Volume 1 photovoltaic solar energy 1 27 – luminescent solar concentrator

Volume 1 photovoltaic solar energy 1 27 – luminescent solar concentrator Volume 1 photovoltaic solar energy 1 27 – luminescent solar concentrator Volume 1 photovoltaic solar energy 1 27 – luminescent solar concentrator Volume 1 photovoltaic solar energy 1 27 – luminescent solar concentrator Volume 1 photovoltaic solar energy 1 27 – luminescent solar concentrator Volume 1 photovoltaic solar energy 1 27 – luminescent solar concentrator

1.27 Luminescent Solar Concentrator JC Goldschmidt, Fraunhofer Institute for Solar Energy Systems ISE, Freiburg, Germany © 2012 Elsevier Ltd All rights reserved 1.27.1 Introduction 1.27.2 Theory of Luminescent Solar Concentrators 1.27.2.1 The Factors that Determine the Efficiency of Luminescent Concentrator Systems 1.27.2.2 Thermodynamic Efficiency Limits 1.27.2.3 Thermodynamic Models of the Luminescent Concentrator 1.27.2.4 Ray Tracing Simulations of Luminescent Concentrators 1.27.3 Materials for Luminescent Solar Concentrators 1.27.3.1 Organic Dyes 1.27.3.2 Inorganic Luminescent Materials 1.27.4 Luminescent Solar Concentrator System Designs and Achieved Results 1.27.4.1 System Designs 1.27.4.2 Achieved System Efficiencies 1.27.5 The Future Development of Luminescent Solar Concentrators 1.27.5.1 Extending the Used Spectral Range into the IR 1.27.5.2 Controlling Escape Cone Losses 1.27.5.2.1 Photonic structures for increased efficiencies 1.27.5.2.2 Controlling the angular emission 1.27.6 Conclusion Acknowledgments References 587 588 588 589 590 592 593 593 594 594 594 595 596 596 596 596 598 599 599 599 1.27.1 Introduction In a luminescent collector, a luminescent material embedded in a transparent matrix absorbs sunlight and emits radiation at a different wavelength than the incident one Total internal reflection traps most of the emitted light and guides it to the edges of the collector (Figure 1) This underlying principle was first used in scintillation counters [1, 2] A geometric concentration is achieved if the area of the edges is smaller than the illuminated front surface of the collector, that is, when the area from which light is collected is larger than the area from which light is emitted When solar cells are optically coupled to the edges, they can convert the guided light into electricity The application to concentrate solar radiation onto a solar cell was proposed in the late 1970s [3, 4] If the solar cell is illuminated with a higher intensity than it would be in direct sunlight, a real concentration is achieved For real concentration, high geometric concentration, as well as high collection efficiency, is necessary Probably, the most outstanding feature of luminescent solar concentrators is their ability to concentrate both direct and diffuse radiation This ability is a great advantage for the application of luminescent concentrators in temperate climates, such as in middle Europe, or in indoor applications with relatively high fractions of diffuse radiation Additionally, luminescent concentrators not require tracking systems that follow the path of the sun, in contrast to concentrator systems that use lenses or mirrors in combination with trackers This facilitates, for instance, the integration of luminescent concentrators in buildings Luminescent concentrators were investigated intensively in the early 1980s, for example, [5, 6] Research at that time aimed at cutting costs by using the concentrator to reduce the need for expensive solar cells However, several problems led to reduced research interest First, the used organic dyes had only relatively narrow absorption bands Second, although the organic dyes showed high quantum efficiencies, defined as the ratio between absorbed and emitted photons, above 95% in the visible range of the spectrum, quantum efficiencies remained at 50% and lower in the infrared (IR) because of fundamental physical reasons Furthermore, the dyes that were sensitive in the IR were unstable under long-term illumination Reabsorption of the emitted light due to overlapping absorption and emission spectra further reduced efficiencies [5] A fundamental problem was the escape cone of the internal reflection, which caused losses of at least 26% Finally, efficient solar cells of which the spectral response matched the emission spectra of the dyes were hardly available After more than 20 years of progress in the development of solar cells and luminescent materials, and with new concepts, several groups such as those of References 7–25 are currently reinvestigating the potential of luminescent concentrators High efficiencies have been achieved [19, 25, 26] and there has also been considerable progress in the understanding and theoretical description, for example, [22, 24, 26–28] However, efficiencies are still too low and system sizes too small for a commercial application Comprehensive Renewable Energy, Volume doi:10.1016/B978-0-08-087872-0.00132-3 587 588 Technology E1 Escape cone Radiation θc Solar cell E2 Collector plate Reabsorption and emission Solar cell Dye Figure Principle of a luminescent concentrator A luminescent material (dye) in a matrix absorbs incoming sunlight (E1) and emits radiation with a different energy (E2) Total internal reflection traps most of the emitted light and guides it to solar cells optically coupled to the edges Emitted light that impinges on the internal surface with an angle steeper than the critical angle θc is lost due to the escape cone of total internal reflection A part of the emitted light is also reabsorbed, which can be followed by reemission The concept of luminescent concentrators has been given many different names; among them are fluorescent collectors [4], quantum-dot solar concentrator [20–22, 29], and organic solar concentrator [30] In this work, I will use the term ‘luminescent concentrator’ for the overall concept For clarity, ‘luminescent collector’ will refer to the collector plate without attached solar cells and ‘luminescent concentrator system’ will refer to a system constructed from a collector plate with solar cells attached In Section 1.27.2, I will give an overview of the theoretical understanding on luminescent concentrator I will start with identifying the factors that determine luminescent concentrator system efficiencies I will then shortly review theoretical models and simulation approaches In principle, all kinds of luminescent materials can be used in a luminescent concentrator: fluorescent materials that show a Stokes shift of the emission to longer wavelengths; phosphorescent materials; upconverters that emit one high-energy photon after the absorption of at least two low-energy photons; and quantum-cutting materials that emit two low-energy photons after the absorption of one high-energy photon In Section 1.27.3, I will review which materials are investigated for their application in luminescent concentrators Many different system configurations have been proposed for luminescent concentrators Variations include deposition of the luminescent material in a film on a transparent slab [31], stacking of different luminescent collectors [4], cylindrical collectors [32], solar cells coupled to the bottom of the collector [13, 15], and many more The most important system configurations and results of their experimental investigation will be presented in Section 1.27.4 In Section 1.27.5, I will discuss the possibilities for the future research on luminescent concentrators 1.27.2 Theory of Luminescent Solar Concentrators 1.27.2.1 The Factors that Determine the Efficiency of Luminescent Concentrator Systems Several factors determine the efficiency of a luminescent concentrator system Most of these factors are wavelength dependent By integrating over the respective relevant spectrum, a description of the overall system efficiency with a set of efficiencies for individual processes is possible [33, 34] Following their representation, the important parameters are the following: ηtrans,front transmission of the front surface with respect to the solar spectrum absorption efficiency of the luminescent material due to its absorption spectrum with respect to the transmitted solar ηabs spectrum QE Quantum efficiency of the luminescent material ηStoke ‘Stokes efficiency’; (1 − ηStoke) is the energy loss due to the Stokes shift fraction of the emitted light that is trapped by total internal reflection ηtrap ηreabs efficiency of light guiding limited by self-absorption of luminescent material; (1−ηreabs) is the energy loss due to reabsorption ‘matrix efficiency’; (1 − ηmat) is the loss caused by scattering or absorption in the matrix ηmat ηtref efficiency of light guiding by total internal reflection ηcoup efficiency of the optical coupling of solar cell and luminescent collector efficiency of the solar cell under illumination with the edge emission of the luminescent collector ηcell The overall system efficiency ηsystem can be calculated from the single parameters via ηsystem ¼ ηtrans ηabs QEηStoke ηtrap ηreabs ηmat ηtref ηcoup ηcell ½1Š Luminescent Solar Concentrator 589 Several aspects are of importance for the different efficiencies: – The transmission of the front surface is determined by its