Professional Documents
Culture Documents
Water Research
journal homepage: www.elsevier.com/locate/watres
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.
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 281
cost and low energy consumption (Wang et al., 2010; Chen, 2012; obtain bio-ceramic/FeZn-LDHs composites.
Lian et al., 2014). During the previous research, by virtue of using
natural sand, gravel, ceramic, coal gangue and other substrates to 2.2. Characterization methods
replace the conventional natural soil medium, CRIS have been
proved to be efficiently produce better penetrability, increasing Morphological features were acquired using a scanning electron
hydraulic loading (1.0e2.0 m d1) and decreasing the space occu- microscopy (SEM, JSM-5610LV, Japan). Chemical compositions
pation (Zhang et al., 2007; Jiang et al., 2011). However, limited by were analyzed by X-Ray Fluorescence Spectrometer (XRFS, Axios,
pollution tolerance and adsorption ability, natural substrates have Panalytical. B. V., Holland). Brunauer Emmett Teller (BET) surface
been one of the worst operation problems for CRIS. Herein, it is area of the natural bio-ceramic were analyzed by N2 adsorption
imperative that effective measures should be taken to improve the at 196 C using a sorptometer (ASAP-2020, Micromeritics, USA).
substrates adsorption ability. The average voidage of the natural bio-ceramic can be obtained by
Layered double hydroxides (LDHs) are a series of compounds the following equation (Eq. (1)):
derived from the structure of mineral brucite (Cavani et al., 1991).
The general formula of LDHs can be expressed as 4V
f¼ 100% (1)
½M1X 2þ MX 3þ ðOHÞ2 ½An X=n nH2 O, where M2þ and M3þ are diva- pD2 H
lent and trivalent metal cations in the octahedral positions of the
brucite-like layers. Due to part of the divalent metal cations can be where f is the average voidage of the substrate (%), D is the inner
isomorphously substituted by trivalent metal cations, the layers diameter of the column (cm), H is the height of the natural bio-
acquires positive charge with the compositions ceramic filled into the column. V is the volume of the H height
ð½M1X 2þ MX 3þ ðOHÞ2 Xþ Þ. An, are the anions incorporated in the water flowed out of the column (mL).
interlayer region for charge neutrality, resulting in the formation of
LDHs (Radha and Kamath, 2003). LDHs have relatively weak 2.3. Water quality monitoring
interlayer bonding, as a consequence, exhibit excellent ability to
capture organic and inorganic anions. The large surface and high The pH value was monitored with a Sartorius pH meter (Model
anion exchange capacity make LDHs as promising materials for PB-10, Germany) using a combined glass electrode. The DO (dis-
phosphorus uptake from wastewater (Novillo et al., 2014; Sun et al., solved oxygen) value was measured by a DO meter (HQ30d, HACH,
2014; Yan et al., 2015). However, LDHs have been normally utilized USA). Laboratory analysis was performed on the water samples for
as dispersed powder form for the adsorption of phosphorus. total phosphorous (TP); total dissolved phosphorous (TDP) and
Herein, it presented appreciable problems for the application in soluble reactive phosphorous (SRP). All the parameters mentioned
CRIS due to the small size, low specific density and subsequently above were determined, based on standard methods (APHA, 2005).
solid/liquid separation process.
As a continuation of our previous work (Zhang et al., 2013, 2014, 2.4. Characterization of the system
2015), nine kinds of different LDHs were synthesized by co-
precipitation method and in-situ coated on the natural anthracite, A simulated lab-scale constructed rapid infiltration system was
zeolites and bio-ceramic, respectively. The results indicated that the constructed and operated indoor, as part of a larger experiment
Zn-LDHs coated on the natural bio-ceramic surface were efficiently containing 2 units (one in parallel). Each unit contains 4 test col-
removed the phosphorus. Therefore, a series of natural bio-ceramic umns (PVC, inner diameter 8 cm and height 25 cm). The first three
substrates have been utilized with various LDHs to synthesized and fourth test columns contained 20 cm thick bio-ceramic/Zn-
core-shell substrate/LDHs composites and LDHs were synthesized LDHs (FeZn-LDHs, CoZn-LDHs and AlZn-LDHs) composites and
through co-precipitation method. Subsequently, column tests were natural bio-ceramic substrates, respectively. The average diameter
used to study the removal performance of nitrogen and phosphorous. of the natural bio-ceramic were 4~8 mm. The true and bulk density
Herein, the objectives of this study are: 1) to prepare bio- was 2.034 and 1.028 g/cm3. The Brunauer Emmett Teller (BET)
ceramic/Zn-LDHs (FeZn-LDHs, CoZn-LDHs, AlZn-LDHs) compos- surface area was 0.57 m2/g. The average voidage of the natural bio-
ites and to characterize the composites by SEM and XRFS in details; ceramic in the column tests was 30.96%.
