Graphene oxide nanocomposite magnetic microbeads for the remediation of positively charged aromatic compounds

Article abstract of DOI:10.1039/c9dt04605d

Integrating graphene as an inorganic nanostructure within a hydrogel matrix enables the creation of a unique hybrid composite combining the peculiar chemical and physical properties of graphene with the high porosity and stability of hydrogels as for example agarose gel. As a consequence, the resulting material forms a double-network system providing advantages deriving from both the components. In this study, we present the synthesis of novel magnetic porous agarose-based graphene oxide microbeads for the adsorption and separation of positively charged aromatic molecules. The hydrogel-based graphene oxide beads revealed an ultrafast adsorption kinetics for positively charged aromatic dyes. We tested this material for the purification of fluorescent-tagged biomolecules. In addition, reduced graphene oxide microbeads were decorated with palladium nanoparticles, showing a high catalytic activity towards the reduction of dyes by sodium borohydride. Our results show that magnetic agarose based graphene microbeads with enhanced physical-chemical properties can be used for several biochemical applications.

Full text of DOI:10.1039/c9dt04605d

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Graphene oxide nanocomposite magnetic
microbeads for the remediation of positively
charged aromatic compounds†
Cite this: Dalton Trans., 2020, 49,
3333
a
b
b
c
a
L. Minati, * G. Speranza, V. Micheli, M. Dalla Serra and M. Clamer
Integrating graphene as an inorganic nanostructure within a hydrogel matrix enables the creation of a
unique hybrid composite combining the peculiar chemical and physical properties of graphene with the
high porosity and stability of hydrogels as for example agarose gel. As a consequence, the resulting
material forms a double-network system providing advantages deriving from both the components. In
this study, we present the synthesis of novel magnetic porous agarose-based graphene oxide microbeads
for the adsorption and separation of positively charged aromatic molecules. The hydrogel-based gra-
phene oxide beads revealed an ultrafast adsorption kinetics for positively charged aromatic dyes. We
tested this material for the purification of fluorescent-tagged biomolecules. In addition, reduced gra-
phene oxide microbeads were decorated with palladium nanoparticles, showing a high catalytic activity
towards the reduction of dyes by sodium borohydride. Our results show that magnetic agarose based gra-
phene microbeads with enhanced physical–chemical properties can be used for several biochemical
applications.
Received 2nd December 2019,
Accepted 9th February 2020
DOI: 10.1039/c9dt04605d
rsc.li/dalton
1
0–12
reported in literature.
porous systems hampers the molecules flow into the inner
Graphene-based materials have been recently reported to have parts, causing a non-optimal contact between the analyte and
The high lateral dimension of these
1
. Introduction
1
–3
potential applications in environmental fields.
Numerous graphene oxide. A practical approach to overcome this
studies have demonstrated that graphene has excellent ability problem is the miniaturization of the system, with the gene-
4
,5
to bind planar aromatic molecules such as dyes and anti- ration of a micrometer-sized structure that strongly accelerates
6
biotics through a combination of electrostatic interaction, the analyte flow in the porous matrix, thus increasing the
hydrogen bonding and π–π stacking. For practical applications, adsorption kinetics. Here, we report the synthesis of hybrid
the single nanosheets must be integrated into
a
3D superparamagnetic hydrogel-based graphene oxide micro-
7
–9
structure
composed by multiple sheets assembled by π–π beads for the adsorption and separation of positively charged
stacking. The homogeneous functionalization of these struc- aromatic molecules. Hybrid porous microparticles of
tures can be particularly difficult in highly porous hydrophobic 40–60 μm in size were successfully synthesized and tested,
materials like reduced graphene aerogel or graphene hierarchi- showing unprecedented performance in term of surface reac-
cal structures. A partial solution to this problem involves the tivity. These beads can be further decorated with metal nano-
incorporation of graphene sheets into porous agarose or algi- particles for a wide range of applications such as material
nate matrix. This approach has the advantage of avoiding the chemistry, catalysis, sensing and biological separation.
formation of compact graphene multilayers, thereby increasing
the total available surface area. This method is efficient for the
separation of dyes and aromatic molecules, however only
macroscopic structures (e.g. mm to cm in size) have been 2. Materials and methods
2.1 Materials
Graphene oxide suspension was purchased from Graphenea
Inc. 20–40 nm NanoArc® spherical Maghemite superpara-
magnetic nanoparticles were purchased from AlfaAesar.
a
Immagina Biotechnology s.r.l., Via Sommarive 18, Trento, Italy.
