processes

Article
Modifying Nanoporous Carbon through Hydrogen
Peroxide Oxidation for Removal of Metronidazole
Antibiotics from Simulated Wastewater
Teguh Ariyanto 1,2, * , Rut Aprillia Galuh Sarwendah 1 , Yove Maulana Novirdaus Amimmal 1 ,
William Teja Laksmana 1 and Imam Prasetyo 1,2
1 Department of Chemical Engineering, Faculty of Engineering, Universitas Gadjah Mada, Jl Grafika No 2,
Yogyakarta 55281, Indonesia; rutruut23@gmail.com (R.A.G.S.); yove.maulana.n@mail.ugm.ac.id (Y.M.N.A.);
william.teja.l13@gmail.com (W.T.L.); Imampras@ugm.ac.id (I.P.)
2 Carbon Material Research Group, Department of Chemical Engineering, Faculty of Engineering,
Universitas Gadjah Mada, Jl Grafika No 2, Yogyakarta 55281, Indonesia
* Correspondence: teguh.ariyanto@ugm.ac.id; Tel.: +62-274-649-2171




Received: 6 October 2019; Accepted: 5 November 2019; Published: 8 November 2019


Abstract: This study examined change in pore structure and microstructure of nanoporous carbon
after surface oxidation and how it affects the adsorption performance of metronidazole antibiotics.
The surface oxidation was performed by hydrogen peroxide at 60 ◦ C. The properties of porous
carbon were investigated by N2 -sorption analysis (pore structure), scanning electron microscopy
(surface morphology), the Boehm titration method (quantification of surface functional group), and
Fourier transform infrared spectroscopy (type of surface functional group). The results showed
that the oxidation of porous carbon by hydrogen peroxide has a minor defect in the carbon pore
structure. Only a slight decrease in specific surface area (8%) from its original value (973 m2 g−1 )
was seen but more mesoporosity was introduced. The oxidation of porous carbon with hydrogen
peroxide modified the amount of oxide groups i.e., phenol, carboxylic acid and lactone. Moreover, in
the application the oxidized carbon exhibited a higher the metronidazole uptake capacity of up to
three-times manifold with respect to the pristine carbon.

Keywords: adsorption; metronidazole; porous carbon; surface modification; wastewater treatment

1. Introduction
The occurrence of pharmaceutical compounds in the aquatic environment has been a high concern.
When water containing pharmaceutical compounds is released into river, it can influence the quality
of water and harm the ecosystem balance. Studies showed that pharmaceutical compounds in an
aquatic system could have serious effects such as antibiotic-resistant microbes [1], the change of
fish reproduction [2], and inhibition of photosynthesis in aquatic organisms [3]. When a waterbody
containing antibiotics flows to agriculture land, this could affect the maturity of plants [4]. One of
the main sources of pharmaceutical effluents is hospitals which, in Europe, typically discharges
pharmaceutical components in the range concentrations of 0.2–11 µg L−1 [5]. This could be even worse
in other countries when using improper wastewater treatment, resulting in a higher concentration of
pharmaceutical substances released to the environment.
Metronidazole is one of the most used pharmaceutical compounds. This is an antibiotic which
hinders the growth of parasitic microbes like bacteria and protozoa [6]. Metronidazole is highly soluble
in water and persist even after biodegradation [7]. The molecular structure of metronidazole is given
in Figure 1.

Processes 2019, 7, 835; doi:10.3390/pr7110835 www.mdpi.com/journal/processes

Processes 2019, 7, 835 2 of 9
Processes 2019, 7, 835 2 of 9

Figure 1. Molecular structure of metronidazole.

Figure 1. Molecular structure of metronidazole

Adsorption is a popular technique for removal of contaminants from wastewater. The method has
Adsorption is a popular technique for removal of contaminants from wastewater. The method
the advantages of a simple design and ease of operation. Among adsorbents, porous carbon has been
has the advantages of a simple design and ease of operation. Among adsorbents, porous carbon has
regarded as the most promising materials, taking advantages of its excellent physical and chemical
been regarded as the most promising materials, taking advantages of its excellent physical and
stability and high specific surface area (500–2000 m2 g−1 ) [8–10]. Studies showed porous carbon
chemical stability and high specific surface area (500–2,000 m2g−1) [8–10]. Studies showed porous
could work to remove metronidazole from aqueous systems [11–13]. The capacity of adsorption of
carbon could work to remove metronidazole from aqueous systems [11–13]. The capacity of
organic compounds is influenced e.g. pore textural properties and surface functional groups of porous
adsorption of organic compounds is influenced e.g. pore textural properties and surface functional
carbon [10,14]. Therefore, there is possibility to further increase adsorption capacity by modification of
groups of porous carbon [10,14]. Therefore, there is possibility to further increase adsorption capacity
pore- and surface chemistry of nanoporous carbon. A high uptake capacity is beneficial providing
by modification of pore- and surface chemistry of nanoporous carbon. A high uptake capacity is
longer adsorption time, hence more efficient. However, no detail study yet is presented so far to modify
beneficial providing longer adsorption time, hence more efficient. However, no detail study yet is
pore- and surface chemistry of porous carbon for adsorption of antibiotic substances.
presented so far to modify pore- and surface chemistry of porous carbon for adsorption of antibiotic
A presence of functional groups like acidic sites could increases wettability of pore surfaces.
substances.
To introduce acidic groups onto nanoporous carbon, surface oxidation using sulphuric acid, nitric
A presence of functional groups like acidic sites could increases wettability of pore surfaces. To
acid and hydrogen peroxide can be performed [15,16]. Oxidation using strong acids is effective to
introduce acidic groups onto nanoporous carbon, surface oxidation using sulphuric acid, nitric acid
increase functional groups of oxygen, but this relatively harsh oxidation induces a destruction of pore
and hydrogen peroxide can be performed [15,16]. Oxidation using strong acids is effective to increase
structures hence remarkably decreasing specific surface area [17,18]. Oxidation of nanoporous carbon
functional groups of oxygen, but this relatively harsh oxidation induces a destruction of pore
using hydrogen peroxide can give a good trade-off between introduction of oxygen-functional groups
structures hence remarkably decreasing specific surface area [17,18]. Oxidation of nanoporous carbon
and change in pore structures. Furthermore, this oxidation method does not require a further step of
using hydrogen peroxide can give a good trade-off between introduction of oxygen-functional
washing of nanoporous carbon.
groups and change in pore structures. Furthermore, this oxidation method does not require a further
This work presents a study of ease modification of nanoporous carbon and its efficacy on
step of washing of nanoporous carbon.
metronidazole adsorption. Oxidation was performed by hydrogen peroxide. Effects of this mild
This work presents a study of ease modification of nanoporous carbon and its efficacy on
oxidation on the pore structure, surface morphology and functional groups and their influence on the
metronidazole adsorption. Oxidation was performed by hydrogen peroxide. Effects of this mild
performance of metronidazole are described.
oxidation on the pore structure, surface morphology and functional groups and their influence on
the performance
2. Materials and of metronidazole are described.
Methods

