Advanced materials and characterization techniques for solar cells III

Resume : Spectroscopic methods based on synchrotron radiation in the energy range of soft to tender x-rays yield valuable information on chemical, structural and electronic surface, interface and bulk properties of thin film solar cells and their components. This knowledge is necessary to understand and improve these photovoltaic devices, which consist of stacks of several different materials, in a systematic way. In this presentation we show some synchrotron based state-of-the-art methods for the analysis of polycrystalline thin film materials and entire photovoltaic devices and their limitations. Using results obtained with different end stations (CISSY, HIKE, Suricat, In-situ PVD) at the BESSY II synchrotron in Berlin, Germany, we show how synchrotron excited photoelectron spectroscopy (PES), x-ray absorption spectroscopy (XAS), x-ray diffraction (XRD) and soft X-ray emission spectroscopy (XES) have increased our knowledge of the properties of bulk phases, surfaces and “hidden” interfaces of these systems. Examples of recent results are the determination of composition-dependent conduction band shifts of absorber materials by XAS, depth-dependent composition analysis by high kinetic energy PES and identification of different phases by in-situ XRD measurements during the formation of compound semiconductor absorber layers from precursor stacks.

Resume : Organic-inorganic perovskites like CH3NH3PbI3 films represent a new paradigm for photovoltaics, which have the potential to overcome the performance limits of current technologies and achieve low cost and high versatility. Although the power conversion efficiency of the CH3NH3PbI3-based perovskite solar cells exceeded already 20%, a number of key issues must be solved before the wide-spread commercialization will be possible. A study of vacancy-mediated migration of I-, Pb2+ and CH3NH3- ions and their relative activation energies suggests, that the migration of halide vacancies to and from the interfaces in the solar cell during its operation is the main conduction mechanism [1,2]. In this work, we are presenting laboratory- and high resolution synchrotron-based spectroscopic studies of the CH3NH3PbI3 perovskite films. For example, the resonant X-ray photoelectron (resPES) study at the N1s absorption edge indicates that the contribution of nitrogen into the conduction mechanism of the CH3NH3PbI3 perovskite films should not be neglected. Moreover, different core level binding energies are observed in the X-ray photoelectron spectra (XPS) of the perovskite when different excitation energies are used. This result indicates that different potentials are at the surface and in the CH3NH3PbI3 film bulk. Detailed XPS and resPES data analysis will be presented. [1] Y. Zhang et al., Mater. Horiz. 2, (2015) 315. [2] J.M Azpiroz et al., Energy Environ. Sci. 8, (2015) 2118.

Resume : In recent years, organic-inorganic metal halide perovskite materials have evolved as highly efficient photovoltaic competitors to Silicon. Another option is to utilize the variety of bandgaps found in this class of materials and combine wide-bandgap thin-film photovoltaic perovskites in tandem with silicon solar cells. With careful opto-electronic engineering, these tandem solar cells can yield power conversion efficiencies beyond 30%. Here, we present a systematic material study on CH3NH3Pb(I0.6Br0.4)3 which has an optimal band gap of 1.77 eV. Crystallization dynamics in spin-coated CH3NH3Pb(I0.6Br0.4)3 films were studied by varying the annealing conditions of the films. The impact of annealing temperature and duration on crystallinity, carrier lifetime and average grain size of the active layers was studied using X-ray diffraction, time-resolved photoluminescence and scanning electron microscopy. The solar cells fabricated with these films show a clear correlation between crystallization dynamics in the perovskite films and device performance. Analysis of grain growth kinetics with extraction of activation energy for grain boundary mobility, and grain growth exponent for the perovskite layers annealed at different temperatures provide a quantitative analysis of quality of the layers. These insights linking the fundamental material characteristics of the perovskite films with device performance are crucial for further development of high-performance perovskite photovoltaics.

Resume : The physical properties of semiconducting thin films are influenced by the course of their growth process. Hence, a detailed understanding of the mechanisms taking place during film growth is important for a targeted development and optimization of thin film solar cell fabrication processes. This is particularly true for complex compound systems such as Cu(In,Ga)Se2 or Cu2ZnSn(S,Se)4, where the process parameter space is huge. Microstructural properties such as compositional gradients, grain size and extended defects may crucially influence the resulting performance of the solar cell. In-situ methods such as x-ray diffraction, x-ray fluorescence analysis, or white light reflectometry are able to provide time-resolved information on the evolution of these properties, potentially pointing to new process pathways or revealing which process steps are decisive for reaching high film quality. However, due to the complexity of the reactions and the interplay of different mechanisms, it is often not possible to directly extract all information contained in the data. In this contribution, we show that a more complete picture and understanding of the formation of compositional gradients, stress-relaxation, and grain growth in Cu(In,Ga)Se2 or Cu2ZnSn(S,Se)4 can be gained when the experimental real-time data are complemented by numerical modelling.

Resume : Solar cells with diffused homo-emitter on p type crystalline silicon (c-Si) are a mature technology in mass production. As an alternative, a-Si/c-Si heterojunction (SHJ) cells based on n-type Si wafers are considered a technology with higher efficiency potential and better energy yield. Moreover, a lean cell production process at temperatures <200°C is compatible with very thin wafers. Here, the challenge lies in finding industry compatible processes and materials to allow low-cost production of high efficiency cells. Highly transparent front-side electrodes and ultra-low contact resistance materials are required. In our contribution we discuss: (1) standard indium-oxide as well as zinc oxide as a low-cost alternative electrode material, and their contribution to minimize optical and resistive losses [1, 2]; (2) Combinations of doped nano-crystalline silicon and silicon oxides, which have been developed as contact layers to minimize both the parasitic absorption and the contact resistance [3, 4]; (3) High work-function metal oxide layers (e.g. MoOx, WOx), which have recently been suggested as possible candidates to replace the doped silicon [5]. References [1] T. Koida et al., Solar Energy Materials & Solar Cells 93 (2009) 851–854 [2] H. Scherg-Kurmes et al., Thin Solid Films 594 (2015) 316–322 [3] L. Mazzarella et al., Appl. Phys. Lett. 106 (2015) 023902 [4] L. Mazzarella et al., Energy Procedia 77 (2015) 304 – 310 [5] C. Battaglia et al., Appl. Phys. Lett. 104 (2014) 113902

Resume : We report on the application of multijunction solar cells in photoelectrochemical devices for hydrogen production. A chemical reaction to generate hydrogen through photoelectrolysis of water requires photovoltages over 1.5 V to run autonomously. The solar-to-hydrogen (STH) efficiency is determined by the photocurrent at the respective required voltage. We have shown successful application of multijunction solar cells in PEC systems with 9.5% STH efficiency [1]. Here we investigate photoelectrochemical system in more details, where the multijunction solar cells made of amorphous (a-Si:H) and microcrystalline (μc-Si:H) silicon cover up a wide photovoltage range in combination with high photocurrents. We study a-Si:H/µc-Si:H and a-Si:H/a-Si:H tandem junction, a-Si:H/µc-Si:H/µc-Si:H and a-Si:H/a-Si:H/µc-Si:H triple junction, and a-Si:H/a-Si:H/µc-Si:H/µc-Si:H quadruple junction solar cells, which provide open-circuit voltages Voc ranging from 1.5 to 2.8 V, along with initial efficiencies up to 13.6 %. The influence of varied illumination intensities on the performance of both multijunction solar cells and PEC devices is addressed. The trade-offs between Voc, short circuit current density and photoelectrochemical performance are studied over a wide range of parameters (such as photovoltages-photocurrents, various catalysts and overpotential losses) in an integrated photoelectrochemical device configuration. 1. F. Urbain, V. Smirnov et al, Energy Environ. Sci. 9, 145 (2016).

Resume : Backside-contact solar cells were fabricated on 10 µm thick, boron doped, liquid phase crystallized silicon (LPC-Si) on glass substrates, with interlayers between the glass and the silicon based on either silicon oxynitride (SiON) or aluminium oxide (Al2O3). The SiON layer contains positive fixed charges, which results in a charge inversion layer on the silicon side of the Si-SiON interface. On the other hand, the Al2O3 layer contains negative fixed charges which results in a charge accumulation layer at the interface. In this contribution we show our results on the fixed charge density in the SiON and Al2O3 layer determined with C-V measurements. By employing large area external quantum efficiency measurements we show that, for the cells with SiON based layers, minority charge carriers are collected from outside the cell area through the inversion layer. Light beam induced current measurements reveal that the loss due to incomplete collection below the absorber contact (electrical shading) is reduced from 10.5% for the Al2O3 samples to 7% for the SiON samples, presumably due to the collection of charge carriers through the inversion layer. This effect can also explain the previously observed large electrical shading loss of 11% for n-doped compared to p-doped LPC-Si solar cells on SiON based interlayers. Finally, our results point out a way to collect charge carriers below the absorber contact of backside-contact solar cells despite a relatively small diffusion length.

Resume : Minority carrier lifetime is the key materials parameter for multicrystalline silicon (mc-Si) substrates for solar cells. Lifetime in mc-Si is limited by recombination associated with metallic impurities in many forms, including point-like defects, precipitates and impurities bound to or precipitated at structural defects such as dislocations or grain boundaries. Some metallic impurities, such as interstitial iron, are sufficiently mobile that they can be redistributed by annealing at low temperatures (< 500 °C) and such internal gettering processes can be used to enhance lifetime. We have conducted a detailed study into the effects of long low temperature annealing on the lifetime and interstitial iron distribution in mc-Si from different regions of a directionally-solidified block. Systematic sets of experiments are conducted at different stages in the cell process (as-received and after phosphorus diffusion gettering). Photoluminescence imaging is used to monitor the evolution of lifetime and interstitial iron concentration with processing time. The results are strongly dependent on the choice of surface passivation, with possible hydrogenation from silicon nitride films having a strong effect. We find long low temperature annealing can result in substantial improvements in carrier lifetime with and without possible hydrogenation. The impact is greatest in red zone material from the bottom of the block, in which lifetime improvements of a factor of ~7 have been achieved.

Resume : It is demonstrated that polycrystalline free standing Si powder based wafers can be processed using hot pressing of ingots at temperatures around 1300 oC followed by the wire saw based wafering of such ingots. Properties of Si powder based wafers have been studied by electrical, SEM, EDS, Raman and EBSD analysis. It has been demonstrated that resistivity of Si powder based substrates can be varied by "in-situ" doping before the sintering process, by mixing Si and Boron powders together. Possible application of Si powder sintered substrates for PV is discussed. Moreover, it has been demonstrated that using an optical furnace that consists of an array of laser diode bars, it is possible to produce a floating molten zone and to realize full crystallization of Si powder sintered samples. As a result, large grain poly-Si substrates can be fabricated, which were analyzed by SEM and EBSD. It is concluded that combination of sintering and laser recrystallization steps being applied to Si powder based wafers, can be considered as a low-cost alternative for Si based wafers and layers for photovoltaic applications. The work described in the paper is finished so the past tense is most appropriate. Further, definite language should be used wherever possible, see below: In this work, polycrystalline free standing Si powder based wafers were produced using hot pressing of ingots at temperatures around 1300 oC, followed by wire saw wafering. The properties of the Si powder based wafers were studied by electrical, SEM, EDS, Raman and EBSD analyses. The resistivity of Si powder based substrates was varied by "in-situ" doping before the sintering process, by mixing Si and Boron powders together. Possible application of sintered Si powder substrates for PV was discussed. An optical furnace consisting of an array of laser diodes was used to produce a floating molten zone and to realize full crystallization of sintered Si powder samples. Large grain poly-Si substrates were produced, which were analyzed by SEM and EBSD. It was concluded that the application of sintering and laser recrystallization steps to Si powder based wafers could be considered as a low-cost alternative for Si based wafers and layers for photovoltaic applications.

Resume : Thin film polycrystalline silicon solar cells combine the advantages of both thin film technologies, i.e. low material and energy consumption for manufacturing and high crystalline quality of wafer-based solar cells. Crystalline quality of silicon thin films can be characterized by both open-circuit voltage of the solar cell and according to our experiments also by transient terahertz spectroscopy [1]. Optical pump terahertz probe spectroscopy enables a contactless photoconductivity inspection of solar cells giving a detailed insight into recombination processes that take place at defects, i.e. interfaces, grain boundaries, dislocations. A series of polycrystalline Si thin film solar cells with different silicon quality was prepared by their treatment in hydrogen plasma under various technological conditions. A clear correlation was found between the carrier trapping rate in the absorber layer of the cells and their open-circuit voltage. In this contribution we will develop this new approach to analyse also alternative passivation approaches, e.g. using the water vapour passivation [2]. The origins of the correlation of the trapping rates probed on picosecond scale and quasi steady state open circuit voltage will be discussed. [1] P. Pikna, V. Skoromets, C. Becker, A. Fejfar, P. Kužel, Thin film polycrystalline Si solar cells studied in transient regime by optical pump–terahertz probe spectroscopy, Applied Physics Letters. 107 (2015) 233901. doi:10.1063/1.4937388. [2] P. Pikna, M. Müller, A. Fejfar, Passivation effect of water vapour on thin film polycrystalline silicon solar cells, 26th International Conference on Amorphous and Nanocrystalline Semiconductors, Aachen, Germany, 2015. This research was supported by Czech Science Foundation project 13-12386S.

Resume : Organic solar cells have demonstrated efficiencies above 10% recently, though their practical implementation is restricted mainly due to insufficient operational stability. In particular, fullerenes were shown to undergo facile photodimerization, which is currently considered as the main source of burn-in degradation effects in organic solar cells [A. Distler et al., 2014, 4, 1300693; T. Heumueller et al., Energy Environ. Sci., 2016, 9, 247]. In this talk, we will present results of our systematic study of ~20 fullerene derivatives bearing different organic addends attached to the carbon cage. The main factors influencing the photodimerization of different materials including standard [60]PCBM will be discussed. Some details of the mechanism of the photoinduced cross-linking reaction will be unraveled and few types of the photochemically stable fullerene derivatives will be presented. In the second part of the talk, we will focus on the alternative photodegradation pathway of fullerene derivatives leading to the generation and accumulation of stable radical species. These paramagnetic centers behave as traps for mobile charge carriers and affect severely the performance of organic solar cells. Finally, we will compare the impacts of the radical and non-radical (dimerization) photodegradation pathways of fullerenes and their derivatives and outline future research directions for designing fairly stable n-type materials for efficient and durable organic photovoltaics.

Resume : It is now well established that the degree of crystallinity affects the lifetime of polymer based organic solar cells and air exposure is mainly thought to be responsible for the degradation. Here we show that in the absence of air (or light) but in the presence of electron irradiation similar effects to that known as burn-in in solar P3HT:PCBM based solar cells can be observed for P3HT with relatively low crystalline order. Highly ordered P3HT, and a novel high efficiency polymer (PffBT4T) do not show such charge induced “burn-in effect”. The above observation is based on the changes in the low loss (plasmon) region of electron energy loss spectra induced by prolonged electron-beam exposure. Detailed analysis of this region suggests slight changes in the band-gap of the P3HT, depending on P3HT morphology. We further establish through secondary electron spectroscopy in a low-voltage SEM that this change in band gap has a profound influence on charge transport which is reflected in substantial differences in the emission of low energy secondary electrons. We also demonstrate that the distinctive differences in secondary electron spectra shape can be exploited to map the local crystallinity distribution in P3HT, by using energy filter low voltage scanning electron microscopy, thus providing a new tool that can aid efficient morphology engineering. Our findings are in agreement with many reports that advocate a higher degree of crystallinity in order to increase device lifetimes but our observation also implies that preventing charge built -up in the active layer through appropriate device design & morphology engineering needs to be considered as a life extending measure in addition to reducing air exposure.

Resume : Organic and hybrid organic-inorganic optoelectronics are the subject of intensive research efforts partially motivated by their potentials to achieve low processing cost devices, for instance via roll-to-roll and inkjet printing processes, as well as their promises to deliver exciting mechanical properties.[1] In the present work,[2] the hole-transport layer of two organic solar cells was doped and shown to strongly influence the device efficiency. A combination of near-field microscopy, Kelvin probes, UV photoemission spectroscopy, conductivity, steady state and time resolved optical spectroscopy, as well as ellipsometry were combined with electro-optical modelling to gain insight into the mechanism involved in photovoltaic structures. We show how a simple PEDOT:PSS doping can alter simultaneously interfacial work-functions, built-in potentials and optical constants to impact on organic solar cell efficiency.

