Luminous Glass: A Study on the Optics Governing Luminescent Solar Concentrators and Optimization of Luminescent Materials through Combinatorial Gradient Sputter Deposition

A luminescent solar concentrator (LSC) is a concept from the 1970s that can find novel application as an electricity-generating window. An LSC converts sunlight to light of a different color by a process called luminescence. This light is transported to the edges of the LSC, where photovoltaic cells convert this incoming light to electricity. Since only a small part of the incoming sunlight is absorbed, most sunlight will still illuminate the rooms behind the LSC-window. Turning buildings and offices into nearly zero-energy buildings (nZEBs) is unlikely to happen by using electricity from rooftop photovoltaics (PVs) alone. Turning the envelope of a building, especially the large amount of glass used as windowpanes or facades, into a source of electricity by using LSCs can go a long way towards making these nZEBs a reality. Why then is not every window already an LSC? As will be explained in Chapter 2 and Chapter 3, current LSCs can be efficient at converting sunlight, but suffer from strong coloring, or are not compatible with large-scale industrial processes.
To solve the issue of coloring, one solution is to dope halides, such as table salt, with rare-earth elements, specifically divalent Thulium (Tm2+). This combination absorbs the entire visible spectrum. Another strategy is to dope insulating nitride or oxynitride materials with divalent or trivalent Europium (Eu2+ or Eu3+). Eu2+ or Eu3+ are strong absorbers of ultraviolet light.
In this thesis, optimizing the luminescent properties of these rare-earth-doped materials is researched using combinatorial synthesis methodology and a novel, fast but detailed characterization setup. The combinatorial synthesis methodology implies that a continuum of rare-earth-doped compositions is deposited on a single 5 × 5 cm2 piece of glass. This composition spread is equivalent to many hundreds of individual samples. The novel characterization setup can characterize the luminescence and other optical properties of these compositions in a matter of minutes.
In Chapter 4, this technique is used to form and analyze solid solutions of Eu2+-doped halides. The broad-band Eu2+-emission is sensitively susceptible to its local environment, unlike the infrared line-emitter Tm2+. Researching such solid solutions is of great importance for Tm2+-doped halide LSCs. A solid solution can combine the luminescent properties of its constituents, potentially yielding uniform absorption of the entire visible spectrum, which would make an LSC-window only dimming, without coloring the incident light. Unfortunately, while these halide materials solve the problem of coloring, they are very sensitive to water and are not used in large scale industrial production.
This is why the focus is shifted in Chapter 5 to materials composed of silicon (Si), aluminum (Al), oxygen (O) and nitrogen (N): the SiAlON material family. These SiAlONs are chemically stable, scratch-resistant and, because of their likeness to amorphous glass, do not scatter light. These SiAlONs are sputtered on a large-scale by industrial glass manufacturers.
Next to fabricating all these materials and characterizing their luminescence, it is also important to predict how they would behave if they were applied as large-scale LSCs. This is done through modeling all optical processes that occur within an LSC, presented in Chapter 3 and Chapter 6. In Chapter 3, a new way of modeling the optical processes within an LSC is presented. The industry-standard is ray-tracing, which can get slow when an LSC absorbs more light, or becomes larger in size. The model presented in Chapter 3 calculates all efficiency steps in an LSC in the same amount of time, regardless of the LSC’s size or transparency. In Chapter 6, we use all these methodologies—fast synthesis and characterization of luminescent thin-films, and modeling of light transport through an LSC—to simulate how efficient an LSC based on AlN:Eu3+,O2– would be. Such an LSC would be transparent in the visible spectrum, as it only absorbs ultraviolet light. AlN:Eu3+,O2– emits red luminescence. Therefore, AlN:Eu3+,O2– will not parasitically absorb the emission that makes its way to the LSC’s edges. The methodology to predict the performance of an LSC used in Chapter 6 is not specific to AlN:Eu3+,O2–, but applicable to all combinatorially synthesized luminescent thin-films.
As mentioned before, halide-type materials doped with Tm2+ have been often suggested as promising materials for LSCs. In the final chapter, Chapter 7, sputtered thin-films of NaI, CaBr2, and CaI2 doped with Tm2+ have therefore been evaluated on their performance as LSC; both in terms of simulated optical efficiency, as well as in terms of aesthetic appeal. Our Tm2+-based thin-film LSCs absorb the entire visible spectrum and emit a line of near-infrared radiation centered at 1140 nm. Chapter 7 demonstrates the universality of the techniques presented in Chapters 5 and 6. These techniques are adapted to take hygroscopic nature of the halides into account. The chapter does forgo on fully taking industrial compatibility into account: halides are not often sputtered, and the water-sensitivity will be a hurdle for implementation on window glass. By combining theory and modeling, we see that 10 μm thick films which transmit 80 % of the visible spectrum would be able to achieve optical efficiencies of 0.71 %. This efficiency already compares favorably to the maximally achievable optical efficiency of 3.5 % at those transmission constraints. Further research will have to show whether the photoluminescent quantum yield of the sputtered thin-films can be increased to achieve unity photon conversion.