Production of p -cymene from crude sulphate turpentine with commercial zeolite catalyst using a continuous fixed bed reactor

Article abstract of DOI:10.1021/op500160f

Crude sulphate turpentine (CST), containing -pinene, 3-carene, other terpene isomers, and organic sulphur compounds, was used as feedstock in the production of cymenes with commercial zeolite catalysts. The catalysts used were not sensitive to sulphur, and the desulphurisation step was not needed prior to cymene synthesis. The experiments were performed in a continuous fixed bed reactor. Promising results were obtained after initial optimization at temperatures close to 300 °C with faujasite Y zeolite under 5 bar N₂ pressure. Cymenes were produced with 54% yields from model compound -pinene and with 28% yields from CST. The mechanistic route of the reaction can be divided into three steps: (1) ring opening, (2) rearrangement, and (3) dehydrogenation. The final step in the formation of p-cymene can occur via dehydrogenation or disproportion; here the effect of reaction conditions and catalyst plays a crucial role.

Full text of DOI:10.1021/op500160f

Article
pubs.acs.org/OPRD
Production of p‑Cymene from Crude Sulphate Turpentine with
Commercial Zeolite Catalyst Using a Continuous Fixed Bed Reactor
Juha A. Linnekoski,*,† Martta Asikainen,† Harri Heikkinen,† Reetta K. Kaila,† Jari Rasanen,‡
̈
̈
Antero Laitinen,† and Ali Harlin†
†VTT Technical Research Centre of Finland, Biologinkuja 7, FI-02101 VTT, Espoo, Finland
‡StoraEnso, Packing Boards, FI-55800 Imatra, Finland
ABSTRACT: Crude sulphate turpentine (CST), containing α-pinene, 3-carene, other terpene isomers, and organic sulphur
compounds, was used as feedstock in the production of cymenes with commercial zeolite catalysts. The catalysts used were not
sensitive to sulphur, and the desulphurisation step was not needed prior to cymene synthesis. The experiments were performed
in a continuous fixed bed reactor. Promising results were obtained after initial optimization at temperatures close to 300 °C with
faujasite Y zeolite under 5 bar N2 pressure. Cymenes were produced with 54% yields from model compound α-pinene and with
28% yields from CST. The mechanistic route of the reaction can be divided into three steps: (1) ring opening, (2)
rearrangement, and (3) dehydrogenation. The final step in the formation of p-cymene can occur via dehydrogenation or
disproportion; here the effect of reaction conditions and catalyst plays a crucial role.
formation, too high temperature (>400 °C) enhances the
thermal cracking of the products.11
INTRODUCTION
■
The petrochemical industry is strongly dependent on the
depleting crude oil resources. Targeting 100% biobased
chemicals and polymers as replacements for the crude oil
derived equivalents is a hot topic in chemistry research.
Terpenes, naturally occurring compounds that can be used as a
renewable feedstocks, have a high potential in this field.1
Terpenes can be modified by isomerisation, oxidation,
esterification, carbonylation, and (de)hydrogenation reactions
to access valuable chemical products.2 Monoaromatic com-
pounds produced from terpenes are valuable for petrochemical
industry.2,3 Of these, p-cymene is of particular importance as it
has potential use in the synthesis of terephthalic acid4 and
already established main uses in p-cresol production.5 p-
Cymene has also uses as a solvent for dyes and varnishes, as a
heat transfer medium, as the intermediate for the production of
pesticides, fungicides, flavours, and pharmaceuticals.6 Industri-
ally p-cymene is produced by Friedel−Crafts alkylation of
benzene or toluene with propylene or 2-propanol, and
hazardous acid catalysts such as AlCl3 or HF are used as
catalysts in the process.5,6 Due to the involvement of highly
toxic substances, this process is becoming a restricted method
by environmental legislation.7−11
However, using pure α-pinene or limonene for the
production of p-cymene is limited due to their high market
price in comparison with toluene and propylene. Especially
limonene has high demand as it is emerging as a biodegradable
solvent.21 Crude sulphate turpentine (CST) is a cheap
byproduct of Kraft pulping process of pine wood.12 Crude
sulfur turpentine contains mainly monoterpenes α-pinene and
3-carene (Table 1, Scheme 1, this work), and it can be used as a
starting material for the production of monoaromatics.
Table 1. CST composition (this work)
CST component
α-pinene
wt %
65
3-carene
24
p-cymene
0.46
2.0
R-(+)-limonene
other terpene isomers (FW 136)
terpene dimers (FW 204)
other compounds
7.4
0.83
0.29
The production of p-cymene from terpenes has emerged as
an environmentally benign alternative.12 Several heterogeneous
and homogeneous catalysts have been used for the conversion
of monoterpenes and terpene mixtures to monoaromatics.
