Journal of Hazardous Materials 396 (2020) 122631

Contents lists available at ScienceDirect

Journal of Hazardous Materials

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

Simultaneous removal of iron and manganese from acid mine drainage by T

acclimated bacteria
Dongmei Hou, Pan Zhang, Dongning Wei, Jiachao Zhang, Binghua Yan, Linying Cao,
Yaoyu Zhou*, Lin Luo*
College of Resources and Environment, Hunan Agricultural University, Changsha, 410128, PR China

GRAPHICAL ABSTRACT

ARTICLE INFO ABSTRACT

Editor: R Teresa A bacterial consortium for efficient decontamination of high-concentration Fe–Mn acid mine drainage (AMD)
Keywords: was successfully isolated. The removal efficiencies of Fe and Mn were effective, reaching 99.8 % and 98.6 %,
Bacteria community dynamics respectively. High-throughput sequencing of the 16S rRNA genes demonstrated that the microbial community
Biogenic Fe-Mn oxides had changed substantially during the treatment. The Fe–Mn oxidizing bacteria Flavobacterium, Brevundimonas,
Metal removal Stenotrophomonas and Thermomonas became dominant genera, suggesting that they might play vital roles in Fe
Biofilter and Mn removal. Moreover, the pH of culture increased obviously after incubation, which was benefit for de-
positing Fe and Mn from AMD. The specific surface area of the biogenic Fe–Mn oxides was 108–121 m2/g, and
the surface contained reactive oxygen functional groups (–OH and −COOH), which also improved Fe and Mn
removal efficiency. Thus, this study provides an alternative method to treat AMD containing high concentrations
of Fe and Mn.

1. Introduction aquatic and terrestrial ecosystems (Aguinaga et al., 2018; Auld et al.,
2017). Available methods for AMD treatment include traditional neu-
Acid mine drainage (AMD) is severely threatening the safety of tralization techniques, magnetic nanoparticles, lignite and zeolite, and

⁎
Corresponding authors.
E-mail addresses: zhouyy@hunau.edu.cn (Y. Zhou), luolinwei0@163.com (L. Luo).

https://doi.org/10.1016/j.jhazmat.2020.122631
Received 10 January 2020; Received in revised form 31 March 2020; Accepted 31 March 2020
Available online 09 April 2020
0304-3894/ © 2020 Elsevier B.V. All rights reserved.
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

membrane methods, among others (Rand and Ranville, 2019). Tradi- Qibaoshan mine AMD (Table S3). Being cultured in the synthetic AMD
tional neutralization techniques require additions of commercially medium 10 days, the final culture was maintained as an active culture
produced alkali, which is expensive and generate a large volume of for further use.
sludge, especially in AMD that contains high levels of iron/ferrous ions Incubation experiment was carried out in a series of 500-mL
(Johnson and Hallberg, 2005). Moreover, some of the new treatments Erlenmeyer flasks. AMD water was sampled at Qibaoshan in October
are too complex or are not suitable for wide application (Kefeni et al., 2018 and stored at 4 °C. Each flask contained 200 mL of AMD water,
2017). In the last few decades, biological methods have drawn con- 20 mL active culture, 0.1 g/L peptone, 0.1 g/L glucose. In the control,
siderable attention, and a variety of bacteria have been investigated to AMD water was filtered through 0.22 μm filters to remove bacteria and
determine their functions in AMD remediation. Sulfate reducing bac- the active culture was replaced by sterile water. Incubation conditions
teria (SRB) are commonly used in AMD remediation (Dale et al., 2015). were as follows: work volume, 220 mL; temperature, 30 °C; rotation
However, these bacteria have specific environmental requirements (i.e., speed, 170 rpm. Each flask was monitored daily for iron and manganese
an anaerobic environment, pH > 5, an organic substrate, and an ap- concentration, pH, cell density and dissolved oxygen (DO). After cul-
propriate sulfur species), and Mn is not effectively removed by biogenic tured 2 days, 7 days and 15 days, the sediments were collected for
sulfide treatment (Gibert et al., 2002). Hence, the removal effect of Fe analyzing dynamic diversity of microbial community. The structure
and Mn were not ideal by previous methods. It is urgent to find a more characterizations of sediments were analyzed after cultured 2 days and
effective way to deal with this problem. 15 days. The sampling loss was compensated with added sterile water.
Iron (Fe) and manganese (Mn)-oxidizing bacteria (FMOB) survive
and flourish in the extremely harsh environments of AMD ecosystems. 2.2. Analysis of physicochemical parameters
Fe-oxidizing bacteria (FeOB) oxidize Fe (Ⅱ) to Fe (Ⅲ) and generate
ferric iron precipitates that can be used as a sorbent for heavy metals The pH was measured immediately using a pH meter (Shanghai REX
removal from aqueous solutions (Larson et al., 2014). Mn-oxidizing Instrument Factory, Shanghai, China). The cell density was counted
bacteria (MnOB) oxidize Mn (Ⅱ) to Mn (Ⅲ) or Mn (Ⅳ) and produce with hemocytometer (Wang et al., 2018). DO was analyzed by Mik-
biogenic Mn oxides that can further oxidize and adsorb other pollutants DO700 (Asmik, Hangzhou, China). The concentrations of total Fe and
(e.g., heavy metals and organic contaminants) (Zhang et al., 2015). In Mn were determined using inductively coupled plasma mass spectro-
previous studies, combined MnOB and FeOB were added to a biofilter to metry (ICP-MS). And the samples used in ICP-MS were centrifuged
increase the treatment efficiency.(Bai et al., 2016b; Zhou et al., 2016) (10,000×g) to remove residue and bacteria, and then digest in a gra-
More importantly, bacteria that had both Fe and Mn-oxidizing activities phite digestion apparatus. More details were described in the previous
were isolated, and their ability to remove Fe and Mn ions in ground- study (Hou et al., 2019). Fe(Ⅱ) was analyzed by ferrozine assay
water was explored (Fan et al., 2016; Li et al., 2016). However, most of (Kappler and Newman, 2004), Mn(Ⅲ/Ⅳ) was measured by leuco-
the isolated MnOB were neutrophiles, because they only oxidized Mn berbelin method (LBB) (Cerrato et al., 2010), Fe(Ⅲ), Mn(Ⅱ), and re-
(II) and precipitated Mn oxides at pH > 7. Thus, FMOB are mostly used moval efficiencies of Fe and Mn were calculated by formulas as follows.
in the treatment of groundwater in which the pH of the working con- All samples were tested three times, and results were reported as the
dition is neutral (Bai et al., 2016a; Qin et al., 2009). Few studies have mean of the data.
explored the ability of these bacteria to treat AMD, and little is known
about the biological oxidation of Fe and Mn at acidic pH. Mn(Ⅱ)=total Mn-Mn(Ⅲ/Ⅳ) (1)
In this study, Fe–Mn rich water of the well-documented AMD site in
Fe(Ⅲ)=total Fe-Fe(Ⅱ) (2)
Qibaoshan Mine was sampled to assemble a Fe–Mn oxidizing bacterial
consortium (Hou et al., 2019). Then, the consortium was adapted to Removal efficiency of Fe/Mn =
C0 C1
× 100%
simultaneously remove Fe and Mn from AMD. Moreover, the changes of C0 (3)
the bacterial community during treatment were analyzed by sequencing
(C0: concentrations of total Fe and Mn before treatment; C1: con-
16S rRNA gene amplicons using the Illumina MiSeq platform. Finally,
centrations of total Fe and Mn after treatment)
the characteristics of biological formations were determined. Thus, this
study offers an alternative method for AMD treatment and ultimately
contributes to increase understanding of the biogeochemical cycling of 2.3. Analysis of microbial community
Fe and Mn.
2.3.1. DNA extraction and high-throughput sequencing
2. Materials and methods The bacteria were harvested from 50 mL samples (including AMD
water and sediment) by centrifuged (10,000×g) for 10 min. Total
2.1. Batch experiments genome DNA was extracted using CTAB/SDS method. After determi-
nate the concentration by 1% agarose gels, DNA samples were diluted
2.1.1. Enrichment of Fe-Mn-oxidizing consortium to 1 ng/μL using sterile water.
The Fe-Mn-oxidizing consortium was enriched from acid mine The 16S rRNA V3-V4 gene was amplified using primers 314 F
drainage in Qibaoshan mine, Liuyang City, Hunan province, China. (5′-CCTAYGGGRBGCASCAG-3′) and 806R (5′-GGACTACNNGGGTAT
After sample collected, 30 mL sample was inoculated into 150 mL of CTA AT-3′) (Wu et al., 2016). Purified PCR amplicons were carried out
modified PYCM medium (Li et al., 2016) (Table S1) contained in 500- with Illumina HiSeq platform (Hou et al., 2019).
mL Erlenmeyer flasks. After 40 days of enrichment in this medium (10
days a cycle), the enrichment culture was obtained. The enrichment 2.3.2. Data analysis
culture described above was adapted to high concentrations of Mn and Raw fastq files were optimized by following steps: remove se-
Fe by serial sub-culturing (data not shown). After adaption, the effects quences with low quality (with a length <220bp), with ambiguous base
of pH, temperatures, Fe and Mn contents on bacteria growth were “N”, and score lower than 20. After that, high quality sequences with ≥
carried out (details in the supported materials). 97 % similarity were clustered into the same operational taxonomic
units (OTUs) for further annotation (Caporaso et al., 2010). The de-
2.1.2. Incubation experiment tailed analyses of OTUs were provided in the supporting information. In
At the beginning of incubation experiment, the acclimated bacteria this study, the relationship between bacterial community structure and
was inoculated in 500-mL Erlenmeyer flasks containing 150 mL syn- environmental factors was analyzed by the redundancy analysis (RDA)
thetic AMD medium (Table S2), simulating the composition of using Canoco 5.0 (Ren et al., 2018). All original sequences were

