Ethanol conversion into olefins and aromatics over HZSM-5 zeolite: Influence of reaction conditions and surface reaction studies

Article abstract of DOI:10.1016/j.molcata.2016.03.005

Ethanol is an important renewable feedstock that represents an attractive alternative resource for petrochemical industry. In this work, the conversion of ethanol into hydrocarbons, particularly light olefins and aromatics, catalyzed by HZSM-5 was studied under different conditions of temperature (300–500?°C), partial pressure of ethanol (0.04–0.35?atm), and space velocity (165–0.65?g_ethanol?g_cat^?1?h^?1). At 500?°C and partial pressure of ethanol equal to 0.12?atm, the formation of propene was favored by an intermediate space velocity (6.5?g_ethanol?g_cat^?1?h^?1), whereas aromatics were favored by the lowest space velocity (0.65?g_ethanol?g_cat^?1?h^?1). The reaction route inferred from the catalytic tests was consistent with that found in the literature and was supported by the surface reaction studies performed by TPD of ethanol and TPSR. In situ DRIFTS of adsorbed ethanol confirmed that the alcohol is adsorbed as ethoxy species on the Br?nsted acid sites of the zeolite. With the increase in temperature, the adsorbed ethoxy species form diethyl ether and subsequently ethene confirming that the conversion of ethanol into ethene involves two consecutive steps.

Full text of DOI:10.1016/j.molcata.2016.03.005

Accepted Manuscript
Title: Ethanol conversion into olefins and aromatics over
HZSM-5 zeolite: Influence of reaction conditions and surface
reaction studies
´
Author: Zilacleide S.B. Sousa Claudia O. Veloso Cristiane A.
Henriques Victor Teixeira da Silva
PII:
S1381-1169(16)30072-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.03.005
MOLCAA 9804
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
12-10-2015
16-2-2016
1-3-2016
´
Please cite this article as: Zilacleide S.B.Sousa, Claudia O.Veloso, Cristiane
A.Henriques, Victor Teixeira da Silva, Ethanol conversion into olefins
and aromatics over HZSM-5 zeolite: Influence of reaction conditions
and surface reaction studies, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.03.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Ethanol conversion into olefins and aromatics over HZSM-5
zeolite: influence of reaction conditions and surface reaction
studies
Zilacleide S. B. Sousaa,1, Cláudia O. Velosob, Cristiane A. Henriquesa,b and Victor
Teixeira da Silvaa*
a Universidade Federal do Rio de Janeiro, NUCAT, Programa de Engenharia Química,
COPPE, P.O. Box 68502, 21941-914, Rio de Janeiro, RJ, Brazil.
b UERJ, Instituto de Química, Programa de Pós-graduação em Engenharia Química,
Rua São Francisco Xavier, 524 Maracanã, 20550-900, Rio de Janeiro,RJ, Brazil.
* Corresponding author: victor.teixeira@peq.coppe.ufrj.br, Phone: +55 21 3938 8344
Fax: +55 21 3938 8300. Email address: victor.teixeira@peq.coppe.ufrj.br
1 Present address: UERJ, Instituto de Química, Rua São Francisco Xavier, 524
Maracanã, 20550-900, Rio de Janeiro,RJ, Brazil
Authors e-mails: zilasousa@gmail.com (Z. S. B. Sousa); co.veloso@hotmail.com (C. O.
Veloso); cah@uerj.br (C. A. Henriques)

Graphical abstract
Highlights




Ethanol can be converted into ethane, propene and aromatics over HZSM-5.
DRIFTS and TPD data allowed to confirm the reaction route.
For temperatures lower than 300 oC the major products are ethane and diethyl ether.
At T = 500 oC the ratio propene/aromatics depends on space velocity.

Abstract
Ethanol is an important renewable feedstock that represents an attractive
alternative resource for petrochemical industry. In this work, the conversion of ethanol
into hydrocarbons, particularly light olefins and aromatics, catalyzed by HZSM-5 was
studied under different conditions of temperature (300–500 ºC), partial pressure of
-1
ethanol (0.04–0.35 atm), and space velocity (165–0.65 gethanol gcat h-1). At 500 ºC and
partial pressure of ethanol equal to 0.12 atm, the formation of propene was favored by
an intermediate space velocity (6.5 gethanol gcat-1 h-1), whereas aromatics were favored by
-1
the lowest space velocity (0.65 gethanol gcat h-1). The reaction route inferred from the
catalytic tests was consistent with that found in the literature and was supported by the
surface reaction studies performed by TPD of ethanol and TPSR. In situ DRIFTS of
adsorbed ethanol confirmed that the alcohol is adsorbed as ethoxy species on the
Brønsted acid sites of the zeolite. With the increase in temperature, the adsorbed ethoxy
species form diethyl ether and subsequently ethene confirming that the conversion of
ethanol into ethene involves two consecutive steps.
Keywords: Ethanol; Light olefins; Aromatics, HZSM-5, TPD of ethanol; DRIFTS

