MWTP-237 FINAL REPORT—IMPROVEMENTS IN ENGINEERED BIOREMEDIATION OF ACID MINE DRAINAGE MINE WASTE TECHNOLOGY PROGRAM ACTIVITY III, PROJECT 24 Prepared by: MSE Technology Applications, Inc. 200 Technology Way P.O. Box 4078 Butte, Montana 59702 Prepared for: U.S. Environmental Protection Agency Office of Research and Development National Risk Management Research Laboratory Cincinnati, Ohio 45268 IAG No. DW89938870-01-1 and U.S. Department of Energy Savannah River Operations Office Aiken, South Carolina 29802 Contract No. DE-AC09-96EW96405 September 2004

REVIEWS AND APPROVALS (MWTP-237):

Prepared by:

Project Engineer Approved by: Program Manager

September 2004

Mine Waste Technology Program

Improvements in Engineered Bioremediation of Acid Mine Drainage

By:

Marek Zaluski, Ph.D, Project Manager MSE Technology Applications, Inc.

Mike Mansfield Advanced Technology Center Butte, Montana 59702

Under Contract No. DE-AC09-96EW96405 Through EPA IAG No. DW89938870-01-1

Diana Bless, EPA Project Manger Sustainable Technology Division

National Risk Management Research Laboratory Cincinnati, Ohio 45268

This study was conducted in cooperation with

U.S. Department Energy Savannah River Operations Office

Aiken, South Carolina 29802

National Risk Management Research Laboratory Office of Research and Development

U.S. Environmental Protection Agency Cincinnati, Ohio 45268

Notice The U.S. Environmental Protection Agency through its Office of Research and Development funded the research described here under IAG DW89938870-01-1 through the U.S. Department of Energy (DOE) Contract DE-AC09-96EW96405. It has been subjected to the Agency’s peer and administrative review and has been cleared for publication as an EPA document. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement or recommendation. The views and opinions of authors expressed herein do not necessarily state or reflect those of the EPA or DOE, or any agency thereof.

ii

Foreword The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions leading to a compatible balance between human activities and the ability of natural systems to support and nurture life. To meet this mandate, EPA's research program is providing data and technical support for solving environmental problems today and building a science knowledge base necessary to manage our ecological resources wisely, understand how pollutants affect our health, and prevent or reduce environmental risks in the future. The National Risk Management Research Laboratory is the Agency's center for investigation of technological and management approaches for preventing and reducing risks from pollution that threaten human health and the environment. The focus of the Laboratory's research program is on methods and their cost effectiveness for prevention and control of pollution to air, land, water, and subsurface resources; protection of water quality in public water systems; remediation of contaminated sites, sediments, and groundwater; prevention and control of indoor air pollution; and restoration of ecosystems. The NRMRL collaborates with both public and private-sector partners to foster technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's research provides solutions to environmental problems by developing and promoting technologies that protect and improve the environment; advancing scientific and engineering information to support regulatory and policy decisions; and providing the technical support and information transfer to ensure implementation of environmental regulations and strategies at the national, state, and community levels. This publication has been produced as part of the Laboratory's strategic long-term research plan. It is published and made available by EPA's Office of Research and Development to assist the user community and to link researchers with their clients.

Lawrence W. Reiter, Acting Director National Risk Management Research Laboratory

iii

Abstract Acid mine drainage (AMD), often generated by active and abandoned hard-rock mines, represents a significant environmental threat. Because many abandoned mine sites have difficult access, a passive remedial technology is needed. Ubiquitous sulfate-reducing bacteria (SRB) have the ability to increase pH and to immobilize dissolved metals through sulfide precipitation. Previous research showed that conditions appropriate for SRB can be engineered in a bioreactor driven only by hydraulic gradient and containing organic matter delivering carbon as a nutrient for SRB. This research also showed that such bioreactors often plug, and their design lacks operational-time prediction. The research reported in this document addressed the problem of bioreactor plugging by: (1) documenting that the horizontal flow in a bioreactor is superior to the vertical flow, (2) developing a replaceable cartridge for a modular SRB bioreactor, and (3) inventing a new reactive medium, a mix of walnut shells and cow manure. This medium does not settle and includes organic matter with both high and low biodegradation rates desired for a quick start and a long-term performance of a bioreactor. The operational-time prediction was addressed by developing a bioreactor economics, size and time of operation (BEST) simulator. BEST is a spreadsheet model used to design the bioreactor depending on the AMD composition, sulfate reduction rate, and an assumed rate of organic carbon depletion. BEST includes a routine for using the PHREEQC geochemical model to simulate the chemical reactions occurring as AMD flows through the bioreactor.

iv

Contents Page

Notice...................................................................................................................................................... ii Foreword.................................................................................................................................................. iii Abstract ................................................................................................................................................... iv Contents .................................................................................................................................................. v Figures .................................................................................................................................................... vi Tables...................................................................................................................................................... vii Acronyms and Abbreviations .................................................................................................................. viii Acknowledgments................................................................................................................................... x Executive Summary ................................................................................................................................ ES-1 1. INTRODUCTION .......................................................................................................................... 1 1.1 Background............................................................................................................................. 1 1.2 Principles of the Sulfate-Reducing Bacteria Technology....................................................... 1 2. PROJECT OBJECTIVES ............................................................................................................... 3 3. SELECTION OF THE MEDIUM WITH ORGANIC CARBON.................................................. 4 3.1 Literature Search Results........................................................................................................ 4 3.2 Substrate Selection for the Project.......................................................................................... 4 4. REPLACEABLE CARTRIDGE .................................................................................................... 6 4.1 Engineering Features .............................................................................................................. 6 4.1.1 Flow Configuration...................................................................................................... 6 4.1.2 Replaceable Cartridge Design ..................................................................................... 9 4.2 Organic Matter........................................................................................................................ 11 4.2.1 Laboratory Experiments for Sulfate Reduction Rate................................................... 11 4.2.2 Results ......................................................................................................................... 12 5. DESCRIPTION OF THE BEST SIMULATOR............................................................................. 28 5.1 Organization of the BEST Simulator...................................................................................... 28 5.1.1 ExcelTM Workbook ...................................................................................................... 28 5.2 Input Data Summary and Explanations .................................................................................. 33 5.3 Modeling Process ................................................................................................................... 33 5.3.1 Treatment Scope Selection .......................................................................................... 33 5.3.2 Treatment System Design............................................................................................ 34 5.4 Comments on the BEST Simulator and its Limitations.......................................................... 36 6. SUMMARY OF QUALITY ASSURANCE ACTIVITIES ........................................................... 57 6.1 Background............................................................................................................................. 57 6.2 Data Evaluation ...................................................................................................................... 57 6.3 Validation Procedures............................................................................................................. 57 6.3.1 Analytical Evaluation .................................................................................................. 57 6.3.2 Duplicate Experiments................................................................................................. 57

v

Contents (Cont'd)

Page 6.3.3 Logbook Review.......................................................................................................... 58 6.3.4 Data Review ................................................................................................................ 58 6.4 Summary................................................................................................................................. 58 7. CONCLUSIONS ............................................................................................................................ 60 8. REFERENCES ............................................................................................................................... 63 Appendix A: CD Containing Report "Evaluation of Organic Substrates for the Growth of

Sulfate-Reducing Bacteria to Treat Acid Mine Drainage" (MWTP-188) and MS AccessTM Database................................................................................................... A-1

Appendix B1: Engineering Drawing: SRB Cell Construction.............................................................. B1-1 Appendix B2: Engineering Drawing: SRB Cell Components and Construction Details...................... B2-1 Appendix B3: Mechanical Adaptation of a Typical Tank and Construction of the

SRB RC .......................................................................................................................... B3-1 Appendix C: CD Containing Bioreactor Economics, Size, and Time of Operation (BEST)

Simulator ........................................................................................................................ C-1

Figures Page

4-1. Test Sump with Vertical Pipes used for Long Duration Field Test of Hydraulic Conductivity

for 0.5/0.5 W/M Organic Medium.............................................................................................. 14 4-2. Long Duration Field Test of Hydraulic Conductivity for Two W/M Organic Media

Configured for Upward Vertical Flow........................................................................................ 15 4-3. Al and Fe Concentrations in W/M Organic Medium of the Plugged Reactor ............................ 15 4-4. Distribution of Hydraulic Head within the W/M Organic Medium in the Plugged Reactor ...... 16 4-5. Laboratory Assembly for Long Duration Tests of Hydraulic Conductivity for the 0.8/0.2

and 0.5/0.5 W/M Organic Media Configured for Upward-Vertical and Horizontal Flows........ 16 4-6. Results of the Long Duration Tests of Hydraulic Conductivity for the 0.8/0.2 and

0.5/0.5 W/M Organic Media Configured for Upward-Vertical Flow......................................... 17 4-7. Results of the Long Duration Tests of Hydraulic Conductivity for the 0.8/0.2 and

0.5/0.5 W/M Organic Media Configured for Horizontal Flow................................................... 17 4-8. Conceptual Picture of the RC Installed at the Mine Site ............................................................ 18 4-9. Conceptual Three-Dimensional Drawings Showing Main Components of the RC.................... 18 4-10. Conceptual Cross Section of the RC with Respective Details .................................................... 19 4-11. Bag with Walnut Shells and Manure .......................................................................................... 19 4-12. Decrease in Sulfate Concentration for Bucket Bioreactors......................................................... 20

vi

Figures (Cont'd) Page

4-13. Sulfate Reduction Rates.............................................................................................................. 20 4-14. Dissolved Zn Trends for Bucket Bioreactors.............................................................................. 21 4-15. Dissolved Cu Trends for Bucket Bioreactors ............................................................................. 21 4-16. Dissolved Cd Trends for Bucket Bioreactors ............................................................................. 22 4-17. pH Trends for Bucket Bioreactors .............................................................................................. 22 4-18. Dissolved K Trends for Bucket Bioreactors ............................................................................... 23 4-19. Dissolved Na Trends for Bucket Bioreactors ............................................................................. 23 5-1. Chart I: Treatment Scope Selection and Preliminary Design of the SRB System..................... 38 5-2. Chart II: PHREEQCI Modeling and the Final Design of the SRB Treatment System................. 39 5-3. Worksheet 0: Input and Output of the BEST Simulator ............................................................ 40 5-4. Worksheet A: Stoichiometric Estimation for Carbon Needed for the Given AMD .................. 41 5-5. Worksheet B1: Number of RCs in the SRB Treatment System for Metals Precipitating as

Sulfides ....................................................................................................................................... 42 5-6. Worksheet B2: Number of RCs in the SRB Treatment System for all Metals Precipitating...... 43 5-7. Worksheet C: Cost Estimate for Adaptation of a Typical Tank ................................................. 44 5-8. Worksheet D1: Preliminary Cost for the SRB Treatment System for Metals Precipitating as

Sulfides ....................................................................................................................................... 45 5-9. Worksheet D2: Preliminary Cost for the SRB Treatment System for all Metals Precipitating . 46 5-10. Worksheet E: Calculation of SRR ............................................................................................. 47 5-11. Worksheet F: PHREEQCI Input File............................................................................................ 48 5-12. Worksheet G: Number of RCs in the SRB Treatment System Based on PHREEQCI ................. 49 5-13. Worksheet H: Cost for the SRB Treatment System Designed using PHREEQCI ....................... 50 5-14. Worksheet I: Number of RCs in the SRB Treatment System Designed using PHREEQCI

with Carbon Adjusted ................................................................................................................. 51 5-15. Worksheet J: Cost for the SRB Treatment System Designed using PHREEQCI with Carbon

Oxidation Adjusted ..................................................................................................................... 52 5-16. Worksheet K: Seepage Velocity and Resident Time for SRB Treatment System with the

Increased Number of RCs to Meet the Velocity and Residence Time Criteria .......................... 53 5-17. Worksheet L: Cost for the SRB Treatment System with the Correction for Flow Velocity...... 54 5-18. PHREEQCI Selected Output File (R-V example.out) Saved as ExcelTM Worksheet.................... 55 5-19. Results of PHREEQCI Modeling with Different Ratios of Carbon Oxidation and the Feed pH .. 56

Tables Page

4-1. Results of the Initial Hydraulic Conductivity Tests.................................................................... 24 4-2. Summary of Hydraulic Conductivity Testing............................................................................. 25 4-3. Target Composition of Synthetic AMDs and Chemical Compounds used for their Production 25 4-4. Decrease of Sulfate Concentration and Sulfate Reduction Rates ............................................... 26 4-5. Dissolved Metal Concentrations and pH for Bucket Reactor ..................................................... 27 6-1. Data Quality Indicator Objectives .............................................................................................. 58 6-2. Summary of Qualified Data for MWTP Activity III, Project 24 ................................................ 59 6-3. Comparison of Results for Duplicate Column Experiments....................................................... 59

vii

Acronyms and Abbreviations ADS advanced drainage system AES atomic emission spectroscopy Al aluminum AMD acid mine drainage ASCII alphanumeric file ASTM American Society for Testing and Materials BEST bioreactor economics, size, and time of operation simulator Ca calcium Cd cadmium cm/s centimeter per second Cu copper DOE U.S. Department of Energy DS decrease in sulfate concentrations EH oxidation-reduction potential expressed with reference to standard hydrogen electrode EPA U.S. Environmental Protection Agency Fe iron FeS2 iron disulfide ft/d feet per day g/cm3 grams per cubic centimeter gpm gallons per minute H2S hydrogen sulfide HDPE high-density polyethylene IAG Interagency Agreement ICP inductively coupled plasma I-O input and output K potassium L liter LOI loss on ignition MCL maximum contaminant level MDL method detection limit Mg magnesium mg milligram mg/L milligram per liter mL milliliter mL/min milliliter per minute mmol millimole mmol/L millimole per liter mmol/(d*L) millimole per day per liter mmol/(d*m) millimole per day per meter Mn manganese mol/(d*m3) mole per day per cubic meter MS Microsoft MSE MSE Technology Applications, Inc. MWTP Mine Waste Technology Program Na sodium NPV net present value

viii

Acronyms and Abbreviations (Cont'd) O&M operation and maintenance p porosity of the medium in a 5-gallon reactor pE oxidation-reduction potential expressed as negative logarithm of electron activity,

conversion: pE = 16.9 EH (volts) pH measure of hydrogen ion activity PHREEQCI pH, redox, equilibrium interactive geochemical modeling program PVC polyvinyl chloride Q flow rate for a 5-gallon reactor QA/QC quality assurance and quality control QAPP quality assurance project plan RC replaceable cartridge RPD relative percent difference S sulfur SMCL suggested maximum contaminant level SO4 sulfate SRB sulfate-reducing bacteria SRR sulfate reduction rate TOC total organic carbon TM trademark V volume of the medium in a 5-gallon reactor W/M walnut shells and cow manure Zn zinc

ix

Acknowledgments This document was prepared by MSE Technology Applications, Inc. (MSE) for the U.S. Environmental Protection Agency’s (EPA) Mine Waste Technology Program (MWTP) and the U.S. Department of Energy’s (DOE) Savannah River Operations Office. Ms. Diana Bless is EPA’s MWTP Project Officer, while Mr. Gene Ashby is DOE’s Technical Program Officer. Ms. Helen Joyce is MSE’s MWTP Program Manager.

x

Executive Summary This document is a final report for Mine Waste Technology Program (MWTP) Activity III, Project 24, Improvements in Engineered Bioremediation of Acid Mine Drainage. The MWTP is funded by the U.S. Environmental Protection Agency (EPA) and jointly administered by EPA and the U.S. Department of Energy (DOE) through an Interagency Agreement and under DOE contract number DE-AC09-96EW96405. MSE Technology Applications, Inc. (MSE) implements the MWTP. Investigations reported in this document focus on the improvements of engineered features and the predictability of a passive technology that could be used for remediation of thousands of abandoned mine sites existing in the Western United States. This passive remedial technology, a sulfate-reducing bacteria (SRB) bioreactor, takes advantage of the ability of SRB that, if supplied with a source of organic carbon, can increase pH and alkalinity of the water and immobilize metals by precipitating them as metal sulfides or hydroxides. The remoteness of acid mine drainage (AMD) sites, their abundance, and related economical aspects require that the design of an SRB bioreactor is simple and relatively inexpensive and that the bioreactor is capable of treating any AMD flow rate. Therefore, bioreactors need to be prefabricated and designed to a size that allows for transportation using backcountry roads in mountain regions. These conditions require that the design of the bioreactor be modular so the treatment system can be assembled at the mine site and consists of the number of modules as required by the AMD flow rate and the metals load. This modular configuration of the SRB treatment system needs to support the prime functional aspects of a bioreactor such as high permeability, ample supply of organic carbon, ability to maintain anaerobic conditions, capacity to accumulate precipitated metals, and means for their periodical removal, if needed. In addition, the configuration of the bioreactor should allow for an easy replacement of organic carbon, if needed. Obviously, the design and sizing of such a treatment system needs to be streamlined and its performance predictable. Therefore, the two main objectives of the project were:

• developing an SRB bioreactor whose reactive material (organic matter) would not be prone to plugging and, if exhausted, would be easy to replace; and

• quantifying reactivity of organic material and developing a simulator that would facilitate optimization of the design parameters and the size of the SRB treatment system.

These project objectives were achieved through the implementation of the following work tasks:

• selection of the medium with organic carbon; • design of an organic carbon replaceable cartridge (RC); and • development of a computer simulator to design and size an SRB treatment system for the given

dissolved metal concentration in AMD and its flow rate. Selection of the medium with organic carbon was accomplished through a literature search. All information gathered during the literature search is contained in the database assembled using Microsoft (MS) AccessTM. The report on this investigation (MSE report # MWTP-188) includes the list and references regarding substrate mixture components used in SRB treatment systems and their effectiveness. The report identified 36 organic substrates and lists the main conditions that need to be considered for the selection of the most appropriate organic medium. Based on the results of this search, a new organic substrate, a mix of walnut shells and cow manure, was developed and selected for the project. Some advantages of using this mix are listed below.

ES-1

Cow manure is an easily biodegradable organic matter that ensures a quick startup of the bioreactor.

Cow manure includes nitrogen needed by other microorganisms for the initial decomposition of manure. Moreover, the nitrogen is in the form of ammonium that is easier for microorganisms to use than nitrates.

Walnut shells are more recalcitrant to biodegradation, thus supporting good long-term operation of a bioreactor.

Walnut shells provide a solid matrix structure because individual shells actually rest on each other. This structure prevents time-driven compaction (settling), thus works toward preservation of the initial permeability of the medium.

Walnut shells contain a high percentage (56%) of total organic carbon (TOC). The TOC of manure is lower and varies, depending on the manure source, from 8% to 20%.

The sustainable hydraulic conductivity of the mixture is 0.01 centimeter per second (cm/s) or higher, based on the results of experiments conducted for the project.

The mix of walnut shells and cow manure is referred to in this document as W/M organic medium. A ratio value that often follows or precedes this term is the ratio of a bulk volume of walnut shells to the bulk volume of manure used for the given mix. The convention used in this document is to express these ratios as decimal fractions of the bulk volume of walnut shells, e.g., 0.4, and the bulk volume of manure, e.g., 0.6, used for the mix before they were combined. A modular SRB treatment system consists of a number of RCs that are configured in parallel or in series depending on the AMD flow rate and its quality (metal load and pH), cleanup objectives, and space available at the given mine site. These RCs are filled with the W/M organic medium of the selected volumetric ratio of walnut shells and cow manure. The number of RCs and the system configuration is determined through modeling conducted using the bioreactor economics, size and time of operation (BEST) computer simulator developed for this project. The investigations leading to the recommended design of the RC included testing of the W/M organic medium for its permeability and sulfate reduction rates (SRR). The permeability tests were conducted in two configurations: (1) an upward vertical flow and (2) a horizontal flow. Several long-duration permeability tests, in the field and under laboratory conditions, were conducted to determine adequacy of each configuration. Results of these tests indicate that the long-term permeability of this medium is significantly higher for flow in a horizontal plane. This phenomenon is attributed to the deformation of the W/M organic medium in which the finest particles are mobilized by the flowing water and migrate downward by gravity to settle at a certain level, usually at the bottom of the container, blocking the flow. In the case of a horizontal configuration, the migrating particles also settle in the bottom of the container; however, they do not block the entry of water that flows above them as it is fed laterally. The experiments conducted showed that for a horizontal flow configuration the sustainable hydraulic conductivity (K) of the mixture is 0.01 cm/s or higher. In general, the hydraulic conductivity value of the 0.5/0.5 W/M organic medium was 1 order of magnitude smaller than the K value for the 0.8/0.2 mix.

ES-2

The laboratory work to determine SRR included six experiments conducted for two synthetic compositions of AMD at three temperatures [i.e., 44 °F (6.7 °C), 58 °F (14.5 °C), and 77 °F (25 °C)]. The two synthetic AMDs, referred to in this document as medium and strong, had pH values of 4.2 and 2.6, respectively. This laboratory experiment, conducted in 5-gallon bucket bioreactors, lasted 5 months and was monitored monthly. The SRR values ranged from 0.17 mole per day per cubic meter [mol/(d*m3)] to 0.79 mol/(d*m3) with the overall mean value of 0.40 mol/(d*m3). Sulfate reduction rate values in these experiments seemed to be independent of the strength of the influent and the temperature at which the experiments were conducted. The large range for SRR values indicates the need for conducting the experiment at least in triplicate. The recommended design of the RC uses a commercially available cylindrical or cuboidal plastic tank most often constructed of high-density polyethylene (HDPE) or polypropylene. Such a tank needs to be equipped with necessary features to accommodate the W/M organic medium and serve as one SRB RC. These modifications will be made in a machine shop, and the tank will then be transported to a mine site. The tank will be installed either aboveground or belowground at the mine site, as required by the site conditions. An appropriate piping system will convey the AMD into the RC. Figure ES-1 included below illustrates a cylindrical RC that could be installed below or aboveground at the mine site.

Figure ES-1. Illustration of a cylindrical RC.

The 5-gallon bags with W/M 0.8/0.2 organic medium shown in this figure may be prepared in advance and then transported to the mine site, or they may be made at the mine site. The bags, which are made of plastic netting that is commonly used by grocery shops for prepacked fruits, have loops in their top portion to facilitate the placement and the removal of the bags from the RC using a rod with a hook. A plastic tarp (not shown in the picture) placed on the top of the bags creates anaerobic conditions. The cost of production and installation (excluding transportation to the site) of such an RC housed in a 2,500-gallon HDPE tank is $8,081. The cost may vary depending on local supply and labor rates applicable at the given location.

