Shortfalls in Electron Production Dim Hopes for MEG Solar Cells

See allHide authors and affiliations

Science  19 Dec 2008:
Vol. 322, Issue 5909, pp. 1784
DOI: 10.1126/science.322.5909.1784a

This article has a correction. Please see:

At the Materials Research Society Fall Meeting, several teams reported setbacks in efforts to sharply increase the electrical output of future solar cells using light-absorbing nanoparticles that can generate more than one electron for every photon of light they absorb.


Four years ago, researchers were delighted to discover that light-absorbing nanoparticles could readily generate more than one electron for every photon of light they absorb. Those extra charges, they hoped, would sharply increase the electrical output of future solar cells. But at the meeting, several teams reported setbacks in reaching that goal.

In typical solar cells, when a semiconductor such as silicon absorbs a photon of light with the right amount of energy, it generates an exciton: an electron paired to a positively charged electron vacancy called a hole. The solar cell then separates those opposite charges and collects them at the electrodes. In 2004, researchers led by Victor Klimov of the Los Alamos National Laboratory in New Mexico reported that when lead sulfide (PbS) nanocrystals were hit with high-energy photons from a laser, they could generate up to seven excitons. Other groups jumped in and found a similar multiple exciton generation (MEG) effect in a variety of other nanocrystals, including cadmium selenide (CdSe) and silicon.

Whither MEG?

When semiconductors absorb high-energy photons, they typically create an excited electron (left). In nanocrystals (top), MEG uses leftover energy to excite more electrons (right).


Then doubts began to creep in. Last year, Moungi Bawendi, a chemist at the Massachusetts Institute of Technology in Cambridge, and his Ph.D. student Gautham Nair reported that when they used a different technique, they spotted only a negligible MEG effect in CdSe nanocrystals, a result they later extended to PbS and lead selenide (PbSe). This year, a Dutch group that had previously reported a sharp increase in MEG in indium arsenide nanocrystals reported it couldn't reproduce the result. “The more results that came in, the more controversy there was,” Klimov says.

At the meeting, John McGuire, a postdoctoral assistant from Klimov's group, reported new evidence that MEG in nanocrystals is far weaker than originally thought. In contrast to the 700% initially reported, the new Los Alamos results suggest the increase is likely about 40%, only slightly higher than the 25% increase seen by Bawendi's group. The upshot, both Bawendi and Klimov agree, is bad news. “These numbers at this point are not of practical use for solar energy,” Bawendi says.

So what changed? Klimov says that for their current experiment, the results of which were also published online 12 November in Accounts of Chemical Research, the Los Alamos team stirred the samples to keep the nanoparticles from absorbing more than one photon at a time—a potential source of false-positive results.

Hope for a strong MEG effect isn't entirely lost, Klimov says. In some samples, the MEG effect was more than three times as high as in other samples. Synthetic differences between samples may have left some with surfaces that enhance the effect, he says—and if so, researchers may learn to engineer particles to optimize it.

Even if a large MEG turns out to be real, however, two separate teams found that getting those charges out of the nanocrystals won't be easy. Randy Ellingson of the National Renewable Energy Laboratory (NREL) in Golden, Colorado, reported that his group had made simple solar cells containing a layer of PbSe nanocrystals, with electrodes above and below. Creating the nanocrystals leaves them decorated with organic groups around the outside. When they are put straight into the films, the crystallites are too far apart to pass charges to the electrodes, where they can be sent through a circuit to do work. So in their current study, Ellingson and his team treated their nanocrystals with hydrazine, which shortened the organic groups and allowed the nanocrystals to sit closer to one another. The solar cells worked. But spectroscopic studies suggested that the hydrazine treatment killed the MEG effect.

Meanwhile, Byung-Ryool Hyun, a graduate student in Frank Wise's group at Cornell University, reported another challenge in getting charges out of PbS. In this case, the nanocrystals were linked to titanium dioxide (TiO2) nanoparticles. When electrons are generated in the nanoparticles, they should move readily to the TiO2. But Hyun reported that electrons moved so slowly that the charges typically recombined with holes and gave up their energy before the TiO2 could snag them.

So is this the end of the road for MEG? Arthur Nozik, who has helped lead the MEG effort at NREL, says he hopes not. “It's kind of a messy situation,” he says. He's hopeful that further research will reveal ways to produce a large MEG effect. For now, however, hopes are dimming for MEG solar cells.


Navigate This Article