Special attention has been given to the production of p-cymene
from α-pinene and limonene over various metal catalysts such
as Pd/C,13 Pd/Al2O3,13 Pd/Si,7,14,15 Pd−Ce/zeolite,7,16,17 Pd/
zeolite,18 Pt/C, Cu/Cr2O3, and ZnO2/Cr2O3.19 In addition
silica−alumina supports,20 modified clays,21 and Pd nano-
cluster22 catalysts have been used. There are also examples of
catalyst-free aromatization reactions.23 A wide temperature
range (150−500 °C) has been used in the aromatizations. It is
known that, as high temperature is advantageous for p-cymene
The literature on the aromatization of CST is very scarce.
This can be partly due to the chemical composition of CST as it
also contains sulphur compounds in high concentrations
(methyl mercaptan 0.2−5 wt %, dimethyl sulfide 1−12 wt %,
and dimethyl disulfide 0−1.3 wt %).24 The Pd catalysts
traditionally used for α-pinene aromatization are sensitive to
sulphur and are therefore not suitable for CST aromatization. A
similar conclusion can be drawn with several other metal
Special Issue: Continuous Processes 14
Received: May 21, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 1. Terpene structures
catalysts. There are some examples in the literature on using
zeolites for the isomerization and aromatization of pure α-
pinene and 3-carene.25,26 The development of sulphur-tolerant
catalyst systems for the production of biobased chemicals is
vital for the emerging bioeconomy, as various biobased
feedstocks, such as CST and crude tall oil, are rich in sulphur.
In this work, the potential of zeolite catalysts for aromatics
production from CST was studied. Besides a good acidic
platform, zeolites offer shape selectivity for the aromatization.
In preliminary experiments, α-pinene was used as a model
compound for CST. Three different zeolite catalysts (Molecular
Sieve 13X (FAU X), Mordenite (MOR), Faujasite Y (FAU Y))
were compared at 300 °C. The most promising catalyst, FAU
Y, was chosen for further experiments with CST.27−29
Figure 1. A schematic picture of the reactor equipment.
The analysis of the liquid samples and the feed CST was
performed by GC-MS instrument trace GC ultra gas chromato-
graph, trace DSQ mass spectrometer (Thermo Finnigan, USA)
column: HP-1MS dimensions 30 m × 0.25 mm, 0.25 μm film
thickness. Oven temperature program: 35 °C/hold 3 min/15
°C min−1/60 °C/hold 0 min/3 °C min−1/130 °C/hold 0 min/
25 °C min−1/300 °C/hold 8 min. Carrier gas He, constant flow
1.3 mL/min; injection volume 1 μL, split 1:30; mass
spectrometer scan range m/z 35−600.
The results were calculated using external calibration
method. p-cymene and α-pinene were calculated quantitatively.
Terpenes and aromatics were calculated semiquantitatively with
the responses of α-pinene and p-cymene, respectively. The
uncertainty of the measurements was 15% for p-cymene and
α-pinene and 30% for semiquantitatively calculated com-
pounds. The analysis results were confirmed with GC-FID and
GC-MS. All of the calculated results are based on the analysed
organic fraction. The results calculated in Tables 3 and 4 are
from 4 to 6 h samples on-stream. Selectivity is calculated as
moles of component in the product divided by reacted amount
of feed components α-pinene and 3-carene. Conversion is
calculated for α-pinene and 3-carene as reacted molar amount
divided by molar amount in the feed.
EXPERIMENTAL SECTION
■
CST was supplied by Stora Enso (Finland) and used as
received. Other reagents and catalysts were obtained from
commercial sources; (−) α-pinene (Fluka, 98% (mixture of
enantiomers) and 99% (GC grade)), R-(+)-limonene (97%,
Sigma-Aldrich), p-cymene (Sigma-Aldrich, 99%, GC grade),
13X (Faujasite X, Union Carbide), Mordenite (Zeolyst), and
Faujasite Y (Zeolyst). The main characteristics of the catalysts
used are shown in Table 2.
A continuous tubular fixed bed reactor was used in the
aromatization experiments (Figure 1). The catalyst bed was
fixed with quartz wool (0.8 g) inside the reactor. Each
experiment was run for 6 h to achieve a steady state. A pressure
test was carried out with 5.5 bar (N2) followed by catalyst
pretreatment at 500 °C (2 h) under vacuum for removal of
moisture. The reactor pressure (5 bar N2) and catalyst bed
temperature were adjusted prior the feed. The terpene feed was
set to 10 g/h. The WHSV was between 1.5 and 6 L/h. The feed
was preheated in the feed line to 160 °C, which is near the
boiling point of monoterpenes. The product flow was cooled
down after the reactor, and the liquid sample was collected in a
cold trap for GC-MS analysis.
