EPA/600/R-14/249
July 2014
Pilot-Scale Treatment of Virginia Canyon
Mine Drainage in Idaho Springs,
Colorado, USA Using Octolig®
by
Barbara A. Butler
Land Remediation and Pollution Control Division National Risk
Management Research Laboratory Cincinnati, OH 45268
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Notice
This report is based on pilot scale testing conducted by Arcadis U.S., Inc. (contract EP-C-10-028),
with sampling and analyses by EPA Region 8 and Arcadis. The research was funded under the
Region 8 Regional Applied Research Effort (RARE) program. The RARE program is a mechanism
used by the Office of Research and Development's (ORD) Regional Science Program to respond
to high-priority, near-term research needs of EPA's regional offices. The Octolig® adsorption
technology was proposed by Arcadis in response to a Request for Proposals (RFP). The
information presented and the views expressed herein are strictly the opinion of the author
and in no manner represent or reflect current or planned policy by the USEPA. Mention of trade
names or commercial products does not constitute endorsement or recommendation of use.
This document has been reviewed in accordance with U.S. Environmental Protection Agency
policy and approved for publication.
II
-------
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 ground water; 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 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.
Cynthia Sonich-Mullin, Director
National Risk Management Research Laboratory
III
-------
Table of Contents
NOTICE ii
FOREWORD Ill
LIST OF FIGURES VI
LIST OF TABLES VII
ACKNOWLEDGEMENTS VIII
1 INTRODUCTION 1
1.1 Research Need 1
1.2 Purpose and Objectives 2
1.3 Study Site 3
1.4 Process Tested 2
2 METHODS 4
2.1 Preliminary Studies 4
2.2 Field Pilot Test 5
2.2.1 System Design 6
2.2.2 Water Treatment Configuration 6
2.2.3 System Control 7
2.2.4 Sampling and Analysis 7
2.2.5 Quality Assurance/Quality Control 9
2.2.6 Data Analysis 9
3 RESULTS AND DISCUSSION 11
3.1 Quality Assurance 11
3.1.1 Field Replicate Samples 11
3.1.2 Analytical QA 12
3.1.3 Comparisons of Data between Labs 12
3.2 Water Samples 13
3.2.1 Parameters 13
IV
-------
3.2.2 Comparison to Water Quality Criteria (Primary Objective 1) 17
3.2.3 Removal efficiencies for target metals (Primary Objective 2) 18
3.2.4 Removal efficiencies for ions of secondary interest (Secondary Objective 1) 21
3.2.5 Regeneration process effectiveness (Secondary Objective 2) 23
3.2.6 Metal removal due to pH adjustment versus sequestration by Octolig® 23
3.2.7 Anions 26
3.3 Regeneration Concentrate Samples 29
3.4 Sludge 32
3.4.1 Sludge comparison (Secondary Objective 3) 32
3.5 Recovery and Reuse (Primary Objective 4) 34
3.6 Capital and O&M Costs (Primary Objective 5) 37
4 CONCLUDING REMARKS 38
5 REFERENCES 40
6 APPENDICES 42
6.1 Appendix A: Arcadis (EPS) Operational SOPs for Treatment System Operation and Regeneration 42
6.2 Appendix B: Equipment/Parts List from Arcadis 46
6.3 Appendix C: Process Flowand Equipment Placement 47
6.4 Appendix D: Equipment Installation and Operation Notes from Arcadis 49
6.4.1 Installation and Fall 2011 Operations 49
6.4.2 Spring 2012 Operations Notes 49
6.5 Appendix E: Field Parameters and Analytical Data for Water and Sludge Samples 56
6.6 Appendix F: Estimates of Costs for 50 GPM Octolig® and Conventional Lime Treatment Systems 62
V
-------
List of Figures
Figure 1. Argo Tunnel Water Treatment Plant location, 2300 Riverdale Drive, Idaho Springs, CO. Figure
from https://maps.google.com/ accessed September 2, 2014. Map data ©2014 Google; image ©2014
Google date: June 2008 3
Figure 2. Virginia Canyon (1st and 2nd from left) and Big Five Tunnel (far right) outfalls. Photo by
Arcadis 4
Figure 3. Argo Tunnel on right, sump region in lower left, and Big Five Tunnel and Virginia Canyon
discharges in background left. Photo by Arcadis 4
Figure 4. Branched polyethyleneimine (BPEI) structure (copied from Lindoy et al, 1999) 3
Figure 5. Schematic diagram of Octolig® showing attachment of BPEI to silica gel. Figure prepared by
Jon Forbort, Arcadis, based on Lindoy et al., 1999 3
Figure 6. Schematic diagram showing tanks and sampling points through the treatment system 8
Figure 7. Hardness through the treatment system overtime 14
Figure 8. Calculated total dissolved solids (TDS) through the treatment system over time 15
Figure 9. pH through the treatment system over time. EPA data are plotted for 4/5, 4/9, 4/16, 4/20, and
5/30; all other data were provided by Arcadis. The dashed lines represent the pH set-points (4.0 for the
lst-stage pH dosing and 8.0 for the 2nd-stage pH dosing) 16
Figure 10. Sodium concentration at each sampling port overtime 20
VI
-------
List of Tables
Table 1. Average chemistries of the drainages from the Argo Tunnel, the Big Five Tunnel, and Virginia
Canyon (modified from U.S. EPA, 2007) 4
Table 2. Stability constants for Octolig® (Martin, 2010)1 4
Table 3. Water quality criteria 11
Table 4. Comparison of effluent (SP3) concentrations with water quality criteria (WQC); gray highlighted
cells indicate data meeting the WQC. Based on dissolved concentrations, except where noted for Al and
Fe 17
Table 5. Comparison of effluent (SP3) concentrations with influent (SP1) concentrations for primary
target metals. Gray highlighted cells indicate data meeting the desired 90% removal 18
Table 6. Comparison of effluent (SP3) concentrations with influent (SP1) concentrations for secondary
interest elements 22
Table 7. Comparison between pre- and post-regeneration removal efficiencies for Cu and Fe in the 1st-
stage of the treatment system 23
Table 8. Percentage of metals removed by each step in the overall pilot treatment system, percentage
removed over the whole system, and pH and alkalinity at each step 25
Table 9. Comparison of effluent (SP3) concentrations with influent (SP1) concentrations for F, Cl, and
SO4 27
Table 10. Percentage of Cl, F, and SO4 removed and corresponding pH at each step and over all steps in
the pilot system 27
Table 11. Concentrations and masses of ions in regeneration concentrate samples 29
Table 12. Estimated masses of ions associated with each Octolig® bed 30
Table 13. Percent recovered during regeneration of the Octolig® beds 31
Table 14. Theoretical sludge composition expected from lime treatment of Virginia Canyon water with
removal efficiency equivalent to that observed in the Octolig® pilot system 33
VII
-------
Acknowledgements
The project was funded through U.S. EPA Region 8's Regional Applied Research Effort (RARE)
program. The RARE program is a mechanism used by the Office of Research and Development's
(ORD) Regional Science Program to respond to high-priority, near-term research needs of EPA's
regional offices. The research team comprised a number of individuals from Arcadis (Jon
Forbort, Chris Lutes, and Carl Singer), EPS (subcontractor to Arcadis) and EPA (Region 8:
Kathleen Graham, Tim Rehder and Christina Wilson; ORD: Barbara Butler and Robert Ford). The
research team thanks Mary Boardman from the Colorado Department of Public Health and
Environment for arranging access to the Argo Treatment Plant site for the pilot study, Mike
Holmes (EPA Region 8, retired) the Remedial Project Manager for the Argo Tunnel Superfund
site, employees working at the existing Argo Treatment Plant for their coordination during the
pilot treatment study, and Jeff McPherson and Michael Bade in the EPA Region 8 lab in Golden,
CO for field supplies and water quality analyses. The author thanks EPA team members and
peer reviewers Richard Graham (EPA Region 8) and Diana Bless (EPA ORD) for their comments
and suggestions for improving earlier drafts of this report.
VIII
-------
1 Introduction
1.1 Research Need
Thousands of abandoned mines are scattered throughout the Rocky Mountains in EPA Region
8. The mining activities that disturbed the lands at these sites left exposed metals-laden rock
and the interaction of water with these rocks has resulted in acidic rock drainage and other
environmental problems. In particular, these waters contain high concentrations of metals and
have low pH, each of which serves to damage the aquatic ecosystems and pose a threat to
human health. Many of these sites are in remote locations having steep topography, limited
access to power supplies, limited land on which to construct a treatment facility, and limited
access for maintenance and monitoring.
To combat aquatic ecosystem damages and human health threats, water treatment facilities
are in use or are planned at a dozen Superfund cleanup sites in Region 8. These treatment
facilities extract heavy metals such that the water will not cause further problems downstream.
Most of Region 8's treatment systems are expensive to operate, and they generate large
quantities of sludge. For example, the Argo Treatment Facility, located in Idaho Springs,
Colorado, generates 10 cubic yards of sludge per day from alkaline chemical treatment of the
water. Non-alkaline chemical treatment technologies exist for removal of metals, but few have
been implemented in the field specifically for the use of treating acid-mine drainage water.
Additionally, few have examined reuse of the metal(s) recovered. To date, only one site has
implemented a system to recover metals and reuse them: a sulfide precipitation system
(BioteQ Environmental Technologies Inc., http://bioteq.ca/operations/wellington-oro-co/) is
operated by Breckenridge Colorado's Water Divison to treat drainage from the Wellington-Oro
mine.
While the primary focus of water treatment is to remove metals from the environment to allow
recovery of the affected ecosystem and to reduce the threat to human health, failing to reuse
the metals and simply disposing of them represents a missed recycling opportunity, including
the potential for recovery of some costs through income generated by recycling, and results in
additional mining activities to acquire these metals. The additional mining activities then have
the potential to exacerbate the environmental issues that EPA currently is addressing.
Knowledge of the chemical and physical parameters of the water is critical when considering
treatment options in mining-influenced watersheds. Typically, pH and alkalinity are low;
acidity, hardness and ionic strength are high; and metal (e.g., aluminum and iron)
oxyhydroxides precipitate rapidly upon aerobic mixing of the mine drainage with natural
-------
streams having near neutral pH. Water temperature, which typically is a function of both
climate and altitude, also is an important variable for many treatment processes.
As well as removing contaminants from the water to meet water quality standards, other
technical requirements of treatment systems for remote locations in EPA Region 8 include
being compact (e.g., < 54' x 9', size of an 18-wheel trailer), requiring minimal
maintenance/monitoring, the ability to function in low temperatures owed to high altitudes,
and having a recyclable waste stream . Applicability of a given technology typically depends on
the specific site's water quality characteristics (e.g. specific metals and concentrations present),
and therefore it is desired to determine technologies that will both remove contaminants to
meet water quality standards and produce waste residuals that may be recycled and reused.
Treatment technologies are needed for removing metals from mining-influenced water that
would allow for reuse of the metals, while producing less sludge than traditional methods.
These capabilities would reduce operation and maintenance (O&M) costs at mines by
decreasing the volume of sludge requiring disposal, and potentially offset some costs through
an income-generating waste stream. This Region 8 RARE project performed a small scale pilot
field study (non-permanent structure) of the Octolig® adsorption technology that was proposed
by Arcadis in response to a Request for Proposals (RFP) to address these technology treatment
and reuse needs.
1.2 Purpose and Objectives
The purpose of the pilot study was to evaluate the utility of the Octolig® technology at a
representative Superfund site in the Clear Creek Watershed, based on assessment of the
project objectives. The project was a 6-week field pilot study to extract metals, including
aluminum (Al), cadmium (Cd), copper (Cu), iron (Fe), lead (Pb), and zinc (Zn), from mining-
impacted water into a form that makes them readily reusable, through either smelting or
another recovery process.
Primary Objectives:
1. To meet or exceed (i.e., be better than) site-specific (watershed) water quality criteria
for Al, Cd, Cu, Pb, and Zn (see Table 3)
2. To attain > 90% removal efficiencies for Al, Cd, Cu, Fe, Pb, and Zn
3. To produce a minimal waste stream, with attainment of a sludge volume of at least 30%
less than traditional treatments, such as with lime
4. To evaluate reuse of the metals recovered via smelting or another process
5. To obtain capital and O&M costs associated with the technology's use at this
representative site and a cost estimate for scaling up to 50 and 300 GPM
-------
Secondary Objectives:
1. To determine removal of silver (Ag), arsenic (As), calcium (Ca), potassium (K),
magnesium (Mg), manganese (Mn), sodium (Na), nickel (Ni), and selenium (Se)
2. To determine the effectiveness of the regeneration process for the Octolig® medium
(i.e., evaluating removal overtime to see if efficiency of the medium changes)
1.3 Study Site
The Argo Tunnel Water Treatment Plant (Argo WTP) in the Clear Creek Watershed in Idaho
Springs, Colorado (Figure 1) treats mine drainage from three sources: the Argo Tunnel, the Big
Five Tunnel, and Virginia Canyon.
Figure 1. Argo Tunnel Water Treatment Plant location, 2300 Riverdale Drive, Idaho Springs, CO.
Figure from https://maps.google.com/ accessed September 2, 2014. Map data ©2014 Google;
image ©2014 Google date: June 2008.
The average chemistries and flows of the three sources, approximated based on historical data,
are presented in Table 1 (EPA, 2007). The water from the Argo Tunnel feeds directly into the
-------
WTP equalization basins, and discharges from the Virginia Canyon groundwater and the Big Five
Tunnel drain from pipes in front of the Argo Tunnel into the equalization basins (EPA, 2007).
Flows from the Big Five Tunnel and Virginia Canyon can be controlled.
Table 1. Average chemistries of the drainages from the Argo Tunnel, the Big Five Tunnel, and
Virginia Canyon (modified from U.S. EPA, 2007).
Parameter
Aluminum (mg/l)
Copper (mg/l)
Iron (mg/l)
Manganese (mg/l)
Zinc (mg/l)
PH
Ave rage f low (GPM)
Argo Tunnel
20
4
120
90
40
3
200 to 450
Big Five Tunnel
5
1
65
30
8
5.5
15 to 40
Virginia Canyon
80
9
3
90
92
3
5 to 180
The photos in Figures 2 and 3 show the collection areas of these drainages prior to the water
being treated at the Argo WTP.
Figure 2. Virginia Canyon (1st and 2nd from
left) and Big Five Tunnel (far right) outfalls.
Photo by Arcadis.
Figure 3. Argo Tunnel on right, sump region
in lower left, and Big Five Tunnel and
Virginia Canyon discharges in background
left. Photo by Arcadis.
-------
During the preliminary stages of the project (see Section 2.1), it was noted that precipitation
began as soon as these three water sources mixed; therefore, it was decided to use water from
a single source. The Virginia Canyon drainage was chosen primarily due to it having the lowest
relative concentration of Fe and the highest relative concentration of Cu, which currently has
the highest value of the metals present in the drainages.
1.4 Process Tested
Octolig® is a pH-responsive immobilized ligand that has a strong affinity for heavy metals. The
technology is based on the chelation of heavy metals using a branched polyethyleneimine
(BPEI) ligand that has been bonded to the surface of a silica gel with a silane linking group
(Lindoy 1993; 1999). Figure 4 shows an example of the BPEI structure from Lindoy et al. (1999)
and Figure 5 is a schematic diagram showing attachment of BPEI to silica gel to form Octolig®.
Octolig® retains high amounts of metal ions from low concentration solutions in a pH range
from 2 to 10. An acidic solution is used to release and concentrate the ions, which regenerates
the Octolig® for further metals removal.
It is believed that the ligand selectivity of Octolig® follows the Hard Soft Acid Base (HSAB)
principle - i.e., that a metal ion is a Lewis acid and the Octolig® ligand is a Lewis base. Octolig®
is a soft base comprising primary, secondary, and tertiary amines (Figure 4). In the Irving-
Williams series for select alkaline earths and divalent transition metals, the following trend in
strength of complexation is observed:
Ba2+ < Sr2+ < Ca2+ < Mg2+ < Mn2+ Zn2+
The proton and metal-binding stability constants for Octolig® have been provided for a single
deprotonation reaction only in the form of a conference presentation (Table 2), and thus have
not been subjected to external technical peer review; however, the reported constants
generally follow the Irving Williams series.
-------
JIN NH
\ w
NH H —
J C,
NH:
}
UN
NHi
o
NH H*N
c
Figure 4. Branched polyethyleneimine (BPEI) structure (copied from Lindoy et al, 1999).
*•
Step One:
Silane Linking Group Grafted To
Surface Silanol Groups of Silica
Gel
<
'-/-?
•*»—'
&mtc*Ga
Silane Linking Group
(3 -chloropropy I )-trimethoxy»lbJte
<
Mica Gel wf unwn0 oraup
\
-»v
1 I
Ligand
Branched
Polyethyleneimine
(BPEI)
Step Two:
BPEI Ligand Grafted To Linking
Group
-------
Table 2. Stability constants for Octolig® (Martin, 2010)1
Cation
H+
Cd2+
Co2+
Cu2+
Fe2+
Mn2+
Ni2+
Zn2+
Equilibrium Species
HL
ML
ML
ML
ML
ML
ML
ML
LogK
10.2
None provided
15.6
22.4 (24)
11.1
None provided
19.1
16.2
1 ACS presentation not subjected to external technical peer review
Octolig® has been shown to take up arsenate, chromate, polymolybdate, selenious acid and
fluoride, but not boric acid, in individual anion bench tests (Martin, 2010). Although nitrate and
sulfate apparently do not interfere with the uptake of metals by Octolig®, since initial bench
scale tests for patent development used metal salts having these forms (Lindoy et al., 1993;
1999), in a mixture study Stull and Martin (2009) observed removal of nitrate and sulfate from
solution, as well as phosphate and nitrite. The mechanism proposed for uptake of anions by
Octolig® is one of encapsulation (Stull and Martin, 2009; Martin et al., 2010), with a weak
association between multiple protons on the nitrogen atoms of the ethyleneamine chain and
the anion. This encapsulation likely is a weaker association than the metal ligand complexes by
which the Octolig® is proposed to bind reversibly with metals (Lindoy, 1993).
2 Methods
2.1 Preliminary Studies
Preliminary testing of the Octolig® media was conducted by the supplier: EPS of Hebron, KY,
subcontractor to Arcadis. Water was collected from the combined water source at the Argo
Tunnel site in June 2010 and shipped to EPS's testing facility in Cincinnati, OH. The mixed water
had a nominal pH of 2.5 and was saturated with Fe3+, which resulted in the precipitation of
large amounts of ferric hydroxide when sodium hydroxide was added to raise the pH (ferric
hydroxide precipitates at pH > 3.5). Because it was desired to evaluate the Octolig® technology
for its ability to remove metals through chelation, rather than through precipitation and/or
adsorption of ions onto precipitates in the pH modifying steps, this source water was deemed
inappropriate to meet the project goals. Instead, the Virginia Canyon water source was chosen
for use due to it having the lowest historical average Fe concentration (3 mg/l, Table 1) and the
4
-------
highest historical average concentrations of both Cu and Zn (9 and 92 mg/l, respectively, Table
1).
Aluminum hydroxide precipitates at pH > 4.5. The Virginia Canyon water contains an average of
80 mg/l of Al (Table 1) and a white precipitate, assumed to be aluminum hydroxide mixed with
some ferric hydroxide (solid phase was not tested for positive identification), was observed to
begin to form when the pH was adjusted to 4.3. These results indicate that careful control of
pH is necessary to maximize sequestration of metals by the media rather than by sorption onto
any aluminum and/or iron hydroxides that might form during the pH adjustment stage of the
process prior to the water being run through the Octolig® media.
To allow for pilot system sizing, time for regeneration, and pH set-points, tests were conducted
to determine the mass transfer zone, determine the breakthrough for major ions in the water,
and to determine the media capacity at anticipated field pilot conditions. These preliminary
tests indicated that Fe and Cu were successfully removed at low pH (~ 4) and that Al, Cd, Cr, Co,
Ni, and Zn were removed at higher pH (~ 6), which supported the use of a two-stage design for
pilot testing.
2.2 Field Pilot Test
The desire was for the pilot treatment system to meet the objectives (see Section 1.2),
including producing a product that had the potential to be processed to recover the metals
sequestered. The approach used was to treat the water in two treatment beds having differing
pH values to selectively remove the metals based on their affinity toward the Octolig®. The first
bed of Octolig® was designed to operate at pH 3.5 - 4, followed by pH adjustment to pH 6.0 -
8.0, and then by an additional bed of Octolig® designed to operate at pH 6.0 - 8.0. These pH
ranges were chosen due to the following:
• Minimum pH values at which metal sorption was expected with Octolig®, based on
bench-scale testing
• The magnitude of changes in pH values expected to result in differing metals sorption
for divalent cations
• They were below pH values expected to lead to the formation of hydroxide precipitates
(other than Fe and Al)
• pH values that would not require neutralization prior to being discharged
-------
This section describes the system design, water treatment configuration, and water sampling
locations and methods. Standard operating procedures (SOPs) for the system in both the water
treatment and regeneration modes are provided in Appendix A; major equipment and parts are
listed in Appendix B; figures showing the process flow diagram and spatial arrangement within
the trailer are provided in Appendix C; and notes regarding system deployment and operations
are provided in Appendix D.
2.2.1 System Design
In Appendix C, Figure Cl presents the process flow diagram (the water treatment configuration
of this diagram is described in Section 2.2.2) and Figure C2 presents a visualization of the spatial
arrangement of the equipment within an 18-wheel trailer, shown in 2 orientations. Equipment
located outside the trailer included the water conveyance lines, a 2,300 gallon clarifier tank
(T23), and bulk chemicals stored in a small shed (not shown in Figure C2). Supplies needed
during testing were moved into the trailer in smaller portable containers.
2.2.2 Water Treatment Configuration
This paragraph describes the water treatment process flow diagramed in Figure Cl (the
sampling ports and associated tanks are shown in Figure 6). The pilot began March 26, 2012
and ended May 7, 2012, with the first day being dedicated to filling the 1st tank with sufficient
source water to begin running the treatment system. Raw influent water from Virginia Canyon
was gravity fed into the 1st stage pH-adjustment tank (T10). Flow was controlled by a
mechanically-actuated float valve. Using a dosing pump, 25% NaOH solution (on and after April
27, a 50% solution was used) was added to the water in Tank T10 mixed with a mixing pump) to
adjust the pH with a set-point of pH 4.0, before being directed to Tank T13 (1st stage feed tank)
and then fed into the 1st stage Octolig® treatment tank (T18), with flow rates ranging from 4.2
to 15.2 GPM over the testing time. The Octolig® treatment tank was operated in upflow mode.
From Tank T18, the water was conveyed to the 2nd stage pH-adjustment tank (T20), with a set-
point of pH 8.0 and mixed with a mixing pump. Overflow from Tank T20 was conveyed to the
clarifer settling tank (T23). Overflow from Tank T23 flowed to the 2nd stage feed tank (T24).
Water then was pumped (flow ranged from 4.2 to 15.3 GPM over the testing time) to the 2nd
stage Octolig® treatment tank (T25), which was operated in upflow mode. Following treatment
in Tank T25, the water was conveyed to the effluent clear-well tank (Til). Effluent from Tank
Til was pumped back to the headworks of the Argo WTP.
During the field pilot-scale test, the system was taken off line for media regeneration. In a
larger scaled operation, parallel beds would be used to perform regeneration without causing
downtime. Scheduling of regeneration was based on the results from the bench testing and
6
-------
attempted to balance maximum loading onto Octolig® with maximum water treatment
efficiency. Also considered was the desire to conduct this part of the process and restore the
system to operation within the available time on site.
