at donor-acceptor interface via charge transfer excitions in bulk heterojunction (BHJ) photovoltaic. The influence of the molecule doping upon interface energetics is critical for understanding the process, in turn a prerequisite for enhancing power efficiency conversion. Here we will explore the formation mechanism of the molecule-doped conjugated polymer/electrode interface to span the range of low-intermediate-high doping concentration, and conclusively establish its universal energy level alignment holding for all doped organic semiconductor systems except intermolecular hybridization.[1, 2] We will also highlight the effect of the energetics at the BHJ on device efficiency, especially open circuit voltage loss attributed to trap-assisted recombination via integer charge transfer states using a variety of experimental techniques and interface modeling.[3]
[1] Q. Y. Bao et al, Adv. Mater. Interface (accepted)
[2] Q. Y. Bao et al, Adv. Energy Mater. 2013, DOI: 10.1002/aenm.201301272
[3] Q. Y. Bao et al, Adv. Funct. Mater. 2014, DOI: 10.1002/admi.201401513

Tandem polymer solar cells have seen a remarkable boost in power conversion efficiency (PCE) in the past few years, with some of the state-of-the-art devices achieving greater than 10% efficiency. Tandem architectures are typically comprised of a two or more complimentary photoactive layers separated by a thin inter-connecting layer. Effective design of this inter-connecting layer is of critical importance to device performance, but is not trivial because the layer must be optically transparent and efficiently harvest electrons and holes from photoactive layers of disparate materials. We have developed a new tandem cell inter-connecting layer comprised of a thermally deposited Cr and MoO₃ bilayer, which is 65-70% transparent over a range of 300 nm to 800 nm and achieved device efficiencies of more than 6% when paired with a high band-gap (e.g. PCDTBT) and a low band-gap (e.g. iso-indigo based P(T3-iI)) polymer. With further optimization of process conditions, layer thickness and interface engineering, higher efficiencies are expected from this system.

Achieving high performance in organic optoelectronic devices increasingly relies on electrical and optical matters of both the active materials and the attendant charge injection/extraction layers. Even with the best-performing active materials, unwelcome limitations in device performance frequently arise when attempting to achieve maximal light management within the device. Light management can typically involve either the in-coupling of incident radiation, in the case of photovoltaic cells and photo-detectors^1,2, or out-coupling in the case of light-emitting diodes. The challenge in reaching highest device performance is to optimize simultaneously both the electronic attributes along with the overall refractive index profile, frequently hampered by additional constraints. One such example in solution-processed devices lies with the thickness of the ubiquitous conductive poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) layer - often kept at a minimum due to optical losses. Such constraints place limits when effective management of the light distribution within the structure is desired - recalling it is the optical length, i.e. the product of the refractive index and physical thickness that is useful here. In this regard developing materials with the ability to tune optical properties (e.g. the refractive index) whilst maintaining sufficient conductivity, would prove useful for contemporary device design.
We present here a versatile, solution-processable inorganic/organic hybrid material, composed of PEDOT:PSS and titanium oxide hydrates^3,4. We show that both the optical and electronic properties can be manipulated and tuned through a change of composition. A number of characterization techniques are used to demonstrate material systems exhibiting higher refractive indices and lower absorption loss than PEDOT:PSS while still maintaining a good conductivity. The new hybrid materials can be readily deposited from solution to provide high-quality coatings with negligible roughness scattering. We discuss the underlying chemistry of this novel class of materials, the thin-film fabrication and processing supported by a number of optical and electrical characterization techniques. We will also demonstrate their operation as interlayers in multilayer optoelectronic devices such as organic solar cells and light-emitting diodes.
1. N. Yaacobi-Gross, N. D. Treat, P. Pattanasattayavong, H. Faber, A. K. Perumal, N. Stingelin, D. D. C. Bradley, P. N. Stavrinou, M. Heeney and T. D. Anthopoulos, Adv. En. Mater. (2014; DOI: 10.1002/aenm.201401529).
2. Z. Tang et al., Adv. En. Mater. 2 (12), 1467-1476 (2012).
3. M. Russo, M. Campoy-Quiles, P. Lacharmoise, T. A. M. Ferenczi, M. Garriga, W. R. Caseri and N. Stingelin, J. of Pol. Sci. Part B: Pol. Phys. 50 (1), 65-74 (2012).
4. M. Russo, S. E. J. Rigby, W. Caseri and N. Stingelin, Adv. Mater. 24 (22), 3015-3019 (2012).