Resume : Great improvement in the performance of organic bulk-heterojunction solar cells was achieved mostly due to successful design of novel photoactive materials. Recently, we have developed promising electron-donor copolymers (X-TTBTBTT-)n (X – fluorine, carbazole; B –benzothiadiazole; T – thiophene), which demonstrate good performances both in spin-coated (up to 7%) [1-3] and roll-to-roll processed (up to 6.2%) [4] devices. In the present work we were aiming to improve further the optoelectronic characteristics of the designed copolymers (X-TTATATT-)n via a systematic variation of the acceptor (A) units. We synthesized and investigated materials based on 2-alkylbenzotriazole, several different quinoxaline derivatives, benzoxadiazole and 5,6-bis(octyloxy)benzoxadiazole. We will discuss how the chemical structure of the acceptor building blocks affects the optoelectronic and physicochemical properties of the materials as well as their performance in organic solar cells. The revealed correlations provide useful guidelines for designing novel copolymers based on extended TTATATT units for efficient organic photovoltaics. [1] A. V. Akkuratov et al., Macromolecules, 2015, 48, pp 2013–2021 [2] I. E. Kuznetsov et al., Chem. Commun., 2015, 51, 7562-7564 [3] A. V. Akkuratov et al., J. Mater. Chem. A, 2016, submitted [4] I. Burgués-Ceballos et al., ChemSusChem, 2015, 8, 4209

Resume : A method to directly determine doping profiles in sandwich-type thin-film devices of low mobility materials is presented. The method is an extension of the Charge Extraction by a Linearly Increasing Voltage (CELIV) technique which is one of the most common methods to measure the charge transport properties in (undoped) low mobility materials. We recently showed that when the doping concentration in the active layer is high enough, so that the depletion region is smaller than the device thickness, the charge extraction transients become purely capacitive when a slowly enough linearly increasing voltage pulse is applied [1]. In this doping-induced capacitive regime of CELIV (doping-CELIV) the doping concentration and built-in potential of the device can be obtained, assuming a constant doping profile. This method has thus far been successfully applied to polymer:fullerene, organic small molecule, and CZTSSe-based solar cells. In this work, we have extended the method of using the capacitive regime of CELIV to an arbitrary doping profile. The method is demonstrated both with computer simulations based on a drift-diffusion approach and experimentally on organic thin-film diodes. Furthermore, using a similar approach, but on undoped devices with one ohmic contact, we show that the density of injected charge due to Fermi level alignment at the ohmic contact can be determined as well. [1] Oskar J. Sandberg, Mathias Nyman, and Ronald Österbacka, Organic Electronics 15 3413–3420 (2014)

Resume : Light management is the most promising route to develop thin film PV. Recently, several strategies were proposed employing the scattering and/or anti-reflection properties of nanophotonic elements, with dimensions comparable or below the illuminating wavelengths, to trap light in solar cells. For the cells’ rear, resonant plasmonic nanoparticles (NPs) can enhance the path length of the near-infrared light reaching the cells’ back contact. High-performing plasmonic back reflector (PBR) structures, produced by an innovative wet-coating method [1], were incorporated in thin film Si cells. The strong scattering properties of the optimally-sized colloidal NPs allow remarkable enhancements, of 43% and 41%, respectively in the cells’ current and efficiency [2]. For the cells’ front, high-index wavelength-sized dielectric features are shown to produce pronounced anti-reflection and forward-scattering, in the regime of wave-optics. Optimized TiO2 half-spheroids scatter as strongly as plasmonic NPs but without the associated parasitic absorption [3], generating superior absorption in the cells, mainly in the UV-visible range, and current and efficiency enhancements up to 50%. Therefore, broadband light trapping close to the theoretical limits can be expected for thin film cells with a combination of plasmonic (at the rear) and high-index dielectric (at the front) photonic structures. [1] M.J. Mendes et al. Nanoscale (2014) [2] M.J. Mendes et al. Nanotechnology (2015) [3] M.J. Mendes et al. Optics Express (2011)

Resume : Silicon photosensitisation using light harvesting structures represents an attractive solution for reducing the amount of semiconductor material needed by up to two orders of magnitude in the manufacture of solar cells [1-3]. Furthermore separating the photovoltaic process into two separate steps, an absorption step and a charge separation step, can provide a solution for low cost solar electricity. The direct attachment of alkyl layers [4-6] and chromophores [7-9] represents an attractive solution to combine the excellent optical properties of dye molecules with the electronic properties of thin silicon converters. The above approach divides the photovoltaic conversion process into two processes consisting of an energy collector (light harvesting) which absorbs light with a high optical absorption cross-section, and transfers energy to a semiconductor converter (silicon) which efficiently separates the photo-generated charges and produces electricity. The resulting organic-inorganic structure presents a step change in thinking in photovoltaic energy conversion. A chlorination–alkylation procedure has been investigated with a view to improving the surface passivation properties on silicon [4-6]. We have found that increasing surface coverage of the alkyl monolayer raises the measured Surface Photovoltage (SPV) and increases the electron-hole recombination lifetime. We investigate the potential of photosensitization of silicon surfaces using covalently attached protoporphyrin IX derivative molecules. The illuminated surface results in the Interaction between the near field of the molecular dipole and the transition dipole moment of the electron transition between the valence and conduction band of the semiconductor. Experimental verification of the near field interaction of dyes on surfaces has been carried out in the past by monitoring the fluorescence lifetime quenching near the surface. We present time resolved fluorescence quenching measurements for dyes directly attached on the silicon surface. In particular, we investigate the near field interaction of the excited dye at distances to the silicon surface less than 2nm. References 1. N. Alderman, L. Danos, L. Fang, T. Parel, and T. Markvart, in Photovoltaic Specialist Conference (PVSC), 2014 IEEE 40th, 2014, pp. 17–21. 2. T. Markvart, L. Danos, N. Alderman, L. Fang, and T. Parel, in 27th European Photovoltaic Solar Energy Conference, Frankfurt, 2012, pp. 1–6. 3. T. Markvart, L. Danos, L. Fang, T. Parel, and N. Soleimani, RSC Adv., 2012, 2, 3173. 4. N. Alderman, M. Adib Ibrahim, L. Danos, M. C. Grossel, and T. Markvart, Appl. Phys. Lett., 2013, 103, 081603. 5. N. Alderman, L. Danos, M. C. Grossel, and T. Markvart, RSC Adv., 2013, 3, 20125. 6. N. Alderman, L. Danos, C. Grossel, and T. Markvart, RSC Adv., 2012, 2, 7669–7672. 7. L. Fang, K. S. Kiang, N. P. Alderman, L. Danos, and T. Markvart, Opt. Express, 2015, 23, A1528–A1532. 8. L. Danos and T. Markvart, Chem. Phys. Lett., 2010, 490, 194–199. 9. L. Danos, R. Greef, and T. Markvart, Thin Solid Films, 2008, 516, 7251–7255.

Resume : One of the solutions for improving the energy conversion efficiency of silicon solar cells (SC) consists of using a down-conversion (DC) layer. Thus, DC layers doped with Tb3 and Yb3 ions allow the emission of two IR photons (1.26 eV) for each absorbed UV ones (>2.58 eV) through a cooperative energy transfer between Tb3+ and Yb3+ ions. Since IR photons energy is more suited to the Si band gap energy (1.11 eV) of SC than the UV ones, conversion efficiency is increased while the photogenerated carriers thermalization is reduced. Succeeding to dope the SiNx antireflective layer with such a couple of ions is consequently a key challenge. Tb3+ -Yb3+ co-doped SiNx layers deposited by reactive co-sputtering approach show an internal DC conversion quantum efficiency as high as 200%. To increase the number of IR photons converted, different approaches have been carried out: (i) a multilayer approach based on alternative Tb and Yb doped sub-layers has been developed for tailoring the Tb3+ -Yb3+ coupling rate and improving the down-conversion efficiency, (ii) silver spherical- and triangle-shaped nanoparticles have been embedded in DC layer to enhance the effective rare earth ions absorption cross-section. Modeling of light-matter interactions coupled to experimental results achieved in such different structures will be presented. An increase of the number of IR photons emitted by the DC layer into the SC has been demonstrated, which is promising for a future efficient SC.

Resume : 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 disorder, which will segregate hole and electron populations and so reduce recombination [3]. Both of these effects are driven by dynamic disorder. We have 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 (King's 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).

Resume : CH3NH3PbI3-based perovskite solar cells (PSC) are an important class of photovoltaic devices with high efficiency, low cost and easily scale-up fabrication process. Among the different deposition processes introduced for PSC fabrication, the two-step method is easy to scale up for large area modules. In the two-step method, deposition of PbI2 layer is crucial because this layer can affect the nucleation process of perovskite. In the present work, we investigate the effect of dopants in PbI2 solution on the performance of PSCs. Different amounts of various dopants like dimethyl sulfoxide (DMSO), t-buthyl pyridine (TBP) and Li-salt are used in PbI2 solution containing dimethylformamide (DMF) as solvent. Furthermore, effect of PbI2 preheating process is also investigated in perovskite nucleation and crystal growth process. The effects of these variants are evaluated by UV-Vis, IV measurement and cyclic voltammetry and standard light and moisture stability test methods. The results show higher overall efficiency and stability of the devices containing small amounts of DMSO as dopant and cold dipping of PbI2 layer in methylammonium iodide solution.

Resume : Perovskite / silicon-heterojunction tandem solar cells are interesting candidates to exceed the Shockley-Queisser limit for single junction solar cells due to the complementary energy band gaps and the high conversion efficiency of the two subcells. Optical simulations have indeed shown the potential in both monolithic and 4-terminal configuration [1]; first experimentally realized devices also show promising results [2]. Currently, experimental monolithic tandem devices have planar configuration only and are therefore partially limited by non-optimal photocurrent generation, due to reflection losses and imbalance between the photocurrents generated in the two subcells. Further gains in photocurrents can be achieved by introducing structures for light scattering and anti-reflection (AR). However, commonly used random pyramid texturization of silicon wafers is not yet suitable for the deposition of thin perovskite layers when implementing deposition techniques such as spin coating that yield the highest efficiencies to date. An alternative way to reduce optical losses on planar wafers is utilizing UV Nanoimprint Lithography (UV NIL). With UV NIL an additional transparent textured layer on top of the planar tandem (or single junction) cell can be created that enables light scattering or/and AR. In our contribution, we focus on numerical modeling for the optical optimization of monolithic perovskite/silicon-heterojunction tandem solar cell using the optical simulator CROWM which is based on a combined ray and wave optics model [3]. First, random pyramids of the silicon wafer either at the back or on both sides are considered. Second, different textures, such as randomly and periodically distributed pyramids, concave microlenses and spheroidal shapes, deposited on the top of the flat tandem cell (ITO) by UV NIL process are investigated. For the selected device architectures we optimize the thickness of the perovskite layer to achieve current matching in monolithic perovskite/silicon-heterojunction tandem solar cells. We present the experimentally relevant optimization for the different thicknesses of recombination layer (ITO) and hole transporting layer (spiro-OMeTAD). Our simulations show that using UV NIL textured layers on top of the planar tandem cell, more than 2 mAcm 2 in photocurrent density can be gained and front reflection is strongly reduced. Finally, the suitability of this process will be shown by comparison of simulations and experiments for perovskite single junction solar cells with a top UV NIL layer. Thus, our results give insight into the light management improvements in perovskite single junction and perovskite/silicon tandem solar cells. Literature [1] M. Filipič, P. Löper, B. Niesen, S. De Wolf, J. Krč, C. Ballif, and M. Topič, “CH_3NH_3PbI_3 perovskite / silicon tandem solar cells: characterization based optical simulations,” Opt. Express, vol. 23, no. 7, p. A263, Apr. 2015. [2] S. Albrecht, M. Saliba, J. P. C. Baena, F. Lang, L. Kegelmann, M. Mews, L. Steier, A. Abate, J. Rappich, L. Korte, R. Schlatmann, M. K. Nazeeruddin, A. Hagfeldt, M. Grätzel, and B. Rech, “Monolithic perovskite/silicon-heterojunction tandem solar cells processed at low temperature,” Energy Environ. Sci., vol. 9, no. 1, pp. 81–88, Jan. 2016. [3] B. Lipovsek, J. Krč, and M. Topič, “Optical model for thin-film photovoltaic devices with large surface textures at the front side,” Inf. MIDEM, vol. 41, no. 4, pp. 264–271, Dec. 2011.

Resume : One direct method for the measurement of the diffusion length of photogenerated charge carriers in semiconductors is based on the dependence on the absorption length of the intensity required for keeping a surface photovoltage (SPV) signal constant (Goodman 1961). Transport or diffusion lengths in CH3NH3PbI3 samples were measured with high accuracy by applying several SPV signals for each sample. Values of the transport lengths varied between tens of nm and µm depending on preparation and treatments. The reliability and the simplicity of the method allow its application also for in-line characterization under well controlled ambience.

Resume : High conversion efficiency has been recently demonstrated for Perovskite thin film photovoltaic devices. Perovskite thin film solar cells are multilayer opto-electrical structures in which light interference occurs. This phenomenon can be used to maximise the light transmission into the absorber material and increase the device efficiency. Fine tuning of the layer thicknesses within the stack can be used to control interference at the interfaces. Optical reflection losses can be reduced by achieving destructive interference within the structure of the cell. The light transmission to the Perovskite absorber of a thin film solar cell on a fluorine doped tin oxide transparent conductor has been modelled using the transfer matrix method. Alternative transparent conductor materials have been also investigated including AZO and ITO. The modelling showed that replacing FTO with ITO could increase the photocurrent by as much as 4.5%. The gain can be further increased to 6.5% by using AZO as the TCO material. Fine tuning of the TiO2 layer thickness can increase current by 0.3%. Furthermore, the current of a Perovskite solar cell can be increased by application of a multilayer anti-reflective coating by another 3.5%. Optical optimisation of the stack design offers a significant increase in conversion efficiency.

Resume : Chalcopyrite semiconductors are commonly used as materials on solar cell devices. Built as tandem solar cells, the study of the heterointerfaces between different semiconductors which are components of the devices should be crucial to understand their operation. Band alignments of the heterojunctions between CuGaS2 chalcopyrite and different semiconductors have been theoretically obtained using density functional theory with hybrid functionals. Band alignments have been determined using an average electrostatic potential as reference level. We have also studied the strain between the heterointerfaces which plays an important role in the electronic properties of them.Results show that CuAlSe2/CuGaS2 and CuGaS2/ZnSe heterointerfaces show band alignments where holes and electrons are localized on different sides of the heterojunction. This condition is necessary for their application on photovoltaic devices. If suitable transition metals are used to substitute for Gallium atom in the CuGaS 2 chalcopyrite, an intermediate band solar cell could be made, which could potentially improve the performance of a solar cell.

Resume : Thin film solar cell approach aims to reduce cost of solar electricity by reducing material usage. Due to the limited film thickness, maximizing the absorption of the solar radiation is considered to be crucial. Scattering of the light provides longer optical path inside the thin film; thus, absorption of light can be enhanced. Aluminum induced texturing (AIT) is a promising method to increase the efficiency of thin film solar cells by promoting the scattering of the transmitted light through glass substrate. There are various parameters affecting the final texture of the AIT glass, such as aluminum thickness, annealing time, annealing temperature, etching time, etc. In this study, we focus on the N2annealing step of the AITprocess. During annealing, aluminum reacts with SiO2 and forms c-Si clusters and Al2O3. The effect of glass composition has been reportedin literature.Here, we aim to investigate on the evolution of the Al/Al2O3 layer and elucidate on the chemical processes that occur during N2 annealing. For the annealing experiment, two different study is made. In one of them,600 °C is set as fixed and samples are annealed in N2 ambient at different durations. To monitor the chemical nature of the films, annealed samples were examined by depth resolved X-ray photoelectron spectroscope (XPS). Surface morphology of the annealed samples before and after the removal of the reaction products were studied using optical microscope,scanning electron microscope and atomic force microscope.The scattering behavior of the textured glass samples were studied by goniometric optical scatter instrument. In the second study, annealing made at different temperatures and different durations. Evaluation of layer is examined by scanning electron microscope in EDX mode and optical microscope. After removing reaction products from surface, all samples are investigated by goniometric optical scatter instrument, atomic force microscope and scanning electron microscope.

Resume : Kesterite (CZTS) is a promising candidate for low-cost and high efficiency earth abundant, thin film solar cells. With the recent reported rapid improvement of the efficiency of CZTSSe solar cells, it is highly desirable to develop a deposition method that is compatible with low cost and scalable processing of kesterite thin film soalr cells towards commercialization. In our group, a non-vacuum, low cost and eco-friendly Electrostatic Spray Assisted Vapour Deposition (ESAVD) method has been adopted to produce CIGS and CZTSSe absorbers for thin film solar cells[1-2]. ESAVD is a promising method for industrialization due to its high deposition speed and close to unity deposition efficiency. In this work, DMSO with a good complex with metal salts was adopted as solvent for CZTSSe precursors to better control the composition in the ESAVD deposited absorber. In addition to the composition and crystallization of absorber, the interface between absorber and Mo glass also has significant influence on the device performance of CZTSSe solar cells. In order to improve the efficiency of ESAVD deposited CZTSSe solar cells, an ultrathin ZnO (circa 10 nm) layer was employed as an intermediate layer between CZTSSe and Mo back contact to avoid the direct contact between Mo and CZTSSe and reduce the decomposition of CZTSSe during the annealing process. XRF and EDX were used to characterize the chemical composition of CZTSSe before and after selenization, respectively. SEM and Raman results showed the improved absorber morphology and the reduced direct interfacial reaction between CZTSSe and Mo. The improvement of the CZTSSe/Mo interface due to the presence of the ZnO intermediate layer was also reflected in the quality of the derived photovoltaic devices, leading to an improved efficiency of ESAVD deposited kesterite solar cells from 3.25% to 4.03%. Different alkali metals like Na, Li and Rb were incorporated into CZTSSe compounds to further improve the photovoltaic performances of the related devices. Photovoltaic properties results showed that Li, Na and Rb incorporation could increase the power conversion efficiency of CZTS devices up to 5.5%. The introduction of a thiourea treatment, has improved the quality of the absorber| buffer interface, and pushed the device efficiency up to 6.3% which is at the moment the best reported result for ESAVD deposited CZTSSe solar cells. References: [1] Giovanni Altamura, Mingqing Wang, Kwang-Leong Choy,Thin Solid Films(acceptted) [2] Mingqing Wang, Xianghui Hou, Junpeng Liu, KwangLeong Choy*, Paul Gibson, Elhamali Salem,Demosthenes Koutsogeorgis, and Wayne Cranton, Phys. Status Solidi A 212, No. 1, 72–75 (2015). Acknowledgements: This work has been funded by the European Union’s Seventh Framework Programme Scalenano, FP7/2007-2013 under grant agreement nº 284486.

