It has been previously demonstrated that thin single p-n junctions can be advantageous for obtaining higher open circuit voltages (Voc). However, a thin junction may not absorb all the input photons and the reduction in short-circuit current (Isc) typically nullifies the gain in Voc unless complicated architectures or manufacturing processes are engineered. In addition, high performance tunnel junctions have been successfully developed for concentrated photovoltaic (CPV) solar cell applications. High peak tunneling currents and optical transparency are key requirements for such tunnel junctions. CPV solar cells having conversion efficiencies exceeding 40% for concentrations up to and in excess of 1000suns (~100W/cm²), are now routinely obtained, including the quantum dot based QDEC® triple junction CPV cells. These high bandgap tunnel junctions are therefore interesting technological building blocks for other photonic and optoelectronic devices operating with similar photocarrier current densities.
Based on the above building blocks, Azastra Opto Inc has engineered a vertical heterostructure design capable of high efficiencies and high tolerances with respect to the alignment and the non-uniformity of the input light. In this work, we study phototransducer devices intended for an optical input with wavelengths centered near 830nm. A monolithic III-V phototransducer is optimized for applications utilizing a narrow-band light source. It has been demonstrated with high conversion efficiencies based on the vertical stacking a number of partially absorbing GaAs n/p junctions connected in series with tunnel junctions. The novel phototransducers, based on 5 thin GaAs n/p junctions, have been used to obtain an electrical output of up to 2.5W with a Voc output of about 5.8V with a conversion efficiency of 50% or better. The temperature dependence of the device has been characterized and is compared to the temperature behavior of CPV solar cells. Devices with active areas of ~3.4mm² were fabricated and tested with different emitter gridline spacings. The open circuit voltage (Voc) of the electrical output is 5 times or more that of a single GaAs n/p junction under similar illumination. The device architecture allows for improved Voc generation in the individual base segments due to efficient carrier extraction while simultaneously maintaining a complete absorption of the input photons with no needs for complicated fabrication processes or reflecting layers. Fill factor (FF) varying between 88 and 89% have been measured. The properties of the individual thin GaAs n/p junctions will also be discussed for different base thicknesses. Interesting photocarrier properties are observed for the thinner GaAs photovoltaic devices.

Conjugated polymers with metal oxide semiconductor nanocrystals can be one of the most attractive combinations for low cost photovoltaic materials. However, the interfacial charge transfer mechanisms, i.e. key parameters controlling solar cell functions, of such hybrid films has not been explicitly elucidated. In this presentation, we have successfully controlled charge carrier dynamics with two different parameters, i.e. (1) by employing different interfacial molecules, and 2) by fine tuning conjugated polymer structure. For the former parameter, the interfacial charge transfer dynamics for conjugated polymers and ZnO nanoparticle (NP) hybrid were investigated by utilizing sub-micro to millisecond transient absorption spectroscopy. Interfacial modification of ZnO-NP surface has been performed by employing different interfacial molecules including oleic acid, fluorescein, MK2, erythrosine-B and eosin-B; all of these molecules possess a carboxyl functional group attaching to the ZnO-NP surface. Interestingly, compared to the hybrid film without introducing interfacial molecules, all hybrid films with dye capping molecules indicate relatively high charge separation efficiency (>70 %) and prolonged charge separation lifetime extending by one order of magnitude. With respect to the device performance, interfacial modification has enhanced the open circuit voltage owing probably to retardation of charge recombination processes. For the latter parameter, we have developed three different types of conjugated molecular wire attached to the TiO₂ nanoporous film surface. We found that the electron injection efficiency for these hybrid films is relatively high (>90 %) and all polymer attached films have shown extremely long charge separated state lifetimes (half lifetimes: >100 ms). These studies provide promising pathways to continuously improve the overall performance of conjugated polymer/inorganic nanocrystals hybrid solar cells.
Keywords: hybrid solar cell, conjugated polymer, ZnO nanoparticle, TiO₂ nanohybrid films, interfacial modification, charge recombination

Organic photovoltaic (OPV) devices are considered to be a promising addition to inorganic solar cells because they offer the possibility to be fabricated on large scales at a low production cost. Especially the concept of an intermixed Bulk-Heterojunction (BHJ) between organic donor and acceptor materials forms the basis for devices with high power conversion efficiencies (PCE) [1] because it allows to handle the separation of photo-generated excitons efficiently. Since it is a challenging task to investigate the morphology and the internal processes experimentally [2], simulations on different scales can be viable tools to guide the optimization of BHJ cells. One way to model organic solar cells is to solve the Drift-Diffusion (DD) equations. They offer an approach at a macroscopic continuum level, low computational effort, and with good agreement to experimental data [3]. A common approximation of DD simulations is the effective medium approach (EMA) [4], i.e. the lack of incorporation of the real blend morphology. In the EMA, donor and acceptor material are treated as one effective material. The donor/acceptor interfaces to split excitons into polaronic charges are assumed to be everywhere across the photoactive layer and no exciton dynamics are considered. For this purpose, kinetic Monte Carlo (kMC) simulations offer a suitable tool to implement the desired morphology [5]. We have developed a full device kMC simulator for organic BHJ devices [6] able to generate intermixed morphologies and including exciton and charge dynamics. Despite good reproduction of experimental measurements the kMC method comes at high computational cost and is not suitable to simulate large device structures. Thus it is relevant to be able to investigate under which conditions the EMA can approximate the physics of a BHJ and how accurate it is. We will show that this question is related to the short range electric field at the interface between donor and acceptor materials and we highlight the important role of interface electrostatics in controlling most of the fundamental processes in a BHJ.
[1] You, J. et al. A polymer tandem solar cell with 10.6% power conversion efficiency. Nature communications 4 (2013): 1446.
[2] Moon, J. et al. 'Columnlike' structure of the cross-sectional morphology of bulk heterojunction materials. Nano letters 9, no. 1 (2008): 230-234.
[3] Koster, L. J. A. et al. Device model for the operation of polymer/fullerene bulk heterojunction solar cells. Physical Review B 72, no. 8 (2005): 085205.
[4] Fallahpour, A. H. et al. Modeling and simulation of energetically disordered organic solar cells. Journal of Applied Physics 116, no. 18 (2014): 184502.
[5] Meng, L. et al. Dynamic Monte Carlo simulation for highly efficient polymer blend photovoltaics. The Journal of Physical Chemistry B 114, no. 1 (2009): 36-41.
[6] Albes, T. et al. Kinetic Monte Carlo modeling of low-bandgap polymer solar cells. 40th Photovoltaic Specialist Conference (2014): 0057-0062.

Photovoltaic cells based on organic materials represent a promising technology for novel applications such as large-scale integration of energy harvesting devices at low fabrication cost. The device architecture with the currently highest power conversion effciencies (PCE) [1] is based on the bulk-heterojunction (BHJ) structure which comprises an intermix of donor and acceptor materials. However, the PCE of organic BHJ devices needs to be improved for a successful market entrance. Because the BHJ is a complex composition of donor and acceptor molecules it is difficult to get insight in the morphology and the charge dynamics by experiments [2]. Especially loss processes due to electron-hole recombination at the donor-acceptor interface are a key factor for device performance and need to be understood in more detail. Organic materials typically have a low dielectric constant (ε = 3 to 5) and Coulomb forces are not well screened. It is assumed that after exciton separation the electron and hole form a Coulomb-bound charge transfer (CT) state [3] and only after splitting from this state, the charges can move through their respective transport phase to the contacts. In many continuum simulations recombination is treated by the Langevin model which descibes a bimolecular process based on the probability of opposite (free) charges finding each other [4]. Simulations based on the kinetic Monte Carlo (kMC) method make it possible to consider a realistic geometry for the BHJ and to analyze recombination events in detail. We have implemented a full device 3D kMC model and investigated the charge distributions of charges during device operation. Even slight modfication of the dielectric constant shows a major change in where charges are located. For low ε = 3 the interaction acts on larger distances, pulling charges from within their transport phases towards the donor/acceptor interface. This accumulation of charges leads to an increase in recombination. Only increasing to ε = 5 shows that charges are evenly distributed in their transport phases with no accumulation and therefore decreased recombination. If charges are accumulated along the boundary, recombination is not controlled by the probability of finding each other anymore and can therefore not be described by a bimolecular process, i.e. the Langevin model becomes invalid. Recombination is then rate controlled and must be treated fundamentally different. Hence, we can support the theoretical [5] and experimental [6] studies about the development of polymers with high dielectric constants to reduce recombination in organic BHJ devices.
[1] You, J. et al. Nature Communications 2013, 4, 1446.
[2] Moon J. S. et al. Nano Letters 2009, 9.1, 230-4.
[3] Burke, T. M. et al. Advanced Energy Materials (2015).
[4] Fallahpour, A. H. et al. Journal of Computational Electronics 2014, 1-10.
[5] Koster L. J. A. et al. Physical Review B 2005, 72.8, 085205.
[6] Cho, N. et al. Advanced Energy Materials 4, no. 10 (2014).

The role of excess photon energy on charge generation efficiency in bulk heterojunction solar cells is still an open issue for the organic photovoltaic community. After reviewing our recent work done by applying the ultrafast pump probe technique with 20 fs time resolution we report a detailed study on the spectral dependence of the internal quantum efficiency (IQE) for a PCPDTBT:PCBM- based solar cell. The experimental data are obtained combining accurate optoelectronic characterization and comprehensive optical modeling. We demonstrate that photons with energy higher than the band gap of the donor material can effectively contribute to enhance the IQE of the solar cell. This holds true independently from the device architecture, reflecting an intrinsic property of the active material. We point out that the nano-morphology of the bulk heterojunction has a crucial role in determining the IQE spectral dependence: the coarser and more crystalline, the lesser the gain in IQE upon high energy excitation.

Organic-inorganic hybrid perovskites are a special class of semiconductors that have revolutionized the prospects of emerging photovoltaic technologies, in forms of both light harvesters and hole transport materials. These organic-inorganic hybrid perovskite compounds adopt the ABX₃ perovskite structure, which consists of a network of corner-sharing BX₆ octahedra, where the B atom is a divalent metal cation (typically Ge²⁺, Sn²⁺ or Pb²⁺) and X is a monovalent anion (typically Cl^minus;, Br^minus;, I^minus;); the A cation is selected to balance the total charge and it can be a Cs⁺ or a small molecular species. Such perovskites afford several important features including excellent optical properties that are tunable by controlling the chemical compositions, ambipolar charge transport, and long electron and hole diffusion lengths. Solar cells based on MeNH3PbI3 perovskites have progressed significantly and reached power conversion efficiencies of around 20%, approaching the efficiency of commercialized c-Si solar cells. Because of the toxicity of Pb however the search is on for lead free alternatives. In this talk we report on the chemistry of tin-based perovskite compounds their chemical and physical properties as well as phase transitions.

In terms of conversion of efficiency of solar radiation into electricity, which is arguably the most important challenge of an emerging, new solar cell technology, perovskite solar cells have made astonishingly rapid strides over the past five years. There is a however a second, equally crucial, yet less glamorous challenge to be mastered, which is the aspect of stability. For solar cells based on organic-inorganic hybrid perovskites this is certainly one of the most challenging tasks simply because the material itself degrades rapidly when in contact with moisture. The origin for this instability is the hydrophilic organic component of the material. As a first line of defense the top layer of the solar cell - the hole-transporting layer in a n-i-p structure - must therefore be able to minimize moisture ingress. In its most common and most efficient incarnation these types of devices employ a doped organic material. In particular, the dopant Li-TFSI, has been shown lead to an increase in degradation of the perovskite absorber mainly due to its hygroscopic nature. Previously, we have demonstrated that polymer-functionalized single-walled carbon nanotubes embedded in an inert polymer, namely PMMA, matrix can be employed as a highly efficient p-type charge extraction layer. The absence of dopants and the moisture-barrier characteristics of the polymer matrix are credited with the significantly enhanced resilience of perovskite devices against moisture and heat. Here, we show how, by optimizing the individual components of this structure, the solar cell performance (with efficiencies as high as 16.5%) as well as the stability of devices can be significantly improved. The components of this hole-transporting structure include the functionalizing polymer, single-walled carbon nanotubes and a polymer matrix.
The functionalizing polymer wraps in single sheaths around the carbon nanotubes thus individualizing them and making them dispersible in chlorinated solvents. Additionally, the polymer-nanotube interaction leads to p-doping of the carbon nanotubes. We show the importance of this functionalizing polymer in conjunction with the properties of the single-walled carbon nanotubes for efficient charge extraction and high photovoltages. The third component is the encapsulating polymer matrix. Being uncoupled from the actual charge transfer, its role is mainly to block moisture. By choosing materials with a very low water vapor-transmission rate, the stability of devices could further be improved. In thermal, moisture and full sunlight stressing, devices with this structure completely outperform their counterparts with conventional hole-transporting materials such as spiro-OMeTAD.
In conjunction with the right encapsulation techniques, we believe that this hole-transporting structure with strong moisture blocking properties, can be an important innovation for the required long-term stability of perovskite solar cells.

The Organolead trihalide perovskites (OTPs) are shown to be ion conducting materials. We previously reported that the photocurrent direction can be switched repeatedly by applying a small electric field of <1 V/mu;m due to the ion electromigration.^[1] The switchable photocurrent reached ±20.1 mA/cm² under one sun illumination in vertical OTP devices, and the open circuit voltage reached 47 V with 125 tandem devices in lateral structure. However, what kinds of ions can move in the material has not been identified yet, and the ion electromigration on the solar cell performance and stability remains unclear. We will report the direct observation of electromigration of methylammonium (MA) ions in MAPbI₃ perovskite films and the consequent formation of a p-i-n structure in macroscale. ^[2] The activation energy, migration mobility of the MA ions will also be discussed. In addition, we will report the influence of the ion electromigration on the device performance and stability of regular perovskite solar cell device. To our surprise, our results show that the migration of ions under working condition is beneficial for the device performance and stability.
^[1] Zhengguo Xiao et al. Giant switchable photovoltaic effect in organometal trihalide perovskite devices. Nature materials 14, 193-198
^[2] Yongbo Yuan et al. Photovoltaic Switching Mechanism in Lateral Structure Hybrid Perovskite Solar Cells. Adv. Eng. Mater. DOI: 10.1002/aenm.201500615

A new generation of thin-film photovoltaic cells based on organic-inorganic metal halide perovskites has recently emerged with extraordinary power conversion efficiencies. We have shown that the prototypical materials CH₃NH₃PbI₃ and CH₃NH₃PbI_3minus;xCl_x exhibit long charge-carrier diffusion lengths [1], owing to a combination of high charge-carrier mobility and abnormally low bi-molecular charge recombination rates that defy the Langevin limit by at least four orders of magnitude [2,3].
However, a large number of materials adopt the perovskite crystal structure with the general stoichiometry AMX₃. Here we discuss how the parameter essential for photovoltaic operation, such as charge carrier recombination mechanisms and diffusion lengths are altered with substitutions of the organic A cation (e.g. methylammonium versus formamidinium), the metal M cation (e.g. Pb²⁺ or Sn²⁺) and the halide X anion (iodide versus bromide).
For example, changing the metal cation at the M site from Pb²⁺ to the less toxic Sn²⁺ to form CH₃NH₃SnI₃ shifts the optical bandgap from 1.55 eV to 1.3 eV into the range of the ‘ideal&’ single-junction solar cell band gap. Using transient THz spectroscopy, we find that the ultrafast charge-dynamics in CH₃NH₃SnI₃ are dominated by a first-order decay that may originate from electron recombination with an unintentional hole doping density of ~10¹⁸ cm³. We establish an effective charge-carrier mobility of 1.6 cm² V^minus;1 s^minus;1, yielding a charge-carrier diffusion length of 30nm [5]. However, we show that the low effective bi-molecular recombination constant allow for charge diffusion lengths in excess of one micron if mono-molecular processes arising from unintentional doping or charge trapping are to be resolved.
In addition, mixed organic lead iodide/bromide system such MAPb(Br_yI_1-y)₃ and its formamidinium relative FAPb(Br_yI_1-y)₃ have recently gained strong interest as the band gap can be tailored between 1.55 eV (MAPbI₃) and 2.3 eV (MAPbBr₃), which results in the coverage of much of the visible spectrum and paves the way for the development of tandem solar cells. Here we show that while some of the fundamental charge-carrier recombination and transport mechanisms strongly correlate with phase disorder in the system, others reflect the inherent electronic structure of the material.
[1] Stranks, Eperon, Grancini, Menelaou, Alcocer, Leijtens, Herz, Petrozza, Snaith, Science 342 (2013) 341.
[2] Wehrenfennig, Eperon, Johnston, Snaith, Herz, Adv. Mater. 26 (2014), 1584.
[3] Wehrenfennig, Liu, Snaith, Johnston, Herz, Energy Environ. Sci. 7 (2014) 2269.
[4] Noel, Stranks, Abate, Wehrenfennig, Guarnera, Haghighirad, Sadhanal, Eperon, Johnston, Petrozza, Herz, Snaith, Energy Environ. Sci. 7 (2014) 3061.

The unprecedented rapid ascending efficiency of perovskite-structured CH₃NH₃PbX₃ solid-state solar cells have raised a storm of research effort. Despite the success in boosting their efficiency, perovskite solar cells are still facing several critical challenges, including the instability of CH₃NH₃PbX₃, especially in moisture, the use of environment-hazardous lead, the costly and unstable complex organics as hole transport materials, the use of precious metals as back cathode, and the hysteresis in current-voltage scans, the fabrication engineering of good-quality perovskite films, the basic science understanding of the unusual excellence in both charge transport and light harvesting of CH₃NH₃PbX₃, etc.¹ To tackle these critical challenges, we have recent demonstrated that nickel, as an industrial commodity metal with work function of 5.04 eV, can replace gold as the back cathode in perovskite solar cells with competitive performance to gold-cathoded cells. We show that the work function and the conductivity of the back cathode are both fundamentally important factors impacting the photovoltage and fill factor of the perovskite solar cells.^2,3 Furthermore, we show that when two iodides in CH₃NH₃PbI₃ are replaced with two pseudohalide ions thiocyanate (SCN^-), the resulting CH₃NH₃Pb(SCN)₂I perovskite films strikingly rival the conventional CH₃NH₃PbI₃ films in moisture-tolerance as evidenced by their tolerance in 95% relative humidity in air for over 4 hours without significant degradation, in contrast to the CH₃NH₃PbI₃ films that degraded in less than 1.5 hours.⁴ The solar cells based on CH₃NH₃Pb(SCN)₂I thin films exhibit efficiency comparable to that of CH₃NH₃PbI₃-based cells fabricated in the same way. We show that engineering of the precurosry solution can reduce the pinholes in the spin-coated films to improve the efficiency.⁵
1. Liu, X.; Zhao, W.; Cui, H.; Xie, Y.; Wang, Y.; Xu, T.; Huang, F. Organic-inorganic Halide Perovskite Based Solar Cells - Revolutionary Progress in Photovoltaics, Inorganic Chemistry Frontiers, 2015, 2, 315-335. Top Ten most-read Inorganic Chemistry Frontiers articles in Q1 of 2015.
2. Jiang, Q.; Xia, S.; Shi, B.; Feng, X.; Xu, T. Nickel-Cathoded Perovskite Solar Cells, J. Phys. Chem. C.,2014, 118#65279;, 25878-25883. News Report by Yahoo Finance, Solar Novus, and Gigaom.
3. Jiang, Q.; Sheng X.; Li, Y.; Feng, X.; Xu, T. Rutile TiO₂ Nanowires Perovskite Solar Cells, Chem. Commun., 2014, 50#65279;, 14720-14723. Featured as Inside Front Cover Page.
4. Jiang, Q.; Rebollar, D.; Gong, J.; Piacentino, E. L.; Zheng, C.; Xu, T. Pseudohalide-Induced Moisture Tolerance in Perovskite CH₃NH₃Pb(SCN)₂I Thin Films. Agnew. Chem. Int. Ed., 2015, in Press. DOI: 10.1002/anie.201503038. Highlighted as Very Important Paper (VIP) by editor.
5. Xie, Y.; Shao, F.; Wang, Y.; Xu, T.; Wang, D.; Huang, F. ACS Applied Materials and Interfaces, DOI: 10.1021/acsami.5b02705

In order to increase the use of photovoltaic solar cells, these devices need to be produced both more cheaply and with higher power conversion efficiency. We are introducing a new anti-reflection coatings based on silicon dioxide (SiO₂) nanospheres that improve solar cell absorption and ultimately yield high efficiency, low cost devices. The fabricating process is based on a Meyer rod rolling which can be performed at room-temperature and applied to mass production, yielding a scalable and cheap manufacturing process. The deposited monolayer of SiO₂ nanospheres leads to a significant increase in the light absorptivity, on the order of 15-20% based on experiment and calculations, which ultimately increases spectral current density. Due to the field coupling between the SiO₂ nanospheres originated from Whispering Gallery Modes (WGMs) inside each sphere, incident light can be strongly coupled into the high-index absorbing material (Si) and light absorption is increased. Because the layer can be made with an easy, inexpensive, and scalable process, this anti-reflection coating (ARC) is an excellent candidate for substituting conventional ARC technologies relying on complicated and expensive processes.

Perovskite solar cells have rekindled a strong interest in the development of highly efficient and affordable solar cells, owing to excellent materials properties of organic lead halide and its low cost and simple fabrication process. TiO₂ mesoporous layer is an important component of the perovskite solar cell and delivers photogenerated electrons from lead halide to transparent electrode. TiO₂ nanoparticles are widely used in the mesoporous films, however, the partially sintered nanoparticle layer has a problem of carrier trapping at the grain boundaries. In addition, it has been argued that point defects of TiO₂ induce space charge at TiO₂/perovskite interface layer and produce undesirable hysteresis of voltage-current curve. To understand the origin of these problems (i.e. carrier trapping and hysteresis), it is necessary to study the mesoporous layer with reduced grain boundary density and controlled defect concentration.
In this study, we report transport behavior of photogenerated electrons in vertical array of TiO₂ nanorods doped with donor impurity (Nb) and acceptor impurity (Al). Vertically grown single crystalline TiO₂ nanorods reduced electron-trapping sites and impurity-doping changed type and concentration of point defects.
Single crystalline TiO₂ nanorod arrays were grown via a microwave method, which was followed by thermal annealing in different ambience. Nb and Al doping concentration was in a range of 0~0.5% and nanorod length was from 0.25µm to 5µm. Perovskite solar cells were fabricated on the nanorods using a modified-two-step method. Also TiO₂ nanoparticle samples were prepared as a reference cell. Materials properties and device performance were characterized a suite of tools such as scanning electron microscopy (SEM), x-ray diffraction (XRD), electrochemical technique, transient current method and incident photon-to-current efficiency (IPCE).
Comparison of nanorod- and nanoparticle-based photoanode quantitatively shows the effect of the grain boundary trapping on the transport and recombination of photogenerated carriers. High electron density of Nb doped nanorods and low electron density of Al-doped nanorods provide a clue on the role of donor impurities on space charge and hysteresis of I-V curve. When point defects were annihilated by changing the annealing ambience, a difference between Nb doped and Al doped samples was decreased. We also used this fundamental understanding of defects, carriers and space charge to optimize the carrier collection efficiency and suppress the carrier recombination and the hysteresis in the perovskite solar cells. Our results guide the design of mesoporous TiO₂ layer with controlled grain boundary and point defect for high efficiency perovskite solar cells.

As an excellent new class of photovoltaic material, the perovskite solar cells have attracted great attention with the efficiency jumping from 3.8% to 20.1% in the past five years. Recently, ferroelectric properties have been proposed to exist in such inorganic-organic hybrid perovskite. In semiconducting ferroelectric photovoltaic materials system, the interplay between ferroelectric polarization and charge separation/recombination has been found vital in modulating solar cell performance.
In the presentation, we demonstrate the presence of ferroelectric domains in CH₃NH₃PbI₃ by piezoresponse force microscopy and quantify the coercive field to switch the polarization of ferroelectric CH₃NH₃PbI₃. For CH₃NH₃PbI₃ perovskite solar cell, negative electric poling decreases the net built-in electric field, driving potential and width of depletion region inside the absorber layer, which hinders charge separation and deteriorates photovoltaic performance; while positive poling boosts these electrostatic parameters and therefore improves the charge separation inside the absorber. Low coercive field (8 kV/cm) enables the switching of CH₃NH₃PbI₃ polarization during the current density-voltage (J-V) measurement. Forward scan initially activates the negative poling, whereas reverse scan first activates the positive poling, which can lead to the J-V hysteretic behavior. Comparative analysis with a traditional ferroelectric 0.25BaTiO₃-0.75BiFeO₃ solar cell is conducted to confirm the impact of ferroelectric polarization and J-V scanning direction on photovoltaic performance. Switchable photovoltaic effect was achieved by symmetric ferroelectric solar cells. The ferroelectric solar cells based on CH₃NH₃PbI₃ thin film demonstrated power conversion efficiency close to 7%. Our ferroelectric device uses the classical electrode/ferroelectrics/electrode structure (instead of the popular p-i-n structure) and its major driving force for charge separation comes from the ferroelectric field. The ferroelectric polarization provides a new perspective for tailoring the working mechanism and photovoltaic performance of the perovskite solar cells.

Two significant barriers to highly efficient thin-film perovskites / crystalline silicon (c-Si) tandem photovoltaics are imperfect current matching and incomplete light trapping. Recent efforts to address this gap through advanced light trapping structures do not always clearly demonstrate how much useful absorption is enhanced, as opposed to parasitic absorption that cannot improve efficiencies. Even small errors in this regard can substantially degrade experimental tandem efficiencies, which has led to a small gap between single junction c-Si and perovskite/c-Si tandems. In this work, we present a new method suitable for capturing these parasitic losses with accurate dispersion and rapid calculation of losses within a finite-difference time-domain simulation. In order to validate our predictions, we have compared to experimentally-measured parasitic losses in three classical types of c-Si light trapping cells (e.g., periodic, random and plasmonic). For experimental parameters previously employed, we found that we could reproduce the overall absorption curves in each case, and that combining random front texturing with plasmonic nanoparticles (NPs) in back generally displays the minimum parasitic loss. Therefore, in order to approach the performance limits, we proceeded to optimize a perovskite/c-Si cell with a correlated random front texturing and a plasmonic back reflector. It is shown that 600 nm diameter NPs strongly enhance light absorption for wavelengths ranging from 700 nm to 1100 nm, while a correlated random surface also provides much broader absorption enhancement. As an optimized result, 18.30 mA/cm² J_sc is achieved after subtracting 3.74 mA/cm² of parasitic losses. This result appears promising for achieving tandem cell efficiencies exceeding records for single-junction crystalline silicon HIT cells (just above 25%).

This article describes the electrical and physical properties of polysilicon doped with novel N+ and P+ screen printed inks using a thermally activated process. Unique ink formulations for N and P type doping of silicon are evaluated in volume production in order to enable a low cost, high throughput process. Inks can be used with multiple substrate types and form factors. The concentrated doping source combined with thermal drive in and activation results in degenerately doped layers of polysilicon. Inks are semiconductor grade which is demonstrated by their use in fabricating high mobility, low leakage Thin Film Transistor (TFT) devices on 300 mm stainless steel substrates. Reproducible sheet resistance values (700 A polysilicon) can be engineered from levels typically ranging from 200 - 2000 ohm/sq. The additive approach substitutes the use of high capital cost ion implantation and lithography processes. The ink formulation results in screen printed widths capable of ranging from 100-300 um. As both N and P type layers can be printed adjacent to each other, it is critical to prevent cross doping using surface preparation techniques. Post doping cleaning of surfaces can be achieved in-situ or by plasma removal depending on process integration and product considerations. Reproducibility and uniformity data to demonstrate manufacturability in a production environment is shown. In summary, a simple, low cost, high throughput additive process based on proprietary inks that can be used in multiple product flows (CMOS TFT, Solar etc.) is demonstrated.

With an ever-expanding, energy-hungry global population, and climate change strongly linked to the exploitation of fossil fuels for energy, development of the next-generation of environmentally friendly, renewable energy sources is now more critical than ever.
One of the most exciting recent developments in the field of solar energy generation is the emergence of low-cost, high efficiency lead perovskite solar cells. For decades lead has been utilised in energy materials applications, from the lead acid battery, to PbTe in thermoelectric devices, and now perovskite solar cells. Whilst lead is a excellent material for use in these devices, it suffers from one major drawback: high toxicity. Lead exposure has been linked to many physical ailments and developmental issues, particularly in children. Clearly, if lead perovskite photovoltaics are to find application in the energy generation sector, end-of-life recycling of these solar cells, to negate the environmental issues of lead contamination, is vital.
We have demonstrated a low-cost, high-efficiency method for the solution-based electrochemical recycling of lead perovskites using deep eutectic solvents. By dissolving lead perovskite materials, including CH₃NH₃PbI₃ (MAPbI₃), HC(NH₂)₂PbI₃ (FAPbI₃) and CH₃NH₃PbI_3-XCl_X (MAPbI_3-XCl_X), and selectively extracting the Pb through electrochemical deposition we can recycling lead perovskite photoactive layers with Pb extraction rates of up to 99.8%. The solvent can then be re-used after the deposition is complete. The results from ICP-MS, XRD, SEM and EXAFS will be discussed. These results clearly demonstrate the applicability of deep eutectic solvents to the recycling of lead perovskite solar cells.

Carrier multiplication (CM), whereby multiple electron hole pairs can be generated from the absorption of a single photon, offers a strategy to surpass the traditional thermodynamic limits governing the power conversion efficiency in photovoltaics. However, CM is inefficient in bulk semiconductors due to the strict requirement of momentum conservation imposed on bulk band structure, resulting in large thresholds of activation. Semiconductor nanocrystals have been demonstrated to have greatly reduced thresholds and improved efficiencies due to the suppression of translational momentum conservation. Yet, the CM yields afforded by current nanomaterials will not greatly increase the power conversion efficiency of a photovoltaic device, stressing the need to understand the factors governing CM in order to achieve further improvements in the CM efficiency. Here, we explore the role of the geometry of nanocrystals in controlling the multiplication yield. In particular we explore the carrier dynamics in two-dimensional PbSe nano-platelets and one-dimensional hetero-structured PbSe-CdSe core shell nano-rods. Transient absorption and time resolved photoluminescence experiments are utilized to measure the multiplication yield and Auger lifetimes. We find that unlike nano-rods, where strong interactions between carriers induces the formation of Coulombically bound excitons, the nano-platelets exhibit carrier dynamics largely similar to zero-dimensional quantum dots where electrons and holes can be viewed as independent particles. Theoretical considerations indicate that binding energies in rods are enhanced relative to both quantum dots and platelets due to a combination of dielectric and geometrical effects. For the hetero-structured nano-rods, our results suggest that there is a synergistic effect of the increased carrier-carrier interactions due to the one-dimensional nature of charge carriers as well as the extended lifetime of shell localized holes resulting in enhanced CM yields beyond those found in core only nano-rods.

We report the experimental demonstration of a low-cost paradigm for photovoltaic power generation that utilizes a prismatic Fresnel-like lens to simultaneously concentrate and separate sunlight into laterally spaced spectral bands, which are then fed into spectrally matched single-junction photovoltaic cells. The optical element was designed using geometric optics and the dispersive properties of the employed material and its performances were simulated with a ray-tracing software. The device was then fabricated by injection molding, suitable for large-scale mass production, and its properties were experimentally characterized. We report an average optical transmittance above 85% over the VIS-IR range, concentration factors above 3x and a spectral separation in excellent agreement with our simulations. Finally, the system was tested with a pair of copper indium gallium selenide based solar cells. We demonstrate an increase in peak electrical power output of 160% under outdoor sunlight illumination. Given the ease of manufacturability and the potential of the proposed solution, the reported spectral splitting approach provides a cost-effective alternative to multi-junction solar cells ready for mass production.

Models based on continuum elasticity theory are discussed to calculate the necessary thickness of a strain compensation (SC) layer for a superlattice (SL) of strained quantum wells (QW) or quantum dots (QD). Determining the optimal thickness of a SC layer relies on knowledge of the size, shape, and areal density of the QDs, as well as properties of the substrate, QD, and SC material systems. Using and expanding upon previously established work, this paper compares three methods of compensating strained SCs, two of which are specific to QDs. The methods for balancing 3D structures with a 2D SC layer result in a global strain balance condition, however some areas are necessarily over-compensation while the rest of the film is under-compensated. A method of determining the critical number of repeat units before the onset of misfit dislocations based on these two methods are shown and discussed, and then used to further refine the calculated SC thickness to maximize the number of SL layers before the onset of misfit dislocations. These models are then expanded to cover material systems (substrates, QW or QD, and SC) composed of AlP, AlAs, AlSb, GaP, GaAs, GaSb, InP, InAs, or InSb, as well as the ternaries and quaternaries possible in the Al/Ga/In/P/As/Sb systems. The results of these models show a match to within 500 ppm strain in the InAs/GaAsP/GaAs QD/SC/Substrate system. These models have been assembled into a free application on nanoHUB for use by the community.

Hybrid organic-inorganic perovskite structured materials (mostly CH₃NH₃PbI₃ and CH₃NH₃PbBr₃) have shown remarkable opto-electronic properties, including photovoltaic and luminescent ones. While diffraction data suggest cubic perovskite structure, there are several experimental reports indicating that the material is polar, leading to speculations about possible ferro- and piezo-electricity. Specifically ferroelectricity has been suggested to play a role in the (“normal”, rather than the anomalous) photovoltaic effect because ferroelectric domains may induce internal electric field facilitating the separation of photo-excited electron and hole pairs, and reduce of recombination through segregation of charge carriers ^[1].
We attempt to test the possible polar character of CH₃NH₃PbI₃ and CH₃NH₃PbBr₃ by searching for pyroelectricity in them. Pyroelectricity is the ability of polar materials to generate a temporary surface charge upon temperature change. If the material is made a part of a closed circuit, this surface charge can result in an external current. We apply the periodic temperature change technique (modi#64257;ed Chynoweth method ^[2]) to single crystals of methylammonium lead halides. We will report on the results of our experiments, both in the dark and under (surface) illumination, and will combine them with results from impedance spectroscopy on the same samples, to answer the title question.
[1] J. M. Frost et al., Atomistic Origins of High-Performance in Hybrid Halide Perovskite Solar Cells, Nano Lett. 14, 2584-2590 (2014)
[2] I. Lubomirsky, O. Stafsudd, Practical guide for pyroelectric measurements, Rev. Sci. Instrum. 83, 051101 (2012)

Carbon nanotubes have been employed in a wide variety of novel electronics and device applications. In recent studies the use of carbon nanotubes for charge carrier extraction and transport layers in the emerging class of hybrid perovskite based photovoltaics has been proposed.[1] In particular a substitution for the organic hole transport layer in the conventional cell architecture can be envisioned. Therein, printed contacts with carbon nanotubes embedded in an inert and resilient polymer matrix could hold the promise to exhibit better charge transport properties while at the same time improving cell stability and durability.
In the present study we disseminated the electronic alignment and charge transfer at the interface between semiconducting single walled carbon nanotubes (swCNT) with (6,5) chirality and methylammonium lead iodide perovskite films by photoemission and transient absorption spectroscopy. Our results show that the carbon nanotube layer undergoes a unique energy level alignment at the interface. An interfacial dipole is formed which can be attributed to charge transfer between the swCNT and the perovskite film. As a result the swCNT layer becomes n-doped at the interface and is progressing to an intrinsic state further away from the immediate contact to the MAPbI₃. We find this electronic characteristic to be beneficial for hole extraction with fast hole capture rates.
Our observations confirm the potential of swCNT contacts for perovskite based photovoltaic devices and give guidelines on selection and layout of CNT based charge extraction layer utlizing the unique electronic alignment mechanism.
[1] Habisreuter, S. N., Leitjens, T., Eperon, G. E., Stranks, S. D., Nicholas, R. J., Snaith, H. J., Nano Lett. 2014, 14, 5561minus;5568

Defects scatter and trap free carriers; thus play an important role in carrier transport in semiconductors. The diffusion of defects also affects the structural and electronic properties of materials. Here we report the density functional theory (DFT) calculations of charge transition levels and diffusion of native defects in CH₃NH₃PbI₃. We show that self-interaction error and the neglect of spin-orbit coupling (SOC) in many previous DFT calculations resulted in incorrect positions of valence and conduction band edges although their difference, which is the band gap, is in good agreement with the experimental value. This problem has led to incorrect predictions of defect level positions. Hybrid density functional calculations, which partially correct the self-interaction error, in combination of the SOC, show that, among native point defects (including vacancies, interstitials, and antisites), only the iodine vacancy and its complexes induce deep electron and hole trapping levels inside the band gap, acting as non-radiative recombination centers. Furthermore, we show the calculations of diffusion barriers of native defects and discuss the impact of defect diffusion in photocurrent hysteresis in CH₃NH₃PbI₃ and phase segregation in mixed-halide hybrid perovskites.

Solar cells based on novel hybrid organic-inorganic halide perovskites have recently reached power conversion efficiencies comparable to silicon solar cells. Nevertheless, the exact details of the charge generation and extraction mechanisms in perovskites are still unclear. Unusual phenomena such as current-voltage hysteresis are still not fully understood. We apply Kelvin Probe Force Microscopy (KPFM) for measuring the local electrical potential and thereby the charge distribution in the different layers of the device under working conditions [1,2]. Therefore, a SPM setup was equipped with a sample illumination and placed in an inert atmosphere to avoid photo-oxidation of the sensitive materials. We prepared smooth cross sections of the solar cell by means of focused ion beam milling. This way, the internal interfaces between the different materials in the fully functional device were accessible for KPFM [1].
New measurements on planar devices revealed distinct differences in the potential distribution compared to mesoporous devices. We observed a mostly flat potential in the planar junction perovskite layer. Upon illumination under short-circuit conditions, the perovskite layer charges up positively in some devices. The timescales of this charging are directly connected with the current voltage hysteresis that we observed in the device. These results help to get a better understanding of these processes in perovskite solar cells.
[1] R. Berger, A.L. Domanski and S.A.L. Weber Electrical Characterization of Organic Solar Cell Materials based on Scanning Force Microscope Methods, Eur. Polym. J., 49, 1907 (2013).
[2] Bergmann, V.W.; Weber, S.A.L.; Ramos, J.A.; Nazeeruddin, M.K.; Grätzel, M.; Li, D; Domanski, A.L.; Lieberwirth, I.; Ahmad, S. and Berger, R. Real-space observation of unbalanced charge distribution inside a perovskite-sensitized solar cell, Nature Communications, 5, 5001 (2014).

The performance of bulk organic and hybrid organic-inorganic heterojunction photovoltaics is often limited by high carrier recombination arising from strong exciton binding and low carrier mobility. Structuring materials to minimize the length scales required for exciton separation and carrier collection is therefore a promising approach for improving efficiency. In this work, first-principles computations are employed to design and characterize nanostructured photovoltaic materials composed of two-dimensional (2D) transition metal phosphate (TMP) sheets covalently bound to organic absorber molecules. Using a combination of transition metal substitution and organic functionalization, both the local electric field and the band alignment between the organic and inorganic regions of these nanostructures are systematically tuned, illustrating the potential for concomitant optimization of photocurrent and open-circuit voltage. Using this approach, a new TMP-based material with high mobility 2D electron and hole conducting channels and a computed open-circuit voltage of 1.7 V is designed. This work suggests that hybrid TMPs constitute an interesting class of materials for further investigation in the search for high efficiency, low cost, and flexible photovoltaics as well as other applications requiring precise energetic alignments.

Organic-inorganic halide perovskites been intensely studied in the past 5 years due to their unprecedented rate of growing power conversion efficiencies. We investigate the structural and electronic properties of ferroelectric domain walls in CH₃NH₃PbX₃ (X=Cl, Br, I). We find that organometal halide perovskites can form both charged and uncharged domain walls, due to the flexible orientational order of the organic molecules. It is found that the presence of charged domain walls will significantly reduce the band gap by 20%-40%, while the presence of uncharged domain walls has no substantial impact on the band gap. We demonstrate that charged domain walls can serve as segregated channels for the motions of charge carriers. Our findings highlight the importance of ferroelectric domain walls in hybrid perovskites for photovoltaic applications.

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.

Cu₂ZnSn(S,Se)₄ (CZTSSe) photovoltaic material could be the earth-abundant (i.e. low cost) and low toxicity alternative compounds for the commercialized Cu(In,Ga)(S,Se)₂ thin-film solar cells. To make this possible, specific efforts applied to the absorbers, front and back interfaces/contacts must be performed for improving CZTSSe performance.
In this work, we demonstrated the high efficiency CZTSSe solar cells by introducing a thin intermediate MoO₃ layer for back contact modification. The intermediate layers increase short#8208;circuit current density (J_sc) from 25.8 mA/cm² to 27.6 mA/cm² and fill factor (FF) from 46.5% to 52.1%, possibly resulting from enhancing grain growth and reducing the thickness of MoSe₂ on the back contact after selenization process. The thin MoO₃ layers could be effective barriers against undesired overselenization of the Mo back contacts. Finally, a 7.7% efficient CZTSSe solar cell with open#8208;circuit voltage (V_oc) of 480 mV, J_sc of 28.8 mA/cm² and FF of 55.83% was obtained.
References
[1] V. Tunuguntla, W.C. Chen, P.H. Shih, I. Shown, Y.R. Lin, C.H. Lee, J.S. Hwang, L.C. Chen and K.H. Chen, J. Mater. Chem. A, (2015), DOI: 10.1039/C5TA02833G
[2] X.Yin, C. Battaglia, Y.Lin, K. Chen, M. Hettick, M. Zheng, C.Y. Chen, D. Kiriya, and A. Javey, ACS Photonics, Vol.1, 1245-1250 (2014)
[3] F. Liu, K. Sun, W. Li, C. Yan, H. Cui, L. Jiang, X. Hao, and M. A. Green, Appl. Phys. Lett., Vol.104, 051105 (2014)

Hybrid halide perovskite solar cells have demonstrated efficiencies approaching those of crystalline silicon. The Cl derived synthesis (forming CH₃NH₃PbI_3-xCl_x) has produced highly efficient devices and has been shown to increase carrier lifetime. However, the role of chlorine is still poorly understood. We have investigated the presence and chemical state of chlorine in PbCl₂ shy;derived methylammonium lead trihalide perovskite thin films (CH₃NH₃PbI_3-xCl_x) using Cl Kshy;edge Xshy;ray absorption near edge spectroscopy (XANES) and X-ray florescence (XRF).
To better understand how the Cl evolves during annealing, in-situ and ex-situ XANES and XRF were performed. XANES measurements were taken in transmission (for quantifying the Cl amount) and in florescence yield. Films were spinshy;cast from a 1:3 molar mixture of PbCl₂ and CH₃NH₃I in DMF. For transmission measurements films were deposited on Si₃N₄ windows, while other measurements were performed on films deposited on FTO coated glass with a TiO₂ layer. The in-situ measurements were done on samples ranging from 90C-105C. All temperatures seem to have an initial rapid decline followed by a much slower decline suggesting that the mechanism for Cl loss from the sample may change. From films annealed at 95C for 120min in dry air, we quantify the amount of Cl remaining in the film and find that there is x=0.05±0.02 in CH₃NH₃PbI_3-xCl_x. By comparing the XANES to Cl standards find that the remaining Cl cannot be accounted for by MACl and PbCl₂ alone. We will discuss possibilities for how the Cl incorporates into the films.

The hypothesis that ion motion is responsible for anomalous hysteresis in the current-voltage curves of perovskite solar cells [1,2] is investigated. A drift-diffusion model that accounts for conduction electrons and holes [3] has been generalised to include slow moving ions. Due to the much greater ion concentration than either free electron or hole densities, the electric field is determined solely by the ion distribution to a good approximation. The resulting simplified model, for ions and electric potential, is distinguished by nanometre scale Debye layers in which most of the charge resides. The electrostatic potential predicted by this model is used to determine the electron and hole distributions and hence to determine the output current. We find that the resulting current-voltage curves have the same qualitative features as measured curves if the electron-hole recombination rate is assumed to be directly proportional to the hole concentration and independent of the electron concentration.
1. W Tress et al Energy Environ. Sci. (2015) 8, 955
2. H J Snaith et al J. Phys. Chem. Lett. (2014).5, 1511
3. J M Foster, H J Snaith, T Leitjens, G Richardson SIAM J Appl Math (2014) 74 1935

In order to investigate alternative absorber materials for inorganic solar cells, a material gaining significant attention in addressing these issues is Cu₂ZnSnS₄ (CZTS), which has an ideal single junction band gap of 1.4-1.5 eV. Several CZTS thin film absorber preparations leading to highly efficient cell (>6% efficiency) have employed hydrazine process (10.1%), co-evaporation (9.15%), hot injection (7.2%) thermal evaporation (6.8%), ion beam sputtering (6.7%), and co-evaporation (6.1%). However, specially designed equipment is required, more expensive techniques and precursors to obtain the compound. In this work thin films of CZTS were obtained by simple chemical bath deposition (CBD) on glass substrates. For CBD reaction was used Copper chloride (CuCl₂), Zinc chloride (ZnCl₂) and Tin chloride (SnCl₂) as a precursor salts, Sodium thiosulfate (Na₂S₂O₃), Thioactamide (C₂H₅NS) as sulfur source and Triethanolamine (TEA) and Sodium tartrate (Na₂C₄H₄O₆) as complexing agent in alkaline middle. The conditions of the CBD such as time, temperature of deposition thermal treatment temperature has been studied to obtain the better experiment in the synthesis of CZTS thin films. The optical, electrical and structural properties of these films were characterized using X-ray diffraction (XRD), Atomic force microscopy (AFM), Electron microscopy scanning (SEM), UV-Vis spectrophotometer, X-ray photoelectron spectroscopy (XPS) and photoresponse. it was obtained the orthorhombic structure of CZTS. The band gap of the films was estimated in the range of 1.4-1.6 eV, the electrical conductivity was measured using it photoresponse and calculated from 10^-4 to 10^-1 S/cm with a thickness around 350 nm.

