Reading time ( words)
Life may be as unpredictable as a box of chocolates, but ideally, you always know what you’re going to get from a quantum dot. A quantum dot should produce one, and only one, photon—the smallest constituent of light—each time it is energized. This characteristic makes it attractive for use in various quantum technologies such as secure communications. Oftentimes, however, the trick is in finding the dots.
“Self-assembled, epitaxially grown” quantum dots have the highest optical quality. They randomly emerge (self-assemble) at the interface between two layers of a semiconductor crystal as it is built up layer-by-layer (epitaxially grown).
They grow randomly, but in order for the dots to be useful, they need to be located in a precise relation to some other photonic structure, be it a grating, resonator or waveguide, that can control the photons that the quantum dot generates. However, finding the dots—they’re just about 10 nanometers across—is no small feat.
Always up for a challenge, researchers working at the National Institute of Standards and Technology (NIST) have developed a simple new technique for locating them, and used it to create high-performance single photon sources.
This new development, which appeared in Nature Communications,* may make the manufacture of high-performance photonic devices using quantum dots much more efficient. Such devices are usually made in regular arrays using standard nanofabrication techniques for the control structures. However because of the random distribution of the dots, only a small percentage of them will line up correctly with the control structures. This process produces very few working devices.
“This is a first step towards providing accurate location information for the manufacture of high performance quantum dot devices,” says NIST physicist Kartik Srinivasan. “So far, the general approach has been statistical—make a lot of devices and end up with a small fraction that work. Our camera-based imaging technique maps the location of the quantum dots first, and then uses that knowledge to build optimized light-control devices in the right place.”
According to co-lead researcher Luca Sapienza of the University of Southampton in the United Kingdom, the new technique is sort of a twist on a red-eye reducing camera flash, where the first flash causes the subject’s pupils to close and the second illuminates the scene. Instead of a xenon-powered flash, the NIST team uses two LEDs.
In their setup, one LED activates the quantum dots when it flashes (so the LED gives the quantum dots red-eye). At the same time, a second, different color LED flash illuminates metallic orientation marks placed on the surface of the semiconductor wafer the dots are embedded in. Then a sensitive camera snaps a 100-micrometer by 100-micrometer picture.
By cross-referencing the glowing dots with the orientation marks, the researchers can determine the dots’ locations with an uncertainty of less than 30 nanometers. The coordinates in hand, scientists can then tell the computer-controlled electron beam lithography tool to place the control structures in the correct places, with the result being many more usable devices.
Using this technique, the researchers demonstrated grating-based single photon sources in which they were able to collect 50 percent of the quantum dot’s emitted photons, the theoretical limit for this type of structure.
They also demonstrated that more than 99 percent of the light produced from their source came out as single photons. Such high purity is partly due to the fact that the location technique helps the researchers to quickly survey the wafer (10,000 square micrometers at a time) to find regions where the quantum dot density is especially low, only about one per 1,000 square micrometers. This makes it far more likely that each grating device contains one—and only one—quantum dot.
This work was performed in part at NIST's Center for Nanoscale Science and Technology (CNST), a national user facility available to researchers from industry, academia and government. In addition to NIST and the University of Southampton, researchers from the University of Rochester contributed to this work.