2
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

deposited into the NCBI and the accession number was SUB6625219. static and dynamic experiments: zeolite < quartz sand < manganese
sand < ceramic particles. Hence, ceramic particles were used as filling
2.4. Characterization of sediment layer in the column experiments. Moreover, due to its rough surface and
porousness ceramic particles could absorb small amounts of iron and
2.4.1. Sample preparation manganese, which enhanced removal efficiency, especially the removal
After cultured 2 days and 15 days, the sediments were obtained by ratio of Mn (Han et al., 2013; Musa et al., 2016). The performance of
centrifuging (10,000×g) 50 mL suspension. After that, the collected the biofilter was monitored for 45 days, and Fe and Mn removal effi-
sediments were dried by a freeze drier and divided into five pieces for ciencies were presented in Fig. S10. After operated 36 days, the biofilter
scanning electron microscopy coupled with energy-dispersive X-ray reached saturation, the average removal efficiencies of Fe and Mn were
spectrum (SEM-EDS), X-ray diffraction (XRD), X-ray photoelectron 99.8 % and 98.6 %, respectively. Consistent with the batch experiment,
spectroscopy (XPS), Brunauer Emmett Teller (BET), and Fourier the removal ratio of Fe was higher than that of Mn, and the time needed
Transform Infra-Red Spectrophotometer (FT-IR), respectively. for Fe steady removing was shorter.

