TECHNICAL TECHNICAL

REPORTS: WASTE
MANAGEMENT
REPORTS

Novel Passive Co-Treatment of Acid Mine Drainage and Municipal Wastewater

William H. J. Strosnider,* Brandon K. Winfrey, and Robert W. Nairn
A laboratory-scale, four-stage continuous-ow reactor system
was constructed to test the viability of high-strength acid mine
drainage (AMD) and municipal wastewater (MWW) passive
co-treatment. Synthetic AMD of pH 2.6 and acidity of 1870
mg L1 as CaCO3 equivalent containing a mean 46, 0.25, 2.0,
290, 55, 1.2, and 390 mg L1 of Al, As, Cd, Fe, Mn, Pb, and Zn,
respectively, was added at a 1:2 ratio with raw MWW from the
City of Norman, OK, to the system which had a total residence
time of 6.6 d. During the 135-d experiment, dissolved Al,
As, Cd, Fe, Mn, Pb, and Zn concentrations were consistently
decreased by 99.8, 87.8, 97.7, 99.8, 13.9, 87.9, and 73.4%,
respectively, pH increased to 6.79, and net acidic inuent was
converted to net alkaline euent. At a wasting rate of 0.69%
of total inuent ow, the system produced sludge with total Al,
As, Cd, Cr, Cu, Fe, Pb, and Zn concentrations at least an order
of magnitude greater than the inuent mix, which presents a
metal reclamation opportunity. Results indicate that AMD and
MWW passive co-treatment is a viable approach to use wastes
as resources to improve water quality with minimal use of fossil
fuels and rened materials.

Copyright 2011 by the American Society of Agronomy, Crop Science

Society of America, and Soil Science Society of America. All rights
reserved. No part of this periodical may be reproduced or transmitted
in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system,
without permission in writing from the publisher.
J. Environ. Qual. 40:206213 (2011)
doi:10.2134/jeq2010.0176
Published online 9 Nov. 2010.
Received 15 Apr. 2010.
*Corresponding author (bill.strosnider@gmail.com).
ASA, CSSA, SSSA
5585 Guilford Rd., Madison, WI 53711 USA

cid mine drainage (AMD) and municipal wastewater

(MWW) are prevalent across the globe, and their treatment
is key to maintaining global water resource quality. Untreated
AMD causes water quality degradation in coal- and metal-mining regions worldwide (Wolkersdorfer and Bowell, 2004a,b,c).
Discharges of untreated MWW degrade water resources in many
developing nations (Gadgil, 1998; Nelson et al., 2001). In developed nations, MWW is generally addressed with active treatment,
consuming considerable nancial, material, and energy resources
(Muga and Mihelcic, 2008). Compared to passive methods, conventional active MWW and AMD treatment are energy intensive
with higher operational and maintenance costs (Nelson et al.,
2001;Younger et al., 2002; Muga and Mihelcic, 2008).
The passive co-treatment of AMD and MWW is a nascent
application of ecological engineering that blends aspects of passive
AMD treatment and conventional active MWW treatment. The
passive treatment of AMD often requires suitable organic substrate electron donors for dissolved oxygen (DO) stripping, bacterial SO42 reduction (BSR), and the bacterially mediated reduction
of metals, such as Fe. Conventional active MWW treatment can
require electron acceptors for bacterial oxidation of carbon substrate, chemicals or ultraviolet light for pathogen removal, and
physical/chemical ltration, coagulation, and/or occulation.
Numerous carbon sources, including horse and cow manure,
chicken litter, ethanol, methanol, and municipal sewage sludge
and compost, have been successfully applied to treat AMD by
BSR, which generates alkalinity via bicarbonate production and
remove metals from solution via sulde precipitation (Neculita
et al., 2007). Metals may also be removed from solution by sorption to the aforementioned organic substrates (Machemer and
Wildeman, 1992; Neculita et al., 2007). Municipal wastewater
contains a wide variety of organic compounds ranging from simple
sugars to cellulose (Metcalf and Eddy, Inc., 1991). Generally, BSR
is coupled with limestone dissolution in passive treatment cells
such as vertical ow bioreactors (VFB) to maximize alkalinity
generation and metals retention (Watzlaf et al., 2004).

W.H.J. Strosnider, Saint Francis Univ., Environmental Engineering Dep., 117 Evergreen
Dr., Loretto, PA 15940; B.K. Winfrey, Univ. of Maryland, Dep. of Environmental Science
& Technology, 1426 Animal Sciences Building, College Park, MD 20742; R.W. Nairn,
Univ. of Oklahoma, School of Civil Engineering and Environmental Science, Center
for Restoration of Ecosystems and Watersheds, 202 W. Boyd St., Norman, OK 73019.
Assigned to Associate Editor Peter E. Holm.
Abbreviations: AMD, acid mine drainage; BOD5, 5-d biochemical oxygen demand; BSR,
bacterial sulfate reduction; DO, dissolved oxygen; HDPE, high-density polyethylene;
IRB, iron-reducing bacteria; LDPE, low-density polyethylene; MPN, most probable
number; MWW, municipal wastewater; SC, specific conductance; SRB, sulfate-reducing
bacteria; theomix, theoretical mix; VFB, vertical flow bioreactor.

