532
A. Piegsa et al.
80
70
60
50
40
30
20
10
0
high selectivity for ring opening products. The group of
Gault et al. [18] carried out experiments on the ring
opening of methylcyclopentane over platinum catalysts.
They presented a correlation between the metal dispersion
on the catalyst surface and the product distribution as well
as the reaction mechanism. It was shown that the product
distribution was non-selective on 0.2 % Pt/Al2O3 with
highly dispersed particles. The possible ring opening
products n-hexane, 2-methylpentane, and 3-methylpentane
were formed with the same probability. When using 10 %
Pt/Al2O3 with low dispersed metal particles, the ring was
only opened between bisecondary (non-substituted) carbon
atoms leading predominantly to the formation of 2-meth-
ylpentane (78 %) and not of n-hexane. Hence, Gault
proposed three different reaction mechanisms. First, the
non-selective mechanism (also known as the multiplet
mechanism). Here methylcyclohexane is supposed to
adsorb planar on the metal surface. The bisecondary and
tertiary–secondary bonds are broken with the same prob-
ability. The second mechanism proposed is called the
selective mechanism (dicarbene mechanism), where the
hydrocarbon adsorbed perpendicularly on the metal surface
via two adjacent bisecondary carbon atoms. Due to steric
hindrance the tertiary–secondary bonds are not cleaved. A
third mechanism, the ‘‘partially selective’’ mechanism,
which competes with the dicarbene mechanism, was pro-
posed to explain the product distribution, where mainly
bisecondary, but sometimes also tertiary–secondary bonds
are broken. The methyl group and a methylene group in
a-position interact with a metal site forming a metallo-
cyclobutane intermediate. This results in cleavage of
tertiary–secondary bonds [2].
n-Paraffins
3-Methyl-
Paraffins
2-Methyl-
Paraffins
Naphthenes
-10
-20
Aromatics
5
6
7
8
9
Number of carbon atoms
Fig. 1 Cetane number of different hydrocarbons versus the number
of carbon atoms
All the above mentioned cases follow the approach to
use (sometimes poorly marketable) aromatics as feedstock
for conversion into valuable intermediates/products.
Before aromatic hydrocarbons can be ring opened to
paraffins they have to be completely hydrogenated to their
corresponding naphthenes [8]. Hydrogenation of aromatics
is well described in the literature [11–14], and it is known
that group-VIII metals (Pt, Pd, Rh…) are suitable catalysts
for hydrogenations [12].
In principle, ring opening of naphthenic hydrocarbons
can be achieved catalytically by cracking on solid acidic
catalysts, hydrocracking on bifunctional catalysts or hy-
drogenolysis on noble metals.
On solid acid catalysts (e.g. zeolites), ring opening of
naphthenes is achieved by a protolytic cracking mechanism
via carbenium intermediates. Due to side reactions and
consecutive cracking reactions these catalysts are not
appropriate to yield high cetane products [2].
Other research groups showed that the size of metal
particles is less important when using iridium or rhodium.
But these metals have higher activity and selectivity to
cleave the bisecondary bonds of the C5-ring [2, 3]. Fur-
thermore it was shown by Resasco et al. [19] that the
dispersion of metal particles is less important for the
reaction mechanism taking place by converting 1,2- and
1,3-dimethylcyclohexane over different Iridium catalyst.
However, they showed the type of support influences the
ring opening mechanism strongly.
Already in 1953, Mills et al. [15] proposed a mechanism
for the hydrocracking of naphthenes over a bifunctional
catalyst based on experiments with cyclohexane, cyclohex-
ene, methylpentane, and methylpentene. The mechanism
proposes three steps [2, 15]: At first, cyclic hydrocarbons are
dehydrogenated to cyclic olefins on a metal site. Secondly,
the olefins are protonated on the acid site, which leads to
formation of carbenium ions, which are either cracked or
skeletally isomerised. The olefins formed after isomerisation
are desorbed from the acid site and finally hydrogenated on
the metal site. This mechanism and the intermediates were
confirmed by Weisz and Swegler [16].
In this work, hydrogenation and ring opening reactions
over commercial noble metal catalysts (Ir, Pt, Rh) were
studied with different model aromatic and naphthenic
hydrocarbons as well as with ‘‘real’’ naphtha. Beside the
kinetics, the impact on the cetane number was analysed.
The ring opening of naphthenic hydrocarbons can also
occur on noble metal catalysts without acidic sites (hy-
drogenolysis). Galperin et al. [17] inhibited the activity of
the acid sites of a bifunctional platinum catalyst by addi-
tion of potassium. They showed that the catalyst acted then
like a monofunctional metallic catalyst, which enabled a
2 Experimental
All experiments were carried out in a continuously operated
fixed bed reactor (Fig. 2), designed to operate at
123