Study on catalytic centres of activated carbons modified in oxidising or reducing conditions

Article abstract of DOI:10.1016/j.molcata.2014.09.014

Three model reactions: decomposition of isopropanol, cyclisation of acetonylacetone and cumene decomposition have been used for characterisation of catalytically active centres in activated carbons modified in oxidising or reducing conditions. Oxidised activated carbon catalysts show high activity in isopropanol decomposition and the dominant product of this reaction is propene, which indicate that oxidation of activated carbons with liquid or gas agents stimulates formation of surface oxygen groups, mainly those of acidic character. Modification of activated carbon by reduction with hydrogen or by nitrogenation with ammonia causes partial removal of the oxygen functional groups leading to a decrease in the catalytic activity towards propene and a simultaneous increase in isopropanol conversion to acetone (basic centres formation). The influence of above modifications on the catalytic activity of activated carbons is confirmed by investigation of cyclisation of acetonylacetone. The dominant product of cumene decomposition performed over activated carbon samples studied is α-methylstyrene, which testifies to the presence of electron-transfer centres in these catalysts.

Full text of DOI:10.1016/j.molcata.2014.09.014

Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Study on catalytic centres of activated carbons modified in oxidising
or reducing conditions
∗
Beata Krzy z˙ y n´ ska, Anna Malaika, Paulina Rechnia, Mieczysław Kozłowski
Faculty of Chemistry, Adam Mickiewicz University in Pozna n´ , Umultowska 89b, 61-614 Pozna n´ , Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 May 2014
Received in revised form 8 September 2014
Accepted 10 September 2014
Available online 28 September 2014
Three model reactions: decomposition of isopropanol, cyclisation of acetonylacetone and cumene decom-
position have been used for characterisation of catalytically active centres in activated carbons modified
in oxidising or reducing conditions. Oxidised activated carbon catalysts show high activity in isopropanol
decomposition and the dominant product of this reaction is propene, which indicate that oxidation of
activated carbons with liquid or gas agents stimulates formation of surface oxygen groups, mainly those
of acidic character. Modification of activated carbon by reduction with hydrogen or by nitrogenation
with ammonia causes partial removal of the oxygen functional groups leading to a decrease in the cat-
alytic activity towards propene and a simultaneous increase in isopropanol conversion to acetone (basic
centres formation). The influence of above modifications on the catalytic activity of activated carbons is
confirmed by investigation of cyclisation of acetonylacetone. The dominant product of cumene decom-
position performed over activated carbon samples studied is ␣-methylstyrene, which testifies to the
presence of electron-transfer centres in these catalysts.
Keywords:
Model reactions
Activated carbon
Isopropanol decomposition
Cyclisation of acetonylacetone
Cumene cracking
©
2014 Elsevier B.V. All rights reserved.
1
. Introduction
et al. [13–16]. On the basis of results of butan-2-ol [13] and iso-
propanol [14] decompositions, these authors claim that the activity
of carbon catalysts towards dehydration is related to the pres-
ence of surface carboxyl groups of different strength, while the
activity towards dehydrogenation is related to the simultaneous
presence of acidic and basic Lewis centres. The activated carbons
modified by oxidation have been shown to effectively catalyse iso-
propanol dehydration and dehydrogenation leading to diisopropyl
ether and propene or acetone, respectively, and to be more active
than unmodified carbons. Moreover, it has been established that
dehydration takes place according to the mechanism of parallel or
subsequent reactions on the outer surface of the catalysts, while
dehydrogenation is realised inside the catalyst pores. According to
literature, such model reactions as cyclisation of acetonylacetone
or cumene decomposition have not been applied in investigation
of catalytic centres in activated carbon materials. To the best of
our knowledge, the cyclisation of acetonylacetone has been so far
used only for characterisation of active centres of inorganic cat-
alysts [24–26]. Depending on the nature of the catalyst applied,
this process can lead to formation of 2,5-dimethylfuran (DMF)
or 3-methylcyclopent-2-en-1-one (MCP), requiring Brönsted acid
centres or basic centres, respectively [25]. Cumene decomposition
has been used in investigation of fluorinated carbons and inorganic
phase supported on carbon materials [27–29], but not activated
carbons. This reaction can proceed in two directions. In the pres-
ence of strong Brönsted acidic centres it leads to benzene and
Model reactions are commonly used for determination of chem-
ical character of the surface of solid-state catalysts such as metals,
metal oxides or zeolites [1–4]. The universality of model reactions
follows from the fact that they are simple, usually take place at well-
defined active centres (acidic, basic or redox) and are carried out in
the conditions of catalyst work. Model reactions permit identifica-
tion of the type and estimation of the strength of active centres on
the catalyst surface. Exemplary model reactions include dehydra-
tion of alcohols or isomerisation of alkenes [5,6], taking place on
weak acidic centres or cumene decomposition that needs strong
Brönsted acidic centres [7–12]. The presence of basic centres is
deduced from measurements of the rate of alcohol dehydrogena-
tion to aldehydes.
So far model reactions have been used to characterise the
catalytically active centres mainly on the surface of inorganic cat-
alysts. However, it has been proved that such processes as alcohol
decomposition [13–22] or Knoevenagel condensation [23] can be
successfully used for identification of acidic/basic reaction centres
in carbon materials. Decomposition of different alcohols over mod-
ified activated carbon catalysts have been studied by Szyma n´ ski
∗ Corresponding author. Tel.: +48 61 829 1471; fax: +48 61 829 1555.
E-mail address: mkozlow@amu.edu.pl (M. Kozłowski).
http://dx.doi.org/10.1016/j.molcata.2014.09.014
1
381-1169/© 2014 Elsevier B.V. All rights reserved.

5
24
B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
Table 1
gas phase was performed in a quartz reactor (heated by a tube fur-
Proximate analyses of precursors (wt%).
3
−1
nace) at the gas flow of 20–50 cm min . The samples obtained
were sieved to the grain size ≤0.06 mm and labelled to include
the following information: initial carbon-modifying agent-time of
modification-modification temperature.
Sample
Moisture Ash (dry basis) Volatile matter
dry ash free)
(
Raw brown coal from Konin
colliery
17.0
22.6
33.3
Demineralized brown coal
Pinewood sawdust
0.0
11.9
0.9
2.4
37.0
83.3
2.2. Catalyst characterisation
Textural properties of the carbon catalysts were determined on
the basis of nitrogen adsorption at 77 K, using a Micromeritics Sorp-
tometer ASAP 2010. The BET equation permitted calculation of the
propene (sometimes to m- and p-diisopropylbenzene), while in
the presence of electron-transfer centres it gives styrene and ␣-
methylstyrene and rarer n-propylbenzene and ethylbenzene (with
involvement of Lewis acidic centres) [11,12]. Another possible
product of this process is toluene, which can appear as a result of
side chain cracking [10]. It is worth noting that cumene decomposi-
tion is interesting not only as a model reaction for characterisation
of active centres but also, according to some authors, as a method
for the production of ␣-methylstyrene – an important monomer
for the use in petrochemical industry [27,30,31].
As the model reactions have been hitherto rarely applied in
investigation of activated carbons, the main aim of this study was to
determine the character of catalytically active centres in activated
carbons, modified in oxidising or reducing conditions, by means of
three selected model reactions: isopropanol decomposition, cycli-
sation of acetonylacetone and cumene decomposition.
apparent surface area (SBET), while the micropore volume (Vmicro
)
and external surface area (of meso- and macropores) (Sext) were
found by using the t-plot method [35]. The total pore volume (Vtot)
was obtained from N amount adsorbed at a relative pressure close
2
to unity. The approximate average pore diameter was calculated
using the formula Dav = 4 Vtot/SBET. Quantitative elemental analy-
sis CHNS was made by an Elemental Analyser Vario EL III. The
total acidity of the activated carbon samples was determined by
potentiometric titration.