Resume : During the last years, the research on Cu(In,Ga)Se2 (CIGS) as a promising absorber candidate for low-cost thin-film photovoltaic cells has steadily increased widening our understanding on its basic materials properties [1] and leading to a high efficiency of 21.7 % for the solar cells [2]. The defect microstructure influences optical and electronic properties of the material. Understanding its evolution during the manufacture and use of solar cells is impossible without knowing the fundamental parameters of its native point defects, the most important one of which is the defect formation energy. Indeed, there exist several recent first-principles studies [3-7] for defect formation energies in CuInSe2 (CIS) and CuGaSe2 (CGS). Their results agree with respect to general trends and orders of magnitudes, but they may differ in some important cases resulting, e.g., in different values for ionization levels within the band gap. It may be even so that some ionization levels arise only in part of the published calculations. Because these details are important in predicting the influence of the defects on doping and charge carrier compensation in the material this kind of discrepancies should be solved. A possible reason for the situation could be the use of inappropriate or insufficient computational parameters and post – treatments. We present the results of a detailed systematic study on the influence of computational parameters on the formation energies. Our approach is based on comparing results obtained by different existing corrections for the supercell method and by different electronic-structure calculation schemes (Projector-augmented wave – plane-wave method, All-electron – numerical-atomic-orbitals method). Another aspect of this presentation is the influence of alkali metal impurities. During the thin film deposition, sodium and potassium diffuse into the CIGS layer [7] and they have also been subjects of several first-principles investigations [8-10]. Inspired by the scatter in the first-principles results for native defects in CIGS materials we have revisited also the formation and migration energies of alkali metals as well as their clustering with other defects. Also we calculated migration barrier for defect in Cu-poor compounds. The migration barriers were determined by the state-of-the-art climbing-image nudged-elastic-band (CI-NEB) method for different types of materials, such as the stoichiometric CIS and the Cu-poor ordered compound. References: [1] Susanne Siebentritt et al.  Solar Energy Materials and Solar Cells,119, 18 (2013). [2]Michael Powalla, Wolfram Witte, Philip Jackson, Stefan Paetel, Erwin Lotter IEEE, Journal of Photovoltaics 4(1):440 (2014). [3] Yu Kumagai and Fumiyasu Oba Phys. Rev. B 89, 195205 (2014) . [4] J. Pohl and K. Albe, Phys. Rev. B 87, 245203 (2013). [5] L. E. Oikkonen, M. G. Ganchenkova, A. P. Seitsonen, and R. M. Nieminen, Phys. Rev. B 86, 165115 (2012); ibid. J. Phys.: Condens. Matter 26 345501 (2014). [6] J. Bekaert,   R. Saniz   B. Partoens Phys. Chem. Chem. Phys., 2014,16, 22299-22308 (2014). [7] F. Pianezzi, P. Reinhard, A. Chirilă. Appl. Phys. 114, 194508 (2013). [8] Tsuyoshi Maeda, Atsuhito Kawabata, and Takahiro Wada Japanese Journal of Applied Physics 54, 08KC20 (2015). [9] L. E. Oikkonen, M. G. Ganchenkova, A. P. Seitsonen, and R. M. Nieminen, J. Appl. Phys 114, 083503 (2013). [10] E.Ghorbani, Janos Kiss, Thomas Gruhn, Guido Roma J. Phys. Chem. C, 2015, 119 (45), pp 25197–25203 (2015).

Resume : Up-Converters implemented in solar cells have theoretical efficiencies of the conversion much higher than the actual and a much smaller price production price. However, the experimental results in the present do not approach the theoretical calculated values, because they have a small increase in photocurrent by incorporating the bifacial solar cell, for example. Hence, it is necessary dispose a quantitative tool for the characterization of candidate materials as up-converter, which permits realize an influence sensibility analyzes of certain parameters and contrasts with the real values founded. Thus, the fundamental objective of this work is to develop a mathematical model to study the behavior of up-converters, based on real parameters found in the literature, when implemented in silicon bifacial solar cells. Besides, exemplifies this model getting comparative results (I x V curve) for different materials used as up-converter when incorporated in silicon bifacial solar cells using the unidimensional software, PC1-D, in order to get the best candidates to be used.

Resume : Combine lower production costs with high efficiency currently it has been a major challenge for the development of solar cells. Then, the objective for this work is to study of optical performance of commercial up converters and quantum dots for application in bifacial solar cells. These materials have properties of absorbing infrared radiation and re-emit visible radiation, and thus the bifacial solar cells can improve the efficiency of solar cell. 24 samples is prepared in a silicone gel, similar to that used in the encapsulation of solar cells for photovoltaic module of the composition, with different concentrations of up converters and quantum dots. To study the performance optical of commercial up converters and quantum dots used the reflectance, transmittance and photoluminesce. PTIR545/UF from Phosphor Tecnology converter showed the most promising results, mainly associated with quantum dots. For exhibit considerable reduction in transmittance and an excellent absortance for wavelengths where the solar cell does not respond. The three measurements indicate similar characteristics of the quantum dots and up converters demonstrating the veracity of the proposal characterization.

Resume : CuInSe2 nanoparticules (CIS-NP) were synthesized on ITO-coated glass substrate by electrodeposition and rapid thermal processing (RTP). The as-deposited films were annealed under argon atmosphere at 250°C, 350°C and 450ºC using RTP for a short time. The relatively short annealing time is proved to be practicable to avoid further losing of the Se content in CIS films. In order to analyze the effect of annealing temperature, the structural, morphological, optical and electrical properties were investigated respectively by means of Xray diffraction, scanning electron microscopy, UV-Visible Spectroscopy and Mott-Schottky plots. XRD results show that elaborated films have a tetragonal chalcopyrite CIS with preferential orientation along (112) orientation. The phase formation of the CIS with good crystallinity was observed at low annealing temperature. Optical absorption studies indicate a direct band gap around 1.02 eV at 250 °C. The optical constants such as refractive index n(λ) and extinction coefficient k(λ) were estimated using an appropriate optical model. To determine the doping type of elaborated semiconductor, its flat band potential and the free carrier concentration we have used the Mott-Schottky plots. A new attempt to anneal the electrodeposited CIS films by short annealing duration using RTP process was proved to be a useful method to synthesize polycrystalline CIS films for solar cell application.

Resume : Screen-printing provides an economically attractive means for making silver electrical contacts to silicon solar cells, but the use of silver substantiates a significant manufacturing cost, and the glass frit paste used in contact formation contains lead. This front contact metallization of Si begins with printing a mixture of an Ag powder, glass frit (mixture of metal oxides such as PbO, SiO2, B2O3, and Bi2O3) and an organic binder over the antireflection coating that is subsequently rapidly fired (<10 secs) up to about 800 ºC. It is known that the frit allows the paste to react with and burn through the anti-reflective coating such that the metal can react with the underlying c-Si during firing. However, the precise phase transformations between Ag, Si, SiNx, and the frit constituents, which happens within few seconds during rapid thermal processing (RTP), giving rise to an Ag-Si contact, are not well understood in absence of in-situ characterization under the actual processing conditions. We have carried out in-situ x-ray diffraction studies on sample mixtures of different components powders (Ag, SiNx, PbO-frit and Si) under realistic processing conditions using an in-situ rapid thermal processing setup. We track the phase progression and reaction pathways at a time resolution of 100 milliseconds. We show the direct evidence of SiNx oxidation by PbO between 550-650 C. On subsequent heating to higher temperature, up to 800 ºC, Ag dissolves into the frit in the form of Ag+ ions, which subsequently etch the c-Si surface and are deposited on etch pits forming intimate electrical contacts. On cooling Ag nanocrystals precipitate in the glass frit allowing electrical contact to the Si. This work clarifies contact formation mechanisms and suggests approaches for development of inexpensive, nontoxic solar cell contacting pastes.

Resume : The Porous Silicon layers were formed on p-type silicon wafers by electrochemical etching in an HF:C2H5OH (1:1 by volume) electrolyte at room temperature at a constant current density 20mA/cm2 and etching duration 5min. This paper investigates the effect of Berberine on the passivation of PS. The immersion of as-etched PS in dilute Berberine solution. The immersion duration of berbrine was variable from 5 to 20min. For the optical and morphological characterization of porous films, Photoluminescence spectroscopy, FTIR spectroscopy and Reflectivity spectroscopy were used. A decrease in the reflectivity to about 6% for Berberine/PS annealed at 20minwas obtained. From Photoluminescence (PL) spectra, a blue-shift of the gap and an intensity were observed when the immersion duration is increased to 20min; we correlate these results to the change in chemical composition of the layers in order to find the optimized conditions for a potential application in silicon solar cells.

Resume : Cu2ZnSnS4 (CZTS) thin films were successfully deposited on soda lime as well as on molybdenum coated soda lime glass substrates by ultrasonic spray technique. After sulfurization treatment under Argon atmosphere at 500°C, the films were polycrystalline and exhibited the kesterite structure which was confirmed by X-ray diffraction and Raman spectroscopy. The composition, surface morphology and optical properties were examined by dispersive X-ray spectroscopy, scanning electron microscopy and photoluminescence spectroscopy respectively. By Hall effect measurements, electrical resistivity and Hall mobility of about 2.4 10-1 Ohm.cm and 6 cm2•V−1•s−1 were reached respectively. Keywords: Cu2ZnSnS4, thin films, ultrasonic Spray process, sulfurization.

Resume : The semiconductor compound Cu2Zn1-xCdXSnS4 (CZCTS) is considered as one of the ideal photovoltaic absorber layer materials for low-cost thin film solar cells, since CZCTS has a large absorption coefficient and all the constituent elements are naturally abundant. A Cu2Zn1-xCdXSnS4 (CZCTS) thin films system (where x = 0 and 0.2) are deposited using chemical bath deposition method on the alluminium and İTO glass substrates.The films deposited onto İTO-glass slides were first cleaned with detergent water and then dipped in acetone. Solution were prepared by mixing 0.2 M aqueous solutions of CuCl2, ZnCl2, CdCl2, SnCl4, and thiourea [CS(NH2)2] at ratio of 2x:1-x: x: 1: 4 (Cu,Zn,Cd,Sn,S) using a magnetic stirrer. The films had a uniform thickness of (800) nm. the structural properties were determined by X-ray diffraction (XRD; Shimadzu) with CuKα radiation (λ = 1.5406 Å). Film morphology was analyzed by atomic force microscope (AFM)- type (CSPM). The optical absorption and transmission spectra were obtained using a UV-vis spectrophotometer within the wavelength range of 300 nm to 1100 nm. The XRD patterns show the major diffraction peaks at 2θ= (28.59) and (28.4) for CZTS at x = 0 and for CZCTS at x = 0.2. The increase in cadmium (Cd) as shown by the shift in the main diffraction peak to a lower value of 2θ is attributed to the increase in lattice spacing of the longer Zn atom (1.71 A°) substation for smaller Cd atoms (1.53 A°). Furthermore, an increase in the main peak intensity is observed in the presence of cadmium. A comparison with ASTM card JSPDS 26-0575 reveals that the CZTS (x = 0) thin film exhibits a crystal structure tetragonal type of kestrits phase with a preferred orientation (112) and other planes, i.e., (220) and (312). For 2θ= (28.59, 47.5, 56.1 The CZCTS film at x=0.2 has a tetragonal phase. The absorbance layers of CZCTS were measured from 300nm to 1100nm. Shows the plot of α (cm-1) versus the wave length λ, which suggests that the two Film exhibits high absorption coefficient ( > 104 cm-1). Thus a very thin layer of film (1-2μm) can absorb over 90% of photons over the spectrum, with higher photon energy in the bandgap. The optical properties of the CZTS layer can be improved with a substitution of Zn atoms by Cd atoms to give lower energy gap gap, because since ZnS has a direct optical band near 3.6 eV that gives a higher energy gap of CZTS. The absorption edge shifts to the NIR region with increased x. The obtained optical gap for CZTS is (1.7)eV which agrees with the CZTS bandgap and 1.66 eV for CZCTS at x = 0.2. Eg decreases with increased cadmium content. The CZCTS films coated on Al substrates were applied to the preparation of CZCTS solar cells. The CZCTS solar cells with a structure of Al /ZnO/CdS/ CZCTS /İn lime glass were fabricated. The performance of the solar cells was evaluated under standard AM 1.5 (100mW/cm2) illumination. The solar cell with the CZCTS absorber layer annealed at 3000C, exhibited a relatively high efficiency of 9.2% (Voc - 0,520 V, Jsc – 22,4 mA/cm2, FF – 0,65). It confirms the effect of preventing the decomposition of CZCTS phase by the addition of Sn during the annealing process. In conclusion, a simple and relatively safe approach for the fabrication of CZCTS nanoparticles has been developed. To the best of our knowledge, this is the first time that this low-temperature colloid approach has been applied to the fabrication of CZCTS nanoparticles. We found that the use of different chalcogenide sources resulted in different products of synthesis. The annealing temperature and special ambient effect on the properties of CZCTS films were investigated.

Resume : Using finite-difference-time-domain (3D-FDTD), we investigate the use of a periodic array of nanocoax within a thin amorphous silicon (a-Si) supported on the glass superstrate (n=1,5), to enhance optical light absorption allowing it to be used in photovoltaic application. In this analysis, the results reveal that the absorption through the coax design is insensitive to the incident and azimutal angle. We evaluate the performance of coax as a function of the normal and oblique incidences of light. By using nanocoax apertures, we have found the way to reduce the thickness of semiconductor film without losing the extraordinary absorption.The aim of this study is to demonstrate that the nanocoax apertures engraved in silicon film, improves the efficiency of absorption light for solar applications. Its insensitive behavior to the azimutal and incident angles and its polarization-insensitive from the light brings a new concept for plasmonic metamaterial engraved in photovoltaic cell for increasing the light absorption. These results are promising for the design of solar cells with nanocoax giving extraordinary absorption.

Resume : CdTe-based thin film solar cells currently represent one of the fastest growing PV technologies, with a superior combination of efficiency, energy payback time and lifecycle environmental impact. However, the current post-deposition annealing treatment is still an energy intensive step of the manufacturing process. A novel method is presented for annealing of CdTe using a high-power diode laser (35 W, 808 nm) for thermal post-processing, combined with holographic optical elements (HOE’s) for laser beam heat flow control. The advantage of a laser for annealing lies in its ability to selectively heat only the surface of the CdTe solar cell; improving energy efficiency, process speed and energy resilience. Heat transfer simulations were used to predict the effects of different laser irradiance profiles on the annealing process thermal cycle, influence the experimental design and predict optimal laser irradiance profiles. Variations in power and process speed on as-deposited and MgCl2-treated close-space sublimated (CSS) CdTe samples have been performed. The results were characterised using scanning electron microscopy (SEM), transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy (XPS). Optical properties were analysed with a spectrophotometer and ellipsometric spectroscopy (SE). The laser annealing treatment was found to be effective in promoting Chlorine diffusion and improving the optical and morphological properties of CdTe thin film devices.

Resume : To response to the world’s urgent need of sustainable energy production, photovoltaic devices, converting the energy of sunlight into electricity, represent one of the most promising avenues. In the search for cost-effective solar cells, solution-processable third-generation solar cells based on organic conjugated materials or organic-inorganic hybrid perovskites are key candidates, currently yielding power-conversion efficiencies of more than 20%. However, many of these organic and hybrid perovskite solar cells can harvest photons from the visible up to only about 800 nm. More than half of the energy of the solar spectrum (in the near-IR) cannot be absorbed. On this issue, lanthanide-doped fluoride upconversion (UC) nanocrystals (e.g. lanthanide-doped NaYF4), if properly incorporated in to the solar cell structure, can be a promising approach to harvest near-IR photons. Such UC process can be further boosted by plasmonic structures. In this work, we perform colloidal synthesis on lanthanide-doped NaYF4 nanocyrstals of different compositions. On solar cells based on methylammonium lead trihalide perovskites, we experimented different strategies to incorporate UC nanocrystals, plasmonic nanostructures, and/or both into the device architecture. Their impacts on the device performance and in particular on near-IR light harvesting will be discussed.

Resume : The surface Plasmon effect of noble metal nanoparticles on the photovoltaic properties of silicon solar cells was investigated. The metal nanoparticles were deposited on the p-type silicon base of the n+/p junction using a chemical deposition method followed by a thermal treatment at 500°C under nitrogen atmosphere. Chemical composition and surface morphology of the deposited metal were examined by energy dispersive X-ray (EDX) spectroscopy and scanning electronic microscopy (SEM). The effect of the deposited nanoparticles on the electrical properties was evaluated by the internal quantum efficiency (IQE) and current-voltage (I-V) measurements. The results indicate that the formation of the metal nanoparticles is accompanied by an enhanced light absorption and improved photovoltaic parameters.

