Ammonia treated Mo/AC catalysts for CO hydrogenation with improved oxygenates selectivity

Article abstract of DOI:10.1007/s12039-017-1269-3

Abstract: A series of ammonia treated Mo/Activated Carbon (AC) catalysts were synthesized by wet impregnation method by nominal incorporation of 5, 10 and 15 wt% of molybdenum. The calcined catalysts (500 ^°C , 4 h, N ₂ flow) were subjected to a stepwise ammonia treatment at temperatures from 25 up to 700 ^°C. This work reports for the first time, ammonia treated different loadings of Mo on DARCO mesoporous activated carbon for CO hydrogenation reaction. These catalysts were tested in the reaction temperature range of 250–325^°C , 7 MPa and 12000/h (GHSV for reactor volume 0.5 mL). At 250 ^°C , all the catalysts showed total oxygenate selectivity of ～ 50%, mainly methanol. At 325 ^°C , total oxygenate selectivity of 16.5% with 18% CO conversion was obtained on 10Mo-N/AC. The result of alcohol distribution revealed high selectivity to propanol (39%) over methanol (34%) at 325 ^°C on 10Mo-N/AC which highlights its unique catalytic behavior in CO hydrogenation. Further, 10MoO ₃/ AC catalyst, with no treatment of ammonia, showed only 4% of CO conversion with 96% hydrocarbon selectivity. Only 0.5% of CO conversion was observed on AC itself at 325 ^°C. The improved oxygenates selectivity on 10Mo-N/AC was associated with Mo ^{δ +} sites on AC generated via ammonolysis. Graphical Abstract: SYNOPSIS A series of ammonia treated Mo/Activated Carbon (AC) catalysts were synthesized by wet impregnation method by nominal incorporation of 5, 10 and 15 wt% of molybdenum. These catalysts were tested for CO hydrogenation reaction at 250–325^°C , 7 MPa and 12,000/h (GHSV) for the first time. At 325 ^°C , total oxygenate selectivity of 16.5% was observed on 10Mo-N/AC. [Figure not available: see fulltext.].

Full text of DOI:10.1007/s12039-017-1269-3

J. Chem. Sci. Vol. 129, No. 5, May 2017, pp. 589–599. © Indian Academy of Sciences.
DOI 10.1007/s12039-017-1269-3
REGULAR ARTICLE
Ammonia treated Mo/AC catalysts for CO hydrogenation
with improved oxygenates selectivity
∗
∗
SHARIF F ZAMAN , NAGARAJU PASUPULETY , ABDULRAHIM A AL-ZAHRANI,
MUHAMMAD A DAOUS, S S AL-SHAHRANI, HITOSHI INOKAWA,
LACHEZAR A PETROV and HAFEDH DRISS
Chemical and Materials Engineering Department, Faculty of Engineering, King Abdulaziz University, P.O. Box
80204, Jeddah 21589, Saudi Arabia
E-mail: zfsharif@gmail.com; sfzaman@kau.edu.sa; pasupulety@gmail.com; nsampathrao@kau.edu.sa
MS received 25 December 2016; revised 21 March 2017; accepted 27 March 2017
Abstract. A series of ammonia treated Mo/Activated Carbon (AC) catalysts were synthesized by wet
impregnation method by nominal incorporation of 5, 10 and 15 wt% of molybdenum. The calcined catalysts
◦
◦
(
500 C, 4 h, N2 flow) were subjected to a stepwise ammonia treatment at temperatures from 25 up to 700 C.
This work reports for the first time, ammonia treated different loadings of Mo on DARCO mesoporous activated
carbon for CO hydrogenation reaction. These catalysts were tested in the reaction temperature range of 250–
◦
◦
3
25 C, 7 MPa and 12000/h (GHSV for reactor volume 0.5 mL). At 250 C, all the catalysts showed total
◦
oxygenate selectivity of ∼50%, mainly methanol. At 325 C, total oxygenate selectivity of 16.5% with 18%
CO conversion was obtained on 10Mo-N/AC. The result of alcohol distribution revealed high selectivity to
◦
propanol (39%) over methanol (34%) at 325 C on 10Mo-N/AC which highlights its unique catalytic behavior
in CO hydrogenation. Further, 10MoO3/AC catalyst, with no treatment of ammonia, showed only 4% of CO
◦
conversion with 96% hydrocarbon selectivity. Only 0.5% of CO conversion was observed on AC itself at 325 C.