206

Although Roetman (1932) rst suggested mixing AMD

with MWW for pathogen removal, few studies have investigated their co-treatment. Fortuitous water quality improvement and BSR has been demonstrated in an evaporation
pond in which high-strength AMD was accidentally introduced to secondary MWW (McCullough et al., 2008). In
addition, algal-based systems have been developed in waste
stabilization ponds that treated synthetic AMD and organicrich tannery euent to achieve relatively high metal removal
eciencies (Rose et al., 1998). Despite promising short-term
performance, these mixed-euent systems failed after a few
months due to algal metal toxicity (Van Hille et al., 1999). In
laboratory microcosm studies, Van Hille et al. (1999) initially
separated waste streams, using MWW to create algae-rich
high-pH euent before mixing with AMD. One eld-scale
single-stage constructed wetland co-treatment system has
been constructed that successfully improved the water quality
of weak secondary MWW euent (5-d biochemical oxygen
demand [BOD5] of ~14 mg L1) and relatively weak (net
alkaline with ~3 mg L1 Fe) AMD (Johnson and Younger,
2006). Strosnider and Nairn (2010) demonstrated that multiple metals of concern could be removed from solution by
incubating high-strength AMD and raw MWW with and
without the presence of limestone. However, the authors are
not aware of any other ow-through high-strength AMD and
raw MWW co-treatment investigation, which is essential to
addressing the approachs feasibility and applicability. The
system described in this manuscript is the rst attempt to
simultaneously co-treat high-strength AMD and raw MWW
in a ow-through regime. Strosnider et al. (2010b) and
Winfrey et al. (2010) documented highly eective MWW
component (BOD5, phosphorus, and fecal indicator bacteria) processing within the same system as described in this
article. To completely evaluate the promise of high-strength
AMD and raw MWW co-treatment, the objectives of this
study were to determine metal removal eciencies and concentrations of metals in the sludge produced.

Materials and Methods

Experimental Design
The experimental setup involved four serial unit processes in
quadruplicate (Fig. 1). First, the MWW and AMD were introduced to the clarier to raise AMD pH to that less inhibitory for BSR and promote metals complexation with organic
ligands. Iron and Al occulation with phosphate, Fe(III)
reduction, DO stripping, and solids settling were other aims of
this unit process. The second and third unit processes together
emulated VFB. The upper column sections, which were lled
with inert biomedia, were designed to encourage further DO
stripping, Fe(III) reduction, and BSR. The bottoms of the
columns were lled with limestone for abiotic alkalinity generation via calcite dissolution and to encourage further BSR.
The nal unit processes were aerobic wetland mesocosms for
sequential Fe then Mn oxidation and precipitation as well as
the removal of remaining As, Cd, Pb, and Zn via sorption to
Fe oxyhydroxides. Each unit process was connected to the next
via clear vinyl tubing and sampled at its outow.

Fig. 1. Conceptual experimental layout for simultaneous cotreatment of

high-strength acid mine drainage (AMD) and raw municipal wastewater (MWW) in a flow-through regime. C1 to C4 represent the clarifiers,
K1 to K4 the Kaldnes K3 biofilm media, L1 to L4 the limestone, and W1
to W4 the wetlands. Black dots indicate sampling points.

Retention times of 1.5 to 2.5 h are typical for conventional

MWW primary clarication systems (Metcalf and Eddy, Inc.,
1991). However, the clarier unit process was sized for a relatively high retention time of 32 h for more complete mixing
through passive diusion, encouragement of BSR, and thorough solids settling. Single transverse baes 4 cm deep and
semicircular weirs 2.5-cm radius served as the physical structures in the low-density polyethylene (LDPE) basins that comprised the clariers. Sludge was wasted from the bottom of the
clariers under gravity ow with a barbed high-density polyethylene (HDPE) T-connector attached to a HDPE valve and
clear vinyl tubing.
The VFB emulation columns were 91.5 cm in height and
12.5 cm in diameter. The bottom 38 cm of the columns were
lled with high-quality (>90% CaCO3) limestone washed of
all nes and separated by sieve analysis adapted from ASTM
D422 with the fraction passing a 2.54-cm sieve yet retained by
a 1.27-cm sieve. The remaining top 53.5 cm of the columns
were packed with Kaldnes K3 biolm media (AnoxKaldnes
Inc., Tnsberg, Norway) to provide sulfate-reducing bacteria
(SRB) biolm attachment surface. Kaldnes K3 biolm media
are polyethylene high-surface-area (500 m2 m3) components
that are typically used in moving-bed biolm wastewater
and drinking water treatment (Saliling et al., 2007). Because
inoculation can be important to SRB performance (Pruden
et al., 2007), the Kaldnes stage was inoculated with 100 mL

Strosnider et al.: Passive Co-Treatment of Acid Mine Drainage and Municipal Wastewater

