24 - 28 October 2016 • Marina Bay Sands Sands Expo and Convention Centre, Singapore
The ultimate limit for solar power conversion stands at 87% so there would appear to be plenty of scope for improving the efficiency of photovoltaic technology. The fastest route to achieving high power conversion efficiency is by stacking multiple photovoltaic junctions. The InGaP/GaAs/Ge multi-junction solar cell represents the industry standard for space and solar concentrator applications. However, the evolution of the technology to a 4J architectures is presently at a crossroads. Options exist to fabricate a lattice-matched 4J cell using dilute nitride semiconductors or strain-balanced quantum wells, or alternatively lattice mismatched and wafer bonding approaches have also proven to be effective; the latter holding the present world record of 46.5%. All these technologies are likely to achieve efficiencies in excess of 50% in the near term. A more difficult question is the extent to which the cost of the multi-junction solar cell can be reduced. While options exist for high throughput manufacturing, it is here that alternative approaches to high efficiency might, ultimately hold an advantage. A perspective on the present status of intermediate band and hot carrier cell concepts will be given.
Absolute electroluminescence (EL) measurements provide a way to evaluate internal properties of individual subcells in multi-junction (MJ) solar cells. To establish methodology of absolute EL measurements for solar-cell characterizations, we studied accuracy of the measurement by comparing independently calibrated absolute EL methods and by evaluating uncertainty of individual methods. The results showed good quantitative agreements within 10% discrepancy among the individually calibrated methods. The absolute EL measurement methods were then applied not only to evaluation of subcells internal voltages, but also to quantitative studies on subcells radiation damages and resulted internal radiative efficiency representing materials quality in MJ solarcells. They were also applied to diagnosis of multiple-quantum-well solar cells with distributed Bragg reflector mirrors, clarifying the advantages of such advanced structures. After the internal radiative efficiency of realistic subcell materials were evaluated, the detailed-balance theory was used to re-analyze the optimized MJ cell designs and the efficiency limit in the presence of non-radiative loss.
In this presentation, we will first discuss the requirements for making efficient two- and four-terminal devices based on silicon bottom cells. C-Si solar cells with passivated contacts (such as a-Si/c-Si heterojunctions or with tunnel oxides interfaces) are shown to be ideal for such applications. We will comment first on devices based on III-V on Si , showing how the 30%, 1 sun, benchmark should be overcome with suitable devices. In the second part of the presentation, we will focus on the combination of perovskite on silicon in various configurations with potential for low production cost. We will show that perovskite cells with strong transparency in the near infrared can be designed for integration into both tandem and four-terminal devices. Whereas 2-terminal devices reach close to 22% efficiency, 4-terminal measurements of tandems with small-size perovskite top cells demonstrate 25.6% efficiency, measured at maximum point for several minutes [2,3 and next]. We finally discuss the limitations and possible industrial implementations of the various approaches.
Characterizing solar cells that have been made can be at times challenging enough, especially when these are heterogeneous, or have unusual transport properties. Could we go one step further, yet, and consider characterizing the performance of solar cells that are not yet completed? There are a number of instances where this possibility could be useful, for instance when promising materials (e.g. molecules, polymers, complex multinary compounds …) have been made but the device technology is not available. There is another, perhaps rarer, instance when this is also desirable: when it is not a material really that is to be tested, but a non-standard photovoltaic conversion process as in e.g. Hot Carrier, or Photothermoelectric, or Intermediate Band, or MEG Solar Cells.
As with any energy conversion process, access to key thermodynamic properties is crucial, and those are essential to evaluate the electric free energy that can be recovered from light. As it turns out, light emission from solids (luminescence), when used properly and in suitable cases, can help recover these thermodynamic quantities, and on top of that, can also be used to measure collection efficiency, non-radiative recombination rates, diffusion length etc… i.e. a number of essential transport properties related to the system investigated [1,2,3].
We will here show how a methodology can be developed to extract relevant information on the performance of exotic photovoltaic conversion processes, even when a device has not been made and will illustrate this with the example of Hot Carrier Solar Cells, as well as other devices (IBSC, Ultrathin devices, nanocells, …). [4,5,6,7]
While perovskites have already reached single junction power conversion efficiencies over 22 %, their tunable bandgap means that they are ideal candidates for tandem solar cells to push PV efficiencies past 30 % at low costs. This presentation will discuss the great possibilities that perovskites provide in terms of making perovskite – silicon as well as perovskite-perovskite tandem solar cells. Our work at Stanford shows that it is possible to combine a planar heterojunction perovskite solar cell with a heterojunction silicon solar cell to make a 23.6 % monolithic tandem. We have also developed an efficient small (1.2 eV) bandgap perovskite which we can pair with a wider gap perovskite to make current matched monolithic tandems of 17 %, and mechanically stacked tandems exceeding 20 %. The stability of the materials is also investigated and is found to be surprisingly good; the solar cells can readily meet the IEC criteria.
Photon management: from Planck to solar cells and beyond
The manipulation of light provides an attractive option to enhance the output of a solar cell. Optical concentration, antireflection coatings and textured surfaces with light trapping are just a few examples of technologies in routine use today. Recent advances in photonics and nanotechnology offer a range of additional tools that are being researched for application in the next generation of solar cells. Starting with a brief historical overview, we shall consider several concepts which are under discussion to improve the light capture at the nanoscale, including light injection into thin films by photon tunnelling, light harvesting energy transfer via the near-field dipole interaction, limitations to light trapping in subwavelength layers by diffraction, and the role of photon recycling in the operation of a solar cell.
On-going challenges and topics in the solar energy conversion using quantum dot solar cells are reviewed. In quantum dot intermediate band solar cells (QD-IBSCs), control of carrier dynamics with particular emphasis on the two-step inter-subband absorption/recombination processes is paramount to achieving higher photocurrent production and hence higher efficiencies. Significant efforts and a rapid progress have been made in the device physics as well as practical demonstration of high-efficiency IBSC using QDs.
We review the electrical characteristics of the record cells of the 16 widely studied photovoltaic materials geometries (efficiencies 10-29%) and compare these to the fundamental limits based on the Shockley-Queisser detailed-balance model. All geometries suffer from incomplete light management. We show how nanostructured dielectric and metallic metasurface and metamaterial architectures can help to control the coupling, trapping and conversion of light in solar cells. Prospects for practical application and large-area fabrication, for which achieving high efficiency is a key factor, are discussed for all materials. Ref.: A. Polman et al., Science 352, 207 (2016)
III-V materials have been used as the building blocks of the highest efficiency single and multijunction photovoltaics to date. However, large-scale adoption of these materials for solar energy harvesting has been hampered by the high cost of the group III elements, Ga and In, and the need for high-cost substrates such as Ge and GaAs. The II-IV-V2 materials provide an alternative pathway: these materials are analogous to III-Vs, but group III elements are replaced by alternating group II and group IV elements, similar to the relationship between CdTe and CIGS. In this case, inexpensive elements such as Zn, Sn, and Si can be used, but the materials offer a similar phase space to the III-Vs, with opportunities for alloying and lattice-matched multilayer stacks. This talk will focus on two materials: ZnSiP2, a wide band gap material that is lattice-matched to Si for tandem PV applications using a Si bottom cell, and ZnSnN2, a narrow-gap, wurtzite-structured material similar to InGaN, for thin film applications.