An Attempt to Sustainably Stabilize EBPR Performance at Meriden, CT with Side-Stream EBPR

Transcription

1 An Attempt to Sustainably Stabilize EBPR Performance at Meriden, CT with Side-Stream EBPR Nick Tooker 1, Paul Dombrowski 2, Frank Russo 3, Annalisa Onnis-Hayden 1, April Z. Gu Northeastern University CEE Dept.; 2 Woodard & Curran; 3 City of Meriden, CT NEWEA Session 1, Innovative Treatment Technologies January 25, 2016 nbtooker@gmail.com

2 2 Acknowledgements Funding provided by WERF, HRSD, W&C City of Meriden staff Clean Water Services Durham Facility Undergraduate research assistants

3 3 Motivation for Improving EBPR EBPR operation is notoriously unstable Sometimes due to lack of carbon Meriden staff are frustrated because this is the only part of plant they can t control And permit limits keep getting lower

4 4 Highly Variable Effluent OrthoP in Meriden 0.7 mg/l TP permit limit (April October) Unstable Stable

5 Low influent rbcod:p Typical EBPR Process Configuration 5 Inconsistent effluent P PO 4 release VFA uptake PHA generation PO 4 uptake PHA consumption Cell growth RAS WAS

6 6 S 2 EBPR Process Configuration Side-Stream RAS 80-95% RAS WAS ~1 day HRT ~2-5 day SRT 5-20% RAS

7 7 Hypothesized Ways S 2 EBPR Improves Stability Is there VFA production in side-stream? Is there active VFA uptake in side-stream reactor *? Is there a shift in microbiological population? To more efficient polyphosphate accumulating organisms (PAOs)? With fewer glycogen accumulating organisms (GAOs)? * - Bi et al, 2013 and Lopez et al, 2006

8 8 What We Currently Know About S 2 EBPR In operation at full-scale facilities 50+ in Europe (mostly Denmark) ~6 in North America No consensus on operation Several different flow schemes Standard models (e.g., BioWin, GPS-X) don t fit observed data Fundamental understanding is lacking

9 9 Testing to Understand S 2 EBPR Simulated S 2 EBPR Batch Testing Meriden, CT Durham (Clean Water Services, Tigard, OR) Westside Regional (West Kelowna, BC) Cedar Creek (Olathe, KS) S 2 EBPR Pilot Testing Meriden, CT

10 10 Simulated S 2 EBPR Batch Testing Reactors Sludge from Meriden (aerobic MLSS) and Durham (TWAS) Similar initial MLVSS of ~ 6,000 mg/l 3-day anaerobic incubation Mixing once per day during sample collection

11 11 VFA Production in Simulated S 2 EBPR Batch Test Durham 218 mg/l as VFA 38 mg/l as VFA 121 mg/l as VFA 5 mg/l as VFA Residual VFA quadrupled and scod tripled after just one day

12 Low ORP allows for fermentation and VFA production 12 Low ORP in Simulated S 2 EBPR Batch Tests Durham Fermentation ORP Methane formation ORP

13 13 Key Organisms in EBPR Processes Accumulibacter Commonly known PAOs Important for effective EBPR Tetrasphaera Lesser known PAOs Widely present in WRRFs (15%+ of population) Some are also fermenters Competibacter Commonly known GAOs Competes with PAOs for VFA

14 14 Microbiological Population Shifts in S 2 EBPR Batch Test Durham Accumulibacter decreased Tetrasphaera didn t change Low GAO proportion

15 15 Sample from Durham Facility (Tigard, OR) EUB mix (general probe) in green; Accumulibacter in yellow

16 16 Sample from Durham Facility (Tigard, OR) EUB mix (general probe) in green; Tetrasphaera in red & orange

17 17 Meriden S 2 EBPR Pilot Goals Effectively implement S 2 EBPR Stabilize EBPR operation Reduce ferric chloride use Improve understanding of process Minimize effort for plant staff

18 18 Meriden S 2 EBPR Pilot Overview (Mar-Aug 2015) Primary Effluent Nitrate Recycle Target SRT ~ 3 days Actual SRT ~ 1-2 days RAS WAS Aerobic MLSS was feed to side-stream reactor (unused clarifier)

19 1-2 day HRT not long enough for VFA generation with aerobic MLSS 19 Delayed VFA Production in Meriden S 2 EBPR Batch Test Durham Meriden

20 ORP not low long enough for residual VFA generation in first 2 days 20 Low ORP in Simulated S 2 EBPR Batch Tests Meriden Fermentation ORP Methane formation ORP

21 21 In S 2 EBPR Pilot Reactor: Low DO, too High ORP DO, mid blanket sludge blanket 4-5 feet deep ORP, mid blanket ORP, blanket bottom DO, blanket bottom Fermentation ORP While DO in sludge blanket was low; ORP was too high for fermentation

22 Possibly due to elevated ORP in S 2 EBPR reactor Minimal Improvements to EBPR Stability w/ Aerobic MLSS S 2 EBPR During Meriden Pilot 22 Secondary Effluent PO 4

23 Increase carbon to anaerobic zone by primary sludge fermentation! 23 Meriden s Fix: Increase PC Blanket Level for VFA PC blanket Effluent PO 4 FeCl 3 dose

24 24 Notes from the Operators Be committed for significant additional sampling, analyses, and labor Communicate constantly with engineers and researchers Be willing to experiment Use your expertise!

25 25 Takeaway Messages VFA production occurred in simulated S 2 EBPR batch reactors But low ORP and adequate HRT required Aerobic MLSS is a poor feedstock for S 2 EBPR reactor Getting ORP low enough is problematic RAS, WAS, or anaerobic MLSS preferred Highly trained and engaged treatment plant staff is critical

26 26 Meriden S 2 EBPR Pilot Next Steps Pilot test #2, March 2016 Alternative operation with RAS or TWAS instead of aerobic MLSS Reduce ORP in reactor Increase VFA production

27 27 References Lopez, C.; Pons, M.N.; Morgenroth, E. (2006). Endogenous processes during long-term starvation in activated sludge performing enhanced biological phosphorus removal. Water Research, 40, Bi, D.; Gou, X.; Chen, D. (2013). Phosphorus release mechanisms during digestion of EBRP sludge under anaerobic, anoxic and aerobic conditions. Water Science & Technology, 67(9),

28 Discussion & Questions Nick Tooker, P.E., Ph.D. Student (April Gu research group) Northeastern University Civil & Environmental ( ) April Z. Gu: april@coe.neu.edu Frank Russo: frusso@meridenct.gov