Study on treatment of natural rubber processing wastewater using integrated physicochemical and biological processes

processes to improve the system performance in the simultaneous removal of organic and nutrients substances in NRP wastewater. 2. Objectives of the thesis The objective of this study is to develop an energy and nutrients recovering wastewater treatment process combining physicochemical and biological methods for NRP wastewater. 3. Main research content of the thesis 1) Overview of current technologies for treatment of NRP wastewater; 2) Study on removal of organic substances and energy recover from NRP wastewater by Expanded Granular Sludge Bed (EGSB) reactor; 3) Study on simultaneous recovery of nitrogen and phosphorus from NRP wastewater by Magnesium Ammonium Phosphate (MAP) precipitation method; 4) Study on simultaneous removal of organic and nitrogen substances from anaerobically treated NRP wastewater in modified Sequencing Batch Reactors (SBRs); 5) Proposal of an energy and nutrients recovering wastewater treatment process for NRP wastewater. 3 CHAPTER 1. LITERATURE REVIEW This chapter presented the following contents: Overview of the natural rubber processing industry; Characteristics of NRP wastewater; The situation of study and treatment of NRP wastewater in domestic and oversea; Wastewater treatment methods related to the thesis; Existing problems in NRP wastewater treatment in Vietnam; and Study orientation of the thesis. The review showed that the study and treatment of NRP wastewater have attracted great attention in past decades in the world and Vietnam as well, and achieved relatively good results. However, previous studies have mainly focused on the treatment of organic matter in wastewater without paying attention to the treatment of nitrogen substances, as well as the recovery of energy and nutrients. CHAPTER 2. REMOVAL OF ORGANIC MATTER AND RECOVERY OF ENERGY BY EGSB REACTOR 2.1. Materials and research methodology 1) Materials, chemicals, and equipment Wastewater: Simulated wastewater (wastewater prepared from coagulation process of natural rubber latex in the laboratory) was used for the start-up period of EGSB reactor. Real NRP wastewater was used for further studies. Seed sludge: Anaerobic sludge from an UASB of Sai Gon - Me Linh Brewery was used as seed sludge. 4 Experimental equipment: An EGSB reactor with a reaction volume of 13.5 L and a height of 155 cm, divided into reaction zone (I) and settling zone (II) as shown in Figure 2.3 was used. 1. Control box 2. Wastewater tank 3. Wastewater supply pump 4. Scum breaking pump 5. Circulating pump 6. EGSB reactor 7. Treated wastewater tank 8. Gas measuring device I. Reaction zone II. Setting zone Figure 2.3. Experimental EGSB system 2) Research methodology Experimental procedure Wastewater from the wastewater tank (2) was pumped into the bottom of the EGSB reactor (6), flowed up through the sludge bed in the reaction zone (I), entered the settling zone (II), then flowed into the treated water tank (7). Volume of generated biogas from the reactor was measured by the gas meter (8). Experimental conditions The EGSB reactor was started up with simulated wastewater (27 days) and real NRP wastewater (60 days) by a gradual increase in organic loading rate (OLR). After the steady state reached, effects of OLR in the range of 7 – 20 kg CODm-3d-1 on COD removal, biogas generation and the system stability were investigated using the real NRP wastewater. 1 2 7 8 6 I . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. II 3 4 5 5 2.2. Results and discussion 1) Development of anaerobic granular sludge On day 27 On day 87 Figure 2.5. Anaerobic granules of EGSB reactor during start-up period The EGSB reactor start-up was performed for 87 days to form anaerobic granules. After 27 days of starting, anaerobic granules appeared, and particles with size of 0.5 - 1.0 mm accounted for 38.5% of activated sludge in the EGSB reactor. After 87 days, the amount and size of anaerobic granules in EGSB reactor has increased significantly: at the lower part of the sludge layer, the particles with dimensions of 0.5 - 1.0 mm and 1.0 - 2.0 mm accounted for 45.5% and 35.4%, respectively; in the upper part of the sludge layer, these percentages were 62.6% and 18%, respectively. The image of anaerobic granules on days 27 and 87 is shown in Figure 2.5. 2) COD removal COD removal efficiency of the EGSB reactor after start-up period is shown in Figure 2.8. The results show that COD treatment efficiency was quite stable in the experimental modes and tended to decrease in the first days after an increase of OLR, but quickly stabilized in each experimental mode (about 5 days). In particular, when changing from mode (II) to mode (III) with a 6 large increase in OLR (from 11.3 to 17.7 kg CODm-3d-1), COD removal efficiency dropped sharply, then gradually ascended but fairly slowly. Sludge concentration in the EGSB reactor at this period was still not high, therefore the effect of OLR was very clear. COD removal efficiency in the modes (I), (II), (IV) and (V) are all over 80%, and that at OLR 19 kg CODm-3d-1, was 82.5%. Although this efficiency was lower than those at OLRs of 7.7 and 10.8 kg CODm-3d-1, it was also a relatively high. Figure 2.8. COD removal efficiency of EGSB reactor during steady operation period OLR (kg CODm-3d-1): (I) = 7.7; (II) = 11.3; (III) = 17.7; (IV) = 19.0; (V) = 10.8 3) Biogas generation yield Figure 2.13 shows that generated biogas amount increased when OLR increased. The average generated biogas amount at different OLR modes from (I) to (V) at standard conditions was 33,6, 44.5, 63.9, 76.6, and 44.2 L/day, respectively. The results in Figure 2.15 show that generated biogas amount at standard conditions was directly proportional to the amount of removed COD. The average biogas conversion yield 0 20 40 60 80 100 0 2000 4000 6000 8000 10000 85 95 105 115 125 135 145 155 C O D r e m o v a l e ff ic ie n c y , % C O D , m g /L Operation time, day COD in influent COD in effluent Efficiency I II III IV V 7 at the standard conditions for all study modes was 0.37 L/kg removed COD. Figure 2.13. Generated biogas amount in the steady period OLR (kg CODm-3d-1): (I) = 7.7; (II) = 11.3; (III) = 17.7; (IV) = 19.0; (V) = 10.8 Figure 2.15. The relationship between the generated biogas amount and the amount of removed COD y = 0.371x R² = 0.9174 0 20 40 60 80 100 50 100 150 200 250 G en er a te d b io g a s a m o u n t a t st a n d a r d c o n d it io n , L /d a y Amount of removed COD, g COD/day 20 40 60 80 100 5 10 15 20 25 80 90 100 110 120 130 140 150 160 G en er a te d b io g a s a m o u n t a t st a n d a r d c o n d it io n , L /d a y O L R , k g C O D m -3 d a y -1 Operation time, day OLR Amount of biogas I II III IV IV 8 CHAPTER 3. RECOVERY OF NUTRIENTS BY MAP PRECIPITATION 3.1. Materials and research methodology 1) Materials, chemicals and equipment Wastewater: Wastewater used in this study was the effluent of the EGSB reactor. Chemicals: MgCl26H2O and H3PO4 were used as additional magnesium and phosphate sources. Equipment: A Jar-Test was used for MAP precipitation. 2) Research methodology Experiments: MAP precipitation were performed at different pH values and Mg2+ : NH4+ : PO43- molar ratios of (molar ratios were changed by adding MgCl2 and H3PO4 solutions). After reaction, the reaction solution was let to settle and filtrated to obtain the precipitate. P-PO43-, N-NH4+ and Mg2+ in the filtrate was analyzed to determine treatment efficiency. The precipitate was washed, dried, and used to determine MAP amount and composition of Mg, N, and P elements in the MAP precipitate. Analysis: MAP crystal dimension was determined by SEM images. MAP composition was determined through EDX spectrum analysis. 