Chemosphere 153 (2016) 244e253

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Biochemical passive reactors for treatment of acid mine drainage:

Effect of hydraulic retention time on changes in efciency,
composition of reactive mixture, and microbial activity
Yaneth Vasquez a, b, Maria C. Escobar b, Carmen M. Neculita c, Ziv Arbeli b, Fabio Roldan b, *
a
, Colombia
Departamento de Ciencias Naturales, Universidad Central, Cra. 5 No. 21-38, Bogota
b
Unidad de Saneamiento y Biotecnologa Ambiental (USBA), Departamento de Biologia, Ponticia Universidad Javeriana, Cra. 7 No. 40-62, Bogota
,
Colombia
c
Canada Research Chair on Passive Treatment of Contaminated Mine Water, Research Institute on Mines and Environment (RIME), University of Quebec in
Abitibi-Temiscamingue (UQAT), 445 Boulevard de l'Universite, Rouyn-Noranda, QC J9X 5E4, Canada

h i g h l i g h t s

Biochemical passive treatment represents a sustainable approach for treating acid mine drainage.

Hydraulic retention times had a strong inuence in the sulfate-reducing bacteria and dehydrogenase activity.

Long hydraulic retention time (4 d) generated higher alkalinity and excess residual suldes in the efuent.

The hydraulic retention time strongly inuenced performance of biochemical passive reactors.

a r t i c l e i n f o a b s t r a c t

Article history: Biochemical passive treatment represents a promising option for the remediation of acid mine drainage.
Received 4 August 2015 This study determined the effect of three hydraulic retention times (1, 2, and 4 days) on changes in
Received in revised form system efciency, reactive mixture, and microbial activity in bioreactors under upward ow conditions.
11 March 2016
Bioreactors were sacriced in the weeks 8, 17 and 36, and the reactive mixture was sampled at the
Accepted 12 March 2016
Available online 25 March 2016
bottom, middle, and top layers. Physicochemical analyses were performed on reactive mixture post-
treatment and correlated with sulfate-reducing bacteria and cellulolytic and dehydrogenase activity.
rg E. Drewes
Handling Editor: Jo All hydraulic retention times were efcient at increasing pH and alkalinity and removing sulfate (>60%)
and metals (85e99% for Fe2 and 70e100% for Zn2), except for Mn2. The longest hydraulic retention
Keywords: time (4 days) increased residual suldes, deteriorated the quality of treated efuent and negatively
Acid mine drainage (AMD) impacted sulfate-reducing bacteria. Shortest hydraulic retention time (1 day) washed out biomass and
Sulfate-reducing bacteria (SRB) increased input of dissolved oxygen in the reactors, leading to higher redox potential and decreasing
Metal removal metal removal efciency. Concentrations of iron, zinc and metal suldes were high in the bottom layer,
Efciency
especially with 2 day of hydraulic retention time. Sulfate-reducing bacteria, cellulolytic and dehydro-
Hydraulic retention time (HRT)
genase activity were higher in the middle layer at 4 days of hydraulic retention time. Hydraulic retention
Biochemical passive reactor (BPR)
time had a strong inuence on overall performance of passive reactors.
2016 Elsevier Ltd. All rights reserved.

1. Introduction from prolonged exposure of sulde ores to air and water

The major type of contaminated efuent generated by the
mining industry is acid mine drainage (AMD), which originates
* Corresponding author.
E-mail address: fabio.roldan@javeriana.edu.co (F. Roldan).
(Nordstrom et al., 2015). First, there is formation of sulfuric acid and
progressive acidication of water, which then dissolves metals in
the rock, resulting in low pH and high metal and sulfate loading.
These characteristics make AMD responsible for physical, chemical,
and biological degradation of aquatic habitats and surrounding soil
quality (Adams et al., 2014).

