The motion of a single atom in an optical trap has been controlled by the use of a fast feedback logic loop developed at Max Planck Institute of Quantum Optics by scientists who also showed that the detection of just a few photons is all that is needed to keep the atom under control. The system developed by the team, which was led by professor Gerhard Rempe, Max Planck director and head of the Quantum Dynamics Div., reacts in real time to the motion of a single atom orbiting in an optical cavity. Artist’s view of the feedback scheme: A single atom placed between two highly reflecting mirrors reveals information about its position by emitting single photons (yellow wave packets). These photons are converted into digital electrical pulses (yellow spheres) and processed in an electronic feedback circuit. The circuit emits an electric current (blue spheres), which alters the intensity of a blue laser (blue valley). This feedback loop swings the atom, depending on its measured position. (Image: MPQ-Quantum Dynamics) Individual photons, emitted by the atom and carrying information about the atomic position, trigger a feedback mechanism that pushes the atom in a direction determined by the researcher, enabling control of the motion of the atom. It increases the time the atom spends in the cavity by a factor of four, but the ultimate goal is to control an atom’s trajectory with a precision limited only by Heisenberg’s uncertainty principle. In their experiment, the researchers first used laser light to cool a dilute cloud of neutral rubidium atoms down to a temperature of a few microkelvin. The cold cloud was then launched toward a high-finesse optical resonator made of two highly reflecting mirrors separated by about one-tenth of a millimeter. Upon arrival in the resonator, one single atom is captured by suddenly turning on optical tweezers in the form of a focused standing laser lightwave reflected back and forth between the mirrors. Because the trapped atom is sensitive to a variety of forces, its motion consists of a regular oscillation around the resonator axis superimposed by a strong erratic motion in all directions. This makes the trajectory of the atom unpredictable on timescales not much longer than the oscillation period, typically less than one-thousandth of a second. Besides the laser field that formed the optical trap, the system has two other lasers: a “probe,” which detects the presence of the atom, and an “actuator,” which either gives the atom a push or leaves it alone, depending on whether the atom is moving away from or toward the center of the trap. When switched on, the actuator laser creates a doughnut-shaped repulsive field that pushes the atom toward the cavity’s center. If an atom is placed in the center of the resonator, the light is blocked, and the photon flux drops to rates as low as 0.03 photons per millionth of a second. When the atom moves away from the resonator axis, trying to leave the resonator, more light is transmitted, making the position of the atom encoded in the intensity of the transmitted light. To make this information readable, the photons leaving the resonator are registered by a sensitive detector for two consecutive time intervals of equal duration, the so-called exposure time. If more photons are detected in the second time interval than in the first, the conclusion is that the atom is trying to escape the resonator. To prevent this, the light intensity of the optical tweezers is ramped up, pushing the atom back to the resonator axis. In case fewer photons are detected in the second interval, the atom is assumed to approach the cavity axis and the power of the optical tweezers is lowered. This reduces the energy of the atom and leads to its efficient cooling. The atom also can be heated by inverting the feedback logic, which rapidly expels the atom out of the resonator. “It is important to note that the feedback is triggered by each detected photon. If the number of detected photons goes up from 0 to 1, the intensity of the optical tweezers is ramped up almost immediately, in a time interval that is 70 times shorter than the oscillation period of the atom,” said Alexander Kubanek, a doctoral student in the Quantum Dynamics Div. “Actually, we have to pay attention that the exposure time is neither too short nor too long. For very short times, the information about the position of the atom is insufficient to trigger the desired feedback. If, on the other hand, the exposure times are too long, the feedback is delayed, leading to a reaction out of phase with the atomic oscillatory motion. So we have to choose exposure times that are long enough to give information on the position of the atom but are yet much shorter than the oscillation period of the atom in the optical tweezers,” he said. The feedback mechanism increases the storage time for a single atom from about 6 ms without feedback to 24 ms with feedback. Longer storage times exceeding 250 ms are achieved by a more sophisticated technique. But more important than the mere prolongation of the storage times are the quantum mechanical implications of the experiment. “It proves that reliable position information can be obtained by quasi-continuous measurements,” Rempe said. “In the future, this might allow us to steer an individual quantum trajectory with a precision ultimately determined by Heisenberg’s uncertainty relation or even protect the quantum state of a trapped particle against the disastrous influence of fluctuations stemming from the atom’s environment.” The work, “Photon-by-photon feedback Control of a Single-atom Trajectory,” was described in the Dec. 17 edition of the journal Nature. For more information, visit: www.mpq.mpg.de.