The Mechanism And Mathmatical Model Study On The Interaction Between Iron And Phosphate In Water

As important nutrient elements and critical determinants of water quality, iron and phosphorus exist widely in natural world. It has great significance to study their migration and transformation in natural waters. In this thesis, on the basis of literature research, the complex reactions, redox reactions, precipitation-dissolution reactions and adsorption-desorption reactions between iron and phosphate were treated as objects of study. The kinetics and thermodynamics study of these reactions were deeplyinvestigated. The interaction mechanism between iron and phosphate in both natural water and wastewater was clearly undertsood. The effect of natural organic matters on the interaction was discussed. Based on the understanding of mechanism, appropriate mathmatical models were chosen to describe the interaction process. Beside the theoretical research, the treatment of simulated eutrophic water by iron addition and the migration of phosphorus in MBR system with iron dosage were also researched. The main contents and results are as following:1. The oxygenation kinetics of nanomolar concentrations of Fe(Ⅱ) in aqueous solution have been studied in the absence and presence of millimolar concentrations of phosphate over the pH range6.0-7.8. At each phosphate concentration, the overall oxidation rate constant varied linearly with pH, and increased with increasing phosphate concentration. A model based on equilibrium speciation of Fe(Ⅱ) was found to satisfactorily explain the results obtained. From this model, the rate constants for oxygenation of the Fe(Ⅱ)-phosphate species FeH2PO4+, FeHPO4and FePO4-have been determined for the first time. FePO4-was found to be the most kinetically reactive species at circumneutral pH with an estimated oxygenation rate constant of21.9±2.44M-1·s-1. FeH2PO4+and FeHPO4were found to be less reactive with oxygen, with rate constants of0.0316±0.02M-1·s-1and0.123±0.08M-1·s-1respectively. In addition, the calculated values using Marcus formula were closed to the modeled values in our study, which means the oxidation of Fe(Ⅱ) followed Marcus principle.2. The formation rates of FePO4in buffer solutions with pH=6.0-8.0were studied and the formation rate constants were obtained through competition kinetics experiments, in which sulfosalicylic acid (SSA) was used as competition ligand. It was found that the formation rate constant was significantly affected by pH. When pH is equal to7.5, the highest rate constant which is0.980±0.063×104M-1·s-1was gotten. In addition, in the pH range investigated, the formation rate constants between FePO4and FeSSA were close to each other, but both of them were two orders of magnitude less than Fe(Ⅲ) hydrolysis precipitation rate constant.3. The kinetics study of phosphate adsorption onto hydrous ferric oxide (HFO) in in buffer solutions with pH=6.0-8.0was investigated. It was found that phosphate adsorption onto HFO followed pseudo-second-order kinetics model. According to this model, under the same initial concentration of phosphate, the saturated adsorption capacity qe decreased following the pH increase. Meanwhile, the adsorption rate constant ksec also slightly decreased following the pH increase. While the initial adsorption rate constant h decreased significantly following the pH increase. But under the same pH value, qe increased, while ksec and h decreased significantly, following the increase of phosphate concentration.4. The kinetics study of "in situ"(in which Fe(Ⅲ) was added directly) phosphate removal by Fe(Ⅲ) in different pH was researched. We found that the "in situ" phosphate removal had much faster phosphate removal rate than phosphate adsorption onto HFO, even half an hour was enough to get the equilibrium. The pseudo-second-order kinetics model could also be used to explain the "in situ" process. In the model, qe increased following the the increase of phosphate concentration and slightly decreased following pH inrease. Meanwhile, ksec and h decreased significantly following the increase of phosphate concentration, while were not sensitive to pH changes. Compared with HFO adsorption, qe in "in situ" phosphate removal was almost two times higher than HFO adsorption for the same concentration of Fe(Ⅲ), and ksec and h were also much higher than HFO adsorption.5. Effects of pH, floc age and organic compounds on the removal of phosphate by pre-polymerized hydrous ferric oxides were investigated. The extent of adsorption of phosphate onto pre-polymerized HFO is strongly dependent upon pH, HFO age and concentration of competing anions with the phosphate adsorption capacity decreasing significantly with increase in pH, HFO age and concentration of competing anion. A semi-empirical DLM and a molecular CD-MUSIC model which take these factors into account have been developed and provide good description of the phosphate adsorption behavior over a wide range of conditions. The specific surface areas for different aged HFO were obtained from there two models, which were S(5min)=1198±50/1200±50m2/g, S(8.5h)=608±30/760±30m2/g, S(24h)=476±30/620±30m2/g. When