Reform The Enhanced Biological Phosphorus Removal Design

Civil & Environmental Engineering

Reform The Enhanced Biological Phosphorus Removal

Design to Enable Sustainable Nutrient

and Carbon/Energy Recovery

April .Z. Gu

Professor

School of Civil and Environmental Engineering

Cornell University

Ithaca, NY


Current Challenges in EBPR Practice

▪ Increasingly stringent permits demand higher consistency

and stability

▪ Backup chemical systems often required

▪ Sporadic metal salt addition negatively impacts P recovery

processes

▪ External carbon may be required to obtain desired C/P

ratio; increases carbon footprint

▪ Conflict between P and carbon diversion & N optimization

(i.e A- B stage, PN/PDNA)


Influent rbCOD/P Ratio Correlates with EBPR Stability

0

1

2

3

4

5

6

7

0 20 40 60

ffCOD/P (mg/mg)

Sta

bil

ity

Ra

nk

rbCOD/P >20-25 recommended

Gu et al., 2008, WER- survey of EBPRs in US

Sufficient rbCOD required for EBPR

Carbon supplement – external C, fermentate

5

rbCOD/P (mg/mg)

50

PAOs dominance

PAO & GAOsco-exist

GAOs dominance

WWTP rbCOD/P = 10-40

Liu et al. 1994; Schuler et al., 2003

EBPR Is Considered as Unfavorable for:

- Low influent C/P

- Fluctuating loading

- Not compatible with short-cut N removal processes

- Stringent limits: Chemical back-up needed for compliance


Enhanced Biological Phosphorus Removal (EBPR)Alternative Technology: Side-Stream EBPR (S2EBPR)

4

Influent

WAS

Aerobic

RAS

Anaerobic

Conventional EBPR

S2EBPR

Influent

WAS

AerobicAnoxic

RAS

Side-stream

Anaerobic

Reactor

5-30%

Anoxic

S2EBPR

• Emerging technology integrating on-

site sludge fermentation

• Offer advantages over conventional

▪ Influent C/P-independent

▪ Controlled anaerobic zone

▪ Favorable condition for PAOs

▪ Flexible implementation

S2EBPR outperform conventional EBPR

Full-scale pilot study

(Vollertsen et ;a.,2006; Barnard et al.,2017; Gu et al, 2019)

Full-scale implementation

Civil & Environmental EngineeringSlide 5

S2EBPR Survey in US - Various S2EBPR Configurations

Side-Stream RAS plus Carbon (SSRC)Side-Stream RAS (SSR)

Unmixed In-Line Fermentation (UMIF)Side-Stream MLSS (SSM)

Guetal.,WERFreport2019


0.001

0.01

0.1

1

10

0% 20% 40% 60% 80% 100%

Conc

entration(m

g/)

%ofValuesLessThanorEqualtoIndicatedValue

0.01

0.1

1

10

0% 20% 40% 60% 80% 100%

Conc

entration(m

g/)


0.001

0.01

0.1

1

10

0% 20% 40% 60% 80% 100%

Conc

entration(m

g/)


0.01

0.1

1

10

0% 20% 40% 60% 80% 100%

Conc

entration(m

g/)


12

TP target 1.5 mg/L

TP summer target 0.22 mg/L

TP target 2 mg/L

TP target 0.25 mg/L

TP_FE OP_SE

OP_SE

OP_SE

OP_FE

South Cary Westside Regional

Cedar Creek Henderson

Performance Survey of S2EBPR

3-year performance data

Conventional

EBPR*

50th

percentile0.05-0.8 [0.26]

90th

percentile0.2-2.5 [1.6]

90th/50th

ratio2-24 [11.5]

South

Cary

Westside

Regional

Cedar

CreekHenderson

50th

percentile0.28 0.04 0.82 0.32

90th

percentile0.89 0.10 1.10 1.00

90th/50th

ratio3.17 2.39 1.34 3.13

• Relatively stable performance were shown for all the 4 S2EBPR facilities,

as indicated by the 90th to 50th percentile ratio (90%/50%) for effluent P

levels.


