High-sensitivity measurement of ${}^3\mathrm{He}-{}^4\mathrm{He}$ isotopic ratios for ultracold neutron experiments

Research efforts ranging from studies of solid helium to searches for a neutron electric dipole moment require isotopically purified helium with a ratio of ${}^3\mathrm{He}$ to ${}^4\mathrm{He}$ at levels below that which can be measured using traditional mass spectroscopy techniques. We demonstrate an approach to such a measurement using accelerator mass spectroscopy, reaching the ${10}^{-14}$ level of sensitivity, several orders of magnitude more sensitive than other techniques. Measurements of ${}^3\mathrm{He}/{}^4\mathrm{He}$ in samples relevant to the measurement of the neutron lifetime indicate the need for substantial corrections. We also argue that there is a clear path forward to sensitivity increases of at least another order of magnitude.

Figure 1

Floor plan of the ATLAS Accelerator Facility. The ECR-1 ion source and an associated RF discharge source were used for ${}^3\mathrm{He}$ AMS and the beam was accelerated to 8 MeV in the “PII” and “Booster” linac sections. The ${}^3\mathrm{He}$ produced was detected, after transport, in the Enge split-pole spectrograph.

Figure 2

Physical orientation of the RF discharge source, Faraday cup, and the ECR source used in this work.

Figure 3

The mini-ECR source developed early in this work to reduce the ${}^3\mathrm{He}$ background seen in the main ECR ion source. The quartz tube is mounted on the extractor electrode for the ECR source and extends into the ECR cavity. At the bottom of the quartz tube holding flange is the 1 mm extractor hole and the offset gas feed for the helium gas.

Figure 4

The observed ${}^3\mathrm{He}/{}^4\mathrm{He}$ ratio observed with an ultrapure sample as a function of the ${}^4\mathrm{He}$ beam current from the source. The helium current is viewed as a surrogate to helium partial pressure in the source plasma. The asymptotic value of ${}^3\mathrm{He}/{}^4\mathrm{He}\approx 3\times {10}^{-13}$ observed at high ${}^4\mathrm{He}$ source current (for which source background becomes less significant) is consistent with that obtained using the current source configuration as described in Sec. 4b. The highest ratio data points were obtained with only oxygen flowing into the source. The ${}^4\mathrm{He}$ current for those data points was $<0.1 e\mathrm{nA}$.

Figure 5

RF discharge source in operation with a helium plasma. Gas flow from the helium sample enters from the left and ions are extracted though the electrode on the right. RF power is supplied using the coils visible to the center right of the source.

Figure 6

The new RF source plasma chamber constructed with materials having a low helium diffusivity. This source operates by inductive coupling from an external coil surrounding the chamber oscillating at approximately 80 MHz with RF power less than 100 W.

Figure 7

Schematic of the gas handling system used with the RF discharge source. Two computer-controlled leak vales allow simple switching between samples while the source remains in operation. The nitrogen supply and pump were used to pump and purge the manifold assembly.

Figure 8

${}^3\mathrm{He}$ energy loss in anode 4, dE4, of the focal plane detector. Events are required to be between channels 175 and 350.

Figure 9

Comparison of results. Good agreement is seen between the Argonne measurements, $▵$, and the calculated concentration of the reference samples, $\blacksquare$. Error bars show the combined standard uncertainty.

Figure 10

Count rate in spectrograph after running pure hydrogen in the source for roughly 45 minutes. Natural ${}^3\mathrm{He}$ backgrounds fall in the region from 175 to 350 in dE4