minutes of march 12, 2014 at littleton/englewood wwtpcwwuc.org/minutes/2014/mar.pdfmarch 12, 2014 ....

81
1 Minutes of March 12, 2014 at Littleton/Englewood WWTP In Attendance: Martha Rudolph CDPHE Dick Parachini WQCD Michael Beck WQCD Jim Dorsch Metro Al Baker CWSD Donna Davis WQCD Bobby Anastasov Aurora Water Nancy Keller (via phone) City of Pueblo Dennis Stowe L/E WWTP Colleen Young Greeley Water & Sewer Mary Gardner L/E WWTP Richard Leger Aurora Water Dan DeLaughter BHCCSD/Leonard Rice Engineers Dave Meyer Westminster Bill Veydovec HMM Shelley Stanley Northglenn Jim Kendrick (via phone) Monument Sanitation District Martha Hahn PCWRA Ginny Johnson (via phone) Colorado Springs Utilities Tad Foster Tad Foster Law Firm Mike Bittner Silverthorne/Dillon Julie Tinetti L/E WWTP Jim Edwards Eagle River Water & Sanitation District Paul Ferraro CWWUC I. State Environmental Program, Legislative Overview, Martha Rudolph (CDPHE) Martha thanked the Council and was very pleased that over 50 WQCD staff visited water and wastewater treatment plants. She said that it is good for the staff. Paul mentioned the new Council program, Building Relationships, and Martha said that she saw it on the Council’s websites. She would like a tour of the L/E WWTP. Legislation o SB 14-134 was introduced but pulled since it did not give any of those affected a full opportunity to study the full implication of the legislation. The Department and the Division will conduct a stakeholder process between now and the Fall when the Division reports back to the Legislature recommending how the funding of the Division should be handled. Presently, the fees for funding the Division are in legislation. SB 14-134 was to allow the WQCC to set the fees and remove them from legislation. Three years ago a new JBC analyst recommended that the WQCC set the wastewater fees for the Division staff. Finally, the JBC approved the concept in 2014 as SB 14-134. Tad Foster

Upload: hadan

Post on 21-Apr-2018

216 views

Category:

Documents


2 download

TRANSCRIPT

1

Minutes of

March 12, 2014 at Littleton/Englewood WWTP

In Attendance:

Martha Rudolph CDPHE Dick Parachini WQCD Michael Beck WQCD Jim Dorsch Metro Al Baker CWSD Donna Davis WQCD Bobby Anastasov Aurora Water Nancy Keller (via phone) City of Pueblo Dennis Stowe L/E WWTP Colleen Young Greeley Water & Sewer Mary Gardner L/E WWTP Richard Leger Aurora Water Dan DeLaughter BHCCSD/Leonard Rice Engineers Dave Meyer Westminster Bill Veydovec HMM Shelley Stanley Northglenn Jim Kendrick (via phone) Monument Sanitation District Martha Hahn PCWRA Ginny Johnson (via phone) Colorado Springs Utilities Tad Foster Tad Foster Law Firm Mike Bittner Silverthorne/Dillon Julie Tinetti L/E WWTP Jim Edwards Eagle River Water & Sanitation District Paul Ferraro CWWUC

I. State Environmental Program, Legislative Overview, Martha Rudolph (CDPHE)

• Martha thanked the Council and was very pleased that over 50 WQCD staff visited water and wastewater treatment plants. She said that it is good for the staff. Paul mentioned the new Council program, Building Relationships, and Martha said that she saw it on the Council’s websites. She would like a tour of the L/E WWTP.

• Legislation

o SB 14-134 was introduced but pulled since it did not give any of those affected a full opportunity to study the full implication of the legislation. The Department and the Division will conduct a stakeholder process between now and the Fall when the Division reports back to the Legislature recommending how the funding of the Division should be handled. Presently, the fees for funding the Division are in legislation. SB 14-134 was to allow the WQCC to set the fees and remove them from legislation. Three years ago a new JBC analyst recommended that the WQCC set the wastewater fees for the Division staff. Finally, the JBC approved the concept in 2014 as SB 14-134. Tad Foster

2

recommended that a credible process is needed. The first stakeholders’ meeting will be by the end of March.

o HB 14-1002 – Water Infrastructure Natural Disaster Grant Fund. The Bill provides $12 million for facilities (water and wastewater) damaged in the flood and wild fires. The Division supports the Bill. Funds not spent by 9/1/2015 will go to the Nutrient Fund.

o Simple Project. Martha mentioned that the Department is working on putting online records, permits, bill paying, etc. Someone will have to pay for building the online system. The CDPHE is presenting working on it and then they will develop a pilot project to insure it works. Martha will attend a fall Council meeting to update the Council.