Microbial diversity in S2EBPR plants is higher than those

in conventional EBPRs

7

Gu et al., WERF report2019, Onnis-Hayden et al., WER, 2019

• Plant-specific community

fingerprint

• Separation of S2EBPR vs

conventional plants

S2EBPR exhibit consistently

higher community diversity index


S2EBPR vs A2O

Full-scale pilot testing

Slide 8

Conventional S2EBPR

Phase Period Conventional S2EBPR

I-A Apr26th,2016-Jun21st,2016 A2OWestbank

(mixed,~1.5h,~100%)

I-B Jun22th,2016-Aug31st,2016 A2OWestbank

(inter.mixed,~1.5h,~100%)

conventional

S2EBPR

Performance

comparison

Gu et al., 2019 WERF, Wang et al., 2019 WR

Oligotype 5 and 6

were predominant

in Conventional

EBPR

Oligotype 2 was

predominant in

S2EBPR

Micro-diversity of Accumulibacter

Conventional

S2EBPR

Beforepilot

Temporal microbial community changes


0 20 40 60 80 100

0.125

0.25

0.5

1

2

4

8

16

32

Ram

an

Inte

nsity

Ratio

of G

lyco

ge

n (IG

LY /Ia

mid

e I )

S2EBPR-PP

EBPR-PP

Ram

an

In

ten

sit

y R

ati

o o

f P

oly

P (

I PP/I

am

ide

-I)

Cumulative frequency (%)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

S2EBPR-Gly

EBPR-Gly

Enhanced Biological Phosphorus Removal (EBPR)Phenotypic Changes at Single-Cell and

Functionally-Relevant Population Level

Increase in polyP, glycogen intensity among

individual PAO cells

Improved performance and stability in S2EBPR maybe associated with:

▪ Higher polyP and glycogen storage

▪ Higher PHA available for P uptake

PolyP increase

in PAOs in

S2EBPR

S2EBPR EBPR S2EBPR EBPR0

1

2

3

4

5

6

Ram

an

In

ten

sit

y R

ati

o o

f P

HA

(I P

HA/I

am

ide I)

PHA in PAO

PHA in GAO

PHA in PPB

PHA in all

UFAT feedingHAc feeding

Normalized PHA intensity in individual cells

(PPB: PHA-

producing

bacteria)

Glycogen

increase in

PAOs/GAOs in

S2EBPR


Conventional S2EBPR

C-source:

-Influent-dependent acetate-

dominant

- Acetate-using PAOs/GAOs

- Susceptible to influent changes

PAO/GAO competition:

- Hac uptake kinetics

- Ks based competition

Anaerobic Zone:

-impacted by recycles/influent

Configuration flexibility:

- Requires rbCOD/anaerobic

- Not compatible with carbon

diversion (A/B)

C-source:

- In situ fermentation, more

complex substrates mixture

- Diverse PAOs/GAOs using

various substrates

PAO/GAO competition:

- Other VFAs (propionate) favors

PAOs

- Differential decay

-Better controlled

-Larger anaerobic biomass % due to

higher MLSS & small split RAS flow

- Flexible implementation

- Compatible with carbon diversion

or short-cut N process

S2EBPR Reforms EBPR Design Strategy


▪ Lack of isolated key culture (e.g.

Accumulibacter PAO)

▪ Bulk-level studies cannot reveal

diversity of metabolic pathways

▪ Limitation of phylogenetic

methods

• Target known PAOs, GAOs

Enhanced Biological Phosphorus Removal (EBPR)Challenges in Studying EBPR Systems

▪ Candidate methods linking

phenotype to phylogeny

▪ MAR-FISH

• Isotope-based approach (e.g.