ES-3

An SRB treatment system usually includes RC, configured in parallel. However, for a site with a low flow rate but high metals load, the RCs may be configured in series. Both the configuration and the number of RCs are determined using the BEST simulator. This simulator is a spreadsheet-based model that is used in conjunction with a public domain computer software package, PHREEQCI geochemical modeling program. While PHREEQCI calculates geochemical equilibrium for the advective-reactive transport of AMD through the bioreactor, the spreadsheet portion of the simulator handles issues of AMD flow rate, size of the bioreactor, its operational time, and its economics. In general, the BEST simulation process is based on the chemical composition of the AMD and its flow rate, TOC content in the organic matter, cost of material and production of a typical RC, the SRR of the organic matter used in the treatment system, and the discount rate and operation and maintenance (O&M) cost for calculation of the net present value (NPV). The BEST simulator was developed and formulated so that a user with minimum modeling experience can operate it. The BEST simulator operation requires basic knowledge of the ExcelTM program and some familiarity with the geochemical model PHREEQCI. Of course, a good chemical background is a bonus. Within this report, the BEST simulator is saved as a Microsoft ExcelTM workbook, BEST V1.xls, and consists of 17 worksheets. Two of these worksheets (I and II) include charts showing the navigation between the 14 worksheets that are identified with letters A through L, and numbers 1 and 2 for the worksheet series B and D, and their interaction with the PHREEQCI model and its input file. Another worksheet (0) entitled "input and output" (I-O) allows for entering the majority of input data and having the most important results also printed on the same page. However, details of the design specification of the material, etc., are not listed in the I-O worksheet, and the user needs to refer to worksheets A through L to examine these details. Most worksheets are linked together, i.e., any change of input data causes appropriate changes of the results calculated by the respective worksheet. However, the PHREEQCI model and its data input file are not automatically linked with the rest of the worksheets, thus required changes need to be input manually. The time of operation calculated by BEST is based on the available carbon present in the W/M organic medium divided by the safety factor of 4. This safety factor is used because the investigations conducted for the project did not focus on confirmation of whether the organic carbon present in the medium is entirely available for the SRB. An example simulation provided in the report considers AMD flowing at 1 gallon per minute (gpm) and laden with 17.78 milligrams per liter (mg/L), 6.12 mg/L, 0.08 mg/L, and 40.4 mg/L of zinc (Zn), copper (Cu), cadmium (Cd), and aluminum (Al), respectively. An SRB treatment system to remove Zn, Cu, and Cd as sulfides would require three RCs and the capital cost of $24,244. The NPV is $37,768 based on a discount rate of 3.2%, O&M at $1,000/year, and the operational time of 18 years. Had the goal for the SRB treatment system been the removal of all the metals including Al that precipitates as hydroxide, the system would have included 17 RCs at a capital cost of $137,385 and an NPV of $156,488 for the system to operate for 39 years.

ES-4

1. Introduction This document is a final report for Mine Waste Technology Program (MWTP) Activity III, Project 24, Improvements in Engineered Bioremediation of Acid Mine Drainage. The MWTP is funded by the U.S. Environmental Protection Agency (EPA) and jointly administered by the EPA and the U.S. Department of Energy (DOE) through an Interagency Agreement and under DOE contract number DE-AC22-96EW96405. MSE Technology Applications, Inc. (MSE) implements the MWTP. 1.1 Background Acid mine drainage (AMD) emanates from many abandoned mines in the Western United States, causing significant environmental problems by contaminating surface waters and groundwater with dissolved metals and raising their acidity. Conventional treatment of AMD is often not feasible due to the remoteness of the site, lack of power, and limited site accessibility. Thus, for such sites, there is a need for a passive remedial technology to immobilize metals and increase the pH of the AMD. Sulfate-reducing bacteria (SRB) have the ability to increase pH and alkalinity of the water and immobilize dissolved metals by precipitating them as metal sulfides, provided that a favorable biochemical environment is created (Ref.s. 1 and 2). This technology was investigated by the MWTP through a field demonstration of SRB bioreactors placed at two abandoned mine sites. One SRB bioreactor (bioreactor) was placed in the mineshaft of the Lilly/Orphan Boy Mine near Helena, Montana. The AMD in this bioreactor was flowing vertically, upwards through the reactive medium placed on shelves suspended in the shaft. At another abandoned mine site, the Calliope site near Butte, Montana, three bioreactors were placed at the land surface

(Ref. 3). These bioreactors, fed by the AMD flowing out of the old mine adit, were configured for horizontal flow. Both demonstrations were successful, as indicated by the metal removal rates and an increase of the AMD pH, but they revealed the necessity of further investigation that would address the appropriate design and sizing of the bioreactor and the need for developing a typical bioreactor that could be easily brought and replaced at a given abandoned mine site to treat AMD for the predicted period of time. The objective of Project 24 was to address these issues. 1.2 Principles of the Sulfate-Reducing Bacteria Technology Acid mine drainage is a typical result of mining sulfide-rich ore bodies. Acid mine water is formed when sulfide-bearing minerals, particularly pyrite [iron disulfide (FeS2)], are exposed to oxygen and water as described by the following overall reaction (Equation 1-1). FeS2 + 15/4 O2 + 7/2 H2O --->

Fe(OH)3 + 2SO42- + 4H+ (1-1)

This reaction results in increased acidity of the water (lowered pH), increased metal mobility, and the formation of sulfate. When provided with an organic carbon source, SRB are capable of reducing the sulfate to soluble sulfide by using sulfate as a terminal electron acceptor. Acetate and bicarbonate ions are also produced. The soluble sulfide reacts with the metals in AMD to form insoluble metal sulfides (Equations 1-2 and 1-3). The bicarbonate ions increase pH and alkalinity of the water.

SO42- + 2CH2O -------> H2S + 2HCO3

- (1-2) H2S + M2+ ---> MS + 2H+, where M = metal (1-3)

1

Organic carbon, the organic electron donor, is represented in Equation 1-2 by the formula 2CH2O, may be provided either by feeding a bioreactor with a chemical compound like lactate or methanol that delivers carbon directly or can be obtained from a selected organic matter that, if not

used for this purpose, may be classified as waste. Because of the remoteness of many abandoned mine sites, the latter option is more appealing as it requires less maintenance and does not pose the risk of the misuse of methanol by irresponsible parties.

2

2. Project Objectives The lifespan of a properly designed SRB passive bioreactor and its capacity to remove metals depends on the conditions given below. The chemical properties of AMD, mainly

pH and metals load. The higher the metal concentration for the same flow rate, the larger the reactor and often the shorter the lifespan of the bioreactor. As the pH of AMD decreases, the bioreactor must become more complex and larger.

Volume of the AMD treated (flow rate). As the flow rate of the treated AMD increases, the bioreactor must become larger.

Longevity of the organic carbon supply system. When the source of organic carbon is depleted or becomes unavailable because of settling processes or physical or chemical encapsulation, the bioreactor will cease operation. To reactivate such a bioreactor, the organic carbon source has to be replenished. It is thus desirable to maximize the time interval between such operations and to design the bioreactor so the replenishment of organic matter is technically easy without breaking or dismantling the bioreactor.

Capacity of the bioreactor to accumulate precipitated metal sulfides and hydroxides or the ability to periodically clean the bioreactor. As the amorphous metal sulfides form during the treatment process, they need room for settling or they will be removed from the bioreactor with the effluent (an undesirable effect).

Preservation of the organic matter bioreactor permeability. A decrease of the permeability of the bioreactor may limit the flow rate for the treated AMD and may cause physical encapsulation of organic

carbon, thus making it unavailable for the sulfate reduction process.

The remoteness of AMD sites, their abundance, and the related economical aspects require that the design of a bioreactor be simple and relatively inexpensive and that the bioreactor be capable of treating any AMD flow rate and the metals load. Therefore, bioreactors need to be prefabricated and designed to a size that will withstand the hardships of transportation using backcountry roads in mountain regions. These conditions require that the design of the bioreactor be modular so the treatment system can be assembled at the mine site and consists of the number of modules as required by the AMD flow rate. In summary, the two main objectives of MWTP Activity III, Project 24 were to:

− develop an SRB bioreactor whose reactive material (organic matter) will not be prone to plugging and, if exhausted, will be easy to replace; and

− quantify reactivity of organic material and develop a computer simulator that would facilitate optimization of the design parameters and the size of the SRB treatment system.

These project objectives were achieved through the implementation of the following work tasks:

− selection of the medium with organic carbon; − design of an organic carbon replaceable

cartridge (RC); and − development of a computer simulator to

design and size an SRB treatment system for a given dissolved metal concentration in AMD and its flow rate.

3

3. Selection of the Medium with Organic Carbon This task, which was to determine the effectiveness of various organic substrates (matters) used in systems designed to treat AMD, was accomplished through a literature search. All information gathered during the literature search is contained in the database assembled using Microsoft (MS) AccessTM. The report on this investigation (Ref. 4) was completed and can be found in Appendix A of this document. The report includes the list and references regarding substrate mixture components reportedly used in SRB treatment systems and their effectiveness. The following section includes a brief description of main findings of the literature search and the description of the selected medium. 3.1 Literature Search Results More than 90 publications that dealt with the use of organic substrates as mixtures for SRB-mediated treatment of AMD were identified and entered in MS AccessTM database. These publications identified 36 organic substrates that included 7 direct and 29 indirect substrates. The direct substrates are those that do not require decomposition by other microorganisms to provide SRB nutrition. Such substrates include:

− alcohols, e.g., methanol and ethanol; − organic acids, e.g., acetate, lactate, formate,

and pyruvate; and − sugars, e.g., sucrose.

Indirect substrates are those requiring decomposition by other microorganisms to provide SRB nutrition. These substrates require complex microbial communities to degrade the organic matter and support SRB growth. The publications examined reported quite a variety of such substrates. They can be classified as:

− composts, e.g., spent mushrooms, leaf; − wood/paper wastes, e.g., sawdust, leaf

mulch, wood chips;

− food production byproducts, e.g., molasses, cheese whey, potato processing waste;

− agricultural products, e.g., hay, straw; − manure, e.g., cow, horse, dried poultry

waste; and − sewage, e.g., digested sludge, sewage

sludge. The use of direct substrates promises to allow more stringent control of biofouling but requires more complicated reactor design and may not be suitable for remote mine sites. The use of some direct substrates, such as ethanol, at remote mine sites is also complicated by public safety concerns. Indirect substrates are more feasible than direct substrates for low maintenance systems at remote mine sites requiring more long-term operation. The choice of an effective substrate mixture is dependent on the composition of the AMD and the types of substrates available at low cost in the vicinity of the system installation. Overall, substrate mixture containing both easily biodegradable materials and more recalcitrant materials are the most effective for supporting SRB growth. The easily biodegradable substrate ensures a quick start of a bioreactor. More recalcitrant materials provide the best long-term bioreactor performance. The substrate mixture should also provide adequate surface area for biofilm development, buffering and adsorption capacity, and adequate hydraulic conductivity. The suitability of a substrate mixture for treating a particular composition of AMD is best determined empirically using laboratory-scale tests. Overall, the literature search indicates that a wide range of organic substrate materials can be used to effectively treat AMD using SRB technology. 3.2 Substrate Selection for the Project A mix of walnut shells and cow manure was selected for the project. This indirect organic substrate satisfies most of the conditions defined

4

through the literature study (Section 3.1) for an efficient mixture to treat AMD. The advantages of using this mix are listed below. Cow manure is an easily biodegradable

organic matter that ensures a quick startup of the bioreactor.

Cow manure includes nitrogen needed by other microorganisms for the initial decomposition of manure. Moreover, the nitrogen is in the form of ammonium that is easier for microorganisms to use than nitrates.

Walnut shells are more recalcitrant to biodegradation, thus they will support good long-term operation of a bioreactor.

Walnut shells provide a solid matrix structure because individual shells actually rest on each other. This structure prevents time-driven compaction (settling) and thus works toward preservation of the initial permeability of the medium.

The concave shapes of walnut shells provide "housing" for manure, thus preventing the bulk portion of manure from migrating down to the bottom of the bioreactor. Therefore, this easily biodegradable organic matter remains evenly distributed in the bioreactor.

As determined through the analytical work conducted for the project, walnut shells contain a high percentage (56%) of total organic carbon (TOC). The TOC of manure is

lower and varies, depending on the manure source, from 8% to 20%.

This nonsettling medium, with even distribution of manure and walnut shells throughout the reactive cell of a bioreactor, provides a large contact area of the organic carbon with AMD and its dissolved sulfate that needs to be reduced to form insoluble metal sulfides.

The sustainable hydraulic conductivity of the mixture is 0.01 centimeter per second (cm/s) or higher, based on the results of experiments conducted for the project.

Although walnut shells are not available in Montana and need to be shipped from out of state (California for this project) their price of $54 per cubic yard (including shipment) is only 2.7 times more expensive than that for cow manure. This difference is not important considering that the reactive material, including walnut shells, comprises less than 8% of the bioreactor capital cost (see Section 5, Figures 5-8 and 5-9).

The mix of walnut shells and cow manure is referred to in this document as W/M organic medium. A ratio value that often follows or precedes this term is a ratio of a bulk volume of walnut shells to the bulk volume of manure used for the given mix. Section 4 includes a description and the results of tests conducted on the W/M organic medium.

5

4. Replaceable Cartridge A modular SRB treatment system consists of a number of RCs that are configured in parallel or in series, depending on the AMD flow rate and its quality (metal load and pH), cleanup objectives, and space available at the given mine site. These RCs are filled with the W/M organic medium of the selected volumetric ratio of walnut shells and cow manure. The number of RCs and the system configuration is determined through modeling conducted using the bioreactor economics, size, and time of operation (BEST) computer simulator detailed in Section 5. This section includes results of RCs permeability tests, a technical specification for an individual RC, description of the laboratory experiment to determine sulfate reduction rates (SRRs), and pertinent analytical results. 4.1 Engineering Features 4.1.1 Flow Configuration Two configurations of flow for an individual RC were considered: (1) an upward vertical flow and (2) a horizontal flow. Several long-duration permeability tests were conducted to determine adequacy of each configuration for the W/M organic medium. Results of these tests indicate that the long-term permeability of this medium is significantly higher for flow in a horizontal plane. Details of these tests are presented in the following sections. 4.1.1.1 Permeability Testing Hydraulic conductivity tests of the W/M organic medium included mixes with different ratios of walnut shells to manure. The convention used in this document is to express these ratios as decimal fractions of the bulk volume of walnut shells, e.g., 0.4, and the bulk volume of manure, e.g., 0.6, used for the mix before they were combined. The following permeability tests were conducted to determine hydraulic conductivity of the W/M organic medium (in chronological order):

− initial tests to determine the original hydraulic conductivity of the W/M organic medium with W/M mixes ranging from 0.33/0.67 to 0.5/0/5;

− long-duration field test for hydraulic conductivity for the 0.5/0.5 W/M organic medium configured for upward vertical flow;

− long-duration laboratory tests for hydraulic conductivity of 0.5/0.5 and 0.8/0.2 W/M organic media configured for upward-vertical and horizontal flows; and

− long-duration field test for hydraulic conductivity of the 0.8/0.2 W/M organic medium configured for upward vertical flow.

4.1.1.1.1 Initial Tests for Hydraulic Conductivity Most of the tests for initial hydraulic conductivity of the W/M organic medium were conducted in 6-inch-diameter polyvinyl chloride (PVC) columns filled with an approximately 4-foot-high layer of W/M organic medium. Eight of the tests were conducted in 10-inch-diameter advanced drainage system (ADS) pipes filled with an approximately 3.5-foot-high layer of W/M organic medium. The 10-inch-diameter ADS pipes were used because such pipes were included in the preliminary design of the RCs at that phase of the investigation. All columns were in a vertical position with tap water flowing downward. These tests were conducted following the procedure recommended in American Society for Testing and Materials (ASTM) D2434-68 (Ref. 5). Because, the preliminary design of the RC focused, at that stage of the project, on the W/M organic medium placed in bags and housed in vertical pipes, most of the initial tests for hydraulic conductivity determination was conducted as couples, with one test run on the W/M organic medium contained in bags made of plastic mesh and another test performed on the W/M organic medium placed directly in a vertical pipe. Such a

6

procedure was applied to examine the impact of plastic netting on the hydraulic conductivity measurements. The W/M organic medium was compacted after placing it in the column to prevent the development of a preferential flow along the wall of the column. The compaction ratio varied from 13% to 27% with respect to the original length of the W/M organic medium placed in the test column. After the flow and the hydraulic head equilibrated for the given tested column, the hydraulic conductivity was measured at least three consecutive times, and the result reported in Table 4-1 is the average of these measurements. The results obtained indicate that hydraulic conductivity values ranged from 0.0069 cm/s to 1.81 cm/s. The lowest value was obtained for W/M organic medium consisting of the lowest ratio of walnut shells, i.e., for the mix of 0.33/0.67. The highest value was calculated for one of the columns packed with the of 0.5/0.5 mix. The mean hydraulic conductivity value was 0.354 cm/s and the standard deviation was 0.383 cm/s. 4.1.1.1.2 Long Duration Field Test for Hydraulic Conductivity of the 0.5/0.5 W/M Organic Medium with Vertical Flow Configuration The long duration field test for hydraulic conductivity of 0.5/0.5 W/M organic medium with upward vertical flow configuration was conducted using a concrete sump saved for this purpose at the Calliope site (Ref. 3). That sump was equipped (for the reported project) with sixteen 9-foot-long 10-inch-diameter high-density polyethylene (HDPE) corrugated pipes as shown in Figure 4-1. Fifteen of these pipes were filled with the 0.5/0.5 W/M organic medium, and one pipe was left empty to measure the hydraulic head at the bottom of the system. The pipes were perforated in their lowest 1-foot portion to allow for inflow of the AMD and filled with a 16-inch-tall bag of walnut shells followed by an 88-inch-tall column of the 0.5/0.5 W/M organic medium bagged in four sacks. The space between the pipes was filled with an 18-inch-thick layer of walnut shells (placed at the sump bottom) followed by a 6-inch-

thick layer of fine gravel, 18-inch-thick layer of sand, and 54-inch-thick layer of bentonite. Acid mine drainage was fed through the inlet pipe located in the side wall of the sump 6 inches above the bottom of the sump. This 90-day test was conducted using the AMD emanating from the abandoned Calliope mine and feeding the sump initially at the rate of 0.21 gallon per minute (gpm). This flow rate declined with time as the overall hydraulic conductivity of the medium in the column was decreasing. The test documented a radical decline of the hydraulic conductivity value for this flow configuration from the original value of 0.022 cm/s to 0.002 cm/s after only 10 days of flow and to 0.0005 cm/s at the end of the 90-day test (Figure 4-2). After the test was completed, the bags with W/M organic medium from one of the test columns (column No. 2) were removed, visually inspected, and sampled for chemical analysis of aluminum (Al) and iron (Fe) concentrations in the solid matrix. These analyses were performed to investigate the possibility of the Al and Fe hydroxides precipitating within the hydraulic-upgradient portion of the W/M organic medium and blocking the flow similarly to the occurrence found during the autopsy of the Calliope bioreactor and reported by MSE in 2002 (Ref. 3). The results show no pattern of an increasing concentration of Al and Fe in the W/M organic medium towards the bottom of the column (Figure 4-3). However, the visual inspection of the W/M organic medium showed a slightly higher content of fine particles within the bottom portion of W/M organic medium. This indicates the possibility that some small particles of manure migrated downward by gravity and formed a less permeable layer in the lower portion of the column. Because conducting particle-size analysis for this W/M organic medium was not feasible, the distribution of hydraulic head along the flow through the W/M organic medium filling two other corrugated pipes (columns No. 3 and 9) of the tested sump was examined using 0.5-inch-diameter metal pipe

7

perforated within its lowest 2-inch portion. This pipe was equipped with a conical tip placed below the perforation. The pipe was pushed incrementally into the W/M organic medium, and the water level inside the pipe was recorded. Results of this examination showed that the hydraulic head declined rapidly in the lowest portion of the W/M organic medium (Figure 4-4). A dotted line in Figure 4-4 depicts an expected distribution of the hydraulic head for a homogeneous medium that is not plugged. This investigation, together with the chemical analysis shown in Figure 4-3, confirms the conclusion that small particles migrate to the bottom of the W/M organic medium and plug the reactor. This discovery made it clear that the selected W/M organic medium deforms when subjected to the flow of liquid and, therefore, the direction of the flow needs to be parallel to the surface where the smallest particles of the W/M organic medium settle. 4.1.1.1.3 Long Duration Laboratory Tests for Hydraulic Conductivity of the 0.5/0.5 and 0.8/0.2 W/M Organic Media Configured for Upward Vertical and Horizontal Flows To test the impact of the flow orientation on long-term hydraulic conductivity of the W/M organic medium, long duration laboratory tests were performed. These tests were conducted in four 6-inch-diameter, 3.5-foot-long PVC pipes oriented in vertical and horizontal directions. There were two horizontal and two vertical pipes with the latter flowing upward. Each pair included one pipe filled with the 0.5/0.5 W/M organic medium and another pipe filled with 0.8/0.2 W/M organic medium. All pipes were fed with tap water delivered to the test system by a peristaltic pump equipped with four heads, hence providing a separate water line to each test pipe (Figure 4-5). The flow rate of the peristaltic pump was set at 23 milliliters per minute (mL/min) through each line. This flow rate, if accepted by a column, would mimic a 12-hour residence time maintained in the bioreactors at the Calliope site (Ref. 3). The feed

from the peristaltic pipe accumulated in an intake vessel positioned above the test pipe and equipped with an overflow outlet and a piezometric tube to record the upgradient hydraulic head. The overflow outlet was installed to remove water from the system if it was not accepted by the column at the rate of 23 mL/min. Water from this vessel flowed through the test pipe to a receiving vessel similarly equipped with a tube to measure hydraulic head downgradient from the test pipe, thus providing information about the hydraulic head difference. The actual flow rate from the receiving vessel was also measured. This measurement was especially necessary when the intake vessel was overflowing due to the reduction of hydraulic conductivity. The test was conducted for 96 days. The results of the experiment are given below (Figure 4-6). The 0.5/0.5 W/M organic medium in a vertical

configuration plugged immediately (as in the previously conducted field test).

The 0.8/0.2 W/M organic medium in a vertical configuration plugged after 12 days.

Highest hydraulic conductivity for the laboratory experiment was for the 0.8/0.2 W/M organic medium in the horizontal configuration. The plugging on day 83 of operation was not associated with the plugging of the W/M organic medium but with the plugging of a small-diameter water supply tube. After the tube was cleaned, the hydraulic conductivity value returned to the value of 0.005 cm/s.

On termination of the experiment, the test pipes were dismantled, and their interior was visually examined. It was discovered that the inlet and outlet fittings of the test pipes were partly plugged, which indicates that part of the plugging syndrome can be attributed to the plugging of the water delivery system rather than plugging of the W/M organic medium. Thus, it is concluded that the actual long-term hydraulic conductivity for the 0.8/0.2 W/M

8

organic medium oriented for the horizontal flow was 0.01 cm/s as indicated by its value after 30 test days. A stipulated hydraulic conductivity value for the 0.5/0.5 W/M organic medium with the horizontal flow is 0.0038 cm/s.

In general, the W/M organic medium configured for a horizontal flow performed superior to the flow in the vertical direction.

4.1.1.1.4 Long Duration Field Test for Hydraulic Conductivity of the 0.8/0.2 W/M Organic Medium Configured for Upward-Vertical Flow The long duration field test for hydraulic conductivity of the 0.8/0.2 W/M organic medium configured for upward-vertical flow was again conducted at the Calliope site. Four bags with 0.8/0.2 W/M organic medium were placed and compacted in one of the previously emptied 10-inch corrugated tubes. The flow was set at 0.018 gpm, which corresponds to 0.27 gpm had all 15 corrugated tubes been used for the test as they were for the 0.5/0.5 W/M organic medium (see Section 4.1.1.1.2). The hydraulic conductivity values for this test are presented in Figure 4-2 and Figure 4-6. Although it appears that the hydraulic conductivity value declined from the highest of 0.024 cm/s to 0.007 cm/s throughout the 98-day-long test, it turned out that most of the test hydraulic head losses were caused by the plug that developed within the valve that controlled the flow to the sump. After the plug was removed, the calculated hydraulic conductivity value was 0.016 cm/s. 4.1.1.1.5 Summary for the Hydraulic Conductivity Testing Table 4-2 includes a summary of the results obtained for W/M organic medium hydraulic conductivity tests that were conducted for the project. The following conclusions may be stated. The hydraulic conductivity value of the

0.5/0.5 W/M organic medium was 1 order of magnitude smaller than the hydraulic conductivity value for the 0.8/02 mix as

shown by the results for the laboratory long duration tests using columns oriented for the horizontal flow.

The long-term hydraulic conductivity value for the flow in a vertical direction decreases very rapidly and may approach 0 value. This decrease is attributed to the deformation of the W/M organic medium of which the finest particles are mobilized by the flowing water and migrate downward by gravity to settle at a certain level, usually the bottom of the container, blocking the flow. In the case of a horizontal configuration, the migrating particles also settle in the bottom of the container, but they do not block the entry of water that flows above them as it is fed laterally.

One reason that the 0.8/0.2 W/M organic medium did not plug during the field test is probably a positive "scale effect" for hydraulic conductivity tests (the larger the cross-sectional area, the smaller the chance that it will become plugged). Another reason is that the 10-inch test column in the field test was hydraulically connected with the large area (the bottom of the 6-foot-diameter sump) filled with highly permeable walnut shells that could accommodate the migrating particles of manure. However, the column used for the laboratory experiment did not have such a reservoir.

The sustainable hydraulic conductivity of the 0.8/0.2 W/M organic medium configured for the horizontal flow is 0.01 cm/s or higher.

The short duration laboratory tests provided the highest hydraulic conductivity values that proved to not be applicable for the deforming W/M organic medium used for these tests.

4.1.2 Replaceable Cartridge Design Replaceable cartridges can be built using commercially available cylindrical or cuboidal HDPE or polypropylene tanks. Such a tank will be equipped with necessary features to accommodate the W/M organic medium and serve

9

as one SRB bioreactor. These modifications can be made in a machine shop and then the tank will be transported to a mine site. At the mine site the tank can be installed, either aboveground or belowground as required by the site conditions. An appropriate piping system can convey the AMD into the RC. Figure 4-8 illustrates a cylindrical RC installed belowground at the mine site. Such installation can take advantage of a small pond where AMD accumulates and then can be piped by gravity to the RC. 4.1.2.1 Typical Tank Adaptation Any tank of a suitable size and shape can be adapted for the RC; its suitability is determined by the mine site conditions, transportation restrictions, availability, price, etc. The RC design presented in this report includes an 8-foot-diameter, 8-foot-tall HDPE or polypropylene tank of the nominal size of 2,500 gallons. The 8-foot-diameter tank was selected for the design because of U.S. Department of Transportation regulations limit the width of the standard load to 8 feet. The engineering design drawing of an SRB RC and the adaptations details are presented in Appendices B1 and B2. Appendix B3 provides a description of tasks for a typical tank adaptation and construction of the SRB RC. Figure 4-9 and Figure 4-10 are conceptual drawings showing main components of the RC. The cost of the adaptation is addressed in Section 5. Main components of the RC are listed below. The round cylindrical tank with its top cut off.

A 4-inch-diameter PVC vertical pipe with perforation. This pipe, located at the inlet portion of the RC, serves as a distribution system to "spread" the AMD throughout the entire height of the RC. This pipe is also a cleanout access for agitating and removing Al and Fe hydroxides that are notorious for precipitating at the inlet of an SRB bioreactor. A jetting tool and/or a suction pump could be used for agitating the precipitate and removing it, respectively. This pipe is referred to in this document as an AMD distribution pipe.

A 10-inch-diameter PVC vertical pipe with perforation. This pipe, located at the outlet of the RC, is a sump for metal sulfides precipitating in the RC. The large diameter of this pipe facilitates easy access to the bottom of the pipe for precipitate removal using a suction pump.

The inlet pipe to convey the AMD from the mine audit, pond, or from another hydraulically upgradient source to the RC. The inlet pipe is equipped with a globe valve to control the flow rate.

The outlet pipe to evacuate treated AMD from the bioreactor. The outlet pipe is located high on the wall of the RC to allow for "deposition" of metal sulfides within the lower portion of the RC.

The overflow pipe, which is connected to the 4-inch pipe, to accommodate excess water in case the hydraulic conductivity of the W/M organic medium decreases beyond the transmitting capacity for the preset flow rate.

The drainpipe, with a valve, is used to drain the RC installed aboveground, if needed.

W/M organic medium bagged in 5-gallon socks of plastic mesh (active W/M organic medium) (Figure 4-11).

Bags (5-gallon) with walnut shells placed around the distribution pipe (Figure 4-10). These bags with material of higher hydraulic conductivity reduce the impact of Al and Fe hydroxides deposition.

A plastic tarp placed on the top of the bags with active W/M organic medium tucked between these bags and the tank walls to create anaerobic conditions in the RC.

Bags (5-gallon) with walnut shells placed above the active W/M organic medium to support the usually dome-shaped cover and prevent it from caving from incidental or vandalistic events.

A bioreactor cover made from the top of the tank that was previously cut off. The lid is fastened to the tank using appropriate latches.

10

Caps for the distribution pipe and the sump to prevent atmospheric oxygen from entering the RC.

A metal culvert (optional) if the soils around a belowground-placed tank are instable and may cause the tank walls to collapse during replacement of the W/M organic medium. (The tank will not collapse when filled with W/M organic medium bags and water.)

4.1.2.2 Organic Media Bags W/M organic media are bagged in a plastic net, e.g., Peacon Bag Co. braided net, which when laid flat is 14 inches wide with 3/8-inch openings (Figure 4-1). This netting material is available in the form of a sleeve. This sleeve is cut to the length of 10 feet, folded to the length of 5 feet to create a double layer sleeve, tied at one end using duct tape to make it a sack, placed in a 5-gallon bucket, filled with appropriate W/M organic medium, and tied at the top in the form of a loop (also using duct tape). The loop is made to facilitate lowering or removing the bag from the RC using a rod with a hook end. 4.1.2.3 Logistics of Construction and Field Installation After the tank adaptation work is completed, the RC tank or tanks (depending on the SRB system design, see Section 5) are transported to the mine site using a flatbed pickup. At the site, the tanks are installed above- or belowground depending on the site topography, winter temperature, feasibility of excavation for the tanks, etc. The AMD is piped to the tanks by gravity, preferably from a detention pond that would create a constant (or close to) hydraulic head differential between the AMD source (the pond) and the RC tank. The SRB treatment system usually includes RCs configured in parallel. However, for a small drainage rate but high metals load, the RCs may be configured in series. Bags with W/M organic medium may be prepared in advance and then transported to the mine site, or they may be made at the mine site. Bags are lowered to an empty RC in layers. After each

layer is completed, the AMD is allowed to flow into the RC and saturate the W/M organic medium and enhance tight packing of the bags. Then, the next layer of bags with the W/M organic medium is placed on top of the previous layer, and again more AMD is introduced to the RC. A total of 494 bags of W/M organic medium and 26 bags of walnut shells are needed to fill up one RC as designed in this document (Figures 4-7, 4-8 and 4-9 and Appendices B1 and B2). Bags with walnut shells are placed in the RC adjacent to the AMD distribution pipe to "envelope" it by the more permeable and porous medium that would facilitate agitating and removal of Al and Fe hydroxides precipitate. Some walnut shells bags are also placed just under the RC lid (Figure 4-9) to protect the lid from caving in. Details on cost and volumes of the organic medium needed are included in Section 5. 4.2 Organic Matter Properties of W/M organic medium were examined through a series of experiments that included:

− laboratory experiment to determine SRR; − chemical analyses for metals; and − chemical analyses for TOC in W/M organic

medium. 4.2.1 Laboratory Experiments for Sulfate Reduction Rate The laboratory work to determine SRR included six experiments conducted using two synthetic compositions of AMD at three temperatures, 44 °F (6.7 °C), 58 °F (14.5 °C), and 77 °F (25 °C). The composition of synthetic AMD, referred to in this document as medium AMD and strong AMD, replicate the AMD that once fed the bioreactors at the Calliope site (Ref. 3). The medium AMD emulates the Calliope AMD composition the summer of 1999, and the strong AMD duplicates the AMD of the lowest pH and highest metal concentration during the time when the Calliope bioreactors were in operation.

11

Table 4-3 includes the target compositions for the synthetic AMD and information about chemical compounds used for their production. Although the tanks with synthetic AMDs were equipped with nitrogen headspace purge to prevent oxidation of the solution by atmospheric oxygen, the AMD composition fluctuated slightly during the 5-month-long experiment. In general, the pH values were 4.2 and 2.6 for the medium and strong AMDs, respectively. The laboratory experiments were conducted in six 5-gallon bucket bioreactors identified by numbers I, II, III, IV, V, and VI. The bioreactors were fed with the synthetic AMD delivered to the test system by two peristaltic pumps equipped with a total of four heads, providing a separate water line to each test pipe. The flow rate from the peristaltic pump was set at 2.7 mL/min through each line, which corresponds to an approximately 1-day residence time in the W/M organic medium in each bucket bioreactor. The bucket bioreactors were filled with a 2-inch layer of No. 6-8 silica sand, (placed at the bottom of a bucket) followed by an 8-inch layer of 0.5/0.5 W/M organic medium and a 0.5-inch-thick layer of No. 6-8 silica sand. The flow was configured vertically upward with the synthetic AMD entering the bucket wall within the lower 2-inch layer of silica sand that was used to distribute the AMD evenly across the bucket. The AMD drained from the mini bioreactor through an overflow pipe whose outlet was located 2 inches above the upper layer of the silica sand. The bucket bioreactors were sealed on the top by a lid equipped with an o-ring. Four bucket bioreactors were fed by the medium strength AMD with two of them operated at 58 °F (bioreactors II and IV) and the other two at 44 °F (bioreactor I) and 77 °F (bioreactor III). Two remaining bucket bioreactors that were fed by the strong AMD operated at 58 °F (bioreactors V and VI). The bucket bioreactors operated for 5 months and were monitored for the following parameters:

− dissolved metals [zinc (Zn), copper (Cu), cadmium (Cd), Al, Fe, manganese (Mn), calcium (Ca), magnesium (Mg), sodium (Na), and potassium (K)], sulfur (S) sulfate, alkalinity, and pH in the influent and effluent monthly; and

− TOC in W/M organic medium at the start and end of the experiment.

4.2.2 Results 4.2.2.1 Sulfate Reduction Rate The calculation of SRR was based on the difference in the sulfate concentrations in the influent and the effluent of the bucket bioreactors. Because the ion chromatography and colorimetric analytical methods provided inconsistent results of sulfate concentrations, the latter were calculated from S concentrations obtained using the inductively coupled plasma (ICP) method and assuming that all dissolved S present was in sulfate form. This is a reasonable assumption because it is likely that any dissolved sulfide present in a sample after filtering would evolve as hydrogen sulfide (H2S) (gas) when nitric acid is added to preserve the sample. Disregarding the first set of measurements, the decrease in the sulfate concentrations (DS) was from 18.6 mg/L to 87 mg/L (Figure 4-12). The DS was calculated as a difference between the influent and effluent concentrations. Negative or abnormally high values of DS that occurred mostly during the first month of the operation are attributed to unstable conditions in the initial period. The SRR values, calculated using equation 4-1 and shown in Figure 4-13 and Table 4-4, ranged from 0.17 mole per day per cubic meter [mol/(d*m3)] to 0.79 mol/(d*m3) with the overall mean value of 0.40 mol/ (d*m3).1 The mean value was calculated disregarding negative SRR values that occurred mostly during the first month of the experiment. The SRR values acquired from the laboratory experiment seemed to be independent of the

1 SRR values expressed in mol/(d*m3) are numerically the same as those expressed in millimole per day per liter [mmol/(d*L)].

12

strength of the influent and the temperature at which the experiment was conducted. The large range for SRR values indicates the need for conducting the experiment at least in triplicates.

pVQDSSRR

**

= (4-1)

Where Q is flow rate in L/d , V is volume of the reactor, and p is porosity of the medium. 4.2.2.2 Performance of the Bucket Bioreactors Analytical results for all dissolved metals concentrations are presented in Table 4-5. Data for selected metals are also shown in figures to facilitate the interpretation of the results. As documented by MSE (Ref. 3), SRB bioreactors effectively remove Zn, Cu, and Cd from AMD. Concentrations of these metals were also reduced in bucket bioreactors as seen in Figure 4-14, Figure 4-15, and Figure 4-16 for Zn, Cu, and Cd, respectively. However, disregarding the first two sampling events, the removal of these metals was more efficient for the medium AMD than for the strong AMD of low pH value. For example, Zn concentration (Figure 4-14) in the medium AMD decreased from the level of 10 mg/L to 4 mg/L and much lower, while for the strong AMD with low pH and Zn concentration of 20 mg/L to 15 mg/L, the removal of Zn was approximately 2 mg/L. It is concluded that this low removal rate is caused by low pH of the strong AMD that (as shown in Figure 4-17 and Table 4-5) increased only by 0.2 unit, i.e., to a pH of 3. For the medium AMD with an original pH of 4.2, the pH increased to 7 at the beginning of the experiment and then decreased to approximately 6. The decreasing trend of the pH for the effluent from the bucket bioreactors that were fed by the medium AMD may be explained by the initial exchange of protons in the solution with alkali and alkaline metals present in the W/M organic medium. This can be deduced, for instance, from Figure 4-18 and Figure 4-19 that depict solution concentrations of K and Na, respectively. Concentrations of these metals in the influent

AMD were lower than in the effluent, thus indicating the removal of these metals from the W/M organic medium, though the rate of the removal was declining with time. Alkaline metals, Ca and Mg, behaved similarly; however, the difference between the inflow and the effluent concentrations was much smaller. The behavior of the alkali and alkaline metals, together with the decreasing trend for pH, may point to the proton exchange. This process, occurring in the initial phase of the bioreactor operation, should not be misconstrued with the bioreactor failing but rather with the bioreactor approaching an equilibrium condition at the lower but long lasting pH. Because the bucket bioreactor experiments lasted only 5 months, these equilibrium conditions were yet to be achieved. 4.2.2.3 Total Organic Carbon Contents Although the experiment with bucket bioreactors was conducted primarily to determine SRR, an attempt was made to determine the TOC in the fresh and spent W/M organic medium. The analyses were conducted using the loss on ignition (LOI 4002) laboratory method. The TOC in the fresh organic matter was 56.5% and 8% by weight for walnut shells and manure, respectively. Therefore, in the 0.5/0.5 W/M organic medium, there was 47% by weight of TOC, based on 0.32 grams per cubic centimeter (g/cm3) and 0.21 g/cm3 for dry bulk density of walnut shells and manure, respectively. In the spent manure, the TOC values were 18.9%, 18.5%, 15.1%, 15.9%, 12.3%, and 14.4% for bucket bioreactors I, II, III, IV, V, and VI, respectively, or 15.85% on average. Based on these results, approximately half of the organic carbon would have been oxidized during the SRR laboratory experiment. This result is in discrepancy with the stoichiometric evaluation that indicates the TOC oxidation is below 1%. As it turned out, the laboratory experienced technical problems with analysis of the mixed

2 LOI 400 and the Walkley-Black methods for organic carbon determination were taken from the Methods of Soil Analysis, Part 2.1982, number 9 (Part 2) in the series. The LOI is method 1-4, and the Walkley-Black TOC is method 29-3.5.2.

13

In conclusion, the values of TOC in the spent W/M organic medium reported above are not reliable and are presented in this document to identify the problem rather than to draw conclusions regarding the depletion rate of organic carbon.

matrix of walnut shells and manure; this was confirmed by the additional analyses conducted on fresh material. The TOC in fresh walnut shells analyzed for the second time was 52.7% and 61.3% using LOI 400 and Walkley Black methods, respectively. However, the TOC value for fresh 0.5/0.5 W/M organic medium was reported at 17.4%, i.e., within the range of the previously analyzed spent 0.5/0.5 W/M organic medium.

Figure 4-1. Test sump with vertical pipes used for the long duration field test of hydraulic conductivity for 0.5/0.5 W/M organic medium.

14

Long-duration field test of hydraulic conductivity for two W/M organic media configured for upward vertical

flow

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0 10 20 30 40 50 60 70 80 90 100

Time ellapsed (day)

Hyd

raul

ic c

ondu

ctiv

ity (c

m/s

)0.8/0.2 vertical (field)0.5/0.5 vertical (field)

Figure 4-2. Long duration field test of hydraulic conductivity for two W/M organic media configured for upward vertical flow.

Al and Fe concentrations in W/M organic medium of the plugged reactor

0

1000

2000

3000

4000

5000

6000

7000

8000

0 1 2 3 4 5 6 7 8

Depth of the column (ft)

Con

cent

ratio

n (m

g/kg

)

Al in column No. 2Fe in column No. 2

Flow direction

Figure 4-3. Al and Fe concentrations in W/M organic medium of the plugged reactor.

15

Distribution of hydraulic head within the W/M organic medium of the plugged reactor

0

1

2

3

4

5

6

0 1 2 3 4 5 6 7 8

Depth of the column (ft)

Hyd

raul

ic h

ead

(ft) Column No. 3

Column No. 9

Column not plugged

Flow direction

Figure 4-4. Distribution of hydraulic head within the W/M organic medium in the plugged reactor.

Figure 4-5. Laboratory assembly for long duration tests of hydraulic conductivity for the 0.8/0.2 and 0.5/0.5 W/M organic media configured for upward-vertical and horizontal flows.

16

Long-duration tests of hydraulic conductivity for the 0.8/0.2 an0.5/0.5 W/M organic media configured for vertical flow

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0 10 20 30 40 50 60 70 80 90 100

Time ellapsed (day)

Hyd

raul

ic c

ondu

ctiv

ity (c

m/s

) 0.5/0.5 vertical (lab)

0.8/0.2 vertical (lab)

0.8/0.2 vertical (field)

Figure 4-6. Results of the long duration tests of hydraulic conductivity for the 0.8/0.2 and 0.5/0.5 W/M organic media configured for upward-vertical flow.

Long-duration tests of hydraulic conductivity for the 0.8/0.2 and 0.5/0.5 W/M organic media configured for horizontal flow

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0 10 20 30 40 50 60 70 80 90 100

Time ellapsed (day)

Hyd

raul

ic c

ondu

ctiv

ity (c

m/s

) 0.5/0.5 horizontal (lab)

0.8/0.2 horizontal (lab)

Figure 4-7. Results of the long duration tests of hydraulic conductivity for the 0.8/0.2 and 0.5/0.5 W/M organic media configured for horizontal flow.

17

Figure 4-8. Conceptual picture of the RC installed at the mine site.

Figure 4-9. Conceptual three-dimensional drawings showing main components of the RC.

18

Figure 4-10. Conceptual cross section of the RC with respective details.

Figure 4-11. Bag with walnut shells and manure.

19

Decrease in sulfate concentration(calculated from sulfur by ICP)

-250

-200

-150

-100

-50

0

50

100

150

200

250

Medium

Influent

I (medium

@ 44F)

II (medium

@ 58F)

III (medium

@ 77F)

IV (medium

@ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F

Bucket bioreactor I.D.

Dec

reas

e in

con

cent

ratio

n (m

g/L)

16-Oct-0116-Nov-0117-Dec-0118-Jan-0218-Feb-0218-Mar-02

Strong AMD sulfate concentration795 mg/L, 777 mg/L, 762 mg/L, 723 mg/L, 747 mg/L

Medium AMD concentration

Figure 4-12. Decrease in sulfate concentration for bucket bioreactors.

Sulfate Reduction Rate(converted from sulfur by ICP)

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Medium

Influent

I (medium

@ 44F)

II (medium

@ 58F)

III (medium

@ 77F)

IV (medium

@ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F

Bucket bioreactor I.D.

Red

uctio

n ra

te m

ol/(d

*m3)

16-Oct-01

16-Nov-01

17-Dec-01

16-Jan-02

18-Feb-02

18-Mar-02

Figure 4-13. Sulfate reduction rates.

20

Dissolved Zn trends

0

5

10

15

20

25

10/6/01 11/5/01 12/5/01 1/4/02 2/3/02 3/5/02 4/4/02

Date

Con

cent

ratio

n (m

g/L)

Medium Influent

I (medium @ 44F)

II (medium @ 58F)

III (medium @ 77F)

IV (medium @ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F)

Figure 4-14. Dissolved Zn trends for bucket bioreactors.

Dissolved Cu trends

0

1

2

3

4

5

6

7

10/6/01 11/5/01 12/5/01 1/4/02 2/3/02 3/5/02 4/4/02

Date

Con

cent

ratio

n (m

g/L) Medium Influent

I (medium @ 44F)

II (medium @ 58F)

III (medium @ 77F)

IV (medium @ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F)

Figure 4-15. Dissolved Cu trends for bucket bioreactors.

21

Dissolved Cd trends

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

10/6/01 11/5/01 12/5/01 1/4/02 2/3/02 3/5/02 4/4/02

Date

Con

cent

ratio

n (m

g/L)

Medium Influent

I (medium @ 44F)

II (medium @ 58F)

III (medium @ 77F)

IV (medium @ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F)

Figure 4-16. Dissolved Cd trends for bucket bioreactors.

pH Trends

0

1

2

3

4

5

6

7

8

10/6/01 11/5/01 12/5/01 1/4/02 2/3/02 3/5/02 4/4/02

Date

pH

Medium Influent

I (medium @ 44F)

II (medium @ 58F)

III (medium @ 77F)

IV (medium @ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F)

Figure 4-17. pH trends for bucket bioreactors.

22

Dissolved K trends

0

100

200

300

400

500

600

700

800

10/6/01 11/5/01 12/5/01 1/4/02 2/3/02 3/5/02 4/4/02

Date

Con

cent

ratio

n (m

g/L) Medium Influent

I (medium @ 44F)

II (medium @ 58F)

III (medium @ 77F)

IV (medium @ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F)

Figure 4-18. Dissolved K trends for bucket bioreactors.

Dissolved Na trends

0

5

10

15

20

25

30

35

40

45

10/6/01 11/5/01 12/5/01 1/4/02 2/3/02 3/5/02 4/4/02

Date

Con

cent

ratio

n (m

g/L) Medium Influent

I (medium @ 44F)

II (medium @ 58F)

III (medium @ 77F)

IV (medium @ 58F)

Strong Influent

V (strong @ 58F)

VI (strong @ 58F)

Figure 4-19. Dissolved Na trends for bucket bioreactors.

23

24

Table 4-1. Results of the Initial Hydraulic Conductivity Tests Results of the initial hydraulic conductivity tests

Test No.

Date Column specification

Organic medium

mix#

K* (cm/s)

Bag material Moisture content (by

weight)

Compaction ratio&

Compaction method

1 1/30/01 6" PVC 0.5/0.5 9.87E-02 none dry ? static loading Mean 0.3545492 2/8/01 6" pvc 0.5/0.5 6.16E-02 none dry ? static loading Standard Error 0.0723383 3/23/01 6" PVC 0.5/0.5 4.48E-01 0.5" extruded net 61% 20% static loading - length 15" to 12" Median 0.27754 3/26/01 6" PVC 0.5/0.5 1.77E-01 none 61% 20% static loading - length 15" to 12" Mode #N/A5 3/26/01 6" PVC@@ 0.5/0.5 5.69E-02 none 61% 20% static loading - length 15" to 12" Standard Deviation 0.3827756 3/29/01 6" PVC 0.4/0.6 2.80E-01 0.5" extruded net 67% 20% static loading - length 15" to 12" Sample Variance 0.1465177 3/29/01 6" PVC 0.4/0.6 2.18E-01 none 65% 20% static loading - length 15" to 12" Kurtosis 7.0592278 3/30/01 6" PVC 0.33/0.67 3.33E-02 0.5" extruded net 67% 20% static loading - length 15" to 12" Skewness 2.3036789 4/2/01 6" PVC 0.33/0.67 6.88E-03 none 67% 20% static loading - length 15" to 12" Range 1.80312

10 4/6/01 6" PVC 0.4/0.6 4.98E-01 0.5" extruded net 68% 20% static loading - length 15" to 12" Minimum 0.0068811 4/11/01 6" PVC 0.4/0.6 7.80E-02 none 64% 20% static loading - length 15" to 12" Maximum 1.8112 4/12/01 6" PVC 0.4/0.6 6.03E-01 0.5" extruded net 64% 27% static loading - peak load of 175 lb Sum 9.9273813 4/13/01 6" PVC 0.4/0.6 4.71E-02 none 68% 27% static loading - peak load of 175 lb Count 2814 4/20/01 6" PVC 0.5/0.5 5.80E-01 0.5" extruded net 67% 20% static loading - peak load of 175 lb15 4/23/01 6" PVC 0.5/0.5 1.04E+00 none 67% 20% static loading - peak load of 175 lb16 4/24/01 6" PVC 0.5/0.5 1.05E-01 0.5" extruded net 64% 20% static loading - peak load of 175 lb17 4/24/01 6" PVC 0.5/0.5 1.81E+00 none 65% 20% static loading - peak load of 175 lb18 4/25/01 6" PVC 0.5/0.5 3.72E-01 0.5" extruded net 65% 20% static loading - peak load of 175 lb19 4/26/01 6" PVC 0.45/0.55 3.40E-01 0.5" extruded net 65% 22% static loading - peak load of 175 lb20 4/26/01 6" PVC 0.45/0.55 4.68E-02 none 66% 22% static loading - peak load of 175 lb21 5/15/01 10" ADS@ 0.5/0.5 2.41E-02 0.5" extruded net 67% 17% dynamic loading - 8 blows**22 5/17/01 10" ADS 0.5/0.5 2.75E-01 0.5" extruded net 67% 13% dynamic loading - 3 blows23 5/18/01 10" ADS 0.5/0.5 4.42E-01 none 64% 13% dynamic loading - 3 blows24 5/22/01 10" ADS 0.5/0.5 7.78E-01 0.5" extruded net 64% 13% dynamic loading - 3 blows25 5/29/01 10" ADS 0.5/0.5 3.48E-01 none 64% 13% dynamic loading - 3 blows26 8/26/01 10" ADS 0.5/0.5 3.56E-01 0.6" braided net 65% 12% dynamic loading - 3 blows27 8/27/01 10" ADS 0.5/0.5 2.14E-01 none 65% 12% dynamic loading - 3 blows28 9/4/01 10" ADS 0.5/0.5 5.90E-01 two layers of 0.6" braided net## 65% 12% dynamic loading - 3 blows

# Ratio of walnut shells to manure* K = Hydraulic conductivity& Compaction calculation = [(original length of bag or lift) -(compacted length of bag or lift)] /(original length of bag or lift)@ ADS = advanced drainage system, i.e. corrugated pipe made of HDPE** Compacted using 28 lb slide weight - 3 foot stroke for each blow

@@ The only test conducted with the upward flow. Other tests were conducted with water flowing downward## Final selection of netting (mesh) material

Statistics

Table 4-2. Summary of Hydraulic Conductivity Testing Summary of hydraulic conductivity tests conducted for the project

Hydraulic conductivity (cm/s) Tests description

Ratio of walnut shells and manure Flow orientation Maximum Minimum* Mean or most probable#

Lab-short duration

0.33/0.67 to 0.5/0.5 Vertical 1.81 0.0069 0.354

Vertical 0.0034 0 NA 0.5/0.5 Horizontal 0.0061 0.0018 0.0038#

Vertical 0.0052 0 NA Lab-long duration 0.8/0.2 Horizontal 0.017 0.005 0.01#

0.5/0.5 Vertical 0.02 0.0005 NA Field-long duration 0.8/0.2 Vertical 0.024 0.007 0.016#

* Not accounting for the cease of flow due to plugging of the external water supply system # The most probable value is based on the diagram in Figure 4-6 (see Section 4.1.1.1.3 for explanation)

NA Not Applicable

Table 4-3. Target Composition of Synthetic AMDs and Chemical Compounds used for their Production

Target composition of synthetic AMDs

SaltMolecular

weightMaterial Medium Strong Medium Strong Medium Strong

Al 14.1 42 Al2(SO4)3.18H2O 666 230.07 342.66 75.20 224.00Cd 0.02 0.08 CdSO4 208 0.049 0.098 0.017 0.069Ca 18 36.8 CaCl2 110.8 65.96 67.43Cu 1.2 6.3 CuSO4.5H2O 249 6.27 16.47 1.83 9.60Fe 7.22 39.8 FeSO4.7H2O 278 47.42 130.70 12.38 68.23Mg 8 17.5 MgSO4 120.3 52.40 57.31 31.60 69.14Mn 3.77 58.6 MnSO4.H2O 169 15.33 119.11 6.58 102.28K 2 4.1 K2SO4 174.2 7.92 8.12 3.30 6.76

Na 5 9.17 Na2SO4 142 20.42 18.73 10.43 19.14Zn 11.1 20.7 ZnSO4.H2O 179 41.72 38.91 16.91 31.54

SO4 229 595 158.26 530.76pH 4.2 2.6

Volume (gal) 350 175

Needed Salts (g)Targets (mg/L) Theoretical SO4 from salts (mg/L)

25

Table 4-4. Decrease of Sulfate Concentration and Sulfate Reduction Rates

Date I (@ 44F) II (@ 58F) III (@ 77F) IV (@ 58F) V (@ 58F) VI (@ 58F)10/16/01 143.7 -13.5 -98.4 -48.6 -23.1 795 168 12911/16/01 212.7 35.4 24.6 55.2 62.7 777 66 6312/17/01 209.4 35.4 -165.6 88.5 33.9 795 36 481/18/02 198.3 35.7 36.9 64.2 45 723 42 362/18/02 203.7 44.7 32.1 47.7 23.7 No feed #N/A #N/A3/18/02 205.2 36.3 18.6 45.9 30.3 747 33 87

Average (sans negative) 37.5 28.1 60.3 39.1 44.3 58.5

I (@ 44F) II (@ 58F) III (@ 77F) IV (@ 58F) V (@ 58F) VI (@ 58F)10/16/01 143.7 -0.12 -0.88 -0.43 -0.21 1.50 1.1511/16/01 212.7 0.32 0.22 0.49 0.56 0.59 0.5612/17/01 209.4 0.32 -1.48 0.79 0.30 0.32 0.431/18/02 198.3 0.32 0.33 0.57 0.40 0.38 0.322/18/02 203.7 0.40 0.29 0.43 0.21 No feed #N/A #N/A3/18/02 205.2 0.32 0.17 0.41 0.27 0.29 0.78

Average (sans negative) 0.33 0.25 0.54 0.35 0.40 0.52

Average for all bioreactors 0.40

Bioreactor ID Bioreactor ID

Sulfate reduction rate for bioreactors fed by medium AMD [mol/(d*m3)]

Sulfate reduction rate for bioreactors fed by

strong AMD [mol/(d*m3)]

Bioreactor ID Bioreactor ID

26

Table 4-5. Dissolved Metal Concentrations and pH for Bucket Reactor Dissolved metal concentrations and pH for bucket bioreactors

Metal Date

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 8.76 0.818 0.486 0.929 0.764 41.5 1.45 1.2811/16/2001 12.8 0.426 0.669 0.623 0.133 40.6 27.7 23.112/17/2001 12.9 0.315 24.2 0.156 0.583 38.7 38.4 40.31/16/2002 8.41 0.0963 0.145 0.0898 0.19 40.6 37.2 37.72/18/2002 9.34 0.217 0.14 0.12 0.0974 #N/A #N/A #N/A3/18/2002 10.2 0.11 0.191 0.174 0.024 40.6 38.8 31.9Average 10.40 0.33 4.31 0.35 0.30 40.40 29.03 25.70

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 0.019 0.004 0.004 0.004 0.004 0.093 0.025 0.02411/16/2001 0.022 0.004 0.009 0.004 0.004 0.085 0.065 0.0512/17/2001 0.022 0.004 0.031 0.004 0.004 0.076 0.074 0.0751/16/2002 0.0178 0.0049 0.0044 0.0044 0.0044 0.0598 0.0583 0.05422/18/2002 0.013 0.0044 0.0044 0.0044 0.0044 #N/A #N/A #N/A3/18/2002 0.012 0.005 0.005 0.005 0.005 0.06 0.05 0.037Average 0.018 0.004 0.010 0.004 0.004 0.075 0.053 0.045

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 18.5 52.7 39.5 75.7 49.3 45.2 68.8 73.111/16/2001 27.7 53.1 40.2 57.3 66.5 44 59.1 85.112/17/2001 26.2 44.9 38.6 69.4 51.1 40.9 47.7 41.21/16/2002 32.1 41.9 42.5 76.6 52.9 48.4 52.6 52.22/18/2002 30.9 48.8 50.5 67.5 39.8 #N/A #N/A #N/A3/18/2002 31.7 43.1 40.5 64.7 41.2 47.6 51 58.1Average 27.9 47.4 42.0 68.5 50.1 45.1 55.5 63.1

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 0.832 0.213 0.133 0.103 0.115 6.37 0.934 0.6111/16/2001 1.18 0.08 0.159 0.018 0.021 6.15 3.88 2.8312/17/2001 1.17 0.029 2.07 0.002 0.056 5.77 5.17 51/16/2002 1.08 0.0069 0.0312 0.0022 0.0487 6.12 4.79 5.12/18/2002 1.1 0.0066 0.0022 0.0035 0.0016 #N/A #N/A #N/A3/18/2002 1.14 0.002 0.004 0.002 0.002 6.16 5.16 4.16Average 1.084 0.056 0.400 0.022 0.041 6.122 4.061 3.352

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 0.371 1.22 0.484 2.45 1.17 41.7 2.13 2.0411/16/2001 0.065 0.183 0.212 0.551 0.511 39.4 11.6 8.712/17/2001 0.044 0.159 1.42 0.141 0.31 36.1 36.1 34.51/16/2002 4.3 0.0408 0.508 0.0985 0.0653 41 34.4 36.62/18/2002 0.0966 0.0208 0.437 0.101 0.0408 #N/A #N/A #N/A3/18/2002 0.035 0.061 0.107 0.086 0.031 40.1 33.4 24.2Average 0.819 0.281 0.528 0.571 0.355 39.480 23.326 18.728

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 7.65 32.5 26.4 30.2 30.6 20.1 38.7 38.411/16/2001 11.8 20.9 16.8 25 20.1 20.2 29.8 33.212/17/2001 11.5 17.2 16.7 25.4 17.4 19.4 22.1 22.41/16/2002 13.5 16.5 18.7 24 21.2 22.5 24.1 23.72/18/2002 13.3 20.4 22.5 19.3 16.1 #N/A #N/A #N/A3/18/2002 13.5 17.5 17 18 17.1 22.7 24.7 27.4Average 11.9 20.8 19.7 23.7 20.4 21.0 28.0 29.8

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 2.55 1.03 0.643 1.56 0.783 57.5 30.1 30.511/16/2001 3.72 1.98 2.87 2.32 2.49 53.7 46.6 54.112/17/2001 3.64 2.68 19 1.59 49.3 50.3 50.8 2.691/16/2002 3.52 3.21 2.91 18.4 2.83 54.5 51.7 522/18/2002 3.44 3.16 2.71 2.23 3.29 #N/A #N/A #N/A3/18/2002 3.49 3.23 3.22 2.47 3.27 53.6 50.9 46.8Average 3.39 2.55 5.23 4.76 10.33 53.7 45.9 36.2

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 3.49 740 622 576 607 8.57 480 40911/16/2001 5.3 116 93.2 166 84.2 8.78 158 12412/17/2001 5.05 44.1 29.6 60 29.1 8.56 34.3 82.71/16/2002 5.55 25 49.9 25.9 68.9 8.87 34.3 242/18/2002 5.4 30.9 60.7 11.9 17 #N/A #N/A #N/A3/18/2002 5.57 15.1 17.6 9.51 20.4 8.81 29.8 51.8Average 5.1 161.9 145.5 141.6 137.8 8.7 146.4 143.9

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 12.6 41.8 38.7 37.1 39 12.5 30.7 27.411/16/2001 18.7 22.6 22.4 24.7 20.7 12.5 18.3 16.612/17/2001 17.9 19 16.8 19.2 13.3 12.2 13.7 20.71/16/2002 15 15.5 16.1 16.5 18 13.4 13.8 13.62/18/2002 14.7 16.4 16.4 14.3 15.3 #N/A #N/A #N/A

Mg

Mn

K

Na

Cd

Ca

Cu

Fe

Medium AMD Strong AMDDissolved metal concentrations (mg/L)

Al

3/18/2002 15.1 15.7 15.2 15.4 15.5 14 16.2 20.2Average 15.7 21.8 20.9 21.2 20.3 13.0 19.0 21.0

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 6.33 1.99 0.516 0.913 0.901 20.3 5.84 7.1911/16/2001 11.5 1.57 3.04 0.996 0.598 19.2 15.6 15.512/17/2001 10.1 1.19 10.9 0.081 1.76 18 18 17.21/16/2002 10.5 3.3 2.95 0.046 0.62 16.1 14.5 14.72/18/2002 8.5 0.495 0.363 0.0281 4.26 #N/A #N/A #N/A3/18/2002 8.31 1.1 4.09 0.033 2.27 15.7 14.2 12.2Average 9.21 1.61 3.64 0.35 1.73 17.78 13.57 12.86

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 47.9 52.4 80.7 64.1 55.6 265 209 22211/16/2001 70.9 59.1 62.7 52.5 60 259 237 23812/17/2001 69.8 58 125 40.3 58.5 254 253 2491/16/2002 66.1 54.2 53.8 44.7 51.1 241 227 2292/18/2002 67.044 53.033 57.15 52.003 60.065 #N/A #N/A #N/A3/18/2002 68.4 56.3 62.2 53.1 58.3 249 238 220Average 65.0 55.5 73.6 51.1 57.3 255.2 235.0 229.8

Weak Influent I (weak @ 44F) II (weak @ 58F) III (weak @ 77F) IV (weak @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 143.7 157.2 242.1 192.3 166.8 795 627 66611/16/2001 212.7 177.3 188.1 157.5 180 777 711 71412/17/2001 209.4 174 375 120.9 175.5 762 759 7471/16/2002 198.3 162.6 161.4 134.1 153.3 723 681 6872/18/2002 201.132 159.099 171.45 156.009 180.1953/18/2002 205.2 168.9 186.6 159.3 174.9 747 714 660Average 195.1 166.5 220.8 153.4 171.8 765.6 705.0 689.4

Medium Influent I (medium @ 44F) II (medium @ 58F) III (medium @ 77F) IV (medium @ 58F) Strong Influent V (strong @ 58F) VI (strong @ 58F)10/16/2001 3.94 7.37 6.66 6.95 7.26 2.45 2.76 3.5911/16/2001 4.16 7.11 6.43 6.95 7.33 2.85 3.09 3.2712/17/2001 4.18 6.53 4.31 6.97 6.58 2.65 2.78 2.81/16/2002 4.16 5.7 6.17 6.63 6.73 2.64 2.98 2.932/18/2002 4.24 6.04 6.45 6.23 6.24 #N/A #N/A #N/A3/18/2002 4.09 5.62 5.91 6.51 6.02 2.71 2.96 3.06Average 4.13 6.40 5.99 6.71 6.69 2.66 2.91 3.13

pH

S

Zn

SO4 from S values

pH

27

5. Description of the BEST Simulator The computer simulator for the design of the BEST is a spreadsheet-based model that is used in conjunction with a public domain computer software package, PHREEQCI (pH, redox, equilibrium interactive geochemical modeling program) (Ref. 6). While PHREEQCI calculates geochemical equilibrium for the advective-reactive transport of AMD through the bioreactor, the spreadsheet portion of the simulator handles issues of AMD flow rate, size of the bioreactor, its operational time, and its economics. In general, the BEST (Appendix C) modeling process is based on the chemical composition of the AMD and its flow rate, TOC content in the organic matter, cost of material and production of a typical RC, the SRR of the organic matter used in the treatment system, and the discount rate (Ref. 7) and operation and maintenance cost (O&M) for calculation of the net present value (NPV). The BEST simulator was developed and formulated so it could be operated by a user with minimum modeling experience. The BEST simulator operation requires basic knowledge of the ExcelTM program and some familiarity with geochemical model PHREEQCI. Of course, a good chemical background would be a bonus. 5.1 Organization of the BEST Simulator The BEST simulator is saved as Microsoft ExcelTM workbook, BEST V2.xls (Appendix C) and consists of 17 worksheets. The first two worksheets (Figure 5-1and Figure 5-2) are Charts I and II that show the navigation between the 14 worksheets that are identified with letters A through L and numbers 1 and 2 for the worksheet series B and D. Chart I covers the navigation among worksheets A, B1, B2, C, D1, and D2 to select the treatment scope based on the size and associated cost of the preliminary design of the SRB treatment system. Chart II shows navigation between worksheets E, G, H, I, J, K, and their interaction with the PHREEQCI model and its input file assembled as worksheet F.

Worksheet 0 entitled "Input and Output" (I-O) (Figure 5-3) allows for entering the majority of input data and having the most important results also printed on the same page. However, details of the design-like specification of the material, etc., are not listed in the I-O worksheet, and the user needs to refer to worksheets A through L to examine these details. Most worksheets are linked together, i.e., any change of input data causes appropriate changes of the results calculated by the respective worksheet. However, the PHREEQCI model and its data input file (Worksheet F) are not automatically linked with the rest of the worksheets, thus required changes need to be entered manually. 5.1.1 ExcelTM Workbook The ExcelTM portion of the BEST simulator allows for entering data only in the I-O worksheet. This means, for example, that the AMD flow rate is entered only once in the I-O worksheet despite the fact its value is also used in calculations performed by worksheets B1, B2, G, I, and K. The only exceptions are unforeseen items associated with the typical tank adaptation that will require a direct entry in Worksheet C rather than in the I-O worksheet. Cells for input data are marked with green "fill color" and are typed in italics. For clarity of the simulator logistics, three additional "fill colors" are used in worksheets A through L; these are yellow, orange, and red. Yellow (with italic font) is used to mark cells with values that were transferred directly from the I-O worksheet. Orange denotes results that were calculated within the given worksheet and were exported to another worksheet. Red indicates values that were imported to the given worksheet from any other worksheet except the I-O worksheet. To prevent an accidental entry to cells that include formulas or links with other worksheets, all worksheets are protected so that the only active cells are those that allow data entry (green "fill color"). However, a user can easily unprotect each worksheet or the entire workbook because there is no password used for protection. Although the

28

BEST simulator operates and requires entry data in specific units (mostly English), selected values are also converted to the metric system and various time units. 5.1.1.1 Description of the Worksheets Chart I (Figure 5-1) shows how to select the treatment scope, it has two paths. The main path, designated with the shapes drawn with a solid line, estimates the preliminary cost of the SRB treatment system that would remove only metals precipitating as sulfides. The right branch of the chart, which includes shapes drawn with a dashed line, estimates the preliminary cost for the SRB treatment system that would remove metals by precipitating them as sulfides and hydroxides. This worksheet neither requires any input data, nor does it calculate any values. Chart II (Figure 5-2) shows the interaction of the ExcelTM workbook with the PHREEQCI program and is a road map to the final design of the SRB treatment system. There are three possible routes in this chart that are designated by shapes drawn with solid, dashed, and dotted lines. This worksheet neither requires any input data, nor does it calculate any values. Worksheet 0 (Figure 5-3) enables the user to enter data and read the most important results using a one-page printout. The following color convention (not seen in a black and white copy of the report) is used in this worksheet:

– portion of the worksheet with violet colored fonts designates the main data entry;

– fields filled with green are for data entry (fonts appearing in these fields are in red);

– portion of the worksheet with blue colored fonts designates the output for the preliminary design (Worksheets A, B1, B2, D1, and D2 that are used exclusively for the preliminary design also use blue colored fonts); and

– portion of the worksheet with black colored fonts designates the treatment system design (this portion includes four areas shaded with different colors to accentuate four

successive "check points" in the process to produce the final design of the SRB treatment system; nevertheless, there are a few fields in this portion of the worksheet that are filled with green and use red colored fonts that require data input).