Triple junction polymer solar cells can offer higher power conversion efficiencies than single junction or tandem polymer solar cells. To achieve this goal three different organic semiconductors with finely balanced optical band gaps and high performance are required. With the successful development of semiconducting polymers that have optical band gaps between 1.4 and 1.6 eV and power conversion efficiencies up to 9%, there is now a need for efficient materials that have relatively wide (i.e. about 1.75 eV) and very low (i.e. about 1.1 eV) optical band gaps. Such materials have received less attention in recent years, because they are less attractive for single junction cells. New results towards efficient materials with band gaps above 1.7 eV and below 1.2 eV will be presented.

During last years polymer solar cells have attracted a lot of attention as a promising technology for renewable energy production because of their potential low cost, reduced weight and appealing aesthetic features. By using the bulk heterojunction concept and state of the art low bandgap electron donor polymers, the 10% power conversion efficiency threshold was recently surpassed.[1,2] An established technique to further boost the efficiency is the addition of fluorine atoms onto the backbone of the conjugated polymers. Devices based on these materials often show an increased open-circuit voltage, decreased charge recombination and improved active layer morphology. In this respect, quinoxalines are attractive (electron poor) building blocks because they allow for selective introduction of either one or two fluorine atoms.[3] Moreover, they also enable smooth fine-tuning of solubility and processability via the alkyl side chain pattern.
In this contribution, we report on the incorporation of fluorinated quinoxaline derivatives in push-pull low bandgap copolymers - in combination with electron rich 4H-cyclopenta[2,1-b:3,4-b&’]dithiophene (CPDT) or N-acyl-dithieno[3,2-b:2&’,3&’-d]pyrrole (DTP) units - and the investigation of the optical, electronic and photovoltaic properties of the resulting polymers and blends (in comparison with the non-fluorinated analogues).
[1] He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photon. 2012, 6, 551.
[2] You, J.; Dou, L.; Yoshimura, K.; Kato, T.; Ohya, K.; Moriarty, T.; Emery, K.; Chen, C.-C.; Gao, J.; Li, G.; Yang, Y. Nat. Commun.2013, 4, 1446.
[3] Chen, H.-C.; Chen, Y.-H.; Liu, C.-C.; Chien, Y.-C.; Chou, S.-W.; Chou, P.-T. Chem. Mater.2012, 24, 4766.

