EPA/600/R-08/097
June 2008
Mine Waste Technology Program
Passive Treatment
For Reducing Metal Loading
By:
Diane Jordan
MSB Technology Applications, Inc.
Mike Mansfield Advanced Technology Center
Butte, Montana 59702
Under Contract No. DE-AC09-96EW96405
Through EPA IAG No. DW89-92197401-0
Norma Lewis, EPA Project Manager
Systems Analysis Branch
National Risk Management Research Laboratory
Cincinnati, Ohio 45268
This study was conducted in cooperation with
U.S. Department of Energy
Environmental Management Consolidated Business Center
Cincinnati, Ohio 45202
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
Disclaimer
This publication is a report of work conducted under the Mine Waste Technology Program that was
funded by the Environmental Protection Agency and managed by the Department of Energy under the
authority of an Interagency Agreement.
Because the Mine Waste Technology Program participated in EPA's Quality Assurance Program, the
project plans, laboratory sampling and analyses, and final report of all projects were reviewed to ensure
adherence to the data quality objectives. The views expressed in this document are solely those of the
performing organization. The views and opinions of authors expressed herein do not necessarily state or
reflect those of the United States Government or any agency thereof
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, recommendation, or
favoring by the United States Government or any agency thereof or its contractors or subcontractors.
Neither the United States Government nor any agency thereof, nor any of their employees, nor any of
their contractors, subcontractors, or their employees, makes any warranty, express or implied, or assumes
any legal liability or responsibility for the accuracy, completeness, or any third party's use or the results
of such use of any information, apparatus, product, or process disclosed, or represents that its use would
not infringe privately owned rights.
-------
Foreword
The U.S. Environmental Protection Agency (EPA) 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 (NRMRL) 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. 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 project was conducted under the Mine Waste Technology Program. It was funded by the EPA and
administered by the U.S. Department of Energy (DOE) in cooperation with various offices and
laboratories of the DOE and its contractors. It is made available at www.epa.gov/minewastetechnology
by EPA's Office of Research and Development to assist the user community and to link potential users
with the researchers.
Sally Gutierrez, Director
National Risk Management Research Laboratory
in
-------
Abstract
This report summarizes the results of Mine Waste Technology Program (MWTP) Activity III, Project 48,
Passive Treatment Technology Evaluation for Reducing Metal Loading, funded by the U.S.
Environmental Protection Agency (EPA) and jointly administered by EPA and the U.S. Department of
Energy. MSE Technology Applications, Inc. performed the technology demonstration.
The overall project objective was to evaluate passive treatment media for a given water chemistry that
could provide information to identify potential treatment systems to reduce the dissolved metals loading
contribution from Canyon Creek by 50%.
A 50/50 blend of groundwater retrieved from Canyon Creek sampling site wells CC1508 and MW-
CCTW01S was used in column and batch testing. The data from batch testing was used to determine
parameters for the column testing. Initial tests determined the equilibrium loading of each passive media.
The three media with the highest heavy metals loading were used to determine passive treatment bed
design characteristics (mass transfer zone, breakthrough curves, precipitation issues, etc.) during the
column study.
In EPA Region 10, as well as other EPA regions, research has been initiated to implement and evaluate a
variety of reactive media for water treatment at a number of sites. It should be noted that information
from this study is not directly transferable to other sites because water chemistry is site-specific and
performance of media will therefore vary from site to site.
IV
-------
Contents
Page
Disclaimer ii
Foreword iii
Abstract iv
Contents v
Figures vi
Tables vi
Acronyms and Abbreviations vii
Acknowledgments viii
Executive Summary ES-1
1. INTRODUCTION 1
1.1 Project Description 1
1.2 Background 1
1.3 Site Description 2
1.4 Project Objectives 2
1.5 Experimental Overview 2
1.5.1 Batch Testing 2
1.5.2 Column Testing 2
2. QUALITY ASSURANCE 5
3. RESULTS 6
3.1 Bench-Scale Testing 6
3.1.1 Batch Testing - Rate of Removal 6
3.1.2 Batch Testing - Equilibrium Capacity 7
3.1.3 Column Test - Equilibrium Capacity 8
4. CONCLUSIONS 22
5. REFERENCES 23
Appendix A: Summary of Quality Assurance Activities A-l
-------
Figures
Page
1-1. Batch testing 3
1-2. Laboratory columns with media 4
3-1. Titration curve for Canyon Creek groundwater 11
3-2. Iron coated sand isotherm 11
3-3. Iron-aluminum coated sand 12
3-4. Manganese dioxide isotherm 12
3-5. Juniper bark isotherm 13
3-6. Granular ferric hydroxide isotherm 13
3-7. Apatite II™ isotherm 14
3-8. Bauxsol™ isotherm 14
3-9. Control columns - Zn concentration vs. bed volumes 15
3-10. GFH columns -Zn concentration vs. bed volumes 15
3-11. Apatite II™ columns -Zn concentration vs. bed volumes 16
3-12. Bauxsol™ columns - Zn concentration vs. bed volumes 16
Tables
3-1. Rate Test Results - Reagent Group 1 17
3-2. Rate Test Results - Reagent Group 2 17
3-3. Equilibrium Capacity Test Results 18
3-4. Column Test Results - Flow Rate - 2 mL/min 19
3-5. Column Test Results - Flow Rate - 5 mL/min 20
3-6. Column Loading Capacities - by Reagent 21
VI
-------
Acronyms and Abbreviations
Cd cadmium
DOE U.S. Department of Energy
EPA U.S. Environmental Protection Agency
GFH granular ferric hydroxide
Mn manganese
MSE MSE Technology Applications, Inc.
MWTP Mine Waste Technology Program
NRMRL National Risk Management Research Laboratory
ORP oxidation-reduction potential
Pb lead
QA quality assurance
SC specific conductivity
SRB sulfate-reducing bacteria
SU standard unit
Zn zinc
vn
-------
Acknowledgments
This document was prepared by MSB Technology Applications, Inc. (MSB) for the U.S. Environmental
Protection Agency's (EPA) Mine Waste Technology Program (MWTP) and the U.S. Department of
Energy's (DOE) Environmental Management Consolidated Business Center. Ms. Diana Bless is EPA's
MWTP Project Officer, while Mr. Gene Ashby is DOE's Technical Program Officer. Ms. Helen Joyce is
MSB's MWTP Program Manager. Ms. Norma Lewis is the EPA Project Manager, Ms. Lauren Drees is
the EPA Quality Assurance Officer, and Ms. Diane Jordan is the MSE Project Manager. The project
would not have been as successful without the assistance of the following individuals:
Bill Adams, EPA Region 10
Gary Hickman, CH2M HILL
Rebecca Maco, CH2M HILL
Basin Environmental Improvement Project Commission
Gary Wyss, MSE Technology Applications, Inc.
Martin Foote, MSE Technology Applications, Inc.
MSE Laboratory
Vlll
-------
Executive Summary
Mine Waste Technology Program (MWTP), Activity III, Project 48, Passive Treatment Technology
Evaluation for Reducing Metal Loading, was funded by the U.S. Environmental Protection Agency
(EPA) and jointly administered by EPA and the U.S. Department of Energy (DOE). MSE Technology
Applications, Inc. performed the technology demonstration. This project evaluated passive treatment
systems to reduce the metals loading from Canyon Creek in Burke, Idaho.
