Study of metal-semiconductor-metal (MSM) structures based on cadmium zinc telluride and cadmium manganese telluride for roomtemperature X- and γ-rays detectors
Artem Brovko
PhD student under the supervision of Prof. Arie Ruzin
Abstract
High resistivity cadmium zinc telluride (Cd1-xZnxTe) or CZT has emerged in the 90s’ as a prominent material for the room-temperature X-ray and gamma-rays radiation detectors. The favorable combination of CZT properties includes high average atomic number (for radiation absorption efficiency), wide energy bandgap (for low generation-recombination current at room-temperature), sufficient carrier mobility (for fast and efficient charge collection). The ternary alloy provides bandgap increase over the previous generation CdTe, but it is not free of challenges (e.g., inhomogeneity, yield, etc.). The deployment of Cd1-xZnxTe radiation detectors is widening in many fields including medicine, security, ecology, and industry. More recently, Cd1-xMnxTe (another II-VI ternary compound) was suggested to overcome the shortcomings of CZT, but it was not studied as much as CZT, yet.
Despite the commercialization and the fact that these II-VI group detectors are studied for many years, the fabrication technology is far from mature and the basic understanding is insufficient.
The overall performance of the CZT detector is affected by every fabrication step from bulk and surface preparation to contact deposition and passivation. The basic understanding of the physics behind these steps in insufficient, which hinders further device optimization.
This research has two main objectives, to promote the understanding of physics associated with the various fabrication steps, and to assess the differences between devices based on Cd1-xZnxTe and Cd1-yMnyTe crystals, grown by the same technology. Both crystal types were kindly provided by GE Healthcare (Rehovot, Israel).
First, the impact of various commonly used surface treatments, including mechanical, mechano-chemical polishing and chemical etching was studied. All treatments were found to affect surface chemistry, stoichiometry, and topography. Some effects were not limited to the near-surface layer, and it was demonstrated that structural defects from mechanical polishing penetrate several micrometers into the bulk. This introduces a new “history-effect”. Correlations were found between the physical properties (chemistry, stoichiometry, topography, and “history”) and the resulting device performance.
Another part of the work was devoted to the post-treated surface passivation with oxygen plasma. The effect of such passivation on physical properties and their correlation to device performance were investigated. Clear indications were found showing that the effect of such surface treatment extend deep into the bulk (contrary to the common belief).
In course of this study several metals were tested for contact formation. For the first time ever printed contacts (from silver particle ink) were deposited on CZT and tested. Such experimental technology simplifies the process and precludes the use of expensive clean room facilities and mask fabrication. The results were found to be very promising, yet further optimization is desirable.
Last, but not least, novel CdMnTe crystals grown by similar technology were tested and assessed. The material uniformity in these samples is improved relative to CZT and the resistivity is comparable. The novel material has great potential, and a dedicated further research is in order. As a final touch, CZT material grown at Washington State University was acquired and tested. These crystals were grown by a unique, proprietary experimental method without post-growth annealing. The results were compared to the Bridgman grown samples.
Various techniques were used in course of this research, including I-V, noise PSD, atomic force microscopy, transient current techniques, deep level transient spectroscopy, TSC, SEM, XRD, XPS.