1. Introduction
The catalytic conversion of ethanol into valuable chemicals such as
hydrocarbons, higher chain alcohols, aldehydes, and esters has attracted significant
attention from academic and industrial researchers. In the petrochemical industry,
ethanol can be used as a green raw material for the production of light olefins (ethene
and propene) and aromatics (benzene, toluene and xylenes - BTX) replacing the
conventional processes, which are high-energy consuming and based on fossil resources
(oil and natural gas). Although the oil price currently shows a downward trend, the
political instability in producing countries causes uncertainty in the supply of oil and
natural gas. Moreover, environmental protection policies of different countries have
encouraged the use of renewable raw materials. Thus, ethanol that can be obtained from
the fermentation of molasses (1st generation ethanol) or from lignocellulose biomass,
which includes agricultural and forestry residues, by-products of wood transformation
industry, and herbal or ligneous plants (2nd generation ethanol), represents an attractive
alternative resource for petrochemical industry [1].
In recent years, the transformation of ethanol into hydrocarbons, particularly C3 -
C4 olefins and BTX aromatics, has attracted considerable attention. Zeolites have been
widely studied as catalysts for these reactions. Among them, ZSM-5 zeolites are the
most promising catalyst for the transformation of ethanol into petrochemical products
[1-6]. These zeolites exhibit acidic and structural characteristics that favor the
transformation of ethanol into not only ethene but also C3 - C8 hydrocarbons. Moreover,
some of their physicochemical properties like acidity, texture, and crystallite size can be
tailored by different synthesis routes or pos-synthesis treatments (such as metal
incorporation, basic leaching, and steaming) to enhance the selectivity to desired
reaction products [1].
The acid properties of ZSM-5 zeolites played an important role in the reaction
selectivity to specific hydrocarbon compounds. When ZSM-5 zeolites are compared
under similar experimental conditions (reaction temperature, ethanol partial pressure,
and contact time), it is observed that a moderate surface acidity favored the production
of propene, whereas higher strength and acid sites density favor the formation of
aromatic hydrocarbons. On the other hand, ethene is the only product formed over
HZSM-5 zeolites with lower acid sites density [5,7-9].

The influence of porous structure on product distribution and catalyst
deactivation was discussed by different authors [2, 10, 11]. Madeira et al. [2] and Phung
et al. [11] compared HZSM-5 with zeolites having similar Brønsted acid sites density
but different porous structure (HFAU, HMOR, HBEA). Despite the differences in the
reaction conditions (temperature and pressure), both groups reported that the higher
+
selectivity to C3 hydrocarbons was observed for HZSM-5 whereas the higher
production of ethene and diethyl ether was reported for large pore zeolites. Similar
trends were reported by Sousa et al. [10] who compared the performance of HZSM-5
and HMCM-22 zeolites with similar SiO2/Al2O3 molar ratios. Although both materials
were 10-MR and usually considered as medium pore zeolites, the latter has big cages
with 0.71 nm diameter and 1.82 nm height. Again, the zeolite with the more closed pore
+
structure (HZSM-5) presented the higher selectivity to propene and C3 hydrocarbons,
ethene being the main product on HMCM-22. The authors mentioned above also
observed that the deactivation by coke was slower on HZSM-5 due to the steric
restriction imposed by its pore structure on the growth of the coke molecules.
In addition to the physicochemical properties of the zeolites, reaction conditions
such as temperature, pressure, and weight hourly space velocity (WHSV) significantly
influence catalyst activity and product distribution.
The effect of reaction temperature (300 – 600 ºC) on the conversion of ethanol
was investigated by several authors [5,7,9,12,13,14]. Their results indicate that the
formation of light olefins (ethene and propene) was favored by the increase in reaction
temperature since higher temperatures promoted the dehydration of ethanol to ethene
and its oligomerization to higher hydrocarbons, which yielded more ethene and propene
by subsequent cracking. On the other hand, the selectivity toward liquid hydrocarbons
passed through a maximum at temperatures between 400 ºC and 500 ºC because above
this temperature, liquid hydrocarbons decreased as a result of cracking reactions, and
methane and coke content increased. However, the optimum temperature to produce
each fraction varied among the different authors probably due to differences in the
chemical composition of the studied zeolites (silica/alumina ratio - SAR) and
experimental conditions (WHSV and ethanol partial pressure).
The influence of contact time (1/WHSV) was strongly dependent on the acid
sites density of the zeolite. For short contact times, ethene was the main product

