Lu 2020
Lu 2020
Lu 2020
a r t i c l e i n f o a b s t r a c t
Article history: Extensively used antibiotic ciprofloxacin increases in water environment, which is not efficiently
Received 10 August 2019 removed by means of conventional adsorbents. MCM-41 has great potential for ciprofloxacin removal.
Received in revised form There has been little research on ciprofloxacin removal by MCM-41 and their corresponding adsorption/
12 March 2020
desorption behaviors. In this study, amorphous SiO2, MCM-41 and diamine-functionalized MCM-41 were
Accepted 10 April 2020
Available online 16 April 2020
used as adsorbents to remove ciprofloxacin. In the adsorption process, their adsorption capacity followed
the order of SiO2≪MCM-41< diamine-functionalized MCM-41. Ciprofloxacin adsorption behaviors were
Handling Editor: Prof. S Alwi revealed through the analysis of adsorption isotherms and kinetics model: i. Adsorption process for
MCM-41/diamine-functionalized MCM-41 mainly included three steps; and ii. Diamine-functionalized
Keywords: MCM-41 had higher adsorption capacity because of its more and stronger chemical-monolayer
MCM-41 adsorption. In the desorption process, strong acid facilitated desorption by inhibiting the interaction
Diamine-functionalized MCM-41 between ciprofloxacin and diamine-functionalized MCM-41, and the mechanism was proposed and then
Adsorption supported by salt elution experiment. Last, the breakthrough bed volume and capacity of these absor-
Desorption
bents were tested in column operation. Diamine-functionalized MCM-41 had better performance with
Ciprofloxacin
550 bed volume and 18.3 mg/g adsorption capacity in the treatment of 5.0 mg/L ciprofloxacin. Both batch
Water treatment
and column results demonstrated that diamine-functionalized MCM-41 was highly efficient for cipro-
floxacin removal. This study indicates great application prospect of mesoporous silica materials for
ciprofloxacin removal and reveals their adsorption/desorption behaviors, which could provide guidance
on adsorbent modification/regeneration for antibiotics removal.
© 2020 Elsevier Ltd. All rights reserved.
https://doi.org/10.1016/j.jclepro.2020.121644
0959-6526/© 2020 Elsevier Ltd. All rights reserved.
2 D. Lu et al. / Journal of Cleaner Production 264 (2020) 121644
present water plants showed that most drugs could be fully elim- grade. Cetyltrimethyl ammonium bromide (CTAB, 99%), 3- (2-
inated through the ozonation and two-step granular activated aminoethyl aminopropyl) trimethoxysiliane (ATMS, 97%) from
carbon filtration. However, the high-use fluoroquinolone antibiotic Sigma-Aldrich, and tetraethoxysilane (TEOS, 99%) from Shanghai
ciprofloxacin (CIP) was found to directly pass through the whole Chemical were used for experimental studies without further pu-
treatment process with extremely low adsorption (Vieno et al., rification. CIP (99% purity) was purchased from Wuhan Yuancheng
2007; Morasch et al., 2010). It is imperative to explore an efficient technology development company limited (China). An amorphous
absorbent for the removal of CIP fluoroquinolone antibiotic from SiO2 was supplied by Sinopharm Chemical (Shanghai, China). Other
drinking water and wastewater. reagents were commercially available and used as received.
As an absorbent, the ordered mesoporous silica material, so-
called mobile composition of matter (MCM-41), with 2D hexago- 2.2. Synthesis and characterization methods
nal p6m structure, has multiple advantages including a large sur-
face area, uniform pore distribution (about 2.5 nm) and ordered The preparation process of the material consists of two parts:
mesoporous channel structure. Its mesoporous surface has preparation of the parent MCM-41 with spherical morphology and
amounts of hydroxyl functional groups (SieOeH) which could be diamine-functionalized MCM-41 via the post-synthesis method
modified with different functional groups through OeH grafting (Liu et al., 2008; Pauwels et al., 2001). Briefly, 2.5 g of CTAB was fully
and substantially expands its application. For instance, the reported dissolved in deionized (DI) water, followed by adding 13.2 g of an
mesoporous silica-based materials modified with amine and thiol ammonia solution (32 wt %) and then mixing with 60 g of ethanol.
functional groups were proved particularly efficient for seques- TEOS (4.7 g) was added after the mixture solution was stirred
tering heavy metals from water (Mercier and Pinnavaia, 1997). vigorously for 15 min. The obtained synthetic gel was further stir-
Moreover, unmodified and modified mesoporous materials using red at room temperature for an additional 2 h. The solid product
alkyl groups had achieved a good performance on adsorption of was obtained by filtration and repeated washing (DI water, 3 times),
organic pollutants (Baccile and Babonneau, 2008; Wang and Tan, and drying at 80 C for 12 h. The surfactants in the as-synthesized
2007). The mesoporous MCM-41 can be used as matrix to design sample (1.0 g) were removed by extraction in a mixture of
an implantable delivery system for drug delivery (Vallet-Regi et al., concentrated hydrochloric acid (1.0 mL) and anhydrous ethanol
2001; Munoz et al., 2003). It has also been applied for catalytic (100 mL) for 12 h the process is repeated again for the completely
cracking because of its sufficient acid sites and high cracking ac- removal of CTAB. The product was filtered, washed several times
tivity. Up to now, the MCM-41 has been mainly applied to the with anhydrous ethanol, and dried at 60 C. The N-MCM-41 was
adsorptive elimination of heavy metals and common organic pol- prepared as followed: MCM-41 (1.0 g) with the free surfactant was
lutants (Walcarius and Mercier, 2010; Arica et al., 2017), medical dispersed in prepared anhydrous toluene solution containing 0.3 g
delivery and catalytic cracking (Manzano and Vallet-Regi, 2010; ATMS, followed by a reflux for 6 h. The resultant materials were
Roik et al., 2017). Considering the special physical and chemical then obtained by filtration, solvent-washing anhydrous toluene,
properties of MCM-41, it would be a promising adsorbent for CIP and drying process for solvent removal.