E-mail: lminati@immaginabiotech.com
b
Fondazione Bruno Kessler, Via Sommarive 18, Trento, Italy
c
Agarose (M
w
∼ 120 000 Da) was purchased from Thermo
CNR-IBF, Via Sommarive 18, Trento, Italy
Fischer Scientific. Other chemical reagents and solvents were
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9dt04605d
purchased from Sigma Aldrich.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans., 2020, 49, 3333–3340 | 3333

View Article Online
Paper
.2 Synthesis of graphene oxide-doped magnetic agarose
Dalton Transactions
2
in fresh ethanol. The composite hydrogel beads were then
transferred from ethanol to chloroform by washing twice with
chloroform. MaGO bead were incubated with 1 mL of Nile
Blue solution (50 µg mL− ) in chloroform by vortexing for 10
seconds and mixing in heat block at 1000 rpm for 1 minute at
room temperature. The beads were separated with a magnetic
support and the supernatant was recovered for analysis. The
beads were then washed in 1 mL of ethanol. After a few
seconds, the alcohol solution returns to the original blue color
indicating a quantitative desorption of the dye from the beads.
beads (MaGO)
MaGO beads were synthesized by water-in-oil emulsion.
Briefly, 1 ml of a mixture of agarose, superparamagnetic nano-
particles and graphene oxide in water (80, 10 and 10% in mass
respectively, 2% final in agarose) pre-heated at 90 °C was
stirred for 5 minutes to obtain a homogeneous suspension.
The suspension was added dropwise to 3 ml of a mineral oil-
Span 80 solution (2% w/w) at 90 °C and mixed for 10 minutes.
The system was cooled down to room temperature and the
beads were washed several times with acetone to remove the
1
oil. Finally, the beads were suspended in water to a concen- 2.7 Purification of protein
tration of 10% v/v of gel.
Bovin Serum Albumin (BSA, 1 mg) and ATTO648N (10 µg) were
mixed together in 1 ml of PBS (phosphate buffered saline) for
minute. The protein–dye mixture was incubated with 1 mL of
The beads were cross-linked by using divinyl sulfone 2% v/v
in aqueous solution for 60 minutes. A sample obtained
without the addition of graphene oxide (MAG sample) was also
synthesized as a reference.
1
MaGO beads for 15 seconds vortexing, then the beads were dis-
carded and the supernatant placed in a new plastic vial. To
test the protein adsorption on MaGO beads, 1 ml of BSA sus-
pension was incubated with 1 ml of MaGO beads for 15
seconds vortexing, then the beads were discarded and the
supernatant placed in a new plastic vial. The procedure was
repeated five times. The amount of unbound BSA was quanti-
fied by Bradford assay.
2
.3 Synthesis of reduced graphene oxide-doped magnetic
agarose beads (MarGO)
1
0 ml of MaGO sample (10% v/v of gel) were incubated in
shaking water bath with 100 mg of hydroxylamine and the
temperature was raised to 80 °C for 4 hours. The reduced
MaGO beads (MarGO) were then purified by washing four
times with water.
2
.8 Catalytic tests
In a typical procedure, 0.9 mL of 4-NP, MB and CR (10− M)
were added to 100 µL sodium borohydride solution (NaBH ) in
water (3 mg mL− ). Then 30 µL of MarGO-Pd gel suspension
10% v/v gel) in water were quickly added to the dye solution
3
2
.4 Synthesis of palladium-doped reduced graphene oxide/
magnetic agarose beads (MarGO-Pd)
4
1
10 ml of MarGO beads (10% v/v of gel) were incubated in
10 ml of 2 mM solution of K PdCl in water for 2 hours at
1000 rpm. The beads were separated, the supernatant was
(
2
4
under mixing. The beads were separated and the supernatant
analyzed by UV-vis absorption spectroscopy at different times.
For all the reactions, the Pd concentration was 1% mol/mol
removed and the beads were washed five times with water. The
beads were then incubated in 1 mM sodium ascorbate solution
in water for 10 minutes, then washed five times with water
compared to the reagent, while NaBH was in 10 times molar
4
excess compared to the dye.
(MarGO-Pd beads).