3.2. Diversity of microbial community during the treatment

2.4.2. Characterization
The SEM-EDS was carried out to investigate the morphology, size,
To investigate the changes in the microbial community diversity
and element distribution of the sediment (Hitachi SU-8020). For XRD
during the treatment, nine representative samples were collected after
patterns of sediment, the samples were thoroughly ground in a mortar,
cultured 2, 7, and 15 days. After the original reads were subjected to
pressed to be a disk, finally analyzed by a Rigaku D-MAX2500/PC ad-
quality control, a total of 770,707 raw sequences and 726,667 effective
vance instrument with Cu-Ka radiation, λ = 1.5418Ǻ. The data were sequences with an average length of 406 bp were obtained from the
collected at a 0.71°2Th./min of scanning speed in the angle range from nine treatment samples (Table S5). The rarefaction curves indicated
5° to 90°. XPS was carried out on a PHI 5000 VersaProbe with a that these numbers of sequences were sufficient to reach saturation
monochromatic Al Ka X-ray source. The BET method was performed by (Fig. S4).
N2 adsorption−desorption measurement employing a surface area The shannon index was used to estimate microbial community di-
analyzer ASAP-2020. FT-IR spectroscopy analysis of the sediment was versity in this study, and the higher index indicated higher diversity. As
conducted by a Vertex 70v spectrometer (Bruker, Germany) with a shown in Fig.S5, the Shannon indexes of 7 days (4.4) and 15 days (4.8)
resolution of 4 cm−1. were significantly higher than that of 2 days (3.7) (P < 0.05, t-test),
which indicated that the α-diversity of the bacterial community in-
3. Results creased during the incubation. Moreover, for β-diversity, the commu-
nity structures of all samples were compared using principal component
3.1. Environmental tolerance and remediation capacity of bacteria analysis (PCA) with euclidean distances (Fig. 2). The bacterial com-
munities that are similar in structure are close to one another on a PCA
In this study, the Fe-Mn oxidizing consortium was enriched from map. Hence, as shown in Fig. 2, the bacterial community structure at
AMD containing high concentrations of Fe and Mn. After acclimation, different stages of treatment was significantly different (P < 0.05, t-
the effects of different pH, temperatures, Fe and Mn contents on bac- test), indicating a succession of bacterial communities during the in-
teria growth were determined, and the results were shown in Fig. S2 cubation.
and S3. It was demonstrated that the acclimated bacteria could survive To further elucidate the dominant taxa in the bioreactors, the
in the pH range of 4–8, temperature range of 10−30 °C, Fe con- overall phylum and genus-level taxa characteristics are compared in
centration range from 100 to 1300 mg/L and Mn concentration range Fig. 3. Proteobacteria, Bacteroidetes, and Actinobacteria were the pre-
from 100 to 700 mg/L. Beyond these limits, these metals may be toxic dominant phyla at all stages of the incubation, accounting for more
to the bacterial consortium and the removal ratios decreased sig- than 99.8 % of the total reads in each library. The relative abundance of
nificantly. Proteobacteria and Actinobacteria deceased after 2 days, whereas that of
After inoculated into a real AMD, the removal efficiencies of Fe and Bacteroidetes increased sharply at 7 and 15 days samples, from 0.44 %
Mn and the changes of pH, cell densities and dissolved oxygen (DO) (2 days) to 20.9 % (7 days) and 20.1 % (15 days), respectively. At the
were shown in Fig. 1. The concentrations of total Fe, total Mn, Fe2+, genus level, the predominant genera also changed during the treatment.
Fe3+, Mn2+, and Mn4+ were provided in supporting information For example, Pandoraea, Azospirillum, and Pseudomonas were more
(Table S4). At the beginning of the treatment, the removal efficiencies abundant at 2 days than at later stages. Compared with 2 days, Flavo-
of Fe and Mn, the cell densities, pH and dissolved oxygen of incubation bacterium, unidentified_Rhizobiaceae, Brevundimonas and Steno-
were not changed obviously. After cultured for 2 days, the cell density trophomonans increased and were highly abundant in 7 and 15 days.
of bacteria increased sharply and entered logarithmic phase (2–7 days). Moreover, biomarkers that were selected by linear discriminant ana-
Following the increase of cell densities, the removal efficiency of Fe lysis (LDA) effect size (LEfSe) were significantly different among groups
increased to 69.7 %, and Mn to 25.6 %, the DO decreased from 4.9 to (Fig. S6). At 7 days, families Rhizobiaceae and Caulobacteraceae and the
1.6 mg/L. During days 8–12, when the bacteria were in a phase of genus Brevundimonas were the biomarkers.
stagnated growth, the removal efficiencies of Fe and Mn were 95.4 % The relationships between the dominant genera of the bacterial
and 72.3 %, respectively, the DO increased to 3.8 mg/L. The maximum community and environmental factors were analyzed by Spearman
removal efficiencies of Fe (99.8 %) and Mn (87.5 %) were reached at rank correlation (Fig. S7). Among the top ten abundance genera,
the late stage of treatment (after 12 days). Notably, the pH increased Flavobacterium, unidentified_Rhizobiaceae, and Brevundimonas were sig-
from 4.2–7.4 in the first 7 days and then stabilized at approximately 7 nificantly positively correlated with pH and Mn4+, and Massilia was
until the end of the treatment. In contrast to the inoculated flasks, the significantly positively correlated with Fe3+ (P < 0.05).
pH values did not change in the control groups (Fig. S9) and stabilized What’s more, Redundancy analysis (RDA) was applied to identify
at approximately 4.2. the key environmental factors that affected the structure of microbial
To further verify the removal ability of bacteria in the dynamic community (Fig. 4 and Table S6). The eigenvalues for the first two
study, column experiments were carried out (Supported information). canonical axes were 0.923 and 0.052. The correlation coefficients be-
The related results were shown in Table S7 and Figure S10. Different tween these two canonical axes and environment factors were 1.000
filter media were measured to identify the optimum filler material. The and 0.999, which respectively explained 92.3 % and 5.2 % of change in
removal ratios of Fe and Mn (Table S7) increased in the same order in bacterial structure, which also corresponded the relationship between

3
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

Fig. 1. Changes in acid mine drainage (AMD) treated with a bacterial Fe-Mn oxidizing consortium.
(a). Removal efficiencies (%) of Fe and Mn. (b).Changes in cell density (c) Changes in pH (d). Changes in dissolved oxygen (DO). The error bars represent the standard
deviations of the mean of three independent cultures.

distribution of the sediments (Fig. 5). When cultured for 2 days, the
surface of the sediment was primarily composed of some granular
particles, which resulted in a rough and porous surface structure,
thereby increasing its specific surface area. After 15 days, more ag-
glomerates and massive structures were on the surface of sediment,
leading to substantial changes in morphology. Furthermore, the surface
element compositions and the distribution of elements were char-
acterized by EDS line scan spectra and elemental mappings. The EDS
spectrum showed that the relative proportions of Fe and Mn in 2-day
sediment were 8.04 and 3.84 wt%, respectively (Fig. 5). By contrast,
after 15 days, Fe increased to 56.58 wt% and Mn increased to 11.58 wt
%. The elemental mappings further confirmed that Fe and Mn contents
in the sediment at 15 days were much higher than those at 2 days. In
addition, Fe and Mn were relatively densely and uniformly distributed
in fluorescence images (Fig. 5), which indicated that more Fe and Mn
were absorbed in the sediment after culture for 15 days.
XPS can identify the chemical states of elements and their surface
contents. Fig. 6a showed the Fe 2p XPS region for the sediments on days
2 and 15. Two peaks were observed at approximately 709 and 724 eV,
which respectively attributed to Fe 2p3/2 and Fe 2p1/2. The binding
energies of Fe 2p for Fe2+ species were located at 709.8 and 715.5 eV,
and the binding energies related to Fe3+ species were at 711.5, 713.5,
Fig. 2. Principal components analysis (PCA) of bacterial community structure
at different treatment stages. and 724.0 eV (Duan et al., 2017; Meng et al., 2016; Nenning and Fleig,
2019). Fig. 6b showed the Mn 2p spectra. For the 2-day sample, the
center of the Mn 2p3/2 peak was well fit by three peaks at 640.5, 641.7,
93.2 %, 5.3 % of the bacterial community structure and environment
and 645.6 eV, corresponding to the chemical states Mn2+, Mn3+, and
factor. Fig. 4 showed the biplots for variance partitioning, which in-
Mn4+ species, respectively (Baran et al., 2016; Beyreuther et al., 2006;
dicated that pH (F = 81.407, P = 0.002) and total Fe (F = 12.315,
Gang et al., 2013; Xia et al., 2018). When sampled on the 15th day, the
P = 0.002) explained the most variation in bacterial community
Mn 2p3/2 spectrum appeared as two peaks: 641.2 eV (Mn3+ species)
structure during the incubation.
and 645.0 eV (Mn4+ species) (Xia et al., 2018). And the peak at
653.0 eV was Mn4+ species in Mn 2p1/2 of the two samples (Tan et al.,
1991). The relative amount of Mn2+, Mn3+and Mn4+ were measured,
3.3. Characterization of the sediment
and the results were shown in Table 1. The O 1s XPS spectrum showed
two peaks (Fig. 6c), which belong to the M–OH (hydroxyl bonded to
The XRD patterns of sediments were illustrated in Fig. S8. The ab-
metal), M–O (oxygen bonded to metal) with binding energies of 531.1
sence of crystalline peaks in the XRD spectrum indicated that the bio-
and 529.6 eV, respectively (Lu et al., 2014).
logical Fe–Mn oxides were primarily in amorphous form. Moreover, the
The relative amounts of element species were measured (Table 1).
BET specific surface area values of the sediments at days 2 and 15 were
The surface concentration of Fe3+ was 67.94 % on the 2nd day and
108 and 121 m2/g, respectively.
75.76 % on the 15th day. The relative amounts of Mn2+, Mn3+, and
SEM was used to characterize the morphology and the element

