Optimizing a Pure Oxygen Wastewater Treatment Plant Randal W. Samstag, PE, BCEE

CarolloTemplateWaterWave.pptx

Principal Technologist

Outline • HPO Basics • HPO Upgrade Case Studies – West Point WWTP Pilot Plant – Bellingham Post Point WWTP

CarolloTemplateWaterWave.pptx

• Conclusions

2

The High Purity Oxygen (HPO) Process • Invented by Union Carbide Corp. in early 1970s • “High Purity Oxygen”

CarolloTemplateWaterWave.pptx

– Cryogenic Plant (95 – 99%) – Pressure Swing Adsorption (88 – 93%) – Vacuum Swing Adsorption (88 – 93%)

3

HPO Upgrade Case Studies • West Point WWTP HPO Pilot Plant

• Bellingham Post Point WWTP

CarolloTemplateWaterWave.pptx

– HPO Modifications Project – Nitrogen Removal Upgrades

4

CarolloTemplateWaterWave.pptx

West Point WWTP Pilot Plant

5

West Point HPO Test Facility • 4-stage HPO pilot facility (Lotepro) – 20-30 gpm – SRT: 1-2 days

• Mar - Dec 1988

• 3 modes evaluated: – Plug Flow – Contact/Reaeration

CarolloTemplateWaterWave.pptx

– Anaerobic Selector w/ Plug Flow

Wef2003Samstag.ppt

CarolloTemplateWaterWave.pptx

HPO Test Facility Schematic

Wef2003Samstag.ppt

7

CarolloTemplateWaterWave.pptx

SVI, mL/g

Settleability Data from HPO Test Facility 900 800 700 600 500 400 300 200 100 0

Min Ave Max

8

The Culprits • Microscopic evaluation found filaments – Thiothrix II – Type 021N

CarolloTemplateWaterWave.pptx

• Sulfur oxidizing aerobes

Wef2003Samstag.ppt

Influence of SRT and Temp on PAO

Washout Aerobic SRT (Days)

Nitrifier Washout

PAO Washout

14 12 10 EBPR and Nitrification

8 6

EBPR

4 2

No EBPR

0 CarolloTemplateWaterWave.pptx

Nocardia Washout

10

15

20

25

30

Temperature (Degrees C)

Mammais and Jenkins, 2001

PAO Have Higher Floc Density • PAO form Neisser-positive cell clumps • Compact flocs settle faster than highly filamentous sludges

CarolloTemplateWaterWave.pptx

• Bio P organisms have a higher density than typical zoogleal or filamentous organisms

Schuler and Jang, 2007

CarolloTemplateWaterWave.pptx

Bellingham Post Point WWTP

12

The Bellingham Post Point WWTP

• 20 mgd capacity HPO activated sludge plant • 10 ton/day pressure swing absorption HPO system

CarolloTemplateWaterWave.pptx

• Three HPO basin stages

Settleability History • Average SVI of 170 mL/g from 1999 to 2004 • Filament caused settleability problems – Type 1701 (Low DO) – Type 1863 (Low SRT)

CarolloTemplateWaterWave.pptx

• Periodically high VFA feed from influent sewer and from solids dewatering operation 3-4 days per week leading to slime bulking

14

Idea for Improvement – Anaerobic Selector • Provide zone for uptake of VFA

• Encourage growth of phosphorus accumulating organisms (PAO) • Increasing population distribution of PAO increases floc density

CarolloTemplateWaterWave.pptx

• PAO have a compact morphology and higher density than other typical activated sludge bacteria (Schuler and Jenkins)

• Experience at three other HPO plants (SE Essex SD, Hyperion, SE San Francisco)

15

The HPO Modifications Project • Improve settleability by converting first stage to anaerobic selector • Replace aerators to improve oxygen transfer • Add variable speed drives to mixers to improve control and save power

CarolloTemplateWaterWave.pptx

• Selectors added capacity for interim operation until major capacity upgrade could be completed

16

SVI History Before and After Conversion to Anaerobic Selector Comparison of SVI and MCRT 350

3.5

Aerobic

CarolloTemplateWaterWave.pptx

3.0

250

2.5

200

2.0

150

1.5

100

1.0

50

0 07/17/05

Anaerobic SVI 7/Avg ml/g

12/14/05

0.5

MCRT 7/Avg Days

05/13/06

10/10/06

03/09/07

08/06/07

01/03/08

0.0 06/01/08

MCRT (Days)

SVI (mL/g)

300

The Original Aerators • Conventional conically shaped low speed aerators with bottom impellors

CarolloTemplateWaterWave.pptx

• Estimated standard aerator efficiency (SAE) 3.2 pounds of oxygen per hour per horsepower

New Aerators

• m2t Technologies • Flat-bladed surface aerators without bottom impellers

CarolloTemplateWaterWave.pptx

• Initial estimate of SAE (second stage conditions) – 4.0 pph/hp

Low Speed Testing for First Stage Aerators

CarolloTemplateWaterWave.pptx

• Tested to deliver 0.5 ft per second velocity at the bottom of the test tank at low speed in first stage (anaerobic / aerobic)

High Speed Testing

CarolloTemplateWaterWave.pptx

•

Test-tank tested at approximately 3.9 pph/hp under high speed conditions simulated for second stage

Simulation of Variable Speed Control

• Savings only occur when aerators are under-loaded

16 Dissolved Oxygen Concentrations (mg/L)

• Simulations by Michael Stenstrom of UCLA indicated up to 35% savings in power with VSD control

14

DO1 DO2 DO 3

12 10 8 6 4 2

0

5

10

15

CarolloTemplateWaterWave.pptx

Time (hours)