2.3. Catalytic measurements
The catalytic tests in the reaction of isopropanol decomposi-
tion were performed by the pulse method in a glass reactor with a
fixed bed catalyst (0.02 g) at temperatures from 423 to 723 K. Iso-
propanol was dosed by a microsyringe in the amount of 0.2 ␮l. The
products of dehydration and dehydrogenation were analysed using
a gas chromatograph (connected on-line with reactor) equipped
with a FID detector and a 2 m column packed with 30% Emulphor
O supported on Chromosorb W. The carrier gas was helium.
Cyclisation of acetonylacetone was conducted in a glass reac-
tor with a fixed bed catalyst (0.05 g) at 623 K using procedure
of Michalska et al. [36]. Upon increasing temperature and dur-
ing the reaction helium was blown through the reaction system
2
. Experimental
2.1. Preparation of activated carbon catalysts
The activated carbon samples studied were obtained from
pinewood biomass (sawdust) and brown coal (from Polish colliery
Konin”). Parameters characterising the precursors (the contents
“
of ash, moisture and volatile components) are given in Table 1. The
precursors were subjected to preliminary mechanical processing
3
−1
(30 cm min ). After reaching a desired temperature, acetony-
3
(
grinding and sifting to the mesh size of 0.4 mm), while the brown
lacetone (0.5 cm ) was introduced into the reaction system by
3
−1
coal was also demineralised by the Radmacher and Mohrhauer
method with the use of concentrated hydrochloric acid and
hydrofluoric acid [32]. Then the samples obtained were subjected
to chemical activation with potassium hydroxide (the weight ratio
of KOH:precursor was 1:1) at 1073 K in a neutral gas atmosphere
an infusion pump at the flow rate of 0.5 cm h . These condi-
tions correspond to liquid hourly space velocity (LHSV) of about
−
1
3
−1 −1
1.52 h (or 10.0 cm gcat h ). The outlet of the reactor was con-
nected to a glass U-shaped tube submerged in a cooling mixture
of isopropanol-dry ice, in which the reaction products were con-
densed. After introduction of the whole amount of acetonylacetone
the contents of the U-shaped tube was analysed on a gas chromato-
graph equipped with a FID detector and a 3 m SE-30 column. The
carrier gas used was helium.
Cumene decomposition tests were performed for the activated
carbon catalysts obtained from brown coal from “Konin” colliery.
This process was realised by the pulse method in a glass reactor
with a fixed bed catalysts (0.05 g) at 623 K. After reaching a desired
temperature, cumene was dosed by a microsyringe in the amount
of 1 ␮l to the reactor. The outlet of the reactor was connected to the
chromatograph. The products were analysed using a FID detector
and a 3 m column packed with 5% SE-30 deposited on Chromosorb
G AW-DMCS. Helium was used as a carrier gas.
3
−1
(
argon) supplied at the flow rate of 50 dm h , for 45 min. The
product was treated with a 5% HCl, washed with distilled water
and dried at 383 K till constant mass. Because of the low content
of mineral components, pinewood sawdust was not subjected to
demineralisation.
The initial activated carbons were further subjected to chemical
modifications using different times and temperatures of these pro-
cesses. The oxidation was applied for introduction of new surface
oxygen functional groups, while the aim of reduction was the par-
tial removal of oxygen groups from the activated carbon surface.
The oxidation was realised by exposure of carbon to liquid or gas
oxidisers. The treatment in solution was performed with 30% H O ,
2
2
concentrated HNO , CH COOOH (peroxyacetic acid, PAA, a mixture
3
3
of acetic acid and hydrogen peroxide [33]) or 2.5 M (NH ) S O in
For all model reactions used in this study the blank experiments
(without catalyst) have been performed. The results obtained indi-
cate that the extent of the noncatalytic reactions is negligible as
only traces of products were obtained.
4
2
2
8
1
M H SO (ammonium peroxydisulphate, APS) [34]. In the pro-
2 4
cesses of wet oxidation the sample of 3 g of activated carbon was
3
mixed with 90 cm of the oxidant solution and after completion
of the process the product was carefully washed with warm dis-
tilled water and dried at 383 K overnight. The oxidising agent used
in gas phase was the atmospheric air. The reduction of activated
carbons was realised by their reaction with hydrogen, ammonia
or by high temperature annealing in argon atmosphere of carbon
samples previously oxidised with ammonium peroxydisulphate at
3. Results and discussion
3.1. Catalyst characterisation
The results of proximate analysis of raw brown coal, demineral-
ised brown coal and pinewood sawdust are given in Table 1. The raw
3
03 K. The modification of samples (0.3–1.0 g) with the agents in

B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
525
Table 2
introduced into the carbon structure [37]. The results obtained in
our study confirmed the above relations as the content of oxygen
increased with increasing temperature of oxidation. The influence
of reaction time on results of oxidation is not so well defined and
extension of the reaction time only slightly affects the content of
oxygen in the oxidised samples. The greatest increase in the con-
tent of oxygen was noted after the reaction of the initial activated
carbon samples (KD and S) with a solution of ammonium perox-
ydisulphate at 333 K for 24 h, while the lowest amount of oxygen
was introduced by exposing these carbon samples to air for 4 h. The
second most effective oxidiser was concentrated nitric acid. Our
results differ from those reported by the group of Moreno-Castilla
Elemental analyses of samples prepared from demineralized brown coal (wt%, dry
basis).
Sample
Ash
C
H
N
S
Oa
KD
5.5
1.8
2.8
2.1
2.0
1.2
1.2
1.2
1.3
4.0
2.1
2.0
1.5
2.1
3.1
1.9
87.4
85.2
85.4
84.6
82.5
79.0
76.3
70.5
67.5
88.6
87.2
92.1
92.7
95.4
95.2
94.4
0.8
1.7
1.6
1.0
1.0
0.9
1.3
1.5
1.5
0.6
0.6
0.8
0.9
0.8
0.8
0.8
0.5
0.4
0.5
0.6
0.4
1.0
0.9
0.7
0.8
0.9
0.8
3.3
3.1
0.6
0.8
1.3
1.5
0.4
0.5
0.9
0.4
0.3
0.3
0.6
0.4
0.4
0.4
0.4
0.1
0.1
0.1
0.3
4.3
10.5
9.2
10.8
13.7
17.6
20.0
25.5
28.5
5.5
8.9
1.4
1.7
1.0
0.0
1.3
KD-PAA-4h-333 K
KD-PAA-8h-333 K
KD-H2O2-24h-303 K
KD-H2O2-24h-333 K
KD-HNO3-4h-333 K
KD-HNO3-8h-333 K
KD-APS-24h-303 K
KD-APS-24h-333 K
KD-AIR-4h-573 K
KD-AIR-8h-573 K
KD-NH3-4h-923 K
KD-NH3-4h-1123 K
KD-H2-4h-923 K
KD-H2-8h-923 K
KD-APS-Ar-4h-1123 K
[
38], who claims that the greatest amount of oxygen is incorporated
into the carbon structure by oxidation with concentrated nitric acid
and the lowest – by oxidation with ammonium peroxydisulphate.
The differences can originate from different conditions of oxida-
tion and the use of different precursors and different activation
methods (KOH vs. steam). The data presented in Tables 2 and 3
also show that the content of oxygen introduced upon the oxida-
tion of activated carbon samples with different oxidisers decreases
as APS > HNO > H O > PAA > AIR. Our earlier investigations prove
a
Calculated by difference.
3
2
2
brown coal shows a high content of mineral components (22.6%), so
to eliminate a possible catalytic effect of mineral compounds it was
subjected to demineralisation. Inorganic substances are removed
from coal upon its exposure to concentrated acids as evidenced by
a low content of ash in demineralised coal (0.9%).
