Surface-treated activated carbons as catalysts for the dehydration and dehydrogenation reactions of ethanol

Article abstract of DOI:10.1021/jp981861l

Two activated carbons were prepared from olive stones with different degrees of activation and oxidized with a (NH₄)₂S₂O₈ solution for variable periods of time to introduce different amounts of oxygen surface complexes. Samples so prepared were characterized to know their surface area, porosity, and surface chemistry and then were used as catalysts in the conversion reactions of ethanol. The dehydration reactions to obtain ethene and ether only occurred on the oxidized samples, where carboxyl groups placed on the external surface of the particles were responsible for these reactions. The dehydrogenation reaction to yield acetaldehyde took place on either acid or basic surface sites placed not only on the external surface, but also on some part of the internal surface. During the reaction some surface active sites for dehydrogenation were lost because some hydrogen remained bound to them. The presence of air in the reactant mixture cleaned some of these sites, increasing the dehydrogenation activity and decreasing the dehydration activity. Also studied was the effect on their activity of the pretreatments of the catalysts in He or H₂ flow at different temperatures up to 1273 K.

Full text of DOI:10.1021/jp981861l

J. Phys. Chem. B 1998, 102, 9239-9244
9239
Surface-Treated Activated Carbons as Catalysts for the Dehydration and Dehydrogenation
Reactions of Ethanol
F. Carrasco-Mar´ın, A. Mueden, and C. Moreno-Castilla*
Departamento de Qu´ımica Inorga´nica, Facultad de Ciencias, UniVersidad de Granada, 18071 Granada, Spain
ReceiVed: April 15, 1998; In Final Form: June 30, 1998
Two activated carbons were prepared from olive stones with different degrees of activation and oxidized
with a (NH4)2S2O8 solution for variable periods of time to introduce different amounts of oxygen surface
complexes. Samples so prepared were characterized to know their surface area, porosity, and surface chemistry
and then were used as catalysts in the conversion reactions of ethanol. The dehydration reactions to obtain
ethene and ether only occurred on the oxidized samples, where carboxyl groups placed on the external surface
of the particles were responsible for these reactions. The dehydrogenation reaction to yield acetaldehyde
took place on either acid or basic surface sites placed not only on the external surface, but also on some part
of the internal surface. During the reaction some surface active sites for dehydrogenation were lost because
some hydrogen remained bound to them. The presence of air in the reactant mixture cleaned some of these
sites, increasing the dehydrogenation activity and decreasing the dehydration activity. Also studied was the
effect on their activity of the pretreatments of the catalysts in He or H2 flow at different temperatures up to
1273 K.
Introduction
tones, are preferentially formed on basic sites, whereas the
dehydration products, olefins and ethers, are favored on acid
sites. However, these reactions catalyzed by carbon materials
have not been studied as extensively as with other materials.
Thus, Szymanski et al.24-26 studied the conversion reactions of
different alcohols and suggested that the dehydrogenation
reaction takes place simultaneously on Lewis basic and acid
sites located at the external surface of the particles. The
dehydration reaction takes place, according to the same authors,
on acid sites essentially located at the external surface of the
particles.
Grunewald and Drago27 also studied the conversion reactions
of ethanol catalyzed by carbon. When ethanol, diluted in N2,
passed over the catalyst at 503 K, the main product obtained
was ethene, but the activity of the catalyst fell to zero in a few
hours. However, if the carrier gas was switched to air, the
catalyst regained activity immediately, and in these conditions
acetaldehyde and ethyl acetate were the major products. These
authors indicated that the surface active sites were reduced by
ethanol and regenerated by air. The mechanism invoked was
acid catalysis. Thus, a Lewis site subtracted a hydride and a
Bro¨nsted site protonated the alcohol. The subsequent loss of a
proton from the former species yielded an aldehyde and loss of
water, and a proton from the latter species yielded ethene.
Therefore, it seems that certain contradictions exist between
the mechanism suggested by Szymanski et al.24-26 for the
dehydrogenation reaction and that of Grunewald and Drago.27
Thus, for Szymanski et al. the reaction involved the simultaneous
action of basic and acid Lewis sites, whereas for Grunewald
and Drago it involved Lewis acid sites, which were completely
deactivated with the reaction time when the carrier gas was inert.
Therefore, the aim of the present work was to gain more
insight into the conversion reactions of ethanol on activated
carbons with different surface properties. For this purpose, two
activated carbons with different surface area and porosity were
chosen. These were oxidized with (NH4)2S2O8 solutions for
The use of carbon materials as catalysts and as catalyst
supports has been reviewed recently.1 Applications of these
solids in catalysis are due to the versatility of their surface
properties, such as surface area, porosity, and surface chemistry.
