The manganese bath is the primary standard for neutron emission rate at NPL
It measures the number of neutrons per second emitted by sealed radionuclide neutron sources such as 241Am-Be and 252Cf. The sources can then be used to calibrate neutron sensitive devices such as area survey instruments and personal dosemeters.
The neutron source is placed inside a large spherical bath containing almost 500 litres of manganese sulphate solution. The manganese nuclei capture neutrons to form an unstable isotope of manganese (56Mn) which decays with the emission of a gamma ray. By measuring the gamma radiation it is possible to determine accurately the neutron emission rate of the source. Corrections are made for neutrons escaping from the bath and for those captured by other nuclei in the solution and in the source mounting assembly.
The NPL manganese bath is one of about 10 such facilities worldwide and is arguably the most advanced and most extensively used. A neutron source calibration service is offered to customers from around the world many of which are themselves national standards laboratories. International comparisons of neutron source measurements are held regularly and NPL has always demonstrated excellent agreement with other laboratories.
When using a radionuclide neutron source to calibrate an instrument or a dosemeter, in addition to knowing the neutron emission rate of the source, the degree to which the source emission is not isotropic, or the same in all directions, also has to be known. As sources are nearly always in cylindrical rather than spherical capsules there is usually significant deviation from isotropic emission. This is called the anisotropy of the source and NPL has the capability to measure it. This is offered as a measurement service to outside customers.
With the cylindrical axis of the source horizontal, measurements are made in a low scatter area at 10° intervals around the source. This enables the relative emission from the waist of the source to be determined which can be several percent higher than if the source was isotropic. Source capsules have also been modelled using the Monte Carlo codes MCNP and MCBEND to determine the anisotropy. Agreement with measured values is generally very good except for a small number of sources where there is uncertainty as to the exact location of the active material within the encapsulation.