Field Zn testing suggested breakthrough from the 1st Octolig® bed on April 18, but testing on
April 20 and 23 suggested resumed Zn removal. Due to both time constraints and field Zn
results, the 1st Octolig® bed was regenerated April 24-26. Both the 1st and 2nd Octolig® beds
were regenerated at the end of the demonstration (May 7-8). A pH 1.8-2.2 solution of sulfuric
acid (H2S04) was used to regenerate the 1st Octolig® bed (Tank T18) and a H2S04 solution with
pH 3.8-4.2 was used for the 2nd Octolig® bed (Tank T25). Sulfuric acid was used due to the
potential for recycling of the regenerant via hydrometallurgical processing (see Section 3.5).
After the first bed was regenerated in April, the pH of the bed was adjusted back to the
operating pH (4.0) before water treatment resumed. Steps for regeneration are included in
Appendix A.
2.2.3 System Control
Zn was monitored using a HACH colorimetric test (www.hach.com, pocket colorimeter II) to
monitor real-time system performance. Zn was chosen because it showed moderately weak
binding to Octolig® in preliminary work and was present at high concentration in the source
water. Mn, Fe, and Cu concentrations also were monitored using field HACH kits
(www.hach.com, pocket colorimeter II) to evaluate system performance, along with monitoring
of pH.
2.2.4 Sampling and Analysis
Figure 6 is a simplified schematic of the tanks and water quality sampling ports from Figure Cl
relative to the two stages of the system.
-------
NaOH added
SF
rginia
anyon —
fluent
i
•»
^- —
'^--^ _
T
•~-— _
,
_^- "
LO
-^
,
'-*-^^
' — ~~,
-^
T13
•*, _w-"
SP
Lb
Feed
tank
f
•••-»_
' — "^
~*
T18
SP
2
Octolig®
NaOH added
s-
• — __
T;
— -%.
— *
>o
^-
'
— --.
__----
T23
SF
'5
Clarifier
s^
1
-^
_— --
T24
Feed
tank
,-—
**
-,
__---
T25
1 — — -
Octolig8
Til
SF
3
To existing WTP
influent
Effluent
tank ,
>i lst-stage of
treatment system
2nd-stage of
treatment system
Figure 6. Schematic diagram showing tanks and sampling points (in red) through the treatment
system.
Grab samples were collected by both EPA and Arcadis to attempt a greater temporal coverage
with the limited funding for the project. EPA collected grab samples on 4/5, 4/9, 4/16, 4/20,
4/30, and 5/3 from sampling ports SP1, SP2, and SP3. QA duplicates were collected on 4/9 and
4/30. To evaluate any removal of metals through precipitation and/or adsorption to
precipitates following the pH-modification steps, versus removal by Octolig®, EPA collected
additional samples at locations SPlb and SP5 on 4/5, 4/16, and 5/3.
Arcadis collected grab samples at SP1 and SP3 on 4/7, 4/9, 4/11, 4/16, 4/18, 4/27, 4/30, and
5/4; and at SP2 on the same dates, except 4/27 and 4/30. Regenerant solutions and sludge
were sampled also by Arcadis. Composite regenerant samples were collected from three
depths from Tanks T26 (1st stage regeneration tank) and T28 (second stage regeneration tank)
on 5/8 and the remaining solution was blended back into the water stream to be treated by the
Argo WTP. Tanks T26 and T28 are shown in Figure Cl. A sludge sample was collected from
Tank T23 on 5/4; sludge not needed for analyses was disposed of within the main Argo WTP's
routine sludge management system by a manually controlled pump.
EPA-collected samples were analyzed at the EPA Region 8 laboratory in Golden, Colorado for
chloride, fluoride, and sulfate (EPA 300.0), total recoverable and dissolved Al, Cu, Fe, Mn, Ni,
and Zn and dissolved Ca, K, Mg, and Na by ICP-OES (EPA 200.7/6010); and total and dissolved
-------
Ag, As, Cd, Pb, and Se by ICP-MS (EPA 200.8/6020). Dissolved ions were measured on samples
that had been filtered in the field at 0.45 u.m. Field parameters measured included pH, specific
conductance (SC), and temperature. Alkalinity (EPA 310.1) and hardness (SM 2340B) were
measured and calculated, respectively, in the laboratory. Hardness was calculated using the
following equation: CaCOs
Hardness (mg/l as CaC03) = [2.497 x Ca2+ (jig/I) + 4.18 x Mg2+ (jig/l)]/WOO
Arcadis-collected samples were analyzed by Test America in Denver, Colorado for total
recoverable Ag, Al, As, Ca, Cd, Cu, Fe, K, Mg, Mn, Na, Ni, Pb, Se, and Zn (EPA 6010C).
Parameters measured on grab samples collected by Arcadis included pH (field), specific
conductance (SC), and total suspended solids (TSS). Sludge depth was measured in TankT23
and samples collected were analyzed for TSS, and total metals analysis (EPA 3050/6010C).
2.2.5 Quality Assurance / Quality Control
EPA field duplicate samples (collected within 10 minutes of one another) were compared using
relative percent difference (RPD). EPA and Arcadis each collected samples on April 9, 16, and
30; while these samples can't be considered as split samples, they were collected within two
hours from one another and their RPDs were compared to determine the applicability of using
total recoverable (dissolved analysis was not conducted on Arcadis-collected samples) data
from both labs for analytes measured using similar analytical techniques (i.e., total Al, Ca, Cu,
Fe, Mg, Mn, Ni, K, Na, and Zn measured by ICP-AES) to evaluate system removal efficiency over
time.
Calculation of RPD: %RPD =100*^ -JT2)/[(JT1 +JT2)/2] where Xi = value from replicate 1;
X2 = value from replicate 2.
Analytical methods included the use of laboratory method blanks, control standards and lab
duplicates, matrix spikes and duplicates, and dilution series samples.
2.2.6 Data Analysis
Data were provided by EPA Region 8 laboratory in both Excel and Adobe PDF formats; data
were transposed to another Excel spreadsheet for data analysis and presentation of results.
Test America provided data in Adobe PDF format, which also was transposed into Excel
spreadsheets for data analysis and presentation of results.
Observed effluent (SP3) concentrations of Al, Cd, Cu, Fe, Pb, Mn, and Zn (all but Fe and Mn
were target metals for Primary Objective 1) were compared to available acute and chronic
9
-------
water quality criteria (WQC) to determine if the concentrations in the effluent from the
treatment process were equal to or less than the WQC values. State of Colorado criteria for the
South Platte River Basin (Colorado State Regulation 38) and site-specific WQC for the Clear
Creek Watershed segment where Argo is located and downstream (Segment 11 of the Clear
Creek Basin in the Colorado State Regulation 38) are provided in Table 3. There is no state or
site-specific criterion for Al; therefore, the National WQC is included in Table 3 and was used for
comparison. Some metals have criteria based on average water hardness. Tim Steele of TDS
Consulting [personal communication July 29, 2010] provided an average high-flow hardness
value of 56.5 mg/l as CaCOs for calculations of criteria in this stream segment1.
Influent (untreated mine-drainage: SP1) and effluent (following all treatment steps: SP3)
samples were compared to determine the percentage removal (Primary Objective 2) for target
metals of interest (Al, Cd, Cu, Fe, Pb, and Zn). The equation used was: 100 x (Effluent - Influent)
/ (Influent). This was done also for Ag, As, Ca, K, Na, Ni, Mg, Mn, and Se to evaluate Secondary
Objective 1. Additionally, removal of ions was compared between SP1 and SPlb, SPlb and SP2,
SP2 and SP5, and SP5 and SP3 for the three days the additional sites (SPlb and SP5) were
sampled to evaluate the potential for removal by precipitation and/or adsorption processes
resulting from pH-modification.
1 When asked what hardness value was appropriate for use at this site, Mary Boardman of the Colorado Department of
Public Health & Environment (CDPHE) suggested we contact Tim Steele, who has done work in this watershed for
many years. Although a footnote to the equations states the hardness value should be "based on the lower 95 per cent
confidence limit of the mean hardness value at the periodic low flow criteria as determined from a regression analysis
of site-specific data" (Colorado State Regulation 38), Tim Steele stated that the average value should be used instead
[July 29, 2010].
10
-------
Table 3. Water quality criteria.
Metal
Aluminum (ac) tree
Aluminum (ch) tree
Cadmium (ac)
Cadmium (ch)
Copper (ac)
Copper (ch)
Iron (ch) diss
Iron (ch) tree
Lead (ac)
Lead (ch)
Manganese (ac)
Manganese (ch)
Zinc (ac)
Zinc (ch)
Water Quality Criterion (ug/l)
State of Colorado1
none given
none given
1.7
0.3
8
5
none given
1000
35
1.3
2469
1364
88
76
Colorado Site-specific1
none given
none given
none given
1.42
none given
17
300
1000
35
1.3
2469
50
215
187
National2
750
87
ac = acute; ch = chronic; tree = total recoverable
Hardness-based values were calculated using the seasonal high-flow average hardness value (56.5
mg/l as CaCO3, n=263); non-hardness based values are highlighted in gray
Site-specific Cd (ch) is a temporary modification to expire 7/1/15
For metals without a state or site-specific criterion, the national value is presented
1 http://www.Colorado.gov/cs/Satellite?c=Page&childpagena me=CDPHE-
Main%2FCBONLayout&cid = 1251595703337& pagename=CBONWrapper (Site-specific is Segment 11
of the Clear Creek Basin)
2 http://water.epa.gov/scitech/swguidance/waterquality/standards/current/index.cfm
3 Results and Discussion
3.1 Quality Assurance
3.1.1 Field Replicate Samples
Only flouride and silver in the replicated filtered samples for SP1 collected on April 9 exceeded
20% RPD, with values of 27% and 33%, respectively. The RPD values for the majority of
analytes for all other sampling dates and all locations were less than 5%. The values for Ag in
samples collected 4/9 were 0.5 and 0.7 u.g/1, which are close to the reporting limit of 0.5 ug/l;
therefore it is likely the high RPD for the field replicates is related to their values being low. The
first replicate for each sampling date was used for subsequent analyses and evaluation of
objectives.
11
-------
3.1.2 Analytical QA
Values for total Ag in SP1 samples for sampling dates 4/5 and 4/9 and SP3 collected 4/5 were
not used because Ag was found in the blank at concentrations exceeding those in the samples;
these are flagged with a "B" and data are not reported or used. Total Ag values for SP1 samples
collected 4/16 and 4/20 and SPlb and SP5 collected on 4/16 were not used because control
sample criteria were not met. The RPD for the laboratory duplicate sample for Ag was 23.3%
for filtered SP1 samples collected 4/5 and 4/9; this slightly exceeded the 20% criterion and the
samples are flagged as estimates. The RPD limit of 10% was slightly exceeded (10.4%) in the
serial dilution for the total Cd concentration on the sample from SP1 collected on 4/16.
Because this deviation was small, the sample result is reported, but flagged as an estimate. The
matrix spike criterion was not met for fluoride (F) in SP1 collected on 4/20 and the sample
result is flagged as an estimate. K, Mg, Ni, and Zn serial dilution results for SP2 collected on
4/16 exceeded the RPD criteria of 10% with values of 14.4, 11.6,10.3, and 16.1%, respectively;
these analyte results are flagged as estimates. No other QA issues were observed for the EPA
data.
Cu concentration in the laboratory blank for samples collected 4/16, 4/18, and 5/4 exceeded
the method detection limit (MDL), but was below the reporting limit (RL) and was present at
concentrations below 10% of the measured sample concentrations; therefore, samples were
considered usable. Fe, Mn, and Na concentrations were identified in the laboratory blank for
samples collected on 5/4, but concentrations were much less than 10% of the values measured
in the samples and the samples were considered usable. As, Cd, Pb, and Ni failed the matrix
spike and/or matrix spike duplicates for the SP1 sample collected 4/7. Batch matrix spikes
conducted by Test America in subsequent analytical runs did not capture the Arcadis-collected
samples from SP1, SP2, or SP3; therefore, it is not possible to determine if the observed
negative matrix effect was evident in any other sample and/or sampling date. As, Cd, and Pb
were measured by ICP-OES by Test America and by ICP-MS by EPA and therefore are not
comparable; results from the more sensitive ICP-MS method were used for data analysis. No
other QA issues were observed for the Test America data.
3.1.3 Comparisons of Data between Labs
Ca, Mg, Na, and K are conservative ions generally found in the dissolved phase in water
samples. EPA measured these ions on filtered samples and Test America measured these ions
12
-------
on un-filtered samples. Because they were expected to be predominantly in the dissolved
phase, the results from the unfiltered and filtered samples were compared. The RPDs for Ca
and Mg were < 5% for SP1, SP2, and SP3 samples from 4/9 and 4/16 and the RPDs for K and Na
ranged from 6 to 21% for the same samples. All but Mg, Mn, and Zn exceeded 20% RPD in the
comparison between analytes measured via ICP-OES (Al, Ca, Cu, Fe, Mg, Mn, Ni, K, Na, and Zn)
on samples collected on 4/30 for SP1 and SP3; SP2 was not sampled by Arcadis on 4/30 so could
not be compared. Except for K and Fe, analytes in SP1, SP2, and SP3 samples collected on 4/9
and 4/16 compared well, with RPD values < 20%; however, because the majority of analytes
exceeded 20% RPD on 4/30 for both the SP1 and SP3 samples, the majority of analytes
compared across 4/9, 4/16, and 4/30 appear biased low compared to concentrations in the EPA
samples, and the reason for these deviations is not known, it was decided to exclude the Test
America ICP-OES data for the water samples. Exclusion of these data did not alter the
outcomes from assessment of project objectives or conclusions made.
3.2 Water Samples
Field parameters and water quality results for each sampling port are presented in Appendix E.
Sampling ports and their corresponding locations are described below and shown in Figure 6;
all but sampling point SPlb are shown also in Figure Cl:
• SP1 = Virginia Canyon raw influent, pre-pH adjustment in Tank T10
• SPlb = effluent from the lst-stage pH adjustment, prior to lst-stage Octolig® treatment,
port located on Tank T13
• SP2 = effluent from the lst-stage Octolig® treatment, port located on Tank T18
• SP5 = effluent from 2nd-stage pH adjustment, prior to 2nd-stage Octolig® treatment, port
located on Tank T23
• SP3 = effluent from the 2nd-stage Octolig® treatment, port located on Tank Til
3.2.1 Parameters
TSS was measured by Arcadis fora few of the sampling dates (Appendix E). TSS was below
detection in SP1 samples analyzed and ranged from 2 to 8 mg/l at SP2 and from 8.8 to 18 mg/l
at SP3. This increase through the system likely is due to the transport of unsettled precipitates
formed from addition of NaOH.
Hardness remained essentially constant between about 1,050 to 1,100 mg/l as CaCOs through
the lst-stage of the treatment system (SP1 through SP2, Figure 7). At SP5, hardness was 947
13
-------
mg/l as CaCOs on 4/5, but returned to ~ 1,000 mg/l as CaCOs for the other sampling events.
Hardness at SP3 was ~ 900 mg/l as CaCOs for four of the six sampling events and was lower
than the corresponding influent concentrations on all sampling dates. That hardness decreased
below the influent concentrations only at SP5 and SP3 suggests that removal of Ca and Mg
occurred predominantly in the 2nd-stage of the treatment system.
1,200 T
1,000 --
o
u
ro
U
1/1
ro
in
OJ
ro
800 --
600 --
400 --
200 --
SP1
SPlb
SP2
Sampling Port
SP5
SP3
D 4/5/2012 m 4/9/2012 H4/16/2012 04/20/2012 Q4/30/2012 H 5/3/2012
Figure 7. Hardness through the treatment system over time.
While total dissolved solids (IDS) is not a water quality parameter in the watershed where the
pilot was conducted, and its assessment was not an objective in this work, whether the process
removes or adds IDS is of interest for potential use of the treatment at other sites in EPA
Region 8 where there might be a IDS limit. IDS was calculated from dissolved concentrations
of ions measured (Appendix E) and concentrations are plotted in Figure 8. IDS concentration
remained essentially constant through the Ist-stage of the system, with the exception of
decreases observed at SP2 on 4/5 and 4/16, and a smaller (relatively) decrease observed on
4/9. Over the entire treatment system, IDS concentration was decreased on 4/5 and 4/9, but
was similar to, or higher than, the influent concentration for the remainder of the study. The
14
-------
higher IDS concentration observed at SP3 on 5/3 likely is due to the sulfate concentration being
higher than was typically observed (see Section 3.2.7 and Appendix E).
3,500 -i
oo 3,000 -
2,500 - r-1™_,c,.1 = n n H
OJ
u
U 2,000 -
Q
^ 1,500 -
"o
tn
T3
OJ
-| 1,000 -
to
1/1
T3
1 500 H
n
-
I
-
=
_ i_
-
=
_ i—
n
i— i
-
i — r
=
_ i—
n
-
B
-
i— i
~
-
rj7
—
— 1
SP1
SPlb SP2
Sampling Port
SP5
SP3
D4/5/2012 m4/9/2012 S4/16/2012 B4/20/2012 D4/30/2012 B5/3/2012
Figure 8. Calculated total dissolved solids (IDS) through the treatment system over time.
The pH was measured in the field by Arcadis when collecting samples for metals analysis and
for monitoring system performance. Some pH data from system performance checks were
provided to EPA, with other pH data discussed in operations notes (see spring 2012 notes in
Appendix D). EPA measured pH on dates when water samples were collected. Results are
plotted in Figure 9 for EPA-measured pH values on dates when EPA collected water samples
and results from Arcadis are plotted for dates when EPA did not collect samples and for dates
when pH was monitored (and provided) without any sample collection.
The higher pH observed at SP2 (following the Ist-stage Octolig® treatment) versus SPlb (pH
set-point 4.0) for measurements obtained from 4/5 to 4/16 is believed to be due to pH
buffering effects of the Octolig®, which was experienced in bench scale testing by EPS
(subcontractor for Arcadis), as well as during testing by MSE (1998) using Berkley Pit mine
water. Being an intermediate sampling location, the pH at SPlb was not measured on all dates,
15
-------
but this phenomenon was not observed in measurements on 4/27 or 5/3. This increase in pH
was not seen in the data for the 2nd-stage of the treatment, however, except for the sample
measured on 4/23 when the pH was 4.51 at SP5 and 4.85 at SP3.
7 -
6 -
±! 5 -
"S
ro /i
-a 4 •
n;
5.
x 3 -
a.
2 -
1 -
n
VJ
r
-
n
^
.
-
i
E
I
I
I
E
E
~
;
E
E
*
^
*
*
*
*
*
*
*
r
n
E
E
i
i
i
i
i
=
E
~
;
E
E
r
"
'
;•
I
y
>
>
f
>
>
>
>
>
>
>
E
i
i
E
E
T
':•<
:<
X
^
^
<
^
n
j
1:
.-
1
i,
".
V
V
••••
;•••
V
-...
V
V
•f
'
:••
:••
:••
:••
...
...
...
...
:::
...
...
...
:
..
...
r
-
7
SP1 SPlb SP2 SP5
:
_
:
1
1
I
s
j
:
E
i
E
i
i
E
E
E
I
E
=
E
7
\
...
^ 's ':
^ '" ';
^ -v 1
SP3
Sampling Port
d 4/5/2012 S 4/6/2012 134/7/2012 d 4/9/2012 Q 4/11/2012 B 4/13/2012 E 4/16/2012 Q 4/18/2012
0 4/19/2012 D 4/20/2012 D 4/23/2012 E 4/27/2012 D 4/30/2012 D 5/3/2012 0 5/4/2012
Figure 9. pH through the treatment system over time. EPA data are plotted for 4/5, 4/9, 4/16,
4/20, and 5/30; all other data were provided by Arcadis. The dashed lines represent the pH set-
points (4.0 for the lst-stage pH dosing and 8.0 for the 2nd-stage pH dosing).
The pH at SP2 was above the lst-stage pH set-point for most of the study, except when it
dropped to 3.7 on 4/20 and at the end of the study (4/27 and beyond). Based on the
monitoring results at the intermediate sampling location SPlb, the pH in the lst-stage feed tank
(T13) did not reach the pH set-point of 4.0 after April 13th. On May 2nd, an overheating failure
of the lst-stage base pump was diagnosed, but field repair efforts did not succeed in returning
the pump to service before the scheduled end of the demonstration (Appendix D).
The pH in the 2nd-stage of treatment never met the set-point of 8.0. The April 5 and 9 samples
from SP3 and the April 5 sample from SP5 were the only samples having measurable alkalinity,
which correspond to pH values above 6. The pH at SP3 was 7.52, 7.46, and 6.27 on April 6, 7,
16
-------
and 27, as measured by Arcadis in the field; however, alkalinity was not measured on those
dates.
3.2.2 Comparison to Water Quality Criteria (Primary Objective 1)
Observed effluent (SP3) concentrations of Al, Cd, Cu, Fe, Pb, Mn, and Zn (all but Fe and Mn
were target metals for Primary Objective 1) were compared to available acute and chronic WQC
to determine if the treatment process met or exceeded them (i.e., concentrations were less
than the WQC values). The comparison is presented in Table 4. Unless otherwise noted, the
comparison is based on dissolved concentrations.
Table 4. Comparison of effluent (SP3) concentrations with water quality criteria (WQC); gray
highlighted cells indicate data meeting the WQC. Based on dissolved concentrations, except
where noted for Al and Fe.
Metal
Al (ac) tree
Al (ch) tree
Cd (ac)
Cd(ch)
Cu(ac)
Cu(ch)
Cu (ch)
Fe (ch) diss
Fe (ch) tree
Mn (ac)
Mn (ch)
Pb (ac)
Pb (ch)
Zn (ac)
Zn (ch)
Criterion
(MB/I)
750 N
87 N
17H,co
1.42 H
8.0H'CO
5.0H'CO
17
300
1000
2469 H
50
35 H
1.3 H
215 H
187"
Concentration in treatment system effluent (ug/l)
Date
4/5/2012
255
255
54.5
54.5
524
524
524
<100
<100
27600
27600
<1
<1
2880
2880
4/9/2012
1460
1460
163
163
109
109
109
<100
<100
33200
33200
<1
<1
14300
14300
4/16/2012
70400
70400
321
321
2470
2470
2470
687
744
76700
76700
18.5
18.5
74500
74500
4/20/2012
15600
15600
247
247
1150
1150
1150
<100
116
53100
53100
<1
<1
46500
46500
4/30/2012
39500
39500
309
309
2430
2430
2430
<100
379
68600
68600
1.3
1.3
68000
68000
5/3/2012
25800
25800
233
233
2470
2470
2470
<100
286
52900
52900
1.1
1.1
49300
49300
ac = acute; ch = chronic; tree = total recoverable; diss = dissolved
H = hardness-based; N = national standard; CO = state standard
Chronic criteria are lower than acute criteria because consideration is taken for the period over
which an organism is exposed to a lower concentration, while acute toxicity occurs over a short
time from exposure to a higher concentration. There is no site-specific acute criterion for Cu,
only a numeric chronic criterion (17 u.g/1), which is higher than the state's hardness-based acute
and chronic criteria (8 and 5 u.g/1, respectively; see also Table 3). Both the state and site-
specific chronic criteria are compared to the treated effluent Cu concentrations in Table 4.
17
-------
Concentrations of dissolved and total-recoverable Fe, and dissolved Pb, sometimes were below
the reporting limit. In these cases, the reporting limit was compared to the water quality
criteria values to determine if the effluent met the WQC.
