The Silicon Vertex Detector for b-tagging at Belle II

Autoren: Manfred Valentan
Kategorie: PhD-Thesis

The Belle experiment at KEK (Tsukuba, Japan) was successfully operated from 1999 until 2010 and confirmed the theoretical predictions of CP violation. In order to increase the beam intensity, a major upgrade of the KEKB collider is foreseen until 2015. The final goal is to reach a luminosity of 8×10^35 cm^-2 s^-1, which is about 40 times higher than the previous peak value. This also implies changes to the Belle detector and its innermost tracking subdetector, the SVD, in particular. The SVD will be completely replaced, as it had already operated close to its limits in the past. All other subsystems will also be upgraded. This leads to the new Belle II experiment.


The aim of Belle II is to search for deviations from the Standard Model of particle physics by providing extremely precise measurements of rare particle decays, thus representing a complementary approach to the direct searches performed at high energy hadron colliders. The upgraded SuperKEKB machine will collide electrons and positrons at the center-of-mass energy of excited states of the Upsilon particle, which hereafter decays to a B meson and its anti-particle. The decay vertices of these mesons have to be precisely measured by the Belle II SVD, together with the PXD and the CDC. This allows the measurement of time-dependent, mixing-induced CP asymmetry. In addition, the SVD measures vertex information in other decay channels involving D meson and tau lepton decays.


Since the collision energy is quite low (around 10 GeV), the emerging particles have low momentum and are subject to strong multiple scattering when traversing material. Therefore, all sensors of the Belle II SVD have to be optimised in terms of material thickness, while preserving high signal yield and position measurement accuracy. This will be possible by the development of thin, double-sided silicon microstrip sensors.


This PhD thesis includes the physics motivation for building a high luminosity B factory and a high precision particle detector, and an introduction to the Belle II experiment, outlining purpose and working principle of the involved subdetectors. More details are given on the Belle II SVD, including mechanical structure, sensors, electrical readout and cooling. Furthermore, the basics of semiconductor physics and silicon processing are reviewed, and the principles of single-sided and double-sided silicon microstrip sensors are explained in detail.


The author?s main task was to develop a trapezoidal double-sided silicon microstrip sensor for the forward region of the Belle II SVD, from the initial CAD drawings to the production. He developed a software framework aiming at fast and flexible design of double-sided silicon microstrip sensors, both for rectangular and trapezoidal shapes. Using this framework, a whole wafer was equipped with a full-scale trapezoidal sensor, several test sensors for optimising the layout, and test structures. Several batches of prototype sensors were produced by Micron Semiconductor Ltd. in England, in close collaboration with the author.


The wafer contains small test sensors dedicated to investigating the strip insulation on the n-side, featuring the p-stop blocking method (in three geometry patterns: atoll, common and a combined variant) and of the p-spray blocking method. These sensors have been extensively tested by the author in particle beams and gamma irradiations, showing that the atoll p-stop pattern is best suited for application at Belle II.


The full-scale prototype sensors were thoroughly tested by the author in the semiconductor laboratory and in particle beams, long-term stability has been demonstrated by irradiation and thermal cycling campaigns. The knowledge gained by examining the test sensors and full-scale sensors led to an update of the design of the full-scale sensor. After production of another prototype batch the updated design was evaluated, compliance with the requirements of the Belle II SVD were shown, and the sensor layout was released for production. In the course of the sensor tests the author went to four beam tests at CERN, and performed the analysis of the data taken.