【作者】 李朝丽；

【导师】 周立祥；

【作者基本信息】 南京农业大学 ， 植物营养学， 2007， 博士

【摘要】 通常认为，强吸附性污染物的迁移风险是很低的，但关于它们在土壤和沉积物中长距离迁移的报道却屡见不鲜。这种现象激发了关于促进运输（facilitated transport）和优势流（preferential flow）的研究，目前普遍认为这是提高污染物在地下环境中运移速度的两种主要机制。胶体（包括矿质胶体、有机胶体、生物胶体等）与优势流对污染物环境行为影响的研究起源于胶体促进的放射性核素的迁移，进而发现胶体亦能影响其它污染物的迁移。围绕这一命题国内外已展开了大量对地下环境中胶体的来源、胶体对污染物的吸附、胶体促进的污染物迁移、胶体的固定的研究。然而，胶体对污染物在土壤中环境行为的影响是多方面的。关于胶体对土壤吸附污染物的贡献、胶体对污染物生物有效性的影响的研究还不多见。另外，对于在生命活动旺盛的土壤中普遍存在的优势流研究也仅停留在室内小型土柱或肉眼可见的几种示踪剂上，并且鲜有关于污染物在土壤剖面中究竟如何分布及优势流对污染物迁移的直接影响的报道。因此有必要开展对模拟大孔隙和自然条件下优势流的研究。为此，本文选取Cd作为供试污染物，采用等温吸附试验、动力学试验、生物盆栽试验、模拟大孔隙土柱淋洗试验、田间淋洗试验等方法，系统研究了土壤胶体和优势流对镉在土壤环境中的吸附、迁移和生物有效性等环境行为的影响及其可能机制。主要结果如下：1、黑土、黄棕壤、水稻土、红壤不同粒级组分（粘粒<2μm，粉粒2～20μm，细砂粒20～200μm，粗砂粒200～2000μm）对镉的吸附均表现为：粘粒>粉粒>粗砂粒>细砂粒，且粒级间的差异都达到显著水平；各粒级组分对镉的吸附均符合Langmiur方程和Frendlich方程，吸附量最大的粘粒组分对镉的平均最大吸附量为202．8±64．6 mmol·kg^-1，粉粒、细砂粒、粗砂粒对镉的最大吸附量分别为135．0±39．2mmol·kg^-1、47．0±9．4 mmol·kg^-1、81．3±33．7 mmol·kg^-1，粘粒分别平均是粉粒、细砂粒、粗砂粒的1．53±0．30倍、4．23±0．67倍、2．57±0．55倍；各组分对镉的吸附量之间的差异随着镉的浓度增大而增大。各组分对土壤吸附镉的贡献率为：粉粒（52．2±15．3）>粘粒（32．3±14．9）>细砂粒（13．8±3．4）>粗砂粒（1．7±1．4）（只有红壤中粘粒>粉粒），土壤80％～90％的镉吸附在<20μm的组分上，粗砂粒的贡献基本可以忽略不计。各因素对吸附的影响大小为：粒级>有机质>平衡pH，而游离氧化铁无显著影响。建立了最大吸附量和分配系数与各影响因子的多元回归方程：最大吸附量=-239．098+54．935×粒级赋值+34．261×平衡pH-2．844×游离氧化铁含量+2．215×有机质含量分配系数=-174．182+26．332×粒级赋值+29．795×平衡pH+1．675×游离氧化铁含量+1．604×有机质含量2、黄棕壤不同粒级组分（粘粒、粉粒、细砂粒、粗砂粒）对镉的吸附动力学与热力学研究表明，两种温度下（25℃和45℃）各粒级组分对镉的吸附均可分为快反应和慢反应两个阶段，0-15min内为快反应阶段，吸附量达到饱和吸附量的95％以上，此后为慢反应阶段；随着温度由25℃升高到45℃，各组分对镉的饱和吸附量增加了4．86％～25．3％；各组分对镉的吸附动力学符合拉格朗日假二级动力学方程，吸附过程以化学吸附为主；二级动力学吸附速率常数表明，随着各组分粒级增大，吸附速率降低；在试验温度范围内随着温度升高，吸附速率加快；吸附过程的限速步骤为颗粒间扩散；各粒级组分对镉的吸附为吸热反应，反应能自发进行。3、盆栽试验结果表明：（1）各处理黑麦草株高、地上部干重、根干重、总生物量都表现为胶体>原土>去胶组分，胶体上总生物量分别平均是原土和去胶组分的1．31±0．02倍和1．82±0．21倍。（2）黑麦草地上部与根中Cd浓度、地上部与根对Cd的富集系数都表现为胶体<原土<去胶组分，表明Cd的生物有效性胶体<原土<去胶组分，这主要是各组分的表面性质、有机质含量、pH等的差异引起的。（3）加入EDTA增加了Cd的解吸，导致黑麦草地上部与根中Cd浓度显著增加，黑麦草地上部干重、根干重、总生物量降低，但植株Cd总量与CK相比还是有所上升。EDTA对Cd的活化作用受到各处理pH的强烈影响，表现为去胶组分>原土>胶体，黄棕壤>红壤，EDTA对各处理植株Cd总量的影响与此吻合。至于地上部与根部Cd浓度对EDTA的响应与上述顺序不完全一致则反映了EDTA对不同组分上黑麦草Cd的迁移系数的影响差异：EDTA使黄棕壤胶体和原土Cd的迁移系数显著增加，而对黄棕壤去胶组分和红壤三种基质Cd的迁移系数无显著影响。（4）两种土壤在各方面差异显著，一次平衡试验表明，EDTA浓度为0时，黄棕壤各组分Cd的解吸率分别表现为胶体和原土约为0，去胶组分组分为10．5±3．5％，红壤各组分平均为20．8±1．9％。可以推测这是各处理黑麦草长势及体内Cd浓度差异的直接原因。4、室内人工大孔隙土柱试验结果表明，大孔隙的存在使Cd的迁移量增加了75％，在10个孔隙体积内大孔隙．对照处理淋滤液中镉的浓度平均是对照处理的106倍。淋洗液中加入胶体和DOM使大孔隙土柱Cd的迁移量分别增加了16％和50％。胶体和DOM对Cd迁移的促进作用在3个孔隙体积内表现得特别明显，淋出液中镉的浓度平均分别是不加胶体或DOM的大孔隙处理的2．47（胶体）和2．66倍（DOM）。5、田间试验结果表明：土壤剖面具有很大空间变异性，优势流对Cd的迁移起主导作用；Cd在土壤基质中的扩散被限制在几厘米范围内，大孔隙的存在使亮蓝和Cd的迁移距离增加了数十倍；胶体通过在土壤基质毛细管中的堵塞效应和在优势流区域大孔隙壁上的沉积抑制了Cd在土壤中的迁移（约10 cm），而DOM与对照相比使Cd的向下迁移增加了约10-20 cm；三个剖面中亮蓝浓度与交换性镉和全镉浓度间有强烈正相关关系，表明亮蓝与Cd在剖面中迁移的优势流路径非常相似。总之，胶体对Cd的强烈吸附作用能在一定程度上控制Cd的生物有效性，进而制约Cd进入生物圈，从而降低Cd的环境风险。优势流对Cd的迁移起主导作用，而胶体的不稳定性使得其在真实环境中对Cd的促进迁移作用具有很大的偶然性和瞬时性，特别是不饱和多孔介质中的复杂环境条件对预测胶体及污染物的环境行为将是一巨大挑战。更多还原

【Abstract】 The risk of leaching of strongly adsorbing contaminants had been assumed to be low, but such compounds have been reported to move large distances in soils and sediments. This has motivated research towards facilitated transport and preferential flow, two mechanisms that could enhance transport rates. Laboratory column studies have demonstrated co-transport of contaminants sorbed to suspended colloids, or simultaneous leaching of in situ colloids and contaminants, while field studies have revealed the association of contaminants with colloids in drain or groundwater. Facilitated transport and preferential flow are interrelated because colloid transport velocities are higher in macropores, where preferential flow occurs. While the direct effect of preferential flow on the transport of colloid and the colloid-facilitated contaminiants transport is unavailable. The objectives of this