Latest publications 

  • Search for Axionlike Dark Matter through Nuclear Spin Precession in Electric and Magnetic Fields
    C. Abel et al., Phys. Rev. X7, 041034 (2017).
    Magnetic field uniformity in neutron electric dipole moment experiments
    C. Abel et al., to appear in Phys. Rev. A. arXiv:1811.06085
  • C2D8: An eight channel CCD readout electronics dedicated to low energy neutron detection
    O. Bourrion et al., Nucl. Instrum. and Meth. A880 (2018) 28-34. arXiv:1710.08786
    Boron-10 conversion layer for ultra-cold neutron detection
    B. Clément et al., submitted to JINST arXiv:1902.09232

Past publications











  • Quantum states of neutrons in the earth's gravitational field: State of the art, applications, perspectives
    V. Nesvizhevsky et K. Protasov, Trends in Quantum Gravity Research (2005) 65-107.
  • Comment on: "UCN anomalous losses and the UCN capture cross section on material defects" (Phys. Lett. A 335 (2005) 327)
    A. Barabanov et K. Protasov, Physics Letters A 346 (2005) 378-380
  • Study of the neutron quantum states in the gravity field
    V. Nesvizhevsky et al, European Physical Journal C 40 (2005) 479-491.



Warning The english content of this web site is currently under construction



Measuring magnetic resonant transitions between neutron quantum states in the gravity field.




Bouncing neutrons

Ultracold neutrons (with a velocity smaller than 30 km/h) can bounce perfectly at many surfaces, such as a glass mirror. Contrary to a bouncing ball, neutron bounces are not damped.




Quantum music of bouncing neutrons

 A bouncing neutron behaves according to the strange rules of quantum mechanics. The frequency of the bounces is quantized, as a guitar chord that vibrates at certain well defined frequencies. This phenomenon occurs for neutrons bouncing very close to the mirror, typically 0.01 mm. As for electrons in atoms, bouncing neutrons have a spectrum of discrete energy states. The transitions between quantum states occur for given specific frequencies. For atoms the quantum frequencies are in the range of hundreds of Tera Hertz (these are the emission or absorption spectral lines), for bouncing neutrons the quantum frequencies lie in the audio range  (254 Hz, 462 Hz et 645 Hz for the transitions 1->2, 1->3 et 1->4 respectively). These frequencies are displayed on the stave below.




The GRANIT experiment at the l'ILL

The pioneering experiments with bouncing neutrons (2000 - 2005) at the Institut Laue Langevin (ILL) have shown the existence of the quantum states. The GRANIT experiment (GRAvitational Neutron Induced Transitions) is a follow up of these early experiments. The goal is to measure precisely transition frequencies between quantum states.

The GRANIT instrument at ILL.


For the purpose of the instrument, a dedicated ultracold neutron source has been built.

A bright cold neutron (CN) beam from the ILL reactor is directed towards a volume filled with 10 L of superfluid Helium cooled down to 0.8 K. A fraction of the beam will basically stop in this medium, thus producing ultracold neutrons.


Ultracold neutrons produced in the source are guided to the GRANIT spectrometer, installed in a clean room (to manipulate the delicate mirrors).



View of the clean room containing GRANIT.The stone on the foreground is now installed inside the vacuum chamber, it serves to support the mirrors. The coils around the chamber can induce a constant magnetic field in any direction.

Searching for new physics with GRANIT

The quantum frequencies to be measured with GRANIT can be calculated from the quantum theory. Therefore, GRANIT allows to test the validity of the standard theory for describing a subatomic bouncing particle. If the measured frequencies departed from the calculated ones, this would be a hint towards new phenomena.

In particular, the mysterious Dark Energy, responsible for the accelerated expansion of the universe, could be due to a new force called the "chameleon". According to this theory, only subatomic particles like the neutron are sensitive to the chameleon force (if it is strongly coupled). If this new force exists, it should modify the quantum frequencies of the bouncing neutron.

Logo nED?

Search for new sources of CP violation via the measurement of the neutron electric dipole moment (nEDM).



Physics motivations

As shown in the figure below, the existence of a nonzero electric dipole moment for the neutron - as for any non-degenerate particle - implies the breaking of two discrete symmetries: the parity P and the time reversal T. In accordance to the CPT theorem, the latter is equivalent to a violation of the combined symmetry CP, C being the charge conjugation operation. This simple observation has deep consequences, connecting cosmology and particle physics. In an attempt to understand the origin of the disappearance of antimatter in the Universe, known as BAU (Baryon Asymmetry of the Universe), Sakharov has indeed shown that three conditions are necessary, one of them being the existence of CP symmetry breaking mechanisms. In the standard model of particle physics (SM), a single source of CP breaking (the δ phase of the CKM matrix) exists, but its intensity is too low to explain the BAU. In contrast, the SM extensions generically predict the existence of new phases hence new sources of CP violation.