regardless the framework SAR of the zeolite. For an HZSM-5 with lower acid sites
density (framework SAR of 280, for example) the yield of ethene decreases
continuously with increasing contact time, whereas the yields of propene and butenes
increased. The formation of aromatics was not observed over this zeolite [10]. On the
other hand, for zeolites with lower SAR, ethene was the only product for short contact
times, but an increase in contact time decreased ethene yield, whereas the yield of
+
propene and butenes increased passing through a maximum, and those of C5 aliphatics
and aromatics increased [5,8,9,12]. Thus, these results suggested that secondary
reactions of oligomerization, hydrogen transfer and cracking were favored by enhanced
contact time. These reactions were also favored by the increase in reaction pressure as
reported by Costa et al. [9] who observed the raise in paraffin and aromatic contents and
the reduction in olefin content with the increase in this parameter. Pressures around 30
bar were used to produce distillate-range hydrocarbons with a petrochemical
characteristic through the transformation of ethanol into hydrocarbons over zeolites. On
zeolite HZSM-5 (SAR = 40), the main products were paraffins and aromatics from C5 to
+
C11. High yields of C3 hydrocarbons and low amounts of ethene and diethyl ether were
obtained [2,15,16].
Based on the product distribution observed in the catalytic tests, different
reaction routes have been proposed in the literature for ethanol conversion over zeolites
[1]. The most commonly accepted route involves the intermolecular dehydration of
ethanol to ethyl ether, followed by the dehydration of the latter to ethene (at higher
temperatures, ethene can also be formed directly by intramolecular dehydration). Then,
ethene undergoes successive reactions like oligomerization, cracking, cyclization and
aromatization to form higher hydrocarbons [1,2,5,17,18]. Some difference among the
routes found in the literature relies on the mechanism of propene formation. Ingram and
Lancashire [19] proposed that propene is formed from the cracking of C6 olefins
(produced by ethene oligomerization) whereas Takahashi et al. [8] proposed that
carbene species be the transient intermediate in the direct production of propene from
ethene.
As part of an extensive research on the conversion of ethanol into hydrocarbons,
in this work we investigate ethanol conversion over HZSM-5 zeolite (SAR = 25) under
different reaction conditions such as reaction temperature, ethanol partial pressure

(pEtOH), and space velocity (WHSV). The results were discussed regarding ethanol
conversion and product distribution. They allowed the selection of the best condition to
produce light olefins (mainly propene) or aromatics (BTX). Additionally, temperature
programmed desorption (TPD), temperature programmed surface reaction (TPSR) and
in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) of ethanol
were performed to provide further insights to support the reaction route proposed from
the catalytic tests and to confirm the nature of the species adsorbed on HZSM-5 active
sites.
2. Experimental
2.1 Catalyst
The HZSM-5 zeolite, supplied by CENPES/PETROBRAS (Rio de Janeiro,
Brazil), had both chemical and framework SAR (SiO2/Al2O3 molar ratio) values equal
to 25 (nominal value), as well as a micropore volume of 0.165 cm3 g-1. More
information on the physicochemical characteristics of this zeolite can be found in a
previous work [10].
2.2 Catalytic Evaluation
Ethanol conversion was studied in a fixed-bed flow micro-reactor under
atmospheric pressure. Ethanol vapor was generated by passing an N2 stream through a
saturator maintained at a constant desired temperature by means of a thermostatic bath,
and then to the reactor containing the catalyst. The catalytic tests were carried out at
200, 300, 400, and 500 °C with a partial pressure of ethanol of 0.04, 0.12, 0.20 and 0.35
-1
atm and a space velocity of 0.65, 6.5, 65 and 165 gEtOH gcat h-1. The reactor effluent
was analyzed by an online gas chromatograph (with a 30-m HP-Plot/Q capillary column
and a flame ionization detector). Before the catalytic evaluation, the zeolite was
thermally pre-treated to eliminate water and other adsorbed species, using a flow of N2
(50 mL min-1) ramped from room temperature up to 500 °C at a heating rate of 2 °C
min-1 and then kept at this temperature for 1 h.
2.3 Temperature-programmed desorption of ethanol (TPD-ethanol)

A home-made instrument equipped with an online quadrupole mass
spectrometer (Balzers QMS 422) was used in this study. The samples were pretreated at
500 C for 1 h under flowing He (30 mL min-1) and then cooled down to 25 C. Ethanol
was chemisorbed at room temperature by passing He (60 mL min-1) through a saturator
containing ethanol for 30 min. Afterward, the sample was purged for 1 h in He flow (60
mL min-1) to remove any physically adsorbed ethanol, and then the temperature was
increased at a heating rate of 10C min-1 up to 500 C. The desorbing species were
continuously monitored by their characteristic mass fragments (m/e): 46 (ethanol), 24
(ethene), 59 (diethyl ether), 18 (water), 2 (hydrogen), 43 (acetaldehyde), 40 (olefins C3-
C4), 78 (benzene), and 91 (toluene). The intensities were corrected to account for the
contributions of different compounds.
2.4 Temperature-programmed surface reaction (TPSR ethanol)
Temperature-programmed surface reaction measurements were carried out using
the same apparatus used for the TPD of ethanol. Initially, the sample was pretreated in
situ, as described in section 2.3. Then, the sample was cooled down to 50 °C and the
ethanol/He mixture (60 mL min-1) was fed to the reactor. The temperature was increased
at a heating rate of 5 ºC min-1 up to 500 C. The reaction products were collected at
temperatures of 50, 100, 150, 200, 250, 300, 350, 400, 450 and 500 C. The desorbing
species were monitored by their characteristic mass fragments (m/e): 46 (ethanol), 24
(ethene), 59 (ethyl ether), 18 (water), 2 (hydrogen), 43 (acetaldehyde), 40 (olefins C3-
C4), 78 (benzene), and 91 (toluene). The results were corrected to eliminate
interferences from other compounds.
2.5 In situ Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS)
In situ DRIFTS analyses were carried out using a Perkin Elmer Spectrum 100
spectrometer equipped with an MCT-A detector and a high-temperature chamber
(Harrick) with ZnSe windows. Spectra were acquired at a resolution of 4 cm-1 and 150
scans. The zeolite was pre-treated under He flow (30 mL min-1) from room temperature
to 500 °C at a heating rate of 5 °C min-1. Then, the zeolite was cooled down to 40ºC,
and a stream of saturated ethanol/He was introduced into the chamber, which was
maintained at a temperature of 40 °C for 30 min. After removing the reversibly

