J.M. Escola et al. / Journal of Catalysis 270 (2010) 34–39
35
corresponding ketones. The b-cyclodextrin is soluble in water and
2.3. Analysis of the reaction products
the well-known molecular recognition between the host cavity of
cyclodextrins and organic compounds allowed for higher oxidation
rate and ketone selectivity. In addition, the solvent also affects the
activityoftheWackersystems.Inthisregard,Mitsudomeetal.[23]re-
ported the remarkable oxidation of higher 1-olefins (around 85%
yields) with molecular oxygen using palladium chloride and N,N-
dimethylacetamide as solvent, without the need for copper chloride.
In a previous work [20], a model higher 1-olefin (1-dodecene)
was oxidized to 2-dodecanone and other ketones in a modified Wac-
ker system without copper chloride and employing tert-butylhydro-
peroxide (TBHP) as oxidant. Acetonitrile and palladium chloride
were the solvent and the catalyst of choice, respectively, due to their
high selectivity towards 2-dodecanone. In addition, the most
adequate operation variables were determined ([CH3CN]/[1-dode-
cene] = 10, [TBHP]/[1-dodecene] = 7). In this work, the performance
of the modified Wacker system in the oxidation of the heavy 1-ole-
fins C12, C16, C18 and C20 is carried out. These olefins can be easily
obtained in the feedstock recycling of polyethylene by thermal
cracking at 350–400 °C. In addition, this Wacker oxidation would al-
low obtaining more valuable products than the initial 1-olefins. On
the other hand, these olefins are obtained in the thermal cracking
in an almost equimolar mixture with the corresponding n-paraffin
of each fraction [24]. Owing to this, the performance of the modified
Wackersystemwas studiedunderthepresenceofthecorresponding
n-paraffins. Henceforth, the obtained results are reported which
indicate that the oxidation of the heavy olefins is feasible with this
modified Wacker system and takes place with high rate for the dif-
ferent olefins, regardless of the molecular weight. In addition, the
presence of the n-paraffin led to an unexpected increase in the selec-
tivity of the corresponding 2-methyl ketone.
The reactions were run for 7 h, with the samples being taken at
regular intervals and analyzed with a Varian 3900 gas chromato-
graph provided with a CP8907 methylsilicone column of 15 m
length ꢂ 0.25
lm width, using a flame ionization detector (FID).
Identification of the different reaction products was performed
by using commercial standards. Product distribution and overall
mass balances (closure was >98%) were determined using appro-
priate reactant and product response factors, derived from multi-
point calibration curves. Prior to the analyses, tetraline was added
as internal standard to the reaction mixture previously dissolved in
tetrahydrofurane. Subsequently, both the 1-olefins conversion and
the selectivity towards the obtained products were determined.
The conversion was defined as (mol of reacted 1-olefin) ꢂ (mol of
starting 1-olefin)ꢀ1 ꢂ 100. The selectivity was divided into the
three obtained product groups: 2-ketone (S2-ketone), other ketones
(Sother ketones) and alkene isomers (Sisomers).
3. Results and discussion
3.1. Mechanisms in the modified Wacker TBHP oxidation
Wacker oxidation using TBHP may proceed according to two
different mechanisms. Firstly, the water from the TBHP reagent
(70 wt.% aqueous solution) can act as oxygen source for the olefin
oxidation according to the conventional hydroxypalladation mech-
anism of the Wacker–Schmitt oxidation [25] leading to the follow-
ing scheme of reactions:
CH3ðCH2ÞnCH@CH2 þ H2O þ PdCl2
! CH3ðCH2Þ COCH3 þ Pd0 þ 2HCl
ðaÞ
ðbÞ
n
ðCH3Þ C—O—OH þ Pd0 þ 2HCl ! PdCl2 þ ðCH3Þ COH þ H2O
2. Experimental section
3
3
However, an alternative mechanism has also proved wherein
TBHP ketonizes the olefin through a peroxypalladation step,
according to Mimoun et al. [26] and Cornell and Sigman [27]. In
a previous work [20], we proposed the scheme of reactions shown
in Fig. 1 for the oxidation of 1-dodecene. The first step comprises
the complexation of the reacting olefin to the palladium (II) com-
plex (step 1). Subsequently, the complex proceeds giving rise to a
2.1. Chemicals
The chemicals used in the present research were as follows: 1-
dodecene (95 wt.%, Aldrich), n-dodecane (99%, Aldrich), 1-hexade-
cene (99 wt.%, Aldrich), n-hexadecane (99%, Aldrich), 1-octadecene
(90 wt.%, Aldrich), n-octadecane (99%, Aldrich), 1-eicosene (90
wt.%, Aldrich), n-eicosane (Aldrich, 97%), tert-butylhydroperoxide
(TBHP, 70 wt.% aqueous solution, Aldrich), acetonitrile (99.5 wt.%,
Scharlau) and palladium (II) chloride (99 wt.%, Fluka). Additionally,
other used chemicals were cyclohexane (99.5 wt.%, Aldrich),
dichloroethane (99.8 wt.%, Aldrich), 2-propanol (99.5 wt.%, Al-
drich), N,N-dimethylformamide (99.8 wt.%, Aldrich) and b-cyclo-
dextrin (C42H70O35ꢁxH2O, Aldrich).
p-allylpalladium species (step 2) which drives to the formation
of the different dodecene isomers (step 3). Tert-butyl hydroperox-
ide is incorporated inside the palladium complex of the dodecene
isomers forming a five-membered pseudocyclic peroxypalladium
complex (step 4). Afterwards, 1,2-hydride shift takes place and
the release of both tert-butanol and the desired product (other ke-
tone) occurs (step 5). In step 6, the peroxypalladium complex is
formed with 1-dodecene instead of the dodecene isomers, releas-
ing the desired product 2-dodecanone (step 7).
2.2. Experimental installation and oxidation reactions
The catalytic experiments were carried out in a stirred glass
batch reactor equipped with a reflux column and propeller stirrer.
The reaction temperature was controlled by a thermostatic bath
where the reactor was placed. In a typical experiment, 5.0 g of
the 1-dodecene was loaded inside the reactor with 11.6 g of aceto-
nitrile. Later, 21.6 g of TBHP and 0.1 g of palladium (II) chloride
were also loaded and the reaction started. The mixture was stirred
at 300 rpm at the reaction temperature (80 °C) during 7 h. In the
case of using b-cyclodextrin, this was added with a molar ratio of
(1-olefin)/(b-cyclodextrin) = 40. If the experiment was carried out
in the presence of the corresponding n-alkane, the mixture loaded
was (1-alkene)/(n-alkane) molar ratio = 1. After finishing the reac-
tion and in order to homogenize the medium, 179.0 g of tetrahy-
drofurane was added to the mixture for the subsequent analysis
of the obtained products by gas chromatography.
3.2. Effect of the used solvent
The reactions taking place in the studied Wacker system,
according to the previous research, are the following ones:
2-METHYL KETONE
1-OLEFIN
(3)
OLEFIN ISOMERS
OTHER KETONES
Oxidation is a slow reaction (pathway 2) which competes with the
fast isomerization of 1-olefin (pathway 1). However, isomerization