This PDF file contains the front matter associated with SPIE Proceedings Volume 7772, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

Luminescence imaging techniques are increasingly used in photovoltaics (PV) related research and development and in the production of solar cells and modules. Intense research in this area has revealed a variety of material and device parameters that can be measured, generally with very short measurement times and high spatial resolution. While the focus of luminescence imaging R&D has so far been on traditional wafer based silicon solar cells, the principles of luminescence imaging, and its inherent benefits apply generally to other solar cell concepts and can therefore be expected to accelerate progress also with the further development and realization of advanced, so-called third generation solar cell approaches. This paper reviews some fundamental aspects of luminescence, specifically the relation between the luminescence intensity and both the minority carrier lifetime and the diode voltage. Some resulting specific luminescence imaging applications for silicon solar cells will be discussed.

We have developed a method to design multi-junction horizontally-oriented solar cells using single-layer photonic metamaterials. These metamaterial light harvesting templates are capable of separating white light into discrete wavelength ranges and trapping it efficiently into different, separately wired cavities. Any number of different wavelength-tailored charge separation complexes can be fixed to the walls of these tuned cavities. To design the metamaterials we have developed a coupled wave analysis of 2D periodic metamaterials. Past results with 1D gratings have shown that this is a very effective method for designing periodic structures and we have generalized the approach to 2D periodic cavities.

We review the use of nanometer-sized particles (including quantum dots) in the conversion of parts of the solar spectrum incident on solar cells to more usable regions. The modification of the solar spectrum ideally would lead to a narrowbanded incident spectrum at a center wavelength corresponding to an energy that is slightly larger than the band gap of the semiconductor material employed in the solar cell, which would lead to an enhancement of the overall solar energy conversion efficiency. Modification of the spectrum requires down and/or up conversion or shifting of the spectrum, meaning that the energy of photons is modified either to lower (down) or higher (up) energy. Nanostructures such as quantum dots, luminescent dye molecules, and lanthanide-doped glasses are capable of absorbing photons at a certain wavelength and emitting photons at a different (shorter or longer) wavelength. We will discuss down and up conversion and shifting by quantum dots, luminescent dyes, and lanthanide compounds, and assess their potential in contributing to ultimately lowering the cost per kWh of solar generated power.

We have investigated a photochemical up-conversion system comprising a molecular mixture of a palladium porphyrin to harvest light, and a polycyclic aromatic hydrocarbon to emit light. The energy of harvested photons is stored as molecular triplet states which then annihilate to bring about up-converted fluorescence. The limiting efficiency of such triplet-triplet annihilation up-conversion has been believed to be 11% for some time. However, by rigorously investigating the kinetics of delayed fluorescence following pulsed excitation, we demonstrate instantaneous annihilation efficiencies exceeding 40%, and limiting efficiencies for the current system of ≈60%. We attribute the high efficiencies obtained to the electronic structure of the emitting molecule, which exhibits an exceptionally high T₂ molecular state. We utilize the kinetic data obtained to model an up-converting layer irradiated with broadband sunlight, finding that ≈3% efficiencies can be obtained with the current system, with this improving dramatically upon optimization of various parameters.

Weak absorption of light near the band gap is one limiting factor on the efficiency of photovoltaics. This is particularly true for thin-film solar cells because the short optical path lengths and limited options for texturing the front and back surfaces. Scattering light laterally is one way to increase the optical path length to increase the chance that a given low energy photon is absorbed. We investigate the use of a periodic array of bowtie apertures to couple incident light to parallel plate waveguide modes supported between two conductors. We show that this increases the efficiency of solar cells by 39% and explain the physical mechanisms. This architecture has potential for thin film photovolatics or for forming an intermediate conductor in multi-junction solar cells.

We report progress in developing optimized diffraction gratings for coupling solar radiation from the airmass 0 spectrum into waveguide modes of ultrathin quantum dot solar cells (QDSCs). Electromagnetic simulations have been used to optimize the grating geometry and to analyze the nature of diffraction within the device structure. These results suggest that increases in photocurrent of over 100% at wavelengths of QD absorption, corresponding to over 10% improvement in short-circuit current, can be achieved in optimal ultrathin devices by incorporating gratings in the rear contact.