Various passive treatment systems have been emplaced in the Coeur d'Alene Basin. Those systems
consist primarily of both open wetland systems and closed contained systems represented by the
following:
• EPA's MWTP Nevada Stewart Permeable Reactive Barrier Demonstration Project - A passive,
Apatite II™ (a fishbone apatite) system was used to treat approximately 20 gallons per minute at the
Nevada Stewart Mine reducing zinc, cadmium, manganese, lead, and iron to instrument detection
limits through certain cells.
• Idaho Department of Environmental Quality's Success Mine Seep Treatment - This system
demonstrated the effectiveness of Apatite II™ to reduce the total metals loading, mainly zinc,
cadmium, and lead, by 75% at the system outflow and, to date, the system has met its objective.
• Bureau of Land Management's Sydney Adit Drainage Treatment - This project consists of several
field pilot water studies including: 1) a bioreactor, 2) a reactive medium system, and 3) biochelators
to reduce the metals loading to receiving waters.
• EPA Region 10 - This treatability study was performed by CH2M HILL, as EPA Region 10's
Remedial Action Contractor, on active and passive processes to investigate sulfate-reducing bacteria
(SRB) and lime high-density sludge processes for treatment of these waters.
These systems were able to reduce the metals loading in the waters treated; however, in some cases it was
reported that short-circuiting and clogging restricted the systems ability to function efficiently.
Additionally, the designs that were implemented limited the ability of the systems to handle fluctuating
seasonal flows.
The purpose of this project was to evaluate passive treatment media to reduce the metals loading
contribution from Canyon Creek by 50%. This goal was established under the Bunker Hill Mining and
Metallurgical Complex Operable Unit 3 Record of Decision in September 2002.
It should be noted that information from this study is not directly transferable to other sites because water
chemistry is site-specific and performance of media will therefore vary from site to site.
Of the seven media tested, three were determined to be applicable for treating Canyon Creek
groundwater: granular ferric hydroxide (GFH), Bauxsol™, and Apatite II™. Of these three media,
Apatite II™ showed a higher capacity to remove zinc (the primary element of concern) from this water
when compared to GFH and Bauxsol™.
ES-1
-------
1. Introduction
1.1 Project Description
Mine Waste Technology Program (MWTP),
Activity III, Project 48, Passive Treatment
Technology Evaluation for Reducing Metal
Loading, was 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. EPA contracted MSE
Technology Applications, Inc. (MSE) through
the MWTP to evaluate passive treatment media.
The overall project objective was to evaluate
passive treatment media to reduce the metals
loading contribution from Canyon Creek. This
goal was established under the Bunker Hill
Mining and Metallurgical Complex Operable
Unit 3 Record of Decision in September 2002.
1.2 Background
Historical mining practices and the naturally
occurring geochemistry can result in the heavy
metal contamination of soil, sediment, surface
water, and groundwater in drainages. One
method of mitigating these sources of
contamination is through the implementation of
passive treatment systems intercepting flow
emanating from the source. A passive treatment
system is one that requires minimal
maintenance. Passive treatment systems, using a
variety of media (i.e., phosphate-based material
such as fishbone apatite, sawdust based organic
mix using anaerobic bacteria, synthetic polymer
adsorbents, and zero valent iron) have been
proposed and, in some cases, have been applied.
Selection of an appropriate media to treat a
given water chemistry at a specific site is
sometimes very difficult. Most passive
treatment systems that have been installed have
been successful at reducing dissolved metal
loadings from contaminated waters. However,
in general, passive treatment systems have been
problematic due to reduced permeability over
time caused by reactive media clogging with
precipitates and foreign debris.
Various passive treatment systems have been
emplaced in the Coeur d'Alene Basin. Those
systems consist primarily of both open wetland
systems and closed contained systems including:
• EPA's MWTP Nevada Stewart Permeable
Reactive Barrier Demonstration Project - A
passive, Apatite II™ (a fishbone apatite)
system was used to treat approximately 20
gallons per minute at the Nevada Stewart
Mine reducing zinc (Zn), cadmium (Cd),
manganese (Mn), lead (Pb), and iron to low
dissolved concentrations.
• Idaho Department of Environmental
Quality's Success Mine Seep Treatment-
This system demonstrated the effectiveness
of Apatite II™ to reduce the total metals
loading, mainly Zn, Cd, and Pb, by 75% at
the system effluent.
• Bureau of Land Management's Sydney Adit
Drainage Treatment - This project consists
of several field pilot water studies including:
1) a bioreactor, 2) a reactive medium
system, and 3) biochelators to reduce the
metals loading to receiving waters.
• EPA Region 10 - This treatability study was
performed by CH2M HILL, as EPA Region
10's Remedial Action Contractor, on active
and passive processes to investigate sulfate-
reducing bacteria (SRB) and lime high-
density sludge processes for treatment of
Canyon Creek waters.
All systems listed above were able to reduce the
metals loading in the waters treated; however, in
some cases it was reported that short-circuiting
and clogging restricted the ability of these
systems to function efficiently. Additionally,
the designs that were implemented limited the
ability of the systems to handle fluctuating
seasonal flows.
-------
1.3 Site Description
Laboratory testing of the passive treatment
media for removal of heavy metals from Canyon
Creek groundwater was performed at the MSB
Testing Facility in Butte, Montana.
1.4 Project Objectives
The project objective was to evaluate passive
treatment media for a given water chemistry that
would provide information to identify additional
treatment options. These treatment options
would be applicable for reducing the metals
loading contribution from Canyon Creek.
1.5 Experimental Overview
A 50/50 blend of groundwater was retrieved
from a sampling site well CC1508 and MW-
CCTWO1S for use in the column and batch
testing. The data from batch testing was used to
determine parameters for bench-scale studies.
These tests are designed to characterize the
ability of the media to act as a passive treatment
system for the removal of heavy metals from
groundwater. Initial tests centered on
determining the equilibrium loading of the
passive media. The top five media with the
highest heavy metals loading were used for
additional studies designed to determine passive
treatment bed design characteristics (mass
transfer zone, breakthrough curves, precipitation
issues, etc.).
1.5.1 Batch Testing
To characterize and quickly evaluate various
passive treatment medium, equilibrium
isotherms were developed for each media. An
initial sample of the groundwater was analyzed
for total suspended solids, total recoverable
metals, and dissolved metals. pH and oxidation-
reduction potential (ORP) measurements were
conducted initially and at the end of each test.
For each media an isotherm was constructed by
varying the mass of media for a fixed volume of
groundwater added. To insure equilibrium was
achieved, equilibrium rate studies were carried
out on each of the media. The equilibrium
capacity studies were run at five masses for each
while keeping the groundwater volume constant
at 500 milliliters (mL). An Erlenmeyer flask
with constant agitation during the reaction
period was used as shown in Figure 1-1. The
media was added in the amounts of 1 gram (g),
2 g, 5 g, 10 g, and 50 g. Samples were taken at
the expiration of equilibrium time as determined
in the rate studies. The samples were filtered
through a 0.45-micron filter to determine the
final equilibrium dissolved concentrations of Cd,
Pb, and Zn.
The amount of metal adsorbed from solution
divided by the mass of media added to the flask
was plotted against the final equilibrium
concentration found in solution are shown in
Figures 3-2 to 3-8. This plot was used to
determine adsorption characteristics of each
media and determine if the media was suitable
for passive treatment of Canyon Creek
groundwater.
