ARD REMEDIATION WITH SLAG: AN APPLICATION TO BERKELEY PITLAKE WATER. Courtney A. Young

Transcription

1 ARD REMEDIATION WITH SLAG: AN APPLICATION TO BERKELEY PITLAKE WATER Courtney A. Young Dept Head and Lewis S. Prater Professor Metallurgical & Materials Engineering Montana Tech Butte MT Denver, CO Monitoring and Treatment Session 8 April 3-5, 2012

2 Dick Berg, State Geologist, MBMG Montana Tech, Butte MT Larry Twidwell, Professor Emeritus, M&ME Montana Tech, Butte MT Krag Filius, Project Engineer MSE Technology, Butte MT Eric Streich, Process Engineer Holcim Cement, Trident MT

3 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

4 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

5 Continental Pit Treatment Plant Horseshoe Bend Water MR Concentrator Viewing Stand Butte Walkerville Interstate 15/90

6

7 Butte was an early copper-mining town: - Referred to as The Richest Hill on Earth - One of the world s largest sulfide ore deposits ACC controlled most mines ACC began phasing out underground mining ARCO purchased all operations Operations halted and pumps turned off Water first appeared in the pit - Listed as a Superfund site - Part of the largest mining Superfund site

8 Concentrations change with position, depth and time Berkeley Pitlake Water: - is acidic near ph contains metals at high concentrations (99% Water): SO 4 (7500 ppm) Fe (1000 ppm) 1984 Zn (650 ppm) Al (300 ppm) Mn (250 ppm) Cu (200 ppm) Cd (2.5 ppm) As (0.5 ppm) Berkeley Pitlake Water: - encompasses ~700 acres - is ~1,000 feet deep - contains ~40 billion gallons - fills at 2.6 million gallons per day - will reach critical level in 2023

9 Continental Pit Horseshoe Bend Water

10

11 HSBW also diverted to pond (3M GPD) - MR starts operating Continental Pit to the east - ARCO and MR named responsible parties MR halts operations including diversion HSBW Treatment Plant is commissioned - two-stage lime precipitation process - diversion of treated water begins - sludges are disposed into the BPL MR reopens and begins full operations MR pumps BPL water to Cu-cementation

12 Horseshoe Bend Water Treatment Plant

13 Horseshoe Bend Water Treatment Plant

14

15 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

16 Participate in a series of 5 studies to summarize available information Generate new information to formulate conceptual environmental models for the Berkeley Pitlake from all of its features Provide data for the development of advanced treatment technologies

17 Fe As Mn Cu Cd Zn Al Initial BPL Water Stage 1A - H 2 O 2 /UV 8.43 < Stage 1B - KMnO < Stage 2 - Na 2 S 0.27 < Stage 3 - Na 2 S 0.22 < < 0.05 < Stage 4 - NaOH < 0.04 < < 0.05 < Drinking Standard Stage 1A: H 2 O 2 = 2OH ; Fe 2+ + OH = Fe 3+ + OH - ; Fe OH - = Fe(OH) 3 Stage 1B: 3Mn MnO H2 O = 5MnO 2 + 4H + Stage 2: Stage 3: Cu 2+ + S 2- = CuS Cd 2+ + S 2- = CdS; Zn 2+ + S 2- = ZnS Stage 4: Al OH - = Al(OH) 3

18 U V S o u r c e G a s E l e c t r o d e G l a s s F r i t M a g n e t M a g n e t i c S t i r r e r

19 Berkeley Pitlake Water Fe/Mn/As Precipitation Cu Precipitation Zn/Cd Precipitation Al Precipitation SO Remediation UV/H O /NaOH 2 2 H S 2 H S 2 NaOH Vacuum Na, K, Mg, Ca Solution To Softening and Discharge? Selective metal recovery is possible A 7-stage process has been envisioned and shown to work (in batch mode) Fe, As, Cu and Cd met DWS Al almost met DWS Mn and Zn did not meet DWS KMnO 4 addition needs to be precise Zn may have precipitated amorphously SO 4 removal was not done but options are Freeze Crystallization Reverse Osmosis Gypsum Precipitation SRB Bioreduction Chemoreduction Photoreduction Reductive Precipitation Softening to remove Na, K, Mg and Ca

20 p H D i s o E ( m V ) H o T e m p e r a t u r e ( C ) p H B e r k e l e y P i t l a k e S e p t. 5, l v e d O x y g e n ( m g / l ) D e p t h ( m ) E H D O T e m p