Worksheet A (Figure 5-4) includes a stoichiometric evaluation of carbon oxidation required to precipitate metals either as sulfides or as sulfides and hydroxides. This evaluation is based on the concentration of metals in the AMD as analyzed for the given site. Input data (entered in the I-O worksheet) includes the name of the site, concentration of metals as listed [milligrams per liter (mg/L)], and entry for other metals if needed. Atomic mass for the new metals needs also to be entered. Specifically, this worksheet calculates the following main values:

− carbon oxidation required to precipitate all metals; and

− carbon oxidation required to precipitate only metal sulfides.

Worksheet B1 (Figure 5-5) includes preliminary calculation of the number of RCs in the SRB treatment system that is required to precipitate metal sulfides for the given AMD flow rate and metal concentration, the RC dimensions, and the SRR. This worksheet also calculates the amount of organic carbon present in this preliminary SRB treatment system design and the maximum number of years for the carbon depletion. Input data includes SRR [mmol/(d*L)] assumed based on literature sources, dimensions of the RC (feet), flow rate (gpm), porosity of W/M organic medium (dimensionless), TOC in fresh manure (%), TOC in walnut shells (%), volumetric moisture content of manure (g/cm3), volumetric ratio of manure in the organic matter, which are all entered through the I-O worksheet. Calculations in this worksheet also use a preliminary value of carbon oxidation required for metals removal as sulfides, which is calculated in Worksheet A. The main information provided by this worksheet includes the following:

29

− volume of the RC; − preliminary number of RC in the treatment

system; − stoichiometric check of organic carbon

supply; and − AMD residence time and seepage velocity.

Worksheet B2 (Figure 5-6) includes preliminary calculation of the number of RCs in the SRB treatment system that is required for precipitation of metals as sulfides and hydroxides for the given AMD flow rate and metal concentration (Worksheet A), the RC dimensions, and an assumed SRR. This worksheet also calculates the amount of organic carbon present in this preliminary SRB treatment system design and the maximum number of years for the carbon depletion. Input parameters are the same as already entered for Worksheet B1; however, the preliminary value of carbon oxidation required for metals removal uses the value calculated for metals removed as both the sulfides and hydroxides (Worksheet A). The main information provided by this worksheet includes:

− volume of the RC; − preliminary number of RC in the treatment

system; − stoichiometric check of organic carbon

supply; and − AMD residence time and seepage velocity.

Worksheet C (Figure 5-7) includes the cost estimate of material and labor to adapt and modify a typical tank for RC. A cylindrical 2,500-gallon nominal size tank is used as an example in this worksheet. Input data includes date of the estimate and labor rate (dollar per hour) entered in the I-O worksheet. Unpredicted modification to a typical tank needs to be entered separately in Worksheet C. If any unit prices used in this worksheet need to be modified, such modifications may be done after unprotecting the worksheet. Worksheet D1 (Figure 5-8) includes the cost estimate and NPV for the preliminary design of the SRB treatment system for the removal of

metals precipitating as sulfides. It also includes the quantities and cost of materials needed to produce bags with the organic mix and installation cost of the SRB system (preliminary design for metals precipitating as sulfides) at the mine site. Calculations conducted within this worksheet are based on the number of RC determined in Worksheet B1. Input data includes the volumetric ratio (dimensionless) of manure in the organic material mix, hourly rates (dollar per hour) for bagging the organic matter and installing the system at the mine site, and labor (hours) needed to make one bag with W/M organic medium and install one RC at the site. The discount rate for calculation of the NPV and the assumed O&M cost are also entered in this worksheet. All data input is made via the I-O worksheet. The main information provided by this worksheet includes:

− list of materials needed for one RC; − quantities of materials needed to construct

an SRB treatment system; − cost of materials needed to construct an SRB

treatment system; − labor involved in constructing an SRB

treatment system; − capital cost of an SRB treatment system; and − NPV of total cost.

Worksheet D2 (Figure 5-9) includes the cost estimate and NPV for the preliminary design of the SRB treatment system for removal of all metals precipitating as sulfides and hydroxides. It also includes the quantities and cost of materials needed to make bags with the organic mix and installation cost of the SRB treatment system (preliminary design for all metals precipitating) at the mine site. Calculations conducted within this worksheet are based on the number of RC determined in Worksheet B2. Input parameters are the same as already entered to the I-O worksheet for the calculations performed for Worksheet D1. The main information provided by this worksheet includes:

− list of materials needed for one RC;

30

− quantities of materials needed to construct an SRB treatment system;

− cost of materials needed to construct an SRB treatment system;

− labor involved in constructing an SRB treatment system;

− capital cost of an SRB treatment system; and − NPV of total cost.

Worksheet E (Figure 5-10) includes calculations of the SRR [mmol/(d*L)] and related carbon oxidation [millimole per liter (mmol/L)]. These calculations are conducted based on the data acquired from a laboratory experiment conducted for determination of the SRR for the specific organic mix used for the treatment system. Input data includes influent and effluent sulfur concentration determined using ICP method [milligram per liter (mg/L)], volume of the laboratory bioreactor (L), and the laboratory flow rate [milliliter per minute (mL/min), all entered in the I-O worksheet. Calculations performed in this worksheet use the same porosity value as previously used for worksheets B and D couples. If no laboratory experiment for the SRR was conducted, the user needs to use default values in Worksheet E with the exception of the S concentration in the effluent, which should be set to 246 mg/L, which corresponds to 575 mmol/L and 0.25 mmol/(d*L) for carbon oxidation and SRR, respectively. The main values calculated by this worksheet include:

− SRR; − required carbon oxidation; and − annual carbon use.

Worksheet F (Figure 5-11) is an input file for the PHREEQCI model that includes values of pH, pE, the concentration of metals in the AMD as analyzed for the given site, and the carbon oxidation as determined in Worksheet E. Input parameters are pH; pE; temperature (oC); concentration of Al, Ca, Fe2+, K, Mg, Mn2+, Na, Zn, Cd, Cu2+ and sulfate (SO4) (all mg/L); carbon oxidation [millimole (mmol)] as calculated in Worksheet E and divided by 10 steps; and the

name of the output file. Caution is recommended when entering data in Worksheet F where only red color entries can be altered despite that the entire cell contains yellow "fill color." An alternative method of entering or changing an input to the PHREEQCI model is to alter it while working within the PHREEQCI program. Worksheet G (Figure 5-12) calculates the number of RCs in the final SRB treatment system for the selected treatment scope (usually for metal removal as sulfides) provided that the PHREEQCI simulation, conducted using the carbon oxidation value as calculated in Worksheet E and the actual AMD pH, delivers acceptable results for metal removal. Shapes drawn with a solid line in Chart II designate such a path. This worksheet is analogous to Worksheet B1 or B2 and uses data that have been already entered in the I-O worksheet. The main information provided by this worksheet includes:

− volume of the RC; − preliminary number of RC in the treatment

system; − organic carbon supply; and − AMD residence time and seepage velocity.

Worksheet H (Figure 5-13) calculates the cost and NPV for the final design of the SRB treatment system of the selected scope for the number of RCs as calculated in Worksheet G. It also includes the quantities and cost of materials needed to make bags with the organic mix and cost of installation of the SRB treatment system at the mine site. This worksheet is analogous to Worksheet D1 or D2 and uses data that have been already entered in the I-O worksheet. The main information provided by this worksheet includes:

− list of materials needed for one RC; − quantities of materials needed to construct

an SRB treatment system; − cost of materials needed to construct an SRB

treatment system; − labor involved in constructing an SRB

treatment system;

31

− capital cost of an SRB treatment system; and − NPV of total cost.

Worksheet I (Figure 5-14) calculates the number of RCs in the final SRB treatment system for the selected treatment scope when the PHREEQCI simulation conducted using carbon oxidation value calculated in Worksheet F and the AMD pH does not deliver acceptable results for metal removal. In such a case, either the amount of carbon oxidation and/or pH needs to be increased to produce acceptable results of metals removal. Such a scenario needs to be simulated again using PHREEQCI software. Shapes drawn with a dashed line in Chart II designate such a path. The only data input is the adjusted carbon oxidation that was used in the latest PHREEQCI simulation. This value is entered in the I-O worksheet. The main information provided by this worksheet includes:

− volume of the RC; − preliminary number of RC in the treatment

system; − organic carbon supply; and − AMD residence time and seepage velocity.

Worksheet J (Figure 5-15) calculates cost and NPV for the final design of the SRB treatment system of the selected scope for the number of RCs as calculated in Worksheet I. It also includes the quantities and cost of materials needed to make bags with the organic mix and installation cost of the SRB treatment system at the mine site. This worksheet is analogous to Worksheet D1 or D2 but uses the scenario as simulated in Worksheet I. All input data have already been entered in the I-O worksheet. The main information provided by this worksheet includes:

− list of materials needed for one RC; − quantities of materials needed to construct

an SRB treatment system; − cost of materials needed to construct an SRB

treatment system; − labor involved in constructing an SRB

treatment system;

− capital cost of an SRB treatment system; and − NPV of total cost.

Worksheet K (Figure 5-16) is applicable only for a treatment system with a parallel configuration of RCs. This worksheet calculates the seepage velocity and the residence time of AMD in the final SRB treatment system with the increased number of RCs to satisfy the flow velocity and the residence time requirements, if not matched by the earlier designs. Shapes drawn with a dotted line in Chart II designate such a path. The only data input is the increased number of RCs in the treatment system that is again entered in the I-O worksheet. The main information provided by this worksheet includes:

− volume of the RC; − preliminary number of RC in the treatment

system; − organic carbon supply; and − AMD residence time and seepage velocity.

Worksheet L (Figure 5-17) is applicable only for a treatment system with a parallel configuration of RCs. This worksheet calculates the cost and NPV for the final design of the SRB treatment system of the selected scope for the number of RCs as selected in Worksheet K. It also includes the quantities and cost of materials needed to make bags with the organic mix and the installation cost of the SRB treatment system at the mine site. No input data is required for this worksheet because all information was entered earlier in the process. The main information provided by this worksheet includes:

− list of materials needed for one RC; − quantities of materials needed to construct

an SRB treatment system; − cost of materials needed to construct an SRB

treatment system; − labor involved in constructing an SRB

treatment system; − capital cost of an SRB treatment system; and − NPV of total cost.

32

5.2 Input Data Summary and Explanations Most of the input parameters and their dimensions are listed in the I-O worksheet (Figure 5-3) of the BEST simulator. The numerical values for these input parameters are those used for the example problem as included in Appendix C. Section 5.1.1.1 briefly addressed these parameters and described their associations with individual worksheets A through L. Figure 5-3 also lists two criteria for the AMD flow in a bioreactor, the AMD flow velocity and its residence time. These criteria do not directly take part in any calculations performed by the worksheet but are used to determine the adequacy of the AMD flow conditions through a bioreactor.3 Other data input includes values entered in Worksheet F, the PHREEQCI model input file. These data may be entered manually using the input subroutine of the PHREEQCI software. For the example given in this document, the following data (written in italic in braces) would have been manually entered, had Worksheet F not been used:

− temperature (oC) {15}; − pH {2.66}; − pE {0}; − concentration of Al, Ca, Fe2+, K, Mg, Mn2+,

Na, Zn, Cd, Cu2+, and SO4 (mg/L) {40.4, 45.1, 39.48, 8.7, 21, 53.7, 13, 17.78, 0.075, 6.122, 765.6};

− carbon oxidation (mmol) divided for ten steps {0.126}; and

− name of the output file {R-V example.out}. Some parameters used in the BEST simulator are not included in the I-O worksheet but are used in calculations performed by the individual worksheets. Such parameters like unit price for cow manure, walnut shells, netting, etc., are listed 3 For the example data set used in this document with the listed values of 2.5 feet per day (ft/d) and 0.5 days for the flow velocity and residence time, respectively. The flow velocity value is too conservative and was used solely to carry on the modeling process from its beginning to the last step, i.e., through the adjustment for the flow velocity. Actually, as stated by MSE, (Ref. 1), this value should be 10 ft/d.

in an appropriate worksheet (A through L), and can be altered after unprotecting the given worksheet. Such an alteration will automatically affect the results calculated. However, if, for example, the unit price for manure was altered in Worksheet B1 it would only affect calculations performed by this worksheet but would not affect the analogous calculations performed in Worksheet B2. Thus, to alter the given parameter with the intention that the alteration is carried out throughout the BEST simulator, it is necessary to alter each individual worksheet that includes the parameter in question. 5.3 Modeling Process 5.3.1 Treatment Scope Selection The process of the treatment scope selection (Worksheet I) starts with the collection of data to characterize the chemistry. These data, entered through the I-O worksheet, are used in Worksheet A to calculate organic carbon oxidation required to reduce the adequate amount of sulfate to precipitate all metals that may form sulfides as shown by the net reaction 5-1 written for Cd. Cd+2 + SO4-2 + 2CH2O = CdS + 2H2CO3 (5-1) Worksheet A also calculates organic carbon oxidation required to generate excess of H2S that is needed to precipitate other metals as hydroxides as shown by the net reaction 5-2 written for Al.

2Al+3 + 3SO4-2 + 6CH2O + 6H2O =

3H2S + 6H2CO3 + 2Al(OH)3

(5-2)

The results of these simplistic calculations, based only on concentrations of dissolved solids rather than their activities and pH, are used to initially estimate the size and cost of the treatment system. This estimation also uses information on the AMD flow rate, the results of the chemical analysis for TOC in the organic mix used for the treatment system, and the dimensions and cost of production of a typical RC (Worksheets B1 or B2 and Worksheet C).

33

The first estimation of the treatment system size and cost is initially done for the alternative that precipitates only metals sulfides. This procedure follows the shapes drawn with the solid line in Figure 5-1, and the results are automatically recorded in Worksheets B1 and D1. Next, the alternative of precipitating all metals is evaluated. This process follows the shapes drawn with the dashed line in Figure 5-2, and the results are recorded in Worksheets B2 and D2. For the AMD whose chemical composition is presented in Worksheet A, the cost of the treatment system would more than triple (compare Worksheets D1 and D2) if metal hydroxides were to be formed. The capital cost is $40,407 and $137,358 for the "metal-sulfides" and "all-metals" alternatives, respectively. Assuming $1,000 annual O&M cost, related NPVs are $59,511 and $156,488, respectively. Such a treatment system would include 5 RCs and 17 RCs for the "metal-sulfides" and "all-metals" alternatives, respectively. Additional negative effects of the treatment system for "all-metals" would be the production of carbonic acid that would contribute to the overall pH of the treated AMD and the excess of H2S that could affect air quality. Consequently, in the example as included in the BEST model (Appendix C), the selected alternative for the treatment system is the option of precipitating metals as sulfides. 5.3.2 Treatment System Design The modeling process of the treatment system final design (Figure 5-2) starts with the selected alternative for the treatment scope. In the example, as included in the BEST model (Appendix C), the selected alternative is to precipitate only metal sulfides. Therefore, the criterion for a successful operation is an adequate decrease of only those metals that precipitate as sulfides, i.e., Zn, Cd, and Cu in the example given in Appendix C. In addition to the dissolved metal concentration, pH, and pE of the AMD, the input data requires

quantifying the carbon oxidation required for sulfate reduction. This value can be acquired through the laboratory experiment for SRR as described in Section 4.2.1. Data from such an experiment are entered in the I-O worksheet and used in Worksheet E to calculate SRR [mmol/(d*L)] and carbon oxidation (mmol/L) that will be used for PHREEQCI modeling. In the example, as included in the BEST model (Appendix C), values calculated in Worksheet E are 0.541 mmol/(d*L) and 1.263 mmol/L for the SRR and carbon oxidation, respectively. If no experiment was conducted, the value 0.25 mmol/(d*L) for SRR and the corresponding value of 575 mmol/L for carbon oxidation can be used for PHREEQCI.4 These values are considered conservative for the organic mix of manure and walnut shells. It is recommended that the laboratory experiment for SRR also include analytical work for dissolved metals. This information can assist in establishing realistic criteria for metals removal by the treatment system being designed. Certainly, criteria like maximum contaminant level (MCL), suggested maximum contaminant level (SMCL), secondary maximum contaminant level, or any other industry project-specific requirements may be used. 5.3.2.1 PHREEQCI Modeling 5.3.2.1.1 About the PHREEQCI Program PHREEQCI is a public domain geochemical modeling program developed and supported by the U.S. Geological Survey. This version of the program, where I stands for "interactive," is a complete Windows-based graphical user interface to the geochemical computer program PHREEQC (Version 2). PHREEQCI can be used interactively

4 Because in the BEST simulator the value of carbon oxidation is automatically exported from Worksheet E to Worksheet G, it must be manually generated in Worksheet E by manipulating its default data. Setting (in I-O worksheet) the effluent sulfate concentration to 246 mg/L will produce the recommended conservative value for carbon oxidation of 575 mmol/L.

34

to perform all the modeling capabilities of PHREEQC—speciation, batch-reaction, one-dimensional reactive-transport, and inverse modeling. The acronym PHREEQ stands for pH, redox equilibrium, and C stands for C programming language. The latest version of PHREEQCI, 2.8.0.0 dated April 15, 2003, can be obtained from the web site: http://wwwbrr.cr.usgs.gov/projects/GWC_coupled/phreeqci. The download is free of charge, and the downloading process is user friendly. PHREEQCI provides a friendly subroutine to create and modify the input file within the program. This option to handle the input data is preferred and recommended for the BEST users. Worksheet F (Figure 5-11) is an archived file created within the PHREEQCI environment and stored within the ExcelTM workbook. However, users that do not feel comfortable with PHREEQCI may start the modeling process by copying Worksheet F directly from the ExcelTM workbook and pasting it in the PHREEQCI program.5 Any alteration to the PHREEQCI input file could also be done within the ExcelTM workbook before Worksheet F is copied to PHREEQCI. 5.3.2.1.2 Description of the Scenario Modeled by PHREEQCI For the example of the BEST simulator, as presented in Appendix C, the PHREEQCI program models the one-dimensional flow and the advective-reactive transport of AMD through one RC. The quality of the AMD (the RC feed) is specified in the PHREEQCI input file as Solution 0. This solution flows through the RC pushing out (a "piston" flow) an imaginary initial solution that is represented in the PHREEQCI input file as Solution 5To copy the input file (Worksheet F) from the Excel worksheet to the PHREEQCI program, do the following: (1) highlight column A in the ExcelTM worksheet, (2) open PHREEQCI, (3) using the Edit menu, paste the worksheet to the right window of PHREEQCI, (4) expand directory Phrqc1 in the left window, and (5) save the file under a different name (in this case R-V example.pqi). While saving, check (in the right window) whether the first line of the copied program listing the location of the database has disappeared. The program will not run if this line is still visible in the right window.

1-10. For the modeling process, the RC is divided into 10 advective cells (1-10) for which the quality of the treated AMD is calculated. While the AMD is flowing through the RC, its quality changes due to the chemical reaction and the oxidation of organic carbon, whose ratio (1.263 mmol/L) is specified in the PHREEQCI input file. Actually, because there are 10 "advective" cells in 1 RC, the ratio of carbon oxidation for 1 cell is one-tenth of the total 1.263 mmol/L. The PHREEQCI program calculates the quality of AMD in the modeled RC as it leaves each of these 10 imaginary advective cells and, therefore, allows tracing changes of the AMD composition as it flows through the RC. 5.3.2.1.3 PHREEQCI Simulations After the value of carbon oxidation (as calculated in Worksheet E and divided by 10) is entered in the PHREEQCI program (a line in the REACTION subroutine), the program needs to be run, and the selective output file (the name of this file specified in the input file) is examined. The examination can be done by opening the ASCII file within the ExcelTM workbook (Figure 5-18). The examination usually includes concentrations of dissolved target metals (expressed in this file as mmol/L) and pH. In the example, as included in the BEST simulator (Appendix C), concentrations of all metals, except Zn, are drastically reduced. The concentration of Zn in AMD leaving the modeled RC is 1.60E-04 mmol/L, which corresponds to 10.5 mg/L. Considering that the secondary maximum contaminant level for Zn is 5 mg/L, it is apparent that an effective SRB treatment system for Zn removal needs to oxidize more carbon. Therefore, following the logistics shown in Chart II (Figure 5-2), additional PHREEQCI modeling is required that will involve either a higher ratio of carbon oxidation or an increase in pH of the influent AMD. Such a route is marked in Chart II by the shapes drawn with a dashed line. Numerous PHREEQCI modeling might be performed before the right combination of feed pH and/or ratio of carbon oxidation is found. For the example of the BEST simulator (as presented in

35

Appendix C), eight PHREEQCI models were conducted. The results of this modeling (Figure 5-19) show that either an increase of the feed pH to 4 or doubling the carbon oxidation ratio to 2.52 mmol/L will satisfy the criterion to lower Zn concentrations below the secondary maximum contaminant level of 5 mg/L. Raising the feed pH to 4 would involve an additional cell with lime, limestone, or other medium that provides alkalinity. Such a cell could be designed and fabricated as described by Desmier et al (Ref. 8). An increase of the carbon oxidation ratio will increase the size of the SRB treatment system. The latter option is calculated in Worksheet I after the 2.52 mmol/L carbon oxidation ratio is entered in the I-O worksheet. The dashed-line-drawn shapes in Chart I directly below the loop for the PHREEQCI modeling indicate this process. The number of RCs calculated in Worksheet I is three, and the relative capital cost automatically calculated in Worksheet J is $24,244. These values are much lower than those calculated during the preliminary simulation as per Chart I (see Section 5.3.1). The main reason behind this discrepancy is a higher SRR that was used for calculations performed in Worksheet I. Had the assumed SRR entered in the I-O worksheet for calculations conducted in Worksheet B1 been the same [0.54 mmol/(d*L)] as the one used in Worksheet I, the size of the SRB treatment system would have been the same for the simulations conducted according to Charts I and II. However, these coincidental results must not discourage the user to venture into the PHREEQCI modeling because the PHREEQCI modeling provides much more sophisticated computational algorithms (that are adequate for the chemical reactions in the bioreactor) than a simplistic stoichiometric approach. The sophistication of the PHREEQCI program embraces algorithms that include function of pH, pE, chemical activities, and the ionic strength of the solution that addresses interaction between the modeled concentrations of metals.

