Although first-generation solar cells using silicon or gallium arsenide provide efficiencies up to around 30%, improvements to production costs appear to be reaching a ceiling and limit their potential to compete with fossil fuel prices. In response, many researchers are looking into thin film photovoltaics that provide lower efficiencies – on the order of 10% – but with much lower manufacturing costs. A new paper in Nature Photonics by researchers at École Polytechnique Fédérale de Lausanne in Switzerland have found another solution – the use of a novel photoabsorber made of a lead halide perovskite (CH3NH3PbI3) that demonstrates ultrafast carrier extraction and efficiency up to 16%. This makes the architecture immediately competitive with thin film solar cells!
Perovksite refers to a particular type of crystal structure that a whole host of compounds can take – they have found their uses in fuel cells, ferroelectrics, and other applications due to their low costs. However, perovskite-based solar cells are rather new, and this paper reports one of the most efficient to date.
This paper by Marchioro et al is especially important because it describes the physical mechanisms behind the lead halide solar cell’s high efficiency. The diagram above shows the basic setup for most first- or second-generation solar cells. A photoabsorber (here it is the lead halide perovskite in brown) is sandwiched between two other materials – the titanium oxide (TiO2) on the left, and the Spiro on the right. You can imagine the lower brown box of the perovskite as the place where all the electrons reside when no light is shining – this is known as the valence band. When the perovskite absorbs light, it actually gives energy to an electron equal to the light energy, enough to knock it up to the upper brown box, known as the conduction band. Once electrons are in the conduction band, they move around quickly and can be extracted for useful work in a circuit.
But here’s where the other materials are necessary. Remember that electrons are negatively charged. Once the electron is excited, it leaves behind a space where it used to be, what is called a hole, which can be treated as a positively charged particle (the absence of a negative charge moving around a bunch of other electrons). From basic electrostatics, we know that positive and negative charges like each other. So, if this electron in the conduction band happens to come close to the hole it left behind, the two will rush towards each other and recombine. When this happens, we lose the excited electron and the potential to do extract energy from the device. So, in a solar cell, we want to prevent recombination as much as possible! Also, notice that the Spiro, another organic material that is on the other side of the perovskite, has a band close to where holes are in the valence band of the pervoskite. Electrons can then easily move to TiO2, and the holes left behind can move to the Spiro. This is the type of clever architecture that scientists develop to entice electrons and holes away from each other in opposite directions.
And that’s exactly what the authors found they could do so well with this solar cell architecture. Titanium dioxide is happy to take excited electrons from the perovskite – seen in the diagram above by the fact that TiO2’s conduction band (blue) overlaps the pervoskite’s conduction band. And Spiro’s band (gold) is in just the right region to accept holes from the perovskite. The authors found that electrons and holes are extracted to these other materials on a subpicosecond timescale – a picosecond is 10^-12 seconds, which is very fast – whereas recombination appears to take place on the microsecond timescale (10^-6 seconds). Any solar cell must operate under conditions where extraction occurs faster than recombination, but this solar cell does so extremely well, as seen by the orders of magnitude difference of charge transfer vs. recombination.
The figure above also shows that TiO2 works better than other tested materials, such as alumina (Al2O3) or glass. The graph is a measure of photoconductance over time. Photoconductance, on the y-axis, will be higher when more electrons are being excited and extracted out of the perovskite, and will become lower either due to recombination or lack of light. Clearly, using the TiO2 (black line) provides higher photoconductance over a longer time period compared to the other materials, demonstrating that the TiO2 provides faster electron extraction (the holes are found to extract to the Spiro equally fast). So these charges (either electrons or holes) are extracted to the TiO2 and Spiro, respectively, before they can recombine, leading to a well-performing photovoltaic device.
This paper is the first to show that ultrafast carrier extraction is the primary reason for such high efficiencies using this pervoskite-based solar cell. This is a great indication that perovskites are an important material to continue to pursue. Future studies will hopefully see if this ultrafast extraction is a general property of perovskite-TiO2 interfaces, or what specific characteristics are necessary for such high performance.
Arianna Marchioro, Joël Teuscher, Dennis Friedrich, Marinus Kunst, Roel van de Krol, Thomas Moehl, Michael Grätzel & Jacques-E. Moser (2014). Unravelling the mechanism of photoinduced charge transfer processes in lead iodide perovskite solar cells Nature Photonics DOI: 10.1038/nphoton.2013.374
ga(‘create’, ‘UA-50429946-1’, ‘gnightearth.com’);