Calculating core level EELS of doped nanocarbons?

[Toma Susi]

Dear exciters,

I am involved in research on nanocarbon systems, specifically graphene and carbon nanotubes, that are doped with heteroatoms such as nitrogen, boron and phosphorus. An important tool in this research is electron energy loss spectroscopy in an atomic resolution scanning transmission electron microscopy (STEM-EELS) in a state-of-the-art Nion100 machine. To interpret and analyze the experimental spectra, a tool to calculate core level EELS spectra accurately is needed. 

I have been looking at a number of ways of doing this, and have some ideas how to do this at certain levels of approximation with other codes. However, I've understood that solving the Bethe-Salpeter equation in an all-electron calculation would be the most accurate way of doing this.

Does the exciting community have experience with these systems? Specifically, how computationally heavy are the Bethe-Salpeter calculations and how well does the code scale; that is, can exciting simulate sufficiently large unit cells to model e.g. graphene with a realistic doping level such as 1 dopant atom per 50 or 100 carbon atoms? Or is this completely unrealistic to expect?

Many thanks in advance,

Toma