The Science Payload and Advanced Concepts Office of ESA has an extensive laboratory program to develop custom X- and gamma-ray detection systems for future ESA space missions. These systems have to operate in a variety of diverse and esoteric environments; ranging from the relatively benign and stable near-Earth orbit environment to the hostile radiation environments of the Jovian system - the corrosive atmosphere of Venus - or the extreme thermal environment of Mercury. Therefore SCI-A has worked for a number of years to produce monolithic compound semiconductor detectors and large area arrays for future space missions.
Semiconductor materials have already been used in alpha, proton and X-ray spectroscopy, in both remote and direct sensing experiments. However, while remote sensing missions have studied the moon, they suffer from limited sensitivity and crude spatial resolution. This is because, up till now, planetary science has lacked short focal length optics matched to a thin but highly absorbent detection plane, which for interplanetary missions must also be free of the power, weight and cryogenic constraints normally associated with traditional detection technologies. New developments in slumped micro channel plate technology and wide band-gap compound semiconductors offer the potential of circumventing these problems, providing spatially resolved, Fano limited spectroscopy across the energy range 1-200 keV.
What are compound semiconductors?
The term compound semiconductor encompasses a wide range of materials, most of which crystallize in either the zincblende (ZB), wurtzite or rocksalt crystal structures. They were first investigated as radiation detectors in 1945 by Van Heerden, who demonstrated that silver chloride crystals when cooled to low temperatures were capable of detecting γ-rays, α-particles and β-particles. This work was later extended to the thallious halides by Hofstadter in 1949. Strangely, at this time, the elemental semiconductors were not anticipated to be responsive to radiation and it was not until several months later than McKay managed to measure the polonium α-ray spectrum using a Ge surface barrier detector. While the subsequent development of Ge and Si radiation detectors was rapid, largely due to the explosive rise of the semiconductor industry, mixed-element and compound semiconductors, such as the alkali-metal halides, CdTe, CdZnTe and HgI2, were plagued by material problems caused by severe micro-crystallinity, high defect densities, impurities and stoichiometric imbalances. Until the 1960s very little additional work was carried out. Since this time, progress has been incremental, but hardly spectacular.
Compound semiconductors are generally derived from elements in groups II to VI of the periodic table. They are so useful because of the shear range of compounds available, compared to the elemental semiconductors, Sn-α, C, Si and Ge. Most elements in these groups are soluble within each other, forming homogeneous solid solutions. These solutions occur when atoms of a particular element are able to substitute a given constituent of a different material without altering its crystal structure.
In order that atoms can form solid solutions over large ranges of miscibility, they should satisfy the Hume–Rothery Rules, namely that,
The rationale behind the Hume-Rothery Rules is:
In addition to binary materials (such as GaAs or InP), most compounds are also soluble within each other, making it possible to synthesize ternary (e.g., AlGaAs, HgCdTe) quaternary (e.g., InGaAsP, InGaAlP) and higher-order solutions, simply by alloying binary compounds together.
Why use compound semiconductors?
Apart from its on-going development of micro channel plate optics, the Advanced Technologies Section is studying the suitability of several compound semiconducting materials, including GaAs, CdZnTe, TlBr, InP and HgI2. These all have band-gaps sufficiently high so they do not require cryogenic cooling but low enough that sub-keV spectral resolution may be achieved at hard X-ray energies. In addition, the materials are drawn from group III and V compounds and so have high enough atomic numbers to ensure high detection efficiencies above 10 keV.