In an age of global warming and declining fossil fuel supplies, one of the technological hopes for the future security of our energy supplies lies with solar power. Nanophotonics – the application of photonics technologies at the nanometre level – is one route to reducing the costs of manufacturing conventional silicon-based photovoltaic solar cells.
However, before these techniques find their way into manufacturing plants, there is a lot of work to be done. The underlying problem with photovoltaic technology at the present moment is that silicon – the most widely used material – has a number of limitations in terms of its light-absorption characteristics. This is where nanophotonics can come to the rescue.
While silicon itself is abundant and cheap, the high-purity material used in photovoltaic cells is expensive to produce. So one aspect of the drive towards cheaper photovoltaic cells is a reliance on reducing material costs by reducing the thickness of the cells. However, this falls foul of the fact that silicon is an indirect band-gap semiconductor, meaning that it is a relatively poor absorber of photons. Solar cells made of silicon must be at least 200μm thick (often 300μm or more) so that the path length of the light within the material is sufficiently long for a useful amount of the energy to be absorbed. If the cell is any thinner, the longer, near-infrared wavelengths of the spectrum are not absorbed efficiently, thus negating the benefit of decreased manufacturing costs.
Several research groups around the world are using nanophotonics to improve the efficiency of thin solar cells, including a group led by Dr Kylie Catchpole at the Australian National University, in Canberra. Dr Catchpole’s technique involves adding metal nanoparticles to the surface of a silicon photovoltaic cell by a self-assembly process typical to nanoengineering. A thin layer of metal is deposited onto the silicon, and then heated in such a way as to allow it to form droplets approximately 100nm in diameter at regular intervals. Incident sunlight will excite resonances in these particles, generating a quantum mechanical effect known as ‘surface plasmons’, which cause the light to be scattered into the silicon. The plasmons and associated scattering can be tuned by altering the size and composition of the nanoparticles. They are also dependent on the wavelength of the incident light.
Pol van Dorpe, a post-doctoral researcher into surface plasmonics at IMEC research institute in Belgium, explains the practical upshot of this work: ‘This scattering becomes useful when producing a thin-film solar cell stack, for example,’ he says. ‘It could be thin-film silicon in the order of tens of microns – very thin compared to the thickness of regular silicon solar cells. If it’s thinner, then it’s cheaper,’ he explains, ‘as the silicon is a major cost factor in these cells.’
The issue of light trapping is sufficiently important even for standard (thick) silicon solar cells that, in manufacture, their surfaces are textured with features at the micrometre scale. Enhancements to single-crystal silicon solar cells, for example, can take the form of pyramidal prisms, 10μm wide and 10μm deep. Other surface texturing techniques exist for polycrystalline silicon. The features serve to refract incident light so that it takes a longer path through the silicon, thereby maximising the energy that can be captured. But such geometrical optics (where light is treated as rays) cannot be incorporated into the new generation of very thin solar cells – hence the recourse to nanophotonic methods such as the light-plasmon interactions described above.
Silicon isn’t the only option however. A range of thin and therefore cheap solar cells are under development, based on organic dye molecules printed onto polymer substrates. Van Dorpe states that the nanoparticles that scatter light so effectively in the silicon case can serve a similar role in these organic cells. ‘The particles can be incorporated into the inside of the solar cell,’ he explains. The nanoparticles again serve to concentrate light in an enhanced near field, and ‘when there is a more intense field, there is also more absorption,’ states van Dorpe. The efficiency of the organic solar cell can therefore be increased by confining the light to a smaller area using the near field effects of nanoparticles.
Nanometre-sized structures are commonplace in electronics and computing, where a single transistor may be only a few tens of nanometres across, but photonics at a nanoscale has thus far proved to be a challenge. Visible light has a wavelength in the range of 380-750nm, and when dealing with components that interact with light, this represents a minimum length scale for engineers.
A plasmon is a quantum of plasma oscillation in a material, and they can be treated as if they are particles, or quasiparticles. In surface plasmonics, plasmons can be confined to the surface of a material, where they move at the boundary between a metal and the air or a vacuum, for example. These surface waves can be excited by light; hence their importance to photonics. Furthermore, the waves themselves are sensitive to the properties of the surface across which they propagate, and these properties can be tuned by adding nanoparticles to the surface.
Seeing smaller
Photovoltaic applications use metal nanostructures to confine light at a nanometre scale. This confinement can be achieved in such a way that the interaction volume of the light is comparable to the size of a molecule. Thus nanophotonics makes possible the creation of detectors with single-molecule resolution for applications in detection and spectroscopy. One such example is that of surface-enhanced Raman spectroscopy, or SERS.
Raman spectroscopy uses infrared light, usually produced by a laser, to excite electrons in a sample. Detectors then look for scattering that corresponds to characteristic transitions between different electron states within the sample. The technique is able to provide useful information about the structure of a molecule, and it is particularly good at resolving low frequency oscillations such as those generated by the chemical bonds that make up the backbones of larger molecules. Conventionally, the technique is capable of resolutions approaching 500nm, but the use of SERS techniques can greatly improve upon this.
SERS involves applying metal nanoparticles to the surface of a sample, to produce a similar effect to that seen in the enhanced solar cells. Again, the intense electromagnetic fields generated by the plasmons can enhance an otherwise weak effect such as Raman scattering, leading to greatly increased sensitivity. The technique is capable of detecting single molecules. Raman spectroscopy finds applications in analysis of the purity of anaesthetic gases in medicine; pollution measurements; purity in pharmaceuticals and so on. (For further information on Raman spectroscopy, see the feature in Electro Optics April/May 2009).
Cutting out the middleman
While practical uses of nanophotonics in the immediate future are focused on enhancements of existing applications such as power generation and spectroscopy, developments on the horizon promise novel technologies. Plasmonic waveguides are analogous to those designed for photons in free space but, whereas optical waveguides have dimensions of 400 to 500nm (commensurate with the wavelengths used), plasmonic waveguides can be many times smaller, at 50 to 200nm. This opens the way to analysis of light not in a large – and comparatively expensive – spectrometer, but on a chip. Not only is there the promise of a cheaper technique, but also of broadening the range of applications that have hitherto been unaffordable.
In the applications above, plasmons are created when photons or light are absorbed by a metal nanoparticle on the surface of a material. It is possible, however, to produce the same effect in a device through direct excitation of a coherent plasmon. Van Dorpe, whose group has been working on such techniques, calls this ‘the missing link’ of photonics. The technique for producing these coherent plasmons relies upon silica nanoparticles doped with a dye that responds to a short wavelength laser. The particles absorb the short wavelength before re-emitting at the resonant wavelength of the surface plasmon. ‘This is essentially a stimulated emission of plasmons,’ states van Dorpe, meaning that the device acts as a ‘plasmon laser’.
Nanophotonics and the exotic delights of plasmonics are relatively new, but already commercially important applications can be envisaged in spectroscopy, detection, and photovoltaic solar energy. With ‘plasmonic lasers’ now on the agenda, who knows where this technology will lead us?