II. Funding: State Revolving Fund, Donna Davis and Michael Beck (CWQCD) Highlights of their presentation are:

• Flood damage to water and wastewater treatment plants is estimated at $170 million.

• For every dollar of Federal funds, the State needs to match it with $.20. The Water & Power Authority receives the Federal funds and provides the match.

• Staff prepares a 2-year Workplan for the SRF that is approved by the CWQCC and the Water & Power Authority Board.

• Federal funds increase by 5% for 2014 to $1.4 Billion and Colorado’s share is $11.2 million for Clean Water and $15.4 million for Drinking Water. Buy America is back in the program.

• The Water & Power Authority can fund up to $100 million of projects for Clean Water and $70 million for Drinking Water.

• Rulemaking Hearing to approve the Division’s Plan is set for May 12, 2014 for projects damaged in the flood for water and wastewater projects.

• The Legislation for the Natural Disaster Fund will only need funding in future years.

• HB 14-1002 passed the House and is going to the Senate.

III. Nitrogen Removal 1.0 to 3.0: Sidestream and Mainstream, Charles Bott Charles gave a detailed and technical presentation on nitrogen removal which is attached.

IV. Updates

• Sediment Workgroup: Nancy updated the Council regarding activities of the workgroup. She is recommending that the Council continue funding GEI’s work with an additional $5,000 to cover 3 more workgroup meetings in the next month. The Council has funded $4,510 for workgroup meetings that needed GEI’s technical review.

Jim Kendrick moved, seconded by Richard Leger, that the Council provides $5,000 in support for the sediment project. Approved.

• NACWA Integrated Planning Workshop is March 31, 2014. Dennis Stowe moved, seconded by Richard Leger, that the Council contributes $300 to NACWA for the Planning Workshop. The motion passed unanimously.

• RM Arsenal Aquatic Life Hearing. Dennis attended the CWQCC hearing regarding the RM Arsenal. The hearing was to classify the lakes on the Arsenal as “Use protected for Aquatic Life.” The US Parks Service presented the proposal. The Division and EPA

3

opposed the recommendation. The Commission did not approve the recommendation from the US Park Service. The vote was 5 against and 4 in favor. Dennis said that this is a strong case against reuse. Denver Water participated in the hearing. Copper level is the issue. The designation remains as “reviewable.” Tad Foster recommenced that the Council send a letter to the Division and Denver Water asking that an antidegradation review be conducted.

• Colorado water Plan. The Colorado Water Plan will be discussed at the Colorado Water Quality Forum next week. Attached are the Council’s comments.

• Colorado Water Utility Council – No report.

• Stormwater – No report. Meeting adjourned at 4:00 pm.

Next meeting:

April 9, 2014 at 1 PM

Transitioning from Nitrogen Removal 1.0 to 3.0 --

Sidestream and Mainstream

Denitrification

NH4+

N2

NO2-

Anammox

Nitrification

NO3-

N-fixation

Charles B. Bott, PhD, PE, BCEE

Hampton Roads Sanitation District

Hampton Roads Sanitation District

• Created in 1940

• Serves 1.6 million people

• Includes 17 jurisdictions – 3,100 square miles

• 9 major plants, 4 small plants

• Capacity of 249 MGD

HRSD’s Bubble Permit - 2011

• James River

– 6,000,000 lbs/yr TN

– 573,247 lbs/yr TP

• York River

– 288,315 lbs/yr TN

– 33,660 lbs/yr TP

• Rappahannock River (one plant)

– 1,218 lbs TN

– 91 lbs/yr TP

Chesapeake Bay TMDL & VA WIP

• Nitrogen – James River – 2011 – 6.0 million pounds/year

• Major upgrades ongoing at Nansemond, James River, Williamsburg, Army Base

– 2017 – 4.4 million pounds/year • VIP - biological process upgrade for improved denitrification

• Small upgrade at Williamsburg possible

– 2021 – 3.4 million pounds/year (possible?) • Upgrade Chesapeake-Elizabeth (full plant)

• Nitrogen – York River – Rapid upgrade to add denite filters for 2011 compliance

– Additional upgrade needed for cost-effective BNR and reliability

• 1% of Total Plant Influent Flow

• Rich in Nitrogen & Phosphorus

• 15 to 25% of the Total Plant TN load

• Ammonium Conc. 800 to 1,500 mg-N/L

• Temperature 30 - 38C

• Alkalinity insufficient for complete

nitrification

• Insufficient carbon for denitrification

• For a Bio-P plant with no iron addition:

