W. Du et al. / Catalysis Communications 41 (2013) 56–59
57
placed in an ice bath and vigorously stirred for 1 h. After that, freshly
prepared aqueous solution of NaBH (0.01 M, NaBH /metal = 5:1
molar ratio)) was added under vigorous stirring, and then the activated
surfactant (PVA) could get into the porous windows and occupy partial
pores. The calculated BET surface areas of MIL-101 and Pt@MIL-101 are
2869 and 2500 m /g, respectively (Fig. 1). The reduction in surface area
4
4
2
(
MIL-101 was added immediately. The suspension was stirred at 0 °C for
another 4 h, followed by washing thoroughly with deionized water. The
sample was finally dried under vacuum at 100 °C for 2 h to obtain the
Pt@MIL-101 catalyst. The loading amount of Pt in the fresh prepared
sample was 0.994 wt.%, based on AAS analysis.
is also attributed to the incorporation of Pt and/or surfactant in the
pores of MIL-101.
Fig. 2 shows the time-resolved TEM images of Pt@MIL-101 catalyst.
The fresh Pt@MIL-101 particles were sized in 150–350 nm. Pt cannot be
identified in the frameworks of fresh Pt@MIL-101 (Fig. 2(a)–(c)). The
oriented and ordered pore channels can be seen when the microscope
was focused on a single particle (Fig. 2(b)). There exist two kinds of
fringes in the particle which correspond to interchannel spacing of
2
.2. Characterizations
Powder X-ray diffraction (XRD) patterns of the samples were
obtained by a Rigaku diffractormeter (D/MAX-IIIA, 3 kW) using Cu Kα
radiation (40 kV, 40 mA, λ = 0.1542 nm). The surface area and pore
4.323 nm (d
and 2.714 nm (d ,
1
, wall thickness = 1.636 nm, pore diameter = 2.687 nm)
wall thickness = 0.985 nm, pore diameter =
2
1.729 nm). These bright spots (in Fig. 2(c)) arranged in orderly rows
would be the cross-sections of the microporous channels with a uniform
pore-size (1.7–1.9 nm), which agrees well with the XRD pattern and
pore size distribution curves (Fig. 1). The thermal stability of Pt@
MIL-101 is low when it was exposed under the electron beam in TEM
analysis. The external layer of the framework melted when the micro-
scope was focused on a single particle for a time longer than 10 min.
Fig. 2(c) and (d) are images of a same corner of a Pt@MIL-101 particle
acquired at 2 and 10 min under the electron beam. A clear image and
stable framework could be identified in Fig. 2(c). After 10 min under
the electron beam in TEM analysis (Fig. 2(d)), the Pt@MIL-101 particle
shrank. Pt nanoparticles (NPs) agglomerated at the edge and surface of
the particle (Fig. 2(d)). The Pt NPs distribution diagram shown as an
insert in Fig. 2(d) indicated that most Pt adatoms agglomerated to
2.0 ± 0.5 nm particles when Pt@MIL-101 was exposed for 10 min
under the electron beam in TEM analysis. Fig. 2(e) is the image of a
Pt@MIL-101 particle (10 min under the electron beam) from another
orientation. XPS analysis confirmed that all Pt(IV) were reduced to
Pt(0) in the prepared Pt@MIL-101 (see Fig. S2 in Supplementary data).
distribution measurements were performed with N
2
adsorption/
desorption isotherms at 77 K on a Micromeritics ASAP 2020 instrument.
Transmission electron microscopy (TEM) images were obtained using
an accelerating voltage of 200 kV (TEM, JEOL-2020F). X-ray photoelec-
tron spectroscopy (XPS) measurements were performed on a Kratos
−
9
Axis Ultra DLD system with a base pressure of 10 Torr.
2
.3. Hydrogenation of nitroarenes
In a typical procedure [28], catalyst (containing 0.1 mg Pt) was
dispersed in 30 mL ethanol, and then a certain amount of substrate
was added. The reactor was sealed, purged with hydrogen and pressur-
ized to 1.0 MPa, followed by stirring with a magnetic stirrer at a rate of
1
000 rpm at 0–30 °C. After reaction, the mixture was separated by cen-
trifugation. The reactants were analyzed by GC (HP 5890, USA), and all
products were confirmed by GC–MS (Agilent 6890–5973 N).
3
. Results and discussion
3
.1. Characterizations of the catalysts
3.2. Catalytic activity
The XRD pattern of the synthesized MIL-101 and Pt@MIL-101 is
Table 1 summarized the reaction results of the hydrogenation of
nitrobenzene (NB) over Pt@MIL-101 at different ratios of NB molecule
to Pt atom. The catalyst exhibited inconceivable activity even at a
high ratio between NB molecule and Pt atom (72000:1). It was found
that NB was converted completely within 60 min at a NB/Pt ratio of
36000:1, and the selectivity of aniline (AN) reached 68.0%. Beside AN,
intermediates of the consecutive hydrogenation process such as
nitrosobenzene (NSB) and azoxybenzene (AOB) were also detected in
the reaction mixture. The conversion of NB reached 78.6% even when
the NB/Pt ratio increased to 72000:1, and the calculated turnover
frequency (TOF) of each Pt atom for NB conversion reached
similar to the simulated pattern of MIL-101 reported by Férey et al.
see Fig. S1 in supplementary data) [15]. Both of them are all of high
crystallinity and the sharp and strong peaks at 2–4° reveal the presence
of both micropores and mesopores. N sorption isotherms of both sam-
ples are of type I model with two uptake peaks at 0.0–0.02 and 0.21–
.24 (of P/P ), which is the characteristic of two kinds of microporous
(
2
0
0
windows [15]. Both pore-size distribution curves of MIL-101 and Pt@
MIL-101 that calculated from their adsorption isotherms show two
peaks at 2.2, 3.0 nm and 1.8, 2.7 nm, respectively (Fig. 1). This decrease
in pore diameter (about 0.3–0.4 nm) infers that Pt and/or some
−
1
9
43.2 min . When the reaction temperature was raised to 30 °C, the
−
1
calculated TOF of Pt atom further increased to 1548 min . And this
value is higher than that of multiwall carbon nanotube supported
ultrafine Pt NPs (1115 min ) [28,29], Pt/SiO
−
1
−1
2
(37 min ) [30], Pt
−
1
−1
NPs (53 min
for AN) [31], Pt/MCM-41 (1 min ) [32], Pt/C and
−
1
Pt/PMO-SBA-15 (around 10–400 min ) [33] in the similar or more
critical reaction condition. Besides, the deactivation of the catalyst was
inconspicuous even after five cycles (see Fig. S3 in supplementary data).
The predominant activity of Pt@MIL-101 could be attributed to the
highly dispersed Pt adatoms in the framework of MIL-101, and the
good distribution of Pt@MIL-101 in the reaction mixture. In addition,
Pt@MIL-101 has a mesoporous zeotype architecture with mesoporous
cages (1.8 and 2.7 nm) and large free apertures (1.2 nm from pentag-
onal windows and 1.6 × 1.45 nm from hexagonal windows) [34],
providing free access for NB and AN molecules [35].
Taking the excellent performance of Pt@MIL-101 in NB hydrogena-
tion reaction into consideration, it is possible that the catalyst would
also exhibit high activity for other nitroarene derivatives. As shown in
Table 2, these reactions proceed very well with Pt@MIL-101 despite of
the group substituting on aryl nitro including electron-donating groups
Fig. 1. Nitrogen adsorption isotherms at 77 K: (a) MIL-101; (b) Pt@MIL-101. The inset
image is pore-size distribution curves for MIL-101 (a) and Pt@MIL-101 (b).