D-shell of iron atom of the amorphous FeCr 15 B 15 alloy effective charge change during the crystallization

. An amorphous metal alloy of the FeCrB system was studied during the crystallization by thermal annealing. Such an alloy is a perspective candidate for the role of an intermediate layer in multilayer covering for cutting tools. By the using of the thermoelectric voltage measurement, positron annihilation spectroscopy, and X-ray photoelectron spectroscopy, the conjoint research was performed for the study of the conduction and d-electron band state in the amorphous metallic alloy FeCr 15 B 15 , which intersects each other by the energy. The results of all the studies agree with each other and indicate the change in the effective charge of the d-shell by 1 electron during crystallization.


Introduction
Nowadays high magnetic and electric properties of metal glasses are extensively used in technical branches. Power transformers and magnetic heads production based on the amorphous metal alloys and Finemet-type alloys, manufactured on their basis, is growing rapidly every year. At the same time, the development of new ways of for obtaining an amorphous state such as gas-phase deposition [1][2][3][4] and selective laser melting [5,6] provides a wider usage of high strength properties of metal glasses. The increase of an amorphous alloys temperature stability leads to a broadening their application range, including the using as a covering for cutting tools [7][8][9]. Possible using of metal glasses as an intermediate layer in multilayer nanostructured coatings is also actual [10][11][12]. An important advantage, in this case is the absence of crystal structure that makes it possible [13][14][15][16] for them to serve as a transition layer between the substrate and the nanostructured covering, damping the difference between crystal structures of the substrate and the covering, and the difference between their thermal expansion coefficients. In many cases, these processes lead to destroying, cracking and chipping of the covering [3][4][5].
The results for the amorphous metal alloys of other compositions were obtained earlier [17][18][19] by the similar methods. This research supplements the works [17,18]. This research was performed on the material of the same system as [18], but other composition, and was supplemented by X-ray photo-electron spectroscopy (XPS). It should be noted that the results of all the research of materials with different composition correlate with each other.

Methods and materials
Samples of FeCr 15 B 15 alloy obtained by rapid quenching (casting the melt on a cooled substrate [17][18][19]) were prepared to perform this research. The samples were annealed in a vacuum furnace at the different temperature for one hour. Annealing was controlled by the X-ray diffractometer. Before the crystallization, the samples showed a halo typical for the amorphous alloys and after crystallization À diffraction lines of FeCr solid solution, which testifies of the nanocrystalline phase formation [17][18][19][20].
The samples were studied by positron annihilation spectroscopy, in particular by measuring angular correlation of annihilation radiation (ACAR).
Positron annihilation spectroscopy methods are based on irradiation of the test sample by positrons and registration of annihilation photons. Measuring of the photon emission delay concerning positron getting into the medium allows one to measure electrons' density in the annihilation area. The deviation of annihilation photons from 180°makes it possible to measure annihilated electron momentum projection on the chosen axis. It should be noted that due to its positive charge the positron is actively captured by negatively charged matrix traps, such as bulk defects as dislocations, vacancies, nanoparticles with high positron affinity, etc. It causes annihilation of more than 50% of positrons in defects even at small defects concentration. Therefore, this method is highly sensitive towards the state of the defects structure. A more detailed application of positron annihilation spectroscopy for studying defects structure is described in [19]. It was shown earlier [17,20,21] that the study of ACAR spectra makes it possible to judge about the occupancy changes in the d-shell of iron group metals. ACAR spectra were measured at the 1-D ACAR apparatus with the angular resolution of 1 mrad of the Institute for Theoretical and Experimental Physics named by A.I. Alikhanov. A detailed description of methods for data acquisition, spectra treatment and interpretation of transition metal alloys spectra is provided in [17][18][19].
XPS was used in this research to study energy distribution of the valence band electrons and to determine the position of the characteristic lines, which allows one to judge about the environment of the atom under study. XPS spectra were acquired by K-alpha spectrometer manufactured by Thermo fisher scientific.
Thermoelectric voltage measurement is highly sensitive towards the changes in density at Fermi level and in the population of upper electron shells [20]. Modified microhardness tester with controlled contact effort was used for this measurement. A tungsten needle was placed into the holder. The needle was heated up to the set temperature by resistive heating. The samples were studied within the temperature range from 50 to 250°C. Contact potential difference was measured by a bridge circuit with the error of not more than ±0.01 mV. Calibration was performed for pure metals. Thermo electromotive force changes drastically at 400-450°C which corresponds, in line with the results of X-ray diffractometry, to the formation start of nano-sized nucleus of FeCr solid solution [21][22][23].