3.2. Results and discussion 1) MAP recovery without magnesium addition NRP wastewater, in addition to ammonium and phosphate, contains significant amounts of magnesium. Therefore, increasing wastewater pH to the appropriate value can result in MAP precipitation by reaction (1.3). 9 Mg2+ + NH4+ + HPO42- + 6H2O  MgNH4PO46H2O↓+ H+(1.3) The results in Table 3.2 show that phosphate-P removal increased from 15.6% at pH 7.5 to 44.7% at pH 9.5 and tended to slightly increase as pH continued to increase to 11.5. On the other hand, ammonium-N removal increased from 3.6% at pH 7.5 to a maximum value of 13.1% at pH 9.5, and then gradually decreased to 5.0% as the pH continued to increase to 11. Table 3.2. P, N and Mg removal without magnesium addition at different pH pH Concentration after MAP precipitation (mg/L) Removal efficiency (%) PO43-–P NH4+–N Mg2+ PO43-–P NH4+–N Mg2+ 7.5 119 214 15.0 15.6 3.6 70.0 8.0 107 209 8.8 24.1 5.9 82.4 8.5 97 204 6.5 31.2 8.1 87.0 9.0 90 197 4.0 36.2 11.3 92.0 9.5 78 193 1.8 44.7 13.1 96.4 10.0 77 198 1.6 45.4 10.8 96.8 11.0 76 211 1.3 46.1 5.0 97.4 Because MAP is soluble in acidic environment, the suitable environment for MAP precipitation is alkaline. However, in an alkaline environment, Mg2+ also reacts with PO43- and OH- to form precipitates of Mg3(PO4)2 and Mg(OH)2. Therefore, at the higher the pH, the more these precipitates form, resulting in reduction of the magnesium source for the MAP forming reaction. Thus, when pH is over a certain value, ammonium-N 10 removal efficiency will decrease. This result explains why the MAP precipitation optimally occurs at a certain pH range. The initial molar ratio of Mg : PO43- in the wastewater was 0.46 : 1.0 (in MAP, it is 1.0 : 1.0), therefore, theoretically, the maximum phosphate-P removal via the MAP precipitation is 46%. At pH 9.5, the removal efficiency of P-phosphate is 44.7%, rather high in comparison to theorical value. at pH 9,5 at pH 11 Figure 3.3. SEM image of MAP crystal at pH 9,5 at pH 11 Figure 3.4. EDX spectrum of the MAP precipitate In this study, suitable pH for MAP recovery was about 9 - 10. At this range, MAP precipitation was clearly observed. MAP crystals were easy to settle and could be observed with the naked eyes, and had a large length of 300 - 383 µm. The 70,6 µm 383 µm 328µm 11 removal efficiencies of phosphate-P and ammonium-N at pH 9.5 were 44.7% and 13.1%, respectively. At pH 11, beside MAP crystals with small size, hardly settable fine flocks also appeared (Fig. 3.3). EDX spectrum (Figure 3.4) and data of chemical composition show that: the mass percentages of the P, Mg and O elements in MAP at pH 9.5 are 14.3%, 10.8% and 54.3%, respectively. These percentages are similar to those in pure MAP. At pH 11, beside the above main elements, there were many other elements such as C, Na, K and Ca in the MAP precipitate. The mass percentages of P, Mg and O elements were 7.4%; 6.0% and 44.2%, respectively, quite lower than those in MAP obtained at pH 9.5. 2) MAP recovery with magnesium addition In NRP wastewater used in this study, the molar ratio of Mg2+ : NH4+ : PO43- was 0.46 : 3.5 : 1.0, while this ratio in pure MAP is 1.0 : 1.0 : 1.0. Therefore, to improve the recovery of phosphate-P, addition of external magnesium source is required. Table 3.5. Effect of Mg2+: PO43- molar ratio on P and N removal Mg2+ : PO43- molar ratio Concentration after MAP precipitation (mg/L) Removal efficiency (%) PO43-–P NH4+–N Mg2+ PO43-–P NH4+–N Mg2+ 0.6 : 1 65.3 185 3.2 53.7 16.7 95.1 0.8 : 1 34.2 173 3.3 75.7 22.1 96.3 1.0 : 1 15.6 163 2.8 88.9 26.6 97.4 1.2 : 1 9.4 159 2.5 93.3 28.4 98.1 1.4 : 1 9.2 161 3.4 93.5 27.5 97.8 12 Mg2+ : PO43- molar ratio Concentration after MAP precipitation (mg/L) Removal efficiency (%) PO43-–P NH4+–N Mg2+ PO43-–P NH4+–N Mg2+ 1.6 : 1 9.1 174 3.6 93.5 21.6 97.9 Table 3.5 show that, removal efficiencies of phosphate-P and ammonium-N reached the best value of 93.3% and 28,4%, respectively, at the Mg2+ : PO43- molar ratio of 1.2 : 1.0. 