Eur. Phys. J. AP 2, 223–226 (1998) T HE EUROPEAN P HYSICAL JOURNAL APPLIED PHYSICS c  EDP Sciences 1998 Ion implantation-reduced diffusion length in HgCdTe ⋆ P. L´ evˆ eque a , A. Decl´ emy, and P.O. Renault Laboratoire de M´ etallurgie Physique b , Universit´ e de Poitiers, UFR Sciences, SP2MI, Bd3, T´ el´ eport 2, BP 179, 86960 Futuroscope Cedex, France Received: 7 November 1997 / Revised: 22 January 1998 / Accepted: 20 February 1998 Abstract. Al 2+ (320 keV) ion implantations at room temperature for doses ranging from 3 × 10 11 to 3 × 10 14 ions cm -2 have been performed in p-type Hg0.78Cd0.22 Te. The samples were analyzed by infrared reflectivity providing quantitative results on electrically active defects. A change of type of the free carriers is observed after implantation, for all doses, in a region of thickness greater than the projected range of ions. The dominant phenomena in this change of type seems to be the diffusion of Hg interstitials which takes place during implantation. From 3 × 10 11 to 3 × 10 13 ions cm -2 , the diffusion length of Hg interstitials follows a classical equation with a linear dependence on the squareroot of the implantation time. For doses greater than 3 × 10 13 ions cm -2 , Hg interstitials diffuse less efficiently. This could be due to the capture of some single interstitials by interstitial dislocation loops as observed by diffuse X-ray scattering. PACS. 61.72.Vv Doping and impurity implantation in III-V and II-VI semiconductors Introduction Design and fabrication of HgCdTe detectors have been greatly improved in recent years, but it involves more and more well-controlled process steps. In order to re- duce expensive and time-consuming iterations during pro- cess development, simulators are used especially to model junction formation by ion implantation in p-type HgCdTe [1,2]. The different mechanisms involved in the junction for- mation are now well established [3]. Unlike conventional implantation, the electrical activity of the implanted ion is not the most important factor. The purpose of the im- plant is to create a source of mobile Hg interstitials near the surface of the sample. The interstitials can diffuse into the material and annihilate Hg vacancies, single or dou- ble acceptors, which are present in large concentrations in p-type HgCdTe grown by T.H.M. These recombina- tions reveal low concentrations of donors species (∼ 0.5 to 1 × 10 15 cm −3 ) grown into bulk samples. When the con- centration of the acceptors is reduced below the donor’s one, the region is converted to n-type. This image of the junction formation by ion implan- tation, included in all simulators, is accepted during an- nealing but it might be extrapolated to the implantation itself. Nevertheless, the production of electrically active defects during implantation and before annealing, as the ⋆ This paper was presented at D.E.S. 97 (Poitiers, France) September 4 and 5, 1997. a e-mail: Patrick.leveque@lmp.univ-poitiers.fr b UMR 6630-CNRS. influence of structural defects on doping efficiency have not yet been studied in great detail. In this work, Al 2+ (320 keV) implantations have been performed on p-type Hg 0.78 Cd 0.22 Te with doses ranging from 3 × 10 11 to 3 × 10 14 ions cm −2 . Implanted samples were analyzed by infrared reflectivity [4]. This technique is contactless and non-destructive and give some infor- mation on electrically active defects. In the first part of this paper, we present the experimental procedure for im- plantations and infrared reflectivity spectra analysis. The results on the depth distribution of electrically active de- fects are presented in the second part. The influence of structural defects as observed by diffuse X-ray scattering [5] on electrically active ones is discussed in the last part. 1 Experimental procedure and infrared reflectivity spectra analysis The samples used in this study were p-type Hg 0.78 Cd 0.22 Te (with a uniform hole concentration around 10 16 cm −3 ) bulk single crystal (111) supplied by SAGEM S.A. (St-Benoˆ ıt). Prior to implantation, the samples were etched in a dilute bromine methanol solu- tion at room temperature. Al 2+ (320 keV) implantations were performed in the laboratory using doses ranging from 3 × 10 11 to 3 × 10 14 ions cm −2 . All the samples were implanted at room temperature with crystal axis tilted by 7 ◦ from the ion beam to prevent channeling effects. The ion current was kept constant below 20 nA cm −2 in order to prevent the temperature rise during implantation.