Positron annihilation studies of recrystallization in the subsurface zone induced by friction in magnesium—effect of the inhomogeneity on measured positron annihilation characteristics

Jerzy Dryzek 0 1 0 J. Dryzek Institute of Nuclear Physics PAN , ul. Radzikowskiego 152, 31-342 Krakw, Poland 1 J. Dryzek ( ) Institute of Physics, Opole University , ul. Oleska 48, 45-052 Opole, Poland The discussion of the positron annihilation studies of crystal structure defects, like vacancies, dislocations, grain boundaries and the defect depth profile, is presented. The role of the positron implantation depth and positron diffusion in such studies has been considered in detail. For description of the measured annihilation characteristics the proposed theoretical models take into account both effects. The annealing studies of defects created in pure magnesium by compression or dry sliding-wear were used for demonstration of the discussed thesis. The positron lifetime measurements were applied for monitoring open volume defects behavior. It was demonstrated that annealing at the temperature of about 300 C removes the defects created by compression. Application of the proposed model to description of the data obtained allows to determine the activation energy of the grain boundary mobility in pure magnesium equal to Q = 0.56 0.18 eV. However, defects created by the dry sliding are not completely annealed up to the temperature of 500 C. The defect depth profile induced by dry sliding evolves with the annealing temperature in such a way that at the worn surface concentration of defects gradually decreases but at the depth between 60 and 100 m the generation of new defects takes place at temperature of 150 and 225 C. Above 300 C the defects still are extended up to the depth of about 80 m. - Application of the positron annihilation spectroscopy to studies of matter is based on the unique properties of a positron as a probe at the atomic level. The positron annihilates with an electron and mainly two energetic quanta are emitted almost in opposite directions, which take over the total energy and momentum of the pair. Because thermalized positron occupies the lowest energy level in its band, the measurement of the total quanta momentum, i.e., the angular correlation of these quanta, is determined by the electron momentum in the annihilation place. Due to the positive charge, the positron is also sensitive to the local distortion of electron density, which is when the open volume defects, vacancies and their clusters are present in the structure. The angular correlation of annihilation quanta (ACAR) and the complementary Doppler broadening (DB) of the annihilation line are the main measured positron annihilation characteristics which are related to the local electron states in a matter. Another characteristic, the positron lifetime (PALS), depends mainly on the local electron density. Nevertheless, before annihilation the positron, as a mobile particle, scans a certain volume of the implanted sample, first as an energetic particle, in the implantation and thermalization process, and then during a random walk as a thermalized one. Thus, the annihilation characteristics reflect the local properties of the sample but they are averaged over the volume. This is not taken into account in the simple positron trapping model proposed by Brandt [1], Bergersen and Stott [2], and Connors and West [3], which is commonly used in the analysis of the PALS or DB spectra. In this model it is assumed that the defects which trap positrons with a certain rate are distributed homogeneously. Additionally, they localize only thermalized positrons. The extension of this model to the diffusion trapping model allows taking into account the fact that defects are distributed in a certain way in the volume. This problem was attacked theoretically and experimentally by many authors; some of them are listed in Refs. [48]. The size of the scanned volume depends on the positron diffusion length which is equal about 0.1 m. However, positrons used for PALS or DB measurements are mainly emitted from the radioactive sources into matter and distributed over the implantation profile. Its total depth is about hundreds of micrometers and depends on the positron energy and density of the matter [9, 10]. If the studied sample exhibits a non-homogeneous defect distribution across the implantation profile and/or the path of random walk, then the measured annihilation characteristics must be sensitive to it. The aim of the paper is to illustrate both effects in experimental studies. They concern a so-called subsurface zone (SZ) in the pure magnesium which is formed by dry sliding against another body. The SZ occurs below the surface exposed to the different technological processes and can extend up to hundreds of micrometers. Properties of the SZ are interesting because it is created directly by the processes at the worn surface which are difficult to characterize and because it is then an entering surface for positrons during measurements. We intended to find out the thermal stability of the SZ and to show how it is affected by the annealing process. In the first part of the paper we discuss the theoretical considerations of influence of the defect distribution on the positron annihilation characteristics. In the second part the experimental results concerning our studies of the SZ in pure magnesium are presented. 2 How the annihilation characteristics reflect the inhomogeneity of the probed samples 2.1 Inhomogeneity across the implantation profile Let us consider a flat semi-infinite sample and positrons which enter through the plane xy and penetrate the depth in the z direction, as depicted in Fig. 1a. Commonly in the experiments, positrons are emitted from the + decay nuclides located directly on the plane xy. They are emitted in the full solid angle. Additionally, we assume that only the layer of the sample between z and z + dz is uniform. The positron annihilation characteristics do not depend on the x and y coordinates. They change only in the z-direction. It is helpful to define the positron implantation profile p(z) as the probability of finding a thermal positrons at the end of the implantation and thermalization process, initially implanted with the energy E at the depth between z and z + dz from the entrance surface of the sample. Thus the total number of Sm(d) = p(z d)S(z) dz, this strategy needs application of the high energetic positron beam, which is not commonly used. The commonly used low-energy positron beam of tens of keV allows to scan only the depth of few micrometers. The last strategy is much simpler. One can remove from the top of the sample a layer of a certain thickness by etching. Sequenced etching and measurement let us to obtain the dependency of Sm on the total thickness of the etched layers, denoted as d, which is expressed as d These give us a possibility to deduce information about the depth profiles (z) and S(z). However, this strategy is not often used by authors [12, 13]. It was shown experimentally [9] and confirmed by Monte Carlo simulations [10] that for positrons e (...truncated)