removal. However, the adsorptive removal of CIP antibiotic using The samples were characterized by a complementary combi-
MCM-41 and the corresponding adsorption-desorption behaviors nation of X-ray diffraction (XRD), N2 adsorption-desorption, scan-
between antibiotics and mesoporous silica materials were barely ning electron microscopy (SEM), FT-IR spectroscopy and surface
investigated in depth, which is significantly critical to select, zeta potential measurement. Powder X-ray diffraction patterns
modifying and regenerating absorbent to efficiently remove CIP were recorded using a Bruker diffractometer with data collected at
fluoroquinolone antibiotic from water. 2q from 1.5 to 8 . Nitrogen adsorption/desorption experiments
In this study, the MCM-41 with perfect sphericity and mono- were performed at 77 K to examine the specific surface area and the
dispersity was prepared, which is beneficial to on-going research pore size distributions. The surface zeta potential was measured
on adsorption/desorption study, especially on column operation. using a Particle Size & Zeta Potential Analyzer. The IR analysis of the
Amorphous SiO2, unfunctionalized MCM-41 (both as references), samples prepared with KBr pellets were performed on a Bruker FT-
and diamine-functionalized spherical MCM-41 (N-MCM-41) were IR spectrometer in the range of 400e4000 cm1. The morphology
used as adsorbent to remove CIP from water. Their performances and structure of the materials were investigated by SEM observa-
were investigated in batch and column operations. In batch ex- tion (FFESEM, JSM-6700). For further analysis, the samples were
periments, CIP adsorption behaviors on these adsorbents were also examined by transmission electron microscopy (TEM, JEM-
proposed through the fitting analysis of the adsorption isotherms 2010).
and adsorption kinetics models (Ho et al., 2000; Song et al., 2018).
The adsorption of CIP on N-MCM-41 at different pH was investi-
2.3. Adsorption experiments
gated to determine the optimized desorption condition, and
possible desorption behaviors were proposed and then verified by
A stock solution of CIP (1000 mg/L) was prepared using deion-
salt elution experiment. A continuous fixed-bed column packed
ized water. CIP concentrations from 10 to 300 mg/L in our study
with these absorbents was then applied for CIP removal, and their
were prepared by diluting the stock solution. In the batch pro-
breakthrough bed volume and breakthrough capacity were inves-
cedure, 20 mg adsorbent material was added to a 20 mL CIP solu-
tigated in detail. This study highlights the potential of mesoporous
tion with the different concentrations. The suspension was stirred
MCM-41 for antibiotic CIP removal and proposes the adsorption-
for 2 h using a water shaker bath at 25 C. Then the adsorbents were
desorption behaviors between CIP and mesoporous silica mate-
filtrated through a syringe filter (0.22 mm) and the CIP concentra-
rials. The results should provide basic knowledge on how to select/
tion was measured by UVeVis spectrometer at 277 nm. The CIP
modify/regenerate adsorbent for antibiotic removal from water.
amount adsorbed at equilibrium was calculated using Eq. (1):
equilibrium time, M (g) is the mass of the added adsorbent. the mesoporous channel surface of MCM-41 led to the decrease of
The initial CIP concentration were prepared in the range of the peak intensity at 2.2 and the less resolved peaks of the higher
10e300 mg/L. In addition, for the experiments of adsorption ki- odder (110) and (200) diffractions, which is most likely due to the
netics, initial CIP concentration was fixed to be 200 mg/L and the overlap of diffraction peaks with higher refractive indices caused by
volume of solution was enlarged to 150 mL. At time intervals of 0.5, the decreased structural ordering. The phenomenon could be
1.0, 2.0, 5.0, 8.0, 10.0, 15.0, 20.0, 25.0, 30.0, 40.0, 50.0, 60.0, 90.0 and explained by that ATMS would be successfully grafted on the
120.0 min, 0.5 mL of sample was taken each time for analysis of mesoporous channel surface and reduced electron contrast in the
residual CIP concentration. electron density. The nitrogen adsorption/desorption isotherms
were type IV and exhibited a sharp step increase at P/P0 ¼ 0.3e0.4
2.4. Desorption experiment (Fig. S2 Left), indicating their uniform-size mesoporous structure.
The sample exhibited a less sharp step when ATMS was introduced,
The influence of pH on the interaction between CIP and N-MCM- and the pore size of N-MCM-41 was slightly smaller than that of
41 was firstly investigated to establish the optimal desorption MCM-41 shown in Fig. S2 right. Table S1 listed the calculated spe-
condition as the following procedure: 20 mg samples and 20 mL cific surface area, pore size and pore volume. As expected for N-
solution with different initial CIP concentration and pH pre- MCM-41, its surface area and pore sizes decreased with the addition
adjusted using HCl or NaOH were prepared at 25 C for 2 h. The of ATMS, which was ascribed to the occupancy of structural chan-
residual CIP concentration was measured by UVeVis spectrometer nels by amino functional groups (Liu et al., 2009).