2) to investigate the efficient decontamination of phosphorus from The CRIS unit was fed with municipal wastewater from the top
municipal wastewater by the column tests; 3) to study the of the columns and flowed out from the bottom. All the experi-
adsorption and desorption behavior of phosphate; 4) to propose ments were carried out in duplicate and continuous operated for 6
the mechanisms for the removal of phosphorus. months. The hydraulic retention time (HRT) was 24 h and the hy-
draulic loading rate (HLR) was 70 L/(m2$d), respectively. The raw
2. Material and methods water was collected after the thin grill of Gongda-Road Lifting
Pump Station and the characteristics were shown in Table 1. The
2.1. Preparation of bio-ceramic/Zn-LDHs experiments started at March and ended at September 2015 (last
about 28 weeks).
Core-shell bio-ceramic/Zn-LDHs (FeZn, CoZn, AlZn-LDHs) com-
posites were synthesized via conventional co-precipitation method 2.5. Adsorption studies
under ambient atmosphere. In a typical synthetic run, such as bio-
ceramic/FeZn-LDHs composites, aqueous solution of FeCl3 (0.75 M) For the adsorption of phosphate, stock solution were prepared
and ZnCl2 (1.50 M) were simultaneously added into a beaker which from KH2PO4 (Guarantee reagent, GR). The adsorption experiments
contain the clean natural bio-ceramic at temperature of 80 C. were executed in a series of 250 mL conical flasks by taking 100 mL
Simultaneously, appropriate volume of 10%wt NaOH was added at a KH2PO4 solutions at the initial phosphate concentration from 0 to
controlled rate to maintain the pH at 11e12 under continuous 512 mg P/L with the substrates dosage of 10g. The mixtures were
stirring for 4 h. After complete precipitation, the mixed compounds shaken at a speed of 120 rpm thermostatic shaker at 25 ± 1 C for
were centrifuged at 1000e1500r/min for 10min and washed thor- 24 h. The substrates were separated from the mixture by the
oughly with de-ionized water until the pH of effluent solution was filtration at once after shaking. The residual concentration of
neutral. Finally, the washed samples were oven-drying for 16 h to 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.
ðCo Ce ÞV where C0 and Ct are the initial and given time t concentration of
qe ¼ (2) phosphate in solution (mg/L), qt is the amount of phosphate
m
adsorbed at time t (mg/g), m is the adsorbent dry weight (g), and V
where C0 and Ce are the initial and equilibrium concentration of is the suspension volume (L).
phosphate in solution (mg/L), qe is the equilibrium adsorption ca- In order to clarify the adsorption kinetics of phosphate onto bio-
pacity (mg/g), m is the adsorbent dry weight (g), and V is the ceramic/Zn-LDHs and natural bio-ceramic, the pseudo-first-order
suspension volume (L). (Eq. (6)) and pseudo-second-order (Eq. (7)) were used. The liner
Based on the above optimized condition, the adsorption iso- form of the models equation could be expressed as follows:
therms of the bio-ceramic/Zn-LDHs and natural bio-ceramic sub-
strates to phosphate were studied at room temperature. The k1
lgðqe qt Þ ¼ lgqe t (6)
commonly model of Langmuir (Eq. (3)) and Freundlich (Eq. (4)) 2:303
were used to describe the adsorption equilibrium:
t 1 t
¼ þ (7)
Ce 1 Ce qt k2 q2e qe
¼ þ (3)
qe KL qm qm
where qe and qt (mg/g) are the amount of phosphate adsorbed on
1 the adsorbents at equilibrium and at time t (h), respectively. k1
lgqe ¼ lgKf þ lgCe (4) (h1) and k2 (g/(mg$h)) are the constant of the pseudo-first-order
n
and pseudo-second-order model.
