Effective Use of Filter Pilot Testing and Chlorine Dioxide - PowerPoint Presentation

Slide1

Effective Use of Filter Pilot Testing and Chlorine Dioxide Demonstration Testing to Optimize Treatment Plant PerformanceJohn Civardi, PE – Hatch Mott MacDonaldSeptember 19, 2013Slide2

OutlineBackground of Aqua Shenango Water Treatment PlantOperational Issues and Treatability Study

UV Peroxide at ShenangoDAF Pilot Testing

Filter Testing Phases 1, 2, and 3Chlorine Dioxide Testing, bench and demonstrationWhere We Are NowSlide3

Plant Background

Capacity 16 MGD

Raw water from the Shenango River in Sharon, PA. 

Treatment: ballasted flocculation and dual media filtration, chlorine for primary disinfection and chloramines for secondary disinfectionPlant commissioned in early 2000.

Slide4

Plant Schematic

Pilot Area

Shenango

River

Raw Water Pump

Filters

KMnO4

Clearwell

ACTIFLO

Chlorine

Alum

Lime

Soda Ash

PAC

Chlorine

Caustic Soda

Filter Aid Polymer

Chlorine

Corrosion Inhibitor

PAC

Ammonia

To Distribution

Caustic Soda

Soda Ash

FluorideSlide5

Plant Operational IssuesDisinfection By-ProductsRange HAAs (53.7-97.7 ppb)

Range TTHMs (39.0-85.1 ppb)UFRVs< 10,000 gal/

sf, Short Filter RunsAlgaeT&O Geosmin

and MIB at 160 ng/LManganeseSlide6

Water Quality Raw & Treated (Average Conditions)

Turbidity (NTU)

TOC (mg/l)

Algae (Counts/ml)

Manganese (ppm)

Raw9.6

5.6716535

0.11Filter Influent

0.853.46

8800.03

Treated0.08

3.1211

ND

Shown Data collected from Pilot Study from 9/15/11 to 1/23/2012

Water Quality/Performance Issues (Maximum

Conditions)

Turbidity (NTU)

TOC (mg/l)

Algae (Counts/ml)

Manganese (

ppm

)

Raw

102.1

7.1

37440

0.33

Filter Influent

3.39

6.4

2480

0.08

Treated

0.16

4.1

36

0.08Slide7

Treatability StudyOptions Matrix

River

4 MG Pre

Sed

Basin

Replace Filter Media

Option 1:

PAC KMnO4

UV-H

2

O

2

Clearwell

Several variations were also considered

Option 2:

River

ACTIFLO

Ozone

Replace Filter Media

CLO

2

UV-H

2

O

2

Clearwell

Option 3:

River

ACTIFLO

DAF

Replace Filter Media

CLO

2

UV-H

2

O

2

Clearwell

ACTIFLOSlide8

Historical T&O Treatment and ProblemsPowdered Activated Carbon

Residuals Generation

PAC provided limited removal especially with Actiflo

Competitive effects of alumCould PAC be optimized and is AOP a suitable option?Slide9

Bench Testing at Aqua’s Neshaminy

WTP

Removal of up to 90%

Geosmin & MIB is desired at maximum plant capacity

Aqua and Carbon Supplier performed jar tests with Geosmin to assess :

Potential competitive effects of alum on carbon usage – literature contained limited dataOptimum type of PAC

Optimum dose and detention timeSlide10

PAC Testing ResultsDosing PAC together with alum results in significantly lower MIB removal (28% removal Alum/PAC

vs

55% PAC then alum)Testing found that PAC should be added prior to alumMinimum PAC detention time is 45 minutes

Min./Max. PAC dosage is 30 mg/L - 60 mg/LSlide11

Plant Impacts of Testing45 Minutes of Detention Time at 15 MGD requires at 500,000 gallon pre-carbon contact tank with mixers

30 mg/L dosage results in an additional 3,800

ppd of Dry Solids

This would double the plant solids production and require additional residuals treatment equipmentSlide12

Cost Comparison AOP vs PAC

UV – H2O2

PACCapital

$2.5 Million$2.2 Million

O&M$200,000

$310,000Equivalent Uniform Annual Cost

$384,000

$475,000Slide13

A Bit About Carbon FootprintSlide14

Comparison with PACNo additional sludge handling is needed whereas the PAC process will generate approx 1.5 tons per day of dry solids (

100% increase in solids production

)Ability to provide 1 log and higher removal of MIB and Geosmin

Ability to achieve additional microbial disinfection

Smaller footprint than the PAC optionProduces less than 25% CO2 compared to UV/Peroxide

