Metals Removal from Groundwater Using Permeable Reactive ...
Transcript of Metals Removal from Groundwater Using Permeable Reactive ...
David J.A. Smyth1, David W. Blowes1, Carol J. Ptacek1, Jeff G. Bain1
Department of Earth Sciences, University of Waterloo
Waterloo, ON, N2L 3G1, Canada BC MLARD Workshop, December 2002
Metals Removal from Groundwater Using Permeable Reactive Barriers
(PRBs)
© University of Waterloo 2002
PRBs for Removal of Inorganic Contaminants from Groundwater
• University of Waterloo experience • Blowes, Ptacek and Robertson • Metals, nutrients and water-borne
pathogens • Plume remediation or control • U.S. Patents 5,362,394 5,514,279
5,876,606
© University of Waterloo 2002
Geochemical Barriers for Metals
• Zero-valent iron: reductive precipitation on grain surfaces
• Organic carbon: sulfate reduction, denitrification • BOF Slag: sorption and co-precipitation
phosphate and arsenic • U.S. Patents 5,362,394 5,514,279 5,876,606 • Activated carbon • Limestone (neutralization)
Pilot Scale Installations
USA
Other
Canada
USA
Other
Canada
Full-Scale PRB Installations
n=14 n=21
PRBs for Inorganic Contaminants
United States
Canada
Pilot-Scale Full-Scale
Europe
Australia Europe
Inorganic PRB Sites
Contaminants Treated
Arsenic
PO4
NO3 U
Cr(VI)
Metals and
AMD
Arsenic
PO4
U
Cr(VI)
Metals and
AMD
Pilot Scale Installations Full-Scale PRB Installations
Zero-Valent Iron for Electroactive Metals
• Field Installation: Chromium (VI), Elizabeth City, NC
• Radionuclides (DOE Facilities) • Arsenic, selenium, mercury • Reductive precipitation on grain
surfaces; precipitation or co-precipitation
© University of Waterloo 2002
Elizabeth City Site U.S. EPA Project
• Reference • Blowes, D.W., et al., 1999. An In-Situ Permeable
Reactive Barrier for the Treatment of Hexavalent Chromium and Trichloroethylene in Ground Water: Volume 1 Design and Installation. Volume 2 Performance Monitoring. Volume 3 Multicomponent Reactive Transport Modeling United States Environmental Protection Agency, Cincinnati, OH, Report EPA/600/R-99/095abc.
• http://www.epa.gov/ada/pubs/reports.html
© University of Waterloo 2002
0 10
meters
N
Cr (mg/L) June 1994
Building 78
Pasquotank River 1.6
<0.01
<0.01 <0.01
<0.01
<0.01
<0.01
<0.01
<0.01
5.6
8.2
0.22
3.8
0.08 <0.01
T.A. Bennett, M.Sc. Thesis, University of Waterloo, 1997
Bui
ldin
g 78
HANGAR 79
© University of Waterloo 2002
Reactive Material
• 150 m3 zero valent iron (280 tons) • 46 m long, 7.3 m deep and 0.6 m wide barrier
© University of Waterloo 2002
One-Pass Continuous Trencher
© University of Waterloo 2002
One-Pass Continuous Trencher
• Depths of < 30 ft • Width 1-2 ft • Very rapid installation • Big equipment • Mobilization
© University of Waterloo 2002
USCG Wall Installation
© University of Waterloo
USCG Wall Installation
-6
-4
-2
Transect 1 (November 1996)
Depth (m)
Groundwater flow direction 0 50 cm
5
6
7
8
9
10
Wall
pH
T.A. Bennett, M.Sc. Thesis, University of Waterloo, 1997
Dep
th (m
)
© University of Waterloo 2002
-6
-4
-2
Transect 1 (November 1996)
Depth (m)
Groundwater flow direction 0 50 cm
0
0.05
1.0
2.0
Wall
Cr(VI) (mg/L)
T.A. Bennett, M.Sc. Thesis, University of Waterloo, 1997
Dep
th (m
)
© University of Waterloo 2002
Mineralogical Characterization
• Increased solid-phase carbon • Carbonate mineralogy
• Iron oxyhydroxides • goethite • ferrihydrite • green rust
• Iron Sulfides © University of Waterloo 2002
Long-Term Efficiency (Mayer)
0.0 0.1 0.2 0.3 0.4 0.5distance [m]
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0ϕ F
e[-
]0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
φ[-
]
ϕFe - T = 5 yearsϕFe - T = 10 yearsϕFe - T = 20 yearsφ - T = 5 yearsφ - T = 10 yearsφ - T = 20 years
porosity
volume fraction of secondary minerals
volume fraction ofzero-valent iron
© University of Waterloo 2002
1) Reduction and Co-precipitation with Goethite i) 4Fe0
