Production of Cycloalkanes in Hydrodeoxygenation of Isoeugenol Over Pt- and Ir-Modified Bifunctional Catalysts

Article abstract of DOI:10.1002/ejic.201800391

Hydrodeoxygenation of isoeugenol was investigated at 200 °C under 3 MPa total pressure in dodecane as a solvent, in hydrogen, over bifunctional Pt- and Ir-modified Beta zeolites and mesoporous materials. As a comparison, Pt and Ir supported on Al₂O₃, SiO₂ and mesoporous MCM-41 were also tested. The catalysts were characterized by XRD, CO pulse chemisorption, transmission electron microscopy, scanning electron microscopy, nitrogen adsorption and FTIR pyridine adsorption desorption. The results revealed that the most active and selective catalyst was Pt-H-Beta-300, which exhibits the lowest acidity and largest crystal size of Beta zeolite among the studied Pt- and Ir-modified Beta zeolites. Complete conversion of isoeugenol and 89 % selectivity to propylcyclohexane was obtained with this catalyst in 240 min. The overall deoxygenation selectivity was 100 %, giving dialkylated cyclohexanes as the second major product. The catalyst was regenerated, reduced and reused in the hydrodeoxygenation of isoeugenol with almost the same performance as the fresh catalyst. Thermodynamic analyses and kinetic modelling of the data were also performed.

Full text of DOI:10.1002/ejic.201800391

A Journal of
Accepted Article
Title: Production of cycloalkanes in hydrodeoxygenation of isoeugenol
over Pt- and Ir-modified bifunctional catalysts
Authors: Louis Bomont, Moldir Alda-Onggar, Vyacheslav Fedorov, Atte
Aho, Janne Peltonen, Kari Eränen, Markus Peurla, Narendra
Kumar, Johan Wärnå, Vincenzo Russo, Päivi Mäki-Arvela,
Henrik Grenman, Marina Lindblad, and Dmitry Murzin
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201800391
Link to VoR: http://dx.doi.org/10.1002/ejic.201800391

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Production of cycloalkanes in hydrodeoxygenation of isoeugenol over Pt-
and Ir-modified bifunctional catalysts
a
a
a
a
b
Louis Bomont , Moldir Alda-Onggar , Vyacheslav Fedorov , Atte Aho , Janne Peltonen ,
a
c
a
a
d
Kari Eränen , Markus Peurla , Narendra Kumar , Johan Wärnå , Vincenzo Russo, Päivi
a
a
e
a*
Mäki-Arvela , Henrik Grénman , Marina Lindblad , Dmitry Yu. Murzin
aJohan Gadolin Process Chemistry Centre, Åbo Akademi University, Turku/Åbo, Finland
bLaboratory of Industrial Physics, University of Turku, Turku, Finland
cLaboratory of Electron Microscopy, University of Turku, Finland
d
Università di Napoli Federico II. Via Cinthia, 4. IT-80126 Napoli, Italy
e
Neste Corporation, Porvoo, Finland
*
corresponding author: dmurzin@abo.fi, web page: http://users.abo.fi/dmurzin/
Hydrodeoxygenation of isoeugenol was investigated at 200°C under 3 MPa total pressure in
dodecane as a solvent in hydrogen over bifunctional Pt- and Ir-modified beta zeolites and
mesoporous materials. As a comparison, also Pt- and Ir-supported on Al
2
O
3
, SiO and
2
mesoporous MCM-41 were tested. The catalysts were characterized by XRD, CO pulse
chemisorption, transmission electron microscopy, scanning electron microscopy, nitrogen
adsorption and FTIR pyridine adsorption desorption. The results revealed that the most active
and selective catalyst was Pt-H-Beta-300 exhibiting the lowest acidity and largest crystal size
of Beta zeolite among the studied Pt- and Ir-modified Beta zeolites. Complete conversion of
isoeugenol and 89% selectivity to propylcyclohexane was obtained with this catalyst in 240
min. The overall deoxygenation selectivity was 100% giving also dialkylated cyclohexanes as
the second major product. This catalyst was also regenerated, reduced and reused in
hydrodeoxygenation of isoeugenol showing nearly the same performance as the fresh one.
Termodynamic analyses and kinetic modelling of the data were also performed for
hydrodeoxygenation of isoeugenol.
1
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Introduction
Pyrolysis of biomass and lignin extraction have been intensively studied for production of bio-
oil and lignin fractions. [1-4] Phenolic compounds contain, however, still large amounts of
oxygen not being suitable for a direct use as fuels, thus these compounds have been
deoxygenated over various catalysts. Hydrodeoxygenation (HDO) of phenolic compounds is
not very easy due to strong adsorption of multioxygenated compounds on the catalyst surface
causing also coking and catalyst deactivation. Moreover, repolymerisation of phenolic
compounds at higher reaction temperatures is still feasible resulting in lower catalyst activities
and a lack of the liquid phase mass balance closure determined by GC. [5] In order to produce
fully deoxygenated hydrocarbons, bifunctional catalysts are needed exhibiting both metal and
acidic function.
Several model compounds, such as guaiacol, [5-13] anisole, [12, 14-15] and phenol [16] have been
intensively used in HDO using different catalysts. HDO of eugenol has, however, been less
[
17-27]
intensively investigated.
dihydroeugenol over Pt-alumina silicate at 250°C under 3 MPa, whereas more strongly acidic
catalysts, such as Ni-SiO -ZrO gave fully deoxygenated propylcyclohexane as the main
It has already been shown that the main product was
2
2
product at 300°C under 5 MPa total hydrogen, when a mixture of phenol, cresol, eugenol, trans-
anethole, vanillin and guaiacol was used as a feedstock. [19] Several very recent works in eugenol
HDO show clearly that acidity is required for production of hydrocarbons [20, 21] and high
temperatures using Ru/C as a catalyst. [22] Higher metal dispersions also promoted HDO as was
the case with different Ru-supported catalysts. [23] In another work [18] a physical mixture of
Pd/C with alkali leached mesoporous H-ZSM-5 gave in HDO of eugenol 73% hydrocarbons at
8
7% conversion at 250°C under 5 MPa hydrogen after 4 h showing a need of acidity.
Hydrodeoxygenation of eugenol was also investigated over Raney nickel together with H-ZSM-
5
at 220°C under 0.5 MPa nitrogen in methanol – water mixture. The main product was
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
propylcyclohexane with the yield of 86.1% after 7 h. [20] Nickel phosphide supported on HZSM-
was giving at 250°C 60% conversion of eugenol after 2 h. The degree of product
5
deoxygenation was 44% with propylcyclohexane being the main product. [24] Cobalt catalyst
supported on HZSM-5 was active in eugenol HDO giving 100% conversion and 62% selectivity
towards propylcyclohexane at 200°C under 1 MPa hydrogen in 2 h, whereas Co/CeO resulted
2
in only 14 % of this product. [25] High yields of propylcyclohexane, 90% were also reported in
eugenol HDO over Ru/CNT catalyst at 220°C under 5 MPa hydrogen in 3 h using undecane
water mixture as a solvent. [26] Ru/HZSM-5 as a catalyst gave 83% yield of propylcyclohexane
in water at 170°C under 4 MPa hydrogen in 4 h. In this case the carbon balance was stated to
be 91%. [21] On the other hand, Ru/MW-CNT produced only 28% propylcyclohexane when
eugenol was transformed under 4 MPa hydrogen at 270°C in water as a solvent after 1 h. [23] It
should be mentioned that the reaction time was, however, short and thus this result is not fully
comparable with that of Wang et al. [21] Ru/C was used as a catalyst in eugenol HDO with
hexadecane as a solvent showing that temperature above 275° C favored deoxygenation. [22] In
minor amounts 1-methyl-3-ethylcyclohexane [20] and products of ring contraction such as
alkylated cyclopentane [13] have been reported. In addition, skeletal isomerization of the side
chain in main product is also possible forming isopropylcyclohexane.