reflection R(λinc), where λinc is the incident wavelength This is usually the Fresnel reflection, which is   nðλinc ị1 R ị ẳ ẵ2 ninc ị ỵ for normal incidence and the surface between a medium with refractive index of and a medium with refractive index n(λinc) The reflection of typical materials is in the range of 4% for one surface Special layers or structures applied to the front can decrease or increase reflection Interestingly, an antireflection coating that reduces the Fresnel reflection does not affect the total internal reflection This can be understood by considering that total internal reflection is an effect strongly linked to refraction Total internal reflection occurs when the light from inside the high-index material impinges on the surface with an angle sufficiently shallow that the light would be refracted back into the medium again As the antireflection coating does not change the refraction, total internal reflection is not affected either – The absorption spectrum Abs(λinc) determines the absorption efficiency A large fraction of the solar spectrum is lost, because many luminescent materials absorb only a narrow spectral region The absorption range of typical fluorescent organic dyes is only about 200 nm in width – The quantum efficiency QE of the luminescent material is defined as the ratio of the number of emitted photons to the number of the absorbed photons For organic dyes, the luminescent quantum efficiency can exceed 95% – The energy of the emitted photons is usually different from the energy of the absorbed photons For most luminescent materials, a Stokes shift to lower energy occurs This means that the emitted photons possess less energy than the absorbed ones Therefore, the wavelength of the emitted photons λemit is different from the wavelength of the incident photons λinc As we will see in Section 1.27.2.2, this Stokes shift is of critical importance to the ability of the luminescent concentrator to concentrate light – The luminescent material emits light isotropically in a first approximation All light that impinges on the internal surface with an angle smaller than the critical angle θc(λemit) leaves the collector and is lost (Figure 1) The critical angle is given by   c ị ẳ arcsin ½3Š nðλemit Þ This effect is also called the escape cone of total internal reflection The light that impinges with greater angles is totally internally reflected Integration gives a fraction p trap ị ẳ 1nemit ị ẵ4 of the emitted photon flux that is trapped in the collector [35] For polymethyl methacrylate (PMMA) with n = 1.5, this results in a trapped fraction of around 74%, which means that a fraction of 26% is lost after every emission process The 26% ratio accounts for the losses through both surfaces An attached mirror does not change this number, as with a mirror the light leaves the collector through the front surface after being reflected – The absorption spectrum and the emission spectrum overlap For principal reasons, absorption must be possible in the spectral region where emission occurs Therefore, part of the emitted light is reabsorbed Again, the energy loss due to a quantum efficiency smaller than occurs, and again radiation is lost into the escape cone – Realistic matrix materials are not perfectly transparent They absorb light and they scatter light, so it leaves the collector – Total internal reflection is a loss-free process However, the surface of the luminescent collector is not perfect Minor roughness at the surface causes light to leave the collector, because locally the light hits the uneven surface with a steep angle Fingerprints and scratches can seriously harm the efficiency of light guiding – The luminescent collector and the solar cell have to be optically coupled Otherwise, reflection losses occur at the interface between collector and air and again at the interface between air and solar cell However, the optical coupling can also cause losses: light can be scattered away from the solar cell or parasitic absorption can occur – Finally, the solar cell has to convert the radiation it receives from the collector into electricity Again, a whole set of parameters determine this process, ranging from reflection and transparency losses to thermalization and electrical losses This description is not very relevant for actually calculating the efficiency of luminescent concentrator systems, because some of the involved efficiencies are neither easy to calculate nor directly accessible by measurement Nevertheless, this description illustrates very well the effects that affect the efficiency of luminescent concentrator systems 1.27.2.2 Thermodynamic Efficiency Limits The ability to as well concentrate diffuse radiation sets the luminescent concentrator apart from all other types of concentrators Systems utilizing only geometrical optics cannot concentrate diffuse light For a discussion of this difference, the concept of étendue 590 Technology and its links with entropy are very helpful For light incident from a cone, with θinc being half of the opening angle, and a flat, not tilted illuminated area Ainc, the étendue ε can be calculated to be ẳ sin inc Ainc ẵ5 From the definition and this result, the étendue can be understood as a measure of how ‘spread out’ a light beam is in terms of angular divergence and illuminated or emitting area The étendue can be calculated for an emitted beam as well In this case, the area of the emitting surface and the opening angle of the emission cone must be applied in the above relation The étendue is closely linked to entropy If an optical system increases the étendue, entropy is generated Following Markvart in Reference 36, if εinc is the étendue of the incident beam and εemit the étendue of the emitted beam, the entropy per photon σ that is generated is σ ¼ kB ln εemit εinc ½6Š with kB is the Boltzmann constant A conservative system does not generate entropy, so in a conservative system, the étendue is constant If there are no other sources of entropy, the étendue cannot be reduced, because the entropy cannot decrease That is, any concentration with geometrical optics that decreases the illuminated area must increase the angular divergence Because for diffuse radiation, angular divergence is already at its maximum, diffuse radiation cannot be concentrated with a system using only geometrical optics As a luminescent concentrator is able to concentrate diffuse radiation, there must be another source of entropy This source of entropy can be found in the Stokes shift The dissipation of part of the excitation energy as heat generates entropy and leads to photon emission at longer wavelengths Therefore, the extent of the Stokes shift determines the maximum concentration that is achievable with a luminescent concentrator This relationship between Stokes shift and concentration has been theoretically described in Reference 37 The theoretical model considers two distinct photon fields: one that is incident on the luminescent collector and one that is emitted from the collector The entropy change Δσ1 associated with the loss of a photon from the incident Bose field is   n2 2inc ẵ7 ẳ kB ln ỵ c Fp;;inc where Fp,v,inc is the flux of photons (i.e., photons per unit time) per unit area, per unit bandwidth, and per 4π solid angle of the incident field The other parameters are the frequency of the photon νinc, the speed of light c, and the refractive index n [37] The emission of a photon with the frequency νemit increases the entropy of the emitted field Additionally, due to the Stokes shift, the energy h (νinc − νemit) is dissipated as heat at the ambient temperature T The entropy generated from these two processes is   8n2 2emit hinc emit ị ỵ ẵ8 ẳ kB ln ỵ T c Fp;ν;emit with the parameters defined for the emission field corresponding to the parameters of the incident field According to the second law of thermodynamics, it must hold ỵ Δσ ≥0 ½9Š In the argument of the logarithm in eqns [7] and [8], the ‘1’ can be neglected under illumination with sunlight and for frequencies in the visible spectral range With this approximation and the eqns [7]–[9], the concentration ratio C can be calculated to be   Fp;;emit 2emit hinc emit ị ẵ10 C :ẳ exp kB T Fp;ν;inc νinc Figure illustrates these results The maximum possible concentration has been calculated with eqn [10] for three different wavelengths of the incident light It becomes obvious that from an entropic point of view, higher concentrations can be achieved for shorter wavelengths than for longer wavelengths For short wavelengths, the maximum concentration is very high and constitutes no practical limit For longer wavelengths, however, a sufficiently large Stokes shift is necessary in order to avoid limitations for principal reasons The fact that there is a maximum concentration has one more consequence: when the maximum concentration is reached, increasing the collector area will not increase the output at the edges of the concentrator Already before the maximum concentra­ tion is reached, increasing the collector area of a large concentrator will not increase the output in the same way as increasing the area of smaller collector As a consequence, the light collection efficiency of luminescent collectors