2. Materials
2.1. Material and Methods
Nanoporous carbon from coconut shell (CSC) was obtained from PT Home System, Indonesia.
2.1. Material
CSC was then crushed and sieved to obtain a particle size in the range of 355–710 µm. Hydrogen
Nanoporous
peroxide solution carbon
of 50% from coconut shell
was obtained from(CSC)
Merck,was obtainedHydrogen
Germany. from PT Home System,
peroxide Indonesia.
solution of 30%
CSC was then crushed and sieved to obtain a particle size in the range of 355–710 µm.
was prepared by dilution of highly concentrated hydrogen peroxide of 50%. Metronidazole 99% Hydrogen
purity
peroxide
was solution
a product of 50%
of Sigma was obtained
Aldrich, from Merck, Germany. Hydrogen peroxide solution of 30%
Singapore.
was prepared by dilution of highly concentrated hydrogen peroxide of 50%. Metronidazole 99%
2.2. Oxidation
purity of Nanoporous
was a product Carbon
of Sigma Aldrich, Singapore.
Fifty milliliters of 30% hydrogen peroxide solution was poured in a three-neck flask. After this,
2.2. Oxidation of Nanoporous Carbon
the solution was heated to the desired oxidation temperature of 60 ◦ C. After oxidation was finished
(2 h),Fifty milliliters
carbon of 30% hydrogen
was separated peroxide
from the solution solution
and was poured
dried overnight in in
anaoven (60 ◦ C).flask. After this,
three-neck
the solution was heated to the desired oxidation temperature of 60 °C. After oxidation was finished
2.3.h),Characterization
(2 of Nanoporous
carbon was separated Carbon
from the solution and dried overnight in an oven (60 °C).
Morphologies of material CSC, both before and after the oxidation process, were probed by
2.3. Characterization of Nanoporous
scanning electron microscope (SEM)Carbon
of JSM-6510LA (JEOL, Tokyo, Japan). Pore analysis of nanoporous
carbonMorphologies of material CSC, both before Nova
was performed by using Quantachrome 2000the
and after (Quantachrome Instruments,
oxidation process, Boynton
were probed by
scanning electron microscope (SEM) of JSM-6510LA (JEOL, Tokyo, Japan). Pore analysis of
nanoporous carbon was performed by using Quantachrome Nova 2000 (Quantachrome Instruments,
Processes 2019, 7, 835 3 of 9

Beach, FL, USA) through the nitrogen sorption method. The multipoint Brunauer–Emmett–Teller (BET)
method was employed to evaluate the specific surface area (SSA) from the isotherm data. Micropore
volume and micropore SSA were evaluated using the t-plot method, which is a correlation between
the adsorbed volume of nitrogen and statistical thickness. BET and statistical thickness of the Hasley
model are shown in Equations (1) and (2) [19]. Calculation of pore textural parameters were carried
out using a software provided by Quantachrome NovaWin version 11.03 (Quantachrome Instruments,
Boynton Beach, FL, USA). !
P 1 C−1 P
= + (1)
V (P0 − P) Vm C Vm C P0
" #1/3
5
t(A) = 3.54 (2)
2.303 log(Po /P)
where P is the pressure, V is the adsorbed volume of nitrogen, Po is the saturated pressure of nitrogen at
77 K, Vm is the monolayer adsorbed volume of nitrogen, t is the statistical thickness, and C is constant.
The qualitative analysis of functional groups in porous carbon was evaluated through
Fourier-Transform Infrared Spectroscopy (FTIR) using Nicolet iS10 (Thermo Fisher Scientific, MA,
USA), while the quantitative analysis was performed through the Boehm titration method [20].
Fifty milligrams of carbons were soaked in 50 mL of 0.001 N Na2 CO3 (prepared from solid Na2 CO3 ,
Merck, Darmstadt, Germany), 0.001 N NaHCO3 (prepared from solid NaHCO3 , Merck, Darmstadt,
Germany), 0.001 N NaOH (prepared from solid NaOH, Merck, Darmstadt, Germany), for 24 h. Next,
the carbons were separated and the solutions of Na2 CO3 , NaHCO3 , and NaOH of the separation’s
result were titrated with 0.001 N HCl (prepared from concentrated HCl, Merck, Darmstadt, Germany).
The number of functional groups present were determined through a balance of acid in the solution.