Application of expensive noble methods with low loading is often done in oil refining for
such processes as catalytic reforming, isomerisation and hydrocracking. Precious metal-
containing catalysts are also used in production of bulk and specialty chemicals. Despite high
costs, superior catalytic performance of noble metals in comparison with non-precious
counterparts as well as established technologies for the recycling of the spent catalysts justify
their application in different industrial catalytic processes ensuring overall profitable
performance.
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
The aim of this work was to investigate kinetics in HDO of isoeugenol over bifunctional
Pt- and Ir-modified H-Beta catalysts with different acidity and compare their performance with
Pt- and Ir-supported on MCM-41, since mesoporosity can have beneficial effects in HDO of
phenolic compounds. [18] Ir can also be beneficial in HDO, since it is known to exhibit
hydrogenolysis activity. [28] In addition mildly acidic Pt/Al
O
and non-acidic Ir/SiO were also
2
3
2
studied in order to elucidate the role of acidity, support structure and metal in HDO of eugenol.
For the best Pt-H-Beta-300 catalyst, also catalyst reuse after regeneration and reduction was
successfully demonstrated. Kinetics of (iso)eugenol hydrodeoxygenation has been scarcely
investigated.[22,
25,
26]
In the current work thermodynamic analyses for isoeugenol
hydrodeoxygenation was performed using a thermodynamic approach [29] and the Gibbs-
Helmholtz equation. The intention was to use the reaction network shown in Scheme 1, which
was supported by the literature data, and to take into account reaction stoichiometry in the liquid
phase. Furthermore, kinetic modelling of HDO of isougenol was also made in the current work.
According to our knowledge kinetic modelling of HDO of eugenol has only been made in ref
[22],
where also mass transfer limitations and hydrogen solubility were considered.
Results and discussion
Catalyst characterization results
XRD results showed for Pt-H-Beta-300 and Ir-H-MCM-41 both X-ray powder diffraction
patterns clearly showed that both Pt [32] and IrO
[33]
were observed in the diffractograms in
2
addition to Cu [34] in some cases originating from the sample holder due to a low amount of
catalyst. Furthermore, X-ray powder diffraction patterns clearly showed that the structure of H-
Beta-300 (Figure 1) remained intact after loading of Pt and the same is valid for Ir-HMCM-41
(
Figure S1). [29-36]
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Pt-H-Beta-300 catalyst exhibited the highest specific surface area (Table 1). The spent
Pt-H-Beta-300 catalyst, after its use in hydrodeoxygenation of isoeugenol showed almost the
2
3
same surface area (681 m /g) and the pore volume (0.24 cm /g). The reason for a minor decrease
in the surface area for the spent Pt-H-Beta-30 catalyst as compared to the fresh Pt-H-Beta-300
catalyst can be attributed to coke deposition in the pores of the catalyst, although the pore
volume remained constant. Pt-H-Beta-300 spent zeolite catalyst was regenerated in an oven
using the step calcination procedure for 120 min. The regenerated Pt-H-Beta-300 catalyst
2
exhibited the surface area of 650 m /g. The decrease in the surface area of the regenerated
catalyst as compared to the spent could be due to the generation of mesoporous during the
calcination procedure. However, it should be mentioned here that the obtained surface area and
2
pore volume for the Pt-H-Beta-300 regenerated is higher than that of Pt-H-Beta-150 (621 m /g)
2
and Pt-H-Beta-25 (592 m /g) fresh catalysts (Table 1).
2
The Pt-H-MCM-41 catalyst also exhibited a high specific surface area (679 m /g) and a
2
large pore volume (0.45 m /g) (Table 1). The specific surface areas of Pt- and Ir-supported on
silica and alumina were the lowest ones for all studied catalysts.
The results from the acidity measurements of Pt- and Ir-modified zeolites and mesoporous
materials are shown in Table 2. The amount of Brønsted acid sites measured for pyridine
desorbed at 250 °C indicates the sum of weak, medium and strong acidic sites. The highest
amount of Brønsted acid sites was observed in Pt-H-Beta-25, which was more than four fold
that of Pt-H-Beta-150 and nearly 9 fold compared to Pt-H-Beta-300. These results are in line
with Si/Al ratio. Mesoporous Pt-H-MCM-41 and Ir-H-MCM-41 exhibited mild acidity, which
is expected from MCM-41 in comparison with zeolite beta. Noteworthy is that the Brønsted
acidity of Ir-H-Beta-150 was 1.6 fold higher than that of Pt-H-Beta-150. These catalysts did not
contain any strong Brønsted acid sites (Table 2).
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
The highest amount of Lewis acid sites was also measured for Pt-H-Beta-25. Interestingly
the second and third highest amounts of Lewis acid sites were determined for Pt-H-MCM-41
and Ir-H-MCM-41.
The scanning electron microscopy was used to study the shape, size and distributions of
the crystals of Pt- and Ir-modified catalysts (Figure 2, 3, S2, S3). The Pt-H-Beta-300 exhibited
spherical shaped crystals with the size ranging from 200 – 600 nm and the spent particles
resemble the fresh ones (Figure 2). The Pt-H-Beta-150 and Pt-H-Beta-25 exhibited smaller
crystal sizes (Figure 3), than the Pt-H-Beta-300. It is important to mention that the Pt- and Ir-
modified H-Beta-300, H-Beta-150 and H-Beta-25 zeolite catalysts maintained the crystal
morphology after the modification with hexachloroplatinic and iridium chloride solutions.
The Pt-H-MCM-41 mesoporous catalysts showed similar crystal morphology (Figure
S2a) as the pristine H-MCM-41 (not shown here), indicating that the modifications of the
mesoporous support using hexacloroplatinic acid and iridium chloride did not change the basic
structure. The size of Ir-H-MCM-41 was ranging from 100 nm to 1000 nm (Figure S2b).
Pt-, Ir- particle size, structures of pores and periodicity of the pores for Pt-H-Beta-25, Pt-
H-Beta-150, Pt-H-Beta-300, Ir-H-Beta-150, Pt-H-MCM-41 and Ir-H-MCM-41 catalysts, both
0
fresh and used in hydrodeoxygenation of isoeugenol, at 200 C and 3 MPa, were also
characterized by transmission electron microscopy. TEM images revealed that the fresh and
spent Pt- and Ir-modified catalysts exhibited similar structures to that of pristine Beta and
MCM-41 catalysts (Figures 4-6, S4-5). The modifications of the microporous Beta zeolite and
mesoporous MCM-41 with hexachloroplatinic acid and iridium chloride did not influence the
parent structures of the supports. Similar observations were obtained in characterization of Pt-,
Ir- modified Beta and MCM-41 mesoporous catalysts using X-ray powder diffraction. Pt
particles were spherical, square and triangular, with the majority of them being spherical.
