Upload
buithu
View
217
Download
3
Embed Size (px)
Citation preview
10/20/2017 1
Pathogen Removal Mechanisms and Pathogen Credits in MBR-Based Potable Reuse Trains
Ufuk G. Erdal, PhD, PE
2017 NWRI Clarke Conference
Outline
• Background
• Objectives
• Comparison of Pathogen Credits
• MBR Pathogen Removal Mechanisms
• Approaches to Improve Pathogen Removals in an MBR Based AWTF
• Discussion
• Questions and Comments
2
Background
• Currently full advanced treatment (FAT) consisting of RO and AOP is required to meet GWRR via subsurface injection
• MF for pretreatment of RO
• Very effective; meets all primary and secondary MCLs, notification level chemicals, and other requirements
3
SODIUM HYPOCHLORITE
SULFURIC ACIDH2O2
MICROFILTRATIONREVERSE OSMOSIS
UVAOP
RO CONC.
OCSD
PLANT #1
SEC. EFF
LIME
BW Waste to
WWTP Influent OCSD OCEAN
OUTFALL
ANTISCALANT
DECARBONATOR
TO BARRIER INJECTION WELLS AND
SPREADING BASINS
Comparison of Pathogen Credits
4
Primary and SecondaryTreatment1
MF/UF RO UVAOP6-Month
Retention in Ground
TotalMinimum
LRV for GWR via Injection
Crypto 1.2 4 1-2 6 0 12.2-13.2 10
Giardia 0.8 4 1-2 6 0 11.8-12.8 10
Virus 1.9 0 1-2 6 6 14.9-15.9 12
1Based on lower 10th percentile values, Rose (2004)
MBR RO UVAOP6-Month
Retention in Ground
TotalMinimum
LRV for GWR via Injection
Crypto 0 1-2 6 0 7-8 10
Giardia 0 1-2 6 0 7-8 10
Virus 0 1-2 6 6 13-14 12
FAT
MB
R A
T
Background and Objective
• Due to lack of DIT or other approved methods to assess membrane integrity, no pathogen credits have been given to MBR in many states.
• Several questions arise.• Can we apply DIT to MBR systems for pathogen credits?• Can MBR, in reality, achieve better pathogen removal?
• What surrogate we can use to assess membrane integrity?
• Can MBR deserve pathogen credit?
• How can we improve pathogen credits in an MBR based AWTF?
• Objective is to provide answers to above questions.
5
MBR Pathogen Removal Mechanisms
6
1. Direct Removal via Size Exclusion - Molecules larger than membrane pore sizes will be rejected regardless of their surface properties (i.e., charge, polarity)
4 µ0.1 µ
2. Absorption into Biomass (MLSS) and Sequential Removal through Membrane Filtration
Floc
0.1 µ
3. Pore Blocking - Large molecules block the pores of the membranes and restrict passage of small viruses
CH0.1 µ
4. Reduction of Effective Pore Size due to Biofilm Growth, Gel and Cake Layer Formation on Membrane Surface and Pores
0.1 µ
5. Self Healing of Damaged Membranes
0.1 µ
Can We Apply DIT to MBR Systems?
• DIT begins by pressurizing membrane fibers from inside to approximately 12-20 psi about 30-45 seconds.
• Once the pressure is stabilized the pressure source was isolated and the decay test started. The pressure was recorded over a 5-minute, or till the pressure decreased to the minimum permissible pressure as required by the test resolution, whichever occurred first.
12
Dailypressure
decay test
Calculate pathogen removal
Concerns with DIT in MBR Systems
• DIT test pressure is relatively high and cannot be applied to flat sheet MBR membranes
• DIT test pressure also exceeds most of the MBR hollow fiber membrane suppliers pressure requirements (3-5 psi)
13
Concerns with DIT in MBR Systems
• Lack of correlation between PDT and LRV in MBR; due to the action of mechanisms other than pure size exclusion
– Pore blocking, cake and gel layer formation
– Presence of predator organisms that consume pathogenic organisms
– Absorption of pathogens to MLSS and their removal thru membrane filtration and periodic sludge wasting
– VCF?
14
Continuous Monitoring of Turbidity
15
Limited data collected from
NV tests indicate
good correlation between
MBR permeate turbidity and
MS-2 LRV
Australian Experience
• In Australia, Amos Branch and Pierre Le-Clech (2015) documented LRVs for MBR systems.