• Centrate TP = 200-800 mg/L

Influent Primary

Clarifier Secondary

Clarifier

Effluent

Centrate

Primary Sludge WAS

Dewatering

Thickening

RAS

Anaerobic

Digestion

Biosolids

Aeration

Tank

Recycle Streams with High Ammonia - Sidestream

6

Sidestream Treatment Options

Biological - N Physical-Chemical – N&P

Ion-Exchange • ARP

Struvite Precipitation • Ostara Process • PhosPaq Process

Nitrification / Denitrification & Bioaugmentation

• With RAS & SRT Control • With RAS • Without RAS

Nitritation / Denitritation • Chemostat • SBR • Post Aerobic Digestion

Deammonification • Suspended Growth SBR • Attached Growth MBBR • Upflow Granular Process

Ammonia Stripping • Steam • Hot Air • Vacuum Distillation

7

1.0

2.0

3.0

Conventional Nitrification-Denitrification

1 mole Ammonia

(NH3 / NH4 +)

½ mol Nitrogen Gas

(N2 )

1 mole Nitrite

(NO2-)

1 mole Nitrite

(NO2-)

1 mole Nitrate

(NO3-)

Autotrophic Bacteria

Aerobic Environment

Heterotrophic Bacteria

Anoxic Environment

75% O2 (energy)

~100% Alkalinity

25% O2 (energy)

40% Carbon (BOD)

60% Carbon (BOD)

Ammonia Oxidizing Bacteria (AOB)

Nitrite Oxidizing . Bacteria (NOB)

8

Sidestream Treatment with Nitrifier Bioaugmentation

BABE, AT-3, BAR, CaRRB, Maureen, etc.

PC

Influent Sec. Effluent

Activated Sludge Tank

RAS

WAS

Centrate

(NH3-N) Nitrification

Reactor

~250C

Nitrifiers

NO3-N

Bioaugmentation is the incentive for 1.0

9

Sidestream Treatment Options

Biological - N Physical-Chemical – N&P

Ion-Exchange • ARP

Struvite Precipitation • Ostara Process • PhosPaq Process

Nitrification / Denitrification & Bioaugmentation

• With RAS & SRT Control • With RAS • Without RAS

Nitritation / Denitritation • Chemostat • SBR • Post Aerobic Digestion

Deammonification • Suspended Growth SBR • Attached Growth MBBR • Upflow Granular Process

Ammonia Stripping • Steam • Hot Air • Vacuum Distillation

10

1.0

2.0

3.0

Nitritation-Denitritation = “Nitrite Shunt” (2.0)

1 mole Ammonia

(NH3 / NH4 +)

½ mol Nitrogen Gas

(N2 )

1 mole Nitrite

(NO2-)

1 mole Nitrite

(NO2-)

1 mole Nitrate

(NO3-)

Autotrophic Bacteria

Aerobic Environment

Heterotrophic Bacteria

Anoxic Environment

75% O2 (energy)

~100% Alkalinity

25% O2 (energy)

40% Carbon (BOD)

60% Carbon (BOD)

Ammonia Oxidizing Bacteria (AOB)

Nitrite Oxidizing . Bacteria (NOB)

Advantages:

• 25% reduction in oxygen demand (energy)

• 40% reduction in carbon (e- donor) demand

• 40% reduction in biomass production 11

Nitritation

Denitritation

Sidestream Nitritation – NOB Repression

• Control

– Elevated temperature (30-35 deg C)

– Low SRT (1-2 days)

– Low DO (~0.5 mg/L)

• NOB Repression Mechanisms (all the possibilities)

– AOB max growth rate > NOB max growth rate at high temp

– Free NH3 inhibition of NOB > AOB

– AOB DO affinity > NOB DO affinity (perhaps only at high temp)

– Nitrous acid inhibition of NOB > AOB

12

SHARON Experience

WWTP Capacity

(pe)

SHARON

kgN/day

Operational

Utrecht 400.000 900 1997

Rotterdam-Dokhaven 470.000 850 1999

Zwolle 200.000 410 2003

Beverwijk 320.000 1,200 2003

Groningen-Garmerwolde 300.000 2,400 2005

The Hague - Houtrust 430.000 1,300 2005

New York-Wards Island 2,000,000 5,770 2009

Whitlingham, UK 275.000 1,500 2009

MVPC Shell Green, UK - 1,600 2009

Geneva – Aïre 2 600.000 1,900 2010

Paris Seine Grésillons 3,500 2010

• 6 operational >10 years experience

• 5 planned

• NYC DEP Wards Island

– First in USA & largest in world

30 - 40% TKN-load

Sidestream Treatment Options

Biological - N Physical-Chemical – N&P

Ion-Exchange • ARP

Struvite Precipitation • Ostara Process • PhosPaq Process

Nitrification / Denitrification & Bioaugmentation

• With RAS & SRT Control • With RAS • Without RAS

Nitritation / Denitritation • Chemostat • SBR • Post Aerobic Digestion

Deammonification • Suspended Growth SBR • Attached Growth MBBR • Upflow Granular Process