4
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

Fig. 3. Weighted unifrac distance and Relative abundance in phylum and genus level. (a) Weighted unifrac distance and Relative abundance of phyla. (b) Relative
abundance of genera.

−COOH groups, CeH groups, and H2O, respectively (Ganesan et al.,

2019; Lu et al., 2014; Saroyan et al., 2019). In 2-day sediment, Fe–O–H
bending bond was at 880, but in 15-day sediment, it disappeared
completely (Qu et al., 2011). The presence of Mn–O–H and Fe–OH in-
dicated that Mn and Fe atoms of the sediment might have interacted
with O atoms via a coordination bond, which is consistent with the XPS
results.

4. Discussion

High concentrations of Fe and Mn in AMD are difficult to remove

and can cause an operational problem for the mass addition of alkali as
a chemical treatment. In this study, a novel strategy was explored to
remove Fe and Mn from AMD by using an acclimated Fe–Mn-oxidizing
consortium. Because of pre-incubation, the lag phase of bacterial
growth was very short (2 days). The logarithmic and stagnant phases
were from 3 to 7 days and from 8 to 12 days, respectively, and were the
phases in which most of the Fe and Mn were removed. The removal of
Fe and Mn depended primarily on the FMOB in the flask, and therefore,
Fig. 4. Redundancy analysis of bacterial community structure and environ- the increase of removal efficiencies lagged behind the growth of bac-
mental factors for three treatment stages (2 days, 7 days, 15 days). DO is ab- teria. Only when the abundance of FMOB increased was the bacteria
breviation of dissolved oxygen. capability of removing Fe and Mn evident. Moreover, the removal ef-
ficiency of Mn was also lower than that of Fe, which might be due to the
more stringent conditions required for the removal of Mn (Qin et al.,
Mn4+ in sediment of the second day were 18.89 %, 38.27 %, and 42.84
2009). In the natural environment, the abiotic oxidation of Mn (II) is
%, respectively. After 15 days, the concentrations of Mn3+ and Mn4+
thermodynamically unfavorable under acidic conditions, and it is very
increased to 47.54 % and 52.46 %, respectively, but the peak of Mn2+
slow even under alkaline conditions (Akob et al., 2014). However, Fe
disappeared. M−OH (Fe–OH, Mn−OH) was the primary form of O
(II) can be easily oxidized at pH > 5 and then precipitates as Fe(OH)3 as
species. However, the concentration of M−OH decreased from 83.14 %
the pH of incubation increases (Emerson et al., 2010; Fabisch et al.,
on days 2 to 77.74% on days 15, whereas that of M–O (Fe–O, Mn–O)
2013). Hence, the oxidation of Fe (II) resulted from the combination of
increased obviously from 16.86 % on days 2 to 22.26% on days 15.
bio-oxidation and chemical Fe oxidation, whereas Mn (II) oxidation was
FTIR spectra of sediments on days 2 and 15 were shown in Fig. 7,
primarily by bacteria. Furthermore, a previous study found that the
and similar peaks were observed in the two sediments. The major peaks
development time for MnOB was much longer than that for FeOB
of the two samples in the region of 500–4000 cm−1 at approximately
(Pacini et al., 2005). Therefore, it is reasonable that the Mn removal
1066, 1383, 1639, 2925, and 3412 were attributed to Mn–O–H, CeC,
lagged behind and was lower than that of Fe in this study. At the late

5
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

Fig. 5. SEM-EDS and element mapping diagram of the sediment ((a) SEM-EDS analysis of 2-day sediment and Fe/Mn distribution on the sediment, (b) SEM-EDS
analysis of 15-day sediment and Fe/Mn distribution on the sediment.).

