Fig. 2: Comparison between experimental values and predicted values via the Machine Learning algorithm we label “rescaled 10-CV” (a Geometric Harmonics algorithm).
From: Machine learning-assisted crystal engineering of a zeolite
a Si/Al ratio via ICP analysis; the green square represents outcome from an input region we selected to explore guided by our surrogate model to enhance the Si/Al ratio to values higher than 3, (b) log10(Particle size) measured from SEM/TEM images, (c) log10(Crystal size) via XRD patterns peak-broadening analysis in accordance with the Scherrer equation using FAU(311) and FAU(331) reflections (only crystal sizes smaller than 60 nm are included), (d) FAU fraction via deconvolution of the first main diffraction peak in XRD patterns, namely the peak area ratio of FAU(111)/(FAU(111) + EMT(100)), (e) Uptake values at P/P0 = 0.01 via Ar-adsorption isotherms, (f) log10(particle size to crystal size ratio), lower value means lower aggregation of FAU particles. Blue dots represent training points for model identification, red dots represent testing points for the identified model, and green dots represent prediction points toward obtaining high-silica FAU zeolites. Entry numbers for dots in (a–f) are labelled in Supplementary Fig. 28. Fewer points were involved in (c) and (f), since we only consider entries with crystal sizes smaller than 60 nm for the crystal size analysis.
