Modelling Polarization Effects in a CdZnTe Sensor at Low Bias

Abstract

Semi-insulating CdTe and CdZnTe crystals fabricated into pixelated sensors and integrated into radiation detection modules have demonstrated a remarkable ability to operate under rapidly changing X-ray irradiation environments. Such challenging conditions are required by all photon-counting-based applications, including medical computed tomography (CT), airport scanners, and non-destructive testing (NDT). Although, maximum flux rates and operating conditions differ in each case. In this paper, we investigated the possibility of using the detector under high-flux X-ray irradiation with a low electric field satisfactory for maintaining good counting operation. We numerically simulated electric field profiles visualized via Pockels effect measurement in a detector affected by high-flux polarization. Solving coupled drift-diffusion and Poisson's equations, we defined the defect model, consistently depicting polarization. Subsequently, we simulated the charge transport and evaluated the collected charge, including the construction of an X-ray spectrum on a commercial 2-mm-thick pixelated CdZnTe detector with 330 µm pixel pitch used in spectral CT applications. We analyzed the effect of allied electronics on the quality of the spectrum and suggested setup optimization to improve the shape of the spectrum.

Keywords: CdZnTe; high flux; polarization; radiation detector.

Figure 1

X-ray spectrum (red) with attenuation coefficient in CdZnTe (inset).

Figure 2

Schematic view of the electric field distribution in a detector with a positive space charge.

Figure 3

Measured electric field profiles in a 2-mm-thick CZT sensor (a), numerical simulation (b); solid lines represent electric field with no incoming X-ray (dark mode), dashed lines are for X-ray 16 Mcps/mm², dotted lines are for X-ray 80 Mcps/mm², blue, red, and green color mark 300 V, 500 V, and 700 V bias, respectively.

Figure 4

Profile of X-ray excitation, which shows dominant excitation under cathode with fast decrease toward anode. Weighting field was calculated according [30] in the middle of the pixel with 330 µm pitch.

Figure 5

Scheme of energy levels with parameters determined by the fit of electric field. Hole trap is red; electron trap is blue.

Figure 6

Calculated space charge density from the numerical simulation. In the dark regime, there is no space charge in the sample.

Figure 7

Measured counts in 2-mm-thick CZT sensor under X-ray conditions with a count rate of 20 Mcps/mm² for a typical pixel. Procedure to extract critical bias $U_c$ is shown at the intersection of linear fit for bias $U<U_c$ with the horizontal line at maximum CCE.

Figure 8

Distribution of $U_c$ values for all pixels.

Figure 9

Simulated current waveforms for different biases and 20 Mcps/mm² X-ray.

Figure 10

Comparison of the collected charge of X-ray photon absorbed near the cathode (red), and deep inside the detector, 1.44 mm (blue). U = 300 V.

Figure 11

X-ray spectrum simulated for common parameters characterizing the electronic circuit, $t_s=16\mathrm{ns}$ (blue). The original X-ray spectrum is plotted for comparison (red).

Figure 12

Analysis of sampled charge depending on photon energy. In case (A), the photon energy is lower than the threshold and the charge is not detected. In case (B,C), the threshold is triggered and the charge is sampled.

Figure 13

Dependence of the collected charge on the excited charge by X-ray photon absorbed near the cathode (blue) compared with the ideal full collection plotted with a straight line (red). Labels point to the threshold levels shown in Figure 12.

Figure 14

X-ray spectrum simulated for extended $t_s=32\mathrm{ns}$ (blue). The original X-ray spectrum is plotted for comparison (red).