Multidimensional Spectroscopy Data

Photoemission-based experiments provide a large amount of useful information. However, compared to microscopy techniques, the analysis of the results is often more complex and requires a pipeline of several data recording, conversion, and manipulation stages before they can be fitted with a model.

multi-1

multi-2The above figures represent the results of a photoemission experiment on a bismuth surface. In the left plot, the photoemission intensity encoded in the colormap is shown as a function of energy (y axis) and momentum (x axis). This plot involves normalization and averaging of the data as well as a non-linear axis transformations. The overlaid white markers represent features in the spectrum. These peak positions were determined by fitting a model consisting of three Gaussian peak profiles and a Fermi cutoff function. These models are shown as red lines in the right-hand plot. I wrote a script to perform all these data handling and fitting procedures.

The figure on the right side shows another example of a multidimensional data set. In this case, angle-resolved two-photon photoemission experiments reveal information about the unoccupied band structure of the same bismuth surface.

multi-3

Two-photon photoemission also provides access to the dynamics of the same Bi(111) surface. In the above plots, the photoemission intensity is shown as a function of energy and time on the scale of femtoseconds. On these timescales, we can observe the dynamics of electrons in solids. To the right and below the colormap are vertical and horizontal normalized slices of the data. The horizontal slices were fitted with a convolution of the Gaussian profile of the impeding laser pulse and the exponentially decaying population of the observed states.

Under certain circumstances, the dynamics reveal not only a decay in intensity with time, but also a stabilization in energy. This behavior is observed in the plot below. In order to quantify the dynamics, I wrote a script that fit the three-dimensional data to an exponential model.

multi-4

 

The data shown on this page were published in two separate articles: