Detection of Nucleation during Cooling Crystallization through Moving Window PCA Applied to in Situ Infrared Data

Abstract

The method here proposed based on the moving window principal component analysis can be employed in timevarying processes tracked through in situ spectroscopy. In the case under investigation, it aimed to detect the nucleation during crystallization processes. For this purpose, statistical indexes were employed, and the contribution plot helped identifying the spectral variables that were changing due to nucleation. Isonicotinamide was here considered as a model active pharmaceutical ingredient, and its cooling crystallization was monitored by means of in situ infrared spectroscopy. The procedure allowed to overcome issues that may be encountered with static principal component analysis, since it could distinguish the slow-varying changes due to external perturbations (temperature) from abnormal events such as the sudden concentration decrease related to the crystallization. The proposed method was demonstrated to correctly detect nucleation without any a priori knowledge of the peaks involved in the process, leading the false alarm rate from 77.38% (obtained with the static principal component analysis) to 6.9%.

Solubility and Crystallization of Piroxicam from Different Solvents in Evaporative and Cooling Crystallization

“…Nevertheless, other process parameters should be considered as these will also influence the polymorphic form. It was observed in our previous study that the formation of INA polymorphs also depended on the solute concentration as well as the temperature at which nucleation was onset [7,8]. The complex nature of the nucleation of polymorphs has been demonstrated in our previous work by the cooling crystallization of piroxicam from acetone-water solutions [9].…”

Applications of Artificial Intelligence and Machine Learning Algorithms to Crystallization

“…performed PCA on Raman spectroscopic data of vitamin B12 to assess the purity of crystals from their spectra. Other uses of PCA include solvent selection and in situ nucleation detection …”

“…Other uses of PCA include solvent selection 92 and in situ nucleation detection. 93 PCA identifies the subspace in which the data approximately lies. An example of (preprocessing) workflow starts with a data set denoted by {x (i) ,i = 1, .…”

An adaptive soft sensor deterioration evaluation and model updating method for time-varying chemical processes

“…Additionally, high-dimensional auxiliary variables in chemical processes result in highly complex models and often incomplete model structures [56]. The moving window method can obtain the latest data representing the current process [57] and adaptively update that data [58], and the cosine similarity correlation calculation method can effectively measure the correlations between vectors [59] and identify the effectiveness of the influences [60]. Therefore, a sample data updating method using moving window--cosine similarity-based soft sensor modeling is proposed to update the sample datasets of soft sensor models for chemical processes and improve their prediction performance.…”