Transparent, rare-earth doped fluorozirconate-based glasses and glass ceramics are attractive systems as up- and downconverters to increase solar cell efficiency. For down-conversion applications, the efficiency of a silicon solar cell could be significantly increased in the ultraviolet spectral range by placing a europium-doped glass ceramic on top. High transparency is a key issue here to avoid scattering losses and to obtain high light output. Transmission spectra of fluorozirconate glasses, which were additionally doped with chlorine ions to initiate the growth of BaCl₂ nanoparticles therein upon thermal annealing, show that the absorbance in the visible spectral depends significantly on the annealing conditions. For up-conversion applications, erbium-doped fluorozironate glasses have been investigated. 2-dimensional intensity mapping of the up-converted fluorescence yielded information on the homogeneity of the glass sample and the erbium distribution therein. Depth scan experiments showed that the position of the focus of the excitation laser beam plays a crucial role since saturation of the 2-photon up-conversion occurs for high excitation power.

Materials used as luminescent down shifters (LDS) have to absorb light effectively in the spectral area where solar cells have poor internal quantum efficiency. At the same time these materials have to emit most of the absorbed spectral powers at lower energies where the internal quantum efficiency of the solar cell is close to the maximum. The effects of silicon nanocrystals prepared by thermal treatment of a silicon-rich-oxide (SRO) layer on the efficiency of c-Si cells are investigated in this paper. The SRO layer is characterized by a high photoluminescence peak at around 800 nm. Influence of the active layer on light transmission and on the modification of the optical spectra due to photoluminescence generation has been determined with the help of optical measurements and transfer matrix simulations. The solar cell efficiency for cells with and without down-shifting layer were measured under illumination with AM1.5G solar spectrum and compared with the simulations. Finally, we model the behavior of cells with and without LDS layer showing that a cell with LDS suffers less from bad surface passivation.

State of the art photovoltaics exhibiting conversion efficiency in excess of 30% (1-sun) utilize epitaxially grown multijunction III-V materials. Increasing photovoltaic efficiency is critically important to the space power, and more recently, the terrestrial concentrator PV communities The use of nanostructured materials within photovoltaic devices can enable improved efficiency, potentially in excess of the Shockley-Queisser limit. The addition of nanostructures such as quantum dots (QDs) to photovoltaic devices allows one to extend the absorption spectrum of the solar cell and "tune" the bandgap to the spectral conditions. Multi-junction (MJ) solar cells would benefit from the additional short-circuit current within the middle current-limiting (In)GaAs cell via QD spectral tuning. While QD tuning is a potentially direct approach to increased efficiency of MJ solar cells, it has been reported that significant improvements can be achieved using QDs to form an intermediate band within the bandgap of a suitable matrix. We will discuss the potential for QD photovoltaic devices and examine the challenges associated with multi-junction device growth with the inclusion of quantum dot arrays. GaAs p-i-n solar cells, with and without InAs QD superlattices are used to demonstrate the potential benefits of QDs. The unique challenges associated with the characterization of this type of device will also be presented. Using strain-balanced Stranski-Krastanov QD formation, we have demonstrated sub-gap photon collection and increased current in QD-enhanced GaAs solar cells containing up to 100 periods. Finally, we will discuss the opportunities that these devices hold for high photovoltaic conversion efficiency.

Quantum dot (QD) luminescent solar concentrator (LSC) uses a sheet of highly transparent materials doped with luminescent QDs materials. Sunlight is absorbed by these quantum dots and emitted through down conversion process. The emitted light is trapped in the sheet and travels to the edges where it can be collected by photovoltaic solar cells. In this study, we investigate the performance of LSCs fabricated with near infrared QDs (lead sulfide) and compared with the performance of LSCs containing normal visible QDs (CdSe/ZnS), and LSCs containing organic dye (Rhodamine B). Effects of materials concentrations (related to re-absorption) on the power conversion efficiency are also analyzed. The results show that near infrared QDs LSCs can generate nearly twice as much as the output current from normal QDs and organic dye LSCs. This is due to their broad absorption spectra. If stability of QDs is further improved, the near infrared QDs will dramatically improve the efficiency of LSCs for solar energy conversion with lower cost per Wp.

Si quantum dots in SiO₂/Si₃N₄ hybrid matrix on quartz substrates were fabricated by magnetron sputtering of alternating silicon rich oxide and Si₃N₄ layers followed by different post-deposition anneals. X-ray diffraction results indicate that the average dimension of the Si QDs ranges from 1.6 to 5.2 nm. The size and crystallisation of the Si nanocrystals are dependent on a number of factors, including the annealing condition, the SRO thickness and the Si3N4 barrier thickness, as evidenced in XRD and Raman measurements. In particular, thicker Si3N4 barrier layers seem to be able to suppress the growth of Si nanocrystals more effectively. PL measurements indicate that the apparent bandgap of the samples investigated in this work is in the range 1.12-1.67 eV, which demonstrates the effect of quantum confinement. More interestingly, analysis of the PL data reveals that the PL peak energy does not only depend on the size of the nanocrystals, but also affected by other details of the nanocrystal formation. A simple core-shell model is constructed to illustrate our explanation. These findings offer a preliminary understanding of the nanocrystal growth and radiative recombination processes in this newly synthesized material.

