Eur. Phys. J. AP 13, 83–87 (2001) T HE EUROPEAN P HYSICAL JOURNAL APPLIED PHYSICS c  EDP Sciences 2001 Redistribution of Ni implanted into InP T.K. Chini 1 , a , S.K. Ghose 2 , B. Rout 2 , B.N. Dev 2 , M. Tanemura 3 , and F. Okuyama 3 1 Saha Institute of Nuclear Physics, 1/AF Bidhannagar, Calcutta 700064, India 2 Institute of Physics, Sachivalaya Marg, Bhubaneswar 751005, India 3 Nagoya Institute of Technology, Graduate School of Engineering, Gokiso-cho, Showa-ku, Nagoya 466, Japan Received: 31 July 2000 / Revised: 26 October 2000 / Accepted: 7 December 2000 Abstract. The redistribution of Ni in InP is studied by annealing samples of InP implanted with 0.9 MeV Ni at 60 ◦ angle of ion incidence with respect to target surface normal as a function of dose (8.5×10 12 – 4.5×10 15 cm −2 ). Ni profiles are measured by secondary ion mass spectrometry (SIMS) and implantation induced damage by Rutherford backscattering spectrometry in channeling (RBS/C) condition. The highest dose sample is characterised by remarkable Ni accumulation near the surface (at ∼ 0.3Rnp) that has not been observed earlier along with two other distinct accumulation zones at Rnp+ΔRnp and 2.2Rnp after annealing at 650 ◦ C for 30 min. Here, Rnp is the normal component of the projected range for oblique angle bombardment. PACS. 61.72.Vv Doping and impurity implantation in III–V and II–VI semiconductors – 61.72.Yx Inter- action between different crystal defects; gettering effect – 68.35.Fx Diffusion; interface formation 1 Introduction A serious problem [1–9] of Fe or other 3d transition metal (TM) implantation in InP or GaAs is the thermal insta- bility (except Ti) of the implanted profile which includes redistribution, outdiffusion, accumulation and or gettering of the implanted species at different region in the specimen during post implantation annealing, which is partly over- come in some cases [4–6] by changing the implantation or annealing condition. Thermal stability of the implanted species is an important criteria to achieve semi-insulating behaviour of InP so as to make it favorable material for electronic device application. Till now most of the studies [4–8] concern a maximum dose region of 10 14 cm −2 for transition metal implantation in InP. But to achieve semi-insulating InP with maximum possible concentration of implanted species which is re- quired for highly doped (say, n + type InP) substrates, high dose (more than the above mentioned) implantation is desirable. Moreover, high dose implantation in InP may influence the implantation profile in different way than the low dose implantation because of the difference in the implantation induced damage between low and high dose. Thus, it is of importance to check the degree of thermal instability of high dose implantation in InP for which few studies [1–3] exist. To our knowledge, Ni (a member of the 3d TM series with suitable deep acceptor level of 0.48 eV [10]) implanta- tion in InP has not been tried earlier. Thus, in the present a e-mail: tapashp@hp2.saha.ernet.in work, we examine the thermal behaviour of Ni in InP af- ter the 0.9 MeV Ni implantation in InP as a function of ion fluence with the emphasis to observe the effect of high fluence. The as-implanted and the annealed samples were characterized by employing secondary ion mass spectrom- etry (SIMS). To correlate the SIMS profiles and implan- tation induced damage, combined Rutherford backscat- tering spectrometry and channeling (RBS/C) were performed for damage estimation. 2 Experimental Mirror polished, liquid encapsulated Czochralski (LEC) grown, (100) oriented, 365 μm thick n-type (Sn doping, 1.8×10 18 cm −3 ) InP wafers were implanted with 58 Ni at room temperature using the 3 MV tandem Pelletron accel- erator (9SDH-2, NEC) facility of the Institute of Physics, Bhubaneswar, India. The samples were mounted on a stainless steel sample holder using a heat conducting tape to ensure good thermal contact. The energy of the ions was 0.9 MeV and the implanted dose was varied in the range of 8.5 × 10 12 –4.5 × 10 15 cm −2 . The ion current den- sity ranged from 8 to 18 nA cm −2 which was low enough to keep the sample at room temperature avoiding the chance of phosphorus evaporation from InP due to beam induced heating. Part of each sample was masked off to provide an unimplanted reference region and beam homogene- ity on the bombarded spot was maintained using raster scan. During implantation the substrates were kept tilted with an angle of 60 ◦ between the beam and the sample