Real New Age crystal: Sandia’s John Reno fabricates the best terahertz crystals in US
The relatively unexplored terahertz frequency range – higher than microwaves, lower than the far infrared – has long intrigued researchers. Lasers and detectors in the THz frequency regime have wide-ranging applications in spectroscopy, astronomy, medical and other types of imaging, and in remote sensing. They are expected to be useful in chemical sensing systems for the detection of molecular absorption lines associated with trace gases.
While lasers with detectors already exist to send and receive signals from that realm, commercial lasers weigh hundreds of pounds, contain long fragile glass tubes, and cost a lot of money, says Jerry Simmons, Manager of Sandia’s Semiconductor Materials & Device Sciences Department (1123).
What’s wanted are lightweight, inexpensive, robust devices. Such devices are now appearing. Their generating sources are semiconductor crystal films. Perhaps surprisingly, the very best crystal films for terahertz work in the US are made on a 10-year-old molecular beam epitaxy machine in Sandia’s aging Compound Semiconductor Research Lab.
Credit, many agree, goes to the expert hand of John Reno (1123).
Sandia is the best and (until very recently, through a transfer of Sandia technology) the only place in the US to grow a crystal film that can be made into such lasers and sensors.
One difference between the older, clunkier lasers and modern attempts at penetrating the terahertz range are obvious. Instead of requiring the space of a table top, THz lasers from crystals grown by John are smaller than a cubic millimeter.
Achieving a laser that operates close to room temperature is still an issue. Currently the highest operating temperature – 130K – has been achieved by Professor Qing Hu at MIT, using crystals grown at Sandia. While the required cooling system for this device is bigger than a cubic millimeter, the overall size is still small compared to the older lasers.
Mike Wanke (1743) at Sandia is developing a terahertz semiconductor laser that should be easier and cheaper to maintain than the MIT version. Aided by tiny thermoelectric coolers, it should operate in higher ambient temperatures.
In addition to growing the best THz laser crystals, John grows another type of crystal structure – one of such ultrahigh purity that, at temperatures of 1 K and below, electrons can travel up to 0.1 millimeter in the crystal without scattering. Only four places in the world have successfully grown these structures.
Because only two of these places are in the US, John’s ultrapure semiconductor layers are in high demand at universities and "are a centerpiece of the resources being assembled at Sandia’s Center for Integrated Nanotechnologies," says Neal Shinn, CINT User Program Manager. John’s materials already play key roles in several CINT collaborations.
This material also is being used to develop revolutionary THz detectors, in work again led by Mike Wanke. According to Mike, "Having a grower who can create the structures required for either the lasers or the detectors is incredibly rare, but one able to grow both is amazing." All these gadgets, which weigh almost nothing and whose materials cost will be trivial when scaled up to mass production, use crystals fabricated by John Reno.
"What’s wanted is a solid-state source that is reliable, depends on semiconductor technology, and delivers a reasonable amount of power," says Mark Lee (1123). He says that except for people John has taught, "John Reno is the only grower of semiconductor material in the US who has grown terahertz crystals that work."
What’s needed, says Mark, is several milliwatts in continuous output. "John’s output of 30 milliwatts at 2.5 terahertz is an order of magnitude higher than any other at that frequency," he says.
How does John achieve this on his aging MBE machine?
"It’s not the machine, it’s the skill of the person operating it," says Mark. There are, he says, a half-dozen molecular beam epitaxy growers in the US who could deliver as good or a bit better high-purity material.
"Of the other two things that really matter, one is precision in layer thicknesses; equally important is alloy precision. John has been tuning his machine for years." Nobody in the US does it better, he says.
To see bearded John at work in his lair – gowned, booted, gloved, masked, and capped, facing a 10-foot-long, seven-foot-high, molecular beam epitaxy machine – is a stirring sight. John is as suited as any armored knight facing a dragon.
On his commands, the MBE’s black electrical wires, thin silver tubes delivering nitrogen, thick cylindrical cooling pumps, and still-larger arm of an electron beam gun all work together to deposit atoms, atomic layer by structured atomic layer.
John’s position is so central that in the strange deck of cards called the Tarot, whose face cards are claimed to represent archetypical human situations – the Hanged Man, the Empress, the Huntsman and so on – John might be a new archetypal entity . . . . the Crystal Grower.
But not, obviously, of the glitzy kind hawked by merchants in Santa Fe. THz sources based on semiconductor crystals have been achieved at very few other sites in the world – among them, the Cavendish Lab in England and the University of Neuchatel in Switzerland.
Each crystal for terahertz lasers takes approximately 17 hours to grow and is composed of 175 "steps," so-called because the declining energies at each step resemble a set of stairs going downhill. Indeed, this type of semiconductor laser is called a "quantum cascade laser," since the electrons act like water as they "cascade" down the steps, emitting a THz photon at each one. These steps consist of about 10 layers built of different thicknesses and materials, each composed of five to 35 atomic layers. John can grow one crystal each day.
The amount of care resembles that taken by a Japanese blacksmith forging a classical samurai ceremonial sword, rearranging and elongating his material’s crystal domains by continual refolding. "The process has to remain the same, whether after five hours or 17 hours, or the quantum levels [of the materials] change," says John. The work requires the opening and closing of precise shutters, he says. To do this, he heats and boils off potentially dangerous materials like arsenic, gallium, aluminum, and indium at roughly 1,100 degrees C, and allows them to coalesce as a crystal film on a mirror-smooth gallium arsenide substrate, 625 microns thick, at nearly 600 degrees C. The work must be accurate down to the atom.
"Some sites are gallium surrounded by four arsenic; or an arsenic surrounded by four gallium," John says. "Different metals are different distances apart." A deposited atom has to have time to find a home before being barraged by new atoms, he says. "They need to be given the right amount of energy to move on this hot surface."
John sees himself as facing two challenges. One is the chemical/mechanical challenge to put what he wants into, say, four atomic layers. The second challenge is the precise growth that will produce the desired output. For that, he says, he relies on researchers to provide him with a working set of specifications to build to.
The specifications are in, and more are on their way.