adsorbed ethanol using a flow of He (30 mL min-1), the catalyst was heated to different
temperatures (100, 150, 200, 250, 300, 400 and 500 °C) under He flow. Spectra were
acquired at the end of the ethanol adsorption step and at each of the temperature stages.
The spectrum of the treated zeolite was used as background.
3. Results and Discussion
3.1. Effect of temperature
The influence of reaction temperature was evaluated in catalytic tests in which
the partial pressure of ethanol (0.12 atm) and space velocity (6.5 gEtOH gcat.-1 h-1) were
kept constant. At temperatures of 300, 400, and 500 C, the ethanol conversion was
complete, while at 200 C the conversion was only approximately 40 %. Figures 1 to 4
show the product distribution for each temperature. Table 1 compares the distribution of
products formed at a reaction time of 10 min, when the possible effects of deactivation
of the catalytic sites by coke formation were insignificant.
Figure 1 and Table 1 show that when the reaction was conducted at 200 C,
ethyl ether (DEE) was the main product, followed by the formation of ethene. At this
temperature, no change was observed in the conversion of ethanol, which remained at
40 %, or in the distribution of products (only ethene and ethyl ether), throughout 15 h of
reaction. To evaluate the effect of ethanol conversion on the product distribution, an
-1
additional catalytic test was performed at 200 C and 1.3 gEtOH gcat h-1. Under these
conditions, ethanol conversion reached 60 %. Although the amount of ethene increased,
DEE was still the main product formed. These results suggested that at this temperature,
the conversion of ethanol into ethene mainly took place via the route including ethyl
ether as an intermediary.
At 300 C, the conversion of ethanol was complete and the formation of DEE
was no longer observed (Table 1). Ethene was the main product formed (87 %), along
+
with small amounts (<5 %) of propene, butenes, and C6 , which decreased along the
reaction and became negligible after 6 h. Thereafter, ethene was the only product
(99 %).
At 400 °C and after a short time on stream (TOS = 10 min) the formation of
=
propene, paraffins (C2 – C4), and the C5 (C5 – C5 ) fraction was favored at the expense
of ethene (Table 1). Figure 3 shows that the formation of ethene declined slightly during

the first 6 h of the reaction, passed through a minimum and then increased, while the
formation of propene, butenes, and paraffins showed the inverse behavior.
The product distribution at the beginning of the reaction at 500 °C indicated that
the formation of propene, butenes, aromatics, and paraffins was favored with increasing
reaction temperature. The increase of propene yield with increasing temperature up to
500 °C has also been reported by different authors [5,7] for the conversion of ethanol
catalyzed by HZSM-5. Concerning the yields of aromatics and paraffins literature
claims that they pass through a maximum at intermediate temperatures [5,7,9,12,13,14].
However, in the present work these maxima were not observed probably due the large
intervals between the studied temperatures.
Figure 4 reveals that the formation of light olefins, ethene, propene, and butenes
decreased smoothly during the first 6 h of the reaction, while the formation of aromatics
and paraffins increased slightly. After 6 h, the formation of ethene increased, reaching
approximately 70 % of the total amount of product formed at TOS = 15 h. On the other
hand, the formation of propene and other hydrocarbons decreased with increasing
reaction time.
The results obtained in this work confirm that temperature plays an important
role in the product distribution over HZSM-5 zeolites. Assuming the reaction scheme
proposed by Inaba et al. [17], at low temperatures (200 C) the intermolecular
dehydration of ethanol occurs, forming ethyl ether, and the subsequent dehydration of
ethyl ether produces ethene. An increase in temperature to 300 C favors the selective
formation of ethene, which is apparently formed from the intramolecular dehydration of
ethanol, although its formation from DEE cannot be completely ruled out. Considering
that ethyl ether was not identified among the reaction products, its formation and
subsequent conversion into ethene might occur at very high rates at temperatures above
300 °C. Further increases in temperature favor the oligomerization of ethene to form
higher olefins, which then undergo aromatization, hydrogen transfer, and cracking
reactions. Thus, the formation of higher hydrocarbons (olefins, paraffins, and aromatics)
increases considerably, while the formation of ethene decreases. This effect was
enhanced with a further increase in temperature to 500 C. At this temperature, there is
+
a reduction in the formation of the C5 and C6 fractions (non-aromatic), probably due to
their cracking to lighter fractions.