1.5.2 Column Testing
Three media, Apatite II™, Bauxsol™ and GFH,
were selected from the equilibrium loading tests
and further tested in a column configuration.
The specific column design was determined
based upon loading data provided from
equilibrium testing. The limits of the design
were to size the columns such that breakthrough
would be observed after passing a maximum of
15 liters through the bed. Based upon the
loading observed in the batch testing, an
appropriate volume of passive media was loaded
into the column. Denstone, a ceramic bed
support media, was used at the bottom and top
of each column to act as a flow dispersion
media. The columns were approximately 1-inch
inside diameter by 24 inches long with packings
varying from 50 to 250 mL of passive media.
The column configuration is shown in
Figure 1-2. The object of the column studies
was to characterize the breakthrough curves
associated with Zn, Cd, and Pb for each of the
three passive media.
Groundwater obtained from Canyon Creek was
introduced to the bottom of the columns and
-------
samples were taken at the outflow until
breakthrough was complete or the test was
discontinued due to time/budgetary constraints.
Samples were analyzed for Zn, Pb, and Cd until
breakthrough was achieved. Breakthrough was
determined with the first appearance of Zn in the
effluent.
The pressure drop across the bed was monitored
daily using a manometer to determine if any
precipitates or other bed fouling phenomena
occurring. This information was used to
determine the best media for the Canyon Creek
groundwater chemistry.
Figure 1-1. Batch testing.
-------
Figure 1-2. Laboratory columns with media.
-------
2. Quality Assurance
A summary of the quality assurance (QA) Project 48, Passive Treatment for Reducing Metal
activities associated with MWTP Activity III, Loading can be found in Appendix A.
-------
3. Results
3.1 Bench-Scale Testing
The first test that was conducted on the
groundwater collected near Canyon Creek was
the development of a titration curve. This curve
is illustrated in Figure 3-1 and shows that an
inflection occurs near a pH of 9.0, which is most
likely related to the hydrolysis of the dissolved
Zn in the water.
3.1.1 Batch Testing — Rate of Removal
Samples of the water in contact with a specific
media were taken at specific times during the
test, filtered, and analyzed for dissolved Zn to
determine the concentration that remained from
the initial 41,500 micrograms per liter (|o,g/L).
As such, this test determined the rate of removal
of Zn from the water by the specific media.
The results of the rate test showed that the media
could be classified into two groups. The first
group included those media that rapidly
increased the pH of the groundwater to levels
greater than 10.0 and maintained that level. This
group of media consisted of partially charred
dolomite; Bauxsol™; partially calcined
limestone; a 50:50 mixture of limestone and
calcium oxide; a 50:50 mixture of limestone and
magnesium oxide; and sodium hydrosulfide.
The rate test results for the first group of media
are contained in Table 3-1.
As is denoted in Table 3-1, all of the media in
the first group removed greater than 96% of the
dissolved Zn from the groundwater within one
hour of contact and increased the pH of the
groundwater to levels greater than 10 within the
first two hours of the test. pH values for the
groundwater were not taken after one hour of
contact time. However, fundamental knowledge
of the chemical processes that occur with these
highly alkaline media can be relied upon to state
that in all likelihood the increase in pH occurred
within the first few minutes of contacting the
water with each media. . An inflection point
was defined in the pH titration curve for the
Canyon Creek water near a pH value of 9.0,
which is shown in Figure 3-1. It is probable that
the primary method of Zn removal from the
Canyon Creek groundwater for this first group
of media was due to hydrolysis of Zn and the
precipitation of Zn hydroxide solids formed by
that process. In the case of sodium hydrosulfide,
it is probable that a significant portion of the Zn
was removed from solution by the formation of
Zn sulfide. The Zn removal results produced by
the first group of media generally tended to
increase with time of contact. All of the media
in the first group removed greater than 99% of
the dissolved Zn from the groundwater within
eight hours of contact, which was the time to
reach equilibrium determined in the rate test.
It is possible that this increased removal of
dissolved Zn was due to a number of secondary
processes, which included increased
precipitation of Zn hydroxides over time;
adsorption onto the surfaces of the fine-grained
portions of the media that did not dissolve
during the test; co-precipitation from the water
with other elements; adsorption onto the surface
of previously precipitated Zn hydroxides; and
adsorption onto the surface of other metallic
hydroxides formed during the test. It is not
possible with the limited amount of data
developed during the rate test to determine
which, if any, of these secondary processes
contributed in a significant manner to the
removal of Zn from the water.
The second group of media included those that
did not significantly increase the pH of the
groundwater. This group of media consisted of
Apatite II™, ferrihydrite coated sand, GFH,
aluminum-iron coated sand, Juniper bark, and
manganese oxide coated sand. The test results
for the second group of media are contained in
Table 3-2.
As shown in Table 3-2, the Apatite II™ media
removed significantly more of the Zn from the
water than any of the other media of the second
group. Apatite II™ removed more than 98% of
-------
the dissolved Zn from the groundwater in the
first hour of contact. During the second hour of
the test, additional Zn was removed to the
minimum concentration of 71.1 |og/L, which
translates to 99% removal. At completion of the
8-hour test, the final concentration of Zn was
228 ng/L. Overall, Apatite II™ performed very
effectively during the batch tests and showed the
fastest rate of removal of the second group of
media.
The remaining five media in the second group
removed between 12% and 47% of the dissolved
Zn from the groundwater within the first hour of
contact. These five media continued to remove
dissolved Zn from the groundwater through the
remaining period of the 8-hour test. The final
removal percentages for these media varied
between 76.5% for GFH and 18.5% for the
aluminum-iron coated sand. None of the media
in the second group raised the pH to a level
where significant hydrolysis of Zn would occur.
As such, the method of removal for these media
was probably some form of adsorption or ion
exchange. Again, it is not possible, due to the
limited amount of data developed during the rate
test, to determine the specific process or
processes by which Zn was removed from the
groundwater by these media.
Evidence has been cited to show that all of the
media contained in Group 1 removed large
quantities of Zn from the Canyon Creek water
by a process that was probably hydrolysis
followed by precipitation. By definition, this
sequence of processes is found in active not
passive processes. As such, the media that were
tested as Group 1 were eliminated from further
testing with the exception of the Bauxsol™
medium. All other media in Group 1 do not
exhibit passive tendencies, as they would require
frequent media replacement and/or post-
treatment pH adjustment for discharge. The
Bauxsol™ media is designed to combine both
hydrolysis processes and adsorptive processes.
In an attempt to determine the adsorptive
capabilities of the media, Bauxsol™ was
included, along with all the media from Group 2,
in the next test sequence.
3.1.2 Batch Testing - Equilibrium
Capacity
The second type of treatability testing conducted
on the groundwater collected near Canyon Creek
was a batch test in which an attempt was made
to measure the equilibrium capacity of each
medium to remove the specific metals from the
water. As previously described, the test
involved contacting varying quantities of media
with a fixed quantity of water for a period of
time necessary for the system to come to
equilibrium. The contact time varied as
determined in the rate test. Samples of the water
were then taken and analyzed for the various
metals of concern. Once the data was collected,
calculations were made that determined the
amount of each metal that was sorbed per unit
mass of each media. The equilibrium
concentration of each metal remaining in the
water, the mass of sorbent used for each test run,
and the calculated quantity of metal sorbed per
unit mass of media are shown in Table 3-3.