2.4. Adsorption of Antibiotic Substances

The metronidazole adsorption was performed through a preparation of metronidazole solution
with various concentrations, from 20 to 200 ppm. This initial concentration range was employed to
determine the equilibrium curve between the amount of adsorbate in the solid phase and the amount
of adsorbate in the liquid phase, until the maximum adsorption capacity was reached. Fifty milliliters
of metronidazole solution were put into Erlenmeyer of 125 mL and 50 mg of carbon samples was
added for each of the solution samples. The Erlenmeyer was shaken for 24 h in the water bath shaker,
(Memmert, Schwabach, Germany) at room temperature. After adsorption, the metronidazole solution
was then analyzed by the method described in literature [21]. The solution was mixed with reagents of
1 N HCl, zinc dust, 1% NaNO2 (prepared from solid NaNO2 , Merck, Darmstadt, Germany) solution,
2% sulfamic acid solution, 1% N-(1-naphthyl)ethylenediamine dihydrochloride or NEDA (prepared
from ~100% NEDA, Merck, Darmstadt, Germany) solution, and 10% NaOH solution until it was
colored. The metronidazole concentration in the solution after the adsorption was measured with
UV–Vis spectrophotometer Shimadzu UV Mini 1240 (Shimadzu, Kyoto, Japan) with a wavelength of
510 nm. The experimental set-up and adsorption conditions were based on the reports in [22,23].

2.5. Adsorption Model of Antibiotic Substances

Adsorption of metronidazole was modeled using Langmuir and Freundlich isotherm models, as
shown in Equations (3) and (4) [19].
Qm KL Ce
Qe = (3)
1 + KL Ce
Qe = KF Ce1/n (4)

where KL is the Langmuir equilibrium constant (L/mol), Qm is the maximum adsorbate adsorption
capacity of the adsorbent (mg adsorbate/g adsorbent), KF is the Freundlich equilibrium constant (mg
adsorbate/(g adsorbent (mol/L)n )), n is the correction factor, Qe is the adsorbate adsorption capacity
 Qe = K F Ce1/n (4)

where KL is the Langmuir equilibrium constant (L/mol), Qm is the maximum adsorbate adsorption
capacity of the adsorbent (mg adsorbate/g adsorbent), KF is the Freundlich equilibrium constant (mg
adsorbate/(g
Processes adsorbent (mol/L)n)), n is the correction factor, Qe is the adsorbate adsorption capacity
2019, 7, 835 4 of 9
(mg adsorbate/g adsorbent) in the solid state in an equilibrium condition with a concentration of
adsorbates in the liquid, and Ce is the adsorbate concentration in the solution (mol/L).
(mg adsorbate/g adsorbent) in the solid state in an equilibrium condition with a concentration of
The mass balance of metronidazole in the solution and adsorbent were derived to get the
adsorbates in the liquid, and Ce is the adsorbate concentration in the solution (mol/L).
adsorption variable. The variable was used to evaluate the adsorption parameter of Equation (3) and
The mass balance of metronidazole in the solution and adsorbent were derived to get the
Equation (4). The mass balance derivative of metronidazole for the batch system was written as
adsorption variable. The variable was used to evaluate the adsorption parameter of Equations (3) and
Equation (5).
(4). The mass balance derivative of metronidazole for the batch system was written as Equation (5).
( Co − Ct )V = ( Qt − Qo ) w (5)
(Co − Ct)V = (Qt − Qo)w (5)
In the initial condition, before metronidazole was adsorbed by carbon, Qo = 0, so Equation (5)
In the
became initial condition,
Equation (6). before metronidazole was adsorbed by carbon, Qo = 0, so Equation (5)
became Equation (6).
((Co )V) V
Co−−CeCe
Qe =
Qe = (6)
(6)
ww
where Co is the initial concentration of metronidazole in the solution (mg adsorbate/L), V is the solution
where Co is the initial concentration of metronidazole in the solution (mg adsorbate/L), V is the
volume (L), and w is the adsorbent mass (g).
solution volume (L), and w is the adsorbent mass (g).
3. Results and Discussion
3. Results and Discussion
3.1. Pore Structure of Carbon Material
3.1. Pore Structure of Carbon Material
The pore structure of CSC, before and after oxidation with hydrogen peroxide was characterized
The pore structure
by low-temperature of CSC,
nitrogen before and at
physisorption after oxidation
77 K. with hydrogen
The sorption isotherms peroxide was characterized
are displayed in Figure 2A.
by low-temperature nitrogen physisorption at 77 K. The sorption isotherms are
The CSC and CSC-OX feature an identical isotherm of Type I, according to the IUPAC classificationdisplayed in Figure
[24],
1A. The CSC
indicating andmaterials
that the CSC-OX possess
feature substantial
an identical isotherm
pores of Type regime.
in a micropore I, according to the
Of note IUPAC
is that the
classification [24], indicating that the materials possess substantial pores in a micropore
amount of nitrogen adsorbed on the CSC-OX was low. This indicated a decrease in the specific surface regime. Of
area (SSA). The pore-sized distributions (PSDs) determined through the QSDFT model are shownin
note is that the amount of nitrogen adsorbed on the CSC-OX was low. This indicated a decrease
the
in specific
Figure 2B.surface
PSD of area
CSC (SSA). The pore-sized
and CSC-OX distributions
are dominantly (PSDs) determined
in micropore regime aroundthrough the QSDFT
(<2 nm). When
model are shown in Figure 1B. PSD of CSC and CSC-OX are dominantly
comparing the PSDs of CSC and CSC-OX in the mesopore range (2–50 nm), it could be seen in micropore regime around
that
a(<2 nm).fraction
higher When comparing
of mesoporesthe PSDs of CSC and
were present CSC-OX
in the CSC-OX. in the
Thesemesopore range
pores were (2–50created
likely nm), itwhen
could
be seen that
performing thea oxidation
higher fraction
process ofusing
mesopores were
hydrogen present in the CSC-OX. These pores were likely
peroxide.
created when performing the oxidation process using hydrogen peroxide.