For reactions carried out at or above 300 °C, although the conversion of ethanol
continues to be complete along the reaction, the formation of ethene continuously
increases, while the formation of higher hydrocarbons decreases, particularly after 6 h.
These results suggest that the strongest acid sites responsible for oligomerization
reactions, cracking, aromatization, and hydrogen transfer undergo gradual deactivation
during the reaction due to the formation of coke. Weaker acid sites are less susceptible
to coking and would remain active to promote the conversion of ethanol into ethene.
3.2. Effect of partial pressure of ethanol
The effect of the partial pressure of ethanol on the product distribution was
evaluated in tests conducted at 500 °C and a WHSV of 6.5 gEtOH gcat-1 h-1. The influence
of reaction time at each partial pressure is shown in Figures 5 to 7, while Table 2 shows
the distribution of products at TOS = 10 minutes. Under the studied conditions, ethanol
conversion was complete throughout the reaction (15 h) for all partial pressures
investigated. Regardless of the partial pressure of ethanol, ethene was the primary
product formed, and the formation of ethyl ether was not observed.
At a partial pressure of 0.04 atm (Figure 5), ethene was the main product. With
increased TOS, a slight decrease in ethene formation occurred, which was accompanied
by the production of olefins with 3, 4, and 5 carbon atoms. These results were consistent
with the fact that oligomerization reactions, which are polymolecular, were not favored
at a low partial pressure of the reactants.
The rise in ethanol partial pressure to 0.12 atm favored the formation of butenes,
aromatics, paraffins and propene at the expense of ethene. As observed in Figure 4, with
increased TOS, the formation of light olefins, ethene, propene, and butenes decreased
during the first 6 h, while the formation of aromatics and paraffins slightly increased.
After 6 h, the formation of ethene increased, reaching approximately 70 % of the total
products formed at the end of the reaction (15 h). On the other hand, the formation of
propene and other hydrocarbons decreased with increasing reaction time. A further
increase in the partial pressure to 0.20 atm promoted the formation of aromatics and
paraffins (C2-C4), but there was a slight decrease in the formation of propene. The
+
formation of methane and higher hydrocarbons (C6 ) was negligible.

At a partial pressure of 0.35 atm, a significant reduction in the formation of
ethene accompanied by a large increase in the formation of paraffins and aromatics was
observed. The formation of propene, butenes, and the C5 fraction was not influenced by
the increase in partial pressure.
As observed in Figures 6 and 7, the effect of TOS on product distribution was
similar at partial pressures of 0.20 and 0.35 atm. Thus, with increasing reaction time, the
formation of propene and butenes was maximized at approximately 6 h, whereas that of
C2 - C4 paraffins and aromatics passed through a maximum at shorter reaction times. On
the other hand, the formation of ethene continuously increased over time.
The obtained results confirmed that the increase in partial pressure of ethanol
favored bi- and polymolecular reactions such as those associated with the formation of
higher paraffins and aromatics, as well as reactions related to coke formation. Thus, at
shorter reaction times, the formation of propene and aromatics was greatly affected, but
at longer TOS, coke formation begins to dominate, deactivating the strong acid sites.
The weaker acid sites were less influenced by coke and remained active to promote the
conversion of ethanol into ethene.
3.3. Effect of space velocity
The influence of the space velocity on the distribution of products was evaluated
in tests conducted at 500 °C and a partial pressure of ethanol equal to 0.12 atm. The
influence of TOS on the product distribution is shown in Figures 8 to 10.
Under the studied conditions, ethanol conversion was complete throughout the
reaction (15 h). Regardless of the space velocity employed, ethene was the main
product, and the formation of ethyl ether was not observed.
Regarding the effect of reaction time on the product distribution, for a space
velocity of 165 gEtOH gcat-1 h-1 (Figure 10), ethene was the only product formed during
the reaction (15 h).
-1
For a space velocity of 65 gEtOH gcat h-1 (Figure 9), the product distribution
remained stable during the first 5 h of the reaction. Thereafter, the oligomerization of
ethene was inhibited, and the formation of propene, butenes, and pentenes decreased
simultaneously with the increase in ethene production.

For the space velocity of 6.5 gEtOH gcat-1 h-1 (Figure 4), as discussed earlier, an
increase in TOS led to a decrease in the formation of light olefins, propene and butene.
The formation of aromatics and paraffins increased slightly up to TOS = 6 h and then
decreased smoothly in an opposite trend to that observed for ethene.
The effect of increased reaction time was significant at a space velocity of 0.65
gEtOH gcat-1 h-1 (Figure 9). The formation of light olefins (ethene, propene, butenes, and
pentenes) increased slightly with time, while the production of aromatics and paraffins
slightly decreased. These results reflected the effect of coke formation, which had a
significant influence on hydrogen transfer and aromatization reactions.
Figure 11 shows the product distribution as a function of contact time
(1/WHSV) under conditions in which the effect of coke deposition was minimized
(TOS = 10 min). The formation of ethene continuously decreases with an increase in
contact time, whereas the formation of propene and butenes increases, passes through a
maximum at 0.2 gEtOH gcat-1 h-1 and then decreases. The production of C2-C4 paraffins
and aromatics is favored at higher contact times.
As shown in Figure 11, the gradual decrease in space velocity under conditions
in which coke formation can be neglected (TOS = 10 min) initially favored the
formation of olefins containing 3 and 4 carbon atoms, and subsequently the production
of paraffins and aromatics. Under these conditions (500 °C, 0.12 atm), ethene is the
only primary product of the reaction (Figure 11). With increasing contact time, ethene
undergoes oligomerization, which leads to the formation of higher olefins that rapidly
undergo cracking and form propene and butenes.
For lower space velocities, there is a reduction in the formation of light olefins
(which behave as reaction intermediates) and a consequent increase in paraffins and
aromatics. This fact confirms that these latter products were generated from olefin
oligomerization followed by aromatization, cracking, and hydrogen transfer, in
agreement with the scheme proposed by Inaba et al. [17].
Similar trends have been reported by Aguayo et al. [14], who studied the effect
of space velocity on the conversion of ethanol to hydrocarbons over an HZSM-5 zeolite
with an SAR equal to 48, at 450 ºC. A reduction in space velocity led to an increase in
+
the formation of C5 compounds and paraffins, whereas the production of ethene
decreased. The authors also observed that depending on the space velocity, the

