In the summer of 2008, I was contacted by a member of Dr. Alexander Moewes' beamteam to diagnose and repair software used for their research. This software, designed to simulate the spectra from Resonant Inelastic X-Ray Scattering (RIXS) measurements at the Canadian Light Source, contained an error producing results not in agreement with the experimental measurements. Was the problem a bug in the code, bad physics, or both? I was hired to examine the software and resolve the issue.
The software was written in C++ and had very little supporting documentation. As the original developer was unavailable I had to decypher what the program did and whether it was doing it properly. After analyzing and testing the code I was able to determine that the problem was an incorrectly labelled energy state during the calculation of the Kramers-Heisenberg formula.
After making the necessary corrections the software was able to produce results which agree with experiments. The figure to the right, from Sanjukta Choudhury's M.Sc. thesis, compares measured XAS spectra with those calculated with the revised software.