PROPERTIES OF THE ZEOLITES
■
All zeolites, FAU Y, FAU X, and MOR, are large pore zeolites
having pore size >7.4 Å (Table 2). FAU Y and FAU X have
crystalline structures formed by spherical super cages of 14 Å
diameter interconnected tetrahdrally through smaller openings
of 7.4 Å.30 MOR consists of two pore channels: an elliptical 12-
membered ring (MR) channel (6.7 Å × 7.0 Å) which runs
parallel to the c-axis and a 8 MR side pocket (3.4 Å × 4.8 Å)
that runs in the b-axis direction.31
Table 2. Main characteristics of the catalysts used
a
a
catalyst
zeolite type
manufacturer
SiO2/Al2O3
Na2O (wt %)
BET (m2/g)
pore size (Å)
b
CBV600
CBV21A
13X
Faujasite Y (FAU Y)
Mordenite (MOR)
Faujasite X (FAU X)
Zeolyst
5.2
0.2
660
500
412
7.44
c
Zeolyst
20
0.08
9.56
7.0 × 6.7
b
Union Carbide
2.5 0.5
7.4
a
b
c
Value obtained from the manufacturer. See ref 30. See ref 31.
B
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Table 3. Calculated Si/Al ratios and percentages of different Al species of the studied zeolite catalysts
amount (%)
a
sample
type
SiO2/Al2O3
Al(IV)FR
Al(V)NFR
Al(VI)NFR
AlFR/AlNFR
MOR
Mordenite
Faujasite Y
Faujasite X
15.7
8.4
34.6
37.9
96.9
38.6
8.4
26.8
53.7
3.0
0.346
0.379
0.969
FAU Y
FAU X
1.4
0.1
a
Calculated based on the integrals seen from the 29Si MAS NMR spectra.
Al sites as determined from the peak area. The results indicate
that in zeolite FAU Y only 38% of the total amount of
aluminum in the catalyst is located in the framework, whereas
in zeolite FAU X aluminum is mainly located in the catalyst
framework (97%).
Table 4. Quantitative hydroxyl group amounts (acid OH,
AlOH, and SiOH) and the total amount of H in each sample
as calculated from the H MAS NMR spectra
1
amount (μmol/g)
In this study, 1H MAS NMR was used to measure
total
amount
of H
residual
quantitatively Bronstedt acid sites (see beginning of this
̈
sample
type
acid OH AlOH SiOH template
chapter) of the zeolites. The results are shown in Table 4.
According to the results, the FAU Y zeolite (Si/Al 8.4) shows a
MOR
Mordenite
Faujasite Y
Faujasite X
n.m
450
68
n.m
262
73
n.m
489
46
n.m
11580
2589
1972
FAU Y
FAU X
1388
1785
Bronstedt acidity of 450 μmol/g and the FAU X zeolite (Si/Al
̈
1.4) only 70 μmol/g. This lower acidity of the FAU X zeolite
(Si/Al 1.4) may be explained by the high amount of Na2O
(9.56 wt %) present in the structure, where most of the Na is
not removed during the dehydration procedure, respectively.
Acidity of the Zeolites: Analysis with Solid-State NMR.
All NMR measurements were carried out at 293 K using
Agilent narrow-bore 600 MHz (14.1T) NMR spectrometer
1
Also, a broad and very intense signal in the H MAS NMR
1
equipped with a 3.2 mm HXY MAS probe. For the H NMR
spectra (Figure 3) at chemical shift around 7−8 ppm was
observed and tentatively assigned to originate from residual
template indicating the presence of hydrocarbons on the
experiments, the samples were dehydrated at 673 K for 4 h
under dry nitrogen gas flow at atmospheric pressure. In the 1H
MAS NMR experiments, 128 transients were acquired using a
2.85 μs pulse (π/2), 30 s recycle delay, and a MAS rate of 20
kHz. After dehydration the samples were loaded inside a
glovebox into 3.2 mm zirconia rotors. For the quantitative OH
group determination, a full rotor of Q8M8 (octakis-
(trimethylsiloxy)silesquioxane), Chymos GmbH, CAS No.
51777-38-9) was used as an external quantitation standard. In
all cases a background signal, run under identical conditions
with an empty rotor, was subtracted from the spectra. The
amount of OH groups (μmol/g) in each sample was
determined by comparing the experimental spectral area to
that of external standard and normalizing by weight (spin
counting).32 27Al and 29Si MAS NMR experiments were carried
out on the samples without drying. In the 27Al MAS NMR
experiments, 4000 transients were acquired using a 1.0 μs pulse
(π/12), 2 s recycle delay, and MAS rate of 20 kHz. In the 29Si
MAS NMR experiments, 20000 transients were acquired using
3.0 μs pulse (π/4), 10 s recycle delay, and MAS rate of 10 kHz.
