Journal of Water Process Engineering 49 (2022) 103116

Contents lists available at ScienceDirect

Journal of Water Process Engineering

journal homepage: www.elsevier.com/locate/jwpe

Effective treatment of acid mine drainage by constructed wetland column:

Coupling walnut shell and its biochar product as the substrates
Junjun Chang a, *, Shengjiong Deng a, b, Xuan Li a, b, Yutong Li a, b, Jinquan Chen a,
Changqun Duan a, *
a
Yunnan Key Laboratory for Plateau Mountain Ecology and Restoration of Degraded Environments, School of Ecology and Environmental Science, Yunnan University,
Kunming 650500, China
b
Institute of International Rivers and Eco-security, Yunnan University, Kunming 650500, China

A R T I C L E I N F O A B S T R A C T

Keywords: Acid mine drainage (AMD) can cause severe risks to the surrounding environment and ecosystems because of its
Constructed wetland (CW) model high metal content and acidity and thus should be disposed of properly. In this study, a novel constructed
Acid mine drainage (AMD) wetland (CW) model system was established by packing a mixture of walnut shell (WS) and its biochar product as
Composite substrates
substrates to treat AMD with a pH of 4.0. Sucrose and plant straw broth (PSB) were successively supplemented as
Metal precipitation
external carbon sources to fuel the bacterial sulfate reduction (BSR) process. Heavy metals in the AMD were
Microbial community
efficiently removed in the CW column (Cu > 99 %, Cd > 97 %, Zn > 94 %, Cr > 93 % and Fe > 76 %), and the pH
increased from 4.0 to 6.5 despite an unsatisfactory sulfate removal efficiency (<40 %). Among the substrates,
biochar exhibited a notably higher metal retention capacity, followed by WS. The relative abundance of sulfate-
reducing bacteria (SRB) in the biochar substrate was the highest, followed by that in the WS. The cooperation of
organic-degrading bacteria mainly consisting of Cellulomonas, Clostridium and Bacteroides and SRB consortia was
responsible for biotic AMD treatment in the CW. A mixture of organic solid waste and biochar is a preferable
filler candidate for CW systems treating AMD although further evaluation of its capacity under varying condi­
tions is required.

1. Introduction can be operated easily and cost effectively, showing high applicability at
a large scale, especially in some remote mining locations [5,6]. Among
Acid mine drainage (AMD) is an acid solution with high levels of passive methods, constructed wetland (CW), a popular sustainable eco-
sulfate and various toxic heavy metal(loid)s that is primarily generated technology, has high environmental compatibility for AMD remediation
during and after mining and metallurgical activities when sulfide min­ through pathways including substrate adsorption and precipitation,
erals are oxidized in water and air [1,2]. The resulting acidic waste bacterial sulfate reduction (BSR) and subsequent metal precipitation,
stream has a high capability of leaching various elements in rocks and and plant uptake [7,8]. Using the metabolism of sulfate-reducing bac­
ores for a long time. Then, the produced toxic contaminants can teria (SRB) to simultaneously convert sulfate into sulfide and generate
percolate and be transported to the environment by runoff over long biogenic alkalinity for effective metal precipitation and acidity
distances [3]. AMD can seriously and persistently impact surrounding neutralization appears to be a sustainable and attractive mode of in AMD
soil and aquatic ecosystems and ultimately damage human health; thus, treatment and metal resource recovery [2,9]. However, AMD contains
pollution derived from AMD has become a worldwide environmental little organic carbon, which is needed as an electron and energy donor
concern [2,4,5]. Therefore, effective treatment of hazardous AMD, for SRB, thus limiting the BSR process. Nutrients for microbial growth
especially the elimination of multiple heavy metals, is imperative for and metabolism are also lacking. Furthermore, the high acidity of AMD
protecting the environment from extensive risk. and its elevated metal contents exert toxic stress on microbial consortia,
Many methods, including active and passive methods, have been including SRB [10].
established in recent decades to treat AMD [2]. Passive technologies Some solutions have been developed to address these issues, such as
have the advantages of relatively low energy and resource inputs and adding organic carbon sources or elemental sulfur to drive sulfidogenic

* Corresponding authors.
E-mail addresses: changjunjun@ynu.edu.cn (J. Chang), chqduan@ynu.edu.cn (C. Duan).

https://doi.org/10.1016/j.jwpe.2022.103116
Received 20 May 2022; Received in revised form 28 August 2022; Accepted 31 August 2022
Available online 8 September 2022
2214-7144/© 2022 Elsevier Ltd. All rights reserved.
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

