Byproducts formation in the ethyl tert-butyl ether (ETBE) synthesis reaction on macroreticular acid ion-exchange resins

Article abstract of DOI:10.1016/j.apcata.2013.09.012

Ethyl tert-butyl ether (ETBE) production is one of the industrial processes of major importance today in Europe. However, the study of side reactions in this synthesis reaction appears scarcely in the open literature. Side reactions take place along with the etherification of C4 olefinic cuts with ethanol, catalyzed by acidic ion-exchange resins. In this work, byproducts formation is studied in a batch reactor. The presence of diethyl ether (DEE), ethyl sec-butyl ether (ESBE), dimers of isobutene (2,4,4-trimethyl-1-pentene (TMP-1) and 2,4,4-trimethyl-2-pentene (TMP-2)) and tert-butyl alcohol (TBA) has been studied in terms of production and selectivity. The effect of temperature, ranging from 323 to 383 K, and the influence of initial molar ratio ethanol/isobutene (R _A/O), ranging from 0.5 to 2.0, on byproducts formation have been analyzed. All byproducts formation was favored by high temperatures. A low initial molar ratio ethanol/isobutene favored the formation of DEE, ESBE, TMP-1 and TMP-2, whereas high molar ratios favored TBA formation.

Full text of DOI:10.1016/j.apcata.2013.09.012

Applied Catalysis A: General 468 (2013) 384–394
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Byproducts formation in the ethyl tert-butyl ether (ETBE) synthesis
reaction on macroreticular acid ion-exchange resins
J.H. Badia, C. Fité, R. Bringué, E. Ramírez, F. Cunill∗
Chemical Engineering Department, Faculty of Chemistry, University of Barcelona, C/Martí i Franquès 1-11, 08028 Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 May 2013
Received in revised form 5 September 2013
Accepted 10 September 2013
Available online 19 September 2013
Ethyl tert-butyl ether (ETBE) production is one of the industrial processes of major importance today
in Europe. However, the study of side reactions in this synthesis reaction appears scarcely in the open
literature. Side reactions take place along with the etherification of C4 olefinic cuts with ethanol, catalyzed
by acidic ion-exchange resins. In this work, byproducts formation is studied in a batch reactor. The
presence of diethyl ether (DEE), ethyl sec-butyl ether (ESBE), dimers of isobutene (2,4,4-trimethyl-1-
pentene (TMP-1) and 2,4,4-trimethyl-2-pentene (TMP-2)) and tert-butyl alcohol (TBA) has been studied
in terms of production and selectivity. The effect of temperature, ranging from 323 to 383 K, and the
influence of initial molar ratio ethanol/isobutene (R◦A/O), ranging from 0.5 to 2.0, on byproducts formation
have been analyzed. All byproducts formation was favored by high temperatures. A low initial molar ratio
ethanol/isobutene favored the formation of DEE, ESBE, TMP-1 and TMP-2, whereas high molar ratios
favored TBA formation.
Keywords:
ETBE
Byproducts formation
Selectivity
Acidic ion-exchange catalysts
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
reactor followed by final conversion in a catalytic distillation
column. In both units, a bed of ion exchange resin catalysts is used
The use of ethyl tertiary-butyl ether (ETBE) became widely
spread in the European Union (EU) since oxygenated ethers
were selected as preferred octane enhancers due to consecutive
directives regulating gasoline composition. The current Renew-
able Energy Directive (2009/28/EC) and Fuel Quality Directive
(2009/30/EC) still are the main drivers for changes in fuel compo-
sition for the foreseeable future. These directives propose a 10%
value for biofuels in the market share in the EU, by the end of
2020, and an increase in the oxygen content limit up to a 3.7 wt%,
allowing a higher oxygenates content in the base pool for gasoline
[1,2].
Industrially, ETBE is obtained under mild conditions by
means of a heterogeneous catalytic etherification reaction of
2-methylpropene (isobutene) with ethanol, in which macroretic-
ular ion exchange acidic resins of polystyrene–divinylbenzene
matrix are used as catalysts [3–5]. Reaction between ethanol and
isobutene to produce ETBE (reaction (1) in Fig. 1) is reversible and
moderately exothermic (heat of reaction of −35 kJ/mol at 298 K)
[6,7]. In catalytic fixed-bed tubular reactors, low inlet temperatures
and a slight excess of ethanol are typical operating conditions in
order to avoid possible byproducts formation. Likewise, ETBE can
be formed by means of catalytic distillation technology, based on
a two-step reactor design, consisting of a boiling point fixed bed
[8–10].
By-products formation in the industrial process is related to
the feed ratio alcohol/olefin, to the sources for the two reactants
and to the hot-spots that could take place within the catalytic bed,
where temperatures above 373 K could be reached. With regards
to the reactants sources, isobutene comes mainly from the efflu-
ent of two petrochemical processes – the fluid catalyzed cracking
(FCC) and the steam cracking (SC) – along with other four-carbon
compounds, resulting in what is known as the C4 cut. A typical FCC
C4 cut composition includes isobutene, isobutane, butane, linear
butenes and others. With respect to ethanol, currently trend is to
produce it from renewable sources, as from biomass by hydrolysis
and sugar fermentation. In such process, an important amount of
water within the alcoholic stream can be present [11].
Main side reactions involve (Fig. 1): dimerization of isobutene
to form 2,4,4 trimethyl-1-pentene (TMP-1) and 2,4,4 trimethyl-2-
pentene (TMP-2), and thereof double bond isomerization (reactions
(2)–(4) in Fig. 1), ethanol dehydration to form diethyl ether (DEE)
and water (reaction (5) in Fig. 1), isobutene hydration to form of
tert-butyl alcohol (TBA) (reaction (6) in Fig. 1), and formation of
ethyl sec-butyl ether (ESBE) by reaction between ethanol and linear
butenes of C4 olefinic cuts (reaction (7) in Fig. 1).
In general, byproducts are inconvenient because some of them
attenuate the desired properties towards the resulting gasoline
mixture (Table 1). The presence of TMP-1 and TMP-2 along with
the ETBE stream product does not pose any problems for blending,
since both compounds have a high octane number. However, on
∗
Corresponding author. Tel.: +34 93 402 1304; fax: +34 93 402 1291.
E-mail address: fcunill@ub.edu (F. Cunill).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.09.012

J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
385
could cause some problems in the separation units, and in the later
use of the unreacted C4 stream.
Nomenclature
As literature on ETBE byproducts is very scarce, the aim of the
present work is to determine the conditions of formation of DEE,
isobutene dimers, TMP-1 and TMP-2, ESBE and TBA in the synthesis
of ETBE on macroreticular acid resins in order to reduce their for-
mation. Finally, as relatively fast equilibrium reactions take place
within the reaction network, their corresponding pseudo equilib-
rium conditions were studied.
