Portland, Ore. -- The future of semiconductors is not chips: Instead of fabricating circuits on chips and soldering them to printed-circuit boards, Canadian researchers propose painting transparent "solution processed" circuits directly onto a device's surface. Such semiconductor circuits--from emitters for large-area displays to detectors for spray-on solar cells--could drastically lower the cost of electronic devices, the group says.
The first beneficiaries could be night vision goggles for the military that would be 10 times more sensitive, yet less expensive, than today's models. But that is just the beginning, according to the team at the University of Toronto, because now spray-on circuits no longer have to sacrifice performance to attain low cost.
"We are reporting the first high-performance semiconductor device made by solution processing," said research group leader Ted Sargent, a University of Toronto professor in electrical engineering and the Canada Research Chair in Nanotechnology. "Our solution-processed detectors now have about 10 times greater sensitivity than traditional, grown-crystal epitaxial semiconductor devices."
In 2003, Sargent's group demonstrated their first solution-based infrared emitters. Then, last year, they reported the first solution-based infrared detector. Unfortunately, the photodetector had dismal performance compared with traditional infrared detector chips.
"When we reported the first solution-processed infrared photodetectors, they were for proof of concept. Their efficiency was lower than [what would be] necessary in any imaginable application," said Sargent. "Now we've achieved much-improved electrical conduction, improving the sensitivity of our detectors by two orders of magnitude [a hundredfold]."
Besides being inexpensive to mass-produce, said Sargent, the photodetector material could post a tenfold sensitivity increase for military night vision systems, which image in infrared, as well as for biomedical imaging systems that use infrared to see through skin.
Sargent enlisted Gerasimos Konstantatos, a doctoral researcher at the University of Toronto, to perform much of the work leading to the discovery.
Organic semiconductors for light-emitting diodes offer better colors, thereby justifying the expense of hermetically sealing them to prevent deterioration of their polymer base. Likewise, conductive polymers are widely used today, but only where their physical flexibility offsets their lower conductance compared with metal.
In just the same way, organic circuits born in solutions and then "painted on" through patterned "silkscreens" defining circuitry have offered easier fabrication and lower cost. But their dismal performance compared with today's silicon chips has lured few companies into developing organic circuits beyond the research lab.
Now Sargent's research group promises to speed adoption of solution-based circuitry by eliminating the speed/ cost trade-offs from the equation. Sargent claims that by taking out the organic polymer and using colloidal lead sulphide nanoparticles instead, inexpensive solution-based circuitry can be made to be as high in performance as silicon chips.
Sargent's original formula for the group's organic-semiconductor solution was one part polymer and one part nanoparticles--tiny "quantum dots," measuring 2 to 6 nanometers in diameter and evenly distributed throughout the conductive-polymer matrix. At those ex- tremely small sizes, the quantum-mech- anical properties of electrons cannot be ignored as they pass from nanoparticle to nanoparticle, because the wavelength of an electron--when considered as a quantum-mechanical wave instead of a particle of matter--is about 20 nanometers.
Consequently, since the nanoparticles are 10 times smaller than the wavelength of the electrons passing through them, the electrons have to squeeze down to fit inside the tiny nanoparticles. That process evokes energetic semiconducting behaviors, including the photovoltaic, photoluminescent and photoconductive effects.
Unfortunately, the relatively low conductance of the polymer matrix itself, compared with the semiconducting nanoparticles, subtracted from the efficiency and sensitivity of Sargent's organic solution-based circuitry. Sargent tried adding more nanoparticles to the mix to mitigate the "organic problem," until he hit upon a method of eliminating the polymer matrix altogether.
Now Sargent mixes pure semiconducting nanoparticles with oleic acid and spin-coats the resulting brew onto glass substrates. After a two-hour bath in a methanol, evaporation of the solvent leaves an 800-nanometer layer of pure semiconducting nanoparticles.
"We went into an entirely new geometry and materials mix," said Sargent. "Before, we worked with a nanocrystal-polymer composite. But this is a pure-nanocrystal device with no polymer; the organic molecules originally used to protect the nanoparticles are largely gone in the finished device."
Sargent previously reported that his detectors could also be configured as infrared solar cells. But to achieve the high sensitivity of the pure nanocrystalline device, the researchers have abandoned the photovoltaic approach-- where electricity is generated from light--and instead configured the nano- particles for a photoconductive mode of operation, in which resistance is diminished and conductance is enhanced by light.
"We believe that this materials system is very well-suited to the photoconductive mode, wherein the conductance of the device is enhanced under illumination," Sargent said.
The research was funded by the Natural Sciences and Engineering Research Council of Canada, the Canada Foundation for Innovation, the Province of Ontario, the Ontario Centres of Excellence and the Canada Research Chairs program.
See related image