Aqua Selected UV-PeroxideSlide15
Slide16

UV Process Layout

Shenango

WTPSlide17

UV Process Layout

Shenango

WTP

Hydrogen Peroxide

UV Reactor

Flow Meter

Chlorine

Cooling Water Return

Cooling Water Supply

NCSlide18
Slide19
Slide20
Slide21

Treatability Study to Optimize Filtration

Technical Experts

Workshops

Selected DAF, Filters, and Chlorine Dioxide

DAF minimizes residuals, low polymer use, algae performance

ACTIFLOSlide22

Recommended Option SchematicShenango River

Raw Water Pump

Filters

KMnO4

Clearwell

ACTIFLO

Chlorine

Alum

Lime

Soda Ash

PAC

Chlorine

Caustic Soda

Filter Aid Polymer

Chlorine

Corrosion Inhibitor

PAC

Ammonia

To Distribution

Caustic Soda

Soda Ash

Fluoride

Pilot: Dissolved Air Flotation

Pilot: Filter Optimization

To Waste

Integration of DAF into the PlantSlide23

Pilot TestingControl/BaselineComponentsCoordination with RegulatorsVendors

Why Pilot Test?Proof of Design ConceptSlide24

Pilot Testing ContinuedThe TeamOwner: Aqua, PA Plant Staff and Main Office Water Quality, Engineering, and Laboratory in Bryn

Mawr, PAEngineer: HMM Pilot Engineer (Pittsburgh, PA), Data Review & Coordination (Millburn, NJ)

Vendors: DAF Supplier (IDI), Filter ConstructorPADEP – Protocol ApprovalCost – New Filter Columns, DAF Rental, Power, ChemicalsData Management

Communication: Weekly Conference CallsSlide25

Pilot TestingContinued

Backwash Controls

Filter Columns

Online AnalyzersSlide26

ProtocolThree seasons: (1) High Algae (2) Cold Water (3) High TurbidityPhase 1: Filter Optimization (9/27/11 - 10/13/11)

Phase 2: DAF and Filters at Control Steady State (10/20/11 - 11/23/11)Phase 3: Filter Optimization for Ballast Flocculation (12/6/11 – 1/24/2012)

Phase 4: Chlorine Dioxide AdditionControl Column with Same Media as Existing Plant Filters High Turbidity Modifications with Pre-Sedimentation TankIntegrated Chlorine Dioxide & Sulfuric AcidSlide27

Pilot Testing Schematic

pH Adjustment

Raw Water

Coagulant

Flocculation Chambers

Sample Point

Polymer (if needed)

Air Compressor & Saturator

DAF Tank

Clarified Water

Sludge Scraper

Recycle Pump

To Filters

To Waste

Filter Column 60

Filter Column GAC

Filter Column 47

Filter Column 72

To Waste

From ACTIFLO

Same Configuration as PlantSlide28

Testing Matrix & Lab Coordination

Regular

Analytical Schedule

Measurement

Sample Point

Frequency

Field or Lab

1

1

2

3

4

Raw Water

Pilot Filter Column Effluent

Plant Clarified Water

Plant Combined Filter Effluent

Turbidity

X

X

X

X

Online and manually once per day

Field

Particle Counts

X

X

X

X

Online

Water Temperature

X

Once per day

Field

pH

X

X

X

X

1,2 Online and all manually once per day

Field

True Color

X

X

X

X

Once per day

Field

UV

254

X

X

X

X

Once per day

Field

TOC/DOC

X

X

X

X

Twice per week

Lab

Alkalinity

X

X

Once per day

Field

Filterability

X

Once

per day

Field

Aluminum (total and dissolved)

X

X

X

X

Once per day

Lab

Iron

X

X

X

X

Once per day

Lab

Manganese

X

X

X

X

Once per day

LabSlide29

Testing Schedule

Optimizing: pH, Acid & CLO

2

Dosage, Flow Rates.

Filter

Optimization with

Chlorinated ACTIFLO Water

Chlorine Dioxide

Optimize Filter

DAF Running at Steady StateSlide30

Phase 1: Optimize Filter using Treated ACTIFLO Water

Filter Media Configuration

Pilot

Filter

Column

Sand

Media

Depth (in)

Effective Size (mm)

Uniformity CoefficientType

Depth (in)Effective Size (mm)

Uniformity CoefficientTotal Media Depth (in)

