Water Research 96 (2016) 280e291

Contents lists available at ScienceDirect

Water Research
journal homepage: www.elsevier.com/locate/watres

Removal of phosphorus by the core-shell bio-ceramic/Zn-layered

double hydroxides (LDHs) composites for municipal wastewater
treatment in constructed rapid infiltration system
Xiangling Zhang*, 1, Lu Guo 1, Hualing Huang, Yinghe Jiang, Meng Li, Yujie Leng
School of Civil Engineering and Architecture, Wuhan University of Technology, Wuhan 430070, China

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

Article history: Constructed rapid infiltration systems (CRIS) are a reasonable option for treating wastewater, owing to
Received 23 November 2015 their simplicity, low cost and low energy consumption. Layered double hydroxides (LDHs), novel ma-
Received in revised form terials with high surface area and anion exchange capacity, faced the problem of the application in CRIS
3 March 2016
due to the powdered form. To overcome this shortcoming, Zn-LDHs (FeZn-LDHs, CoZn-LDHs, AlZn-LDHs)
Accepted 26 March 2016
were prepared by co-precipitation method and in-situ coated on the surface of the natural bio-ceramic to
Available online 30 March 2016
synthesize the core-shell bio-ceramic/Zn-LDHs composites. Characterization by Scanning Electron Mi-
croscope (SEM) and X-ray Fluorescence Spectrometer (XRFS) indicated that the Zn-LDHs were successful
Keywords:
Layered double hydroxides
loaded on the natural bio-ceramic. Column tests experiments indicated that the bio-ceramic/Zn-LDHs
Constructed rapid infiltration system efficiently enhanced the removal performance of phosphorus. The efficiently removal rates of bio-
Bio-ceramic/Zn-LDHs ceramic/FeZn-LDHs were 71.58% for total phosphorous (TP), 74.91% for total dissolved phosphorous
Bio-ceramic substrates (TDP), 82.31% for soluble reactive phosphorous (SRP) and 67.58% for particulate phosphorus (PP).
Phosphorus removal Compared with the natural bio-ceramic, the average removal rates were enhanced by 32.20% (TP), 41.33%
Municipal wastewater (TDP), 49.06% (SRP) and 10.50% (PP), respectively. Adsorption data of phosphate were better described by
the Freundlich model for the bio-ceramic/Zn-LDHs and natural bio-ceramic, except for the bio-ceramic/
CoZn-LDHs. The maximum adsorption capacity of bio-ceramic/AlZn-LDHs (769.23 mg/kg) was 1.77 times
of the natural bio-ceramic (434.78 mg/kg). The effective desorption of phosphate could achieve by using
a mixed solution of 5 M NaCl þ 0.1 M NaOH, it outperformed the natural bio-ceramic of 18.95% for FeZn-
LDHs, 7.59% for CoZn-LDHs and 12.66% for AlZn-LDHs. The kinetic data of the bio-ceramic/Zn-LDHs were
better described by the pseudo-second-order equation. Compared the removal amount of phosphate by
the natural bio-ceramic, the physical effects were improved little, but the chemical effects were
enhanced for 112.49% for FeZn-LDHs, 111.89% for CoZn-LDHs and 122.67% for AlZn-LDHs. Therefore, the
way of coating Zn-LDHs on the bio-ceramic efficiently improved the chemical effects in phosphate
removal, supporting that it can use as potential substrates for the removal of phosphorus in CRIS.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction attention as phosphorous is responsible for eutrophication of

Point and non-point pollution of phosphorous from agricultural
runoff, industrial discharges and municipal drainage have become a
matter of increasing concern in many countries, especially in
developing countries. Removal of phosphorous has attracted much
* Corresponding author. School of Civil Engineering and Architecture, Wuhan
University of Technology, 122, Luoshi Road, Hongshan District, Wuhan, 430070,
China.
E-mail address: ZXLCL@126.com (X. Zhang).
1
Xiangling Zhang and Lu Guo are co-first authors.
rivers, lakes and damage of the aquatic eco-system (Lau and Lane,
2002; Westholm, 2006; Montangero and Belevi, 2007). To date,
the main commercial processes for removing phosphorous are
chemical precipitation with iron, alum or lime (Ayoub et al., 2001;
Seyhan and Erdincler, 2003; Boujelbena et al., 2008), but the sludge
handling become another matter of concern. Therefore, a more
cost-effective and eco-friendly measures should be taken to
significantly lower the impacts of the phosphorous pollution.
Constructed rapid infiltration systems (CRIS), developed from
traditional rapid infiltration systems (RI), are a reasonable option
for treating wastewater, owing to their simplicity, reliability, low

