生物炭的制备及其对氮肥吸附效果的研究

以玉米秸秆为原材料,分别在350、400、450℃下炭化3 h制备玉米秸秆生物炭(分别命名为B350、B400、B450),利用X射线衍射仪、扫描电镜仪、比表面积及孔径分布仪、红外光谱仪等对材料进行表征,确定材料的相关性能,并探讨其与氮肥吸附的相关性。以平衡实验法研究了溶液pH对生物炭吸附氨氮的影响。实验结果表明:玉米秸秆生物炭B350、B400、B450对氨氮的吸附过程均遵循二级吸附动力学模型,且采用Langmuir方程能更好地描述吸附等温线模型;玉米秸秆生物炭对氨氮的饱和吸附量按从大到小依次为B400、B350、B450,且均为有利吸附;玉米秸秆生物炭对氨氮吸附的最佳pH为8～12。     

集成式铁基质生物膜反应器自养反硝化深度脱氮

以污水厂处理水为研究对象，采用铁基质生物载体与生物膜耦合实现高效自养反硝化脱氮。考察停留时间（HRT）对系统脱氮效能的影响，通过动力学及微生物群落结构分析，揭示耦合技术的脱氮机理。结果表明：HRT为8 h，对一级A和一级B污水厂处理水，总氮(TN)平均去除率分别为95.41%和92.55%，TN处理负荷分别为0.48 kg TN/(m³·d）和0.58 kg TN/(m³·d），硝化过程氨氮(NH₄⁺-N)饱和常数分别为1.17 mg/L和0.72 mg/L，反硝化过程硝氮(NO₃^--N)饱和常数分别为0.87 mg/L和0.67 mg/L。出水水质分别达到《地表水环境质量标准》Ⅲ类、Ⅴ类水质标准。铁基质生物载体与生物膜耦合系统中微生物优势菌属为Maritimimonas、Rhodobacter和Sphaerotilus, 均为自养反硝化菌，证实了铁基质生物载体可为自养反硝化菌提供电子，实现生物自养反硝化脱氮。     

Effects of branched‐chain amino acid addition on chemical constituents in lychee wine fermented with Saccharomyces cerevisiae

This study investigated the impact of single branched‐chain amino acid addition (L‐valine, L‐leucine, L‐isoleucine) on nonvolatile and volatile constituents in lychee wine fermented with Saccharomyces cerevisiae MERIT.ferm. The individual addition of branched‐chain amino acids decreased the consumption of proline with significantly different rates. The lychee wine with added L‐leucine and L‐isoleucine had higher odour activity values (OAVs) of isoamyl alcohol, active amyl alcohol and isoamyl acetate. The lychee wine with added L‐valine had higher OAVs of isobutyric acid and isobutyl alcohol (5.55 and 2.98, respectively), which were lower in the wine with added L‐leucine (2.91 and 1.02, respectively) or L‐isoleucine (3.43 and 1.73, respectively) relative to the control wine (4.18 and 1.77, respectively), suggesting interactive effects among the amino acids. These findings demonstrated that lychee wine aroma can potentially be manipulated via the addition of selected branched‐chain amino acids.     

Effect of a halophilic aromatic yeast together with Aspergillus oryzae in koji making on the volatile compounds and quality of soy sauce moromi

Koji making is an essential step in the production of high‐quality soy sauce. In this study, we inoculated koji with a halophilic aromatic yeast (Zygosaccharomyces rouxii) and Aspergillus oryzae and investigated the effects on the volatile compounds and quality of soy sauce moromi. Our results suggest that the optimal yeast inoculation was approximately 1.5 × 10⁶ yeast per gram of koji, and excessive yeast greatly decreased A. oryzae enzyme activity. In comparison with the control group, there were no significant changes in the concentrations of amino nitrogen, total titratable acids, total soluble nitrogen and reducing sugar by extending the fermentation time during the first fermentation month, whereas the concentration of glutamate in the yeast‐inoculated sample increased by 46.16% and the arginine concentration decreased by 61.07%. After the second fermentation month, 17, 38 and 38 volatile components were identified by GC–MS in C1, C2 and C5, respectively. Moreover, a large number of volatile components, such as 2‐Methoxy‐4‐vinylphenol, ethyl acetate and 1‐Octen‐3‐ol were found in C2 and C5. And the total contents of volatile compounds in C2 were obviously higher than C5. This work furthers our understanding of the traditional multistrain koji making and provides a new method for enhancing the quality and flavour of soy sauce.     