research were:1.To determine the general contribution of colloid on the adsorption of Cadmium in different soils;2.To determine the effect of colloid on the bioavailability of Cadmium to plant;3.To determine the effect of colloid on the transport of Cadmium in a repacked soil column with quartz veins;4.To determine the effect of colloid on the transport of Cadmium in field soil monolith and the extent to which the distribution of Cadmium in the profile were associated with the observed preferential flow paths.To meet these objectives, batch experiments and pot experiments and soil column experiments and a field experiment were conducted. The main results are as follows:1. Batch experiments were conducted to investigate the equilibrium sorption isotherms for Cd²⁺ onto particle-sized fractions（2000-200μm, 200-20μm, 20-2μm and 2-0μm, which were named coarse sand, fine sand, silt and clay according to international system respectively） derived from four typical soils（black soil, yellow brown soil, huangnitu paddy soil, and red soil） in China by the use of a novel technology combined with wet sieving, sedimentation-siphoning and centrifugation method. The results showed that adsorption of Cd²⁺ by these fractions could be described by the Langmiur equation Ce=Xm·Ce/X-k^-1 and Frendlich equation X=Kd·Ceⁿ. The maximum adsorption of Cd²⁺ （Xm） and distribution coefficient （Kd） decreased in an order: clay>silt>coarse sand>fine sand in the four soils. The average maximum adsorption of Cd²⁺by the clays of the four soils was 202.8±64.6mmol·kg^-1, and the maximum adsorption of Cd²⁺ by silt and fine sand and coarse sand were 135.0±39.2 mmol·kg^-1, 47.0±9.4 mmol·kg^-1 and 81.3±33.7 mmol·kg^-1 respectively. Accounting to the contents of the fractions in the soils, the contribution of each fraction followed by the order: silt （52.2±15.3%）> clay （32.3±14.9%） > fine sand （13.8±3.4%）> coarse sand （1.7±1.4%）. The percent of Cd²⁺ adsorbed by silt and clay of the four soils exceeded 80%. By choosing particle size, organic matter, iron oxide and equilibrium pH value as parameters, SPSS software and Multivariate Statistical Analysis were employed to build the regression model of adsorption of Cd²⁺ and to evaluate the effect of these factors on it. The results showed that particle size, organic matter and equilibrium pH value significantly affected the adsorption of Cd²⁺. But iron oxide had no significant effect on the adsorption.2. A batch method was used to investigate the kinetics and thermodynamics of cadmium adsorption onto different particle-sized fractions derived from yellow brown soil by using the same method as described above. The results showed that the reaction of adsorption can be divided into two types: a fast reaction in the first 15min and a slow reaction in the later reaction course. The amounts of Cd²⁺ adsorbed by the fast reaction exceed 95% of the adsorption capacity. As the temperature increased from 25℃to 45℃, the adsorption capacities of the four fractions increase by 4.86%-25.3%. First-order rate equation and pseudo second-order rate equations were applied to express adsorption kinetics. Adsorption processes for Cd²⁺ onto the four fractions are found to follow pseudo-second order type adsorption kinetics. The pseudo second-order rate constants exhibited that adsorption speed to reach equilibrium decreased with the increase of particle size. Intra-particle diffusion might be the major rate-limiting step. Thermodynamic parameters including△H°、△S°and△G°were also calculated from graphical interpretation of the experimental data. Positive△H°values indicated the adsorption