Computational modelling of pixelated CdZnTe detectors for x- and γ- ray imaging applications

Cadmium Zinc Telluride (CdZnTe) detectors are currently used in medical imaging systems employing γ-ray photons. As new imaging techniques such as photon-counting and energy-weighted x-ray imaging are gaining research interest, CdZnTe is seen under a new light for potential use in computed tomography, tomosynthesis and other x-ray imaging applications. However, being relatively expensive, CdZnTe could be favoured by advanced computational modelling to assist in detector and imaging system optimisation. In this work, pixelated CdZnTe detectors are computationally modelled using an integrated framework that combines the Finite Element and Monte Carlo numerical methods to obtain realistic detector models.Various detector thickness and pixel sizes are designed and their performance is investigated in terms of charge induction efficiency, detection efficiency and energy resolution. Detection efficiency and energy resolution are assessed for monoenegergetic photon beams within the energy range used in medical x-ray imaging applications such as mammography and computed tomography. Some of the capabilities of the framework are demonstrated. Small pixel sizes, below 100μm are prone to charge transport effects such as diffusion, especially in larger thickness ( > 0.5 mm) and may have limited use in pixelated geometries. Detection efficiency is affected by fluorescence and photon escape as thickness and pixel size decrease. Energy resolution is affected by beam geometry and can vary from ∼ 3% to 11% depending on the beam width. The framework provides a generic platform and a powerful tool that can be used in the design and optimisation of semiconductor detectors made from any semiconductor material, imaging systems and signal correction techniques.