Z. Shahin et al.
Molecular Catalysis xxx (xxxx) xxx
Scheme 1. Oxidative esterification of furfural into methyl-2-furoate.
Fig. 1. Au4f XPS spectra of Au25(SG)18/ZrO2 before and after calcination.
2.2. Preparation of Au25(SG)18/ZrO2
The material was prepared as previously reported [15]. ZrO2 ob-
tained from calcination of Zr(OH)4 (550 ◦C, 12 h) was wet-impregnated
with a water solution of Au25(SG)18 clusters to give a material with a
theoretical loading of 1 wt% of Au. Calcination was performed at 200,
300 or 400 ◦C under air flow for 12 h.
Fig. 2. Furfural oxidative esterification into methyl-2-furoate with 0.1 wt% of
furfural in methanol: comparison of reaction profiles in the absence or presence
of a base. Furfural 35 mg (0.36 mmol); methanol 44 mL; base Na2CO3 75 mg
2.3. Characterizations
(0.33 mol equiv./furfural); catalyst 40 mg (0.2 mg of Au, 1
μmol); O2 6 bar,
100 ◦C.
XPS studies were performed on a AXIS Ultra DLD KRATOS apparatus
using monochromatized AlKα source (hν =148606 eV).
defunctionalization at this level of calcination.
Note that despite the fact that TGA indicated that some organic
component remained after calcination at 300 ◦C, XPS showed the
absence of N- and S- species. Probably at this stage only carbonaceous
species formed from ligand decomposition at 300 ◦C were still present
however not detectable at the surface.
2.4. Oxidative esterification of furfural
In a typical experiment, furfural (35 mg, 0.36 mmol), catalyst (40
mg, corresponding to 0.2 mg of Au, 1 μmol of Au), Na2CO3 (75 mg, 0.33
equiv./furfural) were added to methanol (35 g, 44 mL) in a 50 mL
autoclave. After pressurization with O2 (6 bar), the mixture was heated
to 100 C (5 C.minꢀ 1) and let for the desired time. The reaction was
monitored by regular samplings (0.5 mL) that were diluted in methanol
in the presence of butyric acid as standard for gas chromatography
analysis (Shimadzu GC-2010, FID detector, CP-WAX 52 CB 30 column,
N2 carrier). After reaction, the catalyst was separated by filtration over
Teflon filter, washed with methanol and dried before reuse in other
experiments.
3.2. Initial experiments
◦
◦
Initial set of experiments was conducted in the condition depicted in
Scheme 1 and Fig. 2. The reactions were performed with the catalyst
obtained after calcination at 400 ◦C, a temperature allowing complete
removal of SG ligands. The amount of introduced catalyst was 1 μmol of
Au corresponding to a ratio Au/furfural of 0.57 wt% (0.27 mol%).
Fig. 2 indicates that in these conditions and in the absence of a base,
the conversion was complete after 4 h with a yield of 100 % into methyl-
2-furoate. Interestingly the presence of a base did not have significant
impact, slightly faster reaction and complete product formation was
TOFs are related to substrate consumption and were calculated as
follows: (n0-nt)/(nAu×t) with t arbitrarily fixed at 0.5 h for all
experiments.
achieved after 3 h (Fig. 2) (TOF at 0.5 h were 250 hꢀ 1 and 320 hꢀ 1
,
3. Results and discussion
respectively). Clearly, the use of a base supposed to help methanol
deprotonation is not of definitive interest in these conditions. Besides,
the influence of the amount on Au was assessed by performing the re-
3.1. Catalyst preparation and characterization
action in the presence of 0.5 μmol of Au still in the absence of a base
Au25(SG)18/ZrO2 catalysts were prepared as described in the
Experimental section following previous report [15], and contain 1 wt%
of Au before calcination steps. Thermogravimetric analyses on the
Au25(SG)18 clusters indicate that calcination at 200, 300 and 400 ◦C led
to 36, 75 and 100 % loss of ligand, respectively [15]. As an additional
characterization for the present work, XPS studies of Au indicate here
that the oxidation state of Au changed upon calcination (Fig. 1) [22]. For
initial Au25(SG)18/ZrO2, the Au4f5/2 and Au4f7/2 bonding energies (sig-
nals at 88.4 and 84.7 eV, respectively) are similar to those after calci-
nation at 200 ◦C (signals at 88.2 and 84.7 eV), all corresponding to Au
(δ+) species as in Au25(SG)18 [23]. After calcination at 300 then 400 ◦C
the signals shifted to 87.8 (then 87.6) and 84.0 (then 83.9) eV respec-
tively, indicating the reduction to metallic Au(0) species, in accordance
with ligand departure [22]. In addition, XPS of N1s and S2p (Fig. S1 in
Supplementary information) shows that the amino and thiolate func-
tions from the SG ligands are present on the clusters calcined at 200 ◦C,
and are absent after calcination at 300 and 400 ◦C, confirming their
(Fig. S2). Here the transformation was overall slower and 5 h were
necessary for complete transformation despite close initial rate (TOF of
245 hꢀ 1). Note that in the absence of oxygen pressure (reaction per-
formed under ambient air, Fig. S3), the conversion was limited to 55 %,
even after running a 24 h reaction (TOF of 85 hꢀ 1), while still being fully
selective.
The results obtained here are completely in line with those reported
in the literature with other Au/ZrO2 catalysts [9,11,13,24]. More
recently Au/UiO-66 [6] and Au/CMK-3 [4] catalysts, also showed
comparable activity in similar conditions, particularly involving less
than 1 wt% Au/furfural and most of the time in the absence of base.
3.3. Influence of calcination temperature
The above initial experiments were realized after calcination of the
◦
catalyst at 400 C, allowing complete SG ligands removing and while
2