Catalysis Today 291 (2017) 58–66

Contents lists available at ScienceDirect

Catalysis Today
journal homepage: www.elsevier.com/locate/cattod

Production of hydrogen by dry reforming of ethanol over alumina

supported nano-NiO/SiO2 catalyst
Barnali Bej, Sujoy Bepari, Narayan C. Pradhan ∗ , Swati Neogi
Department of Chemical Engineering, Indian Institute of Technology, Kharagpur, 721302, India

a r t i c l e i n f o a b s t r a c t

Article history: Hydrogen was produced with very high yield by dry reforming of ethanol over alumina supported nano-
Received 20 June 2016 NiO catalyst in silica synthesized using sol-gel method. The activity of the prepared nano-catalyst was
Received in revised form assessed in terms of ethanol conversion, carbon dioxide conversion, carbon monoxide yield and hydrogen
17 November 2016
yield. The catalyst activity as well as distribution of products was observed to vary with Ni loading in
Accepted 1 December 2016
the catalyst and the reaction conditions. Catalyst containing 10% Ni showed good activity among the
Available online 14 December 2016
screened catalyst for ethanol dry reforming. Detailed experimental investigations were carried out over
a wide range of operating parameters such as reforming temperature from 500 ◦ C–850 ◦ C, feed carbon
Keywords:
Dry reforming
dioxide to ethanol molar ratio from 0.5 to 2 and space-time from 19.0 to 41.8 kg cat h/kmol of ethanol
Ethanol fed. The catalyst was found to be active within the range of parameters studied at atmospheric pressure.
Carbon dioxide However, the most favorable reaction conditions were established at 750 ◦ C with CO2 /EtOH molar ratio of
Hydrogen 1.4 and space-time of 24.90 kg cat h/kmol of ethanol fed. Under the optimum condition, almost complete
Syngas conversion of ethanol, 76.0% conversion of CO2 and 100% yield of hydrogen were obtained. The space-
Nano-nickel catalyst time–conversion data in the temperature range of 500–600 ◦ C were fitted into a power law model and
the activation energy of the ethanol dry reforming reaction was determined to be 97.87 kJ/mol.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction dry reforming) has received considerable attention due to envi-

ronmental concerns and the clear need to reduce green house gas
Hydrogen as a fuel is clean, non-polluting and renewable. Hence, (GHG) emissions. Presently, the study of CO2 reforming for hydro-
it is considered as the “green” energy carrier of the future. Advances gen production focuses utilization of natural gas. Nevertheless,
in fuel cell technology make hydrogen more important as a new natural gas is a fossil fuel and hence unsustainable. Therefore, the
energy source for both stationary and vehicular applications. It need for renewable alternatives has been felt globally.
is also widely used in the production of pharmaceuticals, fine Ethanol is a renewable hydrocarbon as it can be produced
chemicals and bulk chemicals such as ammonia and methanol. through the fermentation of biomass or renewable raw materi-
As such, the demand for hydrogen in fuel cell application and als, including energy plants, waste materials from agro-industries
also in chemical industry is continuously increasing [1]. However, or forestry residue materials, organic fraction of municipal solid
the development of a feasible production method for hydrogen waste, etc. Ethanol can be converted to hydrogen through steam
is necessary as most of the hydrogen is available in nature in reforming, partial oxidation, autothermal reforming (oxidative
bound form (as water or hydrocarbons). A number of catalytic reforming) and dry reforming. Although ethanol steam reform-
reforming technologies, such as steam reforming, partial oxidation, ing has been widely investigated [9–19], very insignificant study
and autothermal reforming, have been used for the production of has been performed on dry reforming [20–24]. The present work
hydrogen from various hydrocarbons [2–8]. Among these, catalytic involves dry reforming of ethanol to produce hydrogen.
steam reforming of natural gas, which contains mainly methane, Dry reforming of ethanol is strongly endothermic and produces
is probably the most important and economic process for the pro- only H2 and CO if ethanol reacts in a most desirable way as follows:
duction of hydrogen and synthesis gas in large scale. Recently, the Ethanol dry reforming:
production of hydrogen via carbon dioxide reforming (also called
C2 H5 OH + CO2 → 3CO + 3H2 H◦ = 296.7 kJ/ mol (1)

However, several other undesirable reactions take place at the

∗ Corresponding author. condition of dry reforming giving rise to product composition dif-
E-mail address: ncp@che.iitkgp.ernet.in (N.C. Pradhan). ferent from that could be obtained from purely dry reforming. A list

http://dx.doi.org/10.1016/j.cattod.2016.12.010
0920-5861/© 2016 Elsevier B.V. All rights reserved.
 B. Bej et al. / Catalysis Today 291 (2017) 58–66 59

of possible reactions that may take place during dry reforming of conversion, carbon dioxide conversion, carbon monoxide yield and
ethanol is presented below. hydrogen yield.
Ethanol decomposition:

C2 H5 OH → CO + CH4 + H2 H◦ = 49.6 kJ/mol (2) 2. Experimental section

C2 H5 OH → CO2 + CH4 (3) 2.1. Chemicals used

Ethanol dehydrogenation:
The chemicals used in catalyst preparation and reforming reac-
C2 H5 OH → CH3 CHO + H2 H◦ = 68.5 kJ/mol (4) tion were of analytical grade. Nickel Nitrate hexahydrate (Merck),
Acetaldehyde decomposition: tetraethyl orthosilicate (TEOS) (Merck), ethanol (99%, Merck),
aluminum oxide (active) (SISCO research laboratory), bentonite
CH3 CHO → CH4 + COH◦ = −18.9 kJ/mol (5) powder (Merck) were used in the present work.
Acetaldehyde CO2 reforming:
2.2. Catalyst preparation and characterization
CH3 CHO + CO2 → 2H2 + 3CO (6)

Ethanol dehydration: The catalyst was prepared by sol-gel method as described in our
previous communications [38,39]. The fresh catalyst samples were
C2 H5 OH → C2 H4 + H2 O (7) characterized by X-Ray diffraction, Scanning electron microscopy
Polymerization of ethylene (Coke formation) with energy dispersive X-ray spectrometry, Transmission electron
microscopy and BET surface area analysis as reported elsewhere
C2 H4 → polymers → 2C + 2H2 H◦ = −52.4 kJ/mol (8) [38]. The BET surface area, pore volume and average pore diameter
of alumina supported NiO/SiO2 catalyst were presented in Table 1
Ethanol steam reforming
[38].
C2 H5 OH + H2 O → 2CO + 4H2 , H◦ = 256 kJ/mol (9) The CHNS analysis was performed with the used catalyst on an
Elementer (Model- Vario Macro CUBE) instrument. The combustion
Methane dry reforming:
and reduction chamber temperatures were maintained 1150 ◦ C and
CH4 + CO2 → 2CO + 2H2 H◦ = 247 kJ/mol (10) 850 ◦ C respectively. About 30 mg sample and 60–70 mg tungsten
oxide (WO3 ) were taken with tin foil and poured in a combus-
Water-gas shift reaction: tion chamber. The whole combustion process occurred in oxygen
CO + H2 O → CO2 + H2 H◦ = −41.17 kJ/mol (11) atmosphere.

Methane decomposition:
2.3. Experimental procedure
CH4 → 2H2 + CH◦ = 74.9 kJ/mol (12)
Dry reforming reaction was carried out in a fixed bed vertical
Boudouard reaction:
tubular reactor (10 mm inner diameter) placed inside a cylindrical
2CO → CO2 + CH◦ = −172.4 kJ/mol (13) furnace. The reactor was loaded with catalyst (2.0 g) in such a way
that the centre of the catalyst bed corresponded to the central heat-
Methane steam reforming: ing zone of the furnace. The catalyst particles were mixed with inert
CH4 + H2 O → CO + 3H2 H◦ = 206 kJ/mol (14) particles so that the bed height was maintained at 60 mm. The cat-
alyst was reduced as well as activated while it was heated to 550 ◦ C
◦
CH4 + 2H2 O → CO2 + 4H2 H = 165 kJ/mol (15) with the flow of hydrogen (40 cm3 /min) for 4 h. The thermocou-
Carbon gasification: ple inserted into the thermowell of reactor recorded the catalyst
bed temperature. After catalyst activation, the reaction tempera-
C + H2 O → CO + H2 H◦ = 131.3 kJ/mol (16) ture was fixed at a desired value, the hydrogen flow was stopped
◦ and preselected CO2 and ethanol flows were started. The ethanol
C + 2H2 O → CO2 + 2H2 H = 90 kJ/mol (17)
dry reforming reaction was performed isothermally at atmospheric
Various catalysts based on transition metals [25–31] and noble pressure. The reactor outlet stream was passed through a condenser
metals [32–34] have been used for the production of hydrogen for separation of condensable components in the gas-liquid sepa-
from ethanol by steam reforming. Among these, nickel-based cat- rator. A schematic diagram of the experimental set-up is given in
alysts are widely studied because of their high catalytic activity, Fig. 1.
comparable to expensive noble metal catalysts. However, nickel The product gas stream flow rate was measured using a wet
catalysts deactivate rapidly due to carbon deposition and metal gas meter. The gas was periodically sent to a Gas Chromatograph
sintering in severe operating conditions [35,36]. The Tamman tem- (Model: Chemito GC 1000 DPR) for analysis. The gaseous products
perature, above which nickel sintering can be expected (590 ◦ C), is were analyzed using a thermal conductivity detector (TCD) with a
less than the normal operating temperatures (ca. 800–900 ◦ C) for Spherocarb column (3.175 mm diameter and 2.4 m length). Liquid
steam reforming [37]. The use of a support not only induces ther- samples were collected from the condenser and analyzed in Gas
mal stability but also offers an opportunity for assistance with coke Chromatograph for its ethanol content.
control. High temperature sintering of nickel may be avoided by A series of experiments were carried out at different tempera-
dispersing the metal in a support having very high metal-support tures, feed CO2 /EtOH mole ratios and space-times. The performance
interaction. It has been reported that dispersing Ni in silica gives of the catalyst was evaluated by determining the conversion of
a stable Ni catalyst in steam reforming of methane and ethanol ethanol, conversion of CO2 , yield of hydrogen and yield of carbon
[38,39]. In this study, nano-nickel highly dispersed into silica and monoxide as defined below:
supported on alumina was used for dry reforming of ethanol. The
FC2 H5 OH, in − FC2 H5 OH, out
nickel loading in the catalyst was kept within 15 wt%. The activ- Conversion of C2 H5 OH, XC2 H5 OH (%) = × 100,
ity of the prepared nano-catalyst was assessed in terms of ethanol FC2 H5 OH, in
60 B. Bej et al. / Catalysis Today 291 (2017) 58–66

Table 1
Physico-chemical properties of catalysts at different Ni loadings [38].

