Water Reuse in Texas
San Marcos, TX
UV/Chlorine AOP for Potable Reuse: Lower Cost Option
Keel Robinson, North America Water Reuse Leader
June 15th, 2016
Agenda
AOP 101
UV-AOP for Potable Reuse
UV-AOP Design Considerations
First Full-Scale UV/Cl2 AOP Design
2
AOP 101
What is AOP?
Advanced oxidation processes (AOPs) are technologies that
generate hydroxyl radicals (OH●)
Technology and/or Chemicals OH●
The goal of an AOP technology is to maximize the production of
hydroxyl radicals (OH●) to provide fast reaction kinetics to most
efficiently destroy specific contaminants at the lowest possible cost
Organic OH● Intermediate OH●
Compounds H2O Compounds H2O CO2
4
Why Do We Need AOP?
Some compounds are not strippable, adsorbable, or biodegradable. Some of these compounds are
regulated (either at a federal level or state level) or are candidates for future regulations. AOP is often the
best solution to destroy these types of compounds of concern.
Contaminant of Concern Typical Source Regulated?
1,4-Dioxane Solvent stabilizer, found in groundwater from past Yes (some states)
industrial releases
NDMA Found in groundwater from past industrial releases, Yes (some states)
or formed in wastewater plants
Atrazine Herbicide, found in surface water bodies from Yes
agricultural runoff
MIB & Geosmin Taste & odor compounds found in drinking water No, but a nuisance to customers
from algal blooms
Endocrine Disrupting Compounds Found in wastewater from human use No, but under consideration
(EDCs)
Pharmaceutical and Personal Care Found in wastewater from human use No, but under consideration
Products (PPCPs)
Industrial Micropollutants including Chlorinated Solvents, Petroleum Hydrocarbons, Fuel Yes
VOCs Additives, Phenols
5
Common AOP Technologies
Ozone + Peroxide
Excellent for a majority of AOP applications due to highly
efficient generation of hydroxyl radicals
UV + Peroxide Ideal for NDMA, excellent for low concentrations of
contaminants in RO effluent
UV + Chlorine At low pH, chlorine reacts with UV to create hydroxyl and
chlorine radicals. In some cases, may be more cost-
effective and implementable than peroxide.
ALSO, “OZONE” AND “OZONE WITH BIOLOGICALLY ACTIVE FILTRATION” CAN
BEHAVE AS AN EFFECTIVE AOP
6
Typical AOP Applications (examples)
Groundwater Remediation/Wellhead Treatment (industrial micropollutants)
Well From Air Granular
Contaminated Stripper AOP Activated Carbon
Aquifer (optional) (optional)
Drinking Water/Surface Water Treatment (taste & odor compounds)
Sedimentation Filtration Disinfection
and AOP
End of Pipe (industrial waste discharge, recalcitrant contaminants)
Industrial
Wastewater AOP
Treatment
Potable Reuse (1,4-Dioxane, NDMA, CECs)
Filtered Membrane
Secondary WW Filtration (MF, UF) Reverse Osmosis AOP
Effluent
7
UV-AOP for Potable Reuse
Why Do We Need UV-AOP for Potable Reuse?
