UC Berkeley

Atomic parity violation has been observed in the 408 nm 1S0--3D1 forbidden transition of ytterbium. The parity violating amplitude is 8.7(1.4)e-10 ea0, two orders of magnitude larger than in cesium, where the most precise experiments to date have been performed. This is in accordance with theoretical predictions and constitutes the largest atomic parity violating amplitude yet observed. This also opens the way to future measurements of neutron skins and anapole moments by comparing parity-violating amplitudes for various isotopes and hyperfine components of the transition.

We present a detailed description of the observation. Linearly polarized 408 nm light interacts with ytterbium atoms in crossed electric (E) and magnetic fields (B). The probability of the 1S0--3D1 transition contains a parity-violating term, proportional to E'B[(E'xE B], arising from interference between the amplitudes of transitions induced by the electroweak interaction and the Stark effect ((E' is the optical electric field). The transition probability is detected by measuring the population of the metastable 3P0 state, to which 65% of the atoms excited to the 3D1 state spontaneously decay. The population of the 3P0 state is determined by resonantly exciting the atoms with 649 nm light to the 3S1 state and collecting the fluorescence resulting from its decay. Systematic corrections due to imperfections in the applied electric and magnetic fields are determined in auxiliary experiments. The statistical uncertainty is dominated by parasitic frequency excursions of the 408-nm excitation light due to imperfect stabilization of the optical reference with respect to the atomic resonance. The present uncertainties are 9% statistical and 8% systematic. Methods of improving the accuracy for the future experiments are discussed.

We further present a measurement of the dynamic scalar and tensor polarizabilities of ytterbium's 3D1 state. The polarizabilities were measured by analyzing the spectral lineshape of the 1S0--3D1 transition. Due to the interaction of atoms with the standing wave, the lineshape has a characteristic polarizability-dependent distortion. A theoretical model was used to simulate the lineshape and determine a combination of the polarizabilities of the ground and excited states by fitting the model to experimental data. This combination was measured with a 13% uncertainty, only 3% of which is due to uncertainty in the simulation and fitting procedure. By comparing two different combinations of polarizabilities, the scalar and tensor polarizabilities of the state $3D1$ were measured to be 0.151(36) Hz/(V/cm)^2 and -0.205(53) Hz/(V/cm)^2, respectively. We show that this technique can be applied to similar atomic systems.

Finally, we propose two methods for improving future measurements of atomic parity violation using two-photon transitions. The first method is characterized by the absence of static external electric and magnetic fields. Such measurements can be achieved by observing the interference of parity-conserving and parity-violating two-photon transition amplitudes between energy eigenstates of zero electronic angular momentum. General expressions for induced two-photon transition amplitudes are derived. The two-photon scheme using the 1S0--1P1--3P0 transition in ytterbium (399 nm and 1280 nm) is proposed as a crosscheck of the APV experiment which uses the single-photon 408 nm 1S0--3D1 transition. We estimate that the signal-to-noise ratio of the proposed experiment is comparable to that achieved in the 408 nm system.

The second method allows for measurement of nuclear spin dependent atomic parity violation without nuclear spin independent background. Such measurements can be achieved by driving parity violating two-photon J=0--1 transitions driven by identical photons in the presence of an external static magnetic field. We discuss two promising applications: the 462 nm 5s^2 1S0--5s9p 1P1 transition in strontium-87, and the 741 nm 7s^2 1S0--7s7p 3P1 transition in unstable radium-225.