Filtern
Erscheinungsjahr
- (1)
- 2023 (1)
Schlagworte
- AI (1)
- Anorganische Analyse (1)
- ICP-Massenspektrometrie (1)
- OCSR (1)
- Polymere (1)
- artificial intelligence (1)
- machine learning (1)
- optical chemical structure recognition (1)
Institut
The conventional quantitative method for the analysis of inorganic elements in polymer matrices is a complex and time consuming process that presents a significant risk for error. Typically, polymers are digested in a microwave oven or other devices under high temperature and pressure for several hours while employing different mixtures of high purity acids. In many cases, particularly when high concentrations of doped elements are present, the digestion is often incomplete and therefore the reproducibility depends strongly on the type of polymer and additives used. A promising alternative technology that allows for the direct analysis of these polymers without digestion is laser ablation ICP-MS. Due to a lack of available reference materials and the presence of matrix dependent effects, a precise calibration cannot be obtained. In order to compensate for the matrix dependent effects the use of internal standardization is necessary. In this study the correlation between the carbon released during the ablation process and the 13C signal detected by ICP-MS and its use as an internal standard are investigated. For this purpose, twenty-one virgin polymer materials are ablated; the released carbon is determined and correlated with the corresponding integrated 13C signal. The correlation resulted in a direct relationship between the ablated carbon and 13C signal demonstrating the potential ability to neglect at least some of the matrix dependent and transport effects which occur during the laser ablation of virgin polymers.
The number of publications describing chemical structures has increased steadily over the last decades. However, the majority of published chemical information is currently not available in machine-readable form in public databases. It remains a challenge to automate the process of information extraction in a way that requires less manual intervention - especially the mining of chemical structure depictions. As an open-source platform that leverages recent advancements in deep learning, computer vision, and natural language processing, DECIMER.ai (Deep lEarning for Chemical IMagE Recognition) strives to automatically segment, classify, and translate chemical structure depictions from the printed literature. The segmentation and classification tools are the only openly available packages of their kind, and the optical chemical structure recognition (OCSR) core application yields outstanding performance on all benchmark datasets. The source code, the trained models and the datasets developed in this work have been published under permissive licences. An instance of the DECIMER web application is available at https://decimer.ai.