We demonstrate experimentally that ensembles of conically shaped GaP nanorods form layers of graded refractive index due to the increased filling fraction of GaP from the top to the bottom of the layer. Graded refractive index layers reduce the reflection and increase the coupling of light into the substrate, leading to broadband and omnidirectional antireflection surfaces. This reduced reflection is the result of matching the refractive index at the interface between the substrate and air by the graded index layer. The layers can be modeled using a transfer-matrix method for isotropic layered media. We show theoretically that the light coupling efficiency into silicon can be higher than 95% over a broad wavelength range and for angles up to 60. by employing a layer with a refractive index that increases parabolically. Broadband and omnidirectional antireflection layers are specially interesting for enhancing harvesting of light in photovoltaics.

The optical and electrical properties of bulk polymer RR-P3HT (Regio-Regular Poly(3-hexylthiophène-2,5- diyl):PCBM (Methanofullerene Phenyl-C61-Butyric-Acid-Methyl-Ester) heterojunction incorporating single wall carbon nanotubes (SWCNTs) have been already reported by a number of research groups. We investigated a new approach to functionalize CarboLex single wall carbon nanotubes (SWCNTs-e) for increasing their dispersion in various solvents. The addition of SWCNTs-e in the matrix of P3HT:PCBM improves the photovoltaic (PV) characteristics. Results show that the photovoltaic parameters depend on the concentration of SWCNTs-e. The incorporation of low concentrations of SWCNTs-e in the photoactive layer increases the current density J_sc before annealing. We attribute the improved performance to partial crystallisation of the RR-P3HT. As revealed by XRD studies and confirmed by the absorbance spectra which exhibit the characteristic 600 nm shoulder. Interestingly, we observed also that doping the P3HT:PCBM system with the functionalized SWCNTs increases V_oc from 0.583 to 0.744 V.

We use a rigorous electromagnetic approach to develop a light-trapping theory, which reveals that the conventional limit can be substantially surpassed in nanophotonic regimes, opening new avenues for highly efficient solar cells.

Intrinsic loss mechanisms are quantified for a single junction device under one sun illumination. Thermalisation, below Eg and Boltzmann losses are shown to be the dominant loss mechanisms in a device with Eg = 1.31eV , accounting for 30%, 25% and 9% of the incident solar radiation respectively. Alternative device designs which target these dominant intrinsic losses are considered. Concentrator (and restricted emission) devices target Boltzmann loss. This loss mechanism is shown to have a logarithmic relationship with concentration and as such, a small increase in absorption solid angle equates to a large increase in fundamental limiting efficiency. A multi-junction device resolves the mismatch between the broad solar spectrum and single threshold absorption, and thus targets below Eg and thermalisation losses. A greater number of junctions allows the device absorption profile to better match the solar spectrum, increasing device efficiency. Boltzmann loss slightly increases with junction number and as such, concentration will be proportionally more effective at increasing efficiency in a multi-junction device. A hot carrier device targets thermalisation loss. This loss mechanism is eliminated in the impact ionisation model used in this paper, allowing for enhanced device efficiency. ----- The on-screen and audio presentation of this paper can be played by clicking the multimedia PDF link at the bottom right hand of this page.

The 'Stress induced LIft-off Method' (SLiM-Cut) is a kerf-free method for thin silicon fabrication, being developed at imec for photovoltaic applications [1]. This method makes particularly efficient use of bulk material, thus cutting down the Si cost. SLIM-Cut uses a metallic layer on top of thick silicon substrate. The bonding is achieved at high temperature. Quenching the assembly down to room temperature builds up stress inside the material. The system relaxes by propagating a crack parallel to the metal-silicon interface. The propagation of this crack over the entire surface allows the formation of a silicon foil. The choice of the stress inducing layer is of the utmost importance: (1) the interfacial strength has to be high enough for the crack to grow in the Si lattice, (2) the metal migration has to be limited in order not to compromise the PV conversion efficiency, and (3) the deposition method of the stressing layer should be compatible with PV cell processing. The focus is laid on the two main compromises of the technology today: the precise control of the stress applied to the substrate, and the metal-Si interface. Regarding the control of the stress applied, we have intentionally initiated the crack. Better control of the crack propagation was demonstrated. Tuning of the temperature becomes therefore possible and lift-off was achieved for temperature processing as low as 700°C. Although all crystal orientations (including <100>) have been successfully lifted-off, the choice of the crystal orientation influences strongly the final result. Regarding the metal-Si interface, a detail elemental study has enabled us to identify the composition of the interface layers responsible for good adhesion to the Si. This investigation is also the first step to the engineering of specific paste and cleaning solutions.