From the data shown in Table 3-3, Zn isotherm
curves were developed using the Freundlich
isotherm for each of the seven media. Zinc
concentration data was used for this calculation
as it is the heavy metal of concern in the Canyon
Creek water. In the development of these
curves, the quantity of metal sorbed per unit
mass of media was plotted versus the
equilibrium concentration of Zn remaining in the
water. These Zn isotherm plots are shown in
Figures 3-2 through 3-8.
The isotherms shown in the aforementioned
figures can be separated into three categories.
The first category contains those isotherms that
exhibit a concave upward shape. The isotherms
for iron-aluminum coated sand and juniper bark
are contained in this category. Materials that
exhibit this form of isotherm have low metal
loading capacities at low equilibrium
concentrations of Zn in solution and are
therefore not considered favorable for the
removal of Zn from the Canyon Creek water.
-------
The second group of isotherms displayed a
downward concave graph. Media with this type
of loading response have a high loading capacity
at lower solution equilibrium concentrations.
These media possess characteristics that are
more favorable for removal of Zn from solution.
The Apatite II™ and ferrihydrite-coated sand
plots are characteristic of this type of isotherm.
The final isotherm category exhibits shapes that
are complex in that the isotherm shows multiple
inflections denoting that the chemical processes
forming the isotherm are not dominated by
sorption or ion exchange. The two media with
isotherms that fall into this category are GFH
and Bauxsol™.
3.1.3 Column Test - Equilibrium Capacity
The third treatability test conducted using the
groundwater collected near Canyon Creek, was a
test using two sets of continuously flowing
columns. Each of the sets of columns contained
a control column and columns filled with the
three media Bauxsol™, Apatite II™, and GFH.
One set of columns was operated at a flow rate
of 2 mL/min and the second set of columns was
operated at a flow rate of 5 mL/min. Samples of
the effluent from each column were taken
periodically throughout the test period and
analyzed for Zn, Cd and Pb. A number of
chemical parameter readings including ORP,
pH, and specific conductivity (SC) were also
taken throughout the test period. The data
generated from the columns with a flow rate of
2 mL/min is contained in Table 3-4 while the
data generated from the columns with a flow
rate of 5 mL/min is contained in Table 3-5. It
should be noted that the feed water to the
columns contained a Zn concentration of
approximately 41,600 |o,g/L, a Cd concentration
of 245 |o,g /L, and a Pb concentration of
187 |o,g /L. Graphical representations of the
column results for Zn are presented in Figures
3-9 through 3-12.
As can be seen from the data presented in the
aforementioned tables and figures, each of the
columns acted differently depending upon the
contained media and the flow rate of the feed
solution. The control columns that were filled
with silica sand were not able to remove
significant amounts of Zn from the influent
solution at flow rates of 2 or 5 mL/min. The
column receiving feed solution at 2 mL/min was
able to remove a small quantity of the Zn from
the influent solution while the column receiving
feed solution at 5 mL/min was not able to
remove an appreciable amount of Zn from the
solution. Neither of these columns was able to
remove an appreciable amount of Cd from the
feed solution.
However, both of the columns were able to
remove a feasible amount of Pb from the feed
solution during the initial portion of the test.
Breakthrough of Pb occurred in the 5-mL/min
column between 1 and 33 bed volumes, while
breakthrough of Pb occurred in the 2-mL/min
column between 35 and 70 bed volumes.
Breakthrough is defined as the first appearance
of contaminant of concern in the effluent. A
slower flow rate and the corresponding
increased residence time of the feed solution in
the column allowed the sand grains (and the
associated impurities) in the 2-mL/min column
to sorb the Pb from solution more effectively
than at the faster 5-mL/min flow rate. The sand
filled columns did not produce a significant
effect or trend on any of the chemical
parameters (pH, ORP, or SC) measured during
the test period. Although not shown in Tables
3-4 or 3-5, the difference in the pressure
between the inlet and outlet of the columns was
also measured periodically throughout the test
period. Both of the sand filled columns showed
no increase in head loss or differential pressure
during the testing.
Visual observations of all the columns were also
made throughout the test period. The sand filled
columns did not change appreciably during the
period of the test.
Both of the columns filled with GFH were able
to remove significant quantities of Zn from the
feed solution at flow rates of 2 and 5 mL/min.
Breakthrough of Zn occurred in the 2-mL/min
-------
column between 99 and 110 bed volumes and in
the 5-mL/min column between 86 and 114 bed
volumes. As shown in Figure 3-10, the column
that was flowing at a rate of 2 mL/min removed
more Zn than did the 5-mL/min column. Like
the previously discussed sand columns, this
result is a function of the residence time of the
solution in the column. Both of the GFH filled
columns required time in contact with the feed
water to remove 90% of the dissolved Zn, which
translates to a concentration of approximately
400 ng /L.
With regards to Cd, the GFH filled column that
flowed at 2 mL/min did not exhibit a
concentration breakthrough during the test
period. The GFH filled column flowing at a rate
of 5 mL/min showed a potential concentration
breakthrough beginning between 114 and 141
bed volumes. Bed volume is the sum of the pore
volume plus the volume of the sorbent. Both of
the GFH filled columns were able to remove the
dissolved Pb in the feed water to low levels
throughout the testing period. No concentration
breakthrough was observed from either column
for Pb. Both of the GFH filled columns lowered
the pH of the effluent solution when compared
to the influent solution after approximately 80
bed volumes had past through the columns.
Prior to this portion of the test, the pH of the
effluent solution was close to that of the influent
solution. The ORP of the effluent solution from
both of the GFH filled columns was erratic
throughout the test period with no discernable
trend. No major changes in the SC of the
effluent solution were observed. Both of the
GFH filled columns maintained similar
differences in pressure between the inlets and
outlets of each column throughout the test
period. The information garnered through visual
observation of the GFH filled columns did not
change appreciably during the period of the test.
The two Apatite II™ filled columns functioned
very differently with respect to the removal of
dissolved Zn from the feed solution depending
on the flow rate of the feed solution. The
Apatite II™ filled column that flowed at
2 mL/min was able to remove dissolved Zn to
levels below the analytical detection limit
throughout the period of the test. However, this
process was not due solely to sorption of the Zn.
As can be seen from the data in Table 3-4, after
approximately 30 bed volumes the ORP of the
effluent solution dropped significantly. The
reduced environment in this column facilitated
the growth of SRB. The presence of SRB was
denoted by the odors of hydrogen sulfide and
other biological materials emanating from the
column during the latter portions of the test. It
should be noted that Apatite II™ is a
biologically formed material that contains
significant amounts of oils and other secondary
compounds. Apatite II™ has been known to
develop growths of SRB in other similar tests
under low-flow conditions (McCloskey, 2003).
One of the byproducts of the SRB is soluble
hydrogen sulfide that can react strongly with
metals such as Zn to form insoluble Zn sulfide
solids. It is speculated that the secondary
process contributed to the removal of Zn, as well
as Cd and Pb, from the influent solution. The
Apatite II™ filled column that had a flow rate of
5 mL/min did not develop large quantities of
SRB during the test. The ORP of the effluent
from the 5-mL/min column did not drop
appreciably throughout the test and no hydrogen
sulfide odor was observed. It is probable that
the increased flow rate was able to deliver
enough dissolved oxygen to the column to
prevent a major drop in ORP and a significant
development of SRB.