3) MAP recovery with addition of both magnesium and phosphate MAP recovery efficiency In order to increase the efficiency of N-ammonium removal and MAP recovery, it is necessary to add external sources of magnesium and phosphate. Table 3.7. Effect of Mg2+ : NH4+ : PO43- molar ratio on P and N removal efficiencies Mg2+ : NH4+ : PO43- molar ratio Concentration after MAP precipitation (mg/L) Removal efficiency (%) PO43-–P NH4+–N Mg2+ PO43-–P NH4+–N Mg2+ 0.6 : 1.0 : 1.0 214.0 154.3 8.2 56.5 30.5 96.4 0.8 : 1.0 : 1.0 117.0 138.6 9.6 76.2 37.6 96.8 1.0 : 1.0 : 1.0 36.5 106.5 11.3 92.6 52.0 97.0 1.2 : 1.0 : 1.0 27.4 61.2 13.2 94.4 72.4 97.1 1.4 : 1.0 : 1.0 19.6 42.4 15.6 96.0 80.9 97.1 1.6 : 1.0 : 1.0 15.2 56.4 22.7 96.9 74.6 96.3 1.8 : 1.0 : 1.0 14.3 97.1 28.4 97.1 56.3 95.9 13 The results in Table 3.7 show that the highest N-ammonium removal efficiency of 80.9% was obtained at Mg2+ : NH4+ : PO43- molar ratio of 1.4 : 1.0: 1.0. As a result, the optimal Mg2+ : NH4+ : PO43- molar ratio for the simultaneous removal of both N- ammonium and P-phosphate was 1.4 : 1.0 : 1.0. Analyzation of obtained MAP product MAP sample obtained at pH = 9.5 and Mg2+ : NH4+ : PO43- molar ratio of 1.4 : 1.0 : 1.0 was taken for SEM and EDX analysises to evaluate the crystal size and composition of the MAP product. The results show that the precipitate was clear crystals, and had white color mixed with dark brown color and many stains on its surface (Figure 3.12), possibly due to the organic components in wastewater and/or other precipitates formed during the MAP precipitation. The mass composition of the P, Mg, O elements was 13.6%, 11.4%, 59.4% respectively. This composition is nearly similar to that in pure MAP (12.6%, 9.9% and 65.3%). In addition, the precipitate also contained 11.2% C and other substances. (a) (b) Figure 3.12. MAP precipitate (a) and its SEM image (b) 14 CHAPTER 4. SIMULTANEOUS REMOVAL OF ORGANIC MATTER AND NITROGEN IN MODIFIED SBRs 4.1. Materials and research methodology 1) Materials, chemicals, and equipment Wastewater: Wastewater in this study was the effluent of the EGSB reactor. Seed sludge: Activated sludge from an oxic/anoxic biological tank of Hanoi Plastic Company was used as seed sludge. Equipment: Two similar modified SBRs with an effective volume and a working height of 15 liters and 1.34 m, respectively, were used (Figure 4.1). 1. Wastewater container 2. Wastewater supply pump 3. Wastewater supply pipe 4. Modified SBR 5. Automatic valve 6. Effluent tank 7. Air blower 8. Air flowmeter 9. Air diffuser 10. Controller I. Oxic zone; II. Anoxic zone Figure 4.1. Modified SBR system 2) Research methodology Experimental procedure Wastewater 2 8 7 1 6 A ir E ff lu en t 3 10 9 4 5 I I 15 Two modified sequencing batch reactors (SBRs), R1 and R2, specially configured to consist of both oxic and anoxic zones, and be operated with only a single simultaneous oxic/anoxic phase in each treatment batch were used. R1 was operated with a constant aeration, by contrast, R2 was operated with an air flow varied from a lower rate in the early period of the reaction phase to a higher rate in the later period. The operating strategy for the reactors was also modified to combine the drawing stage of the treated water from the previous batch and the filling stage for the new batch into the same phase. The reactors was operated in a sequential three-step cycle mode (simultaneous drawing and filling; reacting; and setting) as shown in Table 4.1. Table 4.1. Operation mode of the modified SBRs Reactor Time for simultaneous drawing and filling, min Time of reaction phase, min Setting time, min Air flow 0,4 L/min Air flow 2,0 L/min R1 10 0 145 25 R2 55 90 Experimental conditions Study on effects of OLR, ammonium-N loading rate (ANLR) and nitrogen loading rate (NLR) on the performance of R1 and R2 was performed in the OLR and NLR ranges of 0.52 – 1.61 kg CODm-3d-1 and 0.071 – 0.32 kg Nm-3d-1, respectively. Study on effect of COD/TN ratio on the performance of the reactors was carried out in the COD/TN ratio range of 3.4 – 6.0. 