Table 1 Parameters of the spectrum simulation.

From: Deep learning for automated materials characterisation in core-loss electron energy loss spectroscopy

Parameter

Value

Acceleration voltage

\(\in [60, 100, 200, 300]\) kV

Convergence angle

\(\in {\mathbb {R}} \text {; } \in [1, 20]\) mrad

Collection angle

\(\in {\mathbb {R}} \text {; } \in [1, 100]\) mrad

Energy dispersion

1 eV

Spectrum start

\(\in {\mathbb {N}} \text {; } \in [75, E_{first \, edge}-10]\) eV

Spectrum end

\(\in {\mathbb {N}} \text {; } \in [E_{last \, edge}+25, 3272]\) eV

Spectrum range (zero padding)

0 to 3071 eV

FWHM of zero-loss peak

\(\in {\mathbb {R}} \text {; } \in [1,3]\) eV

Plasmon energy

\(\in {\mathbb {N}} \text {; } \in [3,20]\) eV

Plasmon width

\(\in {\mathbb {N}} \text {; } \in [3,20]\) eV

Number of plasmon peaks

\(\in {\mathbb {N}} \text {; } \in [2,5]\)

Scattering parameter \(\frac{t}{\lambda }\)

\(\in {\mathbb {R}} \text {; } \in [0.1,1]\)

Jump-ratio

\(\in {\mathbb {R}} \text {; }\in [0.2,1.5]\)

Background exponent

\(\in {\mathbb {R}} \text {; }\in [2.0,4.0]\)

Background amplitude

\(\in [10^3,10^4,10^5,10^6,10^7]\)

Fine structure width

\(\in {\mathbb {N}} \text {; } \in [50,100]\) eV

Fine structure degree

\(\in {\mathbb {N}} \text {; } \in [10,25]\)

Maximum fine structure scale

10

Atomic diameter

\(\in {\mathbb {R}} \text {; } \in [25,250]\) pm

Chemical shift

\(\in {\mathbb {Z}} \text {; } \in [-5,5]\) eV

Instrumental shift

\(\in {\mathbb {Z}} \text {; } \in [-5,5]\) eV