stage of treatment, the removal of Fe and Mn slowed, because bacterial Table 1

activity declined and more precipitates covered the cells, obstructing The relative amounts of element species (%) in the surface of sediments.
the interactions between bacteria and heavy metals (Emerson et al., Samples Fe2+ Fe3+ Mn2+ Mn3+ Mn4+ M-O H M-O
2010; Zhang et al., 2018).
The pH is one of the important parameters that greatly affect the 2 days 32.06 67.94 18.89 38.27 42.84 83.14 16.86
15 days 24.24 75.76 / 47.54 52.46 77.74 22.26
feasibility of bacterial remediation because it influences the viability of
bacteria and the solubility of metal ions. The results demonstrated that
M–OH (M, metal) indicates Fe–OH and Mn–OH; M–O indicates Fe–O and Mn–O.
the acclimated bacteria could survive in the pH range 4–8, which means
the bacteria had a highly adaptability to pH changing. Moreover, a
and Mn removal from AMD. In turn, with the reduction in pH stress and
higher pH was benefit for bacteria survive and Fe and Mn removal (Fig.
decrease in Fe and Mn contents, alpha diversity index increased sig-
S2). Interestingly, the pH increased dramatically from 4.2 to 7.2 during
nificantly (P < 0.05), which might indicate that more types of bacteria
the AMD treatment. That might be an adaptive strategy for bacteria to
participated in the removal of Fe and Mn. Thus, it was believed that pH
survive in the extreme conditions of AMD. This result was consistent
was the most important environmental factor affecting the structure of
with the finding of Pacini et al. (2005) that reported a slight pH in-
the bacterial community, which could be further confirmed by the re-
crease in a biological treatment, which was due to CO2 consumption of
sults of RDA.
FeOB metabolism (Pacini et al., 2005). By producing chemical oxidants
So, what were the compositions and structures of bacterial com-
(H2O2 or %OH) some of MnOB can also modify the pH of local en-
munity in this treatment system? And what were the relationships be-
vironment (Akob et al., 2014). Moreover, the dominant genus Steno-
tween bacteria community and environment factors? Our results re-
trophomonas which could induce pH rising was also identified in this
flected that both composition and structure of bacteria community were
study (Barboza et al., 2015). Hence, Fe-Mn oxidizing consortium might
remarkably changed during the treatment processing. For example, at
play a pivotal role in pH increase, which would ultimately enhance Fe

Fig. 6. XPS spectra of sediment on days 2 (top) and 15 (bottom) of a 15-day incubation of acid mine drainage treated with a bacterial Fe-Mn oxidizing consortium. (a)
Fe 2p, (b) Mn 2p, and (c) O1 s. M–OH (M, metal) indicates Fe–OH and Mn–OH; M–O indicates Fe–O and Mn–O.

6
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

Fig. 7. FTIR spectra of sediments on days 2 (top) and 15 (bottom) of a 15-day incubation of acid mine drainage treated with a bacterial Fe-Mn oxidizing consortium.

the genus level, the relative abundance of Flavobacterium, with previous studies, the morphology of FMBO were rough and
Brevundimonas and Stenotrophomonas increased during the incubation, wrinkle which presented a higher specific surface area (Lu et al., 2019;
and they became dominant genera on days 7 and 15. Consistent with Saroyan et al., 2019). Moreover, the BET specific surface area values of
this result, Flavobacterium was commonly detected in Mn removal bio- the FMBO were higher than the previous reported FMBO, which sug-
filter and in Mn deposits, indicating the bacteria of this genus play gested that the FMBO might have more active sites for absorption (Lu
important roles in Mn oxidation and removal (Akob et al., 2014). Bre- et al., 2019). Besides that, the FMBO contained reactive oxygen func-
vundimonas had been demonstrated to be a main functional population tional groups (−OH and −COOH) that were also benefit to the ad-
in an acclimated Mn-oxidizing consortium (Zhou et al., 2016). Some sorption of positively charged metal cations (Lu et al., 2019). Further-
strains (MB-38 and AP5s2-K1) belonging to Brevundimonas were also more, the XPS results indicated that low and high valence ions of Fe and
found have ability to oxidize Mn (Cerrato et al., 2010; Santelli et al., Mn coexisted in the sediments. The high valence ions were formed by
2014). Similarly, members of Stenotrophomonas have been proved have oxidation, while the low valence ions were precipitated by adsorption
ability to remove Mn by a nonenzymatic pathway (Barboza et al., or coprecipitation of FMBO. After 15 days, the Fe and Mn contents of
2015). Furthermore, according to Spearman’s rank correlation, Flavo- sediments were significantly increased, and the surface of sediment
bacterium and Brevundimonas were significantly and positively corre- changed from a dispersed particle structure to a lamellar structure.
lated with Mn4+, further demonstrating the importance of these bac- Previous studies also demonstrated that biosynthetic Fe–Mn oxides
teria in Mn (II) oxidizing. could also potentially oxidize or absorb heavy metals because of their
In this study, Fe was effectively removed by the Fe-Mn consortium, colloidal properties and surface functional groups (Akob et al., 2014;
which implied the essential functions of bacterial in this treatment. Bai et al., 2017; Rand and Ranville, 2019). Such a process could also
Among the top 10 abundance bacteria, Thermomonas was the only de- explain the slightly higher removal efficiency than oxidation ratio of Fe
tected FeOB which had an increasing abundance during the treating and Mn in this study (Qin et al., 2009).
process (Zhang et al., 2018). Although another FeOB, Acidovorax, was Usually, the bacteria loss and changing flows were two major
also detected, the abundance was very low (1.35 % of total OTUs) (Su challenges that biological treatment has to overcome. In recent decades,
et al., 2018). The increase in the relative abundance of these two genera a vast amount of literature has been focused on this topic, and some
was the indication of their vital roles during the treatment. Besides, useful advices were supplied (Shin et al., 2019; Musa et al., 2016). In
bacteria of Brevundimonas were not only benefit for the manganese this study, to maintain bacterial community sustenance, three steps
oxidation, but also contributes to the removal of iron because of its were carried out. Firstly, a single functional strain was replaced by
iron-binding siderophores (Preston et al., 2014). mixed bacteria consortium to enhance the survival rates of bacteria.
After 15 days, the iron and manganese ions were mostly pre- After enrichment, the bacteria consortium has been acclimated for a
cipitated in the sediments. The characteristics of sediments were further long time to make sure that it has strong ability to survive in a
analyzed in this study. The XRD results showed the biological Fe–Mn changeable environment. Moreover, before it was incubated into the
oxides existed primarily in amorphous form. A typical crystalline form biofilter, the bacteria need inoculated in synthetic AMD medium to
may evolve over a relatively long time, and crystallites of the particles adapt the environment conditions of AMD and shorten the start-up
that were too small to be conclusively identified, so early formed bio- time. Thirdly, to overcome washout, ceramic particles were selected as
logical minerals were mostly amorphous (Akob et al., 2014; Belzile filling media to improve microorganism attachment within the bio-
et al., 2001). The amorphous characteristic of FMBO is beneficial for system. Immobilization technique is a vital way to maintain high mi-
higher heavy metal absorption because amorphous structure can pro- crobial density and protect cells against damages caused by the external
vide larger ion accessible surface area and richer redox-active centers environment. It can also increase the stability of bacteria against toxic
(Fabisch et al., 2013; Wang et al., 2019; Xiao et al., 2019). Consistent substances (Shin et al., 2019; Zhao et al., 2019). Once the biofilter is

