Palladium is a suitable material for chemical sensors as a catalytic metal layer [1],[2]. Sensors with a palladium gate are sensitive to hydrogen molecules or hydrogen containing molecules which are dissociated on the catalytic metal surface. The hydrogen sensors with palladium gate can be used, for example, in leak detectors.
In this study, palladium thin films different in thickness were deposited onto Si(111), Si(100) (both single crystalline), glass/Ti (polycrystalline textured) and glass (amorphous) substrates by means of r.f. reactive sputtering under the same sputtering conditions in order to appreciate the influence of substrate structure, as well as palladium film thickness on the X-ray diffraction line profile characteristics, i.e. peak position, intensity, FWHM (full width at half maximum), integral breadth and shape factor for individual observed Bragg reflections. The omega-scan curves for palladium (111) reflections were also recorded.
The X-ray diffraction analysis was carried out on the automatic X-ray powder diffractometer URD-6 with Bragg-Brentano goniometer using CuK$_{\alpha}$ radiation. Ceramic Al$_{2}$O$_{3}$ from NIST was used as an instrumental standard. The method proposed by Langford [3] was used to determine the microstructural properties of Pd thin films.
It has been found from the X-ray diffraction line profile analysis that very strong preferred orientation of crystallites in the [111] direction in palladium thin films occurs in all investigated cases. Only (111) and (222) diffraction lines were observed. We have also observed a considerable shift of all diffraction lines in comparison with Pd standard (JSPDS). The shift is the same for both (111) and (222) diffraction lines which indicates the presence of macrostrains in our thin films which increase with increasing palladium layer thickness. More detailed investigation of positions of diffraction lines showed that the shift of diffraction lines is caused not only by macrostrains but also by stacking faults in the (111) planes.
The comparison of the intensities and widths of diffraction lines indicates that probably due to single-line growth of layers, the X-ray scattering will be not in accord with the kinematical theory of X-ray diffraction. Due to this fact the results of line profile analysis can be considered only as an informative.
It has been found that there are very small microstrains in all investigated thin films that at the beginning decrease with an increase of film thickness and later they increase again. The dimensions of areas of coherent scattering increase with increasing film thickness.
The influence of substrate is more presented in the integral intensities of diffraction lines and in the macrostrains while it does not influence the dimension of areas of coherent scattering and microstrains (The influence of stacking faults on actual dimension of areas of coherent scattering due to their very low probability is small). We assume that the difference in the integral intensities influenced by substrate can be caused by the fact that single areas of coherent scattering do not scatter the X-rays incoherently as it is predicted according to the kinematical theory of X-ray diffraction.
This assumption is confirmed also by the results of STM measurements where we can observe that palladium thin films are created by large grains which consist of small grains (average diameter is about 10 nm in the case of 250 nm thick Pd layer). This is in a good agreement with the results obtained by X-ray diffraction analysis. The maximum values of macrostrains reach up to 100 MPa.
Different shape of the line profiles and the omega-scan curves while using single crystalline [Si(111)] and amorphous (glass) indicate that palladium thin films have inhomogeneous distribution of microdeformations which is different with respect to distance from the substrate. The most appropriate function for fitting the omega-scans is Pearson VII function which indicates that the distribution of orientation of crystallites is not Gaussian.
1. Arbab A., Spetz A., ul Wahab Q., Willander M. and Lundstr”m I.:
Sensors and Materials, 4, 4(1993) 173-185
2. Arbab A.- Spetz A. and Lundstr”m I.: Sensors and Actuators B,
15-16 (1993) 19-23
3. Langford J.I., Boultif A., Auffredic J.P. and Louer D.: J. Appl.
Cryst. (1993) 26, 22-33