where Ce is the equilibrium concentration of phosphate in solution
(mg/L), qe is the amount of phosphate adsorbed at equilibrium (mg/ 2.8. Statistic
kg), qm is the theoretical maximum sorption capacity (mg/kg), KL is
Langmuir adsorption equilibrium constant which is related to the Statistical analyses were determined using Excel 2007 and SPSS
energy of adsorption (L/mg), Kf is Freundlich adsorption equilib- 19.0 software (SPSS Inc., Chicago, IL, USA), including analysis of
rium constant, n is a constant which represent the Freundlich variance (ANOVA), Levene's test for homogeneity of variance and
isotherm curvature. normality, and Post Hoc Multiple Comparisons were also per-
formed to test which bio-ceramic/Zn-LDHs was a statistically sig-
2.6. Desorption studies nificant difference, and used the Bonferroni test. In all tests,
differences were considered statistically significant when p < 0.05
Desorption experiments were carried out in different salt/alkali (SPSS, 2003).
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- 3. Results and discussions
dium for desorption.
The desorbed capability of the adsorbent was assessed as fol- 3.1. Characteristics of bio-ceramic/Zn-LDHs
lows: phosphate loaded adsorbent was separated from the mixture
by filtration after adsorption studies and washed in de-ionized In order to better understand the coated behavior of the bio-
water for three times to remove any un-adsorbed phosphate. The ceramic/Zn-LDHs, a fine characterization of the textural proper-
phosphate loaded adsorbent was mixed with 100 mL different ties and chemical compositions of bio-ceramic/Zn-LDHs were
kinds of salt/alkali solution and kept in a shaker for 24 h. The conducted.
desorbed phosphate was analyzed by standard methods as SEM images in Fig. 1 showed the shape and size of bio-ceramic/
mentioned previously. All the experiments were carried out in Zn-LDHs composites and natural bio-ceramic substrates. It was
duplicate. observed that the morphologies of bio-ceramic/FeZn-LDHs aggre-
gated mass of blossom with small and flaky debris (Fig. 1a), while
2.7. Kinetic studies flaky platelets were mainly observed in bio-ceramic/CoZn-LDHs
and bio-ceramic/AlZn-LDHs materials (Fig. 1b, c). In addition, the
Phosphate adsorption kinetic study was carried out by adding surface of bio-ceramic/FeZn-LDHs was rough and had lots of folds,
10g adsorbents to a 250 ml flask filled with 100 ml phosphate so- which was quite different with bio-ceramic/(CoZn, AlZn)-LDHs
lution at the initial concentration of 4 mg/L. The flask was placed in with a mass of flaky platelets aggregated on the surface. Compar-
a thermostatic shaker at 25 ± 1 C and shaken at 120 rpm. The atively, the surface of the natural bio-ceramic was smoother than
subsequent procedure was similar to that for phosphate adsorp- the bio-ceramic/Zn-LDHs and showed no flaky crystal material.
tion. All the experiments were carried out in duplicate. The amount Those results can be remarked that bio-ceramic/Zn-LDHs formed a
of phosphate adsorbed on the adsorbents at time was calculated by highly porous network and responsible for the high surface of LDHs
the following equation (Eq. (5)): 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 study period was shown in Fig. 2. The TP average removal rates of
for layered double hydroxides. the bio-ceramic/Zn-LDHs (FeZn-LDHs, CoZn-LDHs and AlZn-LDHs)
Chemical analysis (XRFS) results of bio-ceramic/Zn-LDHs and were 71.58% (63.25%e79.41%), 64.92% (57.69%e79.41%) and
natural bio-ceramic were shown in Table 2. The main composites of 69.12% (57.10%e82.35%), respectively, and the removal rates of the
the natural bio-ceramic were silica (70.620%) and alumina bio-ceramic/FeZn-LDHs showed the relative stability among them.