decreases with increasing size 1.27.2.3 Thermodynamic Models of the Luminescent Concentrator The picture of an incident and an emitted light field that was introduced by Yablonovitch in Reference 37 has been subsequently developed into a thermodynamic model of luminescent concentrators, for example, [20–22, 29, 38] These models were success­ fully used to describe luminescent concentrators based on luminescent quantum dots At a microscopic level, further developments Luminescent Solar Concentrator 591 Maximum concentration 106 Maximum concentration under illumination at 250 nm 550 nm 1000 nm 105 104 103 102 101 100 20 40 60 80 100 Stokes shift (nm) Figure Illustration of the maximum concentration from an entropic point of view For short wavelengths, very high concentrations are theoretically possible For longer wavelengths, a large Stokes shift is necessary to avoid limitations were presented in Reference 39 At this point, I would like to present a phenomenological thermodynamic model, which brings together the main ideas from different theoretical discussions and offers valuable insight into the working principles of luminescent concentrators If one considers a conventional luminescent collector with dye molecules embedded in a transparent matrix, the incident light field with the intensity Binc excites the ensemble of luminescent molecules in the collector out of equilibrium with the ambient temperature T Because of the fast thermal equilibration among the vibrational substates of the electronically excited state, the electrons cool down very fast to the ambient temperature But as the molecule remains nonetheless in an electronically excited state, the electrons have a chemical potential µ > 0, just as in an illuminated semiconductor The chemical potential is a measure of how many luminescent molecules are excited The emission of the ensemble of the luminescent molecules is described by the generalized Planck’s law [37, 40] The number of emitted photons per time, per area, per unit solid angle, and per frequency interval Bp,v,emit(νemit,T,μ) is Bp;;emit emit ; T; ị ẳ 2emit n2 αðνemit Þ c2  hνemit − μ exp −1 kB T  ½11Š where α(νemit) is the absorption coefficient, which is equal to the emission coefficient following Kirchhoff’s law Part of the emitted light is lost due to the escape cone of total internal reflection, but most of the light is trapped and guided in the collector to its edges As a consequence, the molecules are illuminated by not only the incident field but also the emitted and trapped light The higher the combined intensity Bint is at a point of the collector, the higher is the chemical potential, and in turn also the emission of light The chemical potential is not constant throughout the collector For instance, close to the front surface, the chemical potential is higher because the luminescent molecules are excited from the full incident field Further away from the surface, part of the incident light has been absorbed and therefore intensity is lower (Figure 3) This picture can explain why there is a maximum possible concentration The larger the collector is, the more the photons from the incident field are collected Thus, the intensity of the trapped light field that travels toward the edges also increases This increases the chemical potential, and consequently the emission of light as well The maximum concentration is reached when the chemical potential has become so high that the emitted light lost in the escape cone equals the incident field The limit obtained from this consideration is stricter than the limit presented in eqn [10] [37] The link of the maximum concentration with the Stokes shift and the problem of reabsorption can be understood considering a simple model system that features an absorption region and an emission region (see Figure 4) [13] The absorption coefficient αabs in the absorption region is much higher than that in the emission region with an absorption coefficient αemit As said before, the absorption and emissions coefficients are equal as described by Kirchhoff’s law In spite of αabs > αemit, the emission in the emission region is much larger than that in the absorption region because of the energy dependency of the generalized Planck’s law, which states that in this regime, the emission at lower energies is considerably more likely than that at higher energies Hence, the larger the Stokes shift, that is, the bigger the energy difference E2 − E1, the less frequent the emission in the absorption range ‘relative’ to the emission in the emission range With less light being emitted in the absorption range, reabsorption becomes less likely Because each reabsorption and reemission again causes escape cone losses, with less reabsorption, the escape cone losses are reduced as well Less escape cone losses mean that a higher internally guided field and a higher chemical potential are possible until the emitted light lost in the escape cone equals the incident field As a consequence, a higher maximum concentration is possible 592 Technology Binc μ (Bint) Bemit (μ) Bint Figure Illustration of the main ideas of the thermodynamic model Incident radiation with the intensity Binc excites the ensemble of luminescent molecules in the luminescent collector The fraction of excited molecules is described by the chemical potential µ of the molecule ensemble The luminescent molecules emit radiation with the intensity Bemit, which depends on the chemical potential The trapped fraction of the emitted light and the incident light combines to the internal intensity Bint This internal intensity again determines the chemical potential As the internal intensity is not constant throughout the collector, the chemical potential varies as well Absorption coeff Reflection band αabs αemit Emission Energy E1 E2 Energy Figure Idealized model of the absorption and emission characteristics of a luminescent concentrator [13] In the absorption region, the absorption coefficient αabs is high, while in the emission region, the coefficient αemit is much smaller Following Kirchhoff’s law, the emission coefficient equals the absorption coefficient Nevertheless, emission in the emission region is much higher, because the generalized Planck’s law favors emissions at lower energies A band-stop reflection filter that reflects in the emission region can increase efficiency and the maximum possible concentration considerably Because absorption and emission are linked by Kirchhoff’s law, it is not possible to eliminate reabsorption entirely Additionally, without reabsorption, the excitation of the molecules would be completely independent of the emitted light This would allow for an infinite concentration, which is a clear contradiction of the second law of thermodynamics However, it is possible to reduce the escape cone losses and therefore to increase the maximum possible concentration with the addition of a band-stop filter The band-stop filter should reflect in the emission range but should transmit in the absorption range The desirable reflection band is sketched in Figure Like this, only the small amount of light emitted in the absorption region can be subjected to escape cone losses Again, this means that a higher internally guided field and a higher chemical potential are possible until the emitted light lost in the escape cone equals the incident field In Reference 13 it was shown that the maximum efficiency of a luminescent concentrator system with such a band-stop filter equals the Shockley–Queisser limit of a solar cell with a bandgap similar to that of the cutoff wavelength E2 of the band-stop filter 1.27.2.4 Ray Tracing Simulations of Luminescent Concentrators For complex geometries with different materials, imperfect surfaces, scattering, etc., most thermodynamic models were not sufficient Therefore, several works have investigated luminescent concentrators using Monte Carlo methods and ray tracing One of the first works is the dissertation of Heidler [41] Heidler modeled absorption of the dye, isotropic emission, total internal reflection, and reabsorption The model was even capable of simulating a diffuse-back reflector and stack configurations Simulated collection efficiencies exceeded experimental data by 15% on average due to the idealized conditions of the model However, good agreement between the simulated and measured edge emission spectrum was achieved Recently, new attempts for simulating luminescent concentrators have been made [13, 14, 42–45] Kennedy et al [43, 44] use a simple model describing absorption of the dye, emission, total internal reflection, and reabsorption to calculate the relative Jsc of solar cells coupled to one edge of the collector and to predict the emitted spectrum leaving at the bottom of the collector The Jsc values were overestimated by about 10%, but Luminescent Solar Concentrator 593 again the emitted spectra agreed well with predictions Burgers et al [45] used a quite similar model He determined relevant parameters, such as the dye concentration or the quantum efficiency, by fitting the model to measured data He also included mirrors at the collector edges With his model, he achieved good agreement between external quantum efficiency (EQE) measure­ ments and the simulation-based predictions; a review was presented in Reference In References 13 and 14, a highly idealized model was presented, which for the first time included a photonic structure This idealized model has then been further developed by Prönneke et al [46] to describe more realistic systems [46] In References 26 and 47, a model was presented that was tested against spectrally resolved experimental data, such as reflection and transmission spectra as well was the emission spectra from different surfaces of the collector The shape of the photoluminescence spectrum and the angular distribution of the emitted light proved to be very important input parameters With the model it was also possible to reproduce the angular distribution of the light leaving the collector at the edges In References 26 and 48, the effect of a dependence of the photoluminescence spectrum on the excitation wavelength was investigated In Reference 48, the impact of reabsorption was examined It was found that reduced reabsorption due to a larger Stokes shift can overcompensate a lower quantum efficiency of the absorption/emission process in respect of the overall system efficiency 1.27.3 Materials for Luminescent Solar Concentrators 1.27.3.1 Organic Dyes Organic dyes were the dominant luminescent material in the first research campaign in the 1980s [5, 6, 31, 33, 49–52] They were applied both distributed in a transparent matrix material and as a thin layer on a transparent slab of material The research in that time resulted in luminescent dyes with high luminescent quantum efficiencies above 95% and good stability Figure shows a photograph of a selection of luminescent concentrator materials produced during that time Today, these luminescent dyes are commercially available and therefore also used in recent works [16, 18, 19, 53, 54] Until now, the highest reported efficiencies [19, 25] were reached with systems based on organic dyes However, high quantum efficiency is achieved only in the visible range of the spectrum, while efficiency remains low in the IR In Reference 33, it is shown that these low quantum efficiencies have fundamental reasons that are difficult to overcome The main reason is that the energy difference between the excited electronic states moves closer to the energy of vibrational transitions within the molecule, which facilitates nonradiative transitions Nevertheless, also in recent works, organic dyes are being developed that extend the efficiently used spectral range to longer wavelength For example, in Reference 55, the synthesis of a dye based on a perylene perinone is described, which extends the absorption wavelength range by more than 50 nm in comparison to the perylene-based dye Lumogen Red 305, which is very frequently used in luminescent concentrators This extended absorption allows for the collection of potentially 25% more photons at a reasonable luminescent quantum yield and photostability Another problem of the organic dyes is the large overlap between absorption and emission spectra This results in reabsorption and reemission with the associated losses Research has therefore been conducted to increase the Stokes shift of the organic dyes One option is to use energy transfer from one absorbing dye to another emitting dye [30, 56] In Reference 57, hybrid dyes were investigated, which were composed of organic antenna and inorganic emitting ions to achieve large Stokes Shifts Another option is to use phosphorescence instead of fluorescence [30] Phosphorescence is associated with a larger Stokes shift, but also with lower quantum efficiency There have also been attempts to increase overall efficiency by bringing metal nanoparticles close to the dyes, in order to increase absorption and luminescence due to plasmonic resonances [58, 59] Figure A selection of luminescent concentrator materials based on organic dyes that were produced during the first research campaign in the 1980s at Fraunhofer ISE [5, 6, 33, 49, 51] that are still among the most efficient luminescent concentrator materials The luminescent collectors consist of PMMA doped with organic dyes produced from BASF The used dyes are perylene derivates The precise chemical structures, however, were not published by BASF 594 1.27.3.2 Technology Inorganic Luminescent Materials Because of the instability of organic materials, especially under ultraviolet radiation, inorganic materials have been investigated as well Promising inorganic materials are glasses and glass ceramics doped with rare-earth ions like Nd3+ and Yb3+, or other metal ions like Cr3+ [60–63] The advantages to these approaches are high stability and a high refraction index of the glasses, which increases the trapped fraction of light One big disadvantage is the narrow absorption bands of the luminescent materials Additionally, these material systems turned out to be quite complex and costly to fabricate With the development of nanotechnology, luminescent nanocrystalline quantum dots (NQDs) have become of interest for luminescent concentrators The most frequently used materials are CdS, CdSe, and ZnS quantum dots [20, 21, 64–68] One big advantage of the NQDs is that absorption and emission properties can be tuned by the size and composition of the nanocrystals Additionally, the NQD features a broad absorption range However, the achieved quantum efficiencies are lower than those of organic dyes, especially if the NQDs are incorporated into polymer matrixes In addition, the low Stokes shift presently prohibits reaching high efficiencies; NQDs with larger Stokes shifts are under development (nanorods [68] and type II heteronanocrystals [69]), thus potentially leading to high light collection efficiencies There are also hybrid approaches; for example, in Reference 70, Er-doped WO6 nanocrystals in composite telluride glasses are investigated 1.27.4 Luminescent Solar Concentrator System Designs and Achieved Results 1.27.4.1 System Designs Absorption Resp emission Many system designs have been proposed for efficient and economic luminescent concentrator systems Probably, the most fundamental one was the concept to stack several collector plates [4] With different dyes in each plate, different parts of the spectrum can be utilized (see Figure 6) At each luminescent collector, a solar cell can be attached, which is optimized for the spectrum emitted from the collector With this spectrum splitting, high efficiencies can be achieved in principle The stack design with the matched solar cells at the edges provides a high degree of freedom for cell interconnection Therefore, there is no forced series connection like in tandem cell concepts, which causes current limitation problems Additionally, no tunnel diodes are necessary Figure shows mirrors at some of the edges of the collector plate as well If some edges are not covered with solar cells, but with reflectors, the geometric concentration is increased This can be beneficial for the costs of the luminescent concentrator system, as solar cells are usually the most expensive component of the system However, the reflection on mirrors is not free of losses Therefore, it should be kept to a minimum and the emitted light should reach the solar cells with as few reflections on mirrors as possible For this purpose, an isosceles and rectangular triangular shape of the luminescent collector is beneficial [4] With solar cells at the hypotenuse and the two other sides covered with mirrors, only two reflections are necessary at most until the emitted light hits a solar cell A reflector underneath the collector increases the collection efficiency as well It reflects transmitted light back into the collector and creates a second chance for absorption When a white reflector instead of a mirror is used, light can also be scattered and redirected toward the solar cells For both reflectors underneath the collector and mirrors at the edges, it is beneficial to maintain an air gap between collector and reflector In this configuration, the reflection of the reflector comes on top of total internal reflection However, with an air gap, the diffuse reflector does not change the direction of light emitted into the escape cone to directions that are subject to total internal reflection The reason for this is that due to refraction, the light that leaves the collector is already λ1 S1 C1 λ2 Wavelength λ S2 Mirrors C2 λ3 S3 λ C3 (a) λ (b) Figure Concept of stacked luminescent concentrators, as presented in Reference (a) The different collectors C1−C3 are connected with different solar cells S1−S3 In each collector, a different dye is incorporated The absorption and emission (shaded) spectra of the different dyes are shown in (b) With a proper alignment of the absorption and emission properties, the recycling of photons lost from one collector to another collector is possible It is important that an air gap between the different collectors is maintained so that each spectral range of light is guided in one collector by total internal reflection and does not get lost in adjacent collectors Luminescent Solar Concentrator 595 distributed over a complete hemisphere, even before it hits the diffuse reflector This is not changed by diffuse