207

of homogenized VFB substrate from two mature passive

coal mine AMD treatment systems in Pittsburg and Latimer
counties, Oklahoma. Each column was wrapped in aluminum foil to emulate the lightless conditions in VFB substrate.
Residence time was 42 and 18 h for the Kaldnes and limestone stages, respectively.
The aerobic constructed treatment wetland mesocosms were
two shallow LDPE storage containers that were bisected longitudinally with plastic to create the necessary four treatment
trains. Hydric soils were collected from an existing constructed
mitigation wetland that receives tertiary wastewater at the
Midwest City, OK, MWW Treatment Plant. The surface ow
mesocosms were planted with Hydrocotyle verticillata Thunb.
and Nasturtium ocinale W.T. Aiton. The wetlands were
placed under timed grow-lights on a 12 h d1 cycle. Wetland
residence time was 67 h.
Raw MWW collected after grit screening at the City of
Norman, OK, Water Pollution Control Facility and synthetic AMD approximating that found at the base/precious
metal mining district of Cerro Rico de Potos, Bolivia, were
introduced to the system at a 2:1 ratio (MWW:AMD) with
the aim of producing clarier pH less inhibitive to SRB
(Strosnider et al., 2010a). In Potos, untreated high-strength
AMD and MWW pollute the headwaters of the Ro Pilcomayo
(Strosnider and Nairn, 2010). The AMD and MWW were
introduced with peristaltic pumps at a combined ow rate
(3.8 L d1). The system was gravity ow from the rst (clarier) to the last (wetland) unit processes, with a calculated total
residence time of 6.6 d. Municipal wastewater was collected
weekly, homogenized during pumping, and kept refrigerated
at 4C before introduction to the system. Acid mine drainage
was prepared weekly and stored at a thermostatically controlled
20C until use, at which all unit processes were maintained.
Sludge was wasted from the clariers in varying amounts at
irregular intervals.
Each treatment train continuously handled the mixed inuent for 135 d. Sampling generally occurred twice monthly
during this standard operational run for a total of 10 sampling events. Dissolved metals samples as well as pH, specic
conductance (SC), and DO were taken throughout the entire
experiment. Total metals samples were only taken during the
nal two sampling events. Water samples for most probable
number (MPN) tests were taken from the columns on the nal
sampling event (Day 133). Sludge metal samples were taken
each time sludge was wasted.

Data Collection
Temperature, pH, and SC were determined using an Orion
1230 multimeter (ThermoFisher, Inc., Waltham, MA).
Alkalinity titrations were conducted in accordance with Hach
Method 8203 (Hach, 2006). An Accumet AR60 multimeter
(ThermoFisher, Inc., Waltham, MA) was used to determine
DO concentrations. Dissolved metals samples were ltered
through 0.45-m nylon lters before preservation. Total and
dissolved metals samples were preserved with concentrated
trace metal-grade nitric acid and stored at 4C until microwave
acid digestion following USEPA Method 3015. Digested total
metals samples were ltered through 0.45-m nylon lters
before analysis. Total and dissolved metals samples were ana208

lyzed via a Varian Vista-Pro simultaneous inductively coupled

plasmaoptical emission spectrometer (Varian, Palo Alto, CA)
following USEPA Method 6010.
Iron-reducing bacteria (IRB) populations were estimated
with the MPN technique following Standard Methods
(APHA, 2005) with a slight growth media modication per
Viswanathan (2005). Triplicate test tubes with Durham tubes
were inoculated in an anaerobic nitrogen atmosphere for a
dilution range of 1:102 to 1:106, then incubated at 25C for 14
d and quantied following Woomer (1994).

Data Analysis
Because direct sampling of the clarier inuent mix was
impossible and the reactivity of these waters, the chemical
composition of the theoretical mix (theomix) was calculated
using the ratio of AMD to MWW. Net acidity and acidity
were calculated following Younger et al. (2002). For acidity
calculations, all Fe was assumed to be Fe(II) when pH > 3
and Fe(III) when pH < 3, because Fe(III) predominates for
most solutions with pH < 3 (Watzlaf et al., 2004; Kirby and
Cravotta, 2005). To facilitate the calculation of means and
application of statistical tests, values below detection limits
were assigned one-half the value of the detection limit (Miller
and Miller, 1986). Before means or median testing, all data
sets were tested for normality with the AndersonDarling
test and homoscedasticity. Following Dean and Voss (1999),
when variances varied by greater than a factor of three data
sets were treated as heteroscedastic. Welchs t test was applied
to test the theomix total vs. dissolved metal concentrations,
and clarier sludge total metal concentrations vs. the theomix
total and dissolved metal concentrations because these data
sets were generally normal and homoscedastic. The nonparametric KruskalWallis multiple-comparisons test was used to
determine statistical dierence between the medians of the
remaining data sets due to the prevalence of heteroscedasticity and nonnormality. All statistical testing was completed
with Minitab version 15 (Minitab, State College, PA) using
= 0.05.

Results and Discussion

Influent Characteristics
Inuent characteristics were relatively uniform throughout
the experiment (Tables 1 and 2). The mean alkalinity, BOD5
(265 mg L1), Cl (69 mg L1), and SO42 concentrations place
the MWW between the medium and strong designations of raw MWW created by Metcalf and Eddy, Inc. (1991)
(Strosnider et al., 2010a). The synthetic AMD was similar to
the high-strength, highly acidic AMD generated at Cerro Rico
de Potos (Strosnider and Nairn, 2010), which has concentrations of acidity and some metals orders of magnitude greater
than more commonly encountered AMD (Watzlaf et al., 2004;
Younger et al., 2002).

Aluminum
Aluminum was primarily removed from solution in the clarier and limestone stages (Fig. 2 and Table 3). At the pH
in the clarier, Al can combine with phosphate to form a
relatively stable solid, complex with particulate organic
Journal of Environmental Quality Volume 40 JanuaryFebruary 2011

Table 1. Mean influent acid mine drainage (AMD) and municipal wastewater (MWW) physiochemical properties.

