A Little Metal Goes a Long Way to Reducing Plastic Solar Cell Costs

Polymer solar cells, devices that convert sunlight into electricity using only plastic materials, require a transparent conducting electrode that can both let light through and transfer charge out of the device. Most conventional devices use indium tin oxide (ITO), which has the benefits of reasonable conductivity and high transparency. However, polymer solar cells are often touted for being cheap and mechanically flexible, but ITO is expensive and brittle. One alternative is to use a highly-conducting polymer called PEDOT:PSS, but that also faces its share of problems. Its conductivity drops off dramatically as device sizes go up, significantly reducing solar cell efficiency.

Writing in Advanced Energy Materials, Galagan et al. demonstrate a metal grid approach that replaces expensive ITO with much cheaper inkjet-printed silver or lithographic molybdenum-aluminum-molybdenum lines. The PEDOT:PSS is deposited on top of the grid to form a transparent conducting electrode. The advantage of the grid is to allow light through yet provide charge collection pathways near the PEDOT even at large substrate sizes. While such metal grids have been reported by other groups, here the authors go further in exploring the parameter space and show that they can model the devices.

The group uses a range of pitch sizes to determine the optimal spacing, in this case about 2.5 mm between metal lines. There is a significant shadow effect, where there is dead, nonconductive space under the metal, and this increases dramatically at smaller pitch sizes. Additionally, the films reported tend to degrade over time. However, replacing ITO entirely with a metal grid resulted in efficiency of ~2.4%, higher than expected for a non-ITO electrode, and the authors think that process optimization would boost that efficiency level even higher.

Galagan, et al. Advanced Energy Materials. DOI: 10.1002/aenm.201100552

Nanocars Go Electric

Since Richard Feynman’s famous “There’s Plenty of Room at the Bottom” lecture, scientists have tried to implement his idea of a device the size of a single molecule. The ultimate goal is a machine at the nanoscale– billionths of a meter– that can be controlled. In an impressive paper reported today in Nature, Ben Feringa’s group in the Netherlands demonstrate a vehicle that is a single molecule, in this case a nanocar, that they can move using only electrical pulses.

The electrical control is the critical result because nanocars themselves are not new. Previous research on nanocars has primarily focused on non-electrical driving. For example, some groups heat up the substrate to provide their nanocars with the thermal energy to move, though this is not a highly controllable process as it drives all nanocars on the hot surface rather than a specific car . Other groups have used a metal tip to pull a nanocar, but this is more like towing than actually driving. With the electrical control, the authors now can start the car when they want and see it drive across the road, in this case a clean piece of copper.

The nanocar is a carefully-designed molecule with a rigid carbon-carbon triple bond for a central chassis, carbon ring axles, and carbon ring wheels. In actuality, the wheels are more like paddles. The paddles propel the nanocar only in a certain direction. The chemistry of these paddles is key. If the authors replace two of the paddles with their chemical mirror image, the cars will simply move around randomly.

Driving the car is a two-step process involving electrical and vibrational energy. First, the electrical pulse gives the paddle energy to initially rotate but not fully. The second process uses vibrational excitation, where part of the molecule essentially changes its conformation in order to complete the paddle’s rotation. The energy for the vibration requires a lower voltage pulse than the electrical signal. The whole system is kept at a frosty 7 K (or -447 °F). Once they apply an electrical pulse to kickstart the paddle, the vibrational energy completes the turn.

To demonstrate this motion, the scientists take images using scanning tunneling microscopy, or STM. In STM, a sharp metal tip is brought close to the surface, maybe at most a nanometer away. If a voltage is applied between the metal tip and the sample underneath, a current flows between them. The larger the voltage, the greater the current. The metal tip is scanned back and forth while electrons are flowing into the surface, and the electron flow can be used to create an image.

The electrical pulse used to drive the nanocars is also applied by the STM tip. The authors take an image at a low voltage to make sure they find a nanocar, then move the tip over the molecule and apply a higher voltage to start the car’s engine. The nanocar then moves, albeit glacially– it takes ten electrical pulses to move 6 nanometers.

In the end, the scientists proved that they can move their carefully-designed nanocars at desired times using only electrical pulses. In the future, molecular machines designed with similar principles in mind could be used to carry cargo, for example other molecules.

Kudernac, et al. Nature 479, 208 (2011). DOI: 10.1038/nature10587

A Golden Method to Improve Solar Cell Efficiency

One of the primary issues plaguing organic solar cells is inefficient light absorption. Polymer solar cells have to be thin to accommodate the relatively low charge mobility — if the films are too thick, the photogenerated charges will simply recombine before reaching the electrodes, limiting the amount of power that can be produced. On the other hand, if the films are too thin much of the incident sunlight will not get absorbed and will be reflected away.

