【作者】 李自刚；

【导师】 黄为一；

【作者基本信息】 南京农业大学 ， 微生物学， 2006， 博士

【摘要】 以奶牛粪便为主要堆肥材料，辅以添加稻壳粉、锯木屑作为堆肥水分调节剂和膨松剂，采用高温好氧堆肥方式研究了堆肥过程中微生物多样性及物质转化关系。利用微生物可培养方法对堆肥过程中微生物群落演替规律进行了定量分析。研究发现农业有机固体废弃物堆肥过程是一个由细菌、放线菌、真菌等微生物共同作用并且在不同堆肥阶段由不同优势微生物群落互相演替的过程。在堆肥起始阶段细菌首先快速增长，迅速成为堆肥起始阶段的优势微生物群落；堆肥进入高温期后，细菌数量下降，放线菌、真菌数量上升，特别是放线菌数量增加比较明显；在堆肥的降温期真菌为堆肥优势微生物群落；在堆肥过程中细菌的数量在8.1×10⁸～1.2×10⁹ CFU g^-1之间，比放线菌1.0×10⁶～1.1×10⁷ CFU g^-1、真菌1.0×10⁵～5.6×10⁷ CFU g^-1高出近两个数量级。利用分子生物学方法——变性凝胶梯度电泳（Denaturing Gradient Gel Eletrophoresis，DGGE）法从分子微生态学角度对堆肥过程的微生物的多样性进行了定性分析。分子微生态学方法定性分析研究表明在堆肥过程中细菌群落的相似性和堆肥过程密切相关，即堆肥取样时间间隔越短的堆肥样品细菌群落相似性越高，如堆肥第一天和堆肥第四天的细菌群落的相似性最高达92％以上；而成熟堆肥阶段的细菌群落和堆肥起始阶段的细菌群落的相似性相差最大，相似性只有30％，说明堆肥过程中的微生物群落的演替是一个逐步变化的过程。就堆肥中放线菌多样性而言，其在堆肥初期就已经有一些放线菌存在，但它们的数量和种类相对较少。随着堆肥进行，特别是在进入堆肥的高温期后放线菌的种类和数量明显增多，放线菌多样性丰富。放线菌群落变化与堆肥时间正相关，即堆肥时间间隔短的放线菌菌落相似性程度高。堆肥过程中真菌群落变化和细菌群落及放线菌群落演替所不同的是堆肥过程中的细菌、放线菌群落的相似性和堆肥过程密切相关，而真菌则与堆肥过程中随时间的变化规律不明显，真菌群落在堆肥后期多样性丰富。利用红外光谱（Infrared Spectrum，IRS）分析技术对堆肥过程中的物质转化进行了分析研究，发现：①随着堆肥过程的进行，堆肥中所含的羟基、甲基和亚甲基的量逐渐降低，而甲氧基的含量增加；且羧基最终大多以羧酸盐的形式存在。甲基和亚甲基含量的降低，表明堆肥原材料的脂肪族化合物特征下降。与此同时，伴随着堆肥原材料脂肪族类化合物性质的降低，堆肥中芳香族类化合物性质则趋于相对升高。这说明堆肥过程中碳水化合物及脂肪族化合物分解较快，而木质素分解相对较慢。②对不同堆肥原材料组分分解形成的物质来说，碳水化合物和脂肪族化合物分解后大部分生成CO₂而放出；蛋白质和氨基酸等含N化合物一部分生成CO₂、NH₄⁺和NO₃^-，另一部分则以酰胺态-N存在；而木质素的分解则主要生成氢醌类物质。堆肥过程中含碳素物质和含氮素物质的变化，致使堆肥过程中堆肥的碳氮比（carbon/nitrogen,C/N）发生了变化，堆肥C/N的变化可能是堆肥微生物群落变化的重要原因之一。利用固相微萃取-气相-质谱联用（solid phase microextraction-gas chromatogram-mass spectrum,SPME-GC-MS）技术对堆肥中的低分子量有机物质进行了定量和定性分析研究。研究表明堆肥中的低分子量有机物质中包含多种低分子有机酸，如乙酸、丁酸、蜡酸、辛酸、苯甲酸等，另还包括乙醛、壬醛、苧烯等。其中乙醛、乙酸、壬醛在堆肥中稳定且含量较高，可作为判断堆肥腐熟度的参考指标。对植物有促生和抗生作用的生物活性物质——苯甲酸（benzoic acid）和苧烯（limonene）首次在堆肥产品中被发现，其中苯甲酸具有化感作用，苧烯对松树虫害——刺胫小蠹具有防治作用。利用SPME-GC-MS技术研究堆肥中低分子量有机物质，开辟了对堆肥中低分子量有机物质进行定量分析和详细分类研究的新领域，对为筛选可在农业有机固体废弃物堆肥过程中积累有益低分子量有机物质的微生物腐熟菌剂作了有意义的探索。堆肥中低分子量有机物质的分析为有机堆肥产品施入土壤后的生物学和化学行为的研究提供了较为详细的资料。利用近红外光谱（near-infrared spectroscopy,NIRS）分析技术分析测定了有机堆肥过程中钾含量的变化，根据所测得的结果对近红外光谱数据进行小波变换，利用处理后的小波系数，采用偏最小二乘法预测了有机肥料中植物营养元素钾的含量，建立了小波变换与近红外光谱技术相结合用于测定以奶牛粪便为主的有机堆肥产品中钾元素的测定模型。结果表明：小波变换充分提取了近红外光谱的信息，数据压缩为原始大小的3.6％，计算量大大减少；本文利用C⁴小波系数对48个有机肥料样本进行建模，对42个预示集样本进行预测，预示集的误差均方根和相关系数r²分别为0.1138％和0.927，优于原始光谱直接建模的0.1672％和0.835。该方法和目前常用的测定有机肥料中植物营养元素钾含量的方法——原子吸收光谱法（Atomic Absorption Spectrometry,AAS）和四苯硼酸钾重量法（Gravimetric method as potassium tetraphenylborate,GPT）测定结果相符，但近红外光谱法分析测定结果的重现性要优于后两者。更多还原

【Abstract】 Cow dung added with a small amount saw dust, and rice hull, after full mixed, wascomposted in the equipment which was designed by us. The relationship of microbialbiodiversity and the transformation of composting materials in the composting process ofagricultural organic solids matters were determined.The microbial biodiversity and the development pattern of microbial group in thecomposting process were studied by culture microbial method and un-culture microbialmethod, molecular biology method （denaturing gradient gel electrophoresis, DGGE）,respectively.The culture microbial method had shown that the bacterium was dominant position todecompose the materials at the beginning composting stage and theirs number was therange of 8.1×10⁸～1.2×10⁹ CFU g^-1 during the composting process. When the compostingbody reached the thermophilic phase, the number of bacterium decreased, then theactinomycetes was superiority to brake down the composting materials and the number ofactinomycetes was the scope of 1.0×10⁶～1.1×10⁷ CFU g^-1. When the compostingprocess got to cooling phase and maturation phase, the fungi was preponderance todecompose the composting materials and the number of fungi was the scope of 1.0×10⁵～5.6×10⁷ CFU g^-1.The culture method had made known that composting was given impetus to forwardunder the common action of bacteria, actinomycetes, fungi, but their dominant positionwere different under the different composting stage.The un-culture microbial method, denaturing gradient gel electrophoresis, had madeclear that the similarity of bacteria, actinomycetes biodiversity communities was connectedwith the composting process; however, the similarity of fungi biodiversity communities wasdifferent with bacteria, actinomycetes. Such as, the biggest percent of the similarity ofbacteria biodiversity communities was 92% between the first day and the fourth daycomposting samples and the smallest percent of the similarity of bacteria biodiversitycommunities was 30% between the second day and the sixtieth day composting samples. At the beginning composting stage, there had been little actinomycetes, however,theirs number and kinds were smaller than that of the latter composting stage. With thecomposting, theirs number and kinds had become more and more. There was a positiveconnection between the percent of the similarity of actinomycetes with composting process.The changes of fungi biodiversity communities were different with the changes ofbacteria, actinomycetes biodiversity communities in the composting process of agriculturalorganic solids matter. On the contrary with the changes of bacteria, actinomycetesbiodiversity communities, the changes of fungi biodiversity communities was no relatedwith the composting process. The fungi biodiversity communities at the end of compostingprocess were more abundant than that of other composting stages.The Infrared Spectrum （IRS） annalistic technology was directly applied to study thedecomposition process of cow dung added with a small amount saw dust and rice hull. Theresult showed that infrared spectroscopy could indicate the dynamic changes of organiccomponents in the composting process. With the progress of composted materialsdecomposition, hydroxyl, ketonic carbonyl, methyl, methylene, methine and aliphaticcompounds decreased, but aromatic compounds increased, which meant that thecarbohydrates and aliphatic compounds decomposed quickly and lignin decomposed slowly.In the composting process, carboxyl existed as carboxylic ions, most carbohydrates andaliphatic compounds were oxidized to CO₂ and H₂O, the proteins and amino acids weredecomposed to CO₂, NH₄⁺, NO₃^- and amides, and the lignin were decomposed tohydroquinones. The change of materials contained with carbon and nitrogen resulted