A-35
Amberlyst® A-35WET is a macroreticular, strongly
acidic, cationic, polymeric catalyst by The Dow
Chemical Company
activity of chemical specie i
4-length hydrocarbons
ai
C4
CT-275 Purolite® CT275 is a macroreticular, strongly acidic,
cationic, polymeric catalyst by Purolite
DEE
dp
diethyl ether
2. Experimental setup and procedure
particle diameter
divynil-benzene
ethyl sec-butyl ether
ethyl tert-butyl ether
ethanol
DVB
ESBE
ETBE
EtOH
FCC
2.1. Chemicals
As reactants, ethanol (max. water content 0.02 wt%), supplied by
Panreac (Barcelona, Spain), and a synthetic C4 mixture, supplied by
Abelló-Linde (Barcelona, Spain), as the isobutene source, were used.
The composition of the C4 mixture was 25 wt% isobutene, 40 wt%
isobutane and 35 wt%. trans-2-butene.
TBA (99.7% G.C.) and DEE (99.5% G.C.), supplied by Panreac
(Barcelona, Spain), ETBE (>95.0% G.C.) and ESBE (>99.7% G.C.), sup-
plied by TCI (Tokio, Japan), and TMP-1 (>98% G.C.) and TMP-2 (>98%
G.C.), supplied by Sigma-Aldrich Química SA (Madrid, Spain), were
used for analysis purposes.
fluid catalyzed cracking
GC/MS gas chromatographer/mass spectrophotometer
IB
IPTBE
K
Kꢀ
2-methylpropene (isobutene)
isopropyl tert-butyl ether
thermodynamic equilibrium constant
ratio of activity coefficients in an equilibrium reac-
tion
Kx
experimental equilibrium constant based on molar
fractions
MON
MTBE
nk
motor octane number
methyl tert-butyl ether
mole of product or byproduct
feed molar ratio of alcohol to olefin
research octane number
Reid vapor pressure
steam cracking
selectivity of reactant j toward product k, expressed
as a percentage
temperature [K]
tert-amyl ethyl ether
2.2. Catalysts
As catalysts, two ion exchange resins, namely, AmberlystTM
35 (A-35) and Purolite® CT275 (CT-275), supplied by Rohm and
Haas France SAS (Chauny, France) and Purolite (Barcelona, Spain),
respectively, were used. Both resins are macroreticular, acidic,
oversulfonated polymers of styrene–divinylbenze (DVB) with a
permanent pore structure that makes them appropriate for a wide
range of organic reactions [21,22]. Both, A-35 and CT-275 are
low-swelling, strongly-acidic ion-exchange catalysts of industrial
interest for etherification reactions that can be actually considered
as two of the most important catalysts in use in this field. Table 2
shows the main properties of the two catalysts.
Before their use, catalyst samples were air-dried at room tem-
perature for 48 h, introduced in an atmospheric oven, at 383 K, for
2.5 h and, afterwards, placed in a vacuum oven, at 373 K, for 12 h.
Final water content in the resin beads after drying were 3–5 wt%
(analyzed by Karl-Fischer titration in the laboratory). Addition-
ally, air-dried catalyst beads were crushed and sieved in order to
obtain particle diameters, dp, ranging 0.25–0.40 mm, what ensures
no internal mass transfer influence on the overall reaction rate as
it was determined in previous works within this same group [7].
R◦
A/O
RON
RVP
SC
Sjk
T
TAEE
TAME
TBA
tert-amyl methyl ether
tert-butylic alcohol
TMP-1 2,4,4-trimethyl-1-pentene
TMP-2 2,4,4-trimethyl-2-pentene
Xj
xi
wcat
conversion of reactant j, expressed as a percentage
molar fraction of chemical species i
catalyst weight on dry basis
Greek letters
ꢀi
ꢁrHm
individual activity coefficient of chemical specie i
standard molar enthalpy change of reaction
(kJ mol−1
standard molar entropy change of reaction
(J mo1−l K−l
standard molar free energy change of reaction
(kJ mol−1
◦
)
◦
ꢁrSm
2.3. Analysis
)
◦
ꢁrGm
Samples inline from the reaction medium were taken through a
sampling valve (Valco A2CI4WE.2, VICI AG International, Schenkon,
Switzerland) that injected 0.2 ␮L of pressurized liquid into an Agi-
lent gas-liquid chromatograph 6890 (Madrid, Spain) equipped with
a HP capillary column (PONA Cross-linked Methyl Silicone Gum
of 50 m × 0.2 mm × 0.5 ␮m, ON, Canada). A mass selective detec-
tor HP5973N was used to identify and quantify the reaction system
components. The oven temperature was set at 308 K during 45 min.
Helium (Abelló-Linde, Barcelona, Spain), with a minimum purity of
99.998%, was used as the carrier gas with a flow of 1.5 mL/min.
Up to 11 different chemical species, namely, ethanol, isobutene,
ETBE, isobutane, trans- and cis-2-butene, TBA, DEE, ESBE, TMP-
1 and TMP-2, were chromatographically identified in significant
amounts depending on the experimental conditions. Besides, other
minor byproducts, formed by either the hydration or the oligo-
merization of 2-butenes, were also detected in the experiments
)
the catalyst they can originate polymers that could act as poison
by accumulating in the pores and lowering the accessibility to
acid centers [12]. DEE formation enhances higher blending vapor
pressure that might increase undesired evaporative emissions, as
its vapor pressure is higher than for the rest of possible products
[13,14]. Regarding TBA, although its octane number is high, TBA
presence in the gasoline blending increases its water solubility,
its blending Reid vapor pressure (RVP) [15–18], and its oxygen
content [19]. Finally, as ESBE has a more linear structure than
ETBE, it is assumed that its octane number is lower than that of
ETBE. Thus, the effect of the ESBE presence would be to reduce the
octane number of the gasoline mixture [20]. Besides, byproducts

386
J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
Fig. 1. Reaction network in the ETBE synthesis using a C4 stream as isobutene source.
Table 1
Relevant properties of involved compounds [47–49].
Category
Compound
Blending RVP [bar]
(RON + MON)/2
Boiling point [K]
Oxygen content [% wt.]
Solubility in water [g/100 gwater]
Main product
Byproduct
ETBE
TBA
TMP-1 and TMP-2
DEE
ESBE
EtOH
IB
0.34
0.70
0.11*
0.60*
0.10*
1.25
2.6*
110
101
∼100
Low
Low
113
345
356
374.5
308
355
15.7
21.6
0
21.6
15.7
34.7
0
1.2
Infinite
0
Partially soluble in cold water
∼1
Reactant
351
266.2
Infinite
3.88 × 10-4
Low
*
As blending Reid vapor Pressure (RVP) was not available, vapor pressure at 293 K is given instead.
carried out in favorable conditions (excess of olefins and highest
temperature, 383 K). In particular, 2-butanol, from the hydration of
2-butene, was detected in less than 1.0% of GC area and dimers of 2-
butene and/or codimers of 2-butene and isobutene were detected
in less than 1.5% of GC area. These minor byproducts were never
detected at other working conditions, and they were therefore not
taken into further consideration in the calculations.
