Investigations of real structure depth distributions in metals

using diffraction techniques

Z. Pala

Department of Solid State Engineering, Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Trojanova 13, 120 00 Prague 2, Czech Republic

zdenek.pala@fjfi.cvut.cz

In the case of metals, X-ray diffraction (XRD) is often regarded as a surface technique which reveals structural information about diffracting body only several micrometers or tens of micrometers in thickness. Yet in comparison with electron diffraction, assumption about surface sensitivity would be hardly tenable. On the other end of the penetration scale, neutron diffraction is firmly anchored because the neutrons penetrate metallic materials more easily and the resulting data can be gained from depths amounting to centimetres.

Notion of penetration depth, which was initially introduced by Cullity [1], is of vital importance from the real structure point of view. The main reason is that metal objects, especially those with surface treatment history, often show significant depth gradients of macroscopic and microscopic residual stress, grain sizes, or even texture. Gradients of these real structure characteristics have direct impact on the object behaviour in real environment, affecting its corrosion resistance, dynamic load stability, crack initiation and propagation and other processes that can contribute to either prolongation or appreciable shortening of fatigue life [2].

The effective penetration depth is given by (i) absorption µ of the given volume for the impinging X-ray beam wavelength, (ii) Bragg angle θ and (iii) geometric alignment of the goniometer f(θ,ψ). Applying the well known Lambert-Beer law for absorption, the ratio of intensities diffracted by a layer of thickness dT on the very surface and in the depth T will be

. (1)

Upon setting this ratio to a convenient constant of 1/e and taking a logarithm of the (1), the term for effective penetration depth T^e is calculated:

. (2)

Hence, T^e determines the thickness of a layer providing 63.2 % out of the entire diffracted intensity. Rigorously said, the structural information gained from classical Bragg-Brentano goniometer or even grazing incidence diffraction changes with the changing 2θ angle and is, therefore, influenced by possible steep structural gradients. However, when the gradients are not to be expected within the comparatively small thickness comparable with T^e, the depth profiling can be performed with combining the chosen XRD technique on a conventional laboratory diffractometers and subsequent layer removal. Inherently destructive layer removal should be done with minimal impact to the structure of the remaining layers; the most widely used technique is electro-chemical polishing [3]. Even the laboratory diffractometers offer possibility to change X-ray tube and, accordingly, the used wavelength. This would allow corresponding alteration of T^e, yet given the variety of X-ray tube selection, the scope of T^e for given metal is quite limited.

Another possibility for depth distribution investigation in metals is to employ synchrotron radiation. Not only is it tuneable with maximal energies peaking at 150 keV, but the intensities increase manyfold. The experimental set-up of 1ID C beamline at APS (Advanced Photon Source) in Argonne, USA depicted in Fig. 1 is almost an ideal tool for defined depth profiling. The incoming X-ray beam can be focused down to 2×5 μm² and with maximal energy of 130 keV can easily penetrate 1 cm of steel, with large detector array High Energy Detector Array or HYDRA© the minimal acceptable transmission reaches 0.1 %.

Figure 1. Experimental set-up of 1 ID C in APS, Argonne; courtesy of Jonathan Almer.

The presentation will deal with detailed description of advantages and pitfalls of diffraction experiment at 1 ID C. Furthermore, a mutual comparison between conventional XRD laboratory and synchrotron experiment results will be offered. Tangibly, macroscopic residual stress profiles in ground steel will be compared. Moreover, depth distributions of structure in plasma sprayed tungsten and copper layers used as plasma facing components in tokamaks obtained with use of neutron diffraction [4]and synchrotron radiation will be discussed.

Possibility to measure and study depth distributions of real structures parameters represent a further step in the progress of diffraction techniques and bring them nearer to the wider public. In this case, the industrial applications are, indeed, numerous and significantly broaden horizons of material scientists and physical engineers during new material evolution and manufacturing of metal objects used in hi-tech applications.

1. B.D. Cullity, Elements of X-ray diffraction. Reading: Addison -Wesley. 1956.

2. A. G. Youtsos, Residual stress and its effect on fracture and fatigue. Springer 2006.

3. S. Lee, Y. Lee, M. Chung, Metal removal rate of the electrochemical mechanical polishing technology for stainless steel – the electrochemical characteristics, IMechE Vol. 220, 2006.

4. V. Luzin, J. Matejicek, T. Gnaupel-Herold, Through-thickness Residual Stress Measurement by Neutron

Diffraction in Cu+W Plasma Spray Coatings, Materials Science Forum 652 (2010) 50-56.

The research was supported by the Project MSM 6840770021 of the Ministry of Education, Youth and Sports of the Czech Republic and by the Project SGS10/300/OHK4/3T/14 of the Czech Technical University in Prague.