Ni loading Actual Ni loading Particle size (nm) BET surface area Average pore Pore volume
(Targeted) (m2 /g) diameter (nm) (cm3 /g)

5% 2.46% 9.50 212 5.8 0.31

7% 5.15% 10.22 244 6.0 0.34
10% 8.84% 10.67 260 6.3 0.41
15% 12.76% 14.53 220 6.2 0.37

Fig. 1. Schematic diagram of the experimental set-up.

70
FCO2 , in − FCO2 , out
Conversion of CO2 , XCO2 (%) = × 100,
FCO2 , in 65

FH2 , out 60
H2 yield, YH2 (%) = × 100,
3(FC2 H5 OH, in − FC2 H5 OH, out )
CO2 Conversion (%)

55

CO yield, YCO (%) 50

FCO, out
= × 100, 45 o
2(FC2 H5 OH, in − FC2 H5 OH, out ) + (FCO2 ,in − FCO2 ,out ) Temperature = 550 C
CO2: C2H5OH = 1:1
where, F terms are the molar flow rates of different components as 40 W/F = 24.90 kg cat h/kmol ethanol
indicated in the subscript and at the inlet and outlet as indicated
by the subscript ‘in’ and ‘out’, respectively. 35

30
3. Results and discussion 4 6 8 10 12 14 16
Ni content (wt%)
3.1. Effect of nickel loading
Fig. 2. Effect of Ni loading on carbon dioxide conversion.
Dry reforming of ethanol was carried out over catalysts contain-
ing 5, 7, 10 and 15 wt% Ni loadings (targeted) at 550 ◦ C using carbon
dioxide to ethanol molar ratio 1:1. Carbon dioxide was introduced presented in Fig. 2. It can be seen from the figure that the conver-
at a flow rate of 30 ml/min and ethanol flow rate was maintained sion of CO2 increases with Ni loading upto 10 wt%. At Ni loadings
at 0.078 ml/min. The variation of CO2 conversion, obtained after greater than 10 wt%, the conversion of CO2 is less. This is partly due
1 h of time-on-stream, with nickel loading in the catalyst has been to clustering of nickel particles at high loadings and partly due to
 B. Bej et al. / Catalysis Today 291 (2017) 58–66 61

70 80
H2 CO CH4
60
70
Product composition(% mole)

50

CO2 Conversion (%)

60
40

30 50
5% Ni 7% Ni
20 10% Ni 15% Ni
CO2: Ethanol(molar) = 1:1
40
W/F= 24.90 kg cat h/kmol ethanol
10

0 30
4 6 8 10 12 14 16 450 500 550 600 650 700 750 800 850 900
o
% Ni loading Temperature( C)

Fig. 3. Effect of Ni loading on reformate composition. Fig. 4. Effect of temperature on CO2 conversion over different Ni loaded catalysts.

100 100
deactivation of catalyst by carbon formation. The maximum CO2
conversion of 63.2% was obtained with catalyst containing 10% Ni. 90
98
All further experiments were carried out with the catalyst contain-
80
ing 10 wt% nickel.
96

Ethanol Conversion (%)

The effect of Ni loading on product distribution (dry reformate) 70 YH2 YCO

H2 and CO yield (%)

has been shown in Fig. 3. It was observed that the lowest CO con- Ethanol conversion
60 94
centration and highest H2 concentration was obtained for 10% Ni
loaded catalyst. Higher rates of WGS reaction resulted lowest CO 50 92
concentration among four catalysts in the dry reforming of ethanol
40
at 550 ◦ C. The source of water for WGS reaction may be the dehydra- 90
tion of ethanol [Eq. (7)]. On the other hand, lower concentrations of 30
CH4 (9.5% and & 7.1% mol) were obtained with catalysts containing 88
20
5% and 7% Ni, respectively. Whereas for 10% Ni containing cata-
lyst, comparatively higher concentration of CH4 (15.4% mol) was 10 86
present in the exit product stream. This may be due to the fact that 450 500 550 600 650 700 750 800 850 900
for 10% Ni loading catalyst at 550 ◦ C, the rate of ethanol decompo-
o
Temperature( C)
sition reaction towards the formation of CH4 becomes significant
Fig. 5. Effect of temperature on ethanol conversion and products yield.
than the rates of methane dry or wet reforming reactions.