Only the total-recoverable Fe chronic criterion (1 mg/l) and the acute Pb criterion were met on
all dates sampled (note: the Pb concentration in the Virginia Canyon drainage was lower than
the acute Pb criterion). Dissolved concentrations of Fe and Pb met the respective chronic
criteria (300 and 1.3 u.g/1, respectively) on all days sampled except April 16. The total-
recoverable Al concentration met the acute criterion only on the first sampling date, April 5. No
other WQC were met.
3.2.3 Removal efficiencies for target metals (Primary Objective 2)
To evaluate whether greater than 90% of the primary metals of interest (Al, Cd, Cu, Fe, Pb, and
Zn) were removed in the treatment process, the effluent concentrations for each date were
compared to the influent concentrations using the following formula:
Percentage Removal (%) = 100 * (Concentration in SP1 - Concentration in SP3) / Concentration
inSPl.
Table 5. Comparison of effluent (SP3) concentrations with influent (SP1) concentrations for
primary target metals. Gray highlighted cells indicate data meeting the desired 90% removal.
Analyte / Parameter
Al - Total (Mg/l)
Al - Dissolved (Mg/l)
Cd - Total (Mg/l)
Cd - Dissolved (Mg/l)
Cu - Total (Mg/l)
Cu - Dissolved (Mg/l)
Fe - Total (Mg/l)
Fe - Dissolved (Mg/l)
Pb - Total (Mg/l)
Pb - Dissolved (Mg/l)
Zn - Total (Mg/l)
Zn - Dissolved (Mg/l)
Date
4/5/12
99.6
99. 9A
81.9
82.1
93.5
93.6
95. 7A
95. 6A
94.4A
94. 1A
95.8
96.1
4/9/12
97.8
99.8
45.9
47.6
98.7
98.7
95.9A
96. 0A
94.4
94.2
76.0
79.8
4/16/12
3.2
1.8
-6E
-4.9
72.1
72.0
70.4
73.3
-1.6
0.5
2.7
3.2
4/20/12
77.8
86.8
18.9
18.5
85.6
87.2
95.5
96. 3A
90.7
94. 9A
32.1
37.8
4/30/12
48.3
52.4
1.5
5.2
71.0
72.3
88.0
96. 6A
81.9
93.3
9.1
7.9
5/3/12
62.7
68.3
22.9
25.3
69.3
70.3
90.8
94. 0A
83.3
92.5
25.4
31.2
A = effluent sample below RL, value for RLused for calculations to indicate that
removal is at least to the RL
E = QA criterion not met for either influent or effluent, value is an estimate
18
-------
Percentage removal results are provided in Table 5; negative values indicate that the
concentration in the effluent (SP3) was greater than the concentration in the influent (SP1).
Greater than 90% removal of all metals, except for Cd, was observed on the first sampling date.
Cd removal was never greater than the 82% observed on the first sampling date. Zn removal
exceeded 90% only during the first week of sampling. Only dissolved Fe and Pb removal
remained above 90% for the duration of the study.
For most dates sampled, several ions were present at higher concentrations at downstream
ports in the system relative to their concentrations in the immediate upstream samples
(Appendix E). These increases generally were less than 10%, which is within the variability that
could be expected in a continuously flowing system when times between sample collections
might not match residence times, or could be due to impurities in the NaOH used for pH-
adjustment. The concentration of Zn on 4/9/12, however, was 46% (total) and 52% (dissolved)
higher at SP2 than at SP1. The reason for these larger differences is not known, but possible
explanations include desorption from the lst-stage Octolig® bed due to pH differences over
time with preferential sequestration of Cu or another ion, or to leaching of Zn from some
system part.
Removal efficiencies observed for April 16 show the system was not working well for ions other
than Cu and Fe, although their removal was lower (~ 70%) than at the start of the study and
lower than what was desired for project objectives (> 90%). At SP3, total and dissolved
concentrations of Al, Cu, Fe, Mg, Mn, Ni, Pb, and Zn were higher on 4/16 than on any other
sampling date, while Na concentration and pH (3.58) were lower than on any other date
(Appendix E). Also on this date, total and dissolved concentrations of Al, Cu, Fe, Mg, Mn, Ni, Pb,
and Zn were lower in the preceding tank sampled (Tank T23, SP5 - Figures 6 and 8) than in Tank
Til (SP3), while Na was 3.4 times higher, and pH was higher (4.2), but still acidic. Tank T23 is
the settling tank (Figures 6 and 8), where basic water from Tank T20 traveled after being mixed
with NaOH. Overflow from T23 traveled to the feed tank (T24) for the 2nd stage of Octolig®
(Tank T25). The 2nd-stage pH dosing tank (T20) experienced repeated problems over time,
including the dosing pump hose not remaining below the surface level of the tank of base and
the pH not reaching the set-point of 8. The inconsistencies in dosing of NaOH in the 2nd-stage
are evident in differences in sodium concentration overtime at SP5 and SP3 in Figure 12;
concentrations of Na are consistent overtime within the lst-stage of the treatment system
(SPlbandSP2).
19
-------
350 n
300 -
=^. 250 -
20°
CD
u
o
U
O
l/l
150 -
100 -
50 -
SP1
SPlb
SP2
Sampling Port
SP5
SP3
D4/5/2012 D 4/9/2012 H 4/16/2012 04/20/2012 m 4/30/2012 B 5/3/2012
Figure 10. Sodium concentration at each sampling port over time.
When sampled on 4/5, 4/16, and 5/3, the pH of the water in samples from Tank T23 (SP5) was
6.20, 4.20, and 4.26, respectively, and the pH for samples from Tank Til (SP3) was 6.08, 3.58,
and 4.08, respectively (Appendix E and Section 3.2.1), indicating that the pH 8 set for the dosing
tank was not sustained over subsequent tanks, including when it passed over the 2nd-stage
Octolig® bed. Although it cannot be confirmed, it is possible that higher concentrations of ions
observed in the samples from SP3 on 4/16, versus those in the SP5 samples, are due to the
lower pH water causing release of previously sequestered ions from the Octolig® bed, especially
considering that pH 3.8-4.2 was used for the regeneration of the 2nd-stage Octolig® bed
(Section 2.2.2). The reason for the decreased Na concentration at SP3 relative to SP5 on 4/16 is
not known. As a monovalent ion, Na would be expected to complex more weakly than would
the divalent ions; however, at pH values similar to the regeneration pH values, perhaps Na was
being sequestered by the 2nd-stage Octolig® bed as the metals were being released.
Because alkalinity originated from the base addition step in TankT20 and it remained in the
system through the clarifier (SP5) and the 2nd-stage Octolig® bed only on the dates when
removal efficiencies were highest for metals other than Fe and Cu (removal of these metals was
20
-------
predominantly in the lst-stage Octolig®, Section 3.2.6) is another indication that issues with
base addition and pH control in the 2nd-stage of the process hindered overall effectiveness of
the system. The decreased removal of Cu after April 20 could be due to the issues with the 1st-
stage pH pump (Appendix D) and decreased pH values observed at SP2 (Section 3.2.1), or to
sites in the lst-stage Octolig® bed being unavailable for sequestration due to inefficient
regeneration (see Section 3.3).
An additional potential cause for the lower than target (and sometimes acidic) pH evident in
the SP3 and SP5 samples on all dates (Figure 9 and Appendix E) is a lack of sufficient residence
time to achieve chemical equilibrium to neutralize acidity. The differences in Na concentrations
between SP5 and SP3 (other than on 4/5) suggest that the 2nd-stage was not at equilibrium with
the base added (Figure 10). If precipitation was still occurring after water flowed from the pH 8
adjustment tank (T20) into subsequent tanks (clarifier (T23 - SP5), feed (T24), Octolig® (T25)
and effluent (Til - SP3)), protons released during hydrolysis could have decreased the pH in
any of these tanks (e.g., 2AI3+ + 4H20 -> AI203»H20 + 6H+). It is possible also that the residence
time, or insufficient mixing, did not allow for complete physical exposure of the water to the
added base, leading to only partially pH-treated water, with or without unreacted base, flowing
to subsequent tanks. This potential issue might have been more important after April 27 when
a higher concentration of NaOH (50%, versus 25% used prior to that date) would require a
lower volume to be dosed into the large volume of water.
The tanks for pH adjustment were 330 gallon totes (Appendix A). Average influent flow from
April 5 through April 10 was 5 GPM; average influent flow from April 11 to the end of the study
was 10 GPM (Appendix D, Tables Dl and D2). The flow rate across the Octolig® beds ranged
from 4.2 to 15.2 GPM (Appendix D), which originated from the feed tanks subsequent to each
pH-adjustment tank. It was not documented what water level was maintained in the tanks
preceding pH-adjustment, but if held constant with the influent rate, the residence times within
the tanks can be estimated to have been a maximum of 66 minutes (330 gallon / 5 GPM) and 33
minutes (330 gallon / 10 GPM) for 4/5-4/10 and 4/11 - 5/7, respectively. Data indicate that
better removal was achieved at influent flows of 5 GPM versus 10 GPM, which supports the
potential lack of sufficient residence time under higher flow conditions to allow for equilibrium
in the pH-adjustment tanks.
3.2.4 Removal efficiencies for ions of secondary interest (Secondary Objective 1)
To determine the removal efficiencies for Ag, As, Ca, K, Mg, Mn, Na, Ni, and Se, their
concentrations in the effluent samples (SP3) were compared to their initial concentrations in
the Virginia Canyon influent (SP1). Percentage removals for each element were calculated
21
-------
using the equation provided in Section 3.2.3 and are presented in Table 6. Arsenic was not
detected above the reporting limit (4 u.g/1) in the influent or any system sample on any date and
therefore is not included in Table 6.
Greater than 50% removal was observed for Mn, Ni, and Se on the first two sampling dates.
Similar to the target metals, the best performance for all analytes presented in Table 6
generally was observed during the first week of the pilot test. The negative values observed for
Na are due to the addition of sodium from the use of NaOH to adjust the pH, particularly from
the 2nd-stage of the treatment process, as can be seen by the concentrations plotted in Figure
10. Negative values observed for potassium could be due to it being a contaminant in the
NaOH used for pH adjustment in both the 1st and 2nd stages.
The high concentrations of Mg (92 - 120 mg/l) in the Virginia Canyon water were expected to
result in some being removed by the Octolig®. Magnesium is the stronger of the two weakly
bound divalent alkaline earth metals according to the Irving-Williams series (Section 1.4), and
was observed to be removed by the treatment system at percentages greater than those for Ca.
Table 6. Comparison of effluent (SP3) concentrations with influent (SP1) concentrations for
secondary interest elements.
Analyte / Parameter
Ag- Total (ng/1)
Ag - Dissolved (Mg/l)
Ca - Dissolved (Mg/l)
K - Dissolved (Mg/l)
Mg - Dissolved (Mg/l)
Mn - Total (Mg/l)
Mn - Dissolved (Mg/l)
Na - Dissolved (Mg/l)
Ni- Total (Mg/l)
Ni- Dissolved (Mg/l)
Se - Total (Mg/l)
Se - Dissolved (Mg/l)
Date
4/5/12
n/a
28. 6E
5.2
-70.5
31.5
65.4
63.8
-610.4
73.6
74.6
87.3
84.0
4/9/12
n/a
n/a
1.6
-16.9
32.2
51.1
55.9
-659.5
69.3
69.5
79.9
76.1
4/16/12
n/a
30.0
2.7
-2.6
2.7
3.3
2.4
-10.4
0.2
-0.2
1.6
-0.5
4/20/12
n/a
44.4A
8.3
-2.5
27.6
26.6
30.8
-679.6
37.6
41.9
40.0
40.8
4/30/12
44.4A
37.5A
0.4
-7.4
8.3
10.0
10.8
-407.1
12.2
14.4
14.7
14.8
5/3/12
37.5A
28.6A
5.6
0.4
24.3
25.8
29.6
-276.3
31.8
33.8
38.8
32.6
n/a = influent and effluent samples below RLor concentration in blank exceeded
concentration in the influent sample
A = effluent sample below RL, value for RLused for calculations to indicate that
removal is at least to the RL
E = QA criterion not met for either influent or effluent, value is an estimate
22
-------
3.2.5 Regeneration process effectiveness (Secondary Objective 2)
The lst-stage Octolig® bed was regenerated April 24-26 (Section 2.2.2). Preliminary bench
testing had indicated that Fe and Cu were removed at low pH (~4) and Al, Cd, Cr, Co, Ni, and Zn
were removed at higher pH (~6), which provided the basis for the 2-stage system and
expectation of selective removal. The 2nd-stage Octolig® bed was regenerated only at the
conclusion of the pilot; therefore, only Cu and Fe were used to evaluate regeneration
effectiveness.
To compare effectiveness of the regenerated Octolig® media, the removal efficiencies shown in
Table 6 for Cu and Fe collected on dates prior to regeneration were averaged and compared to
the average removal efficiency for dates post regeneration; results are presented in Table 7.
Because the anomalies in the percentage removals from 4/16 were believed to be due to issues
with the 2nd-stage processes (see Section 3.2.3), and Cu and Fe were expected to be removed in
the lst-stage, 4/16 data were included in average for this comparison.
Table 7. Comparison between pre- and post-regeneration removal efficiencies for Cu and Fe in
the lst-stage of the treatment system.
Cu - Total (ug/l)
Cu - Dissolved (ug/l)
Fe - Total (ug/l)
Fe - Dissolved (ug/l)
4/5/2012
98.1
98.2
92.7
91.7
4/9/2012
96.8
96.7
90.1
91.3
4/16/2012
61.1
61.8
55.8
59.9
4/20/2012
43.0
42.1
45.8
47.2
Pre-Regene ration
Average
74.7
74.7
71.1
72.5
4/30/2012
48.9
49.9
30.4
34.2
5/3/2012
25.5
27.2
28.7
-23.2
Post-
Regeneration
Average
37.2
38.5
29.5
5.5
Difference
(post minus
pre)
-37.5
-36.2
-41.5
-67.0
Data in Table 7 indicate that removal of iron decreased over all dates and was not improved
with regeneration. Percentage removal of copper also decreased with time, but a slight
increase was seen immediately following regeneration (4/30 versus 4/20), although not to the
level observed for fresh media (e.g., 4/5) and not sustained. On average, the regeneration
process appears to have been ineffective. For both metals, percentage removal by the lst-stage
of the system was best during the first week of the pilot testing, similar to the overall system
performance for these and other ions (compare with Tables 5 and 6).
3.2.6 Metal removal due to pH adjustment versus sequestration by Octo//'q®
To understand better the removal mechanisms of precipitation following NaOH addition and
sequestration by Octolig® occurring in the system, the percent of target and secondary-interest
ions removed by each step in the process were determined as follows:
/x remaining = C//Co, where Co is the concentration of analyte in the Virginia Canyon water (SP1)
and Q is the concentration of the analyte at the specific sampling port.
23
-------
• Percent removed by lst-stage pH adjustment step (between SP1 and SPlb) = 100 * (1 -
/spib); labeled SPlb in Table 8
• Percent removed by lst-stage Octolig® step (between SPlb and SP2) = 100 * (/spib
remaining -/sp2 remaining); labeled SP2 in Table 8
• Percent removed by 2nd-stage pH adjustment step (between SP2 and SP5) = 100 * (/$P2
remaining -fsps remaining); labeled SP5 in Table 8
• Percent removed by 2nd-stage Octolig® step (between SP5 and SP3) = 100 * (/SPS
remaining -/SPS remaining); labeled SPS in Table 8
The percent of each analyte removed at each step is presented in Table 8 for each of the dates
when sampling was conducted at all five sampling ports, along with the overall treatment
system removals (labeled SP1-SP3) for comparison. Alkalinity and pH also are shown for
comparison.
Negative values indicate that the concentration increased over the particular step of the
process. In some cases, the percentage removal at a given step was greater than the removal
indicated for the system overall, due to water from some sampling locations having higher
concentrations than at the prior sampling point (see Appendix E). Although the actual reason(s)
for some sampling locations having concentrations of a given metal higher than that at an
upstream sampling location is unknown, some potential reasons for observed higher
concentrations between sampling ports are:
• Impurities in the NaOH used for pH-adjustment (e.g., K)
• Residence time within a given tank in the system being longer (or shorter) than the time
required to collect samples from the tanks in question. In other words, the water
collected might not have been the "same parcel" of water at the prior sampling point,
even if well-mixed, due to inherent temporal differences in concentrations in the
influent water and the volume of the water stored in tanks for treatment.
• Inherent error associated with measurements of very low concentrations (e.g., Se) and
very high (e.g., Na) concentrations
Copper and iron were the only metals consistently removed by the lst-stage Octolig®, although
efficiency decreased over time. Al, Ni, and Pb were removed by the lst-stage Octolig® only at
the start of the pilot with removal on later dates due to precipitation in the 2nd-stage (4/16/12)
or a combination of precipitation and sequestration by the 2nd-stage Octolig® toward the end
of the study (5/3/12). Only for the last date when all 5 ports were sampled (5/3/12) is there
indication of the 2nd-stage Octolig® removing any metals at greater than 10% (except ~ 14% of
Cd on 4/5/12), with the other sampling dates indicating removal predominantly through
precipitation, indicated from the samples collected at SPS.
24
-------
Table 8. Percentage of metals removed by each step in the overall pilot treatment system, percentage removed over the whole
system, and pH and alkalinity at each step.
Analyte / Parameter
Al - Total (ng/l)
Al - Dissolved (u.g/1)
Ca - Dissolved (u.g/1)
Cd- Total (ng/l)
Cd - Dissolved (ug/l)
Cu- Total (ng/l)
Cu - Dissolved (ug/l)
Fe- Total (ug/l)
Fe - Dissolved (ug/l)
K- Dissolved (ug/l)
Mg- Dissolved (ug/l)
Mn- Total (ug/l)
Mn - Dissolved (ug/l)
Na- Dissolved (ug/l)
Ni- Total (ug/l)
Ni - Dissolved (ug/l)
Pb- Total (ng/l)
Pb - Dissolved (ug/l)
Se- Total (ug/l)
Se - Dissolved (ug/l)
Zn- Total (ug/l)
Zn - Dissolved (ug/l)
Alkalinity (mg/l as
CaCO3 )
PH
Date
4/5/2012
SPlb
0.28
-0.28
2.00
2.61
-1.32
-0.47
-1.58
0.43
-1.33
0.20
3.64
1.17
1.70
-0.82
0.49
0.00
-11.67
-12.35
0.67
-4.17
2.59
4.04
<5(RL)
3.05
4/5/2012
SP2
56.40
57.69
-2.00
-4.89
-0.99
98.57
99.73
92.31
93.05
-5.30
-2.73
-10.51
-14.68
-2.19
29.66
29.30
31.11
31.76
-11.33
-3.47
-1.50
-7.94
<5(RL)
4.95
4/5/2012
SP5
42.70
47 4rA(sp5andSP3)
1.20
70.16
70.59
-4.50
-4.64
2 ggA(SP5andSP3)
^ gjA(SP5andSP3)
-33.79
26.18
65.50
66.45
-612.84
38.25
39.28
75 ooA(sp5andsp3)
74.70A(SP5andsP3)
98.00
91.67
90.44
95.85
33.7
6.20
4/5/2012
SP3
0.26
,J\(SP5andSP3)
4.00
14.01
13.78
-0.07
0.11
QA(SP5andSP3)
pjA(SP5andSP3)
-31.63
4.45
9.21
10.35
5.46
5.19
6.06
QA(SP5andSP3)
,J\(SP5andSP3)
0.00
0.00
4.26
4.17
47.9
6.08
4/5/2012
SP1-SP3
99.64
99 86A
-------
To a greater degree than was observed for Cu and Fe, it appears that the majority of removal
for each ion from the base addition in Tank T20 was negated by solubilization at the lower pH in
Tank Til (SP3) on 4/16/12 (Table 8). Without having more information about conditions within
Tanks T20, T24, and T25 (see Figures 6 and 8), however, the mechanism for the increased
concentrations at SP3 versus SP5 (negative removal percentages observed at SP3 in Table 8)
cannot be determined.
Table 8 indicates removal of 53% of Na by the lst-stage pH-adjustment on 5/3/12. Sodium
concentrations in the Virginia Canyon influent were relatively constant over time - ranging
from 36,000 to 39,500 mg/l, but the influent concentration reported for the 5/3 SP1 sample
was 77,600 mg/l (Appendix E). The concentrations of Na on 5/3 at SPlb and SP2 were 36,100
and 36,400 mg/l, respectively, which are similar to concentrations at those sampling ports for
the other dates in the study (range 36,200 to 37,800 mg/l) having the lower Virginia Canyon Na
concentrations. Additionally, concentrations of all other analytes in SP1 on 5/3 were similar to
previous sample dates (Appendix E). Therefore, although no error was noted in the laboratory
analytical report or in the field notes, the Na data for SP1 on 5/3 is considered suspect and the
removal observed at SPlb in Table 8 likely is not accurate.
Observed removal of Mg primarily was due to precipitation from addition of NaOH in the 2nd-
stage (SP5) on 4/5 and to both precipitation and sequestration onto the 2nd-stage Octolig® on
5/3. Removal of Se was similar, with removal only observed at SP5 on 4/5 and 4/16, but at both
SP5 and SP3 on 5/3. Nickel appears to interact with Octolig®, having about half of the total
removed on 4/5 being due to the lst-stage Octolig® treatment and about half of the total
removed on 5/3 being due to the 2nd-stage Octolig® treatment; removal on 4/16 appears to be
solely due to base addition.
Coupled with results presented previously, it appears that the overall system was operating as
expected only during the first week of the pilot and that issues with control of pH in the 2nd-
stage, saturation of ligand sites in the lst-stage, and later pH-adjustment pump issues in the
lst-stage hindered overall performance of the process. While sampling at the additional ports
provided insight to the mechanisms for removal of the ions, results were not consistent across
the three sampling dates. Therefore, it cannot be confirmed which mechanisms were
predominating over the course of time when these intermediate sites were not sampled.
3.2.7 Anions
To understand how some anions interact with Octolig®, F, Cl, and S04 were measured in water
samples. Analytical results are provided in Appendix E and overall system removal efficiencies
26
-------
are provided in Table 9. Table 10 presents percentages removed at each step in the treatment
process using the same formulas provided in Section 3.2.6.
Table 9. Comparison of effluent (SP3) concentrations with influent (SP1) concentrations for F,
Cl, and S04.
Analyte
Chloride
Fluoride
Sulfate
Date
4/5/12
-46.2
14.8
22.9
4/9/12
-4.8
42. 9A
13.9
4/16/12
-0.8
0.0
-0.6
4/20/12
-5.8
0E
4.5
4/30/12
-2.7
8.7
1.1
5/3/12
-0.6
-1050B
-40.7
A = effluent sample below RL, value for RLused for calculations to indicate that removal is at least to the RL
B = influent sample below RL, value for RLused for calculations to indicate the minimum increase in value
E = QA criterion not met for either influent or effluent, value is an estimate
Table 10. Percentage of Cl, F, and SC>4 removed and corresponding pH at each step and over all
steps in the pilot system.
