Factors Affecting Phosphorus Sorption By Lake Sediments

Phosphorus (P) is the limiting nutrient regulating primary productivity in lakes. Phosphorus is considered as a major contributor to eutrophication in aquatic systems. Since 1980s, China has witnessed the surge in consumption of chemical fertilizers, the rapid development of intensive animal husbandry and serious soil erosion. All these have led to the loss of large amounts of phosphorus into the lakes, consequently aggravating water pollution and the phenomenon of eutrophication in these water bodies. Unlike carbon, hydrogen, oxygen and nitrogen, which have much faster cycles, P can be treated as a one-way traffic from rock deposits to agriculture and then to the water and sediments. It will take millions of years to complete the above-mentioned cycle due to less air-borne circulation of P. Once the P has entered the water bodies, it is impossible for automatically elimination. Due to the fact that P has strong affinity to contact with sediment particles, sorption of P on the sediments is one of the main processes to immobilize P in the aquatic systems. Thus, investigation of P sorption on sediments could be useful to describe and predict P fluxes across the sediment-water interface, and corresponding ecological risks. Currently, it is necessary to study the characteristics of P sorption on sediments, for P transport through water eco-systems is largely dependent on P sorption properties of the sediments. In this dissertation, two different sediments from eutrophic lakes (Huajiachi and sub lakes of Taihu) were selected to study the impacts of various factors (Microbes, Temperature, Incubation time, Interaction of pH and Ionic strength) on P sorption of the sediments. In addition, we also compared the differences of pseudo-first and pseudo-second order kinetics in describing P sorption process. It would provide further understanding of underlying mechanisms of P cycle in lakes, as well as theoretical and technical support for harnessing eutrophication in lakes.1. The phosphorus level in sediments is one of the major indicators of the eutrophication of lakes. Little information is available on the effect of microbial activity on P sorption in shallow lake sediments. The effect of microbes on P sorption by lake sediments was examined through an incubation experiment. Sediment samples were collected from eutrophic lakes and sterilized by autoclaving. The sediment samples were inoculated and then incubated at various temperatures. Organic fractions of adsorbed P were analyzed to investigate the distribution of P in sediments. Microbial inoculation resulted in greater P sorption in the sediment than sterilization in oxic conditions. This microbe-enhanced effect increased with temperature. The contribution of microbial inoculation to the maximum P sorption values of the sterilized sediment increased with temperature at increments of 0.2,12.3,22.9, and 33.0 mg/kg for temperatures of 4℃,20℃,28℃, and 36℃, respectively. The concentrations of NaHCO3-IP, NaHCO3-OP, and Microbe-P in the inoculated sediments were significantly higher than those in the sterilized sediments, particularly Microbe-P, which was increased by more than 80%, indicating that P sorption in the sediment was strongly influenced by microbial activity. The microbes release parts of P when oxic conditions change to anoxic conditions. Our results demonstrate that the microbial activity could increase the sorption of P in the sediment, especially under warmer temperatures and oxic conditions. It shows that sediment microbes could play an important role in the cycling of P in shallow lakes. These findings may lead to the development of an alternative technique to restore water quality in shallow lakes through microbial facilitated eco-remediation. Further work is needed to better understand the importance and mechanisms of this P fixation in situ.2. Temperature is a crucial factor affecting the P sorption in sediments. The objective of this part was to evaluate the effect of temperature on sorption of P by sediments from two eutrophic lakes. The study was carried out using short-term batch experiments at 4,20 and 30℃. Phosphorus sorption kinetics, isotherms, fractionation and desorption were investigated. The P sorption was dependent on sediment type and temperature (p<0.001). The Mei sediment showed a higher sorption rate and sorption capacity than Hua sediment. The P sorption kinetics were best described by a pseudo second-order model (R2>0.97). Activation energies derived from the kinetics rate constant indicated that P sorption onto the two sediments was controlled by a diffusion process. For both sediments, Freundlich model fitted the P sorption isotherms well and the calculated apparent sorption heat was 6.37 kJ/mol for Mei sediment and 8.67 kJ/mol for Hua sediment. This indicated that P sorption onto both sediments was endothermic. Adding P significantly increased the soluble and loosely bound P (S/L-P), aluminum bound P (Al-P) and iron bound P (Fe-P) (p<0.05). The amount of Al-P and Fe-P was markedly higher at 30℃than those at 4℃(p<0.05). Subsequent P desorption indicated that adsorbed P was highly