Silicon pixel detectors for synchrotron applications

Stewart, Graeme Douglas (2013) Silicon pixel detectors for synchrotron applications. PhD thesis, University of Glasgow.

Abstract

Recent advances in particle accelerators have increased the demands being placed on detectors. Novel detector designs are being implemented in many different areas including, for
example, high luminosity experiments at the LHC or at next generation synchrotrons. The purpose of this thesis was to characterise some of these novel detectors.

The first of the new detector types is called a 3D detector. This design was first proposed by Parker, Kenney and Segal (1997). In this design, doped electrodes are created that extend through the silicon substrate. When compared to a traditional photodiode with electrodes
on the opposing surfaces, the 3D design can combine a reasonable detector thickness with a small electrode spacing resulting in fast charge collection and limited charge sharing.
The small electrode spacing leads to the detectors having lower depletion voltages. This, combined with the fast collection time, makes 3D detectors a candidate for radiation hard applications. These applications include the upgrades to the Large Hadron Collider (LHC) leading to the High Luminosity Large Hadron Collider (HL-LHC). The limited charge sharing
of the devices can also improve their performance when being employed as imaging sensors. This will provide benefits in X-ray diffraction experiments.

The first experiment to evaluate the 3D detector design analysed for this thesis involved utilising a telescope consisting of 6 calibrated detector planes and a beam of pions from the Super Proton Synchrotron (SPS) at CERN. Once the tracks through the telescope were reconstructed, these gave predicted hits on the 3D detector that could be compared to the recorded energy depositions. By making this comparison, a measure of the detector’s efficiency
in various regions of the pixels was made. The overall efficieny of the pixel was measured at 93.0±0.5%. The detector was also rotated with respect to the incident beam,
increasing the efficiency to 99.8±0.5% for an angle of 10◦, and the detector bias was altered to measure the effect of over-depletion. Measurements of the charge sharing and resolution properties of the device were also reported.
Another detector design that was investigated was a slim edge detector. Instead of the typical guard ring structures that a normal device would employ, this detector reduced the size of these structures to enable easier tiling of the detectors. This was done by scanning the reduced edge and the standard edge of the detector with an X-ray beam with a width FWHM of 7 μm and 15 keV. The noise level of the strip closest to the cleaved edge was twice as large as that of the adjacent strip with no degradation of the charge collection capacity.

The next experiment to evaluate a short, double sided 3D strip detector was a Transient Current Technique (TCT) experiment. The TCT technique allows the electric field in the 3D devices to be probed in a way not possible before. The TCT technique uses the current waveform produced by the detector in response to a near delta function point laser pulse (illumination). The waveforms are recorded as a function of illumination position over the
surface of the device under test as a function of detector bias. This data gives information on the portion of the induced signal from electron or hole motion. From the rise times of the signals the velocity profile of the carriers in the devices and therefore electric fields can be determined. The collected charge was calculated from the
integral of the waveforms.

The detectors were tested prior to irradiation, after irradiating to a dose of 5 x 10^15 1MeV equivalent neutrons/cm^2, and after periods of annealing at elevated temperatures. Annealing was achieved in situ by warming to 60 ◦C for 20 to 600 minutes corresponding to room temperature annealing of between 8 and 200 days. Before irradiation, full lateral depletion between the columns occurs at low bias voltages, at approximately 3 V. A uniform carrier velocity between the columns is not achieved until the bias is equal to 40 V. Both the drift of electrons and holes provide equal contributions to the measured signals. After irradiation there is clear charge multiplication enhancement along the line between columns with a very non-uniform velocity profile in the unit cell of
the device. In addition, charge trapping greatly suppresses the contribution of the holes on the signal produced.

The final novel detector type was an Active Pixel Sensor (APS). Recent developments in CMOS fabrication processes have allowed new sensors to be developed and tailor-made for
specific applications. These challenge traditional Charge Coupled Devices (CCD) in some areas. The characterisation of the APS device took place in an X-ray diffraction experiment at the Diamond Light Source where it was evaluated alongside a CCD. The camera gain and stability had been determined prior to the experiment taking place. During the experiment, the dark current, noise, signal to noise and image lag performance was evaluated and
compared between the APS and the CCD. The signal to noise of the APS and the CCD was comparable (150 and 200 respectively) when the same integration time was used.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Silicon detector, pixel detector, radiation hard, particle detector, TCT, APS, 3D, Planar, Slim Edge
• Subjects: Q Science > QC Physics
• Colleges/Schools: College of Science and Engineering > School of Physics and Astronomy
• Supervisor's Name: Eklund, Dr. Lars
• Date of Award: 2013
• Depositing User: Mr Graeme Stewart
• Unique ID: glathesis:2013-4384
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 13 Jun 2013 13:34
• Last Modified: 13 Jun 2013 13:35
• URI: https://theses.gla.ac.uk/id/eprint/4384