70
3.2. Effect of temperature

In order to study the influence of reaction temperature on con- 60

version and yield of products in ethanol dry reforming reaction, H2 CO
Product composition (% mole)

experiments were carried out at different temperatures, ranging 50 CH4 CO2

from 500 ◦ C to 850 ◦ C using CO2 /EtOH molar ratio 1:1 and space-
time of 24.89 kg cat h per kmol of ethanol fed. Fig. 4 shows the effect CO2: C2H5OH(molar) = 1:1
40 W/F= 24.90 kg cat h/kmol ethanol
of temperature on CO2 conversion for all prepared catalysts (5, 7, 10
and 15% Ni loading). As can be seen from this figure, CO2 conversion
increases with increase in temperature and highest CO2 conversion 30
was obtained at 750 ◦ C for all catalysts. Beyond 750 ◦ C, the CO2 con-
version decreases gradually with temperature. It is clear that 10% 20
Ni loading catalyst gives highest CO2 conversions at all tempera-
tures studied. A maximum CO2 conversion of 73.4% was obtained
over 10% Ni loaded catalyst at 750 ◦ C with CO2 /EtOH molar ratio 10

1:1 and space-time of 24.90 kg cat h per kmol ethanol fed.

The variations of ethanol conversions, H2 yield and CO yield 0
450 500 550 600 650 700 750 800 850 900
with temperature are presented in Fig. 5. The experiments were
o
performed in presence of 10% Ni containing catalyst at differ- Temperature( C )
ent temperatures with CO2 /EtOH molar ratio 1:1 and space-time
24.90 kg cat h per kmol ethanol fed. Ethanol conversion increases Fig. 6. Variation of product distribution with reforming temperature.

with increase in reforming temperature and achieved maximum

conversion (∼100%) at 750 ◦ C and becomes constant. H2 yield and in reformate stream at low temperatures indicates that the occur-
CO yield attain 97% and 36%, respectively at 750 ◦ C. Reformate rence of dehydrogenation of ethanol to acetaldehyde (Eq. (4)) and
composition changes significantly with the change in reaction tem- subsequent acetaldehyde decomposition (Eq. (5)) or direct ethanol
perature as shown in Fig. 6. The presence of large amounts of CH4 decomposition (Eq. (2)). It is evident that increase in temperature
62 B. Bej et al. / Catalysis Today 291 (2017) 58–66

100 100 100

C2H5OH
95 95
95
90 90
CO2 Conversion

Ethanol conversion (%)

o
Temperature=750 C 90

CO2 conversion (%)

Ethanol Conversion
85 CO2:C2H5OH(molar)=1.4:1 85
Conversion (%)

o
Temperature = 750 C
W/F=24.90 kg cat h/kmol ethanol W/F= 24.89 kg cat h/kmol ethanol
80 80 85

75 CO2 75
80
70 70
75
65 65

60 60 70
0 60 120 180 240 300 360 420 480 540 600 660 0.0 0.5 1.0 1.5 2.0 2.5
Feed ratio, CO2:C2H5OH (molar)
Time-on-stream (min)

Fig. 7. Time-on-stream activity of 10% Ni loaded catalyst. Fig. 8. Effect of feed composition on conversions of ethanol and CO2 .

70
to 750 ◦ C results rapid decrease in the production of CH4 . Obviously,
CO2 reforming of not only ethanol but also of the intermediate
product acetaldehyde and by-product CH4 proceeds at this stage. 60
Dry reformate composition (% mole)
Also, it can be seen that the H2 concentration and CO concentra-
tion increase gradually with increase in temperature. At this stage, 50 H2 CO
increased amount of CO basically comes from methane dry reform- CH4 CO2
ing (Eq. (10)) whose rate is higher at higher temperatures as the 40 Temperature = 750 C
o

reaction is endothermic. Also, higher rate of methane dry reforming W/F = 24.90 kg cat h/kmol ethanol
results in decreased CO2 concentration. Further increase in tem-
perature to 850 ◦ C, the productions of CO and H2 reach maximum 30
levels. This result indicates the presence of CHx species adsorbed on
catalyst surface which is continuously reformed with CO2 forming 20
more H2 and CO. Besides that, equilibrium of Boudouard reaction
(Eq. (13)) moves backward and leads to higher CO yield as high 10
temperature promotes gasification of the coke accumulated on the
catalyst surface at lower temperatures with CO2 [40].
0
0.2 0.7 1.2 1.7 2.2
3.3. Time-on-stream activity of the catalyst Feed ratio, CO2: C2H5OH (molar)