Photovoltaic panel performance is improved by means of an optical surface technology that is applied to the front surface of a panel. The technology is comprised of an array of micro lens structures embossed into highly transmissive polymer film or glass. The film version may be applied to new or previously installed photovoltaic panels; the glass version is appropriate for new panels. The structures are designed using a proprietary ray tracing program to concurrently optimize path length (the length of the path of light through the photovoltaic material), total internal reflection and anti-reflection. The structures may be optimized for diffuse light conditions, incoming ray angles and/or time of day. Additionally, the rays may have their paths modified or be redirected by the structures to improve performance at specific band gaps that are customized for particular photovoltaic materials.

We report the design and optimization of plasmonic thin film photovoltaics (TFPV) based on single crystalline Si nanomembranes, transferred onto ITO/plastic solar cells. Due to the presence of Fabry-Perot (FP) interferences, both near and far field plasmonic field enhancements are modulated by thin film thicknesses. The placement of absorption region (junction location and/or quantum well location) is critical in TFPV efficiency improvement. For TFPV thin film thickness between 200-500 nm, we obtained field enhancement up to 100%, and cell efficiency improvement up to 35%, assuming the standard test one sun AM1.5 radiation conditions. The efficiency enhancement for 200 nm silver particles decreases rapidly with the increase of thin film thicknesses, while we observed less change in cell efficiency enhancement for smaller 100 nm silver particles. Additionally, we observed a relatively large process tolerance window for the size, shape, and placement of Ag nanoparticles, formed by the annealing of ultra-thin Ag thin films.

The characteristics of the well-aligned silicon nanowire/poly(3,4-ethylenedioxy- thiophene):poly(styrenesulfonate) (PEDOT:PSS) heterojunction solar cells are investigated. The PEDOT:PSS adheres on the n-type silicon nanowire surface to form core-sheath heterojunction structure. A novel front contact is proposed here to enhance the carrier collection efficiency for the nanostructured emitter. The cells exhibit stable rectifying diode behavior. Such structure increases the area of junctions and shortens exciton diffusion distance, and therefore greatly increases exciton dissociating probability. Compared to the devices without nanowire structure, the power conversion efficiency improves from 0.08 % to 4.99 %.

In this work, the preparation and photovoltaic conversion characterization of 10 μm films of sensitized TiO₂ is reported. The 13 nm TiO₂ nanocrystals with anatase crystalline phase were deposited on an FTO substrate and decorated with Au nanocrystals and sensitized with CdSe Quantum dots (QD) and poly(3-octylthiophene) (P3OT) in different configurations. The photocurrent was measured in a three electrode electrochemical cell. The results exhibited that TiO₂/Au/QD/P3OT films have the largest photocurrent, giving approximately fivefold the photocurrent of TiO₂ films sensitized only with QD and near sevenfold the photocurrent of Au decorated TiO₂ films. These results are attributed to the ability of the Au nanocrystals to extract electrons from QD and the photogeneration and hole transport of P3OT. Both phenomena combined with the QD's photogeneration give a great amount of electrons that increase the photocurrent generation.

Experimental and numerical results concerning the influence of silver nanoparticles on the optical absorption of organic devices are presented. The metallic nanoparticles (NPs) are placed inside an interpenetrated poly(3-hexylthiophene): [6,6]-phenyl-C61-butyric acid methyl ester (P3HT:PCBM) layer using a physical vapor deposition technique. An absorption enhancement by comparison to devices without NPs is shown. An increase of the absorption by annealing is also observed. Moreover, calculations are performed via a numerical analysis based on a Finite Difference Time Domain (FDTD) method. We demonstrate that the light absorption can mainly occur inside the active layer instead of inside the metallic NPs.

