Does NISTMonte work?

A fair question. Let's examine some evidence.

Backscatter coefficients

While there are many subtle pitfalls to measuring the backscatter yield, the backscatter yield does show a characteristic variation with atomic number and can be readily modeled using NISTMonte. Modeling backscatter yield is far easier than measuring it. Assume a planar sample of essentially infinite extent made of a single element. Interestingly the backscatter yield does not depend upon sample density (actually darn little does.) We need to run a fairly large number (tens to hundreds of thousands) of electrons to get sufficiently precise results (think ). The experimental values come from Heinrich KFJ. in X-Ray Optics and Microanalysis. Castaing R, Deschamps P, Philibert J (eds). Herman: Paris, 1966. The inelastic scattering cross section models are: (1) Jablonksi A, Salvat F, Powell CJ. NIST Electron Elastic-Scattering Cross-Section Database – Version 3.1. National Institute of Standards and Technology, Gaithersburg, MD, 2003; (2) Czyzewski Z, MacCallum DO, Romig A, Joy DC. J Appl. Phys. 1990; 68: 3066-3072; and (3) the screened Rutherford model described in Henrich K. Electron Beam X-ray Microanalysis. Van Nostrand Reinhold Company: New York, NY, 1981.

Figure 1:Comparing Heinrich's backscatter yield at 10 keV with various different elastic scattering cross section models.

Figure 2:Comparing Heinrich's measured backscatter yield at 20 keV with various different elastic scattering cross section models.

Measured φ(ρz) curves

Another quantity that can be both modeled and measured is the shape of the φ(ρz) curve. The example presented here compares Henoc's data for Al K-LIII with NISTMonte for various different elastic scattering cross sections. The electron energy loss model also plays an important factor. The data presented here uses Joy-Luo's adaptation (Joy DC, Luo S. Scanning. 1989; 11: 176-180) of Bethe's energy loss expression. NISTMonte uses a continuous slowing down model. This may explain why Henoc's data shows a higher tail at large depths than the modeled result. In reality, electron energy loss shows variability leading to stragglers at large depths. Otherwise there is a gratifying level of similarity.

Quantification

Coming soon...