U. N. Gupta et al.
injector and detector were maintained at 150 and 200 °C
respectively. Calibration curves for reaction products were
obtained using a known amount of pure standard in
acetonitrile.
Table 1 Oxidation of propene
Propene
BPO
Yes
Yes
Yes
Yes
Yes
Yes
None
Yes
Yes Yes No
No
No
Yesa
Oxygen
Catalyst
Yes No
Yes
Yes Yes Yes
Propene oxide yield (mol%) 0.31 0.0069
Products (9 10-6 mol)
0
0
0
3 Results and Discussion
Acetaldehyde
Propene oxide
Acetone
124
214
2.9
3.54
4.56
0.25
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
In our initial experiments we investigated the oxidation of
propene with Au/graphite as catalyst with a range of radical
initiators but no reactivity was observed. We observed that
alloying Pd with Au led to oxidation activity particularly
when BPO was used as the radical initiator. We subse-
quently concentrated our efforts on this reaction system. In
the absence of propene no reaction products associated
with the oxidation of the solvent were observed (Table 1).
Propene does not react with oxygen in the absence of the
radical initiator (BPO) even when the catalyst is present
(Table 1). On the other hand when the reaction was per-
formed without oxygen but in presence of BPO, trace
amounts of products were observed (Table 1). This is an
important observation since it suggests that PhCO-O. rad-
icals can epoxidise alkenes. However, when O2 and BPO
were both present the yield of oxidation products was
significantly enhanced (Table 1). We next investigated the
effect of the reaction time on the oxidation of propene and
the results are shown in Table 2. It is clear that after 2 h
reaction there is no increase in the yield of oxidation pro-
ducts. This suggests that once the BPO has been consumed
the oxidation of propene stops. This is in direct contrast to
our earlier studies concerning alkene oxidation using AIBN
as a radical initiator [7] where the oxidation persists after
the radical initiator has been consumed. In the case of
longer chain alkenes when reacted under solvent-free
conditions we have suggested that BPO will produce
benzoyloxy radicals and we can speculate that these might
attack the C1 position of the 1-alkene to give an interme-
diate radical that can fragment to give epoxide and PhCO.,
the benzoyl radical would then react with O2 to generate
PhCO-OO. and this can then react to produce more epoxide
and benzoyloxy radicals in a radical chain process. Clearly
this is not being observed in the present case. A key dif-
ference in the current experiments is that acetonitrile is
Acrolein
36.2 2.31
14.7 7.82
82.3 52.0
Methanol
COx
Reaction conditions: Acetonitrile (30 ml), BPO (0.168 g, 519 lmol),
0.5 %Au–0.5 %Pd/graphite (0.040 g), Propene (3 bar gauge,
66.2 mmol), O2 (3 bar gauge), N2 to top up to a total reaction pressure
of 30 bar gauge, 90 °C, 4 h
a
Products from the reaction of BPO were observed namely benzene,
phenol and benzoic acid in very minor amounts
graphite as a support (acidified to pH 1 using sulphuric
acid). The amount of support was calculated to give a total
final metal loading of 0.5 %Au and 0.5 %Pd. After 2 h the
slurry was filtered, the solid washed thoroughly with dis-
tilled water to remove Na? and Cl- and dried (110 °C,
16 h). This material has been characterised extensively in
previous studies [8].
2.2 Oxidation of Propene
The reactions were performed in a 50 ml Parr 5500 Series
Compact reactor. As propene is a gas under the conditions
of our reaction we elected to use a solvent. Experiments
using water, methanol or ethanol were found to yield no
epoxide, but experiments with acetonitrile did give epoxide
as product and hence we selected acetonitrile as solvent.
Catalyst (0.040 g), radical initiator (173 lmol of ben-
zoyl peroxide (BPO)) and acetonitrile (30 ml) were added
to the autoclave. Propene (0.0662 mols, 3 bar gauge),
oxygen (3 bar gauge), and nitrogen (24 bar gauge) were
added at 20 °C and the autoclave was heated to 90 °C.
When the reactor attained 90 °C, the solution was stirred
and the reaction started. The reactor was cooled in ice bath
(1–2 °C) to stop the reaction, the reaction products were
collected after removal of the catalyst by filtration. To an
aliquot of the product (5.0 ml) a standard was added
(100 lL o-xylene 99 %, Alfa Aesar) prior to analysis. The
reaction mixture was analyzed using gas chromatography
using a Varian gas chromatograph 3,380 column and a
flame ionization detector. The column was maintained
isothermally at 40 °C for 5 min, followed by a temperature
ramp of 10 °C/min to a final temperature of 200 °C; the
-
used as a solvent. It is apparent that O2 is essential to
observe significant propene oxidation (Table 1) but that the
process is clearly far less efficient in the presence of the
solvent. It is possible that acetonitrile that acts as a radical
terminator as the concentration would be much higher than
that of propene but under our reaction conditions this is
difficult to detect.
We studied the effect of the concentration of BPO and
the results (Table 3) demonstrate that the yield of oxidation
products increases with increasing the amount of BPO
123