Transformer-based deep learning models for adsorption capacity prediction of heavy metal ions toward biochar-based adsorbents

The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data. 15 pages, 5 figures

Biochar produced by thermal decomposition of biomass under oxygen-limited conditions has received increasing attention as a cost-effective sorbent to treat metal-contaminated waters. However, there is a lack of information on the roles of different sorption mechanisms for different metals and recent development of biochar modification to enhance metal sorption capacity, which is critical for biochar field application. This review summarizes the characteristics of biochar (e.g., surface area, porosity, pH, surface charge, functional groups, and mineral components) and main mechanisms governing sorption of As, Cr, Cd, Pb, and Hg by biochar. Biochar properties vary considerably with feedstock material and pyrolysis temperature, with high temperature producing biochars with higher surface area, porosity, pH, and mineral contents, but less functional groups. Different mechanisms dominate sorption of As (complexation and electrostatic interactions), Cr (electrostatic interactions, reduction, and complexation), Cd and Pb (complexation, cation exchange, and precipitation), and Hg (complexation and reduction). Besides sorption mechanisms, recent advance in modifying biochar by loading with minerals, reductants, organic functional groups, and nanoparticles, and activation with alkali solution to enhance metal sorption capacity is discussed. Future research needs for field application of biochar include competitive sorption mechanisms of co-existing metals, biochar reuse, and cost reduction of biochar production.