In recent years, great headway has been made in the development of efficient polymer donors across the community, with published power conversion efficiencies (PCE) >8% in bulk-heterojunction (BHJ) solar cells with fullerene acceptors (single cells), and PCEs >10% in tandem devices. In most reports, the polymer donor involves elaborate repeat unit and side chain patterns, and deviating from those patterns induces substantial drops in device PCE. While the range of polymer design parameters that impact BHJ solar cell performance remains a matter of some debate, our recent developments indicate that the combination of side-chain substituents and the functional groups appended to the main chain critically impacts polymer performance. These effects are particularly pronounced in poly(benzo[1,2-b:4,5-b&’]dithiophene-thieno[3,4-c]pyrrole-4,6-dione) (PBDTTPD) and wide-bandgap analogs blended with phenyl-C61/71-butyric acid methyl ester (PCBM),[1-11] but are also apparent when alternative acceptors are used. Furthering our understanding of how the main chain substitution pattern mediates the interplay between donor and acceptor is a critical step as we look to continue improving BHJ solar cell efficiencies.
[1] P. M. Beaujuge, and J. M. J. Fréchet, JACS 2011, 133, 20009.
[2] C. Piliego, T. W. Holcombe, J. D. Douglas, C. H. Woo, P. M. Beaujuge, and J. M. J. Fréchet, JACS 2010, 132, 7595.
[3] C. Cabanetos, A. El Labban, J. A. Bartelt, J. D. Douglas, W. R. Mateker, J. M. J. Fréchet, M. D. McGehee, and P. M. Beaujuge, JACS 2013, 135, 4656.
[4] J. A. Bartelt, J. D. Douglas, W. R. Mateker, A. El Labban, C. J. Tassone, M. F. Toney, J. M. J. Fréchet, P. M. Beaujuge, and M. D. McGehee, Adv. Energy Mater. 2014, 4, 1301733.
[5] J. Warnan, C. Cabanetos, R. Bude, A. El Labban, Liang Li, and P. M. Beaujuge, Chem. Mater. 2014, 26, 2829.
[6] J. Warnan, A. El Labban, C. Cabanetos, E. Hoke, C. Risko, J-L. Brédas, M. D. McGehee, and P. M. Beaujuge, Chem. Mater. 2014, 26, 2299.
[7] J. Warnan, C. Cabanetos, A. El Labban, M. R. Hansen, C. Tassone, M. F. Toney, and P. M. Beaujuge, Adv. Mater. 2014, 26, 4357.
[8] K. R. Graham, C. Cabanetos, J. P. Jahnke, M. N. Idso, A. El Labban, G. O. Ngongang Ndjawa, B. F. Chmelka, A. Amassian, P. M. Beaujuge, M. D. McGehee, JACS 2014, 136, 9608.
[9] C. Dyer-Smith, I. A. Howard, C. Cabanetos, A. El Labban, P. M. Beaujuge, and F. Laquai, 2014, Submitted.
[10] J. Wolf, F. Cruciani, A. El Labban, P. M. Beaujuge, 2014, Submitted.
[11] M. A. Hamid, J. Wolf, R.-Z. Liang, P. M. Beaujuge, 2014, Submitted.

Polymer solar cells (PSCs) have made a great achievement in the past decades. As a result, a record high power conversion efficiency (PCE) of 9.2% based on single junction PSC and 10.6% based on tandem junctions PSC are reported. Tandem cells by stacking two or more single cells together can absorb more photons from multiple active layers with complementary absorption ranges. In this way, the photon utilization efficiency can be significantly improved and a PCE of 15% can be expected. The challenge is to seek suitable donor materials for those sub cells in terms of their specific requirement, especially for the back cell material due to the trade-off between narrow band-gap and open circuit voltage.
Herein, we have proposed a ternary regioregular copolymer strategy to tune the band-gap and energy levels for PSC applications, which consists of a weak donor component, a strong donor component and a strong acceptor component in each repeat unit. For instance, it is well known that indacenodithiophene (IDT) is a weak donor, cyclopentadithiophene (CPDT) is a strong donor and pyridyl-[2,1,3]thiadiazole (PT) is a strong acceptor. Thus, it is reasonable that the binary copolymer (PIPT-RG) of IDT and PT has a relative wide band-gap and the binary copolymer (PCPDTPT) of CPDT and PT has a very narrow band-gap. Normally, synthesizing random copolymers by mixing these three monomers with various ratios could tune the band-gap from the narrow band-gap (E_g of PCPDTPT) to wide band-gap (E_g of PIPT-RG). According to our previous study, regioregularity significantly affect the material performance in FET and OPV applications. Based on this knowledge, we have synthesized two regioregular binary copolymers PIPT-RG, PCPDTPT and a structurally precise regioregular ternary copolymer PIPCP. The regiorandom counterpart PIPC-RA with the same monomer ratio was also synthesized to compare their chemical properties and device performance. We have found that the band-gap of PIPCP was well controlled (1.47 eV) and the absorption was desirable for back layer in tandem cell applications. Hence the short current is high. Furthermore, the mobility is decent for OPV applications due to the regioregularity. Most importantly, we have obtained surprisingly high V_oc (0.85~0.90 V) based on the PIPCP devices in considering of the low band-gap. Finally we obtained the power conversion efficiency around 6%~7%. These results make the polymer PIPCP is a promising candidate for the future high performance PSCs.

