A permanent electric dipole moment (EDM) of the electron along its spin axis would violate parity and time reversal symmetries, and is of considerable interest in elementary particle physics. Because of the CPT theorem, which states that the combined symmetry operations of parity, time, and charge conjugation must be conserved in any Lorentz-invariant theory, violation of time symmetry implies violation of CP symmetry. The CP operator is effectively the operator that translates particles into anti-particles and vice versa. Therefore, any violation of CP symmetry would imply a difference in behavior of particles vs. antiparticles. Such a mismatch in behavior is required to explain the matter/anti-matter asymmetry of the universe.

The standard model of particle physics contains CP symmetry violation, but not enough to explain the current matter/anti-matter asymmetry of the universe. Moreover, the level of CP violation is so small that the standard model prediction for the electron EDM, d_e<10^-40 e·cm, is far too tiny to be detected by current methods. However many favored extensions of the standard model predict de within 3 orders of magnitude of the latest experimental limit d_e<1.08×10^-27 e·cm. With any improvements in sensitivity, some of these models may be excluded, or evidence of new physics beyond the standard model may be discovered.

We embed our electrons inside the polar molecule PbO, where they experience a huge effective intramolecular electric field Eeff ~ 26 GV/cm. Our experiment takes advantage of several unique properties of the metastable a(1)³Σ⁺ excited state of PbO, including closely spaced levels of opposite parity, long coherence times, and suitability for use in a vapor cell (enabling high counting rates). Over the past 12 years, the experiment has been tested and improved in a variety of different ways. Currently, the shot-noise limited sensitivity of the experiment is approximately 3×10^-27 e·cm/(day^1/2).

The experiment works by populating the a(1)³Σ⁺ state using a pulsed laser. Parallel electric and magnetic fields are applied to the molecules. The closely-spaced opposite parity levels are split by the Stark effect and form two sets of states – one with the dipole moment of the molecule aligned with the electric field and the other anti-aligned. The laser’s polarization aligns the electron spins, which then precess around the magnetic field. As they decay back to the ground state, the fluorescence signal exhibits quantum beats, which correspond to the precession frequencies. The EDM will affect the precession frequencies of the two sets of states in the opposite way, so that one will precess faster and one slower. By looking at the difference of the precession frequency between the two sets of states, the EDM can be measured.

Currently, the experiment is investigating many possible sources of systematic errors . We hope to get an understanding of these systematics and publish a result before the end of the summer of 2012. However, that will not be the end of the search for the electron EDM in the DeMille group. The next generation experiment, ACME, (which uses a cryogenic beam of ThO molecules rather than a high-temperature vapor cell) is already under way.

How strongly do electrons interact with the atomic nucleus via the weak force? For electromagnetism the answer is easy: the electron and nuclear charges determine the strength. However, the weak force is more complicated, having two effective types of charge for each particle. The strength of the weak force must be measured in parts.

The parity-violating parts of the weak force in an atom or molecule can be split into two groups: nuclear spin dependent effects (NSD-PV) and nuclear spin independent effects. The spin independent effects are much easier to measure since they grow proportionally with the number of nucleons, while NSD-PV effects arise only from the unpaired spins in a nucleus—of which there is typically only one. Consequently, nuclear spin independent parity violation is stronger and has been measured well while NSD-PV effects are weaker and largely unmeasured.

The two NSD-PV effects that we are principally interested in are Z boson exchange between electrons and nucleons, and the nuclear anapole moment. Z boson exchange is a fundamentally simpler process (it corresponds to a tree level Feynman diagram) and directly addresses the question of how strong the electron-nucleon weak force is. However, the strengths of these two effects are roughly the same order of magnitude and the contribution of the nuclear anapole moment cannot be distinguished from Z boson exchange in a measurement from a single isotope.