processes and adding alkaline neutralization agents to bioreactors/CWs for microbial inoculation. The inoculated sludge contained a large
[8,11,12]. Sulfur reduction is a cost-efficient and effective process for number of facultative microbial species and was used as it was easily
AMD treatment, especially under neutral conditions [13–15]. Inexpen­ available even in mining area. The microbial community could evolve
sive organic solids are also an attractive CW filler for AMD treatment with the alteration of CW environment [7]. The CW system was placed
because they cannot only leach dissolved organic carbon to drive the outdoors under a rain shelter, and an acclimatization period of 20
BSR process with the cooperation of other microbes but also serve as d followed to facilitate the development of plants and microorganisms.
carriers for biofilm development and as biosorbents for metals
[7,11,16]. Alternatively, organic solids can be converted into biochar by 2.2. Influent AMD and external carbon sources
pyrolysis under an oxygen-deficient atmosphere; biochar is also an
attractive biofilter/CW substrate for AMD treatment due to its alkaline Artificial AMD was used to reduce the variability in influent water
nature and high adsorption capacity for dissolved metals [8,17]. quality in this study. The AMD was prepared in accordance with the
Moreover, biochar can indirectly facilitate AMD treatment by favouring typical properties of sulfide metal mine drainage recorded in the liter­
the growth of plants and functional microbes in CWs [18,19]. Never­ ature [4,6,9]. Specifically, CuCl2⋅2H2O, CdCl2⋅2.5H2O, ZnSO4⋅7H2O,
theless, only a few studies have employed biochar as an alternative K2Cr2O7 and FeSO4⋅7H2O were dissolved in tap water to obtain average
substrate in continuous-flow biofilters/CWs for treating AMD, although concentrations of 5.0 mg L− 1 Cu, 5.2 mg L− 1 Cd, 5.1 mg L− 1 Zn, 6.2 mg
biochar has been widely used as a promising remediator in soil eco­ L− 1 Cr and 56.0 mg L− 1 Fe. Na2SO4 was added to achieve a mean SO2− 4
systems or as a biosorbent for removing various pollutants from aquatic concentration of 920 mg L− 1. The pH of the AMD was adjusted to 4.0 by
solutions [17,20]. Organic solids and biochar products are expected to adding HCl.
work synergistically for effective AMD treatment by multiple routes, External carbon sources of sucrose and PSB were supplemented into
including by neutralizing acids, adsorbing metals, providing an organic the influent tank in operation phases I and II to enhance the BSR activity.
carbon supply and serving as biofilm carriers. But this assumption re­ Specifically, sucrose was added to obtain a COD concentration of 330
quires verification, and the involved mechanisms should be explored. mg L− 1, and NH4Cl and KH2PO4 were supplemented as nutrient sources
In the present study, a CW model system was established by filling a to achieve NH+ 4 -N and total phosphorus (TP) concentrations of 28.1 and
mixture of walnut shell (WS) and its biochar product as substrates to 3.0 mg L− 1 in phase I; phase I lasted for 65 d, with a stable but unsat­
treat AMD. External carbon sources, including sucrose and plant straw isfactory sulfate removal rate obtained. We simulated an addition of
broth (PSB), were successively supplemented to intensify microbial domestic wastewater into AMD deficient in organic matter in this phase.
removal pathways. The main aims were to (1) develop an efficient and To further drive the BSR process, PSB made from easily accessible plant
cost-effective CW-based AMD treatment model system by employing the litter and rich in organic compounds was added in phase II, which lasted
synergetic effect of an organic solid (WS) and its biochar product as over 50 d. The PSB was made by soaking and fermenting reed litter
substrates; (2) evaluate and compare the metal retention capacity and collected from a wetland near Dianchi Lake. The reed litter (1.5 kg, dry
speciation in different substrates; and (3) profile and compare the bac­ weight) was cut into 5 cm pieces after washing and then mixed with tap
terial community structure and functions associated with CW perfor­ water and anaerobic sludge. The mixture was placed in an 80 L tank and
mance in the substrates. sealed for anaerobic fermentation over 15 d. Then, the fermentation
broth supernatant was harvested as a concentrated carbon source. After
2. Materials and methods adding PSB, the COD, NH+ 4 -N and TP concentrations of the influent were
888.2, 62.8 and 5.8 mg L− 1, respectively. The influent (7.0 L) was
2.1. Experimental CW microcosm manually dosed to the CW surface once a day at 8:00–9:00 a.m. after the
treated wastewater was quickly drained from the outlet at the column
A lab-scale vertical subsurface-flow CW system was established using bottom by gravity. The hydraulic retention time (HRT) of the CW system
an acrylic column with dimensions of 20 × 20 × 60 cm for treating AMD, was 24 h, and the CW system was operated over 120 d.
and a schematic of its configuration is presented in Fig. S1. WS (an
agricultural waste) and its biochar product were employed as the CW 2.3. Sample collection and analysis
fillers. WS with a particle size of 2–4 mm was recycled as a solid organic
carbon source and biofilm carrier and filled in the top layer of the col­ Mixed effluent water samples of the CW system were collected, and
umn (15 cm in height). The biochar was fabricated by pyrolysing WS the pH and oxidation reduction potential (ORP) were measured imme­
under a nitrogen environment for 2 h and was layered in the middle of diately by using a pH meter (ST300, OHAUS, Parsippany, USA) and ORP
the column at a height of 15 cm. The hard texture of WS makes it suitable meter (PHBJ-260F, INESA, China), respectively. COD, NH+ 4 , NO3 , total
−
2− 2− 6+ 2+
for use in CWs to mitigate undesirable matrix collapse and fast clogging. nitrogen (TN), TP, SO4 , S , Cr and Fe were determined by spec­
The carbon contents in the WS and biochar were determined by an trophotometric methods according to standard methodologies described
elemental analyser (Vario EL III, Elementar, Germany), and the values in SEPA [22]. After acidification using HNO3 to pH ≤ 2, the dissolved
were 51.0 % and 88.9 %, respectively. The specific surface area was heavy metal concentrations in the water samples were measured by
measured by a surface area porosity analyser (3H-2000PS1, BeiShiDe using atomic absorption spectrophotometry (AAS, Varian AA140/240,
Instrument, China); the values were 3.0 m2 g− 1 for the WS and 4.3 m2 USA).
g− 1 for the biochar. A layer of fine gravel with a grain size of 2–3 mm After an operation time of 120 d, the substrates (including WS,
was filled below the biochar substrate. In addition, a 5 cm layer of gravel biochar and fine gravel) in the column were collected and freeze dried.
with a grain size of 5–8 mm was packed at the bottom of the column to Metal speciation in the substrates was determined according to the
facilitate water collection and prevent clogging of the effluent port. The sequential extraction method developed by Tessier et al. [23]. Specif­
effective volume for wastewater treatment of the CW column was 7.0 L. ically, 1.0 M MgCl2 (pH = 7) was added to tubes with the substrates to
Four healthy seedlings of Iris pseudacorus L. with a height of 15–20 extract the exchangeable metal fraction, and 1.0 M Na-acetate⋅3H2O
cm were planted in the CW. This macrophyte was chosen because it has (pH = 5) was subsequently added to extract the carbonate-bound frac­
high ornamental value and relatively strong resistance to environmental tion. Then, 0.04 M NH2OH⋅HCl in 25 % (v/v) CH3COOH and HNO3
stress, including toxic AMD [7,21]. Then, a mixture (100 mL, pH = 7.0 (0.02 M) + H2O2 (30 %) + CH3COONH4 (3.2 M) at a volume ratio of
± 0.1, MLVSS = 94.8 ± 33.6 g L− 1) of sludge from a sequencing batch 3:8:5 were successively used to extract the Fe/Mn oxide-bound and
reactor treating campus sewage with alternating aerobic and anaerobic organic matter-bound fractions. Finally, a HNO3-HCl-HClO4 mixture
environments and anaerobic sediment from a pond (v:v = 1:1) on was added to obtain the residual fraction. After each extraction step, the
campus was diluted with AMD (Section 2.2) and dosed into the column metal concentrations in the supernatant were measured using AAS after