Resume : The light absorption of planar(Pl) junctions can be improved using nanostructured top surfaces due to their enhanced light trapping properties. Nevertheless, associated with the higher nano-textured surface, the concentration of the recombination surface centers increases, leading to electrical performance degradation. We present comparative studies of the electrical yield of planar and nanostructured solar cells, passivated by SiNx-H or Al2O3 layers. Firstly we realized n - p junctions using (180 keV, 1018 cm-3) implantation of phosphorous in 2x1015 cm-3 boron p type silicon material. Monte Carlo simulation was first performed to optimize the junction profile for nanostructured surface. SIMS analysis are in agreement with the simulations. Dense and uniform arrays of silicon based nanopillars (NPs) and nanocones (NCs) were fabricated by self-assembled silica particles as etch masks. The reflectivity of the Pl, NPs and NCs surface decreases typically by a factor of 3 from 350 to 800 nm, the best one being associated with NCs. SiNx:H and Al2O3 passivation layers increase the electrical yield of the cells typically by a factor of 2 and 6 respectively. The best result is obtained for NPs based cells with a 5 % yield. Efficiency of the passivation layers is discussed versus the morphology of the surfaces. Acknowledgments: This work was partly supported the “GENESE” contract (Ref: 13-BS09-0020-03) from the Agence Nationale de la Recherche, ANR, . T. Xu acknowledges support from the National Natural Science Foundation of China (61204014).

Resume : Boron and phosphorous doped thin film Si solar cells were prepared by (effusion cell equipped) electron beam evaporation method. Doping levels, crystallization conditions, and the point defect centres were investigated by Time of Flight Secondary Ion Mass Spectrometry (ToF-SIMS), electrical properties measurements, Raman Spectroscopy, X-Ray Diffractometer, and temperature dependent electron paramagnetic resonance (EPR) spectroscopy. Especially, EPR spectroscopy gives detailed information about the electronic state of the paramagnetic system, point defect centres in Si-based photovoltaic semiconductor solar cells [1,2], changes in the crystal symmetry (reflected in the Landé g factor of the conduction electrons) [3], hyperfine interaction of the unpaired electron with nearby nuclei, etc. By means of X-band EPR characterization at different microwave powers, it was shown that, upon transition from amorphous to polycrystalline, the nature of the defect centres change from delocalized-like (with fast relaxation) to localized-like (with slow relaxation) electron behaviour. Both room temperature and low temperature (10-70K) measurements revealed crucial information about the changes in the electronic structure depending on the substrate, type of doping, and doping level which will help to construct solar cells with improved efficiency. [1] M. Jivanescu, A. Stesmans, and M. Zacharias, Journal of Applied Physics 104 (2008) 103518. [2] O. Astakhov, R. Carius, Yu. Petrusenko, V. Borysenko, D. Barankov, and F. Finger, Phys. Stat. Sol., 2 (2007) R77–R79. [3] G. Hendorfer and J. Schneider, Semicond. Sci. Technol. 6 (1991) 595-601.

Resume : Due to their controlled band gap, quasi two dimensional CdSe nanoplatelets (NPLs) are interesting alternatives to spherical quantum dots (QDs) as sensitizers in quantum dot solar cells (QDSCs). Also, from a modeling point of view, their well-defined thickness can provide a good base for building stacked models and establishing direct links between experiment and theory. In this study, we combine theoretical and experimental methods to analyse a lattice-mismatched NPL-oxide heterostructure. Experimentally, CdSe NPLs of different thicknesses have been linked to hydrothermally grown ZnO nanorods by SH- ligands. These systems were characterized by SEM, Raman and UV-VIS spectroscopy. The results confirm the presence of NPLs on the substrate, and suggest significant structural changes upon the interface formation between them. The latter was confirmed by a theoretical analysis using a low-cost and highly accurate periodic density functional theory-based computational method. We also calculated the vibrational and electronic properties of this system, and the electron injection efficiency from the NPL towards the ZnO surface. The photogenerated charge transfer, in line with the working principle of QDSCs, turned out to be favored, but only partial. The proposed combined experimental/theoretical approach should be applicable for other semiconductor heterostructures in a wide range of photovoltaic devices to help rationalize their operating principles and thus design new systems.

Resume : Metal-semiconductor Schottky diodes provide a way to separate photo-generated electron-hole pairs near the Schottky junction, which may be useful for collecting solar energy. These photo-excited electrons in metal can travel ballistically over or tunnel through the Schottky barrier into semiconductor to produce currents, which are respectively known as internal photo-emission (IPE) and field-emission (FE). The transport characteristics of these carriers are investigated in an Au-TiO2 Schottky diode via I-V measurements under photo illumination from visible to ultraviolet (UV). Based on the effective-mass approximation modeling, the IPE injected carrier concentration is related to the forward current and short current enhancements comparing to those in dark condition. An external bias can modify the IPE carrier concentration by tuning the available density of states in TiO2 for the hot carriers in Au to inject into. The FE injection becomes important under reverse bias. By fitting the I-V curves according to the WKB approximation, the estimated Schottky barrier height decreases nonlinearly as excitation power increases, which is significant under UV light illumination. The quick recovery of reverse current at the termination of irradiation indicates that the FE process is strongly assisted by the photo-excitation. The fast conversion between rectifying and ohmic behaviors may find applications in optically controlled electronic switches.

Resume : Solution based dip coating method was used to deposit Cu2ZnSnS4 thin films on Mo and/or ITO covered or bare glass substrates. The influence of pre-annealing and post annealing temperature as well as the influence of used substrate on the morphology, phase composition and elemental composition of the resulting films was studied. Copper (II) Chloride, Zinc Chloride, Tin (II) Chloride and Thiourea were used as precursors for synthesis of Cu2ZnSnS4 (CZTS) in methanol as solvent. Thin films were produced by repeated dipping of substrates into gelled precursor solution. The films were dried at 75oC and annealed in two steps. Pre-annealing of dried as-deposited thin films was performed at various temperatures (180oC, 210oC, 240oC, 270oC and 300oC) in air, nitrogen atmosphere and in sealed vacuum ampoules. Post annealing was done in sealed vacuum ampoules at 550oC. Thin films pre-annealed at 240oC had near-stoichiometric elemental composition while rest of the films showed different deviations from stoichiometric CZTS. Thin films deposited onto Mo and ITO substrates were more close to the stoichiometric CZTS than the films on glass substrates, probably due to more effective adhesion of metal thiourea complex gel with Mo and ITO substrates. The thickness of one time dipped, dried and annealed film was found to be around 300 nm. Raman spectra and EDS analyses of the thin films deposited onto Mo substrate pre-annealed at 240oC and heated at 550oC showed Raman dominating peak at 337 cm-1 which corresponds to Cu2ZnSnS4 and a composition close to the stoichiometric composition of Cu2ZnSnS4.

Resume : II-VI compounds have been of great interest to both the research and commercial solar cell applications because of their many advantageous properties in optoelectronic and photovoltaic technologies. In this family, CdS and ZnSe are the most popular n-type layer in semiconductor device applications. However, toxic risks of CdS causes environmental concerns in the large scale solar cell applications due to usage of Cd. Moreover, for ZnSe films, the nature of the defects in the structure, and common high resistivity problem obstruct their photovoltaic applications. Alloying polycrystalline wide band gap thin film semiconductors with elements in group III is a familiar application to optimize the resistivity of these materials. Among these, ZnInSe2 (ZIS) ternary semiconductors have been researched in various fields. In this work, the device properties of the fabricated n-ZIS/p-Si junction were analyzed in detail. This hetero-junction diode was fabricated by physical vapor deposition by using (111) mono-crystalline 600 μm Si wafers with the resistivity value of 1 - 3 (Ω.cm). Initially, the structural, optical and electrical properties of the ZIS thin film layer deposited on the soda lime glass substrate were analyzed. Then, detailed electrical characterization of the hetero-junction was performed by the help of temperature dependent current-voltage (I-V) measurements. The forward I-V behavior of the hetero-junction diode was investigated under the two possible current transport mechanism as thermionic emission and space charge limited current, in addition to determining basic device parameters from the ohmic region. The reverse I-V characteristics of the junction were found to be dominated by tunneling characteristic. The contribution of the p- and n-layer in the junction was studied under the spectral photo-response measurement.

Resume : The insertion of a superlattice (SL) structure in the absorbing layer of solar cell is paid attention for increasing more the conversion efficiency. Since, the built-in electric field in the SL structure disturbs the miniband formation and increases the carrier recombination, we investigated these effects by using a photoreflectance (PR) and a photoluminescence (PL) spectroscopies. We prepared two kinds of samples with different built-in electric field. The SL structures were both inserted in the i-layer of p-i-n GaAs solar cell samples and the n-layer of n-n GaAs structure samples. The barrier thicknesses were changed from 2 to 6 nm for discussing the miniband formation in more detail. For the PR modulus spectra, two critical energies were observed at 1.27 and 1.31 eV. We have already reported the peaks of low and high critical energies are due to the Γ and π point of miniband edges, respectively. However, the obtained miniband widths were smaller than the calculated value by a transfer-matrix method. This is because the built-in electric field in the SL structure disturbs the wave function overlap between the adjacent wells. On the other hand, we observed a peak around 1.30 eV from the PL spectra for all samples. This peak shifted to the low energy side as the barrier width became thinner. Since the PL peak energy corresponds to the energy at the Γ point in the miniband, this shift is explained by the expansion of the miniband width. At the same time, the PL intensity decreased with decreasing the barrier width. Supposing that the carriers are transported mainly by tunneling though the miniband in the SL structure, this is attributed to the increase of non-radiative recombination probability.

Resume : Chemical vapour deposited (CVD) graphene has attracted significant interest as a promising transparent top electrode due to its unique optical and electrical properties applicable for example in photovoltaics (PV) [1]. In comparison with conventional transparent electrodes, graphene has higher transparency, namely in the infra-red region, and low sheet resistance (down to 10–30 Ω/square for high quality layers). Therefore, CVD graphene can replace metal layer in Schottky-barrier based solar cells [2]. In this paper, we describe local PV properties of metal-insulator-semiconductor (MIS) Schottky barrier solar cell composed of CVD graphene transferred on hydrogenated microcrystalline silicon (µc-Si:H) with native silicon oxide insulating layer. We prepared series of graphene/µc-Si:H samples with different thickness of SiO2 and studied its influence on PV properties. Surface morphology and local photoconductivity maps were probed by conductive atomic force microscopy (C-AFM). Moreover, PV quality was characterized by I-V curves obtained locally by C-AFM as well as by standard macroscopic I-V measurements. As the results gained by both techniques are consistent, we have demonstrated the C-AFM to a valuable tool how to measure PV properties at the nanoscale. [1] Li et al., Adv. Mater. 22 (2010) 2743–2748. [2] Song et al., Nano Lett. 15 (2015) 2104–2110. This research was supported by Czech Science Foundation project 14-15357S.

Resume : Germanium nanoparticles or quantum dots (QDs) embedded in transparent dielectric matrix have properties radically different from the bulk semiconductor and present a great potential for application in electronic and optoelectronic devices. Due to quantum confinement the optical bandgap of QDs based materials can be tuned by varying the nanoparticle size. These properties may be exploited for the fabrication of nanoscale electronic devices or advanced solar cells. In this work we explored the structural properties of QDs based superstructures for advanced solar cells. Magnetron cosputtering was used for deposition, and upon suitable thermal treatment a superstructure of QDs was formed. The structural properties were explored by GISAXS/GIWAXS analysis. Both the GISAXS and GIWAXS techniques were used to obtain the size of the grown objects and in addition, the Porod tail was analyzed in the GISAXS pattern in order to obtain information on the layer close to the Ge QD / matrix interface. The interface transition Ge QD / matrix will be discussed. We shall show that such a layer affects the time resolved PL properties of the Ge dots.

Resume : The micro and nanostructured hyperdoped silicon, known as black silicon, is a highly absorbing surface with extended spectral sensitivity [1-2]. This material offers new opportunities and dramatically enhances the infrared sensitivity of silicon-based optoelectronic devices. However, and despite the potential increase in optical absorbance there is an important drawback caused by increase of the charge carrier recombination at the nanostructured surface [3]. This decrease in the lifetime of the carriers is mainly due to the defects created during the femtosecond laser irradiation of the material. A post-laser annealing (LA) of the nanostructured black silicon surface can reduce these defects [4, 5] by creating high crystallinity of the amorphous and polycrystalline surface induced in the black silicon fabrication [6]. Therefore, one important challenge is to increase the crystallinity of the black silicon structure, required to reduce these centres of recombination without reducing optical absorbance [7]. With this objective, this work shows a comparative study of Laser Annealing and simultaneous Laser Annealing and Plasma Immersion Ion Implantation (PIII) [8] process in order to solve the issue of surface recombination in black silicon solar cells, by providing good crystallinity and electrical performance. We demonstrate that either electrical or optical capabilities can be reached with LA and PIII. For this purpose several characterization measurements of the processed materials was carried out by means of Laser beam induced current (LBIC), four-point probe resistivity measurements, reflectivity and absorbance spectroscopy, Scanning electron Microscopy (SEM) and Energy Dispersive X-Ray Spectroscopy (EDX) microanalysis.

Resume : For now, commercial c-Si solar cells have a limited efficiency of about 25%; one of the reasons is the thermalization of high energy photogenerated carriers. Thus, a way for improving this efficiency is to introduce a frequency conversion layer in a cheap and compatible Si PV technology approach. The goal of our study consists in replacing the antireflective SiNx layer of commercial Si PV cell by a down converter layer based on a Ce3 -Yb3 co-doping and embedded in Bragg Mirrors. This system has been produced by reactive co-sputtering technique and their optical properties optimized through the deposition and annealing parameters. SiOxNy composition affects the 5d band of Ce3 ions which presents a wide and bright photoluminescence due to the 5d-4f transition. Excitation photoluminescence measurements show a wide excitation domain for the Ce3 ions ranging from 300nm to 400nm wavelength. Moreover, such an ion benefits from an efficient direct excitation of its 5d levels without any requirement of sensitization from the matrix. By introducing the Yb3 ions, we evidence an effective cooperative energy transfer with an internal quantum efficiency higher than 180 %. To improve the Ce3 -Yb3 coupling rate, Bragg mirrors on top and bottom of the doped SiOxNy layer have been designed. Simulations and experimental results are compared to define the optimized structure favoring the trapping of UV photons as well as the maximum of IR photons redirected to the solar cell.

Resume : AIIBVI thin film semiconductors based photovoltaic cells proved to be suitable candidates for both, terrestrial and space applications. Their wide bandgap, good chemical and mechanical stability and relatively low production price recommend these materials as important candidates for optoelectronic devices, especially for photovoltaic cells. Zinc sulfide (ZnS) thin films were deposited by rf-magnetron sputtering onto optical glass substrates. Deposition pressure was varied in the range 4×10-3 mbar – 1.6×10-2 mbar; other working parameters, like working power, substrate temperature and time deposition, were maintained constant. Optical absorption spectroscopy analysis showed a decrease of the bandgap energy of the deposited films with pressure increase. With transmittance higher than 70% for all deposited layers, ZnS thin films are suitable ”window” layers for AIIBVI semiconductors based photovoltaic cells. Structural and morphological features were investigated by X-ray diffraction (XRD) and scanning electron microscopy (SEM), and charge carriers’ mobility was determined by Van der Pauw method in the 30 – 300K temperature range. Keywords: ZnS, rf-magnetron sputtering, photovoltaic cell Acknowledgements: This work was supported by Romanian Executive Unit for Financing Higher Education, Research and Innovation (UEFISCDI) under PN-II-PCCA program, grant no. 288/2014.

Resume : ZnO thin films are frequently used as transparent layer for photovoltaic applications but a recent approach is to obtain ZnO based solar spectrum converters by rare earth doping. Among these rare earth dopants, a particular interest has been shown on neodymium (Nd) with its near-infrared emission at about 900 nm which is compatible with silicon solar cells. In this work Nd doped ZnO films were grown by pulsed electron beam deposition on Si, glass and and c-cut single crystal substrates at different substrate temperatures and under oxygen gas (10-2 – 2 10-2 mbar). We demonstrate that a simple way to tune the physical properties of Nd doped ZnO thin films is given by a precise control of the growth conditions. The composition and structure of these films were correlated to their electrical (resistivity, mobility and carrier density) and optical properties (UV-VIS absorption and photoluminescence in the near infrared domain of the Nd3 ions). The results show that a slight difference in oxygen pressure has drastic effects on the film properties, leading from transparent conducting to photon down-shifting thin films.

Resume : The energy band structure of highly mismatched alloys (HMAs) is strongly modified due to a band anticrossing (BAC) interaction of localized levels of isovalent dopants with the extended states of either conduction (CBAC) or valence (VBAC) band [PRL 82, 1221 (1999)]. This interaction leads to a formation of a narrow intermediate band in the semiconductor band gap. CBAC HMAs have been used to demonstrate an intermediate band solar cell concept [PRL 106, 028701 (2011)]. The key requirement for the IBSC operation is a presence of optical transitions between all bands. So far it has been only demonstrated for CBAC HMAs. Here we show a first observation of optical transitions between all VBAC split valence bands (VBs) and the conduction band of ZnO1-xSex HMA. We have studied optical properties of ZnO1-xSex alloys with up to 8% of Se. The ZnO1-xSex layers were grown on Al2O3 substrate by pulsed laser deposition technique [JAP 111, 113505 (2012)]. Using combination of optical techniques we clearly show that the observed 3 photoluminescence (PL) bands correspond to the transition from the conduction band to the lower valence subband and two spin-orbit split upper valence subbands. The results indicate relatively long lifetimes of photoexcited hole in the lower valence subband. The composition dependence of the PL peaks is explained by a 12x12 kp VBAC model describing the interaction between the VBs of ZnO and localized Se levels located 0.85 eV above the VB maximum of ZnO host matrix.

