Outline 1. Introduction 2. Overview of chlorine disinfection - PDF document

11/14/2017 Research & Development Forum Drainage Services Department November 14, 2017 Mixing of Chlorine Jets for Sewage Disinfection Joseph Hun-wei Lee Hong Kong University of Science and Technology Outline 1. Introduction 2. Overview of chlorine disinfection process in Stonecutters Island Sewage Treatment Works 3. Field scale model for study of mixing and rapid chlorine demand in disinfection of primary treated (CEPT) effluent 4. Theoretical modelling of dense chlorine jet – chlorine dosage optimization 5. Conclusions 1

11/14/2017 HATS Stage 1/2A  Chemically-enhanced primary treatment (CEPT) at Stonecutters Island Sewage Treatment Works (SCISTW)  Pollutants removal rate:  70% organics (BOD)  80% suspended solids  60% heavy metals  25% total nitrogen  50% phosphorus  E.coli :  before disinfection: 50%  after disinfection: 污水消毒 10 7 → 10 5 counts/100mL Choi et al, Environmental Science and Technology , 2009 Harbour Area Treatment Scheme (HATS) 香港淨化海港計劃 Screening Stonecutters Island Plants/pumping Sewage Treatment Submarine stations Work (SCISTW) outfall 昂船洲污水處理廠 排污口 23.6 km deep tunnels (>100m below ground level) Stonecutters Chemically Enhanced Primary Island STW Treatment (CEPT) since 2001; 23.6 km of deep tunnels; 九龍 disinfection since March 2010 Stage 1: Q = 1.4 x 10 6 m 3 /d 香港島 Stage 2A: Q = 1.8 x 10 6 m 3 /d 2

11/14/2017 Stonecutters Island Sewage Treatment Work (SCISTW) General Layout of ADF - Advance Disinfection Facilities – Stage 1 Main Pumping Station Flow distribution Chamber Sedimentation Tanks Chlorine Dosing Chamber 15 Effluent Box Culverts Dechlorination 2 x 2.5 m x 2.5m Dosing Final Disinfection Facilities at SCISTW FDC No. 2 Flow distribution Chamber Drop Shaft (FDC) - New Effluent Tunnel 6 nos. Sodium Hypochlorite Storage Tanks Riser Shaft Decommissioned box culvert Dechlorination Compound (DC) Chamber 15A Overflow Culvert Extension of Chamber 15 6 3

11/14/2017 Motivation for Chlorine Dosage Control 1. Excessive Total Residual Chlorine (TRC) is toxic to aquatic organisms 2. Disinfection by-products (chlorinated organic compounds) are harmful to marine environment 3. Complex interaction between chlorine dosage and CEPT sewage; unknown chlorine demand and disinfection efficiency at high concentration 4. Enhance sustainability: energy and operation costs 5. Environmental protection should be set at an adequate but not unnecessarily severe level Pollution belt resulting from indiscriminate discharge: natural turbulent mixing is slow – Yangtze River, 1995 4

11/14/2017 SCISTW Chlorine Disinfection Dosing of sodium Dosing of sodium Average chlorine hypochlorite bisulphite Concentration = 12 mg/L (dry) Existing box culvert = 18 mg/L (wet) as the chlorine contact system Flow Chamber 15 Distribution Chamber CEPT Discharge Effluent to Outfall Dechlorinated Effluent Cl 2 conc. (ppm) 10 5 Initial Chlorine decay mixing and 2 order Req. Dil. after disinfection: bacterial kill ~ 10000! 10 7 → 10 5 count/100mL 10-20 x (m) 10m 1km What happens actually … • Most of the chlorine consumption has already taken place before Chamber 9; very low TRC concentration at Chambers 9 and 15. • Significant E.coli reduction is observed at Chamber 9; insignificant E.coli reduction between Chamber 9 and 15 • The 1km culvert does not act as a chlorine contact chamber as expected. Cl 2 conc. (ppm) Chamber 9 Chamber 15 FDC 10 5 Few E.coli kill Required 80% Chlorine Dilution consumed ~ 10000! Expected TRC << 1 10-20 ppm x (m) 10m 1km 1km culvert 5

11/14/2017 Chlorine Flow Distribution Chamber dosing unit (Chlorine dosing unit) • An inclined weir of 1.8m height in the middle of FDC • 10% chlorine solution is injected into the sewage flow Weir through an array of dense jets in two layers 12.5m Dosing 2  3.5m  3.5m Plan view Unit inlet culvert from sedimentation tanks Submerged flow Dense Jets 1.7 m opening Free surface flow 2.65mPD 0.85mPD Vertical section view Beaker Test vs Field Dosing • The mixing processes are very different, inducing very different reaction processes • Beaker test results may not represent the field condition Field Dosing Beaker test - Jet mixing, distance and time are required. - Near-instantaneous mixing - Unlimited reactants - Limited reactants CEPT Sewage 20 m 3 /s Conc. Cl 2 = 10 5 mg/L Conc. 6

