Synthesis and Performance of Iron Oxide-based Porous Ceramsite

download Synthesis and Performance of Iron Oxide-based Porous Ceramsite

of 9

Transcript of Synthesis and Performance of Iron Oxide-based Porous Ceramsite

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    1/9

    Synthesis and performance of iron oxide-based porous ceramsite in a

    biological aerated filter for the simultaneous removal of nitrogen and

    phosphorus from domestic wastewater

    Teng Bao a,b, Tianhu Chen a,⇑, Juan Tan a, Marie-Luise Wille c, Dan Zhu a, Dong Chen a, Yunfei Xi b,d,⇑

    a Laboratory for Nanominerals and Environmental Material, School of Resource and Environmental Engineering, Hefei University of Technology, He Fei, Chinab Institute for Future Environments, Queensland University of Technology (QUT), 2 George Street, GPO Box 2434, Brisbane, QLD 4000, Australiac Institute of Health & Biomedical Innovation, Queensland University of Technology (QUT), Brisbane, QLD 4000, Australiad School of Chemistry, Physics and Mechanical Engineering, Science and Engineering Faculty, Queensland University of Technology (QUT), 2 George Street, GPO Box 2434, Brisbane,

    QLD 4000, Australia

    a r t i c l e i n f o

     Article history:

    Received 21 January 2016

    Received in revised form 9 May 2016

    Accepted 9 May 2016

    Available online 10 May 2016

    Keywords:

    Iron oxide-based porous ceramsite

    Nitrogen and phosphorus removal

    Nitrification

    Denitrification

    Biological aerated filters

    a b s t r a c t

    A novel iron oxide-based porous ceramsite (IPC) was synthesized and applied as a microbial biofilm car-

    rier in a biological aerated filter (BAF) to treat domestic wastewater and compared to commercially avail-

    able ceramsite (CAC). The results indicate that IPC has a higher porosity in comparison to CAC. The

    uniformity and interconnectivity of pores, as well as the rough surface of the IPC are suitable for microbial

    biofilm growth. Biofilm growth occurs in the internal pores of the media and promotes nitrogen and

    phosphorus removal. The effect of air-water ratio (A/W) on the removal of total organic carbon, ammonia

    nitrogen, total nitrogen, and phosphorus were investigated. The results show that the performance of IPC

    BAF is much better than CAC BAF. For instance, at an A/W ratio of 3:1, the total nitrogen removal was

    46.26% with IPC and 15.64% with CAC, and the PO43 removal was 72.25% with IPC compared with only

    33.97% with CAC. An analysis of the microbial community in the IPC BAFs by polymerase chain reaction

    denaturing gradient gel electrophoresis identified  Dechloromonas  sp.,  Sphaerotilus  sp., and  Nitrospira  sp.

    microbes. The diversity on microbial population, along with the attached growth benefit from the mor-

    phological properties of IPC, allows enhancement in the simultaneous nitrification and denitrification

    performance in IPC BAF. Hence, IPC can be considered a very effective novel media material in BAF for

    the simultaneous removal of nitrogen and phosphorus.

      2016 Elsevier B.V. All rights reserved.

    1. Introduction

    Eutrophication can occur when excessive nitrogen and phos-

    phorus reach water bodies through runoff or discharge of wastew-

    ater   [1]. Eutrophication can lead to the deterioration in water

    quality damaging aquatic ecosystems. An outbreak of harmful cya-

    nophytes, commonly known as cyanobacteria, will eventually

    result in increased mortality of aquatic organisms. Hence, the dis-

    charge of treated wastewater must meet effluent quality standards

    for phosphorus and nitrogen. Novel wastewater treatments are

    required for the simultaneous removal of phosphorus and nitrogen

    (SPN) since conventional treatments have commonly been

    designed to focus on removal of one nutrient. Traditionally, tech-

    nologies such as biological and chemical treatment for phosphate

    or nitrate have been widely used   [2,3]. Chemical treatment

    employs a metal salt to coagulate and remove phosphate by sedi-

    mentation. The excess amount of sludge produced is a major lim-

    itation to this process  [2]. Adsorption is the most widely used

    and suitable method for phosphorus removal. Of all the adsorbents,

    iron oxides and untreated clays are considered the most effective

    sorbent materials   [4,5]. Previous studies have shown that raw

    goethite, nano zero-valent iron (Fe0), and hematite are very effec-

    tive materials adsorbing phosphorus  [6,7]. Simultaneous nitrifica-

    tion and denitrification (SND) for the removal of nitrogen can be

    achieved through biological processes. SPN removal from wastew-

    ater can be achieved by the combination of different biological

    and/or chemical technologies (i.e., sequential treatment). A biolog-

    ical aerated filter (BAF) is a fixed-film biosystem with a small foot-

    print that employs filter media with a high specific surface area

    and porosity. The filter can promote in-growth biofilm(s) for

    wastewater treatment. BAFs can reduce chemical oxygen demand

    http://dx.doi.org/10.1016/j.seppur.2016.05.011

    1383-5866/  2016 Elsevier B.V. All rights reserved.

    ⇑ Corresponding authors at: School of Chemistry, Physics and Mechanical

    Engineering, Science and Engineering Faculty, Queensland University of Technology

    (QUT), 2 George Street, GPO Box 2434, Brisbane, QLD 4000, Australia (Y. Xi).

    E-mail addresses:   [email protected] (T. Chen),  [email protected] (Y. Xi).

    Separation and Purification Technology 167 (2016) 154–162

    Contents lists available at  ScienceDirect

    Separation and Purification Technology

    j o u r n a l h o m e p a g e :   w w w . e l s e v i e r . c o m / l o c a t e / s e p p u r

    http://dx.doi.org/10.1016/j.seppur.2016.05.011mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.seppur.2016.05.011http://www.sciencedirect.com/science/journal/13835866http://www.elsevier.com/locate/seppurhttp://www.elsevier.com/locate/seppurhttp://www.sciencedirect.com/science/journal/13835866http://dx.doi.org/10.1016/j.seppur.2016.05.011mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.seppur.2016.05.011http://crossmark.crossref.org/dialog/?doi=10.1016/j.seppur.2016.05.011&domain=pdf

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    2/9

    (COD), suspended solids and ammonia [8,9]. Microbial biofilm sup-

    ported filter media, which are made of raw clay, wastewater

    sludge, construction waste and raw zeolite, have been evaluated

    in recent years and have achieved good results for nitrogen

    removal  [2,10,11]. Dissolved oxygen (DO) plays a key role during

    the BAF process. Denitrifiers are active in areas of very low DO con-

    centration whereas nitrifiers are active in areas of high DO concen-

    tration [12]. SND has been proven to be the most effective way to

    remove total nitrogen (TN) which occurs naturally inside the

    microbial biofilms of filter media  [13]. Porous filter media involv-

    ing a mixed population and upholding oxygen diffusion limits cre-

    ate both anoxic and aerobic micro-environments. As such, the filter

    media allows the sequential use of electron acceptors, such as oxy-

    gen and nitrate. Such mass transfer leads to the stratification of 

    microbial species in the microbial biofilm of filter media  [15,16].

    BAF technology is applied to wastewater treatment; hence, the

    selection of filter media plays a significant role in maintaining a

    variety of microbial populations and a high amount of active bio-

    mass. Recently, some waste materials, such as waste ceramics

    and polyethylene plastic, have been utilized as filter media for bio-

    logical aerobic filters, providing a new approach for disposing the

    waste sawdust   [2]. This strategy serves as a ‘‘win–win” solution

    by treating wastes with wastes. Furthermore, several new filter

    media for BAF have been recently investigated to enhance

    biodegradation capability. For example, combined up-flow anaero-

    bic biofilters and up-flow BAF that employ filter media prepared by

    sludge, coal cinder, and straw have been used to treat synthetic

    wastewater [17]. These filter media achieved favorable results for

    nitrogen removal. However, only a few studies investigated in

    detail SPN with BAF.