δ +
The improved oxygenates selectivity on 10Mo-N/AC was associated with Mo sites on AC generated via
ammonolysis.
Keywords. Molybdenum nitride; activated carbon; CO hydrogenation; liquid oxygenates; ammonolysis.
1
. Introduction
products such as aldehydes, esters and ethers are also
produced depending on the type of the catalyst used
and operating conditions. A large number of pub-
1
One of the most important challenges that mankind will
face in the upcoming years, is to secure the needed
energy supply for the increasing energy demand, while
at the same time minimizing the negative environmental
impact of mass energy production. Syngas utilization is
one of the interesting alternative routes for producing
liquid fuels and a variety of chemicals. Syngas can be
produced from different carbon containing materials,
such as coal, natural gas and biomass by means of gasi-
fication and reforming. Syngas can be converted into
hydrocarbons, methanol and higher alcohols depending
on the type of catalyst used and reaction conditions.
Higher alcohol synthesis (HAS) is an exothermic pro-
cess and low temperature and high pressure favor the
alcohol formation. The side reactions taking place are
water gas shift and formation of hydrocarbons. Synthe-
sized hydrocarbons usually are dominated by methane
together with certain amount of other high molecular
weight paraffins and olefins. Non-alcoholic oxygenated
lished reports dealt with the preparation of oxygenates
or hydrocarbons from syngas using bulk catalysts like
1
2
3,4
1
Ni-Mo S, Co-Mo S, Mo C and MoP. Further-
2
2
2
more, Cu mixed Co catalysts were also extensively
5
studied for HAS preparation from syngas.
Supports like Activate Carbon (AC), Al O , SiO and
2
3
2
ZrO have been studied for HAS by CO hydrogena-
2
tion. The support effect strongly depends on the state of
the active phase such as metal, oxide, and sulfide. Mo
based catalysts with acidic supports such as Al O and
2
3
ZrO were found to suppress the alcohol formation path-
2
ways and increased the hydrocarbon formation during
6
CO hydrogenation.
Activated carbon as a support shows some potential
advantages such as large surface area, high thermal sta-
bility and low tendency of carbon deposition. Spreading
the Mo active phase on a high surface area support was
found to be advantageous as it increases the number of
activesitesduetoimproveddispersionanddistribution.7
*
For correspondence
589

590
Sharif F Zaman et al.
Several studies were performed using activated resulting solution was continuously stirred for 1 h. To this
8
–10
carbon
(
and multi-walled carbon nanotubes solution, approximately 5 g of AC was added and the mix-
ture was aged for 12 h. The excess water was evaporated
MWCNT),11,12 as supports for MoS that showed
2
using a hot plate. The resulting solids were dried overnight
enhanced catalyst activity due to higher dispersion of
the sulfide. K-promoted Ni-Mo catalysts supported on
microporous activated carbon was used for syngas to
HAS process by Liakaku et al.13 Ma et al., studied
the influence of acid treated AC as a support with K-Ni-
Mo active phase for CO hydrogenation reaction. The
◦
in a preheated oven at 100 C under static air. All the pre-
◦
pared catalysts were calcined at 500 C for 4 h under N2 flow
(30mL/min). The calcined samples were subjected to step-
14
wise temperature programmed ammonia treatment. One gram
of the calcined sample was placed in a quartz reactor fitted
with quartz frit. The nitridation procedure was as follows: In
K-Ni-Mo/AC catalyst was more selective to oxygenates
>40%, CO -free basis at 5 MPa and 265 C) over bulk
◦
the first step, catalyst was kept at 120 C for 30 min under
◦
(
2
continuous He flow. In the second step, gaseous ammonia
K-Ni-Mo catalyst. The interaction of activated carbon was introduced to the catalyst bed with an approximate flow
with the Ni-Mo species prevents the complete reduc- rate of 400mL/min and the reactor temperature was increased
0
0
◦
◦
tion of the active metals (Ni and Mo lead to methane from 120 to 350 C with a ramping rate of 1 C/min. In the
third step, the catalyst temperature was increased from 350
formation) and preserves partially oxidized Mo and Ni
species, which are thought to be active sites for alcohol
◦
◦
to 700 C with a ramping rate of 0.5 C/min and aged at this
temperature for 2 h under continuous flow of ammonia. In the
fourth step, the reactor was cooled down to room tempera-
ture under He flow at a rate of 30mL/min. In the fifth step,
the sample was passivated with 1 vol.% O2-balanced He at a
flow rate of 30mL/min for 1 h.