formation of propene and butenes could be maximized. At higher space velocities,
ethene was the main product.
Taking into account the improvement in the production of light olefins (mainly
propene) and aromatics (BTX), the above results indicate that at 500 °C and a partial
pressure of ethanol equal to 0.12 atm, two space velocities can be selected: (i) 6.5 gEtOH
gcat-1 h-1 to maximize the yield of light olefins, particularly propene, and (ii) 0.65 gEtOH
,
gcat-1 h-1, to obtain the highest yield of BTX aromatics.
3.4 Surface reaction studies
Aiming at providing further insights to support the reaction route inferred from
the catalytic tests (sections 3.1 and 3.3), the compounds desorbed from zeolite surface
throughout the ethanol conversion were investigated via temperature-programmed
desorption of ethanol (ethanol TPD) and temperature-programmed surface reaction of
ethanol (ethanol TPSR). The nature of the species adsorbed on HZSM-5 active sites
after ethanol adsorption as well as along the reaction were followed by in situ diffuse
reflectance spectroscopy (DRIFTS).
3.4.1 Temperature-programmed desorption of ethanol (ethanol TPD)
Temperature-programmed desorption of ethanol was performed to evaluate the
primary products associated with the conversion of ethanol catalyzed by HZSM-5
zeolite. Figure 12 shows the TPD profiles after adsorption of ethanol at room
temperature. It can be observed that the desorption of molecular ethanol (unreacted
ethanol) starts at 70 °C, passes through a maximum at 150 °C, and ends at 250 °C. At
150 °C, the intermolecular dehydration of adsorbed ethanol begins with the
simultaneous formation of H2O and ethyl ether. Until 190 C, this is the main reaction
on the surface of the zeolite.
The simultaneous desorption of ethene and ethyl ether is observed between 190
°C and 250 °C, which confirms the results found in catalytic tests. Thus, at low
temperatures (T < 300 °C), the formation of ethene from ethanol would mainly occur
via intermolecular dehydration, with ethyl ether as an intermediate. The preferential
formation of ethyl ether at lower temperatures was also observed by Decanio et al. [20]
using TPD of ethanol adsorbed on Al2O3/F. The desorption of propene or higher

+
hydrocarbons (C3 ) was not observed, probably due to the absence of ethanol adsorbed
at temperatures higher than 300 ºC.
Acetaldehyde and H2 were not identified among the desorbed species, indicating
that ethanol dehydrogenation did not occur on the studied zeolite. The higher density
and strength of the acid sites of this zeolite (SAR= 25) should explain these trends,
since acetaldehyde and H2 desorption in TPD of ethanol was reported for catalysts with
metallic/oxide sites [21] or HZSM-5 with lower acid sites density (higher SAR) [22].
3.4.2 Temperature-programmed surface reaction (ethanol TPSR)
The conversion of ethanol was also studied by TPSR over the HZSM-5 sample,
as shown in Figure 13. It can be observed that between 150 °C and 250 C, ethanol was
transformed into ethyl ether via intermolecular dehydration and ethyl ether successively
underwent dehydration to form ethene. Ether formation reached a maximum at 200 °C
and subsequently decreased with increasing temperature, whereas ethene gradually
increased. Above 300 C, ethanol was directly converted to ethene and small amounts
of higher olefins (propene and butenes).
The conversion of ethanol into hydrocarbons at temperatures below 300 C
begins with the dehydration of ethanol to ethyl ether, which is then dehydrated to
ethene, as previously inferred by catalytic tests and confirmed by TPD of ethanol.
2 C2H5OH  C2H5 – O – C2H5 + H2O
C2H5 – O – C2H5 2 C2H4+ H2O
At higher temperatures (> 300 ºC), the direct dehydration of ethanol to ethene
occurs and ethene is subsequently converted to higher olefins by oligomerization and
cracking.
C2H5OH  H2O + C2H4
C2H4  (CH2)n  C3H6 + C4H8
Differently to what was observed in the catalytic tests, the formation of
aromatics was not identified in TPSR experiments at 400 ºC or 500 ºC. The lower
ethanol partial pressure (0.08 atm) employed in these experiments could explain these
results.
3.4.3 In situ DRIFTS