at 277 nm. The SEM image of MCM-41 and N-MCM-41 showed perfect
Desorption experiments were conducted immediately after sphericity and monodispersity in Fig. S3A and Fig. S3B. TEM image
absorption experiments. Deionized water, pH 3.0 acidic solutions confirmed that the N-MCM-41 material still maintained well or-
pre-adjusted using HCl and 3.0 mol/L NaCl solution were employed dered, one-dimensional pore structure (Fig. S3C), which suggested
as eluents. The desorption procedure was as followed: 15 mL of that the introducing ATMS into the mesoporous channel surface of
supernatant was removed after centrifuge, and the same volume of MCM-41 do not lead to the decrease of mesoporous ordering and
eluent solution was added and then shaken for an additional 24 h. destruction of spherical morphology of product. The spherical
The above procedure was repeated, and the CIP concentrations of morphology of MCM-41 with well monodispersity was beneficial
the first and second elution solutions were determined. The release for adsorbent, especially as adsorbent on adsorption/desorption
ratio was calculated according to Eq. (2): study in column adsorption/desorption research. The spherical
geometric shape facilitated a homogeneous column packing and
15C1 þ 15C2 effectively reduced the operating pressure and pressure drop. The
Release ratio ð%Þ ¼ 100% (2)
20qe particle sizes of MCM-41 and N-MCM-41 were about 2.0 mm from
the SEM images, which had better performance and lower oper-
where C1 (mg/g) and C2 (mg/g) are the CIP concentration in first and ating pressure than the common mesoporous nanoparticles for
second elution solution; qe (mg/g) is the equilibrium adsorption removal of pharmaceuticals (Walcarius and Mercier, 2010).
amount. In order to explore the surface character of mesoporous parti-
cles, the relationship between adsorbent and ciprofloxacin, the FT-
2.5. Column experiments IR spectra of materials were detected and shown in Fig. 1. FT-IR
spectra of the as-MCM-41, MCM-41 and N-MCM-41 showed the
A cylindrically shaped microcolumn (12 mm length and 8.8 mm typical characteristic peaks of silica at ~808 cm1, 978 cm1 and
i.d.) was used. SiO2 and the unmodified and functionalized MCM-41 1088 cm1, corresponding to the vibrations of the SieOeSi sym-
adsorbents of 300 mg were prepared in suspension, and then metric bond, SieOH stretching bond and SieOeSi asymmetric
packed into the microcolumn. The CIP solution at a concentration of bond. In the MCM-41 spectrum, the absorption peaks of 2900 cm1,
5.0 mg/L was fed on the gravitation. The breakthrough point was 2858 cm1, and 1480 cm1 in the as-MCM-41 material disappeared,
defined as the bed volume at which the normalized CIP concen- indicating that the surfactant CTAB was removed clearly (Liu et al.,
tration exceeded a predefined value (i.e. 10 %.). The normalized CIP 2005). Comparing the to that of MCM-41, absorption peaks of N-
concentration was calculated according to Eq. (3). The break- MCM-41 were observed at 2980 cm1, 1480 cm1 and 965 cm1,
through capacity was the corresponding adsorption capacity at the corresponding to the grafted amino groups in the material.
breakthrough point. Compared FT-IR spectrum of CIP with MCM-41-CIP, the
stretching vibration of OeH band at 3446 cm1 and the stretching
Normalized CIP concentration (%) ¼ Cout/Ci ✕100% (3) vibration peak of C]O at 1627 cm1 originated from CIP still pre-
sented, indicating that CIP was adsorbed to the mesoporous
where Cout and Ci are the CIP concentration at the outlet of the channel of MCM-41. In the N-MCM-41-CIP spectrum, the vibration
column and the initial feed CIP concentration. peaks at 3445 cm1, 1680 cm1 and 1629 cm1 were normally be