http://dx.doi.org/10.1016/j.watres.2016.03.063
0043-1354/© 2016 Elsevier Ltd. All rights reserved.
 X. Zhang et al. / Water Research 96 (2016) 280e291
cost and low energy consumption (Wang et al., 2010; Chen, 2012;
Lian et al., 2014). During the previous research, by virtue of using
natural sand, gravel, ceramic, coal gangue and other substrates to
replace the conventional natural soil medium, CRIS have been
proved to be efficiently produce better penetrability, increasing
hydraulic loading (1.0e2.0 m d1) and decreasing the space occu-
pation (Zhang et al., 2007; Jiang et al., 2011). However, limited by
pollution tolerance and adsorption ability, natural substrates have
been one of the worst operation problems for CRIS. Herein, it is
imperative that effective measures should be taken to improve the
substrates adsorption ability.
Layered double hydroxides (LDHs) are a series of compounds
derived from the structure of mineral brucite (Cavani et al., 1991).
The general formula of LDHs can be expressed as
½M1X 2þ MX 3þ ðOHÞ2  ½An X=n nH2 O, where M2þ and M3þ are diva-
lent and trivalent metal cations in the octahedral positions of the
brucite-like layers. Due to part of the divalent metal cations can be
isomorphously substituted by trivalent metal cations, the layers
acquires positive charge with the compositions
ð½M1X 2þ MX 3þ ðOHÞ2 Xþ Þ. An, are the anions incorporated in the
interlayer region for charge neutrality, resulting in the formation of
LDHs (Radha and Kamath, 2003). LDHs have relatively weak
interlayer bonding, as a consequence, exhibit excellent ability to
capture organic and inorganic anions. The large surface and high
anion exchange capacity make LDHs as promising materials for
phosphorus uptake from wastewater (Novillo et al., 2014; Sun et al.,
2014; Yan et al., 2015). However, LDHs have been normally utilized
as dispersed powder form for the adsorption of phosphorus.
Herein, it presented appreciable problems for the application in
CRIS due to the small size, low specific density and subsequently
solid/liquid separation process.
As a continuation of our previous work (Zhang et al., 2013, 2014,
2015), nine kinds of different LDHs were synthesized by co-
precipitation method and in-situ coated on the natural anthracite,
zeolites and bio-ceramic, respectively. The results indicated that the
Zn-LDHs coated on the natural bio-ceramic surface were efficiently
removed the phosphorus. Therefore, a series of natural bio-ceramic
substrates have been utilized with various LDHs to synthesized
core-shell substrate/LDHs composites and LDHs were synthesized
through co-precipitation method. Subsequently, column tests were
used to study the removal performance of nitrogen and phosphorous.
Herein, the objectives of this study are: 1) to prepare bio-
ceramic/Zn-LDHs (FeZn-LDHs, CoZn-LDHs, AlZn-LDHs) compos-
ites and to characterize the composites by SEM and XRFS in details;
2) to investigate the efficient decontamination of phosphorus from
municipal wastewater by the column tests; 3) to study the
adsorption and desorption behavior of phosphate; 4) to propose
the mechanisms for the removal of phosphorus.
2. Material and methods
2.1. Preparation of bio-ceramic/Zn-LDHs
Core-shell bio-ceramic/Zn-LDHs (FeZn, CoZn, AlZn-LDHs) com-
posites were synthesized via conventional co-precipitation method
under ambient atmosphere. In a typical synthetic run, such as bio-
ceramic/FeZn-LDHs composites, aqueous solution of FeCl3 (0.75 M)
and ZnCl2 (1.50 M) were simultaneously added into a beaker which
contain the clean natural bio-ceramic at temperature of 80  C.
Simultaneously, appropriate volume of 10%wt NaOH was added at a
controlled rate to maintain the pH at 11e12 under continuous
stirring for 4 h. After complete precipitation, the mixed compounds
were centrifuged at 1000e1500r/min for 10min and washed thor-
oughly with de-ionized water until the pH of effluent solution was
neutral. Finally, the washed samples were oven-drying for 16 h to
281
obtain bio-ceramic/FeZn-LDHs composites.
2.2. Characterization methods
Morphological features were acquired using a scanning electron
microscopy (SEM, JSM-5610LV, Japan). Chemical compositions
were analyzed by X-Ray Fluorescence Spectrometer (XRFS, Axios,
Panalytical. B. V., Holland). Brunauer Emmett Teller (BET) surface
area of the natural bio-ceramic were analyzed by N2 adsorption
at 196  C using a sorptometer (ASAP-2020, Micromeritics, USA).
The average voidage of the natural bio-ceramic can be obtained by
the following equation (Eq. (1)):
4V
f¼ 100% (1)
pD2 H
where f is the average voidage of the substrate (%), D is the inner
diameter of the column (cm), H is the height of the natural bio-
ceramic filled into the column. V is the volume of the H height
water flowed out of the column (mL).
2.3. Water quality monitoring
The pH value was monitored with a Sartorius pH meter (Model
PB-10, Germany) using a combined glass electrode. The DO (dis-
solved oxygen) value was measured by a DO meter (HQ30d, HACH,
USA). Laboratory analysis was performed on the water samples for
total phosphorous (TP); total dissolved phosphorous (TDP) and
soluble reactive phosphorous (SRP). All the parameters mentioned
above were determined, based on standard methods (APHA, 2005).
2.4. Characterization of the system
A simulated lab-scale constructed rapid infiltration system was
constructed and operated indoor, as part of a larger experiment
containing 2 units (one in parallel). Each unit contains 4 test col-
umns (PVC, inner diameter 8 cm and height 25 cm). The first three
and fourth test columns contained 20 cm thick bio-ceramic/Zn-
LDHs (FeZn-LDHs, CoZn-LDHs and AlZn-LDHs) composites and
natural bio-ceramic substrates, respectively. The average diameter
of the natural bio-ceramic were 4~8 mm. The true and bulk density
was 2.034 and 1.028 g/cm3. The Brunauer Emmett Teller (BET)
surface area was 0.57 m2/g. The average voidage of the natural bio-
ceramic in the column tests was 30.96%.
The CRIS unit was fed with municipal wastewater from the top
of the columns and flowed out from the bottom. All the experi-
ments were carried out in duplicate and continuous operated for 6
months. The hydraulic retention time (HRT) was 24 h and the hy-
draulic loading rate (HLR) was 70 L/(m2$d), respectively. The raw
water was collected after the thin grill of Gongda-Road Lifting
Pump Station and the characteristics were shown in Table 1. The
experiments started at March and ended at September 2015 (last
about 28 weeks).
2.5. Adsorption studies
For the adsorption of phosphate, stock solution were prepared
from KH2PO4 (Guarantee reagent, GR). The adsorption experiments
were executed in a series of 250 mL conical flasks by taking 100 mL
KH2PO4 solutions at the initial phosphate concentration from 0 to
512 mg P/L with the substrates dosage of 10g. The mixtures were
shaken at a speed of 120 rpm thermostatic shaker at 25 ± 1  C for
24 h. The substrates were separated from the mixture by the
filtration at once after shaking. The residual concentration of
phosphate was measured by the standard methods (APHA, 2005).
282 X. Zhang et al. / Water Research 96 (2016) 280e291
Table 1
Characteristics of the raw water.
Parameters pH DO (mg/L)
Concentration 6.69e7.33 3.28e6.78
Average 7.02 3.29
Standard deviation 0.24 1.33
All the experiments were carried out in duplicate. The phosphate
adsorption capacity qe was given from the following relations (Eq.
(2)):
ðCo  Ce ÞV
qe ¼
m
where C0 and Ce are the initial and equilibrium concentration of
phosphate in solution (mg/L), qe is the equilibrium adsorption ca-
pacity (mg/g), m is the adsorbent dry weight (g), and V is the
suspension volume (L).
Based on the above optimized condition, the adsorption iso-
therms of the bio-ceramic/Zn-LDHs and natural bio-ceramic sub-
strates to phosphate were studied at room temperature. The
commonly model of Langmuir (Eq. (3)) and Freundlich (Eq. (4))
were used to describe the adsorption equilibrium:
Ce 1 Ce
¼ þ
qe KL qm qm
1 the adsorbents at equilibrium and at time t (h), respectively. k1
lgqe ¼ lgKf þ lgCe
n
where Ce is the equilibrium concentration of phosphate in solution
(mg/L), qe is the amount of phosphate adsorbed at equilibrium (mg/
kg), qm is the theoretical maximum sorption capacity (mg/kg), KL is
Langmuir adsorption equilibrium constant which is related to the
energy of adsorption (L/mg), Kf is Freundlich adsorption equilib-
rium constant, n is a constant which represent the Freundlich
isotherm curvature.
2.6. Desorption studies
Desorption experiments were carried out in different salt/alkali
solution. 5 M NaCl, 0.1 M NaOH and different volume ratio of 5 M
NaCl and 0.1 M NaOH were examined to find an appropriate me-
dium for desorption.
The desorbed capability of the adsorbent was assessed as fol-
lows: phosphate loaded adsorbent was separated from the mixture
by filtration after adsorption studies and washed in de-ionized
water for three times to remove any un-adsorbed phosphate. The
phosphate loaded adsorbent was mixed with 100 mL different
kinds of salt/alkali solution and kept in a shaker for 24 h. The
desorbed phosphate was analyzed by standard methods as
mentioned previously. All the experiments were carried out in
duplicate.
2.7. Kinetic studies
Phosphate adsorption kinetic study was carried out by adding
10g adsorbents to a 250 ml flask filled with 100 ml phosphate so-
lution at the initial concentration of 4 mg/L. The flask was placed in
a thermostatic shaker at 25 ± 1  C and shaken at 120 rpm. The
subsequent procedure was similar to that for phosphate adsorp-
tion. All the experiments were carried out in duplicate. The amount
of phosphate adsorbed on the adsorbents at time was calculated by
the following equation (Eq. (5)):
TP (mg/L) TDP (mg/L) SRP (mg/L)
1.620e3.328 1.030e2.396 0.747e2.101
2.631 1.886 1.578
0.642 0.445 0.417
ðCo  Ct ÞV
qt ¼ (5)
m
where C0 and Ct are the initial and given time t concentration of
(2) phosphate in solution (mg/L), qt is the amount of phosphate
adsorbed at time t (mg/g), m is the adsorbent dry weight (g), and V
is the suspension volume (L).
In order to clarify the adsorption kinetics of phosphate onto bio-
ceramic/Zn-LDHs and natural bio-ceramic, the pseudo-first-order
(Eq. (6)) and pseudo-second-order (Eq. (7)) were used. The liner
form of the models equation could be expressed as follows:
k1
lgðqe  qt Þ ¼ lgqe  t (6)
2:303
t 1 t
¼ þ (7)
qt k2 q2e qe
(3)
where qe and qt (mg/g) are the amount of phosphate adsorbed on
(4) (h1) and k2 (g/(mg$h)) are the constant of the pseudo-first-order
and pseudo-second-order model.
2.8. Statistic
Statistical analyses were determined using Excel 2007 and SPSS
19.0 software (SPSS Inc., Chicago, IL, USA), including analysis of
variance (ANOVA), Levene's test for homogeneity of variance and
normality, and Post Hoc Multiple Comparisons were also per-
formed to test which bio-ceramic/Zn-LDHs was a statistically sig-
nificant difference, and used the Bonferroni test. In all tests,
differences were considered statistically significant when p < 0.05
(SPSS, 2003).
3. Results and discussions
3.1. Characteristics of bio-ceramic/Zn-LDHs
In order to better understand the coated behavior of the bio-
ceramic/Zn-LDHs, a fine characterization of the textural proper-
ties and chemical compositions of bio-ceramic/Zn-LDHs were
conducted.
SEM images in Fig. 1 showed the shape and size of bio-ceramic/
Zn-LDHs composites and natural bio-ceramic substrates. It was
observed that the morphologies of bio-ceramic/FeZn-LDHs aggre-
gated mass of blossom with small and flaky debris (Fig. 1a), while
flaky platelets were mainly observed in bio-ceramic/CoZn-LDHs
and bio-ceramic/AlZn-LDHs materials (Fig. 1b, c). In addition, the
surface of bio-ceramic/FeZn-LDHs was rough and had lots of folds,
which was quite different with bio-ceramic/(CoZn, AlZn)-LDHs
with a mass of flaky platelets aggregated on the surface. Compar-
atively, the surface of the natural bio-ceramic was smoother than
the bio-ceramic/Zn-LDHs and showed no flaky crystal material.
Those results can be remarked that bio-ceramic/Zn-LDHs formed a
highly porous network and responsible for the high surface of LDHs
materials. Furthermore, the surface of the bio-ceramic/Zn-LDHs
 X. Zhang et al. / Water Research 96 (2016) 280e291 283