(s) + 3O2(g) + 6H20(l) 4Fe3+(aq) + 120H-
(aq)
ii) Fe3+(aq) + H3AsO3(aq) + 2H20(l) FeO(OH,H2AsO4)(s) + 5H+
(aq)
2) Sulphate Reduction
i) 4Fe0(s) + SO4
2-(aq) + 10H+
(aq) H2S(aq) + 4Fe2+(aq)
+ 4H2O(l)
ii) 2As3+(aq) + 3H2S(aq) As2S3(s) + 6H+
(aq)
3) Adsorption
ARSENIC Mechanisms for Removal
© University of Waterloo 2002
McRae (1999): Arsenic Removal Mechanisms
• Energy Dispersive X-Ray Analysis • As present in grain coatings and
possibly on zero-valent iron grain surface
• X-Ray Photoelectron Spectroscopy • As(III) predominant in solid phase • Reductive precipitation and co-
precipitation with goethite in coatings © University of Waterloo 2002
Column Experiments
© University of Waterloo 2002
Zero Valent Iron Column
Pore Volumes
0 20 40 60 80 100 120
As
(µg/
L)
0
200
400
600
800
1000
Input = 1000 µg/L
© University of Waterloo 2002
Pore Volumes0 5 10 15 20 25 30
Ars
enic
(mg/
L)
0
5
10
15
Arsenic Concentration in 100% Iron Column
Note: Detection Limit is 0.005mg/L Influent Effluent
© University of Waterloo 2002
Total Arsenic Concentration Profiles in ZVI Column Mine Groundwater at Velocity of 6.75 cm/day
Distance into Iron (cm)
0 5 10 15 20 25 30 35
Ars
enic
Tot
(mg/
L)
0
1
2
3
4
5
6 Profile after 29.6 PV Profile after 37.5 PV
© University of Waterloo 2002
Sulfate-Reduction PRBs • Metals in sulfate rich water- AMD • Sulfate reduction is microbially mediated
process • Purpose of PRB is to intercept
groundwater flow and enhance sulfate reduction
• Generation of H2S and precipitation of metal sulfide minerals
• Decrease acid-generating potential; remove dissolved metals © University of Waterloo 2002
Acid Mine Drainage and Sulfate Reduction
FeS2(s) + 7/2O2 + H2O => Fe2+ + 2SO42- + 2H+
Fe2+ 1/4O2 + 5/2H2O => Fe(OH)3(s)+2H+
SO42- + 2CH2O => H2S +2HCO3
-
Fe2+ + H2S => FeS + 2H+
Tailings Dam
Sulfate Reduction
Sulfide Oxidation
Iron Oxidation
Reactive Wall
© University of Waterloo 2002
Nickel Rim Mine, Sudbury, ON
• Laboratory batch and column study • Predictive groundwater flow
modelling • Field installation (1995) • Benner et al., 1997; 1999 • Waybrant et al., 1998
© University of Waterloo 2002
Reactive Mixture Composition for PRB
Leaf Compost Municipal
Compost
Wood Chips
Limestone
© University of Waterloo 2002
Bedrock
Sand
Sand
Reactive Material
Direction of Groundwater flow
4 m 8 m
3.6 m
15 m
Porous Reactive Wall Installation
© University of Waterloo 2002
© University of Waterloo
Nickel Rim Wall Materials
© University of Waterloo
NR Wall Installation
Sand
Sand
Compost
© University of Waterloo
Nickel Rim Wall
Clay Cap
© University of Waterloo
Nickel Rim Wall
>350 250-350 150-249 50-149 <50
>350 250-350 150-249 50-149 <50
surface water recharge
? ?
material sand sand reactive
meters
0 5 m
Chloride (mg/L) June 96
Chloride (mg/L) Nov. 95
groundwater flow direction
Benner et al., 1997
Groundwater Flow
© University of Waterloo 2002
Benner et al., 1997
Acid Generating Potential (meq/L)
Sulfate (mg/L)
Iron (mg/L)
groundwater flow direction
meters
0 5 m reactive material sand sand
>20 10 - 20 0.0 - 10 -10 - 0.0 <-10
>2000 1500-2000 1000-1499 500-999 <500
>350 250-350 150-249 50-149 <50
Treatment Results
© University of Waterloo 2002
Sulfate Reduction in PRB
• Decreasing sulfate concentrations • Sulfate-reducing bacteria • Dissolved sulfide present • Isotopic enrichment of 34S in remnant
sulfate • Iron monosulfides identified in cores
© University of Waterloo 2002
reactive material sand sand
groundwater flow direction 0 15 feet
0 5 meters
Sulfate (mg/L)
August 2001
4900
4720
3400
2940
4370
2950
2834
3430
1561
2188
1086
379
2497
57
1995
2036
1960
1992
2746
2558
3708
2652
2483
2870 2687
4190
2578
3408
2620
2549
2439
2229
2156
2726
2212
2473
3054 2289
2510
2240
2424
2551
3026
2812
> 3500 2500- 3500 1500- 2499 500- 1499 < 500
2900
3202
2000
1936 3
R W
2 1
R W
2 3
R W
2 9
R W
3 0
R W
2 6
R w
3 1
1
2