invariance of ratio of carbon and nitrogen （C/N）. The change of C/N might be one of themain reasons of microorganism evolution in the composting process of agricultural organicsolids matters.An experiment was conducted to analyze on the low molecular weight organic matters（LMWOM） in composted production with/without NMF inoculants by Solid PhaseMicroextraction - Gas Chromatogram - Mass Spectrum （SPME - GC - MS）. The results hadshown that there were a large number of low molecular weight organic matters incomposted productions. As far as these organic matters were concerned, there were as muchsixteen kinds of low molecular weight organic acids and Aldehyde, Alcohol, Ketone,Carboxylic acid, Carboxylic acid, Furan, Pyrazine, Fat hydrocarbon, sulphide, benzoic acid,limonene and so on. This study had also shown that kinds and content of. LMWOM incomposting samples with inoculants were much more than that in composted samples without NMF inoculants.The amounts of acetaldehyde, acetic acid, and nonaldehyde acide in compostedproduction was stable and high, which meant they might be one of the judgments ofcompost maturity. This study had extracted the two new bio-active material, limonene andbenzoic acid from composted production. Benzoic acid could promote the growth of plantor control development. Limonene could prevent and cure of plant diseases and insect pestof pine tree.The analysis of the low molecular weight organic matters in composted production bySPME - GC - MS technology supplied the detail data for the function of biology andchemistry when the composted production was applied into soil. At the time, such studymight be regarded as a significant explore for selected the composting micro-organisminoculants which could produce and accumulate such low molecular weight organic mattersin this field.A method for the determination of crop nutrition element - Potassium iron （K⁺） inorganic fertilizer samples was established based on the combination of discrete wavelettransform （DWT） and near-infrared spectroscopy （NIRS） technique. Potassium in organicfertilizer can be determined by NIRS technique, because they were combined with organicgroups with NIRS absorption. In the proposed method, the raw NIRS data and their waveletcoefficients are used for modeling and prediction of the contents of potassium in organicfertilizer by partial least square （PLS） method. It was shown that there was almost no lossof spectral information with the NIRS data compressed to 3.6% of its original size. Themodel based on wavelet coefficients was better than that based on the full NIRS spectralrange. With the improved method, accurate prediction could be achieved.The results had shown that the method of NIRS was as the same as the method ofGravimetric potassium tetraphenylborate （GPT） and Atomic Absorption Spectrometry（AAS） in determination of potassium of composted product, but the reappearance of NIRSmethod was much more than that of AAS and GPT method in the detection of Potassiumiron （K⁺） in composed production.更多还原