Of these three properties, the surface chemistry is probably the
most important in the catalytic processes taking place on the
surface of these solids.
Surface chemistry of carbon materials is basically determined
by the acidic and basic character of the carbon surfaces. The
acidic behavior generally is, associated with oxygen surface
complexes or oxygen functionalities such as carboxyls, lactones,
and phenols. Other oxygen functionalities such as pyrone-,
chromene-, ether-, and carbonyl-type structures give basic
properties to the carbon surface.2-9 All these oxygen surface
complexes can be introduced on carbons by treatments with
different oxidants either in liquid or gas phase. However,
although some functionalities are able to act as basic sites, the
basic properties of the carbon surfaces are not yet well
understood, because some authors3,10-13 have identified Lewis
basic sites located at the π-electron-rich regions within the basal
planes of graphitic microcrystals, away from the edges. Thus,
the number of these basic sites would decrease with the fixation
of oxygen surface complexes, essentially on the edges of the
microcrystals, because these would decrease the delocalized
π-electrons. Support for the existence of these Lewis basic sites
comes from results that show a decrease in acid adsorption for
a small increase in the oxygen content of the carbons3 and more
recently from measurements of heats of neutralization.14,15
Dehydration and dehydrogenation of alcohols are among the
reactions most commonly used to test for acid-base catalysts,
and these reactions have been studied extensively in inorganic
oxides.16-23 The dehydrogenation products, aldehyde or ke-
* Corresponding author. Fax: 34-958-248526. E-mail: CMORENO@
GOLIAT.UGR.ES.
10.1021/jp981861l CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/27/1998

9240 J. Phys. Chem. B, Vol. 102, No. 46, 1998
Carrasco-Mar´ın et al.
different periods of time to introduce variable amounts of oxygen
surface complexes. It has been shown elsewhere,28,29 that this
oxidant treatment introduces strong acid groups and does not
significantly modify the surface area and porosity of the original
carbon. Thus, the relationship between the oxygen surface
complexes and the activity and selectivity of the catalysts was
studied.
Experimental Section
The preparation and characteristics of the activated carbons
used in this work were reported elsewhere.29 The raw material,
olive stones, was carbonized in N2 flow at 1273 K and steam
activated at 1103 K to yield two activated carbons with 20%
and 46% burnoff, which will be referred to in the text as AZ20
and AZ46, respectively. These activated carbons were oxidized
with a saturated solution of (NH4)2S2O8 at 298 K for different
periods of time up to a maximum of 24 h.29 All oxidized
samples were washed with distilled water until there was an
absence of sulfates in the washing water. The oxidation time
is added to the name of the activated carbon, thus, AZ20-2
means activated carbon AZ20 oxidized for 2 h. The original
and oxidized activated carbons were characterized by N2 and
CO2 adsorption at 77 and 273 K, respectively. Mercury
porosimetry up to 2600 kg‚cm-2 and He and Hg densities were
determined. The surface chemistry of the samples was studied
by temperature-programmed desorption followed by mass
spectrometry (TPD-MS), Fourier transform infrared, titrations
with bases and HCl, and measurements of the pH of the point
of zero charge. All the results obtained with these techniques
have been widely discussed elsewhere.29
The decomposition reactions of ethanol were conducted in a
plug-flow microreactor working at atmospheric pressure and
using He or air as the carrier gas. The concentration of ethanol
in the reactant mixture was generally 0.32 vol %. Other ethanol
concentrations were also used. For this purpose, the carrier gas
was saturated with ethanol that was kept in a cold trap at
different temperatures.
The amount of catalyst used was approximately 0.5 g, which
was heat treated in He flow at 453 K for 2 h before studying its
activity. The effect of other heat treatments up to 1273 K in
He or H2 flow on the activity of the catalysts was also studied.
After the heat treatments the samples were cooled to the reaction
temperature (which ranged from 413 to 453 K) and the He or
H2 flow changed to the reactant mixture, 60 cm3‚min-1. This
mixture was allowed to flow through the catalyst for 30 min
before the analysis of the reaction products. After this, only
the carrier gas (He or air) flowed through the catalyst for 30
min, and at the same time a new reaction temperature was
stabilized; then the reactant mixture was again introduced into
the reactor to study the reaction at different temperatures.30 In
all cases the space velocity was 20 h-1, which was calculated
from the ethanol flow through the catalyst and its particle density
from mercury porosimetry.29
Figure 1. Micropore size distribution for activated carbons AZ20 (s)
and AZ46 (- - -).
where rx is the activity to obtain the species x; FEtOH is the
ethanol flow through the catalyst in moles per s, Cx is the ethanol
conversion to the species x, and W is the weight of the sample
in grams.
Results and Discussion
Characteristics of the Catalysts. The micropore size
distributions (MSD) for samples AZ20 and AZ46 are depicted
in Figure 1. These MSD were obtained from the CO2 adsorption
data by applying the equation deduced by Dubinin and Sto-
eckli.31 The MSD of this Figure indicates that the most activated
sample, AZ46, has a wider microporosity and greater micropore
volume than sample AZ20.
Other surface characteristics of the original and oxidized
activated carbons from the series AZ20 and AZ46 are compiled
in Table 1. The oxidation of AZ20 for different periods of time
up to a maximum of 24 h did not affect either the surface area
or the pore texture of the samples, although the oxygen content
of the original sample increased from 1.20% to 6.64% after 24
h of oxidation.
For the same time of oxidizing treatment the samples of series
AZ46 were able to fix a larger amount of oxygen than samples
from series AZ20, increasing from 1.30% to 11.20% after 24 h
treatment. This was due to a greater surface area and more
developed meso- and macroporosity and a wider and greater
microporosity of the samples from series AZ46 than from series
AZ20.
The oxidation of AZ46 now produces a very slight decrease
in the surface area and porosity of the samples, probably because
the original carbon has a higher degree of activation than AZ20,
and, therefore, the pore walls are thinner and destroyed more
easily, as reported previously for other oxidizing treatments.28
Some of the parameters that characterize the surface chemistry
of the catalysts are also compiled in Table 1. Thus, the total
surface acidity, as measured by NaOH titration, increases with
oxidation time for both series of activated carbons because of
the fixation of carboxyl, lactonic, and phenolic groups as shown
elsewhere.26 For the same oxidation time, samples from the
series AZ46 have a greater total acidity than samples from the
series AZ20 because of their higher oxygen content. As a result
of the increase in surface acidity of the activated carbons with
oxidation time there is a marked decrease of their pHPZC. Thus,
the original activated carbons have a basic pHPZC, around 11,
which sharply decreases during the first 30 min of treatment,
and more gradually thereafter. The total surface basicity, as
titrated with HCl, is also given in Table 1, and it decreases when
the degree of oxidation or total surface acidity increases,
indicating that basic surface sites are essentially associated with
the absence of oxygen-containing groups which are predomi-
nantly of acidic nature.
Analysis of the reaction products was done by on-line gas
chromatography using a Perkin-Elmer gas chromatograph,
model 8500, with flame ionization detection and a column
Carbopack B80/120. The flame detector was previously
calibrated by injecting known amounts of the reaction products.
The activity to obtain acetaldehyde, ethene, and ether was
estimated from eq 1
FEtOH‚Cx
rx )
(1)
W

Activated Carbons as Catalysts for Ethanol Reactions
J. Phys. Chem. B, Vol. 102, No. 46, 1998 9241
TABLE 1: Surface Characteristics of Activated Carbons
a
SN2,
W0 ,
Vmeso
,
Vmacro
,
NaOH,
HCl,
sample
O, %
m2‚g-1
cm3‚g-1
cm3‚g-1
cm3‚g-1
meq.g-1
meq‚g-1
pHPZC
AZ20
1.19
2.48
5.89
6.64
1.30
5.57
10.50
11.20
631
608
625
637
914
902
815
810
0.310
0.314
0.313
0.304
0.514
0.510
0.442
0.436
0.238
0.231
0.230
0.206
0.251
0.252
0.254
0.240
0.099
0.089
0.080
0.080
0.130
0.129
0.125
0.120
0.21
0.41
0.96
1.09
0.57
0.93
2.40
2.72
0.76
0.37
0.20
0.18
0.70
0.25
0.14
0.21
10.99
3.26
2.38
2.31
10.85
3.04
2.30
2.20
AZ20-0.5
AZ20-10
AZ20-24
AZ46
AZ46-0.5
AZ46-10
AZ46-24
a W0 ) volume of micropores obtained from DA equation applied to CO2 adsorption at 273 K.
TABLE 2: Total Conversion and Product Distribution (%)
at 453 K from a He/Ethanol Mixture
sample
conversion Aceta Ethene Ether 1,3-Butb EtAcc
AZ20
1.0
17.3
21.0
22.6
1.5
8.9
14.0
16.0
100.0
10.6
11.5
10.8
100.0
25.6
22.0
19.8
AZ20-0.5
AZ20-10
AZ20-24
AZ46
AZ46-0.5
AZ46-10
AZ46-24
25.6
25.2
26.3
62.2
60.3
59.5
0.2
0.4
0.3
1.4
2.6
3.1
17.3
17.0
16.2
53.4
52.3
55.6
0.4
0.9
0.5
3.3
7.9
7.9
a Acetaldehyde. b 1,3-Butadiene. c Ethyl acetate.