observed, probably due to the OeH vibration and C]O stretching
3. Results and discussion vibration of amide bond formed by N-MCM-41 and CIP.
According to the zeta potential data (Fig. S4), MCM-41 showed
3.1. Materials characterization negatively charged from pH 3.0 to pH 11.0, which was consistent
with the results of earlier researcher (Kecht et al., 2008), because of
Fig. S1 gave the XRD analysis of the diffraction peaks for MCM- the SieOH groups presentation on MCM-41 surface. When the
41 and N-MCM-41. Three diffraction peaks at 2.2 , 3.8 and 4.4 , ATMS was grafted on mesoporous channel surface of MCM-41, the
corresponded to (100), (110) and (200) diffractions could be incorporated aminoethyl and aminopropyl moieties were proton-
observed, which indicated that hexagonal mesostructure phase ated at acidic pH values and carried a positive charge. A higher
existed in MCM-41. For N-MCM-41, the intensive (100) accompa- density of such groups on the mesoporous channel surface leads to
nied by weak peaks at (110) and (200) was still obvious and increased zeta potential curves as they affect the overall surface
demonstrated an ordered hexagonal structure as shown in the charge. As result, the zeta potential for N-MCM-41 shifted toward
spectra of MCM-41. compared to MCM-41, introducing ATMS into positively charged below pH 9.0 (Kecht et al., 2008; Rosenholm
4 D. Lu et al. / Journal of Cleaner Production 264 (2020) 121644
40
0
0 50 100 150 200 250
Equilibrium Ciprofloxacin Concentration, Ce (mg/L)
Table 1
Parameters of the isotherms models for CIP adsorptionby the adsorbentsa.
Amorphous SiO2 19.71 0.01 0.11 0.16 0.97 0.92 0.22 0.03 1.00 0.84
MCM-41 69.70 0.02 0.91 2.54 0.65 0.99 1.60 0.10 0.69 0.97
N-MCM-41 139.25 0.02 0.99 6.15 0.59 0.94 2.73 0.02 1.00 0.98
a
Data were obtained from Fig. 2.
120
MCM-41(Experiment) N-MCM-41(Experiment) MCM-41 N-MCM-41
140
MCM-41(First) N-MCM-41(First)
Ciprofloxacin Adsorbed, qe (mg/g)
100 80
80
60
60
40
40
20 First stage
20
A B Second stage
Third stage
0 0
0 20 40 60 80 100 120 0 2 4 1/2
6 0.5
8 10 12
Time (min) t (min )
Fig. 3. Adsorption kinetics of ciprofloxacin on MCM-41 and N-MCM-41in deionized water solution. Adsorption kinetics was fitted by four models, including the pseudo-first-order
(first), the pseudo-second-order (second), the Elovich equation (A) and intraparticle diffusion model (B).
the whole adsorption process based on the values of R2 in Table 2. equilibrium where the adsorption rate was equal to the desorption
The qe value (i.e., adsorption capacities at equilibrium) from the rate.
pseudo-second-order kinetic equation for N-MCM-41 was signifi-
cantly higher than that for MCM-41, demonstrating its higher
3.2.3. Adsorption behavior analysis
adsorption capacity. This result was in good consistence with the
Based on the analysis of the adsorption isotherm and kinetics,
results in Table 1. Both the parameter h in pseudo-second-order
the possible adsorption behaviors of CIP on amorphous SiO2, MCM-
model and the parameter a representing the initial sorption rate
41 and N-MCM-41 were proposed in Scheme 1. The CIP absorbed on
showed that the CIP adsorption rate of N-MCM-41 was higher than
the surface of SiO2, MCM-41and N-MCM-41, while it also absorbed
MCM-41 in the initial stage (Table 2).
the inner pores of mesoporous MCM-41and N-MCM-41 in mono-
Fig. 3B showed the plots for the adsorption using intraparticle
layer and multilayer. The absorption in the pores of MCM-41and N-
diffusion model. The curves showed the three-phase plot suggested
MCM-41 in Scheme 1 largely accounted for their higher adsorption
that the adsorption process included three stages. Based on the
capacity than that of SiO2. The adsorption capacity of N-MCM-41
intraparticle diffusion model and the adsorption theoretics of so-
with low surface area was significantly higher than that of high
lute on porous adsorbent (Badruzzaman et al., 2004), a possible
surface area MCM-41 (496 m2/g vs. 828 m2/g in Table S1), which
adsorption process of CIP on mesoporous MCM-41 materials was
can be explained as followed: According to Fig. S5, the protonation
proposed: i. In the first stage, CIP quickly diffused into the outer
of both carboxylic acid group and amino group of CIP took place
surface of mesoporous MCM-41 materials and absorbed onto their
simultaneously in deionized water, and the zwitterion was main
outer surface. The slop of the first-stage adsorption curve was the
dissociation form of CIP. CIP can interact with SiO2 and MCM-41
largest indicating that this stage was minor rate-controlled step; ii.
through hydrogen bonding between ¼ NHþ 2 of CIP and SiO of
In the second stage, CIP gradually migrated into the pores and
MCM-41, while CIP can interact with N-MCM-41 through conden-
subsequently reached into the pore surface, and consequently
sation interaction between R-COO- of CIP and ReNHþ 3 of N-MCM-41
absorbed on the pore internal surface. The CIP adsorption amount
in Scheme 2. ReNH2 group of N-MCM-41 is protonated more easily
slowly increased suggesting that the second pore diffusion process
into ReNHþ
3 than SiOeH group of MCM-41 into SiO under neutral
was the main rate-controlled step but not exclusive rate-controlled
condition (Leung et al., 2009), indicating more ReNHþ 3 available
step; iii. In the last stage, the absorption reached a dynamic
than SiO to interact with CIP. N-MCM-41 would adsorb more CIP
Table 2
Parameters of the kinetics models for CIP adsorption by the adsorbentsa.
