The surf is up

still remember that brief moment that revealed the astonishment of everyone in the audience. Andrew Cleland had just concluded his talk at the conference ‘Quantum Optics of Nanoand Micromechanical Systems’, which was held last July in Bad Honnef, Germany. He had taken us all by surprise. Andrew’s talk had begun as a review of recent results of his and John Martinis’ team — they had achieved an unprecedented degree of control over individual electromagnetic quanta (photons) in a microwave resonator by coupling the resonator to a superconducting two-state quantum system, or qubit. But the awe came with his last slide: in it, he showed that if the microwave resonator could possibly be replaced with a mechanical oscillator of similar resonance frequency, then the same qubit device could be used to attain quantum control over individual mechanical quanta (phonons). Just eight months later, in a paper published online in Nature today, Cleland, Martinis and their colleagues (O’Connell et al.) report that they have done just that. Over the past few years, impressive progress has been made in studying nanoand micromechanical resonators, and the common aim of exploring the quantum regime of mechanical systems has generated a thriving field that continues to attract an eclectic mix of researchers. O’Connell and colleagues’ results are a remarkable achievement because they involved overcoming two outstanding challenges in the field. The first is bringing the mechanical device reliably to its quantum ground state of motion, and the second is coupling it strongly to a different quantum system. With these challenges met, the future is ripe to use these systems both to test the principles of quantum mechanics and in applications such as quantum information processing. The difficulty in bringing a mechanical device to its quantum ground state lies in the environmental temperatures (T) needed. To suppress residual thermal phonons in the device requires T < hfm/kB, where fm is the device’s resonance frequency and h and kB are Planck’s and Boltzmann’s constants, respectively. Typical mechanical resonators involve the motion of the device’s centre of mass and have resonance frequencies smaller than hundreds of megahertz. The required groundstate temperatures of such devices are below those achievable with standard cryogenic refrigerators. One solution is to use additional cooling schemes analogous to the laser cooling of atoms. This technique has allowed the preparation of mechanical states of mean phonon occupation, n, close to the quantum ground state (n = 0), for both nanomechanical (n ≈ 4) and micromechanical (n ≈ 30) resonators. Another solution is simply to avoid the difficulty, as O’Connell et al. do. Their micromechanical resonator consists of a suspended slab that, owing to its piezoelectric nature (it changes its volume when subjected to an external electric field), can undergo oscillations in its thickness when the two metal electrodes between which it is sand wiched are subjected to a voltage. This acoustic vibration does not involve any centre-of-mass motion and hence allows the authors to achieve ultra-high mechanical frequencies of fm ~ 6 gigahertz, for which they could prepare their system’s ground state using a conventional dilution refrigerator: at temperatures of about 25 millikelvin they obtained n < 0.07. After meeting the first challenge, the authors were set to meet the second: to couple a resonator and another quantum system sufficiently strongly for quantum effects to be observed. Recent experiments have demonstrated mechanical coupling to qubits and strong coupling to optical cavities. However, decoherence mech anisms have thus far prevented quantum effects from being observed. In their experiment, O’Connell et al. overcome this problem. Their mechanical resonator is connected via a capacitor to a Josephson phase qubit, which consists of two superconductors coupled by an insulating (Josephson) junction. The qubit’s ground and excited states represent the two lowest energy states of the superconductors’ wavefunction phase difference across the tunnel junction. By using the piezoelectric nature of the resonator, the electromagnetic energy of the qubit can be converted into the mechanical energy of the resonator, or vice versa. In principle, this interaction allows the authors to coherently transfer an arbitrary state of the qubit to the resonator, and even to generate entanglement — a quantum effect in which the states of the qubit and the resonator are linked together in an inseparable way. To demonstrate coherent quantum-state transfer, O’Connell et al. prepared the qubit in the excited state and switched on the interaction between the qubit and the resonator. They then measured the occupancy of the qubit’s excited state as a function of the interaction time. The observed oscillatory behaviour of this occupancy is a clear quantum effect and indicates reversible qubit–resonator exchange of a single quantum of energy. The typical transfer times for single quanta (the Rabi swap time) was 4 nano seconds, which is smaller than the energy decay times of 17 ns and 6 ns for the qubit and resonator, respectively. The authors observed a similar oscillatory behaviour in the excited-state occupancy when they transferred a qubit’s superposition state, one in which the system is in the ground and excited states at the same time, hence preparing a quantum superposition of the mechanical system. It is also worth noting that, after half the Rabi swap time, the authors’ transfer inter action should create an entangled state between the qubit and the mechanical resonator. They point out, however, that their current experimental performance excludes a direct test of entanglement. Although today it is routine to control the quantum-mechanical motion of individual atoms, controlling that of a nanoor micrometre-sized system is not. Quantum mechanics on such scales has been envisaged since the 1990s. With their experiment, O’Connell et al. have not only set foot firmly in this quantum regime but have also opened the door for quantum control of truly macroscopic mechanical devices. And the prospects are exciting. One QUANTUM MECHANICS