Fig. 1. SEM images of bio-ceramic/Zn-LDHs (FeZn-LDHs (a), CoZn-LDHs (b) and AlZn-LDHs (c)) and natural bio-ceramic (d) (Mag: 10 K, Resolution: 1 mm).

was overlapping crystals like the layer stacked, which was typical
for layered double hydroxides.
Chemical analysis (XRFS) results of bio-ceramic/Zn-LDHs and
natural bio-ceramic were shown in Table 2. The main composites of
the natural bio-ceramic were silica (70.620%) and alumina
(17.551%). After the process of LDHs coated on the natural bio-
ceramic, additional element of Zn was emerged. The content of
the Zn in bio-ceramic/Zn-LDHs were 0.086% for FeZn-LDHs, 0.071%
for CoZn-LDHs and 0.058% for AlZn-LDHs. In particular, compared
with the content of Fe and Co in natural bio-ceramic (5.507% and
0.011%), the natural bio-ceramic coated by FeZn-LDHs and CoZn-
LDHs were also increased 0.021% for Fe and 0.027% for Co,
respectively. Furthermore, based on SEM and XRFS results, the
coated method changed the natural bio-ceramic physical charac-
teristics and chemical compositions, which confirmed the way of
LDHs coated on the surface of the natural bio-ceramic to synthesize
bio-ceramic/Zn-LDHs were feasible.
3.2. Purification experiments
The TP, TDP and SRP removal performance in CRIS during this
study period was shown in Fig. 2. The TP average removal rates of
the bio-ceramic/Zn-LDHs (FeZn-LDHs, CoZn-LDHs and AlZn-LDHs)
were 71.58% (63.25%e79.41%), 64.92% (57.69%e79.41%) and
69.12% (57.10%e82.35%), respectively, and the removal rates of the
bio-ceramic/FeZn-LDHs showed the relative stability among them.
Comparatively, the removal rates of the natural bio-ceramic were
39.38% (26.28%e55.88%) (Fig. 2a, d).
The average removal rates of TDP observed in bio-ceramic/Zn-
LDHs which were FeZn-LDHs, 74.91% (61.17%e91.42%); CoZn-
LDHs, 68.85% (55.83%e87.31%) and AlZn-LDHs, 72.23% (54.37%e
89.18%). However, the average removal rates in natural bio-ceramic
units were 33.58% (15.70%e64.18%) (Fig. 2b, d).
The SRP average removal rates were 82.31% (71.57%e98.29%),
73.88% (62.75%e94.44%) and 77.67% (62.84%e96.15%) for bio-
ceramic/Zn-LDHs (FeZn-LDHs, CoZn-LDHs, AlZn-LDHs), respec-
tively, while in natural bio-ceramic unit it reached 33.25% (18.63%e
72.65%) (Fig. 2c, d) and the removal rates were unstable due to the
high fluctuations (Fig. 2d).
According to the removal rates of the TP, TDP and SRP, the bio-
ceramic/FeZn-LDHs obtained the best purification performance
among the bio-ceramic/Zn-LDHs and natural bio-ceramic.

Table 2
Main chemical compositions of bio-ceramic/Zn-LDHs and natural bio-ceramic/%.

Compositions Substrates

Bio-ceramic/FeZn-LDHs Bio-ceramic/CoZn-LDHs Bio-ceramic/AlZn-LDHs Natural bio-ceramic

Al2O3 17.037 17.071 17.182 17.551

SiO2 70.351 70.628 70.800 70.620
K2O 2.168 2.151 2.108 2.157
CaO 0.606 0.613 0.526 0.681
Fe2O3 5.528 5.532 5.551 5.507
Co3O4 0.022 0.038 0.010 0.011
ZnO 0.086 0.071 0.058 None
Cl 0.199 0.140 0.189 0.018
CO2 0.407 0.201 0.280 0.126
284 X. Zhang et al. / Water Research 96 (2016) 280e291

90 100
(a) Bio-ceramic/FeZn-LDHs Bio-ceramic/CoZn-LDHs (b) Bio-ceramic/FeZn-LDHs Bio-ceramic/CoZn-LDHs
Bio-ceramic/AlZn-LDHs Natural bio-ceramic Bio-ceramic/AlZn-LDHs Natural bio-ceramic
90
80
80

TDP removal rates (%)

TP removal rates (%)

70
70

60 60

50
50
40

40 30

20
30
10
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9
Experiments Experiments

(c) Bio-ceramic/FeZn-LDHs Bio-ceramic/CoZn-LDHs 90

(d) TP TDP SRP
100 Bio-ceramic/AlZn-LDHs Natural bio-ceramic

90 80

Average removal rates (%)

SRP removal rates (%)

80 70

70
60
60
50
50

40 40

30
30
20
20
0 1 2 3 4 5 6 7 8 9 FeZn-LDHs CoZn-LDHs AlZn-LDHs Natural bio-ceramic
Experiments Bio-ceramic/Zn-LDHs and natural bio-ceramic

Fig. 2. Removal rates of (a) total phosphorus (TP), (b) total dissolved phosphorus (TDP), (c) soluble reactive phosphorous (SRP) and (d) average removal rates with the Bio-ceramic/
Zn-LDHs and Natural bio-ceramic.