Moreover, this process causes only a small increase in the con-
tent of volatile matter (from 33.3% to 37%), which means that
demineralisation does not introduce significant changes in the coal
structure. Pinewood sawdust shows high content of volatile matter
that type of oxidising agent determines not only the amount of oxy-
gen introduced to the carbon structure but also the type of oxygen
structures formed [39,40]. The oxidation in solutions of oxidisers
leads mainly to formation of surface carboxyl groups, while oxi-
dation with air leads to formation of both carbonyl and hydroxyl
(phenolic) groups. Similar conclusions have also been drawn by
other scientists [16,41].
As expected, annealing of oxidised carbon in a high temperature
in an argon atmosphere leads to a significant removal of oxygen
from the carbon structure, which should be interpreted as a result of
thermal decomposition of the oxygen functional groups. Even more
pronounced decline of the oxygen content (particularly great in the
samples based on pinewood sawdust) is detected for the samples
treated with ammonia and hydrogen. In such samples the decrease
in the content of oxygen is a consequence of thermal decomposition
of oxygen functional groups and their reduction by hydrogen [42]
(
83.3%) and high moisture content (11.9%) when compared to the
corresponding contents in demineralised brown coal. As it shows
low content of ash, no demineralisation was performed.
To get materials of different chemical structures, the initial acti-
vated carbons (KD and S) were subjected to modifications such as
oxidation with liquid or gas agents, nitrogenation with ammonia,
reduction with hydrogen or thermal treatment in argon atmo-
sphere of carbon samples previously oxidised with ammonium
peroxydisulphate. Results of elemental analysis and ash content
in all catalyst samples studied are collected in Tables 2 and 3.
As follows from the results, all processes of oxidation (realised
with liquid or gas agents) lead to an increase in the amount of
oxygen relative to that in the initial activated carbons. The con-
tent of oxygen in the samples subjected to oxidation was found to
depend on the type of oxidising agent, temperature and duration
of oxidation. According to literature, the higher the oxidation tem-
perature and the stronger the oxidiser, the more oxygen atoms are
(direct reaction with hydrogen applied or with hydrogen generated
by partial decomposition of ammonia [43]).
Further analysis of the results shows that besides introduction
of significant amount of oxygen the oxidation with nitric acid also
leads to a small increase in the content of nitrogen, which is a result
of incorporation of this element into the carbon structure (among
others in the form of nitro groups [44]). Increase in the content
of nitrogen is also observed as a result of modification of activated
carbon samples with ammonia at a high temperature, which can be
explained by formation of pyridine and pyrrole groups or quater-
nary systems [45]. According to the data from Tables 1 and 2, the
reactions of oxidation with liquid oxidisers usually remove min-
eral compounds, while modification with gas agents (air, ammonia
or hydrogen), in contrast to the effects of wet treatment, does
not lead to washing out of mineral compounds and sometimes
even an increase in the content of ash after such a treatment is
observed.
Table 3
Elemental analyses of samples prepared from pinewood sawdust (wt%, dry basis).
Oa
Sample
Ash
C
H
N
S
S
4.0
2.1
3.9
1.6
2.2
2.0
3.3
1.3
1.8
2.5
3.0
5.0
5.1
5.1
5.7
2.1
88.8
84.5
85.6
83.7
81.4
75.2
74.7
69.0
62.8
87.6
86.9
91.1
91.2
93.4
93.2
94.8
1.7
1.9
1.6
1.0
1.1
2.4
3.8
1.7
1.5
0.8
0.8
0.8
0.9
0.7
0.6
0.9
0.2
0.9
0.6
0.2
0.4
0.5
0.5
0.5
0.7
0.6
0.4
3.0
2.7
0.7
0.5
0.9
0.0
0.0
0.0
0.1
0.1
0.0
0.0
0.3
0.2
0.1
0.1
0.1
0.1
0.1
0.0
0.1
5.3
10.6
8.3
13.4
14.8
19.9
17.7
27.2
33.0
8.4
8.8
0.0
0.0
0.0
0.0
1.2
S-PAA-4h-333 K
S-PAA-8h-333 K
S-H2O2-24h-303 K
S-H2O2-24h-333 K
S-HNO3-4h-333 K
S-HNO3-8h-333 K
S-APS-24h-303 K
S-APS-24h-333 K
S-AIR-4h-573 K
S-AIR-8h-573 K
S-NH3-4h-923 K
S-NH3-4h-1123 K
S-H2-4h-923 K
S-H2-8h-923 K
S-APS-Ar-4h-1123 K
Trace amounts of sulphur (0.1–0.3%) were noted in the samples
from pinewood sawdust, while in the samples from brown coal
the content of sulphur was slightly greater (0.1–1.5%). The content
of hydrogen in all the samples was low; in the samples originat-
ing from brown coal it varied in the range 0.6–1.7%, while in the
samples from pinewood sawdust – in the range 0.6–3.8%.
The texture of the activated carbon samples studied is char-
acterised in Tables 4 and 5. As follows, the chemical activation
by potassium hydroxide gives activated carbons of highly devel-
oped microporous structure and large specific surface area (close
2
to 950 m /g for the samples originated from brown coal and almost
a
2
Calculated by difference.
1160 m /g for the samples prepared from pinewood sawdust).

5
26
B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
Table 4
Textural characteristics of carbon catalysts prepared from brown coal.
2
2
3
3
Sample
SBET (m /g)
Sext (m /g)
Vtot (cm /g)
Vmicro (cm /g)
Dav (nm)
KD
953
1041
980
938
958
856
881
744
415
22
22
21
17
20
18
20
17
14
22
27
23
32
22
25
22
0.48
0.52
0.49
0.48
0.50
0.43
0.44
0.38
0.22
0.50
0.55
0.52
0.72
0.54
0.54
0.46
0.44
0.49
0.46
0.45
0.46
0.40
0.41
0.35
0.19
0.46
0.51
0.48
0.66
0.50
0.50
0.42
2.00
2.00
2.00
2.04
2.09
2.00
2.02
2.02
2.09
1.99
2.00
1.99
2.01
2.00
2.00
1.98
KD-PAA-4h-333 K
KD-PAA-8h-333 K
KD-H2O2-24h-303 K
KD-H2O2-24h-333 K
KD-HNO3-4h-333 K
KD-HNO3-8h-333 K
KD-APS-24h-303 K
KD-APS-24h-333 K
KD-AIR-4h-573 K
KD-AIR-8h-573 K
KD-NH3-4h-923 K
KD-NH3-4h-1123 K
KD-H2-4h-923 K
KD-H2-8h-923 K
KD-APS-Ar-4h-1123 K
999
1100
1041
1426
1082
1087
924
Thus, the type of precursor seems to have little effect on the textural
properties of activated carbon catalysts.
discussed above limited accessibility to pores. This interpretation
is confirmed by the fact that the total pore volume also decreases
with increasing time of the reaction and with its temperature. For
the samples obtained from brown coal the analogous relations are
less pronounced (Table 4). It should be mentioned that exposure of
activated carbons to gas reagents gives samples of surface area and
pore volume greater than in the initial material (KD and S). This
phenomenon can be explained by partial gasification of carbon by
oxygen, hydrogen or the hydrogen coming from decomposition of
ammonia. Thus, it can be assumed that the final surface area and
pore volume are resultants of the two opposite processes: decrease
of porosity caused by pore destruction or their limited accessibility
and increase in pore volume as a result of carbon gasification.
All activated carbon samples have well developed microporous
structure as indicated by average pore diameter (Dav), which is
very close to 2 nm, and a dominant contribution of micropores
in the total surface area of the samples and a dominant con-
tribution of micropore volume in the total pore volume of the
samples.