SIME)

• Functional omics approach

• Raman Microspectroscopy


Enhanced Biological Phosphorus Removal (EBPR)Available Microbial Ecology Tools for Full-Scale

S2EBPR Understanding

Neisser/DAPIStaining

AnalysisComplexity$$$

Fluorescenceinsituhybridization(FISH)

qPCR

RamanSpectroscopy

AmpliconSequencing Metagenomics

Raman-FISH-SIP

Gene-chips/RealtimePCR Metatranscriptomics

Metaproteomics

§ Quantificationof“totalPAOs”inBiovolumeFraction

§ QuantificationofknownPAOs”,“KnownGAOs”

§ Quantificationof“knownPAOs”bygenecopy

§ SpecieslevelAccuracy

§ Quantificationof“totalPAOs”,“totalGAO”

§ Single-cellmetabolicfingerprints

§ Cell,population-levelinformation

§ “Overallcommunity”inrelativeabundances

§ IncludingknownPAOs,GAOs

§ Nofunctionalinformation§ 16SrRNAGeneCopy

§ Functionalgenetranscriptionprofile

§ Potentialassembleofpart/wholegenomesinferring

EBPRrelatedpathways

Credit to YueYun Li, BV


inte

nsi

ty

Species 2

Species 1

Species 3

Simultaneous Phylogenetic Identification and Single

Cell phenotyping

Raman Single Cell Phenotyping

Cell-sortingphenotypic clustering analysis

“Single Cell” Genomic analysis: OTUs Raman phenotyping: OPUs

Genotype-function link

(Matrices correlation analysis, Machine learning)

Majed et al., 2009, 2010, 2012, 2020; Li et al., 2018; unpublished


Enhanced Biological Phosphorus Removal (EBPR)

▪ SCRS: a finer resolution phenotyping approach

Look Into Phenotypic Changes of PAOs and GAOs via

Single Cell Raman Spectroscopy (SCRS)

EBPR-related SCRS

fingerprintIdentify total PAOs and GAOsQuantify intercellular polymer dynamics at

population level

Cluster PAOs, GAOs based on

phenotypes, correlate with phylogenetic

diversity

Majed et al., 2009, 2010,2012； Li et al., 2018

OPU- operational phenotypic unit

Cell phenotyping via

metabolic state

Resolved C, P

mass flux

Population-level

Functions


Does S2EBPR Suppress GAOs?

Comparable known GAO abundance in S2EBPR vs conventional

Gu et al., WERF report 2019, Onnis-Hayden et al., WER, 2019

South CaryWestside

Regional

Cedar

CreekHenderson

Total GAOs (Raman) 2.9% 2.9% 2.0% 8.3%

Known GAOs (FISH) 0.7% 0.5% 0.3% 4.5%

Unknown

GAOs?


Enhanced Biological Phosphorus Removal (EBPR)S2EBPRs Enrich for Higher PAOs and GAOs that Use More

Diverse Carbon Sources

S2EBPR EBPR0

5

10

15

20

25

30

Re

lati

ve

ab

un

da

nc

e o

f P

AO

s,

% HAc feeding

UFAT feeding

Implications:

• S2EBPR select for other

substrate-utilizing PAOs

• Acetate-based assessment maybe

biased

When fed with fermentate

Higher PAOs in S2EBPR

Raman phenotype based total PAO and GAO quantification

S2EBPR EBPR0

5

10

15

20

**

HAc feeding

UFAT feeding

Re

lati

ve a

bu

nd

an

ce

of

GA

Os, %Implications:

• Other carbon-utilizing unknown

GAOs?

• More consistent with C/P ratio vs

PAO/GAO relationships

When fed with fermentate

Higher GAOs in S2EBPR


Evidence of “Sequential” Intracellular Polymer Utilization

Implications in Maintenance/Decay

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

PP in PAO GLY in PAO GLY in GAO

No

rmal

ized

Ram

an

inte

nsi

ty

0h 12h 24h 36h 72h

PolyP and glycogen use in PAOs:

Both consumed, then accelerated polyP usage after cessation of glycogen utilization up to 72 hrs

Single cell Raman microspectroscopy reveals temporal trend of polyP and

glycogen utilization in PAOs and GAOs under extended anaerobic condition

Glycogen in GAOs:

Quick glycogen utilization in 12 hours

Li, et al. (2019), unpublished


Model Development

ASM2(Henze et al,

2002)

Agent-

based

model(Bucci et al, 2012)

