Extractive Nutrient Recovery as a

Green Option for Managing

Phosphorus in Sidestreams and

Biosolids

Wendell Khunjar, Ph.D.

Ron Latimer, P.E.

Katya Bilyk, P.E.

Joe Rohrbacher, P.E.

Paul Pitt, Ph.D., P.E.

MR396

Agenda

Sidestream treatment overview

Phosphorus Removal/Recovery Options

Case Study 1 - Nansemond Treatment Plant

Case Study 2 – FWHWRC

Case Studies 3 and 4 – Miami Dade Treatment Plants

Summary

MR396

Biological treatment is a cost effective, robust

option for carbon and nutrient removal

■ Biological nutrient removal uses microorganisms

■ Solids generated must be processed before disposal

■ Anaerobic digestion is a common solids treatment

option

To solids

handling To solids

handling

MR396

Water resource reclamation facilities accumulate

nutrients within the solids process

Primary

Sludge

10-15%

EBPR or

Chem - P

Removal

35-50%

Effluent

10%

Feces

33%

Urine

67%

Secondary

Sludge

25-40%

Sludge

Up to 90%

Adapted from

Cornel et al., 2009

Adapted from

Phillips et al., 2011

Effluent

13%

Feces

20%

Urine

80%

Sludge

20%

Gaseous emission

67%

Up to 90% of the

influent P can be

present in the

solids stream

Up to 20% of the

influent N can be

present in the

solids stream

MR396

Solids stabilization generates nutrient rich liquid

stream

■ Sidestreams are typically

returned to the head of

the plant for treatment

■ Examples of sidestream BFP filtrate

GBT filtrate

Filter backwash

Centrate

Digester supernatant

Screenings Grit10 Sludge

Biosolids

HandlingResidual

10 Treatment 20 TreatmentDisinfection

Effluent20 Sludge

Fe or Alum Fe or Alum

MR396

High nutrient recycle loads can upset the

mainstream process

Description Percent of

Total Influent

Nitrogen

Load

Percent of

Total Influent

Phosphorus

Load

Nansemond,

Suffolk, VA

Centrate

13% 29%

Bowery Bay, NYC

Centrate 17% *

Henrico County, VA

Centrate 15% *

High Point Eastside,

NC Fermenter * 50%

Wards Island, NYC

Centrate 30-40% *

North Durham, NC

Centrate 19% 30%

South Durham, NC

Centrate 21% 25%

7/11/06 Recycle Stream Sampling

Phosphorus

0

100

200

300

400

500

600

700

800

900

Centrate Ash Lagoon Decant Primary Effluent

Sampling Location

TP

Lo

ad

(lb

/day)

Total P

Up to 50% of the total

influent P load can be

present in the

sidestream

MR396

Struvite can be a significant maintenance concern

with anaerobic digestion

Struvite = Mg + NH4 + PO4

NH4 & PO4 released in digestion

Typically Mg limited

Mg addition (i.e. Mg(OH)2) can

promote struvite formation

Miami Dade SDWRF

NYC Newtown Creek

WPCP

MR396

Sidestream treatment is the manipulation of the

return liquid stream for a treatment purpose

Typically focused on nutrient removal or recovery

Usually economical when sidestreams contribute:

■ ≥15% of the influent TN

■ ≥20% P load

Can often reuse existing infrastructure to reduce costs

Thickening

Biological mainstream process

Anaerobic

Digestion Dewatering

Sidestream

treatment

Biosolids

MR396

What options are available for sidestream nutrient

treatment?

