Spectral calibration and modeling of the NuSTAR CdZnTe pixel detectors

TL;DR

This paper presents a spectral calibration and modeling approach for NuSTAR's CdZnTe pixel detectors, combining Monte Carlo simulations with calibration data to characterize detector response and improve scientific measurements.

Contribution

It introduces a novel Geant4-based model integrated with the Shockley-Ramo theorem for detailed spectral response calibration of NuSTAR detectors.

Findings

Determined electron and hole mobility-lifetime products for each pixel.

Compared actual detector performance with ideal defect-free models.

Enhanced calibration accuracy for NuSTAR's spectral measurements.

Abstract

The Nuclear Spectroscopic Telescope Array (NuSTAR) will be the first space mission to focus in the hard X-ray (5-80 keV) band. The NuSTAR instrument carries two co-aligned grazing incidence hard X-ray telescopes. Each NuSTAR focal plane consists of four 2 mm CdZnTe hybrid pixel detectors, each with an active collecting area of 2 cm x 2 cm. Each hybrid consists of a 32 x 32 array of 605 micron pixels, read out with the Caltech custom low-noise NuCIT ASIC. In order to characterize the spectral response of each pixel to the degree required to meet the science calibration requirements, we have developed a model based on Geant4 together with the Shockley-Ramo theorem customized to the NuSTAR hybrid design. This model combines a Monte Carlo of the X-ray interactions with subsequent charge transport within the detector. The combination of this model and calibration data taken using radioactive sources of Co-57, Eu-155 and Am-241 enables us to determine electron and hole mobility-lifetime products for each pixel, and to compare actual to ideal performance expected for defect-free material.