http://dx.doi.org/10.1016/j.chemosphere.2016.03.052
0045-6535/ 2016 Elsevier Ltd. All rights reserved.
 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253
A sustainable approach to remediate AMD-impacted water is
to use biochemical passive reactors (BPRs), which are columns
lled with mixtures of organic and inorganic substrates, referred
to as reactive mixture. BPRs have potential advantages, such as
low costs, ability to work in remote areas, use of recycled or waste
organic materials, and natural appearance (Adams et al., 2014).
Because diverse contaminants, specic environments, and locally
available substrates require appropriate design of BPRs, labora-
tory testing is still necessary (Gusek, 2008). This is usually per-
formed through testing in columns to determine the effectiveness
of the treatment and select essential design parameters,
including hydraulic retention time (HRT) and the composition of
reactive mixtures (Adams et al., 2014). The HRT has a direct in-
uence on the performance of BPRs (Younger et al., 2002;
Neculita et al., 2008a). Flow velocity, which determines the
HRT, plays an important role in the overall activity of microor-
ganisms in the reactors. Very short HRT may not allow adequate
time for bacteria activity to neutralize acidity and precipitate
metals or may result in biomass being washed out of the reactor
(Neculita et al., 2007). Very long HRT may deplete the substrate or
built up bacterial growth and mineral precipitates that eventually
lead to low hydraulic conductivity and short-circuiting of the
treatment system (Younger et al., 2002).
To date, the effect of the HRT on the efciency of BPRs has been
studied with a focus on metal-removing efciencies, mainly in
iron-rich AMD (Chang et al., 2000; Neculita et al., 2008a). Perfor-
mance of these systems in the treatment of AMD, characterized by
high sulfate concentrations with low metal loading, has not yet
been addressed.
The key parameter in an efcient BPR is the activity of sulfate-
reducing bacteria (SRB), because they reduce sulfates to suldes,
which then react with dissolved heavy metals to form solid metal
suldes that are immobilized in the reactor. Although SRB meta-
bolism conducts the sulfate reduction reaction, these bacteria rely
on activity of anaerobic cellulolytic and fermentative bacteria to
break down complex organic materials, such as cellulose, to pro-
vide them with available carbon and energy sources (Pruden et al.,
2007). Therefore, the cellulose and dehydrogenase enzymes can be
used as indicators of general microbial activity in anaerobic systems
(Serra-Wittling et al., 1995). However, there are limited data
available related to the changes in microbial activity under different
HRTs in the long-term operation of BPR, under laboratory
conditions.
The composition of reactive mixture changes during the life of
BPRs by precipitation of oxy-hydroxide, carbonate, and sulde
minerals (Neculita et al., 2007). Metals extracted during the acid-
volatile sulde (AVS) process are known as simultaneously
extracted metals (SEM) and the molar ratios of both (i.e. SEM/AVS)
indicate which are the predominant metal removal mechanisms.
When this ratio is  1.0, it indicates that metals are mainly present
as solids in the form of sulde minerals, whereas at ratio > 1.0, the
oxy-hydroxide and carbonate minerals are predominant (Di Toro
et al., 2005). Although the SEM/AVS molar ratio can provide valu-
able information on reactor efciency and metal removal mecha-
nisms in AMD, limited data is available about the inuence of HRT
on the potential mobility of metals in the reactive mixture evalu-
ated after sacricing the reactor (post-treatment) (Neculita et al.,
2008b).
The main objective of this study was to measure the effect
(temporal and spatial) of HRT (1, 2 and 4 days) on the efciency of
BPRs for the treatment of sulfate-rich AMD. In addition, the cor-
relation between microbial activity (cellulolytic and dehydroge-
nase) and physicochemical characteristics of the reactive mixture
after treatments was determined.
245
2. Materials and methods
2.1. Passive reactors: set-up, operation, and sampling
Seven 5 L laboratory reactors were constructed using acrylic
columns (73  10 cm) and acrylic caps at both ends. The bottom cap
contained a valve used as an inuent port for AMD, and the top cap
contained a double valve: one for collecting treated AMD and
another for instantaneous sampling of efuent. The columns were
packed, using the procedure described by Neculita et al. (2008a)
and Song et al. (2012). Briey, each column was lled with the
same reactive mixture, consisting of 15% cow manure, 10% mush-
room compost, 25% sajo sawdust, 15% gravel, 20% limestone, and
15% wetland sediment as inoculum (dry weight in % w/w); this was
slightly compacted in layers of 10 cm. This reactive mixture (com-
ponents and proportions) was selected because was found highly
efcient for treating synthetic AMD in batch testing (Vasquez et al.,
2016). The reactive mixture was sandwiched in the reactor between
two porous plates coated with a ne geotextile (25 mesh), with a
7 cm layer of gravel (0.6e1.0 cm) at the bottom inuent port and a
3 cm layer of gravel at the top efuent port to prevent loss of
organic matter and clogging. Finally, the columns were sealed and
saturated with synthetic AMD prepared according to the maximum
concentrations for metals and sulfate found in real efuent quality
data found at ve active mine sites in the Zipaquira
Mining District,
Colombia (Vasquez et al., 2016). The synthetic AMD, which was
prepared weekly (Zagury et al., 2006), had the following compo-
sition (in mg L1): 201 44 Fe2; 30 2 Mn2; 19 2 Zn2;
215 11 Ca2; 128 13 Mg2 and 2500 105 SO2 4 , at pH 3.0e3.7.
After the packing procedure, the columns were wrapped in
aluminum foil to exclude light and prevent growth of phototrophic
bacteria (Waybrant et al., 2002; Neculita et al., 2008a). The columns
remained at room temperature in the dark for two weeks. During
this acclimation period, 20 mL samples were collected twice
weekly, from the top port and tested until favorable conditions for
SRB (pH 5e8, ORP < 100 mV) prevailed (Postgate, 1984).
Following acclimation, continuous upward ow of synthetic AMD
was started, with three columns at a ow rate of 0.74 mL min1 (for
2-day HRT), and four columns at a ow rate of 0.37 mL min1 (for 4-
day HRT). Continuous ow was maintained for 36 weeks, using
precision peristaltic pumps (Masterex L/S, Cole-Parmer, Vernon
Hills, IL), which were calibrated prior to starting the tests. After 17
weeks, a strong increase of soluble sulde concentrations in treated
efuents from the columns with 4-day HRT was recorded. These
results justied the decision to change one of the reactors from 4-
day HRT to 1-day HRT, at a ow rate of 1.48 mL min1. Columns
were sacriced throughout the study to monitor the changes in the
post-treatment reactive mixture and microbial activity. Four col-
umns, two of 2-day HRT and two of 4-day HRT, were sacriced at
weeks 8 and 17. The other three columns (one each of 1, 2, and 4-
day HRT) were analyzed at the end of the study (36 weeks). The
reactive mixture from the sacriced columns was removed in three
cores (20  10 cm), one from the bottom (at 0e20 cm), one from the
middle (at 20e40 cm), and one from the top (at 40e60 cm). Each
sample was homogenized and refrigerated at 4  C until analyzed.
2.2. Effect of the hydraulic retention time on efciency of
biochemical passive reactors
To determine the effects of HRT on the efciency of reactors for
increasing pH and alkalinity, as well as promoting reduction of
sulfates and metals during treatment of AMD, 50 mL efuent was
collected with a syringe each week from the valve located at the top
cap of each column. The pH and redox potential (ORP) were
immediately measured with a multi-parameter probe (HI 9828,
246 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253
Hanna Instruments; Woonsocket, RI). Alkalinity was also deter-
mined (TitroLine alpha plus 20, Schott, Mainz, Germany) by using
method 2320B (APHA, 2005). The remaining sample was ltered
(0.45 mm, Sartorius, Goettingen, Germany) before sulfate analysis
by UVeVIS spectrophotometry (Genesys 10, Thermo Scientic;
Waltham, MA) using method 4500eSO4 (APHA, 2005) and sulde
analysis using the methylene blue method (4500 D; APHA, 2005).
Dissolved metals (Fe2, Mn2 and Zn2) were quantied by atomic
absorption spectrometry (Varian 240 FS; Agilent Technologies;
Santa Clara, CA), using method 7000B (USEPA, 2001), with a
detection limit ranging from 0.03 to 1.0 mg L1. Once a month,
dissolved organic carbon (DOC) was analyzed, using high temper-
ature combustion in method 5310B (APHA, 2005), in an external
accredited laboratory.
To identify the potential precipitation mechanisms responsible
for changes in metal concentrations, saturation indices (SI) were
calculated, using the geochemical modeling software Visual MIN-
TEQ 3.0 (Gustafsson, 2011) on the treated efuent collected from
some representative phases of treated water. This modeling soft-
ware considers solution chemistry and equilibrium distribution of
chemical species according to geochemical processes, such as
dissolution/precipitation, oxidation/reduction, complexation, and
ion exchange. However, VMINTEQ does not consider SRB activity
and precipitation of biogenic suldes (Waybrant et al., 2002;
Zagury et al., 2006).
SRB counts were determined once a month, using the most
probable number technique, method D4412e84 (ASTM, 2009).
Formation of black precipitates (after 25 of days inoculation) was