processes to be endothermic. Negative△G°values implied that adsorption reaction was a spontaneous process.3. A greenhouse pot experiment was conducted to investigate the bioavailability of Cd of different soil matrix (colloid, raw soil and de-colloid soil, derived from yellow brown soil and red soil and artificially contaminated by Cd with a concentration of 10.91mg·kg^-1) to ryegrass and the effect of EDTA on it. A laboratory batch experiment was also conducted to study the potential mechanism involved it. The results showed that: （1） the mean shoot height, shoot dry weight, root weight and the total biomass were in the same order of colloid > raw soil > de-colloid soil. The total biomass in the colloid treatment was 1.31±0.02 and 1.82±0.21 times over that in raw soil and de-colloid soil respectively. （2） Cd concentration and bioaccumulation factor of shoot and root were followed by colloid < raw soil < de-colloid soil, indicating the bioavailability of Cd to ryegrass was colloid < raw soil < de-colloid soil. （3） EDTA addition (disodium salt, 27.27 mg·kg^-1) led to significant increase in Cd concentration of shoot and root and decrease in biomass of ryegrass. （4） Batch experiment revealed that the activation effect magnitude of EDTA on Cd was followed by de-colloid soil > raw soil > colloid, and this effect was influenced intensively by the initial pH value.4. A laboratory soil column experiment was conducted to investigate the artificial macropore on the colloid and DOM mediated Cd transport. Compared with the control treatment with no macropore in the soil column, macropore treatment enhanced leaching by 75%. The addition of colloid and DOM increased the loading of Cd by 16% and 50% respectively.5. The potential role of soil colloids and DOM in transporting Cd was investigated through in situ undisturbed paddy soil monoliths. In addition, Brilliant Blue FCF was used as a tracer to assess the effect of preferential flow on Cd down migration. Soil colloid fractionated from the paddy soil and DOM from rice straw were spiked with 5 mg·L^-1 of Cd, and then applied onto the surface of two soil plots （1m×1m） to maintain 150 mm water depth after 72 hour equilibration for Cd²⁺ with soil colloid or DOM. The solution containing Cd but without the addition of soil colloid and DOM was also applied as control treatment. One day later 100 L of dye solution (1g·L^-1) and sequent 100 L of tap water were applied to the plot. After the infiltration was completed, the vertical profiles excavated in the plot were observed for dye movement or preferential flow. Soil samples collected from homogeneously stained and unstained areas with different dye intensities at the various depths were determined for Brilliant Blue, water soluble Cd, exchangeable Cd, and total Cd. The results obtained from classic statistical and geostatistical analyses showed that deep penetration of Cd and Brilliant Blue into the soil profile took place due to preferential flow through macropores, mainly earthworm channels, with much of the chemicals thus bypassing the soil matrix. Dye tracer and Cd distribution within the soil matrix was fairly restricted to several centimeters. Colloid restrained the migration of dye and Cd both in the matrix and preferential flow area. DOM facilitated the transport of Cd and Brilliant Blue in matrix and macropores by about 10 cm over that of the control. Pearson correlation analysis revealed strong associations between Brilliant Blue concentrations and exchangeable and total Cd concentrations in the three plots indicating that they had taken the same preferential flow pathway.更多还原