Fig. 2 is provided as an example chromatogram of an exper-
iment carried out with A-35 as the catalyst, at 383 K, R◦A/O = 2.0,
after 300 min of reaction.
was kept at 2.0 MPa to ensure the liquid phase over the reaction sys-
tem. A detailed description of the experimental setup is described
elsewhere [23].
A set of 28 experiments (Table 3) was performed at four different
temperatures (323, 343, 363 and 383 K) with 3 different initial alco-
hol/olefin molar ratio (R◦A/O = 0.5, 1.0 or 2.0) in order to explore the
effect of different conditions on the side reactions that take place
during the ETBE synthesis reaction in industrial conditions.
For reproducibility evaluation purposes, experiments at 383 and
343 K and at R◦
equal to 0.5 and 2.0, as well as at intermedi-
A/O
ate conditions, were repeated at least three times. Experiments at
the lowest temperature (323 K) were not repeated as they revealed
scarce byproducts formation. The relative standard error for reac-
tants conversion ranged 0.3–4.4% and the mean relative error of the
mass balances was less than 3%.
2.4. Apparatus and procedure
The experimental setup consisted in a 200 cm3 stainless-
steel jacketed batch reactor equipped with a six-blade magnetic
stirrer (Autoclave Engineers, Pennsylvania, US). The reaction tem-
perature range was 323–383 K, controlled within 0.1 K by a
1,2-propanediol–water thermostatic mixture. The system pressure
Depending on the assayed R◦A/O, a calculated amount of ethanol
and 10.08 0.02 g of dried catalyst (wcat) were introduced in the
reactor before the heating and the stirring were switched on.
Table 2
Catalysts properties.
b
c
d
Catalyst
Crosslinkinga
Sulfonation
Acidity (meq H+/g)
Sg (m2 g−1
)
Vg (cm3 g-1
)
dpore (nm)
A-35
CT-275
High
High
Oversulfonated
Oversulfonated
5.32
5.20
34.0
21.8
0.210
0.177
23.6
32.9
a
High crosslinking degree: 14–25% DVB.
b
c
Specific surface area by BET method (N2 for Sg ≥ 1 m2 g-1).
Pore volume estimated from adsorption/desorption of N2 at 77 K.
Mean pore diameter.
d

J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
387
Fig. 2. Example chromatogram at texp = 300 min. T = 383 K, R◦A/O = 2.0, A-35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm.
Catalyst amounts correspond to a 10 wt% of the whole reacting
2.5. Calculations
mixture. Previous experience revealed no effect of the catalyst load
below 11.3 wt% should be expected [24]. Stirring speed was set at
500 rpm to ensure no external mass transfer influence on the over-
all reaction rate. In previous works carried out in the same exper-
imental apparatus, at 363 K and in the range 400–600 rpm, it was
determined no influence of that physical step for the ETBE synthe-
sis reaction [25]. Additionally, the same conclusion was achieved
by other studies at 500–800 rpm stirrer speed [26,27]. Temperature
ranged from 323 to 383 K, because it was considered representative
of the reported temperature profiles in industrial-scale reactors,
including possible hot spots. The amount of C4 mixture, determined
by R◦A/O, was introduced into the reactor from a pressure burette
once the desired temperature inside the vessel was reached. This
instant was considered as the starting point for the reaction.
Results were expressed in terms of conversion (Xj) of the main
reactants, ethanol and isobutene, and of selectivity (Sjk) of products
and byproducts obtained from the two main reactants, as follows:
mole of j reacted
mole of j fed
Xj =
(1)
mole of k produced
=
mole of j reacted
Sjk
(2)
where k is the considered product or byproduct, and j is the reactant
(isobutene or ethanol).
3. Results and discussion
Table 3
List of experiments. Conversion values at t = 300 min, catalyst load = 10 wt%.,
dp = 0.25–0.40 mm, 500 rpm.
Fig. 3 plots molar fraction, xi, evolution of the chemical species
involved in a model experiment, carried out with A-35, at 383 K
and R◦A/O = 0.5. Under such conditions, side reactions were strongly
favored. In Fig. 3a, molar fraction of the main reactants, ethanol
and isobutene, as well as other C4 mixture components (isobu-
tane and linear butenes) were represented. As it can be observed
in Fig. 3b, where ETBE and byproducts molar fraction evolution is
represented, byproducts appeared as soon as the reaction began.
TMP-1 and TMP-2 formed quickly, and after 1 h their molar fraction
increased in an almost linear trend until the end of the experiment.
TBA formation was very low; it achieved a very smooth maximum
concentration level within the first 60 min and then decreased to
almost null values. It can be concluded that, initially, isobutene
reacted swiftly with the available water in the reactive medium
to approach equilibrium (see reaction (6) in Fig. 1) and, then, the
equilibrium position shifted resulting in the observed decrease in
TBA concentration, because of a higher consumption of isobutene in
other simultaneous reactions (especially those of isobutene dimer-
ization, reactions (2) and (3) in Fig. 1). Regarding DEE and ESBE
formation, they seem to describe a linear pattern what suggests a
continuous evolution towards equilibrium, both reactions being far
from that point when the experiment was over. The ETBE concen-
tration drop, reaching null values at the end of the experiment, can
be explained by the multiple demands of ethanol and isobutene in
the different side reactions in which they were also involved, that
led to the ETBE decomposition (reaction (1) in Fig. 1). Note that ETBE
concentration maximum level was reached within the first steps of
Catalyst
A-35
T [K]
R◦
XIB [%]
XEtOH [%]
A/O
323
0.5
1.0
2.0
0.5
88
90
97
94
82
79
78
82
92
91
91
83
84
76
75
74
86
90
89
89
90
74
73
72
72
78
75
85
93
90
47
88
75
87
83
82
46
44
45
74
70
73
72
71
64
92
92
92
94
71
53
52
53
75
50
87
343
1.0
2.0
363
383
0.5
1.0
2.0
0.5
1.0
2.0
CT-275
363
383
0.5
0.5
1.0

388
J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
Fig. 3. Molar fraction evolution of major (a) and minor (b) chemical species.
T = 383 K, R◦A/O = 0.5, A-35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error
bars are referred to the standard error of the mean values. xIB (ꢀ), xEtOH (ꢀ), xisobutane
Fig. 4. Isobutene (a) and ethanol (b) selectivity towards products vs. total conver-
sion. T = 383 K, R◦A/O = 0.5, A-35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm.
ETBE
TBA
Error bars are referred to the standard error of the mean values. SIB
(♦), SIB
(᭹), xtrans-2−butene (ꢁ), xcis-2-butene (ꢂ), xETBE (♦), xTBA (ꢃ), xDEE (ꢄ), xESBE (ꢅ), xTMP-1 (ꢁ),
TMP-1
TMP-2
ETBE
DEE
ESBE
(ꢃ), SIB
(ꢁ), SIB
(ꢆ), SEtOH
(ꢅ), SEtOH
(ꢄ), SEtOH
(ꢇ).
xTMP
- (ꢆ).