interpreted as generation of iron sulde with SRB activity. When
the black color was not obvious, the presence of suldes was
conrmed by a blue color after adding a few drops of 0.005 M ferric
chloride hexahydrate and 0.005 M N-N-dimethyl-p-phenylenedi-
amine dihydrochloride.
2.3. Effect of the hydraulic retention time on changes in
composition of the reactive mixture
Samples of reactive mixtures (n 21) were collected from the
several layers in the bioreactors sacriced at weeks 8, 17, and 36.
Physicochemical analyses were carried out in triplicate. Easily
available substances (EAS), organic nitrogen and total organic car-
bon (TOC) were tested on air-dried and sieved (mesh No 70,
0.210 mm) samples. The pH was measured (Lab 870, Schott; Mainz,
Germany) using method 4972e01 (ASTM, 1995a) and a ratio of
solid to de-ionized water of 1:1. Moisture was determined by dif-
ference in weight between wet samples and samples dried at 105  C
for 20 h (Method D 2216e98; ASTM, 1995b). EAS were analyzed by
forage ber analysis, according to Prasad et al. (1999). The organic
nitrogen was measured by Method 4500eNorg (APHA, 2005), and
cellulose content was determined, according to Harper and Lynch
(1981). The organic carbon was analyzed by the WalkleyeBlack
method (Schumacher, 2002). To ensure complete digestion of
extraction mixtures, 5 mL 1 N K2Cr2O7, 10 mL H2SO4 (Analytical-
grade), and 0.2 g reactive mixture were gently boiled at 140  C for
30 min, cooled overnight, ltered (0.45 mm; Sartorius). Cr3 was
quantied by UVeVIS spectrophotometry (Genesys 10, Thermo
Scientic; Waltham, MA).
Total metal concentrations immobilized in the reactive mixtures
were determined by the digestion method described by Neculita
et al. (2008b), where 0.5 g samples were weighed in 50 mL poly-
propylene centrifuge tubes, 10 mL concentrated HCl was added to
react for 24 h at room temperature. Then, 15 mL HNO3 was added
and the tubes were heated in a water bath at 75 3  C for 50 min,
followed by adding 10 mL HClO4 and 1 mL HF (analytical-grade).
The samples were heated in a water bath at 75 3  C for 2.5 h. After
cooling to room temperature, the supernatant was diluted to 50 mL
with de-ionized water and ltered (0.45 mm; Sartorius). Dissolved
metals (Fe2, Mn2, and Zn2) were then quantied by atomic
absorption spectrometry. Removal mechanisms for metals were
assessed with the SEM/AVS procedure, as modied by Neculita et al.
(2008b).
2.4. Effect of hydraulic retention time on changes in microbial
activity
Dehydrogenase and cellulolytic activity were measured, ac-
cording to Serra-Wittling et al. (1995). Microbial activity was tested
in triplicate with 1.0 g reactive mixture that was air-dried for 24 h
and sieved (mesh No. 50, 0.297 mm). For dehydrogenase analysis,
samples were incubated in darkness for 24 h with 4 mL 1% 2,3,5-
triphenyl-tetrazolium chloride at 30  C, using this chemical as the
electron acceptor and reduced to triphenyltetrazolium formazan
(TPF; displayed as red color), which was extracted with methanol
(10 mL) and quantied by UVeVIS spectrophotometry. For
measuring cellulolytic activity, the sample was incubated at 30  C
for 24 h with 3.5 mL 0.2 M acetate buffer and 3.5 mL 1% carbox-
ymethyl cellulose as the substrate. The Antrone method was used
to quantify total soluble carbohydrates (Dische, 1962).
2.5. Statistical analysis
Statistical analysis were based on SPSS Statistics 19 (IBM,
Armonk, NY), using Pearson correlation between physicochemical
variables and/or microbial activity.
3. Results and discussion
3.1. Effect of hydraulic retention time on changes in efciency
During the initial acclimation period, pH increased from
3.0 0.8 to 7.0 0.5 in efuent from all reactors, regardless of the
HRT. This increase was associated with increased alkalinity,
1500 54 mg CaCO3 L1 for 2-day HRTs and 2400 32 mg
CaCO3 L1 at 4-day HRTs (Fig. 1). Increased alkalinity mostly
resulted from the dissolution of limestone in the reactive mixture
and occurred within the rst several hours. After the beginning of
continuous-ow operations, the pH remained relatively constant,
in the range of 6.5e7.0, and alkalinity decreased to 800 28 mg
CaCO3 L1 for 2-day HRT and 1200 60 mg CaCO3 L1 for 4-day
HRT. Thereafter, alkalinity was relatively constant at both HRTs
until there was a strong increase in the 4-day HRT starting at 21
week until the end of the study (Fig. 1). In previous studies, higher
alkalinity in the reactor was associated with formation of carbonate
species through dissolution of limestone from reactive mixtures or
oxidation of readily available organic substrates by SRB, as well as
from dissolution of surface-bound hydroxyl ions in the substrate
materials under acidic conditions (Neculita et al., 2008a; Robinson-
Lora and Brennan, 2011; Song et al., 2012).
The reactors developed reducing conditions during the accli-
mation period (Fig. 1). The ORP initially decreased to 260 mV and
declined to 360 mV by the fourth week. Afterwards, the ORP
remained relatively constant, without signicant difference be-
tween the 2- and 4-day HRTs, until the end of the study. These
results indicate that the acclimation period was necessary and
important for establishing an appropriate geochemical environ-
ment for anaerobic bacteria, including the SRB. Then, during the
continuous ow, higher pH and alkalinity together with lower ORP
led to a more rapid development of SRB, favoring the consumption
of readily available organic substrates, especially at longer HRT.
Sulfate concentration in synthetic AMD, used to saturate the
 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253
columns during the set-up, was 2500 170 mg L1; however two
weeks later, the sulfate concentration in efuents increased to as
much as 2800 32 mg L1 for 2-day and 3382 31 mg L1 for 4-
day HRT (Fig. 2). This was caused by leaching from the organic
substrate (cow manure and mushroom compost) in the reactive
mixture, as observed during batch testing (Vasquez et al., 2016).
Release of sulfate (27.5 g kg1) from organic substrates, especially
manures and compost, was previously reported (Song et al., 2012).
After the beginning of continuous-ow, sulfate declined to
1300 32 mg/L for 2-day and 1000 50 mg/L for 4-day HRT, and
remained relatively constant in both treatments until week 16. In
following weeks, the sulfate increased to 1700 98 mg/L, probably
due to the reduction of SRB counts caused by increased soluble
suldes until 1952 52 mg H2S L1 for 2-day and 2826 185 mg
H2S L1 for 4-day HRT. Between weeks 23 and 36, sulfate removal
increased, higher in the 4-day HRT columns. There was no differ-
ence between 1-day and 2-day HRT.
The concentration of DOC at the beginning of continuous-ow
operations was 740 mg L1, but decreased to ~10 mg L1 by week
8 and remained nearly constant until the end of the study. This
pattern was similar for the three HRTs tested (Fig. 2). The reactive
substrate released DOC at a nearly constant rate, which was avail-
able for the microorganisms in the reactors to reduce sulfates.
These results are consistent with Song et al. (2012) who observed
that substrates with high initial DOC is likely to leach a considerable
amount of DOC to the efuent, especially at low pH condition,
which is not available for microbial growth.
Suldes were detected by week 4, with signicant differences
between the 2-day and 4-day HRTs (Fig. 2). As expected, the
Fig. 1. Changes in pH, oxide reduction potential (ORP), and alkalinity during treatment of acid mine drainage (AMD) in biochemical passive bioreactors (BPR) over 36 weeks. In week
17, one column of 4 days hydraulic retention time (HRT) treatment was changed to 1 day HRT treatment.
247
concentration of suldes was usually higher in the reactors under
4-day HRT because longer residence time allowed greater oxidation
of available organic carbon and reduction of sulfate by SRB in the
efuents. Sulde concentrations signicantly increased at the 2-
day HRT (1952 52 mg H2S L1) and 4-day HRT (2826 185 mg
H2S L1) after week 16. The increase in concentrations of suldes is
a response to the excessive sulfate added in AMD in the week 15
and the growth of SRB that removed the sulfate leached from the
organic substrates. This eventually created a sharp decline in SRB
counts between weeks 12 and 20 (Fig. 2). Sa
ez-Navarrete et al.
(2009) found that soluble suldes (H2S or HSand S2) reduce
cell size and growth, when present in concentrations > 108 mg
H2S L1. Okabe et al. (2003) found that cell production fell slightly
at a concentration of 100 mg H2S L1, whereas at 180 mg H2S L1,
the rate of growth declined sharply. Concentrations of sulde
decreased in 2-day HRT reactors (357 12 mg H2S L1) and in 4-day
HRT reactors (1822 67 mg H2S L1) after week 24. These changes
favored growth of the SRB in 2-day HRT reactors. Suldes in re-
actors with 4-day HRT remained high (1066 19 mg H2S L1),
whereas in reactors with 2-day HRT were reduced to 218 37 mg
H2S L1, at the end of the study. With these results, after week 17,
one of the 4-day HRT reactors was changed to a 1-day HRT reactor.
As expected, after the change, alkalinity (Fig. 1) decreased from 767
to 510 mg CaCO3 L1 and remained relatively constant until week
36. This change did not affect pH. However, there was an effect on
the ORP, which increased rapidly from 375 to 108 mV from the
dissolved oxygen (DO) in the AMD. The new ow released a large
amount of sulfate (3115 mg L1) (Fig. 2), probably retained in the
reactive mixture as gypsum (CaSO4$2H2O). Another source of
248 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253