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Pt-particle size of Pt-H-Beta-300 fresh calculated from the histograms was in the range
-35 nm (Figure 4). The same catalyst after use in hydrodeoxygenation of isoeugenol exhibited
5
the Pt- particle size in the range 5-40 nm, which is a sign of minor sintering. An average Pt-
particle size calculated from the histogram was 11.1 nm.
Pt-H-Beta-300 spent catalyst was regenerated in an oven for 120 min followed by its
reuse in hydrodeoxygenation of isoeugenol. Pt-particle sizes were in the range of 5-35 nm
(Figure 4e and f), close to that of fresh Pt-H-Beta-300, clearly indicating that the spent catalyst
after hydrodeoxygenation of isoeugenol can be regenerated and that the Pt-particle size was
only slightly increased. In addition, the regenerated Pt-H-Beta-300 exhibited similar activity
as the fresh one as explained below.
The particle size distributions of Pt-H-Beta-150 is given in the histogram of Figure 5. It
is significant to mention that the Pt-particle size distribution of this catalyst was 5-10 nm for
the fresh and spent catalysts, which is much lower than that of Pt-H-Beta-300. The latter was
in the range of 5-35 nm. The plausible explanation for lower Pt-particle size distributions in Pt-
H-Beta-150 catalyst is attributed to a smaller size of Beta-150 (50 -300 nm). The average Pt-
particle size of Pt-H-Beta-150 was 6.2 nm (Figure 5).
The fresh and spent Pt-H-Beta-25 catalysts exhibited the Pt- particle sizes between 3-6
nm (Figure 6), with the average Pt-particle size of 3.4 nm. This catalyst showed the smallest
average Pt- particle size of the three Beta catalysts studied. The average Pt- particle sizes for
the fresh Pt- modified Beta zeolite catalyst were Pt-H-Beta-300 (11.1 nm) > Pt-H-Beta-150 (6.2
nm) > Pt-H-Beta-25 (3.6 nm). From these results, it can be concluded that the crystal size of
Beta zeolite and acidity (Brønsted and Lewis) influenced the Pt-particle size. The largest crystal
size and the lowest amount of Brønsted acid sites (Table 2) in Beta-300 resulted in the largest
average Pt- particle size (11.1 nm). Analogously the high density of the surface hydroxyl
7
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
groups and interaction of metal- metal-oxide supports have been suggested as the reason for the
decreased Rh particle size in Rh/Al
2
O
3
/SiO
2
compared to Rh/SiO with the Rh particle sizes
2
being 1.2 nm and 2.7 nm, respectively. [37]
The metal dispersion determined by CO pulse chemisorption is shown in Table 1. The
highest metal dispersion was obtained in Pt/Al followed by Pt-H-Beta-25, whereas the
lowest dispersions were determined for Ir/SiO and Ir-H-Beta-150. The Pt particle sizes
2
O
3
2
correlated well with TEM data for Pt-H-Beta-150 and Pt-H-Beta-300, except Pt-H-Beta-25
displaying a higher value according to CO chemisorption (Table 1).
The thermogravimetric analysis of the fresh and spent catalysts (Figure 7) was performed
to determine the amount of coke present in the spent catalysts after hydrodeoxygenation of
isoeugenol. Characterization of coke by TGA analysis (Figure 7) showed 3.6 wt% weight loss
from the spent Pt-H-Beta-300 catalyst in the temperature range of 100 -1000 °C after utilization
of this catalyst in isoeugenol HDO at 200°C under 3 MPa. The weight loss for the fresh catalyst
was 1.9 wt% indicating that 1.7 wt% of organic material was adsorbed on the catalyst surface. The
corresponding weight losses for the fresh and spent Pt-H-Beta-25 were 6% and 11.5 %,
respectively, indicating that 5.5 wt% organic material was accumulated on a more acidic catalyst.
Thermodynamic analysis of isoeugenol hydrodeoxygenation
Enthalpy (ΔH0
) and Gibbs free energy (ΔG0
) at standard conditions were calculated
) and Gibbs
r
r
[
29]
starting from the standard enthalpy (ΔH0
using a thermodynamic approach,
free energy (ΔG0
ChemCAD v.5.0 [37] or estimated with the Joback approach, [39]
f
f
) of formation from the elements derived from the database included in
0
0
f ,i
∆H
=
ν ⋅ ∆H
r, j ∑j i, j
(1)
8
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
0
0
f ,i
∆G
=
ν ⋅ ∆G
r, j ∑j i, j
(2)
The equilibrium constant of each reaction was calculated from
ꢈ
∆
ꢅ
ꢀꢁ0
= ꢂꢃꢄ �−
ꢆ,ꢇ
ꢋ
(3)
ꢉꢊ
The dependency of the reaction free Gibbs energy with temperature was included by
implementing the Gibbs-Helmholtz equation valid at P=0.1 MPa (ΔGΦr,j):
φ
r, j
0
r, j
∆
G (T) ∆G
 1 1 
T T 
0
r, j
=
+ ∆H  −

(4)
0
0
T
T
ꢌ
ꢀ
ꢍ
Finally, to calcula2te the ΔGr,j at different pressures Eq. (5) was implemented.
=
ꢎ
 P 
G (P) = ∆G + nRT ln
φ
r, j
∆

(5)
r, j
0

P 
in which n is moles. In thermodynamic modelling the reactions shown in Table 3 and
Scheme 1 were modelled. The stoichiometry used in Table 3 was applied. The calculated
enthalpy and Gibbs free energy formation for each component (i) are reported in Table S1. The
data were retrieved from ChemCAD v.5.0 database directly.
Starting from these values, the enthalpy and Gibbs free energy for each reaction (j) at
standard conditions, equilibrium constants at standard conditions (K0
), enthalpy and Gibbs free
j
energy at different temperatures and pressure were calculated. Two different temperatures were
investigated (T
1
=473.15 K, T
2
=573.15 K) and the energy values were calculated also at P =3
1
MPa.
Thermodynamic analyses shows that thermodynamically not feasible reactions for
propylcyclohexane are the steps 5, 6 and 9 (Table S2) corresponding to de-ethylation and
alkylation of propylcyclohexane forming 1-methyl-3-ethylhexane, isomerization forming
isopropylcyclohexane and its ring contraction forming 1-methyl-3-propylcyclopentane. In
9
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
addition, hydrogenation of the intermediate propylbenzene became unfeasible at a higher
o
o
temperature, 300 C compared to 200 C (step 4). Higher pressure favored demethoxylation of
dihydroeugenol and formation of methanol (step 2).
Catalytic results from the transformation of isoeugenol with Pt- and Ir- modified catalysts
Isoeugenol transformations over Pt- and Ir-modified beta zeolite catalysts
Isoeugenol transformation was investigated with Pt supported on alumina, silica, H-
MCM-41 and Beta zeolites with three different SiO
2
to Al
2
O
3
ratios (Table 4). The
transformation of isoeugenol is very rapid with Pt-Beta zeolite catalysts and thus TOF values
were calculated for the transformations of its hydrogenated product - dihydroeugenol within
the first 30 min. Interestingly the most active catalyst in hydrodeoxygenation was Pt-H-Beta-
3
00 (Figure 11, Table 4, entry 1) exhibiting the lowest Brønsted acidity from the studied Pt-H-
Beta catalysts (Table 2). The second most active catalyst was Pt-H-Beta-150, whereas TOF for
the most acidic catalyst Pt-H-Beta-25 was only 47% of that of Pt-H-Beta-150. These results
indicate that too high amount of acid sites is not beneficial for dihydroeugenol transformation.
