Angle-dependent XPS

A(E) is the inelastic mean free path at the prevalent kinetic energy E through 

0 is the take-off angle with respect to the surface normal d is the thickness of the overlayer. 

Note that 2 (ES) is the inelastic mean free path of electrons formed in the substrate traveling through the overlayer. In the case that the overlayer is a film of Si02 on a silicon crystal (as in Fig. 3.14), Eq. (3-9) reduces to 

It is possible to use Eqs. (3-9) and (3-10) to evaluate the thickness from only one XPS spectrum at a known take-off angle, 8. However, as Fadley already pointed out in his 1976 review [39], several assumptions must be made to derive Eq. (3-9). The most important of these is that the overlayer has a homogeneous composition, a uniform thickness, and a flat morphology. In particular, the latter assumption appears critical. 

It is concluded that XPS is an excellent tool for determining layer thickness in the range of a few nanometers, provided that one measures the full take-off angle dependence, in order to test the applicability of the uniform layer model implicitly assumed in Eqs. (3-9) and (3-10). 

---

In principle, one can take the interpretation further and calculate what oxygen concentration profile fits the measurements of the O/Ag ratio best. In fact, Baschenko et al. [39] did this and concluded that the subsurface oxygen resides mainly in the third and fourth atomic layer below the surface. Although the result appears plausible, it should be noted that such calculations are only permitted when the surface satisfies the requirements of lateral homogeneity and absence of roughness discussed above. As the O/Ag experiments were done with polycrystalline foils, one might wonder whether too detailed an analysis is warranted. Anyhow, the work forms a nice illustration of what angle-dependent XPS can achieve on catalytically relevant adsorbate systems. 

Measurement of such small thicknesses is difficult. Ellipsometry [3] and angle-dependent XPS [5] may be used with care to avoid complications caused by surface roughness. 

XPS spectra were recorded using unmonochromatized Mg K radiation (1253.6 eV), and an unmonochromatized He-resonance lamp was used for ultraviolet photoelectron spectroscopy (UPS). XPS spectra were taken with an analyzer resolution of 0.2 eV, and the net resolution measured as the full width at half-maximum (FWHM) of Au 4f(7/2) was 0.9 eV. The spectrometer is of our own construction and is, e.g., designed to provide optimum angle-dependent XPS or XPS(0) (12,l4). For high 5-values, the photoelectrons leave the sample surface near the grazing angel, and due to the limited escape depth of the electrons, this is a "surface sensitive" mode. In the "bulk sensitive" mode, for low 0-values, the photoelectrons exit near the surface normal, and hence more information from the "bulk" of the sample is obtained (15). 

Alkylthiol-substituted di- and terthiophene do not sclf-assemble on rough Au layers. This results from angle dependent XPS (X-ray photoemission spectroscopy) [105] and NEXAFS measurements [106] which do not indicate any order in the system. One possible explanation is the interaction of the thiophene sulphur with the Au substrate which is not negligible if compared to the thiol sulphur/Au and alkyl-chain interactions. On the other hand, very recent experiments indicate highly ordered self-assembled films prepared under the same conditions but with flatter Au-substrates [107]. Another explanation for the fonner results is therefore a strongly reduced interaction of the relatively short alkyl-chains due to the surface roughness. 

Fig. 5.6 F/C intensity ratio in function of depth in symmetrical blend of fluoro a,(D-end-functional protonated polystyrene (a,(D-/ PS(Rf)2) and deuterated PS, measured by angle dependent XPS. Reproduced with authorization from [57]...

Tyler, B.J., Castner. D.G. and Ratner, B.D. (1989) Regularisation — a stable and accurate method for generating depth profiles from angle dependent XPS ckita. Surface and Interface Analysis, 14, 44,3-450. 

It was found that the protected hexasaccharide forms homogeneous monolayer films with a density of molecules at the surface that is roughly one order of magnitude lower than that of alkanethiols. The concentration and time necessary to deprotect the layer at the surface, as well as the observed contact angle and angle-dependent XPS measurements, indicate contorted molecules at the surface (possibly the onset of the helical amylose structure) with consequent limited accessibility to the protective groups. 

---