Removal Of Odor-Compounds And Control Of Disinfection Byproduct Formation By Uv/Persulfate Oxidation

As a result of the industrial and agricultural development but poor control in wastewater effluent discharge, algal blooms in eutrophic source waters have caused serious taste-and odor-problems in China. Due to the fact that the OTCs of odor-compounds are very low, they are difficult to remove in common drinking water process. Thus, taste and odor in drinking water have received the most complaints from consumer. Besides, organic matter, difficult to be removed by common water production, in eutrophic water is of a high level and can react with chlorine to form disinfection byproducts (DBPs) during chlorine disinfection. Thus, the development of new technology in removing odor-compounds and controlling the formation of DBPs can improve drinking water safety. The aim of this study is to develop an advanced oxidation process, ultraviolet/persulfate (UV/PS), in degradation of 2-methylisoborneol (2-MIB) and geosmin (GSM) and in reducing the formation of DBPs through oxidation of DBP precursors. In addition, a kinetic model by using 2-MIB and GSM as target-compounds has been proposed to fit the UV/PS process. Also, comparative study of permanganate, ozone, ultraviolet/hydrogen dioxide (UV/H2O2) and UV/PS was conducted to evaluate the removal of odor-compounds and their impacts on the formation of DBPs.2-MIB and GSM can be degraded effectively by using UV/PS process and the removal efficiency increased with increasing the dosages of PS. Although pH did not affect the degradation of 2-MIB and GSM directly, different scavenging effects of hydrogen phosphate and dihydrogen phosphate resulted in higher values of the removal efficiencies of the two odor-compounds in acidic conditions. The reaction temperatures and the presence of chloride (C1-) and nitrate (NO3-) played little role in the degradation of 2-MIB and GSM. On the contrary, bicarbonate (HCO3-) and natural organic matter (NOM) were the most vital reasons for the inhibition of 2-MIB and GSM removal when using UV/PS.The reaction rate constants for sulfate radical (SO4·-) reacting with 2-MIB and GSM were determined to be (4.2±0.6)x108 M-1 s-1 and (7.6±0.6)x108 M-1 s-1, respectively, through relative technology with benzoic acid being chosen as reference compound. Using the same method in a UV/H2O2 process, we could calculate that the rate constants of hydroxyl radical (HO’) reacting with 2-MIB and GSM were (4.3±0.2)x109M-1 s-1 and (5.7±0.2)x109M-1 s-1, respectively.A kinetic model for UV/PS process was proposed based on steady-state assumption with respect to 2-MIB and GSM. The modeled values of pseudo-first- order rate constants for both 2-MIB and GSM fit the experimental values all the time under different conditions, including different PS dosages and pHs, and the presence of methanol, tert-butyl-alcohol, HCO3- and NOM. Also, the model was checked in two real water samples from the outlets of filters in two different water productions. The experimental values were about 75%-80% of the modeling values for both 2-MIB and GSM in both the two real water samples. The deviation was might account for the different reaction rate constants between SO4·- and NOM from different sources. Thus, it is reasonable to conduct research focusing on the reaction rate constants of SO4·- reacting with NOM with different characteristics and from different sources. The reaction rate between SO4·- and Suwannee River NOM was estimated to be 2.35x107 Mc-1s-1 through the model.Although none of HO’is produced in UV/PS process, HO’can form through the oxidation of water by SO4·-. Thus, both SO4·- and HO’contributed to the degradation of 2-MIB and GSM and the contribution of HO’was even higher than SO4·- under the condition that no other inhibitors existed. However, the contribution of HO’can be depressed by methanol, tert-butyl-alcohol, CO32-/HCO3- or NOM and the inhibition effect on HO’was much stronger than that on SO4·-. It is not only because that the reaction rate constants of HO’reacting with these compounds are higher than those of SO4·-, but also due to the fact that the formation of HO’can be reduced by the consumption of SO4·-. As a result, when treating real waters, the contribution of HO’to the removal of 2-MIB and GSM in UV/PS can be ignored.UV/PS pretreatment played little role in the increase of carbonaceous disinfection byproducts (C-DBPs), but lowered the formation of nitrogenous disinfection byproducts (N-DBPs) to some extent. Chloral hydrate (CH) and haloketones (HKs) followed an increasing and then decreasing pattern, but dichloroacetonitrile (DCAN) and trichloronitromethane (TCNM) followed an opposite pattern with increasing UV/PS pretreatment time. C-DBPs rose monotonically with increasing dosages of PS, but N-DBPs could be declined at a suitable dosage of PS. Preoxidation pH and the presence of nitrate played little role in DBP formation. In the presence of 0.5 mg/L ammonia, UV/PS pretreatment inhibited the formation of DBPs effectively. However, most of DBPs increased by UV/PS pretreatment when all the free chlorine were converted to chloramine. The conversion of chlorinated DBPs to brominated DBPs in the presence of bromide was barely affected by UV/PS pretreatment.It is difficult to remove 2-MIB and GSM by using permanganate and ozone, but not for UV/PS and UV/H2O2 The degradation of 2-MIB and GSM in nure (?) UV/H2O2.However, the removal rate of both 2-MIB and GSM in real waters by using UV/PS process was lowered than that by using UV/H2O2 due to the more serious inhibition effects in the presence of NOM and CO32-/HCO3-. Preozonation increased the formation of some DBPs significantly and the increase of haloaldehydes (HAs) were the highest among the four oxidation processes due to the fact that oxidation of NOM by ozone produced the highest concentrations of aldehydes which are important precursors of HAs.Nearly all the C-DBPs increased after UV/H2O2.However N-DBP s increased significantly for Suwannee River NOM but showed little variation when treating Songhua River water.Permanganate preoxidation increased the formation of CH and TCNM, but played little role in the increase of other DBPs.The impact of UV/PS pretreatment on the changes of C-DBPs was the lowest among the four oxidation process concerned here.In addition, UV/PS pretreatment could reduce the formation of N-DBPs such as that DCAN, trichloroacetonitrile (TCAN)and TCNM decreased by 26.1%, 34.7% and 25.4%, respectively,when using UV/P S preoxidation of Songhua River water.From aforementioned results and discussions, we could conclude that UV/PS is not only effective in degrading 2-MIB and GSM, but also can control the formation of DBPS,especially for N-DBPs.Thus,it is reasonable to speculate that the combination of UV/PS and chlorine disinfection in d rinking water treatment process has the potential to control odor-compounds and DBPs synchronously.