The stability of the catalyst was tested by determining its dry Fig. 9. Effect of feed ratio on reformate composition.
reforming activity over a longer period of time at the optimum tem-
perature of 750 ◦ C. A 10 h time-on-stream activity was measured in
Feed CO2 /EtOH molar ratio also significantly affects product
terms of conversion of both ethanol and carbon dioxide as shown
distribution as illustrated in Fig. 9. It can be seen from this fig-
in Fig. 7. The catalyst was found to be quite stable as there was only
ure that H2 concentration increases up to feed CO2 /EtOH ratio
about a 4% drop in activity in 10 h. The used catalyst was tested for
of 1.4 and then decreases with increasing CO2 concentration in
its carbon content in CHNS analyzer and it was determined to be
the feed. However, the concentration of CO continuously increases
6.56 mmol C/g cat. The XRD study of the same used catalyst revealed
with increasing CO2 concentration in the feed. These results indi-
almost no change in crystallinity and crystallite size. The catalyst
cate that ethanol decomposition, dry reforming of ethanol, and dry
is, therefore, quite stable even at a high reforming temperature of
reforming of intermediate products occur and lead to the increased
750 ◦ C.
concentration of H2 and CO. Gasification of coke (deposited on the
catalyst surface) is favored at the reaction temperature (750 ◦ C)
3.4. Effect of feed composition with excess CO2 in feed leading to continuous increase in CO con-
centration. The concentration of CH4 drops rapidly with increasing
Dry reforming of ethanol was carried out at different CO2 /EtOH feed ratio and becomes negligible (almost 0.7% mole) at feed ratios
molar ratios at 750 ◦ C and space-time of 24.89 kg cat h per kmol ≥1.4. The presence of excess CO2 in feed facilitates CH4 dry reform-
ethanol fed. Fig. 8 shows the effects of feed CO2 /EtOH molar ratio on ing (Eq. (10)). Beyond CO2 /EtOH molar ratio 1.4, H2 concentration
CO2 and ethanol conversions. It is evident from this figure that the drops remarkably with enhanced production of CO. This may be
ethanol conversion increases with increasing CO2 /EtOH molar ratio due to the occurrence of reverse water-gas shift (R-WGS) reac-
in the feed and becomes almost constant from a feed ratio of 1.4. The tion (Eq. (11)). CO2 concentration increases with further increase
CO2 conversion first increases with increase in feed ratio, reaches in CO2 /EtOH molar ratio beyond 1.4 whereas CH4 concentration
a maximum at a feed ratio of 1.4 and then decreases. Maximum is negligible at higher feed ratios. Obviously, ethanol decomposi-
conversions of both ethanol and CO2 were obtained at a CO2 /EtOH tion (Eq. (3)) and reverse water-gas shift (R-WGS) reaction occur to
molar ratio of 1.4 and these are 99.43% for ethanol and 76% for CO2 . some extent and the water produced is utilized in methane steam
 B. Bej et al. / Catalysis Today 291 (2017) 58–66 63

80 100 80
Prepared NiO/SiO2-Al2O3 catalyst

75
76 90

Ethanol conversion (%)

70

CO2 Conversion (%)

CO2 conversion (%)

72 80
Commercial Ni catalyst
65

68 70
CO2 conversion 60
Ethanol conversion o
o
Temperature=750 C
64 Temperature = 750 C 60 55 CO2:C2H5OH(molar)=1.4:1
CO2: C2H5OH(molar )=1.4:1
W/F=24.90 kg cat h/kmol ethanol

60 50 50
15 20 25 30 35 40 45 10 20 30 40 50 60 70 80 90 100 110 120
Time-on-stream (min)
Space-time, W/F (kg cat h/kmol ethanol fed)
Fig. 12. Comparison of commercial Ni catalyst with prepared catalyst.
Fig. 10. Effect of space-time on conversions of ethanol and CO2 .

70 Table 2
Comparison of developed catalyst with commercial catalyst.

60 Catalyst CO2 Conversion H2 Yield CO Yield

Dry reformate composition (% mole)

(%) (%) (%)

H2 CO
50 NiO/SiO2 -Al2 O3 76.0 100.0 35.0
CH4 CO2 Commercial Catalyst 64.6 96.0 26.0
o
Temperature = 750 C
Conditions: Temperature, 750 ◦ C; Feed CO2 /EtOH molar ratio, 1.4; Space-time,
40 CO2: C2H5OH =1.4:1
24.90 kg cat h/kmol of ethanol fed.

30

higher space-times, the concentrations of all species remain more

20 or less constant. Negligible CH4 concentration at higher space-time
indicates consumption of CH4 by steam reforming. However, it does
10 not affect CO and CO2 concentration much as its rate of formation is
very low. Another point to be noted here is that the ethanol reform-
ing (both dry and wet) kinetics is slower than that of the methane.
0
15 20 25 30 35 40 45 As the methane reforming is faster than the ethanol reforming,
Space time, W/F(kg cat h/kmol ethanol fed) methane concentration drops at a faster rate compared to ethanol
with increasing space-times.
Fig. 11. Effect of space-time on reformate composition.