7
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

mature, the microbial system in the biofilter is stable and has abilities to Akob, D.M., Bohu, T., Beyer, A., Schäffner, F., Händel, M., Johnson, C.A., Merten, D.,
resistant the changes of environment. In biofilter ecosystems, many Büchel, G., Totsche, K.U., Küsel, K., Lovell, C.R., 2014. Identification of Mn(II)-oxi-
dizing bacteria from a low-pH contaminated former uranium mine. Appl. Environ.
bacterial species own same biological functions, when destroyed by Microbiol. 80, 5086–5097. https://doi.org/10.1128/aem.01296-14.
environmental changes some bacteria may loss or died while others Auld, R.R., Mykytczuk, N.C., Leduc, L.G., Merritt, T.J., 2017. Seasonal variation in an acid
who have the same functional genes were being able to complement the mine drainage microbial community. Can. J. Microbiol. 63, 137–152. https://doi.
org/10.1139/cjm-2016-0215.
loss of functions (Harrison et al., 2014; Delgado-Baquerizo et al., 2016). Bai, Y., Chang, Y., Liang, J., Chen, C., Qu, J., 2016a. Treatment of groundwater con-
Hence, although the loss of bacteria and changes of flows were un- taining Mn(II), Fe(II), As(III) and Sb(III) by bioaugmented quartz-sand filters. Water
avoidable, the biofilter can still run stably for a long time. Res. 106, 126–134. https://doi.org/10.1016/j.watres.2016.09.040.
Bai, Y., Yang, T., Liang, J., Qu, J., 2016b. The role of biogenic Fe-Mn oxides formed in situ
To sum up, our results demonstrated that acclimated Fe-Mn oxi- for arsenic oxidation and adsorption in aquatic ecosystems. Water Res. 98, 119–127.
dizing bacteria have potential for AMD treatment. However, to be a https://doi.org/10.1016/j.watres.2016.03.068.
successful technology there were still some challenges to overcome. For Bai, Y., Jefferson, W.A., Liang, J., Yang, T., Qu, J., 2017. Antimony oxidation and ad-
sorption by in-situ formed biogenic Mn oxide and Fe-Mn oxides. J. Environ. Sci.
examples, optimizing the hydraulic retention time and shortening the
(China) 54, 126–134. https://doi.org/10.1016/j.jes.2016.05.026.
start-up time, which should be done prior to implementation of full- Baran, R., Valentin, L., Dzwigaj, S., 2016. Incorporation of Mn into the vacant T-atom
scale application. sites of a BEA zeolite as isolated, mononuclear Mn: FTIR, XPS, EPR and DR UV-Vis
studies. Phys. Chem. Chem. Phys. 18, 12050–12057. https://doi.org/10.1039/
c6cp01713d.
5. Conclusions Barboza, N.R., Amorim, S.S., Santos, P.A., Reis, F.D., Cordeiro, M.M., Guerra-Sa, R., Leao,
V.A., 2015. Indirect manganese removal by Stenotrophomonas sp. and Lysinibacillus sp.
An acclimated Fe–Mn-oxidizing consortium successfully removed Fe isolated from Brazilian mine water. Biomed Res. Int. 2015, 925972. https://doi.org/
10.1155/2015/925972.
and Mn from AMD. During the treatment process, pH increased, which Belzile, N., Chen, Y.W., Wang, Z.J., 2001. Oxidation of antimony III by amorphous iron
was benefit for Fe-Mn removal. Flavobacterium, Brevundimonas, and manganese oxyhydroxides. Chem. Geol. 174, 379–387. https://doi.org/10.1016/
Stenotrophomonans and Thermomonas were dominated genera, in- S0009-2541(00)00287-4.
Beyreuther, E., Thiele, C., Dörr, K., 2006. XPS investigation of Mn valence in lanthanum
dicating that they might play vital roles in Fe-Mn removal. Biogenic manganite thin films under variation of oxygen content. Phys. Rev. B 75, 155425.
sediments were amorphous, with a high specific surface area and re- https://doi.org/10.1103/physrevb.73.155425.
active oxygen functional groups that favored Fe and Mn removal in Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K.,
2010. QIIME allowsanalysisofhigh-throughputcommunity sequencing data. Nat.
turn. Ceramic particle was the optimal filling material for biofilter Methods 7, 335–336. https://doi.org/10.1038/nmeth.f.303.
study. In addition, the removal of Fe and Mn was not only by bio-oxi- Cerrato, J.M., Falkinham 3rd, J.O., Dietrich, A.M., Knocke, W.R., McKinney, C.W.,
dation and chemical oxidation but also by the absorption or co-pre- Pruden, A., 2010. Manganese-oxidizing and -reducing microorganisms isolated from
biofilms in chlorinated drinking water systems. Water Res. 44, 3935–3945. https://
cipitation of biosynthetic Fe-Mn oxides. Thus, the Fe–Mn-oxidizing
doi.org/10.1016/j.watres.2010.04.037.
consortium has potential application in AMD remediation. Dale, J.G., Stegemeier, J.P., Kim, C.S., 2015. Aggregation of nanoscale iron oxyhydr-
oxides and corresponding effects on metal uptake, retention, and speciation: I. Ionic-
Authorship contributions strength and pH. Geochimica Et Cosmochimica Acta 148, 100–112. https://doi.org/