2
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

Fig. 1. Temporal variations in influent and effluent concentrations and removal efficiencies of (a) Cu, (b) Cd, (c) Zn, (d) Cr, (e) total Fe and Fe(II) and (f) SO2−
4 and
dissolved sulfide in the CW microcosm.

filtration using 0.45-μm polyether sulfone filter membranes. In addition,

the mineralogical characteristics of the deposits produced in the column
were analysed by X-ray photoelectron spectroscopy (XPS, K-Alpha+,
Thermo Fisher Scientific, USA) and powder X-ray diffraction (XRD,
TTRIII-18KW, Rigaku, Japan) after vacuum lyophilization.
The plants were harvested and separated into shoots and roots. After
drying to a constant weight in an oven at 60 ◦ C, the plant samples were
digested using a mixture of HNO3 and HCl (v:v = 3:1) and 3–5 mL
HClO4. Then, the metal concentrations in the digestion solution were
determined by AAS, and metal assimilation by the plants was calculated.
The microbial community compositions on different substrates were
profiled using high-throughput sequencing technology at the end of the
experiment. After extracting the total DNA of the biofilms using HiPure
soil DNA kits (Magen, Guangzhou, China), the bacterial V3-V4 regions of
the 16S rRNA gene were amplified using the universal primer pair 341F
Fig. 2. Variations in the pH and ORP values of water samples in the CW column
and 806R. Subsequently, the purified DNA products were sequenced by over the operation time.
Gene Denovo Biotech Co., Ltd. (Guangzhou, China) on an Illumina
HiSeq 2500 PE250 system. The details on sequencing and data pro­
removal efficiencies for AMD in the CW microcosm are presented in
cessing have been described in our previous study [18].
Fig. 1. Metals in the AMD were efficiently removed by the CW column,
except for Fe (averaging 42.5 % in phase I and 75.9 % in phase II), with
3. Results and discussion
removal efficiencies of >99 % for Cu, >97 % for Cd, >94 % for Zn and
>93 % for Cr (Table S1). However, anaerobic conditions favouring the
3.1. AMD treatment performance
BSR process were not effectively established in the initial 40 d, with the
ORP fluctuating above − 100 mV and marginal SO2− 4 removal occurring
The time courses of variations in pollutant concentrations and
(Fig. 1f). This phenomenon could be verified by the increased effluent

3
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

Fig. 3. Influent and effluent concentrations and removal efficiencies of (a) COD, (b) nitrogen and (c) phosphorus in the CW.