    In China, commercially available ceramsite (CAC) has been

    newly used in filter media for domestic wastewater treatment.

    However, such material was reported to exhibit extremely low

    nitrogen and phosphorus removal efficiencies in BAFs. This result

    is probably due to the poor affinity of CAC to microbial biofilm bio-

    mass, which is caused by biocompatibility and suboptimal surface

    hydrophilicity   [2,11]. For this reason, developing a novel filtermedium as a substitute for CAC in the simultaneous removal of 

    nitrogen and phosphorus must be accomplished as quickly as

    possible.

    Previous studies reported that sawdust was treated at 700 C to

    produce a desirable porous material. The formation of interior

    pores improves microbial growth. It was reported that hematite

    and palygorskite calcined at 700 C can effectively absorb phos-

    phorus   [2,5]. In this study, iron oxide-based porous ceramsite

    (IPC) with goethite, sawdust, and palygorskite were synthesized

    by calcination in an O2 atmosphere, used as microbial biofilm sup-

    port in BAF and compared to commercially available ceramsite

    (CAC) [14]. The properties of IPC, CAC and microbial characteristics

    were characterized by scanning electron microscopy (SEM),

    porosimetry analysis (PM) and micro-computed tomography(Micro-CT). The effect of air-water ratio (A/W) on the removal of 

    total organic carbon (TOC), ammonia nitrogen (NH3-N), total nitro-

    gen (TN) and phosphorus (P) was investigated using both supports.

    Finally, the relationship between protozoa and metazoa as well as

    the characteristics of microbial biofilms were determined using

    polymerase chain reaction denaturing gradient gel electrophoresis

    (PCR-DGGE) of the microbial community of the IPC BAF.

    2. Materials and methods

     2.1. Characterization and analytical methods

    Details of reagents and preparation of the proposed filtermedia-IPC are provided in the supplementary material. IPC and

    CAC surface morphologies were examined using a Scanning Elec-

    tron Microscope (SEM, Philips XL30 ESEM-Netherlands Philips

    Company). The physical characteristics of IPC and CAC were mea-

    sured in accordance with the sandstone pore structure method of 

    image analysis. Rubber casting experiments were used to generate

    quantitative size and shape data from pores in the thin section

    (d = 3 cm) [2,14,18]. The porosities of IPC and CAC specimens were

    determined using Micro-CT (l

    CT 40, Scanco Medical, Brüttisellen,

    Switzerland). The two samples were scanned in air at 70 kV and

    57lA, with an isotropic voxel size of 12 lm. A region of interest

    (ROI) was defined for each gray-value image, which corresponded

    to the exact circumference of the sample. A binary threshold was

    applied to each sample and Scanco built in reconstruction algo-

    rithms were used to derive the porosity and to construct the 3D

    image [19–21].

    Microscopic observations for protozoan and metazoan popula-

    tion growth were carried out using a U-RFL-T Olympus Biological

    Microscope (Olympus Corporation, Tokyo, Japan). The multi-point

    BET surface area of IPC and CAC was measured using a Quan-

    tachrome Nova 3000e automated surface area analyzer. The

    growth of biofilm was determined according methods available

    in the literature [2,13]. The microbial populations in IPC BAF were

    analyzed by using domain-specific F341 R907 primers and PCR 

    amplification of 16S rRNA gene fragments for genomic DNA from

    bacteria. The PCR products were applied in the DGGE analysis

    using a BioRad Dcode system (BioRad Co. Ltd., USA). A DGGE gel

    of 8% polyacrylamide with a linear denaturing gradient ranging

    from 35% to 60% (100% denaturing gradient contains 7 M urea

    and 40% formamide) was used. Electrophoresis was conducted at

    a constant voltage of 75 V in 1 TAE buffer at 65 C for 14 h. The

    gels were then stained with 5% Gold view in 1 TAE buffer for

    40 min and imaged using UV trans-illumination. Selected DGGE

    bands were excised using the SK1131 kit (Sangon Biotech (Shang-

    hai) Co., Ltd) and re-amplified by PCR with the aforementioned pri-

    mers. The PCR products were sequenced at Sangon Biotech Co., Ltd

    (Shanghai, China). The obtained sequences were submitted and

    compared to the reference microorganisms in the GenBank data-base using the BLAST program [22,23].

     2.2. Operating conditions of IPC BAF and CAC BAF 

    A schematic diagram of the experimental set-up is provided in

    the supplementary material. Biofiltration was initiated by intro-

    ducing seed sludge obtained from the Hefei City wastewater treat-

    ment plan into BAFs. Each experiment was divided into 3 stages for

    each BAF. During each test stage, the operating conditions of both

    BAFs were identical (see Table 1). A stable hydraulic retention time

    (HRT) of 7 h was used. Three air-water (A/W) ratios of 1:1, 3:1, and

    6:1, respectively, were tested. The two BAFs were monitored for

    4 months after the initiation of biofiltration. Both BAFs were oper-

    ated in summer at wastewater temperatures ranging from 26 C to

    30 C. Samples were analyzed for NH3-N and P in accordance with

    Chinese EPA standards [2,24]. A TOC/TN (Jena Multi N/C 2100) ana-

    lyzer was used to measure TOC and TN. Dissolved oxygen (DO)

    concentrations were measured using a portable digital DO meter

    (Oxi-315, Hao-Yang Biotech Company, Shanghai, China).

    3. Results and discussion

     3.1. Characteristics of IPC and CAC 

    Table 1 shows the elemental properties of IPC and CAC. IPC pos-

    sesses a higher surface area and porosity than those of CAC. Lower

    apparent and bulk densities were also observed in the IPC. All of these properties are associated with the chemical compositions

    T. Bao et al. / Separation and Purification Technology 167 (2016) 154–162   155

    http://-/?-http://-/?-

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    3/9

    of the materials. Sawdust comprised 20% of IPC, which is propitious

    for the manufacture of lightweight porous IPC because of the

    decomposition of sawdust during calcination (Section 1.2, Support-

    ing information). The lower apparent and bulk densities of IPC may

    be ascribed to the addition of sawdust. The gas-forming compo-

    nents from the sawdust also rendered the IPC surface rougher than

    the CAC surface, generating a larger surface area in the former. Fur-

    thermore, the microbial biofilm biomass was greater in amount on

    the IPC than on the CAC because of the former’s rough surface. For

    IPC and CAC, the surface area reached over 79 m2 g1 and 2 m2 g1,

    respectively (Table S1). The rough surface distributed with largepores was beneficial for the microbial biofilms to grow onto the

    IPC. These special surface morphology properties also demon-

    strated that IPC was fit as filter material.

     3.2. Micro-CT based pore characterization

    Micro-CT is a powerful tool for non-destructive and repro-

    ducible analysis of the architecture of porous materials based on

    3D geometrical considerations   [25–27]. In this study, micro-CT

    was used to analyze structural characteristics and porosity of IPC

    and CAC.   Fig. 1A shows the 3D micro-CT reconstruction of IPC.