Since the metal particle sizes vary only slightly in these three Pt-H-Beta zeolite catalysts, it was
also decided to perform experiments with Pt-H-Beta-25 calcined at higher temperatures in order
to increase the Pt particle size (see below).
Isoeugenol was hydrogenated very rapidly to dihydroeugenol and thus conversion of
dihydroeugenol was compared (Table 4, Figure 8) over three Pt-H-Beta catalysts with different
acidity. All these catalysts were active during the whole reaction time and high conversions
were obtained (Table 4, entries 1-3). The sums of the liquid phase masses of the reactant and
products determined by GC were calculated from the liquid phase. A trend of this sum decrease
with an increase of the amount of Brønsted acid sites follows from Figure 9. In fact, this sum
1
0
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
should be ca. 77%, if all dihydroeugenol is deoxygenated and demethoxylated to
propylcyclohexane and the formation of water and methanol is not taken into account. The sum
of the liquid phase masses of the reactant and products determined by GC (Table 4) for Pt-H-
Beta-150 and Pt-H-Beta-25 was lower than the theoretical mass balance. Furthermore, the sums
of liquid phase masses of the reactant and products determined by GC without
hydrodeoxygenation were ca. 100% when only hydrogenation occurred (Figure 9). Incomplete
liquid phase mass balance determined by GC can be in general explained by oligomerisation of
some products, as well as strong adsorption of organic compounds on the catalyst surface and
formation of gas phase products. Oligomers, however, could not be detected by GC analysis.
When considering GC results and by taking into account the accumulated organic material on
the catalyst surface, a slightly higher sum of liquid phase reactant and products determined by
GC was recorded for Pt-H-Beta-300 and Pt-H-Beta-25 (Table 4). Analogous of the data in the
current work and comparison with literature allows to conclude that for isoeugenol HDO at
1
70°C under 4 MPa hydrogen in water as a solvent over acidic catalysts, such as Ru/ZSM-5 [21]
a too high value of the sum of liquid phase masses of the reactant and products determined by
GC was reported (91.4%) as the main product was propylcyclohexane. Analogously the overall
product yields was reported to be 99.5% for isoeugenol HDO over Co/H-ZSM-5 under 1 MPa
hydrogen in dodecane as a solvent at 200°C although 87% of the products were demethoxylated
ones indicating that there might be some inconsistencies in the calculations of the sum of the
liquid phase masses for the reactant and products determined by GC. [25] In deoxygenation of
isoeugenol over Raney nickel (H-ZSM-5) at 220°C under nitrogen in a methanol- water mixture
the sum of calculated products was 96.4% at 97.2 % conversion. In addition, 89.1% of
demethoxylated products were formed, [20] which is too high if methanol was not taken into
account. Furthermore, the propylcyclohexane yield of 100% was reported in ref. [26] in HDO of
o
eugenol at 200 C under 2 MPa hydrogen over CoNx catalyst. These examples show that
1
1
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
quantitative analysis of the reaction mixture is challenging and that the data reported in the
literature should be critically evaluated.
For comparison of Pt- and Ir-catalysts, also Ir-Beta-150 was tested in isoeugenol
hydrodeoxygenation (Table 4, entry 4). This catalyst despite being more acidic than Pt-H-Beta-
1
50 exhibited a low TOF towards dihydroeugenol transformation, most probably due to its large
Ir-particle size. This comparison clearly shows that the metal dispersion has a significant role
in HDO in line with the literature. [23] Because of Ir-H-Beta-150 acidity (Table 2) also some
heavy products, such as dimeric isoeugenol and 2,2’-methylene-bis(1,1-dimethylethyl)-4-
methylphenol were observed by GC-MS (Figures S6, S7). It should be pointed out here, that
these compounds did not elute from the same GC column in quantitative GC analysis in this
study. Since Ir can promote according to the literature hydrogenolysis, [28] it was decided to
explore catalytic activity of highly dispersed iridium supported on non-acidic silica in
deoxygenation of dihydroeugenol.
The effect of the metal particle size in hydrodeoxygenation of isoeugenol was also
investigated using Pt-H-Beta-25 calcined at 650° C and 850 °C (Table 4, entry 5 and 6). As
expected, high temperature treatment was detrimental for catalytic activity and Pt-H-Beta-25
(850°C) exhibited a very low reaction rate compared to Pt-H-Beta-25 calcined at either 450°C
or 650°C, which can be explained by low acidity and a larger metal particle size (Tables 1, 2).
It is also interesting to note that the sum of liquid phase masses of the reactant and products
determined by GC was much lower for a more acidic Pt-H-Beta-25 (650°C), which exhibited a
large number of acid sites. At the same time this sum for Pt-H-Beta-25 (850°C) was 87%, as
the latter catalyst is only mildly acidic due to dealumination occurring for beta zeolite at higher
temperature breaking also the twelve member ring. [40] Such behavior can in part explain low
acidity of Pt-H-Beta-25 (850° C) (Table 2).
1
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
The degree of hydrodeoxygenation with three Pt-H-Beta zeolites was 100% showing that
bifunctional catalysts are required for this reaction. The main product and its selectivity after 4
h are shown for all catalysts in Table 4. Propylcyclohexane was the main product for Pt-H-
Beta-300, Pt-H-Beta-150, Pt-H-Beta-25 (Figure 8c) and the selectivity towards
propylcyclohexane increased also with decreasing the catalyst acidity (Table 4, entries 1-3,
Figure 8c). Dialkylated cyclohexanes were also formed as minor products using Pt-beta zeolites
as catalysts (Table 4). The GC-MS identification showed that most probably the main side
products over Pt-H-Beta-25 were isomers of 1-ethyl-2-methyl-cyclohexane (Figure S8). The
mass spectrum of exactly this molecule was not found in the mass spectra library, however,
data for two other isomers, namely trans-1-ethyl-4-methylcyclohexane and cis-1-ethyl-2-
methylcyclohexane are available in the NIST library. When comparing the MS of the product
formed over Pt-H-Beta-25, it can be clearly seen that it has the cleavage fragments with m/z at
1
11 and 97 corresponding to cleavage of methyl and ethyl groups. Furthermore, these fragments
are not found in propylcyclohexane fragments. It can be thus concluded that two main side
products over Pt-H-Beta-25 are dialkylated cyclohexanes containing nine carbon atoms.