Compared TP, TDP and SRP removal rates with the natural bio-
ceramic, it improved the removal rates of 32.20%, 41.33% and
49.06%, respectively. As for the bio-ceramic/AlZn-LDHs, it out-
performed the natural bio-ceramic 29.74% (TP), 38.65% (TDP) and
44.42% (SRP), respectively. Moreover, the bio-ceramic/CoZn-LDHs
enhanced the removal rates of 25.54% for TP, 35.27% for TDP and
40.63% for SRP.
Meanwhile, the bio-ceramic/Zn-LDHs efficiently enhanced the
SRP removal rates (49.06%, 40.63% and 44.42%) compared to the TP
(32.20%, 25.54% and 29.74%) and TDP (41.33%, 35.27% and 38.65%)
removal performance. In other words, coating Zn-LDHs on the
natural bio-ceramic efficiently improved the SRP removal rates
which in turns to improve the removal rates of TP and TDP.
Accordingly, ANOVA analysis of the TP, TDP and SRP removal rates
showed statistically significant difference (P < 0.05) between the
bio-ceramic/Zn-LDHs and natural bio-ceramic (Post Hoc Test,
Table 3), indicating the coated method to synthesize bio-ceramic/
Zn-LDHs composites were meaningful in TP, TDP and SRP removal
in the study period in our CRIS.
The major processes responsible for TP removal in CRIS attrib-
uted to substrates interception, adsorption, chemical precipitation
and assimilation into microbial (Xu et al., 2013). Compared the
average removal rates with the natural bio-ceramic, the higher
efficiencies of the bio-ceramic/Zn-LDHs should be attributed to the
properties of the LDHs, namely the hierarchical porous structures,
large surface area and high anion exchange capacity. The average
removal rates of TP observed in natural bio-ceramic were better
than Liu et al. (2009) who reported the maximum removal of TP
was 38.40%. The less efficient in TP removal may attribute to the
large average diameter (8e12 mm) and high hydraulic conductivity
(88.8 mm/s) of the bio-ceramic.
The total phosphorus divided in two parts which were total
dissolved phosphorus and particulate phosphorus. In this regard,
the total dissolved phosphorus removal depends on the substrates
physical and chemical adsorption, chemical and microbial reaction.
Compared the main chemical compositions with the natural bio-
ceramic, the coated method increased the content of the metal
ion such as Fe and Zn, which encouraged the conversion of dis-
solved phosphorus into particulate phosphorus. Accordingly, the
coated method under alkaline conditions caused the surface of the
natural bio-ceramic more porous. Most likely, the presence of LDHs
provided a more effective distribution of the porous structure,
which encouraged the more effective adsorption on the bio-
ceramic surface. Furthermore, the coated method may delay the
wastewater passage through the system which in turn to increase
the hydraulic retention time and as a consequence, increase the
removal efficiencies.
The key mechanisms of phosphate removal from wastewater in
CRIS are physic-chemical process. In addition, the adsorption of
phosphate via electrostatic attraction, ion exchange and ligand
exchange (Chitrakar et al., 2006). The outperformed removal rates
of the bio-ceramic/Zn-LDHs may attribute to the process of fixation
of phosphate by Fe in the substrates were enhanced (Netter, 1992),
indicating that process of coating LDHs on the natural bio-ceramic
improved the role of chemical reaction. Besides, it may associate
with the ion exchange process between the interlay anion and
 X. Zhang et al. / Water Research 96 (2016) 280e291 285

Table 3
Levene test and ANOVA analysis of removal rates for TP, TDP and SRP with bio-ceramic/Zn-LDHs and natural bio-ceramic.

Removal rates Test

Levene test Post hoc multiple comparisons (bonferroni)

Levene statistic Sig. Mean(%) Std. Error Sig. 95% confidence

Lower Upper

TP FeZn-LDHs 0.504 0.682 32.197 3.743 0.000 42.722 21.671

CoZn-LDHs 25.541 3.743 0.000 36.067 15.016
AlZn-LDHs 29.742 3.743 0.000 40.267 19.217
TDP FeZn-LDHs 0.714 0.551 41.332 5.388 0.000 56.485 26.180
CoZn-LDHs 35.269 5.388 0.000 50.422 20.116
AlZn-LDHs 38.649 5.388 0.000 53.802 23.496
SRP FeZn-LDHs 0.815 0.495 49.057 5.348 0.000 64.098 34.016
CoZn-LDHs 40.631 5.348 0.000 55.672 25.590
AlZn-LDHs 44.418 5.348 0.000 59.459 29.377

FeZn-LDHs represent Bio-ceramic/FeZn-LDHs composites. Sig. represents Sig. (2 tailed) which also calls P value. Levene's test represents the Test of Homogeneity of Variances.
Mean (%) represents Mean Difference (%). Std. represents mean Standard Error. 95% confidence represents 95% confidence interval of the difference. Lower and upper
represent Lower and Upper bound, respectively. The mean difference is significant at the 0.05 level.

phosphate. Additionally, the three different bio-ceramic/Zn-LDHs
obtained the different purification effects. Due to the three
different LDHs had the same interlayer anions, the varied removal
capacity of LDHs was related to the nature of the precursor metals,
the layer charge density and the orientation of the guest (Goh et al.,
2008). The ionic radius of the precursor metal ions was found to
reduce in the order of: Fe3þ>Co3þ>Al3þ (Cavani et al., 1991). As the
metal ratio of M3þ/M2þ (Zn2þ) decrease, the distance of the cations
was also decrease. Therefore, the basal space of layered framework
was increase. In this regard, the positive charge density of the
layered framework was increased which in turn to enhance the
ionic interactions and as a consequence, exhibited the higher ability
to capture the oxyanions, such as phosphate (Sahoo et al., 2014).
However, the average removal rates of phosphate by the three bio-
ceramic/Zn-LDHs in this study showed the different results. Un-
usual removal behavior is probably controlled by some physical
properties of LDHs such as crystallinity, particle size and surface
area (Halajnia et al., 2012), even the amount of the co-intercalated
water molecule (Costa et al., 2012). Furthermore, those results were
in agreement with Boclair and Braterman (1999) who reported the
corresponding FeZn-LDHs was more stable than the AlZn-LDHs.
3.3. Particulate phosphorus removal
The average removal efficiencies of particulate phosphorus (PP)
of the bio-ceramic/Zn-LDHs and the natural bio-ceramic were
67.58% (FeZn-LDHs), 62.40% (CoZn-LDHs), 63.14% (AlZn-LDHs) and
57.08%, respectively (Fig. 3a). Those results indicated that the way
of coating LDHs on the surface of natural bio-ceramic to synthesize
the new composites of bio-ceramic/Zn-LDHs actually increased the
removal performance of PP.
The removal of PP mainly depends on the physical interception
mechanism and the particulate phosphorus could be intercepted
and trapped in the test columns. Therefore, in order to measure the
actual values of physical interception of PP in the whole purification
experiments, all the substrates in the test columns would be took
out respectively and then be washed thoroughly with de-ionized
water for three times after the whole purification experiments
ended. On the other hand, the purification amount (PP) could be
calculated by the difference of numerical values between the
removal rates of TP and TDP, and the results were illustrated in
Fig. 3b. The results indicated that the purification amount (PP) for
the bio-ceramic/Zn-LDHs and natural bio-ceramic were 1.595 mg
for FeZn-LDHs, 1.402 mg for CoZn-LDHs, 1.567 mg for AlZn-LDHs
and 1.386 mg for natural bio-ceramic, respectively. Compara-
tively, the ratio of the actual physical interception values and the
purification amount of PP were 63.70% for FeZn-LDHs, 86.15% for
CoZn-LDHs, 58.82% for AlZn-LDHs and 83.25% for natural bio-
ceramic. Those results confirmed that the main removal mecha-
nism of PP for the bio-ceramic/CoZn-LDHs and the natural bio-