2
3.1. TMP-1 and TMP-2 formation
the experiment, what means that this reaction proceeded quickly,
due to the high temperature and the catalyst loaded (10 wt% of the
whole reacting mixture).
Isobutene dimerization (reactions (2) and (3) in Fig. 1) gives
TMP-1 and TMP-2. Further oligomerization could originate trimers
and tetramers. Additionally, when linear butenes are present, as it
is the case of feedstocks from FCC or SC units, linear butenes could
react with isobutene to form codimers giving a primary carbocation
which eventually rearranged, forming other trimethylpentenes.
Codimerization reactions, though, are known to be slower than
isobutene dimerization, according to previous studies on the
dimerization/etherification of FCC and SC C4 hydrocarbons in the
presence of methanol [28]. In the present study codimers, trimers
and heavier oligomers were not taken into consideration in the cal-
culations as they were only obtained at the highest temperature in
less than a 1.5% of GC area.
It has to be noted that dimers formation was the most relevant
side reaction in the production of ETBE. This is in good agreement
with the reported data in literature for similar systems, for instance
in MTBE, TAME, TAEE, and IPTBE productions [8,12,29].
Isobutene selectivity towards TMP-1 and TMP-2 was enhanced
by high temperatures and low R◦A/O. This is coherent because high
temperatures and isobutene concentration have been reported to
be enhancers for the isobutene dimerization reaction [30,31]. As
example, Fig. 5 plots the selectivity to TMP-1 against temperature at
Fig. 4 plots selectivity vs. conversion of the two main reactants
in this experiment. Note that, both in Fig. 4a and b, one can clearly
distinguish the points corresponding to the first sample analyzed
(XIB ≈ 55% and XEtOH ≈ 71%) from the rest. This experimental fact
can be explained if the side reactions taking place simultaneously
are considered. It seems obvious, by observation of ETBE molar
fraction evolution in Fig. 3b, that probably even before the first sam-
ple was analyzed ETBE synthesis reaction had progressed greatly.
Then, ETBE decomposition to give isobutene and ethanol took place
during the experiment. This is a quite acceptable hypothesis once
considered the high temperature and amount of catalyst in the
reacting media. In the subsequent analyses during the experiment,
reactants conversion increase is due to its progressively higher con-
sumption in side reactions.
It is worth mentioning that, as this model experiment revealed,
after enough time of reaction, ETBE concentration could reach
null values. As it was the aim of this paper to discuss on
side reactions taking place simultaneously with ETBE produc-
tion, reference time up to 300 minutes was set as the end of the
experiments.

J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
389
Fig. 5. Variation with temperature of the isobutene selectivity towards TMP-1 at
texp = 300 min. A-35, catalyst load = 10 wt%., dp = 0.25–0.40 mm, 500 rpm. Error bars
are referred to the standard error of the mean values. R◦A/O = 0.5 (♦), R◦A/O = 1.0 (ꢀ),
R◦A/O = 2.0 (ꢁ).
Fig. 6. TMP-1 and TMP-2 formation vs. time at different R◦A/O. nk is referred to
number of moles. T = 383 K, A-35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm.
Error bars are referred to the standard error of the mean values. Open symbols: TMP-
1 (R◦A/O = 0.5 (♦), R◦A/O = 1.0 (ꢀ), R◦A/O = 2.0 (ꢀ)). Solid symbols: TMP-2 (R◦A/O = 0.5 (ꢅ),
R◦A/O = 1.0 (ꢈ), R◦A/O = 2.0 (᭹)).
the end of the experiments (t = 300 min). In that figure, a noticeable
difference in isobutene selectivity to TMP-1 is observed between
evolution is presented- that TMP-1 was formed swifter than TMP-
2. The shape of both dimers production against time suggests
irreversible reactions. Reaction rate of the isobutene dimerization
experiments carried out at the lowest R◦
of 0.5 and that at 2.0.
A/O
In the former, values of selectivity towards TMP-1 range from less
than 30% at low temperatures (323 and 343 K) to more than 60%
at 383 K. In the latter, though, almost null values of selectivity are
observed except for the experiment carried out at the highest tem-
decreased as R◦
increased, achieving its maximum values at
A/O
383 K and at R◦A/O = 0.5.
perature. Experimental series with an R◦
of 1.0 constitutes an
A/O
3.2. DEE formation
intermediate situation between the others. Analogous trends were
observed for the selectivity of isobutene to TMP-2, reaching a value
of up to 40% at 383 K and R◦A/O = 0.5.
Regarding ethanol dehydration (reaction (5) in Fig. 1), Fig. 7
reveals that high temperatures greatly enhanced the selectivity
of ethanol towards diethyl ether, reaching levels of almost 50%.
At low temperatures (323 and 343 K), no difference in selectivity
towards DEE was observed at different R◦A/O. At high tempera-
tures (363 and 383 K), low alcohol concentration (R◦A/O = 0.5) led to
higher selectivity to DEE whereas no difference in selectivity was
observed between experiments carried out under stoichiometric
conditions (R◦A/O = 1.0) or when an excess of ethanol was fed to the
reactor (R◦A/O = 2.0). This fact might seem counterintuitive, because
an excess of ethanol could be expected to constitute an enhancer
The enhancement of dimers formation at low molar ratios sug-
gests that isobutene, in a relatively high concentration, is readily
adsorbed on non-dissociated sulfonic groups of the catalyst and
dimerization reaction follows. This is actually a heterogeneous cat-
alytic step, which is faster than an ionic one (specific acid catalysis),
with dissociated protons due to the presence of ethanol in the sur-
roundings of the sulfonic groups (general acid catalysis) [32–36].
With respect to the isomerization reaction between TMP-1 and
TMP-2 (reaction (4) in Fig. 1), the TMP-1 to TMP-2 molar ratio was
monitored throughout each experiment. In the initial stage of the
reaction, this ratio quickly achieved values that ranged from 1.5 to
2.6 and remained stable onwards. Similar ratios were also reported
in analogous systems involving multiple simultaneous reactions
[12]. According to literature, TMP-1 is thermodynamically more
favored than TMP-2, resulting in a molar ratio of about 4:1 at
equilibrium in the same temperature range (323–383 K) as well as
at lower temperatures [37,38] what suggests that, in the present
work, isomerization had not yet reach the chemical equilibrium at
the end of the experiments (300 min).
It is worth mentioning that isomerization of diisobutenes con-
stitute an exception to the general rule in alkene isomers stability.
The general rule states that the more stable alkene is the one in
which the double bond is located in intermediate positions of the
carbon chain, in order to produce the highest possible substitution
of the double bond. However, in isobutene dimers, the more sub-
stituted isomer, the trisubstituted TMP-2, is less stable than the
disubstituted TMP-1. This fact has been already explained in lit-
erature by the internal repulsion between branches in the TMP-2
molecule that arises from the tert-butyl group located in the cis
position to the methyl group [38].
Fig. 7. Variation with temperature of the ethanol selectivity towards DEE at
texp = 300 min. A-35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error bars
are referred to the standard error of the mean values. R◦A/O = 0.5 (♦), R◦A/O = 1.0 (ꢀ),
R◦A/O = 2.0 (ꢁ).