Formation of these products includes breaking of the C-C bond in the propyl chain and
alkylation of the phenyl ring, whereas only alkylation is needed for formation of 1-methyl-4-
ethylcyclohexane. It should, however, be noted that 1-methyl-3-propylpentane exhibiting the
molecular mass of 126 g/mol, i.e. the same as for propylcyclohexane, was stated to be a product
in eugenol HDO over Ru/C.[22] Traces of propylbenzene were also formed over Pt-H-Beta-25,
indicating a slightly lower ring hydrogenation activity after prolonged reaction times. In
addition over this catalyst also 1-methyl-2-propylcyclopentane was formed in small amounts
by ring contraction of propylcyclohexane. Activity of Pt in ring contraction has been reported
previously. [13] The main product over Ru/H-ZSM-5 in isoeugenol HDO in water at 170°C
under 4 MPa hydrogen was propylphenol.[21] Both propylcyclohexanol and 1-methyl-3-
1
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
o
ethylcyclohexane were formed in HDO of eugenol at 200 C under 0.5 MPa hydrogen over
Raney nickel catalyst supported on H-ZSM-5 [20] analogously to the current results. Noteworthy
is also that propylcyclohexanol and propylphenol were the main products in HDO of eugenol
over Co-H-ZSM-5 at 200°C under 1 MPa hydrogen and no propylcyclohexane was formed
despite of the reported acidity of Co-H-ZSM-5. [25] Kinetics of isoeugenol HDO over Pt-H-
Beta-300 and Pt-H-Beta-25 is shown in Figure 10 and will be discussed below.
Experiment with Pt-H-Beta-25 calcined either at 650°C or 850 °C showed the further
transformations of isoeugenol were slow and the main product was dihydroeugenol (Figure 11).
When comparing the current results with the data in [22] when HDO of eugenol was performed
over Ru/C at 270°C in hexadecane as a solvent, it can be noticed that the main products in ref.
[
22]
were dihydroeugenol and propylphenol in line with the current results. Zeolites are more
acidic than Ru/C and thus more propylcyclohexane is formed in the current work.
Isoeugenol transformation over Pt- and Ir-modified MCM-41 catalysts
Isoeugenol was hydrogenated rapidly to dihydroeugenol over Pt-H-MCM-41 and Ir-H-
MCM-41 despite large metal particle sizes (Figure S9). Further transformations of
dihydroeugenol were very slow over Ir-H-MCM-41, whereas over Pt-H-MCM-41 the
selectivity of propylcyclohexane was 35% already after 4 h (Table 4, entry 7). Noteworthy is
also that over Pt-H-MCM-41 with mesopores the selectivity of propylcyclohexane was 1.9 fold
higher than with microporous Pt-H-Beta-25 (650 °C) exhibiting four fold higher Brønsted
acidity. This result indicates that mesoporosity is beneficial for hydrodeoxygenation of
isoeugenol facilitating easier product diffusion from the pores to the bulk liquid. This result
correlates with the kinetic diameter of eugenol equal to 8.97 Å calculated using Chemsketch
[
18]
which is larger than the cavities of beta zeolite exhibiting channels with a cross section of
.6 Å * 6.7 Å and a sinusoidal channel with a cross section of 5.6 Å * 5.6 Å. [42] The pores of
6
1
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
MCM-41 are typically quite large, for example for Al-MCM-41 with Si/Al ratio of 5 the pores
with the size of 3.3 nm were found. [41] Zhang et al. [18] also confirmed that mesoporosity is
beneficial for eugenol hydrogenation.
Isoeugenol transformations over Pt- and Ir-catalysts supported on silica and
alumina
Isoeugenol hydrodeoxygenation was investigated in this work using Pt/Al
2
O
3
bearing
Lewis acidity as well as very mildly acidic Pt/SiO and Ir/SiO catalysts (Table 4, entries 7-9).
2
2
Hydrogenation of isoeugenol per se proceeded rapidly over these catalysts showing also that
further reactions of dihydroeugenol were, as expected, of minor importance due to the non-
acidic character of the catalysts. The main product was thus always dihydroeugenol with a very
high selectivity.
Main parameters affecting hydrodeoxygenation selectivity
Effect of the amount of Brønsted acid sites on selectivity to the main product,
propylcyclohexane after 4 h reaction time in hydrodeoxygenation of isoeugenol is summarized
in Figure 12 showing clearly that a high amount of Brønsted acid sites is not the only parameter
influencing selectivity to propylcyclohexane. One example is a relatively high acidity of Pt-H-
o
Beta-25 calcined at 650 C, which afforded only 18% selectivity towards propylcyclohexane.
For this catalyst a lower HDO activity is explained by a large Pt particle size not favoring even
HDO of the formed dihydroeugenol (Table 4, entry 5).
Reuse of Pt-H-Beta-300
Reuse of the calcined Pt-H-Beta-300 catalyst in dehydrodeoxygenation of isoeugenol
(Figure 13) shows clearly that it is possible to recover the catalytic activity of this catalyst nearly
completely even if the pore volume of the regenerated catalyst was only 78% of the fresh one.
1
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Kinetic modelling
Kinetic modelling of isoeugenol hydrodeoxygenation was performed for three Pt-H-Beta
catalysts with different acidities using pseudo first order simplified kinetic equations. The
reaction scheme is shown in Figure 14. The intermediate C has only been confirmed in the case
of Pt-H-beta-300 exhibiting the mildest acidity, whereas with more acidic catalysts its further
reaction is so rapid, that it has not been identified in the reaction products
Rate equations for Pt-H-Beta-300 have been defined as:
ꢎ = ꢏ ꢐ
(6)
1
1 ꢑ
ꢎ = ꢏ ꢐ
(7)
2
2 ꢒ
ꢎ = ꢏ ꢐ
(8)
3
3 ꢓ
ꢎ = ꢏ ꢐ
(9)
4
4 ꢔ
ꢎ = ꢏ ꢐ
(10)
5
5 ꢑ
ꢎ = ꢏ ꢐ
(11)
7
7 ꢒ
where k
i
correspond to the rate constants of the steps from Figure 14. The concentrations
of products p
1
, p
2
p
3
and p have been lumped, due to their small amounts. Their generation is
4
described by r
7
.
The corresponding mass balances for the batch reactor are:
ꢕꢖꢗ
=
=
=
=
=
=
−ꢎ − ꢎ − ꢎ
6
(12)
(13)
(14)
(15)
(16)
(17)
1
5
ꢕꢘ
ꢕꢖꢙ
ꢕꢘ
ꢎ − ꢎ − ꢎ
6
1
2
ꢕꢖꢚ
ꢕꢘ
ꢎ − ꢎ + ꢎ − ꢎ
8
2
3
4
ꢕꢖꢛ
ꢕꢘ
−ꢎ − ꢎ − ꢎ
9
3
4
ꢕꢖꢜ
ꢕꢘ
ꢎ5
ꢕꢖꢝ
ꢕꢘ
ꢎ + ꢎ + ꢎ + ꢎ
6
7
8
9
1
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
For more acidic catalysts, i.e. Pt-H-Beta-25 and Pt-H-Beta-150 more simplified rate
equations have been used:
ꢎ = ꢏ ꢐ
(18)
(19)
(20)
(21)
1
1 ꢑ
ꢎ′ = ꢏ′ ꢐ
2
2 ꢒ
ꢎ = ꢏ′ ꢐ
5
5 ꢑ
ꢎ′ = ꢏ′ ꢐ
7
7 ꢒ
The differential equations were solved with the backward difference method as a subtask to the
parameter estimation with simplex and Levenberg-Marquardt methods. The numerical tools are
built into the used software Modest, [43] in which the objective function was defined as
2
ꢞ = ∑(ꢟ − ꢟꢡ )
(22)
ꢠ
ꢠ
2
and the degree of explanation R is defined as
ꢦ
∑
(ꢣ ꢥ ꢣꢡ )
ꢤ ꢤ
2
ꢢ = 1 −
(23)
ꢦ
∑
(ꢣ ꢥ ꢣꢧ )
ꢤ
ꢤ
in which ꢟꢧ
in Table 5 and the corresponding data fitting results are given in Figure 10. Overall the
parameters were well defined, apart from k ’ for Pt-H-Beta-25, for which quite high
correlation with k and k’ was observed (Table S3). The degree of explanation was for Pt-H-
R
i
is the mean value of observations. The parameters for different catalysts are shown
5
1
2
Beta-300 98.36%, whereas it was for Pt-H-Beta-25 95.63%. It can be concluded that the
proposed models could describe the kinetic data well.