(a) (b) Puricication values

80 1.8
Actual physical interception values
Average removal rates of PP (%)

1.6
70
Particulate Phosphorus (mg)

1.4

60 1.2

1.0
50
0.8

40 0.6

0.4
30
0.2

20 0.0
FeZn-LDHs CoZn-LDHs AlZn-LDHs Natural bio-ceramic FeZn-LDHs CoZn-LDHs AlZn-LDHs Natural bio-ceramic
Bio-ceramic/Zn-LDHs and natural bio-ceramic Bio-ceramic/Zn-LDHs and natural bio-ceramic

Fig. 3. Average removal rates of (a) Particulate Phosphorus (PP) and (b) physical interception of PP with the Bio-ceramic/Zn-LDHs and Natural bio-ceramic.
286 X. Zhang et al. / Water Research 96 (2016) 280e291

ceramic was physical interception. Furthermore, the combination capacity was also in agreement with the other studies (Das et al.,
of physical interception and adsorption mechanisms attributed to 2006; Cheng et al., 2009).
the bio-ceramic/FeZn-LDHs and AlZn-LDHs.
3.5. Desorption studies

3.4. Adsorption isotherm
Adsorption capacity at different aqueous equilibrium concen-
trations could be illustrated by the adsorption isotherm (Fig. 4) and
the linear plots of Langmuir and Freundlich models were showed in
Fig. 5. The fitted constants and regression coefficients were sum-
marized in Table 4. It can be observed that Freundlich equation gave
more satisfactory fitting to the adsorption isotherms of phosphate
on bio-ceramic/Zn-LDHs and natural bio-ceramic with correlation
coefficient R2 higher than 0.95, indicating the adsorptions on het-
erogeneous surface sites (Radrigues and Pinto da Silva, 2010).
However, the results of the bio-ceramic/CoZn-LDHs indicated that
the Langmuir equation (R2 ¼ 0.9910) fitted better than the
Freundlich equation (R2 ¼ 0.9726), suggesting that the adsorption
of phosphate attributed to monolayer adsorption on the outer
surface of adsorbent (Zhou et al., 2011).
In Freundlich equation, the constant of 1/n for the bio-ceramic/
Zn-LDHs and natural bio-ceramic in this study were 0.4363, 0.4470,
0.4122 and 0.7024, respectively (Fig. 5b), which were in the range of
0e1, demonstrating a good adsorption of phosphate (Cheng et al.,
2010). The higher Kf values indicated the adsorption capacity of
bio-ceramic/Zn-LDHs were higher than the natural bio-ceramic
(Table 4). On the basis of values of qm in Langmuir equation, the
theoretical phosphate adsorption capacity of the bio-ceramic/Zn-
LDHs and natural bio-ceramic followed the order of: AlZn-LDHs
(769.23 mg/kg)>FeZn-LDHs (476.19 mg/kg)> Natural bio-ceramic
(434.78 mg/kg)>CoZn-LDHs (357.14 mg/kg) (Table 4). The
sequence of AlZn-LDHs > FeZn-LDHs in theoretical adsorption
0.6 FeZn-LDHs
qe (mg/g)
The prior desorption experiments results were given in Fig. 6a.
The pH of the solution after mixing 5 M NaCl with 0.1 M NaCl at the
volume ratio of 1:1, 3:1, 4:1 and 9:1 were 12.92, 12.48, 12.36 and
11.97, respectively. The desorption rates were less efficient in 0.1 M
NaOH, 5 M NaCl and the mixed solution of 5 M NaCl and 0.1 M
NaOH with the volume ratio of 4:1 and 9:1. The effective desorption
of phosphate (50.07%) could be achieved in the volume ratio of 1:1
(5 M NaCl þ 0.1 M NaOH), with negligible dissolution of adsorbent
(Fig. 6a). These results coincided with those found by Chitrakar
et al. (2006) who assessed the maximum desorption rates of
MgMn-LDHs could be achieved by using a mixed solution of 5 M
NaCl þ 0.1 M NaOH. Herein, 5 M NaCl þ 0.1 M NaOH (1:1) were used
for the subsequent desorption experiments.
The desorption results of the bio-ceramic/Zn-LDHs and natural
bio-ceramic were presented in Fig. 6b. The amount of desorbed
phosphate from the adsorbent showed an increasing trend with the
increasing initial concentration phosphate on the adsorbent.
However, the desorption rates of the bio-ceramic/Zn-LDHs and
natural bio-ceramic were not obtained in the same initial concen-
tration. Therefore, the simulated curves were used to evaluate the
desorption capacity of the bio-ceramic/Zn-LDHs and natural bio-
ceramic. At the initial concentration of 4 mg/L on the adsorbent,
the desorption rates of the bio-ceramic/Zn-LDHs were 35.09% for
FeZn-LDHs, 23.73% for CoZn-LDHs, 28.80% for AlZn-LDHs and
16.14% for the natural bio-ceramic. Accordingly, the desorption
rates of coated and natural substrates were followed the order of
FeZn-LDHs > AlZn-LDHs > CoZn-LDHs > Natural bio-ceramic.
Furthermore, the way of coating Zn-LDHs on the natural bio-
ceramic improved the desorption capacity, which indicated the
Zn-LDHs coated bio-ceramic were more efficiently in regeneration
for its subsequent application as an adsorbent.

0.4 3.6. Kinetic studies

0.2 Fig. 7 presents the results of fitting experimental data of the

pseudo-first-order and pseudo-second-order models. Likewise it
0.0 can be seen from Table 5 that the correlation coefficient (R2) for the
0.4 CoZn-LDHs bio-ceramic/Zn-LDHs varies in the order: pseudo-second-
qe (mg/g)

0.3 order > pseudo-first-order, suggesting that the adsorption in this

case was in fact a chemisorptions process. However, the correlation
0.2
coefficient (R2) of the natural bio-ceramic showed the pseudo-first-
0.1 order fitted better than the pseudo-second-order.
0.0 Furthermore, it was observed that qe,cal value of the bio-
0.75
AlZn-LDHs ceramic/Zn-LDHs computed from the pseudo-first-order plot
qe (mg/g)

deviated considerably from the experimental values (qe), while the

0.50 values computed from pseudo-second-order were much closer to
the experimental values (qe). These results indicated that the
0.25
pseudo-second-order model gave a better fitted to the adsorption
0.00 of phosphate which was coincided with the results obtained from
0.4 Natural bio-ceramic the correlation coefficient (R2). Comparatively, qe,cal value of the
qe (mg/g)

0.3 natural bio-ceramic showed the opposite results, indicating that

the adsorption kinetic of phosphate on the natural bio-ceramic was
0.2
pseudo-first-order model.
0.1
0.0 4. Discussions
0 50 100 150 200 250 300 350 400 450 500
Ce (mg/L) Desorption studies give an idea of the nature of the adsorption
process. Chemical adsorption exhibits the poor desorption rates,
Fig. 4. Adsorption isotherms of phosphate onto the bio-ceramic/Zn-LDHs and the indicating the adsorbate species are held firmly to the adsorbent
Natural bio-ceramic. with comparatively stronger bonds (Islam and Patel, 2010). The
 X. Zhang et al. / Water Research 96 (2016) 280e291 287

1.4
(a)
(b)
Ce/qe(kg/L)
1.2 FeZn-LDHs
1.0 3.0
y=0.0021x+0.0655 FeZn-LDHs
0.8 2.5
R2=0.9557 y=0.4363x+1.5688

lgqe
0.6 2.0
0.4
1.5
R2=0.9860
0.2
0.0 1.0
CoZn-LDHs 0.5
Ce/qe(kg/L)

1.5 y=0.0028x+0.0538 2.45

CoZn-LDHs
R2=0.9910 y=0.4470x+1.4909
1.0 1.96 R2=0.9726

lgqe
0.5 1.47

0.0 0.98
AlZn-LDHs
AlZn-LDHs
Ce/qe(kg/L)

0.66 y=0.0013x+0.0484 3.0

y=0.4122x+1.7575
0.44 R2=0.9265 2.5
R2=0.9740

lgqe
2.0
0.22 1.5
0.00 1.0
Natural bio-ceramic Natural bio-ceramic
2.0
Ce/qe(kg/L)

y=0.0023x+0.2638 2.36
y=0.7024x+0.8005
1.5
R2=0.8637 R2=0.9546

lgqe
1.77
1.0
0.5 1.18

0.0 0.59
0 50 100 150 200 250 300 350 400 450 500 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0
Ce (mg/L) lgCe

Fig. 5. Langmuir (a) and Freundlich (b) plots for the adsorption of phosphate onto Bio-ceramic/Zn-LDHs and Natural bio-ceramic.