The dissolved Zn concentration in the effluent
from the 5-mL/min column showed a
breakthrough at approximately 140 bed
volumes. No major changes in the pH of the
effluent solution from both of the Apatite II™
filled columns occurred during the test period.
The SC of the column with a flow rate of 5
mL/min did not change in a noteworthy manner
during the test period. However, the SC of the
column with a flow rate of 2 mL/min trended to
higher values after approximately 40 bed
volumes had passed through the column. This
increase was probably due to the presence of
-------
soluble byproducts from the activities of SRB.
The pressure difference between the influent and
effluent portions of both Apatite II™ filled
columns decreased throughout the period of the
test. It was observed that Apatite II™ tends to
swell when it is first wetted which probably
contributed to the increase in the pressure
difference initially, but once the swelling
stopped the pressure difference stabilized. No
discernable changes to the Apatite II™ filled
columns were visually observable during the
period of the test.
A breakthrough in the dissolved Zn
concentration of the effluent solution was
denoted from both of the Bauxsol™ filled
columns during the test. That breakthrough
occurred from the 2-mL/min flow rate column
after approximately 65 bed volumes of solution
had passed through the column, while the
breakthrough from the 5-mL/min column
occurred after approximately 40 bed volumes of
solution had passed through the column. The
dissolved concentrations of Cd and Pb broke
through from both of the Bauxsol™ filled
columns at approximately the same point in the
test, as was the case for the dissolved Zn
concentration. The pH and the ORP of the
effluent solution from both of the Bauxsol™
filled columns also showed marked changes in
value at approximately the same point during the
test that the previously mentioned breakthroughs
occurred. The pH of both columns decreased
and the ORP increased at the point of
breakthrough. This data supports the
conclusions determined about the removal
method of metals from water by Bauxsol™. In
that, elevated values of pH led to hydrolysis
followed by precipitation of solid materials. In
addition, visual observation of the Bauxsol™
filled columns denoted a grey-white gelatinous
solid forming in the column on the surface of the
red-colored Bauxsol™ during the test. Lastly,
the difference in pressure within the column
between the influent and effluent portions of
both Bauxsol™ filled columns increased
dramatically throughout the period of the test.
This increase in pressure differential was
probably due to the precipitation and deposition
of fine-grained, solid materials formed during
the test process.
Subsequent to the completion of the column
tests, a calculation was made in an attempt to
determine the total quantity of Zn that was
removed by each column up to breakthrough of
the dissolved Zn concentration. The results of
that loading calculation are shown in Table 3-6.
The calculation does not take into consideration
the fact that dissolved Zn concentration in the
effluent from the Apatite II™ column with a
flow rate of 2 mL/min did not breakthrough. In
addition, the dissolved Zn concentration
breakthrough for the two sand filled columns
was not identified, as the effluent from these two
columns was always at or near the influent
concentration. Nevertheless, the calculation
does yield some interesting information. The
columns with the slower flow rate yielded larger
capacities for all of the tested medias. The
material in the columns filled with Apatite II™
had considerably higher capacities than the other
media, even at the faster rate of flow. The
materials filling the GFH columns and the
Bauxsol™ columns had relatively similar
capacities for the removal of Zn.
One of the parameters that the test attempted to
discern was the sorptive capacity of Bauxsol™
beyond the Zn removal capabilities of that media
by hydrolysis. Using the same type of
calculation detailed in Table 3-6, it can be
shown that the Bauxsol™ media was able to
remove 5,844 |o,g per g of media at a flow rate of
5 mL/min from approximately 84 bed volumes
of feed water and that the 2-mL/min column was
able to remove approximately 3,547 |o,g per g of
media from approximately 56 bed volumes of
feed water. This calculation was only performed
between breakthrough of the dissolved Zn
concentration and a total volume of
approximately 120 bed volumes. Breakthrough
of Zn occurred sooner in the 5-mL/min flow rate
column with a lower number of bed volumes
when compared to the 2-mL/min flow rate
column.
10
-------
Titration of Canyon Creek Composite Water
4.00 6.00 8.00
Vol. of 0.05024 N NaOH (ml)
Figure 3-1. Titration curve for Canyon Creek groundwater.
Fe-Coated Sand
Zinc Adsorption Isotherm
1
1
u
Q.
8
V
ro
o°-30
N
J
/^
/
/
/
/
'
0 5000 10000 15000 20000 25000 30000 35000 40000
Equil. Zinc Cone. (ug/L)
Figure 3-2. Iron coated sand isotherm.
11
-------
Al / Fe-Coated Sand
Zinc Adsorption Isotherm
ra 0.25
I
f 0.20
re
o
~ 0.15
25000 30000 35000
Equil. Zinc Cone. (ug/L)
Figure 3-3. Iron-aluminum coated sand.
Manganese Dioxide-Coated Sand
Zinc Adsorption Isotherm
ra
i'
O 0.40
N 0.20
0 5000 10000 15000 20000 25000 30000 35000 40000 45000
Equil. Zinc Cone. (ug/L)
Figure 3-4. Manganese dioxide isotherm.
12
-------
Juniper Bark
Zinc Adsorption Isotherm
8
i1 0.40
20000 22000 24000 26000 28000 30000 32000 34000 36000 38000 40000
Equil. Zinc Cone. (ug/L)
Figure 3-5. Juniper bark isotherm.
Granular Ferric Hydroxide
Zinc Adsorption Isotherm
i1
'•5
15000 20000 25000
Equil. Zinc Cone. (ug/L)
Figure 3-6. Granular ferric hydroxide isotherm.
13
-------
Apatite II
Zinc Adsorption Isotherm
7
4000 6000 8000 10000
Equil. Zinc Cone. (ug/L)
12000 14000
Figure 3-7. Apatite D™ isotherm.
Bauxsol
Zinc Adsorption Isotherm
£ 15.00
8.
8
100 150
Equil. Zinc Cone. (ug/L)
Figure 3-8. Bauxsol™ isotherm.
14
-------
Sand Control Columns
Zinc Concentration
c 20000
3
~ 15000
20 40 60
100 120 140 160 180 200 220 240 260
Bed Volumes
Figure 3-9. Control columns - Zn concentration vs. bed volumes.
GFH Columns
Zinc Concentration
5)30000
•S 25000
re
(V
o 20000
o
o
c 15000
Bed Volumes
Figure 3-10. GFH columns - Zn concentration vs. bed volumes.
15
-------
Apatite II Columns
Zinc Concentration
'3>
20000
5
o
100
Bed Volumes
Figure 3-11. Apatite IFM columns - Zn concentration vs. bed volumes.
Zinc Concentration - ug/L
Bauxsol Columns
Zinc Concentration
^^.
/ ZH
//
I/
/
>
,y;
—•—5ml. Per min.
—•—2ml. Per min.
Feed
3 50 100 150 200 250 300 350 400
Bed Volumes
Figure 3-12. Bauxsol™ columns - Zn concentration vs. bed volumes.
16
-------
Table 3-1. Rate Test Results - Reagent Group 1
Reagent Name
Initial Zn
Cone.
