Study could help improve the efficiency of solar cells

UC Berkeley researchers laid the groundwork for potentially developing more energy-efficient solar cells in a new study published Sept. 15 in the journal Physical Review Letters.

In the past, researchers have focused on looking at domains — the crystal surface of a solar cell that can convert sunlight into energy — as a source of electric energy. However, UC Berkeley and Lawrence Berkeley National Laboratory researchers discovered that the walls between the domains — and not just the domains themselves — can also increase energy efficiency.

When sunlight hits a domain, it knocks an electron to a higher energy state, which can then be stored and power a device, said Ramamoorthy Ramesh, a professor in the campus Department of Materials Science and Engineering and co-author of the study.

But between each domain is a wall that until now has been casually disregarded in terms of research, he added.

Ramesh compared domain walls to the walls of an apartment, where the domain is the actual room of the apartment and the domain wall surrounds it, he said. Researchers have been focusing on the contents of the apartment, but they have been ignoring the potential energy that the materials of the walls could be yielding.

The concept of investigating the wall of the domain is scientifically significant, because the walls that separate each domain are only a few nanometers thick but could change the course of energy research, Ramesh said.

“Domain walls are a very interesting animal,” he said. “The wall itself can be doing something.”

A collaboration of three campus research groups focused on using feral oxides — particularly bismuth feral oxide — as a material for the domain wall, which raised the maximum voltage to 15 volts, a dramatic increase from silicon — the material used for powering computers and many electronic devices — which only has a maximum output voltage of just one volt, according to Ramesh.

Although bismuth feral oxide can excite more electrons to a higher energy state than silicon, silicon can produce more electrons and is therefore still more energy-efficient, said Junqiao Wu, a co-author of the study and professor in the campus Department of Materials Science and Engineering.

While bismuth feral oxide can produce a higher voltage capacity, silicon can produce so many more electrons — or a higher current — that the oxide is not likely to replace silicon in the near future as a material in solar cells.

“(Bismuth feral oxide’s) efficiency is lower than silicon,” Wu said. “Even though consumers care about energy efficiency and cost, as a researcher, we need to look if there is another system that can offer a possible breakthrough.”

Even though bismuth feral oxide is not likely to replace silicon in the near future as a semiconductor, studying the domain walls opens a new avenue of research that could eventually yield fruitful results and produce higher energy-efficient solar cells, Wu said.

The discovery of the importance of domain walls not only brings new research topics, but also more questions, Ramesh said.

“I don’t have to look at the domain as a unit. I can now also look at the wall,” he said. “What can we do with the wall? Can we store information? Can you make the wall electrically conductive? There are a lot of possibilities. We still have a whole bunch of research to do.”