MCM-41 0.38 39.92 0.93 44.04 0.01 18.90 1.00 2.66 6.32 0.92
N-MCM-41 0.88 90.41 0.83 108.40 0.003 32.77 1.00 7.18 11.79 0.99
a
Data were obtained from Fig. 3.
6 D. Lu et al. / Journal of Cleaner Production 264 (2020) 121644
Scheme 1. Adsorption behaviors of CIP on the three absorbents in monolayer and multilayer.
than MCM-41 through its more dissociated functional groups amount increased with its initial concentration increasing from
(Scheme 2). The interaction between the adsorbents and the ad- 50 mg/L to 200 mg/L at all pH values. At pH 1.0 and pH 3.0, the CIP
sorbates by the dissociated functional group mainly contributed to adsorption amount (qe) at equilibrium kept low. This phenomenon
chemical monolayer absorption. This explained why the N-MCM- may be explained by that more than 96% CIP changed into the
41 would have more monolayer adsorption (i.e., more ReNHþ 3 ) and protonated form (i.e., R2-N-Hþ 2 ) at strong acid conditions, which
SiO2 have less monolayer adsorption (i.e., less SiO) in the results of was resistant to interact with ReNHþ 3 of N-MCM-41 contributing to
Fig. 1 and Table 1, which in turn further supported our proposed CIP desorption as shown in Scheme S1. With the increase of pH
mechanism in Scheme 2. In addition, the condensation interaction value, the qe increased quickly because of more deprotonated CIP
between CIP and N-MCM-41 was stronger and more likely occurred (i.e., R-COO-) available to interact with ReNHþ 3 of N-MCM-41
compared to the hydrogen bonding between CIP and MCM-41, (Scheme 2), and reached the maximum at pH 7.0. After that, the CIP
which accounted for the higher adsorption energy (KR in Table 1) adsorption decreased despite the increase of deprotonated CIP. Two
and high initial sorption rate (h and a in Table 2) for N-MCM-41. reasons may account for this phenomenon: i). the dissociation of
amino functional groups of N-MCM-41 was inhibited under alka-
line condition (i.e., more ReNH2 of N-MCM-41 resistant to interact
3.3. Desorption analysis
with R-COO- of CIP) in Scheme S1, which can be supported by the
surface zeta potential of N-MCM from positive at pH 9.0 to negative
Desorption process is of significance for the adsorbent regen-
at pH 10.0 in Fig. S4; ii). MCM-41 material (including N-MCM-41)
eration in practical application, which is the reverse process of
cannot be resistant to strong alkaline solutions, due to increasing
adsorption. Weaker adsorption interaction between adsorbent and
deprotonation of silanol groups and hydrolysis of SieOeSi bonds
adsorbate means easier desorption process. Based on the adsorp-
which is catalyzed by mucleophilic attack by OH ions (Xiao and
tion behavior analysis, the dissociation of surface functional groups
Lasaga, 1994; Braun et al., 2016). The structure of MCM-41 may
of SiO2, MCM-41 and N-MCM-41 largely determined their adsorp-
be destroyed in the basic solution (pH value is greater than 11.0) to
tion interaction with CIP. The pH was an essential factor in deter-
some extent, leading to the decrease of its adsorption capacity. That
mining the extent of dissociation, influencing the adsorption
is one of disadvantage of MCM-41 and N-MCM-41 as adsorbents for
interaction and consequently the desorption process. The adsorp-
the removal of the antibiotics. It should be mentioned that N-MCM-
tion interaction between CIP and N-MCM-41 at different pH was
41 had the CIP removal rate great than or close to 80% in different
investigated to determine the optimized desorption condition,
CIP initial concentration solutions at optimum pH 7.0 in Fig. S6,
where the less CIP absorption indicated weaker adsorption inter-
again indicating that N-MCM-41 is an efficient adsorbent for the
action (i.e., easier desorption process).
elimination of fluoroquinolones antibiotic.
Fig. 4 showed the CIP adsorption capacity of N-MCM-41 at
Based on the above results, pH 3.0 acidic solutions could be used
different pH values from 1.0 to 11.0. In general, CIP adsorption
D. Lu et al. / Journal of Cleaner Production 264 (2020) 121644 7
Scheme 2. Interaction between CIP and three adsorbents via monolayer chemical adsorption.
180 the two adsorbents were illustrated in Fig. 5. The release ratio of CIP
pH 1.0 pH 3.0 pH 5.0 pH 6.0 for N-MCM-41/CIP in pH 3.0 solution was significantly higher than
pH 7.0 pH 9.0 pH 11.0
that in deionized water (release ratio 65.8% vs. 5.1% in Fig. 5). The
150 release ratio of CIP for MCM-41 in pH 3.0 solution was also higher
than that in deionized water (release ratio 70.6% vs. 38.6%). This
Ciprofloxacin asorbed (mg/g)
as an efficient eluent for desorption process to regenerate the N- 3.4. Column breakthrough behavior
MCM-41. Deionized water (as blank) and pH 3.0 solution were used
as eluents for MCM-41 and N-MCM-41. Desorption performances of The MCM-41 used conventional synthesis method had very fine
particle size and exhibited a low mechanical strength. Though the
8 D. Lu et al. / Journal of Cleaner Production 264 (2020) 121644
100 100
Amorphous SiO2
N-MCM-41
90 MCM-41
70
60
60
50
40
40
30
20
20
10
0
0 0 200 400 600 800
Salt solution Bed Volume
Acid solution Deionized water
Fig. 6. Column adsorption breakthrough behaviors of ciprofloxacin on three
Fig. 5. Desorption of ciprofloxacin from N-MCM-41 using different eluents.