With regards of dimerization reaction rate at high tempera-
tures, it can be observed in Fig. 6, where number of moles, nk,

390
J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
Fig. 8. Ethanol conversion at texp = 300 min against temperature at different R◦
.
A/O
Fig. 10. Variation with temperature of the isobutene selectivity towards TBA at
texp = 300 min. A-35, catalyst load = 10 wt%., dp = 0.25–0.40 mm, 500 rpm. Error bars
are referred to the standard error of the mean values. R◦A/O = 0.5 (♦), R◦A/O = 1.0 (ꢀ),
R◦A/O = 2.0 (ꢁ).
A-35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error bars are referred to
the standard error of the mean values. R◦A/O = 0.5 (♦), R◦A/O = 1.0 (ꢀ), R◦A/O = 2.0 (ꢁ).
of the DEE formation. But as Fig. 8 illustrates, when an excess of
alcohol was fed, ethanol conversion was lower than operating at
factors, the enhancing effect might be more important than the
inhibiting one in the ethanol dehydration reaction at 383 K.
other R◦
.
A/O
In terms of DEE production at the end of every experiment,
temperature plays a much more important role than R◦
as a
A/O
3.3. TBA formation
DEE promoter. To illustrate it, consider the experiments at 363 K
and lower temperatures depicted in Fig. 9, none of which showed
significant differences in terms of DEE production with respect
to R◦A/O. Nevertheless, at 383 K, an increase in DEE formation is
observed as increasing R◦A/O. The increase of the reaction medium
polarity due to the excess of ethanol affects the catalytic behavior
of the catalyst in several ways [39,40] and can account for the
increase in the DEE production. Polar molecules, as ethanol, can
break the hydrogen network that binds the sulfonic groups and,
as a consequence, more inner active centers become accessible to
the reactants. When ethanol dehydration takes place, the water
released as the reaction proceeds would even enhance this effect.
On the contrary, a polar medium solvates the acidic protons of those
active centers which present a lower acid strength than the protons
bound to the sulfonic groups, thus reducing the reaction rate on
each catalytic center [41]. Taking into account these opposed
Reaction between water and isobutene led to TBA formation
(reaction (6) in Fig. 1). For the present work, water could stem
from different sources: the amount of water contained within
the ethanol (200 ppm), the remaining moisture hold in the cat-
alyst (3–5 wt), and the chemically-produced molecules of water
proceeding from the dehydration of ethanol.
Fig. 10 reveals that, at R◦
other than 0.5, isobutene selec-
A/O
tivity towards TBA at the end of the experiments increased with
temperature. At low temperatures (363 K and lower), no signif-
icant differences in the selectivity to TBA can be distinguished
between experiments at R◦A/O = 1.0 and 2.0, whereas for the experi-
ments at R◦A/O = 0.5 the selectivity was slightly lower. At the highest
temperature (383 K), the higher the molar ratio, the higher the
selectivity towards TBA was. These facts suggest that more polar
media enhanced selectivity to TBA.
Fig. 11 plots the evolution of the TBA production at R◦A/O = 2.0,
that is when the highest amounts of TBA were detected, and at dif-
ferent temperatures. While at lowest temperatures, namely 323 K
and 343 K, the TBA content decreased after a maximum value, at
highest temperatures (363 K and 383 K) no maximum value was
yet achieved by the time the experiment ended. The former behav-
ior can be explained by assuming that, as TBA is known to reach
the equilibrium swiftly [42], water present in the medium reacted
readily with isobutene at the initial steps of the experiment and,
afterwards, the reverse reaction proceeded since isobutene was
consumed by other competitive side reactions, mainly dimeriza-
tion. The later behavior can be related to the major extent of the
dehydration reaction of ethanol at high temperatures, what pro-
duced water that subsequently reacted with isobutene to form
more TBA.
As reported in literature, water shows a strong inhibitor effect
on the ETBE production, even at low contents [42]. This inhibitor
effect would disappear progressively as water was consumed to
yield TBA, which was produced quickly. However, the simultaneous
dehydration of ethanol enhanced by high temperatures, could lead
to higher water production that would account for a sustained
consumption of isobutene. This feasible explanation would imply
Fig. 9. DEE formation against R◦
35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error bars are referred to the
at different temperatures at texp = 300 min. A-
A/O
standard error of the mean values. T = 323 K (♦), T = 343 K (ꢀ), T = 363 K (ꢁ), T = 383 K
(ꢀ).

J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
391
Fig. 12. Variation with temperature of the ethanol selectivity towards ESBE at
texp = 300 min. A-35, catalyst load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error bars
are referred to the standard error of the mean values. R◦A/O = 0.5 (♦), R◦A/O = 1.0 (ꢀ),
R◦A/O = 2.0 (ꢁ).
Fig. 11. TBA formation in time at different temperatures. R◦A/O = 2.0, A-35, catalyst
load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error bars are referred to the standard
error of the mean values. T = 323 K (♦), T = 343 K (ꢀ), T = 363 K (ꢁ), T = 383 K (ꢀ).
that, depending on the experimental conditions, some water can be
redelivered to the reactive media, what arises as a potentially mod-
ifier factor of the catalytic conditions throughout an experiment.
The high polarity of water leads to a combined effect of, poten-
tially, reducing the activity of sulfonic groups, and simultaneously
swelling the resin, what allows reactants to access to active centers
that were initially unreachable [43].
representative experiments carried out with A-35 were repeated
with CT-275. These experiments were performed at high tem-
peratures (363 and 383 K), in order to favor side reactions, and in
excess of olefins and under stoichiometric conditions (R◦A/O = 0.5
and 1.0, respectively), in order not to mask a possible influence
of R◦
by choosing only one initial feed composition. In general,
A/O
significant differences between both catalysts were observed in
terms of ETBE production, total byproducts formation and product
distribution, depending on the assayed conditions.
3.4. ESBE formation
Fig. 14 plots the evolution of the total byproducts formation in
those experiments. Under stoichiometric conditions (R◦A/O = 1.0)
no significant differences between both catalysts were detected.
At 363 K and R◦A/O = 0.5, A-35 led to more byproducts formation
Etherification of linear butenes with ethanol occurs when linear
butenes are present in the C4 cut used as isobutene source, lead-
ing to ESBE formation (reactions (7) in Fig. 1). In the present work,
trans-2-butene was included in the synthetic C4 mixture in order
to emulate the presence of these linear butenes in typical sources
of isobutene, such as FCC or SC C4 cuts. It is worth noting that
isomerization of trans-2-butene present in the initial mixture to
cis-2-butene was detected. Amounts of cis-2-butene reached levels
up to 7% of GC area that were taken into account in the calculations.
According to literature, trans-2-butene is thermodynamically more
stable than cis-2-butene [44,45].
