Possible theses topics
In our group we develop detectors for a mobile, compact neutron tomography facility using fast neutrons (up to 14 MeV). Detectors optimized for efficient detection of fast neutrons are needed. One challenge is to distinguish signals from neutrons and gamma rays, a general difficulty in detecting fast neutrons. If this is successful, however, tomography can be run simultaneously with neutrons and gammas, providing more information about the object under study. The distinction is made by a detailed analysis of the measured pulse shapes, the so-called Pulse Shape Discrimination.
Two detector prototypes have been built so far in our group: a single pixel detector and a 16 pixel neutron camera (pictures). Our prototypes rely on the Stilben organic scintillator (picture), which enables pulse shape discrimination. This scintillator has the special feature that neutron interactions lead to signals with a longer falling edge than, for example, gamma interactions. Thus, neutrons can be clearly identified (picture). For identification, one plots the ratio of the pulse integral over the back part of the pulse to the total pulse integral (so-called PSD variable) against the total pulse integral. Two separate bands are formed, with the upper one assigned to neutron signals and the lower one to gamma signals. The scintillation light is detected with silicon photomultipliers (SiPM) (picture). One picture shows the current test stand, which allows tomographies to be taken (picture).
16-pixel camera. The Stilbene pins are separated by a plastic grid and are connected to an SiPM array via an optical coupling.
Theses topics proposals (Bachelor and Master)
All Bachelor topics (B) can be worked on in teams of two. Master topics (M) are topics for one person. If you have any questions, please do not hesitate to contact me.
A) Quantitative comparison of tomography measurements with simulations (B)
A large number of measurements have already been recorded and simulations performed. They are available in this bachelor thesis and a quantitative comparison between measurement and simulation of identical objects shall be performed. For this purpose different algorithms for image comparison shall be used.
B) Tomography simulations with Geant4 (B)
Existing Geant4 simulations can be extended to be used for neutron and gamma tomography. Different effects on the achievable tomographic image quality are currently investigated in a master thesis, e.g. source geometry and collimation, sample position, detector geometry (pixel size, pixel number). In this bachelor thesis a dynamic transportation and simultaneous measurement (conveyer belt arrangement) will be studied. The Geant4 simulation results will be used to improve an experimental tomography setup. Since Geant4 is C++ based, previous knowledge is recommended.
C) Stepper motor control for neutron tomography (B/M)
In our laboratory we have set up a test stand for neutron and gamma tomography, with which a sample can be moved and rotated between our AmBe neutron source and our detector. The sample can be moved translationally via a stepper motor driven XY-table, a rotation of the sample is realized via another stepper motor (see figure above). The stepper motors are currently controlled via an Arduino. Within the scope of a bachelor thesis this test stand shall be revised and improved. A new concept for the control of the stepper motors (x, y and rotation axis) and for the monitoring of the three coordinates is to be developed. For this purpose the existing system has to be learned and characterized. Afterwards, the new system is to be built and put into operation based on an Arduino. In the context of a master thesis, tomographic measurements of various objects can then be carried out with the setup.
D) New detector geometries (B/M)
Before new, larger prototype detectors are built, the detector geometry and coupling of the scintillator to the silicon photomultipliers must be optimized in simulations. One idea is to couple a simpler, homogeneous scintillator to silicon photomultipliers instead of a pixelated detector and use the weighted light signals for location information. This idea will be simulated in the final thesis. Since Geant4 is C++ based, previous knowledge is recommended.
E) Incorporate Prompt Gamma Neutron Activation Analysis (PGNNA) into the simulation (B/M)
To characterize unknown objects, tomography with fast neutrons and photons can be coupled with Prompt Gamma Neutron Activation Analysis. The unknown object is activated by the fast neutrons and a spectrum characteristic of the material is emitted from the object. In this thesis the existing simulation will be extended by the PGNAA. Since Geant4 is C++ based, previous knowledge is recommended.