10.1016/j.gca.2014.08.029.
Delgado-Baquerizo, M., Giaramida, L., Reich, P.B., Khachane, A.N., Hamonts, K.,
Conception and design of study: Dongmei Hou, Lin Luo. Edwards, C., Lawton, L.A., Singh, B., 2016. Lack of functional redundancy in the
Acquisition of data: Dongmei Hou, Pan Zhang. relationship between microbial diversity and ecosystem functioning. J. Ecol. 104,
936–946. https://doi.org/10.1111/1365-2745.12585.
Analysis and/or interpretation of data: Dongmei Hou, Dongning Duan, J., Chen, S., Zhao, C., 2017. Ultrathin metal-organic framework array for efficient
Wei, Jiachao Zhang. electrocatalytic water splitting. Nat. Commun. 8, 15341. https://doi.org/10.1038/
Drafting the manuscript: Dongmei Hou. ncomms15341.
Emerson, D., Fleming, E.J., McBeth, J.M., 2010. Iron-oxidizing bacteria: an environ-
Revising the manuscript: Dongmei Hou, Binghua Yan, Yaoyu Zhou, mental and genomic perspective. Annu. Rev. Microbiol. 64, 561–583. https://doi.
Linying Cao. org/10.1146/annurev.micro.112408.134208.
Fabisch, M., Beulig, F., Akob, D.M., Kusel, K., 2013. Surprising abundance of Gallionella-
related iron oxidizers in creek sediments at pH 4.4 or at high heavy metal con-
Declaration of Competing Interest
centrations. Front. Microbiol. 4, 390. https://doi.org/10.3389/fmicb.2013.00390.
Fan, X., Wang, S.T., Chun-Yan, L.I., 2016. The isolation and identification of an efficient
Fe/Mn-oxidizing bacterial strain P1, and the optimization of its oxidizing conditions.
We declare that we do not have any commercial or associative in-
Biotechnol. Bull. 32, 175–183. https://doi.org/10.13560/j.cnki.biotech.bull.1985.
terest that represents a conflict of interest in connection with the work 2016.04.023.
submitted. Ganesan, P., Rathnakumar, K., Brita Nicy, A., Velayutham, P., Jeyashakila, R., Anand, S.,
2019. Functional and structural characteristics of extruded snack product in-
corporated with fish powder from a lean fish. Res. J. Biotechnol. 14, 9.
Acknowledgements Gang, L., Feng, B., Song, C., Wang, K., Song, J., 2013. Promoting effect of zirconium
doping on Mn/ZSM-5 for the selective catalytic reduction of NO with NH3. Fuel 107,
This study was funded by National Key research and development 217–224. https://doi.org/10.1016/j.fuel.2013.01.050.
Gibert, O., Pablo, J.D., Cortina, J.L., Ayora, C., 2002. Treatment of acid mine drainage by
Projects of China (2016YFC0403002), Hunan Provincial Key Research sulphate-reducing bacteria using permeable reactive barriers: a review from labora-
and Development (2018WK4007, 2019SK2281), Research and innova- tory to full-scale experiments. Rev. Environ. Sci. Biotechnol. 1, 327–333. https://doi.
tion project for graduate students of hunan province (CX20190504), org/10.1023/A:1023227616422.
Han, W., Yue, Q., Wu, S., Zhao, Y., Gao, B., Li, Q., Wang, Y., 2013. Application and
National Natural Science Foundation of China (21806170 and advantages of novel clay ceramic particles (CCPs) in an up-flow anaerobic bio-filter
21906050). (UAF) for wastewater treatment. Bioresour. Technol. 137, 171–178. https://doi.org/
10.1016/j.biortech.2013.03.124.
Harrison, P.A., Berry, P.M., Simpson, G., Haslett, J.R., Blicharska, M., Bucur, M., Dunford,
Appendix A. Supplementary data R., Egoh, B., Garcial-Llorente, M., Geamǎnǎ, N., Geertsema, W., Lommelen, E.,
Meiresonne, L., Turkelboom, F., 2014. Linkages between biodiversity attributes and
Supplementary material related to this article can be found, in the ecosystem services: a systematic review. Ecosyst. Serv. 9, 191–203. https://doi.org/
10.1016/j.ecoser.2014.05.006.
online version, at doi:https://doi.org/10.1016/j.jhazmat.2020.122631.
Hou, D.M., Zhang, P., Zhang, J.C., Zhou, Y.Y., Yang, Y., Mao, Q.M., Tsang, D.C.W., Nunez-
Delgado, A., Luo, L., 2019. Spatial variation of sediment bacterial community in an
References acid mine drainage contaminated area and surrounding river basin. J. Environ.
Manage. 251, 109542. https://doi.org/10.1016/j.jenvman.2019.109542.
Johnson, D.B., Hallberg, K.B., 2005. Acid mine drainage remediation options: a review.
Aguinaga, O.E., Mcmahon, A., White, K.N., Dean, A.P., Pittman, J.K., 2018. Microbial Sci. Total Environ. 338, 3–14. https://doi.org/10.1016/j.scitotenv.2004.09.002.
community shifts in response to acid mine drainage pollution within a natural wet- Kappler, A., Newman, D.K., 2004. Formation of Fe(III)-minerals by Fe(II)-oxidizing
land ecosystem. Front. Microbiol. 9, 1445. https://doi.org/10.3389/fmicb.2018. photoautotrophic bacteria. Geochimica et Cosmochimica Acta 68, 1217–1226.
01445. https://doi.org/10.1016/j.gca.2003.09.006.

