Arsenic Removal Using Aged Rapid Sand Filter Media

ArsenicArsenic RemovalRemoval UsingUsing AgedAged RapidRapid SandSand FilterFilter MediaMedia

byby C.C. Menard,Menard, D.D. Burt,Burt, M.R.M.R. CollinsCollins WaterWater TreatmentTreatment TechnologyTechnology AssistanceAssistance CenterCenter DepartmentDepartment ofof CivilCivil EngineeringEngineering UniversityUniversity ofof NewNew HampshireHampshire

Summer 2007 OUTLINEOUTLINE

BACKGROUND RESEARCH OBJECTIVE

Assess coating characteristics of ‘aged’ rapid sand filter media.

Quantity Arsenic removal potentials using “aged” rapid sand filter.

Evaluate interferences associated with the adsorption capacity of the metal oxide coating.

CONCLUSION BACKGROUNDBACKGROUND

Reduction/Elimination of Arsenic

Ion exchange

Coagulation / Filtration

Membrane filtration (Reverse Osmosis)

Innovative adsorbents, e.g. metal oxides BACKGROUNDBACKGROUND

Conventional Water Treatment (Pathogen Removal)

– Coagulants Aluminum Sulfate Ferrous Sulfate, Ferric Sulfate, and Ferric Chloride

– Slow sand filtration process

BACKGROUNDBACKGROUND

Natural Aging of Metal Hydroxides to More Stable Metal Oxides

Al(OH)3 →→→→→→Al2O3

Fe(OH)3 →→→→→→Fe2O3

REMOVAL ENHANCEMENT POSSIBILITY!!! Make use of metal oxide coatings that form ‘naturally’ over many years on filter media in WTP from carryover of metal hydroxide flocs produced from “sweep-floc” coagulation RESEARCHRESEARCH OBJECTIVESOBJECTIVES

Explore the Arsenic removal potential of ‘naturally’ coated, regenerable sand filter media. 1) Assess coating characteristics of ‘aged’ rapid sand filter media. sand filter media. 2) Quantify Arsenic removal potentials using ‘aged’ sand filter media. ‘aged’ sand filter media. 3) Evaluate interferences associated with the adsorption capacity of the metal oxide coating. adsorption capacity of the metal oxide coating. METHODOLOGYMETHODOLOGY

Materials used for Objective 1 Filter Media – Portsmouth, NH WTP sand – Philadelphia, PA WTP sand

Characterization Methods – SEM analysis – Electron microprobe analysis Portsmouth, NH Philadelphia, PA

AverageAverage MetalMetal CoatingCoating ContentContent ofof SelectedSelected RapidRapid SandSand FiltersFilters –– SEMSEM picturepicture

Portsmouth, 1996 AverageAverage MetalMetal CoatingCoating ContentContent ofof SelectedSelected RapidRapid SandSand FiltersFilters ––SEMSEM picturepicture

Philadelphia, 2006 AverageAverage MetalMetal CoatingCoating ContentContent ofof SelectedSelected RapidRapid SandSand FiltersFilters ––SEMSEM picturepicture

Philadelphia, 1996 AverageAverage MetalMetal CoatingCoating ContentContent ofof SelectedSelected RapidRapid SandSand FiltersFilters

Electron Microprobe picture

A B C D Silica grain E F G

Philadelphia, 1996 AverageAverage MetalMetal CoatingCoating ContentContent ofof SelectedSelected RapidRapid SandSand FiltersFilters

Weight Oxide Percent

120

100

80 Al2O3 MnO 60 FeO SiO2 40 Percent of Oxid of Percent

0 ABCDEFG Layer Level

Philadelphia, 1996 RESEARCHRESEARCH OBJECTIVESOBJECTIVES

Experimental Set-Up Backwash (BW) Procedure – 50/100 mL sand in 500/1000 mL buffered water – Backwash at target pH’s for 1 hour

Equilibration Procedure – Buffered water at the pH of the Challenge solution

Challenge Procedure – Arsenic Solution according to the array

Backwash/RegenerationBackwash/Regeneration SetSet--UpUp

Motor with paddle

pH controller

Pumps

Buffered Water

Media (Sand)

Base Acid ChallengeChallenge SetSet--UpUp

pH controller Pump

Filter Filter Filter Filter 1 2 3 4

Reservoir

Base Acid

Effluent to Autosampler RESEARCHRESEARCH OBJECTIVESOBJECTIVES

LL 1616 ORTHOGONALORTHOGONAL ARRAYARRAY 5 parameters – EBCT – pH of the challenge solution – Presence of sulfate – Presence of natural organic matter – pH of the backwashing 2 levels RESULTSRESULTS

ARSENIC REMOVED / BED VOLUME at 95 BV (g/m3)

[As]removed/ BV RUN/BV EBCT pH SO4 NOM BW 95 RUN1 2.5 6 0 0 8 3363 RUN2 2.5 6 0 6 8 3176 RUN3 2.5 6 200 6 8 3593 RUN4 2.5 6 200 6 11 3502 RUN5 2.5 8200611 2302 RUN6 2.5 6 200 08 3361 RUN7 2.5 6 0 0 11 3521 RUN8 2.5 8 008 2351 RUN9 5 6 200 6 11 3413 RUN10 5820068 3011 RUN11 580 611 1755 RUN12 5 6008 5873 RUN13 58008 4216 RUN14 580011 3980 RUN15 582000 11 4531 RUN16 5 8 200 6 11 3327 RESULTSRESULTS ARSENIC REMAINING VERSUS BED VOLUME

arsenic restant vs bed volume

1,7 RUN1 1,6 RUN2 1,5 RUN3 1,4 RUN4 1,3 1,2 RUN5 1,1 RUN6 L 1 RUN7 0,9 0,8 RUN8 0,7 RUN9 [As] (mg/[As] 0,6 RUN10 0,5 RUN11 0,4 0,3 RUN12 0,2 RUN13 0,1 RUN14 0 RUN15 0 20 40 60 80 100 120 RUN16 bed volume feed average INTERPRETATIONINTERPRETATION

PREDICTIONPREDICTION PROFILERPROFILER

6004.27 3809.291 ±728.51 BV (mg/L) 1729.2 arsenic removed/ 2 1 2 1 2 1 2 1 2 1

EBCT pH SO4 NOM BW

EBCT(0.0113) > pH (0.0186) > NOM (0.0227) >> SO4 (0.3028) > BW (0.6037) INTERPRETATIONINTERPRETATION MODELMODEL EQUATIONEQUATION

6000

5000

4000

3000

2000

1000 arsenic removed/ BV (mg/L) Actual 1000 2000 3000 4000 5000 6000 arsenic removed/ BV (mg/L) Predicted P=0.0284 RSq=0.67 RMSE=670.91

Y = - 608,5*EBCT + 549,8*pH + (-213,7*SO4) + 526,7*NOM + 99,7*BW + 3454,6 SUMMARYSUMMARY

Sand coating composed mainly with aluminum, manganese oxide with a beginning of small ratio of iron oxides

EBCT, pH and NOM: most important effect Slight positive impact with the presence of sulfate No significant effect of BW pH: in the range of recommended pH for aluminum coated sand RecommendationsRecommendations

Additional experiments to analyze more precisely effect of sulfate but also others anions like phosphate

Additional experiments to determine durability of coatings, test raw water

Pilot or Full scale demonstration

Analyse Portsmouth sand with Electron microprobe