11/14/2017 A “toy experiment” of chlorine jet in sewage coflow (Qiao et al. 2016) CEPT Sewage U a = 0.1m/s Chlorine jet Outflow U j = 0.3m/s Δ x = 0.45m, Δ t = 4.5s 100% About half of the dosed chlorine Chlorine C f = 600mg/L demand is consumed by the chlorinated 80% sewage in 4.5 seconds 60% Similar significant chlorine 40% demand for dosing at 10% or 52.9% 50.3% 44.0% 20% 1% chlorine (s ame chlorine mass flux ) 0% 10% (N=9) 2% (N=8) 1% (N=7) Source chlorine concentration (% w/w) 1:2 Physical Scale Model (with prototype sewage and chlorine) The objectives are to study: • the mixing achieved by the dosing unit in the FDC; • chlorine demand at different key locations in the FDC; • disinfection efficiency in the FDC; and • degree of settling of organic solids in the FDC. Chlorine The 1:2 physical dosing unit model represents a “1/16 slice” of the FDC treated 1/16 slice of FDC 2  3.5m  3.5m sewage flow inlet culvert from sedimentation tanks Weir 12.5m Plan view of FDC 7

11/14/2017 1:2 FDC model 1:2 Physical Scale Model Discharge of sodium Head hypochlorite solution in SCISTW for study of chlorine mixing tank (1/16 slice of FDC sewage flow) Dosing unit Test flume CEPT Sewage inflow from sedimentation tank Outlet tank Chlorine dosing system (10% NaOCl solution) Chlorine Storage tank Air chamber Safety valve Diaphragm pump To dosing unit 8

11/14/2017 Dosing unit Discharge of sugar solution in air Upper port D = 5mm Jet velocity Lower port D = 2.5mm Dosing sugar solution (  = 1.168 g/mL) into cross flow in 1:2 model Initial mixing of upper dosing jet with dyed sugar solution (tap water flow Qs = 60 L/s; U a = 0.2 m/s; total jet discharge q d = 20 mL/s) Qiao et al. ASCE Journal of Hydraulic Engineering , 2017 9

11/14/2017 Head tank Operation of the model flume with CEPT sewage Flow distribution baffle FDC weir Outflow over lateral weirs Sampling apparatus for TRC and bacteria above the weir (15 points) 0.11 0.11 Flume side Free surface wall Left Center Right Measuring y=0.0 point z 0.05 0.05 y 0.07 0.07 0.05 Weir crest Weir plate top (z=0.9 m) Portable TRC photometer 1D motorized vertical traverse for sampling Flume bottom z=0.0 Sampling tube 10

11/14/2017 Summary of experiments on chlorine demand of CEPT sewage (10% NaOCl solution) Oct – Dec 2015 4 key runs with detailed TRC, bacteria and nutrient measurement CFD modeling of the 1:2 scale FDC model • 900,000 grid cells • Free surface determined by volume of fluid (VOF) method U a = 0.36m/s Q u = 15.7mL/s Dosing jet flow = 20mL/s Q l = 4.3mL/s C 0 = 100,000 mg/L Δρ / ρ = 0.2 11

11/14/2017 Jet mixing of the chlorine with the sewage co-flow achieves a rapid dilution in the order of 1000-2000 in the FDC. This high dilution however falls short of the value required to achieve full mixing (i.e. a dilution of 5000-10,000). Only approximately 60-80 percent of the sewage flow over the FDC weir is chlorinated. Flume centreline Cross-sections Measured cross-section TRC profiles: Q s = 100L/s and q d = 20 mL/s, 10% NaOCl solution Lee et al. ASCE Journal of Environmental Engineering , 2017 1.6 100 L/s CEPT sewage flow 1.4  10% chlorine dosage flow 1.2 20 mL/s C m = 12.2 1 17.7 Upper Jet: q u =15.3 (mL/s) FDC weir and U j =1.20 (m/s) 35.0 58.0 20.0 0.8 Level z(m) 8.0 Low jet: q d =4.7 (mL/s) 10.0 and U j =1.50 (m/s) 5.8 0.6 0.4 0.2 Inver level z=0.0 m 0 0 0.5 1 1.5 2 2.5 3 3.5 Distance from jet source x(m) Measured TRC concentration (mg/L) distribution at the centreline section of FDC Measured TRC mass flux and average TRC concentration along the flume 12

11/14/2017 Summary • Approximately 70-80% of the chlorine mass flux is consumed within a very short distance (0.5-1 m, or a matter of several seconds) from the chlorine dosing unit, well upstream of the weir. Only a small fraction of the expected average concentration can be measured in the sewage outflow from the FDC. • The chlorine demand appears to be accompanied by a decrease in ammonia nitrogen of around 4 mg/L. • No settling and accumulation of chlorine solution or organic solids in the FDC has been observed in the model. There is a direct local correlation between TRC and E. coli . Overall there is a one-log E. coli kill above the weir, within a short travel time of about 7-10s from the chlorine injection. After immediate chlorine demand is satisfied, the residual (> 1.5 mg/L) can effectively disinfect the sewage by 2-log kill within contact time of about 5 minutes. Correlation between local E. coli and TRC concentration E. coli concentration variation with contact time above the weir at the downstream of the weir 13