These catalysts were denoted as 5Mo-N/AC, 10Mo-N/AC
and 15Mo-N/AC. Herein, the prefix 5, 10 and 15 represent
the weight percentage of molybdenum impregnated on acti-
14
15
formation. Zhang et al., synthesized Ni-Mo-K sul-
fide doped with 15% carbon nanotubes (CNT) and used
for the syngas-to-HAS reaction. The authors concluded
that CNT assists the hydrogen spillover to Ni-Mo-K sur-
face and stabilizes the partially oxidized Ni-Mo species
which led to higher alcohol formation.
Mesoporous AC, instead of microporous AC, have
received much attention because of their relatively large vated carbon. Until unless stated, these samples were used
pore size (>2nm) and uniform porosity which facilitate for characterization studies. The catalyst with no treatment of
the reactant diffusion, increases the pore accessibility gaseous ammonia with 10 wt% Mo loading was denoted as
10MoO3/AC.
and improves the dispersion of supported active com-
ponent by confining them in the well-defined pores.
16–18
Present work reports the first time, ammonia-treated dif- 2.3 Catalyst characterization techniques
ferent loadings of Mo on DARCO (trade mark for acti-
The BET surface area and pore size distribution of the samples
vated carbon derived from lignite and sold by ICI Amer-
used in the present study were determined by a Quantachrome
ica, Inc., Wilmington, Delaware) mesoporous activated
carbon (written as Mo-N/AC throughout this paper) for
Nova Station adsorption equipment at liquid nitrogen temper-
◦
ature after outgassing the samples at 200 C under the flow of
CO hydrogenation reaction. Further, the characteristics
N2 for 2 h.
of the catalyst were established by BET-pore size distri-
TheX-raydiffractogramsofthesamplesusedinthepresent
bution, CO and H -TPD, XPS and HR-TEM techniques.
2
study were obtained on a EQUINOX 1000 inel XRD instru-
ment using Co Kα = 1.7902 Å, X-ray source generator
settings at 40 kV and 30 mA with real time acquisition over
◦
2
. Experimental
110 of 2 . The FTIR analysis of Mo-N/AC samples was
θ
performed on a Bruker Vertex 70/70v FTIR spectrometer
using KBr pellet method. Transmission electron microscopy
2.1 Materials
(
TEM) images and high resolution TEM images and energy
Ammonium heptamolybdate (NH4)6Mo7O24 · 4H2O
dispersive X-ray analysis (EDS) of Mo-N/AC samples were
collected on Technai 200 kV D1234 Super Twin microscope
with camera length of 97 cm.
XPS results of Mo-N/AC samples were collected on a
SPECS GmbH high vacuum multi-technique surface analysis
system equipped with Mg K α 1253.6 eV X-ray source. Tem-
peratureprogrammeddesorption(TPD)analysisofMo-N/AC
samples were determined using adsorption-desorption of CO
and H2 with Quantachrome pulsar automated chemisorption
instrument. All gases were supplied by Abdulla Hashim Gas
(
≥99.9%)) and DARCO mesoporous activated carbon (AC)
with 20–40 mesh grain size were purchased from Sigma-
◦
Aldrich. The obtained AC was dried at 100 C for 12 h and
used without any further purification. According to the AC
manufacturer, it contains ≤12% moisture, dust ≤0.4% and
mineral impurities (K, Si, Sn and Fe) <1%.
2.2 Catalyst preparation
Ammonium heptamolybdate equivalent to 5, 10 and 15 wt% (AHG), Saudi Arabia, and certified concentration (14.95%
of Mo was dissolved in 50 mL of deionized water and the CO-He and 99.99% H2) was used as a probe gas. In a typical

Ammonia treated Mo/AC catalysts for CO hydrogenation
591
experiment, approximately 0.1 g of catalyst was pretreated in 3. Results and Discussion
◦
He gas flow (10mL/min) at 200 C for 1 h followed by hydro-
◦
gen treatment at 400 C for 4 h. Subsequently, the catalyst was 3.1 BET surface area and pore size distribution
brought to room temperature under the flow of He and satu- studies
rated with probe gas (either CO or H2). Desorption of probe
gas was performed after 1 h of He flushing over a temperature Figure 1a shows the physorption isotherms of acti-
◦
◦
range of 30–400 C at a ramping rate of 10 C/min.