According to the data displayed in Tables 4 and 5, the effect
of the processes of modification applied was different but nev-
ertheless small. The exceptions were the catalysts obtained from
brown coal and pinewood sawdust subjected to a reaction with
ammonium peroxydisulphate at 333 K, for which a drastic decrease
in the surface area was observed, suggesting a destruction of the
porous structure of these samples upon their strong oxidation.
Another possible explanation of this phenomenon is that car-
bon pores can be blocked by bulky oxygen functional groups, e.g.
carboxyl groups (these samples have the highest content of oxy-
gen, see Tables 2 and 3). This supposition has been confirmed by
results of TPD and FTIR measurements (presented in our other
papers [40,46]), which prove that the content of carboxylic groups
in the discussed carbons is definitely the highest for the sam-
ples oxidised with ammonium peroxydisulphate and decreases in
the sequence APS > HNO > H O > PAA > AIR. Similar relations have
3
2
2
been obtained by our group for activated carbons prepared from
peach stones [39]. Moreover, the partial “regeneration” of porous
structure of the APS oxidised carbons after heat treatment in argon
atmosphere (decomposition of most of the oxygen groups) also sug-
gests that the decrease in SBET observed for oxidised samples results
in some extent from the limited accessibility of nitrogen molecules
to pores during porosity measurements.
For the samples originating from pinewood sawdust (Table 5),
increase in the time of oxidation or in the reaction tempera-
ture leads to a small decrease in the specific surface area, which
can be related to partial destruction of the pore system and/or
3.2. Catalytic activity of carbons in isopropanol conversion
Catalytic activity of the activated carbon samples studied was
assessed on the basis of the measured yield of products of iso-
propanol conversion versus the reaction temperature (Figs. 1–9).
According to the results obtained, all activated carbon samples
show catalytic activity in isopropanol dehydration to propene and
its dehydrogenation to acetone, while the degree of isopropanol
conversion to particular products depends on the mode of carbon
Table 5
Textural characteristics of carbon catalysts prepared from pinewood sawdust.
2
2
3
3
Sample
SBET (m /g)
Sext (m /g)
Vtot (cm /g)
Vmicro (cm /g)
Dav (nm)
S
1157
1178
1154
1131
1057
1067
1058
951
23
22
24
19
23
21
22
25
16
29
20
23
44
26
26
26
0.57
0.59
0.58
0.58
0.57
0.53
0.52
0.48
0.24
0.61
0.58
0.60
0.74
0.61
0.60
0.52
0.54
0.55
0.54
0.54
0.52
0.50
0.49
0.44
0.21
0.57
0.54
0.56
0.66
0.56
0.56
0.48
1.99
2.00
2.01
2.05
2.14
1.99
1.98
2.00.
2.04
2.01
2.01
2.00
2.08
2.02
2.02
1.98
S-PAA-4h-333 K
S-PAA-8h-333 K
S-H2O2-24h-303 K
S-H2O2-24h-333 K
S-HNO3-4h-333 K
S-HNO3-8h-333 K
S-APS-24h-303 K
S-APS-24h-333 K
S-AIR-4h-573 K
S-AIR-8h-573 K
S-NH3-4h-923 K
S-NH3-4h-1123 K
S-H2-4h-923 K
S-H2-8h-923 K
S-APS-Ar-4h-1123 K
462
1217
1159
1197
1424
1199
1186
1052

B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
527
100
100
propene
propene
acetone for KD
acetone for S
80
60
40
20
0
80
propene
acetone for KD-HNO -4h-333 K
3
60
40
20
0
propene
propene
propene
acetone for KD-HNO -8h-333 K
3
acetone for S-HNO -4h-333 K
3
acetone for S-HNO -8h-333 K
3
400
450
500
550
600
650
700
750
400
450
500
550
600
650
700
750
Reaction temperature [K]
Reaction temperature [K]
Fig. 1. Decomposition of isopropanol over initial activated carbons prepared from
Fig. 4. Decomposition of isopropanol over activated carbons oxidised with nitric
brown coal (KD) or from pinewood sawdust (S).
acid.
propene
propene
propene
propene
acetone for KD-PAA-4h-333 K
acetone for KD-PAA-8h-333 K
acetone for S-PAA-4h-333 K
acetone for S-PAA-8h-333 K
100
80
1
00
8
6
4
2
0
0
0
0
0
propene
propene
propene
propene
acetone for KD-APS-24h-303 K
acetone for KD-APS-24h-333 K
acetone for S-APS-24h-303 K
acetone for S-APS-24h-333 K
60
40
20
0
400
450
500
550
600
650
700
750
400
450
500
550
600
650
700
750
Reaction temperature [K]
Reaction temperature [K]
Fig. 2. Decomposition of isopropanol over activated carbons oxidised with peroxy-
acetic acid.
Fig. 5. Decomposition of isopropanol over activated carbons oxidised with ammo-
nium peroxydisulphate.
100
100
8
6
4
2
0
0
0
0
0
8
6
4
2
0
0
0
0
0
propene
propene
propene
propene
acetone for KD-AIR-4h-573 K
acetone for KD-AIR-8h-573 K
acetone for S-AIR-4h-573 K
acetone for S-AIR-8h-573 K
4
00
450
500
550
600
650
700
750
4
00
450
500
550
600
650
700
750
Reaction temperature [K]
Reaction temperature [K]
Fig. 3. Decomposition of isopropanol over activated carbons oxidised with hydro-
gen peroxide.
Fig. 6. Decomposition of isopropanol over activated carbons oxidised with air.

5
28
B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
1
00
samples modification. In the reactions over the catalysts studied no
diisopropyl ether was detected among the reaction products.
Fig. 1 presents isopropanol conversion in the reactions over the
initial activated carbon samples originating from brown coal and
pinewood sawdust. The results indicate that isopropanol decom-
position over unmodified activated carbon samples runs mainly
towards propene, which testifies to the presence of acidic cen-
tres in this catalyst. According to literature data, mainly carboxyl
groups are responsible for acidic properties of carbons, although
weak acidic properties are also shown by certain other types of
oxygen functionalities: hydroxyl (phenol), carboxylic anhydrides,
lactone and lactol [47]. For both KD and S samples, the yield of
propene increases with increasing reaction temperature and both
curves reveal a distinct maximum at about 475 K (more pronounced
for KD) and for higher temperatures their activity slowly decreases,
probably because of thermal decomposition of surface active cen-
tres of acidic character, mainly carboxyl groups. This supposition
has been confirmed by Szyma n´ ski et al. [16], who reported that
carboxyl groups decompose in the range 373–673 K. As the max-
imum is higher for the carbon sample obtained from brown coal,
it can be supposed that this sample contains more acidic groups
than that from pinewood sawdust. This has been confirmed by
results of potentiometric titration, which prove that total acidity
propene
propene
propene
propene
acetone for KD-NH3 4h-923 K
-
acetone for KD-NH -4h-1123 K
3
acetone for S-NH -4h-923 K
3
8
6
4
2
0
0
0
0
0
acetone for S-NH -4h-1123 K
3
4
00
450
500
550
600
650
700
750
Reaction temperature [K]
Fig. 7. Decomposition of isopropanol over activated carbons modified with ammo-
nia.
100
propene
propene
propene
propene
acetone for KD-H -4h-623 K
2
+
of KD sample is 0.78 mmol H /g, whereas for S sample it amounts
acetone for KD-H -8h-623 K
2
+
to 0.58 mmol H /g. After drop in the propene yield, this parameter
acetone for S-H -4h-623 K
2
8
6
4
2
0
0
0
0
0
again increases to form some minima in the propene yield curves.