Inverted polymer solar cells (iPSCs) containing high work function metal anodes (e.g. Ag or Au) and modified indium tin oxide (ITO) cathodes exhibit superior efficiency and stability over polymer solar cells (PSCs) with a conventional geometry. A major limitation associated with iPSCs is the large barrier to electron extraction at the photoactive layer-ITO interface. To address this limitation, inorganic materials are implemented as electron transport layers (ETLs). However, organic ETLs possess inherent advantages over inorganic layers for their ease of processing and favorable mechanical properties. Realizing uniform ultrathin films over large areas represents a significant challenge, yet most efficient iPSCs reported to-date require an ultrathin ETL (e. g. ~5 nm of PEIE or PFN). Here, high performance iPSCs were successfully fabricated with an organic ETL composed of zwitterionic fullerenes (C₆₀-SB). Power conversion efficiencies (PCEs) as high as 9.23 % were achieved with an ETL thickness of ~40 nm. Exceptional insensitivity to the ETL thickness, from 5 nm to 140 nm, was found, with PCEs exceeding 8 % across the entire thickness range. C₆₀-SB layers function both as electron acceptor and cathode modification layers in iPSC devices. This dual role of the zwitterionic fullerene contributes to the ETL thickness insensitivity of device performance, which is an important and unique property of C₆₀-SB. X-ray photoelectron spectroscopy (XPS) and ultraviolet photoelectron spectroscopy (UPS) showed that a C₆₀-SB thickness of ~5 nm is necessary for full coverage of an ITO substrate, where upon the work function of ITO was decreased by ~0.6 eV. The orthogonal solubility of C₆₀-SB and the active layer, coupled with its thickness insensitivity, enabled slot-die preparation of iPSCs with PCEs of 7.38% effectively, opening a new route to efficient large area devices that can be fabricated at room temperature.

In organic photovoltaics (OPV) devices, where the active layer is composed of blends of a conjugated polymer and a fullerene such as PCBM, the device efficiency is directly affected by the phase behavior of the polymer and the PCBM. Typically the OPV efficiency decreases rapidly in operation due in part by the high diffusional mobility of the fullerene within the polymer amorphous regions causing large changes in the blend morphology. To enhance the long-term stability of the device, it is predicted that preventing changes in the device morphology during operation are necessary, which can be achieved by use of a modified fullerene. In this study, a PCBM-based thermo-cross-linkable fullerene has been synthesized as an acceptor in a bulk heterojunction based organic solar cell. The cross-linking was achieved using a thermally activated benzocyclobutene (BCB) molecule. The thermo-crosslinking reaction is initiated at temperatures typically used in processing of OPV devices. Compared to PCBM, the cross-linked fullerene is highly insoluble and has a diffusional mobility in poly(3-hexyl thiophene) (P3HT) which is an order of magnitude slower than PCBM. Its electron mobility is comparable to that of PCBM and organic photovoltaic (OPV) devices consisting of bulk heterojunction active layers with P3HT and fullerene show very similar efficiencies. OPV devices with P3HT and either pure cross-linked fullerene or its mixture with PCBM as acceptors have been shown to be extremely stable to thermal annealing with no loss in device efficiency up to 48 hours of annealing. This compares to a loss of 60% of initial efficiency in identically prepared devices when using PCBM as the acceptor. Optical microscopy and grazing incidence wide angle x-ray scattering (GIWAXS) shows that a probable cause for this excellent stability in the cross-linked fullerene containing BHJs is associated with a significant inhibition of formation of crystals of fullerene.

In previous research, electrospinning has been introduced as a technique that may increase polymer solar cell efficiency. However, electrospinning of a conjugated polymer is not possible due to the absence of chain entanglement, which is a prerequisite for electrospinning. In this study, core-shell nanofibers have been fabricated by co-electrospinning of two components, poly(3-hexylthiophene-2,5-diyl) (P3HT) and [6,6]-C61-butyric acid methyl ester (PCBM), as the shell and polystyrene (PS) as the core. Large protrusions have been observed on the core-shell fiber surfaces, which have not been previous observed in fibers of comparable diameters produced by other methods. TEM confirmed the core-shell structure of the fiber. Measurement of the florescence and conductivity of these fibers, by confocal laser scanning microscopy and shear modulation force microscopy (SMFM), indicated that P3HT/PCBM formed homogeneous surfaces on the core-shell nanofibers. Compared to pure PS fibers, DSC measurements indicated a heat capacity change with the addition of P3HT/PCBM. Finally, solar cell devices were fabricated in order to accurately characterize the photovoltaic performance of the electrospun nanofibers.

Quantum dots (QDs) are attractive materials for optoelectronic applications because of the size-dependent tunability of their band gap and their solution processability. Colloidal QDs have been studied actively for applications as light emitting diodes, photodetectors, and photovoltaics. However, further improvements in device performance are required to make them competitive. Although suppressing trap states in QD thin films is key for improving the performance of QD-based optoelectronics, this has remained a major challenge primarily because a fundamental understanding of the source of traps is lacking. Here, we investigate the chemical origins of trap states in PbS QD thin films. Photoluminescence spectroscopy and X-ray photoelectron spectroscopy show that ligand-exchange procedures for device fabrication lead to the formation of sub-bandgap emission features and under-charged Pb atoms. Our experimental results are corroborated by density function theory simulation, which show how the Pb atoms having a lower charge in QDs with iodine ligands contribute to sub-bandgap states. The trap states generated after ligand exchange were significantly reduced by oxidation of under-charged Pb atoms using 1,4-benzoquinone, reducing the density of trap states by a factor of 40, as measured by drive-level capacitance profiling.

We report systematic investigations in the crystallization of the CH₃NH₃PbI₃ thin films in well controlled ambient composition. The key objective of our work is to fabricate photovoltaic cells with high power conversion efficiencies and good device stability through effective control of the trap states. Such localized states, residing in the bulk of the perovskite layer and at the material interface, are known to act as recombination centers leading to short carrier lifetimes and poor charge extraction efficiency. Recent work by the authors demonstrated significant reduction in the trap density in perovskite film by thermal annealing of the material in O₂. In this work, we will present a novel growth process in which the perovskite films were crystallized in a well-controlled ambient with systematically varied ratios between N₂ and O₂. Experimental results clearly demonstrate such growth process is effective in controlling the defect density and the crystallization of the perovskite film. Under the optimized growth conditions perovskite films with grain size > 1 micrometer were obtained. Perovskite solar cells with planar structures, fabricated under the optimized crystallization ambient, demonstrate significant enhancement in the device performance with significant reduction in the hysteresis of the I-V characteristics. The champion device exhibits the following characteristics: V_oc= 1.02 V; J_sc= 23.2 mA/cm²; fill factor = 0.73; and PCE = 17.2%. To examine the underlying mechanisms responsible for the observed enhancements in the PCEs of the devices, systematic investigations including I-V characteristics; external quantum efficiencies; time-resolved photoluminescence; photothermal deflection spectroscopy; and low-frequency noise measurements were performed on the materials and devices. From the experimental results, we demonstrate that the composition of the crystallization ambient has significant impact on the trap densities in the perovskite material with strong effects on the performance of the devices.

In the last years, the organic / inorganic halide perovskites have attracted significant attention in the field of solar cell applications due to the rapid rate in which researchers around the world were able to increase device efficiency. However, large variations in efficiency are observed in similarly prepared devices, which is mostly due to variations in morphology and/or film composition. Even though there are reports showing that the stoichiometry has an influence on the solar cell performance, the induced changes on the density of states and ionization energy (IE) have not been systematically investigated so far. The large variety of published IE values for one and the same material CH₃NH₃PbI₃, which range from approximately 5.1 eV to 6.6 eV, indicates that preparation, treatment, and handling must have a significant influence on this property. A comprehensive investigation is much needed, as exact knowledge of the IE is crucial when it comes to matching organic or oxide transport materials as well as understanding the working mechanism of devices.
In this talk we present an extensive UV photoelectron spectroscopy study of a large number of perovskite films, where we investigate variations in the occupied electronic structure to elucidate the origin of the various ionization energy values reported in the literature. We focus on the perovskite material CH₃NH₃PbI₃ which we prepare using vapor deposition as well as various solution processing methods. We are able to show for the first time that the IE can be intentionally tuned by almost 1 eV by changing the preparation conditions (e.g. molar ratios, deposition methods, annealing procedures). Using X-ray photoelectron spectroscopy these changes can be directly correlated to variations in film stoichiometry. As no additional gap states are observed in these films, we can conclude that only neutral defects are created in the process of composition tuning. The possibility of controlling the ionization energy will open up new opportunities for tailor made perovskite films.

The importance of controlling the crystal growth of the organic-inorganic halide perovskites is evidenced by the recent body of work that explores a variety of processes and chemistries targeting better film and crystal morphology. We have recently devloped a new, one-step, solution process with no volatile organic solvent by using an ionic liquid, methylammonium formate, as the crystallization medium. This process results in a direct crystallization to the perovskite with no crystalline intermediate and dramatically increases the range of processing temperatures and times. The resulting thin films have complete coverage, excellent thickness uniformity, large crystal domains, and a higher degree of crystallographic orientation than films made from previously reported one-step solution routes. We exploit the broad range of processing temperatures to crystallize thin films with domains ranging from 10 to 300 microns and show the impact on optical and electronic properties as a function of grain size.

Texturing the back surface reflector of thin-film solar cells is a promising route to enhance light absorption. Light scattered from a textured surface has a longer path length, and if the scattering angle is large enough, the light will be trapped within the absorber layer. However, deposition of high quality Si films on highly textured surfaces is challenging as the morphological roughness of the underlying substrate leads to formation of cracks and porous areas in the silicon layers, adversely affecting the solar cell efficiency and stability. Therefore, there is a need for techniques that can reconcile the requirement of textured surfaces with the constraint of surfaces that can support thin film silicon growth.
In this work, we demonstrate the use of optically dissimilar TiO₂ nanoparticles (TiO2-NPs) and Si nanoparticles (Si-NPs) to create a light scattering back reflector having a buried scattering interface and a morphologically smooth surface as the growth interface for solar cell fabrication. We fabricated two types of textured surfaces (pyramidal and inverted pyramidal) using TiO2-NPs on a planar reflector in order to scatter an incident light into a wide distribution of angles. Subsequently, a high-index Si-NPs layer was coated on top of the TiO2-NPs textured surfaces so as to flatten the highly textured TiO2-NPs surfaces. A simple and rapid fabrication technique was developed to favor wider adaptability and commercial feasibility. Finally, a-Si:H thin-film solar cells were deposited on such optically textured TiO2-NPs/Si-NPs back surface reflectors and the enhancement of light absorption was studied. Finite-difference-time-domain simulation studies were used to validate and support the experimental findings. Our studies show the usefulness of the proposed low cost technique using nanoparticles with dissimilar refractive indices to create nanotextured optical interfaces while ensuring morphological smoothness for solar cell growth.

Inverted organic optoelectronic devices, in which the metallic electrode serves as the anode, have shown marked improvements in operational lifetime compared to conventional devices, in which the metallic electrode serves as the cathode. For the inverted device configuration, high workfunction metals are required in order to efficiently collect or inject holes in photovoltaic or light-emitting devices, respectively. Ag is an ideal metal in terms of its optical properties for the metallic electrode due to its high reflectivity across the visible spectrum; however, the workfunction of an untreated Ag film (4.3 eV) is not large enough to serve as an efficient anode. The workfunction of Ag can be increased to a value of ~5 eV upon oxidation of the surface to Ag₂O. By forming an ultrathin Ag₂O surface layer (i.e., < 5 nm in thickness), partially-oxidized Ag films can benefit from high workfunctions while maintaining high reflectivity. Here, we expose Ag films to an O₂/Ar plasma for varying times in order to tune the thickness of the surface oxide layer. We study the chemical state, morphology, and thickness of the plasma-treated Ag films using X-ray photoelectron spectroscopy (XPS), scanning electron microscopy, dark-field microscopy, and spectroscopic ellipsometry. For very short exposure times (1 s), the surface structure of the Ag films remains unchanged, although the presence of Ag, Ag₂O, and Ag₂CO₃ forms are all observed in XPS data, suggesting Ag₂O and Ag₂CO₃ form within the top 10 nm of the Ag films, based on the short inelastic mean free path of photoelectrons. For exposure times of 5 - 30 s, the Ag films become porous and exhibit increased nanoscale surface roughness (i.e., more discrete grains), with porosity increasing with increasing exposure time. Further, the intensities of O1s and C1s photoelectrons from Ag₂CO₃ increase on going from an exposure time of 1 s to 5 s, and then decrease for longer exposure times while the intensities of O1s photoelectrons from Ag₂O simultaneously increase. In addition, we expose Ag plasmonic metasurfaces consisting of arrays of Ag nanoparticles on Ag thin films (i.e., AgNPA/Ag) to O₂/Ar plasma for 1 s in order to form an ultrathin oxide/carbonate layer within the top 10 nm of the metasurface. We observe that the AgNPA/Ag structure remains unchanged at short plasma exposure times. We plan to employ these partially-oxidized plasmonic metasurfaces in organic optoelectronic devices in order to simultaneously tune the optical and electronic properties of the metallic electrode. Future work includes determining the thickness of the Ag oxide/carbonate forms on the surface of the Ag nanoparticles, in addition to measuring the workfunction of both planar and plasmonic Ag surfaces before and after exposure to the O₂/Ar plasma.

Zn(O,S) solid solutions are receiving a strong interest for the development of high efficiency Cd-free chalcogenide solar cells. Cd-free Cu(In,Ga)(S,Se)₂ cells with record efficiencies higher than 21% have already been reporting, demonstrating the viability of these materials for the replacement of the CdS buffer layers used in standard chalcogenide devices. Control in the O/(O+S) relative content in the Zn(O,S) solid solution is important, because it gives a way for optimization of the heterostructure band structure, optimizing the conduction band discontinuity in the cell heterojunction, which depends on the chemical composition of the chalcogenide absorber (including chalcopyrite and kesterite absorbers with different Ga/(In+Ga) and/or S/(S+Se) surface content ratios). However, assessment of the chemical composition of these layers grown on the absorbers and/or in the finished devices is challenging because of their reduced nanometric thickness (typically in the 20 - 40 nm region). Non destructive assessment of the composition of the different layers in CIGS based devices is typically made by X-Ray Fluorescence (XRF). However, the XRF is strongly conditioned because of the interaction of sulfur with the Mo back contact, as well as with potential interactions with the ZnO window layer.
In this work, the use of Raman scattering has been investigated for the chemical assessment of the Zn(O,S) buffer layers. Raman scattering is an optical non destructive technique well suited for assessment of CIGS based devices and processes. In this analysis, reference layers with different chemical composition (including ZnS, ZnO and Zn(O,S) alloys with S/(S+O) content ratio between 0.1 and 0.8) were grown on both Mo coated glass substrates and CIGS absorbers by atomic layer deposition, sputtering and chemical bath deposition.
The Raman spectra of both ZnO and ZnS compounds measured in resonant conditions greatly differ from non-resonant spectra, exhibiting mostly the A₁ (LO) mode and its overtones. However, the resonant spectra are quite powerful in detection of these layers in complete devices. Thus, taking into account the strong bang gap bowing of Zn(O,S) solid solutions, we applied three different excitation laser lines (325 nm, 488 nm and 532 nm) to obtain the characteristic Raman spectra for compounds with different oxygen to sulfur ratio. In the obtained spectra it is seen that A₁ (LO) peak exhibits the double mode behavior and is dominating in almost all spectra. It was also found that the ZnS-like A₁ (LO) peak maintains its position at around 350 cm^-1, while the ZnO-like peak, shows a significant red shift. In addition, the relative intensities of the ZnS-like and ZnO-like dominant peaks show a clear correlation with the S/(S+O) content ratio. These data will be discussed and analyzed for the development of a fast and reliable Raman scattering methodology for the quantitative chemical assessment of these layers in the devices.

A photoactive absorber layer, Sb₂S₃ for a hybrid solar cell was deposited onto ITO/TiO₂ using thermal evaporation method and a hole conducting layer, P3HT was deposited onto ITO/TiO₂/Sb₂S₃ by a spin coating method. The Sb₂S₃ and P3HT layers were annealed in a normal N₂ condition on a hot plate at temperature 320°C & 150°C for 30 & 10 min, respectively. SEM, TEM, XRD and Raman results show the as-deposited film is amorphous and annealed film is crystalline in nature. The device is oxidized to be confirmed from XRD and XPS depth profiling studies. Spectroscopic ellipsometry (SE) and UV-vis spectroscopy measurements were carried out to compare the optical properties of the active layer. Ultraviolet photoelectron spectroscopy (UPS) spectrum was collected for the bang alignment of Sb₂S₃. The external quantum efficiency (EQE) is found to be ~ 34% and the photovoltaic conversion efficiency ~1.94%, is the best efficiency of the ITO/TiO₂ (n) /Sb₂S₃ (i) /P3HT (p) /Ag planar hybrid eheterojunctions solar cell device.

We report an accelerated degradation testing of MAPbX₃ films (X = I or Br) by exposure to concentrated sunlight of 100 suns. We present that the evolution of light absorption and the corresponding structural modifications are dependent on the type of halide ion and the sample temperature. The degradation in absorption of MAPbI₃films after exposure to 100 suns for 60 minutes at elevated sample temperature (~45-55^oC), due to decomposition of the hybrid perovskite material, has been demonstrated. No degradation was observed after exposure to the same sunlight concentration but at a lower sample temperature (~25^oC). We therefore postulate that the observed degradation is induced/accelerated by a combined effect of light and heat during concentrated sunlight exposure and is possibly related to penetration of ambient species to the encapsulated samples. No photobleaching or decomposition of MAPbBr₃ films were recorded after exposure to similar stress conditions (light intensity, dose, and temperatures). Reasons for the better stability of MAPbBr3 perovskite films are discussed.

Solar cells using organic semiconductors or organolead-halide perovskite have attracted much attention for their potential advantages of their low cost, light weight and flexible features. Simple process can be of interest from an industrial point of view, since just mixing of lead halide and amine halide results in perovskite and mixing of donor and acceptor molecules results in bulk-heterojunction active layer. Control of the mixing process is an important issue for these devices. We have been focusing on vacuum deposition to fabricate both organic and perovskite solar cells. We developed a method to control the bulk-heterojunction structure based on organic heteroepitaxy, where the donor and acceptor molecules were deposited on self-assembling crystalline buffer layer of biphenyl-bithiophene. We successfully demonstrated the improvement of solar cell characteristics. Moreover, we developed laser deposition method for the fabrication of perovskite. The deposition rate of the lead halide and amine halide can be precisely controlled, although control of the deposition of amine halide has been considered to be difficult because of the gas generation during the deposition.
We fabricated perovskite device based on the structure of Glass/ ITO/ p-buffer/ perovskite/ n-buffer/ Al. As for p-buffer layer, we used PEDOT:PSS or PCDTBT and as for n-buffer layer we used PCBM and BCP. Co-evaporation of PbCl₂ and CH₃NH₃I by laser deposition system resulted in the maximum solar cell characteristics of 14.9 % with forward scan (J_sc = 19.3 mA/cm², V_oc = 1.04 V, FF = 0.741). The backward scan of the device reached an efficiency of 14.4 %, which proves that the device hysteresis was profoundly suppressed with our method. We successfully reduced the issue of the gas generation of the amine halide, which drastically improved the controllability of deposition. Based on this method, we can fabricate finely controlled perovskite film, which opens the way to further improvement of device performance and better understanding of fundamental issue of perovskite materials.

The conventional active layer structure for lead halide-based perovskite solar cells consists of an electron transporting layer (ETL), the perovskite layer, and a hole transporting layer (HTL). The ambipolar charge transport nature of the perovskite layer requires the presence of both the ETL and the HTL for efficient charge extraction. In this presentation, we report a systematic analysis of charge carrier transport in perovskite hybrid solar cells based on deterministic charge carrier transport models. The models describe the transport, i.e., diffusion and drift, of a single type of charge carrier in the ETL and the HTL. In the perovskite layer, the dynamics of both electron and hole transport and of charge generation, in conjunction with the kinetics of bimolecular recombination, are accounted for. In each layer, the charge transport equations are coupled self-consistently with Poisson&’s equation for the electrostatic potential.
Our models predict the influence on the photovoltaic device performance of the charge carrier mobilities, the active layer thicknesses, as well as the energy barriers for charge extraction at the ETL/perovskite interface, the perovskite/HTL interface, and the active layer/electrode interfaces. The computed charge density and electric field distributions provide valuable insights into the multi-physical processes that govern charge transport in perovskite hybrid solar cells. Using the model predictions for photocurrent-voltage (I-V) relations to fit I-V experimental measurements for devices with PCBM and PEDOT:PSS layers used as ETL and HTL, respectively, we find that the electron-hole bimolecular recombination rate is lower by orders of magnitude than that predicted by Langevin recombination theory. We have also demonstrated that the overall device performance is improved by effectively incorporating multi-walled carbon nanotubes (MWCNTs) into the perovskite layer due to the reduction of the bulk recombination rate and the resultant increase in the open-circuit voltage. We have found that the device efficiency is maximized for a certain (optimal) concentration of MWCNTs in the active layer. The model predictions, in full agreement with experimental measurements, provide valuable input toward optimal design of perovskite-based hybrid solar cells for next-generation photovoltaics.

Organic-inorganic perovskites such as CH₃NH₃PbI₃ are highly promising materials for a variety of optoelectronic applications, with certified power conversion efficiencies in solar cells already exceeding 20%¹. A key enabling property of the perovskites is their high photoluminescence quantum efficiency, suggesting that these materials could in principle approach the thermodynamic device efficiency limits in which all recombination is radiative. However, recent reports have demonstrated the presence of non-radiative recombination sites in the form of subgap trap states, which limit performance^{2, 3}.
In this work, we use bulk and confocal photoluminescence (PL) measurements on neat perovskite films without contacts to show that the PL lifetime and intensity increase slowly over time under illumination, with the film eventually reaching a stabilized emission. On the microscale, we find that the emission from dark spots increases substantially whereas the emission from bright spots remains comparatively unchanged. The PL improvements correlate with an order-of-magnitude reduction in trap density. We use energy-dispersive X-ray spectroscopy (EDS) measurements to show that this photo-induced cleaning also correlates with a local reduction in iodide content, which we speculate to be caused by a photo-induced ion migration.
Our work shows that similar improvements in the optoelectronic properties of the materials resulting from chemical passivation treatments^{3, 4} can also be achieved through illumination of the material. Moreover, we provide connections between these slow transient effects under illumination, passivation treatments, and anomalous hysteresis effects observed in full solar cells⁵.
References
1. Stranks, S. D.; Snaith, H. J. Nat. Nanotechnol. 2015, 10, (5), 391-402.
2. Stranks, S. D.; Burlakov, V. M.; Leijtens, T.; Ball, J. M.; Goriely, A.; Snaith, H. J. Phys Rev Appl 2014, 2, (3), 034007.
3. deQuilettes, D. W.; Vorpahl, S. M.; Stranks, S. D.; Nagaoka, H.; Eperon, G. E.; Ziffer, M. E.; Snaith, H. J.; Ginger, D. S. Science 2015, 348, (6235), 683-6.
4. Noel, N. K.; Abate, A.; Stranks, S. D.; Parrott, E. S.; Burlakov, V. M.; Goriely, A.; Snaith, H. J. ACS Nano 2014, 8, (10), 9815-6821.
5. Snaith, H. J.; Abate, A.; Ball, J. M.; Eperon, G. E.; Leijtens, T.; Noel, N. K.; Stranks, S. D.; Wang, J. T. W.; Wojciechowski, K.; Zhang, W. J. Phys. Chem. Lett. 2014, 5, (9), 1511-1515.

With demonstrated open-circuit voltages of 1.5V, solar cells with CH₃NH₃PbBr₃ absorber layers are of special interest for spectrum splitting and photoelectrochemical applications. Current-collection efficiency mapping of solar-cell cross-sections was used to study the working mechanism and minority carrier diffusion length of CH₃NH₃PbBr₃(Cl) perovskite-based solar cells. Such a mapping is possible using the electron-beam-induced current (EBIC) method, where a focused electron beam generates free charge carriers with high spatial resolution. Acquired EBIC profiles indicate that CH₃NH₃PbBr₃(Cl)-based cells operate as a p-n (n-FTO/TiO2 and p-perovskite), rather than a p-i-n junction, as found for the iodide derivative. Furthermore, using bias-dependent EBIC measurements, where the variation in space-charge region width can be monitored directly, combined with impedance spectroscopy, we estimate the doping density to be 3-6 x 10¹⁷ cm^-3. Most interestingly, we find that background illumination of the device during measurement at even just a few percent of 1 sun intensity results in an increase of the electron diffusion length from 100 ± 50 nm to 360 ± 20 nm. This finding demonstrates that e-beam and light irradiation of the device lead to different electrical effects and emphasizes that measurements of CH₃NH₃PbBr₃(Cl) and corresponding devices should be performed under conditions as close as possible to the PV device working conditions in order to extract relevant values.

Via controlling the morphology of bulk heterojunction organic photovoltaic cells, one can control the performance of devices. However, measuring this morphology on multiple length scales is challenging due to the limited depth sensitivity of the methods. We investigate nanostucutral changes of PBDTTT-CF:P₇₁BM BHJ films with different amounts of DIO additives and correlate them to device efficiencies by combining vertical phase morphology and surface imaging. We utilized powerful depth-sensitive neutron reflectometry technique to reveal the internal structure of the films. The morphology of the DIO-free films is not uniform and shows formation of PC₇₁BM enriched layers. AFM and TEM images complement by visualizing surface aggregates of PC₇₁BM embedded in polymer:fullerene matrix. Addition of DIO remarkably changes the vertical phase morphology of the films. Our results show that addition of different concentrations of the solvent additive result into different vertical phase morphologies of the films. Obtained information combined with AFM and TEM images allows reconstituting the details of the rich phase diagram of the polymer-fullerene-solvent system. These results demonstrate the ability to control the morphology and reveal how with a small amount of DIO additive processing one can control the vertical phase morphologies of the films.

In the recent years, perovskite solar cells (PSCs) have emerged as a competitive photovoltaic (PV) technology because of their high power conversion efficiencies (PCEs) and low-cost processing. However, the reported or certified PCEs of perovskite solar cells have been usually based on small areas (~0.1 cm²). This is primarily due to the unprecedented challenge in the formation of high quality organolead trihailde perovskite absorber layers, which is at the heart of high-efficiency PSCs. To address this issue, we have demonstrated here the use of solvent and gas chemistries in achieving perovskite thin films with up-scaled uniformity. First, we show that room-temperature crystallization of perovskites in antisolvent baths results in ultra-smooth thin films, which is due to enhanced nucleation rate and facile antisolvent-solvent extraction process. Second, we show that defects in the as-formed perovskite thin films synthesized using conventional solution methods can be rapidly healed upon reversible interaction with amine gas. Central to the gas-induced defect-healing behavior is formation of an intermediate liquid phase that spreads rapidly. The mechanisms of these perovskite crystallization and re-crystallization behavior are clarified using ex situ and in situ characterization experiments. We also show that it is feasible to tailor the perovskite film properties (composition, thickness, roughness, grain size, etc.) based on modification of the above processes. Large-scale PSCs and PV modules are demonstrated with high efficiencies. The approaches demonstrated here represent a new paradigm in the up-scaling of PSCs.

Organic solar cells (OSCs) are promising candidates for solar energy conversion and have been attracting much attention due to their low-cost, light-weight, mechanical flexibility and their potential for roll-to-roll production. High mobility, low band-gap semiconducting polymers/molecular materials were used to realize high efficiency in OSCs. In addition to new donor materials, the performance of polymer solar cells can also be improved using efficient interface layer between the electrode and active layer. Recently, conjugated polymers or polyelectrolytes have been used as an interlayer to improve electrical of properties of cathode interface. In this work, we studied the effect of novel conjugated polyelectrolytes (CPE), such as sodium poly[9,9-bis(butanesulfonate)-2,7-fluorene) based co-polymers, as a cathode interlayer on the performance of inverted bulk-heterojunction (BHJ) solar cell with a high performance active layer composed of PTB7 and PC₇₁BM. With a CPE interlayer, the photovoltaic performance was significantly enhanced with a high power conversion efficiency (PCE) of 9.5 %, which is about 16% enhancement compared to the cell without CPE interlayer. The enhanced device performance results from the efficient electron extraction and hole blocking ability of CPE interlayer at the cathode/active layer interface. We will also report on the role of morphology and annealing optimization of the active layer on the performance of OSCs, and the use of a CPE interlayer with other active layer materials, including perovskites.

Tin monosulfide (SnS) is a promising candidate for solar cells using earth-abundant-materials because it has a direct bandgap of 1.3 eV and a high light-absorption coefficient (>10⁴ cm^-1). Recently, SnS-related solar cells exhibited conversion efficiencies of 4.36%, which is far below their theoretical efficiency. This is owing to the difficultly of growing SnS films with a high purity and quality and a low defect density.
Recently, post-deposition treatment using alkali metals, such as sodium (Na) and/or potassium (K), has been employed to improve the thin films quality and the solar cell efficiency using chalcogenide semiconductors (CIGS, CZTS, and CTS). The alkali metal depletes the cation (mainly Cu) concentration in the chalcogenide film near the surface, passivates defects, and enables an optical gain through a thinner cadmium sulfide (CdS) buffer layer. Therefore, understanding the effects of alkali metals, such as Na and K, on SnS thin films and solar cells is highly useful for the development of an efficient and environmentally benign solar photovoltaic cell. In this study, we reveal the impact of the alkali-metal treatment on SnS thin films and solar cells.
A Sn precursor containing sulfur was deposited by conventional RF sputtering, and an alkali metal was deposited on the precursor by thermal evaporation as a function of the evaporation time. These precursors were continuously sulfurized at 200-500 °C for 30-120 min to obtain SnS single-phase films. Conventional SnS solar cells were fabricated with an Al/ZnO:Al/ZnO/CdS/SnS/Mo/soda-lime-glass structure.
Compared with conventional SnS films, the grain size of the SnS films containing an alkali metal was small, but the grains were densely packed without extra phases such as SnS₂ or Sn₂S₃. In addition, dot-like particles were observed on the surface of the SnS thin films containing an alkali metal. These particles disappeared after the films were rinsed with water. This tendency is a similar to that previously reported for an alkali-metal post-deposition treatment of CIGS. The carrier concentration of the SnS films containing the alkali metal was on the order of 10¹⁷ cm^-3, which is two orders of magnitude greater than that of a conventional SnS film under the same sulfurization conditions.
The solar-cell performance, particularly the open-circuit voltage, was improved by using SnS film containing an alkali metal. In general, SnS has numerous Sn vacancies that act as acceptors. These results may be the first step toward understanding the effects of alkali metals on the SnS layer for the development of an efficient SnS-related solar cell.

Chalcopyrite type Cu(In,Ga)Se₂ thin film solar cells have recently closed the gap to silicon based technologies, even though the absorbers are inhomogeneous at various length scales. In order to unravel the impact of the inhomogeneity on the electrical properties there is a strong need for spatially resolved investigations. We introduce a new approach by measuring thin lamellas of Cu(In,Ga)Se₂ cross sections via highly focused X-ray fluorescence analysis (XRF) [1]. About 200 to 300 nm thin lamellas were prepared from Cu(In,Ga)Se₂ solar cells using a focused ion beam system. Subsequently, a highly focused X-ray beam scans the lamellas and by analyzing the created fluorescence radiation one obtains elemental maps. Our approach ensures high resolution compositional analysis combined with high spatial resolution. As the spatial resolution is < 100 nm and the typical grain size of the absorber is in the order of microns our approach easily allows the collection of information even from single grains and grain boundaries. We investigated several sequentially produced Cu(In,Ga)Se₂ solar cells. We observe that the temperature during the first selenization stage drastically influences the Ga depth profile. Furthermore, slight lateral variations of the Ga/(Ga+In) ratio in the order of 0.01 can be detected. Additionally, we measured solar cells which were exposed to a post deposition treatment with potassium fluoride. For these samples the composition of some grain boundaries differs from the composition of the surrounding material. Most remarkable is a deficiency of Cu in those grain boundaries. Transmission electron microscopy is also being performed on these lamellas, in order to correlate the grain structure of the lamellas with the XRF data.
[1] Ph. Schöppe et al., Appl. Phys. Lett. 106, 013909 (2015)

With the increasing demand for low-cost solar cells that are truly scalable to the terawatts level, earth-abundant and air-stable semiconductor materials are considered a promising candidate class of materials for this application. Cuprous oxide (Cu₂O) is an earth-abundant semiconductor with a theoretical single-junction maximum power conversion efficiency over 20%. Therefore, it is a promising candidate for the development of solar cells using earth-abundant materials. Cu₂O has a direct bandgap of 2.1 eV, and its absorption coefficient has been reported to be in the range of 10³-10⁴ cm^-1. In addition, its large bandgap of 2.1 eV makes it suitable for use as a top-cell in a tandem device with conventional silicon or CuInGaSe₂ solar cells.
Cu₂O thin films have been obtained by various methods, including the thermal oxidation of Cu films, pulsed laser deposition, vacuum arc plasma evaporation, sputtering, and chemical deposition. Among these methods, sputtering is the most suitable for economically depositing large-area films having well-controlled compositions. In this presentation, Cu₂O thin films were grown by RF reactive sputtering to control the carrier concentration and the crystal structure for fabricating high-conversion-efficiency Cu₂O solar cells. Then, the solar cell was fabricated by a conventional RF sputtering method using n-type semiconductors and transparent conducting oxides (TCOs).
Polycrystalline Cu₂O thin films with a thickness of sim;500 nm were deposited on soda-lime glass substrates by RF reactive sputtering at a comparatively low deposition temperature of 300 °C with additional post-annealing. A metallic Cu with a diameter of 4 in. was used as a sputtering target, and a mixture of Ar and O₂ gases was used as a sputtering gas. In addition, conventional Cu₂O solar cells with a TCO / n-type layer / Cu₂O / Mo / soda-lime glass structure were fabricated.
The crystal structure of the Cu-O thin films depended strongly on the O₂ fraction [O₂/(Ar+O₂)] during the reactive sputtering, because Cu can have several different numbers of valence electrons. The carrier concentration of the Cu₂O thin films was changed from 10¹⁵ to 10¹⁷ cm^-3 by changing O₂/(Ar+O₂) from 1.5 to 3.0, and it also depended on the deposition temperature and post-annealing conditions. The fabricated solar cells exhibited a photovoltaic effect under AM 1.5 illumination.

Visible-light-transparent solar cells with a p-type lithium-doped nickel oxide (NiO:Li) layer were fabricated by a conventional reactive RF sputtering method. A detectable photovoltaic effect was observed from these UV-absorbing solar cells, which are attractive because their optical transparency allows greater flexibility regarding their installation locations.
Whilst most of wide-bandgap transparent conducting oxide (TCO) films exhibit n-type conductivity, NiO exclusively exhibits p-type conductivity. Recently, NiO has been used in visible-light-transparent solar cells installed on a wall or ceiling of a building or greenhouse, because it is composed of inexpensive and less-toxic elements. We thereafter proposed a “NiO-based intelligent window” that can supply a small electrical power from invisible solar cells for an invisible sensor and/or a transistor. Applications of invisible devices using NiO-based solar cells are possible in various environments and limited only by the imagination of researchers.
Controlling the carrier concentration of undoped NiO is an essential issue. In general, a high growth temperature (>800 °C) is required to obtain a NiO film with an appropriate carrier concentration for solar cells (on the order of 10¹⁶-10¹⁷ cm^-3). However, it is difficult to deposit undoped NiO on a plastic or glass substrate at such high temperature. In this presentation, we report fundamental features of NiO:Li thin films deposited without heating while focusing on their electrical and transport properties. Then the optical transparency and carrier transport characteristics of p-NiO:Li / n-ZnO solar cells deposited by a conventional RF sputtering method will be discussed. The influence of their interface properties on the solar cell performance will be shown.
Polycrystalline NiO:Li and ZnO thin films with a thickness of approximately 50-400 nm were deposited by sputtering metallic Ni and ZnO ceramics with a diameter of 4 in., respectively. A mixture of Ar and O₂ gases was used as the sputtering gas. For the Li doping, Li₂O pellets with 0.4 in. square was placed on the Ni metal target. The fabricated solar cells with PEDOT:PSS / NiO:Li / ZnO / TCO / soda-lime glass exhibited a photovoltaic effect under illumination. Moreover, an optical transmittance greater than 70% was obtained in the wavelength range of 400-800 nm. In an external-quantum-efficiency measurement, UV-violet light shorter than 400 nm was absorbed one thousand times more than visible lights. This result indicates that the space charge region around the NiO / ZnO interface acts as a photoabsorbing medium. An open-circuit voltage of 0.30 V, a short-circuit current density of 60 mu;A/cm², and a fill factor of 0.44 were obtained under AM 1.5 illumination. These results indicate that NiO:Li films have a potential for use in invisible solar cells.

III-V/Si heterostructures are thoroughly investigated for applications in photovoltaics domains. To this end, the monolithic coherent MBE-growth of GaP on Si is particularly well suited thanks to the quasi-lattice matching between GaP and Si, leading to a GaP/Si platform suitable for the subsequent growth of a III-V based efficient absorber. For instance, GaAsPN dilute nitride alloy, is particularly well suited thanks to its lattice matching property on Si, and its optimal band gap for tandem solar cell devices. However, coherent monolithic GaP growth onto Si usually gives birth to extended structural defects such as Microtwins and antiphase domains (APD), which deteriorate device efficiency. While novel growth strategies are being developed to yield defect free III-V/Si interfaces, we theoretically investigate the optical and electronic properties of the APD at the GaP/Si interface by means of state-of-the-art GW method. The band structure calculation reveals that GaP APDs present an important upward shift of its valence band maximum, creating an effective quantum well for holes at the antiphase boundaries while the conduction band minimum retains a bulk-like character. This effective band gap shrinkage could lower achievable open circuit voltages.
This work is supported by the French national project MENHIRS (grant N°. ANR-2011-PRGE-007-01).

The organometal halide perovskites are promising solar-cell materials for next-generation photovoltaic applications, due to the long carrier lifetime and diffusion length of these materials. The physical mechanism responsible for the long carrier lifetime remains an open question, with proposals including domain wall induced charge separation, benign defect states, and dynamical band gap fluctuation. In this work, we present first-principles calculations showing that the carrier lifetime is enhanced by the strong spin-orbit coupling in the organometal halide perovskites. Structural fluctuations at room temperature induce opposing spin helicities of the conduction band and valence band. As a consequence, carrier recombination pathways between the conduction band minimum (CBM) and the valence band maximum (VBM) are suppressed, reducing the radiative recombination rate and explaining the long carrier lifetime and diffusion length.
This mechanism is general to semiconductors with strong spin-orbit coupling and broken inversion symmetry, which have a Rashba-type band structure, allowing for the separation of electrons and holes into different spin states at the CBM and VBM. For the organometal halide perovskite MAPbI₃, this is supported by our calculations of the optical phonon scattering rates, which show that hot carriers rapidly thermalize to the CBM and VBM. Using a model derived from first-principles band structure calculations, we demonstrate that the carrier lifetime rapidly increases as the Rashba band splitting is increased beyond the thermal energy scale. We study the impact of disorder on our proposed mechanism using a large-scale tight-binding simulation, and find that it is robust in the presence of sufficiently large polar domains.

Despite 60 years of R&D into hundreds of materials, less than two-dozen exceed 10% laboratory efficiency. How can our community improve the “return on R&D investment,” demonstrating 10 new materials surpassing 10% efficiency each year, instead of a few per decade? What lessons generalize from the remarkable case of methylammonium lead halide perovskites (MAPbX₃), which improved from <4% efficiency in 2009 to >20% in 2015? In this presentation, we discuss the following opportunities:
Opportunity #1: Screen for PV-device-relevant material properties, not just those that are computationally accessible. The advent of MAPbX₃ emphasizes the importance of essential PV-relevant but currently computationally inaccessible properties, including: bulk minority-carrier lifetime and absorber stability. Further, techno-economic analyses emphasize the importance of manufacturing-relevant search criteria, including capital expense (capex), to ensure sustainable industry growth. If these properties are currently inaccessible with today&’s computational tools, engineers must identify and screen for the underlying characteristics that enable these essential properties [1].
Opportunity #2: Accelerate the experimental R&D cycle of learning for novel materials by working smarter and learning faster. With access to high-throughput yet simple synthesis equipment, an individual researcher can synthesize hundreds of unique organic and hybrid organic-inorganic PV devices per day, in contrast to traditional inorganic PV device runs that last one week or longer. Low-capex tools and approaches can accelerate the cycle of learning for inorganic PV materials, inasmuch as they're designed to minimize “false negatives” due to contaminants, intragranular structural defects, and shunting. Leveraging simulations to inform design of test structures (faster to synthesize than full PV devices), PV-relevant material properties can be extracted via novel metrology guided by machine-learning approaches.
Opportunity #3: Develop informatics systems. With accelerated pace of innovation and data generation, data management becomes a priority. This includes developing integrated systems to manage, assimilate, mine, preserve and secure data.
Reducing these principles to practice, we computationally screen for minority-carrier lifetime [1]. Our cycle of learning for inorganic PV materials is now less than a day, thanks to streamlined test-structure design and metrology. Four months of research have yielded room-temperature photoluminescence from two non-traditional PV materials — one of which had not been reportedly synthesized since the 1960&’s. Pending continued affirmation of this approach, we expect lessons to generalize to other novel energy technologies and other traditionally capital-intensive “hard science” problems.
[1] R.E. Brandt et al., Identifying defect-tolerant semiconductors with high minority-carrier lifetimes: beyond hybrid lead halide perovskites, MRS Communications, online, 2015

Organic-inorganic hybrid solar cells based on lead halide perovskites have achieved a record power conversion efficiency (PCE) of over 20%.^[1] Lead halide perovskite solar cells with high efficiencies typically use high-temperature processed TiO₂ as the electron selective layers (ESLs).^[1-2] Though the record efficiency cells use TiO₂ ESLs, the optical and electronic properties of TiO₂ still exhibit some shortfalls, making it a sub-optimal ESL material. For example, the electron mobility of TiO₂ is low when compared to other metal oxides such as ZnO and SnO₂.^[3] In this work, we demonstrate that low-temperature solution-processed nanocrystalline SnO₂ can be an excellent alternative ESL material for efficient perovskite solar cells. Our best-performing planar perovskite solar cell, using a nanocrystalline SnO₂ ESL, has achieved a PCE of 17.21% under reverse voltage scanning, much higher than those of our best reference cell using a TiO₂ ESL measured under the same scan conditions. Furthermore, perovskite solar cells using SnO₂ ESLs exhibited much less degradation than the cells using TiO₂ ESLs. The outstanding performance of SnO₂ ESLs can be attributed to the excellent electro-optical properties of nanocrystalline SnO₂ films, such as good antireflection, suitable bandgap and band edge positions, and high electron mobility.^[4] The low-temperature processed SnO₂ provides a new opportunity for further improving the performance of perovskite solar cells. Additionally, the low-temperature process is compatible with low-cost roll-to-roll manufacturing of perovskite solar cells on flexible substrates.
References
[1] W. S. Yang, J. H. Noh, N. J. Jeon, Y. C. Kim, S. Ryu, J. Seo and S. I. Seok, Science, 2015, 348, 1234-1237.
[2] W. Ke, G. Fang, J. Wan, H. Tao, Q. Liu, L. Xiong, P. Qin, J. Wang, H. Lei, G. Yang, M. Qin, X. Zhao and Y. Yan, Nat. Commun., 2015, 6, 6700.
[3] Snaith, H. J.; Ducati, C. Nano Lett. 2010, 10, 1259-1265.
[4] W. Ke, G. Fang, Q. Liu, L. Xiong, P. Qin, H. Tao, J. Wang, H. Lei, B. Li and J. Wan, H. Lei, B. Li, J. Wan, G. Yang, and Y. Yan, J. Am. Chem. Soc., 2015, 137, 6730-6733.