acidity (Fig. 2) despite the use of an alkaline biochar substrate because amount of S2− produced both benefited metal removal by facilitating
alkalinity is not easily produced by the weak BSR activity of undevel­ metal precipitation as (oxyhydr)oxides, sulfides and carbonates [12,16].
oped microbial consortia. An adaptation period was required for the Thus, Fe removal was substantially increased, achieving a mean effi­
establishment of sulfidogenic conditions in bioreactors treating AMD ciency of 75.8 % in phase II (Table S1). Coprecipitation with Fe oxy­
[7,24]. Therefore, abiotic routes, including substrate adsorption, pre­ hydroxide could also contribute to the efficient removal of other metals
cipitation, complexation and cation exchange, could be predominantly [16,25]. Cr removal was slightly decreased after PSB introduction,
responsible for metal removal during this period [6,16]. Both organic probably because of reductive dissolution of a small fraction of formed
solids and biochar were proven to be excellent adsorbents for heavy Fe-Cr precipitates [27]. The removal efficiency for Cu (>99 %) was
metals [16,17]. The adsorbed Cr(VI) could be reduced to much less toxic higher than that for other metals because Cu precipitates can be easily
Cr(ІІІ), mainly through biochar and WS supplying redox-active electrons formed and are relatively stable [10].
for abiotic Cr(VI) reduction [17]. Fe removal was low (42.5 ± 18.2 % in Although the influent average COD concentration reached 888.2 mg
phase I), with the effluent Fe almost exclusively present in the form of L− 1 in phase II, the SO2− 2−
4 removal efficiency (the influent SO4 con­
Fe2+, probably because of the low oxidation strength of Fe2+ to Fe3+ − 1 2−
centration was 915.2 mg L in phase II, and the COD:SO4 mass ratio of
precipitates in the column and limited adsorption capacity of the sub­ 0.97 was higher than the stoichiometric value of 0.67 for total sulfate
strates for the high level of Fe2+ under the reducing and acidic envi­ removal) was still low (mean value of 26.6 %). The removal rate was
ronments within the CW column [1,25]. much lower than those obtained in some strictly anaerobic sulfidogenic
Then, the BSR process gradually evolved in the CW with the devel­ reactors fed with simple organic carbon sources such as ethanol and
opment of SRB consortia, and the Fe and SO2− 4 removal efficiencies and lactate [4,9], which could be attributed to the following reasons: (1) a
effluent pH values all increased. S2− produced by BSR could react with portion of the added organic substances were complex and not available
metals in the AMD to form stable precipitates, and the neutralized pH to SRB consortia, and the carbon sources were substantially consumed
due to biotic alkalinity favoured metal precipitation and microbial ac­ by other microbial groups such as aerobic heterotrophs and denitrifiers
tivity, which are desirable for AMD treatment [11,26]. After the intro­ instead of by SRB [4]; (2) the favourable anaerobic environment for SRB
duction of PSB, ORP decreased notably, favouring the growth and metabolism in the CW microcosm was disturbed by oxygen penetration
sulfidogenic activity of SRB. The mean SO2− 4 removal efficiency due to the daily addition of oxygen-containing influent, plant oxygen
increased from 7.5 % to 26.6 %, and the pH value increased from 6.2 to secretion and air reoxygenation; (3) the HRT was relatively short for an
6.6 despite PSB containing organic acids, implying that BSR activity was effective BSR process [8,28]; (4) the generated sulfide/sulfur could be
effectively intensified by the addition of abundant plant biomass- converted into sulfate by sulfur-oxidizing bacteria such as Sulfuricurvum
derived organic compounds. This type of carbon source is inexpensive in the CW (Section 3.3); and (5) metals and produced dissolved sulfide
and can be easily available in mining areas and can realize the recycling and organic acids could exert an adverse impact on microbial activities,
of wetland plant biomass. The elevated pH value above 6.5 and high including those of SRB [26,29]. However, toxic metals in the AMD were

4
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

Fig. 4. (a) Metal contents in the wetland plant biomass; (b–f) Metal speciation in the CW substrates based on metal mass per unit substrate volume (mg dm− 3): (b)
Cd; (c) Zn; (d) Cu; (e) Fe and (f) Cr.

efficiently eliminated by the CW model, and acidity was neutralized, because of the high influent Fe concentration and relatively low toxicity
largely reducing the toxicity of the AMD. Moreover, the effluent con­ of Fe. However, the total amount of metals assimilated by the plants was
centration of generated dissolved sulfide was low (0.1–2.8 mg L− 1), marginal compared with the influent loads, which was consistent with
which would have a limited impact on the receiving environment. some other studies [8,12], although the indirect roles of plants in CW
Overall, the CW model packed with WS and its biochar product was treatment performance and ecosystem function are versatile. The
effective in AMD remediation, although the remaining SO2− 4 and Fe
2+
comprehensive effect of plants in CWs on pollutant removal from AMD is
required further elimination through posttreatment units. However, uncertain and dependent on plant species, operating and environmental
there was only one CW model established in this study, thus further conditions, etc. [2,8], and more exploration is needed.
verification and evaluation on its treatment capacity over a long-term Substrate retention of various metal precipitates was the predomi­
operation time are still required and will be conducted in our near nant pathway of metal removal. Microbial consortia played a crucial
future study. role in the formation of metal precipitates (Section 3.3). The results
Organic compounds and nutrients imported with the AMD were also (Fig. 4) showed that notably more metals (except for Fe) were deposited
removed by the versatile CW system (Fig. 3). The COD removal effi­ on the biochar substrate, suggesting that the biochar substrate had a
ciency increased with the operation time and reached 80 % after 20 d. high capacity for metal retention. The metal sequestration capacity of
Effective COD removal was still achieved after the introduction of PSB WS was also higher than that of the conventional gravel substrate. Fe
with a high organic load because a variety of microbes in the CW could was more liable to coat the gravel substrate than the other substrates in a
decompose and utilize the organic substances [18]. Approximately 35 % stable residual form (Fig. 4e). The metals presented different speciation
nitrogen was removed, with almost all nitrogen being in the form of patterns in the three substrates, with high exchangeable and carbonate-
NH+ 4 , indicating that the nitrification process was limited due to a pre­ bound fractions for Zn and Cd and organic matter- and Fe/Mn oxide-
vailing anaerobic environment in the microcosm. TP removal effi­ bound fractions predominating for Cu and Cr. The results suggested
ciencies of 57.4 % and 49.2 % were obtained in phases I and II, that Cu and Cr were present in stable forms in the CW matrix, while the
respectively, primarily through phosphate adsorption and precipitation mobility and bioavailability of Zn and Cd were relatively high. Copre­
in the CW matrix [19]. The generation of FePO4 precipitates could cipitation with Fe might be a significant pathway for Cr removal [27].
contribute to both P and Fe removal [30]. The XPS results (Fig. 5a) show that Fe3O4 was a dominant Fe species
in the CW column (46.9 %). Fe3+ was the main form (50.5 %) and could
3.2. Metals in the plants and substrates be present as Fe2O3, Fe(OH)3, FePO4, etc. FeS2 was also detected in a low
proportion. Cr(III) was the dominant form of Cr (Fig. 5b) at a proportion
Fig. 4 shows the metal assimilation by plants and metal speciation in of 67.8 % and could be present as stable precipitates with low toxicity in
the CW substrates after the operation period of over 120 d. The plant the CW column. Cr(VI) accounted for 32.3 % of the total Cr in the de­
root tissues acted as the major sink of metals. The Fe content in the posits and might be mainly present in chromate form. XRD analysis
biomass was much higher than the contents of other metals, probably detected several Fe forms, including Fe3O4, Fe2O3, Fe(OH)3 and FePO4,

