Spectral conversion can be applied to a wide range of photovoltaic (PV) technologies for a range of benefits. To date, research efforts have focused on using luminescent species to absorb sunlight at one wavelength and then re-emit it at another wavelength where its energy can be more effectively harnessed. Four embodiments of luminescent-PV devices are detailed below:
1) Luminescent Down-Shifting (LDS)
Many PV modules, especially those produced in a production environment, exhibit a poor spectral response to short-wavelength light. LDS can be applied as a passive layer on top of the PV cell that absorbs the short wavelength light where the PV modules have poor EQE and re-emits it at a longer wavelength, where the EQE of the device is high. Figure 1 shows the external quantum efficiency (EQE) of several different commercially-produced PV module technologies at short wavelengths. The poor response limits the photocurrent that can be generated from this part of the solar spectrum.
Two examples of the work carried out by our group are;
Cadmium Telluride (CdTe) PV Modules:With Cadmium sulphide/cadmium telluride (CdS/CdTe) PV modules, the CdS layer is required to form a heterojunction with the CdTe layer. However, since CdS is a direct bandgap semiconductor with Eg = 2.4 eV it strongly absorbs the majority of short-wavelength (violet to blue-green) light before it can be usefully absorbed in the CdTe material. In a production environment it is difficult to reduce the thickness of the CdS layer (< 0.5 um) due to:
Ray-tracing simulations performed using RAYLENE indicated that by coating the CdTe device with a LDS layer an increase in conversion efficiency from 9.6% (similar to a First Solar module) to 11.2% (relative increase of 17%) could be achieved. Experimental confirmation of these results in currently underway and we are also investigating any possible negative effects of illuminating CdTe devices with spectra that possess little or no UV/blue light.
Multicrystalline Silicon (mc-Si) PV Modules:
While the short-wavelength EQE of mc-Si devices is better than the CdTe device described above, there is still room for improvement. The EQE of mc-Si devices decreases at wavelengths <500 nm due to higher reflection and absorption by the anti-reflection coatings that are optimised for longer wavelengths and due to increased emitter recombination caused by the high doping levels [2]. Its EQE decreases further at wavelength below 400 nm due to absorption by the glass and encapsulating layer (EVA). We are investigating the encapsulation of organic dyes into the EVA layer of mc-Si cells thereby allowing the LDS layer to be incorporated into the production process without any need for addition layers or materials apart from the dye. Initial results indicate that high quantum yield can be maintained in the EVA matrix [4]. Ray trace modelling has indicated that a 0.3% improvement in EQE could be achieved using Lumogen dyes produced by Basf [1] and reasonable agreement has been achieved experimentally [3]. The LDS layer could provide additional benefits such as removing the need for UV stabilisers since incident UV light is absorbed. It would also reduce the number of high energy photons that usually create heat by lattice thermalisation, having detrimental effect on the cell efficiency [4].
back to top2) Luminescent Solar Concentrators (LSC)
LSCs are non-imaging optical devices [4,5] used to concentrate sunlight onto PV cells as a means to reduce the amount of expensive PV material required. LSCs consist of polymer sheets such as polymethylmethacrylate (PMMA) doped with luminescent species such as organic dyes, quantum dots or rare earth complexes. Incident sunlight is absorbed by the sheet and emitted isotropically at a specific wavelength, a percentage of this light is waveguided to the edge of the sheet where PV cells are located. Figure 2 shows a cross-sectional diagram of a LSC and illustrates the operation of the device.
The main motivation for implementing a LSC is to replace the large area of expensive solar cells required in a standard flat-plate PV panel, with a cheap polymeric collector, thereby reducing the cost of the module (£/W) and the solar power (£/kWh). A key advantage of LSC technology compared to other concentrating systems is that it can collect both direct and diffuse solar radiation. This means tracking of the sun is not required – adding further potential cost reductions and making LSCs excellent candidates for building integrated photovoltaics (BIPV) – as well as making them the ideal PV technology for cloudier northern European climates.
An ideal fluorophore for LSCs with silicon PV cells attached would have
Traditionally used organic dyes offer high FQY and good stability (up to 10 years) [7] but have narrow absorption spectra and large overlap of the absorption and emission spectra resulting in re-absorption of emitted light. Re-absorption is a significant loss mechanism in LSCs even with 100% FQY materials as each re-emitted photon can be lost through the escape cone. Quantum Dots (QDs) offer broad absorption and tunability due to their size dependent properties but while FQY values are being improved continuously, they exhibit re-absorption losses and poor photostability [8]. Rare earth complexes consisting of lanthanide (Ln) ions such as Ytterbium or Neodymium with highly absorbing ligands attached, offer intense emission in the NIR region and zero re-absorption losses due to the large stokes shift. A high FQY UV absorbing Europium complex has been developed by our group in collaboration with the University of Edinburgh [9]. FQY values of up to 85% have been achieved in a PMMA matrix, the molecular structure and spectral characteristics are shown in Figure 3. Further NIR emitting complexes have also been developed but the absorption range and FQY values need to be improved to make them suitable for incorporation into LSC devices [10]. Meeting all of the criteria required for a high efficiency LSC (>10%) is a challenge and the optimal configuration may involve a combination of materials. For example, stacking of LSC sheets to convert different parts of the spectrum or mixing of luminescent species into polymer matrices.
