Dear Tamás Kolonits , here some of the papers from ScienceDirect

A Kehrel, K.P Lieb, M Uhrmacher,

Diffusion of implanted sodium in nickel and chromium,

Materials Science and Engineering: A,

Volume 115,

1989,

Pages 43-47,

ISSN 0921-5093,

https://doi.org/10.1016/0921-5093(89)90654-0.

(http://www.sciencedirect.com/science/article/pii/0921509389906540)

Abstract: Na+ ions were implanted at 250 keV into polycrystalline samples of chromium and nickel to a dose of 1.5 × 1017 cm−2. Diffusion phenomena during vacuum annealing were observed up to 1000 °C by the resonant nuclear reaction analysis using the reaction 23Na(p, γ)24Mg at the resonance energy ER = 308.75 keV. Above Ta = 750 °C precipitation of sodium in the region of maximal radiation damage was observed. Above the boiling temperature of sodium at T = 883 °C a fast loss of all implanted sodium from the samples is observed. The results are compared with those obtained recently by us in aluminium.

M. Uhrmacher, K.P. Lieb,

Sodium diffusion in metals observed by the 23Na(p,γ) reaction,

Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms,

Volume 68, Issues 1–4,

1992,

Pages 175-179,

ISSN 0168-583X,

https://doi.org/10.1016/0168-583X(92)96073-8.

(http://www.sciencedirect.com/science/article/pii/0168583X92960738)

Abstract: Resonant nuclear reaction analysis (RNRA) provides a convenient and precise method of measuring concentration profiles of light elements in near surface regions. The 23Na(p,γ)24Mg resonance at 309 keV was used to study supersaturated Na-solutions in polycrystalline Al, Cr, Fe, Mo, Ni and Ta after Na+ implantation at 250 keV and thermal annealings. Outdiffusion of Na at high temperatures was observed for all hosts, and different stages during this process were identified. Na diffuses into the region of correlated radiation damage where precipitation occurs and finally Na migrates to the surface where it evaporates. In all metals the loss of Na was observed to occur within a small temperature interval around a critical temperature T1. It turned out that T1 essentially depends on the melting temperature Tm and the structure of the host lattice. Half of the implanted Na-content had left the matrix at 0.53(2)Tm for bcc-metals and at 0.70(2)Tm for fcc-metals. A possible influence of the analyzing proton beam and the role of implanted and dissolved hydrogen in the vicinity of the implanted Na profile was also studied. In fact, hydrogen from the bulk migrates to and is trapped at the Na precipitates.

Thank you Aleksandr Pupynin !