5
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

Fig. 5. XPS spectra of Fe 2p (a) and Cr 2p (b) of the deposits in the CW column; (c) XRD pattern of the deposits.

biochar substrate showed the highest diversity, probably because bio­

Table 1
char provides a high specific surface area and abundant micropores for
Alpha diversity indexes of the bacterial community in different CW substrate
microbial growth [18,19]. High microbial diversity can guarantee the
samples.
strong robustness and versatility of bioreactors/CWs and is particularly
Substrate OTUs Chao 1 ACE Shannon Simpson significant for establishing high performance in bioreactors treating
WS 2668 ± 3519 ± 3607 ± 7.70 ± 0.97 ± 0.01 AMD and fed with complex organics to drive the BSR process [4]. The
150 364 466 0.35 gravel substrate supported higher bacterial community richness.
Biochar 2709 ± 66 3554 ± 3779 ± 77 7.93 ± 0.99 ±
Fig. 6 describes the bacterial community characteristics in the CW
117 0.09 0.001
Gravel 2718 ± 3640 ± 3855 ± 7.65 ± 0.98 ± system. The community compositions on the three different substrates
195 269 209 0.34 0.007 were similar at the phylum level, with the phylum Proteobacteria
dominating, followed by Actinobacteria, Patescibacteria, Bacteroidetes
and Firmicutes. A much higher abundance of Proteobacteria, which
because of the high Fe content in the CW matrix. HCrO2 was also contains a variety of functional species associated with element cycling
identified. Other mineral phases were not detected by XRD, probably and toxic pollutant resistance, was detected on the biochar substrate
because of their relatively low content and/or poor crystallinity [7,8]. than on the WS and gravel, suggesting that the biochar supported more
Metals removed by some pathways, such as hydroxide precipitation and functional microbes [19]. At the genus level, the bacterial community
coprecipitation, would be primarily present in amorphous phases structure on different substrates showed notable distinctions. Specif­
[12,25]. ically, Geobacter was predominant on the biochar substrate (22.6 %);
this genus can not only perform BSR but also reduce iron, Cr and nitrate
3.3. Microbial community characteristics and taxa related to AMD coupled with anaerobic organic decomposition and immobilize other
purification in different substrates metals [31]. Sulfuricurvum (5.5 %) and Zooglaea (2.1 %), in addition to
Geobacter (3.3 %), occupied a high proportion on the WS. Sulfuricurvum
Microbial communities in CW substrates are of significance for belongs to the sulfur-oxidizing chemolithoautotrophs, which can
effective AMD biotreatment [4], which were characterized by high- transform sulfide into sulfate [32], and DO import into the top layer of
throughput sequencing technology at the end of the experiment when the column (packed with WS) might favour its growth. The abundances
the microbial consortia efficient for AMD remediation have been of Chlorobaculum (16.8 %) and Rhodopseudomonas (3.6 %) were rela­
formed. The alpha diversity indexes of microbial communities in tively high on the gravel substrate.
different substrates are listed in Table 1. The microorganisms on the

6
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

Fig. 6. Bacterial community compositions in different substrates at the end of the experiment at the (a) phylum and (b) genus levels (the top 15 phyla and top 20
genera in the samples are presented); relative abundances of (c) SRB and (d) sulfate reduction functional genes identified by Tax4Fun and (e) potential denitrifiers
and nitrifying bacteria in different substrates.