Iron

Iron was removed from solution via multiple means in the clarier and wetland unit
S cm1
mg L1 as CaCO3 equivalent
mg L1
processes (Fig. 2 and Table 3). Flocculation
MWW
7.67
0.98
951
288
287
with phosphate likely removed some Fe
SD
0.12
0.49
66
20
20
from solution in the clarier (e.g., Omoike
and Vanloon, 1999). Combination with
AMD
2.60
7.69
3006
0
1865
organic ligands and the formation of Fe
SD
0.04
0.64
112
0
91
suldes also likely decreased Fe concen DO, dissolved oxygen.
trations in the clarier (e.g., Fletcher and
SC, specific conductance.
Beckett, 1987; Machemer and Wildeman,
1992; Younger et al., 2002). Strosnider et
matter, precipitate as alunite [KAl3(OH)6(SO4)2], and react
al.
(2010a)
documented
the presence of SRB (100 to 104 colwith dissolved organic carbon and precipitate (Omoike and
ony-forming units [cfu] mL1) in the clarier water column
Vanloon, 1999; Munk et al., 2002; Wilkin, 2008). It is likely
and sludge supernatant; populations that reduced sulfate
that all of these removal mechanisms were occurring to vary1
from 680 to 610 mg L1 from the theomix to the clarier
ing extents in the clarier. However, the mean 46.3 mg L
outow for a mean rate of 0.56 mol m3 d1. Aqueous Fe is
total vs. 7.66 mg L1 dissolved Al in the clarier euent was
primarily trivalent at the inuent AMD pH (2.60) (Kirby
indicative of inecient sedimentation. The Al oc did not
and Cravotta, 2005). The increase in pH to 4.47 likely caused
pass through the Kaldnes stage, as there was no statistical difsome remaining Fe(III) precipitation by oxyhydroxide formaference between dissolved and total Al in the Kaldnes eu2
tion because Fe(III) is rapidly hydrolyzed and precipitated
ent. The lack of Al, K, or SO4 removal in the Kaldnes stage
from solution in waters with pH > 4 (Younger et al., 2002).
indicates that alunite formation was likely an insignicant
Minimal Fe(II) oxidation is likely to have occurred in the
removal mechanism in the clarier because the bulk of parclarier due to low DO concentrations. The mean 94.7 mg
ticulate organic matter settled and phosphate was decreased
1
L1 total vs. 56.3 mg L1 of dissolved Fe in the clarier euent
to below detection limits (<0.75 mg L ) in the clarier, leavwas indicative of inecient clarication. Oxidation, hydroing alunite formation the most plausible removal mechanism
lysis, and sedimentation were the likely mechanisms of Fe
in the Kaldnes. In addition, dissolved K concentrations were
removal in the wetland due to the DO and pH.
unchanged in the clarier and K was not concentrated in
Key to the performance of the system, dissolved Fe
the clarier sludge. In the limestone stage pH increased to
remained
unchanged through the Kaldnes and limestone unit
6.72, which dramatically decreased Al solubility, likely formprocesses
and therefore did not armor the limestone. This
ing insoluble amorphous Al(OH)3 (e.g., Munk et al., 2002;
indicates
that
the remaining Fe(III) was reduced to Fe(II) in
Wilkin, 2008). This solid was ushed from the limestone
the
clari
er
and/or
Kaldnes stages, which allowed it to pass
during standard sampling events, resulting in a mean 77 mg
1
1
through
the
limestone
without forming Fe(OH)3 solids.
L total vs. 0.053 mg L dissolved Al, indicating the potenIron-reducing
bacteria
can
use simple sugars that are present
tial of ushing for metal recovery and to retain long-term
in
substantial
concentrations
in MWW (Metcalf and Eddy,
hydraulic conductivity.
Inc., 1991; Lovley, 1991). Iron-reducing bacteria populations
were estimated at 250,000 and 40,000 cells 100 mL1 in the
pH

DO

SC

Alkalinity

Net acidity

Table 2. Mean influent acid mine drainage (AMD) and municipal wastewater (MWW) dissolved (n = 10) and total (n = 2) metal concentrations.
MWW
Metal

Dissolved
Mean

Al
As
Ca
Cd
Cr
Cu
Fe
K
Mg
Mn
Na
Ni
Pb
Zn

0.108
<0.022
39.6
0.0010
0.0036
0.0067
0.315
16.3
21.1
0.056
73.7
0.010
0.015
0.045

SD
0.018
4.5
0.0003
0.0045
0.0021
0.091
0.58
2.99
0.010
3.25
0.0017
0.0051
0.0402

AMD
Total
Mean

Dissolved
Mean

mg L1
0.692
45.8
<0.022
0.25
40.5
82.8
0.0009
2.02
0.0066
0.027
0.029
0.0052
0.719
292
16.6
0.46
18.2
26.1
0.063
54.6
66.9
<0.0006
0.0048
0.145
0.014
1.21
0.526
391

Strosnider et al.: Passive Co-Treatment of Acid Mine Drainage and Municipal Wastewater