reforming (Eq. (15)) with increased production of CO2 . Also, cer- 3.6. Activity comparison with commercial catalyst
tain amount of CO2 comes from ethanol decomposition reaction
(Eq. (3)). Fig. 12 compares the time-on-stream CO2 conversions obtained
with the present catalyst (prepared by sol-gel method), i.e., nano-
3.5. Effect of space-time NiO/SiO2 -Al2 O3 and commercial Ni based reforming catalyst.
Experiments were performed at optimum operating condition of
To investigate the effect of space-time on ethanol and CO2 con- 750 ◦ C temperature with CO2 /EtOH molar ratio of 1.4 and space-
versions and its influence on distribution of products, ethanol dry time of 24.90 kg cat h per kmol of ethanol fed. It is clearly observed
reforming reactions were performed by varying the space-time in from the figure that at the beginning, ethanol reforming occurs with
from 19.0 to 41.8 kg cat h per kmol of ethanol fed at 750 ◦ C using high CO2 conversion over both the catalyst, and then it drops and
CO2 /EtOH molar ratio of 1.4. As can be seen from Fig. 10, the con- attains a constant level for some time. The steady state CO2 con-
versions of ethanol and CO2 increase with increase in space-time version obtained with the prepared catalyst is greater than that
upto 25.0 kg cat h per kmol ethanol fed and become almost con- obtained with commercial catalyst (76% vs. 64.6%). Moreover, the
stant. Maximum 76.0% and almost 100% conversion of CO2 and activity of the commercial catalyst decreased at a faster rate com-
ethanol were obtained at 750 ◦ C with CO2 /EtOH molar ratio of 1.4 pared to the prepared catalyst at longer time-on-stream.
and space-time of 25.0 kg cat h per kmol of ethanol fed. The yield of products hydrogen and carbon monoxide obtained
From Fig. 11, it is seen that H2 concentration increases with with different catalysts are presented in Table 2. The hydrogen yield
rapid decrease in CO2 concentration, whereas CO concentration obtained with commercial catalyst is 96%, whereas 100% H2 yield
first increases with increasing space-time and then becomes con- was achieved in carbon dioxide reforming of ethanol over Al2 O3
stant at higher space-time. CH4 concentration drops significantly supported NiO/SiO2 catalyst synthesized in the present study. Also,
with increase in space-time. From these results, we can conclude less CO yield was observed for dry reforming of ethanol (DRE) over
that at the first stage, ethanol decomposition, ethanol dry reforming commercial Ni based catalyst compared to that observed in DRE
and consequently CO2 reforming of CH4 takes place significantly. At over Al2 O3 supported NiO/SiO2 catalyst.
64 B. Bej et al. / Catalysis Today 291 (2017) 58–66

Table 3
Reaction kinetics data at 500 ◦ C.

PE (atm) PCO2 (atm) Reaction Rate (kmol/kg cat h) Ethanol Conversion (%) W/F (kg cat/kmol h)

0.3838 0.6161 0.0162 30.85 19

0.3823 0.6177 0.0178 35.76 20
0.3797 0.6202 0.0193 40.38 20.91
0.3564 0.6435 0.0228 50.66 22.15
0.3327 0.6673 0.0254 59.12 23.23
0.3279 0.6721 0.0256 61.79 24.06
0.3227 0.6773 0.0258 64.34 24.89
0.3090 0.6909 0.0259 67.73 26.12

Table 4
Reaction kinetics data at 550 ◦ C.

PE (atm) PCO2 (atm) Reaction Rate (kmol/kg cat h) Ethanol Conversion (%) W/F (kg cat/kmol h)

0.3242 0.6758 0.028863 54.84 19

0.3250 0.6749 0.028890 57.78 20
0.3261 0.67385 0.028895 60.42 20.91
0.3248 0.6751 0.029084 64.42 22.15
0.3225 0.6775 0.029234 67.91 23.23
0.3016 0.6984 0.030012 72.21 24.06
0.2787 0.7213 0.030707 76.43 24.89
0.2462 0.7538 0.030965 80.88 26.12

80
90
75
85
70
80
65
75
Ethanol conversion (%)

60
CO2 conversion (%)

70
55
65
50
60 45
55 40
o
50 500 C 35 o
500 C
o
45 550 C 30 o
550 C
o
600 C 25
o
600 C
40

35 20
15
15 20 25 30 35 40 45 15 20 25 30 35 40 45
Space-time (kg cat h/kmol ethanol fed) Space-time (kg cat h/kmol ethanol fed)
Fig. 13. Variation of ethanol conversion with space-time at different temperatures. Fig. 14. Variation of CO2 conversion with space-time at different temperatures.

3.7. Kinetics mated using a non-linear least square regression method based
on Levenberg-Marquart algorithm in POLYMATH software by min-
The experiments were conducted to collect the kinetic data at imizing the sum of residual squares of the reaction rate. The
500, 550 and 600 ◦ C temperature, respectively, keeping CO2 -to- parameters to be estimated by non-linear regression least square
ethanol molar ratio constant at 1.4. The space-time was varied from minimization need to provide proper initial guesses to obtain opti-
19.0 to 41.81 kg cat. h/kmol of ethanol fed. The ethanol and CO2 mized value of k, ␣, ˇ. The algorithm converges within a few
conversion data with varying space-time at 500, 550 and 600 ◦ C iterations if the initial guesses are close to final solution. The func-
are presented in Figs. 13 and 14. The conversion increases with tion to be minimized is of the following form:
temperature as expected. But, the conversions of both the reac-
N  2
tants remain almost constant beyond a space-time of 26.12 kg cat. F= rcali − rexpi . (19)
h/kmol of ethanol fed. i=1