SRB consortia played an essential role in AMD treatment in the CW utilizing SRB and reducing undesirable acetate accumulation in bio­
system in the later operation period. Nine genera of SRB were detected reactors, and are related to the induction of sulfide production from
in the three substrates, with Desulfobulbus, Desulfatirhabdium, Desulfovi­ sulfite [28]. The fermentative genus Bacteroides occupied 0.87 % on the
brio and Desulfobacterium dominating. Among them, Desulfobulbus and WS, 0.43 % on the biochar and 0.64 % on the gravel; this genus can
Desulfovibrio are incomplete organic oxidizers, with acetate being an end convert complex organics such as lignocellulose to acetate [5,29]. A
product, and Desulfatirhabdium is a genus of complete organic oxidizers higher abundance of genera associated with organic decomposition was
that convert organics into CO2 [33]. The syntrophy of SRB groups with present on the WS substrate because it contains complex organic sub­
different substrate metabolisms conducted sulfate reduction in the CW, stances. These metabolically versatile taxa played a crucial role in the
preventing excessive accumulation of acetate that can adversely impact hydrolysis and fermentation of polymeric organics in the WS substrate
microbial activity. Desulfobacterium accounted for a high proportion on and PSB into small molecules and cooperated closely with SRB to carry
the WS and gravel substrates but presented a very low abundance on the out biotic AMD treatment [12,29].
biochar. Overall, the relative abundance of SRB was the highest on the Nitrifying bacteria occupied a very low abundance in the CW (<0.2
biochar substrate but was relatively low on the gravel, implying that %) and were mainly distributed in the top layer because of the anaerobic
biochar favoured the development of SRB consortia. WS was also su­ conditions within the CW column, resulting in limited NH+ 4 removal.
perior to the conventional gravel substrate in supporting SRB growth. Potential denitrifying bacteria, mainly consisting of Zoogloea, Acineto­
The relative abundance of sulfate reduction functional genes was pre­ bacter, Dechloromonas and Proteiniclasticum [30], presented higher
dicted by the Tax4Fun tool. The results showed that the abundance of abundances on the WS substrate. The biochar substrate also favoured
genes encoding dissimilatory sulfate reduction was the highest on the denitrifier attachment relative to the gravel, which facilitated nitrogen
biochar substrate, implying that biochar supported high BSR activity. removal [19]. It is necessary to further profile microbial community on
However, some functional genes related to assimilatory sulfate reduc­ different filling materials in CWs treating AMD under varied conditions.
tion, which generates reduced sulfur compounds for biosynthesis,
exhibited the lowest abundance on the biochar substrate, suggesting 4. Conclusion
that microorganisms on biochar were less capable of removing sulfate
through the assimilatory pathway. A CW system for the effective treatment of AMD was developed by
Organic-degrading microbes were also significant for the effective filling a mixture of WS and its biochar product as substrates and adding
operation of the CW system because they hydrolyse and provide low- liquid organic carbon derived from plant biomass. Metals (except for Fe)
molecular-weight organic substrates for SRB, which cannot directly were efficiently removed even in the initial operation period when the
utilize complex organic carbon. Cellulomonas, a cellulolytic genus that sulfidogenic activity was not well developed. Among the substrates,
can degrade cellulose, exhibited higher abundance on WS (0.85 %) than biochar showed a higher capacity for retaining metals, while gravel
on the biochar (0.37 %) and gravel (0.39 %) substrates. The genus presented the poorest metal interception. Moreover, biochar supported a
Clostridium, containing many members that are fermentative and syn­ higher abundance of SRB, which was favourable for AMD biotreatment.
trophic and can resist low pH by forming spores [34], accounted for WS was also a desirable CW filler for binding metals and supporting
0.62 %, 0.63 % and 0.64 % on the WS, biochar and gravel substrates, various functional microbes, in addition to providing a carbon source.
respectively. Moreover, some members of Clostridium can metabolize Organic-degrading bacteria, dominated by Cellulomonas, Clostridium and
acetate to H2 under anoxic conditions, providing electron donors for H2- Bacteroides, worked in close syntrophy with SRB, mainly consisting of