Zn Cone.
(2> 1 Hr.
Zn Cone. & pH
(2> 2 Hr.
Zn Cone. & pH
(2> 4 Hr.
Zn Cone. & pH
@8Hr.
Partially charred dolomite
Bauxsol™
Partially calcined limestone
50:50 mixture of limestone
and calcium oxide
41,500
41,500
41,500
41,500
221 ng/L
68.8 |ig/L
239 |ig/L
1,290 ng/L
37.7 |ig/L - 12.5
44.8|ig/L-11.4
226 |ig/L - 12.5
885 |ig/L - 12.6
7.34 |ig/L- 12.1
17.2 ng/L-1 1.3
193|ig/L-12.5
469 |ig/L - 12.6
7.34 |ig/L - 12.2
24|ig/L-11.3
160 |ig/L - 12.3
352 ng/L- 12.3
50:50 mixture of limestone 41,500
and magnesium oxide
Sodium hydrosulfide 41,500
39.8|ig/L 22|ig/L-11.6 48.7 |ig/L - 11.1 61.6 |ig/L - 11.3
102|ig/L 413 |ig/LL- 10.6 44.4 |ig/L - 10.4 13.6 |ig/L - 10.3
Table 3-2. Rate Test Results - Reagent Group 2
Reagent Name
Initial Zn
Cone.
Zn Cone.
@lHr.
Zn Cone. & pH
(® 2 Hr.
Zn Cone. & pH
@4Hr.
Zn Cone. & pH
@8Hr.
Apatite II™ 41,500
Ferrihydrite coated sand 41,500
GFH 41,500
Aluminum-iron coated sand 41,500
Juniper bark 41,500
Manganese dioxide coated 41,500
sand
561 |ig/L
30,500 |ig/L
22,000 |ig/L
36,700 |ig/L
32,000 |ig/L
28,100 |ig/L
71.1 |ig/L-6.9
27,800 |ig/L - 6.4
18,500|ig/L-5.7
34,300 |ig/L - 6.3
34,700|ig/L-3.6
27,300|ig/L-5.7
262 |ig/L - 6.7
25,900 |ig/L - 6.2
12,500|ig/L-5.5
36,200 |ig/L-6.1
30,300|ig/L-3.4
24,600|ig/L-5.3
228 |ig/L - 7.0
23,000 |ig/L - 6.2
9,740|ig/L-5.8
33,800 |ig/L - 6.0
28,800 |ig/L - 3.6
23,400|ig/L-5.9
17
-------
Table 3-3. Equilibrium Capacity Test Results
Reagent Name
Apatite II™
Ferrihydrite coated sand
GFH
Aluminum-iron coated
sand
Juniper bark
Manganese dioxide
coated sand
Bauxsol™
Initial
Zn
Concentration
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
40,000
40,000
40,000
40,000
40,000
40,000
40,000
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
44.080
44,080
44,080
44,080
44,080
44,080
44,080
44,080
Equilibrium
Zn
Concentration
-jig/L
13,100
7,150
1,170
606
293
201
36,000
36,000
24,700
10,500
326
29,200
21,100
11,800
19,000
15,800
5,840
6,160
41,000
40,000
37,400
32,300
20,000
38,500
36,700
32,200
27,600
22,500
40,400
38,400
35,300
26,700
7,580
16.4
97.4
29.8
96.2
178
95.1
Equilibrium Cd
Concentration
35.94
18.36
2.77
1.27
0.65
0.50
162.04
151.85
76.03
24.00
1.03
148.00
104.00
53.20
77.90
50.20
17.50
18.20
225.95
214.03
212.34
202.25
141.04
223.41
212.27
182.31
80.23
124.19
242.00
226.59
209.20
164.02
84.32
9.87
6.96
0.14
0.08
0.08
0.16
Sorbent
Mass - g
1.1
2
5.3
10
25
50.2
1.19
2.05
5.31
10.26
50.45
1
2
5
10
25
50
50
1.04
2
5.08
10.25
50.03
1.07
2.15
5.07
10.22
20.74
1
2.15
5.02
10.14
24.67
50.12
1
2
5
10
50
Loading
-HgZn
Sorbed per g
of Sorbent
14.08
9.23
4.05
2.17
0.88
0.44
0.79
0.73
0.73
0.65
0.17
5.40
4.73
2.82
1.05
0.48
0.34
0.34
0.30
0.20
0.13
0.11
0.05
0.78
0.51
0.35
0.24
0.16
0.74
0.53
0.35
0.34
0.30
0.18
21.99
11.01
4.40
2.20
0.44
Loading
-HgCd
Sorbed per g
of Sorbent
99.65
59.20
23.81
12.69
5.09
2.54
15.65
10.08
6.75
4.51
1.01
42.68
33.40
14.74
6.96
3.45
1.94
-
2.81
2.06
0.84
0.52
0.23
4.45
2.99
2.16
2.57
0.95
2.63
2.66
1.83
1.80
1.39
0.98
124.1
63.76
25.51
12.75
2.55
18
-------
Table 3-4. Column Test Results - Flow Rate - 2 mL/min
Reagent Name
Control - Sand
Apatite II™
GFH
Bauxsol™
Bed
Volumes
3.0
13.0
25.3
34.8
71.1
93.4
115.9
3.4
12.5
14.8
25.9
28.5
29.4
39.5
56.6
70.5
80.7
94.2
107.7
120.8
133.7
3.0
11.7
14.0
24.4
26.8
Til
37.0
53.1
66.2
75.9
88.4
98.8
111.3
123.2
14.4
53.5
63.8
109.8
119.8
164.7
PH
(SU)
7.11
6.19
6.14
6.18
6.56
6.53
6.55
7.38
6.79
6.85
6.72
6.85
-
6.78
6.68
6.62
6.57
6.58
6.61
6.59
6.60
7.51
4.79
5.80
5.23
5.90
-
5.09
5.95
6.13
5.16
7.14
7.21
7.10
6.89
11.22
9.48
8.77
6.68
6.71
6.68
ORP
(mv)
154
194.4
173
152.3
95.8
141.7
168
114.8
133.6
74.3
126.6
131.5
-
112.6
94.4
25.8
76.7
-1.0
-.3
-71.1
-71.5
148.6
204.9
251.4
176.1
199.2
-
168.8
163.6
106.4
142.9
73.9
111.8
100.4
150.3
7.4
78.1
47.7
169.7
150.5
134.2
SC
(US/cm)
335
339
339
336
339
339
339
865
371
366
365
371
-
372
386
397
408
469
501
550
579
294
244
240
240
239
-
236
238
241
234
228
218
235
251
917
383
370
353
352
350
Dissolved
Zn-ng
41,000
41,800
39,100
39,500
39,700
39,800
40,000
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
6,990
802
542
185
162
169
84.1
41.6
19.2
15.4
15.4
15.4
1,680
5,800
241
83.4
126
11,300
9,990
13,800
Dissolved
Cd-jig
241
238
241
246
249
242
242
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
2.28
3.30
2.73
1.19
1.03
1.02
0.53
0.25
0.15
0.06
0.03
0.03
0.05
0.09
0.03
0.86
1.89
92.40
87.40
104.00
Dissolved
Pb-Hg
0.16
0.19
0.39
0.37
106
125
145
0.17
0.12
0.13
0.11
0.11
0.10
0.10
0.12
0.21
0.15
0.15
0.19
0.19
0.17
0.04
0.06
0.07
0.07
0.10
0.10
0.08
0.05
0.05
0.04
0.03
0.04
0.05
0.06
0.16
0.14
0.18
3.88
3.63
9.68
19
-------
Table 3-5. Column Test Results - Flow Rate - 5 mL/min
Reagent Name Bed
Volumes
Control - Sand 0.4
33.1
61.6
73.3
159.5
213.3
268.7
Apatite II™ 8.7
38.2
61.7
62.8
69.0
72.0
95.6
137.9
172.4
197.2
GFH 7.8
31.9
51.1
57.1
78.9
86.4
113.7
141.9
Bauxsol™ 21.8
26.1
38.7
122.3
149.2
250.2
274.1
285
377.6
PH
(SU)
7.62
6.80
6.56
6.40
6.41
6.58
6.56
7.13
6.79
-
6.68
6.70
6.73
6.82
6.69
6.72
6.62
6.91
6.33
5.20
5.60
6.53
7.49
6.82
6.63
10.51
9.79
7.52
6.62
6.55
6.30
6.32
6.54
6.37
ORP
(mv)
14.2
249
194.5
141.2
130.4
163.5
152.9
145.6
165.6
-
103.3
118.2
94.8
82.8
107.8
101.9
101.3
178.9
209.3
152.1
173.3
92.3
91.5
112.3
128.0
-28.3
27.6
137.8
152.4
191.7
142.2
154.9
135.7
130.9
SC
(US/cm)
297
338
340
338
338
340
341
417
360
-
360
360
361
365
371
376
363
250
236
238
236
228
229
244
297
522
414
363
347
343
343
343
345
340
Dissolved
Zn-ng
41,200
41,700
41,800
41,700
41,300
41,600
41,800
15.4
15.4