It has been suggested that there is a missing class of materials in organic electronics between the amorphous materials and crystalline organic semiconductors (see Jun-ichi Hanna et al. Thin Solid Films 554, 58-63 (2014)). These solution processed molecular materials would be liquid crystalline with materials performances intermediate between the amorphous and crystalline materials. It has been predicted that nematic liquid crystalline organic semiconductors, offering three dimension transport, should outperform smectic or discotic liquid crystalline materials. We present here the first high performance example of this new class of materials for use in organic solar cells, that is a molecular nematic liquid crystalline hole transport material (BTR) which exhibits excellent bulk mobility leading excellent performance, even for thick films.
The new molecular electron donor material, BTR, with a benzo[1,2-b:4,5-b&’]dithiophene (BDT) core and rhodanine peripheral units was developed and used in OPV devices giving PCEs greater than 9% (max 9.3%, FF from 74-78%). While its pi-conjugated structure is analogous to a high performance compound reported previously (Liu, Y. et al., Sci. Rep.3, 3356 (2013), Zhou, J. et al., J. Am. Chem. Soc.135, 8484-8487 (2013)), the strategic placement of the side chains provided BTR with strong intermolecular interactions, as evidenced by its liquid crystalline behavior. Such interactions were successfully translated into excellent hole transport properties; hole mobilities up to 0.1 cm² V^-1s^-1 (neat film) and 1.6 × 10^-3 cm² V^-1s^-1 (blend film) were recorded by organic field effect transistor (OFET) and space charge limited current (SCLC) methods, respectively. Therefore, BTR-based OPVs with thick active layers (300 to 400 nm) could still afford PCEs of over 8% with high FF of ~70%. This makes BTR a very attractive candidate for roll-to-roll printed OPV modules.

Small-molecule donor/polymer acceptor bulk-heterojunction films with both compounds strongly absorbing have great potential for further enhancements of performance of organic solar cells. By employing a newly synthesized small molecule donor with a commercially available polymer acceptor in a solution-processed fullerene-free system, we report a high power conversion efficiency of close to 4%. Detailed investigation reveals that the record FF value of the small-molecule donor/polymer acceptor system of over 60% is related to the good charge carrier transport properties and low recombination losses; while efficient exciton dissociation/charge transfer comes along with a remarkably low energetic driving force of ~0.06 eV, which ensures the most optimized open-circuit voltage of all organic solar cells studied to date. Despite the more limited exciton dissociation compared to that in a small molecule/fullerene solar cell, the stronger absorption of the N2200 acceptor, allows for a rather comparable photocurrent to be extracted, and thus an overall superior photovoltaic performance.

To date, the role of side alkyl chains in all-polymer solar cells (all-PSCs) is poorly underlined. Here, we demonstrate the dramatic effects of side alkyl chains on the photovoltaic properties of all-PSCs via developing NDI-thiophene copolymers with different lengths of side chain. The blend system of PTB7-Th donor and NDI-thiophene acceptor produces a remarkable power conversion efficiency with 5.96%, which is a record for known all-PSCs so far. The dramatic changes in the morphological behavior and the electron mobility of the polymer blend elucidated the clear trends in the device parameters of the all-PSCs.