21 1.5 (HSO 4 - ) (SO 4 2- ) E H (Volts) Fe(SO 4 ) + Fe 2+ Fe(SO 4 ) 2 - * FeSO 4 ( aq ) Fe 8 O 8 (OH) 6 SO 4. 5H 2 O {H 2 S( aq )} Fe (HS - ) ph

22 Profiles indicated chemoclines/thermoclines existed and were successfully reproduced in lab They have been explained by, but can not be totally attributed to HSBW being less dense than BPLW so, when it enters the pitlake, it floats on top rather than mixes in, and Biological activity which should increase DO as well as ph Experiments showed that the interaction of sunlight (UV radiation) and air with BPL water plays a significant role

23 (Deep Water, Pore Water and Sediment) Collect Core Sample Siphon/Filter Off Deep/Pore Water Split & Section the Core Analyze the Water & Solid Contents

24 Fe3,ppm Concentrations are controlled by the solubility of identified minerals and precipitates! (Deep Water, Pore Water and Sediment) Surface Water (ph ~ 2.5) Deep Water (ph ~ 3.3) Ferric Iron Solubility in Pore Water Schwertzmannite zz ww wvv z w z z Fe8O8(OH)6SO Potassium Solubility Jarosite A x x x x A + B z z z w w w z v v A A + B A KFe 3 ( B KAl 2 A ph ph

25 (Deep Water, Pore Water and Sediment) Muscovite [KAl 3 Si 3 O 10 (OH) 2 ] controls Al 3+ concentration Quartz (SiO 2 ) controls Si 4+ concentration Schwertmannite [Fe 8 O 8 (OH) 6 SO 4 ] precipitate controls the Fe 3+ concentration Jarosite [KFe 3 (SO 4 ) 2 (OH) 6 ] precipitate controls K + concentration Cu 2+, Fe 2+, Zn 2+ and Cd 2+ concentrations could not be associated with a mineral or precipitate are therefore considered to be unsaturated However, Cu 2+, Fe 2+, Zn 2+ and Cd 2+ concentrations were found to increase with depth giving the appearance that supergene deposition is occurring

26 Mineralogy is essentially the same except fine native rock (granite) and gypsum precipitate are more abundant: Native: Granite (38%) Quartz (33%) Muscovite (4%) Muscovite coated with Jarosite Precipitate: K-jarosite (22%) Gypsum (3%)

27 p S i O 2 Chemical controls should be about the same as at depth A m o r p h o u s S i O 2 A B C D E F G A - K a o l i n i t e B - M u s c o v i t e C - K F e l d s p a r D - O r t o c l a s e E - A l b i t e F - A n o r t h i t e G - A n n i t e 4 Deep Water ~p H p H

28 p S i O 2 Silicate (and oxide) slags should do the same thing! H I J K A m o r p h o u s S i O 2 H - F a y a l i t e I - P s u e d o w o l l a s t o n i t e J - A c k e r m a n i t e K - R a n k i n i t e p H

29 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

30 Source of Silicate (and lime) Act as ph-buffers (replace lime) Available in Montana (inactive) Anaconda (ARCO) - Fayalite, Fe 2 SiO 4 East Helena (ASARCO) - Olivine-type, CaFeSiO 4 Rocker (Rhone) - Pseudowallastonite, CaSiO 3 Slag Ca (%) Fe (%) Si (%) Rhone ASARCO Anaconda

31 Rocker/Rhone Poulenc Pseudowollastonite (CaSiO 3 ) East Helena/ASARCO Olivine-type (FeCaSiO 4 ) Anaconda/ACC Fayalite (Fe 2 SiO 4 )

32 Other Global Sources Columbus (Stillwater) Salt Lake City (Kennecott) Trail, BC (Teck Cominco) Dual Ecosystem Enhancement Remove Slag Piles Remediate Berkeley Pitlake In-Situ or Ex-Situ Provide Entertainment Golf Courses Parks & Walkways Sports Complexes Attract Businesses

33 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

34 Conceptual Flowsheet Designs Dry Grinding Slag Ball Mill Cyclone Pneumatic Spray BPL Slag Ball Mill Cyclone Mixer & Pump BPL BPL Water Wet Grinding Pump Slag Ball Mill Hydrocyclone BPL BPL Water Pump