Fig. 2. Changes in dissolved organic carbon (DOC), sulfate, suldes, and sulfate-reducing bacteria (SRB) during the treatment of acid mine drainage (AMD) in biochemical passive
bioreactors (BPR) over 36 weeks. In week 17, one column of 4 days hydraulic retention time (HRT) treatment was changed to 1 day HRT.

residual sulfate may have come from re-oxidation of suldes that
were previously precipitated in the reactor (Dvorak et al., 1992).
From weeks 26 through 33, the reactor stabilized and ORP
decreased from 189 mV to 270 mV. In the last three weeks of the
experiment, the ORP increased; this affected the efciency of sul-
fate reduction. Excessive production of suldes declined at week 20
and decreased to 23 4 mg H2S L1 at week 36 (Fig. 2). The increase
in ow and the consequent increased loading rate of oxygen
affected anaerobic conditions and subsequent capacity of SRB to
produce soluble suldes. Increased ow also washed out the SRB in
the reactive mixture (Fig. 2), which led to clogging the efuent
tubing with leached biomass and metal sulde precipitates. Once
the tubing was replaced, outow was slowly re-established.
Removal of Fe2 and Zn2 in the AMD was observed during the
36 weeks of operation of the reactors, without signicant differ-
ences between 2- and 4-day HRTs (Fig. 3). During the start-up and
acclimation, metals could be removed by precipitation in the form
of hydroxides and carbonates, as well as by sorption on fresh
organic substrate, with respect to the pH of pore water. Then,
during continuous-ow operation, at steady-state condition, metal
removal might be partially explained by the formation of sulde
minerals (Zhang and Wang, 2014). Removing metal is generally
closely related to the ORP. In the reactor changed to 1-day HRT,
efciency of removing Fe2 and Zn2 decreased (75% and 67%,
respectively) when ORP increased from 360 to 108 mV at week
23. The major contributor to ORP is DO and other redox active
species, such as sulfate and Fe2. The reactions that remove metals
in reactors are highly sensitive to redox conditions and are
reversible (Cheong et al., 2010).
During the acclimation period, Mn2 could be removed by
sorption on fresh organic substrate and precipitation as carbonate
(MnCO3 e rhodochrosite). Lewis (2010) found that adsorption on
organic substrates is the predominant mechanism that provides
some Mn2 removal at this phase. After the start of continuous-
ow operations, Mn2 in efuent was 77 4 mg L1 for 4-day
HRT reactors and 60 2 mg L1 for 2-day HRT reactors, which
exceeded levels in the synthetic AMD (31 mg L1). By week 21,
Mn2 concentrations decreased again (efciencies up to 60% were
measured), without signicant differences between the three
retention times. Removal of Mn2 is tied to increasing alkalinity, a
condition present in all reactors. Relatively high alkalinity is usually
necessary to ensure appropriate abiotic oxidation rates of Mn at
pH < 8 (Mariner et al., 2008). Low removal of Mn by bioreactors
results from high solubility of MnS relative to other metal suldes
(Cheong et al., 2010), and the inhibition of its precipitation as hy-
droxides if the concentrations of Fe2 or Zn2 are high (Yoo et al.,
2004). Mn2 has been acknowledged as the most challenging
metal to remove in bioreactors under reducing conditions (Chang
et al., 2000; Neculita et al., 2007; Robinson-Lora and Brennan,
2011). In the present study, after removal of other metals in AMD,
 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253
Fig. 3. Metals removal (Fe2, Mn2, and Zn2) during the treatment of acid mine drainage (AMD) in biochemical passive bioreactors (BPR) over 36 weeks. In week 17, one column of
4 days hydraulic retention time (HRT) treatment was changed to 1 day HRT.
Mn removal as carbonates was probably favored by the presence of
high alkalinity; this coincides with the results from Visual MINTEQ
modeling which did not indicate the formation of MnS.
The SI of the potential minerals in the reactors was calculated by
VMINTEQ modeling, using some representative phases of water
quality. Sphalerite [(Zn, Fe)S] super saturation (SI > 1) over time
was the highest in the column with 1-day HRT. The results also
indicated that Zn concentrations could be controlled by the solu-
bility of a zinc sulde phase. The SI of calcite (CaCO3) indicated that
this mineral could be higher at the beginning of continuous ow,
when alkalinity increased. Mackinawite (FeS) and siderite (FeCO3)
were probably present in the efuents of the reactors with 1- and
4-day HRTs. At the end of the study, rhodochrosite supersaturation
probably occurred in reactors with 1-day HRT.
HRT strongly affected changes in the efciency of reactors
treating sulfate-rich AMD. 4-day HRT, with low up ow velocity,
favored growth and activity of SRB. As a result, high alkalinity and
excess sulde concentrations could leave the reactor, leading to
toxicity of the efuent. However, excessive sulde content from
biological sulfate reduction can result in physical and biological
inhibition of SRB and eventually contributes to failure of the
reactor. 1-day HRT more rapidly washes out SRB and introduces
more DO in the reactor, leading to increased ORP. Anaerobic con-
ditions are disrupted, which adversely affects SRB activity. As a
result, increased ORP reduced removal of metal by up to 30%.
249
3.2. Effect of hydraulic retention time on changes in composition of
the reactive mixture
The analysis of the reactive mixture at weeks 8 and 17 showed
pH > 8.0, with similar trends in all layers (Table 1). At week 36, pH
decreased, especially in the bottom layers (<6.3), probably from
depletion of neutralizing agents caused by constant AMD input and
upward ow in the reactors. The decline in pH in the reactors did
not affect activity of the SRB, which requires a pH 5.0e8.0 range for
optimal growth (Postgate, 1984).
The biodegradable portion of the substrate, composed of soluble
sugars, amino acids, starch, certain proteins, and hemicellulose
(Zagury et al., 2006) rapidly decreased in the rst 8 weeks (Table 1)
because it is the soluble fraction that is readily used by microor-
ganisms. The EAS increased at weeks 17 and 36, and showed a
negative correlation with cellulose content (0.672) and organic
nitrogen (0.754;p < 0.01). The nitrogen was more greatly reduced in
reactors under 4-day HRT, which may be related to the higher count
of SRB in the reactive mixture after treatment at week 36. The
concentrations of organic nitrogen were positively correlated with
total organic carbon (0.764; p < 0.01), increasing the C:N ratio.
Available carbon and nitrogen are among the very important fac-
tors affecting the treatment of AMD with SRB (Waybrant et al.,
2002; Zagury et al., 2006).
Metal (Fe and Zn) in the reactive mixtures were concentrated in
the bottom layer of all reactors, especially in reactors with 2-day
250 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253

