Mercury cadmium telluride, Hg1−xCdxTe (MCT), is a semiconductor widely used as an active material for high-end infrared imaging systems. For this semiconductor, state-of-the-art performances are obtained for p-on-n photodiode architectures [1, 2]. The fabrication process of these photodiodes involves the use of ion implantation for the incorporation of dopants. As this process induces irradiation damage, a specific annealing is used to remove most of it. In previous work, we showed by scanning transmission electron microscopy (STEM) that residual defects such as nano-voids and nano-crystals can still be observed despite of the use of damage correction annealing [3].As the crystal quality of the active layer is critical to ensure the best device performances [4, 5], a better understanding of how this process induces defects is needed to limit their occurrences. However, the small size and low density of these defects make their study difficult.In this work, we used FIB-SEM nano-tomography to better understand these residual damages. Indeed, this characterization technique is well suited to our problem as it allows the visualization of large volumes of interest with a resolution down to a few nanometers.MCT epitaxial layers were grown on CdZnTe lattice-matched substrates. These crystals were then implanted with arsenic ions in a random off-axis direction at an energy of 360 keV. The same standard annealing procedure was applied to all samples. In each case, FIB-SEM nano-tomography allows us to extract the defect density as well as the size and depth profiles of the observed defects. These results are confronted with the depth profiles of the dopants concentration from secondary ion mass spectrometry (SIMS).This procedure was applied to equivalent samples with various fluences of implanted arsenic ions and various molar fractions of cadmium in the semiconductor alloy. These variants allow for insight into defect formation mechanisms in light of the solubility of the dopant in the alloy.[1] W. E. Tennant, “Interpreting mid-wave infrared MWIR HgCdTe photodetectors,” Prog. Quantum Electron., vol. 36, no. 2–3, pp. 273–292, 2012, doi: 10.1016/j.pquantelec.2012.05.001.[2] N. Baier, O. Gravrand, C. Lobre, O. Boulade, A. Kerlain, and N. Péré-Laperne, “HgCdTe Diode Dark Current Modeling: Rule 07 Revisited for LW and VLW,” J. Electron. Mater., vol. 48, no. 8, pp. 5233–5240, Aug. 2019, doi: 10.1007/s11664-019-07299-z. [3] C. Lobre, P.-H. Jouneau, L. Mollard, and P. Ballet, “Characterization of the Microstructure of HgCdTe with p-Type Doping,” J. Electron. Mater., vol. 43, no. 8, pp. 2908–2914, Apr. 2014, doi: 10.1007/s11664-014-3147-9.[4] T. Broult, A. Kerlain, V. Destefanis, P. Guinedor, E. Le Bourhis, and G. Patriarche, “Controlled Dislocations Injection in N/P Hg1−xCdxTe Photodiodes by Indentations,” J. Electron. Mater., vol. 48, no. 10, pp. 6108–6112, Oct. 2019, doi: 10.1007/s11664-019-07139-0.[5] K. Jóźwikowski, A. Jóźwikowska, M. Kopytko, A. Rogalski, and L. R. Jaroszewicz, “Simplified model of dislocations as a SRH recombination channel in the HgCdTe heterostructures,” Infrared Phys. Technol., vol. 55, no. 1, pp. 98–107, 2012, doi: 10.1016/j.infrared.2011.10.003.
Read full abstract