    The red spherical particles in IPC sections denote calcined molded

    composite particles (calcined palygorskite, calcined sawdust, and

    calcined goethite) and pores can also be observed. This 3D imagedemonstrates that IPC has considerable external porosity by vol-

    ume. Fig. 1B shows a 0.72 mm thick cross-section (stack of 20 scan

    slices) of IPC with large pores of diameters approximately 100–

    500lm.  Fig. 1C and D shows a 3D reconstruction and 0.72 mm

    thick cross-section (stack of 20 scan slices) of CAC and CAC filled

    with epoxy, respectively. The porosity was determined to be

    3.46% ± 2.05%. The formation of liquid phases at temperatures

    P1000 C reduces the number of pores in CAC due to a higher per-

    centage of fine particles. This characteristic in CAC hinders the ini-

    tiation of cracks and results in a low porosity  [28].

    Porosity is a key factor in ensuring successful nitrogen removal

    in a given BAF process due to the uneven distribution of DO and

    substrate inside the IPC’s pores or on the rough surface. This fea-

    ture allows simultaneous proliferation of nitrifying and denitrify-ing bacteria. Micro-organisms are immobilized in the IPC’s pores

    which maintain a high biomass concentration while retaining a

    high hydraulic loading rate   [13]. Micro-CT analysis shows that

    when sintered at 700 C, the IPC has a porosity of 40.86% ± 3.26%

    with many unevenly distributed pores (100–500 lm). In principle,

    this characteristic is suitable for the growth of attached micro-

    organisms on rougher surfaces  [13].

     3.3. Cast thin section of IPC and CAC 

    Fig. 2 shows the internal pore structures of IPC and CAC. The IPC

    porosity (40.86% ± 3.26%) rendered the material ideal for microbial

    biofilm in-growth (Fig. 2A). By contrast, CAC possessed much lower

    porosity (3.46% ± 2.05%), which limits microbial growth (Fig. 2C)[14]. Microbial biofilm growth can form on external surfaces and

    within internal pores of the IPC; a rich supply of organic matter

    and nutrient substance allows these microbial biofilms to thrive.

    The microbial biofilm excretes slimy, glue-like exopolymeric sub-

    stances that further improve microbial anchoring to the IPC  [29].

    DO affects the distribution, nitrification, and denitrification rates

    of the microbial biofilm of the IPC as follows. Autotrophs are

    mainly located on the outer layer of the microbial biofilm to meet

    a fundamental requirement for converting NH4+–N. By contrast, the

    location of heterotrophs in the microbial biofilm is less restricted

    because the microorganisms can occupy the microbial biofilm cen-

    ter. In this case, the heterotrophs may grow with an organic sub-strate during the feast phase with the use of NO3

    –N as electron

    acceptor  [30]. Heterotrophs may also grow on the outer layers of 

    the microbial biofilm where DO can be used as electron acceptor

    [30]. The concentration of DO first decreased rapidly within the

    microbial biofilm thickness and then decreased gradually to zero

    at the outer region of the microbial biofilm [31]. Nitration occurred

    outside the microbial biofilm. Given the limited oxygen diffusion,

    gradients of DO concentrations within the microbial biofilm

    became possible. Deep within the microbial biofilm, oxygen trans-

    fer may be blocked, and large external oxygen consumption may

    occur, thus producing anoxic zones. Denitrifying bacteria predom-

    inantly enabled anoxic denitrification within the microbial biofilm.

    Hypoxic microbial biofilms were found within the pores. SND was

    also observed both inside and outside of the flocs  [32–35].Previous research has shown that staining epoxy resin can be

    used to impregnate the pore space of rock samples. This method

    is a very useful technique for identifying mineral components of 

    sedimentary rocks [36,37]. However, few reports have investigated

    the porosity of filter media by using staining epoxy resin. The thin

    sections of CAC and IPC, as depicted in  Fig. 2, were analyzed to

    obtain direct evidence of microorganisms in the porous structure

    of both materials. The rubber casting experiments were used to

    generate quantitative size and shape data from pores in the thin

    section [36,37]. The intergranular pore textures observed in a thin

    section and the red netted texture sections in Fig. 2A are identified

    as calcined palygorskite, goethite (hematite) and sawdust (biomass

    residues), while the gray area (arrowed) represents internal pores

    of the IPC. IPC have a greater variety of internal pores than guest

    microbial biofilm [2]. In addition, the size of those interconnected

    pores is about 100–500 lm (Fig. 2A), which is consistent with the

    results of Micro-CT analyses. It is well known that most microbes

    are approximately 0.5 lm in diameter. Thus, microbes can achieve

    a sustained growth of bacterial population in these open pores. In

    Fig. 2B, the blue section was obtained by vacuum impregnation

    with blue-dyed epoxy (arrowed) in IPC. Fully interconnected pores

    with a diameter of 100–500 lm between the narrowest parts can

    be observed. In fact,  Fig. 2C and D shows that blue-dyed epoxy in

    the porosity system cannot overcome the capillary resistance of 

    the throat(s) and enter into porosity space  [2]. There are also some

    closed pores in CAC (closed pore - arrowed), where microbes can-

    not penetrate the internal pores of the CAC. Previous studies have

    shown that the biofilm biomass was 63.4 mg TN/g for IPC BAF and

    2.2 mg TN/g for CAC BAF [14].

     Table 1

    Operating conditions.

    Sample HRT (h) pH T (C) A/W TOC NH3-N PO43

    AIC ± SD AIC ± SD AIC ± SD

    Water quality indexes and operating conditions

    Stage one 7 6.7 (±0.5) 27–30 1:1 26.06 ± 0.205 9.75 ± 0.196 0.558 ± 0.022

    Stage two 7 6.6 (±0.5) 28–30 3:1 26.10 ± 0.449 9.79 ± 0.329 0.484 ± 0.015

    Stage three 7 7.1 (±0.5) 23–30 6:1 25.29 ± 0.486 9.81 ± 0.175 0.486 ± 0.016

    Note: TOC/NH3-N/PO43 values are influent concentrations mg/L; standard deviation – SD; average influent concentrations – AIC; HRT – hydraulic retention time; A/W – air/

    water ratios.

    156   T. Bao et al. / Separation and Purification Technology 167 (2016) 154–162

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    4/9

     3.4. Comparison of the influence of A/W ratio on TOC removal in the

    IPC BAF and CAC BAF 

    Fig. 3A, Tables 1 and 2 show the IPC BAF and CAC BAF perfor-

    mance for TOC removal with a HRT of 7 h at different A/W ratios

    (i.e., 1:1; 3:1; 6:1). These results confirm that for steady organic

    input loading, microbe bioactivity is enhanced with increase of 

    A/W ratio (1:1? 6:1). On the other hand, diffusivity is strength-

    ened with increase of A/W ratio and contributes to TOC removal.

    Fig. 3A and Table 2  also indicate that the IPC has higher perfor-mance due to enhanced physicochemical properties of IPC as con-

    firmed by the analyses of Micro-CT and porosimetry. The surface

    and internal pore structures of IPC are uniform and well developed

    so that organic matter and nutrient substances are able to reach

    the internal pores of IPC. The available space for microorganism

    growth, mass transfer rate of DO and the dispersion potential of 

    biofilm biomass are also increased by soluble pollutants extending

    to the internal pores of the IPC. These attributes result in a more

    favorable development of the filter layer and improve the effective-

    ness of biofiltration [13].

     3.5. Influence of A/W ratio on NH  3-N and TN removal

    The data in Fig. 3B, Tables 1 and 2 also show optimum removalof NH3-N. The removal rate is higher when the A/W ratio increases

    from 1:1 to 3:1 and then decreases at higher A/W ratio to 6:1. A

    higher air-flow inhibits the microbial biofilm resulting in

    decreased NH3-N removal. IPC BAF had a slightly higher NH3-N

    removal compared to the CAC BAF. A/W is one of the key parame-

    ters responsible for nitrification; therefore, it is the limiting factor

    for nitrification when the A/W ratio in the BAF is lower than

    required for bio-reactions with an A/W ratio of 1:1. Studies by

    Lazarova and Manem [38] have reported that autotrophic bacteria

    compete with heterotrophic bacteria for oxygen, substrates and

    habitation area within a microbial biofilm. This outcome is becauseslow growing autotrophic bacteria inherently compromises the

    nitrification process   [38].   The main cause of this difficulty is a

    decrease in nitrifying activity that affects the competition between

    aerobic heterotrophic bacteria and autotrophic bacteria [12].