Table 1
Physicochemical and microbiological characterization of the reactive mixture post-treatment in biochemical passive reactors.

Week HRT (d) Sample location pH % w/w (dw) SRB  103 (cells/100 mL)

EAS TKN COT Cellulose

Reactive mixture 6.3 0.1 51.8 0.6 7.1 0.1 60.8 1.3 60.9 0.25 7.00
2 Top 8.1 0.6 27.0 0.5 6.0 0.2 34.5 2.1 55.8 2.5 6.0
Middle 8.3 0.1 30.0 0.1 6.6 0.1 33.8 1.6 52.0 0.8 20.0
8 Bottom 8.1 0.4 23.1 0.1 6.0 0.1 32.7 0.8 50.7 1.4 89.1
4 Top 8.5 0.2 30.8 1.2 5.3 0.5 35.5 1.0 56.0 0.2 8.3
Middle 8.5 0.1 26.3 0.5 5.6 0.2 37.1 1.1 51.1 0.4 14.5
Bottom 8.3 0.1 28.3 0.6 5.4 0.2 35.4 1.6 59.4 0.6 122.0
2 Top 7.5 0.5 62.6 5.6 5.6 0.8 31.3 1.2 43.5 0.3 2.0
Middle 7.3 0.9 64.2 0.4 5.3 0.2 29.9 2.1 45.8 1.6 3.7
17
Bottom 7.5 0.2 63.7 4.0 5.1 0.2 30.2 1.0 42.0 1.8 1.9
4 Top 7.5 0.2 58.6 2.1 4.3 0.1 22.8 1.7 41.4 3.6 2.2
Middle 8.0 0.1 55.1 2.0 4.6 0.2 25.7 0.4 43.2 0.5 3.5
Bottom 8.0 0.1 60.0 0.6 4.3 0.1 23.3 1.9 40.2 1.5 2.5
2 Top 7.2 0.7 70.3 3.8 3.1 1.0 27.7 2.7 33.2 3.4 13.4
Middle 7.1 0.3 76.0 1.8 3.5 0.2 22.2 1.0 34.8 2.6 17.5
Bottom 6.3 0.1 70.4 4.2 3.9 0.2 22.4 4.1 29.4 2.1 13.7
36 4 Top 7.4 0.3 65.0 1.2 2.7 0.1 15.3 2.1 36.4 2.4 182.2
Middle 7.4 0.2 61.5 1.0 2.6 0.2 17.3 0.5 31.1 3.4 350.0
Bottom 6.5 0.1 64.5 1.9 2.1 0.1 16.3 0.6 26.5 1.9 99.2
1 Top 7.1 0.3 61.1 1.2 3.4 0.1 22.6 2.5 31.6 0.4 15.3
Middle 6.5 0.9 60.2 1.5 3.4 0.1 25.0 1.6 34.8 0.9 93
Bottom 5.2 0.3 62.6 1.3 3.5 0.3 18.9 3.9 19.2 1.4 109.0

HRT: hydraulic retention time; TKN: Total organic nitrogen; SRB: sulfate-reducing bacteria; EAS: easily available substances; TOC: total organic carbon.