Figure 2. Optimum performance envelope (OPE) curves for ethanol
conversion on catalyst AZ20-0.5. Fractional conversion to a particular
reaction product (Cx) versus total conversion (CTotal). 0, ether; O, ethene.
Product Distribution and Activity of the Catalysts. Total
conversion and product distribution obtained with a He/ethanol
mixture at a reaction temperature of 453 K, with some selected
catalysts from the series AZ20 and AZ46, previously heat treated
at 453 K in He flow for 2 h, are shown in Table 2. The original
carbons AZ20 and AZ46 are dehydrogenation catalysts produc-
ing acetaldehyde as the only reaction product. The oxidized
carbons are able to give both dehydrogenation and dehydration
reactions. The major product obtained is ether. The conversion
increased with the oxidation time, especially at the beginning
of oxidation, due to the formation of the dehydration products.
Samples from series AZ20 have higher conversion than samples
from series AZ46, although the former have a lower oxygen
content than the latter. This behavior will be explained later.
catalysts is given in eq 2.
C2H5
O
C2H5
–H2O
C2H5 OH
–C2H5 OH
(2)
–H2O
C2H4
Moreover, the decomposition reaction of ether from a He/
ether mixture at 453 K was also studied with the catalysts AZ20
and AZ20-20. In the sample AZ20 there was no reaction,
whereas carbon AZ20-20 was able to decompose the ether
giving a mixture of acetaldehyde, ethene, and ethanol. In this
case, part of the ethanol formed underwent the dehydrogenation
reaction to acetaldehyde. Thus, the decomposition of ether only
took place on the oxidized samples, where its formation by
dehydration of ethanol was possible.
Table 2 also shows that 1,3-butadiene and ethyl acetate were
obtained as secondary products; 1,3-butadiene was obtained by
the Prin’s reaction,33 in which acetaldehyde reacted with ethene
to yield an enol that produced 1,3-butadiene by dehydration.
Ethyl acetate is formed by a dehydrogenation reaction
between the acetaldehyde formed initially and ethanol molecules
or surface ethoxy groups. The catalysts containing transition
metal elements, active for dehydrogenation of ethanol, produce
large amounts of ethyl acetate, essentially in the high conversion
range.34 In these catalysts the surface active sites for ethyl
acetate formation are presumed to be the same as for the
dehydrogenation of ethanol, therefore, it is difficult to inhibit
the formation of this compound on the above-mentioned
catalysts. This mechanism also has been suggested for the
reaction of an air/ethanol mixture on a Ni catalyst supported
on an activated carbon.26
Acetaldehyde, ether, and ethene were obtained directly from
ethanol by dehydrogenation and dehydration; i.e., they are
primary reaction products coming from ethanol through a
parallel reaction network. A primary product is defined as that
which is produced from the reactant no matter how many surface
intermediates are involved in its formation. However, ether can
be an unstable product under the reaction conditions. Thus,
Szymanski and Rychlicki25 found that dipropyl ether obtained
by the catalyzed dehydration of propanol with a carbon catalyst
was decomposed during the reaction giving propene and
2-propanol. Campelo et al.23 also found, with an AlPO4 catalyst,
that ether was unstable and decomposed in the corresponding
alcohol and olefin. For this reason, we have developed the
corresponding optimum performance envelope (OPE) curves by
plotting the fractional conversion to ether and ethene against
the total conversion for different weight ratios of the catalyst
to the alcohol introduced as described by Ko and Wojciechow-
ski.32 The OPE curves obtained in the case of the catalyst
AZ20-0.5 are plotted in Figure 2, as an example. Above a total
conversion of 12%, a downward deviation occurs in the ether
conversion, which corresponds with an upward deviation in
ethene formation. This is because ether also participates in the
formation of ethene. Thus, ethene is a primary product plus a
secondary stable reaction product. The formation pathway of
ether and ethene from ethanol dehydration on these carbon
Results given in Table 2 indicate that ethyl acetate was
produced only on the oxidized catalysts, and its percentage
increased with the increase in total surface acidity. Nevertheless,
ethyl acetate was not produced on the original carbons AZ20
and AZ46, whose surfaces have a marked basic character with
a pHPZC of about 11. This means that not all the surface active

9242 J. Phys. Chem. B, Vol. 102, No. 46, 1998
Carrasco-Mar´ın et al.