adsorbents.
materials could perform the adsorption/desorption operation in had higher adsorption capacity and adsorption energy than SiO2/
batch measurement, it could not be packed into the column to MCM-41, probably because of its more and stronger chemical-
perform the adsorption test. This is another disadvantage of MCM- monolayer adsorption through the condensation interaction be-
41 as adsorbents for the removal of the antibiotics. MCM-41 and N- tween R-COO- of CIP and ReNHþ 3 of N-MCM-41 than that through
MCM-41 were prepared in our synthesis method and showed hydrogen bonding between ¼ NHþ
2 of CIP and SiO of SiO2/MCM-41.
almost perfect spherical morphology and monodispersity with The strong acid conditions contributed to the desorption process
particle size of ca. 2.0 mm shown in the SEM and TEM images. The through the inhibition of the interaction between CIP and MCM-41/
spherical geometric shape facilitated a homogeneous column N-MCM-41. Both batch and column adsorption results demon-
packing. Based on the results from the batch experiment, a strated that MCM-41 and N-MCM-41 were efficient for CIP elimi-
continuous fix-bed column was packed with SiO2, MCM-41 and N- nation, and N-MCM-41 had best performance for CIP removal from
MCM-41 to remove CIP from water in the aspect of the practical water. Overall, this research indicates the potential of mesoporous
application. MCM-41 materials for antibiotic CIP removal from water and re-
The bed volume of columns and the liquid flow velocity through veals the adsorption-desorption behaviors between CIP and mes-
column under gravitation were shown in Table S2. The column of oporous materials. The result will be indicative for selecting,
amorphous SiO2 particles had the minimum bed volume and high modifying and regenerating adsorbents to efficiently remove an-
compact, causing the low flow velocity. In contrast, the column of tibiotics/pharmaceuticals for water treatment.
MCM-41 and N-MCM-41 had relatively larger bed volume and
significantly higher flow velocity possibly ascribed to their porous Declaration of competing interest
structures and uniform sphericity. Fig. 6 showed the breakthrough
curves of the materials. The adsorption capacity of amorphous SiO2 The authors declare that they have no known competing
column was too low to keep the normalized CIP concentration (i.e., financial interests or personal relationships that could have
Cout/Ci) less than 10%, and the bed volume on breakthrough was appeared to influence the work reported in this paper.
only about 30. The breakthrough point for N-MCM-41 was 550 bed
volumes, two times than that for MCM-41 (i.e., 290 bed volumes).
CRediT authorship contribution statement
The breakthrough capacities of SiO2, MCM-41 and N-MCM-41
calculated from the breakthrough points were 0.13, 10.35, 18.32 mg/
Dongwei Lu: Conceptualization, Writing - original draft, Inves-
g in continuous fix-bed column (SiO2 ≪ MCM-41< N-MCM-41),
tigation, Methodology. Shu Xu: Data curation, Formal analysis,
which were well consistent with the results in batch experiments.
Investigation. Wei Qiu: Visualization, Validation. Yan Sun:
Taking the surface water from Songhua River in Harbin, China as
Conceptualization, Validation, Writing - review & editing. Xianbin
example (CIP concentration 230 ng/L), 300 mg N-MCM-41 could
Liu: Project administration, Supervision. Jingjing Yang: Writing -
roughly process about 20 m3 water, and the breakthrough bed
review & editing. Jun Ma: Funding acquisition, Supervision.
volume could get 1.0 107, which demonstrates its great applica-
tion prospect for CIP removal from real water.
Acknowledgments
Fund (No. LBH-Z17059). wastewater using lantana camara, forest waste. Int. J. Chem. Eng. 2, 1e11.
https://doi.org/10.1155/2016/5809505.
Ho, Y.S., Ng, J.C.Y., McKay, G., 2000. Kinetics of pollutant sorption by biosorbents:
Appendix A. Supplementary data Review. Separ. Purif. Methods 29, 189e232. https://doi.org/10.1081/SPM-
100100009.
Kamagate, M., Assadi, A.A., Kone, T., Giraudet, S., Coulibaly, L., Hanna, K., 2018. Use
Supplementary data to this article can be found online at of laterite as a sustainable catalyst for removal of fluoroquinolone antibiotics
https://doi.org/10.1016/j.jclepro.2020.121644. from contaminated water. Chemosphere 195, 847e853. https://doi.org/10.1016/
j.chemosphere.2017.12.165.
Kecht, J., Schlossbauer, A., Bein, T., 2008. Selective functionalization of the outer and
Nomenclature inner surfaces in mesoporous silica nanoparticles. Chem. Mater. 20, 7207e7214.
https://doi.org/10.1021/cm801484r.
