Evolution kinetics of elementary point defects in ZnO implanted with low fluences of helium at cryogenic temperature

C. Bhoodoo, A. Hupfer, L. Vines, E. V. Monakhov, and B. G. Svensson

Phys. Rev. B 94, 205204 – Published 28 November 2016

Abstract

Hydrothermally grown $n$ -type ZnO samples, implanted with helium $({He}^{+})$ at a sample temperature of $\sim 40$ K and fluences of $5 \times 10^{9}$ and $5 \times 10^{10} {cm}^{- 2}$ , have been studied in situ by capacitance voltage (CV) and junction spectroscopy measurements. The results are complemented by data from secondary ion mass spectrometry and Fourier transform infrared absorption measurements and first-principles calculations. Removal/passivation of an implantation-induced shallow donor center or alternatively growth of a deep acceptor defect are observed after annealing, monitored via charge carrier concentration $(N_{d})$ versus depth profiles extracted from CV data. Isothermal anneals in the temperature range of 290–325 K were performed to study the evolution in $N_{d}$ , revealing a first-order kinetics with an activation energy, $E_{a} \approx 0.7 eV$ and frequency factor, $c_{0} \sim 10^{6} s^{- 1}$ . Two models are discussed in order to explain these annealing results. One relies on transition of oxygen interstitials $(O_{i})$ from a split configuration (neutral state) to an octahedral configuration (deep double acceptor state) as a key feature. The other one is based on the migration of Zn interstitials (double donor) and trapping by neutral Zn-vacancy-hydrogen complexes as the core ingredient. In particular, the latter model exhibits good quantitative agreement with the experimental data and gives an activation energy of $\sim 0.75$ eV for the migration of Zn interstitials.

Received 7 September 2016
Revised 6 November 2016

DOI:https://doi.org/10.1103/PhysRevB.94.205204

Physics Subject Headings (PhySH)

Capacitance

Semiconductors

Density functional theory Fourier transform infrared spectroscopy Ion implantation

Condensed Matter, Materials & Applied Physics

Authors & Affiliations

C. Bhoodoo^*, A. Hupfer, L. Vines, E. V. Monakhov, and B. G. Svensson

University of Oslo, Department of Physics, Center for Materials Science and Nanotechnology, P.O. Box 1048 Blindern, N-0316 Oslo, Norway

^*chidanand.bhoodoo@fys.uio.no

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 94, Iss. 20 — 15 November 2016

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
$N_{d}$ versus depth profiles with increasing time at 300 K after implantation at $\sim 40$ K. The profiles were measured using 16 kHz probing frequency.
Reuse & Permissions
Figure 2
(a) $N_{d}$ profiles, 1 MHz probing frequency, recorded at different temperatures for a sample implanted with $\sim 5 \times 10^{9} {He}^{+} / {cm}^{2}$ at $\sim 40$ K and then annealed at 325 K for 150 h. The profile at 400 K was recorded off line. (b) DLTS spectrum of the sample in (a), rate window $= (640 {ms)}^{- 1}$ . The inset shows the Laplace DLTS spectrum recorded at 155 K, displayed as signal intensity versus emission rate $(s^{- 1})$ .
Reuse & Permissions
Figure 3
$N_{d}$ (loss), given by the difference between the as-grown charge carrier concentration and the remaining one at $R_{p}$ , versus annealing time for two samples implanted with different fluences at $\sim 40$ K and then annealed at 300 and 325 K, respectively.
Reuse & Permissions
Figure 4
(a) $ln [1 - (\frac{N_{d} (loss)}{N_{d} (saturated loss)})]$ versus annealing time at 290, 300, 310, and 320 K for samples implanted with $\sim 5 \times 10^{10} {He}^{+} / {cm}^{2}$ at $\sim 40$ K. The dotted lines represent results from simulations using the model outlined in Sec. 4c. (b) Arrhenius plot of the reaction rate constant determined experimentally versus the reciprocal absolute annealing temperature. The black circles represent values obtained by least-squares fits of the data in Fig. 4 (ion fluence $= 5 \times 10^{10} {cm}^{- 2}$ ), while the red circle shows the value obtained from samples implanted with a fluence of $5 \times 10^{9} {cm}^{- 2}$ and annealed at 325 K. The latter value is multiplied by a factor of 10 and is not included in the fit yielding an activation energy of 0.69 eV with a prefactor of $1.3 \times 10^{6} s^{- 1}$ . The blue squares (right $y$ axis) show values of the ${Zn}_{i}$ migration $(D_{{Zn}_{i}})$ deduced from the model simulations in Sec. 4c. Error bars indicate the experimental accuracy $(10 % – 20 %)$ .
Reuse & Permissions
Figure 5
Relative formation energies as a function of Fermi level $(E_{F})$ position for $O_{i}$ in split position (reaction coordinate $r = 0$ ), octahedral position $(r = 1)$ , and transition geometries $(0 < r < 1)$ in ZnO. $E_{F}$ equal to zero corresponds to the valence-band maximum. Only segments corresponding to the lowest-energy charge states are shown. The slope of these segments indicates the charge state. Kinks in the curves indicate transitions between different charge states. The inset shows the relative formation energy as a function of $r$ for $E_{F}$ fixed at 3 eV.
Reuse & Permissions
Figure 6
Simulation results for the evolution of interstitial oxygen atoms in the octahedral configuration as a function of annealing time at 300 K employing the model outlined in Sec. 4b. Results are shown for two initial concentrations of $O_{i}$ (split) at $t = 0$ .
Reuse & Permissions

Physical Review B

covering condensed matter and materials physics