Date
4/5/2012
4/16/2012
5/3/2012
Sample Port
SPlb
SP2
SP5
SP3
SP1-SP3
SPlb
SP2
SP5
SP3
SP1-SP3
SPlb
SP2
SP5
SP3
SP1-SP3
Chloride
-3.43
-3.69
-8.71
-30.34
-46.2
-2.39
1.33
0.00
0.27
-0.8
-0.29
-0.29
-0.29
0.29
-0.6
Fluoride
-62.96
3.70
77.778
-3.704
14.8
-6.82
4.55
15.91
-13.64
0.0
n/a
n/a
n/a
-1050.0B
-1050B
Sulfate
-1.68
16.76
5.587
2.235
22.9
-2.81
0.56
1.69
0.00
-0.6
-2.82
1.69
1.13
-40.68
-40.7
pH
3.05
4.95
6.20
6.08
3.14
3.99
4.20
3.58
3.64
3.24
4.26
4.08
B = influent sample below RL, value for RLused for calculations to
indicate the minimum increase in value
n/a = influent and effluent samples below RL
Chloride removal percentages in Table 9 are negative (higher concentration in effluent sample
than influent sample), indicating that it was not being sequestered by the Octolig® and
suggesting that it likely was a contaminant in the sodium hydroxide added to the system.
Some fluoride was removed on different dates (Table 9), but it appears it may have been a
contaminant as well; for example, removal is negative at the 1st pH-adjustment sampling point
27
-------
(Table 10) on all dates when the site was sampled. Removal may have been through some
association with Octolig®, e.g., removal at SP2 on 4/5 and 4/16, or through the precipitation
reactions, e.g., removal at SP5 on 4/5 and 4/16. The data do not provide a consistent trend to
evaluate the potential removal mechanism(s) for F.
Sulfate was the predominant anion present in the Virginia Canyon influent water and
concentrations were nearly constant over time at 1,760 to 1,800 mg/l. Results in Table 9
indicate that there was a trend of decreasing removal over the entire system, with some dates
showing potential of an additional source (negative removal percentages). Table 10 indicates
there was some removal via both Octolig® stages (SP2 and SP3) and the 2nd-stage pH-
adjustment step (SP5), but that no consistent trend exists. On the last sampling date, the
concentration of sulfate remained essentially constant through the system (Appendix E) at
close to 1,800 mg/l; however, the concentration in the effluent from SP3 was 2,490 mg/l, which
is ~ 41% higher than the influent concentration (1,770 mg/l). Potential causes for this
difference are 1) an error in the sulfate measurement, although the lab did not note any issues;
or 2) release of sulfate previously sequestered in the 2nd-stage Octolig® treatment bed. The
concentration at SP5 was identical to the input, which indicates that the increase was not due
to contamination from the NaOH added at that stage, as well as indicating that no sulfate was
removed in previous steps on that date.
Martin et al. (2010) showed Octolig removed anions, including selenious acid and fluoride, in
individual bench tests, but they did not specifically test removal of sulfate or chloride. As
shown in Table 6, 80 to 87% of selenium was removed during the first few days of the study,
but then removal ranged from 14 to 40%, excluding the anomalous data from 4/16. Martin et
al. (2010) showed greater than 99% removal of the selenium, but at concentrations much
higher than those in this study: 55 mg/l Se (as HhSeOs) versus 17 u.g/1 Se, with reported removal
to below the detection limit (not provided) from Dl water and to 173 and 229 u.g/1 from well
and tap water, respectively. These results suggest competition with other ions present, which
could explain lower removals observed in this study, especially at the comparatively lower
starting concentrations in Virginia Canyon water. Fluoride removal seen by Martin et al. (2010),
with a starting concentration of 190 mg/l, increased as the total dissolved concentration (TDS)
increased, with 74% removal at TDS of 194 mg/l and 99.7 % removal at TDS of 602 mg/l
(yielding 49 mg/l and 1 mg/l, respectively, remaining in solution). In a mixture study using
sulfate, phosphate, nitrate, and nitrite, Stull and Martin (2009) obtained > 72% removal of
sulfate with concentration decreasing from 30 mg/l to 8.5 mg/l. The pH of the mixture was
8.65, which is higher than the pH obtained in Octolig® beds in this study where sulfate was
observed to be decreased on 4/5 by 17 and 2% at SP2 and SP3, respectively, with pH values of
4.95 and 6.08.
28
-------
3.3 Regeneration Concentrate Samples
Composite samples were collected from three depths from Tanks T26 (lst-stage regeneration)
and T28 (2nd-stage regeneration) and analyzed by Test America for total concentrations of
target and secondary-interest ions. The lst-stage Octolig® bed was regenerated April 24-26 and
May 7-8, while the 2nd-stage Octolig® bed was regenerated only May 7-8. The lst-stage Octolig®
bed was regenerated with a solution of H2S04 having a pH of 1.8-2.2, and the 2nd-stage Octolig®
bed was regenerated with a solution of H2S04 having a pH of 3.8-4.2; the final solutions
contained approximately 1% sulfuric acid. The April and May regenerant solutions for the 1st-
stage were combined in one tank during the pilot. Concentrations and masses of target and
secondary-interest ions present in each of the final regenerant solutions is provided in Table 11.
The final volumes were 200 and 80 gallons for the lst-stage and 2nd-stage regenerant solutions,
respectively. Masses of analytes in the regenerant solutions were determined using the
following equation:
Mass Ibs = volume of regenerant (gal) * concentration (ug/L) * [(3.785411784 L/gal) -f
(453592.37 mg/lb) / 1000 (ug/mg)]
Table 11. Concentrations and masses of ions in regeneration concentrate samples.
Analyte
Aluminum
Calcium
Cadmium
Coper
Iron
Lead
Magnesium
Managese
Nickel
Potassium
Sodium
Zinc
Ist-Stage Regenerant (T28)
Concentration (ng/l)
270,000
250,000
340
220,000
59,000
<13(MDL)
100,000
75,000
680
9900 (J)
310,000
79,000
Mass(lb)
4.5E-01
4.2E-01
5.7E-04
3.7E-01
9.8E-02
<2.2E-05(MDL)
1.7E-01
1.3E-01
1.1E-03
<1.7E-02(J)
5.2E-01
1.3E-01
2nd-Stage Regenerant (T26)
Concentration (ng/l)
200,000
250,000
310
110,000
19,000
37 (J)
110,000
81,000
610
9000 (J)
380,000
85,000
Mass(lb)
1.3E-01
1.7E-01
2.1E-04
7.3E-02
1.3E-02
2.5E-05(J)
7.3E-02
5.4E-02
4.1E-04
<6.0E-03(J)
2.5E-01
5.7E-02
All values are total recoverable
J indicates the concentration was above the method detection limit (MDL), but below the
reporting limit (RL)
Because the amounts of each ion removed via sequestration on the Octolig® bed and removal
through precipitation varied over time, by amount and potentially by mechanism, it is not
29
-------
possible to determine an accurate mass of metals believed to be sequestered on each Octolig®
bed with which to compare the masses recovered in the regenerant solutions. A rough
estimate can be calculated based on the average influent concentrations (totals), the average
percentage removals onto Octolig® in the two stages using data from Table 8 (removals at SP2
for the first stage and SP3 for the second stage), and the total volume of Virginia Canyon water
treated.
Metal removals by the 2nd-stage Octolig® bed on April 16, 2012 were negative (SP3 in Table 8),
which indicates that mass was removed from the bed on that date. With the exceptions of Cu
and Fe, which were associated primarily with the Ist-stage Octolig® bed (Table 8, SP2) as by
design, overall system analyte removals were observed to be lower on 4/16 than on any other
date (Tables 5 and 6), likely due to the behavior observed in the 2nd-stage Octolig® bed.
Because the negative removals observed for the 2nd-stage Octolig® bed on 4/16 do not appear
to be representative of the performance of the bed at other times, and the average removal
represents the average mass of material expected to be present on the Octolig® bed from the
entire time the pilot operated, the SP3 removal percentages in Table 8 on 4/16 were excluded
from the calculation for the average removal by the 2nd-stage Octolig® step. The resulting
masses of metals estimated to be sequestered (and subsequently expected to be removed from
the media with regeneration) using the following equation are presented in Table 12.
Mass Ibs = volume (gal) of Virginia Canyon water treated by stage * average concentration of
analyte (mg/L) * [(3.785411784 L/gal) -f (453592.37 mg/lb)] * average percentage removal for
Octolig® bed/100
Table 12. Estimated masses of ions associated with each Octolig® bed.
Analyte / Parameter
Aluminum
Calcium
Cadmium
Copper
Iron
Lead
Magnesium
Manganese
Nickel
Potassium
Sodium
Zinc
Average Influent
Concentration (ug/l)
(n=6)
71,050
251,333
311
8,803
2,687
19
109,333
77,117
634
5,200
36,860
74,617
Average % Removed
by Ist-stage
Octolig® (n=3)
14.03
0.25
NR
63.99
61.88
8.86
0.28
NR
9.00
NR
NR
1.14
Average %
Removed by 2nd-
stageOctolig® (n=2)
16.99
4.20
14.58
11.30
23.35
34.34
9.58
10.64
10.43
NR
NR
7.78
Mass Removed
by Ist-stage
Octolig® (Ib)
3.5E+01
2.2E+00
NR
2.0E+01
5.9E+00
6.0E-03
1.1E+00
NR
2.0E-01
NR
NR
3.0E+00
Mass Removed
by 2nd-
stageOctolig® (Ib)
4.1E+01
3.5E+01
1.5E-01
3.3E+00
2.1E+00
2.2E-02
3.5E+01
2.8E+01
2.2E-01
NR
NR
1.9E+01
Volume treated in Ist-stage =426,240 gal Ions; volume treated in 2nd-stage =402,600 (Appendix D, Tables Dl and D2)
NR = negative removal calculated
30
-------
A comparison between the estimated masses provided in Table 12 with the masses provided in
Table 11 indicates very low recoveries of the mass of ions that were expected to be present on
the media (Table 13), suggesting the regeneration process was inefficient.
Table 13. Percent recovered during regeneration of the Octolig® beds.
Analyte / Parameter
% Recovered in 1st-
stage Regenerant
% Recovered in 2nd-
stage Regenerant
Al
1.3
0.3
Ca
18.8
0.5
Cd
n/a
0.1
Cu
1.8
2.2
Fe
1.7
0.6
K
n/a
n/a
Mg
15.3
0.2
Mn
n/a
0.2
Na
n/a
n/a
Ni
0.6
0.2
Pb
n/a
n/a
Zn
4.3
0.3
n/a = either regenerant data was flagged or negative average percent removal was calculated for SP2
orSP3
It is possible also that the values in Table 12 are over-estimated and that the system did not
sequester as much of the ions as was indicated using averages. A quick calculation (shown
below) using the 4/5 data (Appendix E and Appendix D, Table Dl), however, shows that
removal of Cu onto the lst-stage Octolig® bed over only the first few days was 2.3 Ibs, which
also is higher than the amount of Cu recovered in the lst-stage regenerant solution (Table 11).
Mass Ibs = (Cu concentration at SPlb - Cu concentration at SP2) u.g/1 * volume treated in 1st-
stage from start to 4/5/12 (gal) * (3.785411784 L/gal * 453592.37 mg/lb) * 1000 ug/mg
Mass Ibs = [299 - 162] mg/l * 32,760 gal * (3.785411784 L/gal H- 453592.37 mg/lb) = 2.3 Ibs
Therefore, it is probable that the regeneration process was not efficient at removing the ions
from the Octolig® ligand, although some error likely also is due to the use of averages with the
system having variability in efficiency overtime. Additionally, it is possible that the pH of the
solutions used for regeneration was not sufficiently low enough to allow release of the
sequestered ions and for the sulfate not to interact with the Octolig® (see Section 3.2.7). For
example, the pH of the solution used for regeneration (pH 3.8-4.2) of the second Octolig® bed
was within a single pH unit or less to the pH measured on effluent from SP3 on a number of the
sampling dates (Appendix E).
MSE (1998) used 4% nitric acid for regeneration of the Octolig® in their column tests and saw
recoveries ranging from 28 to 60% for Al, 56 to 92% for Cu, 15 to 107% for Mn, 40 to 85% for
Fe, and 31 to 114% for Zn. The authors noted that the "fluctuating and low recovery rates
require further study", but suggested that the regeneration time might not have been adequate
(MSE, 1998). It is not known exactly why recovery was poor in this pilot study, but it is possible
31
-------
that regeneration time was inadequate; however, the time for the regeneration process was
not noted in Arcadis' report.
3.4 Sludge
Lime or other chemical treatments commonly are used to remove metals from mining-
impacted water and result in the production of a sludge waste product. The sludge generally is
dewatered and then trucked off-site for disposal, which adds to the cost of the treatment
system. This project sought to evaluate a treatment system for its ability to remove metals
from water into a form that was recoverable and to produce a sludge volume that was at least
30% lower than that produced by traditional treatment systems. Initially it was intended to
make a direct comparison between the sludge produced in the pilot test and the sludge
produced by the Argo WTP through their lime treatment process; however, there are several
reasons a direct comparison was not possible:
1) Only one of the three water sources treated by the Argo WTP was used for the pilot
(Section 1.3).
2) Approaches to dewatering may differ among operators, so a comparison on a mass of
solids basis is more appropriate.
3) The Octolig® pilot suffered from a number of problems (e.g., maintaining desired pH
across system steps) that affected performance over time, so the efficiency of a properly
operating Octolig® system is not known
a. Some analytes showed a wide range of removal efficiencies over time (Tables 5
and 6)
4) Had pH been maintained at the higher value desired, sludge production likely would
have been increased relative to actual results; therefore, the sludge reduction
calculated in this section is not predictive of future performance.
3.4.1 Sludge comparison (Secondary Objective 3)
While the lime process is more efficient and consistent over time than was the Octolig® pilot, a
rough comparison of sludge production can be made assuming a lime process having the same
average removal efficiencies (based on unfiltered samples, except for Mg) as the pilot system,
based on Tables 5 and 6. Because negative removal percentages by the 2nd-stage Octolig® bed
were observed for most analytes on 4/16, which likely is the cause for much lower removals
observed for the system overall (Tables 5 and 6), and this anomaly does not appear to
represent the typical behavior observed at other times (see other dates in Table 8, and Tables 5
32
-------
and 6), the removal percentages calculated for 4/16 in Tables 5 and 6 were not used to
calculate the average removal percentages in Table 14.
Expected sludge generation from a traditional lime process on Virginia Canyon water was
calculated stoichiometrically (Table 14). The approach was based on the average influent (SP1)
concentrations of each metal over the study period; it assumed all metals were removed as
hydroxide precipitates at pH 10.5, with iron as Fe20s»H20 and aluminum as AbOs'HhO; and it
did not include anion removal. The theoretical concentrations of solids that could be produced
were multiplied by the observed removal efficiencies from the pilot system; the efficiency-
adjusted concentrations of solids are shown in Table 14.
Table 14. Theoretical sludge composition expected from lime treatment of Virginia Canyon
water with removal efficiency equivalent to that observed in the Octolig® pilot system.
Aluminum
Cadmium
Copper
Iron
Lead
Magnesium
Managese
Nickel
Selenium
Zinc
MW
(mg/mmol)
27
112
64
56
207
24
55
59
79
65
Average
Influent
Concentration
(mg/l)(n=6)
71.05
0.31
8.80
2.69
0.02
109.33
77.12
0.63
0.02
74.62
Formula
AI203«H20
Cd(OH)2
Cu(OH)2
Fe203«H20
Pb(OH)2
Mg(OH)2
Mn02
Ni(OH)2
Se(OH)4
Zn(OH)2
Formula Mass
(mg/mmol)
119.96
146.40
97.54
177.69
241.19
58.31
86.94
92.71
147.00
99.37
Metal Oxide/
Hydroxide Mass
Expected (mg/l)
157.95
0.41
13.51
4.27
0.02
262.24
122.04
1.00
0.03
113.43
Total
Total Solids
(mg/l)
157.95
0.41
13.51
4.27
0.02
262.24
122.04
1.00
0.03
113.43
675
Average
Pilot System
Removal
Efficiency
(%) (n=5)
77%
34%
84%
89%
89%
25%
36%
45%
52%
48%
Efficiency-
Adjusted
Total Solids
(mg/l)
122
0.1
11.3
3.8
0.0
65.0
43.7
0.4
0.0
54.1
301
To obtain a mass of solids that could be produced, the following formula was used:
Mass solids (Ibs) = volume of water treated (gal) * total solids concentration (mg/L) *
[(3.785411784 L/gal) -f (453592.37 mg/lb)]
The volume of water treated was 426,240 gallons. The calculated mass of solids expected from
lime treatment is 1,069 Ibs.
The Octolig® treatment process produced 5,500 gallons of sludge over the time of the pilot
(chemical analysis of the sludge in TankT23 at the end of the pilot is provided in Appendix E,
data from Test America), with an average TSS concentration in the sludge of 8,950 mg/l. This
equates to 411 Ibs of sludge produced (5500 gal * 8,950 mg/l * [(3.785411784 L/gal) -f
(453592.37 mg/lb)]. The presence of solids was not assessed in any other tanks in the system
33
-------
aside from the settling tank (T23). Therefore, the sludge mass may be a low estimate for sludge
actually produced during the pilot due to the following:
• It does not take into account any solids that settled in the sludge sample prior to being
analyzed by the laboratory (results were reported as TSS rather than total solids)
• It does not take into account any potential solids that might have formed in the 2nd pH-
adjustment tank (Tank T20) that were not carried over into the settling tank (Tank T23)
• It does not take into account any suspended solids that might have been carried over
into tanks subsequent to Tank T23 (e.g., 2nd stage feed Tank T24 or 2nd stage Octolig®
TankT25)
• It does not take into account any potential solids formed in the 1st pH-adjustment tank
(Tank T10 - SPlb), although data in Table 8 indicate minimal loss of metals between SP1
and SPlb, or in Tanks T13 (1st stage feed tank) or T18 (1st stage Octolig® tank)
• It does not take into account any solids present in the clarifier underflow water (17,860
gal)
The reduction in sludge produced in the pilot compared to what could have been produced by
treatment with lime is: 1,069 - 411 = 658 Ibs, which corresponds to 62% [100 -
100*(411/1069)] less solids mass produced by the Octolig® system and satisfies the third
primary objective. For the reasons mentioned previously, however, this comparative reduction
could be overestimated and is not predictive of future performance.
3.5 Recovery and Reuse (Primary Objective 4)
The fourth primary objective of the study was to evaluate reuse of the metals recovered into
the regenerant from the Octolig® treatment process, via smelting or some other process. Some
things that may influence the feasibility of recovering metals from a waste product include:
• Specific components present and their concentrations
• Physical form of the material
• Amount of material (mass or volume)
• Distance to the recycler from the point of material generation
• The number of sites supporting the recycler
• Whether or not the material would be a constant source and of constant consistency
(composition and concentration)
• Market value of the given metal to be recovered
34
-------
Various methods are used to remove metals from acidic solutions. For example, electrowinning
of sulfuric acid leach solutions is a process that is done at a mine site prior to smelting of some
metals, such as Zn. There are a number of centralized facilities in the U.S. that provide off-site
recycling for the plating industry, some of whom use hydrometallurgical processes, which
should be a sufficient method for the regenerant solution from Octolig® treatment.
The National Metal Finishing Resource Center (NMFRC) has an online reference book for metals
recycling (www.nmfrc.org/bluebook). The 'Blue Book' provides a listing of a number of
companies that recycle materials to recover the metals and discusses the source materials
accepted and the methods used. Some of the companies discussed accept aqueous solutions.
Of the companies surveyed by NMFRC and provided in the NMFRC Blue Book, the three most
likely candidates for recycling small volumes of treatment residual similar to that from the
Octolig® treatment process are 1) Horsehead Resource Development Company, 2) CP
Chemicals, and 3) Encycle/Texas Inc. The following paragraph provides a summary of the
capabilities of these three companies from the Blue Book (www.nmfrc.org/bluebook).
Horsehead Resource Development Company operates six facilities located in IL, PA, TN, OK, and
TX. They accept metal-bearing sludges, filter cakes, bag house dusts and soils, and process
them using two kiln technologies. Only the Chicago, IL and Rookwood, TN facilities accept
electroplating wastes. CP Chemical is a major US producer of inorganic metallic salts and
accepts metal bearing wastes from over 1,000 clients through their Environmental Recovery
Services Division, at six plant locations. Their process includes hydrometallurgical steps, but the
details of their method were not provided. Materials accepted and processed include mostly
segregated metal bearing wastes from plating baths, etchants, pickling solutions, and strippers
containing brass, cobalt, copper, nickel, tin, solder, or zinc. Encycle/Texas Inc. operates in
Corpus Christi, TX for approximately 150 electroplating shops. They accept liquid and solid
wastes containing copper, lead, zinc, nickel, and other metals to a lesser extent. Processing is
via chemical and hydrometallurgical methods with their products being used by primary
smelters and others.
It was hoped that the regenerant solutions produced in this pilot study could be transported
and assessed by operators at smelters or other metals recycling facilities for the potential to
recover the metals present. The low masses of metals recovered (Table 11), however, did not
warrant paying the shipping and processing costs required for such an assessment. Instead,
Arcadis provided details on the regenerant solutions to several companies through discussions.
The following paragraphs provide the feedback on the potential for recycling of the regenerant
solutions obtained in the Octolig® process.
35
-------
PM Recovery, which acquired CP Chemical, and said they could not accept the regenerant or
the sludge wastes from the Octolig® treatment process for recycling. A subsidiary of Horsehead
Resource Development Company, INMETCO in Pittsburg, PA, is set up to receive aqueous
plating wastes, but they require the following: 1) payment of a fee for processing; 2) a sample is
needed prior to agreement or quoted price; 3) feed specifications are Ni > 100 mg/l, Cr > 10
mg/l, Cu < 1800 mg/l, P < 860 mg/l, and Sn < 400 mg/l. Based on these requirements, the
regenerant solution from the Octolig® process cannot be accepted because it is not high
enough in either Ni or Cr.
Environmental Quality Company (EQ), which was a company not included in the NMFRC survey,
has the capabilities to recover metal residues from acidic solutions, but they require a higher
purity of copper (or copper and nickel together, which is their desired combination) than what
was present in the regenerant solutions. The company stated that a 2% Cu solution in 1%
sulfuric acid would be acceptable, if it were relatively pure. The concentration of Cu recovered
in the first regenerant solution was 220 ppm (which was 1.8% of the expected mass to be
recovered). If there had been 100% recovery of the mass of Cu expected to be associated with
the lst-stage Octolig bed, this would have resulted in the regenerant having a concentration of
12,222 ppm Cu [200/x = 1.8/100; x = (220 ppm x 100 / 1.8], which equates to a 1.2% solution of
Cu. EQ stated they would accept solutions of low concentrations for recycling, however, they
would charge a fee for processing that may exceed any value obtained from the recovered
metal(s).
Intec Ltd, an Australian hydrometallurgical firm, stated that a key issue for recovery of any
value is that the waste is either of high grade (i.e., high concentration) or high volume, and
preferably both. The company suggested that revenue from a waste containing 20% Cu would
cover some processing costs, if there was sufficient volume of material, but that the value
recovered from a 0.5% Cu solution would not cover any processing costs, or even the costs to
move the material to the recycler.
Given the response of the recyclers surveyed, both by Arcadis and through the NMFRC survey
(www.nmfrc.org/bluebook), it presently appears that it would be quite difficult to economically
recycle the waste product from the Octolig® treatment system with influent water similar to
the Virginia Canyon mine drainage. To do so would require that 1) the Octolig® treatment
system, including the regeneration process, be much more efficient over time than was evident
in this pilot test; 2) concentrations of influent water treated would need to have higher
concentrations of economically-important metals or a larger volume would need to be treated;
and/or 3) that the regenerant solutions be concentrated further (e.g., by evaporation) to allow
for a greater mass of metals per unit volume prior to being sent for recycling.