The 29Si MAS NMR spectra of the studied zeolites are shown
in Figure 2 (left). 29Si MAS NMR was used to measure and
characterize the Si/Al ratio for the framework.33 In the 29Si
NMR spectra of FAU Y and FAU X zeolites, relative intensities
of the sharp resonances reflect directly the numbers and
occupancies of the crystalloqraphically inequivalent T-sites in
the asymmetric unit of the unit cell. In the case of MOR zeolite,
the shift dispersion is limited and acts mainly as an additional
line-broadening mechanism in the spectrum. The calculated Si/
Al ratios are depicted in Table 3.
1
catalyst surface. To verify the possibility of residual water, H
MAS NMR spectrum of the hydrated zeolite FAU Y was
measured, where the chemical shift of the residual water was at
6.5 ppm. In fact, the intense signal around 7−8 ppm was
dominating the 1H MAS NMR spectrum of the MOR zeolite so
much that we were not able to carry out acid site calculations
for the catalyst due to the measured very high number of total
amount of H in the sample (Table 4).
RESULTS AND DISCUSSION
■
The three zeolites (Table 2) were tested for the aromatization
of model compound α-pinene. The zeolites were tested at 300
°C under vacuum (Table 5). Although all three catalysts gave
high conversions at 300 °C, FAU Y gave clearly higher
selectivity to cymenes (34%) than MOR (17%) or FAU X
(3.9%). The FAU X mostly yielded terpene isomers (78%).
FAU Y was selected for further study. Also, different reaction
temperatures were tested under vacuum but at 200 °C or
below, the catalysts were mainly active for isomerization. The
catalysts were tested without pretreatement, but it was
discovered that the pretreatment is essential for the catalyst
selectivity towards the desired aromatization.
The three zeolites are large pore zeolites having pore size
>7.0 Å. The desired product p-cymene has a smaller kinetic
diameter compared to isomers o- and m-cymene. The shape
selectivity resembles the much-studied xylene isomer case. p-
Xylene has critical diameter of 6.7 Å (molecule size), similar to
benzene and toluene.27 p-Xylene isomers o- and m-xylene have
higher critical diameters of 7.4 Å.27 For cymenes we have a
similar case. p-Cymene has a similar structure compared to p-
xylene, and o- and m-cymenes have structures compared to o-
and m-xylenes. p-Cymene has a kinetic diameter of 6.31 Å
calculated from the molecular weight using eq 1,27 where σ is
the kinetic diameter and Mw is the p-cymene molecular weight.
This indicates that p-cymene can form in all studied zeolites,
The 27Al MAS NMR of zeolites FAU Y, MOR, and FAU X
are shown in Figure 2 (right). In the 27Al NMR spectra, three
different types of Al species were assigned according to their
distinct isotropic chemical shift range: four-coordinated frame-
work aluminum (Al(IV)FR) around 75−50 ppm, 5-coordi-
nated nonframework aluminum (Al(V)NFR) around 30−40
ppm, and six-coordinated nonframework aluminum (Al(VI)-
NFR) around 20−0 ppm.34 Table 3 shows the ratio of different
C
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 2. 29Si MAS NMR (left) and 27Al MAS NMR (right) of the three studied catalysts (FAU Y, MOR, and FAU X) with chemical shifts and
signal assignations.
1
Figure 3. H MAS NMR spectra of the three studied catalysts (FAU Y, FAU X, and MOR) with signal assignations.
D
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Table 5. Comparing catalyst activity with α-pinene
b
selectivity (%)
cymenes
c
terpene isomers
menthenes
(MW 138)
menthanes
(MW 140)
exp. no.
catalyst
conv. (%)
(MW 136)
o
m
p
total
1
2
3
FAU X
MOR
86
78
50
0
0
0
3.9
16
23
3.9
17
34
5.1
20
15
0
98
1.1
1.4
10
0
FAU Y
100
9.3
16
a
Reaction conditions: Catalysts were calcinated at 500 °C at vacuum prior to use, catalyst loading 6 g, T = 300 °C, vacuum (200 mbar), α-pinene
b
c
feed flow rate = 10 g/h. Selectivity is defined in the Experimental Section. Terpene isomers (MW 136) include all terpene isomers with this mass
excluding the feed components α-pinene and limonene.
a
Table 6. CST aromatisation experiments
b
selectivity (%)
terpene
c
conv. (%)
isomers
cymenes
menthenes menthanes
exp.
no.