For each kinetic run, the partial pressures of ethanol and CO2 Here, rcali and rexpi are calculated rate from model equation and
were obtained from product gas composition. The conversion of experimental rate for each run at all temperatures, respectively.
ethanol, reaction rate and the partial pressure of two reactants are N is the number of experimental observations. The calculated reac-
presented in Tables 3–5. The partial pressure data and calculated tion orders (␣, ˇ) appear very close over the temperature range
reaction rate were employed to find the rate equation. The kinetic of 500–600 ◦ C as shown in Table 6. Therefore, the power law rate
data obtained from the experiments have been correlated in the equation can be expressed as
form of a power-law kinetic model:
−r = k(pEtOH )0.96 (pCO2 )3.95 . (20)
−r = k(pEtOH )˛ (pCO2 )ˇ . (18)
Activation energy of the reaction and frequency factor can be
The kinetic parameter (k) at each temperature and orders evaluated from the Arrhenius equation by plotting logarithm of
of reaction with respect to ethanol and CO2 (␣, ˇ) were esti- rate constant (k) to the inverse of reaction temperature as shown in
 B. Bej et al. / Catalysis Today 291 (2017) 58–66 65

Table 5
Reaction kinetics data at 600 ◦ C.

PE (atm) PCO2 (atm) Reaction Rate (kmol/kg cat h) Ethanol Conversion (%) W/F(kg cat/kmol h)

0.3164 0.6836 0.03304 62.78 19

0.3076 0.6924 0.03365 67.3 20
0.2973 0.7027 0.03413 71.36 20.91
0.2794 0.7206 0.03449 76.4 22.15
0.2586 0.7414 0.03480 80.85 23.23
0.2459 0.7541 0.03456 83.15 24.06
0.2322 0.7678 0.03430 85.37 24.89
0.2123 0.7876 0.03362 87.82 26.12

Table 6
0.0 Estimated model parameters.

-0.2 Temperature (K) Rate constant (k) ˛ ˇ R2

-0.4 773 0.1607 0.233 4.02 0.90

823 0.4410 1.170 3.56 0.98
-0.6 873 0.9165 1.480 4.24 0.97

-0.8
activation energy for steam reforming of ethanol over the same
lnk

-1.0
catalyst in the temperature range of 550–650 ◦ C was reported as
-1.2 27.37 kJ/mol [39]. Higher activation energy for dry reforming com-
pared to steam reforming indicates that the dry reforming kinetics
-1.4 is much slower than the wet reforming kinetics. This is further
-1.6 supported by the higher reaction temperature (750 ◦ C) for dry
reforming compared to a temperature of 650 ◦ C for steam reforming
-1.8 for almost complete conversion of ethanol.
-2.0
0.00114 0.00117 0.00120 0.00123 0.00126 0.00129 4. Conclusions
-1
1/T (K )
Alumina supported nano-NiO-SiO2 catalyst was prepared by
Fig. 15. Arrhenius plot for ethanol dry reforming with estimated rate constant. sol-gel technique and used successfully in dry reforming of ethanol.
The crystallite size was in the range of 9–12 nm in the Ni loading
range of 5–15%. Catalyst containing 8.84 wt% Ni, calcined at 400 ◦ C
0.040
was found as optimum catalyst for DRE in terms of CO2 conversion.
o
500 C The catalyst was also found to be quite stable. In fact, the devel-
o
550 C oped catalyst was found to be more stable than the commercial
0.035
o
600 C reforming catalyst tested in the present study.
The most favorable condition of ethanol dry reforming in terms
of conversion was found at 750 ◦ C with CO2 /EtOH molar ratio of 1.4
Predicted rate (kmol/kg cat. h)

and space-time of 24.90 kg catalyst h per kmol of ethanol fed. Under

0.030 this condition, a 76% conversion of CO2 and almost 100% conversion
of ethanol were obtained with 100% hydrogen yield.
A power-law type rate equation was developed and the rate
constant was estimated by fitting the kinetic data at three different
0.025
temperatures. The activation energy of the dry reforming reaction
was determined to be 97.87 kJ/mol.

0.020 References

[1] A. Harayanto, S. Fernando, N. Murali, S. Adhikari, Energy Fuels 19 (2005)