reflection So consequently, when the light enters the collector again, it is refracted into exactly the angles of the escape cone Another idea to increase the geometric concentration was proposed in Reference 71 The angular range of the edge emission of the luminescent concentrator is limited by the critical angle of total internal reflection Therefore, a further concentration is possible until the divergence reaches the full hemisphere Compound parabolic concentrators, which are attached to the edges, are one possibility for this purpose As mentioned before, no luminescent materials that are active in the IR while showing high quantum efficiency, high stability, and broad absorption have been developed so far Therefore, a range of designs were proposed to utilize the IR radiation One option could be to place a cheap solar cell underneath the collector that utilizes IR radiation [15] The IR light transmitted through the collector could be used as well by a thermal collector It was also suggested to use an upconverter to convert the transmitted radiation into light that could be collected by the luminescent collector [34] The transmitted light can also be used to grow plants in a greenhouse [51] A complex geometry for building integrated PV was also presented by Chatten et al in Reference 72, where a luminescent concentrator is integrated into a blind system Finally, the application of luminescent concentrators is also discussed in the context of day lighting, for example, [73], where only the light collection properties are used and no solar cells are involved 1.27.4.2 Achieved System Efficiencies Many materials have been investigated for their potential use in luminescent concentrators However, the number of luminescent concentrator systems consisting of luminescent collectors with attached solar cells that were tested under standard testing conditions used for solar cells remained relatively small The highest efficiencies have been reached based on the combination of different organic dyes Already in the first research campaign in the 1980s, Wittwer et al [51] achieved a conversion efficiency of 4% with a system that combined two mm-thick plates with different dyes in one stack with GaAs solar cells attached to the edges The system was 40 cm  40 cm in size and therefore quite large, so the achieved efficiency can be considered a very good result The geometric concentration ratio, that is, the ratio of the illuminated collector area to the solar-cell area, was 16.7 The system produced around times more energy than that the solar cells would have produced if they had been placed directly in the sun The highest reported efficiency was achieved by Slooff et al [19] In this work, four GaAs were attached to a cm  cm luminescent collector with a thickness of mm The collector consisted of PMMA The collector contained two dyes, 0.01 wt.% Lumogen F Red 305 (Red305) from BASF (a perylene) and 0.003 wt.% Fluorescence Yellow CRS040 (CRS040) from Radiant Color (a coumarine) At the bottom of the collector, a diffuse reflector was placed With this configuration, a system efficiency of 7.1% was achieved The geometric concentration of this system was 2.5 While both described systems used two different dyes, only one type of solar cell was used, therefore not fully exploiting the possibilities of spectrum splitting described in the previous section In References 26 and 74, a system was investigated that used as well different types of solar cells, made of GaInP and GaAs GaInP solar cells have a bandgap of 1.85 eV, which corresponds to a wavelength of 670 nm The typical open-circuit voltage (VOC) of a GaInP solar cell is in the region above 1300 mV The bandgap of GaAs is at 1.43 eV and the typical VOC is above 1000 mV The dimensions of the used collector plates were cm  cm  0.5 cm One plate contained a dye active in the spectral region of 400–550 nm It was used to illuminate the GaInP solar cells The second collector plate contained a dye active between 550 and 650 nm This one was used to illuminate the GaAs solar cells To each of the plates, two solar cells were optically coupled with silicone to adjoining edges In front of the remaining two edges of each collector plate, white reflectors made from polytetrafluoroethylene (PTFE) were placed In this configuration, the geometric concentration defined as the ratio of the luminescent collector area to the area of the used solar cells is 2.5 as well At the bottom of the system, a diffuse reflector was placed Overall, a system efficiency of 6.9% was achieved Figure shows the EQE of the two subsystems 50 EQE (%) 40 GalnP GaAs 30 20 10 300 400 500 600 700 800 900 Wavelength (nm) Figure EQE measurements of the two subsystems, consisting of two parallel interconnected GaInP solar cells and two parallel interconnected GaAs solar cells, respectively, coupled to two different luminescent collector plates The two systems together cover a wide spectral range 596 Technology In all the works [19, 26, 74], it was found that the bottom reflector contributed significantly to the overall system efficiency This can be seen as well in Figure 7, where an EQE of more than 5% is visible at wavelengths longer than the active region of the dyes, which ends at about 650 nm This effect is expected to decrease for bigger systems 1.27.5 The Future Development of Luminescent Solar Concentrators The main task for the future development of luminescent concentrators is to increase system efficiencies based on cheap materials in reasonably sized systems When this task is solved successfully, the integration into applications can be undertaken To increase system efficiencies, there are two main topics: • Extending the used spectral range into the IR • Controlling escape cone losses 1.27.5.1 Extending the Used Spectral Range into the IR From the EQE measurement presented in Figure 7, it is obvious that in the active region of the dyes, the systems reach already quite high quantum efficiencies of approximately 45% As the used solar cells also deliver high voltages and show high fill factors, the main reason for the low overall efficiency is that only the visible part of the spectrum is used If one reached a 45% quantum efficiency also in the range from 650 to 1050 nm, one could expect an extra current density of around 12 mA cm−2 When the luminescent material emits in the region between 1050 and 1125 nm, the emitted light could be used by a silicon solar cell, which reaches a maximum power point voltage of around 580 mV The extra silicon solar-cell luminescent concentrator system then would have an efficiency of nearly 7%, which would result in an overall system efficiency of close to 14% Even with only silicon solar cells attached, which would result in approximately half the efficiency for the higher energy photons because of the lower voltage of the silicon solar cells, an overall system efficiency of 10% appears to be feasible However, these are all highly hypothetical cases, which require a lot of material development for near-infrared (NIR) luminescent materials The concepts presented in Section 1.27.3, especially the use of hybrid materials [57, 70] and the use of luminescent semiconductor quantum dots, might be able to achieve this task However, the already achieved progress in these fields still remains to be demonstrated at a system level 1.27.5.2 Controlling Escape Cone Losses Besides the losses due to an incomplete utilization of the full solar spectrum, the escape cone of total internal reflection is the most important loss mechanism The loss of around 26% does occur not only once but also after every reabsorption and reemission, as already introduced in Section 1.27.2.1 To reduce the escape cone losses, one can identify two different strategies: first, reabsorption can be reduced to reduce the number of emission events The different approaches in the development of luminescent materials with low reabsorption were already discussed in Section 1.27.3 The second approach is to control the path of the light in the luminescent collector In this field, arguably, the biggest progress has been made in the last years 1.27.5.2.1 Photonic structures for increased efficiencies The Stokes shift between absorption and emission opens the opportunity to reduce escape cone losses significantly: a selective reflector, which transmits all the light in the absorption range of the luminescent material and reflects the emitted light, would trap nearly all the emitted light inside the collector [75] The concept is illustrated in Figure In Reference 12, hot mirrors were proposed to serve as selective reflectors and, in Reference 13, photonic structures A possible realization of such a selective reflector is a so-called rugate filter It features a continuously varying refractive index profile that results in a single reflection peak However, some unwanted side lobes remain Optimized rugate filters [76] show only one single reflection peak for a certain wavelength and almost no other reflections In Reference 25, it was shown that overall system efficiencies of luminescent concentrator systems can be increased with the help of such an optimized filter The investigated system used a cm  10 cm, 5-mm-thick