• Based on the data collected, they proposed the following LRV credits:
• For MBR systems, with 95th percentile permeate turbidity ≤ 0.4 NTU, and 95th percentile flux≤16.9 gfd
• Virus LRV: 1.5, Protozoa LRV: 2.0 and Bacteria LRV:4.0
• With membrane nominal pore size <0.1 µ, with 95th percentile turbidity ≤ 0.3 NTU and flux never exceeding 17.7 gfd.
• Virus LRV: 1.5, Protozoa LRV: 4.0 and Bacteria LRV:4.0
16
Now DDW is Providing Conditional Approval for MBR Projects in CA
• If 95th percentile turbidity <0.2 NTU and some other conditions meet:
• Virus LRV: 1.5, Crypto LRV:2.0 and Giardia LRV: 2.0
• Is it enough to meet pathogen LRVs?
17
MBR RO UVAOP6-Month
Retention in Ground
Total
MinimumLRV for
GWR via Injection
Crypto 2 1-2 6 0 9-10 10
Giardia 2 1-2 6 0 9-10 10
Virus 1.5 1-2 6 6 14.5-15.5 12
Approaches to Improve Pathogen Credits in an MBR Based AWTF
1. Get better pathogen credits for RO systems
2. Explore options to get better credit for UVAOP
3. Incorporate additional treatment processes into AWTF
18
1. Get Better Pathogen Credits for RO
19
Why Are We Getting ~2-log LRV for RO?
• Conductivity monitoring is widely used method to assess RO performance (i.e. salt passage)
• Easy to implement with a relatively low cost
• However, resolution of this method is low (up to 2-log)
20
CF CP
LRV RO=log(CF/CP)
TRASAR
21
TRASAR dye dosage in RO feed=40 –100 ppb
TRASAR dye in permeate=0.001 –0.05 ppb
3-3.5 LRV can be reliably demonstrated with TRASAR
DLR
V=2
.3
DLR
V=2
.4
(Fues, 2017)
MS
-2
Ob
se
rve
d
TR
AS
AR
Co
nd
.
2. Explore Better Pathogen Credits for UVAOP
22
Can We Get Better Pathogen Credits for UVAOP?
• Given that UVAOP gets 6-log credit for V/G/C
23
UVAOP Design
24
• UVAOP is designed to meet <10 ng/L NDMA
• Minimum 0.5-log 1,4-dioxane removal
• UV doses >850 mJ/cm2 are needed for 0.5-log 1,4-diaxone removal
• Multiple reactors/banks are operated in series to deliver the target UV dose
• Each reactor/bank is operated and monitored independently
UVAOP Validation at Oxnard
25
UVAOP was designed to meet
Minimum 1.2-log NDMA removal
Minimum 4-log MS-2 Inactivation
Three chambers in series each has two reactors, (total of 6 reactors), each reactor has 72 lamps
At MS-2 RED Dose of 115 mJ/cm2, each reactor can achieve >>4-log Crypto and Giardia inactivation
At an assumed validation factor of 3 for Crypto, five reactors in series could achieve
>20-log Crypto and Giardia Inactivation
• UVAOP + Free Chlorine
• Ozone + UV• Ozone + UVAOP• UVAOP+ Low
Dose UV• Chlorine +
UVAOP
26
Disinfection Processes in Series vs. UVAOP
Disinfection processes in series each may get up to 6 log pathogen credit (up to 12-log total)
For systems with multiple UVAOP reactors in series, may deserve credit for each reactor if all the monitoring requirements are met?
Are these disinfection processes in series different than UVAOP reactors in series?
27
3. Add a UV Disinfection System to MBR Based Train
28
UV Dose Requirements
A validation factor must be applied to these UV doses for full-scale operation
Where Can We Put UV Disinfection?
29
After MF/UF (Before RO) Lower UVT (typically 70-78%) Without residual chlorine does
not provide effective biological fouling control for RO
After RO Higher UVT (typically 95-98%)
Reduces both CAPEX and O&M Costs
MBR UV RO AOP
MBR UVRO AOP
MBR AOPRO UV
Discussion and Conclusions
30
• MBR is a robust secondary treatment technology that produces solids free effluent and effectively removes all particulate material and most of the pathogens
• Using TRASAR, Rhodamine Dye, TOC, Sulfate may provide 2-log additional LRV for RO over conductivity monitoring
• UVAOP may deserve a better pathogen credits than currently given • UV disinfection with <50 mJ/cm2 dose can provide 4-log Crypto and
Giardia inactivation
Questions and Comments