MCM-41 mobile composition of matter, No. 41 Kim, S.D., Cho, J., Kim, I.S., Vanderford, B.J., Snyder, S.A., 2007. Occurrence and
removal of pharmaceuticals and endocrine disruptors in South Korean surface
N-MCM-41 diamine-functionalized spherical MCM-41 drinking and waste waters. Water Res. 41, 1013e1021. https://doi.org/10.1016/
ATMS 3-(2-aminoethyl aminopropyl) trimethoxysiliane j.watres.2006.06.034.
CIP Ciprofloxacin Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B.,
Buxton, H.T., 2002. Pharmaceuticals hormones and other organic wastewater
TEOS tetraethoxysilane
contaminants in US streams 1999-2000: a national reconnaissance. Environ.
CTAB cetyltrimethyl ammonium bromide Sci. Technol. 36, 1202e1211. https://doi.org/10.1021/es011055j.
C0 the concentration of CIP at the initial time in the batch Leung, K., Nielsen, I.M.B., Criscenti, L.J., 2009. Elucidating the bimodal AcidBase
adsorption procedure, mg/L behavior of the WaterSilica interface from first principles. J. Am. Chem. Soc.
131, 18358e18365. https://doi.org/10.1021/ja906190t.
C1 the CIP concentration in first elution solution, mg/g Liu, X., Chang, F., Xu, L., Yang, Y., Tian, P., Qu, L., Liu, Z., 2005. Preparation of ordered
C2 the CIP concentration in second elution solution, mg/g carbon/silica hybrid mesoporous materials with specific pore size expansion.
Ce the concentration of CIP at the equilibrium time in the Microporous Mesoporous Mater. 79, 269e273. https://doi.org/10.1016/
j.micromeso.2004.11.018.
batch adsorption procedure, mg/L Liu, X., Du, Y., Guo, Z., Gunasekaran, S., Ching, C.B., Chen, Y., Leong, S.S.J., Yang, Y.,
Ci the initial feed CIP concentration for column adsorption, 2009. Monodispersed MCM-41 large particles by modified pseudomorphic
mg/g transformation: direct diamine functionalization and application in protein
bioseparation. Microporous Mesoporous Mater. 122, 114e120. https://doi.org/
Cout the CIP concentration at the outlet of the column for 10.1016/j.micromeso.2009.02.023.
column adsorption, mg/g Liu, X., Sun, H., Yang, Y., 2008. Rapid synthesis of highly ordered Si-MCM-41.
KL adsorption equilibrium constant related to the energy of J. Colloid Interface Sci. 319, 377e380. https://doi.org/10.1016/j.jcis.2007.11.025.
Manzano, M., Vallet-Regi, M., 2010. New developments in ordered mesoporous
adsorption of Langmuir equation, L/mg materials for drug delivery. J. Mater. Chem. 20, 5593e5604. https://doi.org/
KR one of the isotherm constants of the Redlich-Peterson 10.1039/b922651f.
equation, which is related to the energy of adsorption Mercier, L., Pinnavaia, T.J., 1997. Access in mesoporous materials: advantages of a
uniform pore structure in the design of a heavy metal ion adsorbent for envi-
M the mass of the added adsorbent in the batch adsorption
ronmental remediation. Adv. Mater. 9, 500e503. https://doi.org/10.1002/
procedure, g adma.19970090611.
R2 the coefficient of determination Mohanta, D., Ahmaruzzaman, M., 2018. Bio-inspired adsorption of arsenite and
h a parameter of pseudo-second-order model fluoride from aqueous solutions using activated carbon SnO2 nanocomposites:
isotherms, kinetics, thermodynamics, cost estimation and regeneration studies.
1/n the adsorption intensity parameter of Freundlich J. Environ. Chem. Eng. 6, 356e366. https://doi.org/10.1016/j.jece.2017.11.076.
equation, dimensionless Morasch, B., Bonvin, F., Reiser, H., Grandjean, D., de Alencastro, L.F., Perazzolo, C.,
pKa the ionization constant Chevre, N., Kohn, T., 2010. Occurrence and fate of micropollutants in the vidy
bay of lake geneva Switzerland Part II: micropollutant removal between
qF the adsorption capacity of Freundlich equation, (mg/ wastewater and raw drinking water. Environ. Toxicol. Chem. 29, 1658e1668.
g)(L/mg)1/n https://doi.org/10.1002/etc.222.
qe the amount of CIP adsorbed at equilibrium time in the Munoz, B., Ramila, A., Perez-Pariente, J., Diaz, I., Vallet-Regi, M., 2003. MCM-41
organic modification as drug delivery rate regulator. Chem. Mater. 15, 500e503.
batch adsorption procedure, mg/g https://doi.org/10.1021/cm021217q.
qm the maximum monolayer adsorption capacity of Pauwels, B., Tendeloo, G.V., Thoelen, C., Van Rhijin, W., Jacobs, P.A., 2001. Structure
Langmuir equation, mg/g determination of spherical MCM-41 particles. Adv. Mater. 13, 1317e1320.
https://doi.org/10.1002/1521-4095 (200109)13:17<1317::AID-
a a parameter of Elovich equation ADMA1317>3.0.CO;2-5.
b one of the isotherm constants of the Redlich-Peterson Rivera-Utrilla, J., Sanchez-Polo, M., Prados-Joya, G., Ferro-Garcia, M.A., Bautista-
equation Toledo, I., 2010. Removal of tinidazole from waters by using ozone and activated
carbon in dynamic regime. J. Hazard Mater. 174, 880e886. https://doi.org/
10.1016/j.jhazmat.2009.09.059.