Resume : The n-i-p photovoltaic junctions based on CdTe/ZnTe has been studied for the presence of defects in their structure and their influence on the electrical properties of the diodes. The heterostructures were grown by MBE technique on the semi-insulating (100)-GaAs substrate. To study the origin of defects in the CdTe/ZnTe junctions the temperature photoluminescence (PL) measurements have been performed and deep level transient spectroscopy (DLTS) has been also applied. The analysis of PL experiments of the investigated samples revealed the presence of recombination centers in the absorber layer. Two types of optical transitions were observed in the PL spectra. Their activation energies of about 6 and 11 meV correspond to the free excitons and donor-acceptor pair radiative recombination, respectively [1]. The DLTS measurements confirmed the presence of recombination centers and several deep traps of activation energies equal to 0.22 eV, 0.45 eV and 0.78 eV. Their possible origin has been discussed. DLTS signal analysis let us assume that the deep traps are located in the absorber layer i-CdTe and at the i-CdTe/ZnTe interface [2]. Summarizing, the PL- and DLTS measurements allowed for characterization of defects present in the CdTe/ZnTe n-i-p junctions. It was found that these defects are responsible for the low efficiency of the studied solar cells. [1] J. Lee, et al., J. Appl. Phys. 78, 5669 (1995). [2] E. Zielony, et al., J. Appl. Phys. 115, 244501 (2014).

Resume : Silicon nanowire arrays have recently become widely studied substrates for thin film solar cells, approaching 10% conversion efficiency [1]. Improving the efficiency further requires improved control and understanding of the metal catalyzed growth of the wires and a subsequent absorber layer. We have studied the growth of silicon nanowires under various deposition conditions (temperature, plasma power, precursor concentration) in PECVD. Layers of catalytic tin or indium metal nanoparticles on Corning glass with sputtered Al:ZnO thin films have been prepared by vacuum evaporation and a dewetting process at elevated temperatures. Silicon nanowires typically with a thicknesses of tens of nm and lengths of microns have grown after few minute silicon deposition; however, we show that under certain deposition conditions nanowires with a thickness below 10 nm can be grown. The nanowire diameter is strongly connected to the size of catalytic nanoparticles. Control of the nucleation by the various approaches (e.g. using the spin-on metal nanoparticles and the graphene coating of the substrate) will be discussed. Deposition conditions as well as the thickness of original metal film have a substantial effect on the nucleation and growth of nanowire arrays as revealed by SEM. These effects are discussed with respect to the improvement of the performance of solar cells based on these nanowires. [1] Misra et.al., IEEE J. Photovoltaics. 5 (2015) Czech Science Foundation 16-12355S supported.

Resume : The most commonly used method for timely obtaining percolation threshold value pc is calculating the number of occupied lattice sites p when the probability of percolation P(p) equals 0.5. This is, however, shown to be unreliable particularly when simulating thin films. In this research, using a simple cubic lattice and discrete percolation models, we show how initial setup parameters and processes implemented in computer simulations of disordered systems can have a significant influence on representation of system characteristics. The most reliable method for obtaining percolation threshold is from the maximum of average cluster size curve (excluding all existent percolation clusters). Percolation probability value equal to unity (for p>pc) shifts towards zero when decreasing one spatial dimension (thin film lattice), while percolation threshold remains approximately equal to the theoretical value when obtained from the average cluster size curve, excluding all existent percolating clusters. This is due to the multiplication of probabilities of each percolating cluster’s appearance, as thin lattice made by multiplying the bulk 3D lattice allows the existence of multiple percolating clusters even for p

Resume : CuZnInSe3 compounds have been recently identified as promising material for PV application for their desirable optical and electrical properties,. Recently, a very promising device efficiency of 7.6% has been reported. Further development of these technologies strongly requires for a deeper understanding of the fundamental properties of these complex materials for the careful assessment of their crystalline structure, band gap, secondary phases, presence of defects and their dependence on the process parameters and layer composition. In this context, this work reports the detailed crystalline, optical and vibrational characterisation that has been performed by XRD, Raman scattering and PL on samples grown with different chemical composition ([Cu]/([Zn]+[In]) between 0.3 and 0.6). XRD shows narrow peaks with a pattern associated to a modified cubic Zinc-Blende structure. Raman spectra show a vibrational characteristic pattern similar to that from chalcopyrite CuInSe2 but with broader and distorted bands at 170–250 cm-1 region that are determined by overlapping of close frequency modes. The relative intensity of these peaks has a strong dependence on the chemical composition of the layers. Finally, temperature dependent PL measurements have also been performed to assess presence of defects in the layers, and the obtained data strongly suggest the existence of a direct bandgap that can be controlled in the 1.16eV to 1.32eV with chemical composition.

Resume : Cu2O is an interesting material for PV solar cell application mostly due to its direct band gap of 2.1 eV, making it a perfect companion to Si solar cells in a tandem structure. Combined, these can reach efficiencies of 41%. The highest efficiency solar cells (>5% of max 20%) utilizing Cu2O as the absorbing layer has been accomplished using thermal oxidation of Cu. Cu2O produced in such a manner is characterized by a high mobility close to single crystal values (~100 cm2/Vs). Although Cu2O is a low-cost material, for practical applications it would be desirable to use thin films instead of thick wafers. In this study we have investigated Cu2O films deposited by reactive magnetron sputtering. To achieve the best possible mobility in Cu2O films, many parameters have been optimized; oxygen partial pressure, deposition temperature, and power on target. However, the mobility remained fairly low, usually around 15-25 cm2/Vs. Solar cells with sputter-deposited Cu2O have so far shown a low degree of rectification and low efficiencies. It is possible that the low efficiency stem from a low mobility or poor crystallinity. Post-deposition rapid thermal processing (RTP) at moderate pressure of Cu2O samples shows a promising and remarkable increase in mobility, up to 53 cm2/Vs (950 °C, 3 min), while retaining a similar carrier concentration. RTP annealings may thus be a step towards higher efficiencies for Cu2O-based solar cells.

Resume : Amorphous/microcrystalline/crystalline silicon (a-Si:H/µc-Si:H/c-Si) triple junction solar cells can have an operating voltage large enough to directly split water, for which >1.6 V (depending on the catalysts) are needed. They may be an inexpensive alternative to high efficiency, high voltage multi junction solar cells based on scarce elements like III-V semiconductors. In addition to standard KOH pyramid texturing of the 140 µm thick n-type Cz wafers, we used a second acid-based polishing step to systematically smoothen the sharp V-shaped valleys. This way, we avoided defective regions in the 2 µm thick μc-Si:H middle and the 250 nm a-Si:H top cell. Quantitatively analyzing the surfaces by atomic force microscopy, we found that a reduction from 47° to 25° in the maximum of the surface angle distribution allowed an increase in open circuit voltage (Voc) of >130 mV and an increase in fill factor (FF) from 58% to >70%. By further optimizing the nanocrystalline silicon oxide (nc-SiOx:H) intermediate reflector layer placed between bottom and middle cell as well as the nc-SiOx:H top cell p-layer, we could increase solar to electricity conversion efficiency to 12.1% (short-circuit current density of 8.7 mA/cm², Voc = 1.95 V, FF = 70.4%), still limited by the middle cell current. The high current density at 1.6 V of >7.5 mA/cm² could allow high solar to hydrogen conversion efficiencies once proper catalysts are applied. We expect further rapid progress by thickness adjustments.

Resume : SnO2 as a standard TCO material have recently attracted special attention, thanks to the particular dual valency of tin and to the reversible transformation of the material that can occurs from p-type SnO to n-type SnO2 with excellent transport properties. Besides, SnO2 can be functionalized with Rare Earth elements (RE), which gives rise to new properties. In the field of solar cells, this class of materials can be used as conversion layer to adapt the incident solar spectrum to the solar cell absorption, reducing the losses due to carriers’ thermalization. In this work we present properties of SnO2 films co-doped with Nd (0.62 at.%) and Yb (1.3 at.%), elaborated by reactive magnetron sputtering. This co-doping process in SnO2 is reported for the first time. The films were elaborated at 100°C for the sake of compatibility with low temperature used for devices fabrication. The XRD analysis showed the phase transformation from amorphous SnO to crystalline SnO2 as a function of the oxygen gaz flow during elaboration. We gain better insight into the oxides proportions as well as the RE profile distribution through chemical analysis by XPS spectroscopy, which also showed that both REs (Nd and Yb) are well inserted in the structure and possess the 3+ valence state predicting that are optically active. Indeed, analysis by PL spectroscopy shows that under UV excitation of 325 nm, the SnO2:Nd:Yb films exhibit a wide and intense emission lines in the infrared region characteristic of Nd and Yb. Thanks to PLE measurements, an efficient energy transfer from the SnOx host matrix to each rare earth has been identified. In addition to that, second energy transfer was observed; Nd3+ ions are found to efficiently sensitize the Yb3+ ions. The films exhibit transparency laying between 75-90% with excellent transport properties, resistivities as low as 0,01 Ohm.cm and mobilities as high as 29,6 cm2/V.s were measured. Such optical and electrical results are of potential interest to solar cells using Nd and Yb co-doped SnO2 films as TCO and photon down shifter.

Resume : Nanometer sized metallic particles, such as Cu, Ag and Au, possess specific optical properties due to the presence of a plasmon band which corresponds to a narrow absorption band in the visible spectral range. The plasmons lead to modification of the properties of the adjacent dielectric material in which the nanoparticles are included. Controlling the chemical environment of nanoparticles and tuning their size and shape, one can modify the macroscopic properties of the host matrix in a controllable manner. In our work, the focus was on Cu nanoparticles, synthesized in or on silica, thus forming a composite material. We produced the samples by high vacuum thermal evaporation of a single Cu layer on top of Si substrate. Some of the samples were also capped with a thin SiO2 layer. The samples were deposited at different substrate temperatures (up to 180°C), and annealed ex situ in high vacuum at temperatures up to 550°C. Nanoparticles development was studied by Grazing incidence small-angle X-ray scattering (GISAXS) and also by Atomic Force Microscopy and Scanning Electron Microscopy when capping layer was not applied. It is shown that higer annealing temperatures result in more spherical Cu nanoparticles, while lower deposition temperatures further enhance their oblateness. Finally, the plasmonic effect was monitored by UV-Vis reflectance spectroscopy, while the oxidation of nanoparticles was further studied by photoluminescence spectroscopy.

Resume : A new process was developed to realise a multicrystalline silicon (mc-Si) solar cell. A O2/porous silicon-based gettering was applied to the mc-Si wafers. As a result an important enhancement of the minority carrier diffusion length L, the recombination velocity at grain boundaries Vr and the minority carrier life time was confirmed by The Light Beam Induced Current (LBIC) mapping. After gettering the oxidized porous silicon (ox-PS) was used as a mask. Micro-periodic fingers were opened on the porous silicon layer using a micro groove machining process to realise a highly doped regions n under the fingers and a local back surface field. The dark I–V curves show the decrease of the serial resistance and the enhancement shunt resistance. The LBIC mapping of the realized device, confirm the presence of a micro periodic local back surface field.

Resume : We present the development of high efficiency single and multi-junction solar cells, focusing on the stability against prolonged illumination. The degradation process, known as Staebler-Wronski effect, is considered as a major limiting factor for high efficiency cells and usually is associated with the stability of amorphous silicon (a-Si:H) absorber layers, while the microcrystalline silicon (μc-Si:H) absorber is generally more stable. We study the stability of single (a-Si:H) and multi-junction (a-Si:H and µc-Si:H based tandem, triple and quadruple junction) solar cells prepared by PECVD at deposition temperatures of 185°C or below. Solar cells were investigated by current-voltage (J-V) measurements under AM 1.5 illumination (intensity of 1000 W/m2, class A spectrum), quantum efficiency (QE) and reflectance measurements. Degradation of solar cell was performed at 55 °C with an intensity of 1000 W/m2 over a period of 1000 hours. High initial efficiencies were achieved for all types of solar cells: 12.4% in the case of a-Si:H single junction cell and above 13.5% for each type of multijunction cells. The degradation behaviour was evaluated with respect to the thickness and bandgap of the a-Si:H absorber layers. After 1000 hrs of illumination low degradation values of as little as 5 % from the initial value were found resulting in stabilized of degradation we achieved efficiencies of 11.8% in the case of tandem and triple solar cell, and of 12.1% in the case of quadruple cells.

Resume : Since the 50’s when the concept of the first semiconducting photovoltaic cell was practically proved, different materials and technologies were tested in order to increase the conversion efficiencies and to reduce the fabrication costs. Today world records conversion efficiencies for single junctions solar cells without concentrators are of 28.8% for thin film GaAs, of 25% for single crystal monocrystalline Si, 21.7% for CIGS thin films, 19.3% for perovskites cells, 13.4% for amorphous silicon thin films solar cells, 11.9% for dye –sensitized cells and of 11.1 % for organic solar cells. For the industrial development of different technologies many aspects should be taken into account: i) the efficiency, ii) the materials cost and iii) the life time of materials and solar energy devices. Due to the discrete band structure of semiconductors, only photons with energies equal or greater than the bandgap energy (Eg) will be absorbed and contribute to the electrical photovoltaic solar cell output. Photons having higher energies than Eg, even they are absorbed, their energies are underutilized due to the thermalization of charge carriers. In order to reduce these spectral losses and increase the energy conversion efficiency, many strategies were considered, such as: multi-junction cells (multiple semiconductors stacked cells), intermediate band semiconductors solar cells, up and down converters. Rare earth doped glass materials gives the opportunity to convert the incident photons wavelength and hence to increase or decrease their energies. In function of the nature of doping materials, the absorption of two low energy photons could contribute to the emission of one high energy photon (up-conversion), and also by the absorption of one high energy photon, one or two low energy photons could be generated (down-conversion). By converting the energies of photons which are not efficiently used, corresponding to the UV and respectively IR part of solar spectrum, the concept of “up and down conversion” is a possible way to increase the efficiencies of all classes of single junction solar cell. Pr3 /Yb3 co-doped ZBLA, Tm3 single-doped and Tm3 /Yb3 co-doped fluoride glasses based on ZLAG and ZBLA glasses showing a mechanism of down-conversion were tested as encapsulation glasses materials for monocrystalline silicon solar cells. The J-V characterizations were done under solar simulator irradiation. The influence of Yb3 and Tm3 concentration on the solar cells performances was investigated.

Resume : Monolayer molecular doping (MD) has recently attracted much attention as an easy and cost-effective method to form p-n junctions in silicon [J. C. Ho et al., Nature Nanomater. 7, 62 (2008)]. It consists in the immersion of the sample in a chemical bath containing dopant precursors molecules diluted in a solvent. During this process the molecules deposit from the liquid all over the exposed surfaces, chemically bond to the target surface with a self-limiting process ruled by their steric properties and work as dopant source during successive thermal annealing. The technique allows the n- or p- type doping by the proper selection of the precursor source [Garozzo et al., MATER SCI ENG B 178 (2013) 686] and it has been successfully applied to structured surfaces [R. A. Puglisi et al., Phys. Stat. Sol. A 210 (2013) 1564]. Recently the method has been successfully demonstrated on Si nanowire array with density of 2×10^10 cm^-2, average length of 500 nm and diameters up to 70 nm [R. A. Puglisi et al., Sol. En. Mat. Sol. Cells 132 (2015) 118]. The doped nanostructures have been then integrated in complete solar cell devices exhibiting photovoltaic properties thus showing promising results for this easy and low cost method to be applied in the next generation of solar cells. We now present results of a study performed on MD applied to solar cells, where different contacts design and materials and passivation processes have been investigated. It is found that the optimized Si nanowire based solar cells present short circuit current as high as 9.3mA/cm2 under 1Sun of solar irradiation.

Resume : Copper Zinc Tin Sulphide (CZTS) nano-powders were synthesised by solution based method in oleylamine at 215oC, 225oC and 235oC. The precursors used were copper pentanedionate, zinc acetate, tin (II) chloride and elemental sulphur. The influence of initial Cu/ (Zn +Sn) concentration ratio on the morphology, phase composition and elemental composition of nano-powders was studied. For every synthesis temperature the initial composition of precursors was varied by changing the Cu/ (Zn +Sn) concentration ratio (i.e. 0.83, 0.76, 0.62, 0.51 and 0.36). SEM analysis confirmed that the formation of nano-particles and/or colloidal nano-particles is depending on the initial composition. The CZTS nano-powders synthesised at 215oC, 225oC and 235oC with decreasing initial Cu/ (Zn + Sn) concentration ratio show increase in the size of colloidal nano-particles from 20 nm to 10 µm. EDS analysis revealed that the elemental composition of the formed nano-powders was stoichiometric CZTS with initial Cu/ (Zn +Sn) concentration ratio (i.e. 0.83 and 0.76) and for other concentrations the elemental composition of the formed nano-powders was close to the used initial composition. Raman analysis revealed that the phase composition of nano-powders depends on the used synthesis temperature and on the initial composition of precursor mixtures. By Raman analysis the nano-powders synthesised with the initial ratios of Cu/ (Zn + Sn) equal to 0.83 and 0.76 at 235oC show Raman peaks characteristic to CZTS.

Resume : Measurements of the photovoltaic (PV) activity of solar cells under low-light conditions and different wavelength-selective filters can supply useful information for the nature, the location and the activation of defects and recombination mechanisms inside the heterojunction and its interfaces. In this work we use this approach for the characterization of crystalsol monograin membrane kesterite (CZTS) solar cells. The temporal dependence of the open circuit voltage, short circuit current, fill factor and conversion efficiency, under low intensity of white, red, green and blue light is studied and related to carrier recombination processes in the bulk of the absorber and its interfaces with the CdS buffer and the electrodes. The low-light behaviour of the cells is discussed in relation to their PV performance under standard test conditions (1000 W/m2) and their external quantum efficiency. Reversible and irreversible contributions to the light-soaking effects are further analyzed. Finally, photoluminescence and electroluminescence imaging was used to elucidate the spatial homogeneity and quality of the monograin cells.