    From Fig. 3C, Tables 1 and 2, it can be seen that TN removal effi-

    ciencies first increase (1:1–3:1) and then decrease with an increase

    of A/W ratio (3:1–6:1). The results confirm that TN removal is

    affected by the A/W ratio in the two BAFs. Denitrification produces

    less energy yield than oxygen respiration. Hence, a bacterial cell

    growing in aerobic conditions will choose to use oxygen as a termi-

    nal electron acceptor. In addition to this competitive effect, oxygen

    controls denitrification at two levels: (i) reversible inhibition of the

    activities of the denitrification enzymes and (ii) regulation of gene

    expression [39]. It can beseen that the IPC BAF has a slightlyhigher

    TN removal rate compared with the CAC BAF (Fig. 3C). Normally, a

    (A) 3 dimensional reconstruction of IPC;

    (B) 20 slices (0.72 mm) cross-section of IPC;

    (C) 3 dimensional of CAC;

    (D) 20 slices (0.72 mm) cross-section CAC;

    Fig. 1.  Micro-CT of IPC and CAC.

    T. Bao et al. / Separation and Purification Technology 167 (2016) 154–162   157

    http://-/?-http://-/?-http://-/?-

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    5/9

    (A) IPC; (B) IPC filled with epoxy (light blue = open pores);

    (C) CAC; (D) CAC filled with epoxy;

    Fig. 2.   Internal porosity of CAC and IPC (cross-section). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this

    article.)

    0

    40

    80

    0 25 50 75

    10

    20   R  e  m  o  v  a   l  e   f   f   i  c   i  e  n  c  y

       (   %   )

       T   O   C   (  m  g   /   L   )

    Time (d)

    A/W=1:1 A/W=6:1A/W=3:1

    CAC removal

    IPC removal

    0 25 50 75

    0

    5

    10

    40

    80

    120

       R  e  m  o  v  a   l  e   f   f   i  c   i  e  n  c

      y   (   %   )

       N   H   3  -   N   (  m  g   /   L   )

    Time (d)

    A/W=1:1

    CAC removal

    IPC removal

    A/W=6:1A/W=3:1

    0

    30

    60

    90

    0 25 50 75

    3

    6

    9

    12

       R  e  m  o  v  a   l  e   f   f   i  c   i  e  n  c  y   (   %   )

       T   N   (  m  g   /   L   )

    Time (d)

    Influent

    CAC effluent

    IPC effluent

    A/W=1:1 A/W=6:1A/W=3:1CAC removal

    IPC removal

    0

    50

    100

    150

    0 25 50 750.0

    0.3

    0.6

    0.9

       R  e  m  o  v  a   l  e   f   f   i  c   i  e  n  c  y   (   %   )

       P   (  m  g   /   L   )

    Time (d)

    A/W=1:1 A/W=6:1A/W=3:1

    CAC removal

     IPC removal

    DC

    BA

    Influent

    CAC effluent

    IPC effluent

    Influent

    CAC effluent

    IPC effluent

    Influent

    CAC effluent

    IPC effluent

    Fig. 3.  Removal efficiency of TOC, NH3-N, TN, and PO43 over time.

    158   T. Bao et al. / Separation and Purification Technology 167 (2016) 154–162

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    6/9

     Table 2

    Average removal rates for IPC and CAC in BAF at different air/water ratios.

    TOC NH3-N

    A/W 1:1 3:1 6:1 1:1 3:1

    Material AEC ± SD ARR ± SD AEC ± SD ARR ± SD AEC ± SD ARR ± SD AEC ± SD ARR ± SD AEC ± SD A

    IPC 19.93 ± 1.21 23.51 ± 4.88 13.53 ± 3.68 47.95 ± 14.96 9.98 ± 2.52 60.50 ± 10.25 7.80 ± 0.75 20.02 ± 7.70 0.57 ± 0.61 9

    CAC 21.87 ± 2.63 16.07 ± 10.13 16.91 ± 2.32 35.06 ± 9.81 12.45 ± 1.34 50.77 ± 5.29 8.25 ± 0.69 15.33 ± 7.46 4.62 ± 2.54 5

    TN PO43

    A/W 1:1 3:1 6:1 1:1 3:1

    Material AEC ± SD ARR ± SD AEC ± SD ARR ± SD AEC ± SD ARR ± SD AEC ± SD ARR ± SD AEC ± SD AR

    IPC 9.11 ± 0 .76 15.86 ± 8.74 5.55 ± 1.84 46.26 ± 1 8.95 7.22 ± 0.98 31.73 ± 8.94 0.16 ± 0.048 73.70 ± 9 .01 0.13 ± 0 .03 72

    CAC 9.97 ± 0 .62 8.09 ± 5 .09 8.76 ± 0 .88 15.64 ± 1 0.56 8.99 ± 0.87 14.96 ± 8.44 0.37 ± 0.07 34.21 ± 12.75 0.32 ± 0 .05 33

    Note: Standard deviation – SD; average removal rates – ARR (%); average effluent concentrations – AEC (mg/L).

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    7/9

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    8/9

    bacteria, Dechloromonas  sp. (No. 6) is a genus in the phylum pro-teobacteria. Dechloromonas   sp. is facultative denitrifying bacteria

    [46]. No. 2 demonstrates 99% similarity with Azonexus sp., a genus

    in the phylum proteobacteria [47]. No. 3 demonstrates 99%similar-

    ity with Hydrogenophaga sp., a genus of comamonad bacteria. Sev-

    eral of these bacterial species were previously classified in the

    genus Pseudomonas [48]. No. 7 and No. 8 demonstrate 99% similar-

    ity with  Sphaerotilus  sp., an aquatic periphyton organism associ-

    ated with polluted water that forms colonies commonly known

    as ‘‘sewage fungus”. However, an aquatic periphyton organism

    Sphaerotilus sp. was later identified as a tightly sheathed filamen-

    tous bacteria [49]. No. 16 demonstrates 100% similarity with Nitro-

    spira sp. which is a phylum of bacteria. This phylum contains only

    one class, nitrospira, which itself contains one order (Nitrospirales)

    and one family (Nitrospiraceae)  [50]. This analysis demonstratesthe biological diversity of microbes found in the biofilm and con-

    firms that nitrifying bacteria and denitrifying bacteria are present.

    This diversity of both bacterial types enhances performance in

    wastewater treatment systems.

    4. Conclusion

    This study reports the performance of synthesized IPC and CAC

    as filter media in a BAF process. SEM and PM results suggest that

    uniform and interconnected pores in IPC are more suited to micro-

    bial growth compared with the compact and closed pore structure

    of CAC. Three-dimensional rendering using micro-CT of the pores

    also reveals critical details of the pore structure and is an accuratemethod to characterize porosity. These observations support

    enhanced microbial growth for IPC compared to CAC. A highermicrobial biofilm production rate is observed in IPC compared to

    CAC when used in the BAF system. The performance efficiency of 

    the IPC BAF is higher than that for CAC BAF, particularly for TOC,

    TN, P, and NH3-N removal. This efficiency also improves when

    the A/W ratio is increased from 1:1 to 6:1. These results indicate

    that IPC is a promising material for the simultaneous removal of 

    nitrogen and phosphorus in wastewater treatment. The biological

    diversity of microorganism communities in the IPC BAF plays a

    key role in the simultaneous removal of nitrogen and phosphorus.