2
.5 Dye adsorption measurements
2.9 Characterizations
MaGO (100 µL of gel suspension) were placed in 1 mL solution
The morphology of the MaGO beads was characterized by SEM
−
1
of dye at pH 7 at 50 mg L concentration. The beads were sus-
pended by vortexing for 10 seconds and mixing in heat block
at 1000 rpm for 1 minute. Beads were then separated and the
supernatant was recovered for analysis.
(
(
Superscan SSX-550, Shimadzu). Wide-angle X-ray diffraction
XRD) analyses were carried out on an X-ray diffractometer (D/
MAX-1200). TEM analysis was performed on
H-800 microscope at 200 kV.
a Hitachi
t
The adsorbed amount (Q ) was calculated according to fol-
lowing equation:
The specific surface area and pore size distribution of
freeze-dried MaGO beads were obtained by nitrogen adsorp-
tion/desorption measurements at 77 K using a Micromeritics,
ASAP 2010 instrument.
Qt ¼ ððC0 ꢀ CeÞÞ=mGO ꢁ V
where C
0 e
(mg L− ), and C (mg L−1) are the dye concentrations
1
XPS measurements were carried out using a Kratos Axis
DLD Ultra. Wide scans were acquired in the BE energy range
1200–0 eV using a 160 eV pass energy, while high resolution
core line spectra were performed setting the analyzer pass
energy at 20 eV and the energy step at 0.05 eV. Spectra were
before and after adsorption, respectively. V is the solution
volume (L) and mGO is the graphene oxide mass in the beads
mg). Dye concentrations were calculated from the UV-Vis
absorbance maximum using a calibration curve.
(
1
3
analyzed using a home-made software based on R platform.
Peak fitting was performed using linear background subtrac-
2
.6 Nile Blue adsorption in hydrophobic solvent
MaGO beads in 1 ml of water were separated and suspended tion and Gaussian components. Optical microscopy images
in ethanol for 10 minutes, then washed and suspended again were obtained with a Leica DM6000CD Microscope.
3334 | Dalton Trans., 2020, 49, 3333–3340
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
beads. MaGO and MarGO beads were crosslinked by using
divinyl sulfone.
3
. Results and discussion
3
.1 Synthesis of graphene-doped agarose beads
Morphological analysis revealed a highly porous structure
Aerogel and graphene hierarchical structures are composed by with the formation of an irregular network composed by gra-
multiple graphene sheets that significantly reduces the total phene oxide sheets and agarose hybrid matrix.
available reactive surface for analyte binding. In addition,
Nitrogen adsorption–desorption isotherms of the MaGO
small pores greatly slow down the diffusion of molecules in beads are shown in Fig. S1.† The specific surface area of the
2
−1
the inner areas of the monoliths, thus decreasing the overall beads was calculated to be 52.37 m g . The pore size distri-
penetration efficiency, a critical step in many applications bution revealed the presence of mesopores of about 30 nm in
such as the decoration of graphene with nanoparticles. To size.
overcome this problem, we prepared a hydrogel matrix for the
encapsulation of dispersed graphene oxide nanosheets. We
3.2 Test of dye adsorption by MaGO beads
used an agarose matrix in the form of microbeads with super- To better characterize the hybrid microparticles and to evalu-
paramagnetic core for better handling and separation of the ate the performance of MaGO we compared the adsorption
particles. Beads were obtained through a water-in-oil emulsion properties of selected dyes on MaGO and MarGO in relation to
in the presence of surfactant. Fig. 1a show the synthesis available published data. Adsorption tests with MaGO were
scheme of the MaGO beads.
