An empirical correction for the influence of low-energy contamination

Low-energy contamination caused by focusing multilayer optics has been known for a long time. So far, the only method to avoid this problem is attenuation by interposition of a low-density material foil into the beam. However, attenuation lowers the intensity, which might be crucial, especially for charge density data collection, and is of course not possible for already measured data. In this article, an empirical correction is proposed as an alternative or addition. The low-energy contamination mainly affects the reflections with indices 3h 3k 3l. In the program SADABS, a 3 lambda correction in analogy to the earlier 2/lambda correction is now implemented. The correction factor can be determined from the model, but for most cases, a default value depending on the experimental settings is acceptable. Alternatively, the correction can be refined in SHELXL, similarly to a twin refinement. An investigation of several data sets collected with and without attenuation revealed that the empirical correction and attenuation lead to similar improvements. The empirical correction is of special interest for charge density investigations, where the loss of intensity could hinder the data collection up to very high resolution. Here, the improvements in the model, especially for the noise level of the residual density, are even more pronounced.