One challenge for efficient organolead trihalide perovskite solar cells is the charge recombination at the grain boundaries and surface, while large grains possessed in the perovskite film yield reduced total grain boundary area that causes less charge recombination. However, the perovskite grain size is usually limited to the film thickness due to rough and wetting surface used to grow perovskite thin film.
Here we show that the design of CH₃NH₃PbI₃ solar cells can resemble that in thin film solar cells by reducing the grain boundaries areas to reduce the charge recombination for higher device efficiency. Clean, smooth and nonwetting plastic surfaces are used to grown the CH₃NH₃PbI₃ film. With that the formation of too dense nucleus is prevented from heterogeneous nucleation, and the grain boundary migration is facilitated during the afterward thermal annealing process. The growth of CH₃NH₃PbI₃ grains with highest average aspect ratio upto 14 is demonstrated using a wide range of nonwetting hole transport layers, which increases nucleus spacing by suppressing heterogeneous nucleation and facilitate grain boundary migration in grain growth by imposing less drag-force ^[1]. The reduced grain boundary area and improved crystallinity dramatically reduce the charge recombination in CH₃NH₃PbI₃ thin films to the level in CH₃NH₃PbI₃ single crystals ^[2]. Combining the high work function of several HTLs, a high device efficiency of 17-18% in low-temperature processed planar heterojunction CH₃NH₃PbI₃ devices under one sun illumination is achieved. To the best of our knowledge this is the first time that low temperature solution processed CH₃NH₃PbI₃ based solar cells with a simple planar heterojunction can reach the stabilized efficiency of 18.3%. This simple method in enhancing CH₃NH₃PbI₃ morphology also paves the way for its application in other optoelectronic devices for enhanced performance.
Reference:
[1]Bi, C. et al. Nonwetting Surface Driven High Aspect Ratio Crystalline Grain Growth for Efficient Hybrid Perovskite Solar Cells. Nat.Commun., Accepted (2015)
[2]Dong, Q. et al. Electron-Hole Diffusion Length Exceeding 3 Millimeters in Low-Temperature Solution Grown CH₃NH₃PbI₃ Single Crystals. Science27, 967-970(2015).

Solar cells based on organic-inorganic halide perovskites have recently shown rapidly rising power conversion efficiencies, but exhibit unusual behaviour such as current-voltage hysteresis and a low-frequency giant dielectric response. Ionic transport has been suggested to be an important factor contributing to these effects; however, the chemical origin of this transport and the mobile species are unclear. Here, the activation energies for ionic migration in methylammonium lead iodide (CH₃NH₃PbI₃) are derived from first principles, and are compared with kinetic data extracted from the current-voltage response of a perovskite-based solar cell. We identify the microscopic transport mechanisms and find facile vacancy- assisted migration of iodide ions, with an activation energy of 0.6eV, in good agreement with the kinetic measurements. The results of this combined computational and experimental study suggest that hybrid halide perovskites are mixed ionic-electronic conductors, a finding that has major implications for solar cell device architectures.
'Ionic transport in hybrid lead iodide perovskite solar cells', C. Eames, J. M. Frost, P. R.F. Barnes, B. C. O&’Regan, A. Walsh, M. S. Islam, Nature Comms., in press (2015)

Solar cells composed of methylammonium lead iodide perovskite (MAPI) are notorious for their sensitivity to moisture. We show using in situ spectroscopic ellipsometry and X-ray diffraction of thin films, single crystals and solar cells: (i) Hydrated crystal phases are formed when MAPI is exposed to water vapour at room temperature. (ii) These phase changes are reversed when the material is subsequently dried. The reversible hydration/dehydration in the presence/absence of water vapour appears to proceed via:
4 (CH₃NH₃)PbI₃ + 4 H₂O #8652; 4 [CH₃NH₃PbI₃bull;H₂O] #8652; (CH₃NH₃)₄PbI₆bull;2H₂O + 3PbI₂ + 2H₂O
In contrast to water vapour, the presence of liquid water results in the irreversible decomposition of MAPI to form PbI₂. MAPI changes from dark brown to transparent on hydration; the precise optical constants of CH₃NH₃PbI₃middot;H₂O formed on single crystals were determined, with a bandgap at 3.1 eV, and are discussed in the context of DFT calculations of the structures.
Using the single-crystal optical constants and thin-film ellipsometry measurements, the time-dependent changes to MAPI films exposed to moisture were modelled. The results suggest that the monohydrate phase forms independent of the depth in the film, suggesting rapid transport of water molecules along grain boundaries. Vapour-phase hydration of an un-encapsulated solar cell (initially Jsc asymp; 19 mA cm^-2 and Voc asymp; 1.05 V at 1 sun) resulted in more than a 90% drop in short-circuit photocurrent and sim;200 mV loss in open-circuit potential; however, these losses were fully reversed after the device was exposed to dry nitrogen for 6 h. Hysteresis in the current-voltage characteristics was significantly increased after this dehydration, which may be related to changes in the defect density and morphology of MAPI following recrystallization from the hydrate. Based on our observations, we suggest that irreversible decomposition of MAPI in the presence of water vapor only occurs significantly once a grain has been fully converted to the monohydrate phase.
Leguy et al., Chemistry of Materials,2015, 27 (9), pp 3397-3407

Photonic management techniques, such as the use of nanophotonic structures, have been widely used to enhance short-circuit currents of the solar cells. On the hand, these photonic management techniques can also have important implications for the voltages of the solar cells. Here we present a detailed-balance based framework to understand both the current and voltage performance of a number of nanophotonic solar cell structures.

The thin-film solar cell community is rapidly acknowledging the role that structured absorbers will play in achieving high efficiency photovoltaic devices with significantly less materials usage. This simulation driven study highlights photon and generated carrier management opportunities for improved thin-film Cu(Inshy;_xGa_1-x)Se₂ (CIGS) device performance. By simulating structures realized via either randomized self-assembly, direct patterning via nanoimprint lithography, or a combination of both, the coupled optoelectronic model predicts increases in both the short circuit current and open circuit voltages of up to 4mA/cm² and ~25mV respectively. A sampling of structures is studied: 1) self-assembled randomized structures that highly scatter incident light and enhance carrier generation near regions of high electric field 2) embedded dielectric gratings for directly fabricated light management structures 3) embedded planar dielectric layers with point contacts to molybdenum to minimize optical and electronic loss mechanisms 4) a combination of these structures. By simulating real cross-sections, the randomizing structures combined with dielectric layers could see efficiencies of 18.3% or greater, compared to the 15.3% predicted efficiency of the 700nm planar equivalent thickness device. The sensitivity of these enhancements to selected materials parameters will also be presented. Importantly, these photon and carrier management structures could have similar implications in other thin-film photovoltaic technologies such as CdTe, Cu₂ZnSnS_shy;4, etc.

In order for thin-film crystalline silicon solar cells to deliver the anticipated cost benefits of dramatically reduced material requirements, it is essential that they also yield power conversion efficiencies comparable to conventional commercial solar cells. A significant hurdle to realizing elevated efficiency in crystalline silicon films thinner than 20 mu;m is the loss of current resulting from reduced photon absorption. A range of light management structures have been proposed in the literature to address this issue and many have been demonstrated to provide high absorption across the spectral range relevant to crystalline silicon, but their promise has yet to be realized in an active photovoltaic device. We present here 10-mu;m-thick crystalline silicon solar cells with a peak short-circuit current of 34.5 mA/cm² and a power conversion efficiency of 15.7%. The record performance for a crystalline silicon solar cell of such thinness is enabled by an advanced light-trapping design incorporating a 2D inverted pyramid photonic crystal and a rear planar dielectric/reflector stack. Coupled with the insights gained from integrated 2D photonic/electronic transport simulations, we assess how additional current and efficiency gains can be realized on the path to a 20% efficient crystalline silicon solar cell using 30-40 times less material than current commercial technologies. Additionally, a new design strategy is also proposed using the least common multiple (LCM) rule to find the optimally mismatched of a front and a back grating. Using numerical simulations, it is shown that LCM rule can lead to photo-generated current nearly 4mA/cm² or 11.74% higher than 34.06 mA/cm² for the case of a single front grating.

III-V compound semiconductors such as gallium arsenide are materials of choice for ultrahigh efficiency photovoltaic systems due to many advantages in materials attributes. However, the excessively high cost of epitaxially growing single-crystalline materials has severely restricted their wide-spread utilization in one-sun, terrestrial applications. In this regard, solar cells having drastically reduced active layer thickness combined with optimized schemes for light trapping and substrate reuse represent promising materials platform that can potentially realize cost-competitive III-V solar cells for terrestrial photovoltaics. Here we present materials and fabrication strategies for ultrathin (~200 nm for emitter and base) GaAs single-junction solar cells that incorporate programmable front-surface dielectric nanostructures and back-surface reflector for bifacial photon management to yield 16.2% one-sun efficiency in the manner that also allows the reuse of the growth substrate. Systematic studies on optical and electrical properties and photovoltaic performance of ultrathin nanostructured GaAs solar cells in experiments as well as optical and electro-optic numerical modeling provide quantitative description of the reported system and design rules for optimal materials and device configurations.

Solution-processed halide perovskite solar cells are presently the forerunner in the next generation photovoltaic technologies, with power conversion efficiencies exceeding 20%. In this talk, I will review some of the developments on the photophysical mechanisms of the workhorse CH₃NH₃PbI₃ system. In addition, I will also highlight some of our latest findings on the interfacial barrier at the CH₃NH₃PbI₃/TiO₂ interface and other investigations on the novel properties of this amazing family of materials beyond the CH₃NH₃PbI₃ system.

Organic solar cells (OSCs) are promising candidates for solar energy conversion and have been attracting much attention due to their low-cost, light-weight, mechanical flexibility and their potential for roll-to-roll production. High mobility, low band-gap semiconducting polymers/molecular materials were used to realize high efficiency in OSCs. In addition to new donor materials, the performance of polymer solar cells can also be improved using efficient interface layer between the electrode and active layer. Recently, conjugated polymers or polyelectrolytes have been used as an interlayer to improve electrical of properties of cathode interface. In this work, we studied the effect of novel conjugated polyelectrolytes (CPE), such as sodium poly[9,9-bis(butanesulfonate)-2,7-fluorene) based co-polymers, as a cathode interlayer on the performance of inverted bulk-heterojunction (BHJ) solar cell with a high performance active layer composed of PTB7 and PC₇₁BM. With a CPE interlayer, the photovoltaic performance was significantly enhanced with a high power conversion efficiency (PCE) of 9.5 %, which is about 16% enhancement compared to the cell without CPE interlayer. The enhanced device performance results from the efficient electron extraction and hole blocking ability of CPE interlayer at the cathode/active layer interface. We will also report on the role of morphology and annealing optimization of the active layer on the performance of OSCs, and the use of a CPE interlayer with other active layer materials, including perovskites.

In this work, we developed the thinner and durable glass-substrate based organic photovoltaics (OPV) for indoor application. OPV has great PV performance under the low to high illumination of room light sources compared to other types of solar cells. The OPV module has 4 cells in series and the module dimension is 29.6mm × 11.8mm × t0.45mm used for the indoor application like a calculator. And we evaluate the OPV performance and durability with the consumer solar cells for the indoor use “amorphous silicon (a-Si), & cadmium telluride (CdTe)” and the dye-sensitized solar cell (DSC) optimized for the indoor use. They have the same specification and external forms as OPV. The maximum output power of each solar cell under the fluorescent light 1000lux(0.1mW/cm²) is OPV:43.5uW/cm², a-Si:36.8uW/cm², CdTe:14.1uW/cm² and DSC:44.6uW/cm². The power conversion efficiency of OPV module under the sunlight(1SUN-100mW/cm²) is 6.0%.
And this solid-state OPV module has excellent thermal durability and tolerance against the light-soak. This OPV is beginning to be applied to the self-charging batteries for mobile devices and wireless sensor networks.

Organic solar cells have attracted much attention recently for their easy fabrication, low cost, solution processability and mechanical flexible. However, the efficiencies of organic solar cells are still much lower than that of a commercialized solar cell. Therefore, effective techniques for improving the efficiencies of organic solar cells are urgently needed. Here, we report a novel technique for improving the efficiencies of organic solar cells by introducing suitable high-mobility conjugated polymers as additives, including alkyldiketopyrrolopyrrole and dithienylthieno[3,2-b]thiophene (DPP-DTT) and poly[2,5-bis(alkyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione-alt-5,5&’-di(thiophene-2-yl)-2,2&’-(E)-2-(2-(thiophen-2-yl)vinyl) thiophene] (PDVT-10). We find that a little amount of DPP-DTT addition (1 wt.%) can lead to a significant increase of the PCE of PTB7/PC₇₁BM-based OPVs from 7.58% to 8.33%, which is mainly due to the remarkable increase in the hole mobilities of the devices. In addition, DPP-DTT could enhance the light absorption of the OPVs in the wavelength region longer than 700 nm, as indicated by the external quantum efficiencies (EQEs) of the devices and the UV-visible absorption spectra of the DPP-DTT polymer. Similarly, the average efficiency of the OPVs based on poly[4,8-bis(5-(2-ethylhexyl)thiophen-2-yl) -benzo[1,2-b;4,5-b']dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3-fluorothieno[3,4-b]thiophene-)-2-carboxylate-2-6-diyl)] (PBDTTT-EFT) and PC₇₁BM can be improved from 8.75% to 10.08% by adding the high mobility polymer PDVT-10. Besides the high hole mobility, the key issue in choosing a suitable high-mobility polymer here is to have the similar HOMO levels of the high-mobility polymer and the polymer donor in OPVs, because the energy levels of the molecular orbitals can influence carrier recombination and lifetimes in the devices. The technique is convenient and cost effective since there are plenty of choices in using high-mobility polymers. This work demonstrates a novel approach for improving the efficiencies of organic solar cells by adding high-mobility polymers.

Understanding the static and dynamic properties of excitons in thin films of conjugated organic molecules is critical to their applications in optoelectronics. However, the presence of mesoscopic disorder in these films and the large system size present particular challenges for conventional electronic structure methods. In order to surpass these difficulties we utilize a general strategy for combining semi-empirical electronic structure method (i.e., PPP method) and classical molecular dynamics simulations. Specifically we investigate the low-lying electronic excitations in films of varying morphologies made up of hundreds of sexithiophene molecules. The accuracy of the model (i.e., the excitonic energy and coupling from the PPP method) is benchmarked against the TDDFT method, the functional of which is chosen based on its comparison to experiment and the CC2 results in the literature. With this model, it is feasible to explicitly simulate Frenkel exciton dynamics in large disordered films (several thousands of non-hydrogen atoms) over time scales of tens of picoseconds. The model is able to qualitatively reproduce the absorption spectra for sexithiophene in solution and the sexithiophene film with standing molecular units. Finally, the effects of disorder on exciton delocalization in several sexithiophene films of different morphologies will be illustrated based on our calculations.

Significant efforts and a rapid progress have been made today in modeling the device physics as well as practical demonstration of high-efficiency quantum dot intermediate band solar cells (QD-IBSCs). These achievements have been possible due to mature and highly uniform III-V QDs and nanostructure material growth and processing technology.
Presently, demonstration of QDs-based IBSCs is undergoing three main development stages. The first challenge is to develop technology to fabricate high-density QDs arrays or superlattice of low defect density and long electron lifetimes. The areal density of QDs has direct influence on the generation and recombination processes via IB states because the density of states of IB (N_IB) is linked to the areal density as, N_IB = N_areal × N_stacks / W, where N_areal is the areal density of QDs array, N_stacks is the number of QDs layer stacks, and W is the intrinsic QD region width, respectively. The strain-compensated or strain-balanced growth technique can significantly improve the QDs quality and characteristics of solar cells even after stacking of N_stacks > 100 QDs layers in lattice-mismatched heteroepitaxy. The second effort is to increase the carrier lifetimes in IB states by controlling the recombination rates such as by using (1) a type-II QD heterostructure that allows a fast spatial separation of electrons and holes, (2) a high potential barrier preventing the thermal electron escape out of QDs, and (3) an electric field damping layer supressing the field-assisted escape of electrons out of QDs. And the last challenge is to realize ideally half-filled IB states to maximize photocurrent generation by two-step IR photon absorption. For this, the doping of QDs, photofilling of QDs by light concentration, as well as photon confinement thin-film structure are all considered important and actively investigated.

Interest in using III-V quantum dots (QDs) to increase the sub-bandgap photocurrent has arisen following the proposal by Luque and Marti (1997) for an intermediate band solar cell (IBSC). We have recently demonstrated the existence and the large contribution of extended tailing states in a QDSC to the overall below-bandgap photocurrent. This continuum distribution of tailing states, superposed with the QD and the wetting layer (WL) confined energy states, facilitates carrier relaxation and acts as an escape pathway for below-bandgap photogenerated carriers. In this work, a much narrower Urbach tail was designed and led to less coupling between quantum states and the background continuum states. Without an efficient carrier extraction pathway, mobile electrons are transported through the QD layers and the strain potential at the interface of In_0.5Ga_0.5As and GaAs drives them towards confined lower states. In this way, QDs can function as “multi-electron” trapping centers. For an InAs/GaAs QD device with a much larger tail width, the trapping effect is much smaller due to the existence of an efficient carrier extraction pathway.
Besides partial filling of the quantum states, electron-doping produces negatively charged QDs that exert a repulsive Coulomb force on the mobile electrons, thus altering the electron trajectory and reducing the probability of electron capture, leading to an improved collection efficiency of photo-generated carriers, as indicated by an absolute above-bandgap EQE measurement. The resulting redistribution of the mobile electron in the planar direction is validated by the observed photoluminescence (PL) intensity dependence on doping. Reduced QD state absorption with increased n-doping has also been observed from high resolution below-bandgap external quantum efficiency (EQE) measurement, and is a direct consequence of the Pauli Exclusion Principle.

Current high-efficiency state-of-the-art photovoltaic (PV) technology depends upon crystalline semiconductor substrates such as gallium arsenide, indium phosphide, or germanium. Such substrates contribute to the bulk of the cost and weight in current high efficiency photovoltaics, which is detrimental to the primary figures of merit for both terrestrial and space PV: cost specific power and mass specific power respectively. Techniques such as epitaxial lift-off and substrate recycling have been proposed to make lighter and cheaper PV, but do not completely eliminate the need for expensive substrates, and still limit the starting lattice constant for epitaxy to that of a material where a crystalline substrate is available. A recrystallized virtual substrate has the potential benefit of possessing an arbitrary lattice constant, not constrained to the common Si, GaAs, Ge, or InP. Molybdenum foil substrates offer a compelling alternative to conventional substrates because the coefficient of thermal expansion is comparable to that of conventional semiconductor materials, but the material grown on such a substrate will be polycrystalline with arbitrary crystal orientations and grain sizes. As such, it is important to understand the effects of crystallinity and of nucleation properties of epitaxy on substrate material. One specific challenge to nucleating GaAs on Ge is the formation of antiphase domains (APDs), or regions of antisite defects in nucleation of a contiguous film on the host substrate. This is particularly an issue in when considering polycrystalline substrates that may have random or slightly textured grain orientation. An enhanced understanding of the impact of nucleation defects can assist in understanding degradation mechanisms between monocrystalline and polycrystalline substrates. In this work we have demonstrated that APD formation is not an insurmountable problem on polycrystalline substrates by growing and fabricating multiple devices on both poly-Ge and poly-GaAs substrates. We have also developed a comprehensive spectral responsivity (SR) and current-voltage (IV) model for crystalline and polycrystalline GaAs solar cells grown on arbitrary substrates, accounting for intragrain material quality as well as the effects of grain size, boundary recombination velocities, and nucleation induced defects. Growth was done by chemical vapor deposition using a low temperature buffer layer to mitigate effects of APD formation versus grain boundary orientation. Device efficiencies over 12% were demonstrated on poly-Ge substrates (no AR coating) with 250 mu;m grain size. In the final presentation, we will show additional characterization results as well as application of our SR and IV models to pinpoint the effects of APDs as well as other nonradiative grain boundary processes.

Semiconductor nanowires (NWs) have received increasing attention in recent years for solar cell applications, due to 1) their intrinsic antireflection property for enhancing light absorption; 2) their small footprints efficiently relaxing the lattice-mismatched strain and thus enabling the construction of multi-junction solar cells with optimal band gap combinations and/or the growth on different substrate materials such as silicon; and 3) significant cost reduction due to much less material usage. With highly suitable bandgap, superior carrier mobility and well-developed synthesis techniques, axial p-i-n junction InP nanowire array solar cells have been reported with significantly improved performances (up to 13.8%) recently. It was found that by simply reducing the length of top n⁺-segment or increasing the length of the bottom p⁺-segment the solar cell performance can be enhanced due to the improved light absorption and carrier collection in the NW solar cell (NWSC) arrays. However in addtion to the length of doped regions, for NWSCs the critical optimisation involves not only a much more complicated junction design in terms of the length, position and doping concentration of different structral segments, but also the highly time-consuming experimental calibration and implementation due to the complexity of dopant incorporation, diffusion, and growth rate variation during the NW growth. In this work we demonstrate the use of nano-scale electron beam induced current (EBIC) technique with combined numerical simulations to facilitate direct characterisation and optimization of axial p-i-n junctions in InP array NWs for solar cell applications.
The InP nanowire array solar cell structures with different axial p-i-n junction designs were grown by selective-area metal organic vapour phase epitaxy. The axial junctions were characterized by EBIC measurements within an FEI Helios 600 nanolab focused ion beam system. Combined with coupled numerical optical and electrical simulations, it has been found that the un-intentional background doping in the nanowire solar cell has a significant effect on junction characteristics. By careful design and modifying the doping profile, the width and position of the p-n junction within the NW can be controlled and placed closer to the top of the nanowires which is favourable for both light absorption and carrier collection in the NWSCs. Solar cell devices have also been fabricated and characterized to confirm the EBIC and simulation results.

The ability to reliably and cost-effectively reuse single crystal substrates to seed epitaxial growth is a virtual requirement of realizing high-efficiency, thin-film, III-V solar cells at production volumes. Controlled spalling is an emerging technique, which results in the rapid exfoliation of a thin, single crystal layer by propagating fracture parallel to the wafer surface. We examine the conditions required for controlled spalling of (100) GaAs, compare experimental results to theoretical spontaneous spalling mechanics, and demonstrate that the fracture process does not generate extended defects that would inhibit quality regrowth or degrade delaminated cell efficiency. Resolving these issues illuminates a viable path forward for reducing III-V device cost through the controlled spalling of (100)-oriented GaAs devices and subsequent wafer reuse when these processes are combined with a high-throughput growth method.

Carbon nanotubes (CNTs) are quasi-one-dimensional materials with a unique set of electronic and optical properties. Today, there is considerable interest in semiconducting CNTs as the base light-absorbing material in organic solar cells, owing to their tunable band gap, mechanical and chemical stability, and solution processability. Exciton dissociation at the CNT/C₆₀ interface has an internal efficiency close to 100%, but the external power conversion efficiency of CNT-based photovoltaics suffers from the poor exciton mobility in the light-absorbing CNT film. While the inter-tube dynamics of photo-excited carriers has been the subject of many experimental studies, as a result of experimental complexity and sample variability, there is no consensus on the role of environment variables such as temperature, impurities, and the morphology of the CNT film in the exciton transfer process between various CNTs.
Here, we calculate the inter-tube exciton transfer rate beyond the conventional Forster theory. We solve the Bethe-Salpeter equation in CNTs to obtain the optically bright and dark excitonic states and use Fermi&’s golden rule to calculate the exciton transfer rate between pairs of semiconducting CNTs. We compare the exciton transfer rate when the donor and acceptor CNTs are nearly parallel to each other (i.e., transfer within a CNTs bundle) with the case of nonparallel CNTs (i.e., transfer between misoriented bundles). We calculate the transfer rate of delocalized E₁₁ excitons between two (7,5) CNTs within a bundle to be ~10¹¹ s^-1. However, we find the transfer rate of excitons from (7,5) CNTs to (8,7) CNTs in the same bundle to be about 3 orders of magnitude lower because of momentum conservation rules. In contrast, we show that the inter-bundle exciton transfer rate is not sensitive to the donor and acceptor chiralities.
Furthermore, we study the effects of sample inhomogeneity, impurities, screening in the surrounding medium, and temperature on the exciton transfer rate. Increased exciton confinement due to inhomogeneities increases the transfer rate. Exciton thermalization into optically inactive branches via phonons and impurities lowers the rate by an order of magnitude, while increased Coulomb screening due to the surrounding environment reduces the exciton transfer rate slightly. We also show the temperature dependence of the exciton transfer rate.

Perovskite solar cells (PSCs) based on organometal halide light absorbers have been considered a promising photovoltaic technology due to the rapidly increasing power conversion efficiency (PCE) along with very low material costs. One of the most commonly used hole transport material (HTM) in the PSCs is SpiroOMeTAD which is quite expensive due to complicated synthetic procedures or high-purity requirement.[1] Recently our group have synthesized a series of triphenylamine-based hole-transport materials (HTMs).[2,3] By increasing the electron-donating strength of functional groups the oxidation potential was systematically shifted resulting in higher open-circuit voltage and efficient PSCs. In the present study a series of novel inexpensive HTMs based upon the triarylamine (TAA) moiety and SpiroOMeTAD analogues have been synthesized and applied in Perovskite Solar Cells revealing competitive performance as compared to SpiroOMeTAD.
References
[1] M. Green, A. H-Baillie, H. Snaith, Nature Photonics, 8, (2014) 506-514.
[2] M. Planells, A. Abate, D. J. Hollman, S. D. Stranks, V. Bharti, J. Gaur, D. Mohanty, S. Chand, H. J. Snaith, N. Robertson, J. Mater. Chem. A2013, 1, 6949.
[3] A. Abate, M. Planells, D. Hollman, V. Bharti, S. Chand, H. Snaith, N. Robertson, Phys. Chem. Chem. Phys.2014, 17, 2335.

This study focuses on the energy transfer and electron transport in the photovoltaic device with uniquely designed nanostructures of an acceptor/donor energy system and an electron carrier delivery system. Furthermore, the effect of both the assembled nanostructures of acceptor/donor energy delivery and the different morphology of nanomaterials on photovoltaic performance are studied.
Förster resonance energy transfer between the xanthene dye (donor of energy) and a new polymethine dye (acceptor of energy) was studied on the surface of TiO₂ films in the dye sensitized solar cells (DSSCs). The sensitization of semiconductor films by the donor acceptor compound leads to a doubling of energy conversion efficiency. Measurements show that this is due to the widening of photo sensitivity of the cell in the blue region of the spectrum. Increasing the total energy absorbed leads to increasing the number of generated charge carriers at the interface of TiO₂ and the dyes.
Zinc Oxide nanoarrays with rod and sheet morphologies were fabricated and assembled for electron transport study of DSSCs. A photoluminescence spectra study indicated that nanosheets had more defect density compared to nanorods, which resulted in a lower open circuit photovoltage. Electrochemical impedance spectroscopy (EIS) and dye absorption amount analysis further explained the high short-circuit current density (J_sc) of the DSSC with ZnO nanosheet structure, based on the dye-loading amount, effective electron lifetime, and electron density.

Colloidal quantum dot (CQD) solar cells have attracted high attention as a new type of solar cells due to the tunable energy band gap by size of dots, the wide absorption spectrum window extending to near infrared (NIR), and the all-solid based cell structure. In addition to these fundamental advantages, structured electrodes and substrates have been received great attentions to improve the cell performance. Our previous study showed a periodic grating structure is one of the structures enhancing the light absorption [1]. In this research, we fabricated substrates with periodic grating structures for PbS CQD solar cell devices and compared the devices with the devices with a flat substrate. The periodic grating structure was fabricated by interference lithography. An indium tin oxide (ITO) was then deposited on a photoresist of grating structure by pulsed laser deposition (PLD) following by TiO₂ deposition by PLD. PbS was deposited by the layer by layer (LBL) deposition. In order to fabricate conformal films, dip coating was selected that allowed us to control the films thickness precisely. After deposition of top electrodes, photoelectrochemical characterizations were employed and the performance of these devices was investigated. The devices with the grating structures showed higher J_sc than the one with the flat structure due to the higher EQE from the exciton peak of PbS.
References
[1]. Y. Fu, A. Dinku, Y. Hara, C. Miller, K. Vrouwenvelder, and R. Lopez, "Modeling photovoltaic performance in periodic patterned colloidal quantum dot solar cells," Opt. Express 23, A779-A790 (2015).

Bulk heterojunction (BHJ) polymer solar cells are an area of intense interest due to their potential to result in printable, inexpensive solar cells that can be processed onto flexible substrates. The active layer is typically spin coated from the solution of polythiophene derivatives (donor) and fullerenes (acceptor) and interconnected domains are formed because of phase separation. However, the power conversion efficiency (PCE) of BHJ solar cell is restricted by the presence of unfavorable morphological features, including dead ends or isolated domains. Here we introduce Poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinylene]:Poly(vinylpolypyrrolidone): [6,6]-Phenyl C₆₁ butyric acid methyl ester (MEH-PPV:PVP:PCBM) electrospun nanofiber into BHJ solar cell for the active layer morphology optimization. Larger interfacial area between donor and acceptor is abtained with electrospinning method and the high aspect ratio of the MEH-PPV:PVP:PCBM nanofibers allow them to easily form a continuous pathway. The surface morphology is investigated with atomic force microscopy (AFM) and scanning electron microscopy (SEM). Electrospun nanofibers are discussed as a favorable structure for application in bulk-heterojunction organic solar cells.

Bulk heterojunction (BHJ) polymer solar cells are an area of intense interest due to their advantages such as mechanical flexibility, low costs, and the easiness of the fabrication. The active layer is typically spin coated from the solution of polythiophene derivatives (donor) and fullerenes (acceptor) and interconnected domains are formed because of phase separation. However, the power conversion efficiency (PCE) of BHJ solar cell is restricted by the disordered inner structures in the active layer, donor or acceptor domains isolated from electrodes and long path conduction. Here we report a self-assembled columnar structure formed by phase separation between poly[N-9Prime;-hepta-decanyl-2,7-carbazole-alt-5,5-(4prime;,7prime;-di-2-thienyl-2prime;,1prime;,3prime;-benzothiadiazole)] (PCDTBT) and polystyrene (PS) for the active layer morphology optimization. The BHJ solar cell device based on this structure is promising for exhibiting higher performance due to the shorter carrier transportation pathway and larger interfacial area between donor and acceptor. Moreover, the absorption spectrum is expanded by additional Poly(3-hexylthiophene-2,5-diyl) (P3HT) in this system, which further enhanced PCE. The surface morphology is investigated with atomic force microscopy (AFM) and the columnar structure is studied by investigation of cross-section of the blend thin film of PCDTBT and PS with the transmission electron microscopy (TEM). The effects of different parameters, including solvents and annealing time, on morphology are discussed. The different morphological structures formed via phase segregation are correlated with the performance of the BHJ solar cells fabricated at the Brookhaven National Laboratory.

We systematically tailored the work function (WF) of conducting polymer hole extraction layer (HEL) using a self-organized HEL (SOHEL) to improve the device performance and clearly understand the energy level alignment of the HEL with the universal photoactive materials as polymers and orgnic-inorganic hybrid perovskites. We employed P3HT(Poly(3-hexylthiophene-2,5-diyl)), PCDTBT(poly[N-9"- heptadecanyl-2,7-carbazole-alt-5,5-(4',7'-di-2-thienyl-2',1',3'-benzothiadiazole)]), MAPbI₃ (methylammonium lead triiodide), and MAPbBr₃ (methylammonium lead tribromide) for the photoactive materials with various ionization potential IP. We measured the midgap state of polymer and perovskite photoactive materials through the energy level pinning effect depending on the conducting substrate with the various work function. In addition, we investigated the interfacial energy state of a hole extraction conducting polymer due to the suppressed density of states extending close to Fermi level by an insulating dopant polymer (e.g. PSS or PFI). Although the Fermi-level of SOHEL becomes pinned to the midgap state of the donor materials, interface energy state of SOHEL further increases to the IP of the donor materials. Therefore the energy offset between photoactive materials with SOHEL can be significantly decreased to achieve high built-in potential and open circuit voltage and power conversion efficiency of PCDTBT, MAPbI₃, and MAPbBr₃. Our SOHEL is promising HEL for a good energy level alignment with universal photoactive mater with deep IP level and the midgap state. Furthermore, our SOHEL significantly prolonged the stability of organic solar cells (half life time : 5700h) compared with pristine PEDOT:PSS (300h) under continuous irradiation of air mass-1.5 global simulated sunlight at 100 mW/cm². We attribute this LT prolongation to the diffusion-blocking ability by the self-organized fluorocarbon layer at the surface of our SOHELs.

The goal of ‘3rd-generation&’ photovoltaics is to identify and realize technologies that could surpass the Shockley-Quiesser efficiency limit for single-junction cells under solar illumination. Here, we present a novel approach using multi-excitonic interactions in nanostructured materials that could enhance existing silicon-based photovoltaics. Specifically, we show that two thin-film excitonic materials—organic semiconductors and colloidal nanocrystals—can be combined to create hybrid, passive devices that achieve broadband, non-coherent up-conversion from the short-wave infrared to the visible. (SWIR; lambda;:1-3mu;m)
To achieve this, we synthetically tune the bandgap of PbS nanocrystals to absorb SWIR photons (lambda;>1mu;m). Photoexcitations then undergo Dexter transfer to sensitize the spin-triplet state in an organic semiconductor (rubrene). Higher-energy (lambda;~612nm) singlet excitons are generated via diffusion-mediated triplet-triplet annihilation, and the emissive efficiency is boosted using a molecular dye (DBP) in a guest:host configuration. We achieve an upconversion efficiency of 1.2±0.2% with lambda;=808nm excitation at 12W/cm².
However, transient spectroscopy indicates that there remains a synthetic challenge to modify the nanocrystal surface to accelerate triplet transfer (tau;~400ns) without creating competitive non-radiative channels. Further work will focus on elucidating the governing physics to achieve higher performance, particularly by enhancing exciton transport to allow the use of thicker—more absorptive—nanocrystal films.
We expect that our hybrid approach may prove broadly applicable to achieve upconversion in solar and SWIR-detection applications, where effective molecular phosphors are lacking—indeed, quantum dots are ideal SWIR sensitizers, as their excitonic spin-states are functionally mixed at room temperature, and both the optical gap and ionization energy can be tuned via colloidal synthesis.

Thin-film photovoltaics are important when the power/weight ratio, flexibility, ease of orientation, and possibly cost and/or shock resistance are critical. Such films have thicknesses typically less than 1 micron, absorb long-wavelength (red) light poorly even with a metallic back reflector, and cannot have standard texturing.
The near-field enhancement around plasmonic structures that decay over nanometers, and far-field scattering from individual or arrayed nanoparticles or patterned nanostructures,^1,2 are well-understood. There is relatively little understanding or control of scattering from nanoparticles/-structures in the complex intermediate-field regime where the substrate thickness is comparable to a wavelength of light -between nanoscale field enhancement and long-range diffraction effects. We investigate front-surface-coated nanoparticles, combined with an Indium Tin Oxide top surface coating, to enhance absorption and current density of thin film solar cells. Random arrays are produced via large-scale chemical approaches appropriate for roll-to-roll processes (such as spray coating) after scale-up, while ordered arrays are produced lithographically.
First, we study scattering from individual and sparse arrays of metallic nanoparticles in baseline polymer films, and model, characterize, and analyze their scattering, mode properties, and absorption. We seek to understand whether scattering into the underlying substrate can be enhanced and optimized with nanoparticles. Classical polarization-dependent dipole scattering profiles are found from non-interacting SiO₂ nanoparticles in a liquid cell, with diameters in the 40-60 nm, when illuminated with 785 nm. We analyze quasi-spherical Ag nanoparticles with diameters of 50, 120, and 300 nm in water (made with hydrogen synthesis³), and then include them in thin (< 300 nm) polymer layers spuncoat onto polyvinyl-4 pyridine-coated glass or silicon substrates. We also analyze, using the same experimental procedure, ~35 nm thick “large” and “small” Ag nanoprisms with in-plane dimensions > 500 nm and < 150 nm, respectively, dielectric, and other nanoparticles. Lithographically-patterned Al nanostructures, with fixed pitches, are optimized for enhanced scattering into submicron thick amorphous silicon layers with an aluminum back reflector. Optical experiments include: total integrated scatter and absorption measurements (higher signal-to-noise ratio but loss of angular information), laser-based angular-resolved scatterometry (angular information but discrete wavelengths), and imaging scattering using a high NA lens. We analyze sparse arrays to verify the single-particle-scattering picture, and then make denser arrays and study/analyze how interactions between the particles influence the scattering.
[1] F. Luekermann, et. al., Appl. Phys. Letts. 100 (2012) 253907.
[2] I. Diukman and M. Orenstein, Solar Cell Mats. 95 (2011) 2628.
[3 M. K. Kinnan and G. Chumanov, J. Phys. Chem. C.114 (2010) 1796.

Metal oxides are important materials systems that enable multiple optoelectronic applications due to the excellent combination of electrical and optical properties as well as their inert nature and non-toxicity. However, the implementation of these materials in flexible devices is somehow restrained by the polymeric substrates&’ ability to withstand thermal processes necessary to obtain specific crystalline phases. Recent studies suggest the use of organic binders along with crystalline nanoparticles for low temperature fabrication of porous metal-oxide structures; other approaches make use of in-situ chemical reactions to generate localized heat and promote crystallization of precursors into thin film metal oxides. The latter requiring multiple deposition-annealing steps at temperature ranges not entirely compatible with most polymeric substrates.
In this work, we study the direct writing of TiO2 and ZnO on polymers at low temperatures. Direct writing allows for high control of the macro and microstructure, is suitable for large-scale applications and reduces waste to minimum by printing exact amounts of ink in specific pre-defined substrate locations. Additionally, the use of sol-gel inks allows for highly controlled stoichiometry and microstructure of the desired functional patterns.
Therefore, we investigate the effect of crystalline phase addition to metal-organic inks in lowering the crystallization temperature during the processing of such inks into directly written patterns.
Ink characterization includes viscosity measurements and thermo-gravimetric analysis. Direct writing of the inks is conducted using nozzle based robotic deposition. The patterned features functionality is investigated optically, electrically and mechanically.
It is shown that the viscosity of the inks is affected by the amount of crystalline phase added and can be altered for specific deposition methods beyond direct writing. The resulting materials have comparable electrical behavior to conventionally crystallized compounds from single-phase sol-gel inks. Also, the resulting microstructure is dependent on the amount of the crystalline phase added as well as the annealing temperature.
It is believed that the combination of engineering of metal-organic inks with both unconventional crystallization routes and direct writing on polymers, may hold the key for manufacturing the next-generation of flexible optoelectronic devices.

Ferroelectric materials have recently attracted much attention as promising candidates for use in photovoltaic devices, and for the coupling of light absorption with other functional properties. Breaking of strong inversion symmetry due to spontaneous polarization in these materials develop a desirable separation of photo-exited carriers and produces voltages higher than its band gap. However a big challenge faced in ferroelectric-photovoltaic devices is to overcome the very low output photocurrent. Ferroelectric oxides are mechanically, chemically, and thermally quite stable. Theoretically, it has been proposed by some research groups that metallic defects and/or doping/substitutions of transition metals and/or metal ions can reduce the band gap of ferroelectric materials which is the basic requirements for their applications in photovoltaic devices. We have therefore studied structural, microstructural, elemental, electrical, and optical properties of Pb(Zr_0.20Ti_0.80)_0.70Pd_0.30O₃ (PZTP:30) bulk samples and synthesized Palladium substituted PZTP thin films on ITO coated glass and oriented (100) LSAT substrates with bottom layer of La_0.67Sr_0.33MnO₃ (LSMO) by pulsed laser deposition technique. X-ray diffraction studies revealed the single-phase formation without any impurity phase and with only (100) peaks in thin films. The existence of ferroelctricity and switching of polarization are confirmed from the band excitation Piezo Force Microscopy (PFM) in PZTP thin films. XPS studies confirmed the existence of Pd in thin films. SEM micrographs of thin films revealed that it has medium grain sizes varying between 5 to 10 mu;m. The frequency dependent dielectric constant and loss tangent of LSAT/LSMO/PZTP films showed almost constant dielectric constant around 3000 and relatively low loss tangent (<0.2) at frequencies below 10 kHz. Well saturated ferroelectric loop with remanent polarization ~40 mu;C/cm² confirmed the presence of ferroelectricity in this material. Optical properties were investigated by UV-visible spectrometer of PZTP films on glass/ITO, it exhibits 60% transmittance at 600 cm^-1, with a reduction of only 30% compared with pure ITO/glass substrate. Direct and indirect band gaps of PZTP films with a thickness of 200 nm were observed ~ 3.4 eV and 2.2 eV respectively. The current vs voltage measurement and transient characteristic of the glass/ITO/PZTP/Au structures will be discussed.

A new high rate deposition method has been used to fabricate thin film CdTe photovoltaic devices using pulsed dc magnetron sputtering. The devices have been deposited in superstrate configuration on to a commercial fluorine doped tin oxide transparent conductor on soda lime glass. The cadmium sulphide and cadmium telluride thin films were deposited from compound targets. The magnetrons were mounted vertically around a cylindrical chamber and the substrate carrier rotates so that the layers can be deposited sequentially. The substrates were held at 200^oC during deposition, a process condition previously found to minimize the stress in the coatings. Optimization of the process involved a number of parameters including control of pulse frequency, power and working gas pressure. The devices deposited using the process are exceptionally uniform enabling the CdTe absorber thickness to be reduced to ~1um. The as-deposited material is dense and columnar. The cadmium chloride treatment increases the grain size and removes planar defects. The microstructure of the films before and after activation has been characterised using a number of techniques including transmission electron microscopy, Energy Dispersive mapping and X-ray diffraction and these measurements have been correlated to device performance. The deposition rate is much higher than can be obtained with radio-frequency sputtering and is comparable with methods currently used in thin film CdTe module manufacturing such as Vapour Transport Deposition and Close Space Sublimation.

Various metal sulfides such as FeS₂, Cu₂S, and Bi₂S₃ show great promise for application as absorber material in thin-film solar cells due to their suitable band gaps accompanied by high absorption coefficients. In most of the cases solar cells have been prepared based on thin films of these materials. We propose the use of these materials in the form of nanoparticles (NP). The sulfurization of metal oxides allows an easy preparation of metal sulfides as well as the preparation of oxide/sulfide heterojunctions. The large surface to volume ratio of NP facilitates the sulfurization compared to thin films.
We sulfurized Fe₂O₃, CuO, and Bi₂O₃ NP by wet-chemical reaction with Na₂S(aq.) or (NH₄)₂S(aq.) as well as by H₂S plasma treatment. The influence of the sulfurization conditions (e.g. sample temperature, exposure time) and subsequent thermal annealing on the structural and electronic properties of the NP was investigated. The microstructure was analyzed by Raman spectroscopy and X-ray diffraction (XRD). Photoluminescence (PL) spectroscopy, and photothermal deflection spectroscopy (PDS) were applied to analyze the electronic properties and the absorption, respectively. In addition the morphology and size of the NP was investigated by scanning electron microscopy (SEM).
Raman spectroscopy and XRD confirm that the oxide NP were sulfurized. After sulfurization at room temperature the NP are amorphous. Subsequent thermal annealing in inert atmosphere leads to the crystallization of the NP; e.g. for Bi₂S₃ below 300C. The CuO and Bi₂O₃ were converted into phase pure Cu₂S and Bi₂S₃ NP, respectively, whereas a phase mixture of marcasite and pyrite was observed for FeS₂.
The opto-electronic quality of the NP was investigated by PDS and PL. Therefore, band edge as well as defect related emissions and the sub band gap absorption were analyzed. For example, the Cu₂S NP showed a strong emission at 1.2eV in good agreement to the band edge deduced from the PDS spectrum.
Furthermore, the stability of the sulfurized NP was investigated. When exposed to ambient conditions the Cu₂S oxidizes. Moreover, the Cu₂S are also not stable under inert atmosphere and converted into substoichiometric Cu_xS phases. In contrast Bi₂S₃ and FeS₂ are very stable under atmospheric conditions.
Therefore, Bi₂S₃ is advantageous on the one hand over FeS₂ due to its phase purity and on the other hand over Cu₂S due to its stability. Thus, we conclude that Bi₂S₃ is the most promising material for photovoltaic application among the three investigated metal sulfides.

