| Mg | Al | Si | Zr | Y | |
|---|---|---|---|---|---|
| Nominal | 16.5 | 33.9 | 41.4 | 4.2 | 4.0 |
| EDXS pristine | 16.8 | 33.7 | 41.1 | 4.5 | 3.9 |
| EDXS annealed | 11.8 | 17.2 | 43.1 | 2.7 | 25.2 |
a Oxygen not considered because of self‐absorption of the low‐energy O‐K X‐rays.
In view of the small excitation volume in TEM experiments, it is rather straightforward to compute the relative intensity ratios of the element‐specific peaks in an EDX spectra and thus to get a quantitative element distribution of tiny areas within a sample once the system is properly calibrated. In this way, crystallization‐induced changes in the composition of the residual glass can, for instance, be investigated, even if the lateral dimensions of the residual glassy areas are a few nm3 only. As shown by Table 1, the STEM‐EDX calculated compositions of the pristine green MAS glass are in excellent agreement with the nominal data, whereas important changes are observed after heat treatment for Zr and especially for Y.
3.3 Electron Energy Loss Spectroscopy
In analytical (S)TEM, an additional useful source of information concerns the nature, valence, and coordination of atoms, which affect specifically the energy loss undergone by the primary beam through inelastic scattering by electron clouds within the sample. As in X‐ray absorption spectroscopy (XAS) or, more precisely, X‐ray near‐edge structure spectroscopy (XANES) (see Chapter 2.2), the interaction of a given atom with its local environment, i.e. the influence of the nearest neighbors on its electronic structure, is thus probed with an appropriate spectrometer mounted below the sample, which filters electrons according to their energy loss. This energy loss near‐edge structure spectroscopy (ELNES) has certain advantages, such as a superior spatial resolution in comparison with XANES. In addition, it is applicable to light elements such as Li, Be, or B, whereas appropriate cross sections for X‐ray generation typically restricts EDXS to elements heavier than B. On the other hand, quantification of electron energy loss (EEL) spectra is not as straightforward as in EDXS, and one is restricted to energy losses lower than ≈2 keV with EELS. Thus, it complements nicely with EDXS, whose typical spectral domain for useful application starts at approximately 2 keV.
Figure 8 Al‐L2,3 edge EELS spectra of MAS sample areas that represent either the residual glassy part or the spinel therein.
As an example, the excitation of aluminum electrons from the 2p level to empty states above the band‐gap energy gives rise to Al‐L2,3 edge EEL spectra where the presence, relative intensity, and energy position of specific features (A–F) bring information on Al coordination (Figure 8). For spinel, the spectrum is, for instance, typical of Al in octahedral coordination [26], which is not observed in the residual glass matrix whose detailed interpretation is actually more complicated because the features in the 100–110 eV energy range are convoluted with the Si‐L edge signal. Although they are beyond the scope of this chapter, elaborate computation techniques nonetheless exist with which information on coordination can be derived with an increasingly high precision from the intensity and positions of the peaks in EEL spectra [5, 27]. Not all ionization edges are well suited for in‐depth analyses, however, the so‐called white lines of transition metals being generally favorable cases.
4 Scanning Probe Microscopy
Owing to the inherently nonperiodic arrangement of atoms in glass, use of scanning probes is of special interest for studying the topography of a sample surface and inspecting, for instance, its roughness after polishing. Besides, information on the structure of a crystal present at the surface can be obtained, especially if some relaxation or reconstruction phenomena have caused a local change of its periodic arrangement.
Among the probe microscopy techniques that have been designed, atomic force microscopy (AFM) is probably the most widespread. Rather than describing its noncontact or tapping modes, we will restrict ourselves to the basic principle of its contact mode, which is sketched in Figure 9. When a cantilever is moved against the sample surface in an x–y raster scan, it is deflected differently depending on the local roughness or composition of the sample surface. Because the deflection can be monitored precisely via the reflection of a laser beam on the cantilever head, it can be kept constant with an appropriate electronic feedback loop (constant‐force AFM). If this force is modulated in the z‐direction with a certain oscillation frequency, the sample surface resists the oscillation, so the cantilever bends according to the stiffness of the probed sample area. With this force‐modulation approach, the information obtained thus goes farther than surface topography since a certain measure of the surface elasticity is also obtained.
Again the MAS sample illustrates the results that can be obtained with AFM in terms of surface topography and hardness (Figure 10). For a polished surface, identification of the dark (glass) and bright (spinel or zirconia) areas apparent in the micrograph has to rely on SEM or (S)TEM experiments, but the force‐modulation micrograph yields information not available by other means on the much greater hardness of the crystalline phases compared with that of the glass matrix.
Figure 9 Working principle of the atomic force microscope in the contact mode.
Figure 10 Topography (left) and hardness (right) contrasts between spinel and the glass matrix of the MAS sample in AFM micrographs. (a) Unetched, polished sample and (b) superficially etched sample. Hardness contrast derived from a mapping of the cantilever amplitude damping.
Because different phases are not attacked by acid solutions at the same rate, etching of glass‐ceramic surfaces by HF or HF–HNO3 solutions can enhance the microstructure