TABLE 3: Activity of the Catalysts for Acetaldehyde, Ethene, and Ether Formation at 453 K from a He/Ethanol Mixturea
racet
,
rethene
,
rether,
sample
µmol‚g-1‚s-1 × 102
µmol‚g-1‚s-1 × 102
µmol‚g-1‚s-1 × 102
Eacet, kJ‚mol-1
Eethene, kJ‚mol-1
Eether, kJ‚mol-1
AZ20
0.28
0.33
0.43
0.41
0.43
0.43
0.55
0.53
58.7 ( 0.9
48.1 ( 0.6
51.9 ( 1.1
49.4 ( 2.6
60.3 ( 1.3
67.5 ( 4.2
63.3 ( 4.2
59.4 ( 1.4
AZ20-0.5
AZ20-10
AZ20-24
AZ46
AZ46-0.5
AZ46-10
AZ46-24
0.81
0.94
1.00
3.92
4.48
4.55
175.5 ( 5.3
153.7 ( 1.6
149.7 ( 1.7
65.7 ( 2.6
60.0 ( 1.8
58.2 ( 3.2
0.29
0.43
0.44
1.81
2.62
2.99
152.7 ( 1.0
143.8 ( 3.5
138.0 ( 5.2
74.0 ( 3.3
69.0 ( 3.7
59.6 ( 4.4
a Activation energies in the temperature range, 413-453 K.
Figure 4. H2 evolution at 453 K against reaction time for catalyst
AZ20-20.
Figure 3. Arrhenius plots for acetaldehyde, ethene, and ether formation
on catalyst AZ46-24. 0, ether; 4, ethene; O, acetaldehyde.
of Eether compared with Eethene. Similar behavior was found in
the dehydrogenation of ethanol on different AlPO4 catalysts.23
These values decreased slightly when the number of acid surface
sites of the Bro¨nsted type increased with oxidation.
centers of the carbon involved in the dehydrogenation of ethanol
are capable of producing ethyl acetate.
The activity values for acetaldehyde, ethene, and ether
formation, racet, rethene, and rether, respectively, at a reaction
temperature of 453 K on some selected catalysts from both series
of activated carbons are given in Table 3, together with the
activation energy to obtain the products above within the
temperature range between 413 and 453 K. These activation
energies were obtained from Arrhenius plots similar to that
shown in Figure 3, as an example. The numbers for the
experimental points indicate the order in which they were
determined and there is as good agreement for data obtained
when either increasing or decreasing temperatures cycles. This
is a good indication that, in the experimental procedure followed
and explained above, the surface active sites remained clean in
each activity measurement and the catalyst did not deactivate.
This behavior is different from that found previously by
Grunewald and Drago.27
If there were no deactivation for acetaldehyde formation, it
would be possible to detect H2 evolution from the catalyst during
the reaction time. Thus, the hydrogen produced during the
conversion reaction of ethanol was followed with a mass
spectrometer as a function of reaction time. The results obtained
with catalyst AZ20-20 are depicted in Figure 4, as an example,
which indicates that there is a large decrease in H2 evolution at
the beginning of the reaction time reaching a steady-state
evolution at approximately 30 min of reaction. These results
show that most of the active sites for the dehydrogenation
reaction are lost at the beginning of the reaction by hydrogen
chemisorption. However, some of these were not deactivated,
and these were responsible for the activity measured on these
catalysts.
The dehydration activities (Table 3), rethene and rether, are
higher in oxidized samples from series AZ20, despite their lower
oxygen content and total surface acidity than samples from series
AZ46 (compared at the same oxidation time). This behavior
could be explained if the dehydration reactions took place only
on the external surface of the particles. Thus, as shown
elsewhere,29 due to the larger molecular size of the CO2-evolving
groups, such as the carboxyl and lactone groups, there will be
some hindrance to their formation in the narrower pores of
sample AZ20 and, therefore, they will be fixed on the sample
more slowly than in sample AZ46, which has a more developed
and wider porosity. Therefore, the relative amount of volumi-
nous acid groups on the external surface of samples from series
AZ20 (especially at the beginning of the oxidation treatment)
would be higher than in the samples from the series AZ46. This
explains the much higher values of rethene and rether in sample
AZ20-0.5 than in sample AZ46-0.5, because in the latter sample,
and in accordance with the arguments above, most of the acidic
groups would be on the internal surface of the particles
inaccessible for dehydration. Likewise, the increase in oxidation
time between 0.5 and 24 h augments the rethene and rether values
for samples from series AZ46 to a greater extent (52% and 65%,
respectively) than for samples from AZ20 (23% and 16%,
respectively), because as oxidation time increases, the samples
from series AZ46 increase the carboxyl acid groups placed on
the external surface to a greater extent than the samples from
series AZ20.
The original carbons from both series, AZ20 and AZ46, had
a basic character (Table 1) and were only dehydrogenation
catalysts (Table 3). When these samples were oxidized the total
surface acidity, as titrated with NaOH, increased and the total
surface basicity, as titrated with HCl, decreased (Table 1).