15.4
15.4
20.5
25
110
656
1,530
2,830
3,760
229
234
57.2
15.4
19.2
5,640
21,900
25.8
23.6
137
7,870
19,900
29,900
28,900
29,300
32,000
Dissolved
Cd-ng
242
246
243
243
247
242
243
0.03
0.03
0.19
0.09
0.03
0.03
0.17
0.09
0.36
0.78
14
1.2
1.2
0.37
0.07
0.06
0.22
5.68
0.05
0.03
33.2
89.4
144
190
189
191
204
Dissolved
Pb-H.g
0.16
12.6
111
96.3
166
164
171
0.12
0.18
0.24
0.15
0.11
0.12
0.13
0.16
0.14
0.15
0.09
0.08
0.07
0.12
0.03
0.04
0.04
0.03
0.52
0.53
0.50
7.49
31.5
77.0
67.5
72.0
94.5
20
-------
Table 3-6. Column Loading Capacities - by Reagent
Reagent Name Total Column Total Zinc Mass of Media in
Effluent Removed Column
mL US 8
Sand - 2 mL/min
Sand - 5 mL/min
Apatite II™ - 2
mL/min
Apatite II™ - 5
mL/min
GFH - 2 mL/min
GFH - 5 mL/min
Bauxsol™ - 2
mL/min
Bauxsol™ - 5
mL/min
Media Removal
Capacity
28,727
65,703
28,399
19,070
24,483
19,540
3,285
1,990
54,161
14,315
1,186,640
796,374
1,014,967
806,864
136,900
83,059
425
425
40
40
250
250
25
25
127.4
33.7
29,666
19,909
4,060
3,227
5,476
3,322
21
-------
4. Conclusions
Of the seven media tested, the majority were not
appropriate for treating Canyon Creek
groundwater in a passive manner. Only three
media, GFH, Bauxsol™, and Apatite II™, were
considered feasible for testing within columns.
Of those three media, Apatite II™ has a much
higher capacity to remove Zn (the primary
element of concern) from this water when
compared to the other two media.
A number of operational parameters associated
with the use of Apatite II™ exist that would
have to be addressed to use the media for the
commercial treatment of this water. These
parameters include, but are not limited to the
tendency of the media to become anaerobic and
foster the development of SRB, the tendency of
the media to swell when it is hydrated, and the
known problems (i.e., high effluent biochemical
oxygen demand and nutrients, hydrogen sulfide,
odors, dissolved Mn, iron, and potentially
arsenic) that occur with plugging when the
media is used without being diluted with an inert
material such as sand (McCloskey, 2003).
It should also be noted that the use of SRB
represents another reasonable method of treating
metal-laden waters of this type.
22
-------
5. References
Benjamin, M. M. and Sletten, R.S., Metals
Treatment at Superfund Sites by Adsorptive
Filtration, U.S. E.P.A., Office of Research and
Development, 1993.
The Dow Chemical Company, DOWEXIon
Exchange Resins and Adsorbents, May, 1997.
McCloskey, L., Final Report - Nevada Stewart
Permeable Reactive Barrier Project, U.S. E.P.A.
Mine Waste Technology Program, 2003.
Millard, S.P., Environmental Stats for S-Plus
Software Program, 2002.
Pourbaix, M., Atlas Of Electrochemical
Equilibria In Aqueous Solutions, National
Association of Corrosion Engineers, 1974
U.S. Environmental Protection Agency,
Guidance for Data Quality Assessment, EPA
QA/G-9.
U.S. Environmental Protection Agency, USEPA
Contract Laboratory Program National
Functional Guidelines for Inorganics Data
Review, Document OSWER 9240.1-45, EPA
540-R-04-004, October 2004.
23
-------
Appendix A
Summary of Quality Assurance Activities
-------
Summary of Quality Assurance Activities
Mine Waste Technology Program
Activity III, Project 48
BACKGROUND
The following is a summary of the quality assurance (QA) activities associated with MWTP Activity III,
Project 48, Passive Treatment for Reducing Metal Loading. Analytical samples and field data were
collected according to the schedule outlined in the approved project-specific QA test plan. All field and
laboratory data available has been evaluated to determine the usability of the data. Data from both critical
and noncritical analyses were evaluated. Critical analyses are analyses that must be performed to
determine if project objectives were achieved, while noncritical analyses provide additional information
about process control, as well as information that is of interest to project participants. The critical
parameters, dissolved Cd, Pb, and Zn, were analyzed to support the project objective of evaluating
passive treatment media that may have potential to reduce metals loading contribution from Canyon
Creek by 50%.
INTERNAL TECHNICAL SYSTEMS AUDIT
An internal technical systems audit of the equilibrium capacity studies and laboratory column studies was
performed by Mindy McCarthy, the MSB MWTP QA Manager, on February 7 and continued through
April 23, 2007. There were no findings, five observations, and one additional technical comment
identified during the audit.
On February 7 and 8, a segment of the equilibrium capacity studies was observed to evaluate the
documentation associated with sampling and testing studies (i.e., field logbook, test sheets, and sampling
documentation), and to evaluate adherence of the test and sampling procedures to the QA test plan in
order to verify data quality. On April 23, a segment of the laboratory column studies was observed and
evaluated for the same criteria as the equilibrium capacity studies portion of the audit. The observations
resulting from the audit included improving field logbook documentation, noting deviations from the QA
test plan in the final report, and adding valuable technical descriptions to the final report. To improve
field logbook documentation, the field/sampling personnel were to ensure that the field personnel
performing the daily study activities initialed each day's logbook entry. The appropriate amendments
were made and implemented to correct the observation.