47

12

0.45-0.55

1.4

Anthracite

35

0.85-0.95

1.4

47

60

12

0.45-0.55

1.4

Anthracite

48

1.15-1.25

1.4

60

72

120.65-0.75

1.5Anthracite60

1.45-1.551.4

72GAC12

0.45-0.551.4GAC60

1.0-1.21.572

Column 47 Represents the Configuration of the Plant’s Filter

Initial Filter Testing ResultsSlide31

Phase 1 Conclusion: Filter GAC had Lower Run Times than Existing Plant Filter Configuration. Filter 72 had the Longest Run Times Compared to All Columns

Replace Filter GAC with Filter 72 Configuration Under ACTIFLO Treated Water for Benchmark Comparison to DAF

Continue to Phase 2

2013 NYC Watershed/

Tifft

Science and Technical Symposium

Initial Filter Testing ResultsSlide32

DAF & Filter Testing

Filter Media Configuration

Pilot

Filter

ColumnSand

Anthracite

Depth (in)

Effective Size (mm)

Uniformity CoefficientDepth (in)

Effective Size (mm)Total Media Depth (in)

4712

0.45-0.551.4

350.85-0.95

47

60

12

0.45-0.55

1.4

48

1.15-1.25

60

72

12

0.65-0.75

1.5

60

1.45-1.55

72

72

12

0.65-0.75

1.5

60

1.45-1.55

72

Phase 2: Dissolved Air Flotation Under Steady State ConditionsColumn 72 is the Optimal Configuration from Phase 1 using ACTIFLO Treated WaterColumn 47 is the Existing Plant Configuration using DAF Treated Water

Column 47 Represents the Configuration of the Plant’s FilterSlide33

DAF TestingSlide34

DAF & Filters During High Turbidity Events

PA

Raw

Water (Primary Axis) & Filter Influent (Secondary Axis)

Turbidity SpikeSlide35

DAF & Filters During High Turbidity Events

PA

Phase 2: Effluent Turbidity with DAF Treated Water

Turbidity Spike

Filter 1: 35” Anthracite

Filter 2: 48” Anthracite

Filter 3: 60” Anthracite

Filter 4: 60” Anthracite (ACTIFLO)

ACTIFLOSlide36

DAF & Filters During High Turbidity Events

Phase 2: Effluent Turbidity with DAF Treated Water

Turbidity Spike

ACTIFLO

Date

47

60

72

72Slide37

Initial Conclusions

Filter

ACTIFLO

Treated Water

DAF Treated Water

Comparison DAF/ACTIFLO

Phase (No.)

Average UFRV (gals/sf)

Phase (No.)

Average UFRV (gals/sf)

UFRV (Ratio)

47

1

4,542

2

11,512

2.53

60

1

7,211

2

15,516

2.15

72

1

8,670

2

15,507

1.79

DAF Improved Filter Runs (UFRV)

DAF Could Not Handle High Turbidity Events and the Addition of Pre-sedimentation was not Cost Effective

Next Step – Optimize Filters for ACTIFLOSlide38

Additional Filter TestingPhase 3: Filter Optimization with Chlorinated ACTIFLO Treated Water Under Cold Temperature Conditions Column 47: Existing Plant’s Configuration

Column 60 & 72: Optimal Configurations from Phases 1 & 2

Filter Media Configuration

Pilot

Filter

Column No.

Sand

Anthracite

Depth (in)

Effective Size (mm)

Uniformity Coefficient

Depth (in)Effective Size (mm)

Uniformity CoefficientTotal Media Depth (in)

47

12

0.45-0.55

1.4

35

0.85-0.95

1.4

47

60

12

0.45-0.55

1.4

48

1.15-1.25

1.4

60

72

12

0.65-0.75

1.5601.45-1.55

1.472

Column 47 Represents the Configuration of the Plant’s FilterSlide39

Filter Data #1

Phase 3:

Turbidity Under Protocol Threshold

Date

47

60

72Slide40

Phase 3: Head-Loss with ACTIFLO Treated WaterBackwash Occurs when Head-Loss Reaches 120 Inches

Filter Data #2

Filter’s 60 & 72 Have Longer Run Times Compared to Filter 47 (Existing Plant)

Date

47

60

72Slide41

Evaluation of Filter Data & Selection of Media

Table 3.1: Filter Performance Evaluation Criteria Summary*

Pilot

Filter

Media Configuration

Filtered Water Turbidity

Filter Run Volume

Filtered Water Color (true)

Filtered Water Manganese:

< 0.1 NTU

> 7,500 gal. at < 10 ft headloss

< 5 Pt-Co color units

< 0.01 mg/L

47

35” Anth./12” Sand (control)





60

48” Anth./12” Sand



+





72

60” Anth./12” Sand







GAC

60” GAC/12” Sand



* A “



” indicates that the evaluation criteria goal was achieved. A “

+

” indicates that the filter media performed better than the existing plant’s filter media. Slide42