有孔内基础平台的竖直孔空心砖结构设计探讨

孔内基础平台设置在现有模数空心砖孔洞的上部,在孔洞内形成小平台,其能阻止砂浆掉落孔洞内,使空心砖的保温功能得以实现。孔内基础平台的设置将砖缝层平面摩擦改变为曲面摩擦,每一个孔洞的孔内基础砂浆与砖缝层面砂浆合为一体,砖缝层面摩擦力提升,墙体抗震能力提升。     

Reactive Oxygen and Nitrogen Species in Defense/Stress Responses Activated by Chitosan in Sycamore Cultured Cells

Chitosan (CHT) is a non-toxic and inexpensive compound obtained by deacetylation of chitin, the main component of the exoskeleton of arthropods as well as of the cell walls of many fungi. In agriculture CHT is used to control numerous diseases on various horticultural commodities but, although different mechanisms have been proposed, the exact mode of action of CHT is still unknown. In sycamore (Acer pseudoplatanus L.) cultured cells, CHT induces a set of defense/stress responses that includes production of H₂O₂ and nitric oxide (NO). We investigated the possible signaling role of these reactive molecules in some CHT-induced responses by means of inhibitors of production and/or scavengers. The results show that both reactive nitrogen and oxygen species are not only a mere symptom of stress conditions but are involved in the responses induced by CHT in sycamore cells. In particular, NO appears to be involved in a cell death form induced by CHT that shows apoptotic features like DNA fragmentation, increase in caspase-3-like activity and release of cytochrome c from the mitochondrion. On the contrary, reactive oxygen species (ROS) appear involved in a cell death form induced by CHT that does not show these apoptotic features but presents increase in lipid peroxidation.     

Effect of lactic acid bacterial fermentation on amino acids and volatile compounds of pork trimming hydrolysate

The objective of this research was to improve the volatile flavour profile of a pork trimming hydrolysate by lactic acid bacteria (LAB) fermentation. Five probiotic strains and one dairy strain were used. The dominant aliphatic aldehydes (80.90%) in the unfermented hydrolysate were greatly reduced to <15% by converting into respective carboxylic acids and alcohols after fermentation. Strain GG was the only LAB that produced a significant level of succinic acid (1.62 mg mL⁻¹) and ketones (14.92%) (mainly diacetyl and acetoin) in the hydrolysate. Strains 299V and GG showed moderate proteolytic activities as their amino acid content reduced to 126.05 and 219.14 mg per 10 mL compared to the control (316.45 mg per 10 mL) after fermentation, respectively. However, little amino acid-derived volatiles were found in all fermented samples. These findings indicate that LAB fermentation modulates the flavour compounds of pork hydrolysates, which has the potential to be developed to novel flavouring materials.     

结合预训练模型和语言知识库的文本匹配方法

针对文本匹配任务,该文提出一种大规模预训练模型融合外部语言知识库的方法。该方法在大规模预训练模型的基础上,通过生成基于WordNet的同义—反义词汇知识学习任务和词组—搭配知识学习任务引入外部语言学知识。进而,与MT-DNN多任务学习模型进行联合训练,以进一步提高模型性能。最后利用文本匹配标注数据进行微调。在MRPC和QQP两个公开数据集的实验结果显示,该方法可以在大规模预训练模型和微调的框架基础上,通过引入外部语言知识进行联合训练有效提升文本匹配性能。