7
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

Desulfobulbus, Desulfatirhabdium and Desulfovibrio, to effectively treat [11] L. Lefticariu, E.R. Walters, C.W. Pugh, K.S. Bender, Sulfate reducing bioreactor
dependence on organic substrates for remediation of coal-generated acid mine
AMD by biotic means. Overall, CW system filled with organic solid waste
drainage: field experiments, Appl. Geochem. 63 (2015) 70–82, https://doi.org/
and its biochar product provides a promising ecological option for AMD 10.1016/j.apgeochem.2015.08.002.
treatment despite further investigation on its capacity and the influ­ [12] H. Wang, M. Zhang, J. Xue, Q. Lv, J. Yang, X. Han, Performance and microbial
encing factors over a long-term operation time is still needed. response in a multi-stage constructed wetland microcosm co-treating acid mine
drainage and domestic wastewater, J. Environ. Chem. Eng. 9 (2021), 106786,
https://doi.org/10.1016/j.jece.2021.106786.
[13] A.P. Florentino, J. Weijma, A.J.M. Stams, I. Sánchez-Andrea, Sulfur reduction in
Declaration of competing interest acid rock drainage environments, Environ. Sci. Technol. 49 (19) (2015)
11746–11755, https://doi.org/10.1021/acs.est.5b03346.
[14] A.P. Florentino, J. Weijma, A.J.M. Stams, I. Sánchez-Andrea, Ecophysiology and
The authors declare that they have no known competing financial
application of acidophilic sulfur-reducing microorganisms, in: Biotechnology of
interests or personal relationships that could have appeared to influence Extremophiles, 2016, pp. 141–175, https://doi.org/10.1007/978-3-319-13521-2_
the work reported in this paper. 5.
[15] J. Guo, Y. Li, J. Sun, R. Sun, S. Zhou, J. Duan, W. Feng, G. Liu, F. Jiang, pH-
dependent biological sulfidogenic processes for metal-laden wastewater treatment:
Data availability sulfate reduction or sulfur reduction? Water Res. 204 (2021), 117628 https://doi.
org/10.1016/j.apgeochem.2005.04.012.
The data that has been used is confidential. [16] O. Gibert, J. de Pablo, J.L. Cortina, C. Ayora, Municipal compost-based mixture for
acid mine drainage bioremediation: metal retention mechanisms, Appl. Geochem.
20 (2005) 1648–1657, https://doi.org/10.1016/j.apgeochem.2005.04.012.
Acknowledgement [17] H. Hu, J. Zhang, T. Wang, P. Wang, Adsorption of toxic metal ion in agricultural
wastewater by torrefaction biochar from bamboo shoot shell, J. Clean. Prod. 338
(2022), 130558, https://doi.org/10.1016/j.jclepro.2022.130558.
This work was financially supported by the National Natural Science [18] B. Ji, J. Chen, J. Mei, J. Chang, X. Li, W. Jia, Y. Qu, Roles of biochar media and
Foundation of China (U2002208), China Yunnan Provincial R & D oxygen supply strategies in treatment performance, greenhouse gas emissions, and
bacterial community features of subsurface-flow constructed wetlands, Bioresour.
Programs (202101AS070033, 2019BC001), the Science and Technology
Technol. 302 (2020), https://doi.org/10.1016/j.biortech.2020.122890.
Program of Yunnan Province, China (202201AT070182), Open [19] S. Deng, J. Chen, J. Chang, Application of biochar as an innovative substrate in
Research Fund from Yunnan Key Laboratory for Plateau Mountain constructed wetlands/biofilters for wastewater treatment: performance and
Ecology and Restoration of Degraded Environments (2018DG005) and ecological benefits, J. Clean. Prod. 293 (2021), 122890, https://doi.org/10.1016/j.
jclepro.2021.126156.
College Student Innovation and Entrepreneurship Program in Yunnan [20] J. Chang, Y. Duan, J. Dong, S. Shen, G. Si, F. He, Q. Yang, J. Chen, Bioremediation
University (202004173). We are also grateful for the Advanced Analysis of Hg-contaminated soil by combining a novel Hg-volatilizing Lecythophora sp.
and Measurement Center of Yunnan University for the sample testing fungus, DC-F1, with biochar: performance and the response of soil fungal
community, Sci. Total Environ. 671 (2019) 676–684, https://doi.org/10.1016/j.
service. scitotenv.2019.03.409.
[21] N. Oustriere, L. Marchand, K. Lizama-Allende, E. Roulet, C. Rousset, F. Bordas,
M. Mench, Selection of macrophytes with Cu-enriched root biomass intended for
Appendix A. Supplementary data
ecocatalyst production, Ecol. Eng. 138 (2019) 88–96, https://doi.org/10.1016/j.
ecoleng.2019.07.001.
Supplementary data to this article can be found online at https://doi. [22] SEPA (State Environmental Protection Administration of China), Methods for
org/10.1016/j.jwpe.2022.103116. Monitoring and Analysis of Water and Wastewater, 4th ed, China Environmental
Science Press, Beijing, 2002 (in Chinese).
[23] A. Tessier, P.G.C. Campbell, M. Bisson, Sequential extraction procedure for the
References speciation of particulate trace metals, Anal. Chem. 51 (1979) 844–851, https://doi.
org/10.1021/ac50043a017.
[24] M.P. Cunha, R.M. Ferraz, G.P. Sancinetti, R.P. Rodriguez, Long-term performance
[1] G. Qiu, Y. Luo, C. Chen, Q. Lv, W. Tan, F. Liu, C. Liu, Influence factors for the
of a UASB reactor treating acid mine drainage: effects of sulfate loading rate,
oxidation of pyrite by oxygen and birnessite in aqueous systems, J. Environ. Sci. 45
hydraulic retention time, and COD/SO42- ratio, Biodegradation 30 (2019) 47–58,
(2016) 164–176, https://doi.org/10.1016/j.jes.2016.01.012.
https://doi.org/10.1007/s10532-018-9863-8.
[2] K. Rambabu, F. Banat, Q.M. Pham, S.H. Ho, N.Q. Ren, P.L. Show, Biological
[25] L. Fernandez-Rojo, M. Héry, P.L. Pape, C. Braungardt, A. Desoeuvre, E. Torres,