3
4
1
2
3
4
3
1
2
4
2
4
3
1
3
4
2
1
1
2
3
4
1
2
3
4
R W
3 6
R W
3 2
R W
3 3
R W
3 4
R W
3 5
2
3
4
1
2
3
4
1
2
3
4
1
2
4
1
2
3
4
R W
2 8
0 0
5 5
meters
reactive material sand sand
groundwater flow direction
712
1258
654
356
978
369
672
19.7
114
105
3.78
299
448
.86
383
222
165
478
404
547
302
893
461
248 358
526
633
323 263
772
441
265
295
148
457 270
290
245
724
383
158
135
309
912
605 528
441
0- 499 500- 999 1000- 1499 1500- 1999 > 2000
3
R W
2 1
R W
2 3
R W
2 9
R W
3 0
R W
2 6
R w
3 1
1
2
3
4
1
2
3
4
3
1
2
4
2
4
3
1
3
4
2
1
1
2
3
4
1
2
3
4
R W
3 6
R W
3 2
R W
3 3
R W
3 4
R W
3 5
2
3
4
1
2
3
4
1
2
3
4
1
2
4
1
2
3
4
R W
2 8
0 0
5 5
meters
0 15 feet
0 5 meters
Iron (mg/L) August 2001
Sulfide Accumulation in Nickel Rim PRB (Daignault 2002, UW B.Sc.Thesis)
0
10
20
30
40
50
60
70
80
90
0 10 20 30 40 50 60 70 80 90
Time (months)
Rate
(µ
mo
l S
cm
-3 a
-1) Up-gradient core (NRW29)
Mid-gradient core (NRW30)
Down-gradient core (NRW31)
Summary of Nickel Rim PRB • The reactive wall is removing significant
portion of the dissolved iron from the plume; full treatment would have required thicker PRB with longer residence time
• Reduced flux of contaminants in groundwater; reduced acid-generating potential of groundwater in receiving surface water
• Cost for materials and installation approximately $25 K (US) in 1995
© University of Waterloo 2002
Issues • Heterogeneities in PRB/ residence
time of contaminated groundwater in PRB is critical to level of treatment achieved
• Some loss of reactivity with time; sustained availability of organic carbon
• Influence of temperature in shallow PRB systems © University of Waterloo 2002
STEEL PRODUCTION WASTES Basic Oxygen Furnace (BOF) Slag • Steel production waste product • Used as aggregate for construction • High Ca and Fe oxides and
hydroxides • Interaction with water: high pH • Removal of phosphate (Baker et al.,
1997; 1998) and arsenic(McRae et al., 1999)
© University of Waterloo 2002
BOF-Oxide
Zero Valent Iron
Activated Alumina
Silica Sand
Limestone
Solid Reactive Mixtures
Aquifer Materials Reactive Materials
Batch Removal Rates
Time (h)
0 2 4 6 8
0
200
400
600
800
1000
As(III) + As(V)
Time (h)
0 2 4 6 8
0
200
400
600
800
1000
Se(VI)
Se(
VI) (µg
/L)
Asto
tal (µg
/L)
Iron Slag
BOF Slag Column
Pore Volumes
0 50 100 150 200
As (µ
g/L)
0
200
400
600
800
1000
1200
As(total)
Second Column Test
• 50 % BOF slag/ 50 % gravel • Low pH site groundwater with 4 mg/L
arsenic • More than 75 pore volumes of flow
(velocity of 0.3 m/day) • Total arsenic concentration in
effluent less than 0.01 mg/L
Arsenic Removal by BOF Slag
• Removal of arsenic oxyanions by sorption iron and manganese oxyhydroxides in BOF
• Removal to low levels (<0.005 mg/L total arsenic)
• Sustained performance for 100 pore volumes of 10 % BOF mixture
North Bay System: pH with Time
Days of Operation0 200 400 600 800 1000
pH
0
2
4
6
8
10
12
14
Septic System pH with Time Sand-Filter pH with TimeBOF Chamber pH with Time
North Bay System: E-Coli with Time
Days of Operation0 200 400 600 800 1000
E-C
oli (
CFU
per
100
mL)
1e-1
1e+0
1e+1
1e+2
1e+3
1e+4
1e+5
E-Coli in Raw Water E-Coli in Sand Filter E-Coli in BOF Filter
BOF-Chamber Performance • Effective removal of phosphorus • Effective removal of E-Coli
• Elevated pH provides environment that eliminates bacteria
• Elevated pH of 12 or higher • Elevated pH is buffered by soils and sediments
upon release to subsurface • pH of groundwater approximately 1 m down-
gradient of discharge gallery was 6.2 to 7 (August 2000)
© University of Waterloo 2002
Zero-Valent Iron and Other Reactive Materials
• Excellent removal of electroactive metals
• Sulfate reduction and AMD treatment • Excellent treatment of nutrients • Performance of field-scale applications • Removal mechanisms • Reactive capacity • Formation of secondary precipitates
© University of Waterloo 2002