The infrared spectrum in the region of hydroxyl vibrations (3800 - 3400 cm-1)
showed the presence of two bands. The band at 3600 - 3605 cm-1 corresponds to OH
groups related to Si-(OH)-Al bridging (Brønsted acid sites), and that at 3740 cm-1 to the
hydroxyl groups of silanol species (Si-OH). After ethanol adsorption at 40 ºC, the band
associated with Si(OH)Al disappeared, whereas the band assigned to the silanol groups
was only slightly affected, indicating that ethanol preferentially adsorbs on the Brønsted
acid sites.
Figure 14 shows the spectra in the region of fundamental vibrations of the
reactive intermediates formed on the surface of HZSM-5. The spectra were recorded
immediately after ethanol adsorption at 40 °C (spectrum (a)) and after increasing
temperature to 100, 150, 200, 250, 300, 400 and 500 °C (spectra (b)-(h), respectively) in
flowing helium. Spectrum (a) shows that ethanol adsorption at 40 °C led to the
appearance of bands located at 2980, 2935, 2907, 2880, 1450, 1180 and 1060 cm-1. As
shown in Table 3, these bands can be ascribed to the different vibrational modes of
ethoxy species formed due to the adsorption of ethanol on Brønsted acid sites, which is
accompanied by the formation of a water molecule [23-30]. However, the unequivocal
rejection of molecular ethanol adsorption was not possible since it presents band at
similar positions.
Thus, ethanol can exist on the surface of the zeolite as: (i) ethoxy species formed
by the dissociative adsorption of ethanol on the Lewis (cleavage of O - H bond) or
Brønsted (cleavage of C - O bond) sites, and (ii) undissociated ethanol molecules.
After the acquisition of the spectrum of adsorbed ethanol at 40 ºC, the
temperature was increased under helium flow. At 100 ºC (spectrum (b)), there were no
changes in the spectrum in the ranges 3200 – 2800 cm-1 and 2800 – 800 cm-1. At 150
ºC, small changes were observed only in the range 2800 – 800 cm-1; the band at 1060
cm-1 decreased, and a band at 1142 cm-1, related to C-O stretching in ethyl ether,
appeared [31]. Ethyl ether could be formed by the interaction between two adsorbed
ethoxy species, as well as by the reaction between an adsorbed ethoxy and an ethanol
molecule adsorbed without dissociation [20].
The C-O stretching bands associated with both ethyl ether (1142 cm-1) and
ethoxy species (1060 cm-1) disappeared at 200 ºC. Simultaneously, the appearance of
the bands at 2988 cm-1 and 950 cm-1 should indicate the transformation of ethoxy

species into ethene [23]. These results confirm that the conversion of ethanol into ethene
involves two consecutive steps, which correspond to ethanol conversion into diethyl
ether and subsequently to ethene through the adsorbed ethoxy species.
At 250 ºC, the bands assigned to ethene became less pronounced, and the band at
1450 cm-1 disappeared (spectrum (e)). New bands also appeared at 1508 cm-1 and 2967
cm-1. According to Tynjälä et al. [24], the band near 2970 cm-1 can be associated with
ethene conversion into higher hydrocarbons. The band at 1508 cm-1, which decreased
with increasing temperature (spectra (e)–(f)), corresponds to polyenic species that
would be formed by the dehydrogenation of oligomeric species, as proposed by Lin et al
[32]. These authors suggested that the cracking of the adsorbed oligomeric species
could lead to the formation of propene and other olefins. With an increase of
temperature to 300 ºC, the intensity of the bands related to CH2 and CH3 stretching
decreased and virtually disappeared at 400 ºC. At 300 ºC, a low-intensity band at 3016
cm-1, assigned to CH stretching of olefinic carbon, was observed. This band disappeared
above 400 ºC, similar to the behavior of bands associated with the stretching of methyl
and methene groups.
DRIFTS study of adsorbed ethanol strongly suggests that at 40 ºC ethanol was
adsorbed on the surface of HZSM-5 as ethoxy species. With increasing temperature,
these species reacted, successively forming ethyl ether, ethene and higher hydrocarbons,
as observed by TPD and TPSR experiments. In the studied conditions, the unequivocal
rejection of dissociative adsorption of ethanol was not possible. Moreover, the
formation of chemically adsorbed water was not observed, as also reported by Golay et
al. [26], who investigated the reaction of ethanol on -Al2O3.
5. Conclusions
The transformation of ethanol into hydrocarbons over an HZSM-5 zeolite was
investigated under different operating conditions (reaction temperature, partial pressure
of ethanol and space velocity). The results show that the studied parameters influenced
the yields of propene and aromatics significantly. Considering propene and BTX
aromatics as the products of interest, at 500 ºC and partial pressure of ethanol equal to
0.12 atm, two different space velocities must be selected: : (i) space velocity equal to

6.5 gEtOH gcat-1 h-1 to obtain light olefins, particularly propene; (ii) space velocity equal to
0.65 gEtOH gcat-1 h-1 to obtain BTX aromatics.
In situ DRIFTS analysis of adsorbed ethanol confirmed that ethanol is adsorbed
as ethoxy species on the Brønsted acid sites of the zeolite. Both catalytic tests and
surface reaction studies suggest that at lower temperatures, the formation of ethyl ether
occurred via intermolecular dehydration, and then it was converted into ethene. On the
other hand, at higher temperatures, the direct conversion of ethanol into ethene
proceeded by intramolecular dehydration. Subsequently, ethene oligomerization forms
heavier olefins, which undergo (i) cracking, producing light olefins (propene, butenes);
or (ii) cyclization and hydrogen transfer, forming aromatic and paraffin.
Acknowledgements
The authors acknowledge CNPq and CENPES/PETROBRAS for the financial
support. ZSBS thanks CNPq for the DSc scholarship. CAH acknowledges UERJ for the
grants (PROCIENCIA Program).

