Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
The Past, Present and Future for Selenium Treatment
14th Annual BC MEND WorkshopNovember 29, 2007Kevin W. Conroy, P.E.
Where is it found?
Paint, pigments, dye formulatingElectronicsGlass manufacturingInsecticide productionPulp and paperAsh piles, FGD blowdown,Coal/oil combustion Agricultural waterPetroleum processing Mining operations
Why is it a problem?
Aquatic life hazard1983 – Kesterson National Wildlife Refuge - CaliforniaBirth defects/death of birds, small animals, fish
Selenium cycle not well understood
Uncertainty on bioavailabilityEven if bioavailable – what is toxic?
Often times low concentration, high volume - makes treatment expensive
How is it regulated?
5 µg/L Freshwater aquatic life 50 µg/L Primary DWS MCLU.S. Fish and Wildlife Service has recommended 2 µg/L to protect fish, waterfowl and endangered aquatic species
Chemistry
3 primary oxidation states-2 selenide+4 selenite (HSeO3
- and SeO3-)
+6 selenate (SeO4-2)
Chemical equilibrium principals don’t really applyDriven by
Redox conditionsBiological activitySorption processes
Past and Present - EPA BAT
Ferric coagulation/filtrationTypically pH <7Coprecipitation effectEffective removal requires reduction of selenate to seleniteProblem if arsenic present
Lime softeningReverse osmosis
Non-preferential processPretreatment due to other typical mine water issues may be required
Past and Present - EPA BAT
ElectrodialysisAlumina
Selenite adsorbed at pH range of 3 – 8Silica can interfere at pH >4Selenate adsorption is poor
Ion exchangeNeed oxidized divalent selenateCompeting ion effects can hinder effectivenessSome specialty resins tested
Past and Present - EPA BDAT
Ferrihydrite precipitation with concurrent adsorption of selenium on the ferrihydrite surfaceFor adsorption – need ferric ion (Fe+3) presentMost effective removal at pH 4-6Somewhat effective up to pH 8Phosphate, silicate, arsenic, carbonate can interfere
Past and Present – EPA/DOE MWTP
Selenium Treatment/Removal Alternatives DemonstrationReport issued in 2001Three technologies tested in field
Ferrihydrite Adsorption (baseline)Catalyzed CementationBiological Reduction
One technology tested on bench scale
Enzymatic Reduction
Past and Present – EPA/DOE MWTP
Objective – treat to <50 µg/LWork done in 1999-2001Basis – KUCC Garfield Wetlands-Kessler Springs site<50 to >10,000 µg/L Se 95%+ selenateTDS 1,000 - 5,000 mg/L
Past and Present – EPA/DOE MWTP
FerrihydriteDid not work on a consistent basisVarious iron types, concentrations and ratios usedCould achieve objective but at prohibitive reagent consumptionQuestions on TCLP stability
Past and Present – EPA/DOE MWTP
Catalyzed cementationDeveloped for arsenic, selenium, thallium removalRemoves metals by cementation on the surface of iron particlesBelieved to work on both selenite and selenateProprietary catalysts usedBench test work had shown favorable resultsDid not work on a consistent basis
Past and Present – EPA/DOE MWTP
Biological Reduction (BSeR™)Used anaerobic solids bed reactors Selenium reduced to elemental selenium by biofilms and proprietary microorganismsMolasses used as carbon sourceWas able to consistently meet objectiveOver 70% of samples less than detection (2 µg/L)
Past and Present – EPA/DOE MWTP
Economics
BDAT Cementation BSeR™Capital $1.0M $1.1M $0.6MO&M $2.1M $1.2M $0.14MNPV $17M $9.5M $1.1M$/1,000 gal $13.90 $8.17 $1.32
Based on 300 gpm plant, 2 mg/L selenium2001 dollars
Past and Present - Nanofiltration
Some test work by USGS in 1996Agricultural drainage
Selenate removal better than selenite
Not surprising – designed for divalent not monovalent ions
95+ % removal at Se<1,000 µg/LMembrane scaling is an issue in high SO4
-2 water
Problems With the Past and Present
Non-selective processesLarge amounts of secondary wasteMultiple reagentsOkay for bulk selenium removal with other metalsCan’t consistently get to <10 µg/L
Biological Reduction – General
Studied for decadesMicrobes degrade/transform contaminant because
Energy sourceDetoxification mechanismResembles another ionCombination of the above
Anaerobic reactorsReduction to elemental seleniumNitrate/sulfate interference?