Nutrient rich

sidestream

Treatment

process

Nutrient free

effluent

Removed/recovered

nutrient

Nitrification/Denitrification

Nitritation/Denitritation

Deammonification

Gas stripping and/or ion

exchange

N Removal and

recovery

Coagulant aided precipitation

Struvite crystallization

P removal and

recovery

MR396

P removal from sidestreams relies on chemical

precipitation

Coagulant aided precipitation o Uses Alum or Ferric

o Non-proprietary

o Traditionally used for controlling

sidestream P at this plant

o High O&M requirement

Struvite recovery o Forms struvite which can be

used as a slow release fertilizer

o Proprietary

• Ostara

• MFH

• Procorp/DHV

• Paques

• Veolia

MR396

Struvite recovery exploits pH dependent chemical

precipitation phenomena

Struvite precipitation

N:P ratio in struvite = 0.45 lbs N required per lb P removed

N:P ratio in filtrate ~ 2.4-2.6, ammonia in excess

Mg+2

NH4+-N

PO4-3 - P

Struvite

Recovery

Reactor

Mg(NH4)PO4(s)

External

NaOH

External

Mg+2

Mg(NH4)PO4(s) = struvite

MR396

Intentional struvite recovery exploits pH dependent

chemical precipitation phenomena

Fluidized bed reactor or CSTR used for struvite

recovery

Magnesium

Caustic

Centrate/Filtrate

Dryer

Effluent

Struvite

Dewatering

Sand (Procorp)

FBR

N:P ratio in filtrate ~ 2.4-2.6, ammonia in excess

N:P ratio in struvite = 0.45

lbs N required per lb P

removed

MR396

There are several commercial options for struvite

recovery

Name of

Technology Ostara Pearl®

Multiform

Harvest struvite

technology

Phospaq Crystalactor® NuReSys

Name of product

recovered Crystal Green ® struvite fertilizer struvite fertilizer

Struvite,

Calcium‐phosphate,

Magnesium‐phosphate

BioStru®

% efficiency of

recovery from

sidestream

80-90% P

10-40% NH3-N

80-90% P

10-40% NH3-N

80% P

10-40% NH3-N

85-95% P

10-40% NH3-N

>80% P

5-20% N

Product

marketing/resale Ostara Multiform Harvest N/A

Third party facilitated by

Procorp N/A

# of full-scale

installations in

design/operation

8 2 2 4 7

MR396

Nansemond Treatment Plant is a 30 MGD facility that

employs a 5-stage BNR for N and P removal

TN = 8

mg/L

TP = 1

mg/L

Nansemond –

HRSD, Virginia

• 30 MGD BNR – 5

Stage

• 8 mg TN/L

• 1 mg TP/L

MR396

Nansemond Treatment Plant is a 30 MGD facility that

employs a 5-stage BNR for N and P removal

TN = 8

mg/L

TP = 1

mg/L

Diurnal Sampling

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

12:00 AM 2:00 am 4:00 am 6:00 am 8:00 AM 9:00 AM 12:00 PM 2:00 PM 4:00 pm 6:00 pm 8:00 pm 10:00 pm

Time

TP

(m

g/L

)

Day 1 Influent TP

Day 2 Influent TP

Day 1 PE TP

Day 2 PE TP

Sidestream load

represents up to

30% of the plant

influent P load

MR396

Two options were considered for sidestream P

Treatment at NTP

Ferric addition o Forms ferric phosphate and ferric

hydroxide

o Non-proprietary

o Traditionally used for controlling

sidestream P at this plant

o High O&M requirement

Struvite recovery o Ostara Pearl

o Treatment fee option

o OSTARA provides facility

and HRSD pays fee for use

o Capital purchase option

o NTP purchases equipment

and receives annual

payments from OSTARA

MR396

Net present worth analyses indicated that the capital

purchase option was the most cost-effective solution

Nutrient recovery option was more cost effective than

Ferric addition option

Item Treatment Fee

Option

Capital Purchase

Option

Ferric Chloride Chemical Cost $ (290,000) $ (290,000)

Sludge Savings $ (155,000) $ (155,000)

Methanol Savings $ (29,000) $ (29,000)

Oxygen Savings $ (19,000) $ (19,000)

Ostara Paybacks $ (87,850) $ (135,850)

Total Annual Savings $ (580,850) $ (628,850)