Figure2.1.(A)
Figure (A)N
N2-sorption
-sorptionisotherm
isothermsorption.
sorption.(B)
(B)The
Thepore
poresize
sizedistribution
distributionof
ofcarbon
carbonevaluated
evaluatedby bythe
the
2
NN-QSDFT
2
2-QSDFT method. In Figure 1A, P is the pressure and Po is the saturated pressure of nitrogen at 77 K.
method. In Figure 2A, P is the pressure and Po is the saturated pressure of nitrogen at 77 K.

Table1 1shows
Table shows
thethe pore
pore parameters
parameters of theof carbon
the carbon
both both
beforebefore andthe
and after after the oxidation
oxidation process,process,
which
which were calculated based on the N -sorption isotherm data. From the analysis result, the
were calculated based on the N2 -sorption isotherm data. From the analysis result, the SSA, total pore
2 SSA,
volume, and percentage of micropore fraction after the oxidation process were decreased when compared
to the pristine one. The decreases occurred likely due to the oxidation process. Oxidation causes a
removal of the carbon element and the formation of functional groups, which then produce bigger
pores [25]. When voids larger than 300 nm are created by the oxidation of pores, they cannot be detected
by nitrogen sorption, resulting in a small decrease in the total pore volume from 0.43 to 0.41 cm3 g−1 .
total pore volume, and percentage of micropore fraction after the oxidation process were decreased
when
when compared
compared to to the
the pristine
pristine one.
one. The
The decreases
decreases occurred
occurred likely
likely due
due to
to the
the oxidation
oxidation process.
process.
Oxidation
Oxidation causes a removal of the carbon element and the formation of functional groups, which then
causes a removal of the carbon element and the formation of functional groups, which then
produce
produce bigger
bigger pores
pores [25].
[25]. When
When voids
voids larger
larger than
than 300
300 nm
nm are
are created
created byby the
the oxidation
oxidation of
of pores,
pores, they
they
cannot be detected by nitrogen sorption, resulting in a small decrease in the total pore
cannot be detected by nitrogen sorption, resulting in a small decrease in the total pore volume from volume from
Processes0.41
0.43 2019, 7, 3835 5 of 9
0.43 to
to 0.41 cm
cm3 gg−1..
−1

Table
Table 1. Analysis result of surface area with the nitrogen sorption method.
Table1.1.Analysis
Analysisresult
resultof
ofsurface
surfacearea
areawith
withthe
thenitrogen
nitrogensorption
sorptionmethod.
method.
Carbon
Carbon SSAaa
SSA Total Pore
Total Pore Percentage of
Percentage of of Percentage of
Percentage
No.
No. Carbon SSA a
Total Pore Volume b Percentage Percentage ofof
Micropore
No. Type (m 2/g)
2 /g) Volume b (cm 3/g) Micropore SSA c (%)c Micropore Volume c (%)
Type
Type (m
(m 2/g) Volume (cm /g)
b 3
(cm 3 /g) Micropore SSA
Micropore c
SSA (%) (%) Micropore Volume
Volume c (%)
c (%)

1. CSC 973 0.43 95 84

1.1. CSC
CSC 973
973 0.43
0.43 95 95 84 84
2.
2.
2. CSC-OX
CSC-OX
CSC-OX 895
895
895 0.41
0.41
0.41 94
94 94 80
80 80
a
aSpecific
Specific surface area
a Specific
surface area surface area; b Pore volume at 0.995 P/Po; c Determined by the t-plot method.
b
b Pore
Pore volume at 0.995 P/Po
volume at 0.995 P/Po
c Determined by the t-plot method
c Determined by the t-plot method

3.2. Surface Morphology

3.2.
3.2. Surface
Surface Morphology
Morphology
The morphology of CSC before and after oxidation was investigated using scanning electron
The
The morphology
morphology of
of CSC
CSC before and after oxidation was
was investigated using
using scanning electron
microscopy. From Figure 3, itbefore
can beand
seenafter
thatoxidation investigated
the microstructures in the CSC’s scanning electron
surface before the
microscopy.
microscopy. From
From Figure
Figure 2,
2, it
it can
can be
be seen
seen that
that the
the microstructures
microstructures in
in the
the CSC’s
CSC’s surface
surface before
before the
the
oxidation process were different. In the 100 times magnification (Figure 3A), the CSC’s surface was
oxidation
oxidation process
process were different. In
In the 100
100 times
times magnification (Figure 2A), the
the CSC’s surface was
quite rough whilewere
in thedifferent.
1000 times the
magnification magnification (Figure
(Figure 3B), the 2A),looked
granules CSC’s surface
smoother was
and no
quite
quite rough
rough while
while in
in the
the 1,000
1,000 times
times magnification
magnification (Figure
(Figure 2B),
2B), the
the granules
granules looked
looked smoother
smoother and
and no
no
void appears.
void
void appears.
appears.

