Surface layers study of bulky samples by X’Pert PRO diffractometer

Z. Pala¹, N. Ganev¹, J. Drahokoupil²

¹Department of Solid State Engineering, Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Trojanova 13, 120 00 Prague

²Institute of Physics of the ASCR, v. v. i., Na Slovance 2, 182 21 Prague

zdenek.pala@fjfi.cvut.cz

Samples for investigation by means of X-ray diffraction are frequently prepared solely and, therefore, suitably for the experimental arrangement, which often imposes stringent conditions to its shape, mass and dimensions. However, real samples from industrial production cannot be usually cut into feasible parts without changing their structural and physical properties. Generally, samples’ amendments can even lead to redistribution of residual stresses by inducing new plastic deformations. Consecutive inspection of such artificially created objects has only limited relevance to the original state which is the centre of interest.

The appropriate attitude to XRD measurements of bulky and heavy samples is, first of all, a choice of convenient goniometer geometry. The theta-theta goniometer configuration offers comparatively large space for sample handling. Preferably, the sample mounting should be external and, thus, allowing placement of large volume beneath the investigated surface. Such external mounting stages should have large travel range with smallest possible position resolution and reading accuracy in both vertical and horizontal directions. Consequently, sufficiently precise position control cannot be omitted. This can be done by a high precision laser sensor for dimensional measurement which has ample resolution up to 0.001 mm.

In the experiment, three types of machined surface layers for guide gibs of dimensions 160×105×45 mm³ were examined. Samples from the steel 11 375.0 were machined by milling, grinding, and scraping. Semiproducts were cut from the steel sheet without any heat treatment by using an acetylene jig-burner. The aim of the research was to characterize each surface by state of macroscopic residual stress on the very surface and in near surface area of ca 200 μm in depth. Moreover, profiles of diffraction line {211} of α-Fe phase were used for calculation of microstrains and domains of coherent scattering by the single line Voigt function method [1]. Microhardness and metallographic measurements provide a supplement to diffraction results.

Figure 1. Mounting of a guide guib for residual stress measurement by X’Pert PRO.

Sample positioning in the X’Pert PRO diffractometer is depicted in Fig. 1. A set of motorized and manual vertical and horizontal stages by Standa [2] was used for sample setting-up. Because the laser sensor was not available, mounting to the desired position was performed by a simple and straightforward alignment procedure recommended by PANALYTICAL. Its basic principle lies in a flat surface positioning to the level when it intersects the incident beam into two equivalent halves [3] which can be verified either by a suitable gauge or by intensity measurement.

State of biaxial macroscopic residual stress (RS) on the surface was established on three chosen areas of each sample in order to find out, in the first approximation, level of RS homogeneity in final surface. RS depth distribution was obtained by successive layer removal by electro-chemical polishing. Sets of diffraction data were evaluated by centre of gravity algorithm and biaxial state was assumed, shear stress was recorded only in ground surface. Results of RS after layer removal were corrected according to Moore and Ewans method [4].

Figure 2a. Residual stress distribution in milled surface.

Figure 2b. Residual stress distribution in ground surface.

Figure 2c. Residual stress distribution in scraped surface.

References

1. Th.H. de Keijser, J.I. Langford, E.J. Mittemeijer, A.B.P. Vogels J. Appl. Cryst., 15, (1982), 308.

2. www.standa.lt

3. X’Pert PRO User’s Guide, Fourth Edition, 2002.

4. Sikarskie D., Trans. of the Metallurgical Society of AIME. 239 (1967) 577 – 580.

Acknowledgements.

The research was supported by the Project MSM № 6840770021 and by the Project № FT-TA4/105 of the Ministry of Industry and Trade of the Czech Republic.