Ammonia Stripping • Steam • Hot Air • Vacuum Distillation

14

1.0

2.0

3.0

The N-Cycle

Denitrification

NH4+

N2

NO2- Anammox

Nitrification

NO3-

N-fixation

15

Partial Nitritation-Anammox = “Deammonification” (3.0)

1 mole Ammonia

(NH3 / NH4 +)

½ mol Nitrogen Gas (N2 ) +

a little bit of nitrate (NO3-)

0.5 mole Nitrite

(NO2-)

Autotrophic Bacteria

Aerobic Environment

Autotrophic Anoxic

Environment 37% O2 (energy)

~50% Alkalinity Ammonia Oxidizing Bacteria (AOB)

Advantages:

• 63% reduction in oxygen demand (energy)

• Nearly 100% reduction in carbon demand

• 80% reduction in biomass production

• No additional alkalinity required

ANAMMOX “Anaerobic” Ammonia Oxidation - (New Planctomycete - Strous et al, 1999)

NH4+ + 1.32 NO2

- + 0.066 HCO3- + 0.13 H+

0.26 NO3- + 1.02N2 + 0.066 CH2O0.5N0.15 + 2.03 H2O

16

One-Step Sidestream Deammonification • SBR + Hydrocyclone Granular Sludge (DEMON)

– Strass, Austria + ~20 others

– Cyklar-Stulz – World Water Works, Inc.

• Upflow Granular Sludge (CANON/ANAMMOX) – Olburgen, Netherlands + ~7 others

– Paques (NL)

• Biofilm process (MBBR-style)

– ANITA Mox -- Malmo & Växjö, Sweden • AnoxKaldnes – Kruger - Veolia

– Deammon -- Hattingen, Germany & Stockholm • Purac

Centrate

NH4+

17

Partial Nitritation and Anammox - combined in a single reactor

Sidestream Deammonification: What’s the benefit?

• Remove ~20% of the N load to the plant by treating the centrate separately

• Do it with: – No chemicals (caustic & methanol)

– < 40% of the energy cost

– (as compared to traditional nitrification-denitrification)

• Risks: – Requires robust process control, particularly during startup

– Process has been adequately demonstrated in Europe

– Seeding required for fast startup

18

S ide st re a m

D e a m m onif ica t ion S t a t us in

N ort h A m e r ica – DEMON – HRSD York River; Started October 2012; operating – ANITA Mox – HRSD James River; Started November 2013; operating

– DEMON – Industrial Project Orlando, FL; started January 2014; operating

– DEMON – Alexandria, VA; in construction, startup 2014 – ANITA Mox – South Durham, NC; in construction – DEMON – Philadelphia, PA; 90% design – DEMON- Guelph, Ontario; 90% design – DEMON – Pierce County, WA; pilot complete; 90% design – DEMON – DCWater Blue Plains; 60% design – ANITA Mox – Chicago Egan MWRDGC; in design – DEMON – Greeley, CO, in design

– DEMON – New York DEP; pilot completion recent – ANITA Mox – LA County San District; pilot completion recent – DEMON – Chicago Egan MWRDGC; pilot completion Feb 2013 – ANITA Mox - Denver MWRD; pilot completion Feb 2013 – MBBR-style process – New York DEP; pilot ongoing – CLEARGREEN (SBR) – Henrico County, VA; pilot during 2012

– DEMON - Alexandria, VA + DCWater pilot (no cyclone) – DEMON – New York DEP + DCWater pilot (no cyclone)

D EM O N at H RS D Yo rk Riv er (1 5

M G D )

DEMON

DENITE FILTERS

HEADWORKS

AERATION BASINS

ANAEROBIC DIGESTION

THICKENING

DEWATERING

Implementation of DEMON at York River

22

ANITA Mox Process

Media

Anammox

Nitritation

NH4+

NH4+

O2 N2

NO2-

Aerobic

Anoxic

AOB

HRSD James River Treatment Plant

S-curve for Sidestream Deammonification

25%

75%

100 Fullscale Installation

Municipal

Industrial

Susanne Lackner, Eva M. Gilbert, Siegfried E. Vlaeminck, Adriano Joss, Harald

Horn, Mark C.M. van Loosdrecht (2014), Full-scale Partial Nitritation/Anammox

Experiences - an Application Survey, Water Research

In the US

Year

2004 2006 2008 2010 2012 2014

Scie

ntific P

ub

lica

tio

ns

0

50

100

150

200

250F

ull-

sca

le I

nsta

llatio

ns

0

2

4

6

8

10

Publications

Installations

Short-Cut Nitrogen Removal Processes: Transitioning to Mainstream 2.0 & 3.0

COLLABORATORS

Conventional Nitrification-Denitrification (1.0)

1 mole Ammonia

(NH3 / NH4 +)