Figure
Figure2.
Figure 3.SEM
2. SEMmicrograph
SEM micrographof
micrograph ofCSC
of CSCat
CSC at100×
at 100×magnification
100× magnification(A)
magnification (A)and
(A) and1000×
and 1000×magnification
1000× magnification(B)
magnification (B).
(B)

In Figure3,4,the
In the surfaceofofCSC-OX
CSC-OX inthe the 100times
times magnification(Figure
(Figure 4A)appears
appears rougher
In Figure
Figure 3, the surface
surface of CSC-OX in in the 100
100 times magnification
magnification (Figure 3A)
3A) appears rougher
rougher
and lookseven
and even moredamaged
damaged thanbefore before theoxidation
oxidation process.InInthe
the 1000times
times magnification
and looks
looks even more
more damaged than than before the
the oxidation process.
process. In the 1,000
1,000 times magnification
magnification
(Figure 4B),the
(Figure the granulesstill
still lookedsmooth
smooth butsomesome voidsstarted
started toappear.
appear. ComparedtotoCSC, CSC,
(Figure 3B),
3B), the granules
granules still looked
looked smooth but but some voids
voids started to
to appear. Compared
Compared to CSC,
CSC-OX
CSC-OX looked rougher and had a bigger void than CSC. The pore structures also looked more
CSC-OX looked
looked rougher
rougher and
and hadhad aa bigger
bigger void
void than
than CSC.
CSC. The
The pore
pore structures
structures also
also looked
looked more
more
crumbled
crumbled than before. The results indicated that performing the oxidation process could alter the
crumbled than
than before.
before. The
The results
results indicated
indicated that
that performing
performing the
the oxidation
oxidation process
process could
could alter
alter the
the
surfacemorphology.
surface morphology.
surface morphology.

Figure
Figure3.
Figure 3.4.SEM
SEMmicrograph
SEM micrographof
micrograph ofofCSC-OX
CSC-OXat
CSC-OX atat100×
100×magnification
100× magnification(A)
magnification (A)and
(A) and1,000×
and 1000×magnification
1,000× magnification(B).
magnification (B).
(B).
3.3. Oxygen Functional Groups
Functional groups in the porous carbon of CSC and CSC-OX were analyzed qualitatively and
quantitatively. The qualitative analysis used Fourier-Transform Infrared spectroscopy (FTIR) while the
quantitative analysis used the Boehm Titration.
From Figure 5, in the CSC before oxidation, there are a group of [C–O] showed by the wave
number around 1200 cm−1 , a group of [C=C] showed by the wave number around 1600 cm−1 , and
3.3. Oxygen Functional Groups
3.3. Oxygen Functional Groups
Functional groups in the porous carbon of CSC and CSC-OX were analyzed qualitatively and
Functional groups in the porous carbon of CSC and CSC-OX were analyzed qualitatively and
quantitatively. The qualitative analysis used Fourier-Transform Infrared spectroscopy (FTIR) while
quantitatively. The qualitative analysis used Fourier-Transform Infrared spectroscopy (FTIR) while
the quantitative analysis used the Boehm Titration.
the quantitative
Processes 2019, 7, 835 analysis used the Boehm Titration. 6 of 9
From Figure 4, in the CSC before oxidation, there are a group of [C–O] showed by the wave
From Figure 4, in the CSC before oxidation, there are a group of [C–O] showed by the wave
number around 1,200 cm−1−1, a group of [C=C] showed by the wave number around 1,600 cm−1−1, and a
number around 1,200 cm , a group of [C=C] showed by the wave number around 1,600 cm , and a
group
groupof
agroup of[C–H]
[C–H]showed
showedby bythe
thewave
wave number around 3,000 cm−1−1. These
Theseindicate the presence of a
of [C–H] showed by the wave number
number around
around 3,000
3,000 cm−1.. These indicate
indicate the
the presence
presence of of a
phenol group
group (at(at ca.
ca. 1,200
1200 cm −1
cm −1) and an aromatic ring of carbon structures
−1 structures (at
(at ca.
ca. 1600 cm−1
1,600cm −1 After the
). After
phenol group (at ca. 1,200 cm ) and an aromatic ring of carbon structures (at ca. 1,600 cm−1). After the
oxidation process, there is a new peak in the wave number around 1700 1,700 cm−1 −1 which shows that after
oxidation process, there is a new peak in the wave number around 1,700cm cm−1 which shows that after
the oxidation process, the group of [C=O] appeared, which is a carbonyl group. When comparing the
the oxidation process, the group of [C=O]appeared, appeared, which is a carbonyl group.When Whencomparing
comparing the
CSC and the CSC-OX spectra, it could be noticed that before the oxidation, the carbon already had
CSC and the CSC-OX spectra, it could be noticed that before the oxidation, the carbon already had
groups of oxygen functional groups. It is understandable because CSC is formed from biomass of
groups of oxygen functional groups. ItIt isis understandable
understandable because
because CSCCSC isis formed
formed from biomass of
lignocellulose material and contains oxygen elements.
lignocellulose material and contains oxygen elements.

C=O C=C
C=O C=C
C-O
C-H C-O
C-H
a.u.
Transmittance/ /a.u.
Transmittance

CSC-OX
CSC-OX

CSC
CSC

4000 3000 2000 1000

4000 3000 2000 1000
-1
Wave number, cm -1
Wave number, cm
CSC and
5. CSC
Figure 4. and CSC-OX’s FTIR analysis
Figure 4. CSC and CSC-OX’s FTIR analysis
Figure 65displays
Figure displaysthethenumber
number of oxygen functional
of oxygen groups
functional in CSC
groups inand
CSCCSC-OX. There are
and CSC-OX. changes
There are
Figure 5 displays the number of oxygen functional groups in CSC and CSC-OX. There are
in the group of lactones, carboxyl, and phenols due to the oxidation process of carbons.
changes in the group of lactones, carboxyl, and phenols due to the oxidation process of carbons. When When the
changes in the group of lactones, carboxyl, and phenols due to the oxidation process of carbons. When
oxidation
the occurs,
oxidation carbon
occurs, carbonloselose
its electron andand
its electron makes
makesit more positively
it more charged.
positively The
charged. occurrence
The of
occurrence
the oxidation occurs, carbon lose its electron and makes it more positively charged. The occurrence
oxygen and hydrogen molecules in the peroxide solution causes the formation of oxygen
of oxygen and hydrogen molecules in the peroxide solution causes the formation of oxygen groups groups in the
of oxygen and hydrogen molecules in the peroxide solution causes the formation of oxygen groups
carbon’s
in surface.
the carbon’s It was It
surface. noticed that carboxylic
was noticed and lactone
that carboxylic groupsgroups
and lactone are more
arepresence in CSC-OX
more presence and
in CSC-
in the carbon’s surface. It was noticed that carboxylic and lactone groups are more presence in CSC-
phenol
OX andgroup
phenolisgroup
decreasing as the oxidation
is decreasing process takes
as the oxidation process place.
takes place.
OX and phenol group is decreasing as the oxidation process takes place.
1.2
1.2
CSC
CSC
1.0 CSC-OX
1.0 CSC-OX