Atomic Layer Deposition (ALD) has been used as a successful technique for depositing composite thin films. Two or more materials can be deposited in alternating discrete layers to create multilayers structures and hence the tenability of material physical properties can be achieved. It was shown previously that ZnO/Al₂O₃ composite film prepared using ALD spans a broad range of physical properties due to the distinct ZnO and Al₂O₃ films characteristic obtained from ALD. The ZnO film grown by ALD tends to be conductive, crystalline and rough however Al₂O₃ grown by ALD is insulating, amorphous and smooth.
In this work three different ZnO/Al₂O₃ cycles ratio of 5:1, 19:1and 35:1 layers are prepared to investigate the electrical properties of the multilayer film. In addition different deposition conditions like deposition temperatures have been varied from 1500-2000 and deposition pressure is swept from (80mtorr-200mtoor) to test the optimum growth condition of the composite films. The starting substrate is p-type Si with resistivity of 0.02Omega;. The native oxide is etched in 1:10 BHF. After the ALD growth, diodes are fabricated using e-beam evaporated aluminum on the samples through a shadow mask. The current-voltage (IV) measurements under forward bias condition of the five samples are tested. In addition, four probe measurements using hall set up are carried out to investigate the resistivity. The results show that the 19:1 ZnO/Al₂O₃ cycles ratio repeated ten times has the lowest resistivity of 10.4 mohm.cm. The IV curve of this sample demonstrates pn diode behavior which mean that 19:1 ZnO/Al₂O₃ acts as conductive n-type layer. Finally the results indicate that composite films grown using ALD can enhance the ZnO electrical properties. These ALD grown layers can be used as low cost TCO&’s and as n-type layers for heterojunction solar cells.

Metallic nanowire networks have been shown to exhibit a low sheet resistivity and simultaneously high optical transmission, the prerequisite for high-performance transparent electrode contacts. In this work we use Pd-Au nanowire networks to fabricate a metal insulator semiconductor silicon solar cell. The metallic nanowires have a manifold capability in that they facilitate charge carrier extraction and separation inside an adjacent semiconductor, while simultaneously allowing for a high optical transmission.
To isolate the effect of the metal nanowire network on the device performance we focus on a simple fabrication process for the front side of silicon solar cell, while using a state of the art contact scheme employed for n-type silicon HIT solar cells on the back side. The Pd-Au nanowire network, which is fabricated by electron beam lithography and metal evaporation, is separated by a 4 nm thin tunnel- and passivation layer from the underlying silicon half cell. This intermediate layer is grown at low temperature by ICP-CVD and ALD and consists of 3 nm a-Si:H and 1 nm Al₂O₃ which provide a high chemical and field effect surface passivation, that lead to minority carrier lifetimes of a few ms.
We show IV traces of solar cells under 1sun illumination, EQE measurements, as well as electron beam induced current measurements which prove that this metal insulator semiconductor solar cell allows for facile charge carrier separation and extraction. Our reflection measurements show a relatively high optical transmission, given the amount of metal used, which we can greatly improve by using an integrated nanowire-nanopyramid structure.
The simple fabrication steps employed are not limited to optically thick Si wafers but can be transferred to other thin film materials and even be applied for materials with short carrier diffusion lengths, due to the narrow spacings of the metal nanowire network front electrodes. Substrate conformal imprint lithography can be used to fabricate those networks over large areas.

Lack of moisture stability of organometal halide perovskites is a serious challenge to overcome before they can be widely used as solar cells. There has been a push to experimentally determine the mechanism of degradation in the presence of moisture, but the complete degradation mechanism is still unknown. To shed light on this problem, an atomistic understanding of the degradation mechanism is required. Towards this goal, we study the nature of water interaction with CH₃NH₃PbI₃ surfaces using density functional theory. We find that the orientation of the methylammonium close to the surface plays a significant role on the water adsorption. There is no indication of the water chemically reacting with the methylammonium cation. A range of water concentrations are explored via ab initio thermodynamics to illustrate the surface hydration process. These results provide important insights into the early stages of the degradation mechanism.

We studied the impact of a buffer layer on the net ionized carrier concentration, electron states in CdTe absorber, and conductivity mechanism in thin film CdTe/CdS solar cells by using temperature dependence of conductivity, ac admittance spectroscopy, and capacitance-voltage methods. We found that the choice of the buffer layer strongly modified the defect density of states in CdTe absorber. Compared to that observed in the CdTe/CdS/ZnO solar cells, the use of ZnS and ZnSe buffer layers results in lower doping densities in CdTe. The energy position of the dominant deep trap levels in CdTe are shifted closer to the midgap when ZnS and ZnSe are utilized as the buffer layers. The analysis of conductivity reveals a strong sensitivity to the type of a buffer layer used. In CdTe/CdS/ZnO solar cells space-charge-limited conduction was observed at forward and reverse bias that allows us to conclude that rectification properties of the CdTe/CdS junction are dependent on trap filling as influenced by electric field and temperature. Complimentary to the trap densities deduced using ac admittance spectroscopy, we were able to obtain the trap density and the energy position of defect levels which influence the charge transport in solar cells. Further, the trap densities deduced from the analysis of space-charge-limited conduction were systematically higher than those determined by means of ac admittance spectroscopy. These observations suggest that CdTe solar cells may function as metal-oxide-semiconductor (MOS) structures in which either a substantial trap density is present in the CdS window layer or the Fermi level is pinned at the TCO/buffer layer interface. Therefore, the Fermi level in a transparent conductive electrode and CdTe are aligned allowing for a modulation of a band bending in CdTe.

Earth-abundant kesterite Cu₂ZnSn(S,Se)₄ (CZTSSe) solar cells have shown significant improvement in material quality and photovoltaic performance over the past few years. However, the growth in efficiency has slowed, due at least in part to the intrinsic limitations imposed by the bulk defects in CZTSSe and in particular the band tailing believed to be caused primarily by Cu-Zn disorder. Multiple spectroscopy techniques have experimentally shown that disorder exists on the Cu-Zn sublattice. One method to reduce the intermixing is to replace Cu with elements that have larger covalent radii such as silver (Ag) in order to increase the energy barrier for site exchange with Zn. Silver has a covalent radius of ~0.153nm which is approximately 16% larger than that of Zn (as opposed to Cu which is only approximately 5% larger than Zn). Ag incorporation into CIGSe has successfully improved the Voc by increasing the band gap and reducing the atomic disorder (and resulting band tailing). In this study, Density Functional Theory (DFT) calculations are employed to evaluate the energetics of Ag_Zn + Zn_Ag defect pair formation and the effect of Ag-Zn intermixing on the electronic properties of Ag₂ZnSn(S,Se)₄ (AZTSSe) alloys. These intermixing energies are compared with the values calculated for CZTSSe. A mixed S-Se AZTSSe alloy was chosen for the calculations to mimic the ratio found to produce the highest-efficiency Cu₂ZnSnS_0.25Se_0.75 solar cells.

Quantum dots offer a powerful solution processable materials platform for third generation photovoltaics (PVs). In particular, the confinement of charge carriers leads to an enhancement of carrier multiplication at solar relevant wavelengths, whereby a kinetically hot carrier may excite additional electrons across the band gap. The appreciable increase in photon to electron conversion could benefit PV applications, permitting power conversion efficiencies beyond a traditional thermodynamic limit for single-junction PVs. However, quantifying the multiplication yield in conducting films of quantum dots have proven challenging. Furthermore, losses in conducting quantum dots films due to recombination and trapping have greatly inhibited the applications of quantum dots and are difficult to directly probe. Here we utilize a transient photocurrent measurements with ~50 picosecond temporal resolution in conjunction with ultrafast optical excitation to directly quantify the multiplication yield as well as interrogate the trapping and recombination processes. Through tuning of the optical photon energy we resolve the presence of multicarrier Auger recombination at low fluence, a direct signature of the presence of the multiplication process. Engineering of the film treatment enables the complete separation of the additional charges generated through carrier multiplication prior to Auger recombination. Analysis of the longer timescale data allows us to follow the evolution in the recombination and trapping mechanisms in time across the range of temperatures. Furthermore, the temperature dependence of the transients allows us to assign the mechanisms of charge transport in our films and to elucidate the effect of various film treatments on their electrical properties.

Diatoms are unicellular, photosynthetic algae that can be found in aquatic systems across the world with numerous species and a variety of shapes, with their sizes ranging 1 um-4 mm. The hierarchical porous structures of their silica micro-shells (also known as frustules) have raised interest in their application to a host of different areas, including bio-sensors, drug-delivery systems, solar energy, and optical sciences, due to their unique mechanical, material and optical/photonic properties.
In most optical research and applications, especially in diatom-based solar cells, frustules are usually viewed as either individual micro-particles with 3D structures and orientations, or generalized as macroscopic porous materials whose performance is not affected by an individual frustule&’s shape or orientation. Due to the lack of an effective way to arrange frustules into uniformly oriented and compact monolayers, little research has taken both the amount and the orientation of frustules into account.
In this paper, diatom frustules of Coscinodiscus sp. were used. To form compact, self-assembled monolayers of frustules on a water surface, the frustule surface was initially be rendered hydrophobic. Hydrophilic thin glass cover slips were prepared for the experiments involving the collection of the frustule monolayers. To this end, the cover slips were soaked in 10% sodium dodecyl sulfate-water solution (SDS) for 2 hours. To form a frustule monolayer, the prepared frustule suspension was slowly added to DI water drop by drop. The frustules floated on the water surface and formed a frustule layer. ~20 mu;l 2% SDS solution was then added to the water and the frustules were pushed together, forming a compact monolayer, which could then be lifted from water surface with a modified cover slip. In order to uniformly adjust the frustule orientation, N₂ bubbles were blown under the water surface for 10 s, prior to the addition of SDS solution.
Whit the method to form large area, close-packed self-assembled monolayers using Coscinodiscus sp. frustules (diameters ranging from 50 mu;m to 70 mu;m), we achieved up to 0.8 cm² of close-packed frustule monolayers on wafer surface. The orientation of these frustules was uniformly controlled with the concave side facing downwards by blowing N₂ gas bubbles below the water-diatom interface. UV-Vis measurement results showed the transmission rates of the uniformly oriented frustule monolayers were lowered compared to randomly oriented frustule monolayers, indicating a potential increase in light absorption efficiency, which may be beneficial in energy applications. Reports have been made about optimizing solar cells&’ performance with diatom frustules. The method introduced herein has the potential for large-scale fabrication of micro/nano-structure monolayers, which may be applicable to a dye sensitized solar cell technological applications, taking advantage of both diatom frustules&’ orientation and hierarchical structures.

Tin mono sulfide (SnS) is one of the most promising materials for the next generation solar cells for sustainable society, because (i) tin and sulfur are earth abundant, nontoxic and inexpensive elements; (ii) the direct band gap energy of SnS (~1.3 eV) is the most appropriate value for high efficient solar cell [1]; (iii) high absorption coefficient over 10⁴ cm^-1 at 1.32 eV [1], which is comparable to other thin films solar cell materials such as CdTe, CuIn_1-x Ga_xSe₂ and GaAs. Nowadays, the highest conversion efficiency of SnS based solar cells are 4.36% in heterojunction [2]. Since undoped SnS usually shows p-type conduction because of Sn vacancy and is difficult to convert into n-type conduction, all of SnS based solar cells have heterojunction structure. For fabricating pn homojunction, a great deal of effort are putting into realization of n-type conduction in SnS and there are two major approaches: One is aliovalent cation doping [3]. The other is isovalent cation doping[4].
The former method was conducted with substitution Sb for Sn. But n-type conduction was not observed [3]. A first principle calculation says that substitutions Sb for Sn do not work as donors [4]. On the other hand, isovalent cation substitution of Pb²⁺ for Sn²⁺ clearly showed n-type conduction [4]. Unfortunately, the large ionic radius difference between Pb²⁺ and Sn²⁺ leads large lattice parameter shift, indicating that lattice mismatch might be occurred at the “homo”-junction between p-type SnS and n-type Sn_1-xPb_xS. In addition, substitutional doping of Pb which is a typical toxic element reduces the advantage of nontoxic SnS. Hence, realization of n-type SnS with a non-toxic dopant whose ionic radius is as same as Sn²⁺ or S^2- is an important issue for high performance SnS homo-junction solar cell. In this study we report effectivity of substitutional doping of aliovalent anions, Cl^-, Br^- and I^- for S^2- for conduction-type conversion: n-type conduction in SnS was realized by Cl^- doping.
Doped and undoped SnS powders were synthesized by solid state reactions in an evacuated SiO₂ tube. The obtained powders were sintered in high density ceramic pellets by spark plasma sintering (SPS). X-ray diffraction (XRD) measurement, which can detect a few tenth of percent impurity, revealed that component phases in samples and lattice parameters. Chemical compositions were determined by wavelength dispersion X-ray fluorescence (XRF). Conduction type was determined by Seebeck and Hall measurements. Both of Seebeck and Hall coefficients of Cl^- doped sample were negative; a typical values were -330 mu;V/K and -160 cm³/C, respectively. In addition, the lattice parameter variation by Cl^- doping was less than 0.1%.
References
[1] K. T. Ramakrishna, et al., Solar Energy Materials & Solar Cells. 90, (2006) 3041.
[2] P. Sinsermsuksakul, et al., Adv. Energy Mater. (2014) 1400496.
[3] P. Sinsermsuksakul, et al., Chem Matter. 24, (2012) 4556.
[4] Z. Xiao et al., Appl Phys. Lett. 106, (2015) 152103.

Dye sensitized solar cells (DSSC) have been attracting attention since their invention. Low-cost production and moderate efficiency values obtained in these cells are main advantages of the DSSCs. However commercialization requires further improvements in dedicated assembly techniques for mass production of modules. During transfering small scale laboratory type cells to large area modules, efficiency of DSSCs is lowered due to insufficient conductivity of FTO (fluorine doped tin dioxide) coated glasses similar to all type of thin film solar technologies. In order to overcome this problem, silver grids are deposited on substrates which in turn decrease the sheet resistance of large substrates. However, these silver lines are strongly corroded by iodide/triiodide redox couple which are the widely used liquid electrolytes in DSSCs. Therefore, these lines are protected by hot melt polymers and large modules are divided into subcells which in turn decreases the active area and efficiency value of the modules. In addition, this strategy results in difficulties for hole drilling and lamination of the modules. An innovative approach for overcoming these problems and increasing the active area, silver lines embedded glass substrate design has been developed in analogous to buried contact technology in silicon solar cells. We employed ultrasonic spray deposition method for production of high quality FTO thin films to be employed in a novel silver embedded grid type dye sensitized solar module. Produced films exhibited dense and crystalline structure with homogeneous coverage on solar glass substrates. Obtained resistivity and light transmission values of FTO are almost same with commercially available FTO coated glasses used widely in the industry. After optimization of the chemistry and deposition conditions, 10x10 cm sized glass substrates could be produced for large area photovoltaic modules. Produced FTO films were used to construct dye sensitized solar modules in comparison with commercial coating. Efficiency value of 3% on active area could be achieved which is slightly higher than commercial substrate. In order to enhance active area, silver grid lines were embedded in glass substrate and FTO coating was deposited on the lines. Due to this novel design, we achieved 2.42% efficiency on the total area of the 5x5 cm sized module, proving that this design is suitable for enhancing efficiency values of parallel type dye sensitized solar modules.

Organometallic halide perovskite are extremely promising novel materials for thin-film photovoltaics, exhibiting efficiencies of over 20%. The low temperature processing of this material (below 130°C) enables one to develop the technology on flexible polymer films if low temperature processes can also be adopted for the other material layers such as compact and mesoporous TiO₂, allowing to fabricate such kinds of cell by high-throughput roll-to-roll production.
In this work we study the fabrication of a thin film perovskite layer on both glass and PET foil. Slot-die coating was used for depositing a homogenous perovskite film on large area substrates (15 cm x 15 cm), and the crystallization mechanism of such layer was studied via X-Ray Diffraction analysis. Starting from PbCl₂ salt mixed with CH₃NH₃I, we observed a CH₃NH₃PbCl₃ intermediate phase in the conversion mechanism and this phase was still present after CH₃NH₃PbI₃ perovskite formation. By comparing the perovskite crystallization on planar and mesostructured TiO₂ we fuond out that the conversion mechanism is the same, however there are differences on the crystals size along the whole process.
An ultra-high gas permeation barrier was then applied on the perovskite film by direct deposition and by lamination, showing a clear increase of crystal stability. Indeed, even after 1800 hour of storage in ambient condition (R.H. asymp; 50%), we cannot identify PbI₂ formation in the encapsulated perovskite layer. On the other hand, PbI₂ precipitation can be detected without encapsulation. Moreover if a shorter annealing time is used, degradation can be detected even for encapsulated samples. These results show that the perovskite film is much more stable than the starting precursor itself and the degradation is mostly due to water and oxygen permeation.
Different stacks were first tested on small area, and the more effectives ones were then up-scaled to 6 inch substrate. Slot die coating was used for the deposition of perovskite, Spiro-OMeTAD, mesoporous TiO₂ and a sol-gel based compact TiO₂. Homogeneous coating was achieved for all the layers by optimizing solution formulation and coating parameters.
Large are devices (2 cm x 2 cm) with all slot die coated active layers (from TiO₂ to Spiro-OMeTAD) were successfully fabricated on 6 inch glass substrates, while on PET foil the best result obtained involved the use of Atomic Layer Deposition for the compact TiO₂. In this way all the issues related to the use of an acidic sol-gel on a low temperature annealed ITO were avoided. Moreover if one uses Spatial ALD, this process is still compatible with atmospheric roll to roll production.
Further development involves ageing test of encapsulated devices, use of roll-to-roll Spatial ALD and use of greener solvents, especially for the perovskite and Spiro-OMeTAD inks. Then a roll-to-roll production of perovskite solar cells will be tested.

Aluminum induced crystallization (AIC) is considered to be a promising technique for fabricating polycrystalline silicon (poly-Si) thin films, particularly for photovoltaic applications. This approach exploits the eutectic Al-Si system to facilitate the crystallization of amorphous silicon (a-Si) grown on Al, thereby enabling crystallization at significantly lower temperatures compared to that required for the solid phase crystallization of a-Si (approximately 600^oC). AIC is commonly realized by annealing a stack of a-Si/Al film grown on glass, at temperatures below the Al-Si eutectic temperature (400-500^oC). It is well known that a number of factors affect the AIC process, including annealing temperature, structure of the AlO_x membrane at the Si/Al interface, and the oxygen content and grain structure of the bulk Al film. The present study aims to investigate the effect of the physical and the compositional nature of the initially deposited a-Si film on the crystallization process during annealing. Accordingly, a-Si films grown by two different methods, e-beam evaporation and PECVD were used to form the a-Si/Al/glass structures and the AIC behavior of the respective stacks, as well as their final poly-Si properties are compared. The influence of processing parameters such as deposition temperature (for e-beam evaporation) and hydrogen dilution (for PECVD) is also studied.

Hybrid organic/inorganic perovskite for use as a light absorber in solar cell has attracted recently great attentions in photovoltaic research. In this study, a novel two-dimensional material, MoS₂, serves as a charge transport layer in lead iodide perovskite (CH₃NH₃PbI₃) solar cell. The MoS₂ thin layer was synthesized by CVD (chemical vapor deposition) and deposited on the perovskite layer by water transfer. The perovskite layer was also deposited on the flexible PET substrate using one-step spin coating and two-step spin coating. X-ray diffractometry, scanning electron microscopy, energy dispersive spectrometry, and UV-visible spectroscopy were used to characterize properties of the materials. The resulting solar cells were evaluated under a sun light simulator by determining current-voltage measurement, incident photon-to-electron conversion efficiency, intensity modulated photocurrent/photovoltage spectroscopy, and electrochemical impedance spectroscopy. The effects of material properties on the cell performance are addressed.

Hybrid organic-inorganic materials, particularly methylammonium lead halide perovskites, have drawn interest in the photovoltaic community, owing to the rapid rise in efficiency. However, the deployment of hybrid perovskite solar cells may be limited by their poor intrinsic stability under ambient conditions and the bioavailable lead iodide salt as a decomposition reaction product. Recent attempts to address stability and toxicity through group substitution have only addressed one part of the problem. These include tin-based perovskites, which are non-toxic but unstable, and some 2D perovskites, which are more stable but retain the lead cation. New semiconductor absorbers are needed that are efficient but overcome the toxicity and stability limitations. Recently, we have identified several classes of materials that could have the high defect tolerance and high minority-carrier lifetimes needed for efficient solar cells.^[1] One class consists of a partially ionized, heavy-metal, non-toxic, p-block cation (e.g. Bi³⁺) and a halide.
We report the synthesis, optoelectronic properties and stability of the hybrid ternary compound methylammonium bismuth iodide ((CH₃NH₃)₃Bi₂I₉, abbreviated as MBI). We synthesized MBI using a vapor-assisted solution process. Through Rietveld refinement of its powder diffraction pattern, we conclude that the structure of MBI has an orthorhombic unit cell comprising CH₃NH₃⁺ and Bi₂I₉^3- groups. Absorption and photoluminescence measurements show that MBI has an onset of absorption at 2.26 eV, making it potentially suitable as a top-tandem absorber. We report the stability of MBI in comparison with methylammonium lead halide perovskites under ambient conditions. X-ray diffraction measurements show that the lead halide perovskites degrade to a mixed phase with PbI₂ after one week, whereas MBI remained phase stable. The higher stability of MBI is confirmed through photoluminescence and absorption measurements. We determine the mechanisms for the improved stability of MBI through thermogravimetric analysis, density functional theory calculations and X-ray photoelectron spectroscopy. Through MBI, we demonstrate a non-toxic and stable analogue to methylammonium lead halide perovskites, from which we can design more stable hybrid organic-inorganic compounds.
[1] R. E. Brandt, et al., MRS Communications. DOI: 10.1557/mrc.2015.26

Cuprous oxide (Cu₂O) solar cells are appealing because the absorber is composed of Earth-abundant, non-toxic elements. Recent developments leading to large efficiency increases have involved vacuum processing, such as the use of vacuum-based atomic layer deposition (ALD) to synthesize the n-type window layer and form the p-n junction. However, the widespread deployment of solar cells is dependent upon reducing the capital expense and increasing the throughput of manufacturing, which in some cases may exclude vacuum-based techniques. In line with this, we have developed an atmospheric pressure chemical vapor deposition (AP-CVD) reactor, which is capable of growing oxide thin films up to two orders of magnitude faster than ALD and at low temperatures. Our AP-CVD reactor consists of a gas manifold that has separate parallel channels of gaseous metal precursor and oxidant, with inert nitrogen gas channels in between.^[1] Using AP-CVD to deposit Zn_1-xMg_xO onto Cu₂O, we have improved the efficiency of Cu₂O solar cells up to 2.2%, the highest at the time of publication for open-air fabricated p-n junctions.^[2] However, these efficiencies remain below those of Cu₂O solar cells with vacuum fabricated p-n junctions, which can reach 6%. Through X-ray photoelectron spectroscopy (XPS), we show that performance-reducing CuO readily forms on the surface of Cu₂O after exposure to ambient air, as CuO is the thermodynamically stable phase at ambient conditions.^[3] We highlight three techniques to avoid the undesired CuO surface-phase. First, the Cu₂O should be stored in an inert environment immediately after deposition. Second, forming gas should be introduced to the inert gas channels in our AP-CVD manifold, to create a reducing atmosphere. Third, the absorber should be scanned under the organometallic metal precursors from the AP-CVD gas manifold to reduce any remnant surface CuO.^[3] These techniques should limit the formation of higher oxidation states on the surface of atmospherically processed Cu₂O solar cells, which can allow the efficiencies to approach those of vacuum-processed solar cells.
[1] R. L. Z. Hoye, et al., ACS Appl. Mater. Interfaces, 7, 10684 (2015)
[2] R. L. Z. Hoye, et al., ACS Appl. Mater. Interfaces, 6, 22192 (2014); Y. Ievskaya, et al., Sol. Energy Mater. Sol. Cells, 135, 43 (2015)
[3] R. L. Z. Hoye, et al., APL. Mater., 3, 020901 (2015)

Organic-inorganic metal halide perovskite solar cells&’ (PSCs) outstanding characteristics have been studied in a variety of research group as the next generation thin-film solar cell. Perovskite is an excellent light harvester which has long diffusion length, broad absorption range, low exciton bonding energy, and ambipolar characteristic. Based on these properties, PSCs matches well with organic hole transport materials, such as N2,N2,N2prime;,N2prime;,N7,N7,N7prime;,N7prime;-octakis(4-methoxyphenyl)-9,9prime;-spirobi[9H-fluorene]-2,2prime;,7,7prime;-tetramine (Spiro-MeoTAD), poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDDOT:PSS), Poly[bis(4-phenyl)(2,4,6-trimethylphenyl)amine] (PTAA), and etc. Carrier transport, one of important factors affecting device performance and life time, can be improved by inserting extraction layer which lowers barrier between transporting layer and electrode. In this work, we used dipyrazino[2,3-f:2&’,3&’-h]quinoxaline-2,3,6,7,10,11-hexacarbonitrile (HAT-CN) as hole extraction layer (HEL) to improve hole migration between interlayer system. HAT-CN is widely used as a hole injection layer in organic light emitting diodes, because of its energy level and surface electrical property. We fabricated PSCs by varying the HAT-CN thickness, and analyzed current density-voltage characteristics depending on different light intensity. We calculated series and shunt resistance of each solar cell to see the improvement of carrier transporting property. From these analyses, we can understandthe effect of HAT-CN as a hole extraction layer on the photovoltaic performance of PSCs.

We study the electroabsorption (EA) spectrum at the fundamental absorption edge of CH₃NH₃PbI₃(Cl) perovskite between 300K and 190K. We extract an exciton binding energy of <10 meV at room temperature by fitting the spectra to a numerical simulation of a Wannier exciton in an electric field. These data suggest that at room temperature and in the presence of an electric field, primary optical transitions occurring at the fundamental absorption edge in CH₃NH₃PbI₃(Cl) solar cells can be described in terms of free carrier generation. In addition, we use the EA data to precisely determine the band gap of CH₃NH₃PbI₃(Cl) and observe an anomalous temperature dependence in which the band gap shifts from 1.647 eV at 300K to 1.616 eV at 190K. Finally, we suggest a novel and computationally simple method for extracting the exciton binding energy and band gap from EA spectra based on a modification of the typical one-electron theory of electroabsorption, demonstrating EA as a relatively simple method for characterizing highly accurate exciton binding energies in perovskite materials.

In the last 6 years organometal halide perovskites have not only proven to be a promising material for solar cells with highest efficiencies exceeding 20 % but they have also shown a huge potential for light-emitting applications. Exploiting the optical properties of specifically tailored perovskite nanocrystals could greatly enhance the efficiency and functionality of applications based on this material. We have investigated the quantum size effect in 2-dimensional lead bromide perovskite nanoplatelets. Their thickness and consequently their photoluminescent (PL) emission can be controlled by varying the ratio of organic cations, methylammonium and octylammonium. We find that an increasing amount of the capping ligand octylammonium leads to the formation of thinner structures and a PL blue-shift of more than 90 nm. Model calculations match well with the experimental values. Stacking of nanoplatelets leads to a further shift in the bandgap energies. Moreover, a comparison of the calculated values with experimental data suggest a large exciton binding energy up to several hundreds of meV in the strongly confined nanoplatelets.

Recent progress in chalcogenide solar cells has placed this technology to the forefront of next generation thin film photovoltaics. Within this family, kesterite absorbers based on earth abundant Cu₂ZnSnSe₄ thin films are promising low-cost alternatives to the established chalcopyrite Cu(In,Ga)Se₂ technologies. However, sample processing and film crystallization in the kesterite material is less well understood and far more complex, especially because of the easy formation of secondary phases such Cu, Zn and Sn selenides.
In this contribution we report on our progress in depositing efficient kesterite solar cells onto commercial ceramic substrates, in an attempt to open new application potentials (e.g. as building integrated elements such as solar tiles). The supply of defined amounts of sodium during or after the chalcogenide film growth is a prerequisite for high quality photovoltaic materials, and for sodium-free substrates an additional sodium supply is hence needed. The ceramic substrates have therefore been coated with a vitreous enamel containing Na₂O, that has been specifically developed for these applications prior to the solar cell deposition. This coating acts twofold: smoothing the surface of the rough ceramic substrate and as a source of sodium. We report here on how a variation of different amounts of Na₂O in the enamel affects surface morphology and device performance of the completed kesterite solar cells. Solar cell efficiencies of up to 4.6% were obtained in a first optimization.
ZnSe and SnSe secondary phases are known to form in general during our selenization process and are removed with a selective etching process. The samples grown on the ceramic substrates showed the additional formation of SnSe₂ secondary phases which are not removed in the standard etching process. Their formation is probably favored because of the different thermal conductivity and/or uneven surface morphology of the used substrates. However, new measurement conditions had to be established, as small amounts of SnSe₂ within or on top of the kesterite material are in general not easily detected with conventional Raman or X-ray diffraction techniques. We find a greatly enhanced sensitivity of the Raman detection for SnSe₂ phases at near resonance excitation with a laser wavelength of 785 nm (direct band gap of SnSe₂: 1.6eV). With this newly developed technique we are able to efficiently detect minor amounts of SnSe₂ on the absorber surface and show the clear correlation between optimal solar cell efficiency and the absence of these secondary SnSe₂ phases. The presence of SnSe₂ is critically related to a decrease of especially the open circuit voltage of the devices.
The finding of up to now undetected SnSe₂ phases might be an important finding for kesterite solar cells in general, as the influence of SnSe₂ secondary phases has not yet been intensively studied in the past and might lead the way to further device improvements for this material.

Reducing silicon thickness in photovoltaic device industry has always been demanded for higher profitability. Further benefits such as short collection lengths, improved open circuit voltages and low surface recombination velocities can also be achieved by silicon thickness reduction. The problem with thin films; however, is the poor light absorption. Therefore, advanced light trapping strategies are essential to attain thin cells with competitive efficiencies. Here, we fabricate solar cells with vertically aligned silicon nanowire arrays on ultrathin silicon (20 to 35 mu;m) wafers. Silicon wafers were thinned down to desired thicknesses using potassium hydroxide etching. Subsequent to this, vertically aligned silicon nanowire arrays were formed on the ultrathin silicon surface via metal assisted etching method. This retrieved the poor light absorption characteristics of the thin wafers. Nanowire decorated ultrathin single crystalline silicon wafers showed enhancement in optical absorption. The effect of silicon nanowires was found to be more significant in thin wafers as expected. Relative improvement in excess of 33% in the reflectivity was observed for 20 mu;m thick silicon wafers by the fabrication of 2 mu;m long silicon nanowires a top. Nanowire textured wafers were then doped through POCl₃ diffusion for the formation of p-n junctions. Evaporation of the front and rear contacts finalized the fabrication of the solar cells. Detailed photovoltaic characterization was conducted and a photovoltaic conversion efficiency of 9 % was obtained for 20 mu;m thick silicon wafers.

PbS is a semiconductor material with an FCC crystalline structure, a 0.41 eV Eg, an 18 nm Bohr radii and it is well known to show quantum confinement effects when in nanoscale dimensions. These properties have made it attractive for applications in optoelectronics, such as Thin Film Transistors (TFTs), IR detectors and more recently photovoltaic devices. However, a reliable methodology for the determination of their electronic structure still is a debate issue among researchers. In this work, PbS nanoparticles were synthesized by the one-pot approach, with different capping ligands (oleic acid, myristic acid, hexanoic acid and L-Cysteine) trying to improve their optical and electrical properties. Afterwards, the nanoparticles were exhaustively characterized by UV-Vis-NIR, RAMAN spectroscopy, cyclic voltammetry and TEM and its related techniques. UV-Vis-NIR was used to determine their optical band gap, whereas EELS was used to find out their electrical band gap and cyclic voltammetry was used to study their LUMO level value. Based on the electronic structure determined, the nanoparticles with the most compatible electronic structure with that of PEDOT:PSS were selected. Selected PbS nanoparticles were dispersed into PEDOT:PSS, which is a conductor polymer; once the nanoparticles were dispersed, thin films were fabricated by spin coating with different concentrations of the nanoparticles; subsequently, electrical response and photoconductivity tests were performed on these thin films. Finally, excitonic solar cells were fabricated with the thin films showing the best electrical performance.

Photovoltaics comprised of Cadmium Telluride (CdTe) represent a growing proportion of the solar cell market, yet the physical picture of charge transport at the meso-scale remains a topic of debate. The difference between actual and theoretical maximum efficiencies has been attributed to recombination at structural defects in the thin-film, among which include: crystal twinning, stacking faults, and grain boundaries. To characterize the contribution of these effects, it is necessary to image electrical performance of individual CdTe grains and grain boundaries. Several studies have used cross-sectional or surface techniques (SEM, TEM, AFM). In this study we use conductive tomography to image photocurrent in a solar cell. This study reveals that stacking faults serve as P-type pathways for charge collection, while grain boundaries operate as N-Type pathways. Instead of eliminating these structural defects, our work predicts that control of this morphology offers a pathway to reach higher device efficiency.

Cu₂ZnSn(S,Se)₄ (CZTSSe) has attracted great interest in recent years as an earth-abundant photovoltaic absorber. While efficiencies have improved steadily, the limiting performance parameter remains a deficit in the open-circuit voltage (Voc). This Voc deficit has been correlated with excessive band-tailing stemming from electrostatic potential fluctuations of the valence and conduction bands and/or band gap inhomogeneity in the material. A number of studies have identified widespread disorder on the Cu-Zn sublattice as a key parameter contributing to the band tailing and hence Voc loss. The covalent radii of Cu and Zn differ by only 5%, which may be why site exchange occurs so readily. Therefore, substituting either Cu or Zn with an element possessing a larger covalent radius (e.g. Ag in place of Cu) is expected to reduce the probability of site-exchange and hence improve the optoelectronic quality of the semiconductor. This contribution will present optoelectronic characterization of Ag₂ZnSnSe₄ (AZTSe) thin films grown by co-evaporation and annealing. We will compare the bulk defect structures of CZTSe, AZTSe, and CIGSe (via photoluminescence characterization at ~6K) and show that the replacement of Cu with Ag significantly suppresses the formation of charged sub-Eg states. This in turn suppresses the harmful band tailing associated with the Voc deficit. We will show that AZTSe is optoelectronically more similar to CIGSe than to CZTSe. Additionally, we will discuss important considerations for device preparation based on AZTSe thin films.

The success of lead-based, inorganic-organic hybrid perovskites in photovoltaics arises from their strong absorption near the band gap, long carrier lifetimes, and simple processing. Additionally, the ability to tune their band gap through halide substitution is attractive for tandem solar cells and light emission applications. The first step toward understanding any material&’s behavior in such an optoelectronic device, however, is knowing its complex refractive indices, n and k. Since optically smooth films of hybrid perovskites are challenging to produce, conventional macroscopic reflectance or ellipsometry studies to determine n and k have yet to be reported for any hybrid perovskite except CH₃NH₃PbI₃ [1].
To address this lack of knowledge, quantitative reflectance and transmittance measurements at the microscale have been performed on individual microcrystals of CH₃NH₃PbBr₃. By measuring crystals of thicknesses varying from hundreds of nanometers to micrometers, unique n and k values at each wavelength have been determined. The single crystals are formed by spincoating a film of precursor solution and then stamping it with polydimethylsiloxane (PDMS) during crystallization. Their surfaces are optically smooth, and exact thicknesses are determined using atomic force microscopy (AFM). In contrast to single crystals commonly prepared by slow precipitation from solution, these crystals are produced in a way similar to the processing often used in thin-film devices. Microscale reflectance and transmittance can be used to compare the optical properties of perovskites deposited from a variety of precursor solutions or to determine how the n and k values evolve as the band gap changes with halide substitution. In principle, this approach can be applied to measure the optical constants of any material that presents challenges in producing smooth films over large areas.
1. Loper, P. et al. J. Phys. Chem Lett. 2015, 6, 66-71.

Dye sensitized solar cells (DSSCs) have received great attention to use as a potential alternative for converting solar energy into electricity. The energy conversion efficiency of DSSCs is mainly dominated by the performance of mesoporous TiO₂ photoanode, which provides an enormous surface area available for dye chemisorption. To further improve the performance of TiO₂ electrode, we develop (1) a scalable synthesis of Nb-doped TiO₂ nanoparticles (NPs) used as the DSSC photoanode, and (2) a facile synthesis of ultrasmall Nb₂O₅ NPs used as an effective underlayer between TiO₂ photoandoe and transparent conductive oxide (TCO) substrate. First, the Nb-doped TiO₂ NPs was made by simply mixing TiO₂ paste with functionalized Nb₂O₅ sol gel. This method is suitable for mass production. By doping 2mol% Nb into TiO₂, the positive shift of conduction band minimum (CBM) enhances the electron injection and the improved electron conductivity facilitates the electron transport, leading to 18.9% improvement of photoconversion efficiency as compared with the standard DSSCs. The long-term performance of DSSCs with Nb-doped TiO₂ photoanode was also reported in this work. Second, a new synthesis route of ultrasmall Nb₂O₅ NPs (2-5nm) was developed by adding NbCl₅ into ethanol with acetic acid as stabilizer via UV ozone treatment. Nb₂O₅ NPs was then coated on TCO as underlayer by electrophoresis, which shows an effective suppression for the electron recombination at TiO₂-TCO interface, resulting in the increase of J_sc and V_oc.

The immense success of group IV and III-V semiconductors has resulted in disruptive new photovoltaic (PV) cell technologies emerging extremely infrequently. For this reason, the recent progress in organo-metallic hybrid perovskite solar cells can be viewed as a highly significant historic event. Not only has the peak reported power conversion efficiency (PCE) increased at an unprecedented rate, to a value now in excess of 20%,¹ but processing techniques employed suggest low-cost, large-area and flexible commercial devices are possible.
Despite the staggering progress made in the PCE of these devices over the last few years, many aspects of device operation remain poorly understood. In particular, debate is intense on the nature of the various instabilities synonymous with these devices. Similarly, despite high reported carrier mobilities,² easily-accessible conduction and valence band energies and previous reports employing other organo-metallic hybrid perovkistes,³ field-effect transistors (FETs) based on methylammonium lead iodide (MAPbI₃) remain notably absent from the literature. Recently it has been suggested that ionic migration could be responsible for device instability, or perhaps even the high PCE observed in these solar cells.⁴
Using various electronic measurements, we here present a body of experimental evidence consistent with the existence of a mobile ionic species within the MAPbI₃ perovskite. Temperature-dependent transistor measurements reveal operating FET devices only below approximately 210K. This is attributed to ionic screening of the (otherwise charge-neutral) semiconductor-dielectric interface. Such ionic motion is anticipated to be temperature-activated, and one should expect any ionic screening to be highly positively-correlated with temperature. Temperature-dependent pulsed gate experiments, reveal a time-dependent source-drain current behavior consistent with this interpretation. Capacitors exhibit a decreasing low-frequency capacitance with temperature and a temperature-independent capacitance at higher frequencies, again as expected for a slow screening process such as mass-transport.
Finally, MAPbI₃ PV cells were found to possess a power conversion efficiency which decreases significantly below 210K. This can once again be interpreted in the context of ionic-motion. Combined, these set of measurements provide an interesting and consistent description of the internal processes at play within the MAPbI₃ perovskite structure.
[1] National Renewable Energy Laboratory, Best Research-Cell Efficiencies; www.nrel.gov/ncpv/images/efficiency_chart.jpg, 2015.
[2] C. C. Stoumpos, C. D. Malliakas, M. G. Kanatzidis, Inorganic Chemistry 2013, 52, 9019.
[3] C. R. Kagan, D. B. Mitzi, C. D. Dimitrakopoulos, Science 1999, 286, 945.
[4] Y. Zhang, M. Liu, G. E. Eperon, T. C. Leijtens, D. McMeekin, M. Saliba, W. Zhang, M. de Bastiani, A. Petrozza, L. M. Herz, M. B. Johnston, H. Lin, H. J. Snaith, Materials Horizons 2015, 2, 315.

Over the last 24 months, we have witnessed a spike in interest in the study of perovskite solar cells, due to a surprising increase in terms of efficiency, going from 3.8% to a certified power conversion above 20%. Nevertheless, the understanding of the optoelectronic properties of such nanostructured materials is still an open problem and issues related to their stability and degradation pathways represent the currently hot topic in this research area.
Organic-inorganic solar cells present a complex composition as well as a composite structure that are strongly related to device fabrication. In this work we investigate four processing methods of the mixed halide perovskite, carried out in different atmospheres, using single step and double step deposition, in order to compare interface quality, morphology, chemical composition and efficiency of the resulting cells. A FTO glass layer was coated first by a compact TiO₂ layer and then by a nanoporous TiO₂ layer. The TiO₂ scaffold was infiltrated and capped by methyl-ammonium lead iodide. Spiro-MeOTAD was spin coated on the perovskite layer; Au contacts were deposited on top. The use of several complementary characterisation techniques allowed us to have a detailed and complete overview of the aforementioned nanostructured systems, but a particular emphasis was placed on the interface elemental migration. The devices were investigated using time of flight secondary ion mass spectrometry (ToF-SIMS), X-rays photoelectron spectroscopy (XPS), and Scanning Transmission Electron Miscroscopy (STEM) used in conjuction with EDX analysis.
Double step deposition of the perovskite component led to higher efficiency and the conversion enviroment was found to strongly affect cell performance and infiltration of the mixed halide perovskite in the TiO₂ mesoporous layer. In particular perovskite synthesis in controlled atmosphere results in enhanced photovoltaic properties. The analysis of elemental distribution in the samples processed in air shows interesting results connected to the iodine migration from the perovskite capping layer to the organic hole transport layer, an elemental diffusion observed for the first time with nanometer - scale resolution. This phenomenon can be connected to the residual moisture within the cell and it could be one of the main responsible of the long term degradation pathway of the cell. Conversely, single step deposition presents a poor perovskite penetration in the TiO₂ scaffold and lower performance.
In conclusion, a multi-approach characterisation of perosvkite based solar cells allowed us to assess a correlation between different deposition methodologies, the performance and the key role plays by the nanostructured architecture and the local elemental composition of the devices. In particular the evidence of the migration of the inorganic component of the mixed halide perovskite represents an important result in the study of the long term stability of this emerging technology.

Efficient solar cells have been obtained using thin films of solution-processed CH₃NH₃PbI₃ prepared from PbCl₂ and CH₃NH₃I, referred to as CH₃NH₃PbI_3-xCl_x. However, there remains limited knowledge about the relationship between preparation route and optoelectronic properties. We use complementary time-resolved microwave conductivity (TRMC) and photoluminescence (PL) to investigate the charge carrier dynamics in thin planar films of CH₃NH₃PbI_3-xCl_x, CH₃NH₃PbI₃ and their meso-structured analogues. High mobilities close to 30 cm²/Vs and microsecond-long lifetimes are found in thin films of CH₃NH₃PbI_3-xCl_x, compared to lifetimes of only a few hundred nanoseconds in CH₃NH₃PbI₃ and meso-structured perovskites. We propose, for the first time, a kinetic model that describes TRMC and PL experiments with one set of global kinetic parameters characteristic for each sample. We find that the trap density is less than 5 ×10¹⁴ cm^-3 in CH₃NH₃PbI_3-xCl_x, 6 × 10¹⁶ cm^-3 in the CH₃NH₃PbI₃ thin film and ca 10¹⁵ cm^-3 in both meso-structured perovskites. Furthermore, our results imply that for all samples, band-to-band recombination is enhanced by the presence of dark carriers resulting from unintentional doping of the perovskite films. Our general approach to determine concentrations of trap states and dark carriers is highly relevant to other semiconductor materials as well.

Etching of semiconductors and formation of defined structures and porous surfaces can be carried out by different methods. Chemical etching, photo-etching, stain-etching, metal-assisted etching and anodic- etching are different names for the application of the induced etching for beneficial applications. Porous silicon layers represent a hot topic of new technology in solar energy conversion and optoelectronics. Porous silicon layers, PSL, on silicon lead to large area solar cells and high solar conversion efficiency. The application of these layers in optoelectronics and their effective optical properties has found great interest. In the last few years PSL of definite pore structures have been prepared by metal-assisted etching of p-Si in aqueous hydrofluoric acid solutions containing different oxidizing agents. Potassium dichromate at definite concentration has shown promising effects. The electrical and electrochemical properties of the formed porous layers have been examined. The effect of etching time, oxidizing agent concentration and HF concentration on the main characteristics of the porous structure was investigated and discussed. In this respect conventional electrochemical techniques and electrochemical impedance spectroscopy, EIS, have been used. The experimental data were fitted to theoretical data according to a proposed electronic equivalent circuit model. The morphology of the formed layers and surface contaminations were investigated by the scanning electron microscopy, SEM, and energy dispersive x-ray, EDX, techniques. The results have shown that PSL, with nano and micro pores were formed on p-Si when etched in HF-K₂Cr₂O₇ aqueous solutions. At 22 mol L^-1 HF and relatively high concentration of K₂Cr₂O₇ [> 0.05 mol L^-1] a passive K₂SiF₆ salt was formed inside the pores. The thickness of the passive layer was affected by the concentrations of both HF and K₂Cr₂O₇. It reduces the effectiveness of the PSL in both the solar conversion process and also its electrical and optical characteristics.