vated carbon (AC) and Mo-N/AC samples. The AC and
Mo-N/ACexhibitedtype-IIBrunauer-Deming-Deming-
Teller isotherms typical for mesoporous solids.19
Adsorption hysteresis was reported due to multilayer
physorption isotherms usually associated with capillary
condensation in mesopores. In the present study, H4
type hysteresis loops were observed, which were asso-
ciated with narrow slit-like pores in all the samples.20
Furthermore, in Mo-N/AC samples, capillary conden-
2
.4 Catalytic activity tests
Catalytic activity tests were performed using PID Eng and
Tech system (Spain) equipped with a flow microreactor, mass
flow, pressure and temperature controllers. The mixture of
CO:H2 = 1:1 volume ratio was selected as model feed at
1
2000/h (reactor volume 0.5 mL) GHSV (gas hourly space
velocity). Catalytic activity tests were carried out in the tem-
perature range of 250–325 C with 25 C step size at the
◦
◦
sation of adsorbate N in the pores occurred at a lower
2
applied pressure of 7 MPa. All the gases for the catalytic relative pressure than that of AC under similar physorp-
◦
activity tests were supplied by Abdulla Hashim Gas (AHG), tion conditions (catalyst 200 mg and 200 C, 2 h, N2
Saudi Arabia.
pretreatment), which is indicative of reduction in the
pore size of AC in Mo-N/AC samples. This might be
For example, 500 mg of Mo-N/AC catalyst was loaded in
to a copper lined stainless steel tube reactor (inside diameter
6
.35 mm and length 9.53 mm) diluted with equal amount of
◦
quartz beads was treated with H2 flow of 50mL/min at 500 C
for 2 h. Subsequently, the H2 flow was replaced by N2 and
reactor temperature was brought back to 250 C. The reac-
(
a)
◦
350
00
250
tant gas mixture CO:H2 = 1:1 volume ratio was introduced
AC
3
5Mo-N/AC
0Mo-N/AC
with flow rate of 100mL/min through an activated carbon trap
1
(which traps iron carbonyls formed inside the gas tank) into
15Mo-N/AC
the reactor to attain the required pressure. After attaining the
steady state at a set temperature and pressure of the reaction,
2
1
1
00
50
00
the reactor exit stream was directed through a heated line at
◦
1
60 C to Agilent 7890A GC equipped with TCD connected
with Restek Plot Q column for CO and CO2 separation and to
flame ionization detector (FID) fitted with HP Plot Q column
for separation of hydrocarbons (methane, ethane, ethylene,
propane, propylene and butane) and oxygenates (methanol,
ethanol, propanol, butanol, acetone, acetic acid, acetaldehyde
and ethyl acetate) estimation. Acetone, acetic acid, acetalde-
hyde and ethyl acetate are reported as other oxygenates in
the activity results. At each temperature, five experimental
points were collected and the results were reproducible within
the error of ± 0.1%. The same catalyst load was used for the
tests at the three reaction temperatures. Negligible (<0.1%)
CO conversion was found on empty SS copper lined reactor
under identical reaction conditions. The syngas conversion
and selectivity of the products were calculated by the follow-
ing equations:
5
0
0
0
.2
0.4
0.6
0.8
1.0
Relaꢀve pressure (P/P
0
)
(
b)
1
0Mo-N/AC
Mo-N/AC
0
.025
.020
.015
.010
.005
5
1
5Mo-N/AC
AC
0
0
0
0
CO conversion (mass %)
CO initial moles − CO final moles
=
× 100
× 100
CO initial moles
2
5
30
35
40
45
50
55
60
65
70
75
ni Ci prodcut
Selectivity (mass%) =
ꢀ
Pore diameter ( )
ni Ci
where, ni is the number the carbon atoms in the molecule of Figure 1. (a) BET isotherms; (b) pore size distribution in
the ith component and Ci is the mole fraction of ith component terms of derivative expression of pore volume with pore size
between r and r+dr for samples AC and Mo-N/AC.
in the reaction mixture at the reactor exit.