    Finally, a large quantity of protozoan and metazoan organisms

    observed in IPC BAF are indicators of good performance in wastew-

    ater treatment systems.

     Acknowledgements

    We gratefully acknowledge the support by the National Natural

    Science Foundation of China – China (41572029 and 41130206).

     Appendix A. Supplementary material

    Supplementary data associated with this article can be found, in

    the online version, at   http://dx.doi.org/10.1016/j.seppur.2016.05.

    011.

    References

    [1]  D.Y. Wu, B.H. Zhang, C.J. Li, et al., Simultaneous removal of ammonium and

    phosphate by zeolite synthesized from fly ash as influenced by salt treatment, J. Colloid Interface Sci. 304 (2006) 300–306.

    IPCBAF for line 1, line 2, and line 3.

    Line 1 Line 2 Line 3

    Fig. 5.  DGGE profiles of bacterial and archea 16S rRNA genes from the IPCBAF for line 1, line 2, and line 3.

    T. Bao et al. / Separation and Purification Technology 167 (2016) 154–162   161

    http://dx.doi.org/10.1016/j.seppur.2016.05.011http://dx.doi.org/10.1016/j.seppur.2016.05.011http://refhub.elsevier.com/S1383-5866(16)30333-1/h0005http://refhub.elsevier.com/S1383-5866(16)30333-1/h0005http://refhub.elsevier.com/S1383-5866(16)30333-1/h0005http://-/?-http://refhub.elsevier.com/S1383-5866(16)30333-1/h0005http://refhub.elsevier.com/S1383-5866(16)30333-1/h0005http://refhub.elsevier.com/S1383-5866(16)30333-1/h0005http://dx.doi.org/10.1016/j.seppur.2016.05.011http://dx.doi.org/10.1016/j.seppur.2016.05.011http://-/?-

  • 8/16/2019 Synthesis and Performance of Iron Oxide-based Porous Ceramsite

    9/9

    [2] T. Bao, T.H. Chen, C.S. Qing, et al., Development and application of palygorskite

    porous ceramsite in a biological aerated filter (BAF), Desalin. Water Treat. 57

    (2016) 107–115.

    [3]  J.J. Xie, T.H. Chen, C.S. Qing, et al., Adsorption of phosphate from aqueous

    solutions by thermally modified palygorskite, Environ. Eng. Manage. J. 12

    (2013) 1393–1399.

    [4]  H.P. Ye, F.Z. Chen, Y.Q. Sheng, et al., Adsorption of phosphate from aqueous

    solution onto modified palygorskites, Sep. Purif. Technol. 50 (2006) 283–290.

    [5]  H.B.Liu, T.H. Chen, R.L. Frost,An overviewof therole of goethitesurfacesin the

    environment, Chemosphere 103 (2014) 1–11.

    [6] H.B. Liu, T.H. Chen, X.H. Zou, et al., Removal of phosphorus using NZVI derivedfrom reducing natural goethite, Chem. Eng. J. 234 (2013) 80–87.

    [7]  H.B. Liu, T.H. Chen, J. Chang, et al., The effect of hydroxyl groups and surface

    area of hematite derived from annealing goethite for phosphate removal, J.

    Colloid Interface Sci. 398 (2013) 88–94.

    [8]  H.A. Hasan, S.R.S. Abdullah, S.K. Kamarudin, et al., Simultaneous NH4+-N and

    Mn2+ removal from drinking water using a biological aerated filter system:

    effects of different aeration rates, Sep. Purif. Technol. 118 (2013) 547–556 .

    [9] J. Werner, B. Besser, C. Brandes, et al., Production of ceramic membranes with

    different pore sizes for virus retention, J. Water Proc. Eng. 4 (2014) 201–211.

    [10]  T. Li, H.J. Wang, W.Y. Dong, et al., Performance of an anoxic reactor proposed

    before BAF: effect of ferrous sulfate on enhancing denitrification during

    simultaneous phosphorous removal, Chem. Eng. J. 248 (2014) 41–48.

    [11]   S.Q. Wu, Y.F. Qi, Q.Y. Yue, et al., Preparation of ceramic filler from reusing

    sewage sludge and application in biological aerated filter for soy protein

    secondary wastewater treatment, J. Hazard. Mater. 283 (2015) 608–616.

    [12]  F. Liu, C.C. Zhao, D.F. Zhao, et al., Tertiary treatment of textile wastewater with

    combined media biological aerated filter at different hydraulic loadings and

    dissolved oxygen concentrations, J. Hazard. Mater. 160 (2008) 161–167.

    [13]  J.L. Zou, G.R. Xu, K. Pan, et al., Nitrogen removal and biofilm structure affected

    by COD/NH4+–N in a biofilter with porous sludge ceramsite, Sep. Purif. Technol.

    94 (2012) 9–15.

    [14]   T. Bao, D. Chen, T.H. Chen, et al., Preparation and characterization of iron

    oxide-based porous ceramsite, Acta Mater. Compos. Sin. 2 (2014) 409–415 (in

    Chinese).

    [15]  B.J. Ni, H.Q. Yu, Y.J. Sun, Modeling simultaneous autotrophic and heterotrophic

    growth in aerobic granules, Water Res. 42 (2008) 1583–1594.

    [16]   K.M. De, J. Heijnen, M. van Loosdrecht, Simultaneous COD, nitrogen, and

    phosphate removal by aerobic granular sludge, Biotechnol. Bioeng. 90 (2005)

    761–769.

    [17]   K.L. Yang, Q.Y. Yue, W. Han, et al., Effect of novel sludge and coal cinder

    ceramic media in combined anaerobic-aerobic bio-filter for tetracycline

    wastewater treatment at low temperature, Chem. Eng. J. 277 (2015) 130–139.

    [18]   Chinese Mohurd, Artifical Ceramsite Filter Material for Water Treatment,

    China Standard Publishing House, Beijing, China, 2008.

    [19]   T. Hildebrand, P. Rüegsegger, A new method for the model-independent

    assessment of thickness in three-dimensional images, J. Microsc. Oxford 185

    (1997) 67–75.[20]   T. Hildebrand, A. Laib, R. Mueller, et al., Direct three-dimensional

    morphometric analysis of human cancellous bone: microstructural data

    from spine, femur, iliac crest, and calcaneus, J. Bone Miner. Res. 14 (1999)

    1167–1174.

    [21]   T. Hildebrand, P. Rüegsegger, Quantification of bone microarchitecture with

    the structure model index, Comput. Methods Biomech. Biomed. Eng. 1 (1997)

    15–23.

    [22]  J. Tan, J. Wang, J. Xue, et al., Methane production and microbial community

    analysis in the goethite facilitated anaerobic reactors using algal biomass, Fuel

    145 (2015) 196–201.

    [23]  X.Q. Zhao, L.Y. Yang, Z.Y. Yu, et al., Characterization of depth-related microbial

    communities in lake sediment by denaturing gradient gel electrophoresis of 

    amplified 16S rRNA fragments, J. Environ. Sci. 20 (2008) 224–230.

    [24]   E.P.A. Chinese, Methods for Water and Wastewater Analysis, fourth ed.,

    Environmental Science Publishing House of China, Beijing, China, 2002.

    [25]  M. Ding, C.C. Danielsen, I. Hvid, et al., Three-dimensional microarchitecture of 

    adolescent cancellous bone, Bone 51 (2012) 953–960.