luminescent collector of PMMA doped with an organic dye One GaInP solar cell was coupled to one short edge with silicone The solar cell had an active area of mm  49 mm Hence, the ratio of illuminated luminescent concentrator area and solar-cell area constitutes a geometric concentration ratio of 20  The solar cell had an efficiency of 16.7% under AM1.5G illumination White PTFE served as bottom reflector and also as reflector at the edges, which were not covered by solar cells The used photonic structure was produced by the company mso-jena by ion-assisted deposition (IAD) and was tuned for high reflection in the emission range of the organic dye (see Figure 9) Without this structure, the system had an efficiency of 2.6 Ỉ 0.1% in reference to the 50 cm2 area of the system The structure increased the efficiency to 3.1 Ỉ 0.1%, which constitutes an efficiency increase of around 20% relative Figure 10 shows the result from a light beam-induced current (LBIC) scan of the system, illustrating how the light collection efficiency is increased over most of the luminescent collector area With the achieved efficiency of 3.1% and the concentration ratio of 20, the realized luminescent concentrator produces about 3.7 times more energy than the GaInP solar cell had produced on its own Luminescent Solar Concentrator Radiation E1 597 Air gaps Photonic structure No loss cone E2 Solar cell Solar cell Reabsorption and emission Dye White-bottom reflector Figure A selective reflector, realized as a photonic structure, reduces the escape cone losses The photonic structure acts as a band-stop reflection filter It allows light in the absorption range of the dyes to enter the collectors but reflects light in the emission range 100 Absorption BA241 PL emission BA241 Refletion filter (%) 80 60 40 20 350 400 450 500 550 600 Wavelength (nm) 650 700 750 Collection efficiency (a.u) Figure Reflection spectrum of the used photonic structure and the absorption and photoluminescence of the luminescent concentrator the filter was designed for The reflection of the structure very nicely fits the emission peak of the dye in the concentrator 600 With Without Photonic bandstop filter 500 400 300 Distance from solar cell (cm) 10 Figure 10 Averaged linescans in x-direction from an LBIC scan with and without photonic structure Close to the solar cell, the efficiency is lower with the photonic structure, because it reduces the effectiveness of the bottom reflector for small distances Over most of the luminescent concentrator, however, collection efficiency is significantly higher with a photonic structure, resulting in a relative efficiency increase of 20% 598 Technology E1 t≈λ E2 Figure 11 Conceptual sketch of a ‘Nano-Fluko’ A very thin layer of luminescent material with thickness t in the range of wavelength λ of the emitted light is placed between two photonic structures, for example, Bragg stacks The photonic structures transmit light in the absorption range of the luminescent material with an energy E1 They are reflective in the emission region (E2) of the luminescent material Because the layer with the luminescent material is so thin, the photonic structures suppress the emission into unfavorable directions The observed efficiency increase of 20% can be already considered as a great success since it shows that photonic structures reduce the escape cone losses significantly However, the used filter is a multilayer system and therefore costly to produce Three-dimensional (3D) photonic structures are a potential alternative to the presented multilayer systems A special three-dimensional photonic structure is the opal The opal has the advantage that it can be produced by a dip-coating process utilizing self-organization of monodisperse PMMA beads [77] This is a potentially low-cost process that could be applied on large-area concentrators However, the achieved quality of opaline films is still too low to achieve optical properties that allow for an increase in efficiency Another, probably more promising options are chiral nematic (cholesteric) liquid crystals that act as spectrally selective mirrors In Reference 78, such chiral nematic liquid crystals were applied onto luminescent collectors, with a small air gap in between It was observed that the highest output is achieved using a scattering background and cholesteric mirror, with a reflection band significantly redshifted (similar to 150 nm) from the emission peak of the luminescent dye The use of an air gap results in light bending away from the waveguide surface normal and, consequently, a redshift of the cholesteric mirrors is required Also, the importance of considering the angular dependence of the spectrally selective mirrors was analyzed in Reference 79 Overall, up to 35% more emitted light exits the luminescent collector edge after application of the cholesteric mirror However, even with a photonic structure, light emitted into the escape cone is more frequently subject to loss events Because it is emitted into a steep angle with respect to the front surface, it has a very long effective path until it reaches a solar cell and therefore suffers more from path length-dependent losses Hence, it would be very beneficial to suppress emission into these unfavorable directions completely This should as well be possible with the help of photonic structures Already the very first works on photonic crystals of Bykov [80] and Yablonovitch [81] dealt with influencing emission with photonic structures Many papers discussed the possibilities subsequently [82–87] For influencing the emission of the dye successfully, it is necessary that the photonic structures are very close to the emission process or that the luminescent material is incorporated into the photonic structures For the luminescent concentrator systems, this means that one has to go from the macroscopic design of the presented systems to a system design in the nanoscale Possible realizations of such a system denoted ‘Nano-Fluko’ were suggested in Reference 88 One possible realization consists of a very thin layer of luminescent material between two photonic structures, for example, rugate filters or Bragg stacks (Figure 11) In such a configuration, the emission of the light would be restricted to a plane parallel to the photonic structure Galli et al [89, 90] showed that the emission of Er 3+ can be strongly enhanced if it is incorporated in a photonic crystal waveguide and that efficient waveguiding occurs Therefore, there is first experimental evidence that such a system can work, and it is an interesting approach to apply this concept to luminescent concentrators However, several layers with the same dye will be needed to achieve sufficient absorption As the guided light is constraint to very thin layers, high intensities will occur in these layers Because of thermodynamic limit for the achievable concentration depending on the Stokes shift of the used dye, one question is which system sizes can be achieved using this approach until the thermodynamic limit reduces efficiency 1.27.5.2.2 Controlling the angular emission An alternative to selective reflectors is to modify the emission characteristic of the dyes in such a way that emission occurs predominantly in favorable directions This can be achieved with an orientation of dye molecules that show a distinct angular characteristic in their emissions depending on their position The dye molecules can be aligned accordingly with the help of liquid crystals [17, 91–93] The dye alignment has to take place such that the optical transition dipole of the luminescent material (the dye molecule) is oriented along the luminescent collector surface normal, directing the maximum possible proportion of luminescence into waveguide modes In Reference 91, it is reported that up to 30% more light is emitted from the edge of a luminescent collector due to the dye alignment Luminescent Solar Concentrator 599 1.27.6 Conclusion Luminescent solar concentrators have the fascinating ability to concentrate both direct and diffuse radiation This ability is directly related to the Stokes shift that occurs between absorption of incoming light and the subsequent emission Up to now, system efficiencies of around 7% have been achieved The efficiency potential that could be achieved based on commercially attractive materials such as silicon solar cells is in the range of 10% To achieve this goal, further progress in the development of luminescent materials that cover the visible and the near-IR range of the solar spectrum, showing high luminescent quantum efficiencies and low reabsorption, is necessary Furthermore, current progress in the research on photonic structures needs to be exploited for its application in luminescent concentrator systems With continuously falling prices for the production of solar cells, however, the cost advantages, which can be achieved by using cheap concentrators such as luminescent collectors, decrease Hence, it is unlikely that luminescent concentrators will play an important role for rooftop or power-plant applications Nevertheless, the unique optical appearance of luminescent concentrators, the possibilities to achieve transparency and to combine different colors, the ‘invisibility’ of solar cells integrated into the frames of modules, etc offer unique design possibilities and might open the way for luminescent concentrators into building integrated photovoltaics and also into the power supply of small appliances and portable consumer electronics Therefore, they could help to spread photovoltaics into fields that are not accessible by conventional solar-cell module technology Acknowledgments Some of the presented work was supported by the German Research Foundation within the Nanosun (PAK88) project I would like to thank all colleagues at Fraunhofer ISE that are or have been involved in the research on luminescent solar concentrators, especially Marius Peters, Armin Zastrow, Volker Wittwer, and Adolf Goetzberger References [1] Shurcliff WA and Jones RC (1949) The trapping of fluorescent light produced within objects of high geometrical symmetry Journal of the Optical Society of America 39(11): 912–916 [2] Birks JB (1964) The Theory and Practice of Scintillation Counting London: Pergamon [3] Weber WH and Lambe J (1976) Luminescent greenhouse collector for solar radiation Applied Optics 15(10): 2299–2300 [4] Goetzberger A and Greubel W (1977) Solar energy conversion with fluorescent collectors Applied Physics 14: 123–139 [5] Wittwer V, Heidler K, Zastrow A, and Goetzberger A (1981) Theory of fluorescent planar concentrators and experimental results Journal of Luminescence 24/25: 873–876 [6] Seybold G and Wagenblast G (1989) New perylene and violanthrone dyestuffs for fluorescent collectors Dyes and Pigments 11: 303–317 [7] van Roosmalen JAM (2004) Molecular-based concepts in PV towards full spectrum utilization Semiconductors 38(8): 970–975 [8] van Sark WGJHM, Barnham KWJ, Slooff LH, et al (2008) Luminescent solar concentrators – A review of recent results Optics Express 16(26): 21773–21792 [9] Luque A, Martí A, Bett AW, et al (2005) FULLSPECTRUM: A new PV wave of more efficient use of solar spectrum Solar Energy Materials and Solar Cells 87(1–4): 467–479 [10] Goldschmidt JC, Glunz SW, Gombert A, and Willeke G (2006) Advanced fluorescent concentrators In: Poortmans J, Ossenbrink H, Dunlop E, and Helm P (eds.) Proceedings of the 21st European Photovoltaic Solar Energy Conference, pp 107–110 Dresden, Germany [11] Richards BS and Shalav A (2005) The role of polymers in the luminescence conversion of sunlight for enhanced solar cell performance Synthetic Metals 154: 61–64 [12] Richards BS, Shalav A, and Corkish R (2004) A low escape-cone-loss luminescent solar concentrator In: Hoffmann W, et al (eds.) Proceedings of the 19th European Photovoltaic Solar Energy Conference, pp 113–116 Paris, France [13] Rau U, Einsele F, and Glaeser GC (2005) Efficiency limits of photovoltaic fluorescent collectors Applied Physics Letters 87(17): 171101–1–3 [14] Glaeser GC and Rau U (2006) Collection and conversion properties of photovoltaic fluorescent collectors with photonic band stop fillers In: Gombert A (ed.) Proceedings of SPIE, pp 143–153 Bellingham, WA: SPIE Press [15] Goldschmidt JC, Peters M, Löper P, et al (2007) Advanced fluorescent concentrator system design In: Willecke G, Ossenbrink H, and Helm P (eds.) Proceedings of the 22nd European Photovoltaic Solar Energy Conference, pp 608–612 Milan, Italy [16] Danos L, Kittidachachan P, Meyer TJJ, et al (2006) Characterisation of fluorescent collectors based on solid, liquid and Langmuir Blodgett (LB) films In: Poortmans J, Ossenbrink H, Dunlop E, and Helm P (eds.) Proceedings of the 21st European Photovoltaic Solar Energy Conference, pp 443–446 Dresden, Germany [17] Debije MG, Broer DJ, and Bastiaansen CWM (2007) Effect of dye alignment on the output of a luminescent solar concentrator In: Willecke G, Ossenbrink H, and Helm P (eds.) Proceedings of the 22nd European Photovoltaic Solar Energy Conference, pp 87–89 Milan, Italy [18] Slooff LH, Budel T, Burgers AR, et al (2007) The luminescent concentrator: Stability issues In: Willecke G, Ossenbrink H, and Helm P (eds.) Proceedings of the 22nd European Photovoltaic Solar Energy Conference, pp 584–588 Milan, Italy [19] Slooff LH, Bende EE, Burgers AR, et al (2008) A luminescent solar concentrator with 7.1% power conversion efficiency Physica Status Solidi—Rapid Research Letters 2(6): 257–259 [20] Chatten AJ, Barnham KWJ, Buxton BF, et al (2003) The quantum dot concentrator: Theory and results In: Kurokawa K, et al (eds.) Proceedings of the 3rd World Conference on Photovoltaic Energy Conversion, pp 2657–2660 Osaka, Japan [21] Chatten AJ, Barnham KWJ, Buxton BF, et al (2001) Novel quantum dot concentrators In: McNelis B, Palz W, Ossenbrink H, and Helm P (eds.) Proceedings of the 17th European Photovoltaic Solar Energy Conference, pp 200–203 Munich, Germany [22] Chatten AJ, Barnham KWJ, Buxton BF, and Ekins-Daukes NJ (2003) A new approach to modelling quantum dot concentrators Solar Energy Materials and Solar Cells 75(3–4): 363–371 [23] Goldschmidt JC, Peters M, Prönneke L, et al (2008) Theoretical and experimental analysis of photonic structures for fluorescent concentrators with increased efficiencies Physica Status Solidi A 205(12): 2811–2821 [24] Peters M, Goldschmidt JC, Löper P, et al (2009) The effect of photonic structures on the light guiding efficiency of fluorescent concentrators Journal of Applied Physics 105: 014909 [25] Goldschmidt JC, Peters M, Bösch A, et al (2009) Increasing the efficiency of fluorescent concentrator systems Solar Energy Materials and Solar Cells 93: 176–182 600 [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] Technology Goldschmidt JC (2009) Novel Solar Cell Concepts, p 273 München: Verlag Dr Hut Peters M, Goldschmidt JC, Löper P, et al (2010) Spectrally-selective photonic structures for PV applications Energies 3(2): 171–193 Peters IM (2009) Photonic concepts for solar cells In: Faculty of Mathematics and Physics, p 202 Freiburg, Germany: Universität Freiburg Barnham K, Marques JL, Hassard J, and O’ Brien P (2000) Quantum-dot concentrator and thermodynamic model for the global redshift Applied Physics Letters 76(9): 1197–1199 Currie MJ, Mapel JK, Heidel TD, et al (2008) High-efficiency organic solar concentrators for photovoltaics Science 321: 226–228 Viehmann W and Frost RL (1979) Thin film waveshifter coatings for fluorescent radiation converters Nuclear Instruments and Methods 167: 405–415 Bose R, Farrell DJ, Pardo-Sanchez C, et al (2009) Luminescent solar concentrators: Cylindrical design In: Sinke W, Ossenbrink H, and Helm P (eds.) Proceedings of the 24th European Photovoltaic Solar Energy Conference, pp 359–362 Hamburg, Germany Zastrow A (1981) Physikalische Analyse der Energieverlustmechanismen im Fluoreszenzkollektor In: Fakultät für Physik, p 209 Freiburg: Albert-Ludwigs-Universität Freiburg Goetzberger A (2009) Fluorescent solar energy concentrators: Principle and present state of development In: Petrova-Koch V, Hezel R, and Goetzberger A (eds.) High-Efficient Low-Cost Photovoltaics: Recent Developments, pp 159–176 Berlin: Springer-Verlag GmbH Keil G (1969) Radiance amplification by a fluorescence radiation converter Journal of Applied Physics 40(9): 3544–3547 Markvart T (2008) Solar cell as a heat engine: Energy–entropy analysis of photovoltaic conversion Physica Status Solidi A 205(12): 2752–2756 Yablonovitch E (1980) Thermodynamics of the fluorescent planar concentrator Journal of the Optical Society of America 70(11): 1362–1363 Chatten AJ, Farrell DJ, Büchtemann A, and Barnham KWJ (2006) Thermodynamic modelling of luminescent solar concentrators In: Poortmans J, Ossenbrink H, Dunlop E, and Helm P (eds.) Proceedings of the 21st European Photovoltaic Solar Energy Conference, pp 315–320 Dresden, Germany Scudo PF, Abbondanza L, Fusco R, and Caccianotti L (2009) Spectral converters and luminescent solar concentrators Solar Energy Materials and Solar Cells 94(7): 1241–1246 Würfel P (1995) Physik der Solarzellen [Physics of Solar Cells] Heidelberg, Germany: Spektrum Akademischer Verlag Heidler K (1982) Wirkungsgraduntersuchungen zur Solarenergiekonversion mit Fluoreszenzkollektoren, in Fakultät für Physik, p 135 Freiburg, Germany: Universität Freiburg Gallagher SJ, Eames PC, and Norton B (2004) Quantum dot solar concentrator behaviour, predicted using a ray trace approach International Journal of Ambient Energy 25(1): 47–56 Kennedy M, Dunne M, McCormack SJ, et al (2008) Multiple dye luminescent solar concentrators and comparison with Monte-Carlo ray-trace predictions In: Lincot D, Ossenbrink H, and Helm P (eds.) Proceedings of the 23rd European Photovoltaic Solar Energy Conference, pp 390–393 Valencia, Spain Kennedy M, Chatten AJ, Farrell DJ, et al (2008) Luminescent solar concentrators: A comparison of thermodynamic modelling and ray-trace modelling predictions In: Lincot D, Ossenbrink H, and Helm P (eds.) Proceedings of the 23rd European Photovoltaic Solar Energy Conference, pp 334–337 Valencia, Spain Burgers AR, Slooff LH, Kinderman R, and van Roosmalen JAM (2005) Modelling of luminescent concentrators by ray-tracing In: Palz W, Ossenbrink H, and Helm P (eds.) Proceedings of the 20th European Photovoltaic Solar Energy Conference, pp 394–397 Barcelona, Spain Prönneke L, Gläser G, Uslu Y, and Rau U (2009) Measurement and