In addition to the model presented in Fig. 14, an alternative model shown in Scheme 1 was
tested. This model implies first demethoxylation of dihydroeugenol followed by
hydrogenation of the phenyl ring in 4-propylphenol, because dehydration of OH-group from
the aromatic ring might be less probable. Thereafter, dehydroxylation of 4-propylcyclohexanol
should proceed rapidly and it is possible that propylbenzene is formed in the presence of
o
hydrogen even at 200 C via dehydrogenation of propylcyclohexane. The thermodynamics of
1
7
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
this reaction is not favorable as the logarithm of the equilibrium constant of a gas phase
o
hydrogenation reaction (i.e. reverse to dehydrogenation) at 200 C is ca. 4. Nevertheless,
kinetic modelling cannot discriminate between these two models as the results using the
alternative model gave an also an adequate description.
1
8
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Conclusions
Pt- and Ir-modified H-Beta-25, H-Beta-150 and H-Beta-300 microporous zeolites, structured
MCM-41 with mesopores and SiO were synthesized by an evaporation impregnation method.
2
Pt-H-Beta-300 catalyst with the smallest amount of Brønsted acid sites exhibited with the
largest crystal size and Pt size.
The highest conversion in hydrodeoxygenation of isoeugenol and selectivity to the desired
cycloalkanes were obtained over Pt-H-Beta-300 catalyst. The highest amount of Brønsted acid
sites was present in Pt-H-Beta-25 followed by Pt-H-Beta-150, Pt-H-MCM-41 and Pt-H-Beta-
3
00.
With the best catalyst the selectivity to propylcyclohexane was 89% at complete
conversion of isoeugenol. When more acidic Pt-H-Beta zeolites were applied, the selectivity
towards propylcyclohexane decreased at the same time increasing the amount of dialkylated
cyclohexanes indicating cleavage of propyl group and alkylation of the cyclohexane. The sum
of the liquid phase masses of the reactant and products determined by GC decreased with
increasing catalyst acidity due to formation of oligomers as well as formation of gaseous
products and accumulation of organic material on the catalyst surface.
Pt-H-Beta-25 calcined at 650°C with lower metal dispersion, but still exhibiting high
acidity, was giving dihydroeugenol as the main product. This result indicates that high metal
dispersion and certain acidity are required for efficient isoeugenol hydrodeoxygenation. The
same conclusion can be drawn from the data for a non-acidic Ir/SiO catalyst exhibiting rather
2
small Ir particles. This catalyst did not lead to efficient deoxygenation of isoeugenol, even if
iridium is known to catalyze hydrogenolysis.
The spent best catalyst, Pt-H-Beta-300 used for hydrodeoxygenation of isoeugenol could be
regenerated regaining the catalytic properties. The regenerated Pt-H-Beta-300 catalyst
1
9
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
exhibited almost the same conversion of isoeugenol and selectivity to cyclohexane as the fresh
Pt-H-Beta-300, indicating catalyst reusability.
Pt-H-MCM-41 catalyst with large mesopores (4.9 nm) exhibited almost twofold higher
selectivity to propylcyclohexane than microporous Pt-H-Beta-25 (650 °C) with pores of 0.7 nm
size and four fold larger amount of Brønsted acid sites. It was found that the particle size of Pt
and Ir, appropriate amounts and strength of Brønsted and presumably Lewis acid sites, pore
size and structure of the catalysts are of importance for hydrodeoxygenation of isoeugenol.
Experimental
Catalyst preparation
Three Pt-modified Beta zeolite catalysts with SiO
2
to Al
2
3
O ratios of 25, 150 and 300 were
prepared from NH
4
-Beta (Zeolyst International) as starting materials. The support materials
o
were calcined at 450 C to obtain the proton form. Pt- and Ir-modified H-Beta-300, H-Beta-150,
H-Beta-25 microporous zeolites, H-MCM-41 structured material with mesopores and SiO ,
2
supported catalysts were synthesized using an evaporation impregnation method in a rotavapor
equipment. Aqueous solutions of hexachloroplatinic acid (Merck) and iridium chloride (Merck)
were used as Pt- and Ir-metal precursors. Ir/SiO was prepared with the same method by
2
addition of iridium to silica gel (Merck) using chloropenta amine iridium(III) chloride (Alfa
Aesar) as an iridium source. The duration of catalyst synthesis was 24 h, after which the
catalysts were removed from the flask and dried in an oven. The thermal treatment of the Pt-
and Ir-modified catalysts was carried out in a muffle oven using the step-calcination procedure
(Table S4). The calcined catalysts were weighed and stored in sealed bottles.
MCM-41 structured mesoporous material was synthesized using the method described in ref
[44].
The silica to alumina ratio of Pt-H-MCM-41 structured mesoporous catalyst was 20.
2
0
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Catalyst characterization methods
X-ray powder diffraction
The Philips X’Pert Pro MPD X-ray powder diffractometer was used for the characterization of
the structure and phase purity of Pt- and Ir- modified microporous zeolites and mesoporous
catalysts. The diffractometer was operated in Bragg-Brentano diffraction mode, and the
monochromatized Cu-Kα radiation (λ = 1.541874 Å) was generated with a voltage of 40 kV
and a current of 45 mA. The primary X-ray beam was collimated with a fixed 0.25° divergence
slit and a fixed 10 mm mask. A 7.5 mm anti-scatter slit was used in the diffracted beam side
prior to the proportional counter.
For all the samples, the measured 2θ angle range was 5.0°
̶ 85.0°, with a step size of
0
.026° and measurement time of 10 s per step. For Ir-H-MCM-41 and 2 wt% Pt-H-MCM-41,
the diffractograms were also determined with smaller 2θ angles in order to observe the
diffraction peaks originating from the MCM-41 phase. In these measurements, the applied 2θ
angle range was 0.8°
̶ 9.8°, with a step size of 0.02° and measurement time of 10 s per step. A
0
.18° parallel plate collimator was used in the diffracted beam side prior to the proportional
counter. The samples were placed on copper sample holders. The phase detection threshold was
ca. 5 %.