Regulations for Indirect Potable Reuse
• California Groundwater Recharge Regulations (Full Advanced Treatment)
o Section 60320.201 (Requires Reverse Osmosis and Oxidation Treatment Process)
o 0.5 log removal of 1,4-Dioxane as an AOP surrogate because it partially passes through RO and is recalcitrant
• Big Spring, TX (TCEQ case-by-case)
NDMA
• Commonly found in tertiary wastewater and partially passes through RO
• California Notification Level (10 ng/l)
• UV-based AOP more effective than O3-based AOP at NDMA removal
Multiple Barrier Treatment
• Treatment redundancy to protect environment and human health
• Low molecular weight compounds (NDMA, 1,4-Dioxane, and CECs) pass through RO
• UV-based AOP provides pathogen barrier with maximum disinfection credit
9
Full Advanced Treatment Train
Indirect Potable Reuse (typical FAT)
Secondary or
Membrane Reverse Environmental
Tertiary UV AOP
Filtration Osmosis Buffer
Effluent
This treatment train is ideal for UV-based AOP because RO produces a high quality
effluent
High UVT (>95%) reduces power demand of UV reactor
Low DOC and alkalinity reduce scavengers, more efficient contaminant removal
Other treatment trains are possible with oxidation processes such as Ozone or O3-BAF
Ozone-based processes are very effective at removing a majority of trace organic
contaminants (TOrCs)
May be used as pretreatment to membranes and UV AOP
May displace membranes and/or UV AOP
10
UV-AOP Design Considerations
Importance of Treatability Testing
Bench-Scale Testing
• Quick and inexpensive, great screening and preliminary
design tool
• Snapshot-in-time, but useful if water sample is representative
of design conditions
• UV Collimated Beam Testing
• Dose-response curves
Pilot-Scale Testing
• Optimize process under a range of real-life conditions
• Optimize equipment sizing variables
• Scale-up tool for full-scale design and performance guarantee
• Demonstration-scale for operators
• Regulatory and public acceptance
12
Dose-Response Curve
Dose-response curve
(CBD-Test)
Log reduction
LOG reduction
(Pilot reactor)
UV dose [J/m²]
13
Design and Operation of UV Reactor: Dose vs. EED
P [ kW ]
Dose = I *t EED =
60 * Q [ gpm ]
UV Dose Setpoint Electrical Energy Dose (EED) Setpoint
Independently measured and verified through Collimated Does not directly measure UV output of lamps
Beam Testing
Provide a common design basis for all manufacturers Specific to a reactor type
Can be used to scale up Difficult to scale unless same reactor used for both pilot-
scale and full-scale
Ensures regulatory compliance when using validated Good parameter for comparing power efficiency of
dose equation (PSS) and on-line sensors different UV reactors and different operating conditions,
but not measuring output of lamps
Allows for energy savings by turning down power to May waste energy when conditions change
lamps when conditions change
14
Validate Performance With Collimated Beam Testing
Contaminant or Collimated Beam
Surrogate Spiking Testing Device
UV Dose and Log
Removal Comparison
RO
Permeate Pilot- or Full-Scale
UV Reactor
H2O2 or NaOCl
UV Dose may be calculated via Point Source Summation (PSS) Method
15
Why PSS for AOP Applications?
• PSS equation developed through extensive validation testing of UV reactors by 3rd party consultant
• PSS, CFD, and RED are different methods for calculating the average UV dose
• PSS accounts for real-time sensor inputs such as flow rate, UVT, and UV intensity
K143 12-17 (with 12 600W lamps per row, 17 rows)
Flow Rate UVT PSS CFD (mJ/cm2) RED
(MGD) (%) (mJ/cm2) (mJ/cm2)
12 96 920 902 942
PSS calculation method is validated by CFD and RED
16
UV Dose Control
S = UV intensity sensor reading (per row) P = ballast power (same to each lamp)
INPUTS UVT = UV transmittance Number of rows on OUTPUTS
Q = Flow Rate
PLC
17
CBT Dose vs. PSS Dose
LP UV/H2O2 AOP
1,4-Dioxane Removal
CBT vs. Pilot-Scale UV Reactor
5 ppm H2O2
1.40
1.30
1.20
1.10
Log Reduction
Collimated Beam Dose
1.00
0.90
0.80 PSS Dose from Pilot
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
UV Dose mJ/cm2
18
Upscaling and Sizing with UV Dose
The reactor data and PSS model are validated using CBT data ideally
generated in parallel on site
The reactor UV dose is calculated in real-time based on the measured
flow rate, UVT, and UV intensity
The dose-response curve is used for upscaling using the same PSS
approach with full-scale reactor
19
UV/H2O2 AOP For Potable Reuse Has Been The Status Quo
