This is an Accepted Manuscript of an article published by Taylor & Francis in Journal of Environmental Science
and Health, Part A on 18 Sept 2017, available online: http://
www.tandfonline.com/10.1080/10934529.2017.1357406
1
Metronidazole removal in powder-activated carbon and concrete-
containing graphene adsorption systems: Estimation of kinetic,
equilibrium and thermodynamic parameters and optimization of
adsorption by a central composite design
S. V. Manjunath
a
, S. Mathava Kumar
a,
*, Huu Hao Ngo
b
, and Wenshan Guo
b
a
Department of Civil Engineering, Indian Institute of Technology Madras, Chennai, Tamil Nadu, India;
b
School of Civil and Environmental Engineering, University of Technology Sydney, Sydney, Australia
* Corresponding authors. Mathava Kumar (mathav@iitm.ac.in; mathavakumar@gmail.com);
Department of Civil Engineering, Indian Institute of Technology Madras, Chennai 600036, Tamil Nadu,
India.
Abstract
Metronidazole (MNZ) removal by two adsorbents, i.e., concrete-containing graphene (CG) and
powder-activated carbon (PAC), was investigated via batch-mode experiments and the outcomes
were used to analyze the kinetics, equilibrium and thermodynamics of MNZ adsorption. MNZ
sorption on CG and PAC has followed the pseudo-second-order kinetic model, and the
thermodynamic parameters revealed that MNZ adsorption was spontaneous on PAC and non-
spontaneous on CG. Subsequently, two-parameter isotherm models, i.e., Langmuir, Freundlich,
Temkin, Dubinin–Radushkevich and Elovich models, were applied to evaluate the MNZ
adsorption capacity. The maximum MNZ adsorption capacities (q
m
) of PAC and CG were found
to be between 25.5–32.8 mg/g and 0.41–0.002 mg/g, respectively. Subsequently, the effects of pH,
temperature and adsorbent dosage on MNZ adsorption were evaluated by a central composite
design (CCD) approach. The CCD experiments have pointed out the complete removal of MNZ at
a much lower PAC dosage by increasing the system temperature (i.e., from 20°C to 40°C). On the
other hand, a desorption experiment has shown 3.5% and 1.7% MNZ removal from the surface of
PAC and CG, respectively, which was insignificant compared to the sorbed MNZ on the surface
by adsorption. The overall findings indicate that PAC and CG with higher graphene content could
be useful in MNZ removal from aqueous systems.
Keywords: Adsorption, isotherm, kinetics, metronidazole, thermodynamics
1. Introduction
Pharmaceuticals constitute a diverse group of organic compounds and are considered to be one of
the most important emerging contaminants in the recent years. These include different compounds
such as antibiotics, hormones, analgesics and anti-inflammatory drugs, antiepileptic drugs, blood
lipid regulators, β blockers, contrast media and cytostatic drugs.
[1]
The usage of antibiotics has been
increasing considerably day by day owing to the personal requirements of people, and demand in
industries such as animal pharmaceuticals, aquaculture, poultries, piggeries, etc. The discharge of
antibiotics into the environment could produce antimicrobial-resistant genes and they can also
produce eco-toxicity.
[2]
Antibiotics and/or their metabolites were found to have longer stability or
half-life in aqueous and soil environments. Recent studies have reported the persistence of
2
antibiotics of different classes in aqueous environments
[3–5]
and the release of antibiotics from
wastewater treatment plant effluents.
[6,7]
Metronidazole (MNZ), an antibacterial and anti-
inflammatory agent,
[8]
is one of the heavily used antibiotics worldwide,
[9]
which has been used to
treat diseases caused by anaerobic bacteria, bacteroides and protozoa. MNZ has very high solubility
(9.8 g/L) and molecular diffusivity (8.48 × 10
6
cm
2
/s) in water, and is expected to be highly mobile
in aqueous systems. Recently, Rosal et al.
[10]
reported the presence of MNZ concentration in the
influent (165 ng/L) and effluent (127 ng/L) samples of a sewage treatment plant.