A neutron with an electric dipole moment (assumed to be parallel to the magnetic moment in the figure) is transformed under both parity P and time reversal T symmetries in a twin neutron, for which the two dipole moments are inverted with respect to the initial situation. The existence of such a state is excluded since it would radically alter the known properties of atomic nuclei (Pauli principle). This assertion, in the hypothesis of a non-zero EDM of the neutron implies the non invariance of physical laws under the P and T symmetries.

The current limit on the EDM of the neutron is 3 x 10-26e cm, i.e. 5 to 6 orders of magnitude above the SM prediction. The observation of a nonzero signal at 10-28-10-27e cm, goal of the next generation of experiments, would be the sign of a new physics and probably a decisive progress in our understanding of the disappearance of antimatter. On the contrary, pushing down the limit in the 10-28e cm range, would make the electroweak baryogenesis scenario, where the BAU is created at the electroweak transition, highly unlikely.

Measurement principle

The neutron EDM measurement is based on the analysis of the precession frequency of neutrons submitted to magnetic and electric fields either parallel or antiparallel. For both configurations, the precession frequency ν is expressed as

hν = 2 μn B ± 2 dn E

where μn and dnare the magnetic and electric dipole moments of the neutron and h is the Planck's constant. The difference provides a direct access to the neutron EDM:

dn = h δν /4 E

To measure the precession frequency of the neutron, we use the Ramsey's technique of separated oscillatory fields, invented in 1951 for the first measurement. The main experimental challenge is to control the temporal stability of the magnetic fied (dB/B ~ 10-7 over 100 s) and to optimize its homogeneity over a large volume (dB/B ~ 10-5 in a 20 l volume). Current experiments use ultracold neutrons (UCN) whose long storage times allow to maximize the measurement sensitivity.

The PSI UCN source

The source of the Paul Scherrer Institute in Switzerland is a spallation source based on the high intensity proton beam (590 MeV, current > 2 mA) and a solid deuterium converter to generate ultracold neutrons. UCNSourceThe following figure shows the main components of the source. The neutrons are produced by spallation by steering the entire proton beam (power > 1.2 MW) on a lead source with a duty cycle of 1 % (typically 4 s every 400 s) to limit the heating of the deuterium. The neutrons are then thermalized in a 3.5 m3 pool of heavy water surrounding the deuterium and then transformed in cold neutron in the solid deuterium crystal, within a volume of 30 l maintained at a temperature close to 5 K. Some of these neutrons can then be converted into ultracold neutrons by interacting with the crystal (phonon generation mechanism) and then be guided to the storage volume through a 1.2 m high chimney, sitting just above the SD2 volume. The storage volume has two guides that can provide UCNs to two experimental areas, one of which is occupied by the nEDM experiment and the other one is used for tests.

The commissioning of the source took place in 2010 and 2011. The source now works with a good reliability but the UCN density obtained so far remains below the expected values.

For more information:

The experiment at PSI

We are currently using the spectrometer developed in the 90s by the RAL-Sussex-ILL collaboration, which holds the best limit on the neutron EDM, published in 2006. This device, moved from the ILL to PSI in 2009, operates to room temperature. The figure below shows the precession chamber, of a volume of 20 l and located in the center of the vacuum chamber. It consists of a hollow cylinder made of polystyrene (PS) and two disk-shaped electrodes. nEDMExperimentThe upper electrode is connected to a high voltage generator (HV ≈ 140 kV) to produce a vertical electric field, parallel or antiparallel to the magnetic field. The measurement of the magnetic field is provided by atomic magnetometers, respectively based on 199Hg and 133Cs. The former operates as a co-magnetometer: during the precession of the UCNs, an atomic vapor of 199Hg, polarized by optical pumping, is injected into the storage chamber and allows a measurement of the same space-time average of the magnetic field as seen by neutrons. Our collaboration has complemented this device with a set of external Cs magnetometers developed by the Fribourg group, and installed on both sides of the precession chamber. They allow a direct measurement of the vertical magnetic field gradient which is one of the major sources of systematic errors.

With this device and based on the current performance of the UCN source, we expect to approach a limit of 10-26 e cm by 2016.


 For more information:

The n2EDM project

Our collaboration works since a few years on the design of a new spectrometer, the n2EDM project. This apparatus will use the same operating principle as the existing one, namely a detector at room temperature combining a Hg co-magnetometer to external magnetometers (Cs and in addition 3He), and will be optimized for the PSI UCN source.chambre n2EDM One of the main improvements will be to use a double precession chamber; a preliminary layout is displayed on the figure. Such a scheme will allow a simultaneous measurement of both (parallel/antiparallel) configurations, and thus greatly reduce systematic errors related to the magnetic field. Our efforts were so far focused on the magnetic shielding whose performances directly impact the achievable limit and which represents by far the most expensive part of the apparatus. The delivery of this shielding at PSI is expected by the end of 2015 and the availability of the new spectrometer is planned for 2018.
Considering that the UCN density will remain at the present level, a limit close to 10-27 e cm should be reached after 5 years of data taking, around 2023.


The english content of this web site is currently under construction

The neutron electric dipole moment page is under construction.