As shown in Fig. 12, selectivity to ESBE was strongly favored by
high temperatures and low R◦A/O. At low temperatures (T = 343 K
and lower) no significant amounts of ESBE were detected, but at
363 K and higher, the selectivity to TBA values achieved by exper-
iments at R◦A/O = 0.5 were far higher than those achieved under
stoichiometric or in excess of alcohol conditions (R◦A/O = 1.0 or 2.0,
respectively).
than CT-275. At the same R◦
and 383 K, byproducts forma-
A/O
tion increased for both catalysts, as expected, but with a larger
production for CT-275. A similar fact has been reported in the oli-
gomerization reaction of 1-hexene for the same resins [46].
According to literature [28], A-35 is expected to be more
active than CT-275 due to its higher acid capacity, regardless the
difference might be scarce. Accordingly, the higher the activity
of a catalyst, the higher the total formation of products and/or
Fig. 13 reveals that ESBE production evolved linearly with time
under the most favorable condition (R◦A/O = 0.5), what suggests that
the reaction was far from the equilibrium within the assayed tem-
perature range.
Similar to the one described for the dimers formation at low
R◦A/O, a heterogeneous catalytic step can be assumed again. Linear
butenes would be able to take the proton directly from the sulfonic
groups, because they would be less dissociated due to the lack of
ethanol [32–36]. This assumption was already adopted in a simi-
lar study on the side reactions taking place along with the MTBE
synthesis [29].
3.5. Catalysts comparison
Fig. 13. ESBE formation vs. time at different temperatures. R◦A/O = 0.5, A-35, catalyst
load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error bars are referred to the standard
error of the mean values. T = 323 K (♦), T = 343 K (ꢀ), T = 363 K (ꢁ), T = 383 K (ꢀ).
In order to compare possible differences in terms of global
behavior of the reaction system due to the catalyst, three

392
J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
sufficiently high amount of dimers would block partially the pores
of A-35 and, therefore, hinder the species diffusion, especially for
larger molecules as TMP-1, TMP-2 and ESBE. Such phenomenon
would not occur, or it could occur in a much lower extension, with
CT-275 because it presents a larger pore size (Table 2).
With respect to isobutene selectivity towards products (Fig. 15),
strong differences were detected between both resins at 363 K
and R◦A/O = 0.5, where A-35 clearly favored isobutene dimerization
compared to CT-275. This difference was reduced by increasing
the temperature to 383 K, at the same R◦A/O = 0.5, where both
catalysts presented similar isobutene selectivity towards each
byproduct. Consequently, a higher temperature reduces the effects
of the structural catalyst properties on selectivity. Finally, at 383 K
and R◦A/O = 1.0, isobutene selectivity towards dimers was more
enhanced by A-35 than by CT-275. Selectivity towards TBA was
higher for CT-275 than for A-35, what can be explained by the
lesser consumption of isobutene due to the competitive irreversible
reaction of dimerization. Evolution of ethanol selectivity towards
products showed similar trends to those described here.
In order to give an explanation to the results plotted in Fig. 15,
it has to be considered that at high temperatures, namely 363
and 383 K, the fastest reversible reactions, ETBE and TBA, could
reach positions close to chemical equilibrium conditions rapidly,
irrespectively of the used catalyst. Then, as slower side reactions
proceeded, the reversible reactions could move to the left side,
leaving free isobutene that can dimerize. Therefore, Fig. 15 shows
an increasing selectivity towards TMP-1 and TMP-2 through the
experiments.
Fig. 14. Total byproducts formation vs. time with two different catalysts. Catalyst
load = 10 wt%, dp = 0.25–0.40 mm, 500 rpm. Error bars are referred to the standard
error of the mean values. Open symbols: A-35 (R◦A/O = 0.5, T = 363 K (ꢇ); R◦A/O = 0.5,
T = 383 K (ꢁ); R◦A/O = 1.0, T = 383 K (ꢀ)). Solid symbols: CT-275 (R◦A/O = 0.5, T = 363 K
(ꢉ); R◦A/O = 0.5, T = 383 K (ꢆ); R◦A/O = 1.0, T = 383 K (ꢈ)).
byproducts should be expected. Nonetheless, in this particular
case, this premise was not accomplished in the experiments at the
highest temperature.
After analyzing each byproduct individually, it was determined
that the shift of the most active catalyst in the byproducts formation
from A-35 to CT-275 as temperature was raised from 363 to 383 K
was mainly due to an increase in the isobutene dimers produc-
tion as well as, though in minor extension, in the ESBE production.
This shift could be explained by morphological differences between
both catalysts. A-35 would experience internal diffusion problems
due to its small pore diameter that would become apparent at
higher temperature that is when dimerization is more favored. A
From the catalysts properties standpoint, it is necessary to bear
in mind that the most active catalyst for the main reaction is also
the most active one for side reactions. So, in general, A-35 was pro-
gressively more selective than CT-275 towards larger molecules, as
TMP-1 and TMP-2. Additional explanation for such phenomenon
could be the higher specific surface area and pore volume of A-
35 that could lead to a more suitable spatial distribution of acid
centers to form isobutene dimers. To support this hypothesis,
Fig. 15. Product distribution evolution in terms of isobutene selectivity to TBA (ꢈ), ETBE ( ), TMP-1 (ꢀ), TMP-2 (ꢀ) with A-35 and CT-275. Catalyst load = 10 wt%,
dp = 0.25–0.40 mm, 500 rpm.

J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
393
literature data showed that, operating under the same tempera-
ture range (343–383 K), in the isobutene dimerization reaction in
the presence of tert-butyl alcohol, the more selective ion-exchange
catalyst assayed for the isobutene dimerization was the one with
the highest specific surface area [47]. However, more than two cat-
alysts should be assayed in order to establish a proper correlation
between the chemical reaction and the catalyst properties.
With regards to the experiments at 383 K and R◦A/O = 0.5, also
in Fig. 15, the almost identical product distribution evolution
experienced by both catalysts evidences that given such favor-
ing conditions -in terms of dimerization reaction- morphological
differences between the catalysts were not significant enough
so as to lead to different behavior. Perhaps, in excess of olefin,
the resins structures could be partially collapsed and the advan-
tages of A-35 towards dimerization were not showed because
the accessibility to its acid centers decreased more than for
CT-275.
Considering the three experiments, it has to be pointed
out that CT-275 led to higher productions of ETBE than A-35,
within the experimental conditions assayed. At 383 K, though,
CT-275 also produced more byproducts than A-35. Notwith-
standing, to quantify the effects of temperature and initial
molar ratio on the two catalysts more experimental work is
needed.
Fig. 16. ETBE equilibrium constant values from literature data and from the present
work. A-35, catalyst load = 10% wt., dp = 0.25–0.40 mm, 500 rpm. Error bars are
referred to the standard error of the mean values. The line corresponds to the rela-
tion between ln K and 1/T achieved experimentally in the present work. Franc¸ oisse
and Thyrion [6] (ꢅ), Vila et al. [50] (ꢆ), Cunill et al. [42] (ꢈ), Izquierdo et al. [49] (♦),
Jensen and Datta [51] (ꢀ), Gómez et al. [53] (ꢉ), Present work (᭹).
