by Tim Palucka
Materials Research Society | Published: 29 January 2013
Generating electricity in the field to power a laptop or night vision goggles could someday be just as simple as adding water to a cartridge containing silicon nanoparticles and a base. Researchers at the University at Buffalo (SUNY) have demonstrated that nanoparticles of Si in a basic solution can split water to release hydrogen and power a portable fuel cell to produce electricity. The ability to split water on-demand without adding heat, light, or electricity to the system could be a significant advance in fuel cell technology.
“The reaction rate with these very small 10-nm Si particles is so much faster than with the relatively large 100 nm Si particles,” says Mark Swihart, whose team published their results in a recent issue of ACS Nano Letters. “Because of this fast reaction rate and the fact that there’s no delay between when you add water and when the reaction starts, it makes the technology at least practical in terms of being able to power a device instantaneously.”
While there was some scant evidence in the scientific literature that Si could perform this feat of splitting water to release hydrogen, it was largely ignored because the reaction rate was so slow as to be uninteresting. Using Al, Zn, or metal hydrides for this purpose looked so much more promising that Si fell by the wayside.
But Swihart and his group have been working with Si nanoparticles for more than a decade, mostly in the realm of quantum dot research. In doing so, they frequently had to use a base such as hydrazine for etching, and they noticed that hydrogen was released when aqueous hydrazine reacted with Si. Investigation showed that the hydrogen came not from decomposition of hydrazine, but from the oxidation of Si to release hydrogen from water.
Further investigation of the reaction using Si particles of different sizes, focusing on 10-nm and 100-nm-diameter particles with aqueous KOH, showed a particle size dependent liberation of hydrogen from water. But the factor of 150 increase in the reaction rate for the 10-nm-diameter particles compared to the 100-nm-particles was well in excess of the factor of 6 difference in their specific surface area. Thus, the increase in rate is much greater than expected based on increased surface area alone.
Swihart believes the difference is caused by geometry, not surface area. The 111 lattice planes etch much more slowly than other planes of Si, so crystals terminated entirely by 111 planes react slowly. “The 10 nm particles etch isotropically—they just get smaller and go away,” he says. There’s no time for faceting to occur in this case. But the 100 nm particles undergo anisotropic etching. The faster-reacting 100 and 110 planes etch away first, leaving a particle with slower-reacting 111 planes behind in what he describes as a “hollow nano-balloon structure.” “With the bigger particles,” Swihart says, “eventually the unreactive 111 surfaces are the ones that end up being left,” thus slowing the reaction rate.
As a proof-of-concept, the research team tested a small fuel cell with a 20 stack polymer electrolyte membrane, comparing the fuel cell’s power output when fed hydrogen from the Si nanoparticle reaction versus hydrogen from a gas cylinder. Stoichiometrically, two moles of H2 should be generated for one mole of Si. In the tests, the fuel cell powered by H2 generated by reaction with Si produced more current and voltage than when the fuel cell was fed a stoichiometric amount of H2 from a gas cylinder. The difference is due to additional hydrogen, beyond the stoichiometric reaction amount, that terminates the Si surfaces after fabrication of the nanoparticles.
While there is much more work to be done, Swihart believes that if this technology is ever to become practical as a portable electricity generator, the KOH (or other base) would have to be mixed in with the Si in a cartridge, so you would not have to carry around a bottle of KOH solution. Such a device would come with the instructions “just add water.” For a soldier in the field needing to power night vision goggles, water from a nearby stream could be all he needs.
Read the abstract in ACS Nano Letters here.