simulation of enhanced photovoltaic system with fluorescent collectors In: Sinke W, Ossenbrink H, and Helm P (eds.) Proceedings of the 24th European Photovoltaic Solar Energy Conference, pp 385–387 Hamburg, Germany Bendig M, Hanika J, Dammertz H, et al (2008) Simulation of fluorescent concentrators In: Parker S and Reshetov A (eds.) IEEE/EG Symposium on Interactive Ray Tracing, pp 93–98 Los Angeles, CA, USA Wilson LR, Rowan BC, Robertson N, et al (2010) Characterization and reduction of reabsorption losses in luminescent solar concentrators Applied Optics 49(9): 1651–1661 Zastrow A, Heidler K, Sah RE, et al (1980) On the conversion of solar radiation with fluorescent planar concentrators In: Palz W (ed.) Proceedings of the 3rd European Photovoltaic Solar Energy Conference, Cannes, France, pp 413–417 Dordrecht, The Netherlands: Springer Sah RE (1981) Stokes shift of fluorescent dyes in the doped polymer matrix Journal of Luminescence 24/25: 869–872 Wittwer V, Stahl W, and Goetzberger A (1984) Fluorescent planar concentrators Solar Energy Materials 11: 187–197 Mugnier J, Dordet Y, Pouget J, et al (1987) Performances of fluorescent solar concentrators doped with a new dye (benzoxazinone derivative) Solar Energy Materials 15: 65–75 Reisfeld R, Shamrakov D, and Jorgensen C (1994) Photostable solar concentrators based on fluorescent glass films Solar Energy Materials and Solar Cells 33: 417–427 Hammam M, El-Mansy MK, El-Bashir SM, and El-Shaarawy MG (2007) Performance evaluation of thin-film solar concentrators for greenhouse applications Desalination 209: 244–250 Debije MG, Verbunt PPC, Nadkarni PJ, et al (2011) Promising fluorescent dye for solar energy conversion based on a perylene perinone Applied Optics 50(2): 163–169 Calzaferri G (2005) Light-harvesting host-guest antenna materials for thin film solar cells In: Kenny R, Dunlop E, and Luque A (eds.) Proceedings of the Fullspectrum Meeting Ispra, Italy, pp 157–166 Wu WX, Wang TX, Wang X, et al (2010) Hybrid solar concentrator with zero self-absorption loss Solar Energy 84(12): 2140–2145 Wilson HR (1987) Fluorescent dyes interacting with small silver particles; a system extending the spectral range of fluorescent solar concentrators Solar Energy Materials 16: 223–234 Hinsch A, Zastrow A, and Wittwer V (1990) Sol–gel glasses: A new material for solar fluorescent planar concentrators? Solar Energy Materials 21: 151–164 Reisfeld R and Kalisky Y (1981) Nd3+ and Yb3+ germanate and tellurite glasses for fluorescent solar energy collectors Chemical Physics Letters 80(1): 178–183 Reisfeld R (1984) Fluorescence and nonradiative relaxations of rare earths in amorphous media on high surface area supports: A review Journal of the Electrochemical Society: Solid-State Science and Technology 131(6): 1360–1364 Andrews LJ, McCollum BC, and Lempicki A (1981) Luminescent solar collectors based on fluorescent glasses Journal of Luminescence 24/25: 877–880 Neuroth N and Haspel R (1987) Glasses for luminescent solar concentrators Solar Energy Materials 16: 235–242 Schüler A, Python M, Valle del Olmo M, and de Chambrier E (2006) Quantum dot containing nanocomposite thin films for photoluminescent solar concentrators Solar Energy 81: 1159–1165 Reda SM (2008) Synthesis and optical properties of CdS quantum dots embedded in silica matrix thin films and their applications as luminescent solar concentrators Acta Materialia 56: 259–264 Gallagher SJ, Norton B, and Eames PC (2007) Quantum dot solar concentrators: Electrical conversion efficiencies and comparative concentrating factors of fabricated devices Solar Energy 81: 813–821 Wang CH, Shcherbatyuk G, Inman R, et al (2010) Efficiency improvement by near infrared quantum dots for luminescent solar concentrators In: Tsakalakos L (ed.) Next Generation, p 77720G Bellingham: Spie-Int Soc Optical Engineering Bomm J, Buechtemann A, Chatten AJ, et al (2011) Fabrication and full characterization of state-of-the-art quantum dot luminescent solar concentrators Solar Energy Materials and Solar Cells 95(8): 2087–2094 van Sark WGJHM, de Mello Donega C, and Schropp REI (2010) Towards quantum dot solar concentrators using thin film solar cells In: De Santi GF, Ossenbrink H, and Helm P (eds.) 25th European Photovoltaic Solar Energy Conference, pp 289–292 Munich, Germany: WIP-Renewable Energies Balaji S, Misra D, and Debnath R (2011) Enhanced green emission from Er2WO6 nanocrystals embedded composite tellurite glasses Journal of Fluorescence 21(3): 1053–1060 Goetzberger A and Schirmer O (1979) Second stage concentration with tapers for fluorescent solar collectors Applied Physics 19: 53–58 Chatten AJ, Farrell DJ, Bose R, et al (2011) Luminescent and geometric concentrators for building integrated photovoltaics In: Swanson RM, et al (eds.) 37th IEEE Photovoltaic Specialist Conference Seattle, WA, USA Luminescent Solar Concentrator 601 [73] Chen W, Abdul-Rahman H, and Rao SP (2010) Development and testing of a fluorescent fiber solar concentrator for remote daylighting Journal of Energy Engineering 136(3): 76–86 [74] Goldschmidt JC, Peters M, Dimroth F, et al (2009) Developing large and efficient fluorescent concentrator systems In: Sinte W, Ossenbrink H, and Helm P (eds.) Proceedings of the 24th European Photovoltaic Solar Energy Conference, pp 207–212 Hamburg, Germany [75] Smestad G, Ries H, Winston R, and Yablonovitch E (1990) The thermodynamic limit of light concentrators Solar Energy Materials 21: 99–111 [76] Southwell WH (1989) Using apodization functions to reduce sidelobes in rugate filters Applied Optics 28(23): 5091–5094 [77] Jiang P, Bertone JF, Hwang KS, and Colvin VL (1999) Single-crystal colloidal multilayers of controlled thickness Chemistry of Materials 11(8): 2132–2140 [78] Debije MG, Van MP, Verbunt PPC, et al (2010) Effect on the output of a luminescent solar concentrator on application of organic wavelength-selective mirrors Applied Optics 49(4): 745–751 [79] de Boer DKG (2010) Optimizing wavelength-selective filters for Luminescent Solar Concentrators In: Wehrspohn RB and Gombert A (eds.) Photonics for Solar Energy Systems III, p 77250Q Bellingham: Spie-Int Soc Optical Engineering [80] Bykov VP (1972) Spontaneous emission in a periodic structure Soviet Physics JETP 35: 269 [81] Yablonovitch E (1987) Inhibited spontaneous emission in solid-state physics and electronics Physical Review Letters 58(20): 2059–2062 [82] Kweon G and Lawandy NM (1995) Quantum electrodynamics in photonic crystals Optics Communications 118: 388–411 [83] Vats N, John S, and Busch K (2002) Theory of fluorescence in photonic crystals Physical Review A 65(043808): 1–13 [84] Schriemer HP, van Driel HM, Koenderink AF, and Vos WL (2000) Modified spontaneous emission spectra of laser dye in inverse opal photonic crystals Physical Review A 63(011801): 1–4 [85] Nair RV, Vijaya R, Kuroda K, and Sakoda K (2007) Emission studies on photonic crystals fabricated using dyed polystyrene colloids Journal of Applied Physics 102: 123106 [86] Chigrin DN (2009) Spatial distribution of the emission intensity in a photonic crystal: Self-interference of Bloch eigenwaves Physical Review A 79(033829): 1–9 [87] Busch K, von Freymann G, Linden S, et al (2007) Periodic nanostructures for photonics Physics Reports 444(3–6): 101–202 [88] Goldschmidt JC, Peters M, Gutmann J, et al (2010) Increasing fluorescent concentrator light collection efficiency by restricting the angular emission characteristics of the incorporated luminescent material – The ‘nano-fluko’ concept In: Wehrspohn RB and Gombert A (eds.) Conference on Photonics for Solar Energy Systems III, 77250S Bellingham, WA: SPIE Press [89] Galli M, Gerace D, Politi A, et al (2006) Direct evidence of light confinement and emission enhancement in active silicon-on-insulator slot waveguides Applied Physics Letters 89(241114): 241114/1–3 [90] Galli M, Politi A, Belotti M, et al (2006) Strong enhancement of Er3+ emission at room temperature in silicon-on-insulator photonic crystal waveguides Applied Physics Letters 88(251114): 251114/1–3 [91] Verbunt PPC, Kaiser A, Hermans K, et al (2009) Controlling light emission in luminescent solar concentrators through use of dye molecules aligned in a planar manner by liquid crystals Advanced Functional Materials 19(17): 2714–2719 [92] MacQueen RW, Cheng YY, Clady R, and Schmidt TW (2010) Towards an aligned luminophore solar concentrator Optics Express 18(13): A161–A166 [93] Mulder CL, Reusswig PD, Beyler AP, et al (2010) Dye alignment in luminescent solar concentrators: II Horizontal alignment for energy harvesting in linear polarizers Optics Express 18(9): A91–A99  ... further developments Luminescent Solar Concentrator 5 91 Maximum concentration 10 6 Maximum concentration under illumination at 250 nm 550 nm 10 00 nm 10 5 10 4 10 3 10 2 10 1 10 0 20 40 60 80 10 0 Stokes shift... Functional Materials 19 (17 ): 2 714 –2 719 [92] MacQueen RW, Cheng YY, Clady R, and Schmidt TW (2 010 ) Towards an aligned luminophore solar concentrator Optics Express 18 (13 ): A1 61 A166 [93] Mulder CL,... Energy Materials 16 : 22 3–2 34 Hinsch A, Zastrow A, and Wittwer V (19 90) Sol–gel glasses: A new material for solar fluorescent planar concentrators? Solar Energy Materials 21: 15 1 16 4 Reisfeld R