Skip to main content
. 2021 Jun 7;4:84. doi: 10.1038/s42004-021-00512-3

Fig. 1. Experimental comparison of stochastic and monochromatic measurements.

Fig. 1

The recovered Fe 1s2p RIXS planes using the polychromatic SASE beam is shown in panel a while the same obtained with a monochromatized SASE beam is shown in panel b. The spectra were recovered from 42,000 shots corresponding to ~24 min of measurement time at 30 Hz in both cases. Solid black contour lines indicate an increase of RIXS intensity by 2σ, where σ is the standard deviation of the predictive distribution estimated for a given emission/incident energy point. These contour lines can be treated as an estimate of relative significance, where peaks of interest are separated from surroundings by many contours. A slice at constant emission energy (6.4033 keV) is shown in panel c over a much larger incidence energy range. For both measurements, the XFEL central energy is fixed and scanning of the mono was limited to 7.100–7.140 keV. The SASE spectral intensity is centered on the Fe pre-edge region and tails off above 7.130 keV. Error bars in panel c likewise indicate 2σ bounds of the solution.