The conventional wastewater treatment process relies mainly on the function of biological
treatment units. However, these units are ineffective in the removal of a wide variety of antibiotics
including MNZ.
[11]
Therefore, it is vital to include appropriate technology (as a tertiary treatment)
in the treatment process to remove the emerging contaminants including MNZ. Due to the
increasing interest in the removal of antibiotics from aqueous systems, several researchers explored
the possibility of MNZ removal by electrochemical process,
[12]
photocatalysis,
[13]
electro-catalytic
reduction,
[14]
membrane process,
[15]
Fenton process
[16]
and coupled electro-reduction-biological
treatment,
[17]
etc. However, adsorption, one of the oldest techniques, has been found to be very
effective in the removal of a wide variety of trace and gross organics. Moreover, this process has
been used extensively in the wastewater treatment process owing to its ease in operation, low cost,
absence of by-product formation, regeneration potential and sludge-free operation when compared
to other treatment methods.
[18,19]
Due to rapid and excessive urbanization, construction and demolition wastes have become a major
concern in the context of urban solid waste management.
[20]
The disposal of construction and
demolition wastes has become one of the biggest concerns throughout the world, as it is generated
in huge quantities. Landfilling of construction and demolition wastes is the main current practice
in many developing countries. Moreover, it was reported that the wastes were dumped illegally on
land or in natural drainages in most developing countries due to shortage of space for dumping.
However, the construction and demolition wastes containing cementitious material can be reused
or recycled for other processes/applications. Several investigations in the past reported the
application of cementitious materials in the removal of various pollutants from water and
wastewater including phosphorous by composite cement mortars;
[21]
fecal coliforms and
phosphorous by pervious geopolymer concrete;
[22]
p-chloronitrobenzene by the cementitious
catalytic membrane with ozonation;
[23]
Indigo carmine by concrete composite;
[24]
and heavy metals
(Cu, Cd, Zn and Pb) by zeolite-Portland cement mixture.
[25]
In the recent years, graphene and graphene composites have been used as adsorbents to treat water
and wastewater containing heavy metals, organic dyes and antibiotics. To mention a few, iron–
aluminum oxide–graphene oxide composite for fluoride removal,
[26]
graphene oxide for removing
diclofenac and sulfamethoxazole antibiotics,
[27]
polysaccharide-modified graphene oxides for the
removal of cationic dyes (Methylene blue, Rhodamine 6G) and anionic dyes (Orange II, Acid
fuchsin),
[28]
and graphene oxide membranes for Cu
2+
, Cd
2+
and Ni
2+
removal are some examples.
[29]
On the other hand, several investigations are in progress to study the addition of graphene/graphene
oxide in the concrete preparation to increase its strength and other characteristics. The application
of concrete prepared with the addition of graphene for water purification especially for antibiotics'
removal has not been investigated. Moreover, the requirement/suitability of graphene content in
3
concrete for the effective removal of antibiotics was not addressed in the past. At the same time, it
is essential to compare the performance of modified concrete with graphene with commercially
available adsorbents including PAC which has huge importance in the application of modified
concrete for water purification applications. On the other hand, the interaction effect of adsorbent
dosage, pH and temperature on the removal of antibiotics using a central composite design (CCD)
along with isotherm experiments was not carried out in the past. Therefore, this investigation was
focused to evaluate the performance of concrete-containing graphene (CG) in MNZ removal and
its effectiveness was compared with the MNZ removal potential of powdered activated carbon
(PAC) by calculating the kinetic rates, thermodynamic parameters and adsorption capacity using
isotherms. Moreover, this study was extended to explore (a) the interaction effects of temperature,
pH and adsorbent dosage on MNZ removal and (b) the extent of MNZ desorption from CG and
PAC.
2. Materials and methods
2.1. Preparation of adsorbent
PAC was supplied by Merck, India, and it was used without any modification. CG (2%w/w, 1–
2 mm size) was obtained from the Building Technology and Construction Management laboratory
of IIT Madras, India. Prior to the adsorption experiments, the CG specimen was crushed, and the
particles retained between ASTM sieve Nos. 8 and 16 were collected and washed several times
with tap water followed by distilled water. Subsequently, the particles were air-dried and stored in
an airtight container for further use. The surface morphology of the adsorbents before and after
adsorption was analyzed by using a scanning electron microscope (GENESIS-2100 SEM,
EmCrafts, South Korea).