HRT because this rate of ow treated more AMD in less time 3.3. Effect of hydraulic retention time on changes in microbial
(Table 2). Formation of black precipitate and accumulation of iron activity
oxy-hydroxide minerals with yellow-brown color were present in
the bottom layers. Similar results were found in reactive mixtures SRB counts, based on the layer in the reactors, indicated that the
after the operation period and dismantling of reactors (Zaluski middle layer of reactors at 4-day HRTs had as much as twice the
et al., 2003). High concentrations of Fe and Zn in reactive mix- level of 2-day HRT reactors. The maximum number of cultivable
tures after treatment indicated that the reactors were efcient in SRB after treatment in the reactive mixtures was 350  103 cells
removing these metals from the inuent AMD. 100 mL1 in the middle layer in the 4-day HRTs at 36 weeks
The concentration of Mn was similar in all layers. As already (Table 1). Appropriate conditions (ORP, sulfate concentrations,
discussed, this metal was removed only at the beginning of the organic carbon and pH > 5.0) in the 4-day HRT reactors favored
treatment, when it was probably adsorbed on the reactive mixture growth of bacteria, whereas metal suldes accumulated in the 2-
or precipitated as rhodochrosite. After exhaustion of the adsorption day HRT reactors adversely impacted SRB growth (Table 1). Metal
capacity of the reactive mixture, no additional removal of Mn was suldes concentrated in the vicinity of the bacterial cells and hinder
recorded, whereas Fe and Zn were still being removed, probably further metabolism by preventing contact between the reactants
partially as biogenic metal suldes. Sulfate concentrations (sulfate, organic matter) and necessary enzymes (Utgikar et al.,
increased in all reactive mixtures, with higher concentrations in the
bottom layers. Positive correlations with Fe (0.421), Zn (0.750), and
Mn (0.381) were determined (p < 0.01) and negative correlation Table 2
Total metal in reactive mixture post-treatment in biochemical passive reactors.
with pH (0.773; p < 0.01) is explained by AMD and upward ow in
the reactors. Week HRT (d) Sample location Total metal (mg/kg dw)
The interpretation of results from SEM/AVS molar ratios > 1.0 Fe Mn Zn
showed that precipitation as oxy-hydroxide and carbonate min-
2 Top 7.3 0.9 0.5 0.0 0.3 0.0
erals represents the predominant mechanism of removing metals Middle 12.0 2.8 0.4 0.1 0.3 0.0
at weeks 8 and 17 (Table 3). The SEM/AVS molar ratio was higher at 8 Bottom 29.0 2.5 0.1 0.0 0.7 0.1
week 8, when the SRB had not been completely established. Over 4 Top 7.4 0.4 0.6 0.0 0.2 0.0
the next several weeks, the molar ratio decreased, probably as Middle 15.7 1.5 0.5 0.0 0.2 0.0
Bottom 26.0 3.1 0.5 0.0 0.4 0.0
removal of metals as suldes started. Results from the SEM/AVS 2 Top 25.4 1.2 0.9 0.1 7.6 0.0
procedure also showed higher mobility of metals under acidic Middle 33.5 0.4 0.6 0.0 9.0 0.1
17
conditions. In the bottom layer of the reactor under 1-day HRT Bottom 85.7 2.0 0.6 0.1 22.5 0.0
treatment, the pH decreased to 5.2, causing movement of metals 4 Top 11.2 1.7 0.8 0.0 7.4 0.0
Middle 15.5 0.7 0.2 0.1 8.0 0.3
toward the middle layer and reduction of the SEM/AVS molar ratio.
Bottom 51.5 0.1 0.7 0.0 16.2 0.3
Statistical analysis showed a strong positive correlation between 2 Top 31.1 6.2 0.7 0.0 26.1 0.3
the SEM/AVS molar ratio and sulfates, Fe, and Zn (0.556, 0.682, and Middle 43.1 2.8 0.5 0.0 34.3 1.7
0.631, respectively; p < 0.01) over 36 weeks, suggesting that sul- Bottom 87.0 3.0 0.2 0.1 64.6 0.2
des produced from sulfates by reduction following SRB activity is 36 4 Top 32.5 0.0 0.6 0.0 20.1 0.6
Middle 38.0 1.0 0.9 0.1 19.7 2.1
an important mechanism for removing metals. The TOC was Bottom 61.1 2.8 0.3 0.1 33.3 1.7
negatively correlated with the SEM/AVS molar ratio (0.792; 1 Top 36.2 5.5 0.4 0.1 30.1 1.2
p < 0.01) as a major factor controlling adsorption of metal (metal- Middle 61.7 5.2 0.4 0.0 27.0 0.1
binding phases; Di Toro et al., 2005). Bottom 40.3 1.0 0.6 0.0 21.1 2.5
 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253 251

Table 3
The acid-volatile sulde (AVS), simultaneously extracted metals (SEM), and SEM/AVS molar ratios in reactive mixture post-treatment in biochemical passive reactors.

Week HRT (d) Sample location SEM (mg/kg, dw) AVS SEM/AVS (molar)

Fe Mn Zn (mg/kg, dw)

2 Top 0.1 0.0 0.1 0.0 0.2 0.0 0.2 0.0 4.8 0.1
Middle 0.1 0.0 0.1 0.0 0.1 0.0 0.2 0.0 4.8 0.1
8
Bottom 26.1 4.0 0.2 0.0 0.4 0.2 1.3 0.1 12 1.8
4 Top 0.7 0.1 0.2 0.0 0.1 0.0 0.2 0.0 4.8 0.1
Middle 0.8 0.1 0.2 0.0 0.1 0.0 0.2 0.0 4.8 0.1
Bottom 23.5 3.0 0.1 0.0 0.2 0.0 4.1 0.5 3.4 0.1
2 Top 6.7 1.3 0.1 0.0 6.6 0.0 3.7 0.1 2.8 0.3
Middle 14.1 0.7 0.6 0.0 7.0 0.1 3.9 0.1 3.1 0.2
17 Bottom 38.6 10.3 0.7 0.4 12.5 0.0 6.9 0.1 8.6 0.2
4 Top 5.0 0.8 0.9 0.0 7.4 0.0 1.9 0.3 3.0 0.0
Middle 5.7 2.2 0.6 0.0 7.0 0.3 3.7 0.3 2.4 0.6
Bottom 42.4 2.8 0.9 0.0 12.7 0.3 9.3 0.6 4.5 1.0
2 Top 6.3 2.0 0.4 0.0 27.7 0.2 5.4 0.1 1.3 0.2
Middle 35.2 0.4 0.5 0.0 31.8 0.4 5.7 0.3 3.3 0.9
Bottom 51.9 1.7 0.1 0.1 34.4 0.0 15.0 0.9 6.2 1.2
36 4 Top 1.6 0.7 0.6 0.0 20.0 0.0 5.8 0.3 0.7 0.1
Middle 5.9 0.1 0.5 0.1 20.1 0.3 5.0 0.3 2.2 0.0
Bottom 16.4 3.7 0.2 0.0 25.0 3.4 17.1 0.2 3.2 0.1
1 Top 13.9 1.9 0.3 0.0 31.3 0.2 3.2 0.5 3.4 0.1
Middle 32.4 1.0 0.3 0.0 25.2 0.0 3.5 0.0 6.3 0.0
Bottom 20.6 0.5 0.7 0.0 20.1 0.5 5.2 1.1 3.3 1.6