However, for both series of samples racet values increased with
oxidation time as shown in Table 3. This trend could be
The dehydration of ethanol (Table 3) only takes place with
oxidized samples, increasing the activity with the increase in
total surface acidity for both series of samples. This indicates
that the dehydration reactions only occur on acid surface sites
of the Bro¨nsted type. In this process, the formation of ether is
more favored than that of ethene as shown by the lowest values

Activated Carbons as Catalysts for Ethanol Reactions
J. Phys. Chem. B, Vol. 102, No. 46, 1998 9243
TABLE 4: Activity of the Catalysts for Acetaldehyde,
Ethene, and Ether Formation at 453 K from Both He/
Ethanol and Air/Ethanol Mixture
TABLE 5: Amounts of H2, CO, and CO2 Evolved from the
Pretreated AZ46-20 Samples after Heating at 1273 K in He
pretreatment
racet
,
rethene
,
rether,
T (K) atmosphere H2, mmol‚g-1 CO, mmol‚g-1 CO2, mmol‚g-1
µmol‚g-1‚s-1 µmol‚g-1‚s-1 µmol‚g-1‚s-1
453
653
873
1273
653
873
He
He
He
He
H2
H2
H2
0.03
0.10
0.06
0.05
0.20
1.61
0.37
3.85
3.45
2.05
0.05
3.44
0.22
0.05
1.53
0.56
0.16
0.04
0.60
0.06
0.00
sample
mixture
× 102
× 102
× 102
AZ20
He/ethanol
0.28
0.42
0.43
0.41
0.66
0.63
0.70
0.74
AZ20-2
AZ20-6
AZ20-15
AZ20
AZ20-2
AZ20-6
AZ20-15
0.91
1.06
0.96
4.28
4.42
4.46
air/ethanol
1273
0.33
0.50
0.67
1.55
2.00
3.24
TABLE 6: Total Conversion, C, and Activity of the
AZ46-20 Catalyst for Acetaldehyde, Ethene, and Ether
Formation at 453 K after Different Pretreatments from a
He/Ethanol Mixture
explained if the dehydrogenation reaction were also catalyzed
by Lewis acid surface sites which would act by subtracting a
hydride. Thus, the dehydrogenation reaction seems to occur
either on basic or acid surface sites, but not necessarily at the
same time on both sites.
On the other hand, with other solid catalysts the dehydroge-
nation reaction did not proceed on acid-base sites. Thus, it
was reported35,36 that on highly dehydrated silica the reaction
took place on silanol surface groups on which dissociative
adsorption of ethanol occurs. In the oxidized carbons, the ether
groups that are present on these samples29 might act in a similar
way, and alternatively, dehydrogenation could also proceed as
eq 3.
racet
,
rethene
,
rether,
pretreatment
µmol‚g-1‚s-1 µmol‚g-1‚s-1 µmol‚g-1‚s-1
T (K) atmosphere C, %
× 102
× 102
× 102
453
653
873
1273
453
653
He
He
He
He
H2
H2
H2
H2
15.4
1.5
2.7
3.5
12.5
1.3
0.56
0.20
0.52
0.53
0.49
0.29
0.13
0.27
0.45
2.99
0.38
2.24
873
1273
0.7
0.9
effects can affect the surface active sites that are involved in
the dehydrogenation and dehydration reactions of ethanol.
The amounts of H2, CO, and CO2 desorbed from the
pretreated samples, obtained from TPD experiments, are re-
corded in Table 5. Results obtained indicate that samples
pretreated in H2 have a higher H2 content, which depends on
the treatment temperature. Thus, pretreatment at 873 K leaves
the greatest amount of hydrogen chemisorbed, because the
oxygen surface complexes that remained after the pretreatment
in He are reduced during the pretreatment in H2, at the same
time leaving a large amount of hydrogen bound to the nascent
carbon atoms left behind.
Conversion and activity of the pretreated catalysts to produce
acetaldehyde, ethene, and ether at 453 K are compiled in Table
6. Furthermore, all the pretreated catalysts at a temperature
equal to or greater than 653 K also gave acetal (1,1′-
dietoxyethane) and the pretreated samples at 1273 K gave
butanol among the reaction products, which comes from
secondary reactions of acetaldehyde.
Activity for acetaldehyde formation, racet, largely decreased
when the pretreatment was carried out at 653 K in He but
increased for higher pretreatment temperatures, indicating that
the carbon sites left behind by the removal of oxygen complexes
above 653 K were able to catalyze the dehydrogenation reaction.