Resume : The kinetic compensation effect or Meyer-Neldel exists as a fundamental property of strongly related processes. It is therefore often used in the electric characterization of semiconductors to evaluate if detected signals have a similar origin. In the past this has for example led to a criterion to evaluate if an anomalous signal capacitance spectroscopy signal can be assigned to the famous N1-signal in CIGS solar cells. In this work we calculate the probability to detect a thermally activated signal in admittance and deep level transient spectroscopy as a function of the activation energy and the pre-exponential factor. Based on these calculations we compare probability of finding a kinetic compensation effect in the spectra with the ideal situation where all signals have the same probability of being detected. We conclude that great care should be taken when using the kinetic compensation effect as an argument that signals have a similar (the same) origin.

Resume : High-temperature solar cells are possible with Photon-enhanced Thermionic Emission (PETE) devices, which represent a novel and very attractive concept for the exploitation of solar radiation, especially if concentrated, characterized by a promisingly high conversion efficiency. PETE converters rely on the concept that engineered semiconductor structures can provide a hot-electron emission, induced by photons with sufficient energy to produce charge couples, combined to a thermionic emission, sustained by the high temperatures induced by every thermalization process. Surface nanotexturing combined to surface-hydrogenation, aimed at achieving negative electron affinity conditions and a work function as low as 1.7 eV with an emitting-layer nitrogen-doping, are here proposed as a radically new and potentially efficient PETE cathode completely based on chemical-vapour-deposited (CVD) diamond, able to ensure an efficient thermionic emission at temperatures up to 780 °C. CVD diamond is transparent to solar radiation due to its wide bandgap, consequently advanced and novel techniques are needed for processing diamond to become an efficient sunlight absorbing material (i.e. black diamond). Surface texturing by fs-laser, boron-implantation, buried and distributed graphitic structures and other technological steps allow for the fabrication of an innovative defect engineered diamond cathode to be efficiently exploited for the conversion of concentrated solar radiation. A diamond-based anode with an even lower work-function can be combined to the efficient black diamond cathode.

Resume : There has been an increased interest in the scattering properties of plasmonic metal nanoparticles to enhance light trapping in opto-electronic devices, such as thin film solar cells. In most cases the nanoparticles are self-assembled over a transparent conductive oxide (TCO) layer of the cells structure. However, until now, little is known about the influence of the substrate (typically glass + TCO) properties on the morphology of the nanoparticles formed. As such, this work presents a complete morphological and optical study of a series of silver nanoparticle structures fabricated on distinct oxides relevant for solar cells application. The results of such comparative study reveal that the TCO conductivity and its surface roughness are key factors that control the morphology of the nanostructures. Therefore, the tuning of such properties allowed the production of remarkably uniform silver nanoparticles with the required sizes (100-300 nm) for efficient light scattering. In addition, a novel and fast method of fabricating highly reproducible plasmonic surfaces is explored, employing a rapid thermal annealing process.

Resume : Semiconductor nanocrystals (known as quantum dots (QDs)) have been the subject of great scientific and technological interest due to their potential optical applications that include photovoltaics, light emitting diodes (LEDs) and lasers. ZnSe and CdSe based quantum dots (QD) are semiconductor nanocrystals that possess unique optical properties and they are the potential candidate for next generation LEDs and photovoltaic devices. Their emission color can be tuned from the visible throughout the infrared spectrum. In this work we report a synthetic route to prepare type-I and type-II heterostructure ZnSe/Zn(Cd)S and CdSe/Zn(Cd)S colloidal core/shell spherical quantum dots, respetively. The synthesized ZnSe/Zn(Cd)S and CdSe/Zn(Cd)S heterostructure core/shell nanocrystals were characterized by using the x-ray diffraction (XRD) for structural properties and UV absorption and fluorescence techniques for optical properties, respectively. The effects of lattice mismatch induced interface strain on the first exciton energy, conduction and valence band energies and diameter of the ZnSe/Zn(Cd)S spherical core/shell quantum dots were calculated by using the effective mass approximation and nearest neigbour sp3s* tight binding theory. It is shown that the interface strain decreases the core diameter in CdSe/ZnS and ZnSe/ZnS type I heterostructures and increases it in the case of the ZnSe/CdS type II heterostructures.

Resume : Knowledge about the specific role of electronic transitions in photo-active materials and their interfaces is of great interest for the further development of solar cells based on thin films and nanocomposites. Transient surface photovoltage (SPV) spectroscopy is proposed as a method allowing the time and spectrally resolved study of electronic states involved in spatial charge separation. In this method, SPV transients are measured at numerous wavelengths and the SPV signals at fixed times are converted into spectra. Examples will be shown for different photo-active materials and their combinations. The separation between shallow and deep electronic states in relation to the relaxation over a space charge region or a very thin defect layer will be demonstrated.

Resume : Separation in space and localization of photogenerated charge carriers can be investigated by surface photovoltage (SPV) techniques. The in-phase and phase-shifted by 90° signals or, in equivalence, the amplitide and phase angle of modulated SPV spectra contain information about the dynamic behavior of processes of charge separation and relaxation of a given sample. However, a detailed analysis of modulated SPV signals is usually not straight forward since localization of different charge carriers can last over different orders of magnitude in time. A model has been developed for the simulation of modulated SPV signals. The model is based on a random walk of a charge carrier across localized states and its interaction with delocalized and recombination states. The correlation between the SPV amplitude and the phase angle is discussed and compared with examples. The method will allow a deeper analysis of localized states at charge-selective contacts.

Resume : Gadolinium oxide is a promising material for the conversion of solar radiation since it provides the possibility of doping TR3+ ions in a wide range of concentrations and creating a donor -acceptor pairs for down- and up-conversion processes. The efficiency of these processes depends on a lot of factors particularly on the structural- phase composition, dispersion, purity of host lattice. The aim of this work is to determine the role of the Gd2O3 matrix intrinsic and impurity-related defects in the excitation energy transport mechanisms. To this purpose we carried out a detailed investigation of micro- and nanostructured Gd2O3, both nominally pure and doped with TR3+ ions by using luminescence and thermo-activation spectroscopy methods. Both intrinsic lattice defects and impurity-related defects were detected and the kinetic parameters of the charge trapping centers were determined. These results allowed us to discuss the nature of the defects states and the effective mechanism of energy transfer processes from host Gd3+-ions to impurity TR3+-ions. The contribution of band charge carriers in energy transfer processes are discussed.

Resume : The goal of this work is to demonstrate that spray pyrolysis technique for deposition of ITO layers can be effectively used for the cost effective processing of ITO/n-type Si based solar cells. This approach does not require formation of an emitter. However, special treatments of n-type Si substrates, to prepare a buffer layer between ITO and Si are still required. Two approaches for the formation of such buffer layers have been considered: (i) chemical etching, which results in a formation of a thin Si porous layer and (ii) low- temperature (~400 ∘C) anneals, which result in a formation of an thin (~1 nm) SiOx layer. High resolution electron microscopy has been used for the analysis of individual ITO layers as well as ITO/Si interfaces. It is established that formation of an ultra-thin SiOx layer at the ITO/Si interface is the most promising route to fabricate heterojunction ITO/Si solar cells with the reasonable (above 14%) efficiencies at a low-cost. Further steps for optimization and improvement of the low-cost processing routes for ITO/Si solar cells will be discussed.

Resume : structured active or absorbing layers of solar cells, including photonic crystals and wire arrays, have been increasingly explored as potential options to enhance performance of thin film solar cells because of their unique ability to control light, using the Rigorous Coupled Wave Analysis (RCWA). The designs are tested in four relevant materials: amorphous silicon (a-Si), crystalline silicon (Si), gallium arsenide (GaAs) and indium phosphide (InP). The parameters for photonic crystals are optimized through computer simulations to obtain the maximum absorption and path length enhancement. We investigated the performance of the considered structure and determined the geometrical parameters that allow a better absorption.

Resume : Zinc indium selenide (ZnIn2Se4) thin films have wide applications in solar cells and optoelectronic devices [1,2]. ZnIn2Se4 is n-type ternary chalcogenide semiconductor of type AIBIIXIV, where A=Zn, Cd, or Hg, B=In or Ga, and X=Se, S, Te [3–14]. In and Ga are rare elements and they can be replaced by earth abundant elements such as Zn or Sn. In this work, in the interest of window layer for possible photovoltaic applications with chalcopyrite and kesterite structures, the device properties for the Zn-Sn-Se thin films were studied. ZnSnSe2 (ZTSe) thin films were deposited by using single crystalline powder of sintered ZnSnSe2. The deposition process was carried out by means of e-beam evaporation on the p-Si substrate and keeping them at the substrate temperature of 200°C. Current-voltage (I-V) measurements and frequency dependent capacitance-voltage (C-V) measurements were carried out to investigate its electrical characteristics. Device parameters such as diode ideality factor, barrier height, series and shunt resistances were obtained and the possible conduction mechanisms were analyzed. The series and shunt resistances were calculated as 3.98x10^2 and 2.71x10^6 Ω respectively. The diode ideality factor evaluated approximately as 2.7 and the barrier height found as 0.77 eV. From C-V measurements, the built-in potential and the acceptor- donor level densities were determined. [1] F.J. Gracia, M.S. Tomar, Thin Solid Films 69 (1980) 137. [7] J. Fillpowicz, N. Romeo, L. Tarricone, Solid State Commun. 38(1980) 619. [3] L. Gastaldi, M.G. Simcon, S. Vitivoli, Solid State Commun. 55 [4] J.A. Buan, R. Nitsche, M. Lichtensteiger, Physica 27 (1961) 448.985) 605.

Resume : The solar industry has been dominated by crystalline silicon (c-Si) solar cells despite the relatively high cost of silicon (~40% of the total module cost). One possible path to decrease the cost of c-Si solar cells is to reduce the thickness of silicon to a few tens of micrometers. For such thin films, superior light-trapping schemes are required to overcome the small absorption coefficient of silicon. Recently, various light-trapping structures operating in the wave optics regime have been introduced to boost the absorption in thin-film c-Si. However, a multiphysics optimization considering both optical absorption and carrier collection efficiencies has not been previously studied in detail. Here, we describe simulations we performed to show the trade-offs when designing advanced light-trapping structures for thin silicon films. Our simulation is capable of quantifying losses due to Shockley-Read-Hall at the bulk and surface, and Auger bulk recombination, as well as optical and contact losses. We found that efficiencies above 20% can be achieved using thin-film (<10 m) c-Si solar cells. Additionally, we simulated thin-film c-Si solar cells with an interdigitated back contact scheme, and demonstrated a super-linear dependence between back contact separation and minority carrier lifetime. Finally, we calculated the optimum height to period ratio of light-trapping structures to be around 0.7 for a fixed period of 700nm that can be obtained in silicon using well-known potassium hydroxide anisotropic etching. The effect of surface recombination was investigated in detail both for the front and back surfaces for this optimum dimensions. It was observed that there was up to 1% efficiency difference between cells which has surface passivated at 5 cm/s and 100 cm/s surface recombination velocity, and up to 3% difference between 5cm/s and 1000 cm/s. These multiphysics simulation results can provide design insights for flexible and high efficiency thin silicon solar cells.

Resume : 2D Molybdenum disulfide layers are of considerable research interest for a variety of optoelectronic applications due to the possibility of tuning the optical and electrical properties by controlling the number of MoS2 layers. Bulk MoS2, a p type material, shows an indirect band gap of 1.2 eV and when prepared in monolayer form it exhibit n type conductivity and direct band gap of 1.8 eV. ZnS is an n type wide band gap material and is used in a number of optoelectronic applications. In this study, heterojuctions of 2D MoS2 layers and ZnS thin films are prepared by combining radio frequency magnetron sputtering technique and sulfurization process. Depositions of ZnS and Mo films have been carried out sequentially followed by sulfurization of Mo coated ZnS thin films in a 2 zone tube furnace. Mo deposition time and amount of sulphur are varied to obtain bulk, few and monolayer of MoS2 on ZnS thin films. Structural, optical and surface photo voltage studies are performed by using x-ray diffraction (XRD), Raman, spectroscopic ellipsometry (SE), photoluminescence (PL) and Kelvin probe force microscopy (KPFM) techniques. XRD, Raman, SE and PL techniques show signatures of both ZnS and MoS2 (bulk, few and monolayer), confirming the formation of heterojunctions. Surface potentials studied under dark and light conditions using KPFM show charge transfer at the interfacial regions under illumination. Surface photo voltage in p MoS2/ n ZnS was observed to be higher than the n MoS2/ n ZnS hetero-interface.

Resume : The conversion efficiency of Si single-junction solar cells is fundamentally limited to <30%. Adding a wide-bandgap top cell with Eg=1.6-1.9eV to the Si cell reduces the themalization losses in the short wavelength region and enables theoretical efficiencies up to 38%. Suitable top cell materials are III-V semiconductors and organic-inorganic metal halide perovskites. Both material systems allow a wide range of bandgap energies; however, their application in photovoltaic devices is at different levels of evolution. Solar cells made of III-V semiconductors are wellestablished for the power generation of satellites and they have been successfully applied in terrestrial concentrator system. In contrast to these relatively expensive devices, perovskite solar cells are a promising low-cost approach, but have only been fabricated on a small research scale and their long-term stability is still under investigation. In our talk, we will discuss the development of Si-based tandem solar cells with both perovskite and III-V semiconductor top cells. A new record one-sun tandem cell efficiency of 29.8% was achieved with a mechanically stacked GaInP/Si device. The rear-heterojunction GaInP top cell and Si heterojunction bottom cells were fabricated independently at NREL and CSEM, respectively, and combined using an optically transparent, electrically insulating epoxy. The 4-terminal tandem cell device structure has been optimized by simulations, and the effect of luminescent coupling between the subcells has been investigated both experimentally and theoretically. PV-Lab’s monolithic perovskite/Si tandem cells will also be discussed; we have recently achieved an efficiency of 21.2% under one-sun, which will be further optimized.

Resume : Nowadays, the best solar conversion efficiencies have been reached thanks to multijunction solar cells consisting of a stacking of III-V semiconductor single junctions on GaAs or Ge. While displaying record efficiencies, these solar cells suffer from the cost of such substrates. Therefore, our strategy is to develop a tandem cell on Si, in order to benefit from both the low cost and technological maturity of Si cells. Indeed, it has been shown that a tandem cell consisting of a 1.7 eV bandgap material on a 1.1 eV Si cell would reach efficiencies up to 37%1. To this aim, we use GaP, grown by MBE, which is quasi lattice-matched with Si. In addition to reaching a perfect lattice matching with Si and improving its optical properties, As and N incorporation in GaP, leading to a GaAsPN absorber,2,3 reduces the bandgap from 2.3 eV to the required 1.7 eV. We have obtained first stage results on GaP/GaAsPN/GaP top pin cells on both GaP and Si substrates, displaying Voc of 0.9 V, and 0.7 V respectively, and Jsc of 3.77 mA/cm² and 3.69 mA/cm² resp. On GaP, a 2.3% conversion efficiency has been obtained. Moreover, a Si(n)/Si(p) tunnel junction, suitable for electrical connexion between the top and bottom cells, has also been obtained. Finally, first-stage development of complete GaAsPN/Si tandem cells is shown. 1Geisz, J. F., and D. J. Friedman, Semicond. Sci. Tech. 17.8 (2002): 769. 2S. Ilahi et al, Sol. Energ. Mat. Sol. C 141 (2015) 3S. Almosni et al, Sol. Energ. Mat. Sol. C 147 (2016)

Resume : III-V/Si(100) tandem absorber structures are promising for high-efficiency PV and direct solar water splitting [1]. Metalorganic vapor phase epitaxy (MOVPE) of such structures commonly involves arsenic, either supplied directly via precursors (e.g. TBAs) or in form of residuals. Here, we study the influence of As on Si(100) surfaces and GaP/Si(100) quasisubstrates in situ during MOVPE [2]. Annealing of Si(100) in TBAs and background As results in an As-modified surface with a characteristic in situ reflection anisotropy spectroscopy (RAS) signal. We show that its spectral features emerge at different stages of a two-step annealing process. LEED patterns of the final surface show a preferential A-type, (1x2) reconstructed surface with dimer rows in parallel to the step edges. These are also clearly visible in STM images. XPS evidences the presence of As at the surface, but also atomic exchange across the interface. After subsequent pulsed GaP nucleation and pseudomorphic GaP heteroepitaxy, we achieve atomically well-ordered GaP/Si(100) quasisubstrate surfaces, which are free of antiphase disorder. The sublattice orientation of the GaP film is inverted compared to GaP grown on H-terminated Si [3]. The atomic structure of the heterointerface is more complex than in the abrupt Si-P case [3] for As-free systems. Ref's: [1] M.M. May et al., Nat. Commun. 6, 8256 (2015). [2] O. Supplie et al., APL Mater. 3, 126110 (2015). [3] O. Supplie et al., J. Phys. Chem. Lett. 6, 464 (2015).

