|CECCARELLI FRANCESCO||Cycle: XXX |
Tutor: RECH IVAN Major Research topic
:Development of high performance Single Photon Avalanche Diode arrays
Advisor: GULINATTI ANGELOAbstract:
Photon counting and time-correlated photon counting techniques have been developed over the years exploiting the ability of Photo Multiplier Tubes (PMTs) to detect a single photon. However PMTs have many limitations: in fact they are bulky, fragile, not suitable to be used in arrays and they have low efficiency in the red and in the near-infrared. Recent years have seen the arising of semiconductor devices, able to overcome many of PMT drawbacks, called Single Photon Avalanche Diodes (SPADs). These new detectors are formed by classical pn junctions, polarized with a reverse voltage higher than the breakdown. The single photon can be absorbed in the active region of the detector, so it is able to trigger an avalanche, which produces a macroscopic current easily detectable by the electronics. However the applied voltage is higher than the breakdown, so the multiplication process is self-sustaining. It is therefore always necessary to couple each SPAD with a circuit that detects the avalanche and reset the device to the initial conditions (Active Quenching Circuit - AQC).
Principally there are two approaches to the fabrication of SPADs: the first is to use a standard CMOS technology. The advantages of this choice are the reliability of the process flow and the possibility of integrating the auxiliary electronics (e.g. AQC) within the same chip. Unfortunately, this technology does not allow enough flexibility for the design of the detector. To obviate this drawback it is possible to use a custom technology, i.e. a technology dedicated to the fabrication of SPAD. This optimizes the main figures of merit of the detector, but it forces you to realize SPAD and electronics on separate chips: therefore, since the number of electrical connections (using wirebonding) between a chip and the other is physically limited, also the maximum number of detectors is limited and this constraint (some tens of devices in each array) is definitely too strict for some applications.
The aim of this Ph.D. is going to be twofold: on one hand new strategies for the fabrication of two-dimensional SPAD arrays will be developed; on the other hand the single detector performance will be improved, taking advantage of the flexibility afforded by the custom technology. In fact typical applications require high quantum efficiency, low dark count rate and high temporal resolution measuring the arrival time of the photon. All these parameters are highly dependent on the electric field, that needs to be finely tuned to satisfy all the requirements. Therefore it is also of paramount importance the development of numerical and physical models, which are able to predict the effect of the electric field on the performance of the detector.