2002). Despite lower counts of SRB, microbial communities in the
waste organic materials are diverse and this is advantageous in
terms of resilience to stress (Hiibel et al., 2011).
At weeks 8 and 36, SRB counts were consistent with dehydro-
genase activity. The 4-day HRT reactors had higher activity
compared to 2-day HRT reactors (Fig 4). Dehydrogenase activity
was signicantly higher (p < 0.05) in the bottom layer of the 4-day
HRT reactors at week 8. At week 17, the second sacrice, there were
no signicant differences at any ow rate. At week 36, dehydro-
genase activity was greater in the middle layer of the 1-day HRT
reactor. The changes in dehydrogenase activity depended of pH and
concentration of metal suldes in the reactive mixture determined
after treatment. Concentration of Zn and dehydrogenase activity
were negatively correlated in the bottom layer of all reactors at
week 36 (0.332; p < 0.05). This metal inhibits enzymes by che-
lation with sulfhydryl groups, which are primarily involved at the
catalytically active sites of the enzymes (Mikanova, 2006). Enzyme
activity may decrease because of physiological adaptations of bac-
teria under stress from heavy metals (Hu et al., 2014). A positive
correlation (0.30; p < 0.05) occurred between dehydrogenase ac-
tivity and pH in the bottom layer of the 1-day HRT reactor at the end
of the treatments (Fig. 4). Passive reactors must maintain a pH
within the optimal range for hydrolytic fermenters, acetogenic
syntrophic bacteria, and SRB (5.0e8.0). Lower pH can cause
cessation of metabolic activity and result in an accumulation of
organic acids and potential process failure (Neculita et al., 2007).
Interpretation of dehydrogenase activity is difcult because mi-
crobes are affected by many factors, including pH and chemical
composition of the reactive mixture (Serra-Wittling et al., 1995;
Chu et al., 2007).
Cellulolytic bacteria are very important for the overall perfor-
mance of anaerobic passive treatments because they provide labile
organic carbon to the SRB (McCauley et al., 2009). Cellulolytic ac-
tivity was not signicantly different (p < 0.05) among the three
retention times, but signicant differences occurred at weeks 8, 17,
and 36. During the acclimation phase, cellulolytic activity was low
because the microorganisms consumed the starch, hemicellulose,
amino acids, some proteins, and soluble sugars in the matrix that
are readily available (Zagury et al., 2006). At 17 and 36 weeks, EAS
sources are scarce and the use of cellulose sources should be
enhanced, thus promoting functional cellulolytic ability. No sig-
nicant correlation was found between cellulolytic activity and pH,
although it has been reported that cellulose degradation can occur
after cessation of bacterial growth at a pH as low as 4.5 (optimal pH

Fig. 4. Dehydrogenase and cellulolytic activity. Whiskers indicate standard deviation. Asterisks indicate the presence of signicant differences (p < 0.05).
252 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253
is 4.75; Lynd et al., 2002).
4. Conclusions
Hydraulic retention time had a strong inuence in reactive
mixture composition and microbial activity during treating acid
mine drainage. Seven passive biochemical reactors used in the
present study were efcient for increasing pH and alkalinity and
removing sulfate (>60%) and metals (85e99% for Fe2 and 70e100%
for Zn2), without signicant differences between the three
retention times. The 4-day HRT generated higher alkalinity and
excess residual suldes in the efuent. The 1-day HRT washed out
biomass and increased input of dissolved oxygen in the reactors,
leading to higher ORP, which decreased metal removal efciency.
Hydraulic retention time had signicant effects on the reactive
mixture composition. Low ow in 4-day HRT reactors favored
growth of microorganisms that rapidly consumed organic nitrogen
and organic carbon in the reactive mixture. High ow in 1-day HRT
reactors decreased the pH (<6.3) at week 36, especially at the
bottom layer, probably from depletion of neutralizing agents
caused by constant AMD input and upward ow in the reactors.
Removal of Fe and Zn, probably in the form of suldes, took place
mainly in the bottom layers of all reactors, mainly in the 2-day HRT.
This led to accumulation of metal suldes and eventually produced
toxicity in the reactive mixture, leading to lower SRB counts and
dehydrogenase activity. The HRT strongly inuenced performance
of reactors and resulted in physical and biological inhibitions of SRB
and failures in the reactors. As a result, HRT must be a balance
between rate and removal efciency.
Acknowledgements
This research was funded by the Ponticia Universidad Javeriana
(ID. 5177) and the Department of Natural Sciences of the Uni-
versidad Central. We thank Carlos Diaz for assistance with labora-
tory analysis.
References
Adams, B., Anderson, R., Bless, D., Butler, B., Conway, B., Dailey, A., Zownir, A., 2014.
Reference Guide to Treatment Technologies for Mining-inuenced Water. vol.
542-R-14-001. US Environmental Protection Agency-OSRTI, Washington, DC.
APHA, AWWA, WEF, 2005. In: Clesceri, L.S., Greenberg, A.E., Eaton, A.D. (Eds.),
Standard Methods for the Examination of Water and Wastewater, twenty-rst
ed. American Public Health Association, Washington, DC.
ASTM, 1995a. Standard test method for pH of soils. In: Annual Book of ASTM
Standards, Section D4972-95a. ASTM International, West Conshohocken, PA,
pp. 27e28.
ASTM, 1995b. Standard test method for laboratory determination of water (mois-
ture) content of soil and rock. In: Annual Book of ASTM Standards, Section
D2216-92. ASTM International, West Conshohocken, PA, pp. 178e181.
ASTM, 2009. Standard methods for sulfate-reducing bacteria in water and water-
formed deposits. In: Annual Book of ASTM Standards, Section D4412-84.
ASTM International, West Conshohocken, PA, pp. 108e122.
Chang, I., Shin, P., Kim, B., 2000. Biological treatment of acid mine drainage under
sulphate-reducing conditions with solid waste materials as substrate. Water
Res. 34, 1269e1277.
Cheong, Y., Das, B., Roy, A., Bhattacharya, J., 2010. Performance of a SAPS-based
chemo-bioreactor treating acid mine drainage using low-DOC spent mush-
room compost, and limestone as substrate. Mine Water Environ. 29 (3),
217e224.
Chu, H., Lin, X., Fujii, T., Morimoto, S., Yagi, K., Hu, J., Zhang, J., 2007. Soil microbial
biomass, dehydrogenase activity, bacterial community structure in response to
long-term fertilizer management. Soil Biol. Biochem 39, 2971e2976.
Dische, Z., 1962. Color reactions of carbohydrates. In: Whistler, R.L., Wolfram, M.L.
(Eds.), Methods in Carbohydrate Chemistry. Academic Press, New York,
pp. 475e514.
Di Toro, D.M., McGrath, J.A., Hansen, D.J., Berry, W.J., Paquin, P.R., Mathew, R.,
Wu, K.U., Santore, R.C., 2005. Predicting sediment metal toxicity using a sedi-
ment biotic ligand model: methodology and initial application. Environ. Toxicol.
Chem. 24, 2410e2427.
Dvorak, D.H., Hedin, R.S., Edenborn, H.M., McIntire, P.E., 1992. Treatment of metal-
contaminated water using bacterial sulfate reduction: results from pilot-scale
reactors. Biotechnol. Bioeng. 40, 609e616.
Gusek, J.J., 2008. Passive treatment 101: an overview of the technologies. In: U.S.
EPA/National Groundwater Association's Remediation of Abandoned Mine,
Lands, pp. 1e13. Denver, CO.
Gustafsson, J.P., 2011. Visual MINTEQ, v.3.0. Royal Institute of Technology, Depart-
ment of Land and Water Resources Engineering, Stockholm. http://www2.lwr.
kth.se/English/OurSoftware/Vminteq/index.htm.
Harper, S.H.T., Lynch, J.M., 1981. The chemical components and decomposition of
wheat straw leaves, internodes and nodes. J. Sci. Food Agric 32, 1057e1062.
Hiibel, S., Pereyra, L., Riquelme, M.V., Reisman, D., Reardon, K., Pruden, A., 2011.
Effect of organic substrate on the microbial community structure in pilot-scale
sulfate-reducing biochemical reactors treating mine drainage. Environ. Eng. Sci.
28, 563e572.
Hu, X.F., Jiang, Y., Shu, Y., Hu, X., Liu, L., Luo, F., 2014. Effects of mining wastewater
discharges on heavy metal pollution and soil enzyme activity of the paddy
elds. J. Geochem. Explor. 147, 139e150.
IBM, 2010. IBM SPSS Statistics for Windows, Version 19.0. IBM, Armonk, NY.
Lewis, A.E., 2010. Review of metal sulphide precipitation. Hydrometallurgy 104,
222e234.
Lynd, L.R., Weimer, P.J., van Zyl, W.H., Pretorius, I.S., 2002. Microbial cellulose uti-
lization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. 66,
506e577.
Mariner, R., Johnson, D.B., Hallberg, K.B., 2008. Characterization of an attenuation
system for the remediation of Mn(II) contaminated waters. Hydrometallurgy
94, 100e104.
McCauley, C., O'Sullivan, A.D., Milke, M.W., Weber, P., Trumm, D., 2009. Sulfate and
metal removal in bioreactors treating acid mine drainage dominated with iron
and aluminum. Water Res. 43, 961e970.
Mikanova, O., 2006. Effects of heavy metals on some soil biological parameters.
J. Geochem. Explor. 88, 220e223.
Neculita, C.M., Zagury, G.J., Bussie re, B., 2008a. Effectiveness of sulfate-reducing
passive bioreactors for treating highly contaminated acid mine drainage: I.
Effect of hydraulic retention time. Appl. Geochem. 23, 3442e3451.
Neculita, C.M., Zagury, G.J., Bussie re, B., 2008b. Effectiveness of sulfate-reducing
passive bioreactors for treating highly contaminated acid mine drainage: II.
Metal removal mechanisms and potential mobility. Appl. Geochem. 23,
3545e3560.
Neculita, C.M., Zagury, G.J., Bussie re, B., 2007. Passive treatment of acid mine
drainage in bioreactors using sulfate-reducing bacteria e critical review and
research needs. J. Environ. Qual. 36, 1e16.
Nordstrom, K., Blowes, D., Ptacek, C., 2015. Hydrogeochemistry and microbiology of
mine drainage: an update. Appl. Geochem. 57, 3e16.
Okabe, S., Santegoeds, C.M., De Beer, D., 2003. Effect of nitrite and nitrate on in situ
sulde production in an activated sludge immobilized agar gel lm as deter-
mined by use of microelectrodes. Biotechnol. Bioeng. 81, 570e577.
Postgate, J.R., 1984. The Sulphate-reducing Bacteria. Cambridge University Press,
Cambridge, UK.
Prasad, D., Wai, M., Berube, P., Henry, J.G., 1999. Evaluating substrates in the
treatment of acid mine drainage. Environ. Technol 20, 449e459.
Pruden, A., Messner, N., Pereyra, L., Hanson, R.E., Hiibel, S.R., Reardon, K.F., 2007. The
effect of inoculum on the performance of sulfate-reducing columns treating
heavy metal contaminated water. Water Res. 41, 904e914.
Robinson-Lora, M.A., Brennan, R.A., 2011. Anaerobic precipitation of manganese and
co-existing metals in mine impacted water treated with crab shell-associated
minerals. Appl. Geochem 26, 853e862.