½ mol Nitrogen Gas

(N2 )

1 mole Nitrite

(NO2-)

1 mole Nitrite

(NO2-)

1 mole Nitrate

(NO3-)

Autotrophic Bacteria

Aerobic Environment

Heterotrophic Bacteria

Anoxic Environment

75% O2 (energy)

~100% Alkalinity

25% O2 (energy)

40% Carbon (BOD)

60% Carbon (BOD)

Ammonia Oxidizing Bacteria (AOB)

Nitrite Oxidizing . Bacteria (NOB)

29

Two-Sludge BNR High Rate Activated Sludge

Advantages Disadvantages

High nitrogen removal possible WW carbon not utilized for denitrification

Low BNR volume Nitrification can be alkalinity limited

SRT 1-3 days

DO 2 mg/L

HRAS

Carbon

Nitrification Denitrification TN 3 - 15 mg/L

AnoxicAerobic

Conventional BNR - MLE

Advantages Disadvantages

WW carbon utilized for denitrification Large BNR volume

Alkalinity recovered Nitrogen removal limited by IMLR

Anoxic Aerobic

IMLR

NitrificationDenitrification TN 10 - 12 mg/L

4-Stage Bardenpho (Better N Removal)

Aerobic

SC

RAS WAS

air

Anoxic

Ae

rob

ic

air

Anoxic

Methanol TN ~ 3-5 mg/L Primary

Effluent

BOD + NH4

Nitrate Recycle (NRCY)

32

NH4+

Probe

Anoxic Aerobic

IMLR

Nitrification/SNDDenitrification TN 10 - 12 mg/L

B-stage BNR

Anoxic Aerobic

IMLR

NitrificationDenitrification TN 16 - 18 mg/L

Adsorption/Bio-oxidation (A-B) Process

Advantages Disadvantages

Low overall volume Requires ammonia-based aeration control

Good nitrogen removal Not operated to achieve complete nitrification

Redirect carbon to anaerobic digestion

Low aeration energy requirement

Anoxic Aerobic

IMLR

NitrificationDenitrification TN 10 - 12 mg/L

SRT 6-12 hours

DO ~0 mg/L

A-stage HRAS

CONVENTIONAL A-STAGE

Simulated COD Balance of the AIZ Strass WWTP

Wett, B.; Buchauer, K.; Fimml, C. (2007) Energy self-sufficiency as a feasible concept for wastewater treatment systems.

Proceeding of the IWA Leading Edge Technology Conference, Singapore, Asian Water, 21-24.

35.4%

80-90%

Waste Sludge 10-20%

Waste Sludge

An Ideal Configuration…

36

Tools for SND-Style Processes (2.0)

• Ammonia-based Aeration Control

– Allows stringent control over DO provided

– Control aerobic SRT to be as long as needed

• NOB Repression

– Must be controlled and confirmed

• Deliver just the right amount of COD and use it for denitritation

37

Objectives for 3.0

• Redirect unnecessary carbon/COD – Primary clarifier (likely insufficient for typical HRSD wastewater) – A-stage HRAS – Chemically Enhanced Primary Treatment (Fe or Al salt + polymer) – Anaerobic treatment (UASB, AnMBR) – (Minimizes B-stage volume required)

• Repress NOBs under difficult conditions – Low temp – Low NH4

• Selectively Retain Anammox (high SRT needed) – Granular sludge – Biofilm process (e.g. MBBR) – Membrane bioreactor

38

Plants with Anaerobic Digestion • Incentive for C-redirection

Key Features:

1. Bioaugment AOB in the sidestream cyclone overflow to the mainstream

2. Bioaugment anammox from sidestream to mainstream

3. Cyclone for anammox retention in mainstream

4. Repress NOB in mainstream (and sidestream)

C-redirectionMainstream

Deammonification

AnaerobicDigester

Sidestream Deammonification

Being piloted at DC Water and tested full-scale at Strass, AT and Glanerland,

CH

WERF-Mainstream Deammonification Project 3 different sites and scales

DC Water WWTP Strass HRSD

Anammox Retention

Approaches

Granulation:

1) Settlers (internal or external)

2) Cyclones

3) Sieves

Biofilm:

4) Plastic Media

43

Strass Demonstration • Carousel type aeration tank at Strass WWTP providing a DO-range of 0 to

1.7 mg/L along the flow-path.

-45-

Cyclones installed at the B-stage in Strass,

Cyclone A (left), Cyclone B since early

September 2011 (right).