(17.551%). After the process of LDHs coated on the natural bio- Comparatively, the removal rates of the natural bio-ceramic were
ceramic, additional element of Zn was emerged. The content of 39.38% (26.28%e55.88%) (Fig. 2a, d).
the Zn in bio-ceramic/Zn-LDHs were 0.086% for FeZn-LDHs, 0.071% The average removal rates of TDP observed in bio-ceramic/Zn-
for CoZn-LDHs and 0.058% for AlZn-LDHs. In particular, compared LDHs which were FeZn-LDHs, 74.91% (61.17%e91.42%); CoZn-
with the content of Fe and Co in natural bio-ceramic (5.507% and LDHs, 68.85% (55.83%e87.31%) and AlZn-LDHs, 72.23% (54.37%e
0.011%), the natural bio-ceramic coated by FeZn-LDHs and CoZn- 89.18%). However, the average removal rates in natural bio-ceramic
LDHs were also increased 0.021% for Fe and 0.027% for Co, units were 33.58% (15.70%e64.18%) (Fig. 2b, d).
respectively. Furthermore, based on SEM and XRFS results, the The SRP average removal rates were 82.31% (71.57%e98.29%),
coated method changed the natural bio-ceramic physical charac- 73.88% (62.75%e94.44%) and 77.67% (62.84%e96.15%) for bio-
teristics and chemical compositions, which confirmed the way of ceramic/Zn-LDHs (FeZn-LDHs, CoZn-LDHs, AlZn-LDHs), respec-
LDHs coated on the surface of the natural bio-ceramic to synthesize tively, while in natural bio-ceramic unit it reached 33.25% (18.63%e
bio-ceramic/Zn-LDHs were feasible. 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-
3.2. Purification experiments
ceramic/FeZn-LDHs obtained the best purification performance
among the bio-ceramic/Zn-LDHs and natural bio-ceramic.
The TP, TDP and SRP removal performance in CRIS during this
Table 2
Main chemical compositions of bio-ceramic/Zn-LDHs and natural bio-ceramic/%.
Compositions Substrates
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
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
90 80
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- was 38.40%. The less efficient in TP removal may attribute to the
ceramic, it improved the removal rates of 32.20%, 41.33% and large average diameter (8e12 mm) and high hydraulic conductivity
49.06%, respectively. As for the bio-ceramic/AlZn-LDHs, it out- (88.8 mm/s) of the bio-ceramic.
performed the natural bio-ceramic 29.74% (TP), 38.65% (TDP) and The total phosphorus divided in two parts which were total
44.42% (SRP), respectively. Moreover, the bio-ceramic/CoZn-LDHs dissolved phosphorus and particulate phosphorus. In this regard,
enhanced the removal rates of 25.54% for TP, 35.27% for TDP and the total dissolved phosphorus removal depends on the substrates
40.63% for SRP. physical and chemical adsorption, chemical and microbial reaction.
Meanwhile, the bio-ceramic/Zn-LDHs efficiently enhanced the Compared the main chemical compositions with the natural bio-
SRP removal rates (49.06%, 40.63% and 44.42%) compared to the TP ceramic, the coated method increased the content of the metal
(32.20%, 25.54% and 29.74%) and TDP (41.33%, 35.27% and 38.65%) ion such as Fe and Zn, which encouraged the conversion of dis-
removal performance. In other words, coating Zn-LDHs on the solved phosphorus into particulate phosphorus. Accordingly, the
natural bio-ceramic efficiently improved the SRP removal rates coated method under alkaline conditions caused the surface of the
which in turns to improve the removal rates of TP and TDP. natural bio-ceramic more porous. Most likely, the presence of LDHs
Accordingly, ANOVA analysis of the TP, TDP and SRP removal rates provided a more effective distribution of the porous structure,
showed statistically significant difference (P < 0.05) between the which encouraged the more effective adsorption on the bio-
bio-ceramic/Zn-LDHs and natural bio-ceramic (Post Hoc Test, ceramic surface. Furthermore, the coated method may delay the
Table 3), indicating the coated method to synthesize bio-ceramic/ wastewater passage through the system which in turn to increase
Zn-LDHs composites were meaningful in TP, TDP and SRP removal the hydraulic retention time and as a consequence, increase the
in the study period in our CRIS. removal efficiencies.
