GModel
CATTOD-8598; No. of Pages7
ARTICLE IN PRESS
2
R. Li et al. / Catalysis Today xxx (2013) xxx–xxx
Table 1
room temperature. The AcH and products concentrations were cal-
culated from the results of GC-FID analysis, and then calculated the
substrate conversion and the selectivity toward AcH (benzene was
used as an external standard). GC-TCD was employed for determi-
nation of the gas-phase products. The conditions for benzyl alcohol
oxidation were similar to the above AcH oxidation process, except
that benzyl alcohol (0.02 M) was added into the solution contain-
AcH (0.08 M) oxidation into AcOH over FeCl3 with a variety of concentration in
CH3CN solvent at room temperature (25 C).
◦
[FeCl3]a
t/min
Conv./%b
Sel./%c
−1
TOF/s
Entry
1
2
3
4
5
6
7
8
9
0
120
120
120
120
60
60
60
140
140
120
0
0
2.1
36.8
54.4
43.4
73.8
74.5
16.1
7.8
–
–
0
0
1.0
2.5
5.0
10
>99
92
88
85
84
80
75
65
0.09
0.82
1.21
0.48
0.16
0.035
ing different amount of AcH, FeCl3 in CH CN (5 mL). The substrate
3
conversion and the selectivity toward benzaldehyde (toluene was
used as an external standard) were calculated from the results of
GC-FID and GC-TCD analysis.
20
100
200
3
−4
2 × 10
8 × 10
4
−5
10
2 × 10
4.4 × 10
2.2. Spectroscopic studies
a
b
c
FeCl3 concentration in CH3CN.
The conversion is based on the results of GC analysis.
During AcH to AcOH oxidation, formaldehyde, formic acid, CO2, etc. were formed
Accurate mass spectrometry data were obtained using Apex
III (7.0 Tesla) Fourier transform ion cyclotron resonance tandem
mass spectrometry (FTICR-MS/MS) (Bruker, Billerica, MA, USA),
equipped with XMASS software (Bruker, version 6.1.1) used for
instrument control, data acquisition and processing. Sodium triflu-
oroacetate was used as an external calibration compound. Solutions
were infused from the ESI source at 3 mL/min with the follow-
ing parameters applied: capillary, −4771 V; end plate, −4413 V;
skimmer 1, 12.00 V; skimmer 2, 6.61 V; offset, 0.98 V; RF ampli-
as by-products, which results in the decrease of selectivity upon elongated reaction.
experiments demonstrate that the auto-oxidation of AcH into AcOH
is not possible at low temperatures, and they also indicate that the
combination of FeCl and MeCN is unique to the AcH oxidation pro-
3
cess. The conversion, selectivity, and TOF value of aerobic oxidation
of AcH (0.08 M) into AcOH with a broad range of FeCl concentration
3
◦
are summarized in Table 1. When the concentration of FeCl3 (abbre-
tude, 582.5 Hz; drying gas temperature, 150 C. Nitrogen was used
viated as [FeCl ]) is less than 1.0 M, no obvious transformation
3
as the nebulizing and drying gas and argon was used as the
collision gas. X-band EPR signals were recorded at ambient tem-
perature on a Bruker EPR A-300 spectrometer. The settings for the
EPR spectrometer were as follows: center field, 3511.39 G; sweep
width, 100 G; microwave frequency, 9.86 G; modulation frequency,
of AcH is observed after 2 h reaction (entry 2, Table 1). Increas-
ing the [FeCl ] has a prominent impact on the conversion of AcH:
3
more than 70% of AcH is oxidized within 1 h when 100 M FeCl3
(0.5 mol) was employed (entry 7, Table 1), while 2 h reaction only
enables 7.8% AcH to be oxidized as [FeCl ] was further increased
3
1
00 kHz; power, 101 mW; conversion time, 10 ms. A glass capillary
to 20 mM (200-fold) (entry 10, Table 1). Notably, the TOF val-
tube containing calculated FeCl , AcH and acetonitrile solution, and
3
ues increased significantly with increasing [FeCl ] to a maximum
3
ice-cooled DMPO solution (0.08 M) was transferred and tested by
EPR spectroscopy at room temperature. No signals are observed in
the following systems: (1) FeCl , AcH and CH CN in the absence of
−
1
(
1.21 s , entry 5, Table 1) at ca. 10 M, with the value dropping
gradually at higher content.