Evaluation of Filter Data & Selection of MediaSummary of PADEP Requirements: Effective Size, Length/Depth RatiosMin. 12 in of Media in Effective Size Range No Greater than 0.45 to 0.55 mm

Ratio Depth (in) to Media Effective Size (mm) Greater than 40Anthracite Effective Size 0.8 mm to 1.2 mm, Uniformity Coefficient No Greater than 1.7

Sand at Least 85% Siliceous Material with an Effective Size of 0.45 to 0.55 mm and a Uniformity Coefficient No Greater than 1.65CostAvailability of MediaSlide43

Evaluation of Filter Data & Selection of MediaSelection of Filter Media: Column 60Improved UFRV by 40%, Meaning Longer Filter Runs

Complies with PADEP StandardsDeeper Anthracite Layer with Larger Effective Size

Additional 2 Inches of Sand for Pathogen BarrierAchieves Water Quality Similar to Existing Filters

Optimal Filter Media Configuration Selected

Pilot

Filter

Column No.

Sand

Media

Depth (in)

Effective Size (mm)

Uniformity Coefficient

Type

Depth (in)

Effective Size (mm)

Uniformity Coefficient

Total Media Depth (in)

60

12

0.45-0.55

1.4

Anthracite

48

1.15-1.25

1.4

60Slide44

Chlorine DioxideSlide45

Benefits of Chlorine Dioxide at the Shenango Plant

Oxidation of manganese before flocculation allows the manganese to be removed during the clarification process (ACTIFLO) and allows the chlorine dosage applied to the filter influent to be reduced or eliminated.

Reduction of the chlorine dosage at the filter influent results in reductions in DBP formation in the

Shenango distribution system.

Chloramination can increase biofilm and cause nitrification in the distribution system. Chlorine dioxide normally breaks down to form chlorite ion. Studies have shown that the presence of chlorite ion in finished water results in better control of organisms in the distribution system, especially those which cause nitrificationSlide46

Approach at ShenangoBench TestingPilot Testing with DAFDemonstration TestingProcurementSlide47

Bench testing with ClO2Slide48

Bench Testing Chlorine DioxideSlide49

Bench Testing OzoneSlide50

How is CLO2 Delivered?ClO2 is best supplied via On-Site Generation SystemsClO2 gas is too energetic to package and ship - heat, light, pressure, shock sensitiveRoad transport of ClO2 gas or solutions is not allowed

Bulk Shipment of CDG 3000 (0.3%) contains 6

lbs chlorine dioxideWhich are then either oxidized or reduced to obtain chlorine dioxide:ClO3- chlorate ion

ClO2 chlorine dioxide ClO2- chlorite ionSlide51

Demonstration TestingThe purpose of this full scale test was to evaluate the effectiveness of chlorine dioxide as a pre-oxidant to improve the following:

Manganese removal in the ballasted flocculation system.

Reduction in filter top chlorine dose while still achieving similar pathogen inactivation. Pre-oxidation with chlorine dioxide was expected to reduce the chlorine demand and allow a reduction in the applied disinfection chlorine dose while still maintaining the same effluent chlorine residual.

Reduction in formation of DBPs in the combined filter effluent.Slide52

Demonstration TestThe CLO2 system delivered 15 to

150 pounds per day (ppd

) of chlorine dioxide. At an average plant flowrate of 9.5 MGD and a dose of 1.0 mg/L, approximately 80

ppd of chlorine dioxide was used. At a maximum plant flowrate

of 16.0 MGD and a dose of 1.0 mg/L, approximately 135 ppd of chlorine dioxide was used.

The chlorine dioxide system was furnished by Siemens and was a Millennium III™ C-150 Auto two chemical flow-pacing chlorine gas/sodium chlorite chlorine dioxide generator capable of producing up to a maximum of 150 lbs

/ day of chlorine dioxide with a 10:1 turn-down and automatic flow-pacing capability. Chlorine gas was supplied using the plant’s existing pre-chlorinator. Sodium chlorite at a strength of 25% was used and was delivered in 250 gallon totes. At the maximum chlorine dioxide usage of 150 ppd

, 84 gallons of sodium chlorite was used and 79 pounds of chlorine was used. As this was a demonstration test treating the entire plant flow, design standards from the Pennsylvania Department of Environmental Protection’s Public Water Supply Manual 383-2125-108 Section IV.B.2 were followed. Slide53