Post-deposition soft annealing (PDA) process, either at low (<200 0C) or high (>200 0C) temperatures, on finished devices is known to improve the characteristics of chalcogenide thin film solar cells. However, there is scarce information on the origin of this behavior, and deeper understanding of PDA is needed for further optimization and improvement of the device efficiency. In the case of kesterite-based solar cells there are no clear evidences on the impact of PDA on the presence of point defects in the surface region of the layers that are relevant for device efficiency. In this framework, this work details an investigation of the impact of PDA treatments on the structural, vibrational, compositional and optoelectronic properties of Cu₂ZnSnSe₄ (CZTSe) solar cells. CZTSe absorber layers were synthesized by selenization of sputtered metallic precursor stacks, and made into solar cells with a CdS/ZnO/ITO window layer. The complete devices were then subjected to a 5 min PDA process at temperatures ranging from 50 to 350 ^oC. This process produced solar cells with efficiencies of up to 10.1%, and open circuit voltages (V_OC) of 470 mV, which is the record V_OC value reported for pure CZTSe-based devices. All solar cells have been characterized by illuminated J-V, photoluminescence, TEM/EELS, and Raman measurements. A deeper analysis of the surface region at different depths of the device cross section was achieved using multiwavelength excitation Raman measurements combining blue (445 nm), green (532 nm) and near-infrared (785 nm) excitation lines.
The results show that with the increase of the PDA temperature in the range from 100 to 300 ^oC, there is general enhancement in the device efficiency, with two local maximums obtained at 175 and 250 ^oC. The V_OC and J_SC are maximized at the higher and lower temperature, respectively. This correlates with the presence of systematic changes in the intensity of the E and B Raman vibrational modes located around the 170, 210 and 250 cm^-1 spectral regions. Changes in the relative intensity of these peaks are explained by the variation in concentration of different point defects. It is observed that with increasing PDA temperature, there is creation of optoelectronically beneficial V_Cu and Zn_Cu point defects in the surface region. This is explained by the migration of Cu cations from the CZTSe surface to the near-surface region and vice-versa for the Zn cations, and confirmed by near-surface Raman and TEM/EELS measurements of representative specimens. Additional variations in the intensity of the Raman peaks will be systematically analyzed in relation to the occurrence of other kinds of defects involving Zn_Sn, Cu_Zn, Sn_Zn, and possible changes in Na concentration in the CZTSe surface. Furthermore, variations in the band gap of the CdS layer with the temperature will be presented and correlated with the improvement in the band alignment between the buffer and the absorber layer.

We report a cadmium selenide tetrapod (CdSe TP) layer as an electron extraction layer for improving the efficiency of planar hetero-junction perovskite solar cells based on lead iodide (PbI₂) and lead chloride (PbCl₂) mixed halides. Insertion of the CdSe TP layer between the titanium oxide (TiO₂) and perovskite film facilitates electron transfer at the TiO₂/perovskite interface, as evidenced by the significant quenching of the steady state photoluminescence of the perovskite film. Absorption in the visible spectral region increased when the CdSe TP layer was introduced. Furthermore, we observed a conductivity enhancement of the perovskite film by the conductive atomic force microscopy (c-AFM) which shows enhanced grain size and current path of CdSe TP induced perovskite film. These effects contribute to an approximately 10% augmentation of the photocurrent. As a result, an efficiency of 13.5% is achieved for perovskite solar cells that incorporate the CdSe TP layer, which is 10% higher than that of the device without the CdSe TP layer.

As the only known direct bandgap semiconductor that is lattice-matched to InP with a bandgap greater than 1.5 eV, (In_0.52Al_0.48As)_z(AlAs_0.46Sb_0.54)_(z-1) is a material with promise for photovoltaic applications. When z = 0.5, its 1.73 eV direct bandgap makes it useful as a subcell in highly efficient InP-based multijunction solar cells with projected efficiency greater than 50% at 500 suns. In addition, it&’s an excellent candidate for the host material for the quantum well- or quantum dot-based intermediate-band solar cell. A few groups have grown this material via MBE with some success. However, the development of this novel quaternary material remains immature and other groups' attempted growths by MOCVD have failed to incorporate the 10% or greater Sb content required to attain a bandgap significantly higher than lattice-matched InAlAs. To our knowledge, this is the first report of this material grown with both high In and Sb content via MOCVD.
Growth was done on a Veeco MOVCD D125 LDM rotating disk reactor. AlAs_0.46Sb_0.54 was used as a growth template, and it became apparent that the material composition of the quaternary was highly sensitive to growth parameters. Specifically, a small window for both temperature and V/III ratio was observed to allow for adequate In and Sb incorporation into the lattice. While AlAsSb requires a V/III ratio between 1 and 2, it was found that In content in the lattice suffered at such low ratios, which resulted in indium-rich deposits on the surface. Sb incorporation efficiency decreased with higher V/III, however, this compromise was necessary to obtain the correct composition. The material studied here was grown at 520 °C with a V/III ratio of 5.24.
Secondary ion mass spectrometry was used to verify that the correct crystal composition had been achieved. Hall results indicated a high n concentration of 4.8×10¹⁶ cm^-3 in unintentionally-doped growths, likely due to deep level donors. Deep-levels and carrier compensation will be investigated using deep-level transient spectroscopy and temperature-dependent Hall experiments. Schottky diodes were fabricated from the uid n-type material using contacts from evaporated Pt/Au and current-voltage and capacitance-voltage characteristics were extracted from these devices. In addition, preliminary spectral response using a 2.41 eV laser will be performed. Initial DLTS results show a dominant electron trap 0.56 eV below the conduction band with a concentration of 8.1×10¹⁵ cm^-3. This is in contrast to lattice-matched n-type InAlAs Schottky diodes, which exhibited a dominant electron trap 0.60 eV below the conduction band with a significantly lower concentration of 1.4×10¹⁴ cm^-3. The effect of the detected impurities on the device metrics (from IV, CV, SR) will be discussed and a comparison of the novel quaternary to the more well-understood InAlAs material presented.

It is well accepted that the open circuit voltage of solar cells is limited by the recombination loss which exhibits first and second order (sometime third) recombination reactions. Here we use a device model (drift, diffusion, Poisson) and incorporate self-consistently the contacts. We find that in the full device model one needs to explicitly include only first order reaction (as trap assisted recombination) and that the signature of higher order recombination arises from the full model. The source of this surprising result is the fact that the upper limit for the open circuit voltage is the energy gap between the anode and cathode. As the position of the electrodes determines the position of the Fermi level at the semiconductor interface it also sets a limit on the maximum allowed charge density. The model shows that space charge at the vicinity of the electrode interface pins the electrode below the energy levels of the bulk of the device thus setting a limit lower than would typically be expected. The validity of the model is verified by reproducing results published by us [1] and others [2].
Having isolated the limiting mechanism we present a new design of the electrode contact region that mitigates the above pinning effect and allows to gain, for state of the art solar cells, 0.2 to 0.3V in open circuit voltage along with some improvement in the fill factor.
[1] L. Tzabari, et. al., J. Phys. Chem. C118, 27681 (2014).
[2] K. Vandewal et al., Adv. Mater.26, 3839 (2014).

Maximization of light harvesting and carrier collection is vital for high efficiency perovskite solar cells. The mesoporous scaffold made up of titanium oxide nanostructures plays a prominent role in enhancing light absorption by scattering and also in providing a pathway for the photogenerated electrons towards the electron collection layer. Hence, controllable engineering of the size and shape of these nanostructures is a key to boost the efficiency of these solar cells. Here a study is presented on mesoporous perovskite solar cells by replacing the traditional nanoparticles by two dimensional square shaped nanoplates as constituent material for the porous scaffold. The nanoplates are as large as 100nm on the sides and less than 10 nm in thickness. The large exposed facets dominantly have a (100) orientation which offers different chemical and electron transfer properties from the commonly used more stable facet. Solar cells have been made by depositing uniform CH₃NH₃PbI₃ perovskite films on the porous layer by two step spin coating technique. Comparison between devices made from nanoparticles and nanoplates under exactly similar conditions was carried out to reveal the difference of charge transfer and collection, and the impact on the device performance.

Recently, Dye sensitized solar cells (DSSCs) of metal free sensitizer have become an academically and industrially important issue.^1,2 Molecularly engineered porphyrin dye of a donor-π-bridge-acceptor has reported with 13% efficiency.³ Ruthenium complexes dyes are expensive, rare metal and not environmental friendly, which may possibly limit the potential inclusive applications of ruthenium dyes.⁴ We report the synthesis and characterization of new series of five dyes based on D35 donor moiety. WS1, WS2, WS3, WS4 and WS5 based on substituted Benzo[1,2,5]thiadiazole, 2,5-dithienylpyrrole (DTP) and dithienopyrrole as a π conjugating linker and bulky moiety as donor. The introducing of bis(2,4-dibutyloxy) benzene substituted on the benzothidizole and dithienopyrrole is to increase the steric hindrance on the dyes. Thus, the electron recombination between redox electrolyte Co(II/III) and TiO2 surface is reduced. Consequently, WS1 device deliver power conversion efficiency reaching 5.9 %, 5.3% and 5.9 % at AM 1.5 simulated sunlight with iodide/triiodide, Co(II/III) and hole contactor material on solid state respectively. While WS2 deliver power conversion efficiency is reaching 6.2% at AM 1.5 simulated sunlight on solid state device.
Refrance
1-Jiabao Yang, Paramaguru Ganesan, Joeuml;l Teuscher, Thomas Moehl, Yong Joo Kim, Chenyi Yi, Pascal Comte, Kai Pei, Thomas W. Holcombe, Mohammad K. Nazeeruddin, Jianli Hua, Shaik M. Zakeeruddin, He Tian, and Michael Grätzel. J. Am. Chem. Soc., Just Accepted Manuscript bull; DOI: 10.1021/ja500280r bull; Publication Date (Web): 24 Mar 2014.
2- Yongzhen Wu, Magdalena Marszalek, Shaik M. Zakeeruddin, Qiong Zhang, He Tian, Michael Grätzel. Energy Environ. Sci., 2012, 5, 8261.
3- Simon Mathew, Aswani Yella1, Peng Gao1, Robin Humphry-shy;Baker1, Basile F.E. Curchod, Negar Ashari-shy;Astani, Ivano Tavernelli, Ursula Rothlisberger, Md. Khaja Nazeeruddin, Michael Grätzel. Nature Chemistry 6, 242-247(2014).
Yongzhen Wu, Magdalena Marszalek, Shaik M. Zakeeruddin, Qiong Zhang, He Tian,aMichael Greuro;atzel and Weihong Zhu. Energy Environ. Sci., 2012, 5, 8261

Solid solutions of Sb₂S_xSe_3-x offer a unique opportunity to design solar cell absorber materials with optical band gap within the extremes of 1.88 eV and 1.1 eV for Sb2S3 and Sb2Se3, respectively. When the value of x is smaller than 1.5, the open circuit voltage V_oc of the solar cell is low (of < 0.4 V) but the short circuit current density J_sc is above of 20 mA/cm². However, if x is larger than 2, the V_oc is high (of 0.6 V) but the J_sc is typically below than 10 mA/cm². In this work we present the fabrication of thin film solar cells made of Sb₂S_2.2Se_0.8 solid solution absorber deposited by chemical bath and thermal evaporation. The Sb₂S_2.2Se_0.8 thin films are made by fully utilizing the products of chemical deposition - thin films formed on SnO₂:F/CdS thin films and the precipitate recovered from the bath as a source for thermal evaporation. The Sb₂S_2.2Se_0.8 thin films have an optical band gap of 1.4-1.5 eV, an absorption coefficient of > 10⁴ cm^-1 in the visible region, and a photoconductivity of 10^-6-10^-5 #8486;^-1 cm^-1. Under these conditions, the solar cells exhibit V_oc of 0.5-0.6 V, and J_sc of 10-15 mA/cm². We find that the fill factor of the cells can be varied by pretreatment of the CdS thin films as well as by changing the chemical deposition to obtain cubic or hexagonal crystalline structure in the CdS film. In the same way, the Sb₂S_2.2Se_0.8/C interface can be changed by chemically modifying the carbon-colloidal paint and by optimizing the temperature of curing (240-320 oC) of the solar cells. These thin film solar cells are among the simplest to prepare and they are stable under continuous sunlight exposure.

What is a large simple cube (CUB)? It is a cubic unit cell containing 8 normal-size simple cubes and 8 times the number of atoms as in a normal rock salt or zinc blende cube. The cube edge is double, and the number of atoms runs into 64 per cell. There are 32 formula units of either SnS or SnSe instead of 4 units per cell in the more common orthorhombic or rock salt structures of these polymorphs. The large simple cubic structure for SnS nanocrystals was reported in February 2015. Soon after, SnS thin films originally assigned to zinc blende structure was re-assigned SnS-CUB as its correct crystal structure. Deposition of thin films of SnS-CUB takes place under particular conditions of chemical deposition, known since 2007. A thin film of SnSe-CUB can be deposited if SnS-CUB layer acts as nucleating-layer; or else it would condense into a thin film of orthorhombic structure. Thin films of SnS-CUB and SnSe-CUB are distinguished by their large simple cube lattice constants of 11.56 Å and 11.98 Å, respectively, and very compact nature of the thin films. They have much larger optical band gaps of 1.74 and 1.35 eV for SnS and SnSe, respectively, compared with 1.1-1.3 eV and 0.94-1 eV, respectively for their orthorhombic (ORT) polymorphs. These thin films are relatively more resistive, 10^-6 Omega;^-1 cm^-1 for SnS-CUB and 10^-5 Omega;^-1 cm^-1 compared with their ORT counterparts. Light-generated current densities as solar cell absorbers for these materials are: 25 and 35 mA/cm². In solar cells of SnS-CUB/CdS/ZnO/ZnO:Al, the short circuit current density is 6.3 ma/cm², conversion efficiency, 1.3% and open circuit voltage 0.470 V as of now. These solar cells are stable under concentrated sunlight of 16 suns. Other than being promising solar cell absorber materials, the large simple cubic SnS and SnSe thin films are interesting specimens for learning basic concepts in crystallography. Based on the more than 50 peaks in the X-ray powder diffraction patterns in the 2theta; interval 5-90^o, the selection rules for Miller indices for a simple cubic structure offers a good exercise.

Diamond nanoparticles, often denoted as nanodiamonds (NDs), possess many unique qualities that are highly advantageous for energy conversion (PV) applications in comparison to other materials (Si, TiO2, etc.). NDs are available in large quantities, are completely non-toxic, can be stored indefinitely or burned into CO2 without any specific disposal procedures. Except for specific surface reactions, structural and electronic properties of NDs are very stable under various conditions [J. Nanopart. Res. 15 (2013) 1568]. Well established colloidal chemistry of NDs enables mixing with polymers without segregation, which is one of the problems in development of organic PV cells [Nanoscale Res. Lett. 6 (2011) 238]. Although diamond is a wide-band gap semiconductor (5.5 eV), in nanostructured form it exhibits light absorption as well as photoluminescence in wide visible spectral range due to sp2 carbon reconstructions and shells on the surface of nanocrystals or at structural defects such as dislocations [J. Appl. Phys. 116 (2014) 223103]. In addition, preliminary research on bulk diamond showed that merging diamond with organic dyes provides an efficient interface for exciton dissociation and electron transfer [Nanoscale Res. Lett. 6 (2011) 238]. In spite of the obvious potential the use of NDs for PV remains largely unexplored.
In this work, we present macroscopic and microscopic photovoltage characteristics of diamond nanoparticles for the first time. We used purified detonation nanodiamonds with residual amorphous carbon shell. At first, we fabricated simple organic photodiodes by spin-coating of P3HT+NDs mixture (50 wt%) onto ITO coated glass and using Al top contacts. The current-voltage characteristics recorded in dark and under standard AM 1.5 illuminations showed clear benefits of NDs: reduced dark current and enhanced photocurrent by about order of magnitude, higher built-in potential, and photovoltage response. To support the macroscopic device data we have studied photovoltage on pristine NDs that were dispersed on high work function metal substrate (Pt). By using novel methodology in Kelvin probe force microscopy (KPFM) [Langmuir 29 (2013) 1634] we show photovoltage generated by a green laser (532 nm) on individual NDs and their nanoscale aggregates. Although the measured photovoltage is small (about 15 mV), it clearly shows that even pristine NDs are able to work as miniature energy conversion devices.This work has been supported by the project 15-01809S (GACR).

Solar cells based on a Cu₂ZnSn(S,Se)₄ (which adopts the kesterite crystal structure) absorber represent an attractive opportunity for photovoltaics technology because of the close match in properties with Cu(In,Ga)(S,Se)₂ (CIGS), with however the use of lower cost and more abundant Zn and Sn rather than In and Ga in the absorber. While the record power conversion efficiency in the kesterite-based devices has grown steadily and has increased from 6.8% to 12.6% since 2008, a fundamental issue still remains with the devices - that of a suppressed open circuit voltage (V_oc) (so-called “voltage deficit” problem). In this talk, we will examine recent studies addressing prospective loss mechanisms, including the presence of high concentrations of anti-site defects leading to band tailing. As an understanding begins to form regarding the performance limits in the current generation of CZTS devices, it is hoped that mechanisms and new device designs can be devised to overcome the V_oc deficit issue, thereby providing a clearer pathway for CZTS as a competitive PV technology.

Until today, it is still unclear where non-radiative recombination takes place within polycrystalline photovoltaic (PV) devices, such as CdTe and CIGS. Despite the remarkable work on characterizing the electrical properties of grains and grain boundaries in both materials, there is still debate weather the boundaries act or not as centers for non-radiative recombination, which ultimately constrains the Voc of these PV technologies. Here we apply scanning photocurrent microscopy and illuminated Kelvin probe force microscopy to resolve the external quantum efficiency (EQE) and the open-circuit voltage (Voc) in both types of device, with nanoscale spatial resolution [1]. For that, we use NSOM probes a local source of excitation, allowing for spatially and spectrally resolved measurements of the local optoelectronic response of the devices, while mimicking the power density operation conditions of real devices [2]. Combined, these new tools provide a full picture of the local response of PV devices, including an indirect measurement of carrier local collection and recombination properties within the material [1]. We apply our new metrology to thin-film compound PV and find that the EQE and Voc in CIGS devices varies locally by more than 50% and 200 mV between grains, respectively. We correlate these variations with the different grain orientations present in the samples, and find a direct correlation between the material and the device local performance. For CdTe, we find that at short wavelengths (<600 nm), when light is absorbed near the exposed surface, the EQE is relatively small because of surface recombination. However, close to the material bandgap (860 nm), the EQE enhancement at the grain boundaries is ×1.5 [3]. These metrologies pave the way to identifying and quantifying where recombination takes place, and enable new insights into the loss mechanisms that hinder solar cells
[1] E. M. Tennyson et al., in review
[2] M. S. Leite et al., ACS Nano11, 11883 (2014)
[3] M. S. Leite et al., IEEE J. Photovoltaics 4, 311 (2014)

Minority carrier lifetime (tau;) is a key figure of merit for any photovoltaic (PV) absorber, and much ongoing research in PV can be fairly described as attempts to improve the lifetime. Reliable lifetime measurements are essential to enable rapid and rational engineering of PV technologies. For legacy PV materials such as Si and GaAs, measurements of tau; are well-established. The bulk lifetimes tend to be fairly long (over 1 ms), and passivation techniques to reduce the surface recombination are widely known. The same cannot be said for thin film PV materials. The bulk tau; is typically on the order of 1 - 10 ns, even for high performing materials such as CdTe [1]. Surface passivation is not well understood, and it is difficult to experimentally distinguish the time scales for minority-carrier recombination in the quasi-neutral, space charge, and interface regions.
We describe advances in optical-pump, THz-probe transient free carrier absorption (THz-FCA) measurements of minority carrier lifetime in an early-stage PV material. We demonstrate methods to deal with important factors including the difference between inter- and intra-granular diffusivity, the effects of sample heating, a spatially non-uniform generation profile, and the case of high injection levels. With an appropriate model of diffusive charge dynamics and a global fitting routine we can extract both the bulk lifetime (tau;) and the surface recombination velocity (S) with good confidence. We then study how these parameters are affected by material processing for the case of SnS, an interesting early-stage PV material. We show that the low record efficiency for SnS solar cells (4.36%) can be explained by a bulk t below 100 ps [2].
We emphasize the need for accurate minority carrier lifetime measurements for PV materials development from an early stage. We hope that our work with THz-FCA will help others to make better use of this and related techniques. To this end, we will freely distribute our data analysis software for use by the community at large.
1. T. A. Gessert, et al., 37^th IEEE Photovoltaic Specialist Conference (2011).
2. P. Sinsermsuksakul, et al., Adv. Energy Mater. (2014). DOI: 10.1002/aenm.201400496

Thin-film photovoltaics utilizing Cu(In,Ga)Se₂ (CIGSe) have exceeded efficiencies of nearly 22% in laboratory-grade devices and are currently being commercially produced in large-area modules at the gigawatt scale. Alternative earth-abundant absorbers like Cu₂ZnSn(S,Se)₄ (CZTSSe) are also showing rapid improvements in efficiency, but their developments are still in relative stages of infancy compared to CIGSe and CdTe devices. Common to all of the record devices is a buffer layer comprised of either cadmium sulfide (CdS) or zinc oxysulfide (Zn(O,S,OH)) that are deposited via a chemical bath and act as the n-type heterojunction partner with the absorber. However, optimization of the buffer for each absorber material is required. Therefore understanding and controlling the properties of the buffer layer is an important step in maximizing the overall solar cell performance.
Here we use calculations based on hybrid density functionals to assess the role of defects incorporated into the buffer layers during CBD growth. Specifically, we characterize the behavior of carbon, nitrogen, oxygen, and hydrogen impurities that are abundant in the bath and focus on their contributions to the electrical and optical properties of CdS and ZnS-derived buffers as a function of chemical conditions. We identify that O, C and N defects and their complexes with H can lead to electrical compensation in n-type buffers that are detrimental to device performance. However, the extent of these effects is highly sensitive to the growth conditions and is expected to be more severe for the ZnS-derived oxysulfide buffers as compared to CdS. We identify process conditions that will minimize their incorporation and discuss signatures of these defects that can help facilitate their detection for diagnostics.
This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and funded by the Department of Energy office of Energy Efficiency & Renewable Energy (EERE) through the SunShot Bridging Research Interactions through collaborative Development Grants in Energy (BRIDGE) program.

Polycrystalline Cu₂ZnSn(S,Se)₄ (CZTSSe) compounds have received wide research interest due to their potential as inexpensive absorber materials composed of earth-abundant elements. Photovoltaic devices fabricated on CZTS,Se have reached conversion efficiencies of 12.6 %. However, higher density of I-II exchange sites can strongly limit the maximum efficiency that can be achieved by this material. A potential earth-abundant substitute for CZTSSe is Ag₂ZnSnSe₄ (AZTSe) where the large covalent radius of Ag compared to Zn can limit the density of exchange sites and potentially lead to higher open-circuit-voltages. One of the key parameters to optimize the photovoltaic performance of polycrystalline solar cells is to control the concentration of recombination sites at the grain boundaries as well as the bulk of the film. To determine the composition and charge at the grain boundaries, this work has employed Auger nanoprobe microscopy (NanoAuger) with 8nm lateral resolution combined with high resolution ambient Kelvin Probe Force Microscopy (KPFM) with dual-lock-in amplifier setup. In order to characterize the grain boundaries as a function of depth within the films, grazing angle cryo-Focused Ion Beam (at -180 °C to -190 °C) is employed. Solution-deposited CZTS,Se grain boundaries showed negative charge (upward band bending) that is associated with the presence of SnOx phase formed during the film&’s air anneal. In contrast, thermally-evaporated Cu(In,Ga)Se₂ (CIGSe) had positively-charged grain boundaries (downward band bending) despite the absence of a distinct compositional difference between the grains and grain boundaries. These results are compared with NanoAuger and KPFM measurements on grazing cross-sections of thermally-evaporated AZTSe. Various grain boundary passivation methods were explored for AZTSe.

Hybrid perovskites are at the center of a flurry of research activity because of their excellent optical and electronic properties and low-temperature, solution- or vapor-phase processing. Although they have rapidly demonstrated their potential in efficient solar cells, many questions about their structural, electronic, and optical properties and how the devices operate still remain unanswered.
This talk will detail how we can apply tools developed largely by the nanoscience community to help us understand hybrid perovskite materials and devices. We will begin with a discussion of both solution and vapor-phase synthetic methods used to make a variety of single-crystalline morphologies such as nanowires (grown through a vapor-liquid-solid mechanism), plates, belts and microcrystals (hundred of microns across and less than a micron thick). Next, we will describe how we apply nanoscale characterization techniques to determine how morphology and processing affect the material&’s properties. Using a home-built laser system we have made maps of reflectance, absorptance, photocurrent, photoluminescence, EQE, IQE and refractive index with sub-micron spatial resolution. Combining this detailed information offers insight into local inhomogeneity within the material. Further we have also developed a platform that electrically contacts individual single-crystalline perovskite plates from the back while leaving the top surface exposed for characterization, allowing the properties listed above to be mapped out on functioning microscale solar cells. We have already demonstrated Voc~0.8V in our all back contact geometry without any special interfacial layers (only metal contacts). Finally, we will show how applying nano-SIMS to these single-crystalline devices can detail the elemental distribution in mixed halide alloy solar cells with sub-micron spatial resolution. Our back contact platform allows us to map changes in the elemental distribution, PL, or photocurrent during operation and after various atmospheric/illumination/heating cycles without destroying the device. The knowledge gained from such experiments on our model devices can then be used to optimize the design of conventional thin-film hybrid perovskite solar cells.

Since the recent emergence of perovskite solar cells and their demonstration as highly promising photovoltaics, much work has been done to tune and improve the device-architecture in the hopes of achieving even higher power conversion efficiencies. One main effort in changing the architecture of these PIN-type solar cells has been to replace the p-type selective contact, most commonly the organic hole-transport material (HTM) Spiro-OMeTAD, with an alternative material. Though there are several desirable reasons to replace Spiro-OMeTAD, the high manufacturing cost and poor device stability being chief among them, one that is often touted is the potential to achieve a higher Voc by switching to a higher ionization potential HTM. In this study we investigate this last point, probing more deeply the effects of varying the doping and ionization potential of the HTM on performance metrics including Voc. We study this change through a combination of experimental and modeling techniques. We first fabricated a series of TiO2/CH3NH3PbI3/HTM devices made with varying HTMs (Spiro-OMeTAD and α-NPD of varying doping concentrations) to empirically test the effect of changing HTM. We then used a drift-diffusion modeling package, SCAPS, to develop a device model that fits the observed trends and provides a greater understanding of the role of a HTM on the photovoltaic performance of these devices. Notably, we find that a perovskite device dominated by recombination at an electron trap best describes our observed trends, and a principal role of the HTM is controlling the population of these traps. Understanding this behavior allows us to develop designs rules for HTMs that minimize recombination, and therefore maximize the Voc. Overall this work lends useful insights into HTM optimization for devices made with CH3NH3PbI3, and looking forward can help us rationally design good architectures for alternative perovskites of varying band gaps.

Perovskite based solar cell meeting efficiencies of up to 20% now and position themselves as a highly promising next-generation photovoltaic technology. However, many of the challenges coming along with the scale-up this technology towards roll to roll manufacturing remain unanswered so far. Until now, the fabrication of high performance perovskite solar cells typically relies on either expensive organic hole transport materials and or high-temperature sintered TiO₂ electron transport layers, which greatly increases the materials and manufacturing costs of the final devices. The proper choice of the architectures or of the interface materials as well as their impact on the device performance and device reproducibility become of outmost importance.
In our first work¹, a novel water-free Poly (3,4-ethylenedioxythiophene) (PEDOT) in organic solvents is introduced into a perovskite solar cell architecture as a low-cost hole transporting material processed on top of perovskite thin films. Using a fairly simple device architecture of indium tin oxide (ITO)/low-temperature processed TiO₂ (LT-TiO₂)/perovskite/Water-free PEDOT/Au, a power conversion efficiency of up to 14.2% is achieved by using a pH neutralized PEDOT (Clevios SEJ 272). We also present a scalable, hysteresis free, planar and low temperature solution processed metal oxides hole transporting layer resulting in high open-circuit voltage, a hysteresis-free photovoltaic response and a hero PCE of 17.5%². All layers of this planar architecture solar cell were processed at temperatures below 140 °C, making this technology fully compatible with the requirements for roll-to-roll processing. Finally, we present a novel hysteresis free electron transporting layer for inverted planar perovskite solar cells³. By incorporating a low-temperature WO_x nanoparticular layer in combination with a mixed fullerene functionalized Self-Assembled Monolayer (SAM), we are able to demonstrate an inverted, planar structure and hysteresis free perovskite solar cell with a maximum efficiency of almost 15%.
All these findings underline the importance of systematically designing and developing customized interface materials for the perovskite technology, rather than adapting interface materials from the organic or dye sensitized photovoltaic community.
References
1 Hou, Y. et al. Inverted, Environmentally Stable Perovskite Solar Cell with a Novel Low-Cost and Water-Free PEDOT Hole-Extraction Layer. Advanced Energy Materials, doi:10.1002/aenm.201500543 (2015).
2 Hou, Y. et al. Overcoming the Interface Losses in Planar Heterojunction Perovskite based Solar Cells. Submited.
3 Hou, Y. et al. Low-Temperature and Hysteresis Free Electron Transporting Layers for Efficient, Inverted and Planar Structure Perovskite Solar Cells. Submited.

Hybrid organic-inorganic perovskite structured materials (mostly CH₃NH₃PbI₃ and CH₃NH₃PbBr₃) have shown remarkable opto-electronic properties, including photovoltaic and luminescent ones. One of the unique characteristics of the former (I-containing) perovskite is what appears to be an unusually high room temperature diffusion of one or more of the ions in the crystals. We investigate the possibility that this phenomenon is due to flexibility in the crystal lattice, facilitating ion movement as all of the candidate ions, apart from protons, i.e., I^-, CH₃NH₃⁺ (or in some cases NH₃NH₂⁺), or Pb²⁺ are (very) large ions. To this end, we have grown high quality single crystals in order to investigate their physical properties: notably nano-indentation was performed to determine Young&’s modulus (E) and Hardness (H) of CH₃NH₃PbI₃ and CH₃NH₃PbBr₃. For both materials, E is similar to previously computed values ^[1] with values in better agreement with computed results for the tetragonal than the cubic phase. We find a mechanical behavior similar to that of characteristic composite materials (e.g., bone), lead alloys and organic molecular solids ^[2].
Based on the measured data, we will discuss implications for possible elastic atomic displacement and for the possibility for local elastic deformation that could be compatible with diffusion of constituent ions, all of which are several times larger in terms of ionic or atomic radius than ions that can diffuse in ordered lattices, but similar to species that can diffuse in, say, polymeric or molecular solids.
[1]- J. Feng, Mechanical properties of hybrid organic-inorganic CH₃NH₃BX₃ (B = Sn, Pb; X = Br, I) perovskites for solar cell absorbers, Apl Materials 2, 081801 (2014)
[2] - M. F. Ashby, Materials Selection in Mechanical Design, Elsevier, 3rd edn, 2005.

We report effects of controlled humidity in ambient condition on grain boundary potential and charge transport within the grains of Pervoskite films prepared by sequential deposited technique. Grain boundary exhibited variation of their electronic properties with change in humidity level from 35% to 75%. The optimum humidity conditions with enhanced charge transport lead to a device efficiency of 12%. Spatial mapping of surface potential in the Perovskite film exhibits higher positive potential at grain boundaries compared to the surface of the grains. Grain boundary potential barrier for electrons and holes were found to vary by ~120 meV on changing humidity level from 35% to 75%. Nanoscale current sensing measurement (Cs-AFM) shows that charge transport in Perovskite solar cell strongly depends in humidity level. X-ray diffraction (XRD) and Raman spectra indicate the formation of PbI₂ phase with increasing humidity level. The degradation of Pervoskite solar cell is mainly assosciated with the increase of PbI₂ phase with increase in humidity level. Transient measurement shows better charge transport in optimum humidity level of 35%. Our results shows strong correlation between humidity level, electronic grain boundary properties and device performance.

Cu₂ZnSnS₄ (CZTS) is an emerging cost-effective and environmental friendly absorber material for thin film photovoltaic technology. While in the sputtering/sulfurization process, the reaction of precursors normally takes place at high annealing temperatures, we have recently developed a novel two-stage annealing strategy, which consists of a low-temperature annealing step followed by a high-temperature annealing step, for co-sputtered SnS₂-ZnS-Cu precursors to separate reaction and crystallization to occur at different temperatures. This method has demonstrated a significant improvement of power conversion efficiency to 8.58% from the previous record 6.77% as reported by Katagiri et al.. As characterized with various techniques, the key for the improvement is that the use of SnS₂ as a precursor allows CZTS phase to readily form at relatively low temperatures before SnS₂ to dramatically decompose into volatile SnS. The high-temperature stage not only allows large CZTS grains to grow but also removes unwanted secondary phases. Hence, this two-stage method makes the fabrication of CZTS thin film more controllable and may provide a new route to further increase the power conversion efficiency of CZTS solar cells.

Device efficiencies of novel Earth-abundant solar cells are often well below the Shockley-Queisser efficiency limit. Through thorough review of literature and spatially resolved electron-beam-induced-current (EBIC) analysis, we find that spatially varying intragranular diffusion length, and not grain-boundary recombination, is often the dominant recombination mechanism. We identify two root causes for low and spatially varying diffusion lengths: impurities and intragranular structural defects. While impurities are assumed to be homogeneously distributed, structural defect densities are known to vary significantly from grain to grain.
Herein, we develop a framework to detect and mitigate intragranular lifetime-limiting structural defects (e.g., dislocations) in novel thin-film materials, by applying metallurgical annealing principles developed in the metals, silicon, and III-V semiconductor communities. We utilize tin sulfide (SnS) as a model thin-film material, because of the recent efficiency progress demonstrated to date, and because an improvement in conversion efficiency from current 4% [1, 2] to approximately 10% is expected with an increase of lifetime from 0.1 to 1 ns. In our SnS thin-film devices, cross-sectional EBIC profiles show evidence for low intragranular bulk lifetime. Transmission electron microscopy (TEM) indicates the presence of a high concentration of extended defects. In a previous study, we demonstrated an improvement of diffusion length with increasing growth temperature [3], up to 285°C (0.48 × T_melt). In this study, we adapt high-temperature post-deposition annealing up to 0.68 × T_melt to reduce intragranular structural defect densities.
We initially use SnS thin-film morphology characterization (mainly through scanning electron microscopy (SEM)) as feedback for optimizing high temperature processing conditions, before investing in detailed microstructural characterization techniques including TEM. We will test the improved SnS thin-films in our previously de-risked device architecture [1]. By combining thorough intragranular structural characterization with advanced device characterization (i.e., temperature, wavelength, and illumination-dependent current-voltage measurements), we aim to evaluate the intrinsic limit of tin sulfide for high-efficiency solar cells.
[1] V. Steinmann et al., Advanced Materials 26, 7488 (2014).
[2] P. Sinsermsuksakul et al., Advanced Energy Materials 4, 1400496 (2014).
[3] R. Chakraborty et al., Applied Physics Letters 106, 203901 (2015).

β-CuGaO₂ is an oxide semiconductor possessing a wurtzite-derived structure that recently discovered [1]. Because it is a metastable material as compared to delafossite α-CuGaO₂ that is the well-known material as a p-type transparent conducting oxide, it is not directly synthesized by a solid-state reaction between Cu₂O and Ga₂O₃. β-CuGaO₂ is generally synthesized by ion-exchange of Na⁺ ions in β-NaGaO₂ possessing a wurtzite-derived structure with Cu⁺ ions in CuCl. Although direct and 1.47 eV of band gap and p-type electrical conduction of β-CuGaO₂ are suitable properties for thin-film solar cell absorbers, β-CuGaO₂ thin-film has not been reported yet. In the present study, we fabricated β-CuGaO₂ thin-films based on the synthesis of β-CuGaO₂ powders.
β-NaGaO₂ powder that was used as a sputtering target was prepared by solid-state reaction of Na₂CO₃ and Ga₂O₃. Na₂CO₃ and Ga₂O₃ were weighed, mixed, and then pressed into disks with a diameter of 17.2 mm at 100 MPa. The disks were fired at 900 °C for 20 h in air. Precursor β-NaGaO₂ films were deposited on a (0001)-Al₂O₃ single crystal substrate heated to 550 °C using an rf-magnetron sputtering technique. β-NaGaO₂ powder was pressed into a 1-inch diameter pellet and used as a sputtering target. Deposition was conducted under an Ar-gas atmosphere (10 sccm flow rate and 0.4 Pa total-pressure); the applied rf-power was 50W and the deposition time was 4 h. β-NaGaO₂ films obtained were placed on the CuCl pellet and heated at 300 °C in a vacuum furnace for 6 h in order to exchange Na⁺ ions in β-NaGaO₂ films to Cu⁺ ions. After the ion-exchange, the films were washed with acetonitrile and methanol to remove the residual CuCl and by-product NaCl, respectively.
Wide-angle X-ray diffraction indicated that the resulting films consisted of β-CuGaO₂, and the orientation of the β-CuGaO₂ films was the same to that of the precursor films. Tauc&’s plot of β-CuGaO₂ film indicated that the optical band gap of the film is 1.45 eV; this value well agrees with that previously reported for the powdered sample (1.47 eV). Consequently, we successfully fabricated β-CuGaO₂ thin-films. Detailed optical and electrical properties of the films obtained will be reported on the meeting.
[1] T. Omata, H. Nagatani, I. Suzuki, M. Kita, H. Yanagi, N. Ohashi, J. Am. Chem. Soc. 136, 3378, (2014)

Nanostructured quantum well and quantum dot solar cells are being widely investigated as a means of extending infrared absorption and enhancing photovoltaic device performance. In some implementations, nanostructured absorbers are employed to generate an intermediate band which can potentially increase photovoltaic efficiency by up-converting sub-band gap photons. In other designs, nanostructured layers are employed to enhance hot carrier extraction and increase photovoltaic efficiency by boosting the limiting voltage output. In this work, we describe the impact of nanostructured layer number on the performance of high-voltage InGaAs/GaAs quantum well solar cells. High-voltage InGaAs quantum well devices with one-sun open circuit voltages greater than 1 V have been demonstrated in a flexible, thin-film format by utilizing structures that employ advanced band gap engineering to suppress non-radiative recombination and expose the limiting radiative component of the diode current. Optical cavity effects result in oscillations in the measured external quantum efficiency of these thin-film quantum well devices. Multiple quantum well structures are observed to have a higher short circuit current but a lower open circuit voltage than similar single quantum well structures. Analysis of the underlying dark diode characteristics indicate that these high-voltage structures are indeed limited by radiative recombination at high bias levels.

Hybrid halide perovskites have rich solid state physics. We have recently reported the rotational activity of the molecular cations [1] using a combination of first principles simulations and quasi-elastic neutron scattering. A key beneficial aspect of these materials is the extremely long-lived charge carriers. One unique feature of hybrid perovskites that may extend charge carrier lifetime is the Dresselhaus splitting (due to large spin orbit coupling) of the conduction band leading to a slightly indirect band gap [2]. This effect is driven by the crystal field present in the bulk material due to the built-in dipole of the cation, and the distortion of the lead-iodide octahedra. Another effect is the fluctuation of electrostatic potential due to dynamic disorder, which will segregate hole and electron populations and so reduce recombination [3]. We have further extended our on lattice Monte-Carlo model StarryNight [4], which can simulate the dimensions of actual thin-film samples, to three dimensions. Analysis of the thermodynamic ensembles as a function of temperature reveal the transition from long-range (low temperature) to short-range (room temperature) domain structures. The electrostatic potential is reconstructed from dipole alignment, and used as an input to a statistical mechanical recombination model [5]. We quantify the beneficial decrease in recombination rate due to segregation of electrons and holes in the 'ferroelectric highways', versus the detrimental decrease in mobility due to disorder. Our new model quantifies the contribution of short-range ferroelectric order on carrier stability and electron-hole recombination in this unique class of materials.
This work has benefited from funding by the EPSRC and close collaboration with the groups of Mark van Schilfgaarde (Kings College London), Piers Barnes (Imperial College London), and Laurie M. Peter (University of Bath).
1. A. M. A. Leguy et al, Nature Comm. 6, 7134 (2015).
2. F. Brivio et al, Phys. Rev. B. 89, 155204 (2014).
3. J. M. Frost, K. T. Butler and A. Walsh, APL Mater. 2, 081506 (2014).
4. J. M. Frost https://github.com/WMD-Bath/StarryNight
5. J. M. Frost (unpublished).

Now that perovskite solar cells surpass 20% in certified power conversion efficiency, large area applicable coating techniques are in focus to scale production capabilities. Additionally, there is a need for straightforward and fast material optimization procedures, since the quest for high performance devices is fraught with numerous material and precursor permutations that are involved in the formation of perovskite thin-film solar cells. Concurrently pumped ultrasonic spray coating offers both scalability and eases material optimization.[1] Spray coating is used in industry to deposit large area and substrate-independent coatings without contact with the substrate. With the concurrently pumped technique, two solutions are differentially mixed inside of an ultrasonically vibrating nozzle prior to forming an aerosol. The differential pumping rate between the two mixed inks facilitates identical deposition and drying condition. This equality in deposition conditions can be used for the precise and exhaustive optimization of precursor ratios, precursor concentrations, and solvent systems.
In this contribution, we demonstrate >13% power conversion efficiency CH₃NH₃PbI₃ thin-film solar cells using the fast and reproducible single pass regime of ultrasonic spray coating. Unlike most other spray coating processes, in the single pass regime the sprayed droplets are allowed to coalesce on the substrate prior to drying into a smooth and uniform layer. Yet to be reported for perovskite devices, the single pass spraying process does not introduce interfaces between dried droplets, which may reduce grain boundaries and transport issues.. Mixed solvents of dimethylformamide (DMF) and γ-butyrolactone (GBL) were tested for coatability on electron beam deposited TiO₂ and on poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS) coated substrates. Coatability was tested via contact angle measurements and preliminary spray coating of layers. Devices based on both the n-i-p (transparent cathode ITO/TiO₂/CH₃NH₃PbI₃/spiro-OMeTAD/Au) and p-i-n (transparent anode ITO/PEDOT:PSS/CH₃NH₃PbI₃/PCBM/ZnO/Al) structures were fabricated. All processing steps were carried out at temperatures le; 100 °C, including the electron beam deposition of TiO₂, further enabling the industrially relevant fabrication for flexible and temperature sensitive substrates. The results of this work take the field one step closer to commercial realization of perovskite solar cells.
[1] J.G. Tait, et al, IEEE Journal of Photovoltaics, 6, 1538 (2014)

Organo-metal trihalide perovskite (OTP) is an emerging class of photovoltaic materials that attract attention because they combine the high efficiency typical of inorganic semiconductors with the low material cost and ease of fabrication of emerging technologies. The improvement of OTP solar cell efficiency has been unprecedented, skyrocketing from 3.8 % to 20.1 % in less than 5 years. However, the knowledge of how the local material properties, such as the chemical composition, the bandgap and the defect density relates to the OTP devices operation and performance is still limited.
Photo Thermal Induced Resonance (PTIR) is a novel, technique that combines the lateral resolution of atomic force microscopy with the specificity of infrared absorption spectroscopy. PTIR employs an AFM tip as a local detector to locally transduce the thermal expansion of the sample induced by light absorption in the sample into large cantilever oscillations. Our PTIR setup combines an AFM microscope with three lasers providing wavelength-tunability from 500 nm to 16000 nm, thus extending PTIR to the visible range for the first time. Local absorption spectra (electronic or vibrational) and maps are obtained with a wavelength-independent resolution as high as 20 nm.
In this talk the working principles of the PTIR technique will be discussed first, followed by its application to elucidate some open questions regarding OTP materials such as: a) their switchable photovoltaic effect, and b) the role and distribution of chlorine in CH₃NH₃PbI_3-xCl_x perovkite films. It is shown that nanoscale absorption maps (electronic and vibrational) provide unique information to engineer perovskite solar cells.
For example, it was observed that for OTP films sandwiched between two identical electrodes, the photovoltaic effect can be switched on and off by applying a small bias for few seconds. Here, PTIR is used map, in situ, the distribution of the methyl ammonium ion component of the perovskite film under applied bias, providing direct evidence of ion electron migration for the first time.
It is also commonly observed that when chlorine ions are present in the precursor solution of mixed Cl/I OTPs they provide several beneficial effects to the perovskite film; however, it is not clear whether those effects are related to chlorine incorporation or just to better crystallization conditions. Here, PTIR is used to map in situ the local bandgap distribution and chemical composition in the OTP film as a function of annealing time.