Resume : We report on the electron beam induced current (EBIC) investigation of GaN nanowires grown on n-type Si (111) substrates. The objective of this study is to acquire information about the modifications of the substrate properties induced by the wire growth in view of the future development of nitride-on-silicon tandem photovoltaic devices. We show that the growth procedure, using a deposition of an ultra-thin AlN film prior to the nanowire growth step, leads to the formation of a p-n junction in the Si substrate with a high surface conductivity. The induced p-n junction exhibits a photoresponse over the spectral range from 360 to 1100 nm. The properties of the induced p-n junction are investigated in cross-section and top-view configurations with EBIC microscopy. For a localized contact of the GaN nanowires , the induced current collection range in Si extends over a few millimeters . The treatment of the surface using reactive ion etching with a CHF3 plasma leads to the inhibition of the surface conductivity and to the appearance of an S-shape in the current-voltage characteristics under illumination. Nevertheless, the conversion efficiency of the plasma-treated sample under AM1.5G solar spectrum is estimated to be in the 2.1-2.7% range.

Resume : The law governing the radiation emitted by a blackbody has been written by Planck. Further developments rooting on this law yielded a somewhat generalized equation for the light emitted by semiconductors where electrons and holes were characterized by Fermi-Dirac distributions at the same temperature but different quasi-Fermi levels. Here we further generalize a non-equilibrium generalized Planck's law accounting for different electron and hole temperatures, different from each other and different from the lattice temperature of the material. With an analytical expression of the absorption coefficient this non-equilibrium generalized Planck's law has been successfully used for fitting photoluminescence spectra recorded at different incident photon fluxes for an InGaAsP multi quantum well structure. Whereas the two carriers are heated during the experiment, electrons with lower effective mass are hotter than holes. The analysis of electrochemical potential variations shows a clear diminution for both carriers at high incident photon flux, which is more important for the holes. This work shows for the first time the relationship between individual carrier temperatures and the emission temperature usually measured with the classical expression of the generalized Planck's law. Moreover the individual absolute electrochemical potentials can also be experimentally extracted. We believe that these results open new perspective for hot carrier solar cell research. More generally, this work sheds light on new approaches for both photovoltaic and thermoelectric research. Planck, Vorlesungen über die Theorie der Wärmestrahlung, 1906 Lasher & Stern, Phys Rev 133, 1964; Würfel J. Phys Chem C (15) 1982 J. Rodiere et al., Appl. Phys Lett. 106, 183901 (2015 F. Gibelli et al. Phys. Rev. Applied, 5 (2), 2016 F. Gibelli et al., J. Phys. Cond. Mat, Submitted

Resume : Epitaxial lift-off is used to create thin-film III-V solar cells without sacrificing the GaAs wafer. It is based on selective etching of an AlAs release layer between the wafer and the cell structure using an HF solution. The wafer can be reused thereby reducing the cost of the cells. The thin-film cell can be transferred to any new carrier, e.g. glass, plastic, silicon or metal foil and offer new cell applications based on their flexibility, low weight and possibility to deposit the cell structure in reverse order. Moreover, thin-film cells with a back-contact that also acts as a mirror yield efficiencies surpassing that of the regular wafer based III-V cells because of an increased photon recycling. In the presentation the influence of junction depth in III-V cell structures will be discussed. Typical III-V cells employ a shallow junction design but our study indicates that for GaAs as well as InGaP cells, a deep junction close to the back of the cell structure performs better than a shallow junction. At the maximum power point the deep junction cells operate mainly in the radiative recombination regime, while in the shallow junction cells non-radiative recombination is dominant. The steeper slope of the IV curve boosts the fill-factor by 3-4%, which is thereby the most improved cell parameter. In order to minimize collection losses in the upper part of the cell, the optimal thickness of the GaAs deep junction cell is only two-thirds of a shallow junction cell. The associated lower cell current is more than compensated by the higher fill-factor and open circuit voltage yielding efficiencies up to 26.5% for a GaAs cell on its native substrate. In the presentation the implications of the deep junction cell design for thin-film cell performance will be discussed.

Resume : With an overwhelming market share of more than 90% in the sales of solar modules, crystalline silicon seems to be the ideal photovoltaic material. However, its success does not exempt silicon technology from the urgent need for further improvements. High efficiency devices like the PERL cell of UNSW or the HIT cell of Panasonic achieved a remarkable level of maturity, but their fabrication is too complicated for mass production on GW scale. Yet, they define ways for the improvement of current mainstream technology. One of them is further thinning of the wafers for reduced material consumption and, assuming excellent surface passivation, for improved open circuit voltage. The optics of the device must be developed accordingly in order to ensure a full absorption of the incident light in continuously reducing wafer volumes. Standard texturing methods are based on anisotropic etching of facets on a length scale of tens of micrometres which are suitable described by tracing of individual rays. In thinner cell designs the feature size will shrink accordingly, pronouncing the effect of scattering centres over refraction at planes.

Resume : Liquid-phase crystallization (LPC) of silicon is an emerging technology for high quality thin-film silicon solar cells on glass substrates. Using this bottom-up approach avoids handling issues associated with thin wafer technology. With 10 µm thin silicon layers with wafer equivalent morphologies, open-circuit voltages well above 600 mV could be achieved, but light absorption and thus short-circuit current densities are still limited [1]. However, the implementation of strong substrate textures for light trapping has been shown to be detrimental for solar cell performance [2]. Therefore, it is highly desirable to design interlayers that act optically rough, but ensure a smooth silicon interface in order to preserve a high material quality. Here, we present SMooth Anti-Reflective Three-Dimensionally textured (SMART) substrates which consist of SiOx-TiOx photonic crystal lattices for LPC silicon solar cells. The structures are fabricated by combining nanoimprint-lithography and spin-coating of silica and titania sol-gels. The optical response of such SMART structures is optimized experimentally and numerically, demonstrating the capability to outperform standard antireflective layers. First SMART substrates show a promising morphology and LPC compatibility and reveal the strong potential to achieve excellent optical properties while maintaining high silicon electronic material quality. [1] Haschke et al., SOLMAT 128, 190 - 197 (2014) [2] Preidel et al., JAP 117, 225306 (2015)

Resume : Record single-junction solar cells are made of 1-2 µm thick GaAs absorbers with short-circuit current (Jsc) close to 30 mA/cm². A thinner absorber would lead to rapid crystalline growth and potential cost-reduction. However, efficient light-trapping is needed to maintain high optical absorption over the whole solar spectrum, and the solar cell architecture has to be revised for optimal charge collection in ultrathin layers. We investigate 200 nm thick GaAs solar cells with a nanostructured back mirror. They exhibit broadband multi-resonant absorption. We have carried out numerical calculations based on RCWA in order to explain the physical origin of resonant modes and to optimize the structure. We predict Jsc up to 26 mA/cm². A Voc improvement is also expected thanks to photon recycling. The nanostructured silver back mirror is fabricated by Nano-Imprint Lithography (NIL), and combined with localized ohmic contacts to ensure efficient charge collection while keeping the high reflectance of the silver back mirror. The GaAs solar cells are subsequently transferred on a glass substrate. Multi-resonant photocurrent spectra are characterized by Fourier-Transform Photocurrent Spectroscopy (FTPS) with high spectral resolution. Jsc up to 22 mA/cm2 is experimentally demonstrated.

Resume : Si-based solar cells remain the main source of solar energy in the world due to low cost of material and relatively simple, industrially well-established technology. In spite of significant progress in increasing the efficiency of these devices, for most of industrially manufactured silicon solar cells’ the efficiency remains around 20%. One of the promising ways to minimize optical absorption losses for high efficiency Si-based photovoltaic device is photonic design of the front surfaces. The surface texturing causes light trapping into the device, due to light interaction with micro- nano-structured surfaces, thus increase in effective light path. However, the most of proposed methods for Si surface modification do not match industrial production requirements: such as fast manufacturing, environmental issues, and low cost per watt for final devices. Recently, it was developed a new method, Nonlinear Laser Lithography (NLL) for laser-induced periodic surface structuring, which allows nanostructuring of indefinitely large areas of many materials by excellent periodicity and uniformity. Given that NLL is a low-cost, single-step and high-speed technique, there is significant potential for solar cell applications. In present work we report the creation of photonic structure on Si-based solar cells by using NLL technique. As a laser source we used home-made 1 MHz femtosecond laser system, which can generate up to 1 μJ of pulse energy with ~120 fs of pulse duration. The laser beam was focused on Si surface and raster scanned with galvo-scanner to obtain area 1x1 cm2. Much larger areas were obtained by translating the sample with translation stage under galvo-scanner. In our work the structured area was 3x3 cm2. From SEM analysis the period of the structure is found to be around 800 nm. The sample was single crystalline p-doped Si wafer, with 4 Ω·cm resistivity and thickness of 275 μm. The solar cells were produced on the same Si wafer for both structured and unstructured areas under the same production conditions. The p-n junction was formed by phosphorous diffusion in doping furnace. The final step was back and front surface metallization in order to form electrical contacts. In conclusion, we proposed and experimentally demonstrated the recently invented NLL technique as a low-cost, high-speed, single-step technique for texturing front surface of Si-based solar cells periodically. After first experiments, the efficiency value of 10.2% were measured for control Si solar cell upon which texturing were not applied. On the other hand, surface textured Silicon solar cells by NLL performed with 12.4% efficiency which shows that the fast and cheap texturing with NLL obviously is efficient in increasing the solar cell efficiency for Si-based solar cells.

Resume : Absorption cross-section (ACS) of silicon nanocrystals (SiNCs) is determined via two completely independent approaches: (i) Excitation-intensity-dependent photoluminescence (PL) kinetics under modulated (long square pulses) pumping and (ii) absorbance measurements combined with morphology information obtained by the high-resolution transmission electron microscopy. This unique comparison reveals consistent ACS values around 10^(-15) cm^2 for violet excitation of SiNCs of about 3–5 nm in diameter and this value is comparable to most of direct band-gap semiconductor nanocrystals; however, it decreases steeply towards longer wavelengths. Moreover, we analyze the PL-modulation technique in detail and propose an improved experimental procedure which enables simpler implementation of this method to determine ACS of various (nano)materials in both solid and liquid states [1]. Our experimental results are compared to available literature data, including theoretical calculations. The results show that we can actually gain absorption per Si volume by forming nanocrystals but the absorption per total nanocomposite (NCs and surrounding matrix) volume is lower than for bulk Si. Therefore we should prepare more complex nanostructures like nanowires or plasmonic structures in order to address applications of Si-nanocomposites in photovoltaics. [1] J. Valenta et al. Appl. Phys. Lett. 108 (2016) 023102.

Resume : Methylammonium Lead Iodide (CH3NH3PbI3) organic-inorganic perovskite films are a promising absorber material with solar cell efficiencies now in excess of 21%. A significant appeal of these materials is their facile 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.

Resume : Defects in printed solar cells such as organic and organic-inorganic perovskite photovoltaics have a significant impact on the device performance, which is critical for the commercial viability of large-area manufacture. The characterisation of the various types of defects that occur in these devices is, therefore, important for identifying those which are critical to device performance and must be controlled during manufacture. This work uses a combination of dimensional surface measurements and functional mapping in order to relate the physical properties of various defects to their impact of photovoltaic properties. In particular, we demonstrate spatially resolved (~ 50 µm resolution) transient photovoltage and photocurrent measurements as a probe for local variations in charge recombination and extraction rates. Reference samples are prepared by depositing silica microspheres prior to deposition of the active layer to simulate dust contamination. These result in a local reduction in the steady state photocurrent generation and an increase in the transient photocurrent decay time, which we interpret as a reduction in the charge extraction rate. By applying this novel technique to real defects in both organic and perovskite photovoltaic solar cells we are able to draw new insights into the local charge carrier dynamics with important implications for the management of defects in large-scale printed photovoltaics.

Resume : Stress tests on third generation photovoltaic systems using inorganic Sn4+ perovskites as hole-transporting materials are reported. Dye-sensitized solar cells were fabricated with air-stable Cs2SnI6 and Cs2SnBr3I3 compounds, deposited upon photosensitized titania substrates with metal-organic dyes. The devices were tested under prolonged exposure to intense light and temperatures up to 80 ºC to simulate realistic operating conditions. Chemical and morphological changes of the perovskites were monitored with powder X-ray diffraction and in-situ micro-Raman spectroscopy whereas the photochemical characteristics of the devices were monitored with J-V measurements and electrochemical impedance spectroscopy. The effects of the light-soaking and thermal tests on the perovskite structure and on the power conversion efficiency are discussed.

Resume : : Photovoltaics based on hybrid perovskite materials have exceeded 20% efficiency in only a few years of optimization, for reasons generally ascribed to large charge carrier mobilities and lifetimes. Although functional properties, such as mobilities, photocurrent generation, and open circuit voltage are influenced by the microscopic structure and orientation of the perovskite crystals, these properties are difficult to quantify on the intragrain length scale and are often treated as homogenous within the active layer. In this work, we mapped the local short circuit photocurrent, open circuit photovoltage, and dark drift current in state-of-the-art methylammonium lead iodide solar cells with sub 30 nm spatial resolution using photoconductive atomic force microscopy. We find, within individual grains, spatially-correlated heterogeneity in short circuit current of up to an order of magnitude and in the open circuit voltage up to 0.6 V. These variations are related to different crystal facets and have a direct impact on the macroscopic power conversion efficiency of these materials. We attribute this heterogeneity to a facet-dependent density of trap states. These results imply that controlling crystal grain and facet orientation will enable a systematic optimization of polycrystalline and single crystal devices for photovoltaic and lighting applications.

Resume : Organic-inorganic hybrid perovskite, MAPbX3 (MA=CH3NH3+, X=Br¬- or I-), single crystals are studied for elucidating their intrinsic physical properties. Concerns on grain boundaries in perovskite solar cells have been alleviated by the reports of several groups [1,2]. However, the grains are yet not fully controlled in terms of compositions, textures, and even believed as a trapping source of ionic migration. We report our observations on the distribution of surface electric potential and current transport on the two single crystals with iodine and bromine by Kelvin probe force microscopy and conductive atomic force microscopy, respectively. Surface potential exhibits about 4.7 eV, which is a similar value for thin-film perovskites [2]. Current level of the grain-boundary free large grains is smaller than thin-films but some spots exhibit large current values at high external voltage bias. Piezoelectric force microscopic measurement on the perovskite single crystals was also performed to decipher the potential ferroelectric properties or inequivalent majority/minority carrier dynamics. [1] Yun, J. S.; Ho-Baillie, A.; Huang, S.; Woo, S. H.; Heo, Y.; Seidel, J.; Huang, F.; Cheng, Y.-B.; Green, M. A. J. Phys. Chem. Lett. 2015, 6, 875. [2] Kim, G. Y.; Oh, S. H.; Nguyen, B. P.; Jo, W.; Kim, B. J.; Lee, D. G.; Jung, H. S.; J. Phys. Chem. Lett. 2015, 6, 2355.

Resume : SnS has attracted the attention of the PV community due to the combination of its desirable optical properties (direct band gap of 1.37 eV, and absorption coefficient of >105 cm-1) with binary and abundant elemental composition, which in theory simplifies its synthesis. However, currently the best SnS based PV device efficiency is 4.36 %. Limited performance is attributed to band gap alignment issues, doping content (controlled by Sn/S ratio) and poor film morphologies (due to the anisotropic structure). In this context Raman spectroscopy (RS) analysis could be useful, as it facilitates the accurate evaluation of material properties, as demonstrated in defect studies of other chalcogenide materials. We present an RS study, supported by XRD and XPS analyses, of cubic and α-SnS films grown by CVD single source precursor. In particular a complete analysis of SnS vibrational properties has been made using 7 excitation wavelengths (from UV to IR). The study describes, with high accuracy, the clear differences in Raman modes of both polymorphs: 7 or 4 peaks for cubic or α-SnS structures, respectively. Depending on the chosen excitation wavelength, pre-resonant behaviors are observed. Additionally the impact of various Sn/S ratios on Raman shift and band broadening has been evaluated and discussed. A clear RS feature dependence with composition has been observed. A methodology for accurate Sn/S ratio evaluation by RS is presented which will be relevant for SnS solar cell improvement.

Resume : Increased efforts are being made to develop measurement methods for novel solar cells especially on thin-film and organic solar cell technologies. In these technologies, transparent and opaque thin-film materials are combined in especially complex stacks. Thickness, composition and adhesion at interfaces must be controlled if possible in a non destructive way. This paper presents some applications to such characterizations of the APiC technology, a unique combination of optics and acoustics, which implements a SONAR at nanoscale using an ultrafast laser. Very high frequency acoustic waves are emitted and detected using such laser pulses. These waves propagate indifferently in transparent or opaque materials. These “hypersonic waves” have such a short wavelength that they suit very well the characterization of thin films, multi-layers, nanostructures and interfaces. High frequency acoustic waves are also very sensitive to interface quality. By analyzing the acoustic transmission or reflection at an interface we can detect a delamination between a film and a substrate or between two films. In this paper, we will present some results to show useful can be the APiC technique for controlling thin-film thickness and adhesion at interface in various samples issued from different thin-film photovoltaic technologies (amorphous and crystalline silicon, CIGS and organic).

Resume : The measurement of photoluminescence for solar cell characterisation has become one of the most fruitful sources of information about material quality and solar cell performance. Its remarkably high sensitivity even for silicon and its non-destructive nature gave way to a wide range of applications in research and industry. Depending on the needs for resolution, measurement time and accuracy camera or microscopy based approaches have been developed. These may not only determine material parameters like carrier lifetime, doping concentration, concentration of several specific impurities, crystal defects, or trap density but also solar cell properties like local voltage and current, series resistance, or interface recombination. This presentation gives insight in the basics and advanced principles of PL based quantitative material and cell analysis with a focus on its application to silicon. It compares the strengths of different modes from steady state to time-correlated single photon counting. Current developments in the field of defect characterisation and cell analysis are presented. Specifically, injection- and temperature dependent carrier lifetime imaging for defect identification and time resolved measurements for surface recombination analysis are discussed which show the great potential of PL.