The diffractograms were analyzed with Philips X'Pert HighScore and MAUD
(http://www.ing.unitn.it/~maud/) programs. HighScore together with MAUD was used for the
phase analysis and MAUD for the Rietveld refinement. The Powder Diffraction File 2 (PDF-2)
database, IZA Structure Commission Database of Zeolite Structures and Inorganic Crystal
Structure Database (ICSD) were used as in refs [36, 45-47]. The crystal structure reference and the
corresponding crystal information (.cif) file of Beta polymorph A (SiO framework) were taken
2
from the IZA Structure Commission Database of Zeolite Structures. [46]
2
1
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Transmission electron microscopy
Transmission electron microscopy was used to determine the size of Pt- and Ir-metal particles
for the fresh and spent catalysts. TEM was also used to study the porous structure and
periodicity of pores of the catalysts. The instrument used for the measurements was JEM 1400
plus with an acceleration voltage of 120 kV and with resolution of 0.98 nm using Quemsa II
MPix bottom mounted digital camera.
Scanning electron microscopy (SEM) and Energy dispersive X-ray micro-analyses
(EDXA)
Morphology of Pt- and Ir-catalysts for namely the crystal shape, size and distributions were
investigated with scanning electron microscopy (SEM), Zeiss Leo Gemini 1530 microscope
equipped with SE (secondary electron) and BSE (backscattered electron) devices. The chemical
compositions of some selected fresh and spent catalysts was investigated using energy
dispersive X-ray-microanalyses (EDXA).
Nitrogen adsorption
The specific surface area, pore volume and pore size distributions of the catalysts were
determined using nitrogen adsorption method with Sorptomatic 1900 (Carlo Erba Instruments).
The sample was outgassed prior to the experiment at 150°C for 3 h. Specific surface area and
the micro-pore volume for the microporous zeolites were calculated with Dubinin and
Horvath/Kawasaki methods, respectively. [45] The specific surface and the pore volume of the
mesoporous catalysts were calculated using BET and Dollimore /Hill methods, respectively.
CO pulse chemisorption
The metal dispersion and particle sizes were determined with CO pulse chemisorption using
Autochem 2910. Prior to measurements about 100 mg of catalyst was prereduced with the
2
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
following temperature programme: 10°C/min- 450 °C (3 h). Thereafter, the catalyst was flushed
at 350 °C for 90 min with argon in order to remove adsorbed hydrogen. Then the catalyst was
cooled to room temperature. CO was adsorbed for 40 min min using 10 vol % CO in helium
(AGA). Thereafter chemisorbed CO was removed by flushing the catalyst with helium. The
pulse chemisorption measurement was performed at 25 °C and the stoichiometry 1:1 for Pt:CO
[
48]
and 1:1.19 Ir:CO [49] was applied for calculating the metal particle size.
Measurement of the acid sites using FTIR-pyridine adsorption desorption
Characterization of the Brønsted and Lewis acid sites, their amount and strength was
determined using pyridine (Sigma Aldrich, ≥ 99.5 % a.r.) adsorption-desorption. The
measurements were performed with ATI Mattson FTIR using 10 mg of self –supported catalyst
pellets. In order to discriminate between weak, medium and strong acid sites, desorption of
pyridine was performed at 250°C, 350°C and 450 °C, respecitvely. Quantification of Brønsted
-1
-
and Lewis acid sites was done by considering intensity of IR signals at 1545 cm and 1455 cm
1,
respectively, using the molar extinction factor given by Emeis. [50]
Thermogravimetrical analysis
Thermogravimetric analysis of the fresh and spent catalysts was performed to analyze the
amount of coke contents in the spent catalysts after use in hydrodeoxygenation of isoeugenol.
The equipment used for the quantitative analysis of the coke contents was a Pyris-TGA, Perkin
Elmer instrument under nitrogen flow of 20 ml/min to avoid oxidation of the sample. The
amount of catalyst was ca. 7 mg. The mass loss was detected as a function of the temperature
increase using the following temperature programme: 25 °C – 10°C/min – 1000°C under inert
atmosphere.
Experimental setup for hydrodeoxygenation of isoeugenol
2
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
Isoeugenol transformation was typically performed in an autoclave using 100 mg of isoeugenol,
0 ml of dodecane (Sigma Aldrich, ≥99%). as a solvent and 50 mg preduced catalyst. All
catalysts were reduced, firstly, started with Ar flushing for ten minutes and followed with H
5
2
flushing (10 min). The temperature programme for reduction was 350°C for 3 h with ramping
of 10℃/min. Thereafter, hydrogen was replaced with argon and 10 ml of the solvent was added
into the reactor containing the pre-reduced catalyst. The experiment was performed by placing
the prereduced catalyst into the autoclave and additionally 40 ml of dodecane. The reactor was
first flushed for 10 min with argon (AGA, 99.999%) and then by hydrogen (AGA, 99.999%).
Thereafter the reactor outlet was closed and the reactor was pressurized at room temperature up
to 2 MPa. Then the temperature was elevated to the desired temperature and finally the pressure
was adjusted to 3 MPa. The reaction was started by turning on the stirrer and using 900 rpm in
order to suppress the external mass transfer limitations. In order to make the experiments in the
kinetic regime, small catalyst particles (below 63 µm) were used. The samples were periodically
o
withdrawn to study the reaction kinetics. The spent catalyst was regenerated at 400 C for 160
min in the presence of air and thereafter reused.
The reuse of the most promising catalyst, Pt-H-Beta-300, was investigated using the
regenerated and reduced catalyst in the second experiment with the same initial isoeugenol
concentration as used in the first experiment. Pt-H-Beta-300 spent catalyst was regenerated in
a muffle oven using step calcination procedure, specifically utilized for Pt containing catalysts
(Table S1).
Analysis of the reaction mixture
The reaction mixture was analyzed by gas chromatography using DB-1 capillary column
(Agilent 122-13e, 30 m internal diameter 250 µm, film thickness 0.50 µm) and a FID detector.
The following temperature programme was used: 60°C (5 min) -3°C/min -300°C. The reaction
2
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
products were identified by GC-MS and quantified by making a calibration curve for the
following compounds: isoeugenol (Fluka, ≥98%), 4-propylcyclohexane (Aldrich, 99%),
dihydroeugenol (2-methoxy-4-propylphenol; Sigma Aldrich, ≥ 99%).
2
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
List of symbols
K0
n
Equilibrium constant at standard conditions for reaction j
Moles, mol
j
P
Pressure, MPa
P0
R
Standard pressure, MPa
Ideal gas constant, J/K/mol
Degree of explanation
R2
T
Absolute temperature, K
Absolute standard temperature, K
T0
Greek symbols
ΔG0
ΔG0
Gibbs free energy of formation at standard conditions, J/mol
Gibbs free energy of reaction at standard conditions, J/mol
f
r
ΔGΦr,j
ΔGr,j
Gibbs free energy of reaction at 0.1 MPa and a chosen temperature, J/mol
Gibbs free energy of reaction at a fixed temperature and pressure, J/mol
Enthalpy of formation at standard conditions, J/mol
ΔH0
ΔH0
f
r
Enthalpy of reaction at standard conditions, J/mol
ν
i,j
Stoichiometric matrix composed by i components and j reactions, -
2
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
References
[
[
[
[
[
1] M. B. Polk, M. Phingbodhipakkiya, Waste Management 1980, 1, 111-118.
2] S. Oh, G. H. Hwan, H. S. Choi, J. W. Choi, Chemosphere,2014, 117, 806-814.
3] A. V. Bridgewater, A. V., Biomass Bioenergy 2012, 30, 68-94.
4] D. E. Kim, X. Pan, Ind. Eng. Chem. 2010, 49, 12156-12163.
