First non-zero measurement of a nuclear electric dipole moment

TL;DR

This work introduces a novel method to probe CP-odd nuclear electric dipole moments via an electrization field that emerges in superconductors when nuclei possess EDMs. By relating the long-range electrization field to nuclear magnetization through Maxwell’s equations and Wigner-Eckart relations, the authors design a Ta/Pb experiment that detects an EDM-driven electromotive force ${\cal E}_L$ as a current in a superconducting wire, with the signal scaling according to Curie’s law and the element’s free magnetization. An extensive set of fixed-temperature and zero-field measurements with Ta (signal) and Pb (control) yields a non-zero Ta EDM: $|d_e^{Ta}|=(3.39\pm0.31_{stat})\times 10^{-32}\,e\cdot\text{cm}$ (99.985% CL), and an upper bound on $|d_e^{Pb}|\lesssim 1.2\times 10^{-31}\,e\cdot\text{cm}$ at 95% CL. The approach promises a large, high-sensitivity nuclear EDM dataset that can constrain CP-odd EFTs, test electron EDM scenarios in superconducting regimes, and potentially inform axion-related CP-violation searches. The method leverages energy conservation in the superconducting loop and precise SQUID readout to achieve sensitivity far beyond conventional techniques, highlighting a path toward extensive CP-odd parameter constraints and novel dark matter probes. All observed effects are shown to be intrinsic to nuclear EDMs rather than external systematics, establishing a new paradigm for EDM studies.

Abstract

This paper reports the first non-zero measurement of a nuclear electric dipole moment using a novel method based on the rate of change of a supercurrent first proposed in 2016~\cite{https://doi.org/10.48550/arxiv.1604.02152} and fleshed out in this current paper. The theory, experimental concept and implementation are described in detail. The non-zero nuclear electric dipole moment measured with over 1000 hours of data was that of $^{181}$Ta producing a best value $|d_\text{e}^\text{Ta}|=(3.39\pm0.31_\text{stat})\cdot10^{-32}e\cdot\text{cm}$ and $|d_\text{e}^\text{Ta}|=(3.39\pm3.18)\cdot10^{-32}e\cdot\text{cm}>0$ at 99.985\%CL. There is an uncertainty on the value of overall multiplicative parameters such as the self-inductance of the superconducting circuit ($\pm4\%$), the mutual inductance between the SQUID pickup coil and the sample wire ($\pm15\%$), and the magnitude of the solenoid current ($\pm5\%$). An upper-limit was estimated for the control element, $^{207}$Pb, $|d_\text{e}^\text{Pb}|\lesssim1.2\cdot10^{-31}e\cdot\text{cm}$ at 95\%CL.

Figures (20)

• Figure 1: Bulk and atomic parameters for sample elements where $a_\text{b}$ is the natural abundance of the isotope and "P" refers to the purity of the metal purchased for the experiment, Ta and Pb. $N_\text{N}$ is given in units of $10^{28} \text{m}^{-3}$. Aluminum was used to build the booster solenoid, but molybdenum may be a good alternative with a much smaller nuclear magnetization but with similar critical magnetic field and temperature; note that $_{42}^{95}\text{Mo}$ and $_{42}^{97}\text{Mo}$ have nearly identical parameters and were combined in a single entry. The critical field of the superconductor is given as it sets an upper-limit on $I_\text{sol}$. Vanadium has the largest free magnetization of all elements that are superconducting in their pure form while lithium has the smallest number of nucleons. Niobium is a widely used in superconductivity applications and research. Pure carbon is not superconducting but graphite intercalated compounds (GICs) are and provide an experimental pathway to measuring the nuclear EDM of other elements that are not superconducting in their pure form like potassium. In a GIC, only 1 out of 4 valence "carbon electrons" are in the conduction band and $_{6}^{13}\text{C}$ is the only stable isotope with a non-zero spin. Similarly, Iron is not superconducting, but in iron-based $p$-wave superconductors, it can provide a pathway to measure the electron EDM.
• Figure 2: Sample assembly components: A) sample wire (Ta or Pb); B) SQUID; C) sample solenoid; D) main solenoid wound around copper tubing (not shown); E) aluminum booster solenoid core (length 5 cm radius 0.5 cm); F) booster solenoid with two layers of winding where one lead forms the first winding layer and the other lead proceeds to form the sample solenoid and returns to form the winding of the second layer; G) BiPb joint of the NbTi booster solenoid wire leads. The currents are indicated with the red arrows: The current applied by the DC power source to the main solenoid, $I_\text{a}$; the induced booster solenoid current, $I_\text{sol}$; the sample current detected by the SQUID, $I_\text{s}$. The dashed rectangles represent shielding: the inner and outer rectangles around the SQUID are made of lead and Hy mu 80 respectively; the shielding around the main and sample solenoids are made of Hy mu 80; the overall shielding is made of lead and aluminum adhesive. The potential and current leads from the SQUID run to the exterior SQUID electronics while the main solenoid leads run to an exterior DC current source.
• Figure 3: (a) Bi56Pb44 joint ALT021 data showing the transition from the superconducting state to the normal state at $T$=8.9 K. (b) SQUID data for the tantalum sample wire showing the transition from the superconducting state to the normal state at $T$=4.5 K.
• Figure 4: PFL circuit detailing the LOCKED mode components in the SQUID electronics, reproduced with the permission of the manufacturer, Starcryo. Labels were added by the author.
• Figure 5: Full REGEN data for Ta at $I_\text{a}$=0.53 A. Each data point is a mean value of 1000 raw data points averaged over 1 second. $T_\text{max}$ is the temperature where the Meissner effect begins while $T_\text{min}$ is the temperature where it ends.