An expansion of the PHREEQCI modeling process may include a check as to whether the treatment system with RCs configured in series would meet the treatment objectives. Such a configuration may be imposed by topographic conditions at the site, e.g., a narrow gulch that would not allow for a parallel RC pattern. In that case, Worksheet F (PHREEQCI input file) would need to be significantly modified. 5.3.2.2 Adjustment for Flow Velocity and Residence Time Results of the MWTP previous field investigation (Ref. 3) indicated that the AMD flow velocity (seepage velocity) through an SRB bioreactor needs to be slower or equal to 10 ft/d, and the corresponding residence time must be longer than 12 hours, otherwise SRB might be flushed out of the medium. This condition is met for the number (three) of RCs calculated in Worksheet I. Therefore, the proper design is as in Worksheet I with the corresponding capital cost of $24,244 and NPV of $37,768 as calculated in Worksheet J. Had the criterion for the maximum velocity been, for example, 2.5 ft/d, the number of RCs needed would have been four to meet this criterion as it is shown in Worksheet K. The related capital cost and NPV would have been $32,326 and $51,429, respectively. The calculations described in this section are applicable only for a treatment system configured with parallel RCs. If the modeled treatment system is configured with RCs in a series and the velocity and/or residence time criteria are not met, the system must be redesigned. This may involve changing the system by placing the RCs in more than one series, switching to a parallel configuration, reducing the flow rate, or relaxing the cleanup standards. 5.4 Comments on the BEST Simulator and its Limitations The user may skip the examination of the

results of the simulation identified in Chart I and define the treatment scope by conducting more modeling using PHREEQCI with various carbon oxidation ratios and examining the

36

results with respect to Al and Fe removal as hydroxides. Nevertheless, the importance of the process defined in Chart I is an immediate realization of the likely costs, if metal hydroxides are to be precipitated by an SRB treatment system.

Although, as stated earlier, only "some familiarity with geochemical model PHREEQCI" is needed to successfully use the BEST simulator for the SRB bioreactor design, it is worth noting that the PHREEQCI software is a powerful tool that, if used more extensively, can significantly enhance the understanding of the geochemical processes occurring in a bioreactor. As an example, provided data availability, PHREEQCI can be used for evaluating adsorption of dissolved metals by the W/M organic medium.

For a user that is not familiar with and possibly intimidated by the PHREEQCI geochemical model, the BEST simulator can be used in evaluating "what if" scenarios in a simplified manner, i.e., without conducting PHREECI modeling. In such a case, the user may take a risk of using stoichiometric

calculations, as described in Chart I, for defining organic carbon needed for the desired sulfate reduction. Then, the user will bypass the PHREEQCI modeling step in Chart II to address the flow velocity and residence time criteria.

It is assumed that the population of SRB is maintained at a sufficiently high level to enable the formation of metal sulfides that can only be produced biochemically.

The BEST simulator includes the design of an SRB treatment system without an additional pretreatment cell for raising the pH of AMD, if desired. If such a pretreatment cell was needed, it could be designed and constructed as described by Desmier et al (Ref. 8).

Capital cost and NPV are calculated for an SRB treatment system that includes only SRB RCs. If lowering of the pH of the AMD feeding the system is needed, the cost of such a pretreatment cell, with lime or other medium that provides alkalinity, will need to be added.

37

38

Enter flow rate

Enter metalconcentrations

Stoichiometricestimate of

carbon needs (Sheet A)

Enter:Volumetric % of manure,

Labor rates

Enter:TOC % in manure,

TOC% in walnut shells,Sulfate reduction rate,

Porosity

Number of RCs in the SRBtreatment system formetals precipitating

as sulfides (Sheet B1)

Cost & Size for the SRBtreatment system formetals precipitating

as sulfides (Sheet D1)

PHREEQCImodeling(Chart I I )

AMD chemistry

AMD flow rate& its variation

Analysis of organicmatter for TOC

Tank cost(Sheet C)

For metals precipitating as sulfides For metals precipitating as sulfides and hydroxides

Enter metalconcentrations

Number of RCs in the SRBtreatment system for

all metals precipitating(Sheet B2)

Stoichiometricestimate of

carbon needs (Sheet A)

Cost & Size for the SRBtreatment system for

all metals precipitating (Sheet D2)

Selection of thealternative for PHREEQCI

modeling

Explanations:Basic input data shown in the bold-line shapesLeft and right sides of the chart need to be processed separately.Text in arrow-shaped blocks lists parameters that are entered manually to a subsequent worksheet.Other needed parameters are automatically transfered from other worksheets.

Chart I: Treatment scope selection and the preliminary design of the SRB treatment system

Figure 5-1. Chart I: Treatment scope selection and preliminary design of the SRB system.

Meets flowvelocity and resident

time criteria

NoNoNo

No

Examine values ofvelocity and resident

time in Sheet G

Yes

SRB treatment systemdesigned as inSheets G & H

Yes

No

No

Yes

Yes

Yes

Chart II: PHREEQCI modeling and the final design of the SRB treatment system

SRB treatment systemdesigned as inSheets K & L

Selected alternativefor final design(from Chart I)

Enter data forPHREEQCI

modeling

PHREEQCIinput parameters

(Sheet F)

PHREEQCI run

Meets metalsremoval criteria

SRB treatment systemdesigned as inSheets I & J

Adjustment of the number ofRCs in the SRB treatment

system

PHREEQCIinput parameters(Sheet F- altered)

PHREEQCI run

Meets metalsremoval criteria

Enter adjusted Carbon use

Number of RCs in theSRB treatment system

(Sheet I)

Meets flowvelocity and resident

time criteria

Cost & size for theSRB treatment system

(Sheet J)

Number of RCs in theSRB treatment system

(Sheet I)

Meets flowvelocity and resident

time criteria

Cost & size for theSRB treatment system

(Sheet J)

Adjustment of carbon neededto decrease sulfate concentrationor the increase of the AMD pH*

Explanations:Text in arrow-shaped blocks lists parameters that are entered manually to a subsequent worksheet,other needed parameters are automatically transfered from other worksheets.Shaded shapes mark the example described in the report.

* An additional compartment with lime or alike will be needed to achieve this goal. Cost estimation of such a compartment is not included in this design.

Sulfate reductionrate laboratory

experiment(includes analyses

for metals)

Setting criteria for:metal removal,flow velocity andresidence time

Enter datafrom the labexperiment

Sulfate reductionrate calculation

(Sheet E)

Figure 5-2. Chart II: PHREEQCI modeling and the final design of the SRB treatment system.

39

SUMMARY WORKSHEET: Input and output of the BEST simulatorExplanations: All fields but green are protected; to unprotect go to: Tools, Protection, Unprotect sheet

Input values are in italic in green fields

Entry data AMD source Strong synthetic AMD Atomic weight Concentration

Al 26.98 40.40

7.56.86.08.0

0.251.0

0.508

560.500.215400.2$60

May,30,2003$40$70

0.1016

$1,0003.2%

mg/LFe+3 55.84 0.00 mg/L

Enter other species if needed 1.00 mg/LCo 58.93 mg/LPb 207.20 mg/LNi 58.69 mg/L

Enter other species if needed 1.00 mg/LEnter other species if needed 1.00 mg/LEnter other species if needed 1.00 mg/L

Cd 112.41 0.08 mg/LCu 63.55 6.12 mg/L

Fe+2 55.84 39.48 mg/LZn 65.39 17.78 mg/L

Max (in center) ftWall ftActive medium ft

Diameter ftAssumed sulfate reduction rate mmol/(d*L)AMD feed flow rate gpm

Porosity dimensionlessTOC in fresh manure of the treatment system %TOC in fresh walnut shells of the treatment system %Volumetric moisture content of manure g/cm3

Dry bulk density of manure g/cm3

Dry bulk density of walnut shells Lb/yard3

Volumetric ratio of manure in organic matter dimensionlessLabor rateDate alteredLabor rate for bagging organic mediumLabor rate for field installationTime to make one bag with organic medium hrTime to install one RC at the site hrAnnual O&M (assumed)Discount rate

Output dataPreliminary design

Cost of a typical tank adaptation 3,813.5Number of RCs 5Years of operation for NPV calculation 30Capital cost $40,407Net present value (NPV) of capital cost $59,511Cost of a typical tank adaptation $3,814Number of RCs 17Years of operation for NPV calculation 30Capital cost $137,385Net present value (NPV) of capital cost $156,488

For metal sulfides

For all metals precipitating

Organic matter properties

Typical tank adaptation

Field installation

Economical factors

BEST simulator is not automatically linked with the PHREEQCI input data file shown in Sheet F, therefore, all alteration to this sheet must be entered manually

RC dimensions HeightMeasured from the bottom to the RC outlet

For a cuboidal tank enter value of 2(A/3.14)0.5

where:A is an area of the cuboidal tank base

Extra items need to be input separately in Sheet C that will then calculate a new value for the cost of a typical tank adaptation

Treatment system designInfluent sulfur concentration mg/LEffluent sulfur concentration mg/LVolume of the laboratory bioreactor LFlow rate mL/minCarbon oxidation required 1.263 mmol/L

PHREEQCI modeling Carbon oxidation for PHREEQCI entry 0.126 mmol/LRun PHREEQCI and check metal removal Number of RCs 1Years of operation for NPV calculation 12Capital cost $8,081Net present value (NPV) of capital cost $17,918

If an inadequate metal removal (repeat this step until an adequate removal is achieved)

Enter to PHREEQCI a larger value of carbon oxidation than before or adjust pH

2.526 mmol/LNumber of RCs 3Years of operation for NPV calculation 18 yearCapital cost $24,244Net present value (NPV) of capital cost $37,768Velocity value from the last design 3.0 ft/dResidence time from the last run 2.35 dayVelocity criterion (maximum) 2.5 ft/dResidence time criterion (minimum) 0.50 day

If velocity and or residence time do not meet requirements (repeat this step until both do)

Enter a larger value for the number of RCs

Corrected velocity 2.3 ft/dayCorrected residence time 3.13 dayYears of operation for NPV calculation 30 yearCapital cost $32,326Net present value (NPV) of capital cost $51,429

Adequate metal removal (last successful iteration)

Ad

255.2235.0

9.12.7

4

justed carbon oxidation for the adequate metal removal

Meets velocity and residence time criteria

Meets velocity and residence time criteria (last successful iteration)

Increased number of RCs with velocity and residence time criteria met

Laboratory experiment

Adequate metal removal

This value must be manually entered to PHREEQCI data input file

If RCs are configured in series, divide and mulitply the calculated residence time and velocity values, respectively, by the number of RCs

This step is applicable only for the RCs configured parallel

Figure 5-3. Worksheet 0: Input and output of the BEST simulator.

40

41

WORKSHEET A: Stoichiometric estimation for carbon needed for the given AMDExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

AMD source:Removal form Metal Atomic weight Sulfide

required to precipitate

metal sulfides(1)

Bicarbonate required to precipitate

metal hydroxides(2)

Excess sulfide made while forming bicarbonate used to

precipitate metal hydroxides(2)

(mg/L) (mmol/L) (mmol/L) (mmol/L) (mmol/L) (mg/L) (mmol/L)Al 26.98 40.40 1.4974 NA(3) 4.4922 2.2461 215.6264 2.2461

Fe+3 55.84 0.00 0.0000 NA 0.0000 0.0000 0.0000 0.0000Enter other species (4) 1.00 0.0000 NA 0.0000 0.0000 0.0000 0.0000

Co 58.93 0.00 0.0000 0.0000 NA 0.0000 0.0000 NAPb 207.20 0.00 0.0000 0.0000 NA 0.0000 0.0000 NANi 58.69 0.00 0.0000 0.0000 NA 0.0000 0.0000 NA

nter other species if neede 1.00 0.00 0.0000 0.0000 NA 0.0000 0.0000 NAnter other species if neede 1.00 0.00 0.0000 0.0000 NA 0.0000 0.0000 NAnter other species if neede 1.00 0.00 0.0000 0.0000 NA 0.0000 0.0000 NA

Cd 112.41 0.08 0.0007 0.0007 NA 0.0007 0.0641 NACu 63.55 6.12 0.0963 0.0963 NA 0.0963 9.2450 NA

Fe+2 55.84 39.48 0.7070 0.7070 NA 0.7070 67.8739 NAZn 65.39 17.78 0.2719 0.2719 NA 0.2719 26.1031 NA

Total 1.0759 4.4922 3.3220 318.9124 2.2461

(1) Net reaction for metals forming sulfides: Cd+2 + SO4-2 + 2CH2O = CdS + 2H2CO3

(2) Net reaction for metals that need pH adjustment to precipitate as hydroxides: 2Al+3 + 3SO4-2 + 6CH2O + 6H2O = 3H2S + 6H2CO3 + 2Al(OH)3

(3) Not applicable(4) All three: symbol, atomic weight and concentration must be entered

Carbon oxidation required to precipitate all metals 6.64400926 mmol C/LCarbon oxidation required to precipitate only metal sulfides 2.1517928 mmol C/L

Background information:

2Al+3 + 3SO4-2 + 6CH2O + 6H2O + 3Cd+2 = 3CdS + 6H+ + 6H2CO3 + 2Al(OH)3

Therefore, it is necessary to account for sulfide-forming metals and alkalinity requiring metals separately

Strong synthetic AMD

Sulfide precipitate

Hydrogen sulfide, a product of reaction (2), can not be used for additional precipitation of metals because such a reaction would release H+ and lower the pH as shown by the following reaction:

Sulfate reduction required to precipitate

metal sulfides or metal hydroxides

Concentration

Hydroxide precipitate

Figure 5-4. Worksheet A: Stoichiometric estimation for carbon needed for the given AMD.

WORKSHEET B1: Number of RCs in the SRB treatment system for metals precipitating as sulfidesExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Calculation of volumes for the replaceable cartridge (RC)Tank size foot inch meter

Max (in center) 7.5 90 2.29Wall 6.8 82 2.07Active Organic medium 6.0 72 1.83

Diameter 8.0 96 2.44Volume ft3 gallon m3

Rim high 341.8 2557 9.68Active organic medium 301.6 2256 8.54Total medium 347.4 2599 9.84

Calculation of the number of RCs in the SRB treatment systemCarbon oxidation required for metals removed as sulfides 2.152 mmol/LDecrease of sulfate concentration related to carbon needs 1.076 mmol/LAssumed or calculated (Sheet E) sulfate reduction rate 0.25 mmol/(d*L)Flow rate of AMD to be treated 1.0 gpmPorosity of organic medium 0.50 dimensionless

12394 gallonNumber of RCs in the treatment system 5.49 RC

Stoichiometric check of organic carbon supplyCarbon oxidation required for metals removed as sulfides 2.152 mmol/LFlow rate 1.00 gpmCarbon used per minute 8.14 mmol/minCarbon used per day 11728.13 mmol/d

4280768 mmol/year51369 g/year

51 kg/year113 Lb/year

TOC in fresh manure of the treatment system 8 % 345 LbTOC in fresh walnut shells of the treatment system 56 % 15760 LbTotal TOC present in organic medium of the treatment system 16105 LbTime of operation from carbon supply in manure 3.1 yearsTime of operation from carbon supply in walnut shells 139.5 yearsTime of operation from carbon supply in organic matter 142.5 years

AMD residence time and seepage velocityResidence time# 5641 min 94.0 hr 3.92 daySeepage velocity< 0.00126 ft/min 1.8 ft/day 0.00064 cm/s# = Volume*Porosity/Flow rate< = Length/Residence time

Carbon used per year

Height

Volume

Required volume of active organic matter

From the bottom to the RC outlet

Calculated as the "rim high" medium plus the medium supporting the lid

= (Decrease of sulfate concentration * Flow rate) / (Sulfate red. rate*Porosity)

Imported from Sheet A

Imported from Sheet A

Figure 5-5. Worksheet B1: Number of RCs in the SRB treatment system for metals precipitating as sulfides.

42

WORKSHEET B2: Number of RCs in the SRB treatment system for all metals precipitatingExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Calculation of volumes for the replaceable cartridge (RC)Tank size foot inch meter

Max (in center) 7.5 90 2.29Wall 6.8 82 2.07Active Organic medium 6.0 72 1.83

Diameter 8.0 96 2.44Volume ft3 gallon m3

Rim high 341.8 2557 9.68Active organic medium 301.6 2256 8.54Total medium 347.4 2599 9.84

Calculation of the number of RCs in the treatment systemCarbon oxidation required for the removal of all metals 6.644 mmol/LDecrease of sulfate concentration related to carbon needs 3.322 mmol/LAssumed or calculated (Sheet E) sulfate reduction rate 0.25 mmol/(d*L)Flow rate of AMD to be treated 1.0 gpmPorosity of organic medium 0.50 dimensionless

38269 gallonNumber of RCs in the treatment system 16.96 RC

Stoichiometric check of organic carbon supplyCarbon oxidation required for the removal of all metals 6.644 mmol/LFlow rate 1.00 gpmCarbon used per minute 25.15 mmol/minCarbon used per day 36212.51 mmol/d

13217565 mmol/year158611 g/year

159 kg/year349 Lb/year

TOC in fresh manure of the treatment system 8 % 1174 LbTOC in fresh walnut shells of the treatment system 56 % 53584 LbTotal TOC present in organic medium of the treatment system 54758 LbTime of operation from carbon supply in manure 3.4 yearsTime of operation from carbon supply in walnut shells 153.6 yearsTime of operation from carbon supply in organic matter 156.9 years

AMD residence time and seepage velocityResidence time# 19178 min 319.6 hr 13.32 daySeepage velocity< 0.000 ft/min 0.5 ft/day 0.00019 cm/s# = Volume*Porosity/Flow rate< = Length/Residence time

Carbon used per year

Height

Volume

Required volume of active organic matter

From the bottom to the RC outlet

Calculated as the "rim high" medium plus the medium supporting the lid

= (Decrease of sulfate concentration * Flow rate) / (Sulfate red. rate*Porosity)

Imported from Sheet A

Imported from Sheet A

Figure 5-6. Worksheet B2: Number of RCs in the SRB treatment system for all metals precipitating.

43

44

WORKSHEET C: Cost estimate for adaptation of a typical tankExplanations: All fields but green are protected; to unprotect go to: Tools, Protection, Unprotect sheet

Other data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Date: Apr, 2003Date altered: May,30,2003

DescriptionNumber Unit Unit price Source Amount Unit M.H.* Total Rate Amount

SRB VESSEL2500 Gal bulk water tank 1 ea $1,842.00 Ryan-Herco $1,842.00 $1,842.00Modifications (cutting tank to form a lid, drilling holes for connections) 1 lot 3 3 $60.00 $180.00 $180.0010" Sch. 40 PVC pipe 8 ft $8.07 Ryan-Herco $64.56 0.085 0.68 $60.00 $40.80 $105.36Modifications (drilling holes) 1 lot 1.5 1.5 $60.00 $90.00 $90.004" Sch. 40 PVC pipe 8 ft $2.20 Ryan-Herco $17.60 0.035 0.28 $60.00 $16.80 $34.40Modifications (drilling holes) 1 lot $0.00 1.5 1.5 $60.00 $90.00 $90.002" Sch. 40 PVC pipe 10 ft $0.74 Ryan-Herco $7.40 0.022 0.22 $60.00 $13.20 $20.6010"x4" PVC pipe saddle 1 ea $25.00 Ryan-Herco $25.00 1.2 1.2 $60.00 $72.00 $97.004"x2" PVC pipe bushing 1 ea $3.91 Ryan-Herco $3.91 0.25 0.25 $60.00 $15.00 $18.914"x2" PVC pipe saddle 1 ea $10.36 Ryan-Herco $10.36 0.5 0.5 $60.00 $30.00 $40.3610" PVC pipe cap 1 ea $121.60 Harrington $121.60 1.2 1.2 $60.00 $72.00 $193.604" PVC pipe cap 1 ea $4.03 Ryan-Herco $4.03 0.48 0.48 $60.00 $28.80 $32.832" PVC gate valve (threaded) 1 ea $43.58 Ryan-Herco $43.58 1.13 1.13 $60.00 $67.80 $111.382" PVC globe valve (threaded) 1 ea $188.00 Ryan-Herco $188.00 1.13 1.13 $60.00 $67.80 $255.802" PVC bulkhead fitting (thread-thread) 4 ea $22.50 Ryan-Herco $90.00 0.45 1.8 $60.00 $108.00 $198.0030 Mil. HDPE tarp 100 ft2 $2.80 McMaster-Carr $280.00 1 $60.00 $60.00 $340.005-5/8" pad-lockable draw latch 3 ea $9.43 McMaster-Carr $28.29 0.75 2.25 $60.00 $135.00 $163.29Enter an extra item $0.00 0 $0.00 $0.00Enter an extra item $0.00 0 $0.00 $0.00Enter an extra item $0.00 0 $0.00 $0.00Total $2,726.33 18.12 $1,087.20 3,813.53* Estimates of man-hours based upon The Richardson's Rapid System Process Plan Construction Estimating Standards.

Total cost ($) mat. & labor

Quantity Material cost ($) Labor time (hours) Labor cost ($)

Figure 5-7. Worksheet C: Cost estimate for adaptation of a typical tank.

WORKSHEET D1: Preliminary cost for the SRB treatment system for metals precipitating as sulfidesExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Material for one replaceable cartridge (RC)Number of bags with organic medium

Total 520Organic medium 494 Volume ft3 "Edge" size (ft)Walnut shells 26 0.67 0.87

Amount of manure and walnut shellsVolumetric Density Weight

ratio yard3 m3 (Lb/yard3) Ton (2000 Lb)Total 12.9 9.8 4.27Manure 0.2 2.4 1.9 1194 1.46Walnut shells 0.8 10.4 8.0 540 2.81

NettingMaterial Feet per 5-gal. bag feet Packaging Feet per roll Number of rolls16" net 12 6,237 Roll 1000 6Duct tape 3 1,559 Roll 180 9String 3 1,559 Roll 200 8

Material for the SRB treatment systemNumber of bags with organic medium

Total 2,599Organic medium 2,468Walnut shells 131

Amount of manure and walnut shellsyard3 m3

Total 64.3 49.2Manure 12.2 9.3Walnut shells 52.1 39.8

NettingMaterial feet Number of rolls16" net 31,187 31Duct tape 7,797 43String 7,797 39

Material cost for the SRB treatment systemMaterial Unit price ($) Number of RCs Cost ($)Tank (includes labor) Nominal design $3,813.53 5 $19,068Manure (yard3) $20.00 $244Walnut shells (ton) $200.00 $2,814Netting (roll) $58.00 $1,809Duct tape (roll) $6.50 $282String (roll) $5.00 $195Subtotal material cost $24,412

Labor for the SRB treatment systemLabor Time (hr) Hourly rate ($) Cost ($)Tank Included in materialBags with medium 0.10 $40 $10,396Field installation 16 $70 $5,600Subtotal labor $15,996

Total SRB treatment system capital cost (no transportation included) $40,407

Net present value (NPV)Operation time based on carbon supply assuming safety factor of 4 35.6 yearsAnnual O&M (assumed) $1,000

Discount rate 3.2% Operation time for NPV calculation 30 years

Net present value of total costs $59,511

Net Present Value:

Number of 5-gal bagsBag size

Number of 5-gal bags

Volume & weight

Volume

Volume & weight

Above the rim and around the inlet

Per vendor's spec.

From Sheet C

US OMB recommends using 30 year time for projects that may last longer

Figure 5-8. Worksheet D1: Preliminary cost for the SRB treatment system for metals precipitating as sulfides.