On the contrary, pretreated samples in H2 above 653 K had a
lower racet, because hydrogen was bound, during the pretreat-
ment, to some of the surface carbon atoms left after reduction
of the oxygen surface groups, making them inactive for
dehydrogenation reaction. Sample pretreated at 873 K had the
lowest activity because the largest amount of hydrogen was
bound at that temperature. The catalyst pretreated either in H2
or He at a temperature equal to or greater than 653 K had no
activity for dehydration, indicating that the carboxyl acid groups
are the only active sites for this reaction, because these groups
are not stable at a temperature greater than 653 K.37
C-O-C + CH3-CH2OH h C-OH + C-OC2H5 f
C-O-C + CH3-COH + H2 (3)
Samples from series AZ46 have slightly higher racet values
than samples from series AZ-20, which indicates that in the
dehydrogenation reaction not only the external surface groups
were available for catalyzing this reaction but also some of the
internal surface groups.
The conversion reactions of ethanol at 453 K were also
studied with an air/ethanol mixture with selected catalysts from
both series of samples. The activity values for samples from
series AZ20 are compiled in Table 4 together with those
obtained with a He/ethanol mixture at 453 K for comparison.
These data show that when the carrier gas was air instead of
He there was a marked increase in the activity for acetaldehyde
formation and a decrease in both rethene and rether. Similar results
were obtained with samples from series AZ46. Thus, the
presence of air in the reactant mixture favors the removal of
hydrogen (coming from the acetaldehyde formation) that
remained chemisorbed on the surface active sites at the first
stage of the reaction (see Figure 4) producing water, which, in
turn, partially inhibited the dehydration reactions.
Effect of the Pretreatments of the Catalysts on Their
Product Distribution and Activity. Sample AZ46-20 was
chosen to carry out these experiments. Different portions of
this sample were pretreated, in the same reactor, under He or
H2 flow at different temperatures up to 1273 K, with a soak
time of 2 h, before carrying out the conversion reactions at 453
K in a He/ethanol mixture. The samples were not exposed to
the atmosphere either after the pretreatment or before the
reaction.
The heat treatment of carbon materials in H2 instead of in an
inert gas, like He, favors the elimination of the oxygen surface
complexes, stabilizes some of the reactive sites of the carbon
surface through the formation of stable CsH bonds, and gasifies
the unsaturated carbon atoms that are more reactive.13 All these
Conclusions
Acetaldehyde, ethene, and ether were the primary products
obtained from the dehydrogenation and dehydration reactions.

9244 J. Phys. Chem. B, Vol. 102, No. 46, 1998
Carrasco-Mar´ın et al.
(6) Bansal, R. C.; Donnet, J. B.; Stoeckli, H. F. ActiVe Carbon; Marcel
Dekker: New York, 1988.
Ethene was also a secondary product obtained from the
decomposition of ether.
(7) Boehm, H. P.; Voll, M. Carbon 1970, 8, 27.
(8) Papirer, E.; Li, S.; Donnet, J. B. Carbon 1987, 25, 243.
(9) Papirer, E.; Dentzer, J.; Li, S.; Donnet, J. B. Carbon 1991, 29, 69.
(10) Studebaker, M. L. Rubber Chem. Technol. 1957, 30, 1400.
(11) Fabish, T. J.; Schleifer, D. E. Carbon 1984, 22, 19.
(12) Leo´n y Leo´n, C.; Solar, J. M.; Calemma, V.; Radovic, L. R. Carbon
1992, 30, 797.
The dehydration reaction took place only on the oxidized
samples. The activity to obtain dehydration products increased
with the total surface acidity, these reactions were catalyzed by
the carboxyl acid groups placed on the external surface of the
particles.
The original activated carbons, which were basic catalysts,
were only dehydrogenation catalysts producing acetaldehyde as
the only reaction product. Formation of this compound
increased with the oxidation of carbons. Thus, the dehydro-
genation reaction seems to take place on either basic or acid
surface groups. Furthermore, in the oxidized samples the
presence of some oxygen surface complexes, such as ether
groups, might also catalyze the dehydrogenation reaction as in
the case of silanol groups in silica samples. Active sites placed
on the external surface and also some of them placed on the
internal surface of the samples participated in the dehydroge-
nation reaction.
At the beginning of the reaction some of the active sites for
the dehydrogenation of ethanol were lost because hydrogen
remained bound to them. However, the presence of air in the
reactant mixture increased the dehydrogenation activity and
decreased the dehydration activities, because the presence of
air kept clean of hydrogen more active sites for dehydrogenation,
producing water which inhibited the dehydration.
(13) Mene´ndez, J. A.; Phillips, J.; Xia, B.; Radovic, L. R. Langmuir
1996, 12, 4404.
(14) Barton, S. S.; Evans, M. J. B.; Halliop, E.; MacDonald, J. A. F.
Carbon 1997, 35, 1361.