T
(°C)
FAU Y catalyst loading
(g)
N2 flow
(mL/min)
α-pin 3-car
(MW 136)
o
m
p
total (MW 138) (MW 140)
4
350
200
300
380
300
350
400
300
350
300
350
35
35
35
35
35
35
35
70
70
70
70
48
38
12
37
14
0
0
0
0
0
0.57
1.4
17
0.57
1.73
23.5
20
0
0
5
3
98
85
0.33
6.5
7.5
2.2
6.8
6.1
7.7
2.6
7.3
6.7
0.76
16
0
6
3
100
100
95
100
100
84
5.6
20
7
3
0.9
12
2.2
7.2
9.6
6.2
16
8
1.5
1.5
1.5
1.5
1.5
3
32
2.4
0.30
11
34
9.8
0
0.20
0.37
0.45
0.31
0.21
0.35
0.46
7.3
18
9.7
25
0.53
6.3
4.8
3.8
0.53
2.9
10
9
98
98
10
11
12
13
14
100
100
94
99
12
19
98
20
28
82
7.2
18
10
6.3
12
100
100
98
25
3
100
14
21
5.4
a
Reaction conditions: Catalyst FAU Y, catalyst was calcinated at 500 °C at vacuum prior to use, pressure 5 bar N2, CST feed flow rate = 10 g/h.
b
c
Selectivity is defined in the Experimental Section. Terpene isomers (MW 136) include all terpene isomers with this mass, excluding the feed
components α-pinene, 3-carene, and limonene.
and the differences in the product distribution are caused by
differences in acidity.
According to the results in Table 5, FAU X shows only high
isomerization activity to terpene isomers and very weak
dehydrogenation activity to cymenes. The low amount of
σ = 1.234(MW)1/3
̈
Bronstead and Lewis sites in FAU X (Table 4) and the lower
(1)
acidity of Bronstead sites is probably the main reason for the
̈
Zeolites possess several types of acid sites, which participate
in the reactions and affect the reactivity. Whereas framework Al
species are always part of the bridging SiOHAl groups (i.e.,
low activity towards cymene production but for high isomer-
ization activity.41
FAU Y has a higher amount of Bronsted acid and Lewis acid
̈
Bronstedt acid sites), the nonframework aluminum species may
sites compared FAU X (Table 4) and a higher concentration of
acidic sites but of lower acidity compared to mordenite.28 FAU
Y shows the highest activity to cymenes, and thus the amount
and strength of acidic sites in FAU Y seems to be optimum
from the tested zeolites. Significantly higher amount of Lewis
sites increases the dehydrogenation activity compared FAU X.
In addition it has to be stated that the reaction conditions
(temperature, WHSV, gas flow) have major effect on the
product distribution, as can be seen for FAU Y in Table 5. In
addition to the cymene formation, isomerization reactions
greatly depend on reaction conditions.
̈
occur as Al3+, Al(OH)2+, Al(OH)2 , AlOOH, Al(OH)3, and
+
Al2O3.35 Of these, Al3+, Al(OH)2+, and Al(OH)2 may act as a
+
Lewis acid sites, which may directly initiate the hydrocarbon
conversion via hydride abstraction.36
It has been observed that, if only Lewis acid sites are present,
no catalytic activity is observed in α-pinene isomerization and
dehydrogenation. In addition also pure Bronstead acidity leads
̈
to low overall activity. The highest catalytic activity was
achieved with catalyst having both Bronstead and Lewis acid
̈
sites.37,38 In addition it has been shown that the transformation
of monoterpenes occur by combination of acid catalysis and
electron transfer.39 On the contrary to monoterpenes,
Narbeshuber et al. (1997) concluded that there is two parallel
mechanisms of dehydrogenation of light alkanes occurring. The
main mechanism is represented by monomolecular protolytic
dehydrogenation. These sites produce the constant steady state
dehydrogenation activity. In addition at the initial stages of the
reaction extra framework aluminum moieties can participate in
high dehydrogenation activity. They speculated that these sites
are later blocked by the reaction product hydrogen.40
Compared to MOR, FAU Y has a lower acidity but a higher
concentration of protonic sites.28 Unfortunately, due to the
impurity in MOR, the activity and acidity measurements are not
entirely comparable to other zeolites. However, it can be
generally concluded that the amount of acidic sites is
comparable to FAU Y. The major difference is in the amount
of nonframework Al(V) and Al(VI) species. Based on the Si/Al
ratio, mordenite should have the strongest acidic sites.
