³He magnetometry for nEDM

To investigate and test fundamental physical theories, many collaborations worldwide search for electric dipole moments of molecules, atoms, and elementary particles.

The nEDM collaboration at the Paul Scherrer Institut, Swizerland is searching the electric dipole moment of the free neutron. The principle of this measurement is by means of observing the free precession of stored ultracold neutrons in a magnetically shielded room, applying the Ramsey-resonance technique.

For the measurement of the electric dipole moment of the free neutron it is important to know the exact value of the magnetic field inside the EDM spectrometer. This field can be measured in-situ by monitoring the spin-precession of polarized ³He, which is the contribution of Mainz to the nEDM experiment.

Like polarized alkali vapor, polarized ³He gas can be used for sensitive magnetometry. Large diameter cells have been shown to operate with polarization time constants of several thousand seconds. Since one measurement cycle of the nEDM apparatus lasts for ~ 200 s this would allow monitoring the magnetic flux integrated over the full neutron precession chamber volume over many measurement cycles of the nEDM apparatus.

For readout and analyzis of the free ³He precession frequency, laser-pumped cesium magnetometers, which reached a sensitivity of ~ 20 fT/√Hz at the group of Antoine Weiss, university of Fribourg, Switzerland, are used. The expected signal amplitude from the precessing ³He spins is 10 pT – 40 pT, resulting in a signal-to-noise ratio (SNR) of up to 2000:1 in a bandwidth of 1 Hz.

The layout of the ³He magnetometer is sketched in following figure:




 Fig.1: Setup of n2EDM with external ³He polarizer


Two flat cylindrical magnetometer vessels (top, bottom) cover the same magnetic flux as the UCN double-chamber. Hence, the average value of frequency measurements ωHe = t,He + ωb,He )/2 gives the normalization signal for the UCN free precession frequency, whereas the frequency difference determines the magnetic field gradient ∂B/∂z = (ωt,He - ωb,He )/(𝛾He•Δz) to a high precision, with Δz being the distance between the upper and lower magnetometers and 𝛾He gyromagnetic ratio of ³He. 

In a first feasibility test of a ³He magnetometer at PSI in 2009, a true signal of ³He magnetization between 10 pT and 20 pT inside the magnetometer vessels we proved by filling polarized ³He from outside to inside the nEDM apparatus.

Now our task is to develope an automatic machine to polarize and fill the ³He inside the nEDM chamber.

The polarizer unit

A polarizer unit located outside the nEDM apparatus will polarize ³He. A solenoid placed around the polarizer will generate a magnetic field of several Gauss, and a mu-metal shield covering the solenoid structure will homogenize the field and reduce the influence from stray fields.

The polarizer unit works in the following way: The laser beam of a fiber laser with a wavelength of 1083 nm passes through an arrangement of several mirrors and beam-splitter cubes and polarizes ³He in two glass tubes (Fig.2) with a combined volume of 6 liters at a pressure of 1 mbar. In a second step, the polarized ³He is compressed by means of a non-magnetic piston compressor into a small storage cell of 0.3 liters volume. By repeating this process, it is possible to reach 40 mbar·l of polarized gas inside the storage cell.

The polarized ³He then can be transferred to the two flat magnetometer vessels by guiding it through a tube inside a magnetic guiding field . Assuming a volume of 20 liters, a ³He pressure of 2 mbar will be reached inside. A 𝜋/2 spin flip pulse then will trigger the ³He spin-precession. In the meantime, the polarizer unit will prepare a new batch of polarized gas.

Fig.3 and Fig.4 show a sketch and the current status of the apparatus. The polarization equipment was completed by the end of 2011.


 Fig.2:  Ultra-compact ³He polarizer with gas discharge in the two optical pumping cells



 Fig.3:  Ultra-compact ³He polarizer, 4 layer test shield and transfer line




  Fig.4:  Current status of the entire system (polarizer, transfer line and measurement shield with Cs magnetometer inside)


³He and Cs Magnetometer test facility

A complete setup of ³He and Cs magnetometers is to be tested in Mainz. For this purpose a lamp-pumped Cs magnetometer (Fig.5) was transferred from university of Fribourg in September 2010 to Mainz and a four-layer mu-metal shield from PSI arrived at Mainz in November 2010. With these components we constructed a test facility in order to investigate future ³He magnetometer arrangements for the nEDM setup.


 Fig.5: Lamp pumped Cs magnetometer from Fribourg (group of A. Weis)

In recent measurements inside the shield, a sensitivity of less than 0.3 pT  was reached using the cesium magnetometer. To polarize ³He in-situ in a field of 1 µT inside the shield, a coil system and setup was built in 2011. After a 𝜋/2 spin flip pulse it is now possible to monitor the ³He spin precession by the lamp-pumped Cs magnetometer.


  Fig.6: Experimental test setup

At the moment a SNR of 50 and a ³He coherent spin-precession time T2 of about 1 hour is reached with one magnetometer.

  Fig.7: One of the first measured ³He spin precession signals

In a joint measurement with our colleagues from Fribourg in November 2011 it was possible to reach a SNR of about 600 by using two laser-pumped Cs magnetometers, operating in gradiometer mode. We as well implemented a nice and efficient tool for testing ³He and Cs magnetometers both together.

The polarizer unit was finished in the end of 2011. At the beginning of 2012 it was shown that it is possible to reach a polarization up to 55% in the optical pumping cell and to compress this gas nearly without losses into the storage cell.



 Fig.8: Polarization curve in the optical pumping cell of polarizer unit

 At 23 January 2012 there was the first successful test to transport polarized ³He from the polarizer unit via the transfer line and to measure the free spin precession of ³He inside the test shield. With a polarization of ~ 50 % and a pressure of 10 mbar inside the storage cell a very clear signal was observed from magnetometer cell in the measurement setup:


  Fig.9: Cs Magnetometer signal before transfer of ³He


   Fig.10: Signal of real ³He spin precession after transfer hyperpolarized ³He into the measurement setup

It was successfully proven that all parts of ³He magnetometer for nEDM are working and main parts of the setup construction are completed.