Caustic Cost Allowance $ 25,000 $ 25,000

Ostara Annual Fee $ 444,000 $ -

Total Annual Operating Cost $ 469,000 $ 25,000

Total Capital Cost $ 1,080,000 $ 4,143,000

Present Worth Operating Costs $ (1,505,750) $ (8,129,160)

Net Present Worth5 $ (425,750) $ (3,986,050)

Present worth cost of line 10 over 20 years at 5% cost of financing

MR396

NTP constructed and has operated the nutrient

recovery facility for ~ 2 years

MR396

Full scale struvite recovery facility at NTP

System has produced ~ 1100 lb struvite/day

MR396

The struvite recovery facility has reduced ortho-P

concentrations by approximately 85%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0

100

200

300

400

500

600

700

800

900

1000

May-10 Aug-10 Nov-10 Mar-11 Jun-11 Sep-11 Dec-11 Apr-12 Jul-12 Oct-12 Jan-13 May-13

Perc

ent R

emov

al

Conc

entr

atio

n m

g/L

Inf.& Eff. Ortho-P and % Removal for OstaraInfluent ortho-P to Ostara Average Eff. ortho-P% ortho-P Removal 7 per. Mov. Avg. (Influent ortho-P to Ostara)7 per. Mov. Avg. (Average Eff. ortho-P) 7 per. Mov. Avg. (% ortho-P Removal)

MR396

Ammonia removal has averaged 25%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0

100

200

300

400

500

600

700

800

900

1000

Jan-11 Mar-11 Jun-11 Aug-11 Nov-11 Feb-12 Apr-12 Jul-12 Oct-12 Dec-12 Mar-13 May-13

Pe

rce

nt

Re

mo

val

Co

nce

ntr

atio

n m

g/L

Inf and Eff Ammonia and % Removal for OstaraInfluent Ammonia to Ostara Average Eff. Ammonia from Ostara% Ammonia Removal 7 per. Mov. Avg. (Influent Ammonia to Ostara)7 per. Mov. Avg. (Average Eff. Ammonia from Ostara) 7 per. Mov. Avg. (% Ammonia Removal)

MR396

F. Wayne Hill Water Reclamation Center

• Gwinnett County, GA

• 60 MGD advanced WWTP

• 0.08 mg/L TP effluent limit

• Bio-P and chemical trim for P-removal

MR396

In 2009, F. Wayne Hill Changed from Bioxide to

Mg(OH)2 in Collection System for Odor Control

Pros: Eliminated need for ALK addition at plant

Cons: Struvite formation in centrate lines, centrifuges, digester

complex

Sludge from 22 mgd Yellow River Bio-P plant coming, which

would substantially increase P load in sidestreams and SFP

Struvite taken from centrifuge

MR396

Limiting effluent P and struvite formation are key

drivers for this plant

Phosphorus outlets: o Effluent (Limit TP = 0.08 mg/L)

o Sludge cake (precipitated complex, biomass, struvite)

o Struvite solids from nuisance formation

Study goal: determine best solution for struvite

issue (Mg continues) or P recycle issue (Mg stops) o Nutrient Recovery

o Metal salts

MR396

Five options were considered for sidestream P

removal from F. Wayne Hill AWRF

Ferric addition with and without

Mg(OH)2 addition

Struvite recovery with and without

WASStripTM

Do Nothing

MR396

WASSTRIP™ concept minimizes nuisance struvite

production

Release P

from sludge

using VFA

rich stream

Low P content of

sludge minimizes

nuisance struvite

formation from

digester onwards

Send P rich sidestream to recovery process

MR396

Bench scale testing of the WASSTRIP™ process

was performed

Determine levels and rates of PO4 release from WAS

Optimize parameters to maximize PO4 release in pilot studies o Anaerobic retention time and WAS:PS blend ratio

MR396

Using the pilot data, Biowin™ process modeling was

used to simulate each alternative

1. Use calibrated whole plant model to simulate alternatives at each flow

scenario

2. “Do Nothing” scenario is modeled for comparison of struvite formation

3. The modeling results were used to assess effectiveness of the nutrient

control strategy and also to estimate costs for the BCE.