45

WORKSHEET D2: Preliminary cost for the SRB treatment system for all metals precipitatingExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Material for one replaceable cartridge (RC)Number of bags with organic medium

Total 520Organic medium 494 Volume ft3 "Edge" size (ft)Walnut shells 26 0.67 0.87

Amount of manure and walnut shellsVolumetric Density Weight

ratio yard3 m3 (Lb/yard3) Ton (2000 Lb)Total 12.9 9.8 4.27Manure 0.2 2.4 1.9 1194 1.46Walnut shells 0.8 10.4 8.0 540 2.81

NettingMaterial Feet per 5-gal. bag feet Packaging Feet per roll Number of rolls16" net 12 6,237 Roll 1000 6Duct tape 3 1,559 Roll 180 9String 3 1,559 Roll 200 8

Material for the SRB treatment systemNumber of bags with organic medium

Total 8,836Organic medium 8,391Walnut shells 446

Amount of manure and walnut shellsyard3 m3

Total 218.7 167.2Manure 41.5 31.8Walnut shells 177.2 135.5

NettingMaterial feet Number of rolls16" net 106,037 106Duct tape 26,509 147String 26,509 133

Material cost for the SRB treatment systemMaterial Unit price ($) Number of RCs Cost ($)Tank (includes labor) Nominal design $3,813.53 17 $64,830Manure (yard3) $20.00 $831Walnut shells (ton) $200.00 $9,569Netting (roll) $58.00 $6,150Duct tape (roll) $6.50 $957String (roll) $5.00 $663Subtotal material cost $83,000

Labor for the SRB treatment systemLabor Time (hr) Hourly rate ($) Cost ($)Tank Included in materialBags with medium 0.10 $40 $35,346Field installation 16 $70 $19,040Subtotal labor $54,386

Total SRB treatment system capital cost (no transportation included) $137,385

Net present value (NPV)Operation time based on carbon supply assuming safety factor of 4 39.2 yearsAnnual O&M (assumed) $1,000

Discount rate 3.2% Operation time for NPV calculation 30 years

Net present value of total costs $156,488

Net Present Value:

Number of 5-gal bagsBag size

Number of 5-gal bags

Volume & weight

Volume

Volume & weight

Above the rim and around the inlet

Per vendor's spec.

From Sheet C

US OMB recommends using 30 year time for projects that may last longer

Figure 5-9. Worksheet D2: Preliminary cost for the SRB treatment system for all metals precipitating.

46

WORKSHEET E: Calculation of sulfate reduction rateExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetData that were transferred directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Calculation of sulfate reduction rateInfluent 255.2 mg/L 2.658 mmol/LEffluent 235.0 mg/L 2.448 mmol/LInfluent 765.6 mg/L 7.975 mmol/LEffluent 705.0 mg/L 7.344 mmol/L

Decrease in sulfur concentration 20.2 mg/L 0.210 mmol/LDecrease in sulfate concentration 60.6 mg/L 0.631 mmol/LCarbon oxidation required 15.2 mg/L 1.263 mmol/LCarbon for PHREEQCI if reaction in 10 steps 0.126 mmol/LReactor volume 9.1 L 0.320 ft3 2.40 gallonsMedium porosity 0.50 dimensionlessFlow rate 2.7 mL/min 3.888 L/d 0.000713 gpm

0.541 mmol/(d*L)0.541 mol/(d*m3)

Calculation of annual carbon useCarbon oxidation required 1.263 mmol/LDaily carbon use 4.909 mmol/d 59 mg/d

21500 mg/y21 g/y

0.021 kg/y0.047 Lb/y

Calculator to convert flow rate in GPM to other units*Instruction: Use only the yellow field to enter the flow rate in gpm

gallons ft3 mL L m3

Flow cubic unit /s 1.7E-02 2.2E-03 4.4E+00 6.3E-02 6.3E-05Flow cubic unit /min 1.0E+00 1.3E-01 2.6E+02 3.8E+00 3.8E-03Flow cubic unit /hr 6.0E+01 8.0E+00 1.6E+04 2.3E+02 2.3E-01Flow cubic unit /day 1.4E+03 1.9E+02 3.8E+05 5.5E+03 5.5E+00Flow cubic unit /year 5.3E+05 7.0E+04 1.4E+08 2.0E+06 2.0E+03* This calculator is not integrated with any other table in this workbook and is given only for the user's convenience

Try until it corresponds to the value in the unit of your measurement

Sulfur concentration (e.g. by ICP)

Sulfate concentration

Sulfate reduction rate

Annual carbon use 1,791.639 mmol/y

SRR=Sulfate concentration decrease [mmol/L] * flow rate [L/d] / (Reactor volume [L] * Porosity)

This value must be divided by 10 before it is manually input in Sheet F

Figure 5-10. Worksheet E: Calculation of SRR.

47

Figure 5-11. Worksheet F: PHREEQCI input file.

48

WORKSHEET G: Number of RCs in the SRB treatment system based of PHREEQCIExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Calculation of volumes for the replaceable cartridge (RC)Tank size foot inch meter

Max (in center) 7.5 90 2.29Wall 6.8 82 2.07Active Organic medium 6.0 72 1.83

Diameter 8.0 96 2.44Volume ft3 gallon m3

Rim high 341.8 2557 9.68Active organic medium 301.6 2256 8.54Total medium 347.4 2599 9.84

Calculation of the number of RCs in the SRB treatment systemCarbon oxidation required for metals as used in PHREEQCI 1.263 mmol/LDecrease of sulfate concentration related to carbon needs 0.631 mmol/LCalculated sulfate reduction rate (form Sheet E) 0.54 mmol/(d*L)Flow rate of AMD to be treated 1.0 gpmPorosity of organic medium 0.50 dimensionless

3360 gallonNumber of RCs in the treatment system 1.49 RC

Check of organic carbon supplyCarbon oxidation required for metals removed as sulfides 1.263 mmol/LFlow rate 1.00 gpmCarbon used per minute 4.78 mmol/minCarbon used per day 6881.13 mmol/d

2511612 mmol/year30139 g/year

30 kg/year66 Lb/year

TOC in fresh manure of the treatment system 8 % 69 LbTOC in fresh walnut shells of the treatment system 56 % 3152 LbTotal TOC present in organic medium of the treatment system 3221 LbTime of operation from carbon supply in manure 1.0 yearsTime of operation from carbon supply in walnut shells 47.5 yearsTime of operation from carbon supply in organic matter 48.6 years

AMD residence time and seepage velocityResidence time# 1128 min 18.8 hr 0.78 daySeepage velocity< 0.006 ft/min 9.0 ft/day 0.00319 cm/s# = Volume*Porosity/Flow rate< = Length/Residence time

Carbon used per year

Height

Volume

Required volume of active organic matter

Measured from the bottom to the RC outlet

Calculated as the "rim high" medium plus the medium supporting the lid

= (Decrease of sulfate concentration * Flow rate) / (Sulfate red. rate*Porosity)

Imported from Sheet E

Imported from Sheet E

Figure 5-12. Worksheet G: Number of RCs in the SRB treatment system based on PHREEQCI.

49

WORKSHEET H: Cost for the SRB treatment system designed using PHREEQCIExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fieldResults that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Material for one replaceable cartridge (RC)Number of bags with organic medium

Total 520Organic medium 494 Volume ft3 "Edge" size (ft)Walnut shells 26 0.67 0.87

Amount of manure and walnut shellsVolumetric Density Weight

ratio yard3 m3 (Lb/yard3) Ton (2000 Lb)Total 12.9 9.8 4.27Manure 0.2 2.4 1.9 1194 1.46Walnut shells 0.8 10.4 8.0 540 2.81

NettingMaterial Feet per 5-gal. bag feet Packaging Feet per roll Number of rolls16" net 12 6,237 Roll 1000 6Duct tape 3 1,559 Roll 180 9String 3 1,559 Roll 200 8

Material for the SRB treatment systemNumber of bags with organic medium

Total 520Organic medium 494Walnut shells 26

Amount of manure and walnut shellsyard3 m3

Total 12.9 9.8Manure 2.4 1.9Walnut shells 10.4 8.0

NettingMaterial feet Number of rolls16" net 6,237 6Duct tape 1,559 9String 1,559 8

Material cost for the bioreactor treatment systemMaterial Unit price ($) Number of RCs Cost ($)Tank (includes labor) Nominal design $3,813.53 1 $3,814Manure (yard3) $20.00 $49Walnut shells (ton) $200.00 $563Netting (roll) $58.00 $362Duct tape (roll) $6.50 $56String (roll) $5.00 $39Subtotal material cost $4,882

Labor for the bioreactor treatment systemLabor Time (hr) Hourly rate ($) Cost ($)Tank Included in materialBags with medium 0.10 $40 $2,079Field installation 16 $70 $1,120Subtotal labor $3,199

Total SRB treatment system capital cost (no transportation included) $8,081

Net present value (NPV)Operation time based on carbon supply assuming safety factor of 4 12.1 yearsAnnual O&M (assumed) $1,000

Discount rate 3.2% Operation time for NPV calculation 12 years

Net present value of total costs $17,918

Net Present Value:

Number of 5-gal bagsBag size

Number of 5-gal bags

Volume & weight

Volume

Volume & weight

Above the rim and around the inlet

Per vendor's spec.

From Sheet C

US OMB recommends using 30 year time for projects that may last longer

Figure 5-13. Worksheet H: Cost for the SRB treatment system designed using PHREEQCI.

50

WORKSHEET I: Number of RCs for the SRB treatment system designed using PHREEQCI with carbon adjustedExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Calculation of volumes for the replaceable cartridge (RC)Tank size foot inch meter

Max (in center) 7.5 90 2.29Wall 6.8 82 2.07Active Organic medium 6.0 72 1.83

Diameter 8.0 96 2.44Volume ft3 gallon m3

Rim high 341.8 2557 9.68Active organic medium 301.6 2256 8.54Total medium 347.4 2599 9.84

Calculation of the number of RCs in the SRB treatment system2.526 mmol/L

Decrease of sulfate concentration related to carbon needs 1.263 mmol/LSulfate reduction rate (from Sheet E) 0.54 mmol/(d*L)Flow rate of AMD to be treated 1.0 gpmPorosity of organic medium 0.50 dimensionless

6723 gallonNumber of RCs in the treatment system 3.0 RC

Check of organic carbon supplyCarbon oxidation based on decrease in sulfate concentration 2.526 mmol/LFlow rate 1.00 gpmCarbon used per minute 9.56 mmol/minCarbon used per day 13767.71 mmol/d

5025214 mmol/year60303 g/year

60 kg/year133 Lb/year

TOC in fresh manure of the treatment system 8 % 207 LbTOC in fresh walnut shells of the treatment system 56 % 9456 LbTotal TOC present in organic medium of the treatment system 9663 LbTime of operation from carbon supply in manure 1.6 yearsTime of operation from carbon supply in walnut shells 71.3 yearsTime of operation from carbon supply in organic matter 72.8 years

AMD residence time and seepage velocityResidence time# 3384 min 56.4 hr 2.35 daySeepage velocity< 0.002 ft/min 3.0 ft/day 0.00106 cm/s# = Volume*Porosity/Flow rate< = Length/Residence time

Carbon used per year

Height

Volume

Required volume of active organic matter

Adjusted carbon required for metals removal based on PHREEQCI

Measured from the bottom to the RC outlet

Calculated as the "rim high" medium plus the medium supporting the lid

= (Decrease of sulfate concentration * Flow rate) / (Sulfate red. rate*Porosity)

Figure 5-14. Worksheet I: Number of RCs in the SRB treatment system designed using PHREEQCI with carbon adjusted.

51

WORKSHEET J: Cost for the SRB treatment system designed using PHREEQCI with carbon oxidation adjustedExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Material for one replaceable cartridge (RC)Number of bags with organic medium

Total 520Organic medium 494 Volume ft3 "Edge" size (ft)Walnut shells 26 0.67 0.87

Amount of manure and walnut shellsVolumetric Density Weight

ratio yard3 m3 (Lb/yard3) Ton (2000 Lb)Total 12.9 9.8 4.27Manure 0.2 2.4 1.9 1194 1.46Walnut shells 0.8 10.4 8.0 540 2.81

NettingMaterial Feet per 5-gal. bag feet Packaging Feet per roll Number of rolls16" net 12 6,237 Roll 1000 6Duct tape 3 1,559 Roll 180 9String 3 1,559 Roll 200 8

Material for the SRB treatment systemNumber of bags with organic medium

Total 1,559Organic medium 1,481Walnut shells 79

Amount of manure and walnut shellsyard3 m3

Total 38.6 29.5Manure 7.3 5.6Walnut shells 31.3 23.9

NettingMaterial feet Number of rolls16" net 18,712 19Duct tape 4,678 26String 4,678 23

Material cost for the bioreactor treatment systemMaterial Unit price ($) Number of RCs Cost ($)Tank (includes labor) Nominal design $3,813.53 3 $11,441Manure (yard3) $20.00 $147Walnut shells (ton) $200.00 $1,689Netting (roll) $58.00 $1,085Duct tape (roll) $6.50 $169String (roll) $5.00 $117Subtotal material cost $14,647

Labor for the bioreactor treatment systemLabor Time (hr) Hourly rate ($) Cost ($)Tank Included in materialBags with medium 0.10 $40 $6,237Field installation 16 $70 $3,360Subtotal labor $9,597

Total SRB treatment system capital cost (no transportation included) $24,244

Net present value (NPV)Operation time based on carbon supply assuming safety factor of 4 18.2 yearsAnnual O&M (assumed) $1,000

Discount rate 3.2% Operation time for NPV calculation 18 years

Net present value of total costs $37,768

Net Present Value:

Number of 5-gal bagsBag size

Number of 5-gal bags

Volume & weight

Volume

Volume & weight

Above the rim and around the inlet

Per vendor's spec.

From Sheet C

US OMB recommends using 30 year time for projects that may last longer

Figure 5-15. Worksheet J: Cost for the SRB treatment system designed using PHREEQCI with carbon oxidation adjusted.

52

Explanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheetAll data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Calculation of volumes for the replaceable cartridge (RC)Tank size foot inch meter

Max (in center) 7.5 90 2.29Wall 6.8 82 2.07Active Organic medium 6.0 72 1.83

Diameter 8.0 96 2.44Volume ft3 gallon m3

Rim high 341.8 2557 9.68Active organic medium 301.6 2256 8.54Total medium 347.4 2599 9.84

Calculation of the number of RCs in the SRB treatment systemDecrease of sulfate concentration related to carbon needs 1.263 mmol/LSulfate reduction rate (from sheet E) 0.54 mmol/(d*L)Enter flow rate of AMD to be treated 1.0 gpmEnter porosity of organic medium 0.50 dimensionless

6723 gallon4 RC

Check of organic carbon supplyCarbon oxidation based on decrease in sulfate concentration 2.526 mmol/LFlow rate 1.00 gpmCarbon used per minute 9.56 mmol/minCarbon used per day 13767.71 mmol/d

5025214 mmol/year60303 g/year

60 kg/year133 Lb/year

TOC in fresh manure of the treatment system 8 % 276 LbTOC in fresh walnut shells of the treatment system 56 % 12608 LbTotal TOC present in organic medium of the treatment system 12884 LbYears of operation from carbon supply in manure 34.5 yearsYears of operation from carbon supply in walnut shells 1576.0 yearsYears of operation from carbon supply in organic matter 1610.5 years

AMD residence time and seepage velocityResidence time# 4512 min 75.2 hr 3.13 daySeepage velocity< 0.002 ft/min 2.3 ft/day 0.00080 cm/s# = Volume*Porosity/Flow rate< = Length/Residence time

WORKSHEET K: Seepage velocity and resident time for SRB treatment system with the increased number of RCs to meet the velocity and residence time criteria

Height

Volume

Carbon used per year

Required volume of active organic matterEnter number of RCs in the treatment system

Meassured from the bottom to the RC outlet

Calculated as the "rim high" medium plus the medium supporting the lid

= (Decrease of sulfate concentration * Flow rate) / (Sulfate red. rate*Porosity)

Must be greater than that determined through PHREEQCI modeling

Imported from Sheet I

Figure 5-16. Worksheet K: Seepage velocity and resident time for SRB treatment system with the increased number of RCs to meet the velocity and residence time criteria.

53

WORKSHEET L: Cost for the SRB treatment system with the correction for flow velocityExplanations: All fields are protected; to unprotect go to: Tools, Protection, Unprotect sheet

All data used in this sheet are entered through BEST summary sheetInput data that were transferred to this sheet directly from the BEST summary sheet are in italic and in yellow fields Results that were calculated within this sheet and were exported to other sheets are in orange fieldValues imported from other sheets are in red fields

Material for one replaceable cartridge (RC)Number of bags with organic medium

Total 520Organic medium 494 Volume ft3 "Edge" size (ft)Walnut shells 26 0.67 0.87

Amount of manure and walnut shellsVolumetric Density Weight

ratio yard3 m3 (Lb/yard3) Ton (2000 Lb)Total 12.9 9.8 4.27Manure 0.2 2.4 1.9 1194 1.46Walnut shells 0.8 10.4 8.0 540 2.81

NettingMaterial Feet per 5-gal. bag feet Packaging Feet per roll Number of rolls16" net 12 6,237 Roll 1000 6Duct tape 3 1,559 Roll 180 9String 3 1,559 Roll 200 8

Material for the SRB treatment systemNumber of bags with organic medium

Total 2,079Organic medium 1,974Walnut shells 105

Amount of manure and walnut shellsyard3 m3

Total 51.5 39.3Manure 9.8 7.5Walnut shells 41.7 31.9

NettingMaterial feet Number of rolls16" net 24,950 25Duct tape 6,237 35String 6,237 31

Material cost for the SRB treatment systemMaterial Unit price ($) Number of RCs Cost ($)Tank (includes labor) Nominal design $3,813.53 4 $15,254Manure (yard3) $20.00 $195Walnut shells (ton) $200.00 $2,251Netting (roll) $58.00 $1,447Duct tape (roll) $6.50 $225String (roll) $5.00 $156Subtotal material cost $19,529

Labor for the SRB treatment systemLabor Time (hr) Hourly rate ($) Cost ($)Tank Included in materialBags with medium 0.10 $40 $8,317Field installation 16 $70 $4,480Subtotal labor $12,797

Total SRB treatment system capital cost (no transportation included) $32,326

Net present value (NPV)Operation time based or carbon supply assuming safety factor of 4 402.6 yearsAnnual O&M (assumed) $1,000

Discount rate 3.2% Operation time for NPV calculation 30 years

Net present value of total costs $51,429

Net Present Value:

Number of 5-gal bagsBag size

Number of 5-gal bags

Volume & weight

Volume

Volume & weight

Above the rim and around the inlet

Per vendor's spec.

From Sheet C

US OMB recommends using 30 year time for projects that may last longer

Figure 5-17. Worksheet L: Cost for the SRB treatment system with the correction for flow velocity.

54

Sele

ctiv

e PH

REE

QC

I out

put f

ile (R

-V e

xam

ple.

out)

save

d as

Exc

el s

heet

(All

info

rmat

ion

are

pres

ente

d as

in th

e or

igin

al fi

le w

ith th

e ex

cept

ion

of fi

rst t

wo

lines

and

Exp

lana

tions

of s

ymbo

ls g

iven

bel

low

the

tabl

e)

sim

sta

te

s

oln

dist

_x

t

ime

ste

p

pH

p

e

pct

_err

A

l

Cd

Cu(

2)

Zn

1

i_

soln

0-9

9-9

9-9

92.

660

1.91

784

1.50

E-03

6.68

E-07

9.64

E-0

52.

72E

-04

1

i_

soln

1-9

9-9

9-9

97

40.

0548

047

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

01

reac

t0

-99

01

2.65

922

3.77

158

1.93

137

1.50

E-03

6.68

E-07

3.23

E-0

52.

72E

-04

1

ad

vect

11

1010

2.65

922

3.77

158

1.93

137

1.50

E-03

6.68

E-07

3.23

E-0

52.

72E

-04

1

ad

vect

22

1010

2.66

717

2.20

281

1.94

586

1.50

E-03

1.89

E-09

1.11

E-1

72.

72E

-04

1

ad

vect

33

1010

2.68

512

2.11

838

1.96

121.

50E-

031.

66E-

098.

53E

-18

2.72

E-0

41

adve

ct4

410

102.

7034

52.

0680

11.

9767

41.

50E-

031.

58E-

091.

17E

-17

2.72

E-0

41

adve

ct5

510

102.

7221

82.

0270

61.

9924

91.

50E-

031.

50E-

091.

30E

-17

2.72

E-0

41

adve

ct6

610

102.

7412

81.

9904

32.

0084

61.

50E-

031.

44E-

091.

35E

-17

2.72

E-0

41

adve

ct7

710

102.

7511

61.

9721

32.

0236

71.

50E-

031.

41E-

091.

36E

-17

2.41

E-0

41

adve

ct8

810

102.

7618

81.

9530

42.

0391

91.

50E-

031.

38E-

091.

37E

-17

2.10

E-0

41

adve

ct9

910

102.

7741

31.

9320

22.

0550

91.

50E-

031.

34E-

091.

38E

-17

1.83

E-0

41

adve

ct10

1010

102.

7880

31.

9089

92.

0713

81.

50E-

031.

31E-

091.

39E

-17

1.60

E-0

4

S(-2

)

S

(6)

Fe(

2)

F

e(3)

ZnS

(A)

d

_ZnS

(A)

Gre

enoc

kite

d_G

reen

ocki

te

Cov

ellit

e d

_Cov

ellit

e A

l(OH

)3(a

)d_

Al(O

H)3

(a)

Ferri

hydr

ited_

Ferri

hydr

ite0.

00E

+00

7.98

E-03

7.08

E-0

40.

00E+

000.

00E

+00

0.00

E+0

00.

00E

+00

0.00

E+00

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

000.

00E

+00

2.00

E-03

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

000.

00E+

000.

00E+

000.

00E

+00

0.00

E+0

00.

00E

+00

0.00

E+00

1.13

E-17

7.92

E-03

7.08

E-0

43.

16E

-12

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

006.

28E

-05

6.28

E-05

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

001.

13E-

177.

92E-

037.

08E

-04

3.16

E-1

20.

00E

+00

0.00

E+0

00.

00E

+00

0.00

E+00

6.28

E-0

46.

28E-

050.

00E

+00

0.00

E+0

00.

00E

+00

0.00

E+00

2.89

E-05

7.85

E-03

7.08

E-0

48.

52E

-14

0.00

E+0

00.

00E

+00

5.99

E-0

66.

66E

-07

3.02

E-0

43.

36E-

050.

00E

+00

0.00

E+0

00.

00E

+00

0.00

E+00

9.19

E-05

7.79

E-03

7.08

E-0

47.

03E

-14

0.00

E+0

00.

00E

+00

1.78

E-0

92.

23E

-10

9.90

E-1

71.