    [26]  K.T. De, M.A. Boone, S.T. De, et al., A pore-scale study of fracture dynamics in

    rock using X-ray micro-CT under ambient freeze-thaw cycling, Environ. Sci.Technol. 49 (2015) 2867–2874.

    [27]  E. Cho, A. Sadr, N. Inai, et al., Evaluation of resin composite polymerization by

    three dimensional micro-CT imaging and nanoindentation, Dent. Mater. 27

    (2011) 1070–1078.

    [28]  G.R. Xu, J.J. Zou, G.B. Li, Stabilization of heavy metals in ceramsite made with

    sewage sludge, J. Hazard. Mater. 152 (2008) 56–61.

    [29]   D.R. Simpson, Biofilm processes in biologically active carbon water

    purification, Water Res. 42 (2008) 2839.

    [30]   J.M. Schöntag, B.S. Pizzolatti, V.H. Jangada, et al., Water quality produced by

    polystyrene granules as a media filter on rapid filters, J. Water Proc. Eng. 5

    (2015) 118–126.

    [31]   T. Li, J. Liu, R. Bai, et al., Membrane-aerated biofilm reactor for the treatment of 

    acetonitrile wastewater, Environ. Sci. Technol. 42 (2008) 2099.

    [32]   C.Y. Wu, S. Ushiwaka, H.Horii, et al., Membrane-attached biofilmas a meansto

    facilitate nitrification in activated sludge process and its effect on themicrofaunal population, Biochem. Eng. J. 40 (2008) 430–436.

    [33]  K.M. De, C. Picioreanu, M. Hosseini, et al., Kinetic model of a granular sludge

    SBR-influences on nutrient removal, Biotechnol. Bioeng. 97 (2007) 801–815.

    [34]  B.W.A. Van, L.M.D. Van, J.J. Heijnen, Control of heterotrophic layer formation

    on nitrifying biofilms in a biofilm airlift suspension reactor, Biotechnol. Bioeng.

    53 (1997) 397–405.

    [35]   H. Satoh, Y. Nakamura, H. Ono, et al., Effect of oxygen concentration on

    nitrification and denitrification in single activated sludge flocs, Biotechnol.

    Bioeng. 83 (2003) 604–607.

    [36] Sandstone Pore Structure Method of Image Analysis (Industry Standard CJ/T),

    2013 (in Chinese).

    [37]   K. Ruzyla, D.L. Jezek, Staining method for recognition of pore space in thin and

    polished sections: research method paper, J. Sediment. Res. 57 (1987) 777–

    778.

    [38]   V. Lazarova, J. Manem, An innovative process for wastewater treatment: the

    circulating floating bed reactor, Water Sci. Technol. 34 (9) (1996) 89–99 .

    [39]   J.M. Galvez, M.A. Gomez, E. Hontoria, et al., Influence of hydraulic loading and

    air flow rate on urban wastewater nitrogen removal with a submerged fixed-

    film reactor, J. Hazard. Mater. B101 (2003) 219–229.

    [40]  M.A. Gomez, E. Hontoria, J.G. Lopez, Effect of dissolved oxygen concentration

    on nitrate removal from groundwater using a denitrifying submerged filter, J.

    Hazard. Mater. B90 (2002) 267–278.

    [41]   K.D. Laursen, S.E. Jepsen, J. Jansen, et al., Denitrification in submerged filters

    exposed to oxygen, Nutrient Removal From Wastewaters, Technomic, Basel,

    Switzerland, 1994.

    [42]  F.Q. Gan, J.M. Zhou, H.Y. Wang, et al., Removal of phosphate from aqueous

    solution by thermally treated natural palygorskite, Water Res. 43 (2009)

    2907–2915.

    [43]   Y. Feng, Y.Z. Yu, L.P. Qiu, et al., The characteristics and effect of grain-slag

    media for the treatment of phosphorus in a biological aerated filter (BAF),

    Desalin. Water Treat. 47 (2012) 258–265.

    [44]   Y. Shen, Z. Zhang, Modern Biomonitoring Techniques Using Freshwater

    Microbiota, China Architecture & Building Press, Beijing, China, 1990 (in

    Chinese).

    [45]   H. Tsuno, T. Hidaka, F. Nishimura, A simple biofilm model of bacterial

    competition for attached surface, Water Res. 36 (2002) 996–1006.

    [46]  M.A. Horn, J. Ihssen, C. Matthies, et al.,  Dechloromonas denitrificans   sp. nov.,Flavobacterium denitrificans   sp. nov.,   Paenibacillus anaericanus   sp. nov. andPaenibacillus terrae strain MH72, N2O-producing bacteria isolated fromthe gutof the earthworm Aporrectodea caliginosa, Int. J. Syst. Evol.Microbiol. 55 (2005)1255–1265.

    [47]   H.B. Reinhold, T. Hurek, Reassessment of the taxonomic structure of the

    diazotrophic genus   Azoarcus sensu  lato and description of three new generaand new species,   Azovibrio restrictus   gen. nov., sp. nov.,   Azospira oryzae   gen.nov., sp. nov. and   Azonexus fungiphilus   gen. nov., sp. nov., Int. J. Syst. Evol.Microbiol. 50 (2000) 649–659.

    [48]   A. Willems, J. Busse, M. Goor, et al., Hydrogenophaga, a new genus of 

    hydrogen-oxidizing bacteria that includes Hydrogenophaga flava  comb. nov.(Formerly   Pseudomonas flava),   Hydrogenophaga palleronii   (FormerlyPseudomonas palleronii),  Hydrogenophaga pseudoflava  (Formerly  Pseudomonas

     pseudoflava   and ‘‘Pseudomonas carboxydoflava”), and   Hydrogenophagataeniospiralis (Formerly  Pseudomonas taeniospiralis), Int. J. Syst. Bacteriol. 39(1989) 319–333.

    [49]   V. Pellegrin, S. Juretschko, M. Wagner, et al., Morphological and biochemical

    properties of a Sphaerotilus  sp. isolated from paper mill slimes, Appl. Environ.

    Microbiol. 65 (1999) 156–162.[50]   S. Ehrich, D. Behrens, E. Lebedeva, et al., A new obligately

    chemolithoautotrophic, nitrite-oxidizing bacterium,   Nitrospira moscoviensissp. nov and its phylogenetic relationship, Arch. Microbiol. 164 (1995) 16–23.