FindingsSlide54

Chlorite LevelsSlide55

SummaryUV-Hydrogen Peroxide is feasible and cost effective for Geosmin and MIB reductionDAF provided significant improvement to UFRV but raw water turbidities above 20 NTU were problematic

Chlorine Dioxide can be purchased and used for 100 gpm

pilot studies for feasibility studiesChlorine Dioxide can be cost effectively tested at plant flowrates on the order of 20 MGD Slide56

AcknowledgementsAqua, PA American WaterPete Kusky, Plant SupervisorBill Young, Plant Chemist

Larry Wehr, Process ControlsMarc Lucca, VP of Production

Craig Lutz, Water Quality SpecialistDoug Crawshaw, Production EngineerBill McGinty, Manager of Treatment/QC

Jack Walter, Division ManagerInfilco (IDI), SiemensHatch Mott MacDonald (HMM)Josef Argenio, EIT

Mark Tompeck, PE

Thank YouSlide57

Phase 1: Head-Loss with ACTIFLO Treated Water

Backwash Occurs when Head-Loss Reaches 120 Inches

Filter’s 60 & 72 Have the Longest Run Times

Existing Plant

Filter

Date

47

60

72

GAC

Initial Filter Testing ResultsSlide58

At room temperature, chlorine dioxide (ClO2) is a light sensitive gas denser than air, yellow/greenish in color, highly soluble in water, with a chlorine like odor.

Cl

O

O

ClO

2

Molecular Structure

Technical OverviewSlide59

Technical OverviewChlorine Dioxide belongs to the family of chemicals known as Oxidizers. CLO2 is also a powerful disinfectant. Ozone

Hydrogen Peroxide

ChlorineChlorine DioxideMeasured by ORP potentialSlide60

Why Chlorine Dioxide?Cost effective disinfectant for regulatory compliancedeclining surface water quality favors CLO2effective @ pH 2-10

enhanced CxT’s for crypto and giardia

reduces THM’s, THAA’s and AOXEPA approved for primary and residual disinfectionSelective oxidant reduces

Mn / Fe, assists on particle count reduction and enhances filter runsImproves taste and odor, oxidizes / destroys sulfides, certain humics

/ fluvics / phenolsSuperior DBP control:residuals measurable, well studied

does not oxidize bromide to bromate; no reaction with ammoniadoes not chlorinate organicsSlide61

Disinfection with ClO2Broad spectrum biocide bacteria, viruses, fungi, algae, mollusks and biofilm

Provides CxT credits effective for Giardia and Cryptosporidium in multiple barrier approach

lower “CxT” values versus Cl2 and chloraminesSynergy with UV (pre or post UV for wastewater and reclaim water)

Effective pH 2 to 10does not oxidize bromide to bromate

does not react with ammonia is often more effective and less costly than permanganate in reducing manganese and iron concentrationsPoints of Application

ahead of chlorination or chloramination to reduce THM’sfollowing UV for residual and enhanced disinfection

ahead of ozone to reduce ozone demand and minimize bromatecompatible with chloramines - controls nitrificationSlide62

Auto Flow-Pacing “Chlorine Free” Chlorine Dioxide Generators (1 to 110 kg/hr)

12% Sodium Hypochlorite

15% Hydrochloric Acid 25% Sodium Chlorite

Chlorine in-situ on demandSlide63

Auto Batch Generator – 3 Chemical Slide64

THM Formation

Naock

&

Doerr

1978

“No THMs

with ClO

2

”Slide65

Biofilm ControlCLO2 penetrates and kills biofilmDoes not react with ammonia Retains disinfection capability with no need to overcome ammonia demand

Shown to reduce planktonic population by 98.72% and biofilm population by 99.88%Additional benefits due the chlorite ion disinfection by-product:

retards nitrification issues extends life of chloramines

Walker, J.T. and M. Morales. 1997. Evaluation of Chlorine dioxide For The Control of Biofilms. Jour. Water. Sci. Tech.. 35:11 (319-323)

McGuire, M.J., N. I Lieu and M. S.

Pearthree

. 1999. Using chlorite ion to control nitrification. Jour. AWWA. 91:10 (52-61)

Gates, D. 1998. Chlorine Dioxide Handbook, Water Disinfection Series.

Denver,CO: AWWA. Pp. 63 - 64.Slide66

Point of AdditionTo eliminate prechlorination (thereby reducing DBP formation) Head of treatment train

Low demand waters Entry to treatment plant (in combination with coagulation)

High demand waters Intake (sufficient time to react)Turbid waters Will not remove TOC, but improves coagulation & particle counts

After sedimentation (reduced demand)Before ozonation to reduce demand on ozone and to help reduce bromate formation by ozone