Table 4 performance of phosphate in the sewage by the physical adsorption

Adsorption isotherm parameters of Bio-ceramic/Zn-LDHs and Natural bio-ceramic. and the desorbed phosphate by the chemical reaction. Desorption
Adsorbent Langmuir isotherm model Freundlich isotherm study were used to simulate the amount of the phosphate by the
model chemical adsorption and the un-desorbed phosphate acted by the
qm KL R2 Kf n R2
chemical reaction. Herein, the theoretical simulated removal per-
formance of phosphate was calculated by the physical and chemical
FeZn-LDHs 476.1905 0.0321 0.9557 37.0510 2.2920 0.9860
adsorption and chemical reaction. The results of actual measured
CoZn-LDHs 357.1429 0.0520 0.9910 30.9671 2.2371 0.9726
AlZn-LDHs 769.2308 0.0269 0.9265 57.2137 2.4260 0.9740 and theoretical simulated values of phosphate were illustrated in
Natural bio-ceramic 434.7826 0.0087 0.8637 6.3168 1.4237 0.9546 Fig. 8.
Due to the part of simulation were used the adsorption iso-
therms which were operated in the ideal conditions, the difference
desorbed phosphate by the chemical reaction depends on the of the actual measured and theoretical simulated values were
species and the concentration of desorption solution, namely, the certainly existed. The purification experiments were affected by the
desorbed phosphate may be incomplete. The mechanisms of characteristic of the raw water, such as the coexisting anions and
phosphate removal are physical adsorption, chemical adsorption heavy metal ions. Therefore, the average percentage of the actual
and chemical reaction. Herein, the Langmuir or Freundlich measured and theoretical simulated values were only 49.62% for
adsorption isotherms were used to simulate the removal FeZn-LDHs, 46.37% for CoZn-LDHs, 47.13% for AlZn-LDHs and

(a) (b)
70 9 FeZn-LDHs
Concentration of phosphate desorbed

CoZn-LDHs
60 8
AlZn-LDHs
7 Natural bio-cermaic
Desorption rates (%)

of adsorbent (mg/L)

50
6

40 5

4
30
3
20 2

1
10
0
1:1 3:1 4:1 9:1 0.1M NaOH 5M NaCl 0 5 10 15 20 25 30 35 40 45
Volume ratio of 5M NaCl and 0.1M NaOH Initial concentration of the phosphate
on the adsorbent (mg/L)

Fig. 6. Phosphate desorption (a) from the natural bio-ceramic with various reagents; (b) desorption experiments of the bio-ceramic/Zn-LDHs and natural bio-ceramic after the
adsorption study with the mixed solution of 5 M NaCl and 0.1 M NaOH at the volume ratio of 1:1.
288 X. Zhang et al. / Water Research 96 (2016) 280e291

-0.5 (a) 600

(b)
FeZn-LDHs FeZn-LDHs
-1.0 y=-0.0777X-1.7109 450 y=27.9909x+32.8769
lg(qe-qt)

R2=0.9584 R2=0.9697

t/qt
-1.5 300

-2.0 150

-2.5 0
CoZn-LDHs CoZn-LDHs
-0.8 y=-0.0784x-1.7096 690 y=33.4555x+60.0426
lg(qe-qt)

R2=0.9106 R2=0.9430

t/qt
460
-1.6
230
-2.4
0
-1.0 AlZn-LDHs 400 AlZn-LDHs
y=-0.0518x-1.9016 y=27.5612x+18.8884
lg(qe-qt)

300
-1.5
R2=0.9195 R2=0.9808

t/qt
200
-2.0
100
-2.5 0
0.00 Natural bio-ceramic 2500
Natural bio-ceramic
y=-0.0762x-1.8659 y=41.3802x+787.0188
lg(qe-qt)

-0.92 2000
R2=0.8948 R2=0.5183

t/qt
1500
-1.84
1000
-2.76 500
0 1 2 3 4 5 6 7 8 9 10 11 12 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
t (h) t (h)

Fig. 7. Pseudo-first-order (a) and pseudo-second-order (b) kinetics for adsorption of phosphate onto Bio-ceramic/Zn-LDHs and Natural bio-ceramic.

25.59% for the natural bio-ceramic (Fig. 8b). for FeZn-LDHs, 0.515 mg for CoZn-LDHs, 0.536 mg for AlZn-LDHs
The theoretical simulated removal performance of phosphate in and 0.505 mg for the natural bio-ceramic (Fig. 9b). Furthermore,
the sewage by the physical adsorption and desorbed phosphate by the removal mechanism of physical adsorption was not efficiently
the chemical reaction were shown in Fig. 9a, b. Those results improved.
indicated that the theoretical simulated removal amount of the According to the initial concentration of the phosphate on the
phosphate for the bio-ceramic/Zn-LDHs and natural bio-ceramic adsorbent after the adsorption study and the equilibrium concen-
showed the small difference. The average amount was 0.536 mg tration of phosphate in solid phase after the desorption study, the
simulated curve showed the good relationship of linear for the bio-
ceramic/Zn-LDHs and natural bio-ceramic (R2 > 0.99, Fig. 10).
Table 5
Herein, the simulated curves were used to calculate the amount of
Kinetic models for phosphate adsorption on Bio-ceramic/Zn-LDHs and Natural bio- the phosphate by the chemical adsorption and the un-desorbed
ceramic. phosphate by the chemical reaction (Fig. 9c, d).
Samples qe Pseudo-first-order Pseudo-second-order
The results depicted in Fig. 9c showed that the phosphate
removal by the chemical adsorption for the bio-ceramic/Zn-LDHs
qe, cal k1 R2 qe, cal k2 R2
were higher than the natural bio-ceramic. The average removal
FeZn-LDHs 0.035 0.019 0.1789 0.9584 0.036 23.8310 0.9697 amount of phosphate was 0.319 mg for FeZn-LDHs, 0.318 mg for
CoZn-LDHs 0.029 0.020 0.1806 0.9106 0.030 18.6413 0.9430
CoZn-LDHs, 0.334 mg for AlZn-LDHs and 0.150 mg for the natural
AlZn-LDHs 0.037 0.013 0.1193 0.9195 0.036 40.2162 0.9808
Natural bio-ceramic 0.011 0.014 0.1755 0.8948 0.024 2.1757 0.5183 bio-ceramic (Fig. 9d).

