 |
| Dr. Peter Böni |
 |
 |
Multi-channel neutron guide for the Spallation Neutron Source SNS in Oak Ridge National Laboratory (USA) as manufactured by SwissNeutronics. The glass plates are coated with supermirror m = 3.6. The casing is made of stainless steel.
|
|
 |
Magnetron sputtering system.
|
|
 |
Reflectivity of one of the best supermirror coatings with m = 3 produced so far by SwissNeutronics on a sputtering plant at PSI.
|
Supermirrors as Wave Guides for the Transport of Neutrons
SwissNeutronics was founded as a spin-off company of Paul Scherrer Institute (PSI), Villigen Switzerland, on April 23, 1999. Last October 1st, SwissNeutronics became a new Industry Partner of MaNEP. The five shareholders of the company are scientists and engineers with many years of experience in neutron scattering. Presently, SwissNeutronics has 7 full time employees at their premises in Klingnau. Besides a construction office and a machine shop, SwissNeutronics runs two large 5-axis CNC milling machines and a large DC-magnetron sputtering plant. Therefore, SwissNeutronics is in the unique position to offer products to the customers being exclusively designed, constructed, manufactured, coated, assembled, and aligned by SwissNeutronics staff. SwissNeutronics offers their products and research contracts to research laboratories and industry all over the world. In Japan and in the US distributors of SwissNeutronics commercialize the products.
A collaboration between SwissNeutronics and MaNEP
Similarly as in light optics, where the photons are guided by means of fibers it is possible to transport neutrons from the source to the neutron scattering instruments by means of neutron guides. They are usually composed of glass elements with a cross section of typically 60x170mm, having a total length of 30-100m. The inner surface of the glass is coated with a layer of Ni, reflecting the neutrons under an angle of for example 0.47° for neutrons with a wavelength l = 4.7Å. The transmission of neutron guides can be increased by the addition of a supermirror coating, i.e. sequences of artificial multilayers composed of Ni and Ti, diffracting neutrons at small angles.
Of course, a high reflectivity can only be achieved if the coating process of the Ni/Ti bi-layers is optimized such that the interface roughness and interdiffusion is minimized. Because the typical layer thickness can be as small as 40Å it is necessary to achieve an interface roughness that does not exceed 5-8Å. Using reactive sputtering we succeeded to achieve at three times the critical angle of Ni (1.41° for l = 4.7Å) a reflectivity R ~ 85%. However, the process leading to this excellent result was not reproducible and in average we only obtained R = 78%. Moreover, we faced problems with adhesion. In order to remain competitive in the field of supermirrors, SwissNeutronics has to achieve soon a reflectivity of approximately 90% near m = 3 in mass production as well as to increase the angle of total reflection to larger m (the factor m indicates the increase of the critical angle compared to natural Ni). This is only possible, if the interface roughness can be reduced to below 4Å.
The aim of the MaNEP project, "Roughness reduction in metallic and multilayered neutron supermirrors" in collaboration with the ETHZ and the University of Geneva, is to enhance the performance as well as the yield of supermirrors made of metallic multilayers that are produced by means of magnetron sputtering. We expect to achieve a considerable improvement of the performance of the mirrors and to speed up their production by novel production procedures, i.e. by introducing the concept of controlled interdiffusion at the interfaces. Of course, a systematic characterisation of the produced multilayers can only be performed in collaboration with research groups because SwissNeutronics does not have the full range of necessary equipment for investigating the products available. The structure and roughness of the multilayers will be characterized by means of neutron and X-ray reflectometry, which are techniques averaging over a large area, whereas scanning electron microscopy (SEM/EDX) and atomic force microscopy will be used for the structure and surface characterization on a local scale. Finally, we shall investigate the possibility of improving present day algorithms for the calculation of supermirror sequences by taking into account interdiffusion and roughness.
|
