Chasing clues about the infant universe in relic light known as the cosmic microwave background, or CMB, scientists are devising more elaborate and ultrasensitive detector arrays to measure the properties of this light with increasing precision.
To meet the high demand for these detectors that will drive next-generation CMB experiments, and for similar detectors to serve other scientific needs, researchers at the Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) are pushing to commercialize the manufacturing process so that these detectors can be mass-produced quickly and affordably.
The type of detector they are working to commercialize incorporates sensors that, when chilled to far-below-freezing temperatures, operate at the very edge of superconductivity—a state in which there is zero electrical resistance. Incorporated in the detector design is transition-edge sensor (TES) technology that can be tailored for ultrahigh sensitivity to temperature changes, among other measurements.
The team is also working to commercialize the production of ultraprecise magnetic field sensors known as SQUIDs (superconducting quantum interference devices).
In the current TES detector design, each detector array is fabricated on a silicon wafer and contains about 1,000 detectors. Hundreds of thousands of these detectors will be needed for a massive next-generation CMB experiment, dubbed CMB-S4.
The SQUID amplifiers are designed to enable low-noise readout of signals from the detectors. They are intended to be seated near the detectors to simplify the assembly process and the operation of the next-generation detector arrays.
More exacting measurements of the CMB light’s properties, including specifics on its polarization—directionality in the light—can help scientists peer more deeply into the universe’s origins, which in turn can lead to more accurate models and a richer understanding of the modern universe.
Berkeley Lab researchers have a long history of pioneering achievements in the in-house design and development of new detectors for particle physics, nuclear physics, and astrophysics experiments. And while the detectors can be built in-house, scientists also considered the fact that commercial firms have access to state-of-the-art, high-throughput microfabricating machines and expertise in larger-scale manufacturing processes.
So Aritoki Suzuki, a staff scientist in Berkeley Lab’s Physics Division, for the past several years has been working to transfer highly specialized detector fabrication techniques needed for new physics experiments to industry. The goal is to determine if it’s possible to produce a high volume of detector wafers more quickly, and at lower cost, than is possible at research labs.
“What we are building here is a general technique to make superconducting devices at a company to benefit areas like astrophysics, the search for dark matter, quantum computing, quantum information science, and superconducting circuits in general,” said Suzuki, who has been working on advanced detector R&D for about a decade.
This breed of sensors has also been enlisted in the hunt for a theorized nuclear process called neutrinoless double-beta decay that could help solve a riddle about the abundance of matter over antimatter in the universe, and whether the ghostly neutrino particle is its own antiparticle.
Progress toward commercial production of the specialized detectors has been promising. “We have demonstrated that detector performance from commercially fabricated detectors meet the requirements of typical CMB experiments,” Suzuki said.
Work is underway to build the prototype detectors for a planned CMB experiment in Chile known as the Simons Observatory that may incorporate the commercially produced detectors.
About 3 miles above sea level, in the Atacama Desert of Northern Chile, researchers have worked on successive generations of TES-based detector arrays for CMB-related experiments including POLARBEAR, POLARBEAR-2, the Simons Array, and the Simons Observatory.
A detector array for two telescopes that are part of the POLARBEAR-2 and Simons Array experiments is now being fabricated at UC Berkeley’s Marvell Nanofabrication Laboratory by Berkeley Lab and UC Berkeley researchers. The effort will ultimately produce 7,600 detectors apiece for three telescopes. The first telescope in the Simons Array has just begun its commissioning run.
The Simons Observatory project, which is now in a design and prototyping phase, will require about 80,000 detectors, half of which will be fabricated at the Marvell Nanofabrication Laboratory.
These experiments are driving toward a CMB-S4 experiment that will combine detector arrays in Chile and near the South Pole to better resolve the cosmic microwave background and possibly help determine whether the universe underwent a brief period of incredible expansion known as inflation in its formative moments.
The commercial fabrication effort is intended to benefit this CMB-S4 experiment, which will require a total of about 500,000 detectors. The current design calls for about 400 detector wafers that will each feature more than 1,000 detectors arranged on hexagonal silicon wafers measuring about six inches across. The wafers are designed to be tiled together in telescope arrays.
Suzuki, who is part of a scientific board working on CMB-S4 along with other Berkeley Lab scientists, is collaboring with Adrian Lee, another board member who is also a physicist at Berkeley Lab and a UC Berkeley physics professor. It was Lee who pioneered microfabrication techniques at UC Berkeley to help speed the production of TES-containing detectors.
In addition to the detector production at UC Berkeley’s nanofabrication laboratory, researchers have also built specialized superconducting readout electronics in a nearly dustless clean room space within the Microsystems Laboratory at Berkeley Lab.
Before the introduction of higher-throughput manufacturing processes, detectors “were made one by one, by hand,” Suzuki noted.
Suzuki labored to develop the latest 6-inch wafer design, which offers a production throughput advantage over the previously used 4-inch wafer designs. Older wafers had only about 100 detectors, which would have required the production of many more wafers to fully outfit a CMB-S4 experiment.
The current detector design incorporates niobium, a superconducting metal, and other uncommon metals like palladium and manganese-doped aluminum alloy.
“These are very unique metals that normally companies don’t touch. We use them to achieve the unique properties that we desire for these detectors,” Suzuki said.
The effort has benefited from a Laboratory Directed Research and Development grant that Lee received in 2015 to explore commercial fabrication of the detectors. Also, the research team has received support from the federally supported Small Business Innovation Research program, and Suzuki has also received support from the DOE Early Career Research Program.
Suzuki has worked with Hypres Inc. of New York and STAR Cryoelectronics of Santa Fe, New Mexico, on the fabrication processes for the detectors, and worked with the University of New Mexico and STAR Cryoelectronics on the SQUID amplifiers. Suzuki said that working with the companies has been a productive process. “They gave us a lot of ideas,” he said, to help improve and streamline the processes.
The industry-produced SQUID amplifiers will be used in one of the telescopes of the POLARBEAR-2/Simons Array experiment, Suzuki noted, and the design of these amplifiers could drive improvements in the readout electronics of a CMB-S4 experiment.
As a next step in the effort to commercially fabricate detectors, a test run is planned this year to demonstrate fabrication quality and throughput.