0.8
0.8
mmole/gram
mmole/gram

0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0 Lactone Carboxyl Phenol
Lactone Carboxyl Phenol
Figure 5. Analysis result of the acid group before and after the oxidation process with the Boehm
Figure6.5. Analysis
Figure Analysis result
result of
of the
the acid
acid group
groupbefore
beforeand
andafter
afterthe
theoxidation
oxidationprocess
processwith
withthe
theBoehm
Boehm
titration method of CSC and CSC-OX.
titration method of CSC and CSC-OX.
titration method of CSC and CSC-OX.

3.4. Metronidazole Adsorption Performance

The oxidation process tended to increase the mesoporosity of porous carbon (shown in Table 1) and
also increased the number of oxygen functional groups (Figures 5 and 6). CSC and CSC-OX were then
 Processes 2019, 7, 835 7 of 9

3.4. Metronidazole Adsorption Performance

The
Processes oxidation
2019, process tended to increase the mesoporosity of porous carbon (shown in Table
7, 835 7 of 1)
9
and also increased the number of oxygen functional groups (Figure 4 and 5). CSC and CSC-OX were
then employed for metronidazole adsorption. It can be seen from Figure 6 that the adsorption
employed for metronidazole adsorption. It can be seen from Figure 7 that the adsorption performance
performance of oxidized carbon is better than the pristine carbon in the term of uptake capacity. At
of oxidized carbon is better than the pristine carbon in the term of uptake capacity. At equilibrium of
equilibrium of 45 mg/L concentration, adsorbed metronidazole in CSC-OX is ca. 310 mg/g adsorbent
45 mg/L concentration, adsorbed metronidazole in CSC-OX is ca. 310 mg/g adsorbent while in CSC
while in CSC is only ca. 100 mg/g adsorbent. This shows that modifications of porous carbon through
is only ca. 100 mg/g adsorbent. This shows that modifications of porous carbon through hydrogen
hydrogen peroxide improve the adsorption performance. The high adsorption capacity of
peroxide improve the adsorption performance. The high adsorption capacity of nanoporous carbon
nanoporous carbon obtained by the simple oxidation process could compete with other activated
obtained by the simple oxidation process could compete with other activated carbons, having a range
carbons, having a range of uptake capacity of 150–300 mg/ g adsorbent [11,12].
of uptake capacity of 150–300 mg/g adsorbent [11,12].

400
MNZ adsorbed in solid / mg g-1carbon

300
CSC-OX

200

100 CSC

0
0 20 40 60
MNZ concentration in liquid / mg L-1
Figure 6. Adsorption isotherm of metronidazole on CSC and CSC-OX (data: square and triangle;
Figure 7. Adsorption isotherm of metronidazole on CSC and CSC-OX (data: square and triangle;
Langmuir-fittedcurve—full
Langmuir-fitted curve—fullline;
line;Freundlich-fitted
Freundlich-fittedcurve—dash
curve—dashline).
line).

Thedistribution
The distributionofofmolecular
molecularadsorption
adsorptionbetween betweenthe theliquid
liquidphase
phaseand andthethesolid
solidphase
phasecould
couldbebe
describedwith
described withanan adsorption
adsorption isotherm
isotherm whenwhen it isitinisthe
in equilibrium
the equilibrium condition.
condition. The parameter
The parameter that
that was
was obtained through the isotherm model is important information
obtained through the isotherm model is important information in the sorption mechanism; its surface in the sorption mechanism; its
surface characteristics and also showed adsorbent affinity [26].
characteristics and also showed adsorbent affinity [26]. The constants values of the metronidazole The constants values of the
metronidazole
adsorption modelsadsorption
of Langmuirmodelsand of Langmuir
Freundlichand areFreundlich
shown in are shown
Table 2. Thein Table 2. Themodels
Langmuir Langmuir is
models is more appropriate for metronidazole adsorption models
more appropriate for metronidazole adsorption models because its R is closer to 1. For Langmuir because
2 its R 2 is closer to 1. For