performed in different conditions and using beads without GO
The particle size distribution calculated by optical (MAG sample) as controls for assessing the effect of graphene
microscopy (Fig. 1e) was between 40–60 micrometers. The in the process of dye adsorption. As shown in Fig. 2a and b,
presence of graphene oxide inside the beads was confirmed by more than 95% of Nile Blue (NB) and Rhodamine B (RhB)
the brown color of MaGO beads (Fig. 1f). After chemical were removed from the aqueous solution using MaGO sample
reduction, the particle color turns to black (Fig. 1g), attesting with an incubation time of 1 minute. The fast adsorption of
the formation of MarGO (Magnetic Reduced Graphene Oxide) the NB and RhB by MaGO beads (Fig. 2c) can be explained by
Fig. 1 (a) Synthesis pathway of MaGO and MarGO particles. (b) Scanning electron microscopy images of a MaGO particle (c and d) high magnifi-
cation images of a MaGO beads showing the porous structure. (e) Size distribution of the MaGO particles. (f) Photograph of a MaGO and (g) MarGO
particle.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans., 2020, 49, 3333–3340 | 3335

View Article Online
Paper
Dalton Transactions
Fig. 2 (a) UV-visible spectroscopy of NB solution and the supernatants after adsorption with MaGO and subsequent washing of the particles with 8
M urea for 1 minute. Inset: Image of a single bead incubated with NB. (b) UV-visible spectroscopy of RhB solution and the supernatants after adsorp-
tion with MaGO and subsequent washing of the particles with 8 M urea for 1 minute. Inset: Image of a single bead incubated with RhB. (c) MaGO
adsorption kinetic profiles for NB (blue squares) and RhB (red circles). (d) Values of adsorbed amounts for different dyes by MaGO (dark grey) and
MarGO (light gray). (e) UV Vis of BSA and ATTO647N dye (red line) incubated with MaGO beads followed by vortexing for 15 seconds (magenta line)
and 30 seconds (blue line). (f) Normalized Bradford quantification of BSA suspension incubated with fresh MaGO beads for 15 seconds for 5 cycles.
A minimum of 3 independent measurements were performed.
(i) the high affinity of the GO for aromatic dyes, (ii) the fast process more likely took place at specific homogeneous sites
diffusion of the dye inside the porous agarose matrix and (iii) within the adsorbents, (ii) the adsorption sites are identical
the compact size of the beads that greatly enhance the and (iii) the dyes are adsorbed as monolayer on the surface of
diffusion speed of the dye inside the gel.
GO.
The slightly large K value of the GO respect to the MaGO
The best fit for our adsorption kinetics data is represented
L
by a pseudo-second order model (eqn (1)) indicating a strong suggests that the graphene oxide suspension would be more
interaction between the dyes and MaGO beads (Fig. S2† and effective adsorbents for MB and RhB. We can explain this con-
Table 1).
sidering the porous structure of MaGO and the presence of the
magnetic nanoparticles that can slightly reduce the surface
ð1Þ area of GO available for adsorption.
t
1
1
¼
þ
kqe2 qe
t
qt
To test whether the dye adsorption properties are depen-
dent on the analyte molecular charge, we used different pH
conditions. The interaction between the dye and agarose gel in
MAG beads is very weak and the NB can be quantitatively
removed from the control beads by washing with an acidic pH
water buffer (Fig. S3 and S4†). The decrease in pH induces the
protonation of the amino groups of the dye, increasing its
water solubility and encouraging the diffusion from the
agarose matrix to the water solution.
e
The kinetic parameters are reported in Table 1. The q cal-
culated using the pseudo second order kinetic equation are
very close to the experimental data (Fig. 2d). The results indi-
cates that the adsorption kinetics is well represented by the
2
pseudo second order kinetics (R ∼ 1).
Langmuir adsorption isotherms fit for NB and RhB are
reported in Fig. S2† while the kinetic parameters are reported
2
in Table S1.† The values of the correlation coefficient R for
On the contrary, the interaction between the dye and gra-
phene in MaGO beads is quite insensitive to pH variations.
This result confirms that the dye is adsorbed on graphene
oxide surfaces dispersed in the agarose gel as a consequence
of the strong π–π interaction between the aromatic portion of
the Langmuir equation were 0.997 for NB and 0.993 for RhB,
indicating that the model correlated well with the experi-
mental data. This result indicates that (i) the adsorption
1
4
Table 1 Kinetic parameters for NB and RhB adsorption on MaGO
the molecules and graphene conjugated rings.
MaGO beads were also tested using other positively charged
aromatic dyes, such as Methylene Blue (MB), Neutral Red (NR)
and Crystal Violet (CV) (Fig. S5†). The difference in the adsorp-
2
Sample
k
q
e
R
NB
RhB
0.00232
0.00959
580.27
327.35
0.998
0.997 tion capacity between the dyes could be attributed to the
3336 | Dalton Trans., 2020, 49, 3333–3340
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
different chemical properties including aromaticity, number of thanks to their unique superparamagnetic graphene–agarose
positive charges and planarity. In particular, NB that is the composition.