References
[1] A. Galadima, O. Muraza, J. Ind. Eng. Chem. 31 (2015) 1-14.
[2] F. Ferreira Madeira, N.S. Gnep, P. Magnoux, S. Maury, N. Cadran, Appl. Catal.
A-Gen. 327 (2009) 39-46.
[3] F. Cherubini, Energy Convers. Management 51 (2010) 1412-1421.
[4] D. Zhang, R. Wang, X. Yang, Catal. Lett. 124 (2008) 384-391.
[5] Z. Song, A. Takahashi, N. Nimura, T. Fujitani, Catal. Lett. 131 (2009) 364-369.
[6] M. Inaba, K. Murata, I. Takahara, K. Inoue, Adv. Mater. Sci. Eng. Sci. (2012) 1-7.
[7] A.K. Talukdar, K.G Bhattacharyya, S. Sivasanker, Appl. Catal. A: Gen 148
(1997) 357-371.
[8] A. Takahashi, W. Xia, Q. Wu, T. Furukawa, I. Nakamura, H. Shimada, T. Fujitani,
Appl. Catal. A 467 (2013) 380-385.
[9] E. Costa, A. Uguina, J. Aguado, P.J. Hernández, Industrial Eng. Chem. Process Des.
Dev. 24 (1985) 239-244.
[10] Z.S.B. Sousa, D.V. Cesar, C.A. Henriques, V. Teixeira da Silva, Catal. Today 234
(2014) 182-191.
[11] T. K. Phung, L. Projetti Hernandez, A. Lagazzo, G. Busca, Appl. Catal. A: Gen
493 (1015) 77-x.
[12] K.K. Ramasamy, Y. Wang, Catal. Today 237 (2014) 89-99.
[13] K. Murata, M. Inaba, I. Takahara, J. Jpn. Petrol. Inst. 51 (2008) 234-239.
[14] A.T. Aguayo, A.G. Gayubo, A.M. Tarrio, A. Atutxa, J. Bilbao, J. Chem. Technol.
Biotechnol. 77 (2002) 211-216.
[15] F. Ferreira Madeira, N. S. Gnep, P. Magnoux, H. Vezin, S. Maury, N. Cadran,
Chem, Eng. J. 161 (2010) 403-408.
[16] F. Ferreira Madeira, K. Ben Tayeb, L. Pinard, H. Vezin, S. Maury, Appl. Catal. A:
Gen. 443-444 (2012) 171-180.
[17] M. Inaba, K. Murata, M. Saito, I. Takahara, Kinetics Catal. Lett. 88 (2006) 135-
142.
[18] K. Inoue, M. Inaba, I. Takahara, K. Murata, Catal . Lett 130 (2010) 14-19.
[19] C.W. Ingram, R.J. Lancashire, Catal. Lett. 31 (1995) 395-403.
[20] E.C. Decanio, J.W. Bruno, V.P. Nero, J.C. Edwards, J. Catal. 140 (1993) 84-102.

[21] S. M. Lima, I. O Cruz, G. Jacobs, B. H. Davis, L. V. Mattos, F. B. Noronha, J.
Catal. 257 (2008) 256-268.
[22] R. Barthos, A. Szechenyi, F. Solymosi, J. Phys. Chem. B, 110, (2006) 21816-
21825.
[23] P-Y. Sheng, G.A. Bowmaker, H. Idriss, Appl. Catal. A: Gen. 261 (2004) 171-181.
[24] P. Tynjala, T. T. Pakkanen, S. Mustamaki, J. Phys. Chem. B, 102 (1998) 5280-
[25] J. Wan, F. Chang, Y. Wei, Q. Xia, Z. Liu, Z., , Catal. Lett. 127 (2008) 348-353.
[26] S. Golay, R. Doepper, A. Renken, Appl. Catal. A: Gen. 172 (1998) 97-106.
[27] H. Knozinger, B. Stubner, J. Phys. Chem. 82 (1978) 1526-1532.
[28] M.A. Natal-Santiago, J.A. Dumesic, J. Catal. 175 (1998) 252–268.
[29] J.N. Kondo, K. Ito, E. Yoda, F. Wakabayashi, K. Domen, J. Phys. Chem. B 109
(2005) 10969-10972.
[30] K. Rintramee, K. Föttinger, G. Rupprechter, J. Wittayakun, Appl. Catal. B-
Environ., 115-116 (2012) 225-235.
[31] G.A.M. Hussein, N. Sheppard, M.I. Zaki, R.B. Fahin, J. Chem. Soc. Faraday
Trans. 87 (1991) 2661-2668.
[32] B. Lin, Q. Zhang, Y. Wang, Ind. Eng. Chem. Res. 48 (2009) 10788–10795

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9