Recently, dynamic disorder in the organic cations in the hybrid lead halide perovskite was invoked to explain the extraordinary photovoltaic performance of these solution processed materials.
We present our study on the structural dynamics of lead halide perovskites with low frequency Raman spectroscopy. We compare hybrid perovksite containing MA ion with inorganic perovskite containing Cs ion. The dynamic disorder fundamentally exists in the Pb-halide inorganic backbone, with and without the presence of the MA ion. The coupling between the organic and inorganic moieties enhance the dynamic disorder, and exhibit significantly influence on the phase stability and the nature of phase transitions.
Our results on the lead-halide perovskite structural dynamics have strong implications on their electronic properties. The wave functions at the band gap reside only on the Pb-X framework, not the MA ion. The fundamental dynamic disorder of the inorganic framework thus directly and strongly modulate the transient local electronic band structures. This structural dynamics could facilitate both charge separation and transport. Our results also shed light on further materials development exploring systems with dynamic disorders beyond lead halide perovskites.

Given the appealing features of a confined system, semiconductor nanostructures (NS) are widely investigated, in particular for light harvesting applications. One of the most interesting features in nanostructures is the quantum confinement effect (QCE), arising in nanostructures smaller than the exciton Bohr radius (5 nm for Si, 24 nm for Ge). QCE is expected to increase the optical bandgap and the oscillator strength from the bulk values, allowing to tailor the light absorption spectrum and intensity. Still, up to now these effects have not been largely exploited, since other factors can hinder the QCE, such as interface related states, large spread in NS size, and defects in the embedding matrix. Quantum confinement theory typically assumes a sharp interface between a nanostructure and its environment, leading to an abrupt change in the potential for confined electrons and holes. When the interface is not ideally sharp and clean, significant deviations from QC model appear. Here we elucidate the role of the interface on QCE in Ge quantum dots (QDs, 2-10 nm in diameter) embedded in SiO₂ grown by rf-magnetron sputtering or plasma enhanced chemical vapor deposition (PECVD). Structural (TEM, RBS, Raman) and optical (absorption spectroscopy, Tauc analysis) characterizations, and effective mass approximation models are employed to describe the QCE-induced variation of optical bandgap (from 1.0 up to 2.5 eV) and of oscillator strength. Through a detailed electron energy loss spectroscopy (EELS) analysis we characterized the structural and chemical properties of QD interfaces. PECVD QDs exhibit a sharper interface compared to sputter ones, which also demonstrates a larger contribution of substoichiometric Ge-oxide states. Such a difference strongly modifies the QC strength, as experimentally verified by light absorption spectroscopy. A large size-tuning of optical bandgap and an increase in the oscillator strength occur when the interface is sharp [1]. A spatially dependent effective mass (SPDEM) model explains the larger reduction in the exciton effective mass in the sharper interface case [2]. Moreover, an unprecedented high light absorption efficiency, 15 times larger than in the bulk, was discovered for smaller Ge QDs grown in a multilayer configuration (where thin films with Ge QDs are separated each other by thick SiO₂ barriers). This effect is explained in terms of a local field effect inducing a polarization within the QD, screening the incoming electromagnetic waves. When QDs are packed in high density configuration this effect reduces the overall absorption. A multilayer configuration ensures a lower density of QD packaging, reducing the self-screening effect and enhancing the light absorption. These results add new insights into the role of interfaces and of packaging on confined systems, and open the route for reliable exploitation of QC effects.
[1] S. Cosentino et al., Nanoscale (2015)
[2] E. G. Barbagiovanni et al., JAP 117 (2015)

GaSb-based solar cells have potential to complement state-of-the-art GaAs-based photovoltaics by absorbing low energy photons as part of a four-terminal, mechanically-stacked architecture. An efficient four-terminal device requires a lateral conduction layer to efficiently extract majority carrier photocurrent through a front-surface metal grid by providing low sheet resistance and minimal optical loss of the light transmitted through the GaAs-based cells. For minimal optical loss in the GaSb-based material system, Al(1-x)Ga(x)Sb alloys have a wider bandgap than a GaSb absorber.
Lateral conduction layers are typically n-type for greater carrier mobility, but Al(1-x)Ga(x)Sb suffers from a low maximum electron concentration and a low mobility from intervalley scattering. Here, we report p+Al(1-x)Ga(x)Sb films synthesized via molecular beam epitaxy with silicon doping. We characterized the structural, electronic and optical properties and demonstrate conductivity greatly exceeding that in n-type layers.
Hall effect measurements reveal that the Si electrical activity is constant in these alloys at concentrations exceeding 1E19 cm^-3. This suggests there are no deep level traps for holes in these alloys, and the Group IV Si is not amphoteric at these high concentrations. High-resolution x-ray diffraction and atomic force microscopy show no degradation in crystallinity and morphology with high Si doping, and ellipsometry confirms a p+Al(0.2)Ga(0.8)Sb band gap wider than GaSb. The hole conductivity in Al(0.2)Ga(0.8)Sb can exceed 900 S/cm, greater than the electron conductivity by more than an order of magnitude, and is promising for the efficient operation of GaSb-based photovoltaics.

Lead iodide perovskite (CH₃NH₃PbI₃) is a promising light absorber for high efficiency solar cells, due to its high extinction coefficient, broad absorption spectrum, and excellent electrical properties.^[1] Mesoscopic perovskite solar cells contain mesoporous (mp) - TiO₂ nanoparticle layer for electron transport and collection, which is filled with light absorbing CH₃NH₃PbI₃. Power conversion efficiency of perovskite has reached 20.1%. ^[2] However, the charge recombination in the mp-TiO₂ based electron transport layer (ETL) is still one of the critical factors that hinder the electron transport. Grain boundary scatters electrons and causes electron-phonon conversion.
Herein, we report reduced graphene oxide (rGO)/mesoporous (mp) TiO₂ nanocomposite based mesostructured perovskite solar cells that show an improved electron transport property owing to the reduced resistance of the composite layer. The impact of the rGO added to the TiO₂ layer on the film resistivity, electron diffusion, recombination time and photovoltaic performances are investigated, thus the amount of the rGO is optimized. The rGO/mp-TiO₂ nanocomposite film, compared with the mp-TiO₂ film, prominently increases short circuit current density, open circuit voltage values and fill factor, which attests to lowered series resistance of the mesocopic layer. The rGO/mp-TiO₂ nanocomposite film with an optimal rGO content of 0.4 vol.% shows 18 % higher photon conversion efficiency than the TiO₂ nanoparticles based perovskite solar cells. Consequently, the rGO/mp-TiO₂ nanocomposite, for the mesostructured perovskite solar cells, exhibited the power conversion efficiency (PCE) of 14.5%.
[1] Hui-Seon Kim, Chang-Ryul Lee, Jeong-Hyeok Im, Ki-Beom Lee, Thomas Moehl, Arianna Marchioro, Soo-Jin Moon, Robin Humphry-Baker, Jun-Ho Yum, Jacques E. Moser, Michael Grätzel& Nam-Gyu Park, Scientific reports 2012, 2, 591
[2] Woon Seok Yang, Jun Hong Noh, Nam Joong Jeon, Young Chan Kim, Seungchan Ryu, Jangwon Seo, Sang Il Seok, Science, 2015, doi: 10.1126/science.aaa9272

Polymer based solar cells have been proposed as a lower cost alternative to Si where techniques such as spin casting and rapid roll-to-roll processing via large scale printing methods are applicable. The bulk heterojunction (BHJ) solar cell is an important example of a polymer solar cell technology that has been proposed in recent years. However, due to the disordered inner structures in the active layer, donor or acceptor domains isolated from electrodes and long path conduction, the power conversion efficiency (PCE) of BHJ solar cell is low. Therefore, control of the inner structure within the active layer is required to enhance the efficiency.
In our research work, we have demonstrated that the performance of BHJ solar cells can be significantly enhanced according to the highly ordered column structure within the active layer. In our approach, a self-assembly of tertiary polymer blend film is confined between the air and solid interfaces. The principal has been proved using a blend of PMMA: P3HT: PCBM where we showed that the PMMA phase formed a column structure in the P3HT, which template the PCBM phase between the electrodes. We also noticed that the performance had a strong relationship with the oxidation of the electrodes. The high vacuum processing condition causes a major challenge in putting it into a large-scale commercial manufacture. The conventional BHJ architecture has been replaced by an inverted structure where air-stable metals are used as the electrode. Inverted-type devices also avoid the requirement for the corrosive and hygroscopic hole-transporting PEDOT: PSS layer which are detrimental to device lifetime. Neutron reflectometry was used to demonstrate the confinement of PCBM at the interface between P3HT and PMMA/PS in the active layer. The columnar structured template is investigated under atomic force microscopy (AFM) and transmission electron microscopy (TEM). The different morphological structures formed via phase segregation are correlated with the performance of the PEV cells fabricated at the BNL-CFN and significant enhancement for the efficiency is observed.

Organic photovoltaics is one of the most promising solar cell technologies due to its potential for low cost manufacturing and relative ease of fabrication of devices on flexible substrates. While plastic substrates are traditionally used for this purpose, their permeability to moisture and oxygen is high. One of the better alternatives could be the use of metal foils such as of steel, which can be sturdy as well as impervious to oxygen and moisture. However, opaqueness of steel substrates requires designing the device in such a fashion so that maximum light can be absorbed in the device. Moreover, relatively higher roughness and conductivity of metal substrates requires use of layers which can lead to a pixelated device structure. In this work, we show the the role of interface layer morphology in improving the efficiencies of organic solar cell with a thin layer of gold as transparent conducting electrode. The device structure used in this study is Steel/Insulator/Al/ZnO/P3HT:PC₆₀BM/MoO₃/Au, with ZnO as an electron transport layer and Molybdenum oxide (MoO₃) as a hole transport layer. In this study, we show that the thickness and morphology of constituent layers plays an important role in maximizing charge extraction and improving the device efficiency leading to devices of ca. 1.4% power conversion efficiency with open circuit voltage of 0.6 V and fill factor of ca. 38% with 45% transparency of thetop electrode.

Methylammonium Lead Iodide (CH₃NH₃PbI₃) organic-inorganic perovskite films are a promising absorber material with solar cell efficiencies now in excess of 20%. A significant appeal of these materials is their synthesis using solution processes. Typically a low temperature anneal (about 100 °C) is involved in film synthesis with subsequent cooling through the cubic-to-tetragonal phase transition near 65 °C. Since the transition temperature is within the range expected in real world device applications, it is therefore important to understand the structural behavior at this transition and its impact on the device performance. In order to better understand this phase transition in thin films, we have developed the capability for operando synchrotron X-ray diffraction by designing a sample stage for simultaneous, temperature dependent measurement of J-V curves and diffraction. This has allowed us to obtain X-ray diffraction data during the operation of CH₃NH₃PbI₃ devices. Here we will present detailed structural characterization of the perovskite crystal structure with increasing temperature, including the tetragonal lattice distortion, octahedral rotations associated with the room temperature tetragonal phase, and thermal (disorder) parameters. The impact of these structural changes on the device J-V characteristics will be described and we comment on potential implications for material and device properties.

Inorganic-organic hybrid perovskite materials have recently shown rapid advancement in their use as the active material of photovoltaic devices, and the field continues to push the bounds of efficiencies to approach those achieved by modern silicon-based devices. With the corresponding increase in the possibility of wide-scale application of these materials, finding a viable alternative to lead in the perovskite material would be advantageous for making these solar cells environmentally friendly. In this work, we use ZnO nanowires as an electron transport layer to overcome short electron diffusion lengths observed in the tin-based perovskite material CH₃NH₃SnI₃. We control the coverage of the perovskite material on the ZnO nanowire scaffold using a hot-casting method and show excellent infiltration between nanowires. Additionally, chemical and thermal treatments of the perovskite films show the potential to manipulate the “self-doping” observed in these films. By controlling this “self-doping,” increased electron conductivity, as determined by the JV curves of electron-only devices, can be achieved to further overcome intrinsically short electron diffusion lengths. This study introduces mechanisms to overcome the fundamental limitations presented by the facile oxidation of tin-based perovskite materials and shows promise for efficient lead-free perovskite solar cells.

Organic-inorganic halide perovskite solar cells have attracted great attention due to promising power conversion efficiency with a potential low-cost manufacturing via cost competitive solution processes. To date, various solution processes such as spin-coating, inkjet printing, spray coating, blade coating and slot-die coating have been used to fabricate lab-scale perovskite solar cells. Among these methods, slot-die coating is regarded as the most scalable and roll-to-roll compatible process with a proven success in producing large scale printed organic solar cells. Therefore, we developed a slot-die coating process for the fabrication of perovskite solar cells using a home-built slot-die coater based on 3D printing platform. Slot-die coated perovskite devices on ITO glass showed up to ~12 % power conversion efficiency. The roll-to-roll compatible process is actually used in roll-to-roll process and it is proven that the process is feasible in roll-to-roll process with a comparable efficiency of roll-to-roll printed perovskite solar cells on flexible substrates. The batch to roll-t-roll translation process as well as analysis of printed perovskite films will be presented in this work.

In Cu(In,Ga)Se₂ (CIGS) thin film solar cells, a buffer layer is able to have an advantageous properties for lattice match between CIGS absorber layer and transparent conducting oxides layer as upper layer. Excess carrier lifetime and quantum efficiency can be increased by adopting the optimized conditions of buffer layer process. The ZnS(O, OH) material has been intensively studied as buffer layers for wide-bandgap chalcopyrite solar cell, resulting in enhanced transmittance and low optical absorption in the range of short wavelength. Cd-free buffer layer has been considered for friendly-environment material, replaced with CdS. During chemical bath deposition (CBD) process, Zn atoms are poorly diffused than Cd into CIGS absorber layer due to low activation energy than Cd atoms. For this reason, post-treatment such as air annealing, light soaking (LS), heat-light soaking (HLS), ammonia rinsing should be applied for the higher efficiency solar cells. Among them, an electron beam irradiation (EBI) can be excellent method for simple and rapid post-treatment with good quality. It can be ascribed to generate electrons from Ar plasma with effectively controlling electron dose and accelerated energy.
In this work, ZnS(O, OH) buffer layer was prepared by CBD process with ZnSO₄ 0.16 M, ammonia 7.5 M, Thiourea 0.6 M aqueous solution. The bath temperature was fixed at 80 ^oC. And the thickness of all samples was controlled above 100 nm. Four post-treatment methods of air annealing, light soaking (LS), heating light soaking (HLS) and EBI were applied for buffer layer with the same condition. An air annealing temperature was fixed at 200 ^oC and the annealing time was controlled. The LS was treated under halogen lamp, the HLS was treated by using the combination of air annealing and LS at above 100 ^oC, and the e-beam was irradiated at 200 W of radio frequency (RF) power and DC power 1~4 kV for 5 min. And then, the impacts of post-treatments such as well-known air annealing, light soaking (LS), heat-light soaking (HLS), but also another method of an electron beam (e-beam) irradiation on cell performances were investigated for ZnS(O, OH) / CIGS solar cell. The as-deposited and post treated buffer layers were investigated by X-ray diffraction (XRD), photoluminescence (PL), transmittance electron microscope (TEM) and GD-OES. The CIGS thin film solar cells were respectively fabricated using as-deposited and post-treated buffer layer and characterized by light induced current-voltage (LIV) and external quantum efficiency (EQE).
EBI can be a good candidate for simple and rapid post-treatment method of buffer layer. The more useful information will be reported on the upcoming conference.

The hybrid perovskite (CH₃NH₃PbX₃, X=Cl, Br, I) materials using organic and inorganic elements have attracted much attention as an efficient light harvester due to their high absorption coefficient and ambipolar transport characteristics. Thus, intensive researches on photovoltaic devices combining the hybrid perovskites with variously nanostructured materials have been carried out to highly increase power conversion efficiency by enhancing light trapping and photo-generated charge transfer.
In this study, we report the efficiency enhancement of CH₃NH₃PbI₃ perovskite solar cell by incorporating flower-shaped mesoporous TiO₂ layer as an electron transport layer. Here the 200 nm thick mesoporous TiO₂ layer was imprinted with a three-dimensionally flower-shaped template prepared by ion-assisted aerosol lithography (IAAL) suitable for making a complex three-dimensional nanostructure. Then, CH₃NH₃PbI₃ perovskite layer was deposited on them by one-step spin coating process, followed by spiro-MeOTAD coating and Au deposition as a hole transport layer and a counter electrode, respectively.
The current density-voltage measurements show enhancement in short circuit current density (J_sc) and fill factor (FF) of the perovskite solar cell with the flower-shaped mesoporous TiO₂ layer compared to the reference device with normal mesoporous TiO₂ layer. The observed enhancement of power conversion efficiency is attributed to increased light scattering in a broad spectral range by the flower-shaped mesoporous TiO₂ layer, as well as improved the rate of charge transfer due to enlarged interfacial area. Therefore, this work demonstrates great potential of hybrid patterning by a combination of soft lithography with IAAL technique for efficient light harvesting in perovskite solar cells.

Recently organic-inorganic hybrid perovskite solar cells have shown tremendous potential for solar energy harvesting applications. Since its discovery in 2009 by Miyasaka et al., power conversion efficiencies skyrocketed beyond 20% (verified, NREL 2014) in liquid electrolyte-free solid state devices within only 5 years of extensive research worldwide. A major drawback of the currently best performing solar cells lies in their chemical composition, namely the lead content. The toxicity of lead containing compounds remains a huge obstacle for commercialization. Although lead analogues 3D tin perovskite CH₃NH₃SnI₃ thin films were already successfully integrated into devices with moderate power conversion efficiencies (PCE~ 6%) by H. Snaith et al., stability issues became even more demanding in comparison to the air and moisture sensitive lead perovskite. In an attempt to replace lead by a non-toxic metal, organic-inorganic bismuth halide perovskite sheets were prepared via spin-coating technique and investigated as an alternative to lead.
Lead-free solution processed solid state (CH₃NH₃)₃Bi₂I₉ heterojunction devices (ITO/PEDOT:PSS/(CH₃NH₃)₃Bi₂I₉/PCBM/Ag) were assembled by solution processing and their J-V characteristics (V_oc= 0.7 V, I_sc= 0.22 mA/cm², FF= 0.49) were recorded under simulated sunlight conditions (100 mw/cm²). A comprehensive investigation of the physicochemical properties of the (CH₃NH₃)₃Bi₂I₉ sheets was performed by means of SEM, XRD, UV-VIS, PDS, XPS, UPS, PL, and Raman spectroscopy. Density functional theory (DFT) and time-dependent DFT (TDDFT) simulations were performed to provide an insight into the structural, vibrational and optoelectronic properties of (CH₃NH₃)₃Bi₂I₉, thus providing a comparison between the obtained experimental- and calculated data.

Heterostructure with Aluminum nitride (AlN), Gallium nitride (GaN) based materials have been interest especially for Light Emitting Diodes (LEDs). Aluminum nitride buffer layers were grown on Patterned Sapphire Substrate (PSS) by reactive radio frequency (RF) sputtering in plasma, which is composed of a mixture of argon and nitrogen gas at different currents with the consistent thickness (~ 200nm). The influence of AlN crystalline quality as the crystalline of AlN gets improved, the achievement uniformity in surface of GaN is possible. For Gallium nitride grown by Metal Organic Chemical Vapor Deposition (MOCVD) of approximately 4.5 um was characterized using a Field emission scanning electron microscope (FE-SEM) and atomic force microscope (AFM). The differences in the GaN growth behavior affect the crystallinity of the GaN epitaxial layer. In this study, results were obtained from high resolution x-ray diffraction (HR-XRD) panchromatic (300 to 800 nm) cathodoluminescence (CL), and the dislocation densities were calculate using full with half maximum (FWHM) values. In this study, the formation of the high quality was studied, which crystallinity of AlN requires to make high quality GaN epitaxial layer on PSS is investigated.

The height and size of nanopillar are significant factors to decide how much the nanopillars contribute to enhancement for PCE (Power Conversion Efficiency). In order to enlarge interfacial area, it is clear that we induce the development for smaller nanopillar if possible. However, too small nanopillars would be damaged by the coating process of top layer, which is coated on nanopillars in sequential coating for bilayer. Further, the narrow width of nanopillars prevents the solution for top layer inserting into the nanopore between nanorods. Accordingly, we should attempt to trade off between process manipulation and size control through its individual manner. Meanwhile, previously reported researches have been focused on how to make nanopillars smaller, though there is no regard to stereoscopic shape of nanopillar. However, it is required to be selected an optimum template or mold which decides the shape of nanopillar in pressure-type method such as nanoimprint lithography and anodic aluminum oxide (AAO) membranes. In chemical reaction method such as phase segregation and removal, it would be harder to control the shape of nanopillar. If we can predict the optimal shape of the nanopillar, further manipulations such as controlling the concentration, the reaction rate and the pre-post treatment can be implemented for the specific shape. Despite these necessities, any of experimental and theoretical studies for optimization of the size and the shape in poly(3-hexylthiophene) (P3HT) nanopillar have not yet been investigated to the best of our knowledge.
We demonstrate a modeling analysis to decide size and shape for most effective P3HT nanopillars, suggesting the analogy between bilayer and nanopillar structures. A regression function for PCEs as to height of nanopillar is derived from experimental results of bilayer solar cells with the thickness of 20/80, 40/60, 60/40, and 80/20 nm in P3HT/[6, 6]-phenyl C₆₁ butyric acid methyl ester (PCBM). We set up an inequation using the function and the integration, satisfying a requirement that amount of energy generated in flank and bottom sides of unit nanopillar is greater than the one obtained at interface of bilayer solar cell. Accordingly, we obtain an optimum and maximum radius for nanopillars with various shapes, depicting a final function via this inequation. As a result, we finally showed that the radius of P3HT nanopillars with rectangular or cylinder, cut-cone, cone shape should be less than 135, 53, 2 nm respectively.

Perovskite solar cells have emerged as next-generation photoelectric conversion devices having authorized photoconversion efficiency of 20.1 %. We have recently reported that using mixed tin-lead halide perovskite solar cells it was possible to extend the photon harvesting window in the near infra-red region up to 1060 nm in contrast to the typically observed 800 nm for lead-perovskite solar cells. Considering the energetic constraints, P3HT was used as hole transport layer in place of most commonly used Spiro-OMeTAD leading to observed photoconversion. However, perovskite solar cells based on mixed lead-tin halides, still lacks to exhibit sufficient photoconversion efficiency and efforts are needed to enhance the photoconversion efficiency further. In this presentation, a flat uniform film fabrication of perovskite thin films based on mixed tin-halides will be discussed. Efforts have been directed to control the crystal growth using tin iodide complexes and its implications on the photoconversion behavior.

Organometallic halides perovskites solar cells are one of the most important developments in recent years, and have been one of the main topics of research in the fields of sensitizing dye solar cells, organic and thin film solar cells. Among the advantages of those materials are the narrow band gap, high absorption coefficient, efficient exciton dissociation and high mobility of carriers. However, perovskites are unstable in air and, as a result, perovskite based solar cells are susceptible to oxygen, moisture, ultraviolet light and solution processes. In this study, thin films of the perovskite were obtained by the method of depositing two-step solutions via spin coating under nitrogen atmosphere. To reduce the rapid degradation of the perovskite layers, one or two protective barriers have been applied to those films: the first was a polymer coating of ethylene vinyl acetate (EVA), and the second one was an organic compound derived from myricin. The permeation rate of moisture and oxygen into each protection layer was measured. Optical density spectra of the perovskite films, which were coated with either one or two protection layers, were monitored as a function of storage time in air. It includes that the perovskite compound degradation rate could be reduced by protection films and the mechanism of protection is under investigation.

Research and development activities on solution based organic or hybrid solar cells have become intensive because of their promising low cost and scalable production technologies. The conventional heterojunction structure of those cells consists in a transparent conductive coating (TCO) as the anode and a metal contact with low work function as the cathode. However, recent studies showed that such architecture generates drawbacks in terms of the Open-Circuit Voltage (V_oc) and long term stability due to oxidation of metal contacts. Inverted Hybrid Solar Cells (IHSCs) with a high work function metal as the anode and a modified TCO as the cathode are more stable in air. Specifically, ZnO thin films can work as Electron Transport Layers (ETLs) or hole blocking layers in IHSCs. In this study compact ZnO thin films were deposited on TCO substrates by a Spray Pyrolysis Technique (SPT) without use of nitrogen as carrier gas. The surface morphology, the roughness and the optical properties of the obtained ZnO films were studied as functions of deposition temperature and time, the solution composition and the distance between the substrate and the spray system. IHSCs were prepared with compact ZnO films at TCO cathodes and organic-inorganic perovskite with nanoparticles of ZnO as active layers. It is found that the stability of the cells was influenced by the thickness and morphology of the compact ZnO layers. The results suggest that the growth of high quality ZnO thin films by solution method is a key challenge in the field of low cost photovoltaics.

In recent years Cu₂ZnSnS₄ (CZTS) based thin film solar cells have attracted much attention as a nontoxic and earth-abundant alternative to more established thin film technologies. Currently, one of the limiting factors for CZTS cell efficiency is a deficit of open circuit voltage, possibly in part due to an inferior band alignment at the front contact. So far, the best results have been reported for device structures containing CdS window layers. A promising Cd-free alternative for a window layer material is In₂S₃, which may also have a better conduction band alignment leading to higher open circuit voltages. For the application of In₂S₃ in CZTS thin film solar cells, the main desired criteria are a low deposition temperature, suitable optoelectronic properties, as well as the scalability of the process.
In this work the reactive sputter deposition of In₂S₃ is investigated utilizing a high throughput combinatorial screening of the temperature-flux parameter space with a variety of characterization techniques including XRD, XRF and UV-VIS transmission measurements. It is demonstrated how changes in sulfur partial pressure controlled by the addition of H₂S to the sputtering gas, as well as deposition temperature influence the composition, structure and orientation of the In₂S₃ films. Films of phase pure In₂S₃ could be deposited over a wide temperature range.
This established UHV deposition process for In₂S₃ enables ongoing work towards an in-depth analysis of the band alignment at the In₂S₃/CZTS hetero-junction using photoelectron spectroscopy (PES) measurements, which will lead to a better understanding of the performance limitations associated with a non ideal conduction band offset in conventional CdS/CZTS based devices.
In addition device libraries were built on CZTS substrates to further investigate the feasibility of the sputter deposited In₂S₃ layers for CZTS solar cell processing. In₂S₃ thickness gradients were applied to determine the ideal buffer layer thickness.

The organic-inorganic hybrid perovskite solar cell have achieved great progress in power conversion efficiency with optimization of device structure and deposition method. As the key component of mesoscopic perovskite solar cells, TiO₂ mesoporous layer plays an important role in electron injection and charge separation. Besides, nanopatterned TiO₂ mesoporous layer can increase light scattering into perovskite layer efficiently. Therefore, in order to further increase the opto-electronic properties of perovskite solar cells, optimization of mesoporous TiO₂ layer is needed.
In this regard, we report a bio-inspired perovskite solar cells which utilize moth&’s eye patterned mesoporous TiO₂ layer. We fabricate PDMS mold by soft lithography using Silicon masters with various size of moth&’s eye pattern and make moth&’s eye patterned mesoporous layer by stamping method. On that layer, CH3NH3PbI3 perovskite layer was deposited by one-step spin coating method and hole transfer layer was deposited on the CH3NH3PbI3 by spin-coating of spiro-MeOTAD. Finally, Au counter electrode was deposited by thermal evaporation.
With the presence of moth&’s eye patterned mesoporous TiO₂ layer, efficiency of perovskite solar cells is improved due to increase of short circuit current density (J_sc) and fill factor (FF). It was because moth&’s eyes patterned mesoporous TiO₂ layer enhances light scattering of incident light into the perovskite layer and increasing perovskite/mesoporous TiO₂ interfacial area improves electron injection.

Hybrid organic-inorganic Perovskite based materials have recently catered the growth of highly efficient low cost photovoltaic devices with energy conversion efficiencies reaching as high as 20%. A variety of organic, inorganic and halide components have been employed in ABX₃ combination, allowing research communities all over the globe to tailor the material properties to a great extent. Here, we report the effect of incorporating pseudo-halide thio-cyanite anion (SCN^-) in MAPbI₃ framework, resulting in a mixed halide perovskite. The pseudo-halide thiocyanite (SCN^-) anion has an effective ionic radii of ~217 pm which is relatively closer to that of iodine&’s effective ion radii of 220 pm and is expected to substitute a fraction of Iodine atoms in the parent material. XRD and FTIR studies confirm the presence of SCN^- anion in the bulk of the material resulting in MAPbI_3-x(SCN)_x configuration. Thin films of MAPbI_3-x(SCN)_x spin coated on porous Alumina substrates exhibited a band-gap of 1.65 eV, a significant 80 meV shift from the parent material (MAPbI₃). Room temperature and cryogenic PL measurements along with absorption measurements show that mixing (SCN)^- causes a significant increase in the stoke shift in comparison to MAPbI₃. Further analysis on the PL emission reveals a possible contribution of the shallow states on the dynamics of excited charge carriers. PL decay measurements (TCSPC) on MAPbI_3-x(SCN)_x accounts ~86% of the radiative recombination to the bimolecular recombination processes, but it should be also noted that in SCN^- incorporated material the slow decay kinetics corresponding to charge extraction seems to be slower than MAPbI₃. Also the room temperature emission revealed a ten-fold increase in the emission intensities for MAPbI_3-x(SCN)_x as compared to the parent material MAPbI₃. Studies on MAPbI_3-x(SCN)_x micro crystals using spatially resolved emission spectroscopy attributed this increase in intensity to a highly edge specific emission. Also the transition energies and spectral line-widths obtained at various (local) nanodomains of MAPbI_3-x(SCN)_x micro crystals were found to be non-uniform. Interestingly certain sites in MAPbI_3-x(SCN)_x micro crystals exhibited unambiguous temporal intermittency in emission (blinking) which is extremely unusual and fascinating. This unusual behaviour of optical instability under photo irradiation was also observed for the first time in MAPbI₃ micro crystals as well, leading us to believe that this could be general phenomenon exhibited by most hybrid perovskite materials. Also, MAPbI_3-x(SCN)_x was employed as a photon absorber material in porous-Titania based configuration with Li-doped spiro-OMeTAD as the hole transport material.

Currently actual questions of increasing the conversion efficiency of solar energy into electricity and reasons of the degradation of solar cells (SCs) are studied continuously. It has great economic importance and it is important for saving the planet's ecology. The conversion efficiency of SCs depends on the structure of the solar module, the type of semiconductor materials and etc. Certain structural defects with deep energy levels (DLs) or deep traps in the band gap in active semiconductor layer affect on the stability of SCs electrical properties. DLs are responsible for the generation-recombination processes and form the recombination component of direct current through the p-n junction.
The subject of this work is the analysis of deep level defects influencing on conversion efficiency of SCs based on Si p-n junction. SCs have different conversion efficiency coefficient in the range of 14-20%. The theoretical evaluation of the influence of DLs in the band gap of the semiconductor material on the efficiency of solar cells was carried out. Samples were studied by capacitance-voltage characteristics at different temperatures (C-V-T characteristics) and current deep level transient spectroscopy (I-DLTS). Free charge carriers distribution profiles in the base of the diode structure were estimated from C-V measurements. DLs energy spectra were studied by I-DLTS. DLs concentration was calculated with taking into consideration C-V data. The data about the DLs energy spectra and its parameters such as activation energy, capture cross section, concentration are obtained. The studies were carried out at the measuring complex specially designed for the study of SCs semiconductor diode structures [1].
The studied samples are varying only by values of the conversion efficiency. Samples were made on the Si substrates from different parties. Technological operations during the manufacturing process of each sample are not differing from each other. As a result of studies by I-DLTS it is shown that the DLs spectra depends on the quality of Si substrate. As a result of measurements the correlation between the total concentration of DLs and the values of the SCs conversion efficiency is found.
This work is executed at financial support of the Ministry of Education and Science of the Russian Federation (Project RFMEFI57414X0006) using of the equipment of Regional Center Probe Microscopy collective use of Ryazan State Radio Engineering University.
References
1. V.G. Litvinov, N.V. Vishnyakov, V.V. Gudzev, V.G. Mishustin, S.M. Karabanov, S.P. Vikhrov, A.S. Karabanov Measuring Complex for Analysis of Recombination Deep Traps in Semiconductor Solar Cells // USB Proceedings 2015 IEEE International Conference on Industrial Technology (ICIT) Seville, Spain 17 - 19 March, 2015. P. 1071-1074.

Organic and inorganic hybrid material of CH₃NH₃PbX₃ with perovskite crystal structure has conceived as emerging light absorbing materials for high efficiency photovoltaic devices due to their higher light absorption coefficient, long absorption wavelength, and excellent electron/hole diffusion length. Here, we demonstrate the screening effect of the polarization states on the charge redistribution related with photovoltaic performance in ferroelectric CH₃NH₃PbI₃ thin films using atomic force microscopy based techniques. Fundamental physical information, such as switchable ferroelectric polarization and the piezoelectric coefficient of CH₃NH₃PbI₃ thin films, was investigated via piezoresponse force microscopy (PFM). On the basis of PFM results, we further observed the interplay between polarization and injected charges to explore screening effects in the as-grown and illuminated states using Kelvin probe force microscopy. The obtained results clearly reveal that the direction and the amount of the charge transfer can be influenced by the screening of the polarization states at the surface. These results can provide fundamental information regarding the influence of ferroelectricity on the CH₃NH₃PbX₃ solar cells.

CuInS₂ (CIS) thin film solar cell is paid attention because of its high absorption coefficient, adjustable band gap, high incident photon conversion efficiency, good stability and so on. The best cell, prepared by using vacuum deposition process, recorded efficiency 20.3%. However, those processes required high vacuum conditions which would lead to high cost and limitation on mass production.
To overcome these obstacles for commercialization, several solution-based, low temperature deposition methods have been investigated, including precursor solution growth and nanoparticle (NP) ink deposition. These process usually use direct coating of solution containing metal (Cu, In, Ga) and chalcogen (Se, S) precursors, and proper heat treatment process for film crystallization. Recently, Liu et al. used a hydrazine-based homogeneous precursor solution, which could coat high quality solution film with a remarkably high cell efficiency of 12%. Although high efficiency, hydrazine solution is so toxic flammable that it is not proper for commercialization.
1-D nanostructures metal oxide (e.g., ZnO, TiO2, ITO) has been consider promising materials for more efficient solar cell. Ordered nanostructures increase light capture via diffused reflections between nanostructure arrays, and they enhance charge separation and transport via their unique inherent morphological, electrical, and optoelectronic properties (e.g., their high surface-to-volume ratio and high electron mobility).
This research tried to produce low cost CIS solar cell through non-hydrazine solution coating, but more efficiently with nanowire structure. Superstrate construction was chosen for nanowire coating. It was composed of glass/ ITO electrode/ ZnO nanowire/ CdS layer/ CIS film/ Au electrode. Simple hydrothermal method made ZnO nanowire on ITO glass, and nano-crystal layer deposition (NCLD) method made ZnO/CdS nanowire core-shell. The solution with Cu, In and S precursor was spin coated on the ZnO/CdS nanowire core-shell for absorption layer. Au electrode was deposited with sputtering.
SEM, XRD measuring proved that CIS film was deposited onto ZnO/CdS nanowire core-shell. UV-transmittance checked the absorption range of each film. I-V curve showed that efficiency of the cell is 6.4%.
In conclusion, superstrate CIS solar was produced using non-hydrazine solution process at low temperature. The cell with ZnO nanostructure could show better efficiency than earlier superstrate flat solar cell. It was because the nanostructure let contact area at each layer larger than one of flat structure so that higher efficiency in light absorption and carrier transfer were possible.

Highly efficient dye sensitized solar cells (DSSCs) based on organic optoelectronic materials have attracted significant attention over the past two decades for their promise as economic alternatives to conventional solar cells based on silicon. Our approach to increasing the photon conversion efficiency is to improve the light-harvesting capability of a photoelectrode film by utilizing optical enhancement effects, which can be achieved by harmonizing the light scattering in a hierarchically structured TiO₂ (HS-TiO₂) electrode with the photon absorption spectrum of the dye. Also, we designed a novel metal-free dye (JH-1) with a broad and high intensity absorption spectrum that was harmonized with HS-TiO₂ aggregate (an average size of 660 nm) to achieve resonant scattering. The incident photon-to-electron conversion efficiency (IPCE) was drastically improved (more than 30%) by the harmonizing the size of the polydisperse HS-TiO₂ with absorption spectrum of JH-1 dye. The resultant resonant multiple scattering enhanced the light harvesting efficiency and charge collection yield. The DSSC yielded an high photocurrent density of 20.9 mA cm^minus;2, reaching the efficiency of 9.18% under 100 mW cm^minus;2, AM 1.5G illumination without an antireflection layer. Using the novel metal-free organic dye, the device also displayed very good long-term stability (efficiency of 7.44% after 800 h under the irradiation light intensity of 100 mW cm^minus;2 at 60°C).

In spite of the tremendous progress in the efficiency of MAPbI₃ perovskite-based solar cells, several fundamental aspects of these cells still lack complete understanding, even though progress has been made. One of them concerns the hysteresis behavior in the current-voltage curve of the solar cells, which is suspected to be also related to the fast degradation and relative instability of the cells over time. Recently there have been several works claiming that the cause for the hysteresis is due to the polarization of the perovskite layer under applied voltage bias and illumination, which is believed to be induced by significant ion migration within the perovskite layer.
To follow the issue systemically, current-voltage characteristics of MAPbI₃ inverted solar cells were measured at different temperatures, ranging from 320K down to 200K with different scan rates (5 mV/s to 1500 mV/s) and illumination intensities (0.1 - 0.6 AM1.5). Hysteresis behavior in current-voltage curves was not observed at room temperature (~ 296 K), as expected for p-i-n perovskite cells in the inverted architecture. Hysteresis behavior gradually increases upon lowering the temperature for fast scan rates, reaching a maximum at 200K. Transient response of short circuit current was monitored as a function of temperature and a slower photocurrent decay response was observed at lower temperature. We suggest that the observed phenomenon is further evidence of ion migration within the perovskite layer, and a model is proposed to explain our experimental observations.

Thin-film solar cells based on Cu₂ZnSnS₄ (CZTS) absorber layers have been shown to benefit from the addition of sodium to the layer during the preparation process. Various positive effects from sodium have been reported, including increase in the grain size, carrier concentration, and/or hole mobility, passivation of deep defect states, and improvements in Jsc, Voc, fill factor, and power conversion efficiency. In this work, we examine a series of CZTS devices grown on Mo-coated glass which did not contain a diffusion barrier between the Mo and glass. The result is a spatially inhomogeneous Na distribution in the CZTS layer. We will show that light beam induced current (LBIC) measurements are an excellent and high-throughput method for evaluating Na homogeneity in a CZTS layer of a solar cell. This is demonstrated by comparing LBIC measurements with secondary ion mass spectrometry (SIMS) measurements recorded over similar length scales to show that LBIC “hot-spots” are spatially correlated with Na “hot-spots” measured by SIMS. We will also show that sodium is also correlated with chemical interactions between adjacent layers in the device stack. Regions of the sample containing higher Na content also exhibited more copper in-diffusion with the underlying MoS₂ layer. Additionally, these regions had a suppressed signal of Cd near the back contact. Neither of these effects are well understood in CZTS devices, and neither have previously been correlated with the presence or absence of Na.

Hybrid organic-inorganic lead halide perovskite photovoltaic cells have already surpassed 20% conversion efficiency in the few years that they have been seriously studied. However, many fundamental questions still remain unanswered as to why they are so good. One of these is: Is the organic cation really necessary to obtain high quality devices? To date, there are no reports on high band gap (> 2 eV) all-inorganic perovskite solar cells. CsPbBr₃, a high bandgap semiconductor with a perovskite structure at standard temperature and pressure, has been estimated to have an electron mobility of ~1000 cm²/(V s) and comparable electron/hole lifetimes of 2.5 µs [1], emphasizing CsPbBr₃ has the potential to be a promising absorber in solar cells. In this presentation we discuss if perovskites with an inorganic A cation can form light absorbers with PV properties, comparable to those of the alkyl ammonium perovskites, in particular in terms of high open circuit voltages that are an important feature of these cells. We show that CsPbBr₃ exhibits a photovoltaic performance that does not fall behind CH₃NH₃Br₃, including relatively high values of open circuit voltage typical of the hybrid perovskite. [2]
References
[1] C.C. Stoumpos et al., Cryst. Growth Des. (2013), 13, 2722-2727.
[2] M. Kulbak, D. Cahen and G. Hodes, J. Phys. Chem. Lett. (2015), 6, 2452-2456.

Aluminum oxide (Al₂O₃) deposited with the atomic layer deposition (ALD) has been known with its excellent surface passivation on low doping silicon wafers. Considerable interest is therefore drawn to see if the excellent passivation remains on high-doping silicon surface, such as the aluminum doped back-surface field (BSF) region. The latter is commonly used in current silicon solar cells, and its passivation with ALD-Al₂O₃ is of interest to know. In this work, we examine the ALD-Al₂O₃ passivation on Al-doped surface formed with screen-printing of Al paste. The sheet resistance and thickness of BSF used in this work is 25 ohm/#9633; and 6 um respectively. When the surface was coated with ALD-Al₂O₃, an effective surface recombination velocity S_eff of 50 cm/s, corresponding to a Jo around 8x10^-14 A/cm², is resulted. This implies the open-circuit voltage of cell can be as high as 690 mV, and the cell efficiency over 21% becomes feasible.

Perovskite solar cells based on organo-lead halide perovskite are being visualized as a promising candidate in contributing to large scale solar energy production due to their high power conversion efficiency, low cost and compatibility with different scalable processing techniques.
Methyl ammonium lead iodide based perovskite (CH₃NH₃PbI₃) have shown their best performance due to its excellent optoelectronic properties and certified efficiency reached up to >18 %. However, the instability of CH₃NH₃PbI₃ due to its structural phase transition at high temperature makes them unsuitable for real applications, as it affects the photovoltaic performance. To overcome this problem, high temperature resist formamidinium organic cation was employed to replace methyl ammonium to increase its temperature stability and optical absorption onset in these mixed cation based perovskite solar cells. The different molar ratios of formamidinium were used to replace methyl ammonium cations in order to keep the narrow band gap and thermal and structural stability. This strategy of mixed organic cations in lead iodide based perovskite improves the photovoltaic efficiency by tuning the optical, electrical, and morphological properties of the absorber material; and also the stability of the devices. Further, the device PV properties was compared using two different methods of perovskite deposition; sequential and one-step method using solvent engineering. Compared to sequential deposition, the solvent engineering approach in one step method leads to the formation of compact, uniform, complete coverage perovskite layer with large crystals which resulted in a suppression of defect-assisted recombination across the film thickness, as observed in SEM images. The devices were subjected to different electro-optical characterization and for stability measurements, they were kept at 50% RH and evolution of PV parameters was studied. It was found that the mixed cation based perovskite was more stable than that of their pure organic cation either methyl ammonium or formamidinium cation based perovskite using one step deposition method.