labile, in particular for Hua sediment. The degree of P mobility that occurred during sediments sorption was inversely related to the temperature at the time of sorption. A significant relationship (R2=0.978) between phosphorus sorption capacity and oxalate-extractable Fe and Al at different temperature reflects the amorphous contents of Fe and Al are responsible for the temperature effect on P sorption.3. Phosphorus could react with particle surface over a long time. Equilibrium time used in sorption studies varied greatly. To elucidate the effect of time on P sorption results, two incubation times (24 h and 360 h) were studied in this chapter. Results shown that P sorption were biphasic, a fast stage followed by a slow one. Higher P concentrations required longer time to reach equilibrium. Phosphorus sorption capacity increased with incubation time. Freundlich provided better description of the isotherm data in comparison with Langmuir model due to the complex reaction mechanism involved. Subsequent P fraction indicated longer incubation time facilitated the increase of sediment S/L-P, Al-P and Fe-P pool. Phosphorus desorption after 360 h sorption was less than that after 24 h sorption, which demonstrated that longer contact time favored P fixation in sediment. Phosphorus sorption results were noncomparable using different time inferred from the above results. Appropriate time scale should be considered for P sorption experiment.4. The effects of pH and ionic strength (KCl) on P sorption by Mei and Hua sediments were studied. In the range of pH 6～9, the amount of P adsorbed by the sediments samples first increased then decreased. For Mei sediment, the sorption peaked at pH 6.97, while for Hua sediment it peaked at 7.49. The sorption was enhanced by increase of ionic strength in Mei sediment, whereas it was weekend by increase of ionic strength in Hua sediment. It was assumed that the differences were related to the sediment particle surface properties. Mei sediment contain more fine particles was made up of Fe and Al minerals and Hua sediment with coarse particles was mainly composed by Ca minerals. It was suggested that the effect of ionic strength on sorption by Mei sediment through its influence on electric potential in the plane of sorption. However, competition between anion and phosphate resulted in the decrease of sorption with ion strength on Hua sediment.5. The first-order and second-order kinetics equations have been widely used to describe sorption data obtained under non-equilibrium conditions. One is often claimed to be better than another according to a marginal difference in correlation coefficient. No attempts were made to comprehensively compare these two empirical equations for P sorption. In this paper, two different sediments (Mei and Hua) were selected to study the effect of initial P concentration (Co) on P sorption kinetics and to identify the corresponding sorption kinetic type. Kinetics studies had been carried out using an agitated batch method. The kinetics of sorption were followed based on the amounts of P adsorbed at various time intervals. Data were analyzed using pseudo-first order and pseudo-second order equations. The results showed that:(1) for the Mei sediment, the process belonged to a two-stage dynamic: an initial fast exchange of P followed by a much slower process, and its sorption quantity was much larger than that of the Huajiachi sediment, which can be seen as chemical sorption. For the Hua sediment, the sorption process was slow and had a small adsorbing capacity, which belonged to physical sorption, and the data points fluctuated greatly; (2) the zero equilibrium P concentration (EPCo) in the Mei and Huajiachi sediment was 0.03 and 0.42 mg/L, respectively. When Co< EPCo, the sediment desorbed P, or else the sediment adsorbed P; (3) the sorption data in the Mei and Hua sediment followed pseudo-second and pseudo-first kinetic model respectively by investigating the theoretical criteria and comparing standard residual error and correlation coefficient; (4) the sorption quantity and sorption rate of sediments increased with Co. Besides, the higher the Co value was, the longer time was required to achieve equilibrium. From the relationship between Co and kinetic parameters, generalized predictive models for P adsorbed onto sediments at any contact time and initial P concentration within the given ranged were proposed. These equations can then be used to derive the amount of P adsorbed at any given ion concentration and the reaction time. It would be helpful to understand the P transfer differences between sediments.6. In this chapter, a model containing native adsorbed P based on classical Langmuir model was developed. By comparing with other models including modified Langmuir and Freundlich models, the advantage and disadvantage of the developed model were discussed. Contained EPCo in the equation expression, the model could be used to determine the sorption direction, e.g. sorption or desorption by quantitative methods. We also deduced a kinetic model including initial concentration and equilibrium sorption quantity. For its application, future test needed to be conducted.