back to top
3) Up-conversion
Upconversion (UC) can be used to generate a single high energy photon out of two or more incident low energy photons, allowing for utilization of sub-band gap energy i.e. light >1100 nm in Si solar cells. This provides an extra 35% of solar intensity with the AM1.5G spectrum [11]. UC can be achieved by placing an UC layer on the rear surface of a solar cell thereby avoiding any interference with the cells response to incident light with energy above or within the bandgap. In addition, a reflective layer can be placed behind the UC layer to prevent the transmission of any incident light and to ensure all UC light is collected. The highest EQE achieved to date is 3.4% using a 20% Er 3+ UC layer, Figure 4 (a) shows a schematic of a UC layer on a solar cell and 4 (b) shows the measured EQE of a Si cell with a UC layer [12].
back to top
4) Down-conversion
Down-conversion (DC) of one high energy photon into two lower energy photons has also been investigated as a means to enhance solar cell efficiency. DC provides access to an extra 32% of incident AM1.5G solar spectrum compared to bare Si [11]. DC layers are applied to the front surface of PV cells and also serve to minimise thermalisation losses and subsequent heating of the device by reducing the number of incident high intensity photons. An example of a DC mechanism is one Terbium ion (2.6eV) giving up its energy to two Ytterbium ions (1.2eV) resulting in a peak close to 970 nm. This second order process shows the most promise for use as a DC material as many other DC phosphors only respond to Vacuum UV light (<190nm) which is not present in the solar spectrum.
back to top
__________________________________________
[1] K.R.McIntosh and B.S. Richards Increased mc-Si module efficiency using fluorescent organic dyes:a ray tracing study, Proceeding of IEEE 4th world conference on Photovoltaic energy conversion, Hawaii USA, May 2006, Vol. 2, pp. 2108-2111.
[2] E. Klampaftis, D. Ross, K.R.McIntosh, B.S.Richards, Enhancing the performance of solar cells via luminescent down-shifting of the incident spectrum, Sol. Eng. Mat and Sol. Cells, 93 (2009) pp. 1182-1194
[3] K.R.McIntosh, G.Lau, J.T.Cotsell, K.Hanton, D.L.Batzner, F. Bettiol, B.S.Richards, Increase in external quantum efficiency of encapsulated silicon solar cells from a luminescent downshifting layer, Pro. Photovoltaics , Res. Appl. (2008)
[4] W. H. Weber and J. Lambe, Luminescent greenhouse collector for solar radiation, Applied Optics, Vol. 15, pp. 2299-2300, 1976.
[5] A. Goetzberger and W. Greubel, Solar energy conversion with fluorescent concentrators, Applied Physics, vol. 14, pp. 123-129, 1977.
[6] B. Rowan, L. Wilson, B.S. Richards, Advanced material concepts for luminescent solar concentrators, IEEE journal of selected topics in quantum electronics, Vol. 14, 2008, Issue 5., pp. 1312-1322.
[7] A. A. Earp, G. B. Smith, J. Franklin, and P. Swift, Optimisation of a three-colour luminescent solar concentrator daylighting system, Sol. En. Mat. and Sol. Cells, Vol. 84, (2004) pp. 411-426.
[8] S.J. Gallagher, B.C. Rowan, J.D. Doran, B. Norton, Quantum Dot Solar Concentrator: Device characterisation using spectroscopic techniques, Solar Energy, 81, (2007) 540-547.
[9] O. Moudam, B. Rowan, M. Alamiry, P. Richardson, B.S. Richards, A. Jones, N. Robertson, Europium complexes with high total photoluminescence quantum yields in solution and PMMA, Chemical Communications, 43, (2009) pp. 6649-6651.
[10] Rowan, B., Wilson, L., Richards, B., Jones, A., Moudam, O., Robertson, N., (2009) Visible and near infrared emitting lanthanide complexes for luminescent solar concentrators, Proceedings of 24th European Photovoltaic Solar Energy Conference and Exhibition, 21-25 September 2008, Hamburg, Germany.
[11] B.S.Richards, Enhancing the performance of silicon solar cells via the application of passive luminescent conversion layers, Sol. Eng. Mat and Sol. Cells 90 (2006) pp. 2329-2337.
[12] A. Shalav, B.S.Richards et al , Applied Physics Letters 86 (2004) 103505