Langmuir
model, the model,
value ofthe thevalue
maximum of the adsorption
maximum adsorption capacity (Qcapacity (Qm) for CSC-OX showcases the
m ) for CSC-OX showcases the 400 mg/g
400 mg/g adsorbent
adsorbent which is more whichthan is more thanthan
3-times 3-times thatthan that (ca.
of CSC of CSC 115(ca.
mg/g115 adsorbent).
mg/g adsorbent). However,
However, the
the value of Langmuir constant of K for oxidized carbon is 25% lower
value of Langmuir constant of KL for oxidized carbon is 25% lower with respect to pristine carbon.
L with respect to pristine carbon.
Forthis
For thisphenomenon,
phenomenon,ititisislikely likelythatthatan anincrease
increaseininuptake uptakecapacity
capacityofofmetronidazole
metronidazoleseems seemstotobebe
more related to the change in pore textural property rather than surface
more related to the change in pore textural property rather than surface chemistry. An increase in chemistry. An increase in the
amount
the amount of of
oxygen
oxygen functional
functional groups
groupsininCSC-OXCSC-OXtends tendsto to decrease
decrease affinity
affinity of of metronidazole
metronidazoleonto onto
surfaceofofpores
surface poresasas cancanbebe seen
seen from fromthethe decrease
decrease of value
of value of Langmuir
of Langmuir constant
constant of KLof(from
KL (from 0.12
0.12 to 0.09to
0.09 L/mol). This negative effects of oxygen functional groups
L/mol). This negative effects of oxygen functional groups to adsorption of organic compounds areto adsorption of organic compounds
are observed
also also observedfromfrom literatures
literatures [14,27].[14,27]. But alternation
But alternation of porosity
of porosity in oxidized
in oxidized carbon carbon (see Table
(see Table 1 and1
and SEM
SEM image) image)
likelylikely
induces induces
moremore occupation
occupation of pore of pore
surfacessurfaces as indicated
as indicated by the byincrease
the increase
of theofQthe
m
Q m value. Freundlich fitted parameters also agree with the findings.
value. Freundlich fitted parameters also agree with the findings. Freundlich constant of KF increases Freundlich constant of KF
increases
for oxidized forcarbon
oxidized butcarbon
the value butofthe value of factor
correction correction factor of nfrom
of n decreases decreases
2.00 tofrom1.64.2.00 to 1.64.
Figure7 6displays
Figure displays thethe Langmuir
Langmuir and and Freundlich
Freundlich fitted
fitted curvescurves of CSC
of CSC and CSC-OX
and CSC-OX suggesting
suggesting that
that Langmuir model is more fitted to the equilibrium data.
Langmuir model is more fitted to the equilibrium data. Because the Langmuir model is compatible Because the Langmuir model is
compatible
(indicated by(indicated
value of Rby value
2 ), this of R2that
shows ), this
the shows that the
adsorption adsorption
process happenedprocessin ahappened in a specific
specific homogenous
homogenous area to the adsorbent that is in the interface layer between
area to the adsorbent that is in the interface layer between adsorbate and adsorbent or called monolayer adsorbate and adsorbent or
called monolayer
adsorption [19]. adsorption [19].
Processes 2019, 7, 835 8 of 9

Table 2. Constant value of various fitting model of metronidazole adsorption.

Langmuir Model Freundlich Model

No. Material Qm, KF

KL ,
mg Adsorbate/g R2 mg Adsorbate/(g n R2
L/mol
Adsorbent Adsorbent (mol/L)n )
1. CSC 0.12 114.94 0.90 17.99 2.00 0.64
2. CSC-OX 0.09 400.00 0.95 39.03 1.64 0.85

4. Conclusions
Modification of nanoporous carbon was conducted using mild oxidation of hydrogen peroxide.
The results showed that the oxidation process alters the porosity of nanoporous carbon and amount of
oxygen functional group. After oxidation, a higher amount of oxygen functional groups of porous
carbon is seen and only a little decrease in the specific surface area, but more porosity is exhibited.
In the application, oxidized carbon showed remarkably higher metronidazole uptake capacity. The
increase in the adsorption capacity seems to be more related to the change in pore textural property
rather than amount of oxygen functional group.

Author Contributions: Conceptualization, T.A. and I.P.; methodology, T.A. and I.P.; formal analysis, R.A.G.S. and
Y.M.N.A.; investigation, R.A.G.S. and Y.M.N.A.; resources, T.A. and I.P.; writing—original draft preparation, T.A.;
writing—review and editing, T.A. and I.P.; visualization, W.T.L.; supervision, T.A. and I.P; funding acquisition,
T.A.
Funding: This research and APC were funded by program of Rekognisi Tugas Akhir (RTA), Universitas Gadjah
Mada, grant number 2129/UN1/DITLIT/DIT-LIT/LT/2019.
Acknowledgments: The authors thank the PT Home System Indonesia for providing the porous carbon synthesized
from coconut shell as a gift.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Gadipelly, C.; Pérez-González, A.; Yadav, G.D.; Ortiz, I.; Ibáñez, R.; Rathod, V.K.; Marathe, K.V. Pharmaceutical
industry wastewater: Review of the technologies for water treatment and reuse. Ind. Eng. Chem. Res. 2014,
53, 11571–11592. [CrossRef]
2. Kidd, K.A.; Blanchfield, P.J.; Mills, K.H.; Palace, V.P.; Evans, R.E.; Lazorchak, J.M.; Flick, R.W. Collapse of
a fish population after exposure to a synthetic estrogen. Proc. Natl. Acad. Sci. USA 2007, 104, 8897–8901.
[CrossRef] [PubMed]
3. Escher, B.I.; Baumgartner, R.; Koller, M.; Treyer, K.; Lienert, J.; McArdell, C.S. Environmental toxicology
and risk assessment of pharmaceuticals from hospital wastewater. Water Res. 2011, 45, 75–92. [CrossRef]
[PubMed]
4. Yakubu, O.H. Pharmaceutical wastewater effluent-source of contaminants of emerging concern: Phytotoxicity
of metronidazole to soybean (Glycine max). Toxics 2017, 5, 10. [CrossRef] [PubMed]
5. Munoz, M.; Garcia-Muñoz, P.; Pliego, G.; Pedro, Z.M.D.; Zazo, J.A.; Casas, J.A.; Rodriguez, J.J. Application of
intensified Fenton oxidation to the treatment of hospital wastewater: Kinetics, ecotoxicity and disinfection.
J. Environ. Chem. Eng. 2016, 4, 4107–4112. [CrossRef]
6. Goolsby, T.A.; Jakeman, B.; Gaynes, R.P. Clinical relevance of metronidazole and peripheral neuropathy:
A systematic review of the literature. Int. J. Antimicrob. Agents 2018, 51, 319–325. [CrossRef] [PubMed]
7. Lanzky, P.F.; Halting-Sørensen, B. The toxic effect of the antibiotic metronidazole on aquatic organisms.
Chemosphere 1997, 35, 2553–2561. [CrossRef]
8. Prasetyo, I.; Rochmadi, R.; Wahyono, E.; Ariyanto, T. Controlling synthesis of polymer-derived carbon
molecular sieve and its performance for CO2 /CH4 separation. Eng. J. 2017, 21, 83–94. [CrossRef]
9. Amelia, S.; Sediawan, W.B.; Prasetyo, I.; Munoz, M.; Ariyanto, T. Role of the pore structure of Fe/C catalysts
on heterogeneous Fenton oxidation. J. Environ. Chem. Eng. 2019, 102921. [CrossRef]
Processes 2019, 7, 835 9 of 9

