No abstract

No abstract

The present work aims to introduce a novel and eco-friendly method, i.e., a water-leaching pretreatment for extracting highly reactive biomass silica from rice husk (RH), for viability as a pozzolanic additive in cement. For comparison, the traditional acid pretreatment method was also employed throughout the experimental study. The silica from RH was extracted using boiled deionized water and acid solution as leaching agents to remove the alkali metal impurities, and then dried and submitted to pyrolysis treatment. The results indicated that potassium was found to be the major contaminant metal inducing the formation of undesirable black carbon particles and the decrease in crystallization temperature of amorphous RHA silica. The boiling-water-leaching pretreatment and acid-leaching pretreatment on RHs significantly removed the metallic impurities and reduced the crystallization sensitivity of RHA silica to calcination temperature. A highly reactive amorphous silica with purity of 96% was obtained from RH via 1 N hydrochloric acid leaching followed by controlled calcination at 600 °C for 2 h. The acid treatments increased the crystallization temperature of silica to 1200 °C and retained the amorphous state of silica for 2.5 h. In the case of water-leaching pretreatment, leaching duration for 2.5 h could yield an amorphous silica with purity of 94% and render the silica amorphous at 900 °C for 7 h. The RHA silica yielded by water-leaching pretreatment presented a comparable enhancing effect to that of acid leaching on hydration and improved the strength of cement. Furthermore, compared with the acid-leaching method, the water-leaching pretreatment method is more environmentally friendly and easier to operate, and hence more widely available.

Text classification is of importance in natural language processing, as the massive text information containing huge amounts of value needs to be classified into different categories for further use. In order to better classify text, our paper tries to build a deep learning model which achieves better classification results in Chinese text than those of other researchers’ models. After comparing different methods, long short-term memory (LSTM) and convolutional neural network (CNN) methods were selected as deep learning methods to classify Chinese text. LSTM is a special kind of recurrent neural network (RNN), which is capable of processing serialized information through its recurrent structure. By contrast, CNN has shown its ability to extract features from visual imagery. Therefore, two layers of LSTM and one layer of CNN were integrated to our new model: the BLSTM-C model (BLSTM stands for bi-directional long short-term memory while C stands for CNN.) LSTM was responsible for obtaining a sequence output based on past and future contexts, which was then input to the convolutional layer for extracting features. In our experiments, the proposed BLSTM-C model was evaluated in several ways. In the results, the model exhibited remarkable performance in text classification, especially in Chinese texts.

No abstract