W. Wu et al. / Applied Catalysis A: General 469 (2014) 483–489
485
this reaction, providing a moderate cyclohexane conversion (43.9
81.8%, see entries 10–11). Notably, this photochlorination did not
easily occur in another DMF medium of strong polarity (ε = 36.7,
entry 12), which is likely due to the capture effect of DMF on the
free radicals produced in the photoreaction [34]. Ethyl acetate with
a weak polarity (ε = 6.0) also provided a low chlorinated efficiency
for the present photoreaction (see entries 13) owe to itself transfor-
mation to acetic acid, which has been confirmed by use of GC-MS
measurement. The formation of acetic acid should originate from
ethyl acetate hydrolysis.
2.5
2.0
1.5
1.0
0.5
0.0
3
6
3.2.2. Effect of various parameters
In the following experiments, the influence of the amount of
chlorinating agent FeCl3·6H2O and irradiation time on the photo-
oxy-chlorination of cyclohexane was checked by using CH3CN or
acetone as a medium. As shown in Fig. 3A, when the photore-
action was carried out in CH3CN, an increase in the amount of
FeCl3·6H2O from 0.5 to 2 mmol resulted in the continuous and con-
siderable increase of cyclohexane conversion from 27.7 to 46.9%
and the gradual decrease of chlorocyclohexane selectivity from
tivity for the chlorinated products were hardly influenced by the
amount of FeCl3·6H2O. The effect of the amount of FeCl3·6H2O
CH3CN (Fig. 3B vs. Fig. 3A), but the selectivity for chlorocyclohexane
was hardly influenced by the amount of FeCl3·6H2O. This illustrates
that a further photo-oxy-chlorination of chlorocyclohexane can be
restrained efficiently in acetone.
Fig. 4A shows that when the irradiation time was between 2
and 12 h, the curve of cyclohexane conversion in CH3CN climbed
sharply with irradiation time; after 12 h, it climbed slowly and
eventually achieved ca. 58.8% at 16 h. The selectivity for chlorocy-
clohexane continuously decreased with time, with a concomitant
increase in dichlorocyclohexane selectivity, while the selectivity for
cyclohexene was hardly influenced by the time. Fig. 4B shows that
cyclohexane conversion in acetone continuously increased with
irradiation time and eventually achieved ca. 60.1% after 16 h. The
effect of irradiation time on the selectivity for various products in
acetone was very similar to that in CH3CN.
1
2
7
200
300
400
500
600
700
Wavelength/ nm
Fig. 2. The UV–vis spectra of various metal chlorides in acetonitrile solvent
(4.0 × 10−4 M, 1: CoCl2·6H2O; 2: CuCl2·2H2O; 3: ZnCl2; 4: VOCl3; 5: SnCl4·5H2O;
6: MnCl2·4H2O; 7: NiCl2·6H2O; 8: FeCl3·6H2O). Inset is the UV–vis spectra of
FeCl3·6H2O in different solvents (4.0 × 10−4 M).
as a representative. As shown in the inset of Fig. 2, the visible
absorption edge of FeCl3·6H2O was influenced obviously by sol-
vent. In comparison with CH3CN, ethyl acetate, and especially
acetone gave a stronger and wider absorption edge, while dimethy-
lamide (DMF) gave a weaker but much wider absorption edge.
The solvent-mediated effect on the visible absorption edge of
FeCl3·6H2O probably originates from their coordination interaction
with FeCl3·6H2O.
3.2.1. Photo-oxy-chlorination performance of various metal
chlorides
Table 1 lists data for the visible light-triggered oxy-chlorination
of cyclohexane with some metal chlorides (1 equiv.) in CH3CN
at 35 ◦C under an N2 atmosphere. Entries 1–5 illustrate that
MnCl2·4H2O, ZnCl2, CoCl2·6H2O and NiCl2·6H2O with a negative
standard redox potential (SRP) and SnCl4·2H2O with a positive SRP
were nearly invalid for this chlorination, providing less than 2%
of cyclohexane conversion in CH3CN solvent. This is likely due to
these metal chlorides having no response to visible light. A few
metal chlorides such as VOCl3, FeCl3·6H2O and CuCl2·2H2O with
a positive SRP, which can respond to visible light, were found to
be efficient chlorination agents to this photoreaction, affording
mono- and di-chlorinated cyclohexane as major reaction products,
with concomitant formation of a small quantity of cyclohexanol,
cyclohexanone and cyclohexene (see entries 6–8). Among them,
CuCl2·2H2O showed the best photoreaction effect, providing about
66.6% of cyclohexane conversion with a relatively low selectivity
for the chlorinated products (72.3%) after 8 h of visible light irra-
diation (see entry 7). FeCl3·6H2O exhibited a good reactivity and
provided an excellent selectivity for the chlorinated products (entry
8). Entry 9 shows that the oxy-chlorination of cyclohexane with
ing that the present oxy-chlorination is indeed triggered by visible
light.
chlorination, we tried to use concentrated HCl as a chlorinating
agent for the FeCl3 or CuCl2-photocatalyzed cyclohexane oxy-
chlorination under visible irradiation and the results are listed
in Table 2A. Entry 1 shows that when a mixture of FeCl3·6H2O
(0.1 mmol) and cyclohexane (1 mmol) in CH3CN (5 mL) was irra-
diated by 35 W halogen lamp for 12 h under N2 atmosphere, the
obtained cyclohexane conversion (15.7%) was higher than the stoi-
chiometric amount of FeCl3·6H2O (10%), and a small amount of
the oxygenated products cyclohexanol and cyclohexanone was
also found in addition to the main product chlorocyclohexane.
This implies that the present reaction possesses some photocat-
alytic cycling character. When the photoreaction described above
was carried out under air atmosphere (see entry 2), the con-
version and the selectivity for the oxygenated products were ca.
21.7 and 36.9%, respectively, being obviously higher than both
those obtained under N2 atmosphere. This is mainly due to a
contribution to the FeCl3-photocatalyzed oxygenation of cyclo-
hexane by molecular oxygen. Adding 0.1 mL of concentrated HCl
to the above-mentioned photoreaction system was invalid to the
improved cyclohexane conversion under N2 atmosphere (entry
3), and contrarily resulted in the decreased cyclohexane conver-
sion under air atmosphere (entry 4). Moreover, its introduction
The influence of solvent on the photo-oxy-chlorination was fur-
ther examined by using 2 equiv. of FeCl3·6H2O as an example and
the results are shown in Table 1. Among the solvents examined,
acetone (ε = 20.7) and CH3CN (ε = 37.5) with a moderate and strong
polarity, respectively, were found to be the best two media for