The goal of luminescent solar concentrators (LSC) is to simultaneously decrease costs and increase efficiency. Traditionally, LSCs are designed to absorb unusable light and re-emit it at wavelengths with the highest efficiency for the underlying PV cell. By increasing the concentration of highest – efficiency light, material costs decrease because fewer PV cells are required for the same power generation. Unlike the aforementioned concentrator techniques, there is no required tracker creating odd-shaped array footprints in residential areas.
LSCs are constructed of a polymeric lens that directs the light into the adjacent PV cells. The flat plate geometry is the most widely discussed, but cylindrical concepts have been proposed and patented since the late 1970s [21-24]. Both geometries are constructed from a polymer, typically PMMA. The light enters the largest face, bounces internally, and exits the edges. The concentration factor (Ggeom) is the geometric ratio of the area of the face (Aface) to the area of the edge (Aedge) (Equation 6.7, Figure 6.7).
 |
 |
 |
|
A
Ggeom = f (6.7)
edge
For a rectangular geometry, 3 mm thick and 1 m long, the concentration factor would be 333x. This theoretical value is not realized in practice. Currently, the highest reported value is a 4% efficiency improvement [17]. In order to become competitive, LSCs must demonstrate a 6% to 10% increase in efficiencies above those of the solar cell. These poor practical demonstrations are due to a number of chemical and optical limitations.
The polymeric LSC relies on a fluorescent dye to convert the incident light into usable wavelengths. These dye molecules must absorb low-wavelength, high-energy light and emit high-wavelength, low-energy light. Each molecule purchased from a chemical manufacturer has a characteristic absorption and emission curve, and both are commonly represented as a Bell curve of intensity versus emitted or absorbed wavelengths. The difference between absorption and emission maxima is defined as the Stoke’s shift.
This is the same phenomena discussed for UV quenchers in Chapter 2. Therefore, some polymer formulations will have this inherent property built into the commercial formulation. Polymers with these UV additives will fluoresce green. However, the additive concentration is typically not high enough to exhibit LSC behavior. Therefore, PV manufacturers will have to request a specific fluorescent additive, with the desired characteristics, be added to the polymer. When choosing a fluorescent molecule, the absorption curve of the encapsulant’s formulation (e. g., UV stabilizers) should not overlap the fluorescent molecule’s absorption or emission curves in order to optimize conversion efficiency.
Similar to photovoltaic cells, fluorescent molecules are characterized by a quantum efficiency curve. The quantum efficiency of the dye is how much
of the absorbed light is converted to emitted light. A quantum efficiency of 100% is ideal, but commonly unrealized. Most commercial organic dyes have an efficiency of 75% to 80%.
At high loadings, adjacent dye molecules molecularly couple, decreasing conversion efficiency. This means they absorb the light emitted from surrounding dye molecules rather than the impinging light, thereby decreasing the efficiency of the conversion of incident light [25]. As an example, Kurian and coworkers demonstrated a 24% decrease in quantum yield (75% to 51%) for a 6x increase in fluorescent rhodamine 6G dye concentration in PMMA (1.5 • 10-4 to 9 • 10-4 mol/l) [26]. Therefore, it is essential to create a homogenous dye distribution at the lowest concentration possible to avoid this conversion inefficiency.
The response of the PV cell and dye pair must be optimized. The emission curve must overlap with the internal quantum efficiency (IQE) of the PV cell, but the absorption curve must not rob useful wavelengths from the cell. Coumarin-6 overlapped with a single crystalline silicon cell IQE response is a commonly proposed pair (Figure 6.8). The absorption peak maximum occurs at 450 nm where the cell is 68% efficient and emitted at 500 nm where the cell is 71% efficient. Ideally, the absorption maximum would be between 350 and 400 nm where the cell is less than 50% efficient and emitted between 550 and 600 nm where the cell is 80% efficient. Unfortunately, this means the
|
Wavelength (nm) FIGURE 6.8 Single crystalline silicon solar cell internal quantum efficiency (IQE) curve overlapped with an absorption and emission curve for coumarin-6. |
desired absorption curve is further than a typical Stoke’s shift away from the desired area of emission.