This increased activity of carbons can originate from other than car-
boxyl groups of acidic character, e.g. mentioned earlier hydroxyl
(phenol), lactone, lactol groups and carboxylic anhydrides. These
functionalities are more thermally stable than carboxyl groups [16]
and they can originally occur in carbons investigated but can also
appear as a result of carboxyl groups decomposition. For exam-
ple adjacent carboxyl groups undergo dehydration to anhydrides,
whereas carboxyl and phenolic groups can react to form lactones
[34].
The activated carbon samples originating from both precursors
also catalyse isopropanol decomposition towards acetone, but the
yield of this product is much smaller than that of propene. The
degree of isopropanol conversion to acetone slightly increases with
increasing temperature. The above results prove that the activated
carbon samples besides acidic active centres (which dominate) con-
tain also basic active centres. According to literature data, the basic
properties of carbons are mainly determined by the presence of
chromene, pyrone, diketone and quinone structures but they are
also related to the presence of a system of delocalised ␲ electrons
from graphene layers [48].
acetone for S-H -8h-623 K
2
4
00
450
500
550
600
650
700
750
Reaction temperature [K]
Fig. 8. Decomposition of isopropanol over activated carbons modified with hydro-
gen.
100
propene
propene
acetone for KD-APS-Ar-4h-1123 K
acetone for S-APS-Ar-4h-1123 K
Figs. 2–6 present results of measurements of the catalytic activ-
ity of activated carbon samples obtained from brown coal or
pinewood sawdust and then modified with liquid or gas oxidis-
ing agents. Oxidation was carried out either in “mild” conditions
8
6
4
2
0
0
0
0
0
(
(
low temperature or short reaction time) or in “drastic” conditions
high temperature or long reaction time). As indicated by the data
shown in Fig. 2, the main direction of isopropanol decomposition
over activated carbon samples oxidised with peroxyacetic acid is
dehydration to propene. It should be noted that the curves illus-
trating the yield of propene are only slightly different from those
showing catalytic activity of unmodified activated carbon samples.
A possible explanation is that peroxyacetic acid is a weak oxidiser
and introduces a small number of carboxyl groups into the acti-
vated carbon structure. This supposition is confirmed by elemental
analysis results, which indicate that the samples oxidised with per-
oxyacetic acid show the lowest content of oxygen from among the
samples modified by liquid oxidisers, see Tables 2 and 3. Compar-
ison of the effects of oxidation with peroxyacetic acid in “mild”
and “drastic” conditions reveals only small differences in the cat-
alytic activity of the activated carbon tested (a bit higher activity is
4
00
450
500
550
600
650
700
750
Reaction temperature [K]
Fig. 9. Decomposition of isopropanol over activated carbons oxidised with ammo-
nium peroxydisulphate and subsequently thermally treated in argon.

B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
529
observed for the samples modified in “drastic” conditions, i.e. at a
longer time of the reaction). For the samples obtained from brown
coal, clear maxima and minima of activity are observed, whose
appearance can be related to the mentioned earlier thermal decom-
position of certain acidic functional groups (carboxyl groups). It is
worth noting that the reaction temperature responsible for these
maxima (about 480 K) is very close to the temperature at which
character of their surfaces. Analysis of the data obtained for the ini-
tial and the oxidised activated carbon samples (shown in Figs. 1–5)
reveals a strong dependence between propene yield and the con-
tent of oxygen introduced into the carbon structure. The content
of oxygen was the lowest in the initial samples (KD and S),
and in the samples oxidised with liquid oxidisers the content
of oxygen increased in the order: PAA < H O < HNO < APS and
2
2
3
CO release starts in TPD experiments [40]. It additionally confirms
in the same order the catalytic activity of the oxidised samples
increased.
2
the supposition on decomposition of carboxyl groups of the cata-
lysts during catalytic test. The above maxima are absent in the case
of curves obtained for the samples prepared from pinewood saw-
dust. A possible explanation is that the concentration of carboxyl
groups is higher in oxidised brown coal-originated samples than in
oxidised samples obtained from pinewood sawdust.
Results of isopropanol decomposition over activated carbon
samples oxidised with hydrogen peroxide are presented in Fig. 3
and show that the catalytic activity of these samples is somewhat
higher than that of the samples oxidised with peroxyacetic acid. The
increased degree of isopropanol conversion to propene means that
Fig. 6 presents the catalytic activity in isopropanol decompo-
sition of activated carbons modified by oxidation in gas phase
(air). The character of the curves in this figure is slightly differ-
ent from that recorded for the carbons oxidised with APS or HNO3.
The samples oxidised with air show initially very low activity
towards isopropanol decomposition to propene, which increases
with increasing temperature. It is known from literature that the
oxidation of carbon with liquid oxidisers favours formation of
carboxyl groups, while the oxidation with air leads mainly to for-
mation of carbonyl and hydroxyl (phenol) groups [41,49]. Of the
latter two only hydroxyl groups are Brönsted acids and can act as
active centres in dehydration of isopropanol.
the oxidation with H O2 is more effective in introduction of oxy-
2
gen functional groups of acidic character than the oxidation with
PAA, which is in agreement with the results of elemental analysis
To sum up the results concerning isopropanol decomposition
over oxidised activated carbon samples, the oxidation by liquid as
well as gas agents leads mainly to formation of surface oxygen func-
tional groups of acidic character. This conclusion is based on the
significant activity of oxidised samples in the reaction towards iso-
propanol dehydration whose main product is propene. The highest
degree of isopropanol conversion to propene was obtained for the
samples prepared from both precursors and oxidised by ammo-
nium peroxydisulphate or concentrated nitric acid. The results are
in agreement with the elemental analysis data showing that oxi-
dation with these oxidisers leads to the greatest increase in the
content of oxygen. Our results have also shown that the type of
oxidation conditions (“mild” or “drastic”) also influences the degree
of isopropanol dehydration. Usually the samples oxidised in “dras-
tic” conditions revealed higher catalytic activity. This observation
is related to the fact that the “drastic” conditions generally favour
introduction of a greater amount of oxygen into the carbon struc-
tures than the “mild” ones (the content of oxygen in the oxidised
carbon samples increases with increasing temperature of oxida-
tion; the effect of increasing reaction time is not so clear as increase
in the reaction time has only small influence on the contents of oxy-
gen). The exceptions are the samples oxidised with APS at 333 K
for 24 h. Although they have the greatest content of oxygen, the
yield of propene in the reactions at 423 K with their use is by
25% lower than that in the presence of the samples oxidised with
APS at 303 K. However, the catalytic activity of the former samples
rapidly increases and then remains at a very high level (close to
100%). The most pronounced influence of the oxidation tempera-
ture on the catalytic activity of carbons in isopropanol dehydration
was observed for the samples oxidised with hydrogen peroxide.
(
Tables 2 and 3).
The results from Fig. 3 also imply that in the “drastic” conditions
higher temperature of oxidation) the yield of propene obtained
(
from dehydration of isopropanol is much higher. This observation
is in agreement with the results of elementary analysis showing
that at higher temperatures of oxidation with H O the amount of
oxygen incorporated into the carbon structure is greater. Another
observation following from Fig. 3 is that the maxima of catalytic
activity of the carbon samples obtained from brown coal and oxi-
dised with H O become less pronounced than those observed for
the initial sample (KD) and the samples oxidised with peroxyacetic
acid.
2
2
2
2
Fig. 4 presents results concerning isopropanol decomposition
over the activated carbons oxidised with concentrated nitric acid
(
which belongs to the group of the strongest oxidisers, along with
ammonium peroxydisulphate [37,40]). The treatment with the
solutions of the above oxidising agents leads mainly to formation of
carboxyl groups [37,39,44,46]. As has been established, the higher
the oxidation temperature and the stronger the oxidiser, the more
oxygen atoms are incorporated into the carbon structure [37]. In
view of the above, the activated carbon samples oxidised with nitric
acid were expected to be highly effective in dehydration reaction.
