1146
R. Klimkiewicz et al. / Catalysis Communications 11 (2010) 1143–1147
1.57 g/h of toluene) should yield 2.97 g/h of product containing
1.43 g/h of ketone fraction and 1.57 g/h of toluene. Since, at 400 °C the
solid product was produced at 1.92 g/h, the difference 1.08 g/h can be
related to evaporated toluene and unidentified by-products of the
reaction. Solid product separated in this way still contained toluene.
Drying in a desiccator resulted in a 36.5% weight loss. Thus, 1.22 g/h of
ketone fraction was obtained at 400 °C. This represents an efficiency
of 85.3% as compared to the theoretical value. Similarly, 1.70 g/h solid
product was obtained at 420 °C. The greater difference (1.3 g/h)
indicates a higher degree of toluene evaporation and a larger share of
side destruction reactions and/or secondary condensations. Drying of
such samples in a desiccator resulted in 38.1% weight loss. Thus, at
420 °C the ketone fraction was obtained with the efficiency 1.05 g/h,
i.e. 73.4% yield in relation to the theoretical value.
Obtained fraction of ketones was characterized by means of GC–
MS and NMR. The 13C NMR spectra (Fig. 1) revealed the 210 ppm
signal of C-carbonyl group. Other signals should be assigned in
descending order, starting from the CO group: 31.95 ppm—second
carbon in chain, 29.65 ppm—third methylene carbon, etc. The last
signal—22.73 ppm is assigned to methyl carbons. The 1H NMR spectra
(Fig. 2) show a triplet at 0.88 ppm (CH3 group) and a large multiplet
1.25 ppm (CH2 groups). These alpha and beta groups are shifted
2.38 ppm and 1.55 ppm, respectively. Additional signal at beta CH2
(probably due to H2O, shift 1.56) increases the integration by 2.
Results of GC–MS measurements have confirmed that material
labeled as “ketone fraction” contains the expected ketones of high
molar mass.
It was found that the catalyst's activity was stable during the run,
exhibiting low tendency to formation of carbon deposits of resin and
tar type accumulating on it. After the test catalyst remained in the
reactor and was regenerated by heating at 470 °C for 4 h under air
flow. Model test with the use of 1-butanol (Table 2) showed that both
the fresh and the regenerated catalysts exhibit the same activity and
selectivity for 1-butanol conversion. Also the XRD patterns of the
LaMnO3/La–Al2O3 material before and after catalytic test do not show
noticeable differences (Fig. 3). LaMnO3 is very well dispersed on the
surface of La-stabilized alumina and only weak signals of lanthanum
manganite are visible on the XRD pattern of both fresh and used
material.
Properties of the LaMnO3/La–Al2O3 catalyst are presented and
discussed elsewhere [31]. Here it is worth to remind that this material
possesses relatively high specific surface area (46.0 m2/g), concen-
tration of basic and acidic sites 0.22 and 0.55 mmol/m2, respectively.
Results of cyclohexanol decomposition on this material show that
dehydrogenation of cyclohexanol to cyclohexanone proceeds in
higher extent than dehydration to cyclohexene (CHON/CHEN selec-
tivity ratio 3.1) which suggests that basic sites are more reactive than
acidic ones. This has significant implications for dehydrogenation.
Nevertheless, the acid centers also play an important role in relation
to the secondary conversion of aldehydes to ketones [30]. From TPD
NH3 analyses, we found that the total acidity of LaMnO3/La–Al2O3 was
as high as 0.284 mmol NH3/g. For comparison the total acidity of La-
doped alumina carrier was measured and the value of 0.362 mmol
NH3/g was found. Higher acidity of carrier alone than catalyst
supported on this material results from participation of acidic sites
of alumina. These sites are then covered by less acidic mixed oxide
deposited on the carrier's surface. It is worth to note that
concentration of acidic sites determined for LaMnO3/La–Al2O3 by
TPD NH3 method is lower than concentration determined previously
by titration method. This difference is related with peculiar character
of each procedure. Nevertheless, acidic sites are more abundant than
basic ones.
Fig. 3. XRD patterns of LaMnO3 perovskite supported on stabilized alumina before and
after test and regeneration.
compounds were present in minor trace amounts, as it was confirmed
by thin-layer chromatography. The best results of tests performed at
400 and 420 °C are shown in Table 1. It was found that at lower or
higher reaction temperatures, conversion or selectivity is lower than
these ones obtained at 400–420 °C. The semi-product, resulting from
the reaction at 380 °C contained over 25% of unreacted alcohols. After
the test at 420 °C, at the full conversion, the ketones efficiency
dropped to 53.6%. The brownish product contained 46.4% of
hydrocarbons and other unidentified compounds resulting from the
cracking and the other side transformations [19,21,23].
The transformation follows the general reaction scheme [19,21]:
Thus, this transformation causes the loss of one carbon atom in
relation to two molecules of alcohols, regardless of the length of the
chain. Therefore, the atom economy of this method increases with
increasing alcohol's carbon chain length. This difference in the carbon
balance has been taken into account in the calculations below.
Since, during ketonization of equimolar mixture of primary
alcohols RCH2OH and R'CH2OH, three ketones are produced in the
quantitative relations resulting from the probability:
RCOR : RCOR' : R'COR'=1/4 : 1/2 : 1/4 [21].
In this event, the average molecular weight of the mixture of
ketones C11COC11, C11COC13 and C11COC13 is equal to 366 Da.
As the molar ratio of alcohols C12:C14 in the raw material
was =3:1, one can assume that the probability of quantitative
relations of received ketones will be close to:
RCOR : RCOR' : R'COR'=5/8 : 1/4 : 1/8.
The expected average molecular weight of the mixture of so
obtained ketones is 352 Da.
If we assume full conversion and selectivity only to ketones, the
3.14 g/h of raw material (including 1.57 g/h of alcohol fraction and
Table 3
Acid strength distribution determined by TPD NH3.
Sample
Concentration of acid centers with equal strength, %
473–
523–
573–
623–
673–
723–
773–
The share of acid strength distribution is shown in Table 3. It can be
clearly observed that deposition of 30 wt.% of LaMnO3 develops
mainly weak and medium acid sites (desorption temperature
Td =473–673 K) at the charge of strong acid sites (Td N773 K).
523 K
573
623
673
723
773
823
La–Al2O3
LaMnO3/La–Al2O3
4.0
5.2
9.1
14.9
18.5
20.9
20.4
20.9
20.1
17.2
15.3
13.5
12.6
7.3