24E-

170.

00E

+00

0.00

E+0

00.

00E

+00

0.00

E+00

1.55

E-04

7.73

E-03

7.08

E-0

46.

28E

-14

0.00

E+0

00.

00E

+00

6.22

E-1

08.

89E

-11

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

002.

18E-

047.

66E-

037.

08E

-04

5.73

E-1

40.

00E

+00

0.00

E+0

04.

21E

-10

7.02

E-1

10.

00E+

000.

00E+

000.

00E

+00

0.00

E+0

00.

00E

+00

0.00

E+00

2.81

E-04

7.60

E-03

7.08

E-0

45.

29E

-14

0.00

E+0

00.

00E

+00

3.17

E-1

06.

33E

-11

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

003.

12E-

047.

54E-

037.

08E

-04

5.07

E-1

41.

27E

-04

3.17

E-05

1.23

E-1

03.

09E

-11

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

003.

45E-

047.

47E-

037.

08E

-04

4.86

E-1

49.

08E

-05

3.03

E-05

9.63

E-1

13.

21E

-11

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

003.

81E-

047.

41E-

037.

08E

-04

4.64

E-1

45.

36E

-05

2.68

E-05

7.04

E-1

13.

52E

-11

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

004.

21E-

047.

35E-

037.

08E

-04

4.41

E-1

42.

34E

-05

2.34

E-05

3.83

E-1

13.

83E

-11

0.00

E+00

0.00

E+00

0.00

E+0

00.

00E

+00

0.00

E+0

00.

00E+

00

Exp

lana

tions

:C

once

ntra

tion

are

expr

esse

d as

mm

ol/k

g (s

ame

as m

mol

/L)

step

Adv

ectio

n sh

ift n

umbe

rsi

mS

imul

atio

n nu

mbe

rpH

Neg

ativ

e lo

garit

hm o

f hyd

roge

n io

n ac

tivity

stat

eTy

pe o

f cal

cula

tion

pER

edox

pot

entia

lI_

soln

Seq

uent

ial n

umbe

r of t

he s

olut

ion

used

pc

t_er

rC

harg

e ba

lanc

e er

ror

reac

tP

rogr

am a

cnow

lege

men

t tha

t a re

actio

n (c

arbo

on o

xida

tion)

was

use

d in

the

sim

ulat

ion

(-2)

vala

nce

(cha

rge)

of t

he io

ndi

st_x

Rea

ltive

dis

tanc

e al

ong

the

flow

with

in th

e m

odel

ed b

iore

acto

rd

Diff

eren

ce in

con

cent

ratio

n be

twee

n tw

o "a

dvec

tive"

cel

ls

-99

Pro

gram

not

atio

n (a

)A

mor

phou

sA

dvec

Adv

etiv

e tra

nspo

rt (fl

ow)

See

PHR

EEQ

C m

auna

l for

mor

e ex

plan

atio

nstim

eTh

he a

dvec

tion

shift

num

ber

Figu

re 5

-18.

PH

RE

EQ

CI s

elec

ted

outp

ut fi

le (R

-V e

xam

ple.

out)

save

d as

Exc

elT

M w

orks

heet

.

55

Zn concentration vs. carbon oxidation

for AMD of pH 2.66

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6

Carbon oxidation (mmol/L)

Zn (m

g/L)

Zn mg/L

Zn concentration vs. carbon oxidationfor AMD of pH 4

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6

Carbon oxidation (mmol/L)

Zn (m

g/L)

Zn mg/L

Zn concentration vs. carbon oxidationfor AMD of pH 5

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6

Carbon oxidation (mmol/L)

Zn (m

g/L)

Zn mg/L

Zn concentration vs. carbon oxidationfor AMD of pH 6

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6

Carbon oxidation (mmol/L)Zn

(mg/

L)

Zn mg/L

Zn concentration vs. pHfor carbon oxidation 1.26 mmol/L

0

2

4

6

8

10

12

14

16

18

20

2 3 4 5 6 7

Initial pH of AMD

Zn (m

g/L)

Zn mg/L

Concentration in the feed

Zn concentration vs. pHfor carbon oxidation 2.52 mmol/L

0

2

4

6

8

10

12

14

16

18

20

2 3 4 5 6 7

Initial pH of AMD

Zn (m

g/L)

Zn mg/L

Concentration in the feed

Zn concentration vs. pHfor carbon oxidation 3.78 mmol/L

0

2

4

6

8

10

12

14

16

18

20

2 3 4 5 6 7

Initial pH of AMD

Zn (m

g/L)

Zn mg/L

Concentration in the feed

Zn concentration vs. pHfor carbon oxidation 5.04 mmol/L

0

2

4

6

8

10

12

14

16

18

20

2 3 4 5 6 7

Initial pH of AMD

Zn (m

g/L)

Zn mg/L

Concentration in the feed

Figure 5-19. Results of PHREEQCI modeling with different ratios of carbon oxidation and the feed pH.

56

6. Summary of Quality Assurance Activities 6.1 Background Following is a summary of the QA activities associated with the project. A binder containing the test plan, copies of field logbooks/data, and analytical data was reviewed. All field and laboratory data available has been evaluated to determine the usability of the data. Sulfate as sulfur measurements were classified as the only critical analysis for this project. A critical analysis is an analysis that must be performed in order to determine if project objectives were achieved. Data from noncritical analyses were also evaluated. Several strategies were used to measure sulfate; however, the one that produced the most reliable results was sulfur determined by ICP-atomic emission spectroscopy (AES) and subsequent calculation of sulfate as sulfur. Sulfate data determined by other methods was removed from further consideration. 6.2 Data Evaluation The data quality indicator objectives for critical measurements were those typically used by analytical laboratories for determining data quality. These guidelines were compatible with project objectives and the methods of determination being used. The data quality indicator objectives were method detection limits (MDLs) accuracy, precision, and completeness. Control limits for each of these objectives are summarized in Table 6-1. 6.3 Validation Procedures Data that was generated throughout the project was validated. The purpose of data validation is to determine the usability of data that was generated during a project. 6.3.1 Analytical Evaluation An analytical evaluation is performed to determine the following:

− that all analyses were performed within specified holding times;

− that calibration procedures were followed correctly by field and laboratory personnel;

− that laboratory analytical blanks contain no significant contamination;

− that all necessary independent check standards were prepared and analyzed at the proper frequency and that all remained within control limits;

− that duplicate sample analysis was performed at the proper frequency and that all RPDs were within specified control limits; and

− that matrix spike sample analysis was performed at the proper frequency and that all spike recoveries (%R) were within specified control limits.

Measurements that fall outside of the control limits specified in the quality assurance project plan (QAPP), or for other reasons are judged to be outlier, were flagged appropriately to indicate that the data is judged to be estimated or unusable. An analytical evaluation was performed to determine the usability data that was generated by the HKM Laboratory for the project. Laboratory data validation was performed using USEPA Contract Laboratory Program National Functional Guidelines for Inorganics Data Review (Ref. 9) as a guide. The QC criteria outlined in the QAPP were also used to identify outlier data and to determine the usability of the data for each analysis. Only one batch of solid samples was flagged for low iron spike recovery (see Table 6-2). For aqueous samples, accuracy could not often be quantified because the matrix spikes added to the samples were not high enough for iron, manganese, and zinc analyses. 6.3.2 Duplicate Experiments Two duplicate experiments were performed. Column IV was a duplicate of column II. Column VI was a duplicate of column V. A comparison of results for the critical sulfur measurement by ICP-AES is presented in Table 6-3.

57

During the course of the project, several issues arose about the best way to obtain reliable data for sulfate. Traditional wet chemistry sulfate procedures and HACH kit analyses did not prove to be reliable with this specific sample matrix. Therefore, these data were not used to support conclusions.

With the exception of the first sample for dissolved sulfur for Column II and Column IV, all of the values show very good agreement for duplicate columns, indicating consistent column preparation and sampling procedures. 6.3.3 Logbook Review

Questionable TOC results on solid sample materials early in the project also resulted in this data being discarded from further interpretation.

The project logbook was reviewed to determine the usability of the field data. In the early stages of the project, the calibration information was not recorded in the logbook. Entries indicated that the instruments had been calibrated but lacked crucial details such as the buffers/standards used for the calibration. During the rest of the study, this information was routinely recorded.

6.4 Summary All data generated at MSE and the HKM Laboratory has been validated according to EPA guidelines. Some of the data was flagged for various reasons and is summarized in Table 6-3.

6.3.4 Data Review The importance of documenting calibration of field

meters should be reiterated to sampling personnel. MWTP Activity III, Project 24 presented unique challenges for the sampling and analytical team, particularly with respect to generating reliable sulfate data and TOC data. On a positive note, the data was very organized, which facilitated the data evaluation process.

The analytical data for the project was well organized. All issues associated with the data were documented on the original data packages from the laboratory, as well as any discussions with the HKM Laboratory, including the proposed path forward to obtain more reliable data.

Table 6-1. Data Quality Indicator Objectives

Parameter Unit MDL1 Precision2 Accuracy3 Completeness4

Sulfate as S mg/L 0.02 20% RPD 75% - 125% recovery 95% 1Method detection limit’s are based on what is achievable by the methods and what is necessary to achieve

project objectives and account for anticipated dilutions to eliminate matrix interferences. Method detection limits will be adjusted as necessary when dilutions of concentrated samples are required.

2Relative percent difference (RPD) of analytical sample duplicates. 3Percent recovery of matrix spike, unless otherwise indicated. 4Based on number of valid measurements compared to the total number of samples.

58

Table 6-2. Summary of Qualified Data for MWTP Activity III, Project 24

Date1 Sample ID Analysis QC Criteria

Control Limit Result Flag2 Comment

02/19/02 WAL021902 MAN021902

Total Iron Matrix Spike

75% - 125% recovery

44.1% J Flag samples “J” for out-of-control matrix spike.

1 Date that the samples were collected.2 Data Qualifier Definitions: U-The material was analyzed for, but was not detected above the level of the associated value (quantitation or detection limit). J-The sample results are estimated. R-The sample results are unusable. UJ-The material was analyzed for, but was not detected, and the associated value is estimated.

Table 6-3. Comparison of Results for Duplicate Column Experiments

Analysis Date Column II Column IV RPD Column V Column VI RPD Dissolved S 10/16/01 80.7 55.6 36.8 209 222 6.0 Dissolved S 11/16/01 62.7 60.0 22.5 237 238 0.4 Dissolved S 1/16/02 53.8 51.1 5.1 227 229 0.9 Dissolved S 2/08/02 57.1 60.0 5.0 No data No data N/A Dissolved S 3/18/02 62.2 58.3 6.5 238 220 7.9 S in solid material 4/08/02 1610 mg/kg 1640 mg/kg 1.8 1890 mg/kg 1570 mg/kg 18.5

59

7. Conclusions The following conclusions were derived based on the project work. Literature data identified 36 organic substrates

that were used for SRB technology. All information gathered during the literature search is contained in the database assembled using Microsoft AccessTM (Appendix A).

Of the 36 substrates, 29 are considered indirect organic substrate since they require decomposition by other microorganisms to provide SRB nutrition. These substrates require complex microbial communities to degrade the organic matter and support SRB growth. Examples of such substrates are composts; wood/paper wastes; food production byproducts like molasses, cheese whey, and potato; hay; straw; manure; and sewage sludge.

The remaining seven substrates are considered direct organic substrates because they do not require other microorganisms to be usable by SRB. Such substrates include methanol, ethanol, acetate, lactate, formate, pyruvate, and sucrose.

Overall, substrate mixtures containing both easily biodegradable materials and more recalcitrant materials are the most effective for supporting SRB growth. The easily biodegradable substrate ensures a quick start of the bioreactor. More recalcitrant materials provide the best long-term bioreactor performance.

The substrate mixture should also provide adequate surface area for biofilm development, buffering and adsorption capacity, and adequate hydraulic conductivity.

The suitability of a substrate mixture for treating a particular composition of AMD is best determined empirically using laboratory-scale tests.

A new organic mix (medium) that contains walnut shells and cow manure was developed for the project. This indirect organic substrate satisfies most of the conditions defined through the literature study for an efficient mixture to treat AMD. The main advantages of using this mix include:

– cow manure is an easily biodegradable organic matter that ensures a quick startup of the bioreactor and includes nitrogen needed by other microorganisms for the initial manure decomposition;

– walnut shells are more recalcitrant to biodegradation, thus they support good long-term operation of a bioreactor. They also provide a solid matrix structure because individual shells actually rest on each other. This structure prevents time-driven compaction (settling), thus works toward preservation of the initial permeability of the medium;

– the new mix is rich in organic carbon since the walnut shells contain 56% of the TOC and the TOC of manure ranges from 8% to 20%, depending on the manure source; and

– the sustainable hydraulic conductivity of the 0.8/0.2 W/M organic medium for the horizontal flow is 0.01 cm/s or higher.

The long and short duration tests for hydraulic

conductivity of the new organic medium revealed and provided the following findings:

– the hydraulic conductivity value of the 0.5/0.5 W/M organic medium was 1 order of magnitude smaller than the hydraulic conductivity value for the 0.8/02 mix;

– the long-term hydraulic conductivity value for the flow in a vertical direction decreases very rapidly and may approach 0 value. This decrease is attributed to the deformation of the W/M organic medium of which the finest particles are mobilized by the flowing water and migrate

60

downward by gravity to settle at a certain level, usually the bottom of the container, blocking the flow. In a horizontal configuration, the migrating particles also settle in the bottom of the container, but they do not block the entry of water that flows above them as it is fed laterally;

– the sustainable hydraulic conductivity of the most permeable, 0.8/0.2 W/M, organic medium configured for the horizontal flow is 0.01 cm/s or higher;

– the short duration laboratory tests provided the highest hydraulic conductivity values that proved to be not applicable for the deforming W/M organic medium used for these tests; and

– in general, for the deformable W/M organic medium that has a potential of its finest particles migrating downward by gravity, the vertical flow orientation is a recipe for plugging the bioreactor. The AMD flow through a bioreactor filled with such a medium must be in a horizontal, or close to horizontal, direction.

The SRR values ranged from 0.17 mol/(d*m3)

to 0.79 mol/(d*m3) with the overall mean value of 0.40 mol/(d*m3). These values were determined by laboratory experiment conducted using six 5-gallon buckets operating as bioreactors at three different temperatures and fed with two different AMDs.

The SRR values acquired from the laboratory experiment seemed to be independent of the strength of the influent and the temperature at which the experiment was conducted.

The large range of SRR values measured during the laboratory experiment indicates the need for conducting the experiment at least in triplicates.

Analytical work for TOC did not provide consistent results regarding the depletion of organic carbon in the 5-gallon bucket bioreactors. As it turned out, the laboratory experienced technical problems with analysis

of the mixed matrix of walnut shells and manure.

The recommended design of the RC uses commercially available cylindrical or cuboidal plastic tanks most often built of HDPE or polypropylene. Such a tank needs to be equipped with necessary features to accommodate the W/M organic medium and serve as one SRB bioreactor. Such adaptation work needs to be performed prior to the transportation of the tank to a mine site. At the mine site, the tank will be installed either aboveground or belowground, as required by the site conditions, and an appropriate piping system will be installed to convey the AMD into the RC.

At the mine site, the RC will be filled with 5-gallon bags of W/M organic medium that can be prepared and then transported to the mine site, or they can be made at the mine site. The bags, made of plastic netting commonly used by grocery shops for prepacked fruits, have loops in their top portion to facilitate the placement and the removal of the bags from the RC using a rod with a hook.

Built into the RC are a cleanup port, a sump for precipitate, an overflow outlet, and a valve to control the flow. Anaerobic conditions are created by a tarp placed on the top of organic medium. A rigid lid with locks placed on the top of the RC protects the RC from atmospheric precipitation and vandalism.

Cost of production (excluding transportation to the site and site installation) of such an RC housed in a 2,500-gallon HDPE tank is $8,081. The cost may vary depending on local supply and labor rates applicable at the given location.

A modular SRB treatment system consists of a number of RCs that are configured in parallel or in a series depending on the AMD flow rate and its quality (metal load and pH), cleanup objectives, and space available at the given mine site. Both, the configuration and the number of RCs are determined through the

61

BEST computer simulator developed for this project.

The time of operation calculated by the BEST is based on the available carbon present in W/M organic medium divided by the safety factor of 4. This safety factor is used because the investigations conducted for the project did not focus on confirming whether the organic carbon present in the medium is entirely available for SRB and the analytical data for TOC in the spent organic medium proved to not be reliable.

The BEST simulator is a spreadsheet-based model that is used in conjunction with a public domain computer software package, i.e., PHREEQCI geochemical modeling program. While PHREEQCI calculates geochemical equilibrium for the advective-reactive transport of AMD through the bioreactor, the spreadsheet portion of the simulator handles issues of AMD flow rate, size of the bioreactor, its operational time, and its economics including NPV.

An example of simulation provided in the report considers the AMD flowing at the rate of 1 gpm and laden with 17.78 mg/L, 6.12 mg/L, 0.08 mg/L, and 40.4 mg/L of Zn, Cu, Cd, and Al, respectively. An SRB treatment system to remove Zn, Cu, and Cd as sulfides would require three RCs with a capital cost of $24,244. The NPV is $37,768, based on a discount rate of 3.2%, O&M at $1,000/year, and the operational time of 18 years.

The BEST simulator was developed and formulated so that a user with minimum modeling experience can operate it. The BEST simulator operation requires basic knowledge of the ExcelTM program and some familiarity with the geochemical model PHREEQCI. Of course, a good chemical background is a bonus. Had the goal for the SRB treatment system

been the removal of all the metals including Al and Fe3+ that precipitates as hydroxide, the system would have included 17 RCs at a capital cost of $137,385 and a NPV of $156,488 for the system to operate for 39 years.

The BEST simulator is saved as Microsoft ExcelTM workbook BEST V1.xls and consists of several worksheets, one of them being an I-O worksheet that allows for entering the majority of input data and having the most imported results also printed on the same page. However, details of the design-like specification of the material, etc., are not listed in the I-O worksheet, and the user needs to refer to one of the individual worksheets to examine these details. Most worksheets are linked together, i.e., any change of input data causes appropriate changes of the results calculated by the respective worksheet. However, the PHREEQCI model and its data input file are not automatically linked with the rest of the worksheets, thus required changes need to be manually entered.

Future considerations could take into account the verification of the accuracy of the BEST model by performing laboratory/field RC installations.

62

8. References 1. Canty, M. "Overview of Sulfate-Reducing

Bacteria Demonstration Project Under the Mine Waste Technology Program," Mineral Processing and Extractive Metallurgy Review, Vol 19, pp. 61-80, 1998.

2. Zaluski, M., J. Trudnowski, M. Canty, and

M.A. Harrington Baker, "Performance of Field-Bioreactors with Sulfate-Reducing Bacteria to Control Acid Mine Drainage," Proceedings from the Fifth International Conference on Acid Rock Drainage, Vol. II, pp. 1169-1175, 2000 Society for Mining, Metallurgy, and Exploration, Inc.

3. MSE Technology Applications, Inc., Final

Report – Sulfate-Reducing Bacteria Reactive Wall Demonstration, Mine Waste Technology Program, Activity III, Project 12, 2002.

4. MSE Technology Applications, Inc.,

Evaluations of Organic Substrates for the Growth of Sulfate-Reducing Bacteria to Treat Acid Mine Drainage, Mine Waste Technology Program, Activity III, Project 24, 2001.

5. ASTM, Standard Test Method for

Permeability of Granular Soils (Constant Head, Designation: D2434-68, 1993.

6. Parkhurst, D.L. and C.A.J. Appelo, User’s Guide to PHREEQC (Version 2) – a Computer Program for Speciation, Batch-Reaction, One-Dimmensional Transport, and Inverse Geochemical Calculations, U.S. Geological Survey, Water-Resources Investigations Report 99-4259, 1999.

7. U.S. Office of Management and Budget, The

Executive Office of the President, Discount Rates for Cost-Effectiveness, Lease Purchase and Related Analyses, Circular No. A-94 – Appendix C, 2003.

8. Desmier, R., B.C.T. Macdonald, T.D. Waite,

and M.D. Melville, "Passive Treatment of Acid Sulfate Soil Drainage Using a Closed Tank Reactor," Proceedings of the Sixth International Conference on Acid Rock Drainage, The Australian Institute of Mining and Metallurgy (AusIMM), 2003.

9. U.S. Environmental Protection Agency,

USEPA Contract Laboratory Program National Functional Guidelines for Inorganics for Data Review, 1994.

63

Appendix A

CD Containing Report "Evaluation of Organic Substrates for the Growth of Sulfate-Reducing Bacteria to Treat Acid Mine Drainage" (MWTP-188) and MS AccessTM Database

Appendix B1

Engineering Drawing: SRB Cell Construction

B1-1

Appendix B2

Engineering Drawing: SRB Cell Components and Construction Details

B2-1

B2-2

B3-1

Appendix B3

Mechanical Adaptation of a Typical Tank and Construction of the SRB RC

Modifications of bulk storage tank for use as SRB RC (cell) Using the actual diameter of the purchased tank, determine the needed height required to achieve the

necessary volume of SRB media.

If the bulk tank has wide corrugated walls, as is common in larger sizes, the tank will be cut on a smaller diameter section above the needed height. Near the top the tank will be cut on a larger diameter section. This will create a lid (CR cover) that fits over the lower section’s edge and prevents any natural water (rain) from entering the cell.

If the tank is smooth sided, the tank will have to be cut in two places to attain the desired height. A lip will have to be fixed to the upper section to create a cap that prevents the introduction of water.

Construction With the lid of the tank off, bore holes into the sides of the tank and install the 2-inch PVC bulkhead

fittings as shown on the Assembly Drawing.

Install the pipe saddles on the sections of perforated piping at the height required to allow the attachment of 2-inch piping between the saddles and the bulkhead fittings previously installed on the tanks.

Install 2-inch pipe stubs into the bulkhead fittings on the inside of the tank.

Connect the perforated pipe sections to the influent and effluent connections; perforated sections shall rest on the bottom of the tank.

While placing bags with W/M organic medium into the tank, care must be taken to ensure the perforated sections of piping remain in a vertical position.

Place 30-mil tarp over SRB material and tuck in at all edges.

Backfill above tarp to provide structural integrity for cell.

Bore holes into lid section to correspond to the positions of the 4- and 10-inch perforated pipe ends. Care should be taken to limit any gap between lid and pipes.

Install draw latches onto the lid and tank with metal screws.

Seal the space between the 4- and 10-inch perforated pipes and the lid with non-drying sealant or a rubber ring, if available.

Install exterior piping to the connections in the side of the tank as determined by the drawings. Support of the piping shall be determined in the field based upon the environmental and topographical conditions.

Appendix C

CD Containing Bioreactor Economics, Size, and Time of Operation (BEST) Simulator