Photocurrent generation in organic photovoltaic devices has historically been thought to occur predominantly via photo-induced electron transfer (PET, Channel I) from a polymer donor to a fullerene acceptor. It is now understood that two concurrent processes operate in a working device; PET, as well as photo-induced hole transfer (PHT, Channel II) from the acceptor to the donor.[1, 2]
Fullerene derivatives are the gold standard for acceptor materials in organic photovoltaic devices, as they possess chemical properties that afford excellent charge dissociation efficiencies and electron transport. Drawbacks of the use of fullerenes in such devices include their low extinction co-efficients in the visible spectrum as well as their tendency to form large crystallites in bulk heterojunction blends.
A method to exploit both Channel I and Channel II photocurrent generation pathways in a photovoltaic device is to develop non-fullerene acceptors, which have in principle a much more diverse range of structural motifs. In this presentation we will discuss recent progress in engineering the properties of non-fullerene acceptor materials that can be used in organic solar cells. We will present the development and our latest understanding of materials in terms of their structure, optical and electronic properties, processing, film structure, and device performance.
[1] Y. Fang, A.K. Pandey, A.M. Nardes, N. Kopidakis, P.L. Burn, P. Meredith, A Narrow Optical Gap Small Molecule Acceptor for Organic Solar Cells, Advanced Energy Materials, 3 (2013) 54-59.
[2] Y. Fang, A.K. Pandey, D.M. Lyons, P.E. Shaw, S.E. Watkins, P.L. Burn, S.C. Lo, P. Meredith, Tuning the Optoelectronic Properties of Nonfullerene Electron Acceptors, Chem Phys Chem, (2014). DOI: 10.1002/cphc.201402568

While organic photovoltaic bulk heterojunction devices employing fullerene based electron acceptors are approaching 10% power conversion efficiency (PCE), very few non-fullerene acceptors reach efficiencies above 3 % even with high performance donor polymers.¹ Here we present a promising new class of rhodanine-flanked small molecule acceptors which achieve up to 4.11% PCE using P3HT as the donor polymer, outperforming [60]PCBM in equivalent devices.² These acceptors are synthetically simple and versatile, with wide scope to integrate different chemical functionalities to the molecule in order to alter their electronic, structural and morphological properties. We will demonstrate how these properties can be modified by replacing the central fluorene unit with indenofluorene or indacenodithiophene units, as well as replacing the original C₈ alkyl chains with other linear or branched chains, and the subsequent impact this has on device properties.
The materials display encouraging electron acceptor characteristics, including a large extinction coefficient in the visible region of the spectrum; the ability to accept multiple electrons reversibly; a LUMO which is localised on the outer, sterically exposed portion of the molecule, making it more accessible for electron transfer; and a non-planar structure which prevents the large-scale crystallization often associated with such non-fullerene acceptors. Devices made with P3HT as the donor polymer and one of the new acceptor matierals, FBR, demonstrate a higher V_oc (0.82 V) than the P3HT:[60]PCBM comparison device (0.59 V) and resultant higher PCE of 4.11% compared to 3.53%. Space charge limited current measurements show that the electron mobility of our P3HT:FBR blend is comparable with that of P3HT:[60]PCBM blends. Furthermore, these blends demonstrate improved morphological stability in comparison to P3HT:[60]PCBM devices as observed by optical microscopy.
Ultrafast transient absorption spectroscopy (TAS) is used to study the polaron generation efficiency and recombination dynamics of the P3HT:acceptor blends and this is related to the morphology of the active layers via time-of-flight secondary ion mass spectrometry (TOF-SIMS) and grazing incidence x-ray diffraction (GIWAXS). It is determined that the P3HT:FBR blend is highly intermixed, leading to increased charge generation relative to P3HT:[60]PCBM devices, but also faster geminal recombination due to a non-ideal morphology in which the acceptor does not aggregate enough to create appropriate percolation pathways. Despite this sub-optimum morphology, the device performance is among the highest reported for non-fullerene acceptors employing P3HT, with potential for further optimization due to its inherent synthetic versatility.
[1] Eftaiha, A. a. F. et al.,Journal of Materials Chemistry A2014, 2, 1201; Lin, Y. et al.,Materials Horizons2014, 1, 470.
[2] Holliday, S. et al., 2014, manuscript in preparation