8
D. Hou, et al. Journal of Hazardous Materials 396 (2020) 122631

Kefeni, K.K., Msagati, T.A.M., Mamba, B.B., 2017. Acid mine drainage: Prevention, oxidation in ashumet pond, a historically polluted freshwater pond undergoing re-
treatment options, and resource recovery: a review. J. Clean. Prod. 151, 475–493. mediation. Geomicrobiol. J. 31, 605–616. https://doi.org/10.1080/01490451.2013.
https://doi.org/10.1016/j.jclepro.2017.03.082. 875605.
Larson, L.N., Sánchez-España, J., Kaley, B., Sheng, Y., Bibby, K., Burgos, W.D., 2014. Saroyan, H., Kyzas, G., Deliyanni, E., 2019. Effective dye degradation by graphene oxide
Thermodynamic controls on the kinetics of microbial low-pH Fe(II) oxidation. supported manganese oxide. Processes 7, 40. https://doi.org/10.3390/pr7010040.
Environ. Sci. Technol. 48, 9246–9254. https://doi.org/10.1021/es501322d. Shin, D.C., Kim, J.S., Park, C.H., 2019. Study on physical and chemical characteristics of
Li, C., Wang, S., Du, X., Cheng, X., Fu, M., Hou, N., Li, D., 2016. Immobilization of Iron- microorganism immobilized media for advanced wastewater treatment. J. Water
and Manganese-oxidizing Bacteria with a Biofilm-forming Bacterium for the Effective Process. Eng. 29, 100784. https://doi.org/10.1016/j.jwpe.2019.100784.
Removal of Iron and Manganese from Groundwater. Bioresour. Technol. 220, 76–84. Su, J., Liang, D., Lian, T., 2018. Comparison of denitrification performance by bacterium
https://doi.org/10.1016/j.biortech.2016.08.020. Achromobacter sp. A14 under different electron donor conditions. Chem. Eng. J. 333,
Lu, J., Liu, H., Zhao, X., Jefferson, W., Cheng, F., Qu, J., 2014. Phosphate removal from 320–326. https://doi.org/10.1016/j.cej.2017.09.129.
water using freshly formed Fe–Mn binary oxide: adsorption behaviors and mechan- Tan, B.J., Klabunde, K.J., Sherwood, P.M.A., 1991. XPS studies of solvated metal atom
isms. Colloids Surf. A Physicochem. Eng. Asp. 455, 11–18. https://doi.org/10.1016/j. dispersed (SMAD) catalysts. Evidence for layered cobalt-manganese particles on
colsurfa.2014.04.034. alumina and silica. J. Am. Chem. Soc. 113, 855–861. https://doi.org/10.1021/
Lu, K., Wang, T., Zhai, L., Wu, W., Dong, S., Gao, S., Mao, L., 2019. Adsorption behavior ja00003a019.
and mechanism of Fe-Mn binary oxide nanoparticles: adsorption of methylene blue. Wang, Y., Chen, X., Zhou, H., 2018. Disentangling effects of temperature on microbial
J. Colloid Interface Sci. 539, 553–562. https://doi.org/10.1016/j.jcis.2018.12.094. community and copper extraction in column bioleaching of low grade copper sulfide.
Meng, F.L., Wang, Z.L., Zhong, H.X., Wang, J., Yan, J.M., Zhang, X.B., 2016. Reactive Bioresour. Technol. 268, 480–487. https://doi.org/10.1016/j.biortech.2018.08.031.
multifunctional template-induced preparation of Fe-N-doped mesoporous carbon Wang, G., Liu, Y., Wu, M., Zong, W., Yi, X., Zhan, J., Liu, L., Zhou, H., 2019. Coupling the
microspheres towards highly efficient electrocatalysts for oxygen reduction. Adv. phenolic oxidation capacities of a bacterial consortium and in situ-generated man-
Mater. 28, 7948–7955. https://doi.org/10.1002/adma.201602490. ganese oxides in a moving bed biofilm reactor (MBBR). Water Res. 166, 115047.
Musa, S., Ariff, N.A., Kadir, M.N.A., Denan, F., 2016. Simplified method for groundwater https://doi.org/10.1016/j.watres.2019.115047.
treatment using dilution and ceramic filter. IOP Conf. Series: Mater. Sci. Eng. 136, Wu, X., Zhang, H., Chen, J., Shang, S., Wei, Q., Yan, J., Tu, X., 2016. Comparison of the
012074. https://doi.org/10.1088/1757-899X/136/1/012074. fecal microbiota of dholes high-throughput Illumina sequencing of the V3–V4 region
Nenning, A., Fleig, J., 2019. Electrochemical XPS investigation of metal exsolution on of the 16S rRNA gene. Appl. Microbiol. Biotechnol. 100, 3577–3586. https://doi.org/
SOFC electrodes: controlling the electrode oxygen partial pressure in ultra-high-va- 10.1007/s00253-015-7257-y.
cuum. Surf. Sci. 680, 43–51. https://doi.org/10.1016/j.susc.2018.10.006. Xia, F., Song, Z., Xin, L., Xi, L., Yang, Y., Zhang, Q., Peng, J., 2018. Improved catalytic
Pacini, V.A., Maria Ingallinella, A., Sanguinetti, G., 2005. Removal of iron and manganese activity and N2 selectivity of Fe–Mn–Ox catalyst for selective catalytic reduction of
using biological roughing up flow filtration technology. Water Res. 39, 4463–4475. NO by NH3 at low temperature. Res. Chem. Intermed. 44, 2703–2717. https://doi.
https://doi.org/10.1016/j.watres.2005.08.027. org/10.1007/s11164-018-3255-x.
Preston, J., Smith, A.D., Schofield, E.J., Chadwick, A.V., Jones, M.A., Watts, J.E., 2014. Xiao, Z., Bao, Y., Li, Z., Huai, X., Wang, M., Liu, P., Wang, L., 2019. The construction of
The effects of Mary Rose conservation treatment on iron oxidation processes and hollow cobalt-nickel phosphate nanocages through a controllable etching strategy for
microbial communities contributing to acid production in marine archaeological high supercapacitor performances. Appl. Energy Mater. 2, 1089–1092. https://doi.
timbers. PLoS One 9, e84169. https://doi.org/10.1371/journal.pone.0084169. org/10.1021/acsaem.8b01627.
Qin, S.Y., Ma, F., Huang, P., Yang, J.X., 2009. Fe (II) and Mn (II) removal from drilled well Zhang, Z., Lai, J., Yin, H., Feng, X., Tan, W., Liu, F., 2015. Absorption mechanisms of
water: a case study from a biological treatment unit in Harbin. Desalination 245, Cu2+ on a biogenic bixbyite-like Mn2O3 produced by Bacillus CUA isolated from soil.
183–193. https://doi.org/10.1016/j.desal.200. Geochem. Trans. 16, 5. https://doi.org/10.1186/s12932-015-0020-6.
Qu, Q., Yang, S., Feng, X., 2011. 2D sandwich-like sheets of iron oxide grown on graphene Zhang, L., Sun, H., Zhang, X.X., Ren, H., Ye, L., 2018. High diversity of potential nitrate-
as high energy anode material for supercapacitors. Adv. Mater. 23, 5574–5580. reducing Fe(II)-oxidizing bacteria enriched from activated sludge. Appl. Microbiol.
https://doi.org/10.1002/adma.201103042. Biotechnol. 102, 4975–4985. https://doi.org/10.1007/s00253-018-8961-1.
Rand, L.N., Ranville, J.F., 2019. Characteristics and stability of incidental iron oxide Zhao, C.S., Liu, J., Deng, Y.H., Tian, Y.Y., Zhang, G.J., Liao, J.L., Yang, J.J., Yang, Y.Y.,
nanoparticles during remediation of a mining-impacted stream. Environ. Sci. Liu, N., Sun, Q., 2019. Uranium(Ⅵ) adsorption from aqueous solutions by micro-
Technol. 53, 11214–11222. https://doi.org/10.1021/acs.est.9b03036. organism-graphene oxide composites via an immobilization approach. J. Clean. Prod.
Ren, L., Cai, C., Zhang, J., Yang, Y., Wu, G., Luo, L., Huang, H., Zhou, Y., Qin, P., Yu, M., 236, 117624. https://doi.org/10.1016/j.jclepro.2019.117624.
2018. Key environmental factors to variation of ammonia-oxidizing archaea com- Zhou, H., Pan, H., Xu, J., Xu, W., Liu, L., 2016. Acclimation of a marine microbial con-
munity and potential ammonia oxidation rate during agricultural waste composting. sortium for efficient Mn(II) oxidation and manganese containing particle production.
Bioresour. Technol. 270, 278–285. https://doi.org/10.1016/j.biortech.2018.09.042. J. Hazard. Mater. 304, 434–440. https://doi.org/10.1016/j.jhazmat.2015.11.019.
Santelli, C.M., Chaput, D.L., Hansel, C.M., 2014. Microbial communities promoting Mn(II)