Hybrid organic-inorganic perovskites with general formula ABX₃ (where A is an organic cation, B: Pb²⁺ and X: I^- or Br^-), have emerge as new absorber material for 3^rd generation photovoltaics. In a regular architecture, perovskite is sandwiched between mesoporous scaffold made of electron transporting material like TiO₂ or other isolating oxide (Al₂O₃ or SiO₂) and hole transporting material (HTM) [1]. Previous attempts have been developed to create a mesoporous photonic crystal alternating TiO₂ and SiO₂ layers conferring different colors to the perovskite solar cells. However, in the attempt of giving color to the devices, the overall power conversion efficiencies of the perovskite solar cells (PSC) were decreased.
In the present work, we report a porous 1-Dimensional photonic crystal (1-DPC) in order to manage the light harvesting properties of the perovskite absorber. These devices consist on several monolayers of tilted nanocolumns deposited by physical vapor deposition at oblique angle deposition (PVD-OAD) [2]. Changing deposition conditions, such as temperature, pressure and angle of deposition, porosity and then refractive index of each monolayer can be tuned. Varying the refractive and thickness of each layer, and total number of layers, the transmission spectra of the full mesoporous structure is controlled providing a minimum in transmission spectra (i. e. maximum in absorbance one), in the same range where the perovskite absorbs better (400-450 nm), producing higher light absorption and current. Employing this technique, we have created two types of 1-DPC photoanodes in PSC: 5 layers of alternated TiO₂ with two different refractive indexes giving a power conversion efficiency of ~11%, and 3 layers of TiO₂/SiO₂/TiO₂ give in excess of ~12%. It is worth to note that non-electron injecting layer like SiO₂ can also be employed as photoanode in perovskite solar cells.
References
[1] S. Kazim, M. K. Nazeeruddin, M. Grätzel and S. Ahmad, Angew. Chem. Int. Ed., 2014, 53, 2812-24.
[2] F.J. Ramos, M. Oliva-Ramirez, M.K. Nazeeruddin, M. Grätzel, A.R. González-Elipe, S. Ahmad, J. Mater. Chem. A, 2015, 3, 13291-98

Perovskite solar cell with thin CH₃NH₃PbI₃ layer shows a power conversion efficiency as high as 20%. In order to surmount limited diffusion length of CH₃NH₃PbI₃ perovskite light absorber and accompanied J-V hysteresis, mesoporous TiO₂ layer (mp-TiO₂) was introduced. The main function of mp-TiO₂ inside of the perovskite device is to prevent electron-hole pair from recombination during transfer through active layer. The traditional method of achieving mp-TiO₂ layer is spin-coating onto the TiO₂ blocking layer using TiO₂ paste solution which is diluted in ethanol. However, spin-coating is not suitable for large-area deposition or continuous roll-to-roll process due to its batch-process feature. Therefore, spin-coating method has been noticed as limitation for mass production so far. In this research, we use electro-spray deposition (ESD) system which enables the process continuous, easy, and high throughput with possibility of large-area deposition. Compared to conventional spin-coated mp-TiO₂ layer, electro-sprayed mp-TiO₂ layer indicated nearly identical morphology. The main difference in morphology between each method is the pore size of mp-TiO₂ layer. Porosity of mp-TiO₂ layer plays an important role in charge separation mechanism of perovskite solar cell due to extent of infiltration of perovskite material into mp-TiO₂ layer. Electro-spray deposition system satisfy the requirements for two main improvement issues presented above, large-area deposition and expand pore size. As a result, we could achieve high power conversion efficiency perovskite solar cells with electro-sprayed mp-TiO₂ layer via ESD system which shows better efficiency than those with mp-TiO₂ layer via spin-coating method. As well as enlarging porosity, electro-spray TiO₂ layer showed less defect and better crystallinity. Simply changing the method of deposition with the same TiO₂ precursor enables both high PCE and large-area deposition.

Lead trihalide alkylammonium, a hybrid organic-inorganic perovskite, has impressed the photovoltaic research community due to the very desirable properties that make it suitable for the direct and clean conversion of solar energy in electricity. In particular, compact thin films of TiO₂ have been commonly used as electron collection layers in perovskite solar cells and the optimization of their electrical conductivity would improve the transport of electrons generated in the perovskite layer. In this work, two subsequently spin-coated compact TiO₂ layers have been prepared by a sol-gel method. Their electrical conductivities were enhanced by chemical doping in the sol-gel solution with the trivalent cations Fe³⁺, Al³⁺ and Bi³⁺. The chemical doping also modified slightly the Fermi level of each TiO₂ layer, which is essential to achieve an efficient charge transport in the cells at the adequate tuning levels. Furthermore, double deposition of TiO₂ layers helped to reduce the density of pinholes formed during the processing of a single blocking layer and to avoid the leakage of electrical current within the cells. The effect of the doping is described through the characterization of the structural, morphological and optical properties of the as-modified TiO₂ photoanodes and their performance is assessed through the incorporation into perovskite-based solar cells. It is found that at certain concentrations of the doping agents the double TiO₂ layers achieve an optimum conductivity that enhances the power conversion efficiency of perovskite-based solar cells compared to those with the undoped TiO₂ compact photoanodes.

Hybrid organometallic perovskite, e.g., methylammonium lead iodide (CH₃NH₃PbI₃) is attracting considerable attention as energy-efficient light absorber materials for photovoltaic applications owing to their solution processability, tunable bandgap, strong absorption coefficients and cost effectiveness. In this study, we demonstrate that CH₃NH₃PbI₃ planar heterojunction (PHJ) solar cell with enhanced efficiency can be fabricated by introducing a fullerene (C₆₀) layer of different thickness (0, 3, 7 and 10 nm) as an interlayer between air-stable amorphous titanium oxide (TiO_x) and CH₃NH₃PbI₃ layer. The highest solar cell performance was found at an optimum C₆₀ thickness of 7 nm. The introduction of C₆₀ interlayer between CH₃NH₃PbI₃ and TiO_x layers can increase photo-produced charge carrier cites, followed by accumulation lowering and trapping of photo generated charges at the TiO_x interface. As a result, a boost in the short-circuit current density (J_sc), fill factor (FF) occurs, and an enhanced power conversion efficiency (PCE) compared to that obtained with the solar cells having no interlayers was observed. All the solar cells have been fabricated using the following structure: ITO/TiO_x/C₆₀ (or without)/CH₃NH₃PbI₃/Spiro-OMeTAD/Ag. A solar cell fabricated using phenyl-C₆₁-butyric acid methyl ester (PCBM) as the interlayer within the similar architecture instead of a C₆₀ interlayer, has been used as the reference material. The incorporation of the C₆₀ interlayer caused an increase in J_sc from 11.90 mA/cm² to 15.17 mA/cm², while the value is limited to 13.47 mA/cm² with the PCBM interlayer. The enhancement in the FF (0.49) due to the C₆₀ and PCBM were, respectively, 0.69 and 0.55. The enhancement of J_sc and FF due to the C₆₀ interlayer was attributable to the reduction of the injection barrier between CH₃NH₃PbI₃ and compact-TiO_x interface, which leads decrease in series resistance from 25.75 to 7.39 Omega;cm². The PCE increases significantly from 5.10 to 9.51%, when C₆₀ interlayer is inserted between the CH₃NH₃PbI₃ and compact-TiO_x layers. We also observed that the surface energy and morphology of C₆₀ layer employed an impact on the resulting device performance. The surface energy of the compact-TiO_x layer can be modified in the range of 43 to 56.5 mJ/m², with the introduction of C₆₀ interlayer. The root-mean-square (RMS) roughness value for the C₆₀ films was 4.21, 4.12, 3.31 and 8.28 nm, respectively, at 0, 3, 7 and 10 nm thickness. The RMS roughness was smoother with 7 nm C₆₀ as compared to the other thickness. Hence, we assume that the surface energy control might give further insight to improve the performance of perovskite solar cells.

Here we report controllable crystallization method to CH₃NH₃PbI₃ lead-halide perovskite by sequential deposition with solution casting and spin-drying of methylamine iodide (MAI) solution. Recently, formation of perovskite thin film solar cells without hole-blocking TiO₂ layer has attracted match attention, with its process ease and industrial productivity. Recent studies showed that two-step sequential deposition method has an advantage in film smoothness for planar devices without hole-blocking layers. In that case, MAI doping is commonly realized by dipping PbI₂ film into MAI solution or dripping (drop-by-drop adding) the solution during spinning the PbI₂ substrate. But dipping method requires large quantity of solution and too long dipping to the solution might damage the thin film substrate, and dripping method could not introduce enough MAI into the film. Thus, there is a strong need for enhancing perovskite film quality in two-step method.
In this study, MAI solution was casted before the spin starts, and spin-dried to control the MAI doping ratio. Doping speed was controlled by the concentration, quantity of the solution, and humidity of the environment. We used PbI₂ solution containing a small amount of MAI as a first precursor layer. With cast-spin method, the doping was done for shorter period of time than dipping method. This could be attributed to increasing concentration of MAI solution by solvent evaporation during the process. As-doped perovskite film showed XRD pattern with significantly reduced remnant PbI₂ peak at 12.3^o and enhanced perovskite CH₃NH₃PbI₃ peak at 13.8^o, compared to films by drip-spin coating method. J-V curve measurement and stability test was conducted on planar structure perovskite solar cells fabricated by drip-spin and cast-spin method. Experimental results suggested that the photovoltaic effect depends on the reaction time of PbI₂ and MAI.

The introduction of plasmonic structure is a viable way for enhancing the performance of optoelectronic devices by improving the coupling of surface plasmons. In this study, by employing block-copolymer lithography and nano-imprinting lithography, we demonstrate highly effective multiple-patterned plasmonic nanostructures of metal (gold or aluminum) electrodes for use as back reflectors in various solution-processable organic optoelectronic devices, such as organic photovoltaics (OPVs), organic photodiodes (OPDs), and organic phototransistors (OPTs). The surface plasmon effects of metal electrodes allow significant additional light absorption compared with metal electrodes without the multiple patterns, leading to remarkable enhancements in the power conversion efficiency (PCE) of OPVs, the external quantum efficiency (EQE) of OPDs, and the EQE of OPTs. These results demonstrate that the developed multiple patterns constitute a versatile and effective route for achieving high-performance organic optoelectronic devices.
KEYWORDS: organic optoelectronics, plasmonic structure, multiple patterns, block co-polymer lithography, nano-imprinting lithography, DDF calculationm, SNOM

Lead halide perovskite solar cells have attracted considerable attention with the potential of achieving high efficiency with low cost fabrication processes. Over the last few years, the cell efficiency has sharply increased, and now exceeds 20 %. However, despite this rapid development, fundamental cell operation mechanism such as a charge separation or recombination process at the perovskite interfaces has not still been explicitly understood.
In this presentation, we will present photo-excitation intensity dependent electron and hole injections in a CH₃NH₃PbI₃ perovskite deposited TiO₂ nanocrystalline film with a spiro-OMeTAD hole conductor. The excitation intensity increase from 10 nJ/cm² to 50 mJ/cm² has accelerated excited state decay from 180 to 5 ns. This accelerated excited state lifetime competes with electron and hole injection processes. Based on the kinetic analysis, the estimated electron and hole injection rate is 11±1 ns and 1.8±0.2 ns, respectively. Increased excitation intensity up to 50 mJ/cm² lowers the electron and hole injection yield to 10 % and 50 %, respectively. However, under AM1.5G one sun condition (100 mW/cm²), both electron and hole injections must occur with almost 100 % efficiency. The charge separated state lifetime was prolonged with the presence of the TiO₂ electron accepter and the spiro-OMeTAD hole acceptor, 20 times longer compared to N₃ sensitized TiO₂ film, suggesting that their presence indeed supports optimizing solar cell performance. The similar electron injection rate (10 ns) was recently observed for CH₃NH₃PbBr₃ perovskite deposited TiO₂ film. Therefore, the similar electron and hole injection mechanism can be expected for other organometal halide deposited TiO₂ film with a spiro-OMeTAD layer.
Key words: perovskite solar cells, CH₃NH₃PbI₃, TiO₂ nanocrystalline film, spiro-OMeTAD,electron and hole injection, photo-excitation intensity dependence

Cubic phase of SnS and SnSe in the rock salt structure has been known since 1967 in thin films formed by thermal evaporation of the source materials on cleaved NaCl crystals. This has been listed as a polymorph of their more common orthorhombic crystalline phase. During 2006-2009, another cubic polymorph in the zinc blende structure for SnS was reported by our group. The material in thin film was obtained by chemical deposition. Solar cell structures using the material reached conversion efficiency of 0.5%. We have recently clarified that this cubic structure is correctly a large simple cube (SnS-CUB) with a cube edge double that of the zinc blende structure and containing 64 atoms in the unit cell. In this work we illustrate that SnSe thin films can also be deposited from chemical bath into such large simple cubic structure, on a SnS-CUB layer. The SnS-CUB and SnSe-CUB condense into very compact thin films, with optical band gap of 1.74 and 1.35 eV, respectively; considerably larger than of their ORT polymorphs, of 1.1 and 0.94 eV, respectively. We present here the solar cell performance of multilayer SnS-CUB/SnSe-CUB absorbers and discuss the thermal stability of such solar cells. Particular importance will be given to the X-ray diffraction patterns of the multilayer films obtained in the grazing incidence mode bringing out the structural determination of these large single cubic polymorphs with cube edge of 11.587 Å for SnS and 11.975 Å for SnSe.

CuSbS₂ is a semiconductor material has received a lot of interest to materials scientists and chemist because of it is practical application to solar cell as absorbing layer. In this work, CuSbS₂ thin films were obtained in one-step by chemical bath deposition on glass substrates at 35 °C for 16 h. The films were thermally annealed at 380 °C and 5x10^-3 torr for 45 min for 1 h. Characterization of the films was performed by X-ray diffraction, atomic force microscopy (AFM), profilometry, UV-Vis spectrophotometry and electrical characterization. CuSbS₂ films showed a thickness of 335 nm. X-ray diffraction analysis shown the crystalline state of the annealed films was calcostibite (JCPDS 44-1417). The band gap energy was estimated around 1.23 eV. The electrical conductivity of the material presented was 0.19 S/cm.
The material obtained was incorporated into the cell with structure glass/SnO₂:F/CdS/Sb₂S₃/CuSbS₂ using C-Ag paints as electric contacts. Measurement of the optoelectronic properties was carried out using a solar simulator (1000 W/m², AM 1.5 filter); from the current density-voltage (J-V) curve, the values #8203;#8203;obtained were for V_oc = 382 mV, J_sc = 5.324 mA/cm², FF = 0.32 and eta; = 0.657% with a thickness of 335 nm of CuSbS₂. Using the same structure and painting as an electric contact only Ag, the following results were obtained: V_oc = 468 mV, J_sc = 1.598 mA/cm², FF = 0.32 and eta; = 0.238%. Values obtained are higher than the values previously reported by Y. Rodríguez-Lazcano et al. (V_oc = 345 mV and J_sc = 0.2 mA/cm²) using the chemical bath deposition as technique for preparation of the absorber layer.

Nanoparticles of oxide semiconductors play crucial roles in mesoscopic solar cells such as dye-sensitized solar cells (DSSCs), quantum dot solar cells and perovskite solar cells. Not only the absolute size of the particles but also their shape and preferentially exposed crystal facets are expected to cause drastic changes in their performance, because it serves as the interface for carrier extraction and also for recombination. Also expected is the influence to the chemical stability of the light absorber (dye, Q-dot, perovskite) to be deposited onto the oxide surface.
While titanium dioxide is widely used, zinc oxide (ZnO) is an attractive alternative due to its high electron mobility and flexibility in its structure control. We have succeeded in synthesis of size and structure controlled ZnO nanoparticles by rapid hydrothermal crystallization employing microwave radiation and structure directing agents (SDAs). Here, the synthetic methods, structural characterizations and their properties in DSSC applications are reported.
Microwave radiation for 30 min to aqueous solutions containing Zn(CH₃COO)₂ and SDAs such as triethanolamine (TEOA) and 1,2,4,5-benzenetetracarboxylic acid (BTCA) with controlled pH by addition of KOH resulted in formation of differently shaped highly crystallizad ZnO. ZnO with hexagonal bi-pyramidal (HBP) shape and with nanosheet (NS) shape in about 50 to 100 nm size were obtained by use of TEOA and BTCA, respectively. Combination of the observation techniques of scanning electron microscope (SEM) transmission electron microscope (TEM), selected area electron diffraction (SAED) revealed that the axis of the HBP ZnO corresponds to the c-axis of ZnO,while the triangular sides in 6-fold symmetry correspond to the (102), making it the predominantly (84%) facet to be exposed. On the other hand, the predominantly exposed facet of the NS ZnO was identified as that of the (100) and it accounted as much as XX% of the total surface area.
Dye adsorption capability of these ZnO nanoparticles was checked by employing D149 and also for commercial polycrystalline ZnO (TAYCA MZ-500) for comparison. The surface area occupied by a single D149 molecule at saturation was determined as 0.68, 0.62 and 1.2 nm² for MZ-500, HBP and NS, respectively, indicating the superior dye stability on the (102) surface. DSSC devices employing these sensitized ZnO photoelectrodes exhibited differences in current and voltage; J_sc (mA cm^-2) = 12.0 (HBP) > 10.8 (MZ) > 8.05 (NS), V_oc (mV) = 615 (NS) > 579 (MZ) > 562 (HBP). Densely adsorbed D149 on the (102) surface appears to be the most efficient for electron injection. On the other hand, either the elevated band position and/or suppressed recombination may be the reasons for the high voltage of the NS. Further studies to elucidate the origin of these differences are under way by employing the techniques such as voltage decay, IMVS and SLIM-PCV.

Organic/inorganic lead halide perovskite (OIP) materials with OrPbX₃ structure (Or being an organic cation, X one of the halides Cl, I or Br) have recently gained unpreceded attention in the scientific community. Driven by the success in high-performance OIP based solar cells reaching a power conversion efficiency greater than 20% the versatility of this material class has been widely acknowledged in optoelectronics, spanning from solar cells over light-emitting diodes, photodetectors to transistors and light-emitting transistors. Whilst OIP solar cells already outperform their organic counterparts the performance in other areas of application, especially light emitting diodes (LED), is still lacking with most reported external quantum efficiencies not exceeding 1% at room temperature. It can be expected that the low-temperature solution process in ambient condition or inert atmosphere, compared to high-temperature vacuum processes for inorganic semiconductors, leads to a significant number of defects explaining the gap between expected and obtained performance in this devices.
In this work, we show, for the first time, an experimental study on the character of electronic trap sites in CH₃NH₃PbBr₃. By thermal admittance spectroscopy (TAS) we construct a trap density of states (tDOS) and compare it with computational results of the physical origin of traps and their respective energies. It is commonly known that traps with energies deep in the band-gap act as quencher and are diametrically opposed to device performance. Via time-correlated single-photon counting (TCSPC) we demonstrate the influence of traps acting as non-radiative decay channel thus drastically shortening the decay lifetime and reducing photo luminescent quantum yield. We develop an effective treatment to reduce the number of deep traps and demonstrate the influence on device performance in LED employing OIP as emitter material.

Anatase and rutile titanium dioxide thin films grown by a low temperature process are investigated for their use as a single layer antireflection coating for GaAs solar cells. The thin films are obtained by spin coating a layer from the TiO₂ sol-gel and subsequently annealing at 150 ^oC. The sol-gel is synthesized by the hydrolysis of titanium isopropoxide in the presence of an acid or a base. By controlling the pH of the sol-gel during growth, pure anatase and rutile phases are obtained. A pH of around 3.0 yields anatase phase while a pH of 9.0 yields pure rutile phase TiO₂. The two different phases of TiO₂ are characterized by measuring the Raman scattering spectra. The optical constants, thickness and reflectance of the thin films on GaAs are obtained using a spectroscopic ellipsometer. The sol-gel is spin coated on GaAs based solar cells and annealed at 150 ^oC to form the anti-reflective layer. The performance of the solar cells is evaluated before and after coating with the TiO₂ films. The anatase TiO₂ anti-reflective films performed better than the rutile with a power conversion efficiency enhancement of 50%. Quantum efficiency enhancement of 63% and 25% are obtained with anatase and rutile phase films respectively. The performance enhancement of the solar cells using these thin films can be attributed to the destructive interference of light associated with a single layer coating on the solar cell surface.

Methylammonium lead halides (MAPbX₃, X = halogen) and their mixed-halide crystal analogs draw a wider range of concern among people because of the outstanding photovoltaic performance, and realizing the fabrication of low cost and high performance solar cell devices. However, perovskite materials are highly sensitive to moisture, most fabrication of high performance perovskite solar cell devices carried out with very low levels of H₂O and O₂, which limits the development and application of this kind of high performance solar cells.
CH₃NH₃PbI₃/ZnO mesoporous solar cell device have been robustly fabricated in 70% relative humidity (RH), which resulting a relative stable pervoskite solar cell device. The devices were stored in 40±3% RH conditions without any encapsulation, and maintained 72% of their initial power conversion efficiency (PCE) after long time storage (about 254 days).
PbI₂ play a positive role for stabilizing the performance of CH₃NH₃PbI₃/ZnO mesoporous solar cell device. In a fresh CH₃NH₃PbI₃/ZnO thin film PbI₂ crystal can be also observe, indicating the robust fabrication of CH₃NH₃PbI₃/ZnO mesoporous thin film can help the PbI₂ formation, and which not only exist on CH₃NH₃PbI₃ surface but also at the CH₃NH₃PbI₃/ZnO interface, due to the decomposition of CH₃NH₃PbI₃ under high RH condition. On one hand, PbI₂ layer can act as a moisture defender to protect the CH₃NH₃PbI₃ in the ZnO nano-rod interspace as hole transfer material (HTM) for stabilizing the performance. On the other hand, PbI₂ layer exist between ZnO and CH₃NH₃PbI₃ interface made the CH₃NH₃PbI₃/ZnO p-n junction become CH₃NH₃PbI₃/PbI₂/ZnO p-i-n structure and improved the diffusion length of the photogenerated carriers, and enhancing the photovoltaic performance of stored perovskite solar cell device. The consequent stabilization of performance has been revealed by transient photovoltaic (TPV) measurement. It shows that the thin films include PbI₂, fresh CH₃NH₃PbI₃ and stored CH₃NH₃PbI₃ are all possess long carrier life time, the stored CH₃NH₃PbI₃/ZnO thin film still maintain relative fast photogenerated carrier transfer stability compare with the fresh CH₃NH₃PbI₃/ZnO thin film. It demonstrate that PbI₂ in the CH₃NH₃PbI₃/ZnO-nano-rod thin film do not substantially affect the dynamic process of CH₃NH₃PbI₃/ZnO-nano-rod thin film which fabricated under high RH condition, and part of the CH₃NH₃PbI₃ can be protected with PbI₂ layer. This robust CH₃NH₃PbI₃ solar cell devices fabrication technique shows a new perspective for performance stable pervoskite solar cell development.

We fabricate mixed halide perovskite films by dual-source vacuum deposition of PbX₂ (X= Cl, Br, and I) and methyl ammonium iodide (MAI) precursors with various deposition rates. The vacuum deposition is an optimal way of film fabrication in the sense that it gives a perovskite film free of contamination such as lead metalic phase, residual solvent, and moisture.The film stoichiometry and crystallization as a function of the precursor type and relative deposition rate are closely related to its electronic properties. The ionization potential and band gap of the MAPb(I_1-yBr_y)₃ film are controlled by halide composition due to electronegativity difference and lattice constant change. However, the MAPb(I_1-yCl_y)₃ films show negligible change in stoichiometry with precursor depostion rate due to MACl formation and evaporation during the film formation. None the less, a tiny incorporation fo Cl in the perovskite film attects surface morphology and possibly carrier conduction.

A CIGS absorber with high gallium contents in the space-charge region can reduce the carrier recombination and improve V_oc. Therefore, controlling Ga grading on top of CIGS thin film solar cells is the purpose of this experiment. To reduce Se vacancy is important that Ga elements of diffusion into Se vacancy between Mo back contack and CIGS absorption layer be decreased. In order to reduce Se vacancy and confirm Ga inter-diffusion, two CIGS solar cells were fabricated by converting CIG precursor without and with Se interlayer. The copper-indium metallic precursors were fabricated corresponding to the sequence CuIn/In/Mo/Stainless Steel Substrate (STS) by sequential direct current magnetron sputtering and Se layer was evaporated by RTA system to obtain a Se/CuIn/Mo/STS stack. CuGa precursor layer was also fabricated on the Se/CuIn/In/Mo/STS stack to cupper-indium-gallium precursors. Finally, both CuGa/Se/CuIn/In/Mo/STS and CuGa/CuIn/In/Mo/STS stacks were selenized at 500 °C for 1 hour. Through the secondary ion mass spectroscopy (SIMS) and X-ray diffraction (XRD), there is as change between fabricated CIGS absorption layers that the amount of Ga elements decreases from the top to bottom layer of the CIGS absorption layer gradually. We will discuss the effect of Se interlayer in CIGS absorption layer and its effect on the solar cell.

The implementation of a front and a back grating is a promising approach towards improving light trapping for thin photovoltaic cells. However, an optimal design strategy has yet to be fully explored. In this study, we investigate light trapping using a simple 2D triangular groove grating design on the front and back of a 10-µm-thick crystalline silicon absorber. Using numerical simulations, it is shown that, to maximize light trapping, the front and back gratings must exhibit a mismatch in period in order to effectively couple incident light into the absorber while simultaneously preventing light from coupling back out. According to these results, a simple design rule was developed, based on the least common multiple (LCM) of the front and back grating periods, to optimally design a double grating light-trapping structure. Using this LCM design rule, it is predicted that for a front grating period of W_1d = 2200 nm and a back grating period of W_2d = 3177.8 nm, the photo-generated current (J_ph) can be as high as 38.02 mA/cm² at normal incidence, which is nearly 4mA/cm² or 11.74% higher than J_ph=34.06 mA/cm² for the case of a single front grating case with a period of 2200 nm. Thus, by optimally choosing the periods of a front and back grating, it is possible to achieve substantial improvement of light-trapping performance even for simple and easily manufacturable grating designs. Furthermore, this approach is not restricted to only photovoltaics, but it can be applied to any materials in different applications such as photodetectors.

This research focuses on the effect of ligand exchanged CdSe quantum dot (QD) interlayer in inverted organic solar cells. The ~4 nm monodispersed CdSe QDs were synthesized using by oleic acid, cadnuim oleate and trioctylphosphine selenide (TOPSe) under solution process. Because the oleic acid ligand has a long alkyl chain which act as an insulator, the QD ligand exchange has been conducted from an oleic acid ligand to pyridine to increase electrical conductivity. To overcome the general problems of self-aggregation of ligand exchanged CdSe QDs, SAM treatment on the ZnO layer was performed by using a 3-mercaptopropionic acid (3-MPA) to graft CdSe QDs uniformly as the n-type monolayer between ZnO and the active layer. The CdSe QD interlayer uniformity has been confirmed from confocal microscope image. In the inverted organic solar cells, the ligand exchanged CdSe QD interlayer contributed to increase device efficiency compared to the device without interlayer. These result can be mainly attributed to the effect of electron transport and hole blocking effects, with increased fill factor from decreased series resistance (Rs) and increased shunt resistance (Rsh). This study will contribute to apply inorganic nanoparticles as thin film fabrication by solving self-aggregation problem.

The issues related to durability of organic solar cells should be addressed before its commercialization. Hence, a conventional solar cell of the poly (3-hexylthiophene) (P3HT) : (6, 6)-phenyl-C₆₁butyric acid methyl ester (PC₆₁BM) blend on ITO substrates was fabricated, and investigated, which shows improved life time by using a combinational Ca/Al cathode, wherein the deposition of calcium layer is extended beyond the aluminum layer. The extended deposition of calcium layer beyond aluminum prevents the edge degradation by its oxidation and hence protecting the active pixel area of the device, resulting in the improvement of life time of device from sim;80 hours to sim; 400 hrs.

Graphene produced by Chemical Vapor Deposition (CVD) is a favorable transparent conductor, with potential to substitute conventional electrode materials, such as indium tin oxide in (opto-)electronic devices.
In this work, we focus on solving the common problem of energy level mismatch between graphene and the organic semiconductor overlayer, by using the molecular acceptor hexafluoro-tetracyano-napththoquinodimethane (F₆TCNNQ) to non-covalently functionalise the graphene sheet.
Using ultraviolet photoelectron spectroscopy, the work function of graphene on glass was found to continuously increase from 4.5 eV up to 5.5 eV upon sequential deposition of F₆TCNNQ. The large work function increase is disentangled into two contributions: a) charge-transfer (CT) induced p-doping of the graphene sheet, shifting the Fermi level -0.6 eV w.r.t the Dirac point and b) an interface dipole that increases the vacuum level by 0.4 eV. The CT is manifested as an additional emission line at the low binding energy side of the neutral N 1s core level, measured with X-ray photoelectron spectroscopy (XPS). This emission indicates that the transferred negative charge is mainly localised in the cyano groups located at the periphery of the molecules. XPS also shows that a mixture of negatively charged and neutral molecules co-exist in the (sub)monolayer regime. Finally, near edge X-ray absorption fine structure spectroscopy measurements indicated that the molecular monolayer consists mainly of flat-lying molecules.
To examine the impact of pre-covering graphene with a molecular acceptor, that increases its work function in a working device, we use the commercially available tetrafluoro-tetracyanoquinodimethane (F₄TCNQ), with an electron affinity similar to that of F₆TCNNQ. The resulting organic solar cell consisted of the prototypical zinc-pthalocyanine (ZnPc) and C₆₀ as donor and acceptor materials, respectively. We could demonstrate improved device characteristics for the photovoltaic device with the F₄TCNQ interlayer due to more efficient hole extraction at the interface between the molecularly modified graphene and ZnPc.

Graphene quantum dot (GQD) layers were deposited as a down-conversion layer on crystalline silicon solar cell surfaces by kinetic spraying of GQD suspensions. A supersonic air jet was used to accelerate the GQDs onto the surfaces. Here we report the coating results on a silicon substrate and the GQDs&’ application as a down-conversion layer in crystalline silicon solar cells, which enhanced the power conversion efficiency (PCE). GQD layers deposited at nozzle scan speeds of 40, 30, 20, and 10 mm/s were evaluated in crystalline silicon solar cells; the results indicate that GQDs play an important role in increasing the optical absorptivity of the cells. The current density (J_sc) is enhanced by about 2.94 % (0.9 mA/cm²) at 30 mm/s. This improved the PCE by 2.7 % (0.4 percentage points) in p-type silicon solar cells compared to a reference device without a GQD down-conversion layer.

Graphene has triggered an immense interest in its application as a transparent conducting electrode (TCE) material, owing to its high optical transmittance, electrical conductivity, flexibility and chemical stability which have positioned graphene as a strong candidate for replacing Indium Tin Oxide against other emerging TCE materials such as metal nanowires and carbon nanotubes. Chemical vapor deposition is currently the most promising route towards large scale production of graphene; however, it typically results in polycrystalline graphene and requires an additional transfer process which further introduces defects and impurities leading to an increase in the sheet resistance. Doping of graphene with foreign molecules that physically absorbs on its surface without disrupting the basal plane has been a popular route for reducing its sheet resistance by increasing the carrier density via charge transfer, which typically comes at a significant loss in optical transmission.
Herein, we report the successful bromine doping of few layers graphene (4-10 layers) resulting in air-stable transparent conducting electrodes with up to 80% reduction of sheet resistance reaching ~180 Omega;/#9633; at the cost of 2-3% loss of optical transmission. The remarkably low tradeoff in optical transparency leads to the highest enhancements in figure of merit reported thus far. The charge transfer p-doping is confirmed by Hall measurements, Raman spectroscopy and work function measurements which showed an increase by up to 0.3 eV. Bromine is mainly physically absorbed on graphene sheets as evident by XPS measurements in which minimal covalently bonded bromine was observed. XRD results and AFM/STM height measurements have confirmed bromine intercalation in between the sheets in addition to being present at the surface, and hence have an increased stability in air as compared to bromine doping of a single layer graphene. These results should help pave the way for further development of graphene as potentially a highly transparent substitute to other transparent conducting electrodes in optoelectronic devices.

ZnO exhibits a wurtzite hexagonal structure and its direct optical bandgap of 3.4 eV at RT is wide enough to transmit most of the useful solar radiation. In our previous work [1], non-doped ZnO films were successfully grown on a polyethylene terephthalate film by a conventional spray pyrolysis at 150 #730;C using a diethylzinc (DEZ)-based solution under an air atmosphere. It is well known that the DEZ reacts with water and/or oxygen at low temperature, and ZnO can be generated.
In this work, novel precursor solutions were prepared by the reaction of DEZ and water in some solvent. Diethoxyethane and/or other ether solvents were used for the solution. In sample preparation, all operations were carried out under inert gas (e.g. nitrogen). The hydrolysis reaction was carried out by addition of water to DEZ in solvent under cooling condition to remove reaction heat and avoid side reactions. After aging the reaction mixture at RT for 2 hours, the novel precursor solution containing Zn-O structure compound was obtained.
A novel precursor for ZnO film deposition with Zn-O structure was synthesized by the reaction of DEZ and water in some ether solvents. Non-doped ZnO films on a glass substrate have been successfully grown by conventional spin coating using the novel precursor solution. Strong and clear X-ray diffraction peaks of ZnO film grown by the novel precursor solution are observed compared to those grown by simple DEZ solution. The samples have an optical transmittance of more than 85%, and a porous surface determined from optical transmittance and scanning electron microscopy, respectively. We also apply for perovskite based solar cell instead of TiO₂.
References:
[1] K. Yoshino, Y. Takemoto, M. Oshima, K. Toyota, K. Inaba, K. Haga, and K. Tokudome,
Jpn. J. Appl. Phys. 50 (2011) 040207.

We report the electrical properties of uniform and large area bilayer graphene chemically doped with tetracyanoethylene (TCNE) molecules intercalated between the graphene sheets. The bilayer graphene sheets were synthesized by hot filament chemical vapor deposition (HFCVD) and they are misoriented as indicated by the symmetric and relatively narrow 2D band and confirmed by the Moore pattern in the high resolution transmission electron microscopy (HRTEM) images. The fast Fourier transform (FFT) of the HRTEM image shows that the graphene layers are rotated by 30 degrees with respect to each other. The high electron affinity of TCNE molecules produces p-type doping and increases the charge carrier concentration up to 2 x 10¹³ cm^-2 per stacking for a doping level of about one monolayer of TCNE. The sheet resistance decreased from around 2800 #8486;/sqr for undoped stacking to 1360 #8486;/sqr for the stacking with the largest possible doping. The optical transparency of the stacked layers remains very high upon doping with TCNE molecules. The results show that chemical doping of graphene with strong electron acceptor molecules can lead to highly conductive transparent electrodes.

Platinum has been demonstrated as a potential counter electrode material candidate for dye sensitized solar cells (DSSCs). However, the production cost limits the application of platinum in energy harvesting technologies. Carbon allotropes such as graphene and carbon nanotubes have shown immense potential in photovoltaic technology due to their stability, low cost and simple synthesis techniques. We demonstrate DSSCs employing vacuum filtration processed self-assembled liquid crystal carbon nanotubes (LC-CNTs) as counter electrodes. The photovoltaic performance is significantly affected by the concentration of CNT used in the counter electrodes. The highest photo-conversion efficiency is observed to be ~7% for 60 µg of CNT. The photo-catalytic activity and the charge transport mechanism are greatly influenced by the concentration of CNT in the electrode. We further investigate the effect of annealing temperature on the optical, electrical properties of LC-CNT and the overall photovoltaic performance of the DSSCs. Compared to as prepared LC-CNT films, the samples annealed at 250^oC and then 400^oC show significant changes in electrical and optical characteristics. The nematic domain size is expected to increase as a result of annealing temperature compared to pristine LC-CNT samples confirmed by the changes in optical transmittance and electrical resistance. This paper will further present the effect of LC-CNT annealing temperature on charge transport mechanisms studied by cyclic voltammetry and temperature dependent charge transport analysis.

We present our understanding, with CH₃NH₃PbCl₃, a 3D semiconducting material as a model system, on the 2D van der Waals growth and growth kinetics of 3D parent materials. We show the successful synthesis of ultrathin (sub-10 nm), large scale (a few tens of micrometers in lateral dimension) single crystalline 2D perovskite thin films on layered muscovite mica substrate by van der Waals epitaxy. Classical nucleation and growth model explaining conventional epitaxy has been modified to interpret the unique 2D results in perovskite under van der Waals mechanism. The generalization of our van der Waals nucleation and growth model shows that a 3D crystal with low cohesive energy tends to favor the 2D growth while the one with strong cohesive energy has less kinetic window for 2D growth. It also demonstrates that ionic and metallic crystals with delocalized bonding characters have more tendency to form ultrathin structures compared to covalent materials with strong localized bonding characters. Finally, with static and kinetic Monte Carlo simulations, we show that the factual 2D morphology in perovskite material precisely manifests the kinetic competition between van der Waals diffusivity and thermodynamic driving force, a unique phenomenon to van der Waals growth, which suggests a fundamental limit on the morphology stability of the 2D form of a 3D material.

Within the last few years organic-inorganic halide perovskites have risen to become a very promising PV material, captivating he research community. In the most efficient devices, which now exceed 20% solar to electrical power conversion efficiency, the perovskite is present as a solid absorber layer sandwiched between n and p-type charge collection contacts. Increasing importance of improving solar cell operation is reliant upon understanding and controlling thin-film crystallisation and controlling the nature of the p and n-type contacts. In addition, understanding and enhancing long term stability of the materials and devices if a key driver. Here I will present our work on developing thin film perovskite solar cells, and specifically highlight recent advances in understanding the thin film crystallisation and the mechanism driving hysteresis, which is often observed in the current voltage curves. I will further present our recent work on all inorganic perovskites, which may offer vastly enhanced stability at elevated temperatures.

Perovskite based solar cells, mostly employ solution processed perovskite layers. Evaporated methylammonium lead iodide perovskite layers have also been reported and been employed in solar cells.^1-3 Our group has developed several perovskite based solar cells, using vacuum based perovskite preparation methods. These metal oxide free p-i-n type perovskite cells exhibit high power-conversion efficiencies.^4-10 We have extended this work to fully evaporated perovskite devices as well as devices employing a multi-layer stack of different perovskite layers prepared using vacuum based processes. These multi-layer devices give rise to enhanced open circuit voltages, above 1.1 V while still having high current densities.
(1) Era, M.; Hattori, T.; Taira, T.; Tsutsui, T. Chem. Mater. 1997, 9, 8.
(2) Mitzi, D. B. Chem. Mater. 2001, 13, 3283.
(3) Liu, M.; Johnston, M. B.; Snaith, H. J. Nature 2013, 501, 395.
(4) Malinkiewicz, O.; Aswani, Y.; Lee, Y. H.; Minguez Espallargas, M.; Graetzel, M.; Nazeeruddin, M. K.; Bolink, H. J. Nature Photonics 2014, 8, 128.
(5) Wetzelaer, G.-J. A. H.; Scheepers, M.; Sempere, A. M.; Momblona, C.; Ávila, J.; Bolink, H. J. Adv. Mater.2015, 27, 1837.
(6) Roldan-Carmona, C.; Malinkiewicz, O.; Soriano, A.; Minguez Espallargas, G.; Garcia, A.; Reinecke, P.; Kroyer, T.; Dar, M. I.; Nazeeruddin, M. K.; Bolink, H. J. Energy & Environ. Sci. 2014, 7, 994.
(7) Roldan-Carmona, C.; Malinkiewicz, O.; Betancur, R.; Longo, G.; Momblona, C.; Jaramillo, F.; Camacho, L.; Bolink, H. J. Energy & Environ. Sci. 2014, 7, 2968.
(8) Momblona, C.; Malinkiewicz, O.; Roldán-Carmona, C.; Soriano, A.; Gil-Escrig, L.; Bandiello, E.; Scheepers, M.; Edri, E.; Bolink, H. J. APL Materials 2014, 2, 081504.
(9) Malinkiewicz, O.; Roldán-Carmona, C.; Soriano, A.; Bandiello, E.; Camacho, L.; Nazeeruddin, M. K.; Bolink, H. J. Adv. Ener. Mater. 2014, 1400345.
(10) Longo, G.; Gil-Escrig, L.; Degen, M. J.; Sessolo, M.; Bolink, H. J. Chem. Commun.2015, 51, 7376.

Organic-inorganic metal halide perovskite solar cells (PSCs) have reached extremely high power conversion efficiency (PCE) of 18~20%. However, all of the high PCEs were achieved by tiny cells (working area << 1 cm²) with very short lifetime. As well known, the measurement errors will be enlarged as the active cell area gets smaller. Moreover, the poor stability and large current-voltage scanning hysteresis of PSCs may further disturb the reliable characterization of their photovoltaic performance. To steadily advance R&D of perovskite solar cells based on reliable data, it is an urgent issue to adopt the practice of measuring conversion efficiencies by international standard testing organizations.
In this presentation, I will demonstrate our recent achievement on PSCs that a certified efficiency of 15% was obtained from devices with working area over 1 cm². To realize this achievement, we put our focus on two aspects: one is improving the reproducibility of device performance through controlling the morphology and uniformity of photoactive and charge transport layers in the solar cells; the other is increasing the long-term stability of device by developing new charge transport materials with low hygroscopicity and high carrier mobility. For example, we developed a new method for fabricating uniform CH₃NH₃PbI₃ layer with using uncrystallized PbI₂ precursor films, which overcame the problem of incomplete conversion and uncontrolled particle sizes of perovskite, thus greatly increasing the film uniformity and reproducibility. We developed a dopant-free carrier transport material that has low hygroscopicity and high carrier mobility, with which we successfully improved the stability of the complete devices. Efforts have also been made to improve the blocking effect of electron and hole interfacial passivation layers in large area devices. As a result, we achieved 15% conversion efficiency in perovskite solar cells (> 1 cm²) for the first time in the world through the official evaluation by the international standard testing organization (Calibration, Standards and Measurement Team at the Research Center for Photovoltaics, National Institute of Advanced Industrial Science and Technology (AIST))
Reference
1. L. Y. Han, et al., Highly compact TiO₂ layer for efficient hole-blocking in perovskite solar cells. Appl. Phys. Express, 2014, 7, 052301.
2. L. Y. Han, et al., Retarding the crystallization of PbI₂ for highly reproducible planar-structured perovskite solar cells via sequential deposition. Energy Environ. Sci., 2014, 7, 2934-2938.
3. L. Y. Han, et al., A dopant-free hole-transporting material for efficient and stable perovskite solar cells. Energy Environ. Sci., 2014, 7, 2963-2967.
4. L. Y. Han, et al., Hybrid interfacial layer leads to solid performance improvement of inverted perovskite solar cells. Energy Environ. Sci., 2015, 8, 629-640.

Hybrid perovskite photovoltaics have reached over 20% efficiency in only a few years of optimization, which is ascribed to enormous mobilities of electrons and holes. The charge mobility, and subsequently photocurrent generation and open circuit voltage, is influenced by the sample&’s crystal structure but believed to be homogenous within the active layer. We employed PinPoint photoconductive Atomic Force Microscopy to map the local dark current, short circuit current, and open circuit voltage with sub 10 nm spatial resolution in a MAPbI_3-xCl_x solar cell. We discover strong intra- and intergrain anisotropy in the mobility, short circuit current, and open circuit voltage and that spatially the heterogeneity in these properties are correlated. Density functional theory band structure calculations reveal anisotropic effective mass in perovskite crystals that can explain the intragrain heterogeneity in mobility. This anisotropy suggests that the local performance depends on the grains&’ crystal orientation - a key insight to enhance device performance of both, polycrystalline and single crystal photovoltaic devices.

Moisture, in the form of ambient humidity, has a significant impact on methylammonium lead halide perovskite films. In particular, due to the hygroscopic nature of the methylammonium component, moisture plays a significant role during film formation. This issue has so far not been well understood, and neither has the impact of moisture on the physical properties of resultant films. Herein we carry out a comprehensive and well-controlled study of the effect of moisture exposure on methylammonium lead halide perovskite film formation and properties. We find that films formed in higher humidity atmospheres have a less continuous morphology but significantly improved photoluminescence, and that film formation is faster. In photovoltaic devices, we find that exposure to moisture, either in the precursor solution or in the atmosphere during formation, results in significantly improved open-circuit voltages and hence overall device performance. We then find that by post-treating films with moisture exposure, we can enhance photovoltaic performance and photoluminescence in a similar way. The enhanced photoluminescence and open-circuit voltage imply that the material quality is higher in films that have been exposed to moisture. We determine that this improvement stems from a reduction in trap density in the films, likely due to the partial solvation of the methylammonium component and ‘self-healing&’ of the perovskite lattice. This work highlights the importance of controlled moisture exposure when fabricating high-performance perovskite devices, and provides guidelines for the optimum environment for fabrication. Moreover, we note that often an unintentional water exposure is likely responsible for the high performance of solar cells produced in some laboratories, whereas careful synthesis and fabrication in a dry environment will lead to lower-performing devices.

We will review progress in solar cells based on solution-processed hybrid organic-inorganic materials such as colloidal quantum dots, perovskites, and combinations thereof.

Single junction solar cells fail to absorb a great portion of the light spectrum, which consists of photons with energy smaller than their bandgap. In this work, we explore the potential for colloidal quantum dots (CQDs) to augment the performance of silicon and perovskite solar cells. We identify through a detailed balance modeling, an optimum sub-1 ev bandgap for the CQDs to harness the transmitted light in the infrared. By using short bromothiol-ligands we were able to improve quantum dots passivation and obtain a strong hole transport. Also bromothiol ligands were shown to be critical to prevent dots agglomeration and retaining favourable injection into the TiO2 electrode. The results were confirmed using photoluminescence spectral and transient studies. The films based on these CQDs yield the highest AM1.5 power conversion efficiency yet reported in a solution-processed material having sub-1 ev bandgap.