Grit tank

BRB 5-10- A1BRB 5-10- A2 BRB 5-10- B1 BRB 5-10- B2 BRB 5-10- C5+C6 BRB 5-10- C8BRB 5-10- C1+C2BRB 5-10- C3+C4 BRB 5-10- C7

Anaerobic Digesters

Dewatered Sludge

Grit

Tertiary Clarifier Metal Salt

Combined Filter Effluent - Pre O3 Infl

Tertiary Caustic

Sec Scum

Bioreactor21

Raw INF

YR Sludge

Influent (SV)7

Influent (SV)17

P release

Bioreactor33

OSTARA struvite

Influent (SV)51

MR396

P recovery provides equivalent struvite reduction

compared with the ferric addition option

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

Do Nothing Ostara + WASSTRIP +

Mag

Ostara + WASSTRIP +

No Mag

Ostara Centrate No Mag

Ferric Digesters

Mag

Ferric Centrate No Mag

PP

D o

f St

ruvi

te in

Slu

dge

Cak

e

Scenario

30 mgd

40 mgd

50 mgd

MR396

Struvite recovery + WASSTRIP has lowest net

present cost and 8-Year Payback

FWHWRC is pursuing nutrient recovery option

$0.0

$5.0

$10.0

$15.0

$20.0

$25.0

20

13

20

14

20

15

20

16

20

17

20

18

20

19

20

20

20

21

20

22

20

23

20

24

20

25

20

26

20

27

20

28

20

29

20

30

20

31

20

32

20

33

NP

C, $

Mill

ion

s

Year

Ostara + WASSTRIP, Mag

Ostara + WASSTRIP, No Mag

Ostara Centrate, No Mag

Ferric Digesters, Mag

Ferric Centrate, No Mag

MR396

Miami-Dade resource recovery

CDWWTP – 143 mgd

AADF

SDWWTP – 112.5 mgd

AADF

■ Pure oxygen facilities

■ Meets secondary treatment standards

■ Disposal through deep injection wells

■ Significant

nuisance struvite

formation issue

MR396

Miami Dade resource recovery - evaluations

In-situ Coupon Testing to

Determine Degree of Struvite

Formation and Confirm

Theoretical Ferric Dose

Bench Scale Testing to

Determine Optimal Ferric Dose

Ostara

Pilot Test

Modeled Struvite Potential and

Reduction for Each Alternative

& Performed Cost Evaluation

MR396

Nutrient Recovery was the most cost effective

option at both plants

Long term recommendation is to implement nutrient

recovery at both facilities

Notes: * 6% interest and escalation

** Could be significantly higher

MR396

Summary

Thickening Anaerobic

Digestion Dewatering

Sidestream

treatment

clarificatio

n

Biosolids 1. Manipulate the nutrient content of the

biosolids

2. Provide plant with alternative revenue

stream

1. Reduce energy and chemical consumption

in the mainstream process

2. Provide factor of safety for mainstream

nutrient removal process

3. Minimize nuisance struvite formation and

reduce O&M costs

To

mainstream

MR396

Summary

Compare struvite crystallization with precipitation with

coagulant (i.e., alum or ferric)

Payback site-specific and dependent on the P load

Tool for Evaluating Resource RecoverY (TERRY)

developed through WERF grant to help facilities perform

high level evaluation for implementing P recovery

MR396

Contact Information

Wendell O. Khunjar, PhD

4035 Ridge Top Road, Suite 400, Fairfax VA 22030

703 267-2759 (direct) | wkhunjar@hazenandsawyer.com

Katya Bilyk

4011 Westchase Blvd., Suite 500, Raleigh, NC 27607

Phone: 919-755-8593| kbilyk@hazenandsawyer.com

Ron J. Latimer, PE

5775 Peachtree Dunwoody Rd. Suite D-520, Atlanta, GA

30342

404-459-6363 (main) | rlatimer@hazenandsawyer.com