Communications
Table 1: Catalytic efficiency of the MWCNT/Ag hybrid in comparison to
the control systems.
of 2-methylhydroquinone to 2-methylbenzoquinone. We
initially assessed the amount of Ag NPs on the nanohybrids
by using TGA in air. We measured a quantity of Ag NPs that
corresponded to 3% of the total amount of the nanohybrids
(see Figure S10 in the Supporting Information). This value
was used for the calculation of the amount of NPs added to
the catalytic reaction. To obtain the reference product, we
began with the oxidation of 2-methylhydroquinone in meth-
anol with 35% aqueous H2O2 in the presence of commercially
available Ag2O NPs, as previously reported.[35] The evolution
of the model reagent 2-methylhydroquinone to 2-methylben-
zoquinone could be easily followed by HPLC analysis. Since
most of the NPs on the nanotubes were constituted of metallic
silver, we also evaluated the efficacy of commercially
available Ag0 NPs of sizes below 100 nm. Indeed, both
silver oxide and metallic silver catalyzed the transformation
of the hydroquinone, although not to completion (see the
Supporting Information).
These control experiments allowed to rationalize the
heterogeneous catalytic reaction with our MWCNT/Ag NP
hybrids. The oxidation of 2-methylhydroquinone was carried
out in methanol in the presence of 35% aqueous H2O2 and
0.03 equivalents of NPs on Ad-MWCNT 1a. The resulting
reaction mixture was sonicated at room temperature to
maintain a good dispersion of the nanotubes at all stages of
the catalytic cycle. After 4 h, we observed the complete
disappearance of the starting material and the formation of
the desired 2-methylbenzoquinone (see Figure 3B, and Fig-
ure S11 in the Supporting Information). The catalytic oxida-
tion was completely selective, as we did not observe the
production of any substituted biaryl because of the compet-
itive oxidative homocoupling of 2-methylhydroquinone.[35,36]
At the end of the reaction, we easily recovered the nanotubes
by simple centrifugation and without any further purification.
The collected hybrids were reused under the same reaction
conditions for two additional cycles to give similar excellent
conversion (Figure 3B). After the third cycle, the NPs were
still present as assessed by TEM (data not shown). We also
confirmed the oxidation state of Ag NPs after the three
catalytic cycles by using XPS. We did not observe significant
variations on the composition of the nanoparticles (see
Figure S12 in the Supporting Information). In addition, as
an alternative control to assess the key role of adenine as a
ligand for the NP coordination, we prepared a hybrid by using
the nanotubes with only the ammonium groups (f-MWCNT 9,
precursors of Ad-MWCNT 1a; see Scheme S2 in the Sup-
porting Information). In this case, we could generate Ag NPs
on CNTs with a wider diameter distribution (see Figure S13 in
the Supporting Information). We then tested the catalytic
capacity of this complex on the oxidation of 2-methylhydro-
quinone without obtaining the desired product (see Fig-
ure S14 in the Supporting Information). After obtaining this
result, we observed the complex by TEM (results not shown),
and surprisingly, we did not find any NP on the grid, thus
confirming the important role of adenine in comparison to
primary amines for stabilizing Ag NPs on the tubes. To
compare the catalytic efficacy of the MWCNT/Ag nano-
hybrid with silver oxide and silver nanoparticles, we calcu-
lated the turnover number (TON) and turnover frequency
Catalyst
Turnover number
Turnover frequency
silver(I) oxide
silver(0) NPs
Ag/MWCNT 1a
10.3
26.3
33.3
2.6 hÀ1
6.6 hÀ1
8.3 hÀ1
(TOF) for all three types of catalysts. Table 1 shows the data
for the different systems.
The catalyst based on CNTs is clearly more efficient in the
conversion of the hydroquinone to benzoquinone than Ag2O
and Ag NPs alone. As shown in Table 1, the heterogeneous
MWCNT/Ag catalysts feature the characteristics that are
necessary for the reactions to run at higher rate in comparison
to the silver oxide or metal nanoparticles. Moreover, the
hybrid material can be easily recovered by simple filtration
and can thus be immediately recycled. CNTs can be
considered an efficient catalyst support because of their
high specific surface area and chemical inertness. Our
synthetic protocol proves that CNTs might become pliant
materials by tuning their functionalization to promote the
assembly of complex systems; this method is useful for the
controlled and cost-effective design of new selective catalysts
and for the production of advanced devices based on CNTs.
In conclusion, we have demonstrated the possibility to
shape CNT functionalization in order to induce catalytic
nanoparticle deposition. Adenine that is covalently bound to
the CNT surface is able to trigger the formation of catalytic
NPs with controlled sizes. This result provides the possibility
for the tailored design of novel catalytic materials based on
CNTs with full exploitation of their properties. Moreover, we
have offered a proof-of-concept protocol by using this novel
material as heterogeneous catalyst to promote the oxidation
of hydroquinones, and by proving the catalytic efficiency of
the nanohybrids on this model reaction and the complete
recycling without loss of activity over multiple cycles.
Experimental Section
MWCNTs, produced by the catalytic carbon vapor deposition
(CCVD) process, were purchased as purified from Nanocyl (thin
MWCNT > 95% C purity, Nanocyl 3100 batch no. 071119), average
diameter and length: 9.5 nm and 1.5 mm, respectively). Details of the
synthesis of the ox-MWCNTs and the three Ad-MWCNTs 1a, 1b,
and 1c are reported in the Supporting Information.
Preparation of Ag NPs on Ad-MWCNTs 1a, 1b, and 1c: 0.1 mg
of the Ad-MWCNTs 1a, 1b, 1c, ox-MWCNT, and f-MWCNT 9 (used
as controls) were dispersed in methanol/water (1:1, 1 mL) by
sonication for 5 min. A solution of AgNO3 (1 or 5 equiv with respect
to the adenine loading) in methanol/water (1:1, 1 mL) was prepared
and added to the different functionalized carbon nanotubes. The
resulting dispersions of 1a, 1b, 1c, ox-MWCNT and f-MWCNT 9
were set aside for 10–12 h. After this period, some CNTs precipitated.
Catalysis: CNT-supported silver nanoparticles (1.00 mg of CNTs
contain 0.0003 mmol of silver) and 35% aqueous hydrogen peroxide
(4.0 mL, 0.05 mmol) were added to a solution of 2-methylhydroqui-
none (1.25 mg, 0.01 mmol) in methanol (0.50 mL). The resulting
suspension was sonicated at room temperature until complete
consumption of starting material and formation of 2-methylbenzo-
quinone was shown by HPLC. This analysis showed a selectivity of
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9893 –9897