A quantum dot single-photon emitter capable of operating at temperatures up to 120K could help pave the way to advances in quantum cryptography, according to a presentation at the Nanotech2004 conference held in March in Boston, MA.
Richard Mirin, electronics engineer at the National Institute of Standards and Technology (NIST; Boulder, CO), detailed his experimental proof of single photon emission, also known as photon antibunching, from quantum dots at temperatures up to 120K. At 135K, Mirin said he still saw evidence of antibunching, however, "It's not unequivocal evidence that it's a single emitter," at that temperature, he noted.
A number of groups are using small islands of semiconductor material about 10 to 20 nm across called quantum dots to produce single photons. Most of the measurements so far have been at around 10K, Mirin said, with some at 40K. Practical single-photon emission at 77K is "eminently doable," he said, and 120K should not be out of the question. Higher-temperature operation would allow for simpler, easier-to-build systems for researchers developing quantum cryptography systems and new methods of metrology.
"I'd like to get to room temperature [293K], but I'm not confident it will make it all the way to room temperature," he said. "Ideally what we'd like to do is have a rack-mounted system with an optical fiber, and you'd get one photon at a time out of your fiber."
Mirin used quantum dots grown epitaxially from layers of indium-gallium-arsenide and gallium-arsenide as the single photon emitters. He etched mesas into the material to isolate various numbers of quantum dots. Mirin then excited a mesa with pulses from a mode-locked Ti:sapphire laser emitting 200-fs pulses at 850 nm at a rate of 82 MHz; a cooled CCD camera recorded the emissions from the quantum dots.
This graph shows the difference between using one quantum dot and using an ensemble of about 100 quantum dots. The strong suppression of the peak at zero delay for the single quantum dot is clearly visible, whereas the peaks associated with an ensemble of independent emitters are all the same height.
Because quantum dots are so small, they can be made to hold only one electron and one hole, so they're only capable of emitting one photon at a time. In practice, they can actually hold two electron pairs with opposite settings of the quantum mechanical property called spin, one pair in the spin up state and another pair in spin down. That allows for combinations that produce a photon with a slightly different emission wavelength than if there's only a single electron-hole pair.
Mirin hopes to address this in future work. "If you can do electrical injection of exactly one electron-hole pair, then you can have the perfect single-photon source," he said.
He'd also like to integrate the quantum dots in semiconductor microcavities. A group at Stanford University (Stanford, CA), led by Yoshihisa Yamamoto, has done that, but not at these higher temperatures. Putting the quantum dots in microcavities makes them more efficient and easier to use. "Just as you can make arrays of VCSELs, you could make arrays of single-photon sources," Mirin said.
Single-photon sources are important for quantum encryption because the polarization states of a string of photons can act as a distributable encryption key for data. At least two companies already sell quantum encryption systems, but their emitters are not perfect because the system must correct for the emission of multiple photons, reducing the bit rate of the data stream.
NIST's main interest in single-photon emission is for metrology, calibrating optical power measurements in the subpicowatt range. There could also be applications in biology for single-molecule detection, Mirin added.
Other methods of single-photon emission are being investigated, but Mirin said each one has some difficulty that reduces its efficiency. It will be quite some time until it's clear which method is the most effective, he said.