In response, researchers have proposed a double-down conversion approach, also known as a two-dye system. In this case, two fluorescent dye molecules are incorporated into the LSC. The absorption curve of one dye has an emission curve that overlaps with the absorption of another. The emission of the second dye molecule overlaps with the IQE of the PV cell. Of course, the limitation is that both dyes must have high quantum efficiencies and the emitted light must perfectly couple inside the LSC to be fully effective.
Fluorescent chemistry has been extensively used in the textile and toy industries for decades; however, these organic molecules have short lifetimes. Some glow-in-the-dark shirts and toys contain fluorescein, an organic molecule that has an absorption peak maximum of 494 nm and emission peak maximum at 521 nm. Most of these commercially available formulations have low stability because they were formulated for commercial goods with short lifetimes. The fluorescent molecules photoxidize under prolonged exposure to UV light and oxygen. Photoxidation results in a loss of fluorescent properties and a color shift, also known as lightfastness or photobleach- ing. This remains a poorly understood and predicted mechanism. The dye industry typically does not guarantee color stability for more than 2 to 10 years of outdoor exposure.
When LSCs were first proposed in the late 1970s, Batchelder and coworkers performed stability measurements on laser dyes, rhodamine-6G tetrafluo – roborate and coumarin-6. The dyes were impregnated into PMMA plaques at relatively low concentration (10-4 moles/liter) and exposed to light and dark cycles at 60°C and 50°C, respectively. The UV source was a fluorescent bulb, and the chamber was kept at 100% Relative Humidity (RH). Photobleaching was monitored by absorption measurements at various exposure intervals. The behavior was characterized by a rapid decrease followed by a plateau, modeled as a decaying exponential. These dyes were projected to last for 2 to 6.8 equivalent years. Admittedly, Batchelder did not identify the chemical mechanism for degradation, or its dependence on moisture or oxygen ingress [23]. The dye’s short lifetime indicates the device efficiency would exponentially degrade during the 25- to 30-year power warranty offered by PV manufacturers. It is important to note these academic studies are not performed on fully packaged modules. Therefore, a PV manufacturer needs to perform his or her own testing to verify these aging characteristics.
In order for LSCs to maximize efficiency, the converted light must be coupled into the adjacent PV cell. To achieve coupling, there must be a number of internal reflections before the light escapes from the LSC into the PV cell. Based on Snell’s law, the refractive index multiplied by the angle of incidence is equal to the refractive index of the second medium and the angle of transmittance through that medium (Equation 6.8):
n1 sin 01 = n2 sin 02 (6.8)
There will be internal reflections when the emitted light is less than the critical angle. To find the critical angle (0c), the exiting angle of light must be 90o to the polymeric interface (02 = 90°). This makes the critical angle equivalent to the inverse sine of the ratio of the refractive indices of the two media (Equations 6.9 and 6.10):
 |
 |
 |
 |
n1sin 0c = n2 sin90o
 |
 |
 |
If the surrounding medium is air, then the equation for internal reflection is solely dependent on the polymer matrix (nx) (Equation 6.11):
The difficulty is a fluorescent dye molecule in the center of the concentrator will emit light in 360°. Therefore, the majority of light will be lost as it exceeds the critical angle of escape. A number of researchers have modeled this loss in flat PMMA plates to find approximately 26% of the emitted light from the point source is lost due to escape from the LSC [23,27].
In order to increase the internal reflections, researchers proposed a series of additional processing layers laminated to the surface of the LSC. These additional layers would contain liquid crystals that would transmit all wavelengths but selectively reflect those emitted wavelengths of highest efficiency for the PV cell. When included in the LSC structure, they are referred to as photo-band stop filters. In addition to this approach, a number of researchers have investigated processing methods to allow the dye molecules to preferentially align, increasing the concentration of emitted light trapped at the critical angle. Unfortunately, each of these new materials and processing steps increase production costs, making LSCs a less competitive alternative to the aforementioned CPV techniques.