most hydrophobic dye in the list have the highest adsorption
To further characterize the efficiency of the beads, we com-
capacity. On the contrary, RhB is not planar, it contains a pared the adsorption performance of a solution of MB
negative charge and have the lower adsorption capacity. In between micrometric MaGO and 1 mm diameter agarose–gra-
addition, the adsorption capacity of the dyes on MarGO beads phene beads with the same chemical composition of MaGO
were generally lower respect to the MaGO. This indicates that (B-MaGO 1 mm). The B-MaGO 1 mm beads were synthesized
the reduction of the GO and the consequent decrease of the by dropping the agarose-magnetic nanoparticles-GO solution
negative charge density affects the dye adsorption capacity on used for the synthesis of the MaGO beads in a cold mineral oil
MarGO.
containing Span 80 as surfactant. The beads were washed in
Finally, we showed that MaGO beads can efficiently capture acetone and redispersed in water after purification. As reported
amphoteric dyes directly dissolved in strong apolar solvents. in Fig. S7,† MaGO beads show a much faster adsorption kine-
NB can be efficiently dissolved in solvents like n-hexadecane tics compared to the B-MaGO 1 mm beads. Because the poro-
and chloroform. After incubation of the MaGO beads with a sity of the beads is the same between the two formulations, we
0 mg L− NB solution in chloroform, a strong decrease of the conclude that the smaller size of the MaGO beads is the only
1
5
UV-vis adsorption of the solution was observed (Fig. S6†), responsible for their fast adsorption kinetics. In conclusion,
while the composite hydrogel beads became uniformly the main advantages of MaGO beads compared to the other
colored, confirming the efficient removal of NB molecules systems presented in the literature stem from the much higher
from the solvent.
adsorption kinetics. This is a consequence of the small MaGO
The adsorbed amount of MaGO beads with apolar solvents size that promotes the penetration of the dyes inside the
−
1
resulted in a value of 956 mg g , higher than those obtained porous structure, thus enabling a selective removal of cationic
for NB in water. Additionally, we observed a NB desorption of dyes from a multicomponent protein-containing solution that
about 90% after only 1 min of incubation of the beads in could not be achieved with less efficient systems.
ethanol (Fig. S6†).
3
.3 Metal-doped agarose/graphene beads for catalysis
The desorption efficiency reaches ca. 98% after the fourth
cycle of incubation of the composite hydrogel beads in fresh The deposition of metal nanoparticles on graphene is impor-
ethanol.
1
5,16
tant in catalysis.
It is well-known that the graphene area
To test possible biochemical application of MaGO, we for metal deposition and metal salt reduction on graphene
1
7
used the beads to separate dyes from a labeled protein–dye oxide are critical limiting factors. To overcome this con-
mixture. Protein labelling with one or more fluorescent tags straint, we investigated the possibility of further implementing
is an essential tool used in modern laboratories to investigate MaGO growing metal nanoparticles (i.e. palladium) on gra-
biological processes by in vitro and in vivo imaging. A phene sheets (Fig. 3a). Graphene oxide was chemically reduced
demanding problem in the biotech sector is the separation of in situ to form a reduced graphene oxide–agarose hybrid. The
excess dye from the conjugated protein. At the end of the reduction of graphene oxide inside the MaGO beads was
reaction, the fluorescent tags must be purified from the carried out using hydroxylamine as a reducing agent. We
unreacted fluorescent dye. Normally this task is accom- monitored the reaction observing the color change of the
plished using time consuming techniques like repeated ultra- beads from dark brown to pale black (Fig. 1f and g), an empiri-
centrifugations or dialysis. A protein routinely used in many cal evidence of the reduction process. The reaction efficiency
biochemical experiments (BSA) and a common fluorescent was confirmed by X-ray photoelectron spectroscopy (XPS,
tag for protein conjugation (ATTO647N) were mixed together Fig. 3a and S8†). The features at about 284.4, 286.6 and 288.8
2
then MaGO beads were added to the suspension followed by eV of the GO C 1s core line were assigned to the sp -C, C–O
1
8
vortexing (15 seconds) and magnetic separation to remove the and COOH bonds, respectively. The C 1s core line of reduced
supernatant. The kinetics of dye depletion was followed by graphene oxide revealed that the features at about 286.6 and
UV-vis (Fig. 2E). After only 15 seconds of incubation with 288.8 eV strongly decrease, indicating the elimination of
2
MaGO, the UV-Vis spectra revealed a strong decrease of the oxygen with the formation of sp -rich graphene. The reduction
2
+
ATTO647N adsorption measured at 650 nm, while only a of the Pd was investigated by following the UV-Vis spectrum
2
+
minimal decrease of the BSA adsorption band (measured at modification of the Pd solution incubated with the MarGO
80 nm) was observed. After a second cycle of incubation beads.