3
3
In order to understand the reaction dynamics more clearly, the
DMPO, (2) FeCl , DMPO and MeCN, (3) DMPO, MeCN and AcH in
3
TOF variation as a function of [FeCl ] in logarithm is shown in
3
open air, and (4) DMPO, CH CN, and other organic substrates (e.g.
3
Fig. 1a. It is noteworthy that the TOF distribution is nearly Gaussian-
HCHO, MeOH, and AcOH) were employed.
shaped (red parabolic curve), peaking at ln[FeCl ] ≈ 2.0. The above
3
results suggest that the initial concentration of FeCl largely affects
3
2.3. Outline of DFT calculation
the AcH-to-AcOH transformation, making the whole process to
be controllable. Such a catalyst-concentration-dependent catalytic
process is rare and interesting, implying that unique catalytic
centers may be formed in such a diluted concentration range
The hybrid type density functional theory method so-called the
Becke three-parameter Lee–Yang–Parr functional was employed
8–11]. The Los Alamos effective core potential was used for the
[
(5–100 M) similar to previous observation of active gold cluster
Fe atom [12], and the 3s, 3p 3d, and 4s electrons of Fe and all the
electrons for the first row atoms were dealt with the valence double
quality basis sets [13] which are implemented in the Gaussian 03W
package [14]. The reactions were simulated in vacuum. However,
a part of calculation was carried out including the solvation effects
employing the self consistent reaction field theory at the level of
polarizable continuum model.
The whole reaction is divided into four elementary steps. There
are four transition states (TS’s). First TS is characterized, and then
the intrinsic reaction coordinate (IRC) is evaluated to both direc-
tions [15–17]. From each end point of the IRC, usual geometry
optimization leads to the reactant or product as the local minimum.
for CO oxidation [18].
The kinetics of AcH oxidation was further analyzed by vary-
ing the concentration of AcH (abbreviated as [AcH]) with [FeCl3]
at a given value of 20 M (0.1 mol). As shown in Table 2, only
limited conversion is observed when [AcH] below 0.02 M (entry 1,
Table 2), but it is intriguing to note that the conversion rate speeds
up upon increasing [AcH] and reaches an optimal concentration at
0
.2 M (entry 6, Table 2). By plotting the TOF data against [AcH], we
Table 2
The oxidation of AcH with concentration from 0.02–1.0 M over FeCl3 (20 M) in
CH3CN at room temperature.
[AcH]a
t/min
Conv./%
Sel./%
TOF/s
−1
3
. Results and discussion
Entry
1
2
3
4
5
6
7
8
9
0.02
0.04
0.08
0.12
0.16
0.2
0.4
0.8
1.0
30
60
30
30
30
30
30
30
30
0.9
12.9
13.2
12.1
22.7
22.9
12.9
7.5
>99
87
86
94
89
89
95
94
0.016
0.072
0.294
0.601
1.01
1.27
1.43
1.66
By accident, we found that AcH can be efficiently oxidized into
AcOH in a homogeneous system containing micromole level FeCl3
◦
and MeCN solvent at room temperature (25 C), while negligi-
ble reaction occurred under the following conditions: (i) in the
absence of FeCl3 or other instead iron salts (e.g. FeSO , Fe (SO ) ,
4
2
4
3
and Fe(NO ) ) were used, (ii) MeCN solvent was replaced by THF,
3
3
CH Cl or H O, and (iii) other analogous metal chlorides (e.g. ZnCl ,
2
2
2
2
a
CoCl , RuCl , and MnCl ) were used as the catalysts. The parallel
AcH concentration.
2
3
2
Please cite this article in press as: R. Li, et al., Dioxygen activation at room temperature during controllable and highly efficient acetaldehyde-