Foundation work

+S2EBPR

extension (Li et al, 2018)

Glycolysis-

TCA shift and

metabolic

versatility

Our previous work This study

• Maintenance precedes

decay

• PAO sequential polymer

usage in maintenance

• Glycolysis-TCA pathway shift

• Agent-based simulation on

PAO phenotypes of

glycolysis-TCA preferences

• PAO-GAO competition with

this pathway shift under

S2EBPR conditions

• PAOs, GAOs and OHOs

• PAO-GAO competition

• Agent-based modelling

framework


Agent-Based Modeling Simulation Showed

Differential Decay for PAOs vs GAOs

*PAOs have much delayed

decay due to its versatile

metabolic ability to use

multiple polymers for

ATP/NADH balance

*Differential decay

contributes to GAO

suppression under

extended anaerobic

condition 85%

90%

95%

100%

0

1

2

3

4

5

6

7

8

0 6 12 18 24 30 36B

iom

ass r

ela

tive

to

be

gin

nin

g o

f in

cu

ba

tio

n

Po

lyP

sto

rag

e (

mo

l P

P-P

/C-m

ol V

SS

)G

lyco

ge

n s

tora

ge

(C

-mo

l G

ly./

C-m

ol V

SS

)

Time during anaerobic incubation batch test (hours)

PAO Biomass


Take Home Messages

▪ S2EBPR is an alternative strategy that can address

some of the current challenges

▪ S2EBPR allows for flexible implementation with more

controllable, less influent carbon-dependent, more

favorable PAO enrichment over GAOs

▪ S2EBPR improves process stability compared to

conventional EBPR

▪ S2EBPR allows EBPR to be compatible with carbon

capture/redirection processes


S2EBPR Enables Sustainable Nutrient Removal and Carbon/Energy Recovery

A/B Process (PN/A) + S2EBPR pilot plant at Hampton Road Sanitation District, US

Anammox MBBR

CarbonAddition

A-stage WAS Fermentation

A-Process B-Process: Intermittant aeration, short-cut N removal

S2EBPR

Credit to: Charles Bott, Stephanie Klaus, HRSD team


Full Scale Implementation and Piloting

• Metropolitan Water Reclamation District of

Greater Chicago, Ill.

• Metro Wastewater Reclamation District, Denver,

Colo.

• Charlotte Water, NC

• Hampton Roads Sanitation District, VA

• Clean Water Services, OR

• Geneva, Ill.

• Western Wake WRD, NC

• Boulder, Colo.

• NEW Water, Green Bay, Wisc.

• Wilson, NC

• Trinity River Authority of Texas

• Madison Metropolitan Sewerage District, Wisc.

• Longmont, Colo.

• DC Water, Va.

• Toronto Water, Canada

• Olathe, KS

On-going WRF project

- More than 15 participating

facilities who will implement

or pilot the S2EBPR

- Develop design guidance and

monitoring strategies


Acknowledgements

Nicholas B. Tooker (Ph.D)

GuangYu Li (Ph.D)

Md Mahbubul Alam (M.S.)

Dongqi Wang (visiting scholar)

Varun Srinivasan (postdoc)

Annalisa Onnis-Hayden (Associate teaching

professor)


Acknowledgements

▪ WE&RF S2EBPR project team

• HRSD – Charles Bott

• Black & Veatch – James Barnard, Andy Shaw

• Woodard & Curran – Paul Dombrowski

▪ WRRF staff at all partner facilities: Rock Creek,

Cedar Creek, Westside Regional, South Cary,

Henderson, Meriden, Westfield, Ayer, North

Attleborough, Upper Blackstone

▪ Undergraduate research assistants at

Northeastern University, and Interns at City of

Olathe and Clean Water Services

▪ Dr. Amit Pramanik (WE&RF), Dr. JB Neethling

(HDR Inc.), Dr. H. David Stensel (University of

Washington), Dr. Glen Daigger (University of

Michigan), and Dr. Cliff Randall (Virginia Tech)

for their advice and support

Documents

Reform The Enhanced Biological Phosphorus Removal Design