Local ordering of nanostructured Pt probed by multiple-scattering XAFS

We present detailed results of a multiple-scattering (MS) extended x-ray absorption fine structure (EXAFS) data analysis of crystalline and nanocrystalline platinum. Advanced MS EXAFS analysis has been applied to raw x-ray absorption data including the background, using the expansion of the absorption cross section in terms of local two-body and three-body configurations. Present EXAFS results on bulk Pt are found to be in agreement with previous structural and vibrational data, and has been used as a reference for reliable structural refinement of nanosized systems. EXAFS structural refinement of Pt nanoparticles has been performed in combination with electron microscopy and x-ray diffraction, showing the importance of considering the actual size distribution and morphology of the samples. Present samples were unsupported and supported Pt nanocrystalline systems with size distributions showing clusters of quasispherical shape in the $1–7\mathrm{nm}$ range. In particular, EXAFS spectra have been analyzed accounting for the reduction of the coordination number and degeneracy of three-body configurations, resulting from the measured size distribution and expected surface atom contributions. The importance of a correct account of the reduction of the number of neighbors for calculating MS contributions is emphasized in the paper. EXAFS results have been found compatible with x-ray diffraction and transmission electron microscopy investigations. We estimate that EXAFS could be used to study cluster shapes only for sizes below $2\mathrm{nm}$ using present methods and quality of the experimental data. We have also shown that the local distribution of distances and angles probed by EXAFS is broader than in bulk Pt, with first-neighbor bond length variance and asymmetry increasing upon reducing the particle size. Methods and results presented in this paper have been found to be successful for a robust structural refinement of monatomic nanocrystalline systems and represents a solid starting point for analyzing subtle structural and dynamical local changes occurring during in situ experiments involving nanomaterials for specific applications like supported nanocatalysts.

Figure 1

(Top) TEM image of the Pt catalytic material scraped from the electrode (Pt2 and Pt3) and (bottom) comparison between nanoparticle size distributions obtained for Pt2/Pt3 and for Pt1.

Figure 2

(Color online) X-ray powder diffraction patterns of the nanocatalysts under consideration (Pt1, Pt2, and Pt3). Experimental data (points, blue) are compared with the calculations using Voigt functions (solid and dashed lines). The agreement with calculated spectra is excellent, and the residual curves are shown below each diffraction pattern.

Figure 3

SEM images of the Pt electrode surfaces: (top) $a\approx 0.1$ and (bottom) $a\approx 0.04$.

Figure 4

(Color online) The best-fit results of GNXAS analysis performed for Pt foil at Pt $L_3$ edge ($k^3$ weighted). Upper curves represent components of the model signal, vertically shifted for clarity: solid line, model signal; dashed blue line, experimental signal. The scale is indicated within the figure $\left(0.1{\mathrm{\AA }}^{-1}\right)$.

Figure 5

(Color online) Magnitude of the Fourier transforms calculated with a $k^3$ weight in the range of $k=[5.0;18.0]{\mathrm{\AA }}^{-1}$. (Top) The successive contributions vertically shifted for clarity (see also Fig. 4); the scale is indicated within the figure $\left(10{\mathrm{\AA }}^{-4}\right)$. (Bottom) The comparison between total theoretical signals and experimental one.

Figure 6

(Color online) Two- and three-atom configuration degeneracies (photoabsorber placed at the vertex between two shortest bonds) as a function of particle diameter, calculated using a model of quasispherical Pt clusters (solid lines) and hemispherical (111)-truncated Pt clusters (dashed lines). The variation ranges for degeneracy values (bands shown on the $y$ axis) used in the case of Pt2 sample are also presented. Two vertical dotted lines define Pt2 nanoparticle size limits resulting from TEM/XRD analysis (see Fig. 1 and Table ).

Figure 7

(Color online) The best-fit results of GNXAS analysis performed for Pt1 sample at Pt $L_3$ edge ($k^3$ weighted). Upper curves represent components of the model signal: solid line, model signal; dashed blue line, experimental signal. The scale is indicated within the figure $\left(0.1{\mathrm{\AA }}^{-1}\right)$.

Figure 8

Comparison between experimental $k\chi \left(k\right)$ spectra and the best-fit calculated signals obtained using multiple-scattering contributions for each sample. The scale is indicated within the figure $\left(0.1{\mathrm{\AA }}^{-1}\right)$.

Figure 9

Fourier transforms of total $\chi \left(k\right)$ signals calculated for the experimental EXAFS data presented in Fig. 8.

Figure 10

Two-dimensional contour plots presenting $E_0\text{−}{R}_1$ correlations obtained for fitting the EXAFS spectra of (top) Pt foil and (bottom) Pt3 (the noisiest spectra). The inner elliptically shaped curves represent 95% confidence interval.