Sunlight Trapped with Si Nanowires

A better way of trapping sunlight using silicon nanowires is behind a new approach that could dramatically reduce the costs of making photovoltaics by allowing them to be manufactured with metallurgical-grade, rather than ultrapure, silicon.

Although current silicon PV technologies can convert sunlight into electricity at impressive 20 percent efficiencies, their cost is prohibitively expensive for large-scale use. Researchers at Lawrence Berkeley National Laboratory have developed an approach that both reduces the cost of silicon solar cells and dramatically improves their light-trapping ability.

This photovoltaic cell is composed of 36 individual arrays of silicon nanowires featuring radial p-n junctions. The color dispersion demonstrates the excellent periodicity present over the entire substrate. (Images courtesy of Peidong Yang)

"Through the fabrication of thin films from ordered arrays of vertical silicon nanowires, we've been able to increase the light trapping in our solar cells by a factor of 73," said chemist Peidong Yang, who led the research. "Since the fabrication technique behind this extraordinary light-trapping enhancement is a relatively simple and scalable aqueous chemistry process, we believe our approach represents an economically viable path toward high-efficiency, low-cost thin-film solar cells."

Yang holds joint appointments with Berkeley Lab's Materials Sciences Div., and the University of California, Berkeley, chemistry department. He is a leading authority on semiconductor nanowires – one-dimensional strips of materials whose width measures only one-thousandth that of a human hair but whose length may stretch several microns.

"Typical solar cells are made from very expensive ultrapure single crystal silicon wafers that require about 100 µm of thickness to absorb most of the solar light, whereas our radial geometry enables us to effectively trap light with nanowire arrays fabricated from silicon films that are only about eight micrometers thick," he said. "Furthermore, our approach should, in principle, allow us to use metallurgical-grade or 'dirty' silicon rather than the ultrapure silicon crystals now required, which should cut costs even further."

Generating Electricity from Sunlight

At the heart of all solar cells are two separate layers of material, one with an abundance of electrons that functions as a negative pole, and one with an abundance of electron holes (positively charged energy spaces) that functions as a positive pole. When photons from the sun are absorbed, their energy is used to create electron-hole pairs, which are then separated at the interface between the two layers and collected as electricity.
Because of its superior photoelectronic properties, silicon remains the photovoltaic semiconductor of choice, but rising demand has inflated the price of the raw material. Furthermore, because of the high level of crystal purification required, even the fabrication of the simplest silicon-based solar cell is a complex, energy-intensive and costly process.

A radial p-n junction consists of a layer of n-type silicon forming a shell around a p-type silicon nanowire core. This geometry turns each individual nanowire into a photovoltaic cell.

Yang and his group can reduce both the quantity and the quality requirements for silicon by using vertical arrays of nanostructured radial p-n junctions rather than conventional planar p-n junctions. In a radial p-n junction, a layer of n-type silicon forms a shell around a p-type silicon nanowire core. As a result, photoexcited electrons and holes travel much shorter distances to electrodes, eliminating a charge-carrier bottleneck that often arises in a typical silicon solar cell. The radial geometry array also, as photocurrent and optical transmission measurements by Yang and Garrett revealed, greatly improves light trapping.

"Since each individual nanowire in the array has a p-n junction, each acts as an individual solar cell," Yang said. "By adjusting the length of the nanowires in our arrays, we can increase their light-trapping path length." Although the conversion efficiency of these solar nanowires was only about 5 to 6 percent, Yang said this efficiency was achieved with little effort put into surface passivation, antireflection and other efficiency-increasing modifications.

"With further improvements, most importantly in surface passivation, we think it is possible to push the efficiency to above ten percent," Yang said. Combining a 10 percent or better conversion efficiency with the greatly reduced quantities of starting silicon material and the ability to use metallurgical-grade silicon, should make the use of silicon nanowires an attractive candidate for large-scale development.

As an added plus, Yang said, "our technique can be used in existing solar panel manufacturing processes."

Yang described the research in the paper, "Light Trapping in Silicon Nanowire Solar Cells," published in the journal Nano Letters, which he co-authored with Erik Garnett, a chemist who was then a member of Yang's research group. The work was funded by the National Science Foundation's Center of Integrated Nanomechanical Systems.

For more information, visit: www.lbl.gov