2098–2106.
[2] J.N. Armor, Appl. Catal. A 176 (1999) 159–176.
0.015
0.015 0.020 0.025 0.030 0.035 0.040
[3] S. Ahmed, M. Krumpelt, Int. J. Hydrogen Energy 26 (2001) 291–301.
[4] M. Serra, P. Salagre, Y. Cesteros, F. Medina, J.E. Sueiras, J. Catal. 197 (2001)
Experimental rate (kmol/kg cat. h) 210–219.
[5] J.R. Rostrup-Nielsen, J. Sehested, J.K. Nørskov, Adv. Catal. 47 (2002) 65–139.
Fig. 16. Experimental rate vs. Predicted rate at different temperatures. [6] M. Nurrunnabi, Y. Mukinakano, S. Kado, B. Li, K. Kunimori, K. Suzuki, K.
Fujimoto, K. Tomishige, Appl. Catal. A 299 (2006) 145–156.
[7] J.G. Seo, M.H. Youn, H.I. Lee, J.J. Kim, E. Yang, J.S. Chung, P. Kim, I.K. Song,
Fig. 15. The activation energy was calculated to be as 97.87 kJ/mol Chem. Eng. J. 141 (2008) 298–304.
[8] M. Levent, D.J. Gunn, M.A. El-Bousiffi, Int. J. Hydrogen Energy 28 (2003)
and the frequency factor as 6.78 × 105 kmol/kg cat. h (atm)4.91 . 945–959.
Finally, the rate equation can be expressed as [9] F. Marino, M. Boveri, G. Baronetti, M. Laborde, Int. J. Hydrogen Energy 26
(2001) 665–668.
−97.87⁄RT [10] A. Therdthianwong, T. Sakulkoakiet, S. Therdthianwong, Sci. Asia 27 (2001)
−r = 6.78 × 105 e (pEtOH )0.96 (pH2 O )3.95 . (21) 193–198.
[11] D.K. Liguras, D.I. Kondarides, X.E. Verykios, Appl. Catal. B 43 (2003) 345–354.
The model predicted rates matched reasonably well with the [12] J. Comas, F. Marino, M. Laborde, N. Amadeo, Chem. Eng. J. 98 (2004) 61–68.
experimental rates at all temperatures as presented in Fig. 16. The [13] Y. Yang, J. Ma, F. Wu, Int. J. Hydrogen Energy 31 (2006) 877–882.
66 B. Bej et al. / Catalysis Today 291 (2017) 58–66

[14] A. Aboudheir, A. Akande, R. Idem, A. Dalai, Int. J. Hydrogen Energy 31 (2006) [27] J.W.C. Liberatori, R.U. Ribeiro, D. Zanchet, F.B. Noronha, J.M.C. Bueno, Appl.
752–761. Catal. A: Gen. 327 (2007) 197–204.
[15] P.V. Mathure, S. Ganguly, A.V. Patwardhan, R.K. Saha, Ind. Eng. Chem. Res. 46 [28] Y. Sekine, A. Kazama, Y. Izutsu, M. Matsukata, E. Kikuchi, Catal. Lett. 132
(2007) 8471–8479. (2009) 329–334.
[16] J.L. Ye, Y.Q. Wang, Y. Liu, H. Wang, Int. J. Hydrogen. Energy 33 (2008) [29] W. Wang, Y. Wang, Int. J. Energy Res. 34 (2010) 1285–1290.
6602–6611. [30] A.F. Lucredio, J.A. Bellido, E.M. Assaf, Catal. Commun. 12 (2011) 1286–1290.
[17] L.P.R. Profeti, E.A. Ticianelli, E.M. Assaf, Appl. Catal. A: Gen. 360 (2009) 17–25. [31] G. Ozkan, S. Gok, G. Ozkan, Chem. Eng. J. 171 (2011) 1270–1275.
[18] F. Wang, Y. Li, W. Cai, E. Zhan, X. Mu, W. Shen, Catal. Today 146 (2009) 31–36. [32] L.V. Mattos, F.B. Noronha, J. Power Sources 145 (2005) 10–15.
[19] C. Graschinsky, M. Laborde, N. Amadeo, A. Le Valant, N. Bion, F. Epron, D. [33] N.R. Peela, A. Mubayi, D. Kunzru, Chem. Eng. J. 167 (2011) 578–587.
Duprez, Ind. Eng. Chem. Res. 49 (2010) 12383–12389. [34] M. Doemoek, A. Oszko, K. Baan, I. Sarusi, A. Erdohelyi, Appl. Catal. A: Gen. 383
[20] K. De Oliveira-Vigier, N. Abatzoglou, F. Gitzhofer, Can. J. Chem. Eng. 83 (2005) (2010) 33–42.
978–984. [35] D.L. Trim, Catal. Today 49 (1999) 3–10.
[21] X. Wu, S. Kawi, Catal. Today 148 (2009) 251–259. [36] J. Sehested, Catal. Today 111 (2006) 103–110.
[22] J.D.A. Bellido, E.Y. Tanabe, E.M. Assaf, Appl. Catal. B 90 (2009) 485–488. [37] D.L. Trimm, Catalyst deactivation 1991, in: C.H. Bartholomew, J.B. Butt (Eds.),
[23] X. Hu, G. Lu, Catal. Commun. 10 (2009) 1633–1637. Stud. Surf. Sci. Catal., 68, 1991, p. 29.
[24] A.M. da Silva, K.R. de Souza, G. Jacobs, U.M. Graham, B.H. Davis, L.V. Mattos, [38] B. Bej, N.C. Pradhan, S. Neogi, Catal. Today 207 (2013) 28–35.
F.B. Noronha, Appl. Catal. B 102 (2011) 94–109. [39] B. Bej, N.C. Pradhan, S. Neogi, Catal. Today 277 (2014) 80–88.
[25] J. Sun, X.P. Qiu, F. Wu, W.T. Zhu, Int. J. Hydrogen Energy 30 (2005) 437–445. [40] M. Cimenti, J.M. Hill, J. Power Sources 186 (2009) 377–384.
[26] M.C. Sanchez-Sanchez, R.M. Navarro, J.L.G. Fierro, Catal. Today 129 (2007)
336–345.