3.6. Equilibrium data
with published literature data on ETBE equilibrium [6,42,49–53].
This suggests that, despite side reactions taking place simulta-
neously, ETBE synthesis reaction was fast enough to reach a pseudo
equilibrium state. The dependence of K was estimated by integra-
tion of the van’t Hoff equation, and the reaction enthalpy change
was considered a function of temperature (Kirchoff equation).
At 298.15 K we obtain the following values:
In the previous discussion it has been affirmed that some equi-
librium reactions could be relatively fast enough to assume that
they were in a pseudo equilibrium state in such a way that the
equilibrium point shifted accordingly to the changes in composi-
tion of the species involved in other reactions. In order to check
if this hypothesis was realistic, activities relation corresponding to
each equilibrium reaction were calculated for each experimental
point. The following relation of activities for the ETBE synthesis is
shown as example.
◦
ꢁrHm = −38.5
6.0[kJ mol−1
17.3[J mo1−l K−l
6.1[kJ mol−1
]
◦
ꢁrSm = −89.3
]
ꢀ
ꢁ
ꢀ
ꢁ ꢀ
ꢁ
aETBE
xETBE
ꢀETBE
K =
=
·
= Kx · Kꢀ
(3)
aIB · aEtOH
xIB · xEtOH
ꢀIB · ꢀEtOH
◦
ꢁrGm = −11.8
]
The individual activity coefficients of all the species involved in
the equilibrium reactions considered were calculated by the mod-
ified UNIFAC (Dortmund) method [48].
Deviations of the calculated equilibrium constant due to pres-
sure effects can be considered as negligible, as it was determined by
calculation of the Poynting factor [54]. In the whole experimental
conditions range, the Poynting factor was always close to the unity
(always higher than 0.9998).
Equilibrium reaction syntheses of ETBE and TBA, as well as the
isomerization reaction of isobutene dimers, can be considered to
be in pseudo equilibrium situations in a number of experiments, as
it can be inferred by the evolution of the implicated species relation
of activities.
Experimental values of the ETBE pseudo equilibrium constant
fit well, within the margin of experimental error, with previously
reported expressions for the temperature dependence of the chem-
ical equilibrium constant such as Eqs. (4) and (5) [42,49,50]:
With regards of TBA formation, experimental pseudo equi-
librium constant values were far lower than the ones reported
as chemical equilibrium constants [55]. It has to be highlighted
that actual amounts of water in the reacting system were esti-
mated by means of the water initially fed to the reactor (the
sources of which were the remaining moisture of catalysts and
the water content in ethanol) and the extension of the reactions
in which water was involved (namely, dehydration of ethanol
and isobutene hydration). So, in relative terms, the indetermina-
tion associated to the water quantification was severe. Therefore,
the commented deviation in the calculated pseudo equilibrium
constants can be explained by its high sensitivity towards the water
quantification indetermination because, accordingly with the low
quantity of water in the reacting system throughout the experi-
ments, very high activity coefficients of water were achieved that,
subsequently account for this low values of TBA pseudo equilibrium
constant.
ln(KETBE) = 1140.876 − 14583.2 T−1 − 232.873 ln(T)
+ 1.086489 T − 1.11385 × 10−3 T2 + 5.53858
× 10−7 T3
(4)
(5)
ꢂ
ꢃ
4226.21
KETBE = 7.4 × 10−5 exp
T
Experimental values of K (Fig. 16), as well as estimates of
standard enthalpy, entropy and free energy change of the synthesis
of ETBE were calculated for comparative purposes. It is worth not-
ing that, except for the experiments at 383 K and R◦A/O = 0.5 (that
is, in the conditions at which isobutene dimerization was more
favored), the calculated values showed a rather good agreement
As for DEE and ESBE formations, their mole production evolution
in time at all conditions (Fig. 3b and 13, for instance) revealed linear
patterns, what suggests far from equilibrium situations for both
reactions through all the experiments.

394
J.H. Badia et al. / Applied Catalysis A: General 468 (2013) 384–394
4. Conclusions
[8] R.A. Meyers (Ed.), Handbook of Petroleum Refining Processes, third ed.,
McGraw-Hill Books, New York, 2004.
[9] H. Järvelin, Handbook of MTBE and Other Gasoline Oxygenates, CRC Press, Boca
Raton, FL, 2004.
[10] A. Kolah, L. Rihko-Struckmann, K. Sundmacher, Handbook of MTBE and Other
Gasoline Oxygenates, CRC Press, Boca Raton, FL, 2004.
[11] F. Aiouache, S. Goto, Handbook of MTBE and Other Gasoline Oxygenates, CRC
Press, Boca Raton, FL, 2004.
[12] J. Tejero, A. Calderón, F. Cunill, J.F. Izquierdo, M. Iborra, React. Funct. Polym. 33
(1997) 201–209.
[13] K.F. Wong, C.A. Eckert, J. Chem. Eng. Data 14 (1969) 432–436.
[14] R.H. Perry, D.W. Green, J.O. Maloney, Perry’s Chemical Engineers’ Handbook,
McGraw-Hill, New York, 2008.
[15] API Bulletin 2500, American Petroleum Institute, January 1955.
[16] API Bulletin 2513, American Petroleum Institute, February 1959.
[17] API Bulletin 2516, American Petroleum Institute, February 1962.
[18] API Bulletin 2518, American Petroleum Institute, February 1962.
[19] F. Ancillotti, V. Fattore, Fuel Process. Technol. 57 (1998) 163–194.
[20] R. Meusinger, R. Moros, Fuel 80 (2001) 613–621.
[21] Rohm & Haas, A-35 Product Data Sheet October 2012, 2003.
[22] Purolite, CT-275 Data Sheet October 2012, 2012.
[23] V.J. Cruz, J.F. Izquierdo, F. Cunill, J. Tejero, M. Iborra, C. Fité, React. Funct. Polym.
65 (2005) 149–160.
Dimerization of isobutene, giving TMP-1 and TMP-2, was the
most relevant side reaction. Low molar ratios and high temper-
atures enhance this reaction. High temperatures enhanced DEE
formation. Excess of ethanol produced scarcely more DEE than low
molar ratios and that was related to slow kinetics of the ethanol
dehydration reaction. At the conditions assayed, TBA was produced
in very low quantities. Although at low temperatures, selectivity
towards TBA at the end of the experiment was almost zero, at high
temperatures, the higher the molar ratio, the higher the selectivity
towards TBA was. Possible explanation implies a combined effect of
the dehydration of ethanol, what produces water, and the readiness
with which isobutene reacts with this water. There was a strong
effect of the loaded water over this side reaction. ESBE formation
was detected only at high temperatures but the selectivity towards
it achieved values up to 50%. Low feed molar ratios also enhanced
this side reaction.
Finally, it can be stated that high temperatures strongly enhance
the formation of all the byproducts studied – kinetics of the side
reactions are strongly affected by temperature – whereas initial
molar ratios have a strong influence on the nature of side reactions
that are favored.