36
-------
3.6 Capital and O&M Costs (Primary Objective 5)
It was desired in this project to obtain capital and O&M costs associated with the technology's
use at this representative site and cost estimates for scaling up to 50 and 300 GPM for
evaluation of potential use at other sites. Partial costs for this pilot were provided by Arcadis
and are listed below:
• Bench-testing:
o Labor = $20,000
o Analytical = $3,317
• Start-up, including labor to the start of monitoring: $14,660
• Monitoring, maintenance, sampling, and operations post start-up: $11,879
o This cost does not include sampling conducted by EPA, for which there was no cost to
the project
• Analytical costs during monitoring: $3,870
o These costs do not include testing conducted by EPA, for which there was no cost to the
project
• Materials:
o Equipment costs were not provided
o Reagent costs were estimated at $100 per week
o Power costs were estimated at $46 per week
Spreadsheets for estimated capital and annual costs were provided by Arcadis for an assumed
50 GPM treatment having water chemistry similar to that of the Virginia Canyon water. Two
estimates were provided: one assuming the use of the Octolig® system and another assuming
the use of a conventional chemical/physical treatment system using lime. Both estimates were
based on removal of Al, Cd, Cu, Fe, Pb, Mg, Mn, Ni, Se, and Zn, and included bi-weekly sampling
and analytical costs. For the Octolig® system, cost was based on the average overall removal
percentages observed in the pilot for each analyte (Table 14) and 75% recovery of sequestered
analytes from the Octolig® media into the regenerant solution (which is higher than what was
observed in the pilot). Because efficiency was observed to be varied over time and variable
within each of the stages of the system, the cost estimate may not be representative of an
actual 50 GPM system. Costs for the Octolig® system also assume there is no cost for disposal
of the regenerant solution, but don't assume there is a payment for regenerated metals; i.e.,
costs are assumed to be net zero for regenerant. For the lime system, 100% removal of
analytes as hydroxide precipitates was assumed. Details are provided as Appendix F.
37
-------
In summary, the total capital investment for the 50 GPM Octolig® system was estimated to be ~
$4.4 million and for the lime treatment system it was estimated to be $3.3 million. Total annual
costs were estimated to be ~ $578,000 and ~ $412,000 for the Octolig® treatment system and
the lime treatment system, respectively, including sludge disposal costs. Therefore, for the
Octolig® system to have an economical advantage over a traditional lime system, costs
recovered from recycling of metals from the regenerant would need to be ~ $166,000 per year,
or higher, and another ~ $1.1 million would need to be obtained over the course of time to
make up for the capital investment cost differences. Because removal efficiency was observed
to be varied over time and variable within each of the stages of the system, the cost estimate
may not be representative of an actual 50 GPM Octolig® system.
A cost estimate for a 300 GPM system was not provided by Arcadis; however, treatment of so
high a flow with Octolig® as the sole treatment likely would not be possible in a remote location
due to the size of the system that would be required.
4 Concluding Remarks
The first primary objective - to meet or exceed (i.e., be better than) site-specific (watershed)
water quality criteria for Al, Cd, Cu, Pb, and Zn - was not met for Cd, Cu, or Zn, and not met
consistently for Al or Pb. The second primary objective of greater than 90% removal of Al, Cd,
Cu, Fe, Pb, and Zn was never met for Cd, but was met for the other ions on at least one
sampling date, with the best system performance (values of percentage removed and numbers
of ions having > 90% removed) occurring during the first week of sampling. Overall, removal
efficiency was not consistent. At the influent concentrations in the Virginia Canyon mine
drainage, with the exceptions of Fe and Pb, removal of metals to the levels of the water quality
criteria in Table 4 would require greater than 95% removal, with most sampling dates having
influent concentrations requiring greater than 99% removal. The site at which to perform the
pilot study was not decided upon prior to award of the contract; therefore, unlike the Virginia
Canyon drainage, treatment at a different site to > 90% removal might result in concentrations
that would meet the WQC.
The pilot test seemed to operate best during the first part of the study, approximately during
the first week. The lack of the system's efficiency over time appears due to a combination of
things: 1) problems with pH pumps and control during operations; 2) potentially insufficient
retention times or incomplete mixing to reach equilibrium within the pH-control tanks and
within the Octolig® treatment tanks; and 3) inefficient regeneration of the Octolig® media,
38
-------
perhaps due to retention time being too short and/or that the pH was not low enough to allow
complete removal of the sequestered ions.
As well as removing contaminants from the water to meet water quality standards, other
technical requirements of treatment systems for remote locations include being compact (e.g.,
< 54' x 9', size of an 18-wheel trailer), requiring minimal maintenance/monitoring, and the
ability to function in low temperatures owed to high altitudes. The Octolig® pilot demonstrated
that such a system can fit into relatively small locations, with all components, except the water
conveyance lines, clarifier tank, and bulk chemicals, fitting within an 18-wheel trailer, for the
flow treated during the pilot (Figure C2). Larger flows, however, would require a larger system
that may or may not fit onto a small remote site (Table El, a 50ft x 50ft pre-engineered building
was suggested by Arcadis to house a system to treat 50 GPM flows). Low temperatures would
require an insulated trailer, and perhaps a heating source. The pilot initially was planned to
start during the summer/fall, but our hosts at the Argo site were unable to accommodate us at
that time. Instead, the system was started in November 2011, but low temperatures were
encountered and testing was cancelled and postponed to spring 2012 due to freezing issues
resulting in leaks in water and chemical lines. The trailer used was not insulated because pilot
testing during the cooler season was not anticipated; however, these colder temperatures
would be anticipated in any full-scale treatment system operating year-round at remote sites in
Region 8. The pilot test required daily monitoring and maintenance, and it isn't clear that such
frequent monitoring and maintenance wouldn't be required even with a more efficiently
operating system, due to the careful control of pH that is necessary to maximize sequestration
of metals. Based on responses of recyclers surveyed, it appears that it would be difficult to
economically recycle the waste product from the Octolig® treatment system, unless
concentrations of economical metals in the regenerant were higher than those observed in this
pilot study.
Based on the results from this pilot study, Octolig® does not appear to be an appropriate sole
treatment system for remote sites where the desire is to consistently meet water quality
criteria, consistently remove > 90% of the metals present, have minimal monitoring
requirements, and recover metals from the waste stream(s) to offset costs of treatment. It is
possible that further refinement of the process could allow for it to be used as a secondary
treatment; however, it may not be cost-effective without the ability to recycle the metals at a
profit.
39
-------
5 References
Gao, Baojiao, An, Fuqiang, and Liu, Kangkai. 2006. Studies on chelating adsorption properties of
novel composite material polyethyleneimine/silica gel for heavy metal ions. Applied Surface
Science, 253:1946-1952.
Lindoy, Leonard F. and Eaglen, Peter L. 1993. Ion complexation by silica-immobilized
polyethvleneimines. US Patent # 5,190,660, March 2, 1993. Available online:
http://www.freepatentsonline.com/5914044.pdf. Last accessed May 23, 2014.
Lindoy, Leonard F., Eaglen, Peter L., and Alldredge, Robert L. 1999. Immobilized branched
polyalkyleneimines, US Patent #5,914,044, June 22, 1999. Available online:
http://www.freepatentsonline.com/5914044.pdf. Last accessed May 23, 2014.
Martin, Dean F. 2010. Robert Alldredge, chemical engineer, inventor, entrepreneur, developer
of Octolig®, American Chemical Society Division of the History of Chemistry, Presentation. 240th
ACS National Meeting, Boston, MA, August 22-26, 2010. Abstract available online:
http://www.scs.illinois.edu/~mainzv/HIST/meetings/2010-
fall/HIST%20Fall%202010%20Abstracts%20Program%200nly.pdf.
Martin, Dean F., Lizardi, Christopher L., Schulman, Eileen, Vo, Bryan, and Wynn, Darius. 2010.
Removal of selected nuisance anions by Octolig. Journal of Environmental Science and Health Part
A, 45:1144-1149.
MSE Technology Applications Inc. and Montana Tech of the University of Montana. 1998. Final
report - Berkley Pit innovative technologies project phase II: Metre General Inc.,
demonstration. Prepared for US EPA and US DOE. Final Report December 1998. Available
online: http://www.epa.gov/ordntrnt/ORD/NRMRL/std/mwt/a4/a4113.pdf.
Smith, Robert M. and Martell, Arthur E. 1975. Critical Stability Constants. Volume 2: Amines.
Plenum Press, New York.
Stull, Frederick W. and Martin, Dean F. 2009. Comparative ease of separation of mixtures of
selected nuisance anions (nitrate, nitrite, sulfate, phosphate) using Octolig®. Journal of
Environmental Science and Health Part A: Toxic/Hazardous Substances and Environmental
Engineering, 44(14):1545-1550.
40
-------
U.S. Environmental Protection Agency (U.S. EPA). 2007. Remediation system evaluation (RSE)
Central City / Clear Creek Superfund site Argo Tunnel water treatment plant Idaho Springs,
Colorado, Report of the remediation system evaluation site visit conducted May 16, 2007. Final
Report Sept. 2007. EPA-542-R-07-019.
41
-------
6 Appendices
6.1 Appendix A: Arcadis (EPS) Operational SOPs for Treatment System
Operation and Regeneration
www.epswastewater.com
Environmental Products & Solutions
4465 Limaburg Rd. Unit 3
Hebron, KY 41048
Octolig9 Pilot System Standard Operation Procedure
The objective of this Operation Procedure (SOP) is to provide a brief description of the Octolig* pilot unit
systems employed for the Recovery and Reuse of Metals from Mining Influenced Water field technology
demonstration at the ARGO Treatment Plan in Colorado Springs, Colorado. This SOP assumes that the
end user has, at a minimum, reviewed the site specific Health and Safety Plan (HSP) and has general
knowledge and understanding of the pilot system operation.
Pilot System Operational Description
Acid Mine Drainage (AMD) waters will be collected from the Virginia Canyon pipe outlet and conveyed
via gravity to 1st Stage pH Neutralization Tank T10. A flexible coupling will connect a SCH 80 PVC Tee and
the flow through branch will be connected to a hose that will be routed to the pilot system. The branch
connection will be left open to facilitate a siphon break as well as provide a means of overflowing raw
influent during periods of variable flow. A control valve (V-l) is provided to set the flow rate to 15 gpm.
A float valve (V-2) is provided within tank T10 to provide a mechanical shutoff to prevent over filling (the
Virginia Canyon stream will overflow into the Argo Tunnel outfall through the Tee connected to the
Virginia Canyon outlet). You cannot count on V2 to prevent overfilling with the setup. VI must be set to
provide less than adequate flow so pump P14 shuts off occasionally.
Tank T10 will be mixed with Mixing Pump Pll with an integral float switch to control operation (e.g., if
the tank level is drawn because the influent flow rate from Virginia Canyon drops below the established
operational flow rate of the pilot unit Based on current drawing, pump Pll is only stopped in the event
of a catastrophic Tank failure or a failure in the drain pipe integrity) to facilitate neutralization of pH.
Sodium hydroxide will be dosed to the Tank T10 via 1st Stage Base Pump P12 and adjust the pH to the
proper set point (the set point will be established at 4.0 at the startup of the pilot unit). The pH of Tank
T10 will be monitored with a pH sensor, pHll. The sensor is connected to a pH monitor/controller that
provides a visual display of the pH within the tank and also controls the operation of Pump P12. The
42
-------
pump controller provides on/off control, the operator will have to adjust the stroke and frequency
settings of the pump manually to optimize delivery of the sodium hydroxide reagent.
The pH neutralized influent will overflow from Tank T10 into 1st Stage Feed Tank T13. The 1st Stage Feed
Pump P14 will convey the pH neutralized influent into 1st Stage Octolig* Unit T18. Note that 1st Stage
Filter F17 is provided to remove total suspended solids (TSS) prior to entering T18. The valve
immediately after the filter (V8) will be used to choke back flow ofP14 to the desired feedrate (15 gpm
until further notice)as indicated at the flow meter. The influent flow will enter and be distributed at the
bottom of the 1st Stage Octolig* Unit (T18) and uniformly flow upward through the Octolig* bed. The 1st
Stage Octolig* Unit (T18) effluent will then overflow into 2nd Stage pH neutralization TankT20.
Tank T20 will be mixed with Mixing Pump P21 with an integral float switch to control operation (e.g., if
the tank level is drawn because the influent flow rate from Virginia Canyon drops below the established
operational flow rate of the pilot unit. Again, the drawing indicates this is only possible in the event of
tank or pipe failure, not problems with flow.) to facilitate neutralization of pH. Sodium hydroxide will be
dosed to the Tank T20 via 2nd Stage Base Pump P22 and adjust the pH to the proper set point (the set
point will be established at 8.0 at the startup of the pilot unit). The pH of Tank T20 will be monitored
with a pH sensor, pH21. The sensor is connected to a pH monitor/controller that provides a visual
display of the pH within the tank and also controls the operation of Pump P22 (as well as 2nd Stage Acid
Pump {No planned service for this pump} - see description for regeneration cycles). The pump
controller provides on/off control, the operator will have to adjust the stroke and frequency settings of
the pump manually to optimize delivery of the sodium hydroxide reagent.
The pH neutralized process water will overflow from Tank T20 into the Clarifier T23, where aluminum
hydroxide precipitates will settle out and form a sludge blanket at the bottom of the vessel. Clarified
effluent will overflow to 2nd Stage Feed Tank T24. The 2nd Stage Feed Pump P25 will convey the clarified
process water into 2nd Stage Octolig® Unit T25. The valve (V20) after the P25 MUST be set to feed faster
than the overflow. T24 must empty and be shut off by the level control periodically to ensure it does
not overflow! The process water flow will enter and be distributed at the bottom of the 2nd Stage
Octolig* Unit (T25) and uniformly flow upward through the Octolig* bed. The 2nd Stage Octolig* Unit
(T25) effluent will then overflow into Effluent Clearwell Tank Til. Overflow from Effluent Clearwell Tank
Til will flow via gravity to a water collection sump within the Argo Treatment Plant. Note that a portion
of this treated effluent will be used in the Regen/Rinse Feed Tank T12 to make up the acid regeneration
solution. Tank T12 should be filled after start-up and lineout but well before regeneration (fill when you
confirm Zn treatment at the outfall) and then left isolated.
Operations sampling must take place before routine shutdown for sludge management and
regeneration. Sludge measurement, sampling and discharge is integrated into routine shut down.
Routine Shutdown
The Influent control valve(Vl) from Virginia Canyon is closed. pH adjusted water from T13 should be
drawn down until level control shuts off P14. Valve V10 feeding T18 should be closed. The liquid
remaining in the 1st stage Octolig* tank (T18) will be pumped out through the bottom and valve V9 using
43
-------
pump P16, to T20 where it will be neutralized and flow to the clarifier and continue treatment. When
T18 is empty, P16 should be turned off and V9 closed.
Sludge in T23 may now be measured and managed.
Regeneration Description
When detection of a metal breakthrough, after an Octolig* stage is found, the system will be shut down
for regeneration of the Octolig* beds. The 1st Stage Octolig (T18) is expected to require regeneration
weekly or semi-weekly. The 2nd Stage Octolig (T25) will likely be regenerated once in the demonstration.
1st Stage Octolig (T18) will always be regenerated first. Liquid in tank T12 will be made acidic to pH 2 by
adding sulfuric acid with the controller fed by P15 Regen Acid Pump while the P16 Regen Feed Pump is
recirculating water in T12. Acid from T12 will then be pumped with P16 into T18. P16 will be turned off
and T18 will be drained back through V9 to T12 where the pH can be re-adjusted to pH 2. This process is
repeated until pH in Tank T12 is < pH2.2 prior to adjustment. Acid in tank T12 is pumped with P16 to
tank T28. Tank T28 is then isolated pending sampling and disposal.
2nd Stage Octolig (T25) is drained (this should be possible without pumping) into T12 with special
attention to stopping flow from T25 before the T12 overflows. Water from T12 will be pumped into 1st
Stage Octolig (T18). Water will be drained from T18 into T12. The pH of T12 will be adjusted up to 4.0
while recirculating though P16. Water will be pumped with P16 into T18. And the process repeated until
pH is >3.8 and <4.5 prior to adjustment in T12.
If no 2nd Stage Regeneration:
• Water will then be pumped though P16 from T12 into pH neutralization tank T20. Begin Startup
Procedure.
With 2nd Stage Regeneration:
• Water will then be pumped through P16 from T12 into 2nd Stage Octolig (T25). T25 is drained
(this should be possible without pumping) into T12 with special attention to stopping flow from
T25 before the T12 overflows. The pH of T12 will be adjusted to 4.0 while recirculating though
P16. Water will be pumped with P16 into T25. The process will be repeated until pH is >3.8 and
<4.5 prior to adjustment in T12. Water in tank T12 is pumped with P16 to tank T26. Tank T26 is
then isolated pending sampling and disposal.
• Water will then be drained from T25 (this should be possible without pumping) into T12 with
special attention to stopping flow from 725 before the T12 overflows. If the level in T12 is
below the fill pipe from the effluent clearwell tank (Til), Til should be allowed to drain into
T12 with special attention to stopping flow from Til before the T12 overflows. The pH of T12
will be adjusted to 8.0 while recirculating though P16. Water will be pumped with P16 into T25.
Water will then be drained from T25 into T12. The process will be repeated until pH is >7.8 and
<8.5 prior to adjustment in T12. Water will then be pumped from T12 into T25. Begin Startup
Procedure.
44
-------
ARCADIS Regeneration SOP for Field Pilot Plant
Start from normal operation treating AMD. T12 should be full of treated effluent. Operations samples in
this mode of operation
1. Move to Step 1 - Drain T18. Stop AMD influent. Empty T18 into T20 and continue treating
a. Measure and Manage sludge in T23 after Step 1 and before Step 7.
2. Step 2 - pH adjust T12. Bring pH to 1.8-2.2 while recirculating adding H2SO4 as necessary
3. Step 3 - fill T18 with solution from T12.
4. Step 4-purge T18 into T12.
5. Repeat 2., 3., and 4. until pH does not require acid adjustment after Step 4. T12 is full of pH
1.8-2.2 water and T18 is empty.
6. Step 5 - Transfer T12 contents (used regenerant) to T28
7. Step 6 - Purge T25 to fill T12. CAREFUL not to overfill T12. Leave remaining water in T25!
8. Repeat Step 3 - fill T18 with solution from T12. This solution should be ~pH 8. This is to rinse
acid from T18.
9. Repeat Step 4-purge T18 into T12.
10. Repeat Step 2 - pH adjust T12. Bring pH to 3.8-4.2 while recirculating adding NaOH as necessary.
11. Repeat Step 3 - fill T18 with solution from T12.
12. Repeat Step 4 - purge T18 into T12.
13. Repeat 10., 11. And 12. Until pH does not require base adjustment after step 4. T12 is full of pH
3.8-4.2 water and T18 is empty.
14. IF NO 2nd Stage REGENERATION proceed to 15. If 2nd Stage REGENERATION skip to 18.
15. Step 7 - Purge T12 to T20. Continue treatment (Neutralization etc) but there will be insufficient
volume to discharge effluent.
16. Restart system Treating AMD. Open VI allowing pH control to operate as tanks fill.
17. Step 10 - Fill T12 with Effluent from Til. CAREFULL not to overfill T12.
18. Step 8 - Fill T25 with solution from T12. Solution should be pH 3.8-4.2.
19. Repeat Step 6 - Purge T25 to fill T12. CAREFUL not to overfill T12. Leave remaining water in T25!
20. Repeat Step 2 - pH ad adjust T12. Bring pH to 3.8-4.2 while recirculating adding H2SO4 as
necessary.
21. Repeat 18, 19, and 20. Until pH does not require base adjustment after step 6. T12 is full of pH
3.8-4.2 water and T25 contains residual 3.8-4.2 water.
22. Step 9 - T12 contents (used regenerant) to T26.
23. Repeat Step 6. Be careful not to overfill T12. There may not be enough water in T25 to fill T12.
24. Step 10 - Fill T12 with Effluent from Til. CAREFULL not to overfill T12.
25. Step 7 - Purge T12 to T20. Continue treatment (Neutralization etc) but there will be insufficient
volume to discharge effluent.
26. Repeat Step 6.
27. Repeat 25. and 26. Until pH in T12 is >7.5.
28. Repeat Step 7 - purge T12 to T20.
29. Restart system Treating AMD. Open VI allowing pH control to operate as tanks fill.
30. Step 10 - Fill T12 with Effluent from Til. CAREFULL not to overfill T12.
45
-------
6.2 Appendix B: Equipment/Parts List from Arcadis
•no
IN
LQ11
Til
Ft*
fcad p^*-Q*X>mf zmnt FOHt if HP, 1|*
•HI
1JH
1JI
PVDI
1JM
1T1*. MM 41
TH
ttt
l.lflHF
Til
IPIO
•PPH
LC11
axil
IX*
•P-a
46
-------
6.3 Appendix C: Process Flow and Equipment Placement
WAto irnoem
^
SUW»)
' .
r-UC }•-•}
•Tel*
1A5C PU*
1
•
m
m
iuc
cuiuncu TU«
JT
«*• ©
WXUOlt
C
pi
1
>• u
>(M
?i
-
•-
V
,
no
PH WUT TAW
•tctxouTt souniON
Hnuonuot
Tx = tank number
Px = pump number
Vx = valve number
LC = controller
SPX = samplinE port
AOO o* CAUSTIC
A1XAUM ICORCUlAriON
SPtNT RCCCNCIIA1IT
I
•
li
II
Figure Cl. Process flow diagram. Figure prepared by EPS, subcontractor to Arcadis and modified by EPA.
47
-------
Tank T23
Figure C2. Visualization of spatial arrangement of equipment in the trailer, in two orientations. The large green tank is T23, which
was located outside the trailer. Figure prepared by EPS, subcontractor to Arcadis and modified by EPA.
48
-------
6.4 Appendix D: Equipment Installation and Operation Notes from Arcadis
6.4.1 Installation and Fall 2011 Operations
The pilot test equipment within the trailer was assembled at the EPS (subcontractor) facility in
Cincinnati Ohio. It arrived at the Argo site September 24th, 2011 but was unable to be
maneuvered into the narrow space behind the large scale treatment building allotted for this
demonstration until September 27th. Work in October 2011 included installing the sample
collection piping, interconnecting the trailer to the exterior settling tank, and the Argo plant for
discharge, connecting electrical power for the system, loading the Octolig media into the
appropriate tankage. Collection piping was designed and installed for approximately 10 gpm,
based on the sustained fall flow reported for Virginia Canyon.
The system was used for limited initial operations in November of 2011. During this
commissioning period, numerous problems in the system were diagnosed and addressed. For
example, efforts were required to balance the flows of pumps within the system to maintain
proper hydraulic levels in tanks, despite the presence of a PLC control system. An air release
point was installed in the system discharge piping to improve flow performance. However,
freezing weather conditions led to multiple freezeups and some leakage from piping, despite
efforts to provide insulated and heat traced exterior lines. It was determined to be impractical
to heat the trailer with electrical resistance heaters during extended periods of cold weather,
since it had very poor insulating properties. A decision was made in early December to delay
further demonstration operations until Spring 2012. The system was drained to the extent
feasible, detached from the water supply, turned off and winterized to the extent practicable.
Vulnerable parts such as pH meter parts were removed for warm storage in an ARCADIS facility.
Total flow in November 2011 was minimal - approximately 5,700 gallons to the 1st Stage Octolig
Unit (T18) and 1,100 gallons to the 2nd Stage Octolig Unit (T25). Therefore, a relatively small
percentage of the bed capacity was consumed and no regeneration cycles were performed.