Resume : Third generation photovoltaics have witnessed the irruption of hybrid organic-inorganic perovskites as an unrivalled light harvester. [1] The rapidly achieved breakthroughs, comprising devices with high power conversion efficiencies, are beginning to be accompanied by fundamental studies necessary for turning the current rush into a mature technology. Research efforts aiming at a better understanding of the electrical, structural and photophysical properties of this material are of utmost importance, as they will bridge the gap between the as-prepared material and realistic marketable devices with excellent performances. In this work we present an in-depth analysis regarding the photophysical properties of the methylammonium lead tri-iodide perovskite (CH3NH3PbI3) as a tool for both, understanding the formation of the material itself and its performance under different environmental conditions. Performing a combined study of in-situ structural and photophysical properties during the annealing of CH3NH3PbI3 upon spin-coating the precursor solution we are able to unveil different stages of the crystallization process where nucleation and growth of perovskite crystals dominate. [2] Our results represent a clear evidence of these two processes, opening a new path for improving the properties of the material by acting on the appropriate formation stage. Beyond its use as a tool to understand its formation, the photophysics of CH3NH3PbI3 have proved fundamental in understanding how the material interacts with its environment under device operation conditions. We have identified two different reversible processes in the form of photoinduced activation and darkening of the photoluminescence, extremely dependent on the atmosphere where the films are placed. [3] We resolve the role of oxygen and moisture in the photoactivation and photodarkening, respectively. Indeed, oxygen is revealed as decisive for the emission recovery after a fatigue of the PL in perovskite films. Our results thus tackle two aspects of the physical properties of CH3NH3PbI3, which strongly determine their performance. The extracted conclusions provide insights concerning crystallization and stability of the CH3NH3PbI3 films, which allow bringing this material closer to technological applications. References: [1]. Stranks, S. D. and Snaith, H. J. Nat. Nanotechnol. 10, 391-402 (2015). [2]. Anaya, M., Galisteo-López, J. F., Calvo, M. E., López, C. and Míguez, H. J. Phys. Chem. C 2016 (Accepted). [3]. Galisteo-López, J. F., Anaya, M., Calvo, M. E. and Míguez, H. J. Phys. Chem. Lett. 6, 2200-2205 (2015).

Resume : GaP/Si heterojunctions are of great interest for photovoltaic applications. GaP has a gap of 2.26 eV with the smallest lattice mismatch beyond III-V binary compounds (less 0.4 %) providing conditions for appropriate wide gap emitter for Si based subcell as well as nucleation layer for III-V multijuction solar cell grown on Si substrate. However, the commonly used techniques for the growth of GaP layers such as molecular beam epitaxy (MBE) and metal organic vapor-phase epitaxy (MOVPE) require high temperatures of 800-900C at least at the initial stage for silicon oxide removing and surface reconstruction. High temperatures affect the Si wafers quality and the properties of III V/Si interface leading to low photovoltaic performance of GaP/Si heterojunctions obtained by epitaxy compared to the high efficiency a-Si:H/c-Si solar cells. The study of defects created in Si during epitaxial growth using MBE, MOVPE and ALD at different temperature will be reported in this paper. The photoluminescence (PL) and deep level transient spectroscopy (DLTS) measurements were used for defect characterization. A significant Si degradation during the high temperature MBE and MOVPE growth was demonstrated. This work was carried out in the framework of the joint Russian-French project LAMSOL supported by Ministry of Science and Education of Russian Federation (14.616.21.0040) and by PHC Kolmogorov Program (35522TL).

Resume : The impressive growth of the world wide PV market went along with significant cost reductions of crystalline silicon photovoltaics (c-Si PV), the market leading PV technology. Besides scale economies, the technological progress in the field of cell technology is rather driven by incremental improvements than by a frequent launch of revolutionary new concepts. This leads to a highly developed and well working quasi-standard in c-Si PV, the screen-printed p-type passivated emitter and rear cell (pPERC), which is already capable of conversion efficiencies above 22% and is likely to remain market-leading for years. Independent of the cell technology, kerfless wafering techniques show a high potential to further reduce the costs of crystalline silicon photovoltaics by reducing the silicon consumption. For example, the stress-induced exfoliation of thin Si layers offers the unique possibility to produce thin, kerf-free wafers directly from standard high-quality ingots. Our paper highlights recent trends in c-Si PV with a focus on pPERC. Beyond that, we discuss the potential of kerfless wafering and present solar cells from exfoliated silicon. We use an Al stressor layer to exfoliate a 70 µm thin Si wafer from a thicker wafer. 9 out of 10 cells show efficiencies above 17 %, with a champion cell exhibiting a Voc of 650 mV and an efficiency of 17.64 %. The cells are not restricted by the quality of the exfoliated material.

Resume : Foil creation by lifting off a thin epitaxial silicon layer is a very promising technique to create substrates with a thickness below 100 µm and with almost no kerf loss, allowing to lower both the material cost per m2 and the energy payback time of silicon solar cells. This contribution first of all deals with the latest optimization of the foil fabrication in order to create foils that combine high material quality with a high detachment yield. The epitaxial foils are grown on a double layer stack of porous silicon that is formed by electrochemical etching and subsequently annealed at high temperatures. This porous silicon stack acts both as template for epitaxy and as detachment layer. The quality of the resulting epitaxial foil strongly depends on the smoothness of the resulting porous-silicon based template. Therefore, adapted annealing processes for the porous silicon stack are proposed, based on the observation that smoother surfaces and hence higher-quality foils can be obtained by annealing at lower temperatures for longer times. Furthermore, we were able to simultaneously improve the detachment yield for large-area foils. Finally, we will review in this contribution the current status of the process development and the limitations in cell performance for devices based on these epitaxial thin foils. We will highlight the challenges for both the freestanding processing into devices of these thin epitaxial foils as well as for processing of bonded foils into devices.

Resume : Thin crystalline silicon (c-Si) solar cells with thickness in the order of few tens of microns possess many attractive applications, for instance, electronic wearables, space probes and satellites thanks to their flexibility and light-weight. However, reducing the thickness of active layer of silicon solar cells leads to poor light absorption within the silicon layer, especially in the near infra-red region of the solar spectrum. The poor absorption becomes problematic for thin c-Si solar cells as it causes substantial photocurrent loss. One method to curtail the absorption loss is to incorporate light trapping structures into thin silicon. Light trapping structure of random upright pyramids has been proved to be efficient for conventional silicon solar cells; it allows for easy and inexpensive method of texturization by alkaline based solution. However, the average size of the randomized pyramids ranges from 4 ? 10 µm which is not suitable geometry for thin silicon with thicknesses less than 20 µm. Recently, periodic submicron inverted pyramids have been shown to enhance absorption in thin c-Si solar cells. In this work, we fabricated flexible thin c-Si solar cells with advanced light trapping of periodic inverted pyramids using relatively low-cost wet etching process as well as optimized random upright pyramids with maximum size of 3 µm for thin silicon. Efficiencies of more than 9% and 12.5% have been obtained for bare and textured silicon solar cells with a thickness of 30 µm, respectively. Thin c-Si solar cells were successfully attached to a polymer and removed from the host wafer. In this presentation, we will discuss the fabrication process of flexible thin c-Si solar cells along with the fabrication of the advanced light trapping structures.

Resume : The photovoltaic solar cell industry is putting a lot of effort to reduce silicon cell thickness as a way to achieve cost reductions while taking advantage of expected high conversion efficiencies for such thicknesses. Indeed there are two positive points in this trend, i) material gains could be obtained and ii) thickness decrease allows some relaxation in terms of silicon bulk quality. Moreover, the classic process of ingot sawing cannot be a conceivable solution to produce thin silicon wafers, since more than 50 % of the silicon material will be lost during this step. In this work, we present solar cells made of thin silicon films formed at room temperature by the SLIM-cut technique using an epoxy layer. SLIMcut method consists in detaching thin monocrystalline silicon layers from a film inducing stress. In our case, a layer of epoxy is manually distributed on the surface of a monocrystalline silicon sample and annealed at 150 °C for 1h. The propagation of crack is activated by cooling the sample on an aluminum plate which is cooled by liquid Nitrogen. A relationship between the thickness of the epoxy layer and the foil has been determinated experimentally. Different structures of solar cells adapted to the silicon thin films have been performed. As an example, using silicon film of around 100 µm thick, conversion efficiencies of 14% and 16.7% were measured respectively using a conventional solar simulator and the "suns-Voc" system. Acknowledgments: The research leading to these results has received funding from the European Union Seventh Frame work programme (FP7/2007-2013) under grant agreement n°608593

Resume : In order to optimize long term stability of CIGS modules, their degradation behaviour should be understood and minimized. Therefore, we have designed and built two ‘hybrid’ degradation setups, in which humidity, temperature, illumination and electrical loads are all used as degradation loads on CIGS solar cells. In these setups, climate chambers are combined with illumination (illuminated areas of 40x40 and 100x100 cm2 for up to 12 and 32 samples calibrated BAA and AAA respectively according to IEC norm 60904-9 [1]), which allows real time monitoring of the degradation behaviour. Previously, experiments under elevated temperature, humidity and illumination allowed real-time monitoring of the increase in series resistance due to ZnO:Al degradation [1] as well as the identification of the shunting effects caused by the migration of alkali-elements in CIGS solar cells [2]. Recent studies showed degradation rates of CdTe modules [3] are strongly impacted by electrical state: modules kept under open circuit degraded rapidly, while modules at maximum power point conditions were stable. Therefore, electrical loads have been added to the setups, allowing exposure to both positive and negative biases. In this presentation, we will show the impact of exposure of CIGS solar cells to open circuit and maximum power point conditions, while future experiments will focus on reversibility of migration of alkali-elements in CIGS solar cells at negative voltages. [1] Theelen et al., Proc. 42th IEEE PVSC (2015) 1-6 DOI: 10.1109/PVSC.2015.7355639 [2] Theelen, et al., Prog. Photovolt: Res. Appl. 23 (2015) 537-545 [3] Sinapis et al., Proc. 42th IEEE PVSC (2015) 1-5 10.1109/PVSC.2015.7355628

Resume : Recently, solution-processable perovskite solar cells (PSCs) have created much excitement with certified power conversion efficiencies (PCEs) of 21.0%[1] challenging the long-standing perception that high efficiencies must come at high costs. One major bottleneck for highest efficiency PSCs is the dearth of suitable hole transporting materials (HTMs) where the best performers are polytriarylamine polymer (PTAA) with 20.1%[2] or the small organic molecule 2,2’,7,7’-tetrakis(N,N-di-p-methoxyphenylamine)-9,9’-spirobifluorene (spiro-OMeTAD) with 19.7%[3]. In this work, we present a molecularly engineered HTM with a simple dissymmetric fluorene-dithiophene (FDT) core substituted by N,N-di-p-methoxyphenylamine donor groups, which can be easily modified, providing the blueprint for an entire generation of novel low-cost HTMs. We use FDT on state-of-the-art devices and achieve PCEs of 20.2%, the highest reported value yet, outcompeting comparable control devices with spiro-OMeTAD. Moreover, we show simulations in which the thiophene groups of the FDT bind to the perovskite in a particularly advantageous way introducing the design principle of dually functionalised HTMs, i.e. with a unit that has a particular interaction with perovskite and the unit responsible for the hole transport. Thus, we present a low-cost HTM which has the potential to replace spiro-OMeTAD[4]. [1] http://www.nrel.gov/ncpv/images/efficiency_chart.jpg [2] Yang, W. S. et al. High-performance photovoltaic perovskite layers fabricated through intramolecular exchange. Science 348, 1234-1237, (2015). [3] Ahn, N. et al. Highly Reproducible Perovskite Solar Cells with Average Efficiency of 18.3% and Best Efficiency of 19.7% Fabricated via Lewis Base Adduct of Lead(II) Iodide. J Am Chem Soc 137, 8696-8699, (2015). [4] Saliba, M. et al. A molecularly engineered hole-transporting material for e cient perovskite solar cells. Nature Energy, (2016)

Resume : Zinc oxide nanorods have been investigated as an alternative to mesoporous TiO2 for use in dye-sensitised solar cells (DSSCs) due to improved charge transport properties and a direct current pathway to the transparent electrode. However, due to the significantly lower surface area and subsequent reduction in dye loading, low current densities generally limit efficiency around or below 1 %. Here we present a method for the growth of lamellae on the surface of ZnO nanorods to produce brush-like structures which allows higher dye adsorption, leading to increased efficiency. By coating ~ 6 um long nanorods using an optimised precursor concentration for lamellae growth, a power conversion efficiency of ~2 % is achieved for cells using N719 dye and iodide-based electrolyte, which is one of the highest efficiencies reported for ZnO nanorod-based DSSCs. A number of methods are used to maximise the device efficiency. For example, it is shown that by increasing the nanorod growth time, greater length and therefore surface area is achieved. In addition, the use of a hydrogel-derived seed layer is shown to lead to a higher density of thinner nanorods, further contributing to the high surface area and therefore efficiency. Correct choice of annealing conditions is also shown to be important to improve the crystallinity of the ZnO lamellae while retaining the high surface area of the structure. Overall this study shows that a number of methods can be used to increase the surface area of ZnO nanorod arrays, including correct choice of seed layer, extended growth time, and the addition of appropriate surface structures. Combining these techniques has produced high efficiency ZnO nanorod-based devices, and higher efficiencies may be possible using these methods, for example by further increasing the nanorod length or using dyes optimised for ZnO-based devices.

Resume : We report perovskite solar cells with an ultra-porous TiO2 electron transport layer fabricated using flame spray pyrolysis (FSP). FSP allows rapid deposition of porous TiO2 films and could be scaled up for large area and high throughput solar cell production. An efficiency of 13.7% was achieved for a flame deposited porous TiO2 perovskite solar cell coated with 2 nm of TiO2, which is a comparable efficiency to perovskite solar cells made with the widely used spin-coated TiO2 porous layer. The porosity of the as-deposited (uncoated) flame deposited porous TiO2 film is 97% which can be reduced to any desired value by coating it with a thin layer of TiO2 that also adds mechanical stability. The perovskite easily infiltrates into the flame deposited porous TiO2 film because of its large pore size and high porosity. The flame deposited porous TiO2 solar cells could be further optimised by adjusting the perovskite deposition to add a perovskite capping layer. The results reported here demonstrate a promising approach for large scale production of perovskite solar cells.

Resume : Hybrid organometal halide perovskites such as methylammonium lead iodide (CH3NH3PbI3) are a promising class of cost- and energy-efficient light absorption materials for thin-film photovoltaics. In this paper, we have discussed concerning two issues, CH3NH3PbI3 nanoparticles (NPs) and CH3NH3PbI3 nanocrystals (NCs) based perovskite solar cells. The preparation and characterization of CH3NH3PbI3 NPs on compact TiOx/indium tin oxide glass substrates using a simple spin-coating technique have been investigated. Compact-TiOx films were prepared by chemical bath deposition (CBD) according to the procedure described by Kuwabara et al. (Organic electronics 11 (2010) 1136). The CH3NH3PbI3/ ionic liquid mixture solution was prepared via adding 1 wt% of 1-hexyl-3-methylimidazolium chloride (HMImCl) in 25 wt% solution of CH3NH3PbI3 in N,N-dimethylformamide (DMF). Analysis of the resulting spin-coated films revealed that uniform spherical CH3NH3PbI3 NPs with an average diameter of 540 nm were homogeneously distributed on the compact-TiOx substrate as shown in Fig. 1 (a, b). However, addition of IL to the spin-coating solution facilitated the formation of homogenous nucleation sites and prevented rapid crystal formation of CH3NH3PbI3. Therefore, the presence of an IL resulted in uniform thin films with good morphology. Fig. 1: (a, b) The SEM images of the CH3NH3PbI3 films prepared without (w/o IL) and with (w/IL) IL. The nanoparticle based device showed power conversion efficiency (PCE) up to 2.44 %. The preliminary results are promising as it is the first report on the fabrication of solar cells based on CH3NH3PbI3 NPs. We also expect that the results will open a pathway towards a better understanding for the fabrication, modification and enhancement of the performance of solar cells with CH3NH3PbI3 NPs. In the present case, we assume a hindering effect followed by impact on charge dissociation, transport, and/or recombination on the device performances due to the residual IL content remained on the CH3NH3PbI3 NP films. Therefore, performance improvement experiments are underway to ensure the complete removal of IL-contents from the CH3NH3PbI3 NP films. And, we fabricated planar heterojunction (PHJ) type PSC with enhanced efficiency by introducing fullerene (C60) interlayers with thicknesses of 0, 3, 7 and 10 nm between air-stable amorphous compact TiOx and CH3NH3PbI3 layers. The modified morphology obtained by inclusion of C60 improved the surface energy properties of the cells in terms of enhanced photocurrent. Atomic force microscopy verified the correlation between the surface energy and phase morphology of the PHJ solar cells. The introduction of a C60 interlayer between CH3NH3PbI3 and TiOx layers increased the content of photogenerated charge carrier sites, as well as lowering the accumulation and trapping of photogenerated charges at the TiOx interface. The optimum thickness of C60 interlayer was 7 nm, for which a maximum PCE of 9.51% was obtained.