(15) Lo´pez-Ramo´n, M. V.; Stoeckli, H. F.; Moreno-Castilla, C.;
Carrasco-Mar´ın, F. Carbon, in press.
(16) Schwab, G. M.; Schwab-Agallidis, E. J. Am. Chem. Soc. 1949, 71,
1806.
(17) Krylov, O. V. Catalysis by Nonmetals; Academic Press: New York,
1970.
(18) Yashima, T.; Suzuki, H.; Hara, N. J. Catal. 1974, 33, 486.
(19) Rudam, R.; Stockwell, A. Catalysis: Specialist Periodical Reports.
The Chemical Society: London, 1977; Vol. 1, p 87.
(20) Tanabe, K. In Catalysis; Anderson, J. R.; Boudart, M., Eds.;
Springer-Verlag: New York, 1981; Vol. 2, p 231.
(21) Tanabe, K.; Misono, M.; Ono, Y.; Hattori, H. In New Solid Acids
and Bases. Their Catalytic Properties; Delmon, B.; Yates; J. T., Eds.;
Elsevier: Amsterdam, 1989; Vol. 15.
(22) Gervasini, A.; Auroux, A. J. Catal. 1991, 131, 190.
(23) Campelo, J. M.; Garc´ıa, A.; Herencia, J. F.; Luna, D.; Marinas, J.
M.; Romero, A. A.; J. Catal. 1995, 151, 307.
(24) Szymanski, G. S.; Rychlicki, G. Carbon 1991, 29, 489.
(25) Szymanski, G. S.; Rychlicki, G. Carbon 1993, 31, 247.
(26) Szymanski, G. S.; Rychlicki, G.; Terzyk, A. P. Carbon 1994, 32,
265.
(27) Grunewald, G. C.; Drago, R. S. J. Am. Chem. Soc. 1991, 113, 1636.
(28) Moreno-Castilla, C.; Ferro-Garc´ıa, M. A.; Joly, J. P.; Bautista-
Toledo, I.; Carrrasco-Mar´ın, F.; Rivera-Utrilla, J. Langmuir 1995, 11, 4386.
(29) Moreno-Castilla, M.; Carrrasco-Mar´ın, F.; Mueden, A. Carbon
1997, 35, 1619.
(30) Moreno-Castilla, M.; Carrrasco-Mar´ın, F.; Mueden, A. In 23rd
Biennial Conference on Carbon; Penn State University: University Park,
PA, 1997; p 190.
(31) Dubinin, M. M.; Stoeckli, H. F. J. Colloid Interface Sci. 1980, 75,
34.
(32) Ko, A. N.; Wojciechowski, B. W. Prog. React. Kinet. 1983, 12,
201.
Pretreatment of the catalysts in H2 at temperatures higher than
653 K decreases the activity for dehydrogenation caused by the
bonding of H atoms with some of the surface active sites created
after reduction of the oxygen surface complexes.
Acknowledgment. The authors acknowledge the financial
support of DGICYT, Project no. PB 94-0754. A.M. acknowl-
edges the Universidad de Granada and ICMA for the fellowship
to carry out his third-cycle studies.
References and Notes
(1) Radovic, L. R.; Rodriguez-Reinoso, F. In Chemistry and Physics
of Carbon; Thrower, P. A., Ed.; Marcel Dekker: New York, 1997; Vol.
25, p 243.
(33) March, J. AdVanced Organic Chemistry, Reactions, Mechanism,
and Structure, 4th ed.; John Wiley & Sons: New York, 1992.
(34) Matsumura, Y.; Hashimoto, K.; Yoshida, S. J. Catal., 1990, 122,
352.
(35) Matsumura, Y.; Hashimoto, K.; Yoshida, S. J. Chem. Soc., Chem.
Commun. 1987, 1599.
(36) Matsumura, Y.; Hashimoto, K.; Yoshida, S. J. Catal. 1989, 117,
135.
(37) Moreno-Castilla, C.; Carrrasco-Mar´ın, F.; Maldonado-Ho´dar, F. J.;
Rivera-Utrilla, J. Carbon 1998, 36, 145.
(2) Boehm, H. P. In AdVances in Catalysis; Eley, D., Pines, H., Weisz,
P. B., Eds.: Academic Press: New York, 1966; Vol. 16, p 179.
(3) Puri, B. R. In Chemistry and Physics of Carbon; Walker, P. L.,
Jr., Ed.; Marcel Dekker: New York, 1970; Vol. 6, p 191.
(4) Mattson, J. S.; Mark H. B. ActiVated Carbon: Surface Chemistry
and Adsorption from Solution; Marcel Dekker: New York, 1971.
(5) Kinoshita, K. Carbon: Electrochemical and Physicochemical
Properties; Wiley: New York, 1988.