2.2. Preparation of adsorbate
MNZ (C
6
H
9
N
3
O
3
) of analytical grade supplied by Sigma-Aldrich was used for the preparation of
stock solution (1,000 mg/L). The prepared stock solution was placed in a volumetric flask, sealed
and stored in a refrigerator. The working solutions of required concentrations were prepared from
the stock solution (1,000 mg/L) by diluting it with distilled water.
2.3. Kinetic and equilibrium adsorption study
2.3.1. Kinetic study
The kinetic study was conducted in a batch mode at an initial MNZ concentration of 10 mg/L.
Exactly 100 mL of solution containing 10 mg/L MNZ was poured into 250-mL conical flasks, and
the adsorbent (either PAC or CG) was added into the flasks at a predetermined adsorbate-to-
adsorbent ratio (basis of wt:wt; 1:100 for PAC and 1:1,000 for CG). Subsequently, the flasks were
kept in a temperature-controlled incubator shaker at 25°C with continuous shaking at 100 rpm for
24 h. At regular time intervals, the samples were withdrawn from the flasks and analyzed for MNZ
concentration.
2.3.2. Equilibrium study
4
Batch-mode equilibrium adsorption experiments were conducted under similar operating
conditions, i.e., at 25°C, 100-mL working volume with continuous shaking with a range of initial
MNZ concentrations (5, 10, 25, 50 and 100 mg/L). The adsorbate-to-adsorbent ratio was fixed as
1:100 and 1:1,000 (on a wt:wt basis) for PAC and CG, respectively. The experiments were
conducted for a pseudo-equilibrium time obtained from the kinetic study. At the end of the
experiment, the samples were collected from the flasks and analyzed for MNZ concentration. Using
the experimental data, MNZ adsorption capacity q
e
(mg/g) at equilibrium was calculated as given
in Eq. (1):
where C
0
and C
e
are MNZ concentrations at the start of the experiment and at equilibrium (mg/L),
respectively. V is the volume of MNZ solution (L) and M is the mass of adsorbent (g).
2.4. Quantification of MNZ concentration
After the adsorption study, an aliquot of MNZ solution was collected, centrifuged and filtered.
Subsequently, the filtered sample was analyzed using high-performance liquid chromatography
(HPLC) fitted with a UV–Vis variable wavelength detector (Dionex UltiMate 3000). The C18
chromatographic column (Acclaim 120, 5 μm, 4.6 × 250 mm) was used to separate the compounds.
The column was operated at a reverse phase mode at a wavelength of 254 nm using
acetonitrile:water (60:40) as a mobile phase. The pump was operated at a flow rate of 1 mL/min.
The HPLC analysis results were used to calculate the MNZ removal as per Eq. (2):
2.5. Determination of rate constants
The rate of adsorption was determined using the adsorption equilibrium data and equilibrium
models (i.e., the pseudo-first-order and pseudo-second-order models).
Pseudo-first-order model
Lagergren's pseudo-first-order rate equation [Eq. (3)] describes the adsorption of liquid–solid
systems based on the concentration of the solution and adsorption capacity of the solid. Equation (3)
states that the rate of adsorption is equal to the distance to equilibrium. At time t = 0 for fresh
adsorbent, q
t
= 0 and the distance of equilibrium is q
e
. The increase in time reduces the distance of
equilibrium, while the distance disappears at equilibrium, i.e., q
e
q
t
= 0.
[30]
Equation (4) depicts
the expression for the pseudo-first-order model in a linearized form:
where q
e
and q
t
are adsorption capacities (mg/g) at equilibrium and at various times (t), respectively.
K1 is the pseudo-first-order rate constant (min
−1
) that was obtained by plotting a graph of ln(q
e
q
t
)
versus t.