Indeed, high yield of propene (80–95%, obtained at temperature as
low as 473 K) points to strongly acidic character of these carbon
samples. This character has been confirmed by results of potentio-
metric titration (total acidity of the above samples was in the range
1
+
.99–2.52 mmol H /g). The degree of isopropanol conversion to
propene fast increases up to 473 K and for higher temperatures the
catalytic activity of activated carbon samples modified with nitric
acid gets stabilised or slowly decreases. A small decrease in the
catalysts activity observed for activated carbon samples obtained
from pinewood sawdust and oxidised with HNO3 is probably a
consequence of decomposition of some acidic oxygen functional
groups.
The data on the degree of isopropanol conversion to propene
over the activated carbon samples oxidised with solution of ammo-
nium peroxydisulphate (APS) are presented in Fig. 5. As mentioned
above, this compound is a very strong oxidiser, which is evidenced
by the results of elementary analysis (the samples oxidised with
it show the highest content of oxygen from among all oxidised
ones, Tables 2 and 3). Also total acidity of the above samples
Extension of the time of oxidation with HNO , PAA or air only
3
slightly improves the catalytic activities of the respective samples
in isopropanol dehydration. As follows from our results the type of
precursor used for preparation of activated carbon has no signifi-
cant influence on the degree of isopropanol conversion to propene.
Only for the samples obtained from brown coal and oxidised with
PAA (both in mild and drastic conditions) and with H O at elevated
2
2
temperature, the catalytic activity in isopropanol dehydration was
much better than that of the samples obtained from pinewood
sawdust.
The effect of carbon modification in reducing conditions on
the catalytic activity in isopropanol decomposition is illustrated
in Figs. 7–9. As follows from these results, modification of acti-
vated carbons by reduction with hydrogen and nitrogenation with
ammonia leads to a decrease in the catalytic activity towards
+
was the highest (3.09–3.81 mmol H /g). In the reactions over
carbon catalysts oxidised with APS nearly 100% conversion of iso-
propanol to propene was obtained, indicating almost totally acidic

5
30
B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
conversion
selectivity to DMF (acidic sites)
selectivity to MCP (basic sites)
100
100
80
60
40
20
80
60
40
20
0
samples prepared from brown coal
samples prepared from pinewood
sawdust
0
1
2
3
4
Total acidity [mmol/g]
Fig. 11. Acetonylacetone transformation over carbons prepared from brown coal.
Fig. 10. Yield of propene in the decomposition of isopropanol reaction versus the
total acidity of the activated carbon samples studied.
3.3. Catalytic activity of carbons in acetonylacetone conversion
Results illustrating catalytic activity of the activated carbon
samples studied in cyclisation of acetonylacetone are presented
in Figs. 11 and 12. As follows, cyclisation of acetonylacetone over
all oxidised carbon samples gives 2,5-dimethylfuran (DMF) as the
main product. This fact confirms literature reports claiming that
oxidation increases the acidic character of carbon as a result of
incorporation of oxygen in the form of different functional groups
propene and the domination of acetone in the products of iso-
propanol decomposition. It is a consequence of partial (or in some
samples even total, see Tables 2 and 3) removal of oxygen func-
tional groups whose majority show acidic character (see the results
in Fig. 1). The appearance of acetone in the products of isopropanol
decomposition testifies to the presence of basic centres on the
surface of activated carbon samples [14]. In the samples modi-
fied with ammonia the basic centres can be amine or pyridine
nitrogen functionalities [50]. This supposition is confirmed by an
increased content of nitrogen in the samples after this modifica-
tion (Tables 2 and 3). It is difficult to surmise what is responsible
for the basic properties of the samples reduced with hydrogen.
Surely not the basic oxygen groups because these samples show
no or very little content of oxygen. A possible explanation is that
the basic character of these samples is related to the presence
of some unsaturated sites [42] or ␲ electrons in graphene layers
[
(
53]. The highest selectivity of acetonylacetone cyclisation to DMF
close to 100%), indicating the presence of almost only acidic
sites, was obtained for the samples prepared from both precur-
sors and oxidised with ammonium peroxydisulphate at different
temperatures or with concentrated nitric acid at two different mod-
ification times. On the other hand, high selectivity towards MCP
was observed for the samples reduced with hydrogen or subjected
to thermal treatment, however, much prevalence of the concentra-
tion of basic centres over acidic ones was noted only for the samples
modified with ammonia. The above relations are well pronounced
for the samples obtained from both precursors used and are similar
to those obtained for isopropanol decomposition.
[
51].
As indicated by the presence of two products of isopropanol
On the basis of measurements of the total degree of acetony-
lacetone conversion it can be concluded that the greatest number of
active centres (both acidic and basic) occurs in the samples oxidised
with ammonium peroxydisulphate at 303 K. The number of active
centres is the lowest in the samples modified with APS at 333 K
decomposition (propene and acetone) the thermal treatment in
neutral gas atmosphere at 1123 K of carbon samples oxidised with
ammonium peroxydisulphate gives catalysts having both acidic
and basic active centres (Fig. 9). The presence of basic centres in the
samples studied can be related to the occurrence of thermally sta-
ble oxygen groups of basic character, e.g. mentioned earlier pyrone
functionalities (which decompose only above 1173 K [16]), and
with the presence of ␲ electrons from graphene layers. The yield
of propene is low up to about 573 K but above this temperature it
rapidly increases. As acidic oxygen groups do not exist at temper-
ature of 1123 K [16] and thermal decomposition of isopropanol is
negligible (blank test), some other phenomenon has to be respon-
sible for the rapid increase in propene yield above 573 K. It cannot
be excluded that some mineral substances contained in the carbons
investigated (see Tables 2 and 3) are involved in decomposition of
isopropanol towards propene. According to Gervasini et al. [52] this
process can proceed not only on Brönsted acids but also on Lewis
acidic centres (which are present on the surface of certain mineral
compounds).
conversion
selectivity to DMF (acidic sites)
selectivity to MCP (basic sites)
100
80
60
4
0
0
0
2
Fig. 10 shows the correlation of carbons activity at 623 K in
the reaction of isopropanol conversion to propene with the cat-
alyst total acidity. Analysis of these data indicates that the activity
of the catalysts studied in the above reaction shows a tendency
to increase with increasing of their acidity, although some clearly
marked deviations from this tendency are noted.
S
Fig. 12. Acetonylacetone transformation over carbons prepared from pinewood
sawdust.

B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
00
531
1
100
benzene
toluene
styrene
α
-methylstyrene
ethylbenzene
total yield
8
0
80
60
40
20
0
samples prepared from brown coal
samples prepared from pinewood
sawdust
60
40
20
0
0
1
2
3
4
1st dose of substrate
2nd dose of substrate
3rd dose of substrate
Total acidity [mmol/g]
Fig. 14. Results of cumene decomposition over initial activated carbon prepared
from brown coal versus the number of substrate pulses.
Fig. 13. Selectivity to DMF versus the total acidity of the activated carbon samples
studied.
domination of electron-transfer centres, while the yield of benzene
oscillating near 10% informs about the presence of a low number of
strong Brönsted acidic centres. The yields of the other products do
not exceed 7%. As there are no literature reports on cumene decom-
position over activated carbon, it is difficult to establish which parts
of carbon structure are responsible for its catalytic properties in
this reaction. It can only be supposed that the electron-transfer
centres are various defects in the carbon structure, such as disloca-
tions, voids, atoms with free valencies. The presence of catalytically
active sites of this type has been postulated by e.g. Muradow’s
group [54]. Similar conclusions have been drawn by Fiedorow et al.