During recent years, intensive researches have been carried out to improve the performance of all-polymer solar cell which are mainly focusing on the energy level and morphology control by backbone engineering and device fabrication process, respectively. On the other hand, side-chain engineering of conjugated polymers has become more and more critical not only for enhancing the solubility but also for controlling the compatibility, crystallinity, morphology, and so on which would further lead to an enhancement of their optoelectronic performance. Recently, we have reported a high performance all-polymer solar cell by utilizing polystyrene (PS) as the side chain. By this polymer side chain engineering, phase separation size could be controlled without fabrication process optimization which shows the new strategy of material design for all-polymer solar cell application.
In order to control the feeding ratio of polystyrene in the donor-acceptor (D-A) type conjugated polymers, the attached PS should have controllable molecular weight and low polydispersity index (PDI). Therefore, we have used commercially available PS with low PDI and controlled molecular weight as a starting material to synthesize PS attached acceptor monomer. However, most of the donor monomers with PS attached are not synthetically accessible by using the commercially available PS. Hence, we have developed a new strategy of utilizing a termination process of living anionic polymerization of styrene to attach low PDI and controlled molecular weight PS to donor monomer. By incorporating less than 10% of the newly synthesized PS attached donor monomer into the D-A type conjugated polymer, we have observed a drastic suppression of the strong aggregation tendency of D-A type conjugated polymers. This suppression is consider to be preferable in the all-polymer solar cell development. Owing to this strategy, systematical study of the effect not only of PS length but also of the position where the PS is attached to the backbone on the device performance could be carried out.

There has recently been a remarkable growth of interest in all-polymer solar cells, in which both the donor and acceptor materials that absorb light and transport charges are semiconducting polymers, and shown a great potential to replace fullerene/polymer devices. Using n-type semiconducting polymers as an acceptor can provide better synthetic flexibility in tuning the electronic structure and optical, charge transport, and photovoltaic properties as well as the environmental stability and durability of the bulk heterojunction solar cells. However, the performance of all-polymer solar cells is still inferior compared to fullerene based solar cells. The lack of suitable n-type semiconducting polymers, low photocurrent, and lack of approaches to optimize polymer/polymer blend morphology are the major factors that limit the performance of all-polymer photovoltaic devices. To overcome these challenges, we have developed a series of new naphthalene diimide (NDI) based acceptor copolymers and their structure-property-performance relationship is studied.^1-5 We investigated the photovoltaic properties of all-polymer solar cells composed of the new acceptor copolymers as the acceptor and poly(3-hexylthiophene) (P3HT) or a thiazolothiazole-dithienosilole copolymer as the donor. The resulting all-polymer solar cells gave power conversion efficiency up to 5 % and provided a new insight into material design for high performance fullerene-free all-polymer blend photovoltaic cells.
1. Hwang, Y. J.; Ren, G. Q.; Murari, N. M.; Jenekhe, S. A. Macromolecules 2012, 45, 9056.
2. Earmme, T.; Hwang, Y. J.; Murari, N. M.; Subramaniyan, S.; S. A. Jenekhe, J. Am. Chem. Soc. 2013, 135, 14960.
3. Hwang, Y. J.; Murari, N. M.; S. A. Jenekhe, Polym. Chem. 2013, 4, 3187.
4. Hwang, Y. J.; Earmme, T.; Subramaniyan, S.; S. A. Jenekhe, Chem. Commun., 2014, 50, 10801.
5. Hwang, Y. J.; Earmme, T.; Subramaniyan, S.; S. A. Jenekhe, Adv. Mater. 2014, 26, 6080.