Three observations pertained to deviations from the QA test plan. The first observation concerned a
deviation regarding the number of passive media that were tested during the equilibrium capacity studies.
The QA test plan indicated that 12 passive media would be tested at five different mass loadings during
the equilibrium capacity study. Based on results from the equilibrium rate study, the passive media used
during the equilibrium capacity testing was reduced to seven. Another observation noted a deviation
concerning the number of passive media tested during the laboratory column studies. The QA test plan
indicated that four passive media would be further tested in a column configuration. However, only three
passive media were selected for the column studies, based on results of the equilibrium capacity studies.
Another observation was that GFH might not have been properly characterized, based on its Freundlich
plot. However, additional runs were performed after the original equilibrium capacity study testing and
the mass loadings were re-plotted to properly characterize the media. There was no variation from the
A-l
-------
output of the GFH data that was reanalyzed from the original plot. The last observation was to include
valuable technical descriptions in the final report to further clarify the study procedures.
Am additional technical comment was made because a statement in the sampling procedures in
Section 5.1 of the QA test plan was not applicable. It was noted in the test plan that collected samples
shall leave no headspace in the container to minimize air entrainment because the entrained air could react
with species in the samples and affect analytical results. For the analyses presented in the plan, this
provision was not applicable, so entrained air would not affect the outcome of the analytical results. This
was necessary to mention because the samples taken during the studies did have headspace in the
sampling containers. Upon observation, the samples were collected in the appropriate manner.
DATA EVALUATION
Data that was generated throughout the project was validated. The purpose of data validation is to
determine the usability of data generated during a project. Data validation consists of two separate
evaluations: an analytical evaluation and a program evaluation. The analytical evaluation focuses on
laboratory data validation, field logbook evaluation, and field data evaluation, while the program
evaluation concentrates on chain-of custody procedures, sampling and data completeness, and field
quality control (QC) samples.
ANALYTICAL EVALUATION
An analytical evaluation of all data was performed to determine the usability of the data that was
generated by MSB Laboratory for the project.
Laboratory Data Validation
Laboratory data validation was performed using USEPA Contract Laboratory Program National
Functional Guidelines for Inorganic Data Review (USEPA, 2004) as a guide. The data quality indicator
objectives for critical measurements were outlined in the QA test plan and 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 A-l. In addition to the data quality indicators listed in Table A-l,
internal QC checks, including calibration, calibration verification checks, calibration blanks, matrix spike
duplicates, interference checks, method blanks and laboratory control samples were used to identify
outlier data and to determine the usability of the data for each analysis.
The validation of laboratory data determined that all analyses were performed within specified holding
times, calibration procedures were correctly performed, and laboratory blanks contained no significant
contamination. Check standards, duplicate sample analysis, and spike sample analysis were performed at
the proper frequencies and were within the specified control limits. In some instances, sample
concentrations were greater than four times the spike concentrations, and the recovery limits were not
applicable. Serial dilutions were within acceptable limits, indicating that there were no matrix effects.
Measurements that fall outside of the control limits specified in the QA test plan or method requirements,
or for other reasons were judged to be outlier, are normally flagged appropriately to indicate that the data
was judged to be estimated or unusable; however, there were no data from this project requiring flags.
A-2
-------
Field Logbook Evaluation
Field data validation began with an examination of the project logbook and project logsheets that were
generated for this project. The general logbook contained logs of the daily study activities as well as the
instrument calibrations for field measurements. Project log sheets were used to document study
information for the passive media equilibrium rate studies data, the equilibrium capacity studies data, and
the laboratory column studies data. Details for sample collection including sample dates and times,
sample identification, sampling personnel, and field measurements were documented on these log sheets.
Information recorded on the passive equilibrium rate studies log sheets included the particle size, pH and
temperature measurements, and media mass and test water volume. The information that was recorded on
the log sheets for the batch testing included the pH, temperature, ORP, and specific conductance (SC)
measurements, as well as the adsorbent masses and test water volumes. Information recorded on the
column testing log sheets included the effluent flow, effluent volume, pressure differential, pH, SC, and
ORP measurements and the sample identification of samples taken.
Field Data Evaluation
Field data validation was performed to determine the usability of the data that was generated during field
activities. The usability was determined by verifying that correct calibration procedures on field
instruments were followed. All of the field measurements were classified as noncritical.
Measurement of pH was manually performed by MSB personnel using the hand-held YSI multimeter.
The pH meters had automatic temperature compensation and were capable of measuring pH to ± 0. 1 pH
units. The meter was calibrated daily prior to analysis using fresh 4, 7, and 10 buffer solutions.
The QA test plan requirements for pH measurements, both the required frequency and the required
sample measurements, were realized.
Oxidation-Reduction Potential
Oxidation-reduction potential was measured using the YSI multimeter with a silver/silver chloride
reference electrode to determine the ORP of samples. The electrode calibration was verified at the
beginning of every sampling event using a solution of known ORP. The measured standard ORP
measurements were documented, and the ORP values were within ± 20% of the known solution value.
The QA test plan requirements for ORP measurements, both the required frequency and the required
sample measurements, were realized.
PROGRAM EVALUATION
The program evaluations included an examination of data generated during the project to determine that
all field QC checks were performed and within acceptable tolerances. Program data deemed inconsistent
or incomplete and not meeting the QC objectives outlined in the QA test plan would be viewed as
program outliers and flagged appropriately to indicate the usability of the data; however, there were no
data flagged.
A-3
-------
Chain-of-Custody Procedures
Information provided on the chain-of-custody was accurate and complete, and any discrepancies noted by
MSB Laboratory were communicated with the project manager and resolved through laboratory
corrective action procedures.
Sampling and Data Completeness
All samples that were supposed to be collected were collected and were analyzed for the required
analyses as outlined in the QA test plan.
Field QC Samples
In addition to internal laboratory checks, field QC samples were collected to determine overall project
performance. All field QC samples were collected at the proper frequency for tests specified in the QA
test plan. None of the field blanks collected for the project showed significant contamination. Field
duplicates showed very good agreement to the original sample. No samples required qualification due to
field QC sample results.
QA SUMMARY
The project personnel conducted QA/QC activities for this project in accordance with the procedures
outlined in the QA test plan. All critical activities were documented in the field logbook and on the
project log sheets, field instrumentation was properly calibrated, samples were properly collected, and
field QC samples were appropriately taken. Based on the quality assurance activities of MWTP, Activity
III, Project 48, the data generated was of sufficient quality to evaluate project objectives.
Table A-l. Data quality indicator objectives for precision, accuracy, MDL, and completeness.
Parameter
Dissolved Cd
Dissolved Pb
Dissolved Zn
Matrix
Aqueous
Aqueous
Aqueous
Unit
|ig/L
|ig/L
|ig/L
MDLa
1
1
10
Precision1"
<20%
<20%
<20%
Accuracy0
75-125%
75-125%
75-125%
Completeness*1
90%
90%
90%
a MDLs 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. MDLs will be adjusted as necessary when dilutions of concentrated
samples are required.
b Relative percent difference of analytical sample duplicates.
0 Percent recovery of matrix spike, unless otherwise indicated.
d Based on number of valid measurements, compared to the total number of samples.
A-4
------- |