One-dimensional and two-dimensional carbon materials, such as carbon nanotubes (CNTs), graphene and graphene nanoribbons (GNRs), show excellent optoelectronic properties; but they have not been made useful as solar energy conversion materials in photovoltaic devices. Challenges for implementing low dimensional nanoscale forms of carbon are in the synthesis of mono-disperse, well-defined species with proper energy level alignment. We herein introduce several atomically-defined GNRs synthesized via chemically bottom-up approaches as highly efficient electron acceptors. The GNRs are constructed by fusing perylene diimide (PDI) units together with a two-carbon bridge.^[1] They have relatively high electron mobilities, good electron-accepting ability, and LUMO levels similar to those of PC₆₁BM and PC₇₁BM.^[1]
We fabricated bulk heterojunction (BHJ) solar cells using GNRs as electron acceptors and commercially available polymers (PTB7 and PTB7-Th) as electron donors.^[2] By varying the lengths of the ribbons and controlling film morphology in BHJs, we achieved a power conversion efficiency > 8% that is on par with PTB7-Th:PC₇₁BM BHJs in a same device configuration. This value is a record high for non-fullerene BHJs.^[3] Femtosecond transient absorption spectroscopy revealed both electron and hole transfer processes at the donorminus;acceptor interfaces, indicating that charge carriers are created from photogenerated excitons in both GNR and polymer phases. This study brings together organic photovoltaics and carbon nanomaterials and describes a new motif for designing highly-efficient acceptors for organic solar cells.^[3]
[1] Zhong, Y. et al. Helical Ribbons for Molecular Electronics. J. Am. Chem. Soc. 136, 8122-8130 (2014).
[2] Zhong, Y. et al. Efficient Organic Solar Cells with Helical Perylene Diimide Electron Acceptors. J. Am. Chem. Soc. 136, 15215-15221 (2014).
[3] Zhong, Y. et al. Graphene Nanoribbons as Electron Acceptors in High Performance, Bulk Heterojunction Solar Cells. Submitted.

Organic dye-sensitized solar cells (DSSCs) with a new indoline-based organic molecule as a dye sensitizer and tetramethylpiperidinoxyl (TEMPO) as a mediator displayed high photovoltaic performance beyond 10% with high voltage of almost 1.0 V under 1 sun irradiation. Skeleton of the indoline dye was similar to the commercial one (D205: Mitsubishi Paper Mills) which was modified with long C₂₂H₄₅ chains using 1-docosanol extracted from “Peanut Oil”. The dye displayed a very high molecular absorption coefficient in visible range. The long C₂₂ alkyl chains favorably interacted with the TEMPO mediator, which was analyzed by quartz crystal microbalance and NMR spectroscopy. The quasi-solid-state solar cell using gel electrolyte yielded also high efficiency >10% without the loss of J_SC and FF, because of the large heterogeneous electron-transfer rate and electron self-exchange reaction rate for TEMPO even in the gel electrolyte. The DSSC displayed high voltage of 0.8 V even under weak light intensities of less than 100 µW/cm² such as a fluorescent lamp and a LED lamp. This result strongly indicated that the DSSC was suitable for supplying electric power to portable or small electronic devices. Indoor use of the high efficiency DSSC is now being tested toward practical applications of a power source for wireless sensor devices.

Dye-sensitized solar cells (DSSCs) have been considered as a promising device for light-energy harvesting due to their low fabrication cost and high conversion efficiency. However, the majority of the research has been focused on the front-side illuminated DSSCs that the light source is directed to glass/TiO₂. Only few researches have been focused on back-side illuminated (BSI) DSSCs which the light source is directed to Pt counter electrodes (CEs) in spite of their possible advantages such as mechanical hardness, flexibility, facile fabrication of nanostructure, and availability in high-temperature process. In particular, BSI-DSSCs is essential part for BSI-DSSCs/Li battery power pack which is proposed by W. Guo et al.. Yet, multiple tandem cells and complicated fabrication process are required to practically design the power pack to obtain necessary voltage. For efficient design for the power pack, the effort to enhance the performance of a single DSSC is necessary. In this research, we fabricate BSI-DSSCs based on TiO₂ nanotube arrays by anodization. The anatase TiO₂ nanotubes are physically evaluated in inner diameters, wall thicknesses, and tube lengths according to the applied voltages ranged 20 to 50 V. BSI-DSSCs with TiO₂ nanotube arrays under 50 V anodization condition are fabricated with Pt films on indium tin oxide (ITO) and fluorine doped indium tin oxide (FTO) glasses to investigate effect by transparency. The BSI-DSSCs with Pt/ITO glasses that possess ca. 70 % of transmittance in visible region exhibit favorable current density and conversion efficiency while the BSI-DSSCs with Pt/FTO glasses that possess > 70% of transmittance in visible region show an advantage in voltage. In the study of the electrochemical impedance spectroscopy under dark condition, it is observed charge transfer resistance (R_ct) at Dye/TiO₂ nanotube/electrolyte interface has influenced on open-circuit voltage (V_oc) of the DSSCs and maximum frequency shift to lower frequency (FTO: 4.329 Hz agrave; ITO: 3.275 Hz) is measured in the DSSCs with Pt/ITO glasses, indicating longer carrier life time of carriers. As an attempt to further enhancement of the performance, the concentration of tert-butyl pyridine (TBP) in electrolyte is varied from 0.5 M to 2M. As the concentration of TBP increase, R_ct is increased from 25.18 [Omega;middot;#13216;] to 44.25 [Omega;middot;#13216;] with enhanced V_oc. Moreover, the deterioration of the CE performance is observed with increasing the concentration of TBP, which results in decrease in DSSC performance with 2M TBP. 0.742 V of the most enhanced V_oc is measured with 2M TBP and the DSSCs with 1M TBP exhibits higher performance of 3.32 % in conversion efficiency.

Limit of light trapping in nanostructured medium could be signficantly higher than the conventional limit in bulk materials. We will discuss the theory and methods to explore more efficient light trapping for solar energy conversion.

Nowadays photovoltaic (PV) market is dominated by crystalline silicon wafer based solar cells with a cost-saving reduction of the wafer thickness reported every year. With the constant reduction of the silicon wafer thickness, PV industry is searching for new approaches which can provide at least as good light trapping as random pyramids, but without the removal of substantial amount of the material. Both requirements can be satisfied by radial junction solar cells based on controlled length silicon nanowire (SiNW) arrays. The solar cells presented in this work have been fabricated around crystalline SiNW arrays by two methods: a metal assisted chemical etching (MACE) of crystalline silicon material (top-down) and a plasma-enhanced vapor-liquid-solid (VLS) growth (bottom-up).
We have studied heterojunction with intrinsic thin layer (HIT) on c-Si NW arrays prepared by the top-down approach either by using standard MACE process alone or coupled with nanosphere lithography (NSL) to achieve ordered NW arrays of controlled diameters. Optical properties of NSL grown NW arrays have been studied and optimized using an in-house developed model based on rigorous coupled wave analysis (RCWA) and compared with experimental data. The c-Si NW surface passivation by hydrogenated amorphous Si (a-Si:H) has been optimized to maximize the performance of SiNW HIT devices, which has reached over 12 % efficiency for randomly organized arrays.
Radial junctions provide an attractive solution also for solar cells with ultra-thin absorbers, as they combine advantages of a very good light trapping with an excellent carrier collection. Such high efficiency solar cells can be prepared by plasma-enhanced VLS growth in plasma enhanced chemical vapor deposition (PECVD) reactor using low-melting point metals, thus allowing the fabrication at a low temperature (below 400 #8304;C) during a single pump down process. In this case, we have studied PIN structures formed on p-type randomly oriented crystalline silicon nanowire arrays, with 100 nm thick a-Si:H absorber layer plus a 10 nm thick n-type a-Si:H, microcrystalline Si or their oxides, respectively. Our PIN radial junction solar cells with a-Si:H absorber have reached energy conversion efficiencies over 9 % for 0.126 cm² device areas.
We will demonstrate the efficiency of top-down and bottom-up approaches not only by measured J-V characteristics, but also by their measured external quantum efficiency (EQE) and absorption. The impact of the NW density on the device performance will be shown: theoretically, for different materials and organizations, and experimentally on selected NW arrays. Furthermore, we will discuss the optimization of the solar cell performance in terms of the passivation quality and optical transparency of doped hydrogenated microcrystalline silicon oxides. Moreover, we will demonstrate low degradation rate and high incidence angle acceptance of solar cells fabricated on the top of VLS grown nanowires

Due to the intermittent nature of renewable energy sources such as solar and wind energy, systems designed to convert grid-scale power from these resources must be coupled with energy storage for off-grid operation. Traditional routes to coupling energy storage with energy conversion systems consists of an energy storage device situated external to the energy conversion system. However, direct integration of both systems offers numerous advantages, such as multipurposing of materials and storage directly at the point of generation to mitigate DC-AC inversion. Our recent efforts have demonstrated the ability to utilize the excess silicon material in a polycrystalline solar cell for supercapacitor energy storage applications. However, a key challenge that remains is to develop chemical passivation routes at low temperatures that can preserve the solar cell P-N junction and yield a passive silicon energy storage interface with a solid-state electrolyte. We address this issue by exploring multiple routes for low-temperature passivation, specifically including (i) wet chemical passivation of porous silicon using thin film metal coatings, and (ii) electrochemical passivation of porous silicon using conductive polymer networks. Our efforts demonstrate these passivation routes as being promising for the development of fully integrated energy storage and conversion systems applicable toward a range of applications ranging from grid-scale storage to remote energy transfer applications.
References:
A. S. Westover et al., "Direct integration of a supercapacitor into the backside of a silicon photovoltaic devices." Applied Physics Letters 104, 213905 (2014)

Vertically grown nanocomposite film creates high expectations for functional device applications owing to an enhanced effect in hetero junction interfaces. We demonstrate self-assembled ZnO-NiO nanosystem of three dimensional structure as a new proposal for ZnO-based nanocomposite. Phase separated ZnO-NiO films with pillar/matrix structure were grown epitaxially on SrTiO₃(111) single crystal substrate by pulsed laser deposition under the process condition above 820°C in substrate temperature and above 10Pa in oxygen gas pressure during deposition. A combination of in-plane and out-of-plane XRD profiles reveals an epitaxial relationship between the film and the substrate as ZnO(001)[100],[110]//NiO(111)[110]//SrTiO₃(111)[110]. Twelve-fold symmetry peaks in the in-plane XRD profile suggest that two domains exist in the ZnO phase with 30° rotating each other. Cross-sectional transmission electron microscope (TEM) images and compositional mapping by energy dispersive X-ray spectrometry (EDS) results show clear phase separation composed of NiO pillars of 20-50nm in diameter embedded in ZnO matrix. A feature to be emphasized in high-resolution TEM image is that the vertical pillar/matrix interface atomically junctures without any amorphous and low-crystallized area. The phase separated epitaxial growth of pillar/matrix is designed based on a lattice matching of film/substrate materials and growth modes as follows: a small mismatch of 1.8% between wurtzite-type ZnO (001) and 30° rotated SrTiO₃(111) enables a layer-by-layer growth by Frank van der Merwe (FM) mode, while relatively large mismatch of 7.2% between rocksalt-type NiO(111) and SrTiO₃(111) brings an island growth by Volmer Weber (VW) mode. Also, a thermal equilibrium phase diagram of ZnO-NiO indicates that a phase separation stably occurs in high temperature region.
In conclusion, self-assembled three dimensional structure of ZnO-NiO has been successfully grown epitaxially on SrTiO₃(111) substrate. Electrical properties of this nanosystem will be elucidated in the near future. Forthcoming issues are to grow pillars having a diameter small enough to generate quantum effect and extending through the entire film thickness up to several hundred nanometers. ZnO and NiO are well known as n-type and p-type conductive material, respectively, so we expect an improvement in conversion efficiency of solar cell devices or in sensitivity of sensor applications due to geometric and quantum effect induced by such three dimensional pn junctions.

III-V compound multi-junction (MJ) solar cells have high efficiency potential of more than 50% due to wide photo response, while limiting efficiencies of single-junction solar cells are 31-32%. In order to realize high efficiency III-V compound MJ solar cells, understanding and controlling imperfections (defects) are very important. This paper presents fundamentals of defects and defect management for III-V MJ, space and concentrator solar cells based on our studies..
Loss mechanisms to be solved for realizing high-efficiency III-V MJ solar cells are 1) bulk recombination loss, 2) surface recombination loss, 3) interface recombination loss, 4) voltage loss, 5) fill factor loss, 6) optical loss, 7) insufficient photon energy loss. The origins of those losses and key technologies for reducing those losses are presented. Key technologies for reducing the above losses are high quality epitaxial growth, reduction in density of defects, optimization of carrier concentration in base and emitter layers, double-hetero (DH) junction structure, lattice-matching of active layers and substrate, surface and interface passivation, reduction in series resistance and leakage current, anti-reflection coating, photon recycling and so forth. In order to solve energy loss problems due to thermalization or lack of absorption, multi-junction (tandem) structure is essential.
The key issues for realizing super-high-efficiency MJ solar cells are 1) sub cell material selection, 2) tunnel junction for sub cell interconnection, 3) lattice-matching, 4) carrier confinement, 5) photon confinement, 6) anti-reflection in wide wavelength region and so forth. Especially, DH structure have been found to effectively prevent from impurity diffusion from tunnel junction and high tunnel peak current density has been obtained by the authors.
InP and InP-related materials such as InGaP and InGaAsP have been found to be superior radiation resistant compared to GaAs and Si and have unique light-illumination defect annealing phenomena by the authors. As a result, InGaP-based MJ solar cells have been industrialized for space use even in Japan. Origins of radiation-induced defects in InP and InP-related materials, and radiation damages to solar cells are discussed.
For concentrator applications by using MJ cells, the cell contact grid structure should be designed in order to reduce the energy loss due to series resistance, and tunnel junction with high tunnel peak current density is necessary.
The conversion efficiency of inverted epitaxially grown InGaP/GaAs/InGaAs triple-junction solar cells has been improved to 37.9% (1-sun, AM1.5G) and 44.4% (250- 300 suns) as a result of proposing double-hetero structure wide-band-gap tunnel junctions, and inverted epitaxial growth. Concentrator photovoltaics (PV) is expected to contribute as one of major PV as well as the first crystalline Si PV and the second thin-film PV.

Widespread use of high efficiency III-V solar cells and intermediate efficiency silicon solar cells is limited due to expensive and scale-confined bottom templates (e.g. single crystal germanium, GaAs and silicon wafers). In order to reduce the cost, we have developed high textured, epitaxial Ge and Si films on inexpensive, flexible and polycrystalline substrates by roll-to-roll continuous deposition. Our approach employs the ion beam assisted deposition technology to achieve biaxial texture in MgO templates deposited on the inexpensive substrate. In order to realize the epitaxial growth of Ge and Si, and intermediate layer of CeO₂/LMO was found to be required between the MgO template and Ge thin film. High quality Ge and Si films are deposited by medium frequency magnetron sputtering or plasma enhanced chemical vapor deposition (PECVD). Doping concentration can be controlled by adjusting the PH₃ and B₂H₆ flow rate in the PECVD. X-ray diffraction results confirmed the single-crystalline-like Ge film growth with a sharp out-of-plane texture of 0.9° and an in-plane texture spread 3.3°. Out-of-plane and in-plane texture spreads of 1.3° and 1.7° were obtained in the epitaxial Si thin films. Full width at the half maximum values of the relevant peaks in the first-order Raman spectrum also indicate the highly crystallinity of Ge (3.5 cm^-1) and Si (5.2 cm^-1) thin films, comparable with Ge (3.4 cm^-1) and Si (4.6 cm^-1) wafers. Hall measurement shows high carrier mobility values of 903 cm²/Vmiddot;s and 109 cm²/Vmiddot;s are measured for Ge and Si thin films respectively. In a further step, n- or p-type single-crystalline GaAs thin films are heteroepitaxially grown on these flexible Ge templates via roll-to-roll metal organic chemical vapor deposition. Fabrication of multi-junction solar cells based on these high quality biaxially-textured GaAs and Ge thin films is underway.
This work was partly funded by the U.S. Department of Energy SunShot Initiative.

Tandem architectures are one of the most promising concepts to yield competitive power conversion efficiencies in organic photovoltaic devices. The deposition of multi-layers for advanced device architectures will allow them to enter the market in a wide field of applications. Besides the commonly studied tandem solar cells that comprise two absorber layers with complementary spectral response, the so-called homo-tandem approach also leads to significantly enhanced power conversion efficiencies for a large variety of absorber materials with moderate charge carrier transport properties. However, the detailed study of the subcells&’ optoelectronic properties, the current matching and the device optimization become much more challenging. In particular, the investigation of homo-tandem solar cells requires advanced external quantum efficiency (EQE) measurements, in contrast to common hetero-tandem solar cells that rely on a certain spectral separation. Here, we study 3-terminal devices that were deliberately designed to match the optoelectronic properties of highly efficient 2-terminal tandem devices. Thereby, we overcome the drawbacks of previously published methods that alter the optical properties by introducing optically active electrodes. In this work, the additional intermediate contact integrated in the recombination zone based on PEDOT:PSS and zinc oxide does not change the optical field distribution. With the 3-terminal devices showing equal optoelectronic properties, detailed EQE measurements lead to deep insight into interdependencies of the subcells&’ and the tandem solar cell&’s EQEs and short circuit current densities. Most importantly we found quasi-complementary spectral response for both subcells which we exploited for measuring the EQE of the subcells without additional intermediate contact directly at the conventional 2-terminal tandem devices. Monochromatic bias wavelengths and bias voltages were determined from 3-terminal devices. Furthermore, we derived a simulative method to estimate the required monochromatic bias light intensity for saturating one subcell of a homo-tandem solar cell although subcell competition for light harvesting dominates the whole spectral range. Altogether, this allowed for transferring the conventional EQE method to homo-tandem solar cells which will allow for proper characterization of organic homo-tandem solar cells in the future.

CuGaSe₂ (CGSE) is an interesting photovoltaik material with respect to tandem devices on CuInSe₂ based systems. CGSE is well lattice matched to GaAs. By MBE we prepared CGSE films with varying Cu/Ga ratios. A 4x1 and 4x2 reconstruction was determined by LEED, and the electronic structure with respect to the Cu/Ga ratio was investigated by XPS and UPS. All preparation and analysis was performed in an integrated UHV/XUHV system. Beneath conventional MBE high quality films could be prepared by using instead of an external Ga source the substrate as the Ga source by preparing the films at 550° to 600° C. In particular this behaviour gives a clue to the understanding of the excellent epitaxial growth of CuInSe₂ on GaAs which was investigated in the past by our group. As an lattice matched film of CuGaSe₂ at the GaAs interface will form. With increasing thickness a graded film from CuGaSe₂ to CuInSe₂ will grow bridging the larger lattice mismatch.

Tandem solar cells overcome conversion efficiency limitations of conventional solar cells by using more than one junction. Each of the junctions utilizes that part of the solar spectrum it can use most efficiently, thus reducing thermalization losses. The boost in energy yield of a tandem PV module comes at production costs that are increased by the additional process steps required. Economically, a tandem PV module make sense if it can generate power at a lower cost than a module made of either of the sub-cells. In this contribution we present a parametric study and define conditions under which tandem PV modules are economically sensible.
The minimum sustainable price (MSP) defines the price for which a manufacturer has to sell his product in order to continue production sustainably. MSP conditions for a manufacturing process are met if the internal rate of return (IRR) equals the weighted average cost of capital (WACC). We have developed bottom up MSP models for different solar cell materials, based on the corresponding module production processes. We use these models to project how a production process and cost-structure for a flat-plate tandem PV module look like. We analyze different components and define cost relations of this process. For a tandem PV module, a key cost relation is the one between the cost of the solar cell and the cost to make the module. Tandem solar cells become economically the more attractive the lower this relation is. Similar considerations also hold for PV systems. We observe that the benefits of tandem solar cells are more emphasized on the system level.
In a second step, we combine the cost relation analysis with a tandem solar cell efficiency calculator. Using this approach we show that several key conditions need to be met for tandem PV modules to be economically attractive: The sub-cells should have comparable single-junction efficiencies, they should reach similar $/W values and they should form ideal band-gap combinations. Ideal bandgap combinations are formed, if the difference between single junction solar cell efficiency limit and tandem solar cell efficiency limit as a function of bandgap is maximized.
In summary, in this contribution economic aspects of tandem PV modules are studied. We define cost relations based on bottom-up cost models for different tandem solar cells. We investigate under which conditions tandem solar cells are economically superior to single junction PV modules made of either of the sub-cells. By adding efficiency calculations we find that the single materials used to form a tandem solar cell should have similar $/W values and should form optimum band-gap combinations.

Polycrystalline silicon thin-film is an ideal candidate for application in low cost silicon based solar cells, since polycrystalline silicon has combination of crystalline structure as well as low cost fabrication technique. The main problems of fabricating polycrystalline silicon thin-film solar cells having conversion efficiency higher than 10% are low carrier mobility (mu;) and low lifetime (tau;) of photogenerated carriers because of huge recombination at grain boundary (GB). Here, we shall talk about two inventions that raise hope for low cost crystalline silicon thin-film solar cells on glass substrate. In first case, we succeeded to grow polycrystalline silicon thin-film at 200°C lower temperature than normally reported (450°C) in literature. The polycrystalline silicon thin-films were deposited by hot wire chemical vapor deposition (hot-wire CVD) technique. We found advanced form of polycrystalline silicon thin-film where grains are distributed not randomly but along crystallographic plane {(111), (311), etc.}. The grain size ~ 1 to 2 mu;m estimated by Scanning Electron Microscopy (SEM). The Raman shift at 517 cm^-1 for these films, indirectly confirms the grain size. Atomic Force Microscopy (AFM) reveals the exotic structure at the film surface, i.e. we found surface texturing has combination of six-fold and five-fold symmetry. In other words, we found Penrose tiling at the film surface, we got quasicrystalline silicon thin-film. Authors will discuss the quasicrsytalline structure of silicon thin-film deposited at 250°C, in details including quasi-unit cell concept, how it is sensitive to atomic hydrogen flux during film growth. Secondary Ion Mass Spectroscopy (SIMS) analysis reveals that the advanced polycrystalline silicon, we developed contains hydrogen content ~ 10²¹ cm^-3 ( i.e. 10at%). It is reported in the literature that hydrogen content in fully polycrystalline silicon is 1 to 2 at%. Higher hydrogen content will act as a defect passivation at the grain boundary (GB), carrier lifetime (t) and solar cells performance is very sensitive to point defects located at grain boundary (GB). Therefore, we found hydrogenated polycrystalline silicon thin-film for solar cells application. We also developed partially ordered polycrystalline silicon thin-film at 450°C substrate temperature by hot-wire CVD technique. The hydrogen content (FTIR) in these films is in the range 1 to 2 at%. We observed dopant atom boron (B) incorporation profoundly influences the structure, we found very ordered (partially ordered to fully ordered) structure, we got nearly (220) oriented epitaxial-like silicon thin-film. We got carrier mobility (mu;) ~ 10 to 20 cm²/V.s for undoped films. We developed hydrogen passivation scheme by hot-wire CVD, where we observed carrier mobility is improved to 40 to 50 cm²/V.s. The dark conductivity and Fermi level for these films are 10^-6 S/cm and 0.45 ± 0.05 eV respectively.

The hybrid halide perovskites have emerged as a breakthrough in the field of solar energy, demonstrating a viable high performance device platform due to a wide flexibility processing conditions and device architectures. Consequently these materials have a great potential to lead to cost-effective solar energy production. We have performed a comprehensive structural study, to elucidate chlorine and substrate role on the stability of perovskite films for solar cell applications. Different substrates are used to deposit CH₃NH₃PbI_3-xCl_x with different concentration of chlorine. Using grazing incidence X-ray diffraction (GIXRD) measurements, we found that the degradation rate of the perovskite that decomposes in PbI₂, depends on the used deposition substrate as well the concentration of Cl. Through first principles calculations, we propose a mechanism on how chlorine affects the perovskite film properties. Furthermore, the power conversion efficiency of the fabricated perovskite planar heterojunction solar cells is investigated and depends on the Cl content in the films. Highest device performance with an efficiency of 15.7% is obtained at the cost of stability. We will show how these findings could lead to a much more controllable processability of the perovskites, and offer a route to controlling the degradation rate and a more favorable way of producing stable hybrid perovskite films and devices.

Kesterite solar cells based on Cu₂ZnSn(S,Se)₄ absorbers have attracted much interest due to the material's relatively low toxicity and earth abundant elemental composition. The maximum efficiency reached by CZT(S,Se) solar cells is 12.6%, but these devices continue to exhibit a large voltage deficit compared to theoretical limits. A main component of these limitations can be attributed to bulk defects and band tails, which despite recent efforts in increasing cation ordering, may have limited potential improvements due to thermodynamic boundaries. An alternative avenue for driving up V_OC - in spite of bulk properties - is to use electrostatic fields via a high work function back contact to increase separation of electrons and holes, limit back contact recombination, and drive collection of carriers.
We use a facile exfoliation method to remove the active device from the Mo/glass substrate after growth, thereby enabling the deposition of a high work function back contact on the resulting superstrate solar cell structure. The process preserves an optimized hydrazine solution growth procedure responsible for past champion devices. In addition, the exfoliation technique enables the direct characterization and identification of significant recombination at the back half of the absorber that can harm V_OC. This recombination was reduced through chemical etching and an optimized oxidative treatment to passivate the back of the exfoliated device. Carrier lifetimes increased when measured from the back of the film, as well as the front, and translated to improved V_OC's by up to 30 mV relative to the pre-exfoliated device. Scanning electron microscopy, X-ray photoelectron spectroscopy, nano-Auger, and photoluminescence imaging were used to characterize the exfoliated film in an effort to understand the high carrier recombination at the back and help elucidate the mechanisms for improved passivation. The benefit of a high work function back contact is demonstrated, and the key role of absorber and MoO₃ thickness are optimized, while minimizing effects on fill factor.

We demonstrate a novel micro-grid top electrode for highly efficient radial-junction Si microwire solar cells. The micro-grid electrode minimizes optical and electrical losses, thus ensuring proper function of the shallow (sheet resistance of ~100 #8486;/sq) junction emitter. This leads to effective collection of the photocarriers from the shallow junction emitter through the top electrode without severe Auger/surface recombination, improving the overall power conversion efficiency of the Si microwire solar cell. With an optimized micro-grid structure, our 1 cm² microwire solar cells show a conversion efficiency of up to 16.5%, with an open-circuit voltage of 565.2 mV and a short-circuit current density of 35.9 mA/cm²; this conversion efficiency is 72% higher than that of solar cells with an edge electrode (9.6%). Further, an approximately 1-mu;m-thick Ni electrode that was formed by electroplating considerably reduced the metal and contact resistances, which reproducibly yielded a fill factor of over 80% (max 81.2%). Thus, the use of a novel micro-grid to construct an ideal metal/emitter interface presents a unique opportunity to develop highly efficient microwire solar cells.

There is an urgent need for efficient photovoltaics based on non-toxic, Earth-abundant materials that can be processed by low-cost, high-throughput methods. Historically, the cycle of learning for discovering new materials has been slow, limited by the need to fabricate and optimize devices. Recently, we have introduced a computational screening approach to inform and streamline experimental identification of defect-tolerant photovoltaic absorbers.^[1]
A critical component of this process is the design of a growth system that enables rapid deposition, but also allows films with high structural quality and phase purity to be synthesized. We chose physical vapor transport because it is flexible, allowing a wide variety of materials to be grown, while also being a system that can be maintained with high cleanliness, which is critical for maintaining low impurity contents. To control phase purity and structural-defect densities, important tunable deposition parameters are the spatial temperature profile, heating and cooling rate, and the system pressure. To avoid false negatives (i.e., low performance as a result of our fabrication process and not inherent to the material) during our search, we target phase-pure films with parts-per-million purity, low intragranular dislocation density, and grain diameter at least on the order of the film thickness.
In this work, we detail our development of two systems for physical vapor transport growth. The first design consists of a single-zone tube furnace that deposits a combinatorial library over the thermal gradient into the cold zone, allowing quick exploration of a range of substrate temperatures. In the second design, we reduce the tube furnace&’s thermal inertia to improve control of the cooling rate. This design consists of two hot zones, where the temperature of the source and substrate can be independently controlled. Using bismuth triiodide (BiI₃) as a test material, we compare the thickness, morphology, and crystalline phase of films produced using the two systems. Reliably and efficiently optimizing the structural parameters in a novel semiconductor will be critical for screening these materials for use in photovoltaic devices as well as potentially other optoelectronic applications.
[1] R. E. Brandt, et al., MRS Communications, 1-12 (2015). DOI: 10.1557/mrc.2015.26

Indium Gallium Nitride (InGaN) test solar cells were made using RF sputtering method. A photo voltage of 0.115V for Batch 1 and 90 to 240mV for batch 2 has been recorded. Photo current values are in micro amperes. Growth conditions, substrate cleaning methods are changed accordingly to make quality InGaN thin films to improve the photo current values and therein efficiency of solar cell. InGaN thin films with varying Indium and Gallium composition were deposited on aluminosilicate glass and silicon (111) substrates using RF magnetron sputtering method under substrate temperatures varying from room temperature to 450 ^oC. An Indium-Gallium (50-50 by atomic %) target was used with varying gas flow ratio to have different compositions of In and Ga. Refractive index values from 1.4 to 2.5, extinction coefficient values and film thickness were measured using ellipsometry measurements. Bandgap values ranging from 1.0 eV to 2.8 eV and absorption depth of thin films were calculated using UV spectrophotometer and tauc plots. Hall Effect measurements were performed to calculate electrical properties. Resistivity of thin films decreased by 1200 times, conductivity increased by 2000 times, mobility increased by 5 times compared to the films grown during fabrication of test solar cells. Crystalline nature of the samples were increased with reduced defects. Stress, strain and lattice constants were measured using XRD measurements. Indium and Gallium compositions were calculated during EDX analysis. These higher quality InGaN thin films were used in fabrication of new HIT solar cells. GaN and AlN buffer layers were used. A huge improvement in photocurrent and efficiency of solar cells both in laboratory conditions and under Sun were recorded.

Tandem solar cells have the potential to reach very high efficiencies by employing absorber materials of different band gaps. Layers of the materials are stacked such that each of them may utilize a different part of the solar spectrum. While this device architecture can in principle yield high efficiencies, losses associated with the stacking need to be minimized. Which losses occur depends on the exact tandem solar cell design. Most tandem solar cells today use an integrated two terminal configuration. This configuration requires current matching between the junctions and poses high demands on interface quality. An alternative is the four terminal (4T) configuration. In the 4T device design each absorber is contacted separately. The 4T design does not require current-matching, but it requires a more sophisticated contacting scheme. The need for a semi-transparent back contact on the top cell adds series resistance, and all top cell contacts add to the shading of the bottom cell. Moreover, each of the three relevant contacted surfaces - the front and back of the top cell and the front of the bottom cell- have different loss sensitivities to shading and series resistance. An optimum contacting scheme, therefore, needs to balance optical and resistance losses.
This study addresses the question of how to best contact a 4T tandem solar cell. Through optical and electrical simulations, we study different methods of contacting double-junction tandem solar cells with a standard 15.6 cm x 15.6 cm silicon wafer bottom cell. We explore the use of different contacts such as metal fingers, TCOs and silver nano-grids. In a first approach, we use an analytical model to assess the shading losses for the different contacts for different angles of light incidence. The different contacting schemes are investigated analytically. Using metal fingers, we find that aligning front and back contacts for the thin-film top cell is important. First simulation results show that the efficiency loss for a <30% GaAs on Si tandem solar cell with optimized metal finger contacts is in the range of 1.5%. This calculation uses a finger width of 50 um and pitch of 2 mm. For comparison, the loss for an unaligned front contacting scheme is 1.8%. We additionally explore design rules for the use of TCO contacts and if there are advantages in combining these technologies. In a later step, the developed design rules for contacts will be combined with available active area optical and device models into a full-area model of the tandem solar cell.

A perovskite solar cell based on Cadmium Sulfide (CdS) layer, as alternative to conventional TiO₂, processed with a scalable chemical deposition method at low-temperature is presented. CdS layer is grown on ITO and FTO substrates and thickness was controlled by deposition time and solution temperature. 10.7% power conversion efficiency was achieved using CH₃NH₂PbI₃ absorber and Spiro-OMeTAD as hole transport layer (HTL), and the device exhibited an open circuit voltage exceeding 1.0 eV. Different parameters were examined such CdS thickness and post-deposition annealing temperature, as well as exploring other organic and inorganic-based HTLs to replace Spiro-OMeTAD. Compared to the as-deposited CdS, devices fabricated with post-deposition annealing of CdS showed lower efficiency. Finally, CdCl₂ treatment and DI water cleaning treatment on the surface of as-deposited CdS film were performed to investigate their effects on the device performance. The morphological, optical and electrical properties of CdS film and devices were characterized.

Perovskite solar cells have made rapid progress in the past several years. Very recently, a power conversion efficiency (PCE) has reached more than 20%. On the other hand, the photovoltaic performance is strongly dependent upon fabrication conditions, which have critical impact on morphologies of perovskite films. Consequently, most of perovskite solar cells exhibit wide variations in the device performance even though they are fabricated by the same methods. This is probably because crystalline growth of perovskite materials is not well controlled for efficient photovoltaic performances. In this study, we have prepared dense CH₃NH₃PbI₃ layers by fast deposition-crystallization method and fabricated planar heterojunction solar cells with a layer structure of FTO/c-TiO₂/CH₃NH₃PbI₃/Spiro-OMeTAD/Au. The grain size of CH₃NH₃PbI₃ was varied from 100 to 500 nm by changing the stock solution concentrations. We discuss the relationship between the photovoltaic performance and the grain size of perovskite in terms of carrier recombination dynamics.

Control over the deposition of quantum dot (QD) on nanostructured semiconductors is very important for the photovoltaic performance of QD sensitized solar cells. The best control to attach the QDs to metal oxides in a specific manner is typically achieved through bifunctional molecular linkers (e.g. mercaptopropionic acid (MPA)). However, some materials such as ZnO, are not compatible with these molecules due to their pH sensitivity. We have developed new linkers, mercaptophosphonic acids of different lengths, which allow for efficient functionalization of ZnO nanowires and also mesoporous TiO₂ without damaging their surface. We found that the efficiency of the QD sensitized solar cells fabricated with such assemblies strongly depends on the linkers used e.g. best device efficiency was achieved with devices having shorter linker.
The efficient charge transfer at interface is one of main barrier which limits the efficiency of QD sensitized solar cells. We investigated the influence of linker lengths on the efficiency of charge transfer from QDs to nanostructured metal oxides (NMOs). For this purpose, CuInS₂ QDs deposited onto TiO₂ and ZnO nanowires substrates were assembled. Both steady state and time-resolved photoluminescence were used to study the charge transfer. First the Photoluminescence quantum yield (PLQY) was measured for all the systems studied. The relatively high PLQY of the QD film on the fused silica substarte (SiO₂) (6.7%) drops significantly to 0.1-0.8% for the systems deposited on TiO₂ and ZnO nanowires substrate. On SiO₂ substrates the excitons generated in the QDs upon the absorption of light can decay either radiatively or non-radiatively on surface or internal defects. In the assemblies with NMOs a third exciton de-excitation pathway appears: injection of the excited electron of CuInS₂ into the conduction band of the TiO₂ or ZnO. This electron transfer pathway competes with radiative recombination thus decreasing the overall PLQY of the QDs. The PLQY decrease in assemblies can therefore serve as an indirect indicator of electron transfer between excited CuInS₂ QDs and TiO₂ or ZnO.
In order to further quantify this charge injection efficiency in these assemblies, transient PL studies were performed using time-correlated single photon counting (TCSPC). We found that rate and efficiency of charge transfer from the QDs to the metal oxides strongly depends on length and nature of linker used e.g. the highest charge transfer rate (9.6x10⁷/s) was found for the system with shorter linker (MPA) and lowest charge transfer rate 0.3x10⁷/s for the systems with longer linker mercaptoundecylphosphonic acid (MUPA). Similarly the efficiency of charge transfer for the QDs with shorter linker MPA was almost double (80%) compared to QDs with longer linker MUPA (47%). Hence we believe that our studies offer new opportunities to overcome charge transfer limitations in QD sensitized solar cells

In this work, we investigated the design and photovoltaic efficiencies of Si/Ge tandem solar cell with backside reflector, which can be a promising structure for future terrestrial application of multi-junction solar cells. Comprehensive simulations have been done over a wide range of thickness using AM 1.5G standard sun irradiance, and multi-junction solar cell theories by means of MATLAB simulation tool. Our simulations predicted the highest efficiency of the design with total thickness of 27 µm is about 23 %, which shows a significant improvement in efficiency compared to previous works. In order to find the optimum thickness of the design, several combination of base and emitter of top and bottom cells have been simulated. It was found that the design could be grouped in three different regions considering total thickness of the device, current matching limit, and effectiveness of the backside reflector. The simulation showed that Si top cell thicker than 6 µm results in a very thick structure due to current matching limitation and hence backside reflector is not effective on efficiency improvement. In fact, Ge photo generated current, limits the efficiency of the devices with Si thicker than 6 µm. For structures with Si thickness from 4 to 6 µm, photo-generated currents of top and bottom cells can be perfectly matched. Furthermore, total thickness of the devices of this group is suitable for utilization of available exfoliation technologies, which enables practical implementation of backside reflector and thickness reduction. The structure of devices with Si top cell thinner than 4 µm is very thin and hence the backside reflector shows the most effectiveness on efficiency improvement.

Hybrid organic-inorganic perovskite structured materials (mostly CH₃NH₃PbI₃ and CH₃NH₃PbBr₃) have shown remarkable opto-electronic properties, including photovoltaic and luminescent ones. One of the favorable properties that attributed to these materials, and which places them at the front of materials research, is the fast crystallization with high crystalline quality of the reaction product from PbX₂ and CH₃NH₃X (X=I or Br) obtained at low temperatures. Likely, much of the surprising properties of these materials are related to its high crystalline quality. However, the ease of formation is, probably, also what makes these materials quite sensitive to changing environments and working conditions. Indeed, reproducing results based on what is written in the experimental section of a published report is not trivial task, with each group having its own procedure details that often turn out to be critical.
CH₃NH₃PbX₃ thin films are usually formed by one of two ways:
(1) By evaporating the solvent of a homogeneous mixture of the PbX₂ and alkyl ammonium halide salt (usually CH₃NH₃X);
(2) By performing a heterogeneous reaction where PbX₂ crystallites are exposed to a solution or the vapor of alkyl ammonium halide salt.
Due to several practical benefits of the heterogeneous reaction (e.g. reproducibility, better morphological control of thin films), and its close relation to the decomposition of the material due to the possible CH₃NH₃X exhaust, we investigate the transformation mechanism of PbI₂ (single crystals) to CH₃NH₃PbI₃ perovskite structure while being exposed to dissolved or vaporized CH₃NH₃I.
Based on high resolution morphological studies (Scanning Electron Microscope) the transformed structure implies on a strong correlation between the crystal lattices (PbI₂ and CH₃NH₃PbI₃), showing a highly ordered CH₃NH₃PbI₃ crystals grown from PbI₂ single crystals. Together with focused photo-luminescence studies, we suggest a two-step mechanism of the transformation mechanism, which implies that the commonly performed heterogeneous reactions (order of seconds to minutes) might be insufficient. We also suggest, based on the observed lattice spatial displacements between the crystals, a correct way of estimating the spatial expansion of PbI₂ to the CH₃NH₃PbI₃ perovskite structure.

Solution-processed perovskite solar cells (PSCs) received wide interest for realizing high efficiency. In this presentation, we report on the basic chemistry, optics and ferroelectrics of perovskite thin films and emphasize the importance of coordination formation process, optics management and ferroelectrics understanding on high efficiency of PSC. We demonstrate efficient planar perovskite solar cells with a PCE as high as 17% without appreciable hysteresis.

Colloidal quantum dots (CQDs) composed of compound semiconductors such as PbS, PbSe, and CdS have been gaining much attention as promising constituent materials for solar cells because the position of intense exciton absorption bands can be tuned by selecting suitable compositions of the semiconductors as well as quantum dot size so as to cover the entire solar spectral region. The possibility of multiple exciton generation (MEG) and the hot-carrier concept toward ultrahigh efficiency solar cells makes QDs even more attractive. In addition to the unique optical properties originating from quantum size effects, CQDs are compatible with solution processes for cost effective solar cell fabrication.
We have reported PbS CQD-based heterojunction solar cells by employing ZnO nanowire (NW) arrays not only to establish carrier pathways but also to increase light harvesting efficiency. The morphology of ZnO NW arrays was systematically investigated to achieve high light harvesting efficiency as well as efficient carrier collection. The solar cells with the PbS-QD/ZnO-NW bulk heterojunction (BHJ) structures made up of densely-grown thin ZnO NWs about 1200 nm long yielded a maximum EQE of approximately 60% in the near-IR region (@1020 nm) and over 80% in the visible region [J. Phys. Chem. Lett., 4, 2455 n(2013).], and achieved long-term stability under continuous light soaking [Physica Status Solidi (RRL) - Rapid Research Letters, 8, 961 (2014).]. Recently, we also studied photocurrent properties of the PbS-QD/ZnO-NW solar cells by focusing on the solar cell structure, and found out that, once good electron pathways are established in the PbS-QD/ZnO-NW BHJ structure, holes in the PbS region can diffuse a distance over 1000 nm. In addition to the morphology control of the BHJ, light management using plasmonic metal nanoparticles is one strategy to improve light harvesting efficiency. Ag nanocubes (NCs) with an average edge length of about 80 nm, which give rise to the light harvesting efficiency enhancement, were employed. Power conversion efficiency can be improved by approximately 35% after the optimization of position and concentration of Ag NCs in the BHJ [ACS Nano, 6, 4165 (2015).]. Our recent progress on PbS-QD/ZnO-NW BHJ solar cells will be presented.

Although most research on methylammonium lead iodide (MAPbI₃) perovskites has focused on producing polycrystalline films by simple and inexpensive solution processing methods, recent results by other workers have shown that large grained materials and single crystals have lower trap density, longer diffusion lengths, and enhanced photoluminescence. Here, we introduce a simple, low-temperature electrochemical/chemical route to deposit both textured and epitaxial MAPbI₃ films. The perovskite films are produced by chemical conversion of lead dioxide (PbO₂) films that have been electrodeposited as epitaxial films onto single-crystalline Au substrates. The out-of-plane orientation can be controlled by the conversion temperature: [001]-oriented films are produced at lower temperature, whereas [110]-oriented films are produced at higher temperatures. Both textured and epitaxial MAPbI₃ films have lower trap densities and higher photoluminescence intensities than polycrystalline films produced by spin coating.

Polycrystalline thin film technology has shown great promise for low cost, high efficiency photovoltaics. To further increase the power efficiency, a firm understanding of microstructural properties of the devices is required. In this work, we investigate the heterogeneities of CdTe solar cells using two local optical techniques. The first approach, uses a near-field scanning optical microscope (NSOM) probe and near-IR lasers (808, 850, 905, and 980 nm), for exciting the sample locally and to generate excess carriers close to CdTe band-gap energy (asymp; 1.45 eV). A sub-micron thickness lamella CdTe sample was prepared by focused ion beam (FIB) and cleaned by low energy ion milling as a post-FIB processing. The lamella was transferred to a high efficiency photodetector and then illuminated with an optical fiber probe (200 nm in diameter) mounted on a tuning fork. The transmitted power throughout the lamella sample was detected by photodetector. The absorption/transmission spatial maps were compared with the data obtained by a novel and complementary technique, photothermal induced resonance (PTIR). PTIR can be used to obtain absorption spectra and maps over a broader range of wavelengths. In PTIR, a wavelength tunable pulsed laser is used in combination with an atomic force microscope tip to measure the light absorbed in the sample by transducing locally the resulting CdTe lamella thermal expansion. The nano-scale high-resolution PTIR images show a variation of local absorption in a range of wavelengths near the CdTe bandgap energy, implying an inhomogeneous composition of the CdTe absorber layer corresponding to a local band gap variation of asymp; 1.45 eV ± 0.04 eV. The spatial variation of the CdTe band-gap is consistent with the transmitted/absorbed data maps obtained via near-filed excitation. The resolution and the sensitivity of these two approaches are compared. Initial experiments to analyze a local absorption associated with the deep levels in the CdTe band-gap will be presented.