Australasian Science: Australia's authority on science since 1938

Silicon Defects Corrected

By Stephen Luntz

A method for correcting defects in low-grade silicon could boost the efficiency of commercial photovoltaic panels by 10% in a few years time, further improving solar power’s competitiveness.

The manufacture of high-grade silicon is expensive. Specialist uses aside, manufacturers use silicon containing a range of defects. This helps to limit maximum efficiencies for most commercial solar panels to 19%.

“The process by which hydrogen atoms can bind to defects in the silicon, where a silicon atom forms a dangling bond and hydrogen can make it electrically inactive, has been known for some time,” says Prof Stuart Wenham of the School of Photovoltaics and Renewable Energy Engineering at the University of NSW.

However, getting hydrogen atoms to travel through the silicon to the sites of defects and then bond and deactivate them has been a challenge. Hydrogen atoms can be positive, neutral and negative. They move through silicon most easily when neutral, but need to be in an appropriate charge state to fix the defects in the silicon.

While Wenham says “it was once expected to be easy” to control hydrogen’s charge state, this has turned out to be a major obstacle to correcting defects. Now, however, UNSW has patented a method to solve this.

“Our research team has worked out how to control the charge state of hydrogen atoms in silicon,” Wenham says. “We have seen a 10,000 times improvement in the mobility of the hydrogen, and we can control the hydrogen so it chemically bonds to things like defects and contaminants, making these inactive.”

Some silicon contains a dominant form of defect, meaning that most problems can be fixed through the application of hydrogen in a particular state. In other cases the defects are different, so sequential application of hydrogen in different states would be required.

While the cost of silicon panels has fallen by 80%, so-called “balance of system costs” including installation and inverters have failed to keep pace. As a result, a 10–20% improvement in efficiency would currently only make a small difference to the overall cost of installing solar.

However, Wenham anticipates that “these costs will fall as part of the maturing of the industry, so we expect that in the longer term the cost of panels will again be the dominant component in price”. He expects the first industry partners will be releasing panels using UNSW’s hydrogen correction mechanism by 2016.

The technique could also have applications for other semiconductor devices, but Wenham says that “these usually have much more expensive processing that dominates the cost, so there is less pressure to drive down the price of silicon”.