(
in particular the sample obtained from brown coal), the samples
reduced with hydrogen and those whose oxidation with APS was
followed by thermal treatment. The relatively low degree of total
conversion of acetonylacetone observed for the carbons oxidised
with ammonium peroxydisulphate at 333 K can be explained tak-
ing into regard the textural parameters of these samples. Although
the oxidation with APS in “drastic” conditions (333 K) leads to the
introduction of the greatest amount of oxygen (Tables 2 and 3), it
also causes a significant decrease in the specific surface area and
in the total pore volume (Tables 4 and 5), related to destruction of
the porous system of carbon samples or/and to limited pore acces-
sibility caused by bulky carboxyl groups. The low total number of
acidic and basic centres in the other above discussed samples can
be explained by destruction of the oxygen functional groups under
the effect of thermal treatment and/or hydrogen.
The catalytic activity of carbons in acetonylacetone conversion
to 2,5-dimethylfuran as a function of catalyst total acidity is pre-
sented in Fig. 13. Similarly as in the case of isopropanol dehydration,
the activity of the catalysts studied in the above reaction shows a
tendency to increase with increasing of their acidity, although some
clearly marked deviations from this tendency are noted. This result
means that besides the acidity also other factors influence the cat-
alytic performance of the activated carbons. In consistence with
earlier discussed observations, such a factor can be the texture of
the catalyst surface.
[
28] who studied fluorinated carbon materials and suggested that
possible carbon active centres towards ␣-methylstyrene are the
paramagnetic centres appearing as a result of cleavage of the C
C
bonds. Also the nature of strong Brönsted acidic centres in carbons
has not been resolved yet. It cannot be excluded that the pres-
ence of these centres is related (at least in part) to the mineral
compounds found in small amounts in the carbon samples stud-
ied (Tables 2 and 3) or to the remains of HCl strongly adsorbed on
the carbon surface during sample preparation (see Section 2).
Figs. 15 and 16 present the yields of the particular products
of cumene decomposition over the activated carbon samples sub-
jected to oxidation with liquid agents, Fig. 17 gives the results
of cumene decomposition over the carbon sample oxidised with
100
benzene
-methylstyrene
total yield
toluene styrene
ethylbenzene
α
3
.4. Catalytic activity of carbons in cumene decomposition
80
60
40
The catalytic activities of activated carbon samples obtained
from brown coal in the process of cumene decomposition per-
formed by the pulse method are characterised in Figs. 14–18. As
follows from Fig. 14, the yields of the majority of cumene decom-
position products decrease from the first to the third substrate
pulse. This tendency was observed for all carbon catalysts tested
in this study and can be explained by the formation of carbona-
ceous deposit blocking the active centres of the catalysts. For this
reason in the following part of the paper only the results for the first
pulse injection of cumene are presented, so as to analyse the results
for the catalysts surface almost not covered with carbonaceous
deposit. Analysis of the results obtained after the first injection of
cumene indicates that the surface of unmodified activated carbon
is characterised by the presence of different types of active cen-
tres. The high yield of ␣-methylstyrene (close to 50%) evidences the
20
0
KD-PAA-4h-333 K KD-PAA-8h-333 K KD-
h-303 K KD-
4h-333 K
Fig. 15. Cumene decomposition over activated carbons prepared from brown coal
and oxidised with peroxyacetic acid or hydrogen peroxide.

5
32
B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
100
100
benzene
-methylstyrene
total yield
toluene styrene
ethylbenzene
benzene
-methylstyrene
total yield
toluene
styrene
ethylbenzene
α
α
80
60
40
20
0
80
60
40
20
0
KD-H
KD-HNO -8h-333 K KD-APS-24h-303 K KD-APS-24h-333 K
Fig. 16. Cumene decomposition over activated carbons prepared from brown coal
and oxidised with nitric acid or ammonium peroxydisulphate.
Fig. 18. Cumene decomposition over activated carbons prepared from brown coal
and modified with ammonia, hydrogen or by thermal treatment.
air, whereas Fig. 18 illustrates changes in the catalytic activity of
activated carbon samples subjected to modification with ammo-
nia, hydrogen or oxidised with APS at 303 K and then annealed in
argon atmosphere. Analysis of these data shows that the dominant
direction of cumene decomposition over all carbon samples investi-
gatedis dehydrogenationandthemainproductis ␣-methylstyrene.
This observation means that the carbon catalysts have a prevailing
content of electron-transfer centres. The greatest amount of ␣-
methylstyrene was obtained over the carbon treated with ammonia
at 923 K and for the sample oxidised with hydrogen peroxide at
Brönsted acidic centres. This observation means that carbon sam-
ples modified by oxidation contain mainly weak acidic centres. On
the other hand, no cumene conversion to benzene over the car-
bon samples modified by treatment with hydrogen or by thermal
treatment in the argon atmosphere means that these modifications
lead to total removal of strong Brönsted acidic centres. Nitrogena-
tion with ammonia also leads to the removal of the majority of
strong acidic centres as the conversion to benzene over the relevant
samples does not exceed 2%. Most probably this effect is related to
the influence of high temperature causing decomposition of the
majority of strong Brönsted acidic centres.
3
33 K, which suggests that these samples have a high content of
the earlier mentioned structural defects. The lowest amount of ␣-
methylstyrene was obtained over a carbon sample subjected to
oxidation with APS at 333 K. However, the low catalytic activity of
this sample can be related to its small specific surface area (Table 4),
and thus a limited contact of reagents with the active centres of the
catalyst.
The greatest number of strong Brönsted acidic centres was found
in the activated carbon samples oxidised with nitric acid (the yield
of benzene close to 18–20%). This result suggests that the pres-
ence of strong Brönsted acidic centres can be partly related to the
The greatest total yield of cumene decomposition products was
observed for the activated carbon samples over which the high-
est degree of conversion to ␣-methylstyrene was obtained, i.e.
the sample oxidised with H O2 at 333 K, the sample treated with
2
ammonia at 923 K and for the initial KD sample. The lowest total
yield of cumene decomposition products was noted for the sam-
ple oxidised with APS at 333 K over which the lowest conversion
to ␣-methyl-styrene was established. As follows from analysis of
our results, there is correlation between the total yield of all prod-
ucts and the conditions of carbon modifications. On the one hand,
for the oxidised samples the total yield increases with increasing
time and temperature of oxidation. The exception is the sample oxi-
dised with APS at 333 K for which the total yield is lower than for
the sample oxidised with APS at 303 K. On the other hand, a reverse
tendency was observed for the samples treated with ammonia and
hydrogen, i.e. with increasing time and temperature of modification
the total yield of cumene decomposition products decreased.
remains of small amounts of the modifying agent (HNO is a strong
3
Brönsted acid), strongly adsorbed on the carbon surface. Oxidation
with the other oxidising liquids leads to lower number of strong
100
benzene
-methylstyrene
total yield
toluene
styrene
ethylbenzene
α
8
6
4
2
0
0
0
0
0
4
. Conclusions
All activated carbon samples studied were found to show cat-
alytic activity in the model reactions of isopropanol decomposition
and cyclisation of acetonylacetone. Oxidation of activated carbon
samples was proved to lead to their increased acidity, while the
treatment with ammonia, hydrogen or thermal treatment lead to
increased basicity of the carbon. The carbon samples oxidised with
ammonium peroxydisulphate or concentrated nitric acid showed
almost fully acidic character. The type of precursor used to obtain
activated carbon samples had insignificant effect on the results of
these model reactions. The results concerning the catalytic activity
of carbon samples in cumene decomposition proved the dominance
of electron-transfer centres over the strongly acidic Brönsted ones.
The highest number of electron-transfer centres was found in the
KD-AIR-4h-573 K
KD-AIR-8h-573 K
Fig. 17. Cumene decomposition over activated carbons prepared from brown coal
and oxidised with air.