16 4.2. Results and discussion 1) Effect of OLR on the COD removal The results in Figure 4.4 show that both R1 and R2 almost reached steady state after only one week from the startup. The average COD removal efficiencies of both R1 and R2 in all phases were quite stable and over 95%. There was no significant difference in COD removal efficiency for both R1 and R2. Figure 4.4. Effect of OLR on COD removal of R1 and R2 OLR (kg CODm-3d-1): I = 0.52, II = 0.73, III = 0.90, IV = 1.19, V = 1.61 Figure 4.7. Effect of NLR on ammonium removal of R1 and R2 NLR (kg Nm-3d-1): I = 0.071, II = 0.096, III = 0.16, IV = 0.21, V = 0.31 96 97 98 99 100 0 100 200 300 400 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 H iệ u s u ấ t x ử l ý N -N H 4 + , % N -N H 4 + , m g /L Thời gian vận hành, ngày effluent N-amoni_R1 Effluent N-amoni_R2 Influent N-amoni 0 20 40 60 80 100 0 500 1000 1500 2000 2500 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 C O D r e m o v a l , % C O D , m g /L Thời gian vận hành, ngày Influent COD Effluent COD_R1 Effluent COD_R2 Removal_R1 Removal_R2 I II I III I IV V I II III IV V 17 2) Effect of ANLR on the ammonium-N removal The results in Figure 4.7 show that the N-ammonium removal of both R1 and R2 became stable after only one week operation. Ammonium removal efficiencies of R1 and R2 were almost the same, averagely over 99%. Average effluent ammonium-N concentration was less than 1 mg/L. Figure 4.10. TN removal efficiency of modified SBRs in different phases NLR (kg Nm-3d-1): I = 0.071, II = 0.096, III = 0.16, IV = 0.21, V = 0.31 3) Effect of NLR on the TN removal Figure 4.10 shows that essential time for R1 and R2 to achieve stable TN removal was 30 and 21 days, respectively. In the stationary period, the TN removal efficiency of R1 was quite high with an average of 88 - 92% in phases III – V. R2 was able to achieve a steady state faster than R1. The removal efficiency of R2 was higher than that of R1, about 97% in phases III and IV, and 94% in phase V. At the same NLR, the TN removal efficiency of R2 was always higher than that of R1. The TN concentration in the effluent of R2 was significantly lower than 0 20 40 60 80 100 0 100 200 300 400 500 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 T -N r e m o v a l, % T N , m g /L Operation time, day Influent T-N Efluent T-N_R1 Efluent T-N_R2 Removal_R1 Removal_R2 I II III III IV 18 that of R1, always less than 25 mg/L in all three experimental phases III - V, meanwhile, the TN concentration in the effluent of R1 fluctuated within the range of 15 - 50 mg/L. 4) Effect of COD/TN ratio on COD removal The results in Figure 4.13 show that no significant differences in the COD removal efficiencies versus COD/TN ratio were observed. The COD removal efficiencies of both reactors were similar, with an average of over 95%. Figure 4.13. Effect of COD/TN ratio on COD removal 5) Effect of COD/TN ratio on ammonium-N and TN removal The N-ammonium removal efficiencies of both reactors were over 99%, and N-ammonium concentration in the effluent was below 1 mg/L (Fig. 4.14). TN removal efficiencies of both reactors tended to increase when increasing in the COD/TN ratio. These results are consistent with expectations, since a low COD/TN ratio leads to a shortage of organic substrates for denitrification, resulting in low TN removal (Figure 4.15). The mean TN removal efficiency 40 50 60 70 80 90 100 0 500 1000 1500 2000 2500 3000 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 C O D r e m o v a l, % C O D , m g /L COD/TN ratio COD Influent COD Effluent - R1 COD Effluent - R2 COD removal - R1 COD removal - R2 19 for reactor R1 increased from 70% to 92% when raising the COD/TN ratio from 3.4 to 6.0; meanwhile, that of reactor R2 increased from 80% to 