592
Sharif F Zaman et al.
associated with the location of Mo in the mesopores of support in Mo-N/AC samples. No XRD reflections were
AC with Mo loadings. observed for Mo phases up to 15 wt% of Mo load-
Figure 1b presents the BJH pore size distribution of ing on AC. No crystalline phases of molybdenum were
AC and Mo-N/AC samples. Unimodal mesopore dis- detected for 10 wt% of Mo on Norit activated carbon
tribution was observed in all these samples with the catalysts and attributed to a high degree of dispersion of
22
pore diameter range of 30 − 50 Å. The AC showed the Mo phase on the support. However, in the present
mesoporous texture in the range of 35 − 50 Å and study, the intensity of the AC related X-ray reflections
averaged at 45 Å. On the other hand, Mo-N/AC sam- were decreased for samples from 5Mo-N/AC to 15Mo-
ples showed a shift in the mesopore average diameter N/AC under identical XRD conditions (30 mg catalyst,
to lower side (Figure 1a hysteresis loop was shifted to 2 h analysis). This phenomenon was likely due to inter-
lower P/P ). For instance, 5Mo-N/AC showed meso- actions of Mo species with activated carbon graphite
0
pores in the range of 34–47.5 Å and averaged at 43 Å structure. The results suggest that the Mo-related crys-
and 10- and 15Mo-N/AC samples showed mesopores tallites are ≤ 4 nm, and/or distributed well on the surface
in the range of 32–45.0 Å and averaged at 41.0 Å. The of the support.
results indicate that the mesopores distribution in a wide
range in support AC, whereas, partial filling of AC pores
by Mo component narrowed the diameter range of mes-
3.3 FTIR studies
pores in Mo-N/AC samples. Further, at 15wt.% of Mo FTIR studies were performed to determine the changes
loading, ACshowedreducedvolumeofmesoporescom- on the surface carbon-oxygen functional groups with
pared to 10Mo-N/AC sample which suggests, to some Mo loading in Mo-N/AC samples and the results are
extent, that mesopores were filled totally by Mo species presented in Figure S2 (Supplementary Information).
at this loading of Mo (Figure 1b, line with filled circle). The FTIR spectra of AC exhibited major bands in the
2
1
−1
Champson et al., also reported reduction of BET sur- fingerprint region (400–1500 cm ) and minor bands
−1
face area and pore volume for temperature programmed in the functional group region (4000–1500 cm ). The
ammonia treated 10Mo-N/AC (DARCO) catalyst over broad band in the region of 950–1300 cm (centered
its support. The authors related that the reduction of at 1100 cm ) is ascribed to C-O stretching vibra-
−1
−1
pore volume and BET surface area was due to Mo-N tion, such as those in ethers, phenols and hydroxyl
23
formation inside the mesopores of AC. In this study, the groups. The weak bands observed at 700 and 780–
−1
systematic transformation of mesopore diameter to nar- 800 cm are attributed to out-of-plane O-H bending
row size upon Mo loading on AC was associated with and out-of-plane=C-H bending vibrations, respectively.
−1
stepwise temperature programmed ammonia treatment, The presence of 1580 cm stretching vibration reveals
which resulted in the formation of Mo-N species in the the carbonyl functional groups (C=O) on AC. The band
−1
mesopores of AC.
appearing around 3400 cm is ascribed to O-H stretch-
The AC showed the BET surface area and mesopore ing vibration of carboxylic acid or physically adsorbed
2
volume as 609.0 m /g and 0.460 mL/g,respectively. On water. Similar bands are pronounced in the region of
−1
the other hand, the BET surface area and mesopore 950–1300 and 700–800 cm for Mo-N/AC samples,
volume for 5-, 10- and 15Mo-N/AC samples obtained which indicate the intact AC surface functional groups
2
as 561.0, 486.0 and 218 m /g and 0.449, 0.420 and even after Mo impregnation and ammonolysis process.
0
.297mL/g, respectively. The decrease in the volume In addition, new bands appeared at 900, 1400 and
−1
of mesopores was associated with partial filling of AC 3100 cm in Mo-N/AC samples. The bands at 900 and
mesopores with Mo-N species. Despite the pore filling, 3100 cm are ascribed to C-H vibrations in aromatic
the integrity of mesopore structure of AC in Mo-N/AC carbon structures. The appearance of these bands after
−1
23
was intact, as observed by the N sorption isotherms.