SD

Total
Mean

3.2
0.14
3.5
0.08
0.034
0.0028
23.8
0.62
1.3
3.0

0.026
0.10
21.6

46.3
0.38
90.6
2.29
0.012
0.088
268
0.02
26.5
54.4
<0.0006
0.192
1.25
388
209

Kaldnes and limestone stages, respectively, indicating

especially active IRB in the Kaldnes stage where reduction of any remaining Fe(III) is crucial. The lower IRB
concentrations in the limestone stage are possibly due
to the eciency of clarier and Kaldnes IRB at depleting Fe(III) concentrations.
Aerobic heterotroph bacterial activity is the likely
mechanism by which DO was driven below the suggested anoxic limestone drain design parameter of <1
mg L1 DO in the Kaldnes stages to limit the oxidation
of Fe(II) within the limestone stages. Limestone in the
presence of Fe(III) or Fe(II) and DO >1 mg L1 will
become coated with Fe(OH)3, dramatically lowering
alkalinity production and porosity which can dramatically reduce the eective lifetimes of passive treatment
systems (Younger et al., 2002; Watzlaf et al., 2004; Fig. 2. Mean of all sampling events for dissolved Al, Fe, Mn, and Zn concenSantomartino and Webb, 2007). The combined activ- trations (bars) with respect to dissolved oxygen (DO) and pH (lines) for the
ity of heterotrophic and Fe-reducing bacterial activity theoretical mix (Theomix) and the outflow of each unit process. Error bars
allowed Fe removal to be isolated in the clarier and represent 1 SD above and below the mean.
wetland stages where precipitate accumulation is prehave a relatively high anity for organic ligands (Fletcher and
ferred while protecting the VFB emulation stages from limeBeckett, 1987; Machemer and Wildeman, 1992). Although
stone armoring or clogging.
signicant concentrations of Zn were removed from solution
in the clarier, there was no signicant dierence between total
Manganese
and dissolved concentrations in the clarier euent, which
Manganese remained unchanged throughout the system until
implied ecient clarication of Zn solids and is contrary to the
the wetland stage, where it decreased by 13.6% (Fig. 2 and
trend noted in Al, Cd, Pb, and Fe. This may also indicate that
Table 3). The oxidation and hydrolysis of Mn was possible
Zn removal was primarily a result of BSR because the majority
because pH was >6 (6.95) and Fe was driven to such low conof Zn removal occurred at the bottom of the clarier in the
centrations (from 45 to 0.18 mg L1) by the outow of the
organo-metallic sludge where SRB activity is thought to have
wetland. It is likely that Mn removal primarily occurred by the
been concentrated (Strosnider et al., 2010a). In the wetland
wetland outow where Fe concentrations were lower because
stage, Zn removal was likely due to sorption to Fe oxyhydroxFe (II) will reduce oxidized forms of Mn when present (Watzlaf
ides. Zinc has a high anity for Fe oxyhydroxides, especially
et al., 2004).
at the pH present in the wetlands (Carroll et al., 1998; Mayes
et al., 2009).

Zinc

Zinc was primarily removed from solution in the clarier and

wetland (Fig. 2 and Table 3). In the clarier, sphalerite (ZnS)
formation, and complexation with organic ligands present
in the MWW likely served as the primary Zn removal reactions. Bacterial sulfate reduction can remove Zn from solution via sphalerite formation (Younger et al., 2002). Zinc can

Cadmium
Cadmium was removed in the clarier (Table 3), likely from
sulde formation via BSR and complexation with organic
matter. Cadmium has a high anity for organic ligands with
which it will complex as well as adsorb to bacterial solids
between pH 4 and 5 (Fristoe and Nelson, 1983). As in the case

Table 3. Dissolved metals removal performance. Unit process removal is the percent difference between unit process inflow and outflow.
Metal

Al
As
Cd
Cr
Cu
Fe
Mn
Ni
Pb
Zn

AMD

Theomix

Wetland

mg L1
45.8
15.2
0.034
0.25
0.090
<0.022
2.02
0.67
0.015
0.027
0.011
0.0011
0.005
0.006
0.0019
292
96.4
0.18
54.6
18.0
15.5
0.145
0.051
0.039
1.21
0.40
0.049
391
129
34.3

Overall
removal
%
99.8
87.8
97.7
90.4
68.3
99.8
13.9
23.2
87.9
73.4

Primary removal
unit process

Unit process
removal

Clarifier/limestone
Clarifier
Clarifier/Kaldnes
Clarifier
Clarifier
Clarifier/wetland
Wetland
Clarifier
Clarifier/Kaldnes
Clarifier/wetland

%
49.5
99.5
87.8
91.1
85.0
66.2
77.8
41.6
99.6
13.6
21.4
83.0
32.2
39.1
47.0

Unit process treatment rate

g m2 d1
0.71
2.5
0.007
0.06
0.01
0.0007
0.0004
3.8
0.34
0.02
0.001
0.03
0.01
4.8
0.23

g m3 d1
5.7
6.6
0.060
0.46
0.02
0.006
0.004
30
16
0.88
0.008
0.25
0.01
38
11

AMD, acid mine drainage.

Theomix, theoretical mix.
Kaldnes, Kaldnes K3 biofilm media.
210

Journal of Environmental Quality Volume 40 JanuaryFebruary 2011

of Al, the mean 0.16 mg L1 total vs. 0.059 mg L1 of dissolved

Cd in the clarier euent was indicative of inecient sedimentation. The Cd-enriched oc did not likely pass through
the Kaldnes stage as there was no statistical dierence between
dissolved and total Cd in the Kaldnes euent. Although Cd
can complex with Fe oxyhydroxides (Carroll et al., 1998), Cd
increased slightly within the wetland, likely due to evapoconcentration and the lack of aerobic removal mechanisms. Zinc
sorption suppresses Cd sorption to Fe oxyhydroxides (Carroll
et al., 1998), and signicant Zn sorption likely occurred in
the wetlands. Therefore, it is unlikely that sorption to Fe oxyhydroxides was an important Cd removal mechanism in the
clarier as well, where Zn concentrations were approximately
3 higher than in the wetland.