However, the impurity can affect the amount and strength of
acidic sites. In addition, mordenite has two types of pore
E
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 2. Mechanism of acid catalyzed α-pinene isomerization, dehydrogenation, and hydrogenation
channels, and the smaller of these (3.4 Å × 4.8 Å) is not
capable to occupy the reactant or product molecules. This
decreases the available reactive space and can have effect on
MOR activity.
original catalyst amount (1.5 g) but with an increased N2 flow
rate from 35 mL/min to 70 mL/min. When the catalyst
amount was increased at temperature of 300 °C (1.5 g → 3.0 g)
at an increased N2 flow rate (70 mL/min), the selectivity to
desired cymenes decreased slightly, and the selectivity to
menthenes and menthanes increases. Temperatures higher than
this caused undesired hydrogenation and cracking.
Besides the products listed in Tables 3 and 4, the reaction
conditions used in these experiments also gave various
byproducts. In the catalytic reaction conditions camphene is
formed, which further reacts to cyclopentenes. Cracking
increases with increasing temperatures, giving coke and other
aromatics besides cymenes, including toluene and tetramethyl-
benzene. Also, xylenes form at higher temperatures. Dimeriza-
tion of terpenes was observed as well.
Experiments with the model compound α-pinene were
carried on to test the effect of carrier gases and catalyst
structure. The aromatization of α-pinene has been studied
extensively in the literature as discussed in the Introduction. In
the presence of air, FAU Y was quickly deactivated by coke.
Although the presence of O2 favored the dehydrogenation/
hydrogenation ratio, a noticeable deactivation of FAU Y was
observed as decreased conversion of α-pinene. When N2 was
used as carrier gas (35 mL/min), the selectivity to cymenes was
dramatically improved, the total yield being 54% (p-cymene
44%, m-cymene 9.3%, o-cymene 0.7%), and 100% conversion
was obtained. The nitrogen flush is likely to facilitate the
selectivity by removing some of the hydrogen released in the
catalyst bed. The dehydrogenation/hydrogenation ratio clearly
increases when nitrogen flush is applied. Using powderized
FAU Y did not improve the selectivity to the desired cymenes
(data not shown).
The aromatization of CST was tested (Table 6) by varying
the temperature (200−400 °C), catalyst loading (1.5 and 3 g),
and nitrogen flow (35 mL/min and 70 mL/min). Also thermal
reactions were studied in catalyst-free reaction conditions
(Table 6, Experiment 4). In the catalyst-free experiment at 300
°C (Experiment 4), conversions remained low, and main
products were terpene isomers. At 200 °C with 3 g of FAU Y
(Experiment 5), high conversions were obtained, but the main
products were again terpene isomers. It seems that 200 °C is
sufficient enough temperature for the isomerization but not
high enough for the dehydrogenation step to take place. A
temperature of 300 °C gives a higher selectivity to cymenes
with 3 g of catalyst (Experiment 6). With half the amount of
catalyst (1.5 g), a higher temperature of 350 °C is needed to
obtain comparable selectivities (Experiment 9), and with 1.5 g
of catalyst at 300 °C, the selectivity to cymenes was rather low
(Experiment 8). The highest selectivities to cymenes (up to
28%, Experiment 11) were obtained at 300 °C with half of the
DISCUSSION OF THE REACTION MECHANISM
■
Several reaction steps are required for the aromatization of α-
pinene and 3-carene to cymenes (Scheme 2). In the presence of
an acid catalyst, cationic intermediates are likely to form (for
example I and II), further reacting to the terpene isomers.26 In
Scheme 2 only the p-isomers are presented, but also o/m-
isomes are forming. This isomerization can also take place by
thermal reactions, as was concluded from Experiment 4 (Table
6). However, the conversion of α-pinene and 3-carene to the
terpene isomers is clearly higher in the presence of an acid
catalyst, and therefore it is likely the acid catalyzed isomer-
izations are dominating under the reaction conditions.
After the isomerization, rearrangement of the monoterpene
isomers can take place, catalyzed by acid.
The monocyclic terpenes can undergo dehydrogenation to
cymenes. As remarked with Experiment 5, 200 °C was not a
high enough temperature for the dehydrogenation to take
place, but higher temperatures are required. Also in non-
catalytic conditions (Experiment 4), no cymenes are formed;
this shows that the dehydrogenation requires the acid catalyst.
As one mole of hydrogen is forming in the dehydrogenation
step, this hydrogen will to react further. There are several
acceptors for the hydrogen. Terpenes can disproportionate by
F
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(4) Berti, C.; Binassi, E.; Colonna, M.; Fiorini, M.; Kannan, G.;
Karanam, S.; Mazzacurati, M.; Odeh, I.; Vannini, M. WO/2010/
2010078328, 2010; CAS abstract number 2010:847387.
accepting the hydrogen and become reduced to menthenes
(FW 138). Menthenes can further accept a mole of hydrogen
to give menthanes (FW 140). The organic sulphur compounds
in the feed can also act as hydrogen acceptors as they can
undergo hydrogenolysis, disproportionation, and reduction
(eqs 2−4).