• Has historically been the standard AOP technology for groundwater recharge/indirect potable
reuse in California with multiple successful installations in operation today (e.g. Orange County,
West Basin, WRD, Big Spring)
• Also used in drinking water for taste & odor and in groundwater remediation applications
• Hydrogen peroxide is relatively expensive and not readily used at WWTPs
• The photolysis of hydrogen peroxide is inefficient as only about 10% of the chemical is
consumed in the UV AOP reaction; thus, incurring significant residual quenching costs
20
Introducing UV/Cl2 AOP
• Recent academic research shows that UV/Cl2 AOP is effective at low pH
Watts, M. & Linden, K., 2007. Chlorine Photolysis and Subsequent OH Radical Production During UV Treatment of Chlorinated Water. Water Res., 41:13:2871
Watts, M., Rosenfeldt, E,. & Linden, K., 2007. Comparative OH Radical Production Using UV-Cl2 and UV-H2O2 Processes. Jour Supply Water Res Technol - AQUA., 56:8:469
Watts et al. 2012. Low pressure UV/Cl2 for advanced oxidation of taste and odor. Journal-AWWA
• Reverse osmosis for FAT produces a low pH permeate (~5.5)
• Sodium hypochlorite (chlorine) is readily used at most WWTPs
• Use of sodium hypochlorite may result in significant life cycle cost savings relative to hydrogen
peroxide
• Residual chlorine may be desirable for additional pathogen credit and/or secondary disinfection
• First greenfield full-scale system under construction by City of Los Angeles
21
UV/Cl2 AOP for Reuse – Chemistry Considerations
Chloramines and Breakpoint Reactions
Residual chloramines may be present in RO
permeate prior to hypochlorite addition
pH and Chlorine Speciation
Lower pH favors free chlorine in hypochlorous
acid form (slow hydroxyl radical scavenger)
Higher pH shifts free chlorine to hypochlorite ion
form (rapid hydroxyl radical scavenger)
Disinfection by-product formation
THMs/HAAs
Ref: Hach
Chlorate
Target pH and free chlorine residual at inlet to UV reactor
22
Terminal Island Water Reclamation Plant
LASAN TIWRP AWPF AOP Design
• 5 MGD to 12 MGD expansion, converting
chloramination to comply with groundwater AOP Design Basis:
recharge regulations 3 to 12 MGD
TOC < 0.25 mg/l
• 18 month bench and pilot scale study UVT >96%
including O3/H2O2, UV/H2O2, and CA Groundwater Recharge Regulations
UV/HOCl led by LASAN, Trussell
Technologies, and Carollo Engineers AOP Specifications:
6-log virus credit
• Selected UV/HOCl based on performance 0.5 log 1,4-Dioxane removal
and life cycle costs <10 ppt NDMA in effluent
UV dose = 920 mJ/cm2
• Awarded AOP system to Xylem/Wedeco, Free chlorine dose = 2-4 mg/l
startup expected in late 2016
UV Dose is the design and operational basis, not EED
First Ever UV/Cl2 AOP Full-Scale Design
24
Wedeco MiPRO AOP Pilot System
Containerized with climate control (HVAC), lighting, sink,
and refrigerator
Fully automated with PLC and Operator Interface
Remote Monitoring & Datalogging
Can run operate in various modes of operation including
Ozone only, UV only, Ozone with Peroxide (AOP), UV
with Peroxide (AOP), and UV with Chlorine (AOP)
State-of-the art instrumentation
25
Terminal Island Treatability Testing Objectives
12 month pilot study to compare 6 different AOPs
UV LP + H2O2
UV LP + NaOCl
UV MP + H2O2
UV MP + NaOCl
Ozone + H2O2
H2O2 + Ozone
26
Bench and Pilot Scale Results
27
Investment Decision: Chemical Savings
• $3.3M in chemical savings
over 20 years
• NaOCl already on-site
• Additional pathogen
barrier/credit with Cl2 for FAT
UV/Cl2 AOP selected
28
Investment Decision: Real-Time Energy Savings
UV Dose control reduces power usage
Allows for variable power control to optimize power consumption and
ensure regulatory compliance
Q UVT UV Dose EED Power Savings Log removal of
[gpm] [%] [mJ/cm2] [kWh/1000gal] [%] 1,4-Dioxane
8333 98 920 0.194 28 >0.5
8333 97 920 0.230 14 >0.5
8333 96 920 0.268 0 >0.5
29
Wedeco Full-Scale UV Reactor
• Wedeco K143 Series LP UV Reactor
• Validated for 6-log virus removal
• 12 lamps per row
• 17 rows of lamps
• 1 UV intensity sensor per row
• 600W lamps
• 34:1 linear power turndown
• Low headloss
• Add more rows for linear expansion
30
Acknowledgments on TIWRP Project
Roshanak Aflaki, Ph.D, P.E. ,
Plant Manager, Water Reclamation Division
Donald C. Tillman and Los Angeles-Glendale Water Reclamation Plant
31
The End of Our Presentation
Thank You! Questions?
Keel Robinson
keel.robinson@xyleminc.com
32
