Researchers at Lawrence Livermore National Laboratory (LLNL), the University of California (UC) Berkeley, UC Riverside and UC Santa Barbara have miniaturized quadrupole ion traps for the first time with 3D printing - a breakthrough in one of the most promising approaches to building a large-scale quantum computer.
Quadrupole ion traps have four electrode poles that create an oscillating electrical potential that traps ions by overriding their natural vibration, similar to how raising or lowering different ends of a playground parachute can keep a soccer ball on its surface. The traps keep ions confined for hours before they escape, and if the ions are cooled to their ground state - where they are at their lowest possible energy - they can function as quantum bits (qubits), the most basic unit of information in a quantum computer.
Made with ultrahigh-resolution, two-photon polymerization (2PP) 3D printing, the millimeter-scale ion traps can confine calcium ions with frequencies, error rates and coherence competitive with the state-of-the-art and can be used to perform single- and two-qubit operations. The team's findings were published in a recent paper in Nature.
"This is the sort of technological change that's going to take ion traps from working well with just a few ions to doing something that we would consider a computation and, hopefully, to something we can start using as a computer," said co-author Kristi Beck, LLNL physicist and director of the Livermore Center for Quantum Science.
New shapes and sizes
Qubits need to remain coherent (in a quantum state) for as long as possible and behave as reliably as possible for researchers to effectively encode data and perform operations. Trapped ions have much longer coherence times and work at higher temperatures than other approaches - requiring a laser to cool ions to their ground state instead of cryogenic refrigeration.
However, there is a tradeoff between performance and scalability. In industry, researchers commonly use "planar" ion traps with surface electrodes that can scale well as building blocks of a large-scale information processing system, but the traditional 3D designs have better performance. The team saw a potential solution to this problem in 3D printing.
"3D printing gives us the confinement we need to trap the ion well and at high frequencies, and we can also make many ion traps on the same chip," said Materials Engineering Division (MED) staff engineer and co-first author Xiaoxing Xia. "This is similar to when people worked with bulky, individual transistors before the integrated circuit was invented. 3D printing can allow us to move beyond these conventional traps into more highly integrated systems like our current processors."
The printed traps can confine calcium ions with much higher frequencies than both the regular 3D traps and planar traps, creating deep harmonic potentials between electrodes that stabilize the system and improve coherence. The traps demonstrated their stability by confining two calcium ions that exchanged positions every few minutes - competitive with the state-of-the-art.
The team also implemented a two-qubit entangling gate with 98% fidelity, performed single qubit rotations and measured motional heating rates, which quantify one of the primary sources of error for trapped ion quantum gates.
"I'm excited by the potential that opens up from just our proof of concept," said MED staff engineer and co-author Abhinav Parakh. "To be able to harness the exponential computational power [from quantum computing], we need to have multiple ions entangled with each other, bring them close, do computation with them, and then pull them apart - something can be done efficiently using 3D printed structures."
The team can reliably print a miniaturized ion trap in 14 hours from scratch, or in 30 minutes if they only print the electrodes on an existing substrate. The ability to rapidly prototype and the flexibility to print in nearly any configuration also gives them a chance to experiment with new designs, such as a planar trap based on the classic 3D design that the team developed, printed, miniaturized and used to trap ions at both cryogenic and room temperatures.
"We dramatically expanded the range of achievable trap geometries and increased the complexity," said UC Berkeley quantum physics professor and co-author Hartmut Haeffner. "With this increased design space, we can now think very differently on how to optimize and miniaturize our ion traps."
As designs evolve, the team aims to integrate photonics and electronics on the same chip to make the entire system more efficient and compact. Beck also wants to explore ways to make quantum computers more reliable and easier to control. The greatest source of error is noise, uncontrolled interactions with the environment that make a quantum system behave unreliably, and the surfaces of ion traps are currently a major source of noise.
"If we can take away more material that is close to the ions, there's going to be fewer places where we know that noise is entering into the system, so we expect to see better performance," she said.
A potential catalyst
The project was originally supported by a UC National Laboratory Fees Research Program, which fostered the collaboration among LLNL, UC Berkeley, UC Riverside and UC Santa Barbara.
"By reinvesting and continually refining how we direct the fees earned from managing our national labs, the University of California is proud to support high-impact collaborations by bringing together talent and expertise that open doors to breakthrough science and speed discoveries," said June Yu, vice president of UC National Laboratories.
Materials Science Division (MSD) researcher Juergen Biener, who was co-PI on the project with Haeffner, credits the project's success to this partnership.
"Getting this world-class expertise together to share ideas and work as a team is the type of thing that really accelerates a field," Biener said. "Within three years, we went from zero to almost state-of-the-art in a fairly competitive field and I think that, in itself, tells a story."
The team hopes their innovative approach will help put LLNL on the map for ion trap quantum computing hardware development and serve as a potential catalyst for future collaborations that will help transform their ideas into commercial products.
The miniaturized ion traps can also be used for sensing and ultra-precise atomic clocks, and if the laser cooling system is successfully scaled and integrated on a chip, they could be the basis for compact, low-power mass spectrometers for precision metrology. Xia also sees it as an opportunity to show what high-resolution 3D printing can do for the field.
"Quantum computing is an ideal early adopter for 3D printing because they want the very high resolution, fine features and intricate 3D geometry that no other fabrication technique can provide," said Xia.
Other co-authors on the paper include Shuqi Xu, Qian Yu, Sumanta Khan, Eli Megidish, and Bingran You from UC Berkeley, Boerge Hemmerling from UC Riverside, and Andrew Jayich from UC Santa Barbara.
-Noah Pflueger-Peters