remediation of acid mine drainage: review of past trends and current outlook,
V. Tardy, E. Resongles, E. Laroche, S. Delpoux, C. Joulian, F. Battaglia-Brunet,
Environ. Sci. Technol. 2 (2020), 100024, https://doi.org/10.1016/j.
J. Boisson, G. Grapin, G. Morin, C. Casiota, Biological attenuation of arsenic and
ese.2020.100024.
iron in a continuous flow bioreactor treating acid mine drainage (AMD), Water Res.
[3] S. Shen, J. Chen, J. Chang, B. Xia, Using bioenergy crop cassava (Manihot
123 (2017) 594–606, https://doi.org/10.1016/j.watres.2017.06.059.
esculenta) for reclamation of heavily metal-contaminated land, Int. J.
[26] M.G. Kiran, K. Pakshirajan, G. Das, An overview of sulfidogenic biological reactors
Phytoremediat. 22 (2020) 1313–1320, https://doi.org/10.1080/
for the simultaneous treatment of sulfate and heavy metal rich wastewater, Chem.
15226514.2020.1768512.
Eng. Sci. 158 (2017) 606–620, https://doi.org/10.1016/j.ces.2016.11.002.
[4] M. Martins, E.S. Santos, M.L. Faleiro, S. Chaves, R. Tenreiro, R.J. Barros, M.
[27] G. Dotro, D. Larsen, P. Palazolo, Preliminary evaluation of biological and
C. Costa, Performance and bacterial community shifts during bioremediation of
physical–chemical chromium removal mechanisms in gravel media used in
acid mine drainage from two Portuguese mines, Int. Biodeterior. Biodegradation 65
constructed wetlands, Water Air Soil Pollut. 215 (2011) 507–515, https://doi.org/
(2011) 972–981, https://doi.org/10.1016/j.ibiod.2011.07.006.
10.1007/s11270-010-0495-9.
[5] H. Chen, T. Xiao, Z. Ning, Q. Li, E. Xiao, Y. Liu, Q. Xiao, X. Lan, L. Ma, F. Lu, In-situ
[28] P. Hernández, G. Recio, C. Canales, A. Schwarz, D. Villa-Gomez, G. Southam,
remediation of acid mine drainage from abandoned coal mine by filed pilot-scale
I. Nancucheo, Evaluation of operating conditions on sulfate reduction from acidic
passive treatment system: performance and response of microbial communities to
wastewater in a fixed-bed bioreactor, Miner. Eng. 177 (2022), 107370, https://doi.
low pH and elevated Fe, Bioresour. Technol. 317 (2020), 123985, https://doi.org/
org/10.1016/j.mineng.2021.107370.
10.1016/j.biortech.2020.123985.
[29] E.W. Nogueira, L.A.G. de Godoi, L.N.M. Yabuki, G. Brucha, M.H.R.Z. Damianovic,
[6] J.O. Ighalo, S.B. Kurniawan, K.O. Iwuozor, C.O. Aniagor, O.J. Ajala, S.N. Oba, F.
Sulfate and metal removal from acid mine drainage using sugarcane vinasse as
U. Lwuchukwu, S. Ahmadi, C.A. Igwegbe, A review of treatment technologies for
electron donor: performance and microbial community of the down-flow
the mitigation of the toxic environmental effects of acid mine drainage (AMD),
structured-bed bioreactor, Bioresour. Technol. 330 (2021), 124968, https://doi.
Process Saf. Environ. Prot. 157 (2022) 37–58, https://doi.org/10.1016/j.
org/10.1016/j.biortech.2021.124968.
psep.2021.11.008.
[30] B. Ji, M. Jiang, Y. Yang, S. Deng, J. Chen, J. Chang, High treatment effectiveness
[7] J. Chen, S. Deng, W. Jia, X. Li, J. Chang, Removal of multiple heavy metals from
for secondary effluent in Fe–C microelectrolysis constructed wetlands with electron
mining-impacted water by biochar-filled constructed wetlands: adsorption and
donor supplementation, J. Clean. Prod. 342 (2022), 130934, https://doi.org/
biotic removal routes, Bioresour. Technol. 331 (2021), 125061.
10.1016/j.jclepro.2022.130934.
[8] S. Singh, S. Chakraborty, Bioremediation of acid mine drainage in constructed
[31] D.R. Lovley, S.J. Giovannoni, D.C. White, J.E. Champine, E.J.P. Phillips, Y.
wetlands: aspect of vegetation (Typha latifolia), loading rate and metal recovery,
A. Gorby, S. Goodwin, Geobacter metallireducens gen. nov. sp. nov., a
Miner. Eng. 171 (2021), 107083, https://doi.org/10.1016/j.mineng.2021.107083.
microorganism capable of coupling the complete oxidation of organic compounds
[9] E. Sahinkaya, F.M. Gunes, D. Ucar, A.H. Kaksonen, Sulfidogenic fluidized bed
to the reduction of iron and other metals, Arch. Microbiol. 159 (1993) 336–344,
treatment of real acid mine drainage water, Bioresour. Technol. 102 (2011)
https://doi.org/10.1007/BF00290916.
683–689, https://doi.org/10.1016/j.biortech.2010.08.042.
[32] A. Giordani, R.P. Rodriguez, G.P. Sancinetti, E. Ayumi-Hayashi, E. Beli, G. Brucha,
[10] T. Yilmaz, A. Yucel, Y. Cakmak, S. Uyanik, A. Yurtsever, D. Ucar, Treatment of
Effect of low pH and metal content on microbial community structure in an
acidic mine drainage in up-flow sulfidogenic reactor: metal recovery and the pH
neutralization, J. Water Process. Eng. 32 (2019), 100916, https://doi.org/
10.1016/j.jwpe.2019.100916.

8
J. Chang et al. Journal of Water Process Engineering 49 (2022) 103116

anaerobic sequencing batch reactor treating acid mine drainage, Miner. Eng. 141 [34] C. Rodrigues, D. Núñez-Gómez, H.V.D.M. Follmann, D.D. Silveira, M.E. Nagel-
(2019), 105860, https://doi.org/10.1016/j.mineng.2019.105860. Hassemer, F.R. Lapolli, M.Á. Lobo-Recio, Biostimulation of sulfate-reducing
[33] D. Zeng, Q. Yin, Q. Du, G. Wu, System performance and microbial community in bacteria and metallic ions removal from coal mine-impacted water (MIW) using
ethanol-fed anaerobic reactors acclimated with different organic carbon to sulfate shrimp shell as treatment agent, J. Hazard. Mater. 398 (2020), 122893, https://
ratios, Bioresour. Technol. 278 (2019) 34–42, https://doi.org/10.1016/j. doi.org/10.1016/j.jhazmat.2020.122893.
biortech.2019.01.047.