(a) (b) Theoretical simulated values

Average removal amount of phosphate (mg)

FeZn-LDHs (simulated) CoZn-LDHs (simulated)

AlZn-LDHs (simulated) Natural bio-ceramic (simulated) Actural measured values
1.4
Removal amount of phosphate (mg)

FeZn-LDHs CoZn-LDHs 1.2

AlZn-LDHs Natural bio-ceramic
1.2
1.0

1.0
0.8
0.8
0.6
0.6

0.4
0.4

0.2 0.2

0.0 0.0
0 1 2 3 4 5 6 7 8 9 10 FeZn-LDHs CoZn-LDHs AlZn-LDHs Natural bio-ceramic
Simulated times Bio-ceramic/Zn-LDHs and natural bio-ceramic

Fig. 8. Removal amount (a) and average removal amount (b) of phosphate by the actual measured and theoretical simulated values.
 X. Zhang et al. / Water Research 96 (2016) 280e291 289

0.8 (a) FeZn-LDHs CoZn-LDHs

AlZn-LDHs Natural bio-ceramic
0.7 (b)

Average amount of phosphate (mg)

0.7
Amount of the phosphate (mg) 0.6
0.6

0.5 0.5

0.4
0.4

0.3
0.3
0.2

0.1 0.2
0 1 2 3 4 5 6 7 8 9 10 FeZn-LDHs CoZn-LDHs AlZn-LDHs Natural bio-ceramic
Simulated times Bio-ceramic/Zn-LDHs and natural bio-ceramic

0.50 (c) FeZn-LDHs CoZn-LDHs (d)

AlZn-LDHs Natural bio-ceramic 0.40

Average amount of phosphate (mg)

0.45
Amount of the phosphate (mg)

0.35
0.40
0.30
0.35

0.30 0.25

0.25 0.20
0.20
0.15
0.15
0.10
0.10

0.05 0.05

0.00 0.00
0 1 2 3 4 5 6 7 8 9 10 FeZn-LDHs CoZn-LDHs AlZn-LDHs Natural bio-ceramic
Simulated times Bio-ceramic/Zn-LDHs and natural bio-ceramic

Fig. 9. The theoretical simulated removal amount (a) and average amount (b) of phosphate by the physical adsorption and desorbed phosphate by the chemical reaction; the
theoretical simulated removal amount (c) and average amount (d) of phosphate by the chemical adsorption and the un-desorbed phosphate by the chemical reaction.

According to the purification experiments results, the bio- phosphate removal indicated that coating Zn-LDHs on the natural
ceramic/Zn-LDHs were more efficiently than the natural bio- bio-ceramic efficiently enhanced the chemical effect, such as the
ceramic for the phosphate removal. However, the theoretical capacity of anions exchange. Therefore, the removal mechanisms of
simulated removal performance of the phosphate by the physical phosphate by the bio-ceramic/Zn-LDHs were physic-chemical
adsorption and desorbed phosphate by the chemical reaction process, but the part of removal amount by chemical process was
showed small difference (Fig. 9b). At the same time, the theoretical efficiently enhanced.
simulated removal performance of the phosphate by the chemical
adsorption and un-desorbed phosphate by the chemical reaction
demonstrated the large improvement, which the phosphate 5. Conclusions
removal amount was twice higher than the natural bio-ceramic
(Fig. 9d). It represented that the LDHs coated on the natural bio-  Three different novel core-shell bio-ceramic/Zn-LDHs adsor-
ceramic to synthesis bio-ceramic/Zn-LDHs were efficiently bents were synthesized by co-precipitation method and char-
increased the chemical effect. This should attribute to the presence acterized by SEM and XRFS, proving the successful synthesis of
of surface hydroxyl groups (MeOH) on the LDHs could be bio-ceramic/Zn-LDHs.
exchanged by the adsorbed phosphate groups (MeOeP), indicating  Bio-ceramic/Zn-LDHs efficiently enhanced the removal perfor-
a ligand exchange process occurred during adsorption period (Yang mance of phosphorus compared to the natural bio-ceramic,
et al., 2014). In addition, the interlayer anions could also be implying the way to coat Zn-LDHs on the natural bio-ceramic
exchanged by the phosphate species, demonstrated ion exchange was meaningful.
reaction presented during the phosphate removal process (He et al.,  Freundlich model was successfully applied to the adsorption
2010; Park et al., 2010). Moreover, the relatively interlayer bonding data of the bio-ceramic/FeZn(AlZn)-LDHs and the natural bio-
of LDHs accelerated the reaction of anions exchange. Herein, those ceramic, indicating the adsorption of phosphate was multi-
results demonstrated the large improvement in the purification layer adsorption.
experiments.  Pseudo-second-order kinetic model was better fitted to bio-
Based on the above results, coating LDHs on natural bio-ceramic ceramic/Zn-LDHs, revealing that the adsorption process of
improved the Fe elements content, implying it may improve the phosphate was chemical adsorption.
process of fixation of phosphate by Fe element. On the other hand,  Coating Zn-LDHs on the natural bio-ceramic to synthesize bio-
the kinetic study indicated that the adsorption of phosphate by the ceramic/Zn-LDHs efficiently improved the chemical effects in
bio-ceramic were chemical adsorption and the simulation of the phosphate removal, enhanced the phosphorus removal rates,
solved the appreciable problem of the application of powder
290 X. Zhang et al. / Water Research 96 (2016) 280e291

Cheng, X., Huang, X.R., Wang, X.Z., et al., 2009. Phosphate adsorption from sewage
25 FeZn-LDHs sludge filtrate using zinc-aluminum layered double hydroxides. J. Hazard. Mater
Equilibrium concentration of phosphate in solid phase (mg/L)

20 y=0.7517x 169 (1/3), 958e964.