with MaGO, the signal at 650 nm disappeared, indicating The intensity of the absorption peak of Pd at about
2
2
+
complete removal of the dye. We quantified the protein loss 410 nm strongly decreased after only 30 minutes of incubation
through multiple cycles of incubation with MaGO beads by with MarGO beads (Fig. 3b), suggesting a strong adsorption of
colorimetric Bradford assay (Fig. 2F), observing only a palladium on the graphene surface. The formation of Pd nano-
minimal loss after four cycles of adsorption, probably due to particles is forced by the difference in the reduction potential
some non-specific adsorption or the dead volume of the between reduced graphene oxide (+0.25 V vs. SHE, Standard
beads. In conclusion, MaGO beads can be used to efficiently hydrogen electrode) and the transition metal salt (+0.6 V vs.
2
+
14
2+
purify dyes from a solution or from a protein–dye mixture SHE for Pd ). Therefore, the rGO/Pd system can undergo
This journal is © The Royal Society of Chemistry 2020
Dalton Trans., 2020, 49, 3333–3340 | 3337

View Article Online
Paper
Dalton Transactions
2+
2
Fig. 3 (a) XPS C 1s core line of GO reduced by NH OH at 80 °C for 1 and 4 h. (b) UV-Vis spectra of Pd before and after incubation with MarGO
beads. (c) XPS analysis of MarGO-Pd, Pd 3d core line. (d) XRD pattern of MarGO-Pd showing the (200) reflection. Inset: Transmission electron
microscopy image of Pd nanoparticles on reduced graphene oxide.
spontaneous reduction to form metal nanoparticles on the gra- bated with NaBH
4
reported in Fig. 4a (red line) showed an
phene surface. intense absorption peak at 390 nm. After incubation of the
The chemical state of the palladium was investigated by 4-NP with MarGO-Pd, the peak intensity decreased and finally
XPS (Fig. 3c). The Pd 3d region showed the presence of a spin– died out, indicating reduction of the nitro group. In our experi-
orbit doublet whose components at 335.5 eV and 341 eV are mental conditions, the 4-NP was completely degraded in about
1
9
4
assigned to palladium in the metallic state. The spectrum 4 minutes. 4-NP solutions incubated with NaBH alone are
does not show any feature that can be assigned to oxidized pal- stable for more than 24 h, confirming that the palladium
ladium, confirming the complete reduction of the palladium nanoparticles present on the graphene surface are the sites
precursor. The Pd content in the MarGO-Pd composite was where the reduction occurs. To investigate the stability and
found to be 4.75% as estimated by the XPS quantitative ana- recyclability of the catalyst, we used MarGO-Pd beads with
lysis. XRD analysis of MarGO-Pd beads reported a diffraction repeated cycles of reduction of 4-NP. After each reaction, the
peak at 69°, attributed to the (200) plane of crystalline palla- catalyst was removed, MarGO-Pd where washed with water and
20
dium (Fig. 3d). The mean nanoparticle diameter calculated dried for the next cycle of catalysis. We observed that
by the Scherrer equation was 2.3 nm. This value is in agree- MarGO-Pd beads did not show a significant decrease of the
ment to that calculated by TEM analysis. The image shows conversion efficiency even after five reaction cycles (Fig. 4b).
that the palladium nanoparticles have a spherical shape with a
mean diameter of about 3 nm.