DO=1.4-1.5 mg/L 0.19 mg/L 0.35mg/L

0.010.09mg/L0.6mg/L 1.6-1.7mg/L

Dutch Approach

Key Features:

1. Large AOB-Anammox granules seeding from sidestream to mainstream

2. Large AOB-Anammox out-compete NOB in mainstream

C-redirectionMainstream

Deammonification

AnaerobicDigester

Sidestream Deammonification

Being piloted by TU-Delft at Rotterdam, NL

For plants with anaerobic digestion

HRSD Mainstream Approach

Anaerobic digestion is not necessary

C-redirection Nitrite-shunt anammox

• Minimum aeration and volume for C-removal

• Reduce volumetric requirement for nitrogen removal

• Promote nitrite shunt pathway to achieve more nitrogen removal for a given influent C/N

• Produce effluent containing ammonia and nitrite for anammox polishing

• Remove remaining nitrogen autotrophically without additional aeration energy and supplemental carbon

• Meet very low effluent TIN limits

RAS

RAS

WAS

RAS

Chiller

Basket

Strainer

IMLR

Hampton Roads Sanitation District

Chesapeake-Elizabeth Nutrient Removal Pilot Study

Process Flow Schematic

Alkalinity

addition

AIR

AIR

A-Stage High Rate Activated

Sludge (HRAS) Process

B-Stage Deammonification Process

B-Stage SND/Nitrite Shunt Process

Forward Flow

Return Activated

Sludge (RAS)

Internal Mixed Liquor

Recycle (IMLR)

Alternative Flow

Anoxic

Aerobic

Anaerobic

Clarification

Preliminary TreatmentWaste Activated

Sludge (WAS)

Compressed Air Process Effluent

LEGEND

= Centrifugal pump

= Peristaltic pump= Membrane

Air diffuser= P.D. pump

= Stainless Steel

Finned Coil

= Immersion

Heater

= MBBR Media

Off-gas

Analyzer

Primary Clarification

Process

= Tank Mixer

AIR

TK-105

HRAS

TK-305

ANAMMOX

MBBR

FROM CETP

GRIT TANK

EFF.

CHANNEL

Preliminary

Treatment

Alkalinity

addition

Chiller

TK-103

TEMP.

CONTROL

TK-101

SCUM AND

GRIT

REMOVAL

TK-106

CLARIFIER TK-107

HRAS

EFFLUENT

TK-401

PRIMARY

CLARIFIER

TK-402

TEMP.

CONTROL

TK-202TK-205

CLARIFIERTK-203 TK-204

TK-302

PARTIAL

NITRITATION

TK-304

CLARIFIER

WAS

WAS

PHASE I PILOT

P204

RAS

CHILLER

B

Hampton Roads Sanitation District

Chesapeake-Elizabeth Nutrient Removal Pilot Study

Process Flow Schematic

AIR

A-Stage HRAS Train BB-Stage Ammonia Versus NOx-N (AVN) Process

Forward Flow

Return Activated

Sludge (RAS)

Internal Mixed Liquor

Recycle (IMLR)

Alternative Flow

Anoxic

Aerobic

Anaerobic

Clarification

Preliminary TreatmentWaste Activated

Sludge (WAS)

Compressed Air Process Effluent

LEGEND

= Centrifugal pump

= Peristaltic pump= Membrane

Air diffuser= P.D. pump

= Stainless Steel

Finned Coil

= Immersion

Heater

= MBBR Media

Off-gas

Analyzer

A-Stage HRAS Train A

= Tank Mixer

AIR

TK-107B

FROM CETP

GRIT TANK

EFF.

CHANNEL

Preliminary

Treatment

TK-103B

TEMP.

CONTROL

TK-101

SCUM AND

GRIT

REMOVAL

TK-108B

CLARIFIERTK-109B

A-STAGE

EFFLUENT

TK-202

TK-205

CLARIFIER

TK-203 TK-204

P105A

WAS

TK-106B TK-105B

TK-201

Off-gas

Analyzer

AIR

TK-107A

TK-108A

CLARIFIER

TK-106A TK-105A

TK-206

ANAMMOX

MBBR

CHILLER

A

TK-103A

TEMP.

CONTROL

TK-102B

BASKET

SCREEN

TK-102A

BASKET

SCREEN

P101

P102B

P102A

P103B

P103A

P104B

P104A

TK-109A

EFFLUENT

BUCKET

P106B

RAS

P106A

RAS P105A

WAS

P201

B-Stage Inf

TK-110A

CLARIFIER

Waste

P107A

CHILLER

C

TK-208

TEMP.