In Applied Physics Letters, Xie and co-workers report an increase in the power conversion efficiency (PCE) of their organic solar cell devices by incorporating gold nanoparticles. The gold nanoparticles exhibit a local surface plasmon resonance that significantly enhances absorption. As a result, more sunlight is absorbed and more charge is generated, even from a thin active layer.

Additionally, Xie et al. went one step further and incorporated the gold nanoparticles in the layer between the photoactive material and the transparent conducting oxide electrode. While this layer does not absorb light, the gold nanoparticles decrease the series resistance by almost 30%. Low series resistance at the electrode is important in preventing charges from recombining instead of being extracted.

As a result of the combined effects of light absorption and low series resistance, the gold nanoparticles increase the PCE by up to 22% in their devices. Further experiments, perhaps using much cheaper silver nanoparticles or using nanoparticles in some of more cutting edge materials with higher efficiencies, may finally push plastic solar cell efficiencies into the industrially viable range.

F-X. Xie, et al. Applied Physics Letters 99, 153304 (2011). DOI: 10.1063/1.3650707

Flexible, High-Efficiency Solar Cells Reaching New Levels

Recently, the solar energy company Solyndra declared bankruptcy. That announcement was quickly followed by reports that the FBI had raided the Fremont, California company’s headquarters to investigate why hundreds of millions of dollars in government loans were unable to keep the company afloat, presumably suspecting fraud. And because Solyndra was heavily promoted by the White House, its bankruptcy was the subject of much debate about the government’s role in industry.

What was lost in the news shuffle was the technology behind Solyndra. Conventional solar cells, like the blue-hued ones seen everywhere from parking meters to rooftops, are made of silicon. Silicon, however, is incredibly expensive to process, a fact that has kept solar energy prices upwards of 5 to 10 times that of fossil fuels. Solyndra had a different approach. The name sounds like “cylinder” because that was their innovation: cylindrical solar panels. They used a material known as copper-indium-gallium-selenide, or CIGS, deposited on materials they could roll into tubes to make them.

CIGS is comparatively simple to produce using equipment costing only in the tens of thousands of dollars — as opposed to the literal billions spent on equipment to process silicon. It falls in the realm of “thin film” technologies, where research and development focuses on efficiency below that of silicon but at a price point that more than compensates. For example, one of the touted advantages of CIGS-based solar cells is that CIGS can be deposited on plastic films to make roll-able solar cells, which has significant advantages from a production standpoint. (Imagine a newspaper machine churning out solar panels.) While that is the ultimate goal, thus far CIGS devices produced on flexible substrates have been comparatively poor at converting the sun’s photons into electrons. In fact, that was one of the issues that did in Solyndra; their cylindrical solar panels were just not efficient enough to justify the cost.

In a new paper published in Nature Materials, Chirilă et al. report a new method for producing CIGS on a plastic substrate with record-setting efficiency for a flexible film device of ~18%. Put another way, about 1 out of every 5 photons — the individual particles of light — that reach the solar cell become electrons that are useable as a source of electricity. The problem they needed to address has to do with the temperatures used before the plastic substrate breaks down. In CIGS films deposited on metal substrates or metal-on-glass, the temperatures used to deposit the constituent materials are often in excess of 600 °C (1112 °F), far beyond the tolerance of plastics.

Why not just lower the temperature? It turns out gallium (the “G” in CIGS) does not like to combine with the other materials and forms a gallium-rich barrier at the top and bottom of the device rather than smoothly blending throughout the film. These gallium-rich regions form an energy barrier that blocks electrons from getting out of the device, which in turn limits the amount of electricity and therefore the efficiency at turning photons into electrons. The traditional way to counter this gallium barrier problem is by depositing at higher temperatures, hence the problem when it comes to flexible substrates.

The significant result from the Nature Materials paper was showing that adding the gallium in a graded amount with time can counter that barrier effect, even at lower deposition temperatures. At high temperatures, most groups use a three-stage approach where the materials are introduced at specific intervals. Instead, Chirilă and co-workers introduce the gallium throughout the process, decreasing the gallium evaporation with time to account for the barrier effect.

This work is not for a production-level system, however. Their tests were done on individually fabricated devices and not ones made on a full industrial line. Additionally, their efficiency tests did not take into account the geometry, so it’s unknown if this would have improved upon the Solyndra design considerably.

It’s unlikely that Solyndra-like technology will suddenly start beating silicon in the race to provide a green alternative to fossil fuels. In fact, Solyndra was done in by two factors that this work may not even address — the fall in polysilicon prices by nearly a factor of 10 and the heavily subsidized Chinese solar industry. On the other hand, it may simply have been that Solyndra was too far ahead of the curve. According to reports, their devices were less than 10% efficient, and if the devices were produced using the graded approach reported here, perhaps they may have been in a more competitive position.

Chirilă, et al. Nature Materials,  DOI: 10.1038/nmat3122