Adaptive Background Correction of Crystal Image Datasets: Towards Automated Process Control

Kiernan, Luke; Jones, Ian; Kurki, Lauri; Cullen, Patrick J.; Arnaout, Toufic El

doi:10.1007/s11220-020-00310-6

Cited by 2 publications

(1 citation statement)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They can also be used to distinguish between different crystal shapes . However, because of the inherent nonuniqueness of the measured chord for a given crystal, it is difficult to evaluate sizes of nonuniform particles. , Most methods yielding shape information rely on two-dimensional (2D) image analysis. , These techniques allow for an estimation of the size and shape of particles. Assumptions regarding morphology may help to make more accurate measurements. , However, because of the limitation to two dimensions, exact results are still difficult to achieve.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Nonideal Shape Evolution during Potash Alum Crystallization Using Microcomputed Tomography and Three-Dimensional Image Analysis

Schiele

Antoni

Meinhardt

et al. 2021

Crystal Growth & Design

View full text Add to dashboard Cite

Nowadays, it is common to analyze crystallization processes and crystalline products using two-dimensional image analysis. Various techniques exist but they are not fundamentally capable of capturing the full morphology of particles due to their limitation in two dimensions. This is particularly true when complex shapes, e.g., through agglomeration or broken crystals, occur. Here, an approach is presented in which potash alum crystals are sampled from a laboratory-scale reactor at six time points over the course of a crystallization process. Three-dimensional (3D) images of all crystals in the samples were obtained by microcomputed tomography and used for morphological characterization. The method directly yields volume and surface area distributions without the need for any assumption regarding particle morphology. Applying geometric crystal models allowed for a more detailed analysis of the crystals. In the example considered, it was shown that most crystals assumed nonideal shapes over the course of the process. The supporting model provides indication that the shapes approach ideality through face-independent crystal growth. Overall, more than 11 000 crystals were analyzed. In general, this work aims at demonstrating the potential of crystal analysis by means of microcomputed tomography and 3D image analysis.

show abstract