20

25

Summary Comparison Performance

Description

Aerobic (2006)

Anaerobic (2007)

Difference (%)

7.2

6.7

-6.8%

1,291

1,212

-6.5%

1.21

1.15

-5.7%

155.71

126.77

-22.8%

Estimated Mixer Power (hp)

222.2

125

-77.9%

Estimated Oxygen Power (hp)

215.2

201.6

-6.8%

Estimated Total Power (hp)

437.5

326.6

-34.0%

O2 (tpd) O2 (lb/mgal) O2 (lb/lb BOD)

CarolloTemplateWaterWave.pptx

SVI (mL/g)

Project Budget • Total cost - $670,000

• Puget Sound Energy grant - $330,000

CarolloTemplateWaterWave.pptx

• City of Bellingham cost - $340,000

24

CarolloTemplateWaterWave.pptx

Comparison of Operating Modes Actual Bellingham Operation Anaerobic Selector Operation - 2007

HPO Operation - 2006 • 100% Aerobic

• 67% Aerobic

• 2.2 Days Aerobic SRT

• 1.6 Days Aerobic SRT

• 157 mL/g SVI

• 126 mL/g SVI

• 2,400 mg/L MLSS

• 2,900 mg/L MLSS

25

Comparison of Operating Modes Actual Bellingham Capacity

CarolloTemplateWaterWave.pptx

HPO Operation - 2006

20 MGD Capacity

Anaerobic Selector Operation - 2007

22 MGD Capacity

10 Percent Increase

26

What About BNR Conversions? • Concerns about HPO

CarolloTemplateWaterWave.pptx

– CO2 accumulation -> low pH -> suppresses nitrification – As SRT increases HPO becomes less and less economical compared to fine bubble – Examples: • • • • •

LOTT (Olympia, Washington) Denver Metro Indianapolis Bellingham All abandoned HPO for BNR

27

Oxygen Supply Cost as function of SRT HPO: New VPSA HPO Plant

AAS: New Diffused Aeration System

$800 $700

Unit Cost of Oxygen ($/ton)

$600 $500 $400 $300 $200

CarolloTemplateWaterWave.pptx

$100 $0

1.5

2.5 SRT (Days)

8.0

28

CarolloTemplateWaterWave.pptx

HPO Conclusions

29

Conclusions from the HPO Test Facility

CarolloTemplateWaterWave.pptx

• Anaerobic selectors at the HPO test facility failed because low SRT, low temperature prevented formation of an environment supportive of EBPR. PAO were washed out.

Wef2003Samstag.ppt

30

Conclusions from the Post Point Project • Anaerobic selector provided a way to control settleability by encouraging PAO and taking up excess VFA • Improved oxygen transfer efficiency of new aerators saved power • Variable speed control of DO provided power savings and improved process control • The upgrade provided a 10 Percent capacity increase for interim operation CarolloTemplateWaterWave.pptx

• For BNR upgrade, fine bubble was more economical

31

Thank you!

Questions?

CarolloTemplateWaterWave.pptx

Randal Samstag [email protected] 206-684-6532 (office) 206-851-0094 (mobile)

Generalized Conclusions about HPO

CarolloTemplateWaterWave.pptx

The Initial Claims

Experience

• Less O2 consumption

• O2 consumption depends more on SRT than on DO

• Better settling sludge

• Little evidence of better settleability at a given SRT

• Can operate at a higher MLSS

• If settling is not better, no justification for higher MLSS

• Smaller tank volume required

• If settling is not better, no justification for smaller tanks

• More power efficient than conventional aeration

• Power efficiency depends on SRT

• Can be used for nitrification

• High CO2 decreases pH and inhibits nitrification

33

CarolloTemplateWaterWave.pptx

MLSS Showed Indications Phosphorus Accumulating Organisms

34

Recent SVI and SRT Data from West Point WWTP West Point SVI and SRT Data

CarolloTemplateWaterWave.pptx

SRT (day)

Linear (SVI)

30 per. Mov. Avg. (SVI)

30 per. Mov. Avg. (SRT (day))

Linear (SRT (day))

450

9.0

400

8.0

350

7.0

300

6.0

250

5.0

200

4.0

150

3.0

100

2.0

50

1.0

0 12/4/05

12/4/06

12/4/07

12/3/08

12/3/09

SRT (Days)

SVI (mL/g)

SVI

0.0 12/3/10 35

Summary Comparison Operation Description

Aerobic (2006)

Anaerobic (2007)

Difference (%)

Flow (mgd)

12.2

12.1

-0.8%

BOD (ppd)

12,852

12,994

1.1%

1,763

2,057

14.3%

Total SRT(days)

2.2

2.4

6.5%

Aerobic SRT (days)

2.2

1.6

-40.3%

13.3

7.4

-78.8%

Temp (Deg C)

17.57

17.24

-1.9%

O2 Purity (%)

88%

89%

1.4%

MLSS (mg/L)

CarolloTemplateWaterWave.pptx

DO (mg/L)

CarolloTemplateWaterWave.pptx

Comparison of Operating Modes Typical Operating Assumptions Classical HPO Operation • 100% Aerobic

Anaerobic Selector Operation • 85% Aerobic

• 1.5 Days Aerobic SRT

• 2.5 Days Aerobic SRT

• 200 mL/g SVI

• 120 mL/g SVI

• 1,700 mg/L MLSS

• 2,200 mg/L MLSS

37

Comparison of Operating Modes Capacity Impact

CarolloTemplateWaterWave.pptx

Classical HPO Operation

13 MGD Capacity

Anaerobic Selector Operation

17 MGD Capacity

30 Percent Increase

38