The UV-vis spectra of MB and CR solutions incubated with
NaBH4 (Fig. S10†) showed an absorption peak that did not
change for up to two hours. The presence of MarGO-Pd in the
samples drastically changed the reaction speed. In fact, after
3
.4 Catalytic test of MarGO-Pd beads
4
-Nithophenol (4-NP), Methylene Blue (MB) and Congo Red few seconds from the addition of the particles the solution
(
CR) were selected as model molecules for testing the catalytic became clear, as confirmed by the complete disappearance of
properties of the MarGO-Pd beads. The reduction pathway the absorption peak of MB at 640 nm (Fig. 4c, ESI video†). The
2
1
includes a nitro-to-amino conversion for 4-NP, a hydrogen- catalytic process of dye degradation can be explained by an
2
2,23
−
ation of the conjugate xanthene ring for MB
and a C–N electrochemical mechanism. At the beginning, the BH
release for CR (Fig. S9†). 4-NP is a and the dyes are adsorbed together onto the surface of palla-
hazard intermediate involved in paracetamol synthesis and is dium nanoparticles. Subsequently, a transfer of electrons from
4
ions
2
4
bond breaking with N
2
−
widely used to test the catalytic activity thanks to the possi- BH
4
ions takes place and MB is then reduced. During this
bility to follow the reduction of the nitro group to amino group process, the structure of conjugated π–π xanthene ring is
2
2
trough spectroscopic analysis. UV-visible spectra of 4-NP incu- broken to form a colorless product.
Another example
3338 | Dalton Trans., 2020, 49, 3333–3340
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Fig. 4 (a) UV-Vis spectra of 4-NP reduction before (red line) and 5 minutes after (black line) incubation with MarGO-Pd beads. (b) Conversion of
-NP in five successive cycles of reduction with MarGO-Pd. A minimum of 3 independent measurements were performed (c) UV-Vis spectra of MB
reduction before (black line) and 15 seconds after (red line) incubation with MarGO-Pd beads. (d) UV-Vis spectra of CR reduction before (black line)
0 seconds after (red line) and 120 seconds after (blue line) incubation with MarGO-Pd beads.
4
6
2
5–27
showing the effectiveness of MarGO-Pd as a catalyst is the among the most active catalysts.
All these results confirm
decomposition of CR by NaBH . In absence of an external cata- that the degradation of the dyes is mediated by the palladium
4
lyst, this reaction is extremely slow and the color of the solu- nanoparticles on the graphene surface.
tion remains unaltered for several days. We monitored the
degradation process of CR with or without MarGO-Pd beads,
observing the change of the maximum absorbance peak of the
dye as a function of time. Upon the addition of MarGO-Pd, the 4. Conclusion
absorption peak intensity of CR displayed a gradual decrease
We provided evidence that graphene-based agarose micro-
without any change in shape and position (Fig. 4d), with com-
beads can be used to selectively separate aromatic dyes from
plete degradation after about 2 minutes. To further assess the
complex mixtures. The fast adsorption of positively charged
catalytic activity of the MarGO-Pd beads, the turnover fre-
aromatic dyes on MaGO beads can be exploited for fast purifi-
quency (TOF, calculated as moles of molecules reacted over
cation of fluorescent-tagged protein or biomolecules. By taking
moles of catalyst × time) for each reaction was calculated and
advantage of the unique chemical and physical properties of
reported in Table 2.
the reduced graphene oxide surface, we demonstrated the
The efficiency of the reduction of 4-NP, MB and CR com-
possibility to grow small palladium nanoparticles on these
pared to those reported in the literature ranks MarGO-Pd
beads for possible applications as fast, reusable and efficient
products in catalysis. MarGO-Pd particles are recommended as
catalysts in the reduction/degradation reactions of aromatic
Table 2 Reaction times and TOF values of MarGO-Pd for each reaction
dyes for the following reasons: (i) the very high adsorption
−1
Dye
Reaction time (min)
TOF (molDYE molPd min−1
)
kinetics allow a uniform decoration of the rGO by Pd nano-
particles (ii) the porous structure allow the reagents to concen-
trate within the active sites, as well as the product to leave the
catalytic sites. (iii) The great affinity of graphene for the aro-
4
CR
MB
-NP
4.1
2.5
0.9
8.4
16.6
14.1
This journal is © The Royal Society of Chemistry 2020
Dalton Trans., 2020, 49, 3333–3340 | 3339

View Article Online
Paper
Dalton Transactions
matic dyes allows to locally concentrate the reagent around the
metal nanoparticle, greatly increasing the reaction kinetics.
9 L. Jiang and Z. Fan, Nanoscale, 2014, 6, 1922–1945.
10 C. Cheng, Y. Cai, G. Guan, L. Yeo and D. Wang, Angew.
Chem., Int. Ed., 2018, 57, 11177–11181.
11 C. Cheng and D. Wang, Angew. Chem., Int. Ed., 2016, 55,
6853–6857.