Results and discussion
ACAR spectra are shown in Figure 2. The spectra were approximated by the least squares method with the combination of three Gaussians and an inverted parabola considering apparatus and background resolution functions. The results are given in Table 1.
One can obtain more detailed information on the interpretation of transition metals ACAR spectra and the approximation method from the articles [24,25]. It was mentioned earlier [17][18][26][27][28][29][30] that the intensity of the 15-18 mrad wide Gaussian-3 correlates with the d-shell occupancy of transition group metals. The above-mentioned works contain the study of positron annihilation in transition metals. A wide Gaussian with the width of 16 ± 1.5 mrad, the intensity of which, as was shown, correlates with the d-shell occupancy, was observed in all spectra of transition metals. Parabola with the 5-6 mrad cutoff angle convolved with the 2-5 wide Gaussian is appeared in the spectra.
Considering the occupancy of this iron atom shell, which is known from the sources [31][32][33], it is possible to estimate the effective charge change during the FeCr 15 B 15 alloy crystallization as 1 electron.
Samples of FeCr 15 B 15 alloy in amorphous and crystallized states were prepared to study the the electron structure of amorphous and crystalline states using XPS. A detailed description of the samples preparation method is provided in [17]. Moreover, samples of pure iron and chrome metal were studied to obtain the basis for the valence spectra interpretation of amorphous and crystalline alloys. Amorphous samples were annealed in vacuum to obtain crystalline state. Before the study, the surface was etched by ions with the energy of 3 keV, the current being 5 mA. The etching continued until the oxygen line intensity on the panoramic spectrum fell below the level of 5%. Spectra of valence band within the range of À3-+20 eV were studied.  Initial spectra are presented in Figures 3 and 4, peak intensities are given in Tables 2 and 3. XPS spectra processing and approximation method are described in [34,35]. Valence spectrum of transition metals is defined by d-electrons, as the 4-s shell has little intensity, broad distribution and merges into the background.
The area of the amorphous sample under the valence band peaks is 14.6% higher considering the setting concerning Fe2p peak. Electron density distribution of the amorphous alloy in the valence spectrum is shifted towards the higher binding energies which correspond to bound states. The peak in the area of 9.5 eV that corresponds to the bound state of the ironbased amorphous alloy is increased. Fe3p level shift is 0.3 eV for the annealed sample, and 1.06 eV for the amorphous one. Binding energy increase of the amorphous iron Fe3s level is 0.45 eV. This value is lower for the 3p chrome peak in the FeCrB alloy and makes 0.3 eV. It demonstrates, taking into account of the research data [35][36][37], that the main changes during the amorphization occur in the electron structure of an iron atom. 3.2 ± 1 3.5 ± 1.5 10.7 ± 0.7 73 ± 4 6.2 ± 0.7 2.1 ± 0.6 9.5 ± 0.7 17.8 ± 0.5 13.6 ± 1 FeCrB-600 2.9 ± 1.5 4.5 ± 2 10.1 ± 0.4 69 ± 2 5.3 ± 0.5 4.9 ± 0.9 9.5 ± 0.9 18.2 ± 0.3 11.3 ± 0.7 FeCrB-700 3.4 ± 2 3.3 ± 0.5 10.4 ± 0.9 76 ± 2 5.4 ± 0.9 4.8 ± 1.5 7.5 ± 0.7 17.9 ± 0.8 13.1 ± 1 This research makes it possible to outline the ways for amorphous metal alloys implementation as intermediate damping layer of multilayer hard-alloy coverings. Results obtained by conjoint methods of X-ray photoelectron spectroscopy, angle correlation positron annihilation radiation and the measurement of thermo electromotive force suggest that during the thermal crystallization of an amorphous metal alloy efficient charge is redistributed between the iron atom d-shell and the conduction band.
Results obtained by all the methods used in this article agree with each other well and point at the change in effective redistribution of the iron atom d-shell charge during the crystallization and the formation of the crystalline phase nano-sized nucleating seeds in the amorphous metal matrix. According to the results of the angle correlation annihilation radiation, 20 ± 5% (1 electron, according to estimates) of iron atom shell effective charge is redistributed. According to the results of the Xray photoelectron spectroscopy such redistribution made 14.6 ± 2%. These measurements in the thermo electromotive force do not allow to estimate the quantity and the direction of the effective charge redistribution; however, they agree with the aforementioned ones and make it possible to specify the process activation temperature: 400-450°C.