    162   T. Bao et al. / Separation and Purification Technology 167 (2016) 154–162

    http://refhub.elsevier.com/S1383-5866(16)30333-1/h0010http://refhub.elsevier.com/S1383-5866(16)30333-1/h0010http://refhub.elsevier.com/S1383-5866(16)30333-1/h0010http://refhub.elsevier.com/S1383-5866(16)30333-1/h0015http://refhub.elsevier.com/S1383-5866(16)30333-1/h0015http://refhub.elsevier.com/S1383-5866(16)30333-1/h0015http://refhub.elsevier.com/S1383-5866(16)30333-1/h0015http://refhub.elsevier.com/S1383-5866(16)30333-1/h0020http://refhub.elsevier.com/S1383-5866(16)30333-1/h0020http://refhub.elsevier.com/S1383-5866(16)30333-1/h0025http://refhub.elsevier.com/S1383-5866(16)30333-1/h0025http://refhub.elsevier.com/S1383-5866(16)30333-1/h0030http://refhub.elsevier.com/S1383-5866(16)30333-1/h0030http://refhub.elsevier.com/S1383-5866(16)30333-1/h0035http://refhub.elsevier.com/S1383-5866(16)30333-1/h0035http://refhub.elsevier.com/S1383-5866(16)30333-1/h0035http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0045http://refhub.elsevier.com/S1383-5866(16)30333-1/h0045http://refhub.elsevier.com/S1383-5866(16)30333-1/h0050http://refhub.elsevier.com/S1383-5866(16)30333-1/h0050http://refhub.elsevier.com/S1383-5866(16)30333-1/h0050http://refhub.elsevier.com/S1383-5866(16)30333-1/h0055http://refhub.elsevier.com/S1383-5866(16)30333-1/h0055http://refhub.elsevier.com/S1383-5866(16)30333-1/h0055http://refhub.elsevier.com/S1383-5866(16)30333-1/h0060http://refhub.elsevier.com/S1383-5866(16)30333-1/h0060http://refhub.elsevier.com/S1383-5866(16)30333-1/h0060http://refhub.elsevier.com/S1383-5866(16)30333-1/h0060http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0070http://refhub.elsevier.com/S1383-5866(16)30333-1/h0070http://refhub.elsevier.com/S1383-5866(16)30333-1/h0070http://refhub.elsevier.com/S1383-5866(16)30333-1/h0075http://refhub.elsevier.com/S1383-5866(16)30333-1/h0075http://refhub.elsevier.com/S1383-5866(16)30333-1/h0075http://refhub.elsevier.com/S1383-5866(16)30333-1/h0080http://refhub.elsevier.com/S1383-5866(16)30333-1/h0080http://refhub.elsevier.com/S1383-5866(16)30333-1/h0080http://refhub.elsevier.com/S1383-5866(16)30333-1/h0080http://refhub.elsevier.com/S1383-5866(16)30333-1/h0085http://refhub.elsevier.com/S1383-5866(16)30333-1/h0085http://refhub.elsevier.com/S1383-5866(16)30333-1/h0085http://refhub.elsevier.com/S1383-5866(16)30333-1/h0085http://refhub.elsevier.com/S1383-5866(16)30333-1/h0090http://refhub.elsevier.com/S1383-5866(16)30333-1/h0090http://refhub.elsevier.com/S1383-5866(16)30333-1/h0095http://refhub.elsevier.com/S1383-5866(16)30333-1/h0095http://refhub.elsevier.com/S1383-5866(16)30333-1/h0095http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0105http://refhub.elsevier.com/S1383-5866(16)30333-1/h0105http://refhub.elsevier.com/S1383-5866(16)30333-1/h0105http://refhub.elsevier.com/S1383-5866(16)30333-1/h0110http://refhub.elsevier.com/S1383-5866(16)30333-1/h0110http://refhub.elsevier.com/S1383-5866(16)30333-1/h0110http://refhub.elsevier.com/S1383-5866(16)30333-1/h0115http://refhub.elsevier.com/S1383-5866(16)30333-1/h0115http://refhub.elsevier.com/S1383-5866(16)30333-1/h0115http://refhub.elsevier.com/S1383-5866(16)30333-1/h0115http://refhub.elsevier.com/S1383-5866(16)30333-1/h0120http://refhub.elsevier.com/S1383-5866(16)30333-1/h0120http://refhub.elsevier.com/S1383-5866(16)30333-1/h0125http://refhub.elsevier.com/S1383-5866(16)30333-1/h0125http://refhub.elsevier.com/S1383-5866(16)30333-1/h0130http://refhub.elsevier.com/S1383-5866(16)30333-1/h0130http://refhub.elsevier.com/S1383-5866(16)30333-1/h0130http://refhub.elsevier.com/S1383-5866(16)30333-1/h0130http://refhub.elsevier.com/S1383-5866(16)30333-1/h0135http://refhub.elsevier.com/S1383-5866(16)30333-1/h0135http://refhub.elsevier.com/S1383-5866(16)30333-1/h0135http://refhub.elsevier.com/S1383-5866(16)30333-1/h0140http://refhub.elsevier.com/S1383-5866(16)30333-1/h0140http://refhub.elsevier.com/S1383-5866(16)30333-1/h0145http://refhub.elsevier.com/S1383-5866(16)30333-1/h0145http://refhub.elsevier.com/S1383-5866(16)30333-1/h0145http://refhub.elsevier.com/S1383-5866(16)30333-1/h0150http://refhub.elsevier.com/S1383-5866(16)30333-1/h0150http://refhub.elsevier.com/S1383-5866(16)30333-1/h0150http://refhub.elsevier.com/S1383-5866(16)30333-1/h0150http://refhub.elsevier.com/S1383-5866(16)30333-1/h0155http://refhub.elsevier.com/S1383-5866(16)30333-1/h0155http://refhub.elsevier.com/S1383-5866(16)30333-1/h0160http://refhub.elsevier.com/S1383-5866(16)30333-1/h0160http://refhub.elsevier.com/S1383-5866(16)30333-1/h0160http://refhub.elsevier.com/S1383-5866(16)30333-1/h0165http://refhub.elsevier.com/S1383-5866(16)30333-1/h0165http://refhub.elsevier.com/S1383-5866(16)30333-1/h0170http://refhub.elsevier.com/S1383-5866(16)30333-1/h0170http://refhub.elsevier.com/S1383-5866(16)30333-1/h0170http://refhub.elsevier.com/S1383-5866(16)30333-1/h0175http://refhub.elsevier.com/S1383-5866(16)30333-1/h0175http://refhub.elsevier.com/S1383-5866(16)30333-1/h0175http://refhub.elsevier.com/S1383-5866(16)30333-1/h0185http://refhub.elsevier.com/S1383-5866(16)30333-1/h0185http://refhub.elsevier.com/S1383-5866(16)30333-1/h0185http://refhub.elsevier.com/S1383-5866(16)30333-1/h0190http://refhub.elsevier.com/S1383-5866(16)30333-1/h0190http://refhub.elsevier.com/S1383-5866(16)30333-1/h0195http://refhub.elsevier.com/S1383-5866(16)30333-1/h0195http://refhub.elsevier.com/S1383-5866(16)30333-1/h0195http://refhub.elsevier.com/S1383-5866(16)30333-1/h0200http://refhub.elsevier.com/S1383-5866(16)30333-1/h0200http://refhub.elsevier.com/S1383-5866(16)30333-1/h0200http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0210http://refhub.elsevier.com/S1383-5866(16)30333-1/h0210http://refhub.elsevier.com/S1383-5866(16)30333-1/h0210http://refhub.elsevier.com/S1383-5866(16)30333-1/h0215http://refhub.elsevier.com/S1383-5866(16)30333-1/h0215http://refhub.elsevier.com/S1383-5866(16)30333-1/h0215http://refhub.elsevier.com/S1383-5866(16)30333-1/h0215http://refhub.elsevier.com/S1383-5866(16)30333-1/h0220http://refhub.elsevier.com/S1383-5866(16)30333-1/h0220http://refhub.elsevier.com/S1383-5866(16)30333-1/h0220http://refhub.elsevier.com/S1383-5866(16)30333-1/h0225http://refhub.elsevier.com/S1383-5866(16)30333-1/h0225http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0250http://refhub.elsevier.com/S1383-5866(16)30333-1/h0250http://refhub.elsevier.com/S1383-5866(16)30333-1/h0250http://refhub.elsevier