B. Krzy z˙ y n´ ska et al. / Journal of Molecular Catalysis A: Chemical 395 (2014) 523–533
533
sample oxidised with H O2 at 333 K and in the sample treated
[23] E. Perozo-Rondón, V. Calvino-Casilda, R.M. Martín-Aranda, B. Casal, C.J. Durán-
Valle, M.L. Rojas-Cervantes, Appl. Surf. Sci. 252 (2006) 6080–6083.
2
with ammonia at 923 K, while the lowest for the sample oxidised
with ammonium peroxydisulphate at 333 K. The greatest number
of strong acidic Brönsted centres was found in the samples oxidised
with nitric acid, whereas the thermal treatment or reduction with
hydrogen resulted in their total removal.
[
24] J. Weitkamp, M. Hunger, U. Rymsa, Microporous Mesoporous Mater. 48 (2001)
55–270.
[25] S. Carre, B. Tapin, N.S. Gnep, R. Revel, P. Magnoux, Appl. Catal. A: Gen. 372 (2010)
6–33.
26] J. Go s´ cia n´ ska, M. Ziółek, E. Gibson, M. Daturi, Catal. Today 152 (2010)
3–41.
27] Y. Sakurai, T. Suzaki, K. Nakagawa, N. Ikenaga, T. Suzuki, Catal. Lett. 69 (2000)
9–64.
28] P. Fiedorow, A. Krawczyk, R. Fiedorow, K.T. Chuang, Mol. Cryst. Liq. Crys. A 354
2000) 435–442.
2
2
[
[
[
3
5
Acknowledgement
(
This work was supported by National Science Centre in Poland
[29] P.M. Boorman, K. Chong, R.A. Kydd, J.M. Lewis, J. Catal. 128 (1991)
537–550.
(
grant no. N N204 179140).
[
[
30] S.M. Bradley, R.A. Kydd, J. Catal. 141 (1993) 239–249.
31] F. Patrick, S. Sugunan, React. Kinet. Mech. Catal. 103 (2011) 125–131.
References
[32] W. Radmacher, P. Mohrhauer, Brennstoff-Chemie 37 (1956) 353–358.
33] R. Pietrzak, H. Wachowska, Fuel 82 (2003) 705–713.
[
[
[
1] M.A. Aramendía, V. Boráu, I.M. García, C. Jiménez, A. Marinas, J.M. Marinas, A.
Porras, F.J. Urbano, Appl. Catal. A 184 (1999) 115–125.
2] C. Lahousse, J. Bachelier, J.-C. Lavalley, H. Lauron-Pernot, A.-M. Le Govic, J. Mol.
Catal. 87 (1994) 329–332.
3] G. Zadrozna, E. Souvage, J. Kornatowski, J. Catal. 208 (2002) 270–275.
4] D. Martin, D. Duprez, J. Mol. Catal. A: Chem. 118 (1997) 113–128.
5] H. Lauron-Pernot, F. Luck, J.M. Popa, Appl. Catal. 78 (1991) 213–225.
6] M. Guisnet, Stud. Surf. Sci. Catal. 20 (1985) 283–297.
[34] C. Moreno-Castilla, F. Carrasco-Marín, A. Mueden, Carbon 35 (1997)
1619–1626.
[35] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, Academic Press,
London, 1982.
[36] A. Michalska, M. Daturi, J. Saussey, I. Nowak, M. Ziółek, Microporous Meso-
porous Mater. 90 (2006) 362–369.
[37] T.J. Bandosz, J. Jagiełło, J.A. Schwarz, Anal. Chem. 64 (1992) 891–895.
[38] C. Moreno-Castilla, M.V. López-Ramón, F. Carrasco-Marín, Carbon 38 (2000)
1995–2001.
[39] I. Gniot, P. Kirszensztejn, M. Kozłowski, Appl. Catal. A 362 (2009) 67–74.
[40] B. Krzy z˙ y n´ ska, A. Malaika, P. Rechnia, A. Toli n´ ska, P. Kirszensztejn, M. Kozłowski
– to be sent.
[41] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Carbon 37 (1999)
1379–1389.
[42] J.A. Menéndez, J. Philips, B. Xia, L.R. Radovic, Langmuir 12 (1996) 4404–4410.
[43] H.P. Boehm, G. Mair, T. Stoehr, A.R. de Rincón, B. Tereczki, Fuel 63 (1984)
1061–1063.
[44] I.I. Salame, T.J. Bandosz, J. Colloid Interface Sci. 240 (2001) 252–258.
[45] J. Przepiórski, M. Skrodzewicz, A.W. Morawski, Appl. Surf. Sci. 225 (2004)
235–242.
[
[
[
[
[
[
[
7] S.E. Tung, E. McIninch, J. Catal. 4 (1965) 586–593.
8] K. Brückman, J. Haber, React. Kinet. Catal. Lett. 10 (1979) 173–177.
9] Z. Sarbak, Appl. Catal. A: Gen. 159 (1997) 147–157.
[
10] N.N. Binitha, P.P. Silija, V. Suraj, Z. Yaakob, S. Sugunan, IOP Conf. Ser. Mater. Sci.
Eng. 17 (2011) 012018.
[
[
[
[
[
[
11] D. Best, B.W. Wojciechowski, J. Catal. 47 (1977) 11–27.
12] A. Corma, H. Farag, B.W. Wojciechowski, Int. J. Chem. Kinet. 13 (1981) 883–895.
13] G.S. Szyma n´ ski, G. Rychlicki, Carbon 29 (1991) 489–498.
14] G.S. Szyma n´ ski, G. Rychlicki, Carbon 31 (1993) 247–257.
15] G.S. Szyma n´ ski, G. Rychlicki, A.P. Terzyk, Carbon 32 (1994) 265–271.
16] G.S. Szyma n´ ski, Z. Karpi n´ ski, S. Biniak, A. S´ wi a˛ tkowski, Carbon 40 (2002)
627–2639.
17] F. Carrasco-Marín, A. Mueden, C. Moreno-Castilla, J. Phys. Chem. B 102 (1998)
239–9244.
18] C. Moreno-Castilla, F. Carrasco-Marín, C. Parejo-Pérez, M.V. López Ramón, Car-
bon 39 (2001) 869–875.
2
[
[
[46] B. Krzy z˙ y n´ ska, A. Malaika, P. Rechnia, M. Kozłowski – to be sent.
[47] J. Lahaye, Fuel 77 (1998) 543–547.
[48] M.A. Montes-Morán, D. Suárez, J.A. Menéndez, E. Fuente, Carbon 42 (2004)
1219–1225.
9
[
[
[
19] E. Jasi n´ ska, B. Krzy z˙ y n´ ska, M. Kozłowski, Catal. Lett. 125 (2008) 145–153.
20] J. Jasi n´ ska, B. Krzy z˙ y n´ ska, M. Kozłowski, Cent. Eur. J. Chem. 9 (2011) 925–931.
21] J. Bedia, J.M. Rosas, J. Márquez, J. Rodríguez-Mirasol, T. Cordero, Carbon 47
[49] V. Strelko Jr., D.J. Malik, M. Streat, Carbon 40 (2002) 95–104.
[50] C.L. Mangun, K.R. Benak, J. Economy, K.L. Foster, Carbon 39 (2001) 1809–1820.
[51] J.P. Chen, S. Wu, K.-H. Chong, Carbon 41 (2003) 1979–1986.
[52] A. Gervasini, J. Fenyvesi, A. Auroux, Catal. Lett. 43 (1997) 219–228.
[53] B.K. Pradhan, N.K. Sandle, Carbon 37 (1999) 1323–1332.
[54] N. Muradov, F. Smith, A. T-Raissi, Catal. Today 102–103 (2005) 225–233.
(
2009) 286–294.
22] J. Bedia, R. Ruiz-Rosas, J. Rodríguez-Mirasol, T. Cordero, J. Catal. 271 (2010)
3–42.
[
3