90%+ removal reported
Biological Reduction – Ponds/Wetlands
Panoche Drainage District – San Joaquin Valley
74-1,400 µg/L due to Se rich soilPrimarily selenate formNumerous bioremediation studiesAlgal-Bacterial Selenium Removal (ASBR)Anoxic ponds – reduce selenate to selenite to elemental and settleGenerally about 80% maximum removal
Biological Reduction – Ponds/Wetlands
Additional California work - 2005Constructed wetlands9 plant species tested63% - 71% removal~20 µg/L influent, 3 – 6 µg/L effluent
Problem with ponds/wetlandsHRT’s in days
Biological Reduction – Advances
More selective microbes isolatedAdvances in fixed film/biofilm mediaBetter understanding of operating conditionsResult – retention times have been reduced from days to hours for active systemsAdvances also made in passive technology
Biological Reduction – Advances
ABMet®Offered by GE Water and Process TechnologiesSame as BSeR™ processSeveral FGD projects at commercial scale3,000 – 5,000 µg/L seleniumUp to 20,000 mg/L chloride98% – 99% removal projectedEffluent as low as 10 µg/L
Biological Reduction – Advances
Passive Selenium Reducing BioreactorTested on Colorado Western SlopeBureau of Reclamation Science and Technology ProgramInfluent typically ~20 µg/LSpike to 70 µg/L1,000-2,000 mg/L SO4
-2 background
Biological Reduction – Advances
Passive Selenium Reducing BioreactorFour reactors with different substrate compositionsOrganic substrate composed of wood chips, hay, manureZVI incorporated; no advantage12 hour detention time adequate, optimization possibleOperated for 20 weeksEffluent typically <2 µg/LUp to 98% removal
Biological Reduction – Advances
Active Anaerobic Bioreactor SystemWaste rock seepage250 gpm capacityFixed film bioreactor with high surface area mediaMolasses used as carbon sourcePhosphate/urea addedReverse osmosis system used during high flow – 700 gpmBioreactor feed switched to RO brine
Biological Reduction – Advances
Active Anaerobic Bioreactor System18 hour retention timeLow flow (raw seepage)
Se ~30 µg/LSO4
-2 ~6,000 mg/LHigh flow (RO brine)
Se ~70 µg/LSO4
-2 ~13,500 mg/L
Biological Reduction – Advances
Creek
Feed Pond
Low Flow Operations
High Flow Operations
Pre-Filtration
HeatRecovery
AnaerobicBioreactor
AnaerobicBioreactor
AerobicBioreactor
Sewer
Pre-Filtration
HeatRecovery
ReverseOsmosis
AnaerobicBioreactor
AnaerobicBioreactor
AerobicBioreactor
Sewer
Permeate
Permeate
Brine
Brine
Biological Reduction – Advances
Biological Reduction – Advances
Active Anaerobic Bioreactor SystemEffluent goal is 10 µg/LPilot plant operated for 7 monthsHigh sulfide was an issue
Discharge qualitySolids fouling
Full scale system designed and constructedOperating for about 18 months Compliant water being produced
Future?
Selenium with other metalsConventional lime-iron based processes for bulk removalBiological polishing process
Low concentration seleniumBiological reductionBoth active and passive options
Combinations with other processes – i.e. membranesReduce costs to <$5.00/1,000 gal? Maybe down to $1.00/1,000 gal?
Biological Reduction – Application Keys
Nutrients are vital in establishing microbial populationUnderstanding of site chemistry and environmental interactionsAnalytical methods
Can get discrepancies in total and dissolvedPossibly related to digestionVolatile selenide
Need for aerobic post-treatmentHigh COD, P, N
Conclusions
There is no silver bullet for selenium removal to low levelsAll sites must be evaluated individuallyPaper designs risky –development work always recommendedSelenium can be removed to <10 µg/LCost of selenium reduction to low levels is decreasing
Questions?