ez-Navarrete, C., Zamorano, A., Ferrada, C., Rodrguez, L., 2009. Sulphate reduc-
Sa
tion and biomass growth rates for Desulfobacterium autotrophicum in yeast
extract: supplemented media at 38  C. Desalination 248, 377e383.
Schumacher, B.A., 2002. Methods for the Determination of Total Organic Carbon
(TOC) in Soils and Sediments. NCEA-C-1282, EMASC-001. Ecological Risk
Assessment Support Center, Ofce of Research and Development, U.S. Envi-
ronmental Protection Agency, Las Vegas.
Serra-Wittling, C., Houot, S., Barriuso, E., 1995. Soil enzymatic response to addition
of municipal solid-waste compost. Biol. Fert. Soils 20, 226e236.
Song, H., Yim, G.J., Ji, S.W., Nam, I.H., Neculita, C.M., Lee, G., 2012. Performance of
mixed organic substrates during treatment of acidic and moderate mine
drainage in column bioreactors. J. Environ. Eng. 138, 1077e1084.
USEPA, 2001. Flame Atomic Absorption Spectrophotometry. Method 7000b. US
Environmental Protection Agency, Washington, DC.
Utgikar, V.P., Harmon, S.M., Chaudhary, N., Tabak, H.H., Govind, R., Haines, J.R., 2002.
Inhibition of sulfate-reducing bacteria by metal sulde formation in bioreme-
diation of acid mine drainage. Environ. Toxicol. Chem. 17, 40e48.
Vasquez, Y., Escobar, M.C., Neculita, C.M., Arbely, Z., Roldan, F., 2016. Selection of
reactive mixture for biochemical passive treatment of acid mine drainage. En-
viron. Earth Sci. http://dx.doi.org/10.1007/s12665-016-5374-2.
Waybrant, K.R., Ptacek, C.J., Blowes, D.W., 2002. Treatment of mine drainage using
permeable reactive barriers: column experiments. Environ. Sci. Technol. 36,
1349e1356.
Yoo, K., Sasaki, K., Hiroyoshi, N., Tsunekawa, M., 2004. Fundamental study on the
removal of Mn2 in acid mine drainage using sulfate reducing bacteria. Mater.
Trans. 45, 2422e2428.
Younger, P.L., Banwart, S.A., Hedin, R.S. (Eds.), 2002. Mine Water, Hydrogeology,
Pollution, Remediation. Kluwer Academic Publishers, Dordrecht, The
Netherlands.
Zagury, G.J., Kulnieks, V.I., Neculita, C.M., 2006. Characterization and reactivity
 Y. Vasquez et al. / Chemosphere 153 (2016) 244e253 253

assessment of organic substrates for sulphate-reducing bacteria in acid mine
drainage treatment. Chemosphere 64, 944e954.
Zaluski, M.H., Trudnowski, J.M., Harrington-Baker, M.A., Bless, D.R., 2003. Post-
mortem ndings on the performance of engineered SRB eld- bioreactors for
acid mine drainage control. In: Proc. Sixth International Conference on Acid
Rock Drainage. Cairns, QLD, Australia, pp. 845e853.
Zhang, M., Wang, H., 2014. Organic wastes as carbon sources to promote sulfate
reducing bacterial activity for biological remediation of acid mine drainage.
Miner. Eng. 69, 81e90.