Advances in X-ray Energy Dispersive Spectroscopy in the Modern Analytical Electron Microscope

The use (and abuse) of computationally mediated imaging and diffraction experiments at the
nano-to-pico scale continues to increase in modern electron optical instruments. This is
particularly relevant in the latest generation of aberration corrected electron optical instruments
where advanced configurations are pushing the limits of what we in the past only dreamed of
routinely accomplishing. However, at the same time, only incremental advances have been
done to augment/improve our analytical capabilities. Spectroscopically, the most common
signals are derived from inelastic scattering creating either electrons (EELS), photons (CL) or
x-rays (XEDS). Each of these three have their own merits and they all complement each other
occasionally having overlapping in functionality. Regardless of which of these three
measurements are chosen, the sensitivity of each is directly tied to the ability to detect a
characteristic signal above some continuum background from which a quantitative measurement
can be realized. Since 2004, Argonne has been working on designs and implementations of
linear arrays of Silicon Drift Detectors (SDD's) for x-ray energy dispersive spectroscopy (XEDS)
in the Analytical Electron Microscope (AEM). The goal to maximize the solid angle, sensitivity
as well as mitigate artifacts. Evolving our original π steradian detector solution [1] and
collaborating with ThermoFisher Scientific [2], we have developed the X-ray Perimeter Array
Detector (XPAD) which when combined with a custom electron optical pole piece (ZTwin) has
improved the detector performance to a record collection angle of 4.5 sR. This magnitude of
this improvement is illustrated in Figure 1 which plots the evolution of the experimentally
measured solid angle of detectors on Analytical TEM/STEM instruments over the last 40 years.