Lead
Lead was removed in the clarier and remained unchanged
throughout the subsequent unit processes (Table 3). In
the clarier Pb likely complexed with particulate organic
matter, formed galena (PbS) via BSR, and/or sorbed to Fe
oxyhydroxides (e.g., Fletcher and Beckett, 1987; Aderhold
et al., 1996; Carroll et al., 1998; Younger et al., 2002). The
mean 0.14 mg L1 total vs. 0.069 mg L1 of dissolved Pb
in the clarier euent was indicative of inecient clarication. Although Pb can sorb to Fe oxyhydroxides at the
pH range found within the wetland and should not be precluded by Zn sorption (Carroll et al., 1998), no signicant
dierence was noted between dissolved Pb in the limestone
and wetland euent.

Arsenic
Due to the rapid removal of As to below detection limits
(<0.022 mg L1) in the clarier (Table 3), the full extent of
removal is unknown throughout the remainder of the system.
The clarier likely had a gradient of DO concentrations
decreasing from the surface to the organo-metallic sludge at the
bottom. This gradient would allow for aerobic [i.e., Fe(II) oxidation] and anaerobic processes (i.e., BSR). It is likely that As
sorbed to Fe oxyhydroxides formed in the clarier. Amorphous
Fe oxyhydroxides such as would have been formed in the clarier have the highest adsorption capacity (Mohan and Pittman,
2007). It is also likely that As2S3 or AsS formed from BSR (e.g.,
Younger et al., 2002; Luo et al., 2008).

Minor Metal Constituents

Chromium, Cu, and Ni were also removed via co-treatment,
despite their low concentrations in the AMD and MWW
(Table 3). Chromium and Cu were concentrated in the clarier sludge and a signicant decrease was noted between the
theomix and the wetland euent. Although there was no statistically signicant dierence between the theomix and wetland euent, Ni was concentrated in the clarier sludge, which
indicates marginal removal. Possible removal mechanisms are
Cr reduction and subsequent hydrolysis, Cu sulde formation or sorption to Fe oxyhydroxides, and Ni sulde formation
and sorption to organic solids and/or Fe oxyhydroxides (e.g.,
Younger et al., 2002; Wilkin, 2008).

Overall Performance
The system produced euent with mean pH 6.97 and net
alkalinity of 10.4 mg L1 as CaCO3 equivalent. The alkalinity
imparted by the MWW and generated within the system supported many of the aforementioned metal removal processes
(Strosnider et al., 2010a). Although the system was bacterially mediated, metals removal performance generally displayed
a lack of temporal variability. The system also consistently
decreased BOD5 to <3 mg L1, phosphate to <0.75 mg L1,
and fecal indicator bacteria to <2 cfu 100 mL1 (Strosnider et
al., 2010a; Winfrey et al., 2010). Johnson and Younger (2006)
reported 40 to 80% Fe removal in a single-stage constructed
co-treatment wetland with a mean residence time of 14 h
receiving AMD with ~3 mg L1 Fe. Although the residence
time of the multistage co-treatment system was approximately
10 that of the Johnson and Younger (2006) system, the multistage system demonstrated much higher percent Fe removal
while handling AMD with an Fe concentration approximately
100 greater.
The multistage co-treatment system also performed well
with respect to conventional AMD passive treatment systems
(Younger et al., 2002; Watzlaf et al., 2004), despite treating
AMD that was far more acidic and metals-rich than that for
which passive treatment systems are generally applied. It should
also be noted that the multistage co-treatment system was not
optimized for maximum sustainable loading rates and its full
potential is unknown. Although 99.6% of Fe was removed
from solution in the wetland stage, there was not sucient Fe
loading (0.34 g m2 d1) to reach typical (1020 g m2 d1)
aerobic wetland design removal rates (Younger et al., 2002).
Overall, the high percent removal of key ecotoxic elements
indicates that greater removal rates are likely achievable in an
optimized system. Additionally, the removal of metals initially
at relatively low concentrations (Cr, Cu, and Ni) indicates that
co-treatment may also be an eective polishing tool for AMD
with lower metal concentrations and acidity.

Waste Generation
Approximately 0.69% of the total ow entering the clarier
was wasted throughout the experiment. This wasting rate led
to sustainable metals removal that did not compromise clarier
residence time. Sludge total Al, As, Cd, Cu, Mn, Ni, Pb, and
Zn concentrations were negatively correlated with wasting rate
(Fig. 3). The sludge contained mean concentrations of total Al,
As, Cd, Cr, Cu, Fe, Mn, Ni, Pb, and Zn of 179, 1.73, 19.6,
0.33, 0.57, 1350, 38, 0.37, 12.4, and 955 mg L1, respectively,
which is much greater than the dissolved theomix. The concentration of Mn and Ni in the sludge indicated that some degree
of Mn and Ni removal was occurring in the clarier which
could not be concluded from solely analyzing the theomix and
clarier outow data. Manganese and Ni removal was likely
primarily due to sorption to organic matter (e.g., Machemer
and Wildeman, 1992; Aderhold et al., 1996).
The high concentrations of multiple ecotoxic metals of
interest in the sludge present a possible reclamation opportunity or if mishandled an environmental hazard. Iron has been
economically recovered from standard passive treatment systems for years and Mn recovery has recently begun (Hedin,

Strosnider et al.: Passive Co-Treatment of Acid Mine Drainage and Municipal Wastewater

211

2003; Denholm et al., 2008). It is important

to note that the sludge was not dewatered
before sampling and analysis. It may be possible to drive concentrations of many reclaimable metals from mg L1 to g L1 by dewatering.
Subsequent smelting or sequential extraction
by acidication may be a sensible reclamation
approach. Reclaimed metals could be a continual source of revenue for future large-scale
passive co-treatment systems.