35 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

36 Characterize Montana Slags Bond Work Index SEM/EDX/MLA Analysis Remediation Potential Model Effects Parameters Slag Type (Fe/Si Ratio) Particle Size ( Mesh) Slag Amount ( g/l) Experimental Design (StatEase)

37 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

38 Date Slag Target Size Mesh (mm) F80 (mm) P80 (mm) Avg Gbp Bond Work Index 1/29/05 ACC 48 (0.295) /17/05 ACC 100 (0.147) /20/05 ACC 200 (0.074) /29/05 ASARCO 48 (0.295) /12/05 ASARCO 100 (0.147) /30/05 ASARCO 200 (0.074) /26/05 RP 48 (0.295) /4/05 RP 100 (0.147) /4/05 RP 200 (0.074)

39 (150 um Asarco Slag Before)

40 (150 um Asarco Slag After)

41 (Bottle Roll Tests) Size = 53 um; Amount = 100 g/l 12 hours

42 (Bottle Roll Tests) Size = 53 um; Time = 12 hrs ARCO ASARCO Rhodia ph Concentration (grams/liter)

43 (Bottle Roll Tests) Amount = 100 g/l; Time = 12 hrs

44 Run (Box-Behnken Matrix) Slag Type (Fe/Si Ratio) Particle Size (μm) Slag Amount (g/l) 1 0 = Rhodia -37 = 400 mesh = ARCO = 100 mesh = 200 mesh = ASARCO

45 ph Al As Cd Cu Fe Mn SO 4 2- Zn Test Run BPL Concentrations (mg/l) Final Responses (mg/l) Drinking Water Standards (mg/l)

46 A = Fe/Si Ratio (0-2); B = Size (um); C = Amount (g/l) ph = A B C AB + 0.7A 2 Log [H] = A B C 0.006AB 0.7A 2 Log [Al] = A B C AC [As] 0.5 = A C 0.006AC A 2 [Cd] 0.5 = A 0.027*C Log [Cu] = A B C AB AC BC A C 2 Log [Fe] = A B C [Mn] = A + 252B 127C AC 86310A 2 [Zn] 0.5 = A 0.49C

47 Size = 37 μm Size = 147 μm

48 C: Amount, g/l Design-Expert Software ph ph 4.57 X1 = A: Fe/Si Ratio X2 = C: Amount, g/l Actual Factor B: Size, um = A: Fe/Si Ratio

49 Fe (ppm) Size = 92 μm Fe/Si Ratio Amount (g/l)

50 C: Amount, g/l Design-Expert Software Original Scale Log10(Fe, ug/l) Fe, ug/l 13.7 X1 = A: Fe/Si Ratio X2 = C: Amount, g/l 650 Actual Factor B: Size, um = A: Fe/Si Ratio

51 a) b) Fe (ppm) Zn (ppm) Fe/Si Ratio Conc (g/l) Fe/Si Ratio Conc (g/l) c) d) As (ppb) Fe/Si Ratio Conc (g/l) Cd (ppb) Fe/Si Ratio Conc (g/l)

52 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

53 Slags can be an effective for remediating ARD Their use could or will: replace lime (pseudowollastonite slag) diminish lime consumption (fayalite/olivine) lead to remediation of two ecosystems Depending on the slag type and particle size: effluent ph from 5-9 can result effluent concentrations can meet DWS Al and Cu concentration profiles are similar to Fe Likewise, Al and Cu redissolution at high ph is minimal similar to Fe

54 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

55 We are always on the lookout for funding Series VI to! This research was previously funded by the MWTP via an Interagency Agreement (IAG) between the U.S. EPA and the U.S. DoE, IAG No. DW Thanks are also extended to the Department of Metallurgical & Materials Engineering at Montana Tech for bearing the costs for some analyses and the MBMG for helping collect samples: James Madison and Ted Duaime. Sincere appreciation is given to the workforce who have worked on these projects over the years. MS students included Ray Ziolkowski, Tom McMillan, Yu Chuan Tai, Eric Streich and Krag Filius. BS students include Sonny Adams, Jennifer Gambill and Brian Ross.

56 Berkeley Pitlake Previous Research Silicate Slags Objectives Procedures Results & Discussions Conclusions Acknowledgements

57 U.S. EPA Hardrock Mining Conference (Advancing Solutions for a New Legacy) Denver CO April 3-5, 2012

58 ?