(5) Fiege, H. In Ullmann’s encyclopedia of industrial chemistry; Wiley-
VCH: Weinheim, 2012; Vol. 10, Cresols and xylenols, pp 419−460.
(6) Eggersdorfer, M. In Ullmann’s encyclopedia of industrial chemistry;
Wiley-VCH: Weinheim, 2012; Vol. 36, Terpenes, pp 29−45.
(7) Buhl, D.; Weyrich, P. A.; Sachtler, W. M. H.; Holderich, W. F.
̈
CH3SH + H2 ⇒ CH4 + H2S (hydrogenolysis)
(2)
Appl. Catal. A: Gen. 1998, 171, 1−11.
(8) Roberge, D. M.; Buhl, D.; Niederer, J. P. M.; Holderich, W. F.
̈
2CH3SH ⇒ (CH3)2S + H2S (disproportionation)
(3)
Appl. Catal. A: Gen. 2001, 215, 111−124.
(9) Wadaani, F. A.; Kozhevnikova, E. F.; Kozhevnikov, I. V. Appl.
Catal. A: Gen. 2009, 363, 153−156.
(10) Binitha, N. N.; Sugunan, S. Catal. Commun. 2007, 8, 1793−
1797.
(CH3)2S + 2H2 ⇒ 2CH4 + H2S (reduction)
(4)
Equations 2−4 show possible reactions with methyl mercaptan
and dimethyl sulfide.
(11) Golets, M.; Ajaikumar, S.; Mohln, M.; Warna, J.; Rakesh, S.;
̈
̊
By reacting with the hydrogen, the sulphur compounds may
increase the selectivity towards cymenes. Oxygen would be a
good hydrogen acceptor,23 but as mentioned earlier, the
presence of O2 resulted in coke formation and deactivation of
catalyst. Using N2 flush seems to somewhat reduce the
hydrogenation of terpenes by flushing away the forming H2.
The importance of the N2 flow is shown by the fact that the
catalyst amount can be reduced by half when the N2 flow is
doubled and the desired cymene selectivity is increased. On the
contrary, when the catalyst amount is increased at increased N2
flow, a reduction in the desired cymene yield is observed due to
increased hydrogenation reaction.
Mikkola, J.-M. J. Catal. 2013, 307, 303−315.
(12) Rasanen, J.; Penttinen, T.; Harlin, A.; Kaila, R. WO/2011/
̈
̈
2011151526 A1, 2011; CAS abstract number 2011:1582420.
(13) Grau, R. J.; Zgolicz, P. D.; Gutierrez, C.; Taher, H. A. J. Mol.
Catal. A: Chem. 1999, 148, 203−214.
(14) Buhl, D.; Roberge, D. M.; Holderich, W. F. Appl. Catal. A: Gen.
̈
1999, 188, 287−299.
(15) Lesage, P.; Candy, J. P.; Hirigoyen, C.; Humblot, F.; Basset, J.
M. J. Mol. Catal. A: Chem. 1996, 112, 431−435.
(16) Weyrich, P. A.; Trevino, H.; Holderich, W. F.; Sachtler, W. M.
̈
H. Appl. Catal. A: Gen. 1997, 163, 31−44.
(17) Weyrich, P. A.; Holderich, W. F. Appl. Catal. A: General 1997,
̈
158, 145−162.
(18) Roberge, D. M.; Buhl, D.; Niederer, J. P. M.; Holderich, W. F.
̈
CONCLUSIONS
■
Appl. Catal. A: Gen. 2001, 215, 111−124.
The suitability of commercial zeolites for aromatization of α-
pinene and CST was studied. From the tested zeolites, FAU Y
gave the best selectivity to cymenes. After initial optimization
with α-pinene, complete conversions were obtained with α-
pinene, giving over 54% selectivities to cymenes. Complete
conversions were also obtained with CST with 28%
selectivities. The major cymene isomer was in each experiment
the desired p-cymene; m-cymene was also often obtained but
with only very small amounts of o-cymene.
(19) Al-Wadaani, F.; Kozhenikova, E. F.; Kozhenikov, I. V. Appl.
Catal. A: Gen. 2009, 363, 153−156.
(20) Martín-Luengo, M. A.; Yates, M.; Martínez Domingo, M. J.;
Casal, B.; Iglesias, M.; Esteban, M.; Ruiz-Hitzky, E. Appl. Catal. B:
Environ. 2008, 81, 218−224.