5] A. Jongerius, R. Gosselink, J. Dijkstra, J., Bitter, P. Bruijnincx, B. Weckhuysen,
ChemCatChem 2013, 5, 2964 – 2972.
6] R. N. Olcese, J. Francois, M. M. Bettahar, D. Petitjean, A. Dufour, Energy Fuels 2013, 27,
75-984.
7] C. Sepuldeva, K. Leiva, R. Garcia, L. R. Radovic, I. T. Ghampson, W. J. DeSisto, J.L.
Garcia Fierro, N. Escalona, Catal. Today 2011, 172, 232-239.
8] V. N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Appl. Catal. B: Environ. 2011, 101, 239-
45.
9] P. M. Mortensen, J.-D. Grunwaldt, P. A. Jensen, G. Knudsen, A. Degn Jensen, Catal Sci.
Technol.2014, 3672-3686.
10] I. T. Ghampson, C. Sepuldeva, R. Garcia, J. L. G. Fierro, N. Escalona, Catal. Sci. Technol.
016, 6, 4356-4369.
11] E. Santillan-Jimenez, M. Perdu, R. Pace, T. Morgan, M. Crocker, Catalysts 2015, 5, 424-
41
[
9
[
[
2
[
[
2
[
4
[
[
[
12] J. E. Peters, J. R. Carpenter, D. C. Dayton, Energy Fuels 2015, 29, 909-916.
13] D. Gao, Y. Xiao, A. Varma, Ind. Eng. Chem. Res. 2015, 54, 10638-10644.
14] M. Shetty, K. Murugappan, W. H. Green, Y. Roman-Leshkov, ASC Sust. Chem. Eng. 2017,
5
(6), 5293–5301.
[
[
15] Q. Lu, C.-J. Chen, W. Luc, J. G. Chen, A. Bhan, F. Jiao, ACS Catal. 2016, 6, 3506-3514.
16] R. Lodeng, C. Ranga, T. Rajkhowa, V. I. Alexiadis, H. Bjorkan, S. Chytil, I. H. Svenum,
J. Walmsley, J.W. Thybaut, Biomass Conv. Bioref. DOI 10.1007/s13399-017-0252-z
[
[
17] A. K. Deepa, P. L. Dhepe, ChemPlusChem 2014, 79, 1573-1583.
18] C. Zhang, J. Xing, L,. Song, H. Xin, S. Lin, L. Xing, X. Li, Catal. Today 2014, 234, 145–
1
52
[
[
[
19] X. Zhang, Q. Zhang, T. Wang, T. Ma, Y. Yu, L. Chen, Biores. Technol. 2013, 134, 73-80.
20] J. Feng, C.-Y. Hse, K. Tang, J. Jiang, Appl. Catal. A. Gen. 2017, 542, 163-173.
21] L. Wang, J. Zhang, X. Yi, A. Zheng, F. Deng, C. Chen, Y. Ji, F. Liu, X. Meng, F.-S. Xiao,
ACS Catal. 2015, 5, 2727-2734.
[22] A. Bjelic, M. Grilc, B. Likozar, Chem. Eng. J. 2018, 333, 240-259.
2
7
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
[
[
[
23] A. Dwiatmoko, L. Zhou, I. Kim, J.-W. Choi, D. J. Suh, Catal. Today 2016, 65, 192-198.
24] J. Qi, S.-F. Tang, Y. Sun, C. Xu, X. Li, Chemistry Select 2017, 2, 7525-7529.
25] X. Liu, W. Jia, G. Xu, Y. Zhang, Y. Fu, ACS Sustainable Chemistry Engineering 2017, 5,
8594-8601.
[26] X. Liu, L. Xu, G. Xu, W. Jia, Y. Ma, Y. Zhang, ACS Catalysis 2016, 6, 11, 7611-7620.
[27] M. Y. Chen, Y.-B. Huang, H. Pang, X.-X. Liu, Y. Fu, Green Chem. 2015, 17, 1710-1717.
[28] A. J. Pamphile-Adrian, P. P. Florez-Rodriguez, F. B. Passos, J. Braz. Chem. Soc. 2016,
7
7, 5, 956-966.
29] M.W. Zemansky, M.M. Abbott, H.C. Van Ness. Basic engineering thermodynamics.
McGraw-Hill 1975.
30] J. M. Newsam, M.M.J. Treacy, W. T. Koetsier, C. B. de Gruyter, Proc. R. Soc. Lond. A
988, 420, 375-405.
[
[
1
[
[
[
[
[
[
31] D. Das, P. J. E. Harlick, A. Sayari, Catal. Commun. 2007, 8, 829–833.
32] O. A. Anunziata, A. R. Beltramone, J. Cussa, Catal. Today 2008, 133-135, 891-896.
33] T. Swanson, Natl. Bur. Stand. (U.S.), Circ. 1953, 539, I, 31. (Pt)
34] Natl. Bur. Stand. (U.S.) Monogr. 1965, 25, 4, 19. (Ir)
35] T. Swanson, Natl. Bur. Stand. (U.S.), Circ. 1953, 539, I, 15. (Cu)
36] Database of Zeolite Structures, http://www.iza-structure.org/databases/, June 2012, IZA
Structure Commission.
[
[
[
37] N. Yang, S. F. Bent, J. Catal. 2017, 351, 49-58.
38] ChemCAD v.5.0, Chemstations (http://www.chemstations.com/)
39] B.E. Poling, J. M. Prausnitz, J.P. O'Connell. The properties of gases and liquids (5th ed.)
2
004. McGraw-Hill: New York.
40] G. J. Wang, Y. Wang, Y. Liu, Z. Liu, Y. J. Guo, G. Liu, Z. Yang, M. X. Xy, L. Wang, L.,
Appl. Clay Sci. 2009, 44, 1-2, 185-188.
[
[
[
41] S.-B. Liu, J.Wang, J. Cat. 1991, 132, 2, 432-439.
42] C. Baerlocher, W. M. Meir, O. H. Olson, Atlas of Zeolite Framework Types, 5th revised
ed.; Elsevier Science: Amsterdam, The Netherlands, 2001.
[
[
[
43] H. Haario, MODEST Users guide, Profmath Oy, Helsinki 2007.
44] T. Kresge, M. E. Leonovitz, W. J. Roth, J. C. Vartuli, US Patent 509,868,4, 1992.
45] Powder Diffraction File 2 (PDF-2), sets 1-46, 1996 release, International Centre for
Diffraction Data (ICDD).
46] Inorganic Crystal Structure Database (ICSD), version 2.1.0, http://www.fiz-
karlsruhe.de/icsd.html, April 2012, Fiz Karlsruhe.
[
2
8
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800391
[
47] R. Ghezini, S. Sassi, A. Benqueddach, Microp. Mesop. Mat. 2008, 113, 370-377.
48] K. Shinichi, Y. Yoshiteru S. Atsushi, H. Tadashi, Journal of Japan Petr. Inst. 2005, 48,
, 173－177.
[
3
[
[
49] S. J. Tauster, S. C. Fung, R. L. Garten, J. Am. Chem. Soc. 1978, 100, 170-175.
50] C. A. Emeis, J. Catal. 1993, 141, 2, 347-354.
2
9
This article is protected by copyright. All rights reserved.