Thursday, May 05, 2011

Nano for solar

Sorry about the delay in this posting.  Real life has been busy.

Solar energy is an obvious candidate for a long-term solution to many of our energy problems.  The amount of power reaching the surface of the earth is on the order of 350 W/m2.  We could meet the world's projected energy needs in 2030 by covering around 250 km by 250 km with 10% efficient solar cells.  Unfortunately, the total surface area of all photovoltaics ever manufactured is less than 0.1% of that.  (This is why being able to produce photovoltaic cells by printing processes would be great.  Hint:  estimate the total area printed by the New York Times in a month.)  There are a number of challenges involved in solar.  Why might "nano" broadly defined be a big help?  Let me give three examples from the large wealth of ideas out there.

1) Semiconductor nanocrystals as absorbers.  Because of the beauty of quantum confinement, it is possible to make semiconductor nanocrystals out of a single material, and use different sizes to capture different parts of the solar spectrum.  Moreover, there is evidence (after some controversy) that nanocrystals may enhance "multiexciton generation" (e.g., here and here).  In a traditional solar cell, a photon with energy twice as large as the semiconductor band gap will generate an electron-hole pair (which must be ripped apart somehow), and inelastic processes will lead to the excess (above the band gap) energy being lost as heat.  However, at some rate, instead you can generate two band-gap-energy pairs.  The idea is that the rate of that process can be enhanced in nanocrystals, since conservation of "crystal momentum" can be relaxed in materials that are so surface-dominated.

2) Nanostructured materials for photoelectrochemical cells.  There are a number of proposals for using electrolytes in solar applications, including dye-sensitized solar cells.  In this case, one would like to use a high surface area anode, such as nanostructured TiO2 or some similar nanostructured material.  Moreover, instead of using organic dyes as the absorbers and sources of photoexcited electrons, one could imagine again using semiconductor nanocrystals.

3) Plasmon-enhanced photovoltaics.  One way to try to boost the efficiency of solar cells is to get the light to hang around the absorber material for longer.  One compact way to do so is to use plasmonically active metal nanoparticles or nanostructures as optical antennas.  The local fields near these structures can enhance scattering and local intensity in ways that tend to boost performance, though resistive losses in the metal may limit their effectiveness.  It's worth pointing out that one can also use plasmonic antennas as sources of hot electrons, also interesting from the photovoltaic angle.

There are many more ideas out there - I haven't even mentioned anything about nanotubes or graphene.  While the odds of any individual idea being a truly transformative breakthrough are small, there are probably more clever things being proposed in this area now that at any time ever before, thanks to our ability to manipulate matter on very small scales.   

6 comments:

Joel Kelly said...

I struggle to understand the complexities of MEG as a chemist (beyond the general concept of a "phonon bottleneck").

However, I have never really understood how MEG would work in a practical device. In order to see evidence for MEG, you need strong confinement, which usually means fairly small nanocrystals. The MEG onset is usually around 2-3Eg, which for most semiconductors with an increased bandgap due to size effects is going to miss the majority of the solar spectrum, so the benefit is going to be reduced. I'm most familiar with Si-NCs where MEG would only help for photons in the blue and near UV.

Secondly, how the heck do you get all those charges out? Wouldn't coupling the NC into some sort of junction seriously muck up the electronic structure that is required to produce the MEG effect?

Anonymous said...

Congratulations, your Nano-goldbar-solar cells are even in the European press, see here: Mark Knights gold solar cells in the European press

Web Design Tampa said...

For rays in the ultraviolet range of the spectrum, the increased efficiency improves the power-performance of sun-to-electricity conversion by 60 percent.

Anonymous said...

I agree with Joel- it's an obvious paradox that a highly confined structure is... highly confined. Solar cells have to allow current flow in order to work.
The plasmonic gold bars in Science are interesting for a very inefficient IR detector in Si. It seems like a wonderfully creative idea and a neat paper. The low quantum efficiency (EQE is estimated at 0.01%) might be useful for some application that absolutely has to be implemented on Si.
For solar, I feel that the team faces an uphill battle. The authors project that the EQE could be raised to 2% over some range of wavelengths. Thus, the additional boost to short circuit from IR photons would be extremely small; a good Si solar cell can turn 50-90% of photons from near-UV to slightly above the bandgap into current. The question is, how much loss do the nanoscale metal objects induce in the rest of the spectrum (visible, etc.) where Si ordinarily performs so well?

Joel Kelly said...

Hi Dr. Natelson, I'd love to hear your comments on this recent perspective on carrier multiplication in nanocrystals from the Bawendi group: http://pubs.acs.org/doi/full/10.1021/nl200798x

fashion accessories said...

I think it is possible to make semiconductor nanocrystals out of a single material, and use different sizes to capture different parts of the solar spectrum.