The major processes responsible for TP removal in CRIS attrib- The key mechanisms of phosphate removal from wastewater in
uted to substrates interception, adsorption, chemical precipitation CRIS are physic-chemical process. In addition, the adsorption of
and assimilation into microbial (Xu et al., 2013). Compared the phosphate via electrostatic attraction, ion exchange and ligand
average removal rates with the natural bio-ceramic, the higher exchange (Chitrakar et al., 2006). The outperformed removal rates
efficiencies of the bio-ceramic/Zn-LDHs should be attributed to the of the bio-ceramic/Zn-LDHs may attribute to the process of fixation
properties of the LDHs, namely the hierarchical porous structures, of phosphate by Fe in the substrates were enhanced (Netter, 1992),
large surface area and high anion exchange capacity. The average indicating that process of coating LDHs on the natural bio-ceramic
removal rates of TP observed in natural bio-ceramic were better improved the role of chemical reaction. Besides, it may associate
than Liu et al. (2009) who reported the maximum removal of TP 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.
Lower Upper
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 of the bio-ceramic/Zn-LDHs and the natural bio-ceramic were
obtained the different purification effects. Due to the three 67.58% (FeZn-LDHs), 62.40% (CoZn-LDHs), 63.14% (AlZn-LDHs) and
different LDHs had the same interlayer anions, the varied removal 57.08%, respectively (Fig. 3a). Those results indicated that the way
capacity of LDHs was related to the nature of the precursor metals, of coating LDHs on the surface of natural bio-ceramic to synthesize
the layer charge density and the orientation of the guest (Goh et al., the new composites of bio-ceramic/Zn-LDHs actually increased the
2008). The ionic radius of the precursor metal ions was found to removal performance of PP.
reduce in the order of: Fe3þ>Co3þ>Al3þ (Cavani et al., 1991). As the The removal of PP mainly depends on the physical interception
metal ratio of M3þ/M2þ (Zn2þ) decrease, the distance of the cations mechanism and the particulate phosphorus could be intercepted
was also decrease. Therefore, the basal space of layered framework and trapped in the test columns. Therefore, in order to measure the
was increase. In this regard, the positive charge density of the actual values of physical interception of PP in the whole purification
layered framework was increased which in turn to enhance the experiments, all the substrates in the test columns would be took
ionic interactions and as a consequence, exhibited the higher ability out respectively and then be washed thoroughly with de-ionized
to capture the oxyanions, such as phosphate (Sahoo et al., 2014). water for three times after the whole purification experiments
However, the average removal rates of phosphate by the three bio- ended. On the other hand, the purification amount (PP) could be
ceramic/Zn-LDHs in this study showed the different results. Un- calculated by the difference of numerical values between the
usual removal behavior is probably controlled by some physical removal rates of TP and TDP, and the results were illustrated in
properties of LDHs such as crystallinity, particle size and surface Fig. 3b. The results indicated that the purification amount (PP) for
area (Halajnia et al., 2012), even the amount of the co-intercalated the bio-ceramic/Zn-LDHs and natural bio-ceramic were 1.595 mg
water molecule (Costa et al., 2012). Furthermore, those results were for FeZn-LDHs, 1.402 mg for CoZn-LDHs, 1.567 mg for AlZn-LDHs
in agreement with Boclair and Braterman (1999) who reported the and 1.386 mg for natural bio-ceramic, respectively. Compara-
corresponding FeZn-LDHs was more stable than the AlZn-LDHs. 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-
3.3. Particulate phosphorus removal ceramic. Those results confirmed that the main removal mecha-
nism of PP for the bio-ceramic/CoZn-LDHs and the natural bio-