(21) Martín-Luengo, M. A.; Yates, M.; Saez Rojo, E.; Huerta Arribas,
D.; Aguilar, D.; Ruiz, E. Appl. Catal. A: Gen. 2010, 387, 141−146.
(22) Zhao, C.; Gan, W.; Fan, X.; Cai, Z.; Dyson, P. J.; Kou, Y. J.
Catal. 2008, 254, 244−250.
(23) Asikainen, M.; Jauhiainen, O.; Aaltonen, O.; Harlin, A. Green
Chem. 2013, 15, 3230−3235.
(24) Weyerhaeuser Crude Sulphfate Turpentine Material Safety Data
Sheet 2010.
These results are promising, as they indicate that no noble
metal catalysts may be needed in the production of biobased
aromatic compounds from terpenes. The development of
sulphur-tolerant catalyst systems is of vital importance to the
bioeconomy, as sulphur-containing feedstocks such as CST or
crude tall oil could be directly used for catalytic transformations
to produce biobased chemical products. As this report describes
the initial optimization studies, further work of the conditions
and scope of these transformations is ongoing in our laboratory.
(25) Eswaramoorthy, M.; Krishnasamy, V. Indian J. Chem. 2001, 40A,
264−269.
̈
(26) Ozkan, F.; Gunduz, G.; Akpolat, O.; Besu
̧
n, N.; Murzin, D. Y.
̈
̈
̈
Chem. Eng. J. 2003, 91, 257−269.
(27) Jae, J.; Tompsett, G. A.; Foster, A. J.; Hammond, K. D.;
Auerbach, S. M.; Lobo, R. F.; Huber, G. W. J. Catal. 2011, 279, 257−
268, DOI: 10.1016/j.jact.2011.01.019.
(28) Cejka, J.; Wichterlova, B. Catal. Rev. 2014, 44, 375−421, DOI: ;
AUTHOR INFORMATION
Corresponding Author
*E-mail: Juha.Linnekoski@vtt.fi.
Notes
■
10.1081/CR-120005741.
(29) Roelofsen, D. P. Molecular sieve zeolites − properties and
applications in organic synthesis. Doctoral Thesis, Technical University
of Delft, 1972.
(30) Meier, W. M.; Olsen, D. H. Atlas of zeolite structure types;
Butterworths: London, 1992.
The authors declare no competing financial interest.
(31) van Laak, A. N. C. Post-Synthesis modifications on zeolites for
improved accessibility and catalytic performance. Doctoral thesis,
University of Utrecht, 2011.
ACKNOWLEDGMENTS
Funding from Tekes and Stora Enso Finland is gratefully
acknowledged.
■
(32) Jiang, Y.; Huang, J.; Dai, W.; Hunger, M. Solid State Nucl. Magn.
Reson. 2011, 39, 116−141.
REFERENCES
(33) Klinowski, J. Chem. Rev. 1991, 91, 1459−1479.
(34) Huang, J.; van Vegten, N.; Jiang, Y.; Hunger, M.; Baiker, A.
Angew. Chem., Int. Ed. 2010, 41, 7776.
■
(1) Gallezot, P. Catal. Today 2007, 121, 76−91.
(2) Swift, K. A. D. Top. Catal. 2004, 27, 143−155.
(35) Martens, J. A.; Souvrins, W.; van Rhijn, W.; Jacobs, P. A. In
Handbook of Heterogeneous Catalysis; Ertl, G., Knozinger, H.,
(3) Maki-Arvela, P.; Holmbom, B.; Salmi, T.; Murzin, D. Y. Catal.
̈
Rev. 2007, 49, 197−340.
̈
G
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Weitkamp, J., Eds.; Wiley-VCH: Weinheim, Germany, 1997; Vol. 1, pp
324−365.
(36) Sommer, J.; Jost, R.; Hachoumy, M. Catal. Today 1997, 38,
309−319.
(37) Rachwalik, R.; Olejniczak, Z.; Jiao, J.; Huang, J.; Hunger, M.;
Sulikowski, B. J. Catal. 2007, 252, 161−170.
(38) Selvaraj, M.; Pandurangan, A.; Seshadri, K. S.; Sinha, P. K.; Lal,
K. B. Appl. Catal. A: Gen. 2003, 242, 347−364.
(39) Stratakis, M.; Stavroulakis, M.; Sofikiti, N. J. Phys. Org. Chem.
2003, 16, 16−20.
(40) Narbeshuber, T. F.; Brait, A.; Kulathuyier, S.; Lercher, A. J.
Catal. 1997, 172, 127−136.
(41) Atkinson, D.; Curthoys, G. J. Phys. Chem. 1990, 84, 1358−1360.
H
dx.doi.org/10.1021/op500160f | Org. Process Res. Dev. XXXX, XXX, XXX−XXX