10. Ariyanto, T.; Kurniasari, M.; Laksmana, W.T.; Prasetyo, I. Pore size control of polymer-derived carbon
adsorbent and its application for dye removal. Int. J. Environ. Sci. Technol. 2019, 16, 4631–4636. [CrossRef]
11. Carrales-Alvarado, D.H.; Ocampo-Pérez, R.; Leyva-Ramos, R.; Rivera-Utrilla, J. Removal of the antibiotic
metronidazole by adsorption on various carbon materials from aqueous phase. J. Colloid Interface Sci. 2014,
436, 276–285. [CrossRef] [PubMed]
12. Çalişkan, E.; Göktürk, S. Adsorption characteristics of sulfamethoxazole and metronidazole on activated
carbon. Sep. Sci. Technol. 2010, 45, 244–255. [CrossRef]
13. Ilomuanya, M.; Nashiru, B.; Ifudu, N.; Igwilo, C. Effect of pore size and morphology of activated charcoal
prepared from midribs of Elaeis guineensis on adsorption of poisons using metronidazole and Escherichia
coli O157:H7 as a case study. J. Microsc. Ultrastruct. 2017, 5, 32. [CrossRef] [PubMed]
14. Liu, G.; Li, X.; Campos, L.C. Role of the functional groups in the adsorption of bisphenol A onto activated
carbon: Thermal modification and mechanism. J. Water Supply Res. Technol. AQUA 2017, 66, 105–115.
[CrossRef]
15. Peng, Y.; Liu, H. Effects of oxidation by hydrogen peroxide on the structures of multiwalled carbon nanotubes.
Ind. Eng. Chem. Res. 2006, 45, 6483–6488. [CrossRef]
16. Rehman, A.; Park, M.; Park, S.-J. Current progress on the surface chemical modification of carbonaceous
materials. Coatings 2019, 9, 103. [CrossRef]
17. Moreno-Castilla, C.; Ferro-Garcia, M.A.; Joly, J.P.; Bautista-Toledo, I.; Carrasco-Marin, F.; Rivera-Utrilla, J.
Activated carbon surface modifications by nitric acid, hydrogen peroxide, and ammonium peroxydisulfate
treatments. Langmuir 1995, 11, 4386–4392. [CrossRef]
18. Ros, T.G.; Van Dillen, A.J.; Geus, J.W.; Koningsberger, D.C. Surface oxidation of carbon nanofibres. Chem. A
Eur. J. 2002, 8, 1151–1162. [CrossRef]
19. Do, D.D. Adsorption Analysis: Equilibria and Kinetics, 2nd ed.; Imperial College Press: London, UK, 1998; ISBN
1860941303.
20. Allwar, A. Characteristics of pore structures and surface chemistry of activated carbons by physisorption,
FTIR And Boehm methods. IOSR J. Appl. Chem. 2012, 2, 9–15. [CrossRef]
21. Saffaj, T.; Charrouf, M.; Abourrche, A.; Aboud, Y.; Bennamara, A.; Berrada, M. Spectrophotometric
determination of metronidazole and secnidazole in pharmaceutical preparations based on the formation of
dyes. Dyes Pigments 2006, 70, 259–262. [CrossRef]
22. Sarwendah, R.A.G. Oksidasi Karbon Berpori dari dari Tempurung Kelapa dengan Hidrogen Peroksida
serta Pengaruhnya untuk Adsorpsi Metronidazol. Bachelor’s Thesis, Universitas Gadjah Mada, Yogyakarta,
Indonesia, 2018.
23. Amimmal, Y.M.N. Oksidasi Karbon Berpori dari Kayu Mlanding dengan Hidrogen Peroksida serta
Pengaruhnya untuk Adsorpsi Metronidazol. Bachelor’s Thesis, Universitas Gadjah Mada, Yogyakarta,
Indonesia, 2018.
24. Thommes, M.; Kaneko, K.; Neimark, A.V.; Olivier, J.P.; Rodriguez-Reinoso, F.; Rouquerol, J.; Sing, K.S.W.
Physisorption of gases, with special reference to the evaluation of surface area and pore size distribution
(IUPAC Technical Report). Pure Appl. Chem. 2015, 87, 1051–1069. [CrossRef]
25. Hasse, B.; Glasel, J.; Kern, A.M.; Murzin, D.Y.; Etzold, B.J.M. Preparation of carbide-derived carbon supported
platinum catalysts. Catal. Today 2015, 249, 30–37. [CrossRef]
26. Li, Z.; Jia, Z.; Ni, T.; Li, S. Adsorption of methylene blue on natural cotton based flexible carbon fiber aerogels
activated by novel air-limited carbonization method. J. Mol. Liq. 2017, 242, 747–756. [CrossRef]
27. Ania, C.O.; Parra, J.B.; Pis, J.J. Influence of oxygen-containing functional groups on active carbon adsorption
of selected organic compounds. Fuel Process. Technol. 2002, 79, 265–271. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).