Concerning the catalysts comparison, CT-275 led to higher pro-
ductions of ETBE than A-35, within the experimental conditions
assayed. However, at 383 K, CT-275 also produced more byproducts
than A-35.
[24] V. Cruz, J. Izquierdo, F. Cunill, J. Tejero, M. Iborra, C. Fité, R. Bringué, React. Funct.
Polym. 67 (2007) 210–224.
[25] E. Gimenez Buendia, Final Project, University of Barcelona, 2001.
[26] M. Umar, A.R. Saleemi, S. Qaiser, Catal. Commun. 9 (2008) 721–727.
[27] M. Umar, A. Saleemi, B. Saha, Nucleus 46 (2009) 101–107.
[28] M. Di Girolamo, M. Lami, M. Marchionna, E. Pescarollo, L. Tagliabue, F. Ancillotti,
Ind. Eng. Chem. Res. 36 (1997) 4452–4458.
[29] M. Vila, F. Cunill, J.F. Izquierdo, J. González, A. Hernández, Appl. Catal., A: Gen.
117 (1994) L99–L108.
[30] M. Honkela, A. Krause, Catal. Lett 87 (2003) 113–119.
[31] J.F. Izquierdo, M. Vila, J. Tejero, F. Cunill, M. Iborra, Appl. Catal., A: Gen. 106
(1993) 155–165.
[32] F. Ancillotti, M.M. Mauri, E. Pescarollo, J. Catal. 46 (1977) 49–57.
[33] F. Ancillotti, M. Massi Mauri, E. Pescarollo, L. Romagnoni, J. Mol. Catal. 4 (1978)
37–48.
[34] J. Tejero, F. Cunill, J.F. Izquierdo, M. Iborra, C. Fité, D. Parra, Appl. Catal., A: Gen.
134 (1996) 21–36.
[35] B.C. Gates, W. Rodriguez, J. Catal. 31 (1973) 27–31.
[36] R. Thornton, B. Gates, J. Catal. 34 (1974) 275–287.
[37] W.S. Gallaway, M.J. Murray, J. Am. Chem. Soc. 70 (1948) 2584–2586.
[38] R.S. Karinen, M.S. Lylykangas, A.O. Krause, Ind. Eng. Chem. Res. 40 (2001)
1011–1015.
[39] H. Panneman, A.A.C.M. Beenackers, Ind. Eng. Chem. Res. 34 (1995) 3817–3825.
[40] H. Panneman, A.A.C.M. Beenackers, Ind. Eng. Chem. Res. 34 (1995) 4318–4325.
[41] C. Fité, J. Tejero, M. Iborra, F. Cunill, J.F. Izquierdo, D. Parra, Appl. Catal., A: Gen.
169 (1998) 165–177.
Minimum ETBE production was detected in experiments at high
temperatures and low molar ratios. Selectivity of the two reactants
towards ETBE was slightly higher at R◦A/O = 2.0 than at R◦A/O = 1.0.
However, at R◦A/O = 2.0, part of the ethanol did not react. Then, in
order to minimize side reactions along with the ETBE reaction syn-
thesis, it is advisable to control the cooling system -so hot-spots
should be minimized-, to work with molar ratios slightly higher
than the stoichiometric one and to keep the feedstock as free of
water as possible.
Acknowledgments
[42] F. Cunill, M. Vila, J.F. Izquierdo, M. Iborra, J. Tejero, Ind. Eng. Chem. Res. 32 (1993)
564–569.
[43] C. Fité, M. Iborra, J. Tejero, J.F. Izquierdo, F. Cunill, Ind. Eng. Chem. Res. 33 (1994)
The authors thank Rohm and Haas France SAS (The Dow Chem-
ical Company) and Purolite for providing Amberlyst and CT ion
exchange resins, respectively. We are also indebted to Mrs. Vanessa
Arellano and Mrs. Laura Cazacu for their indispensable participa-
tion in the experiments that made this work possible.
581–591.
[44] A. Moronta, J. Luengo, Y. Ramírez, J. Quin˜ónez, E. González, J. Sánchez, Appl.
Clay. Sci. 29 (2005) 117–123.
[45] O.S. Pavlov, S.A. Karsakov, S.Y. Pavlov, Russ. J. Appl. Chem. 82 (2009)
1117–1122.
[46] M. Cadenas, R. Bringué, C. Fité, E. Ramírez, F. Cunill, Top. Catal. 54 (2011)
998–1008.
[47] M.L. Honkela, A. Root, M. Lindblad, A.O.I. Krause, Appl. Catal., A: Gen. 295 (2005)
216–223.
References
[48] J. Gmehling, J. Lohmann, A. Jakob, J. Li, R. Joh, Ind. Eng. Chem. Res. 37 (1998)
4876–4882.
[49] J.F. Izquierdo, F. Cunill, M. Vila, M. Iborra, J. Tejero, Ind. Eng. Chem. Res. 33 (1994)
2830–2835.
[50] M. Vila, F. Cunill, J. Izquierdo, J. Tejero, M. Iborra, Chem. Eng. Commun. 124
(1993) 223–232.
[51] K.L. Jensen, R. Datta, Ind. Eng. Chem. Res. 34 (1995) 392–399.
[52] K.G. Sharonov, A.M. Rozhnov, A.V. Korol’kov, S.Y. Karaseva, E.A. Mirosh-
nichenko, L.I. Korchatova, J. Chem. Thermodyn. 27 (1995) 751–753.
[53] C. Gómez, F. Cunill, M. Iborra, F. Izquierdo, J. Tejero, Ind. Eng. Chem. Res. 36
(1997) 4756–4762.
[54] R.C. Reid, J.M. Prausnitz, B.E. Poling (Eds.), The Properties of Gases and Liquids,
fourth ed., McGraw-Hill, New York, 1987.
[55] M. Iborra, J. Tejero, M. El-Fassi, F. Cunill, J.F. Izquierdo, C. Fité, Ind. Eng. Chem.
Res. 41 (2002) 5359–5365.
[1] DIRECTIVE 2009/28/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUN-
CIL of 23 April 2009 on the promotion of the use of energy from renewable
sources and amending and subsequently repealing Directives 2001/77/EC and
2003/30/EC.
[2] DIRECTIVE 2009/30/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUN-
CIL of 23 April 2009 amending Directive 98/70/EC as regards the specification
of petrol, diesel and gas-oil and introducing a mechanism to monitor and
reduce greenhouse gas emissions and amending Council Directive 1999/32/EC
as regards the specification of fuel used by inland waterway vessels and repeal-
ing Directive 93/12/EEC.
[3] A. Chakrabarti, M. Sharma, React. Polym. 20 (1993) 1–45.
[4] M. Sharma, React. Funct. Polym. 26 (1995) 3–23.
[5] M.A. Harmer, Q. Sun, Appl. Catal., A: Gen. 221 (2001) 45–62.
[6] O. Franc¸ oisse, F.C. Thyrion, Chem. Eng. Process. 30 (1991) 141–149.
[7] R. González, Doctoral Thesis, University of Barcelona, 2011.