Conflicts of interest
1
2 S. Lee, B. J. Moon, H. J. Lee, S. Bae, T.-W. Kim, Y. C. Jung,
et al., ACS Appl. Mater. Interfaces, 2018, 10, 17335–17344.
3 G. Speranza and R. Canteri, SoftwareX, 2019, 10, 100282.
4 T. Sun, Z. Zhang, J. Xiao, C. Chen, F. Xiao, S. Wang, et al.,
Sci. Rep., 2013, 3, 2527.
5 S. Yang, J. Dong, Z. Yao, C. Shen, X. Shi, Y. Tian, et al., Sci.
Rep., 2014, 4, 4501.
6 G. Moon, H. Kim, Y. Shin and W. Choi, RSC Adv., 2012, 2,
Massimiliano Clamer is an employer and a shareholder in
Immagina Biotechnology s.r.l., a company engaged in the
development of new technologies for gene expression analysis,
including smart beads. Luca Minati is an employer of
Immagina Biotechnology s.r.l.
1
1
1
1
1
Acknowledgements
2205–2207.
7 H. J. Yin, H. J. Tang, D. Wang, Y. Gao and Z. Y. Tang, ACS
Nano, 2012, 6, 8288–8297.
8 https://srdata.nist.gov/xps/.
9 L. Minati, K. F. Aguey-Zinsou, V. Micheli and G. Speranza,
Dalton Trans., 2018, 47, 14573–14579.
We thank the Laboratory of Biomolecular Sequence and
Structure Analysis for Health (LaBSSAH) for the technical
support.
1
1
This work was partially financed by CARITRO Foundation.
20 T. Teranishi and M. Miyake, Chem. Mater., 1998, 10, 594–
600.
References
2
1 P. Veerakumar, R. Madhu, S.-M. Chen, V. Veeramani,
C.-T. Hung, P.-H. Tang, et al., J. Mater. Chem. A, 2014, 2,
16015–16022.
1
2
3
4
5
6
7
8
K. C. Kemp, H. Seema, M. Saleh, N. H. Le, K. Mahesh,
V. Chandra and K. S. Kim, Nanoscale, 2013, 5, 3149–3171.
F. Perreault, A. Fonseca de Faria and M. Elimelech, Chem. 22 S. Li, H. Li, J. Liu, H. Zhang, Y. Yang, Z. Yang and B. Wang,
Soc. Rev., 2015, 44, 5861–5896. Dalton Trans., 2015, 44, 9193–9199.
Y. Shen, Q. Fang and B. Chen, Environ. Sci. Technol., 2015, 23 Z. Cheng, J. Liao, B. He, F. Zhang, F. Zhang, X. Huang,
9, 67–84. et al., ACS Sustainable Chem. Eng., 2015, 3, 1677–1685.
S. Tang, D. Xia, Y. Yao, T. Chen, J. Sun, Y. Yin, W. Shen and 24 T. N. J. I. Edison, R. Atchudan, M. G. Sethuraman and
Y. Peng, J. Colloid Interface Sci., 2019, 554, 682–691. Y. R. Lee, J. Photochem. Photobiol., B, 2016, 162, 604–610.
T. A. Tabish, F. A. Memon, D. E. Gomez, D. W. Horsell and 25 G. Fu, L. Tao, M. Zhang, Y. Chen, Y. Tang, J. Lin and T. Lu,
S. Zhang, Sci. Rep., 2018, 8, 1817. Nanoscale, 2013, 5, 8007.
F. Yu, Y. Li, S. Han and J. Ma, Chemosphere, 2016, 153, 365– 26 F. Chen, A. S. Gong, M. Zhu, G. Chen, S. D. Lacey, F. Jiang,
4
3
85.
Y. Li, Y. Wang, J. Dai, Y. Yao, J. Song, B. Liu, K. Fu, S. Das
and L. Hu, ACS Nano, 2017, 11, 4275–4282.
27 T. Kim, X. Fu, D. Warther and M. J. Sailor, ACS Nano, 2017,
11, 2773–2784.
X. Wang, L. L. Lu, Z. L. Yu, X. W. Xu, Y. R. Zheng, et al.,
Angew. Chem., 2015, 127, 2427–2431.
J. Zou and F. Kim, Nat. Commun., 2014, 5, 5254.
3340 | Dalton Trans., 2020, 49, 3333–3340
This journal is © The Royal Society of Chemistry 2020