citrate was used as the competing anion, both the inner-sphere monodentate surface complex=FeHCit-and the outer-sphere bidentate surface complex (=FeOH)2HCit2-for DLM and the binding of carboxylic group in citrate with surface sites as one inner-sphere surface complex=FeOOCR for CD-MUSIC model were found to be important contributors to citrate adsorption with site occupancy by these species preventing phosphate adsorption. In addition, the Kint obtained in DLM was found to obey linear free energy relationship (LFER), and according to the changes of available surface site concentration following HFO ages, we got the relationship between the Fe(Ⅲ) hydrolysis precipitation rate constant and time/pH, which is6. Effect of other organic compounds on the removal of phosphate by HFO in different pH was also studied, and DLM or CD-MUSIC model was used to describe this process. For the small organic molecule including citrate and SSA, citrate had greater influence on the phosphate adsoprtion than SSA, in which DLM was used to describe the process. While for natural polymeric compounds including SRFA and HA, the content of carboxyl group in SRFA is higher than HA according to NICA-Donnan model. It has been concluded that carboxyl group was the main group which reacts with HFO. Therefore, the content of carboxyl group in organic compounds determined the effect of organic compound on phosphate adsorption, which was proved by CD-MUSIC model. It was also found that alginate and soluble microbial products (SMP) has little effect on phosphate adsorption, but both of these two models could not describe the process because of the complexity of molecules.7. The "in situ" phosphate removal by Fe(Ⅱ) or Fe(Ⅲ) in different pH was comparativly studied, and the effect of citrate on "in situ" phosphate removal by Fe(Ⅲ) was also researched. DLM model was successful to explain the "in situ" phosphate removal by Fe(Ⅲ) with1.21±0.13mM/mM Fe of available surface site concentration which is much higher than the value used in HFO adsorption. In addition, it was found that low concentration of citrate could passive1mM Fe(Ⅲ) activity of phosphate removal. When Fe(Ⅱ) or Fe(Ⅲ) was used to remove phosphate, at low pH=6.0, because of the slow oxygenation of Fe(Ⅱ), Fe(Ⅲ) or HFO was produced less than Fe(Ⅲ) addition, hence, Fe(Ⅲ) had better phosphate removal efficiency than Fe(Ⅱ), but at high pH=8.0, Fe(Ⅱ) could be oxided in few minutes, which would bring HFO with better dispersion, higher specific surface area and higher phosphate removal efficiency, therefore, Fe(Ⅱ) had better phosphate removal efficiency than Fe(Ⅲ).8. The treatment with iron addition for simulated eutrophic water in different pH which was extracted from MBR system with phosphate and organic matters addition was investigated, and the effect of different organic matters on this process was researched emphatically. It is found that the precence of organic compounds significantly affect the phosphate removal efficiency. The sequence for the effect is citrate>SSA>SRFA>HA>alginate. 9. A bench-scale MBR case was researched, in which different kinds of iron were used as phosphate removal reagent. The changes of phosphate concentration in effluent, supernatant and sluge were analysed. The phosphate removal efficiency of real domestic sewage by MBR with Fe(Ⅱ) was also investigated. After stable effluent obtained, it was found that Fe(Ⅱ) has higher phosphate removal efficiency than Fe(Ⅲ), which means the cheaper reagent Fe(Ⅱ) is more suitable as the iron addition. In the different areas of MBR system, the concentration of phosphate in supernatant was significantly different to each other. The highest concentration of phosphate was found in the first anoxic compartment. And due to the slow oxygenation rate of Fe(Ⅱ) in anoxic condition, the concentration of phosphate in anoxic compartment of MBR-3(Fe(Ⅱ) addition) was higher than MBR-2(Fe(Ⅲ) addition). Concentrated nitric acid, dilute hydrochloric acid and Milli-Q water were used to digest the excess sludge from MBR, respectively. After the digestion, the concentration of phosphate was detected, and it was found that concentrated nitric acid gave highest the concentration of phosphate, while Milli-Q water brought the least one, which means the part of easily dissociation phosphorus in sluge is small, but when pH of the sluge solution decreased dramatically, great amount of phosphate would release. When MBR system with Fe(Ⅱ) was used to disposal real domestic sewage, it was found that the phosphorus removal efficiency could reach99%affer Fe(Ⅱ) addition and microorganism had high resistance to Fe(II).In summary, the basic research of interaction between iron and phosphate in natural water and wastewater supplied the theoretical and data support for the further understanding of the migration and transformation. The development and application of related mathmatical models in this thesis would play important roles in right prediction of iron and phosphate changes in waters. Furthermore, the study of engineering treatment cases would give good guiding significance for real wastewater treatment.