6.4.2 Spring 2012 Operations Notes
Field work to restart the system began March 26, 2012. Pipes and fittings were
reinstalled/repaired. The water supply line to the Virginia Canyon discharge began rapidly on
March 27, since water was needed to commission the system. With an influent flow rate of 6
49
-------
gpm and numerous tanks to fill, an extended period was required for startup2. That time was
also extended because initially operations proceeded only during normal working hours to
ensure that any leaks that developed were rapidly addressed. Monitoring with field test kits
began on March 30th and the first laboratory samples were collected on April 5th. Operational
periods and flow rates are compiled in Tables Cl and C2. Flow through the system and volume
treated is presented graphically in Figures Cl through C3.
On April 13, 2012, it was observed that the second stage pH dosing pump hose had risen above
the level of the base in the supply container. The container was refilled and the hose secured at
a lower position and the system ran well until April 16th. On April 18th it was noted that the
increased flow rate had markedly increased base consumption and so the size of the ready use
base reservoir for Tank T20 was increased. Early in the day, the T20 tank pH was 4.12, which
was well below the set-point. But, after the larger tank was installed and refilled, the T20 pH
returned to near the set-point at 8.05. Because the observed pH at SP5 and in a special sample
drawn directly from T20, were lower than their target range, the pH set-point at Tank T20 was
adjusted from pH 8 to pH 9. Data suggested that pH was dropping across the clarifier, perhaps
due to proton release resulting from precipitation of manganese within the clarifier (rather
than in Tank T20) due to slow oxidation kinetics (greater than one hour) at pH less than 9 and in
the presence of high sulfur (Hemm, 1963). Another possible cause of these pH drops considered
was the buildup of lower than target pH water in the Clarifier Settling Tank (T23), perhaps due
to the temporary base feed outages that had been experienced (discussed above). The set-
point was adjusted to pH 8.7 on April 19th, to attempt balance between the need for pH
control and available reagent vendor delivery schedules. On Friday morning, April 20th, the
operator again noted that the intake hose had moved above the level of the base, despite
efforts to secure it, the flexible tube coiled and pulled away from the bottom of the container.
A piece of rigid PVC tube was then installed to maintain the flexible base intake tube in a
straightened, vertical configuration.
On Monday morning, April 23rd, the ARCADIS operator noted that the entire contents of the
tank of base, about 30 gallons of 25% NaOH, had been consumed over the weekend. It was also
noted that sludge generation had increased, and that the sludge level had risen from
approximately 1.0 to 1.5' depth in the clarifier to 4' to 4.5', which caused some sludge to
overflow the clarifier into succeeding tanks. The sludge that had overflowed into the
succeeding tanks and was cleaned out and added back into the clarifier that day. After
2 There were five 330 gallon tanks in the system, a 1,000 gallon 1st Stage Octolig® Unit, 2,300 gallon Clarifier Settling
Tank and 560 gallon 2nd Stage Octolig® Unit for a total system hydraulic capacity (not counting lines in the main
system or tanks used for regeneration only) of approximately 5,500 gallons.
50
-------
collecting field parameters, the system was taken off line later in the afternoon of April 23rd to
allow the first stage media to be regenerated. As a diagnostic step, the pH read by the dosing
pump controller was compared to the field pH meter and found to agree within 0.02 pH units.
On April 23rd, a drum of 50% NaOH solution arrived. This had been ordered after exhaustion of
the initial supply 25% NaOH solution, since the stronger solution was more quickly available,
and warmer temperatures allowed use of the stronger solution, which is more vulnerable to
crystallization at lower temperatures.
In scheduling the regeneration, ARCADIS wanted to balance the objectives of maximum water
treatment efficiency with maximizing the loading onto the Octolig. High Octolig loading was
desirable both for treatment economics and to maximize the potential to recover a valuable
product. Through April 16th, the field zinc data suggested that treatment was efficient. While
data on April 18th suggested zinc breakthrough, data on April 20th and 23rd suggested
resumption of zinc removal. The decision to go into regeneration on the 23rd was thus guided
in part by field data and by the desire to complete regeneration and restore the system to
operation within the available time on site.
Since this was the first time this unit had been regenerated, and the first time these staff
members had performed a regeneration with strong acid, the work was undertaken slowly for
added caution. Regeneration was completed by the evening of April 26th and the system
restarted early in the morning on April 27th. On April 27th, using the stronger 50% NaOH
solution for pH adjustment, ARCADIS observed some substantial "overshooting" above the pH
set-point in Tank T20. Due to the higher strength of the 50% NaOH solution, the feedback
control loop lagged behind the response in pH at the pumping rate previous used for the 25%
NaOH solution. As a result, the pH in TankT20 tank rose as high as pH 11 before gradually
drifting down to the set-point. To address this, ARCADIS reduced the pumping rate of the 2nd
Stage Base Pump (P22). This reduced the base consumption over the weekend (April 28 and 29)
to 15 gallons of 50% NaOH, which was within the capacity of the ready use reservoir. However,
at this lower pump setting, Pump P22 was not able to achieve the pH set-point. On April 30th,
the pH was observed to range between 7.50 and 7.63 at Tank T20 before the base pump was
adjusted to an intermediate pumping rate.
On Tuesday May 1st, it was noted that after 11 straight measurements of pH at SP 2 (after the
1st Stage Octolig® Unit [T18]) in the expected range (greater than pH 4) two sequential
measurements well below pH 4 had been recorded (on Friday the 27th and Monday the 30th).
This led to efforts to diagnose problems with the 1st Stage pH Neutralization Tank (T10), 1st
Stage Base Pump (P12), and appurtenant controls. On May 2nd it was determined that the
Pump P12 was overheating. It was determined that the repair could not be rapidly completed in
the field. There were project schedule constraints (driven in part by the need of the host facility
51
-------
to be freed of the extra burden of balancing effluent flows from pilot test before their busiest
season - spring snow melt). Therefore, in consultation with EPA, the decision was made to
proceed with sampling by EPA on Thursday May 3rd even with the pH at Tank T10 at
approximately 3.5, below the pH 4 target, because of the malfunctioning pump. ARCADIS
collected its final round of samples on Friday May 4th. The system was shut off on Monday May
7th after the completion of the planned operational period. Regeneration of both Octolig® beds
was completed on May 7th and May 8th.
The total volume of acid used for all of the regeneration was 3 gallons of Sulfuric Acid (66
Degrees Baume 93.2%). The total volume of base used in water treatment was 110 gallons of
25% NaOH solution and 35 gallons of 50% NaOH. The total volume of first stage regenerant
produced was 200 gallons (Tank T28) and 80 gallons of second stage regenerant solution (Tank
T25).
Careful coordination was needed with the Argo Treatment Plant staff, who received our
discharge from the pilot scale unit, to ensure that they were prepared for changes in flow rate.
During this initial operational period, problems with the pH dosing pumps and flow meters
were diagnosed and resolved. Since temperatures in early spring can still dip down below
freezing overnight, foam insulation, heat tape, and electrical space heaters were used to
maintain system temperatures.
Attempts were made to begin overnight operation on the evening of April 5th, but this was
discontinued because constrictions in the system discharging water back to the main Argo
treatment plant prevented proper flow balancing. Overnight operations were successfully
begun as of April 6th. The system was shut down on April 11th for maintenance including leak
repair, pH meter recalibration and installation of a larger discharge line to the Argo treatment
plant sump. The system was restarted the same day. The influent flow rate was increased on
April 13th, taking advantage of the larger discharge line.
52
-------
Table Dl. First Stage flow rates and treatment volumes
Date / Time
3/26/12 9:00
3/27/12 9:00
3/28/12 9:00
3/29/12 9:00
3/30/12 9:00
4/2/12 9:00
4/3/12 9:00
4/4/12 9:00
4/5/12 9:00
4/6/12 9:00
4/7/12 9:00
4/9/12 9:00
4/11/12 9:00
4/13/12 9:00
4/16/12 9:00
4/18/12 9:00
4/19/12 11:00
4/23/12 9:00
4/24/12 9:00
4/25/12 9:00
4/26/12 9:00
4/30/12 9:00
5/2/12 9:00
5/4/12 12:00
5/7/12 12:00
5/8/12 0:00
Period
Time
(min)
1,440
1,440
1,440
1,440
4,320
1,440
1,440
1,440
1,440
1,440
2,880
2,880
2,880
4,320
2,880
1,560
5,640
1,440
1,440
1,440
5,760
2,880
3,060
4,320
Period
Down-Time
(min)
-
960
960
960
960
3,300
960
960
960
0
0
0
0
0
0
0
0
0
1,440
1,440
1,440
0
0
0
0
Period
Run-Time
(min)
480
480
480
480
1,020
480
480
480
1,440
1,440
2,880
2,880
2,880
4,320
2,880
1,560
5,640
0
0
0
5,760
2,880
3,060
4,320
Cumulative
Run-Time
(min)
0
480
960
1,440
1,920
2,940
3,420
3,900
4,380
5,820
7,260
10,140
13,020
15,900
20,220
23,100
24,660
30,300
30,300
30,300
30,300
36,060
38,940
42,000
46,320
Average
Influent
Flow Rate
(GPM)
6
6
6
10
10
10
10
5
5
5
5
10
10
10
10
10
10
0
0
0
10
10
10
10
Total
Influent
Volume
(gal)
2,880
5,760
8,640
13,440
23,640
28,440
33,240
35,640
42,840
50,040
64,440
93,240
122,040
165,240
194,040
209,640
266,040
266,040
266,040
266,040
323,640
352,440
383,040
426,240
Total
Volume
Treated
(gal)
0
2,880
5,760
8,640
13,440
23,640
28,440
30,600
32,760
39,096
45,144
58,104
93,240
122,040
165,240
194,040
209,640
266,040
266,040
266,040
266,040
323,640
352,440
383,040
426,240
Bed
Volumes
19
39
58
90
159
191
205
220
262
303
390
626
819
1,109
1,302
1,407
1,786
1,786
1,786
1,786
2,172
2,365
2,571
2,861
Average Period
Flow Monitor
Flow Rate
(GPM)
0.0
14.5
13.5
13.4
14.0
14.0
14.0
4.5
4.5
4.4
4.2
4.5
14.5
15.2
15.0
15.0
15.0
15.0
0.0
0.0
0.0
14.8
15.2
15.0
15.1
Average
Influent
Flow Rate
(GPM)
-
0.0
0.0
0.0
0.0
0.0
3.0
4.0
5.0
5.0
5.0
5.0
11.0
11.0
11.0
11.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
System Shutdown - regeneration of T18 and T25 tanks; draining of water tanks
T18 regenerated April 24-26
53
-------
Table D2. Second Stage flow rates and treatment volumes
Date / Time
3/26/12 9:00
3/27/12 9:00
3/28/12 9:00
3/29/12 9:00
3/30/12 9:00
4/2/12 9:00
4/3/12 9:00
4/4/12 9:00
4/5/12 9:00
4/6/12 9:00
4/7/12 9:00
4/9/12 9:00
4/11/12 9:00
4/13/12 9:00
4/16/12 9:00
4/18/12 9:00
4/19/12 11:00
4/23/12 9:00
4/24/12 9:00
4/25/12 9:00
4/26/12 9:00
4/30/12 9:00
5/2/12 9:00
5/4/12 12:00
5/7/12 12:00
5/8/12 0:00
Period
Time
(min)
1,440
1,440
1,440
1,440
4,320
1,440
1,440
1,440
1,440
1,440
2,880
2,880
2,880
4,320
2,880
1,560
5,640
1,440
1,440
1,440
5,760
2,880
3,060
4,320
Period
Down-Time
(min)
-
960
960
960
960
3,300
960
960
960
0
0
0
0
0
0
0
0
0
1,440
1,440
1,440
0
0
0
0
Period
Run-Time
(min)
0
0
0
0
0
480
480
480
1,440
1,440
2,880
2,880
2,880
4,320
2,880
1,560
5,640
0
0
0
5,760
2,880
3,060
4,320
Cumulative
Run-Time
(min)
0
0
0
0
0
0
480
960
1,440
2,880
4,320
7,200
10,080
12,960
17,280
20,160
21,720
27,360
27,360
27,360
27,360
33,120
36,000
39,060
43,380
Average
Influent
Flow Rate
(GPM)
6
6
6
10
10
10
10
5
5
5
5
10
10
10
10
10
10
0
0
0
10
10
10
10
Total
Influent
Volume
(gal)
0
0
0
0
0
4,800
9,600
12,000
19,200
26,400
40,800
69,600
98,400
141,600
170,400
186,000
242,400
242,400
242,400
242,400
300,000
328,800
359,400
402,600
Total
Volume
Treated
(gal)
0
0
0
0
0
0
4,800
6,912
8,976
15,024
21,072
34,032
69,600
98,400
141,600
170,400
186,000
242,400
242,400
242,400
242,400
300,000
328,800
359,400
402,600
Bed
Volumes
0
0
0
0
0
32
46
60
101
141
228
467
660
950
1,144
1,248
1,627
1,627
1,627
1,627
2,013
2,207
2,412
2,702
Average Period
FM2
Flow Rate
(GPM)
0.0
14.5
13.5
13.4
14.0
14.0
14.0
4.4
4.3
4.2
4.2
4.5
13.8
14.0
14.8
15.3
14.8
14.7
0.0
0.0
0.0
14.8
14.7
14.8
14.9
Average
Influent
Flow Rate
(GPM)
-
0.0
0.0
0.0
0.0
0.0
3.0
4.0
5.0
5.0
5.0
5.0
11.0
11.0
11.0
11.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
System Shutdown - regeneration of T18 and T25 tanks; draining of water tanks
T18 regenerated April 24-26
20
15
1 10
13
5
0
5-/
••*•• /
• • ^« • ^
Flow Rate
^
•*A*-»A. ^-•••••^« • tf'*-±.
1' '{ //
^ / 1
\ /
A J_^t 1
Vpr 10-Apr 15-Apr 20-Apr 25-Apr 30-Apr 5-May
Date
Wg Influent Flowrate i.^.. 1st Stage Instantaneous Flowrate
.nd Stage Instantaneous Flowrate
Figure Dl. Flow rate versus time. Flow is zero gpm for regeneration April 24-26.
54
-------
4,000
„ 3,000
cu
E
"o 2,000
•a
00 1,000
0
5-/
Bed Volumes
.^••*::
::::**4*=:-
...-X*
•*!••
M • ' A *
*::i::1
AAA"«
\pr 10-Apr 15-Apr 20-Apr 25-Apr 30-Apr 5-May
Date
••A' 1st Stage "A" 2nd Stage
Figure D2. Bed volume versus time. Bed volume was constant during regeneration April 24-26.
Thousand
500
400
Ł 300
_o
$ 200
100
0
5-/
Cumulative Volume Treated
s
«"*::*"
*
.::l*a*:::"'
^ • •
4..'*
.*"
^::4--
-------
6.5 Appendix E: Field Parameters and Analytical Data for Water and Sludge Samples
Sampling Point SP-1: Virginia Canyon Influent Water
Analyte / Parameter
Ag - Total (ug/l)
Ag - Dissolved (ng/l)
Al- Totaling/I)
Al - Dissolved (Mg/l)
As -Total dig/I)
As - Dissolved (ng/l)
Ca - Dissolved (ug/l)
Cd- Total dig/I)
Cd - Dissolved (ug/l)
Cu- Totaling/I)
Cu - Dissolved (ng/l)
Fe - Total (ug/l)
Fe - Dissolved (ug/l)
K - Dissolved (ng/l)
Mg - Dissolved (Mg/l)
Mn - Total (ug/l)
Mn - Dissolved (ug/l)
Na - Dissolved (ug/l)
Ni - Total (ug/l)
Ni - Dissolved (ng/l)
Pb- Totaling/I)
Pb - Dissolved (ug/l)
Se- Totaling/I)
Se - Dissolved (ug/l)
Zn - Total (ug/l)
Zn - Dissolved (ug/l)
Alkalinity (mg/l as CaCO3 )
Hardness (mg/l as CaCO3)
Chloride (mg/l)
Fluoride (mg/l)
Sulfate (mg/l)
TSS (mg/l)
Acidity (test america)
Resistivity (test america)
Specific Conductance
(us/cm)
Temp(°c)
PH
Sampling Date
4/5/2012
B
0.7 (E)
71,100
70,200
<4(RL)
<4(RL)
250,000
307
304
8,540
8,220
2,340
2,260
5,090
110,000
77,100
76,300
36,600
617
611
18.0
17.0
15.0
14.4
73,300
74,300
<5(RL)
1,077
37.9
2.7
1,790
2,663
7.08
3.80
4/7/2012*
<1.1(MDL)
650
3.7
2,700
3.53
4/9/2012
B
0.5 (E)
66,800
71,200
<4(RL)
<4(RL)
249,000
307
311
8,620
8,160
2,410
2,470
5,330
109,000
71,100
75300
39,500
642
614
17.8
17.2
15.4
14.2
68,000
70,900
<5(RL)
1,071
39.9
3.5
1,800
650
3.6
2,729
14.12
2.95
4/11/2012*
<1.1(MDL)
660
3.6
2,800
3.36
4/16/2012
B
1
72,700
71,100
<4(RL)
<4(RL)
256,000
301
306
8,710
8,820
2,510
2,570
5,090
112,000
78,000
78,600
36,600
635
632
18.6
18.6
18.5
18.9
74,800
77,000
<5(RL)
1,100
37.6
4.4
1,780
590
3.8
2,736
7.47
3.23
4/1S/2012*
not measured
520
3.8
2,700
3.56
4/20/2012
B
0.9
70,200
69,500
<4(RL)
<4(RL)
252,000
301
303
8,840
8,970
2,600
2,690
5,280
112,000
75,200
76,700
36,300
644
645
19.4
19.5
18.0
17.9
72,500
74,800
<5(RL)
1,090
36.4
3.1 '
1,790
2,787
3.63
3.19
4/27/2012*
not measured
600
3.5
2,800
3.51
4/30/2012
0.9
0.8
76,400
72,000
<4(RL)
<4(RL)
251,000
333
326
9,290
8,760
3,160
2,920
5,020
109,000
82,600
76,900
35,300
631
616
21.0
19.3
17.0
14.9
80,300
73,800
<5(RL)
1,076
37.2
4.6
1,760
610
3.6
not measured
not measured
not measured
5/3/2012
0.8
0.7
69,100
68,100
<4(RL)
<4(RL)
250,000
315
312
8,820
8,310
3,100
1,680
5,390
104,000
78,700
75,100
77,600
632
598
19.8
14.7
17.0
14.4
78,800
71,700
<5(RL)
1,053
34.1
<0.2(RL)
1,770
2,838
12.39
3.47
5/4/2012*
<1.1(MDL)
630
2,900
3.59
* analyzed by Test America on unfiltered sample; pH and conductivity measured in the field by Arcadis
RL = reporting limit
MDL = method detection limit
E = QA criterion not met, value is an estimate
B = concentration in blank higher than sample value, value not reported
56
-------
Sampling Point SP-lb: Post 1st Base Addition, Pre 1st Octolig® Bed
Analyte / Parameter
Ag - Total (Mg/l)
Ag - Dissolved (Mg/l)
Al - Total (Mg/l)
Al - Dissolved (Mg/l)
As - Total (Mg/l)
As - Dissolved (Mg/l)
Ca - Dissolved (Mg/l)
Cd - Total (Mg/l)
Cd - Dissolved (Mg/l)
Cu - Total (Mg/l)
Cu - Dissolved (Mg/l)
Fe - Total (Mg/l)
Fe - Dissolved (Mg/l)
K - Dissolved (Mg/l)
Mg - Dissolved (ng/l)
Mn - Total (Mg/l)
Mn - Dissolved (Mg/l)
Na - Dissolved (Mg/l)
Ni- Total (Mg/l)
Ni - Dissolved (Mg/l)
Pb - Total (Mg/l)
Pb - Dissolved (Mg/l)
Se - Total (Mg/l)
Se - Dissolved (Mg/l)
Zn - Total (Mg/l)
Zn - Dissolved (Mg/l)
Alkalinity (mg/l as CaCO3 )
Hardness (mg/l as CaCO3)
Chloride (mg/l)
Fluoride (mg/l)
Sulfate (mg/l)
Specific Conductance
(uS/cm)
Temp (°c)
PH
Sampling Date
4/5/2012
B
<0.5(RL)
70,900
70,400
<4(RL)
<4(RL)
245,000
299
308
8,580
8,350
2,330
2,290
5,080
106,000
76,200
75,000
36,900
614
611
20.1
19.1
14.9
15.0
71,400
71,300
<5(RL)
1,048
39.2
4.4
1,820
2,661
6.58
3.05
4/16/2012
1(E)
0.6
72,500
69,700
<4(RL)
<4(RL)
250,000
307
304
8,800
8,860
2,570
2,560
5,080
110,000
78,300
77,000
35,800
635
605
19.2
19.0
18.4
17.8
75,500
75,500
<5(RL)
1,077
38.5
4.7
1,830
2,735
7.11
3.14
5/3/2012
0.8
0.5
73,200
72,600
<4(RL)
<4(RL)
252,000
335
318
9,330
8,740
3,300
3,020
5,390
107,000
81,800
78,200
36,100
656
616
21.6
20.0
17.1
16.1
84,100
74,300
<5(RL)
1,070
34.2
<0.2(RL)
1,820
2,786
11.79
3.64
RL= reporting limit
E = QA criterion not met, value is an estimate
B = concentration in blank higher than sample value
57
-------
Sampling Point SP-2: Post 1st Octolig® Bed
Analyte / Parameter
Ag - Total (ug/l)
Ag - Dissolved (|Jg/l)
Al- Total (ug/l)
Al - Dissolved (ug/l)
As -Total (ug/l)
As - Dissolved (|Jg/l)
Ca - Dissolved (ug/l)
Cd- Total (ug/l)
Cd - Dissolved (ug/l)
Cu- Total (ug/l)
Cu - Dissolved (ug/l)
Fe- Total (ug/l)
Fe - Dissolved (ug/l)
K - Dissolved (ug/l)
Mg - Dissolved (ug/l)
Mn- Total (ug/l)
Mn - Dissolved (ug/l)
Na - Dissolved (ug/l)
Ni- Total (ug/l)
Ni - Dissolved (ug/l)
Pb- Total (ug/l)
Pb - Dissolved (ug/l)
Se- Total (ug/l)
Se - Dissolved (ug/l)
Zn- Total (ug/l)
Zn - Dissolved (|Jg/l)
Alkalinity (mg/l as CaC03)
Hardness (mg/l as CaCO3)
Chloride (mg/l)
Fluoride (mg/l)
Sulfate (mg/l)
TSS (mg/l)
Specific Conductance
(uS/cm)
Temp(°c)
PH
Sampling Date
4/5/2012
<0.5(RL)
<0.5(RL)
30,800
29,900
<4(RL)
<4(RL)
250,000
314
311
162
152
170
187
5,350
109,000
84,300
86,200
37,700
431
432
14.5
13.7
16.6
15.5
72,500
77,200
<5(RL)
1,073
40.6
4.3
1,520