CONTROL

P209

Anammox

Feed

CETP NPW

P210

Chemical

Feed

TK-207

EFFLUENT

BUCKET

P205

Anammox

Bypass

P203

WAS

TK-207

EFFLUENT

BUCKET

P208

PHASE II PILOT

Mainstream NOB Out-Selection

52

Ammonia Pressure

Ammonia levels and temperature are too low for free NH3 inhibition

Residual ammonia allows AOB to grow closer to their maximum growth rate

0

0.2

0.4

0.6

0.8

1

0 0.5 1 1.5 2 2.5 3 3.5 4

Sp

ecif

ic g

row

th r

ate

(1

/d

)

mg/L (AOB:NH4+-N, NOB:NO2

--N)

mu-AOB mu-NOB

Monod Curves for AOB and NOB

Strategy # 1 Maintain Residual NH4-N > 1.5 mg/L

54

DO (mg/L)

0.0 0.5 1.0 1.5 2.0

Act

ivit

y (

mg

N/L

.d)

0

100

200

300

400

AOB rate

NOB rate

Monod model NOB fit

Monod model AOB fit

Monod Curves for AOB and NOB

Strategy # 2 Operate at DO > 1.5 mg/L

Dissolved Oxygen Pressure

High DO (>1.5 mg/L) allows AOB to grow close their

maximum rate and out select NOB (K-strategist)

SRT Pressure

• To operate the system at an SRT close to AOB washout

• To wash out NOB

SRT/(AOB)SRTcritical ~1

This point can be inferred by

0.7< Avg. NLR/Max AOB activity>1 ??

Strategy # 4 Aggressive SRT

AvN Aeration Control

Residual Ammonia

High DO

Transient Anoxia

Aerobic SRT control

D.O.NO2-NNO3-N

NH4-N

Aerobic Duration

Controller/PLC

DOController/

PLC

DO = set point

NH4-N - NOx-N = setpoint

MAirS

AvN aeration control

•Effluent ammonia set-point = Effluent NOx

•Aerobic Fraction control

•Constant DO

•Intermittent aeration

•Ammonia, nitrite, and nitrate sensors needed

AvN Aeration Control in Action

Aer

obic

Fra

ctio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Nit

roge

n (m

g/L

)

0

2

4

6

8

10

Aerobic Fraction

NH4-N

NO2-N

NO3-N

NOx-N

24-hour

Dis

solv

ed O

xyge

n (m

g/L

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

DO

1-hour

Dis

sove

d O

xyge

n (m

g/L

)

0.0

0.5

1.0

1.5

2.0

A

B

AvN Aeration Advantage

Model-based evaluation of mechanisms and benefits of mainstream shortcut nitrogen removal processes (2014), Ahmed Al-Omari, Bernhard

Wett, Ingmar Nopens, Haydee De Clippeleir, Mofei Han, Pusker Regmi, Charles Bott, Sudhir Murthy. WWTMOD

75% Avg Load 125% Avg Load 75% Avg Load 125% Avg Load

AvN aeration control is being tested at Strass WWTP, Austria

“Optimum” Nitrogen Removal

Grady et al (2011) Biological Wastewater Treatment, 3rd Edition

Batchelor, B (1983). Simulation of single-sludge nitrogen removal. Journal of Environmental Engineering

AvN SRT control

• An automated wasting strategy such that the system always operates close to an AOB wash out SRT

• Being tested now…

Nitrogen Removal in AvN CSTR

AOB and NOB rates

NOB out-selection was realized in a single CSTR and maintained without using any external inhibitors at 25 ᵒC

NOB rates vs NOB Population

Pilot Layout

RAS

WAS

AER PCL

A-stage

SCL

RAS

WAS

B-stage (AvN+)

AER ANX

AOB versus NOB (AvN) Anammox MBBR HRAS

Carbon concentration Nitrite-shunt Nitrogen Polish

Anammox polishing

0

2

4

6

8

10

12

14

16

270 285 300 315 330

mg

N/L

Days

Inf TIN Eff TIN

In-Situ Activity Testing NH4

+ + 1.32 NO2- + 0.066 HCO3

- + 0.13 H+

0.26 NO3- + 1.02N2 + 0.066 CH2O0.5N0.15 + 2.03 H2O

Un

itle

ss R

ati

o

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

18-Oct 7-Dec 26-Jan 17-Mar 6-May 25-Jun 14-Aug 3-Oct

NO2 uptake rate/ NH3 uptake rate

NO3 production rate/ NH3 uptakerate

Avg: 0.246

SD: ± 0.07

Avg: 1.38

SD: ± 0.18

Anammox Stability

0

2

4

6

8

10

12

14

16

18

20

0 5 10 15 20

NO

2 r

em

ove

d

NO2 influent (mgN/L)

Anammox polishing is very stable in an unaerated MBBR

Influent NH4-N = 25 mg/L

Limitation of anammox

11% NO3-N production

Therefore, theoretically only 89% N removal is

possible

NO3-N in the AvN effluent > 0 mg/L (100% NOB

out-selection is not possible)

Nitrate removal in COD limited condition

Acetate (COD)

AVN effluent

NO3-N NO2-N

Then,

Anammox can combine

NH4-N and NO2-N to N2

COD/NO3-N <1.5

Kartal, B., Kuypers, M.M.M., Lavik, G., Schalk, J., Op den Camp, H.J.M., Jetten, M.S.M., Strous, M. 2007. Anammox bacteria disguised as

denitrifiers: nitrate reduction to dinitrogen gas via nitrite and ammonium. Environmental Microbiology 9, 635-642.