We propose a metamaterial based absorber design that operates at the infrared regime. The absorption peak was 83.6%. We can incorporate solar-cell layers inside the metamaterial absorber in order to significantly increase solar-cell efficiency.

The optical properties of the silicon naniwires (SiNWs) fabricated by a method of metal-assisted electroless etching have been investigated. The optical parameters of both SiNWs and the Si wafer, including dielectric function (ε (ω) ) and the effective refractive index ( N(ω) ), could be obtained from the light-absorption theory and Kramers-Kronig relation. From the comparation of calculated characteristics, we can observe that the refractive index of the SiNWs is reduced to 1.26 while that of Si wafer is changing from 2.4 to 6.2 in the range of 1.5-4eV, and the imaginary part of complex dielectric constant of Si NWs is about two orders of magnitude lower than that of Si wafer. There are two peak positions in the curve of Si wafer, while just one broad peak position in that of Si NWs. These differences could be considered as the reasons for the special characteristic of the Si NWs.

In this proceeding, we simulate the optical properties of vertically-aligned silicon nanowire and nanohole arrays using the transfer matrix method. We find that the optical absorption in both silicon nanowire and nanohole arrays improves with increasing lattice constant up to 600nm - 700nm. We attribute the observed optical absorption enhancement effect to an increase in the field concentration inside the active silicon region and the excitation of guided resonance modes. For optimized parameters, both structures can be more absorptive than an equally-thick silicon solid film with an optimal single layer Si₃N₄ anti-reflection coating. This conclusion holds true for both optically thin (2.33μm) and optically thick (100μm) structures. For optically thin structures, the enhancement in the optimal nanohole array exceeds the conventional light trapping limit. For optically thick structures, the enhancement in both optimal nanohole and nanowire arrays exceeds the light trapping limit. Additionally, we show that the overall absorption efficiencies for hexagonal and square lattices of nanowires are very similar.

The optical properties of ordered and disordered vertical silicon nanowire arrays, including random position, diameter, and length, are investigated using the finite-difference time-domain method. The ordered array with diameter of 100 nm shows overall small reflection and large absorption. An absorption peak is located at 2.4 eV, which is due to the optical resonance effect. Both randomly-positioned and random diameter arrays remain unreflective as the ordered array. The absorptance of randomly-positioned nanowire arrays has similar frequency dependence as ordered array, while it is enhanced due to the enhanced scattering. Random diameter array has a different absorptance profile and no evident absorption peak is observed, which is explained by the different resonant frequencies of the inclusion nanowires. Random length can create a random rough top surface on the nanowire arrays, which can reduce reflection and enhance absorption compared to uniform top surface.

Ternary systems based on mixtures of polymer, PCBM and CdSe nanoparticles were investigated. The photophysical and electrochemical properties were modulated by changing the size of the inorganic nanoparticles and their effects on the performance of the solar cells were analyzed. At the optimized conditions, the presence of the nanoparticles increased the photocurrent and photovoltage, improving the efficiency of the devices. A complete study on the morphologic effects induced by the presence of these nanoparticles was performed using AFM, HR-TEM and optical microscopy techniques.

The optical absorption of a transparent conductive oxide (TCO), which is often used as the basis for junction or contact layers in thin film photovoltaics, can be tailored by incorporating a nanophase semiconductor (SC) component. Using a, dual-source, sequential R.F. magnetron sputter deposition technique, we manipulate the optical and electronic properties of SC:TCO composites by varying the local and extended nanophase assembly and composition. The present study explores nanocomposite systems based on Ge:ZnO and Ge:ITO. The impact of host material (ITO vs. ZnO) on the evolution of nanostructure is investigated. Heat treatment of the as-deposited films results in an increased crystallinity of the TCO and SC components, confirmed by X-ray diffraction and Raman spectroscopy studies. The presence of the SC phase is found to influence TCO grain growth and crystallographic orientation, and modification of the SC phase distribution is coincident with the morphological development of the TCO phase in both composite systems. Upon heattreatment, the high-energy optical absorption edge of the nanocomposite is blue-shifted compared to that of the corresponding as-deposited material. This indicates the development of quantum-confinement conditions for photocarriers within the Ge phase which leads to an increased energy gap over that expected for the more bulk-like, asdeposited Ge material. Under the deposition and thermal treatment conditions used in the present study, the spectral absorption response is consistent between the ZnO and ITO-based thin films examined. This suggests that carrier confinement conditions are mediated by the development of similar Ge-phase local spatial extent and Ge:TCO interfacial structures in both systems, regardless of TCO identity.