Conclusions
In the co-treatment system described above,
the stage was set for a self-designed microbial
ecosystem to develop and take advantage of
the available energy ows for preferential biochemical reactions (e.g., SO42 and Fe reduction) as well as create suitable conditions (e.g.,
anaerobic and aerobic zones) for important
abiotic geochemical reactions (e.g., calcite
dissolution and Fe oxidation). These results,
coupled with earlier ndings, indicate that
multistage passive co-treatment is capable of
highly ecient processing of both AMD and
MWW constituents. Co-treatment could augment the aordability and applicability of
passive treatment approaches. High-strength
AMD previously suitable only for energyintensive active treatment can be passively
co-treated with MWW. Signicant cost and
energy savings would result from eliminating Fig. 3. Total metals concentrations in sludge (y-axis) with respect to percent clarifier outthe need to purchase and transport organic flow wasted (x-axis). Horizontal bars frame the 95% confidence interval constructed with
substrate, which can be a major cost of pas- the t statistic of the mean dissolved metal concentration in the theoretical mix.
sive AMD treatment. In addition, the use of
resource recovery (A preliminary case study). p. 285297. In ProceedMWW as a substrate consumes an item with negative societal
ings, 2008 Natl. Meeting of the Am. Soc. of Mining and Reclamation,
Richmond, VA. 1419 June 2008.
value in the place of valuable organic substrate, such as comFletcher, P., and P.H. Beckett. 1987. The chemistry of heavy metals in digested
post. Future studies should investigate metals reclamation
sewage sludge: II. Heavy metal complexation with soluble organic matfrom the sludge generated and the eect of varied loading
ter. Water Resour. 21:11631172.
rates/ratios, improved clarication, and inclusion of further
Fristoe, B.R., and P.O. Nelson. 1983. Equilibrium chemical modeling of heavy
metals in activated sludge. Water Resour. 17:771778.
unit processes, to develop design guidance for full-scale pasGadgil, A. 1998. Drinking water in developing countries. Annu. Rev. Energy
sive co-treatment systems.
Environ. 23:253286.

Acknowledgments
Funding was provided by USEPA Agreement FY04 104(b)(3) X797682001-0, USGS Agreement 04HQAG0131, and the University
of Oklahoma Center for Restoration of Ecosystems and Watersheds.

References
Aderhold, D., C.J. Williams, and R.G. Edyvean. 1996. The removal of heavymetal ions by seaweeds and their derivatives. Bioresour. Technol. 58:16.
American Public Health Association (APHA). 2005. Standard methods for
the examination of water and wastewater. 21st ed. Am. Public Health
Assoc., Baltimore.
Carroll, S.A., P.A. ODay, and M. Piechowski. 1998. Rockwater interactions
controlling zinc, cadmium, and lead concentrations in surface waters and
sediments, U.S. Tri-State Mining District. 2. Geochemical interpretation. Environ. Sci. Technol. 32:956965.
Dean, A., and D. Voss. 1999. Design and analysis of experiments. Springer,
New York.
Denholm, C., T. Danehy, S. Busler, R. Dolence, and M. Dunn. 2008. Sustainable passive treatment of mine drainage: Demonstration of manganese
212

Hach. 2006. Digital Titrator Model 16900 manual. Hach, Loveland, CO.
Hedin, R.S. 2003. Recovery of marketable iron oxide from mine drainage.
Land Contam. Reclamat. 11:9397.
Johnson, K.L., and P.L. Younger. 2006. The co-treatment of sewage and mine
waters in aerobic wetlands. Eng. Geol. (Amsterdam) 85:5361.
Kirby, C.S., and C.A. Cravotta. 2005. Net alkalinity and net acidity 2: Practical considerations. Appl. Geochem. 20:19411964.
Lovley, D.R. 1991. Dissimilatory Fe(III) and Mn(IV) reduction. Microbiol.
Rev. 55:259287.
Luo, Q., T.K. Tsukamoto, K.L. Zamzow, and G.C. Miller. 2008. Arsenic, selenium, and sulfate removal using an ethanol-enhanced sulfate-reducing
bioreactor. Mine Water Environ. 27:100108.
Machemer, S.D., and T.R. Wildeman. 1992. Adsorption compared with sulde precipitation as metal removal processes from acid mine drainage in
a constructed wetland. J. Contam. Hydrol. 9:115131.
Mayes, W.M., H.A. Potter, and A.P. Jarvis. 2009. Novel approach to zinc removal from circum-neutral mine waters using pelletised recovered hydrous ferric oxide. J. Hazard. Mater. 162:512520.
McCullough, C.D., M.A. Lund, and J.M. May. 2008. Field-scale demonstration of the potential for sewage to remediate acidic mine waters. Mine
Water Environ. 27:3139.

