IEEE Spectrum – Quantum computers have the potential to solve seemingly intractable problems in no time flat. But a big stumbling block on the path to practical quantum computing is figuring out how to observe the tiny quantum signals that drive computation. In an advance that may make that observation easier, a group at Aalto University, in Finland, has created a new kind of microwave amplifier based on a mechanical resonator—essentially a nanometer-scale tuning fork.
The Aalto amplifier consists of a meandering microwave cavity [blue] coupled to an aluminum resonator. The two components are separated by a gap just a few nanometers wide.
The sensitive measurement of electrical signals is at the heart of modern technology. According to the principles of quantum mechanics, any detector or amplifier necessarily adds a certain amount of noise to the signal, equal to at least the noise added by quantum fluctuations. This quantum limit of added noise has nearly been reached in superconducting devices that take advantage of nonlinearities in Josephson junctions. Here we introduce the concept of the amplification of microwave signals using mechanical oscillation, which seems likely to enable quantum-limited operation. We drive a nanomechanical resonator with a radiation pressure force and provide an experimental demonstration and an analytical description of how a signal input to a microwave cavity induces coherent stimulated emission and, consequently, signal amplification. This generic scheme, which is based on two linear oscillators, has the advantage of being conceptually and practically simpler than the Josephson junction devices. In our device, we achieve signal amplification of 25 decibels with the addition of 20 quanta of noise, which is consistent with the expected amount of added noise. The generality of the model allows for realization in other physical systems as well, and we anticipate that near-quantum-limited mechanical microwave amplification will soon be feasible in various applications involving integrated electrical circuits.
Let us consider the prospects to reach nearly quantum-limited operation of the device in the phase-insensitive mode. Let us first suppose essentially the same setup as presently, but with a short beam only 0.8 micron long and 50 nm thick and wide. This beam has 500MHz, and is nearly in the ground state at 20 mK, and thus would be on par with the best Josephson devices.
Another possible approach towards the quantum limit might be pre-cooling of the mechanics near the ground state by inverting the pump to the red sideband. We note that the cooling tone cannot be applied simultaneously to the amplification pump since they would cancel each other. Hence, the setup should be pulsed: applying a cooling pump pulse followed by an amplification pump pulse might allow to reduce the noise emanating from the mechanical resonator. However, detailed analysis of this would require considering transient effects.
As our amplifier scheme containing a coupling of two oscillators does not explicitely require superconductivity, it could be realized for example in the much sought-after THz regime as well by fabricating a smaller cavity with a resonance frequency within this regime.
Design for the cavity. a, Simulation drawing for the meandering cavity structure, with ideal circuit components inserted between the open ends; b, change of the cavity resonance for 1 fF change of Cg; c, micrograph showing the clamped beam and part of the cavity. The roughly isotropic etch causes about 700 nm undercut also for the cavity.
In the new amplifier, two elements are coupled together: a microwave cavity (a sort of mirrored, walled maze for electromagnetic waves) and a mechanical resonator (a suspended flexible aluminum beam). Only a tiny gap separates the mechanical resonator and the optical cavity; etched by a focused beam of ions, the gap is just 6 to 13 nanometers wide. “It’s one of the keys to the amplification,” says Massel. “The smaller the separation, the higher the coupling between optics [the cavity] and mechanics [the resonator].”
When the microwave cavity is hit with a strong microwave signal (called the pump), radiation pressure from the microwaves reduces the damping of the resonator, allowing it to vibrate with less resistance. When a second microwave signal—the one intended for amplification—hits the cavity, the resonator’s reduced damping lets the beam amplify the signal.
One advantage of the mechanical amplifier over competing designs, says Massel, is that its simple structure isn’t susceptible to “electrical flicker,” a major source of noise found in other kinds of amplifiers. Flicker, normally a low-frequency disturbance, is found in almost all electrical devices, including Josephson junctions. The best amplifier would eliminate all sources of noise except for those from quantum fluctuations—the uncertainty about an object’s position or momentum that arises when it’s measured. The Aalto mechanical amplifier would leave only thermal noise and those unavoidable quantum blips to interfere with measurements. “This is the selling point of this concept,” says IEEE Fellow Fadhel Ghannouchi, a microwave researcher at the University of Calgary, in Alberta, who was not involved in the research.
The Aalto team was able to reduce noise down to 20 quanta, but that isn’t a clear demonstration that a mechanical resonator can reduce noise to the unavoidable quantum limit, says Ghannouchi. “They’re replacing what is electronic with mechanics, and you don’t get flicker noise with mechanics,” he says. “We anticipate that we can therefore measure low-level signals, but I’m not seeing any clear demonstration of that yet.”