4
E. Fisher et al. / Polyhedron xxx (2016) xxx–xxx
supported nanoparticles were filtered out. The reaction flask was
washed with diethyl ether (10 mL) to dissolve any remaining
product. The solid supported nanoparticles were then washed with
We therefore demonstrated that both Au and Ag water
suspendable nanoparticles are able to catalyze the oxidation of
1,1-diphenylethylene affording the product of oxidative cleavage
benzophenone as the major product. It should be noted that in
the report of Shi and co-workers a mechanism involving the first
step of epoxidation and a subsequent oxidation to produce car-
bonyl products was discarded since they could not detect epoxida-
tion products or perform further oxidations on specific epoxides.
[20] Gold NP of sizes 10 and 12 nm (1 mol%) are able to oxidize
1,1-diphenylethylene in 18 h affording conversions of 40–90%
depending on the time. Silver nanoparticles catalyze this oxidative
cleavage but the amount required for a conversion of 90% (18 h of
reaction) is higher (5 mol%).
deionized H
2
O (1 Â 10 mL) and diethyl ether (3 Â 10 mL). Follow-
ing the filtration process, the supported nanoparticles were dried
under vacuum for further use. The organic layer was extracted
from the combined organic and aqueous filtrate with diethyl ether
(
3 Â 20 mL). Following extraction, the organic layer was washed
with a saturated sodium bisulfite solution (3 Â 10 mL), followed
by deionized
(2 Â 10 mL). The organic layer was then
dried over MgSO overnight. The solution was filtered and poured
2
H O
4
into a pre-weighted round bottom flask and dried under high vac-
1
uum. The product obtained was weighed and analyzed by H NMR
spectroscopy. Ketone and epoxide (Scheme 2) were observed in
different percent conversions depending on the reaction conditions
While trying to recycle these nanoparticles and run subsequent
reactions we observed a dramatic decrease in the percent total
conversion. This is due to the decomposition and aggregation of
the water-suspendable nanoparticles in solution under the reac-
tion conditions. We decided to support these nanoparticles on a
solid in order to improve their catalytic activity and their recycla-
bility. Since the nanoparticles used in this study contain tannic
acid/citrate as surface groups, a procedure was developed to dis-
place these groups and coordinate the metal NP to the functional
groups on the solid support. The surface of silica materials can be
functionalized by various methods, e.g. by reaction of the silanol
groups on the silica surface with alkoxysilanes [23].
(Tables 4–6).
2.6. Transmission electron microscopy characterization
TEM images were taken on a JEOL 1010 electron microscope.
Samples of the catalysts were suspended in a 2:1 mixture of iso-
propanol and DI water. 5 L of this solution was dropped onto
l
the formvar side of a carbon coated formvar 300 mesh copper
TEM grid. The sample was then dried under vacuum and imaged
at 100 kV.
Functionalizing the silica with aminopropyltrialkoxysilanes
proved efficient for Au nanoparticles. The primary amine in the
APTMS (Fig. 1) is known to possess a high affinity for Au nanopar-
ticles and to prevent nanoparticles from aggregating. The amine
groups give rise to a positive charge on the surface of the silica
while the gold and silver nanoparticles with either a citrate or a
tannic acid surface inherently have a negative charge, thus bring-
ing the particles in close contact to the silica surface. Amines are
better ligands for noble metal surfaces than the alcohol or acid
groups in the citrate or tannic acid, thus the incorporation of ami-
nes was used to bind the metal nanoparticles to the silica surface.
We tested supported Au NP in the oxidation reaction and
observed that when the mol% of catalyst was increased from 1 to
3
. Results and discussion
We studied the use of commercially available water-stable
Ag/Au nanoparticles for the liquid-phase oxidation of 1,1-dipheny-
lethylene in H O at 90 °C with TBHP as oxidant (2.1 equivalents).
2
The oxidation products are benzophenone, product of the oxidative
cleavage, and 1,1-diphenylepoxide (Scheme 2). We started by
investigating the effect of time in the catalytic activity of 1 mol%
of unsupported 10 nm and 12 nm gold nanoparticles (Table 1).
From the results depicted in Table 1 we infer that the total con-
version and the percent conversion to ketone are directly propor-
tional to the reaction time. Increasing the reaction time resulted
in an increased total conversion and increased conversion to
ketone. We next examined how the conversion percentage was
related to the mol percent of catalyst used in the reactions for Au
NP of size 12 nm (Table 2) with a constant reaction time of 18 h.
We did not observe significant changes in terms of total conver-
sion. However, we observed that increasing the amount of catalyst
increased slightly the amount of conversion to ketone product.
Our group had reported that silver(I) complexes containing a
tripodal bis(imidazole) thioether ligand (10 mol%) could perform
this oxidative cleavage at 90 °C in toluene. We therefore examined
next the catalytic activity of silver nanoparticles of 10 nm in size
and the dependence of the catalytic reaction in terms of time
and mol percent of the catalyst (Table 3).
1
.5, the total conversion increased and the ratio to the product of
oxidative cleavage increased (see entry 2). Also, when comparing
entry 1 in Table 4 to entry 1 in Table 1, we can observe that the
activity of 10 nm Au nanoparticles improves when the Au NP were
supported on silica functionalized with APTMS. In the same man-
ner, Au NP of size 12 mm afforded the same catalytic activity while
supported on silica functionalized with APTMS affording high% of
total conversion and a 97% conversion to benzophenone. In general
the percentage of epoxide produced was considerably lower than
with unsupported Au or Ag NP (never more than 10%).
The 10 nm Au nanoparticles from entry 2, Table 4 (before and
after catalysis) were sent for characterization by TEM imaging
(Fig. 2) and we could observe that there were not considerable dif-
ferences in the samples in terms of NP size.
We examined next the recyclability of the supported Au
nanoparticles (10 mm size). The results are collected in Table 5.
In the recyclability tests we used 1.5 mol% of 10 nm Au nanoparti-
cles that were supported on 0.25 g silica functionalized with
APTMS (entry 1). After the reaction the nanoparticles were
The results in Table 3 show that as the reaction time increases
entry 1 to entry 2), the conversion to ketone and the total conver-
sion increases as well. However, as the mol% of the catalyst
increases the overall percent of conversion doubles but the ketone
and epoxide distributions do not change significantly.
(
O
O
Si
NH
2
Si
SH
O
O
O
O
3
aminopropyltrimethoxysilane (APTMS)
3 mercapto propyltrimethoxysilane (MPTMS)
Fig. 1. Two silanes (MPTMS and APTMS) used in this study to functionalize silica to make solid supports.