Acetate Addition

Influent Ac(COD)/NO3-N 0.83±0.22

NO2-N removed/ NH4-N removed 0.35±0.07

NOx-N removed/NH4-N removed 1.37±0.16

Overall TIN removal (%) 50±12

Anammox contribution 87±6

Ac(COD)/TIN removed = 0.81±0.15

Day

358th 372th 385th

Un

kn

ow

AM

X a

bu

nd

ance (

cop

ies/

mL

)

1e+9

1e+10

1e+11

Ca. "B

rocadia

fulg

ida"

abu

nd

ance (

cop

ies/

mL

)

1e+7

1e+8

1e+9

Unknown AMX

Ca. "Brocadia fulgida"

A Few Conclusions… • NOB out-selection is possible in mainstream (Sensor

and Novel Control systems are unavoidable)

• Separate stage C and N removal – No need to compromise N removal for footprint-energy

• Anammox polishing in the mainstream is feasible and simple

Sidestream and Mainstream

• Anammox are easy as long as they can be selectively retained in the system and fed appropriately

• NOB out-selection is really tough

The people

Olawale Akintayo

Charles Bott

Ryder Bunce

Kartik Chandran

Michael Desta

Norman Dockett

Haydee De Clippeleir

Dana Fredericks

M. Gomez Brandon

Mofei Han

Martin Hell

Becky Holgate

Jose Jimenez

Rebecca Jimenez

Hansa Keswani

David Kinnear

Yi Wei Ma

Matthew Michaelis

Mark Miller

Sudhir Murthy

Geert Nyhuis

Sylvia Okogi

Ahmed Omari

Maureen O’Shaughnessy

Hong Keun Park

Sabine Podmirseg

Pusker Regmi

Rumana Riffat

Andrew Shaw

Beverley Stinson

Imre Takacs

Claire Welling

Bernhard Wett

Questions?

Charles B. Bott

[email protected]

– 757-460-4228

74

RAS

AIR

B-Stage Nitritation-Denitritation through Modulating Aeration

(NiDeMA) Process

Alkalinity

additionTK-202

TK-205

CLARIFIER

TK-203 TK-204

WAS

TK-201

TK-206

ANAMMOX

MBBR

HACH NH4D

HACH LDO

WTW NH4

WTW NH4

SCAN NO3

SCAN NO2

SCAN TS

HACH LDO

pH

HACH LDO

TemperatureHACH LDO

MOV MOV MOV MOV

HACH Nitratax

pH

Temperature

P201

P204

P205

P203

P206

RAS

AIR

B-Stage Nitritation-Denitritation through Modulating Aeration

(NiDeMA) Process

Alkalinity

additionTK-202

TK-205

CLARIFIER

TK-203 TK-204

WAS

TK-201

TK-206

ANAMMOX

IFAS

HACH NH4D

HACH LDO

WTW NH4

WTW NH4

SCAN NO3

SCAN NO2

SCAN TS

HACH LDO

pH

HACH LDO

TemperatureHACH LDO

MOV MOV MOV MOV

HACH Nitratax

pH

Temperature

P201

P204

P205

P203

P206

The future…?

• AOB – Nitritation (NOB out-selection needed)

• Sulfide-driven Autotrophic Denitritation/Denitratation

• Nitrite + Methane – Methanotrophic Denitritation (damo)

• Is Anammox required??

77

RawWastewater

Screening

GritRemoval

UV Disinfection

Anaerobic Treatment

Anaerobic MBR or Biofilm Process

N Removal 3.1CH4

Discharge to Chesapeake Bay

WasteSludge

FeCl3

TertiaryFiltration

WAS

EnergyGeneration

RAS

WAS

NO2 recycle

NitritationAnaerobic Treatment

N-DAMO • Corollary – ammonia

– NH4+ oxidation by ammonia monooxygenase requires O2

– Anaerobic NH4+ oxidation was thought impossible until anammox was

discovered

• NH4+ oxidation through hydroxylamine and hydrazine

• Methane Oxidation – CH4 oxidation by methane monooxygenase requires O2

– Anaerobic methane oxidation was thought impossible – would be problematic…

– Mixed results on methane use for heterotrophic denitrification

• N-DAMO – 3 CH4 + 8 NO2

- + 8 H+ 3 CO2 + 4 N2 + 10 H2O (Raghoebarsing et al, 2006)

(Luesken et al, 2011)