Journal of Environmental Quality Volume 40 JanuaryFebruary 2011

Metcalf and Eddy, Inc. 1991. Wastewater engineering: Treatment, disposal and
reuse. 3rd ed. McGraw-Hill, New York.
Miller, J.C., and J.N. Miller. 1986. Statistics for analytical chemistry. Ellis
Horwood, Chichester, UK.
Mohan, D., and C.U. Pittman. 2007. Arsenic removal from water/wastewater
using adsorbents: A critical review. J. Hazard. Mater. 142:153.
Muga, H.E., and J.R. Mihelcic. 2008. Sustainability of wastewater treatment
technologies. J. Environ. Manage. 88:437447.
Munk, L., G. Faure, D.E. Pride, and J.M. Bigham. 2002. Sorption of trace
metals to an aluminum precipitate in a stream receiving acid rockdrainage; Snake River, Summit County, Colorado. Appl. Geochem.
17:421430.
Neculita, C., G.J. Zagury, and B. Bussire. 2007. Passive treatment of acid
mine drainage in bioreactors using sulfate-reducing bacteria: Critical review and research needs. J. Environ. Qual. 36:116.
Nelson, M., H.T. Odum, M.T. Brown, and A. Alling. 2001. Living o the
land: Resource eciency of wetland wastewater treatment. Adv. Space
Res. 27:15471556.
Omoike, A.I., and G.W. Vanloon. 1999. Removal of phosphorus and organic matter removal by alum during wastewater treatment. Water Res.
33:36173627.
Pruden, A., N. Messner, L. Pereyra, R.E. Hanson, S.R. Hiibel, and K.F. Reardon. 2007. The eect of inoculums on the performance of sulfate-reducing columns treating heavy metal contaminated water. Water Res.
41:904914.
Roetman, E.T. 1932. The sterilization of sewage by acid mine water. M.S.
thesis. West Virginia Univ., Morgantown.
Rose, P.D., G.A. Bosho, R.P. van Hille, L.C.M. Wallace, K.M. Dunn, and
J.R. Duncan. 1998. An integrated algal sulphate reducing high rate
ponding process for the treatment of acid mine drainage wastewaters.
Biodegradation 9:247257.
Saliling, W.J.B., P.W. Westerman, and T.M. Losordo. 2007. Wood chips and
wheat straw as alternative biolter media for denitrication reactors
treating aquaculture and other wastewaters with high nitrate concentrations. Aquacult. Eng. 37:222233.
Santomartino, S., and J.A. Webb. 2007. Estimating the longevity of limestone
drains in treating acid mine drainage containing high concentrations of
iron. Appl. Geochem. 22:23442361.
Strosnider, W.H., and R.W. Nairn. 2010. Eective passive treatment of highstrength acid mine drainage and raw municipal wastewater in Potos,

Bolivia using simple mutual incubations and limestone. J. Geochem.

Explor. 105:3442.
Strosnider, W.H., B.K. Winfrey, and R.W. Nairn. 2010a. Alkalinity generation in a novel multi-stage high-strength acid mine drainage and municipal wastewater passive co-treatment system. Mine Water Environ.
doi:10.1007/s10230-010-0124-2 (in press).
Strosnider, W.H., B.K. Winfrey, and R.W. Nairn. 2010b. Biochemical oxygen
demand and nutrient processing in a novel multi-stage raw municipal
wastewater and acid mine drainage passive co-treatment system. Water
Res. (in press).
Van Hille, R.P., G.A. Bosho, P.D. Rose, and J.R. Duncan. 1999. A continuous process for the biological treatment of heavy metal contaminated
acid mine water. Resour. Conserv. Recycl. 27:157167.
Viswanathan, S. 2005. Microbial analysis of a vertical-ow passive treatment
mesocosm. M.S. thesis. Univ. of Oklahoma, Norman.
Watzlaf, G.R., K.T. Schroeder, R.L. Kleinmann, C.L. Kairies, and R.W. Nairn.
2004. The passive treatment of coal mine drainage. Available at http://
www.netl.doe.gov/technologies/coalpower/ewr/water/pdfs/Passive%20
Treatment.pdf (veried 18 Oct. 2010). USDOE, Natl. Energy Technol.
Lab., Pittsburgh, PA.
Wilkin, R.T. 2008. Contaminant attenuation processes at mine sites. Mine
Water Environ. 27:251259.
Winfrey, B.K., W.H. Strosnider, R.W. Nairn, and K.A. Strevett. 2010. Highly
eective reduction of fecal indicator bacteria counts in an ecologicallyengineered municipal wastewater and acid mine drainage passive cotreatment system. Ecol. Eng. 36:16201626.
Wolkersdorfer, C., and R. Bowell (ed.) 2004a. Contemporary reviews of mine
water studies in Europe, Part 1. Mine Water Environ. 23:162182.
Wolkersdorfer, C., and R. Bowell (ed.) 2004b. Contemporary reviews of mine
water studies in Europe, Part 2. Mine Water Environ. 24:237.
Wolkersdorfer, C., and R. Bowell (ed.) 2004c. Contemporary reviews of mine
water studies in Europe, Part 3. Mine Water Environ. 24:5876.
Woomer, P.L. 1994. Most probable number counts. p. 5978. In R.W. Weaver
et al. (ed.) Methods of soil analysis. Part 2: Microbiological and biochemical properties. SSSA, Madison, WI.
Younger, P., S.A. Banwart, and R.S. Hedin. 2002. Mine water: Hydrology,
pollution, remediation. Kluwer Academic Publ., Boston.

Strosnider et al.: Passive Co-Treatment of Acid Mine Drainage and Municipal Wastewater

213

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.