.com/S1383-5866(16)30333-1/h0250http://refhub.elsevier.com/S1383-5866(16)30333-1/h0250http://refhub.elsevier.com/S1383-5866(16)30333-1/h0250http://refhub.elsevier.com/S1383-5866(16)30333-1/h0250http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0245http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0240http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0235http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0230http://refhub.elsevier.com/S1383-5866(16)30333-1/h0225http://refhub.elsevier.com/S1383-5866(16)30333-1/h0225http://refhub.elsevier.com/S1383-5866(16)30333-1/h0220http://refhub.elsevier.com/S1383-5866(16)30333-1/h0220http://refhub.elsevier.com/S1383-5866(16)30333-1/h0220http://refhub.elsevier.com/S1383-5866(16)30333-1/h0220http://refhub.elsevier.com/S1383-5866(16)30333-1/h0215http://refhub.elsevier.com/S1383-5866(16)30333-1/h0215http://refhub.elsevier.com/S1383-5866(16)30333-1/h0215http://refhub.elsevier.com/S1383-5866(16)30333-1/h0210http://refhub.elsevier.com/S1383-5866(16)30333-1/h0210http://refhub.elsevier.com/S1383-5866(16)30333-1/h0210http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0205http://refhub.elsevier.com/S1383-5866(16)30333-1/h0200http://refhub.elsevier.com/S1383-5866(16)30333-1/h0200http://refhub.elsevier.com/S1383-5866(16)30333-1/h0200http://refhub.elsevier.com/S1383-5866(16)30333-1/h0195http://refhub.elsevier.com/S1383-5866(16)30333-1/h0195http://refhub.elsevier.com/S1383-5866(16)30333-1/h0195http://refhub.elsevier.com/S1383-5866(16)30333-1/h0190http://refhub.elsevier.com/S1383-5866(16)30333-1/h0190http://refhub.elsevier.com/S1383-5866(16)30333-1/h0185http://refhub.elsevier.com/S1383-5866(16)30333-1/h0185http://refhub.elsevier.com/S1383-5866(16)30333-1/h0185http://refhub.elsevier.com/S1383-5866(16)30333-1/h0175http://refhub.elsevier.com/S1383-5866(16)30333-1/h0175http://refhub.elsevier.com/S1383-5866(16)30333-1/h0175http://refhub.elsevier.com/S1383-5866(16)30333-1/h0170http://refhub.elsevier.com/S1383-5866(16)30333-1/h0170http://refhub.elsevier.com/S1383-5866(16)30333-1/h0170http://refhub.elsevier.com/S1383-5866(16)30333-1/h0165http://refhub.elsevier.com/S1383-5866(16)30333-1/h0165http://refhub.elsevier.com/S1383-5866(16)30333-1/h0160http://refhub.elsevier.com/S1383-5866(16)30333-1/h0160http://refhub.elsevier.com/S1383-5866(16)30333-1/h0160http://refhub.elsevier.com/S1383-5866(16)30333-1/h0155http://refhub.elsevier.com/S1383-5866(16)30333-1/h0155http://refhub.elsevier.com/S1383-5866(16)30333-1/h0150http://refhub.elsevier.com/S1383-5866(16)30333-1/h0150http://refhub.elsevier.com/S1383-5866(16)30333-1/h0150http://refhub.elsevier.com/S1383-5866(16)30333-1/h0145http://refhub.elsevier.com/S1383-5866(16)30333-1/h0145http://refhub.elsevier.com/S1383-5866(16)30333-1/h0140http://refhub.elsevier.com/S1383-5866(16)30333-1/h0140http://refhub.elsevier.com/S1383-5866(16)30333-1/h0135http://refhub.elsevier.com/S1383-5866(16)30333-1/h0135http://refhub.elsevier.com/S1383-5866(16)30333-1/h0135http://refhub.elsevier.com/S1383-5866(16)30333-1/h0130http://refhub.elsevier.com/S1383-5866(16)30333-1/h0130http://refhub.elsevier.com/S1383-5866(16)30333-1/h0130http://refhub.elsevier.com/S1383-5866(16)30333-1/h0125http://refhub.elsevier.com/S1383-5866(16)30333-1/h0125http://refhub.elsevier.com/S1383-5866(16)30333-1/h0120http://refhub.elsevier.com/S1383-5866(16)30333-1/h0120http://refhub.elsevier.com/S1383-5866(16)30333-1/h0120http://refhub.elsevier.com/S1383-5866(16)30333-1/h0115http://refhub.elsevier.com/S1383-5866(16)30333-1/h0115http://refhub.elsevier.com/S1383-5866(16)30333-1/h0115http://refhub.elsevier.com/S1383-5866(16)30333-1/h0110http://refhub.elsevier.com/S1383-5866(16)30333-1/h0110http://refhub.elsevier.com/S1383-5866(16)30333-1/h0110http://refhub.elsevier.com/S1383-5866(16)30333-1/h0105http://refhub.elsevier.com/S1383-5866(16)30333-1/h0105http://refhub.elsevier.com/S1383-5866(16)30333-1/h0105http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0100http://refhub.elsevier.com/S1383-5866(16)30333-1/h0095http://refhub.elsevier.com/S1383-5866(16)30333-1/h0095http://refhub.elsevier.com/S1383-5866(16)30333-1/h0095http://refhub.elsevier.com/S1383-5866(16)30333-1/h0090http://refhub.elsevier.com/S1383-5866(16)30333-1/h0090http://refhub.elsevier.com/S1383-5866(16)30333-1/h0090http://refhub.elsevier.com/S1383-5866(16)30333-1/h0085http://refhub.elsevier.com/S1383-5866(16)30333-1/h0085http://refhub.elsevier.com/S1383-5866(16)30333-1/h0085http://refhub.elsevier.com/S1383-5866(16)30333-1/h0080http://refhub.elsevier.com/S1383-5866(16)30333-1/h0080http://refhub.elsevier.com/S1383-5866(16)30333-1/h0080http://refhub.elsevier.com/S1383-5866(16)30333-1/h0075http://refhub.elsevier.com/S1383-5866(16)30333-1/h0075http://refhub.elsevier.com/S1383-5866(16)30333-1/h0070http://refhub.elsevier.com/S1383-5866(16)30333-1/h0070http://refhub.elsevier.com/S1383-5866(16)30333-1/h0070http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0065http://refhub.elsevier.com/S1383-5866(16)30333-1/h0060http://refhub.elsevier.com/S1383-5866(16)30333-1/h0060http://refhub.elsevier.com/S1383-5866(16)30333-1/h0060http://refhub.elsevier.com/S1383-5866(16)30333-1/h0055http://refhub.elsevier.com/S1383-5866(16)30333-1/h0055http://refhub.elsevier.com/S1383-5866(16)30333-1/h0055http://refhub.elsevier.com/S1383-5866(16)30333-1/h0050http://refhub.elsevier.com/S1383-5866(16)30333-1/h0050http://refhub.elsevier.com/S1383-5866(16)30333-1/h0050http://refhub.elsevier.com/S1383-5866(16)30333-1/h0045http://refhub.elsevier.com/S1383-5866(16)30333-1/h0045http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0040http://refhub.elsevier.com/S1383-5866(16)30333-1/h0035http://refhub.elsevier.com/S1383-5866(16)30333-1/h0035http://refhub.elsevier.com/S1383-5866(16)30333-1/h0035http://refhub.elsevier.com/S1383-5866(16)30333-1/h0030http://refhub.elsevier.com/S1383-5866(16)30333-1/h0030http://refhub.elsevier.com/S1383-5866(16)30333-1/h0025http://refhub.elsevier.com/S1383-5866(16)30333-1/h0025http://refhub.elsevier.com/S1383-5866(16)30333-1/h0020http://refhub.elsevier.com/S1383-5866(16)30333-1/h0020http://refhub.elsevier.com/S1383-5866(16)30333-1/h0015http://refhub.elsevier.com/S1383-5866(16)30333-1/h0015http://refhub.elsevier.com/S1383-5866(16)30333-1/h0015http://refhub.elsevier.com/S1383-5866(16)30333-1/h0010http://refhub.elsevier.com/S1383-5866(16)30333-1/h0010http://refhub.elsevier.com/S1383-5866(16)30333-1/h0010