Although the nuclear anapole moment complicates the measurement of the strength of the weak force, it is interesting in its own right. Inside the nucleus, the weak force causes the spin of the unpaired nucleon to point in its direction of motion as it orbits the nuclear core. The magnetic moment associated with the nuclear spin is equivalent to a current loop; its orbit around the nuclear core results in an effectively toroidal current. This toroidal current gives rise to an anapole moment, analogous to how a current loop gives rise to a dipole moment. Atomic electrons then interact magnetically with the anapole moment. Measuring nuclear anapole moments will give valuable insight into the strength of the nucleon-nucleon weak force. Furthermore, the magnitude of the nuclear anapole moment is unique to each nucleus, so having a table of these values may be useful in a way analogous to that in which nuclear magnetic moment measurements have been useful to NMR.

Our novel approach to measuring these small effects uses diatomic molecules. The NSD-PV effects cause levels of opposite parity to be mixed, and it is this mixing that we directly measure in order to deduce the NSD-PV strengths. The mixing is very small, but is amplified if the levels are closely spaced in energy. Due to their large moment of inertia, diatomic molecules have rotational levels that are very closely spaced and are thus an ideal system for measuring NSD- PV. The levels are brought even closer together using a magnetic field and associated Zeeman shifts of the levels. Finally, the mixing is amplified through interference with an oscillating electric field, and detected using laser-induced fluorescence from the molecules.

Recently, we assembled a complex “interaction region” which will allow us to apply an electric field and laser beams to the molecules flying through the small confines of our superconducting magnet. We plan to soon make NSD-PV measurements in ¹³⁸BaF as a test of possible systematic errors in our approach; this system should not exhibit NSD-PV because ¹³⁸Ba does not have an unpaired nucleon spin. Later, after improvements to our molecular beam flux, we will measure NSD-PV in ¹³⁷BaF, where the predicted size of the effect is large enough to detect.

The tunable, anisotropic, long-range electric dipole-dipole interaction that exists between heteronuclear molecules is fundamentally different from the contact interaction that exists in atomic systems. In order to study this interaction, we are interested in creating and trapping a large sample of ultracold polar molecules. To achieve the ultracold temperatures necessary to observe novel behavior, we extend the mature techniques associated with atomic cooling to molecular systems.

We begin with cold (~100 μK), dense samples of atomic rubidium and cesium held in traditional magneto-optical traps. We illuminate these samples with a resonant laser beam that promotes colliding atomic pairs into an electronically-excited molecular state. By a judicious choice of photoassociation state, these molecules spontaneously decay directly to deeply-bound levels of the ground electronic state, including the vibrational ground state. Further, molecular selection rules constrain the available rotational levels such that we can continuously produce rovibronic (v=J=0) RbCs molecules.

As these molecules are created, we intend to accumulate them in an optical trap generated by an intense CO₂ laser. The optical trap also contains unwanted vibrationally and rotationally excited RbCs molecules that can inelastically scatter with the absolute ground state molecules. This process liberates enough energy to eject the molecules and thereby deplete the sample. In order to achieve long trap lifetimes, we will intentionally load a high density sample of cesium atoms into the trap. The atoms will selectively remove the excited molecules via inelastic scattering while leaving the absolute ground state molecules, which only scatter elastically, unaffected. A resonant laser beam can then be used to remove the atoms, leaving a pure absolute ground state molecular sample.

It has been roughly three decades since laser cooling techniques produced ultracold atoms, leading to rapid advances in a wide array of fields. Until recently, laser cooling had not been extended to molecules because of their complex internal structure, which precludes the realization of a true optical cycling transition. However, this complexity makes molecules potentially useful for a wide range of applications. For example, heteronuclear molecules possess permanent electric dipole moments that lead to long-range, tunable, anisotropic dipole–dipole interactions. The combination of the dipole–dipole interaction and the precise control over molecular degrees of freedom possible at ultracold temperatures makes ultracold molecules attractive candidates for use in quantum simulations of condensed-matter systems and in quantum computation. In addition, ultracold molecules could provide unique opportunities for studying chemical dynamics and for tests of fundamental symmetries.

Our group has demonstrated the first laser cooling of a diatomic molecule, in addition to deflection and, most recently, slowing of a molecular beam through radiative forces. Our current work is focussed on extending this technique to allow trapping of these molecules in a magneto-optical trap (MOT) or a microwave trap. This work is enabled by a scheme to create a quasi-cycling transition in strontium monofluoride (SrF), which allows up to ~10⁵ photon scattering events before the accessible molecular population decays by 1/e.