2,307
5.85
4.95
4/7/2012*
8
2,500
5.02
4/9/2012
<0.5(RL)
<0.5(RL)
50,200
50,600
<4(RL)
<4(RL)
249,000
465
464
280
268
239
215
5,310
109,000
73,000
76,600
37,800
762
770
21.6
19.9
18.7
17.2
99,200
108,000
<5(RL)
1,071
42.3
5.2
1,740
2,457
11.51
4.65
4/11/2012*
2.4 (J)
2,600
4.17
4/16/2012
<0.5(RL)
<0.5(RL)
80,900
78,400
<4(RL)
<4(RL)
244,000
308
293
3,390
3,370
1,110
1,030
5,180
106,000 (E)
79,200
75,900
36,400
662
620 (E)
20.7
19.5
18.3
17.8
75,800
70,800 (E)
<5(RL)
1,046
38
4.5
1,820
2,566
6.54
3.99
4/18/2012*
not measured
2,500
4.22
4/20/2012
0.5 (E)
<0.5(RL)
77,000
73,500
<4(RL)
<4(RL)
252,000
304
307
5,040
5,190
1,410
1,420
5,240
111,000
76,300
76,400
36,200
646
625
20.2
20.1
19.2
18.5
71,700
75,000
<5(RL)
1,086
38.1
4.6
1,810
2,641
8.19
3.71
4/30/2012
<0.5(RL)
<0.5(RL)
78,500
72,700
<4(RL)
<4(RL)
250,000
339
316
4,750
4,390
2,200
1,920
5,180
106,000
83,800
76,600
35,900
640
616
21.7
19.3
16.9
14.7
81,000
72,900
<5(RL)
1,061
37.8
4.6
1,810
not measured
not measured
not measured
5/3/2012
<0.5(RL)
<0.5(RL)
75,100
73,500
<4(RL)
<4(RL)
251,000
329
326
6,570
6,050
2,210
2,070
5,410
107,000
80,900
77,800
36,400
646
618
20.9
20.2
16.3
15.9
79,900
74,000
<5(RL)
1,067
34.3
<0.2(RL)
1,790
2,850
12.15
3.24
5/4/2012*
2
2,900
3.54
* analyzed by Test America on unfiltered sample;
RL = reporting limit
E = QA criterion not met, value is an estimate
J = above MDL, but below RL
pH and conductivity measured in the field by Arcadis
58
-------
Sampling Point SP-5: Post 2nd Base Addition, Pre 2nd Octoli;
Analyte / Parameter
Ag - Total (ug/l)
Ag - Dissolved (ug/l)
Al - Total (ug/l)
Al - Dissolved (ug/l)
As - Total (ug/l)
As - Dissolved (ug/l)
Ca - Dissolved (ug/l)
Cd - Total (ug/l)
Cd - Dissolved (ug/l)
Cu - Total (ug/l)
Cu - Dissolved (ug/l)
Fe - Total (ug/l)
Fe - Dissolved (ug/l)
K - Dissolved (ug/l)
Mg - Dissolved (ug/l)
Mn - Total (ug/l)
Mn - Dissolved (ug/l)
Na - Dissolved (ug/l)
Ni - Total (ug/l)
Ni - Dissolved (ug/l)
Pb - Total (ug/l)
Pb - Dissolved (ug/l)
Se - Total (ug/l)
Se - Dissolved (ug/l)
Zn - Total (ug/l)
Zn - Dissolved (ug/l)
Alkalinity (mg/l as CaCO3 )
Hardness (mg/l as CaCOS)
Chloride (mg/l)
Fluoride (mg/l)
Sulfate (mg/l)
Specific Conductance
(uS/cm)
Temp(°C)
PH
Sampling Date
4/5/2012
<0.5(RL)
<0.5(RL)
440
<100(RL)
<4(RL)
<4(RL)
247,000
98.6
96.4
546
533
<100(RL)
<100(RL)
7,070
80,200
33,800
35,500
262,000
195
192
<1(RL)
<1(RL)
1.9
2.3
6,210
5,980
33.7
947
43.9
2.2
1,420
2,599
7.12
6.20
4/16/2012
0.6 (E)
<0.5(RL)
52,000
50,800
<4(RL)
<4(RL)
252,000
264
272
2,220
2,280
595
467
5,140
102,000
65,000
67,600
136,000
553
557
12.7
11.8
14.1
14.0
59,500
63,800
<5(RL)
1,049
38
3.8
1,790
2,687
6.3
4.20
5/3/2012
<0.5(RL)
<0.5(RL)
49,100
46,200
<4(RL)
<4(RL)
247,000
290
281
4,710
4,270
1,010
151
5,610
94,000
67,900
65,600
180,000
530
500
10.1
4.3
13.2
12.1
67,700
62,800
<5(RL)
1,004
34.4
<0.2(RL)
1,770
2,902
13.05
4.26
RL= reporting limit
E = QA criterion not met, value is an estimate
59
-------
Sampling Point SP-3: Post 2nd Octolig® Bed
Analyte / Parameter
Ag- Totaling/I)
Ag - Dissolved (Mg/l)
Al - Total (Mg/l)
Al - Dissolved (Mg/l)
As -Total (Mg/l)
As- Dissolved (Mg/l)
Ca - Dissolved (Mg/l)
Cd -Totaling/I)
Cd - Dissolved (Mg/l)
Cu -Totaling/I)
Cu - Dissolved (Mg/l)
Fe - Total (Mg/l)
Fe - Dissolved (Mg/l)
K - Dissolved (Mg/l)
Mg - Dissolved (Mg/l)
Mn -Totaling/I)
Mn - Dissolved (Mg/l)
Na - Dissolved (Mg/l)
Ni -Totaling/I)
Ni- Dissolved (Mg/l)
Pb- Total (Mg/l)
Pb - Dissolved (Mg/l)
Se - Total (ng/l)
Se - Dissolved (Mg/l)
Zn- Total (Mg/l)
Zn - Dissolved (Mg/l)
Alkalinity (mg/l as CaCO3 )
Hardness (mg/l as CaCOS)
Chloride (mg/l)
Fluoride (mg/l)
Sulfate (mg/l)
TSS (mg/l)
Specific Conductance
(uS/cm)
Temp (°q
pH
Sampling Date
4/5/2012
B
<0.5(RL)
255
<100(RL)
<4(RL)
<4(RL)
237,000
55.6
54.5
552
524
<100(RL)
<100(RL)
8,680
75,300
26,700
27,600
260,000
163
155
<1(RL)
<1(RL)
1.9
2.3
3,090
2,880
47.9
902
55.4
2.3
1,380
2,564
7.7
6.08
4/7/2012*
18
2,700
7.46
4/9/2012
<0.5(RL)
<0.5(RL)
1,460
141
<4(RL)
<4(RL)
245,000
166
163
109
109
<100(RL)
<100(RL)
6,230
73,900
34,800
33,200
300,000
197
187
<1(RL)
<1(RL)
3.1
3.4
16,300
14,300
22.5
916
41.8
<2(RL)
1,550
2,774
12.03
5.8
4/11/2012*
8.8
2,900
5.85
4/16/2012
<0.5(RL)
0.7
70,400
69,800
<4(RL)
<4(RL)
249,000
319
321
2430
2470
744
687
5,220
109,000
75,400
76,700
40,400
634
633
18.9
18.5
18.2
19.0
72,800
74,500
<5(RL)
1,071
37.9
4.4
1,790
2,515
7.1
3.58
4/18/2012*
not measured
2,500
4.36
4/20/2012
<0.5(RL)
<0.5(RL)
15,600
9,160
<4(RL)
<4(RL)
231,000
244
247
1270
1150
116
<100(RL)
5,410
81,100
55,200
53,100
283,000
402
375
1.8
<1(RL)
10.8
10.6
49,200
46,500
<5(RL)
911
38.5
3.1
1,710
2,962
8.66
4.71
4/27/2012*
not measured
3,300
6.27
4/30/2012
<0.5(RL)
<0.5(RL)
39,500
34,300
<4(RL)
<4(RL)
250,000
328
309
2690
2430
379
<100(RL)
5,390
100,000
74,300
68,600
179,000
554
527
3.8
1.3
14.5
12.7
73,000
68,000
<5(RL)
1,036
38.2
4.2
1,740
not collected
not collected
not collected
5/3/2012
<0.5(RL)
<0.5(RL)
25,800
21,600
<4(RL)
<4(RL)
236,000
243
233
2710
2470
286
<100(RL)
5,370
78,700
58,400
52,900
292,000
431
396
3.3
1.1
10.4
9.7
58,800
49,300
<5(RL)
913
34.3
2.3
2,490
3,030
13.28
4.08
5/4/2012*
17
3,000
4.76
* analyzed by Test America on unfiltered sample; pH and conductivity measured in the field by Arcadis
RL = reporting limit
B = concentration in blank higher than sample value, value not reported
60
-------
Sludge Chemistry
Analyte / Parameter
Ag - Total (Mg/l)
Al - Total (Mg/l)
As - Total (Mg/l)
Ca - Total (Mg/l)
Cd - Total (Mg/l)
Cu - Total (Mg/l)
Fe - Total (Mg/l)
K- Total (Mg/l)
Mg - Total (Mg/l)
Mn - Total (Mg/l)
Na - Total (Mg/l)
Ni- Total (Mg/l)
Pb - Total (Mg/l)
Zn - Total (Mg/l)
TSS (mg/l)
Sampling Date
4/26/2012
97100
5/4/2012
8,800
5/8/2012
55 (J)
490,000
<44(MDL)
480,000
1400
35,000
19,000
13,000 (J)
460,000
380,000
800,000
3000
180.0
330,000
8,950
MDL = method detection limit
J = above MDL, but below reporting limit (RL); RL = 100
for Ag and 30,000 for K
Average TSS reported for 5/8/12
61
-------
6.6 Appendix F: Estimates of Costs for 50 GPM Octolig® and Conventional Lime
Treatment Systems
The following assumptions apply to the cost estimates for the 50 GPM Octolig® treatment
system (Tables Fl and F2):
• Costs are based on removal of Al, Cd, Cu, Fe, Pb, Mg, Mn, Ni, Se, and Zn at influent
concentrations observed during the pilot study.
• Costs associated with regulatory negotiations are excluded.
• Capital costs are based on a treatment system with 50 GPM capacity employing two
stages of Octolig® media each with clarification.
• Costs are for the treatment facility and conveyance piping only - collection systems at
the source are excluded.
• Plant water is assumed to be available for washdown, polymer makedown, etc.
• Octolig® media replacement is expected to occur every 5 years, O&M schedule includes
amortized cost.
The following design calculations were used to develop the costs estimated for the 50 GPM
Octolig® treatment system (Tables Fl and F2):
Design Parameter Design Value
Raw Influent Flowrate (gpm): 50.0
Raw Influent Flowrate (MGD): 0.072
Raw Wastewater flowrate (gpd): 72,000
Solids Thickener Decant (gpd): 1,541
Filter Press Filtrate Return (gpd): 1,996
Design Influent Flowrate (gpm): 52
Current Nominal Flowrate (gpd): 75,537
(MGD): 0.0755
Inlet pH: 3.4
Inlet Acidity (mg/L as CaCO3): 613.8
CaCOS Equivalanece (mg/L as CaCO3): 1,122
Inlet Aluminum Concentration (mg/L): 71
Inlet Cadmium Concentration (mg/L): 0.311
Inlet Copper Concentration (mg/L): 9
Inlet Iron Concentration (mg/L): 2.7
Inlet Lead Concentration (mg/L): 0.0
Inlet Magnesium Concentration (mg/L): 109
Inlet Manganese Concentration (mg/L): 77.1
Inlet Nickel Concentration (mg/L): 0.634
Inlet Selenium Concentration (mg/L): 0.017
Inlet Zinc Concentration (mg/L): 74.617
Inlet Calcium Concentration (mg/L): 251
Inlet Potassium Concentration (mg/L): 5
Inlet Sodium Concentration (mg/L): 44
Inlet Chloride Concentration (mg/L): 37
Inlet Flouride Concentration (mg/L): 3.7
Inlet Sulfate Concentration (mg/L): 1,782
Notes
decant water = solids loading, wet minus thickened solids, wet
filter press filtrate = solids loading, wet minus pressed sludge, wet
raw influent, thickener decant, and filtrate return
raw influent, thickener decant, and filtrate return
based on observed influent pH in pilot
Buffering capacity of raw influent to a pH of approximately 8.3
Reagent demand based on equivalent weight of metals and sulfate
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
62
-------
Influent Equalization
Detention Time (days):
EQ Tank Volume (gal):
EQ Tank Diameter (ft):
EQ Tank Sidewater Depth (ft):
EQ Tank Sidewall Height (ft):
1st Stage pH Neutralization Vessel
Reactor Detention Time (hr):
Target Reactor Tank Volume (gal):
Reactor Tank Sidewater Depth (ft):
Reactor Tank Length (ft):
Reactor Tank Width (ft):
Nominal Reactor Tank Volume (ft3):
Nominal Reactor Tank Volume (gal):
1st Stage Sodium Hydroxide Dose
Target pH:
Target Alkalinity Addition (mg/L as CaCO3):
Target Sodium Hydroxide Dosage (mg/L):
Sodium Hydroxide Strength (wt% NaOH):
Sodium Hydroxide Density (Ibs/gal):
Target Dosage (gals/day):
1st Stage Octoliq
Estimated TSS (mg/L):
No. of Vessels:
Vessel Diameter (in):
Vessel Diameter (ft):
Vessel Height (in):
Vessel Height (ft):
Vessel Volume (gal):
Octolig Media Bed (ft):
Octolig Media Volume (ft3):
Octolig Media Volume (gal):
Octolig Media SG:
Media Bed (Ibs):
Empty Bed Contact Time (EBCT) (min):
Hydraulic Loading (gpm/ft2):
Vessel Velocity (ft/min):
Apsorptive Capacity (mol/kg Octolig):
Apsorptive Capacity (Ibmol/lb Octolig):
System Adsorptive Capacity (Ibmol):
Influent Copper Loading (Ibs/day):
Influent Copper Loading (Ibmols/day):
Influent Iron Loading (Ibs/day):
Influent Iron Loading (Ibmols/day):
Total Molar Mass of Metals (Ibmols/day):
Operating Days Until Regeneration (days):
Bed Volumes Until Regeneration (days):
Copper Mass in Media at Regen (Ibs):
Iron Mass in Media at Regen (Ibs):
Percent Recovery at Reneration (wt%):
1.0
80,000
20.0
34.0
36.0
0.2
525
5.0
5.0
5.0
125.0
935
4.0
100
80.0
50%
12.8
7.53
67
3
72
6
96
8
1,692
Each Vessel
4
113
846
0.6
4,233
16.1
0.62
0.083
0.2
0.0004
5.5946
4.4193
0.0696
1.4423
0.0258
0.0954
59
1,756
261
85
75%
assumed time for equalization of influent sources (raw influent, decant
water, filtrate water)
based on flow rate and detention time
size of tank necessary for volume
actual depth of water expected
height of tank necessary for volume
Based on nominal design flowrate
flow x detention time
calculated from tank dimensions
based on pH to remove copper and iron onto the Octolig without
Estimated
Estimated
Based on raw influent flowrate
Based on raw influent flowrate assuming 10% of solids precipitate out
in this step
Series Operation
calculated from vessel dimensions
Total
339
2,538
12,700
media volume * nominal flow rate
Value for entire amount of Octolig within the 3 vessels
based on pilot removal efficiency, assumes all Cu in 1st stage
based on pilot removal efficiency, assumes all Fe in 1st stage
based on capacity and molar mass of metals
flow x operating days T total volume of Octolig
Assumed recovery from the media into regenerant solution
63
-------
1st Stage Regeneration/Rinse Vessel
Target Reactor Tank Volume (gal):
Reactor Tank Sidewater Depth (ft):
Reactor Diamter (ft):
Reactor Tank Sidewall Depth (ft):
Nominal Reactor Tank Volume (ft3):
Nominal Reactor Tank Volume (gal):
Regeneration Cycle
Sulfuric Acid Strength (wt% H2SO4):
Sulfuric Acid Density (Ibs/gal):
Volume of Renerant per Cycle (gal):
Acid Strength (vol%):
Acid added per batch (gal):
Annual Acid Consumption (gal/yr):
Annual Regenerant Volume (gal):
Copper (mg/L):
Iron (mg/L):
Rinse Cycle
Sodium Hydroxide Strength (wt% NaOH):
Sodium Hydroxide Density (Ibs/gal):
Volume of Rinsate per Cycle (gal):
Base Strength (vol%):
Base added per batch (gal):
Annual Base Consumption (gal/yr):
2nd Stage pH Neutralization Vessel
Reactor Detention Time (hr):
Target Reactor Tank Volume (gal):
Reactor Tank Sidewater Depth (ft):
Reactor Tank Length (ft):
Reactor Tank Width (ft):
Nominal Reactor Tank Volume (ft3):
Nominal Reactor Tank Volume (gal):
2nd Stage Sodium Hydroxide Dose
Target pH:
Target Alkalinity Addition (mg/L as CaCO3):
Target Sodium Hydroxide Dosage (mg/L):
Sodium Hydroxide Strength (wt% NaOH):
Sodium Hydroxide Density (Ibs/gal):
Target Dosage (gals/day):
Estimated TSS (mg/L):
Flocculating Aid - Polymer Dosage
Target Concentration (mg/L):
Target Dosage (Ibs/day):
As Delivered Polymer Density (Ib/gal):
As Delivered Polymer Feed (gpd)
Neat Polymer (Ibs/day):
Clarifier
Influent TSS (mg/L):
Solids Load, dry basis (Ibs/day):
Aluminum, dry basis (Ibs/day):
Cadmium, dry basis (Ibs/day):
Lead, dry basis (Ibs/day):
Magnesium, dry basis (Ibs/day):
Manganese, dry basis (Ibs/day):
Nickel, dry basis (Ibs/day):
Selenium, dry basis (Ibs/day):
Zinc, dry basis (Ibs/day):
2,500
6.6
8.0
9.0
452.4
3,384
93%
15.3
5,076
3%
152
942
31,401
4,619
1,508
0.5
12.8
7,614
3%
228
1,413
0.2
525
5.0
5.0
5.0
125.0
935
8.0
395
316.0
50%
12.8
29.7
285
3
1.89
8.340
0.227
0.009
285
171
32.9
0.0638
0.01020
16.28
16.58
0.171
0.00526
21.4
3 Vessels x 1,692 gal Vessel
based on number of days per year of regeneration
based on number of days per year of regeneration
75% recovery from Octolig assumed
75% recovery from Octolig assumed
Assumed 1.5 x volume of Regenerant
based on number of days per year of regeneration
Based on nominal design flowrate
Based on nominal design flowrate
Estimated based on aluminum concentration
80% of target alkalinity
Based on raw influent flowrate
Based on raw influent flowrate, concentrations and pilot observed
As Delivered, made down solution, 0.5% strength
Based on nominal design flowrate
Based on raw influent flowrate, concentrations and pilot observed
Based on raw influent flowrate, concentrations and pilot observed
Based on pilot removal efficiency
Based on pilot removal efficiency
Based on pilot removal efficiency
Based on pilot removal efficiency
Based on pilot removal efficiency
Based on pilot removal efficiency
Based on pilot removal efficiency
Based on pilot removal efficiency
64
-------
Hydraulic Loading (gpm/ft2): 0.400
Surface Area Required (ft2): 131
Clarifier underflow solids content (%): 1.000%
Underflow density (Ibs/gal): 8.34
Underflow wastage, wet basis (Ibs/day): 17,137
Underflow wastage (gal/day): 2,055
Inclined plate clarifier
Assumed
Clarifier Effluent (gpd):
Clarifier Effluent (gpm):
73,482
51
2nd Stage Octolig
No. of Vessels:
Vessel Diameter (in):
Vessel Diameter (ft):
Vessel Height (in):
Vessel Height (ft):
Vessel Volume (gal):
Octolig Media Bed (ft):
Octolig Media Volume (ft3):
Octolig Media Volume (gal):
Octolig Media SG:
Media Bed (Ibs):
EBCT(min):
Hydraulic Loading (gpm/ft2):
Vessel Velocity (ft/m in):
Apsorptive Capacity (mol/kg Octolig):
Apsorptive Capacity (Ibmol/lb Octolig):
System Adsorptive Capacity (Ibmol):
Influent Cadmium Loading (Ibs/day):
Influent Cadmium Loading (Ibmols/day):
Influent Lead Loading (Ibs/day):
Influent Lead Loading (Ibmols/day):
Influent Manganese Loading (Ibs/day):
Influent Manganese Loading (Ibmols/day):
Influent Nickel Loading (Ibs/day):
Influent Nickel Loading (Ibmols/day):
Influent Selenium Loading (Ibs/day):
Influent Selenium Loading (Ibmols/day):
Influent Zinc Loading (Ibs/day):
Influent Zinc Loading (Ibmols/day):
Total Molar Mass of Metals (Ibmols/day):
Operating Days Until Regeneration (days):
Bed Volumes Until Regeneration (days):
Cadmium Mass in Media at Regen (Ibs):
Lead Mass in Media at Regen (Ibs):
Manganese Mass in Media at Regen (Ibs):
Nickel Mass in Media at Regen (Ibs):
Selenium Mass in Media at Regen (Ibs):
Zinc Mass in Media at Regen (Ibs):
3
72
8
96
8
3,008
Each Vessel Total
6
302 904.8
2,256 6,767.7
0.6
11,289 33,865.8
43.0
0.35
0.047
0.2
0.0004
14.9189
0.18655
0.0016597
0.0115
0.0001
46.3070
0.8429
0.3804
0.0065
0.0101
0.000128
44.8058
0.685419
1 .536637
9.7
108
1.810
0.1113
449.18
3.69
0.10
434.62
Percent Recovery at Reneration (wt%):
2nd Stage Regeneration/Rinse Vessel
Target Reactor Tank Volume (gal):
Reactor Tank Sidewater Depth (ft):
Reactor Diamter (ft):
Reactor Tank Sidewall Depth (ft):
Nominal Reactor Tank Volume (ft3):
Nominal Reactor Tank Volume (gal):
75%
4,000
10.6
8.0
13.0
653.5
4,888
Nominal flow rate minus underflow wastage
Series Operation
media volume * nominal flow rate
Value for entire amount of Octolig within the 3 vessels
nominal flow x operating days to regeneration * total volume Octolig
Assumed
65
-------
Regeneration Cycle
Sulfuric Acid Strength (wt% H2SO4):
Sulfuric Acid Density (Ibs/gal):
Volume of Renerant per Cycle (gal):
Acid Strength (vol%):
Acid added per batch (gal):
Annual Acid Consumption (gal/yr):
Annual Regenerant Volume (gal):
Cadmium (mg/L):
Lead (mg/L):
Manganese (mg/L):
Nickel (mg/L):
Selenium (mg/L):
Zinc (mg/L):
Rinse Cycle
Sodium Hydroxide Strength (wt% NaOH):
Sodium Hydroxide Density (Ibs/gal):
Volume of Rinsate per Cycle (gal):
Base Strength (vol%):
Base added per batch (gal):
Annual Base Consumption (gal/yr):
Solids Thickener Tank
Solids Storage Tank Volume (gal):
Sludge density (Ibs/gal):
Sludge solids content (%):
Solids Loading, dry basis (Ibs/day):
Solids Loading, wet basis (Ibs/day):
Solids Loading (gpd)
No. Days Storage (days):
Thickened Solids, wet basis (Ibs/day):
Decant Water (gpd):
Filter Press
Cake solids, dry basis (%):
Cake density (Ib/ft3):
Minimum Filter Press Size (ft3):
Pressed sludge, wet (Ibs/day):
(ton/day):
(ton/yr):
Aluminum (mg/kg):
Cadmium (mg/kg):
Lead (mg/kg):
Magnesium (mg/kg):
Manganese (mg/kg):
Nickel (mg/kg):
Selenium (mg/kg):
Zinc (mg/kg):
Filtrate (gpd):
Coagulant (Sludge Conditioning)
Target Coagulant Concentration (mg/L):
Target Coagulant Concentration (Ibs/day):
Coagulant Feed (gpd):
Polymer (Sludge Conditioning)
Target Polymer Concentration (Ibs/day):
Polymer Feed (gpd):
Effluent Equalization
Detention Time (days):
EQ Tank Volume (gal):
EQ Tank Diameter (ft):
EQ Tank Sidewater Depth (ft):
EQ Tank Sidewall Height (ft):
93%
15.3
9,024
3%
271
10,187
339,550
18
1.1
4,476
37
1.0
4,331
50.0%
12.8
13,535
3%
406
15,280
12,000
9.0
4.0%
171
17,137
2,055
5.8
4,284.3
1,541
35%
70
7
490
0.24
89
67,268
130.3
20.8
33,253
33,858
349
10.74
43,649
1,996
200
7.1
0.86
1.7
0.21
1.0
80,000
20.0
34.0
36.0
3 Vessels x 1,692 gal Vessel
Volume of Regenerant x 1.5
flow x influent TSS
underflow wastage
Based on solids loading, dry
solids loading, dry * by cake solids, %
solids loading, wet minus pressed sludge, wet
based on target concentration and thickened solids, wet
1% per dry Ib solids
66
-------
Table Fl. Octolig® with conventional coagulation / flocculation treatment: Total capital
investment cost.