97%. The results in Figure 4.15 also show that TN removal efficiency of R2 was remarkably higher than that of R1 at all COD/TN ratios investigated. Figure 4.14. Effect of COD/TN ratio on N-ammonium removal Figure 4.15. Effect of COD/TN ratio on TN removal efficiency 94 95 96 97 98 99 100 0 100 200 300 400 500 600 3.0 3.5 4.0 4.5 5.0 5.5 6.0 N -a m o n iu m r e m o v a l, % N -a m o n iu m , m g /L Tỉ lệ COD /TN N-amoni Influent N-amoni Effluent - R1 N-amoni Effluent - R2 Removal - R1 Removal - R2 20 40 60 80 100 3.0 3.5 4.0 4.5 5.0 5.5 6.0 T N r e m o v a l, % COD/TN ratio R1 R2 20 CHAPTER 5. PROPOSAL OF AN APPROPRIATE TREATMENT PROCESS FOR NRP WASTEWATER 5.1. Proposed treatment process Through the results achieved in this study, an energy and nutrients recovering wastewater process for NRP wastewater integrating physicochemical and biological processes as in Figure 5.1 was proposed. Figure 5.1. Proposed appropriate treatment process for NRP wastewater 5.2. Process material balance Based on results of the material balance (Table 5.1 and Figure 5.2), it can be seen that the proposed process could ensure that the effluent quality could meet the discharge requirements according to the Vietnam Regulation QCVN 01-MT: 2015 BTNMT. In addition, with the scale of Ha Tinh Rubber Factory, about 320 kg MAP per day could be recovered each day. The biogas amount that could be recovered daily is about 352 Nm3, equivalent to around 200 Nm3 methane. In terms of energy, it is equivalent to about 200 kg of oil or nearly 2000 kWh. Rubber latex seperating pond Equalizati on tank EGSB MAP prepicipation Product drying furnace Intermediate tank B io g a s Wastewater Efluent Modified SBR Stationa ry pond MAP product 21 Table 5.1: Influent parameters, calculation results for effluent Parameters Unit Influent (1) Effluent (6) COD mg/L 6,480 74.2 TN mg/L 285 31.0 TP mg/L 186 24.8 SS mg/L 500 50.0 Point (1) (2) (3) (4) (5) (6) (7) (8) (9) Q, m3/d 220 232.7 232.7 232.7 232.7 212.1 20.6 12.7 7.9 Q, m3/h 9.2 9.7 9.7 9.7 9.7 8.8 0.9 0.5 0.3 COD, mg/L 6480 6181 1854 1483 1483 74.2 - 1000 - COD load, kg/d 1426 1438 431.5 345.2 345.2 15.7 - 12.7 - TN, mg/L 285 376 376 310 310 31.0 - 100. - TN load kg/day 63 87.5 87.5 72.1 72.1 6.6 - 1.3 - TP, mg/L 186 177 177 35.4 35.4 24.8 - 24.8 - TP load , kg/day 41 41.2 41.2 8.2 8.2 5.3 - 0.3 - SS, mg/L 500 500 200 200 200 50.0 8000 500.0 20000 SS load, kg/day 110 116 46.5 46.5 46.5 10.6 165 6.3 158 Figure 5.2. Results of calculating the material balance 22 CONCLUSIONS AND RECOMMENDATIONS Conclusions 1) Generation of granular sludge in ESGB reactor after 3 months startup was successfully performed. The reactor could be stably operated in the OLR range of 7 - 20 kg CODm-3d-1 with average COD removal of over 80%. Biogas conversion yield for NRP wastewater was 0.37 L/g removed COD. 2) Applicability and appropriate conditions for MAP precipitation in the nutrient recovery from the NRP wastewater were clarified. The optimal pH for MAP precipitation was 9.5, at which nearly 45% of phosphate-P in NRP wastewater could be recovered without addition of magnesium source. In the case of magnesium addition, the optimal Mg2+ : PO43- molar ratio and phosphate-P removal efficiency were 1.2 : 1.0 and 93.3%, respectively. When both magnesium and phosphate were added, the optimal Mg2+ : NH4+ : PO43- molar ratio was 1.4 : 1.0 : 1.0, at which the removal efficiencies of phosphate-P and ammonium- N, were 97.1% and 80.9% respectively. 3) A new modified SBR capable of effectively and simultaneously removing organic and nitrogen substances from NRP wastewater was successfully developed. The performance of the reactor was significantly improved compared to conventional SBRs and nitrification/denitrification reactors. At OLRs of 0.9 - 1.6 kg CODm-3d-1 and NLRs of 0.16 - 0.31 kg Nm-3d-1, average removal efficiencies of COD, ammonium-N and TN were 97 %, nearly 100% and 94 - 97%, respec