15 R2=0.9959 Cheng, X., Huang, X.R., Wang, X.Z., et al., 2010. Influence of calcination on the
adsorptive removal of phosphate by Zn-Al layered double hydroxides from
10 excess sludge liquor. J. Hazard. Mater 177 (1/3), 516e523.
5 Chitrakar, R., Tezuka, S., Sonoda, A., et al., 2006. Selective adsorption of phosphate
from seawater and wastewater by amorphous zirconium hydroxide [J]. J. Colloid
0 Interface Sci. 297 (2), 426e433.
CoZn-LDHs Costa, D.G., Rocha, A.B., Souza, W.F., et al., 2012. Comparative structural, thermo-
24.9 y=0.8234x dynamic and electronic analyses of Zn-Al-An- hydrotalcite like compounds
R2=0.9986 (An ¼Cl, F, Br, OH, CO2 
3 or NO3 ): an ab initio study. Appl. Clay Sci. 56,
16.6 16e22.
Das, J., Patra, B.S., Baliarsingh, N., et al., 2006. Adsorption of phosphate by layered
8.3 double hydroxides in aqueous solutions. Appl. Clay Sci. 32 (3/4), 252e260.
Goh, K.H., Lim, T.T., Dong, Z.L., 2008. Application of layered double hydroxides for
0.0 removal of oxyanions: a review. Water Res. 42 (6/7), 1343e1368.
AlZn-LDHs
40 Halajnia, A., Oustan, S., Najafi, N., et al., 2012. The adsorption characteristics of ni-
y=0.8146x trate on Mg-Fe and Mg-Al layered double hydroxides in a simulated soil solu-
30
R2=0.9969 tion. Appl. Clay Sci. 70, 28e36.
He, H.M., Kang, H.L., Ma, S.L., et al., 2010. High adsorption selectivity of ZnAl layered
20
double hydroxides and the calcined materials toward phosphate. J. Colloid
10 Interface Sci. 343 (1), 225e231.
Islam, M., Patel, R., 2010. Synthesis and physicochemical characterization of Zn/Al
0 chloride layered double hydroxide and evaluation of its nitrate removal effi-
Natural bio-ceramic ciency. Desalination 256 (1/3), 120e128.
30 y=0.8948x Jiang, X., Ma, M.C., Li, J., et al., 2011. Analysis of microbial molecular ecology tech-
20 R2=0.9991 niques in constructed rapid infiltration system. J. Earth Sci. 22 (5), 669e676.
Lau, S.S.S., Lane, S.N., 2002. Biological and chemical factors influencing shallow lake
eutrophication: a long-term study. Water Res. 288 (3), 167e181.
10 Lian, Y.L., Xu, M.Y., Zhong, Y.M., et al., 2014. Ammonia oxidizers in a pilot-scale
multilayer rapid infiltration system for domestic wastewater treatment. Plos
0 one 9 (12), 1e18.
0 5 10 15 20 25 30 35 40 45 Liu, X.Q., Tian, Y., Liu, X.L., 2009. Simulated experiment study on sewage by con-
Initial concentration of the phosphate structed rapid infiltration system. J. Beijing Techno. Bus. Univ. (Nat. Sci. Ed.) 27
on the adsorbent (mg/L) (1), 1e5.
Montangero, A., Belevi, H., 2007. Assessing nutrient flows in septic tanks by eliciting
Fig. 10. Simulated curve of the phosphate by the chemical adsorption and the un- expert judgment: a promising method in the context of developing countries.
desorbed phosphate by the chemical reaction. Water Res. 41 (5), 1052e1064.
Netter, R., 1992. The purification efficiency of planted soil filters for wastewater
treatment. Water Sci. Technol. 26 (9/11), 2317e2320.
Novillo, C., Guaya, D., Avendan ~ o, A.A., et al., 2014. Evaluation of phosphate removal
LDHs in CRIS and could be used as potential substrates for the capacity of Mg/Al layered double hydroxides from aqueous solutions. Fuel 138,
phosphorus removal in the CRIS. 72e79.
Park, K.Y., Song, J.H., Lee, S.H., et al., 2010. Utilization of a selective adsorbent for
phosphorus removal from wastewaters. Environ. Eng. Sci. 27 (9), 805e810.
Radha, A.V., Kamath, P.V., 2003. Aging of trivalent metal hydroxide/oxide gels in
Author contributions divalent metal salt solutions: mechanism of formation of layered double hy-
droxides (LDHs). Bull. Mater. Sci. 26 (7), 661e666.
Radrigues, L.A., Pinto da Silva, M.L.C., 2010. Adsorption kinetic, thermodynamic and
X.L.Z. designed the research; L.G., H.L.H and Y.J.L. performed the desorption studies of phosphate onto hydrous niobium oxide prepared by
research; X.L.Z. analyzed the data; L.G. and H.L.H wrote the main reverse micro-emulsion method. Adsorption 16, 173e181.
manuscript text; Y.H.J, M.L. and X.L.Z. revised the manuscript. All Sahoo, P., Ishihara, S., Yamada, K., et al., 2014. Rapid exchange between atmospheric
CO2 and carbonate anion intercalated within magnesium rich layered double
authors reviewed the manuscript. hydroxides. ACS Appl. Mater. Interfaces 6 (20), 18352e18359.
Seyhan, D., Erdincler, A., 2003. Effect of lime stabilisation of enhanced biological
Acknowledgements phosphorus removal sludges on the phosphorus availability to plants. Water
Sci. Technol. 48 (1), 155e162.
SPSS, 2003. Analytical Software, Statistical Package for the Social Science (SPSS)
This work was funded by the National Natural Science Foun- Headquarters. SPSS, Chicago, IL.
dation of China (NO.31270573, 31400435). The authors also thanks Sun, X.F., Imai, T., Sekine, M., et al., 2014. Adsorption of phosphate using calcined
Mg3Fe-layered double hydroxides in a fixed-bed column study. J. Ind. Eng.
Material Research and Testing Center, Wuhan University of Tech- Chem. 20 (5), 3623e3630.
nology for their technical support of the characteristics of the Wang, D.B., Zhang, Z.Y., Li, X.M., et al., 2010. A full-scale treatment of freeway toll-
natural and modified bio-ceramic substrates. gate domestic sewage using ecology filter integrated constructed rapid infil-
tration. Ecol. Eng. 36 (6), 827e831.
Westholm, L.J., 2006. Substrates for phosphorus removal-potential benefits for on-
References site wastewater treatment? Water Res. 40 (1), 23e36.
Xu, W.L., Xu, X.L., Zhang, J.Q., 2013. Removal performance and mechanism of TP in
APHA, 2005. Standard Methods for the Examinations of Water and Wastewater, improved constructed rapid infiltration system. Fresenius Environ. Bull. 22 (7),
twenty-first ed. APHA and AWWA and WEF, DC, Washington. 1826e1830.
Ayoub, G.M., Koopman, B., Pandya, N., 2001. Iron and aluminum hydroxy (oxide) Yan, L.G., Yang, K., Shan, R.R., et al., 2015. Kinetic, isotherm and thermodynamic
coated filter media for low-concentration phosphorus removal. Water Environ. investigations of phosphate adsorption onto core-shell Fe3O4@LDHs composites
Res. 73 (4), 478e485. with easy magnetic separation assistance. J. Colloid Interface Sci. 448, 508e516.
Boclair, J.W., Braterman, P.S., 1999. Layered double hydroxides stability. 1. Relative Yang, K., Yan, L.G., Yang, Y.M., et al., 2014. Adsorptive removal of phosphate by Mg-
stabilities of layered double hydroxides and their simple counterparts. Chem. Al and Zn-Al layered double hydroxides: kinetics, isotherms and mechanisms.
Mater 11, 298e302. Sep. Purif. Technol. 124, 36e42.
Boujelbena, N., Bouzida, J., Eloueara, Z., et al., 2008. Phosphorus removal from Zhang, X.L., Chen, Q.Z., Guo, L., et al., 2015. Effects of varying particle sizes and
aqueous solution using iron coated natural and engineered sorbents. J. Hazard. different types of LDH-modified anthracite in simulated test columns for
Mater 151 (1), 103e110. phosphorous removal. Int. J. Environ. Res. Public Health 12 (6), 6788e6800.
Cavani, F.F., Trifiro, A.V., Vaccari, A., 1991. Hydrotalcite type anionic clays, prepara- Zhang, X.L., Guo, L., Chen, Q.Z., et al., 2014. Coating modification of anthracite
tion, properties and applications. Catal. Today 11 (2), 173e301. substrate in vertical-flow constructed wetlands by LDHs synthesized from
Chen, J.M., 2012. Comparative experiment study on the total phosphorus removal different compounds and the nitrogen removal efficiencies. Environ. Sci. 35 (8),
efficiency of different infiltration media combinations in the CRI system. Adv. 3012e3017.
Mater. Res. 415e417, 1735e1739. Zhang, X.L., Liu, X.T., Xu, L., et al., 2013. Purification effect of vertical flow
 X. Zhang et al. / Water Research 96 (2016) 280e291 291

constructed wetlands using modified substrates coated by MgFe-LDHs. China and phosphorus. Fresenius Environ. Bull. 16 (11), 1468e1473.
Environ. Sci. 33 (8), 1407e1412. Zhou, J.B., Yang, S.L., Yu, J.G., et al., 2011. Novel hollow micro-spheres of hierarchical
Zhang, X.L., Zhang, S., He, F., et al., 2007. Differentiate performance of eight filter zinc-aluminum layered double hydroxides and their enhanced adsorption ca-
media in vertical flow constructed wetland: removal of organic matter, nitrogen pacity for phosphate in water. J. Hazard. Mater 192 (3), 1114e1121.