References Roik, N.V., Belyakova, L.A., Dziazko, M.O., 2017. Adsorption of antitumor antibiotic
doxorubicin on MCM-41-type silica surface. Adsorpt. Sci. Technol. 35, 86e101.
Arica, T.A., Ayas, E., Arica, M.Y., 2017. Magnetic MCM-41 silica particles grafted with https://doi.org/10.1177/0263617416669504.
poly (glycidylmethacrylate) brush: modification and application for removal of Rosenholm, J.M., Meinander, A., Peuhu, E., Niemi, R., Eriksson, J.E., Sahlgren, C.,
direct dyes. Microporous Mesoporous Mater. 243, 164e175. https://doi.org/ Linden, M., 2008. Targeting of porous hybrid silica nanoparticles to cancer cells.
10.1016/j.micromeso.2017.02.011. ACS Nano 3, 197e206. https://doi.org/10.1021/nn800781r.
Baccile, N., Babonneau, F., 2008. Organo-modified mesoporous silicas for organic Song, B., Xu, P., Zeng, G., Gong, J., Wang, X., Yan, J., Wang, S., Zhang, P., Cao, W., Ye, S.,
pollutant removal in water: solid-state NMR study of the organic/silica in- 2018. Modeling the transport of sodium dodecyl benzene sulfonate in riverine
teractions. Microporous Mesoporous Mater. 110, 534e542. https://doi.org/ sediment in the presence of multi-walled carbon nanotubes. Water Res. 129,
10.1016/j.micromeso.2007.06.056. 20e28. https://doi.org/10.1016/j.watres.2017.11.003.
Badruzzaman, M., Westerhoff, P., Knappe, D.R.U., 2004. Intraparticle diffusion and Sui, Q., Huang, J., Deng, S.B., Yu, G., Fan, Q., 2010. Occurrence and removal of
adsorption of arsenate onto granular ferric hydroxide (GFH). Water Res. 38, pharmaceuticals caffeine and DEET in wastewater treatment plants of Beijing
4002e4012. https://doi.org/10.1016/j.watres.2004.07.007. China. Water Res. 44, 417e426. https://doi.org/10.1016/j.watres.2009.07.010.
Benotti, M.J., Trenholm, R.A., Vanderford, B.J., Holady, J.C., Stanford, B.D., Vallet-Regi, M., Ramila, A., del Real, R.P., Perez-Pariente, J., 2001. A new property of
Snyder, S.A., 2009. Pharmaceuticals and endocrine disrupting compounds in US MCM-41: drug delivery system. Chem. Mater. 13, 308e311. https://doi.org/
drinking water. Environ. Sci. Technol. 43, 597e603. https://doi.org/10.1021/ 10.1021/cm0011559.
es801845a. Vieno, N.M., Harkki, H., Tuhkanen, T., Kronberg, L., 2007. Occurrence of pharma-
Braun, K., Pochert, A., Beck, M., Fiedler, R., 2016. Dissolution kinetics of mesoporous ceuticals in river water and their elimination a pilot-scale drinking water
silica nanoparticles in different simulated body fluids. J. Sol. Gel Sci. Technol. 79, treatment plant. Environ. Sci. Technol. 41, 5077e5084. https://doi.org/10.1021/
319e327. https://doi.org/10.1007/s10971-016-4053-9. es062720x.
Caliskan, E., Gokturk, S., 2010. Adsorption characteristics of sulfamethoxazole and Walcarius, A., Mercier, L., 2010. Mesoporous organosilica adsorbents: nano engi-
metronidazole on activated carbon. Separ. Sci. Technol. 45, 244e255. https:// neered materials for removal of organic and inorganic pollutants. J. Mater.
doi.org/10.1080/01496390903409419. Chem. 20, 4478e4511. https://doi.org/10.1039/b924316j.
Girish, C.R., Murty, V.R., 2016. Mass transfer studies on adsorption of phenol from Wang, S., Tan, L., 2007. Reversible and irreversible adsorption of dye on mesoporous
10 D. Lu et al. / Journal of Cleaner Production 264 (2020) 121644
materials in aqueous solution. Stud. Surf. Sci. Catal. 165, 227e230. https:// Zhao, S., Ba, C., Yao, Y., Zheng, W., Economy, J., Wang, P., 2018. Removal of antibiotics
doi.org/10.1016/S0167-2991 (07)80305-3. using polyethylenimine cross-linked nanofiltration membranes: relating
Xiao, Y., Lasaga, A.C., 1994. Ab initio quantum mechanical studies of the kinetics and membrane performance to surface charge characteristics. Chem. Eng. J. 335,
mechanisms of silicate dissolution: Hþ (H3Oþ) catalysis. Geochem. Cosmochim. 101e109. https://doi.org/10.1016/j.cej.2017.10.140.
Acta 58, 5379e5400. https://doi.org/10.1016/0016-7037 (94)90237-2.