Purchased Equipment Costs
QTY Unit Unit Cost
Extension
Description
Inclined Plate Clarifier
Equalization Tanks
1st Stage Octolig Skid
Includes flash mix tank, flocculation tank,
Lump $75,000.00 $75,000.00 catwalk and appurtenances
20-ft diameter, 36-ft high, bolted glass lined
steel tank with roof, ladders, safety
Lump $120,000.00 $240,000.00 platform, and appurtenances
Three (3) 6-ft Diameter Reactor Vessels
with 4-ft of Octolig, Neutralization Vessel,
Regenerant/Rinse Vessel and
Lump $175,000.00 $175,000.00 appurtenances, chemical day tanks
Three (3) 8-ft Diameter Reactor Vessels
with 6-ft of Octolig, Neutralization Vessel,
Regenerant/Rinse Vessel and
2nd Stage Octolig Skid
Bulk Acid Storage Tank
Bulk Caustic Storage Tank
Octolig Media
Chemical Feed System
Sludge Pumps
Filter Press
Filter Press Feed Pump
Filter Press Polymer and DE systems
Polymer Delivery System (Dynablend or
equivalent)
Sludge Storage Tank
Piping, valves, and appurtenances
Outfall
Instrumentation
Power drop
Motor Control Center
Control Panels / SCAD A System
Taxes (5% of EQ)
Freight (2% of EQ)
1 Lump $225,000.00
1 Lump $30,000.00
1 Lump $30,000.00
1,244 ft3 $300.00
1 Lump $12,500.00
2 Lump $5,000.00
1 Lump $65,000.00
1 Lump $4,000.00
1 Lump $12,500.00
1 Each $12,000.00
1 Each $35,000.00
1 Lump $75,000.00
1 Lump $25,000.00
1 Lump $25,000.00
1 Each $15,000.00
1 Lump $50,000.00
1 Each $150,000.00
Equipment Subtotal (EQ.):
1 Lump $81,961.06
1 Lump $32,784.42
Total Purchased Equipment Cost (EEC.):
Direct Installation Costs
Mobilization, Demobilization, Permits, and
Temporary Controls
Treatment Building Foundation and Building
Erection
Conveyance Forcemains to Distribution
Build.
Imported Backfill
Set Process Equipment (8% of EQ)
Process Piping (15% of EQ)
Heat trace and insulation (2% of EQ)
Painting (1.5% of EQ)
Heating and Lighting (8% of EQ)
Site Security (4.5% of EQ)
Electrical (15% of EQ)
QTY Unit Unit Cost
1 Lump $40,000.00
2,500 sf $200.00
5,280 ft $50.00
750 ton $22.00
1 Lump $131,137.70
1 Lump $163,922.12
1 Lump $32,784.42
1 Lump $24,588.32
1 Lump $131,137.70
1 Lump $73,764.95
1 Lump $245,883.18
Total Direct Installation Cost (DJ):
TOTAL DIRECT COST (DC) [PEC + Dl]:
Indirect Costs
Engineering (6% of DC)
Administration/PM (5% of DC)
Geotechnical
Bonds (1.5% DC)
Construction Oversight (6% of DC)
Start Up
Contractor Profit (10% of PEC)
Contingencies (5% of DC)
QTY Unit Unit Cost
1 Lump $202,661.10
1 Lump $168,884.25
1 Lump $15,000.00
1 Lump $50,665.28
1 Lump $202,661.10
1 Lump $25,000.00
1 Lump $175,396.67
1 Lump $168,884.25
Total Indirect Cost (Ki):
TOTAL CAPITAL INVESTMENT (TCI) FDC + IC1:
$225,000.00
$30,000.00
$30,000.00
$373,221.21
$12,500.00
$10,000.00
$65,000.00
$4,000.00
$12,500.00
$12,000.00
$35,000.00
$75,000.00
$25,000.00
$25,000.00
$15,000.00
$50,000.00
$150,000.00
$1,639,221.21
$81,961.06
$32,784.42
$1,753,966.69
Extension
$40,000.00
$500,000.00
$264,000.00
$16,500.00
$131,137.70
$163,922.12
$32,784.42
$24,588.32
$131,137.70
$73,764.95
$245,883.18
$1,623,718.39
$3,377,685.08
Extension
$202,661.10
$168,884.25
$15,000.00
$50,665.28
$202,661.10
$25,000.00
$175,396.67
$168,884.25
$1,009,152.66
$4,386,837.75
appurtenances, chemical day tanks
Flocculant
Progressive cavity pumps
Nominal 1 4,000 gallon tank with catwalk,
sludge rake, and appurtenances
Description
50 ft x 50 ft Pre-Engineered Building
Pipe Bedding as Necessary
Description
67
-------
Table F2. Octolig® with conventional coagulation / flocculation treatment: Total annual
operating costs.
Direct Annual Operating Costs
QTY
Unit
Unit Cost
Extension
Description
Maintenance and Replacement Parts (4%
EQ)
Laboratory Analytical
Sampling/Operator & Maintenance Labor
(OL)
Octolig Media Replacment
Polymer
Coagulant
Sodium Hydroxide (50%)
Sulfuric Acid
1st Stage Regnerant
2nd Stage Regenerant
Sludge Disposal
Sludge Transportation & Demurrage
Electricity (1,780 kW-hr/day assumed)
1
26
2,080
1
690
2,608
30,297
11,129
31,401
339,550
89
8.9
649,700
Lump
Biweekly
Staff Hours
Lump
Ib
Ib
gal
gal
gal
gal
ton
Load
kW-hr
$65,568.85
$500.00
$85.00
$75,000.00
$3.50
$3.50
$3.50
$3.50
$0.00
$0.00
$75
$700
$0.12
$65,568.85
$13,000.00
$176,800.00
$75,000.00
$2,414.40
$9,129.32
$106,038.06
$38,949.88
$0.00
$0.00
$6,701.90
$6,255.10
$77,964.00
pilot
40 hours per week, 52 weeks per year;
loaded cost
based on target dose of f locculant
coagulant
based on target doses in 2 stages and
regeneration rinsing steps
Assumed based on 600 ppm dosage,
90% strength (-0.2 TPD).
dispose, but no value assumed for
recycling
dispose, but no value assumed for
recycling
Assumed 35% solids filter cake
Equivalent to 1 00 brake horsepower
Indirect Annual Operating Costs
Total Direct Annual Operating Cost (QAC_): $577,821.51
QTY Unit Unit Cost Extension
Description
Overhead & Administrative Charges (1.0%
TCI) 1
Property Taxes (1.0% TCI) 1
Insurance (1.0% TCI) 1
Lump
Lump
Lump
— Excluded -
— Excluded -
— Excluded -
Total Indirect Annual Operating Cost (1AC_):
$0.00
TOTAL ANNUAL COST (TAG) [DAC + IflC]: $577,821.51
68
-------
The following assumptions apply to the comparative cost estimates for a 50 GPM conventional
lime treatment system (Tables F3 and F4):
• Treatment train consists of a 20-ft diameter clarifier/thickener; quicklime (CaO) is dosed
to precipitate metals and gypsum (CaSCU) if/when sulfate concentrations exceed 2,000
mg/l.
• Stoichiometric lime dose to yield a pH of 10-11; recarbonation step to be employed to
reduce pH to < 9 to facilitate discharge to meet NPDES permit.
• Costs are based on removal of Al, Cd, Cu, Fe, Pb, Mg, Mn, Ni, Se, and Zn at influent
concentrations observed during the pilot study.
• Costs associated with regulatory negotiations are excluded.
• Capital costs are based on a treatment system with 50 GPM capacity employing a solids
contact, upflow clarifier.
• Costs are for the treatment facility and conveyance piping only - collection systems at
the source are excluded.
• Plant water is assumed to be available for washdown, polymer makedown, etc.
The following design calculations were used to develop the costs estimated for the 50 GPM
lime treatment system (Tables F3 and F4):
Design Parameter Design Value Notes
Raw Influent Flowrate (gpm): 50.0
Raw Influent Flowrate (MGD): 0.072
Raw Wastewater flowrate (gpd): 72,000
Solids Thickener Decant (gpd): 1,012
Filter Press Filtrate Return (gpd): 1,481
Design Influent Flowrate (gpm): 52
Current Nominal Flowrate (gpd): 74,493
(MGD): 0.0745
Inlet pH: 3.4
Inlet Acidity (mg/L as CaCO3): 613.8
Stoichiometric Equivalanece (mg/L as CaCO3): 1,122
Inlet Aluminum Concentration (mg/L): 71
Inlet Cadmium Concentration (mg/L): 0.311
Inlet Copper Concentration (mg/L): 9
Inlet Iron Concentration (mg/L): 2.7
Inlet Lead Concentration (mg/L): 0.0
Inlet Magnesium Concentration (mg/L): 109
Inlet Manganese Concentration (mg/L): 77.1
Inlet Nickel Concentration (mg/L): 0.634
Inlet Selenium Concentration (mg/L): 0.017
Inlet Zinc Concentration (mg/L): 74.617
Inlet Calcium Concentration (mg/L): 251
Inlet Potassium Concentration (mg/L): 5
Inlet Sodium Concentration (mg/L): 44
Inlet Chloride Concentration (mg/L): 37
Inlet Flouride Concentration (mg/L): 3.7
Inlet Sulfate Concentration (mg/L): 1,782
decant water = solids loading, wet minus thickened
filter press filtrate = solids loading, wet minus pressed
raw influent, thickener decant, and filtrate return
raw influent, thickener decant, and filtrate return
based on observed influent pH in pilot
Buffering capacity of raw influent to a pH of
approximately 8.3
Reagent demand based on equivalent weight of metals
and sulfate
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
based on average influent concentrations from pilot
69
-------
Influent Equalization
Detention Time (days):
EQ Tank Volume (gal):
EQ Tank Diameter (ft):
EQ Tank Sidewater Depth (ft):
EQ TankSidewall Height (ft):
Reactor Vessel
Reactor Detention Time (hr):
Target Reactor Tank Volume (gal):
Reactor Tank Sidewater Depth (ft):
Reactor Tank Length (ft):
Reactor Tank Width (ft):
Nominal Reactor Tank Volume (ft3):
Nominal Reactor Tank Volume (gal):
Quicklime Dose
Target pH:
Target Alkalinity Addition (mg/L as CaCO3):
Target Quicklime Dosage (mg/L):
Quicklime Strength (wt% Ca):
Target Dosage (tons/day):
Estimated TSS (mg/L):
Flocculating Aid - Polymer Dosage
Target Concentration (mg/L):
Target Dosage (Ibs/day):
As Delivered Polymer Density (Ib/gal):
As Delivered Polymer Feed (gpd)
Neat Polymer (Ibs/day):
Clarifier
Influent TSS (mg/L):
Solids Load,
Aluminum,
Cadmium,
Copper,
Iron,
Lead,
Magnesium,
Manganese,
Nickel,
Selenium,
Zinc,
dry basis
dry basis
dry basis
dry basis
dry basis
dry basis
dry basis
dry basis
dry basis
dry basis
dry basis
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
(Ibs/day):
Clarifier Diameter (ft):
Clarifier Sidewater Depth (ft):
No. of Clarifiers:
Hydraulic Loading (gpd/ft2):
Hydraulic Loading (gpm/ft2)
Solids Loading (Ibs/day-ft2):
Weir Overflow (gal/day-ft):
assumed time for equalization of influent sources (raw
1.0 influent, decant water, filtrate water)
70,000 based on flow rate and detention time
20.0 size of tank necessary for volume
29.8 actual depth of water expected
36.0 height of tank necessary for volume
0.3 Based on nominal design flowrate
776 flow x detention time
5.0
5.0
5.0
125.0 calculated from tank dimensions
935
10.5
1,122 lime precipitation
628
90%
0.2
Based on raw influent flowrate assuming all ions
675 precipitate as hydroxides
3 As Delivered, made down solution, 0.5% strength
1.86 Based on nominal design flowrate
8.340
0.223
0.009
Based on raw influent flowrate and concentrations
675 assuming all ions precipitate as hydroxides
Based on raw influent flowrate and concentrations
405 assuming all ions precipitate as hydroxides
42.7 Based on pilot removal efficiency
0.187 Based on pilot removal efficiency
5.29 Based on pilot removal efficiency
1.61 Based on pilot removal efficiency
0.0115 Based on pilot removal efficiency
65.7 Based on pilot removal efficiency
46.3 Based on pilot removal efficiency
0.380 Based on pilot removal efficiency
0.0101 Based on pilot removal efficiency
44.8 Based on pilot removal efficiency
18
15.0
1.0
293 Target <400 gpd/ft2
0.203
1.6 Target less than 20 Ibs/day-ft2
1,317 Target less than 10,000 gal/day-ft
70
-------
Clarifier Volume (MG)
Design basin solids inventory, dry basis (Ibs)
Design sludge blanket (ft):
Design clarifier solids (gal):
Design clarifier solids (Ibs):
Clarifier underflow solids content (%):
Underflow density (Ibs/gal):
Jnderflow Wasteage required, dry basis (Ibs/day)
Underflow wastage (vol% of Influent Flow):
Underflow wastage, wet basis (Ibs/day):
Underflow wastage, wet basis (tons/day):
Underflow wastage (gpm):
Underflow wastage (gpd):
Underflow wastage (MGD):
Underflow wastage (MG/yr):
Upflow Recycle (Recycle to Influent TSS Ratio):
Upflow Recycle (Recycle to Influent Flow Ratio):
Upflow Recycle, dry basis (Ibs/day):
Upflow Recycle, wet basis (tons/day):
Upflow Recycle (gpm):
Upflow Recycle per clarifier (gpm)
Upflow Recycle (gpd):
Upflow Recycle (MGD):
Clarifier Effluent (gpd):
Clarifier Effluent (gpm):
Solids Thickener Tank
Solids Storage Tank Volume (gal):
Sludge density (Ibs/gal):
Sludge solids content (%):
Solids Loading, dry basis (Ibs/day):
Solids Loading, wet basis (Ibs/day):
Solids Loading (gpd)
No. Days Storage (days):
Thickened Solids, wet basis (Ibs/day):
Decant Water (gpd):
0.029
200 Clarifier volume x TSS (mg/L)
2
3,807 Based on volume of sludge blanket
971 Based on % solids and density of sludge
3.0%
8.5
405
2.1%
13,509
6.8
1.1
1,589
0.002
0.6
33.1 20:1 to 30:1 recycle to influent solids ratio, dry basis
6 4:1 to 6:1 recycle to influent flow
13,409
1,900
310.4
310.4
446,960
0.4
72,904
51
12,000
9.0
8.0%
405
13,509
1,589
7.6
5,065.9
1,012
Filter Press
Cake solids, dry basis (%):
Cake density (Ib/ft3):
Minimum Filter Press Size (ft3):
Pressed sludge, wet (Ibs/day):
(ton/day):
(ton/yr):
Aluminum (mg/kg):
Cadmium (mg/kg):
Copper (mg/kg):
Iron (mg/kg):
Lead (mg/kg):
Magnesium (mg/kg):
Manganese (mg/kg):
Nickel (mg/kg):
Selenium (mg/kg):
Znc (mg/kg):
Filtrate (gpd):
Based on underflow wasteage required, dry
solids loading, dry v by cake solids, %
35%
70
17
1,158
0.58
211
36,846
161
4,565
1,393
9.91
56,699
39,992
329
8.72
38,695
1,481 solids loading, wet minus pressed sludge, wet
71
-------
Coagulant (Sludge Conditioning)
Target Coagulant Concentration (mg/L): 200
based on target concentration and thickened solids,
Target Coagulant Concentration (Ibs/day): 8.4 wet
Coagulant Feed (gpd): 1.01
Polymer (Sludge Conditioning)
Target Polymer Concentration (Ibs/day): 4.1 [1% per dry Ib solids]
Polymer Feed (gpd): 0.49
Effluent Egualization
Detention Time (days): 1.0
EQ Tank Volume (gal): 70,000
EQ Tank Diameter (ft): 20.0
EQ Tank Sidewater Depth (ft): 29.8
EQ Tank Sidewall Height (ft): 36.0
72
-------
Table F3. Conventional lime treatment: Total capital investment cost.
Purchased Equipment Costs
QTY
Unit
Unit Cost Extension
Description
Clarifier (Bridge and Internals)
Clarifier Tank
Equalization Tanks
Paste Slaker/Slurry Tank/Slurry
Pumps/Super Sacks
Chemical Feed System
Carbonic Acid System
Sludge Pumps
Filter Press
Filter Press Feed Pump
Filter Press Polymer and DE systems
1
1
2
1
1
1
2
1
1
1
Lump
Lump
Lump
Lump
Lump
Lump
Lump
Lump
Lump
Lump
$50,000.00
$75,000.00
$105,000.00
$115,000.00
$12,500.00
$250,000.00
$5,000.00
$112,500.00
$4,000.00
$12,500.00
$50,000.00
$75,000.00
$210,000.00
$115,000.00
$12,500.00
$250,000.00
$10,000.00
$112,500.00
$4,000.00
$12,500.00
20-ft diameter
20-ft diameter, 15-ft sidewall height, ring
foundation to be installed on a concrete base
20-ft diameter, 36-ft high, bolted glass lined
steel tank with roof, ladders, safety platform,
and appurtenances
2,000-lb supersacks will be employed to deliver
bulk pebbled quicklime
Flocculant
Packaged Pressurized CO2 Delivery System
Progressive cavity pumps
Based on sludge characteristics and amount
Polymer Delivery System (Dynablend or
equivalent)
Sludge Storage Tank
Piping, valves, and appurtenances
Outfall
Instrumentation
Power drop
Motor Control Center
Control Panels / SCADA System
1 Each $12,000.00 $12,000.00
Nominal 14,000 gallon tank with catwalk,
1 Each $35,000.00 $35,000.00 sludge rake, and appurtenances
1 Lump $75,000.00 $75,000.00
1 Lump $25,000.00 $25,000.00
1 Lump $25,000.00 $25,000.00
1 Each $15,000.00 $15,000.00
1 Lump $50,000.00 $50,000.00
1 Each $75,000.00 $75,000.00
Equipment Subtotal (EQ.): $1,163,500.00
Taxes (5% of EQ)
Freight (2% of EQ)
Lump
Lump
$58,175.00
$23,270.00
$58,175.00
$23,270.00
Total Purchased Equipment Cost (EEC.): $1,244,945.00
Direct Installation Costs
QTY
Unit
Unit Cost Extension
Description
Mobilization, Demobilization, Permits, and
Temporary Controls 1
Treatment Building Foundation and Building
Erection 1,250
Clarifier Foundation and Erection 1
Clarifier Roof 314
Equalization Tank Foundation and Erection 2
Conveyance Forcemains to Distribution
Build. 5,280
Imported Backfill 750
Set Process Equipment (8% of EQ) 1
Process Piping (15% of EQ) 1
Heat trace and insulation (2% of EQ) 1
Painting (1.5% of EQ) 1
Heating and Lighting (8% of EQ) 1
Site Security (4.5% of EQ) 1
Electrical (15% of EQ) 1
Lump $40,000.00 $40,000.00
sf $200.00 $250,000.00 25 ft x 50 ft Pre-Engineered Building
Lump $75,000.00 $75,000.00
sq.ft. $45.00 $14,137.17 Fiberglass
Lump $35,000.00 $70,000.00
ft $50.00 $264,000.00
ton $22.00 $16,500.00 Pipe Bedding as Necessary
Lump $93,080.00 $93,080.00
Lump $116,350.00 $116,350.00
Lump $23,270.00 $23,270.00
Lump $17,452.50 $17,452.50
Lump $93,080.00 $93,080.00
Lump $52,357.50 $52,357.50
Lump $174,525.00 $174,525.00
Total Direct Installation Cost(Ql): $1,299,752.17
TOTAL DIRECT COST (DC) [PEC + PI]: $2,544,697.17|
Indirect Costs
QTY
Unit
Unit Cost Extension
Description
Engineering (6% of DC)
Administration/PM (5% of DC)
Geotechnical
Bonds (1.5% DC)
Construction Oversight (6% of DC)
Start Up
Contractor Profit (10% of PEC)
Contingencies (5% of DC)
1 Lump
1 Lump
1 Lump
1 Lump
1 Lump
1 Lump
1 Lump
1 Lump
$152,681.83
$127,234.86
$15,000.00
$38,170.46
$152,681.83
$25,000.00
$124,494.50
$127,234.86
$152,681.83
$127,234.86
$15,000.00
$38,170.46
$152,681.83
$25,000.00
$124,494.50
$127,234.86
Total Indirect Cost (Kl): $762,498.33
TOTAL CAPITAL INVESTMENT (TCI) FDC + IC1: $3,307,195.50 |
73
-------
Table F4. Conventional lime treatment: Total annual operating costs.
Direct Annual Operating Costs
QTY
Unit
Unit Cost Extension
Description
Maintenance and Replacement Parts (4%
EQ)
Laboratory Analytical
Sampling/Operator & Maintenance Labor
(OL)
1 Lump $46,540.00 $46,540.00
26 Biweekly $500.00 $13,000.00
2,080 Staff Hours $85.00 $176,800.00 40 hours per week, 52 weeks per year
Polymer
Coagulant
CO2
Quicklime (CaO)
Sludge Disposal
Sludge Transportation & Demurrage
Electricity (1 ,780 kW-hr/day assumed)
3,764
3,084
55,482
76
211
21.1
649,700
Ib
Ib
Ib
ton
ton
Load
kW-hr
$3.50
$3.50
$0.50
$200.00
$75
$700
$0.12
$13,175.71
$10,794.68
$27,740.89
$15,297.70
$15,848.89
$14,792.29
$77,964.00
Assumed based on 250 ppm dosage
Assumed based on 600 ppm dosage, 90%
strength (-0.2 TPD).
Assumed 35% solids filter cake.
Equivalent to 1 00 brake horsepower.
Total Direct Annual Operating Cost (DAC.): $411,954.15
Indirect Annual Operating Costs
QTY
Unit
Unit Cost Extension
Description
Overhead & Administrative Charges (1.0%
TCI) 1 Lump
Property Taxes (1.0% TCI) 1 Lump
Insurance (1.0% TCI) 1 Lump
- Excluded -
- Excluded -
- Excluded -
Total Indirect Annual Operating Cost (Mi): $0.00
TOTAL ANNUAL COST (TftC) [DftC + \PC]: $411,954.15
74
------- |