Search

Aeolian sand and loess are both natural materials with poor engineering-related properties, and no research has been devoted to exploring aeolian sand–loess composite materials. In this study, we used aeolian sand and loess as the main raw materials to prepare unfired bricks by using the pressing method, along with cement, fly ash, and polypropylene fiber. The effects of different preparation conditions on the physical properties of the unfired bricks were investigated based on compressive strength, water absorption, and softening tests and a freeze–thaw cycle test combined with X-ray diffraction and scanning electron microscope analysis to determine the optimal mixing ratio for unfired bricks, and finally, the effects of fibers on the durability of the unfired bricks were investigated. The results reveal that the optimal mixing ratio of the masses of aeolian sand–loess –cement –fly ash–polypropylene fiber–alkali activator–water was 56.10:28.05:9.17:2.40:0.4:0.003:4.24 under a forming pressure of 20 MPa. The composite unfired bricks prepared had a compressive strength of 14.5 MPa at 14 d, with a rate of water absorption of 8.8%, coefficient of softening of 0.92, and rates of the losses of frozen strength and mass of 15.93% and 1.06%, respectively, where these satisfied the requirements of environmentally protective bricks with strength grades of MU10–MU15. During the curing process, silicate and sodium silicate gels tightly connected the particles of aeolian sand and the loess skeleton, and the spatial network formed by the addition of the fibers inhibited the deformation of soil and improved the strength of the unfired bricks.

Sentiment-controlled text generation aims to generate texts according to the given sentiment. However, most of the existing studies focus only on the document- or sentence-level sentiment control, leaving a gap for finer-grained control over the content of generated results. Fine-grained control allows a generated review to express different opinions toward multiple aspects. Some previous works attempted to generate reviews conditioned on aspect-level sentiments, but they usually suffer from low adaptability and the lack of an annotated dataset. To alleviate these problems, we propose a novel pre-trained extended generative model that can dynamically refer to the prompt sentiment, together with an auxiliary classifier that extracts the fine-grained sentiments from the unannotated sentences, thus we conducted training on both annotated and unannotated datasets. We also propose a query-hint mechanism to further guide the generation process toward the aspect-level sentiments at every time step. Experimental results from real-world datasets demonstrated that our model has excellent adaptability in generating aspect-level sentiment-controllable review texts with high sentiment coverage and stable quality since, on both datasets, our model steadily outperforms other baseline models in the metrics of BLEU-4, METETOR, and ROUGE-L etc. The limitation of this work is that we only focus on fine-grained sentiments that are explicitly expressed. Moreover, the implicitly expressed fine-grained sentiment-controllable text generation will be an important puzzle for future work.