INTEGRATED PROCESSES FOR TREATMENT OF BERKELEY PIT WATER

1
INTEGRATED PROCESSES FOR TREATMENT OF BERKELEY
PIT WATER

• ACTIVITY III, PROJECT 21

2
BACKGROUND

• The Berkeley Pit (Butte, Montana) - is currently
  filling at a rate of 3 million gallons per day
  of acidic, metals laden water
• EPA issued a Record of Decision in 1994 the
  Berkeley Pit will be allowed to fill until
  approximately 2021, at which time the water level
  will approach the Critical Water Level

3
BACKGROUND (cont.)

• Treatment technologies will be revisited
  approximately 2009 treatment required
  essentially forever
• ROD designated hydroxide precipitation with
  aeration (followed by reverse osmosis if
  necessary) as preferred treatment technology
• Over 1000 tons per day of dewatered sludge will
  be produced

4
PROJECT CONCEPT

• Value of contained metals presents opportunity
  for offsetting treatment costs via product
  recovery/resale
• Acid mine drainage a worldwide problem
• Project will evaluate both proven and new
  technologies for optimizing overall economics of
  producing compliant water

5
PROJECT CONCEPT (Contd.)

• All aspects of problem will be included
• Challenges
• Distance of Butte, Montana from markets
• Dilute feed stream (though extremely
  contaminated)
• Low-value base metals present

6
CURRENT PROJECT SCOPE

• Develop two optimized flowsheets
• Water Treatment-Only
• Water Treatment-Plus-Product Recovery
• If results economically attractive, pursue pilot
  testing of optimized product recovery
  process at Berkeley Pit

7
CURRENT PROJECT SCOPE (Contd.)

• Major Tasks
• Prepare standardized cost-estimating methodology
• Develop optimization strategy (identify/prioritize
  potential process improvements)

8
PROJECT STATUS AS OF APRIL 2000

• Work plan complete
• Conceptual design of sludge repository complete
• Cost estimating methodology document complete
• Document verifying technical and cost aspects of
  reference flowsheets complete
• Optimization strategy in development
• Preliminary optimization efforts underway
  (gathering cost/technical data applicable to both
  flowsheets)

9
PROJECT SCHEDULE

• Final report describing optimized flowsheets due
  for publication in November 2000

10
IMPROVEMENTS IN ENGINEERED BIOREMEDIATION OF ACID
MINE DRAINAGE Activity III, Project 24
11
Project Objectives

• Objectives for improvements of engineered
• features of a passive SRB-bioreactor include
• Selection of media
• Design of a permeability and contact time
  enhancing system (PACTES),
• Design of an organic carbon replaceable cartridge
  system (RCS),
• Development of computer software to model SRB
  bioremedial processes in the bioreactor.

12
Scope of Work

• The scope of work of the project includes seven
  tasks
• Task I
• Selection of organic carbon media that
• is permeable when saturated with water,
• contains sufficient mass of organic carbon to
  minimize treatment rates, and
• Could be economically used for passive SRB
  bioreactors.

13
Scope of Work, cont.

• Task II
• PACTES design, evaluation through a bench test
  study, and implementing it in the field.
• Task III
• Designing of an organic carbon RCS that would
  be easy to install and replace in a bioreactor at
  a remote location.

14
Scope of Work, cont.

• Task IV
• Development or adaptation of computer software
  to model SRB bioremedial processes in the
  bioreactor.
• This task includes efforts on
• Software development and validation
• Lab experiments for bioreaction kinetics

15
Scope of Work, cont.

• Task V
• Implementation of the results of the four
  previous tasks in a bioreactor constructed for
  this purpose.
• Task VI
• Project management activities.
• Task VII
• Site selection and characterization

16
Status of Work(as of 03/31/00)

• Task I was initiated in February, 2000.
• Data base structure is 60 developed.
• Search of information is advanced approximately
  30.

17
SLUDGE STABILIZATION

• Activity IV Project 2

18
OBJECTIVE

• Formation, properties and stability of sludge
  generated during treatment of acid mine waste
  water
• Physically and chemically characterize sludges
• Study the stability of sludges created by
  treatment techniques
• Apply to acid mine water
• Point source
• Non-point source

19
Stabilization Techniques will be Developed for
Hazardous Sludge

• Commonly used additives for metallurgical waste
  solids
• Thermal Processing
• Effective for arsenic bearing waste
• Recovery of metal values or removal of hazardous
  constituent/recycling to metallurgical processes
• In particular, sulfide sludge

20
DEMONSTRATION OF ARSENIC REMOVAL TECHNOLOGY

• Activity IV Project 5

21
OBJECTIVES

• Remove Arsenic from Solution
• Characterize Solid Products
• Determine Stability During Storage

22
CONCEPT

• Produce an apatite mineral-like structure with
  the substitution of arsenate for phosphate in the
  structure

23
REMOVAL OF ARSENIC FROM WASTE SOLUTIONS

• WHAT IS WRONG WITH SIMPLE LIME PRECIPITATION??

24
EPAs BDAT FOR As BEARING WASTEWATERS

• Ferrihydrite precipitation is an adsorption
  phenomena
• Potential Problem
• Long-term storage

25
ASARCO DEMO RESULTS

• Scrubber Blowdown Water
• gt3,000,000 ppb As to lt10ppb As
• Thickener Overflow
• 6,000 ppb As to lt15 ppb
• Long-term Aging Presently Being Conducted (ASARCO
  and Mineral Hill Products)

26
MINE WASTE BERKLEY PIT LAKE CHARACTERIZATION
PROJECT

• Activity IV Project 8

27
CHARACTERIZATION PROJECTS

• DEPTH PROFILES
• ORGANIC CARBON
• SRB ACTIVITY IN SEDIMENTS
• SURFACE WATER REACTION KINETICS

28
SUMMARY

• Berkley Pit Lake system is complex and requires
  much more research to fully understand
• Knowledge gained through work on Berkley Pit may
  be used on other pit lakes through out the world

29
Artificial Neural Networks As An Analysis Tool
for Geochemical Data

• Activity IV Project 14

30
WHY USE NEURAL NETWORK?
TO SORT THROUGH OR ANALYZE VERY LARGE DATA
VOLUMES NNs basically think like the human brain
31
ALGAL REMEDIATION DATA OF BERKELEY PIT

• 4 Classes of Data with 15 Samples
• Within each class, 5 subclasses exist with 3
  samples each

32
Self- Organizing Map

• Groups Data According to Trends Within the Data
• For Algae, the SOM Output Compared to Known Data
  Classes
• NOTE Neural Networks can also be used to
  predict data

33
Future Possibilities for NN Analysis of Algae

• Look for behavior trend within Algae species
• Compare similarities and differences
• Train network to recognize different Algae
  species and concentrations
• Develop network to predict Algae types and
  concentrations from pit-water metal concentrations