The average removal efficiencies of particulate phosphorus (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 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
Adsorption capacity at different aqueous equilibrium concen- volume ratio of 1:1, 3:1, 4:1 and 9:1 were 12.92, 12.48, 12.36 and
trations could be illustrated by the adsorption isotherm (Fig. 4) and 11.97, respectively. The desorption rates were less efficient in 0.1 M
the linear plots of Langmuir and Freundlich models were showed in NaOH, 5 M NaCl and the mixed solution of 5 M NaCl and 0.1 M
Fig. 5. The fitted constants and regression coefficients were sum- NaOH with the volume ratio of 4:1 and 9:1. The effective desorption
marized in Table 4. It can be observed that Freundlich equation gave of phosphate (50.07%) could be achieved in the volume ratio of 1:1
more satisfactory fitting to the adsorption isotherms of phosphate (5 M NaCl þ 0.1 M NaOH), with negligible dissolution of adsorbent
on bio-ceramic/Zn-LDHs and natural bio-ceramic with correlation (Fig. 6a). These results coincided with those found by Chitrakar
coefficient R2 higher than 0.95, indicating the adsorptions on het- et al. (2006) who assessed the maximum desorption rates of
erogeneous surface sites (Radrigues and Pinto da Silva, 2010). MgMn-LDHs could be achieved by using a mixed solution of 5 M
However, the results of the bio-ceramic/CoZn-LDHs indicated that NaCl þ 0.1 M NaOH. Herein, 5 M NaCl þ 0.1 M NaOH (1:1) were used
the Langmuir equation (R2 ¼ 0.9910) fitted better than the for the subsequent desorption experiments.
Freundlich equation (R2 ¼ 0.9726), suggesting that the adsorption The desorption results of the bio-ceramic/Zn-LDHs and natural
of phosphate attributed to monolayer adsorption on the outer bio-ceramic were presented in Fig. 6b. The amount of desorbed
surface of adsorbent (Zhou et al., 2011). phosphate from the adsorbent showed an increasing trend with the
In Freundlich equation, the constant of 1/n for the bio-ceramic/ increasing initial concentration phosphate on the adsorbent.
Zn-LDHs and natural bio-ceramic in this study were 0.4363, 0.4470, However, the desorption rates of the bio-ceramic/Zn-LDHs and
0.4122 and 0.7024, respectively (Fig. 5b), which were in the range of natural bio-ceramic were not obtained in the same initial concen-
0e1, demonstrating a good adsorption of phosphate (Cheng et al., tration. Therefore, the simulated curves were used to evaluate the
2010). The higher Kf values indicated the adsorption capacity of desorption capacity of the bio-ceramic/Zn-LDHs and natural bio-
bio-ceramic/Zn-LDHs were higher than the natural bio-ceramic ceramic. At the initial concentration of 4 mg/L on the adsorbent,
(Table 4). On the basis of values of qm in Langmuir equation, the the desorption rates of the bio-ceramic/Zn-LDHs were 35.09% for
theoretical phosphate adsorption capacity of the bio-ceramic/Zn- FeZn-LDHs, 23.73% for CoZn-LDHs, 28.80% for AlZn-LDHs and
LDHs and natural bio-ceramic followed the order of: AlZn-LDHs 16.14% for the natural bio-ceramic. Accordingly, the desorption
(769.23 mg/kg)>FeZn-LDHs (476.19 mg/kg)> Natural bio-ceramic rates of coated and natural substrates were followed the order of
(434.78 mg/kg)>CoZn-LDHs (357.14 mg/kg) (Table 4). The FeZn-LDHs > AlZn-LDHs > CoZn-LDHs > Natural bio-ceramic.
sequence of AlZn-LDHs > FeZn-LDHs in theoretical adsorption 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.
0.6 FeZn-LDHs
qe (mg/g)
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)
lgqe
0.5 1.47
0.0 0.98
AlZn-LDHs
AlZn-LDHs
Ce/qe(kg/L)
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.
(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
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).
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
0.5 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
0.35
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)
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.