the impregnation of Mo on AC might be associated with
the formation of C-H bonds in aromatic rings, as a con-
sequence of the opening of cyclic ether bridges by the
2
3.2 XRD analysis
23
interaction of molybdenum species with ether groups
2
XRD spectra of AC and Mo-N/AC samples are pre- or hydrogenation of sp carbon during the ammonolysis
−1
sented in Figure S1 (Supplementary Information). Sup- process. The band at 1400 cm is related to deforma-
port (AC) related X-ray reflections are pronounced at tional vibration of physically adsorbed water.24
2
5
θ = 22.0, 24.0, 31.2, 42.8, 44.5, 46.25, 49.9, 52.0,
The decrease in the infrared intensity of carboxyl
9.0, 71.0, 76.0, 81.0 and 98.0 which are attributed to or conjugated carbonyl groups at 1720 cm in 15Mo-
◦
−1
activated carbon graphene framework. Identical reflec- N/AC was related to the interaction between molyb-
tions were pronounced at similar 2 θ values due to the denum species with carboxyl or conjugated carbonyl

Ammonia treated Mo/AC catalysts for CO hydrogenation
593
groups.25 In the present study, more prominent C-O 235.7 eV due to the split components of 3d5/2 and 3d3/2,
−
1
vibration was observed on AC at 1100 cm . However, respectively. The existence of these bands suggests that
6+
the intensity of this band was relatively decreased with the Mo valence state is present in all the studied cata-
21
increase of the Mo loading in Mo-N/AC samples. The lysts. The surface molybdenum content (atomic %) on
result suggests C-O groups are chemical anchoring cen- 5Mo-N/AC, 10Mo-N/AC and 15Mo-N/AC was found to
ters for the studied Mo-N/AC samples which led to be 0.40, 0.70 and 0.82 atomic percentage, respectively.
Mo-O-C type of complexes in agreement with XPS data
where C-O signal was grown with increasing of the Mo
loading.
3
.4a CO-TPD studies: CO-TPD results of reduced
AC and Mo-N/AC samples are presented in Figure 2a.
No CO sorption was observed on AC. On the other hand,
a single desorption signal was observed in all the studied
Mo/AC samples. On 5- and 15Mo-N/AC, CO desorption
peak maximum was observed at 165 C with a capacity
of 45 and 50 μ mol/g, respectively. However, for 10 Mo-
N/AC, CO desorption peak maximum was observed at
2
cate that CO sorption is relatively higher at 10 wt% of
Mo loading over 5 and 15 wt%. This is associated with
homogeneous distribution of Mo species at this load-
ing on AC, as observed by TEM analysis. Three types
3.4 XPS measurements
Figure S3 (Supplementary Information) presents C 1s,
O 1s and Mo 3d XPS patterns for AC and Mo-N/AC
samples. The AC itself showed broad pattern for C 1s
and the satellite bands in the binding energy range of
◦
◦
00 C with a capacity of 100μmol/g. These results indi-
2
82.0–293.0 eV are attributed to different types of oxy-
gen containing functional groups on AC surface. The
Shirley deconvolution of C 1s yielded six chemically-
shifted signals as shown in Figure S3. The C 1s peak
3
centered at 284.6 eV is ascribed to sp carbon of AC
27
of CO sorptions were reported on Mo/Al O surface:
2
3
as the major component. The peak at relatively lower
(
i) physically adsorbed CO in the temperature range of
2
binding energy (283.7 eV) is endorsed to C=C (sp
◦
2
5 − 80 C; (ii) molecular CO adsorption in the temper-
carbon) which suggests the presence of unoxidized
domains in AC. Additionally, four more carbon-related
components exist at 285.5, 287.0, 289.5 and 291.0 eV.
They are assigned to -C-O, carbonyl (-C=O), carboxy-
◦
ature range of 100 − 300 C; and (iii) dissociative CO
◦
28
adsorption in the temperature range of 600 − 1000 C .
Bulk Mo N displayed two CO desorption peaks at 165
2
◦
and 300 C. The first CO desorption peak was assigned to
desorption of molecularly adsorbed CO on Mo (Mo
and Mo ) sites (Mo-C ≡ O). The later CO desorption
peak was assigned to desorption of NCO species.29 In
this study, all the Mo-N/AC samples displayed molec-
ular CO desorption peaks with maxima in the range
−
late ( O-C=O) and carboxyl (-COOH) functionalities,
δ +
2+
26
respectively. Identicalcomponents, pronounceddueto
C 1s in Mo-N/AC samples (Figure S3), indicate that all
characteristic functional groups of C 1s remained intact,
even after Mo was impregnated on AC. However, the
changes observed in Mo-N/AC samples with respect to
Mo loading are as follows: (i) The intensity of C-C peak
was reduced and a prominent XPS tail (around 285.5
eV) appeared from AC to 15Mo-N/AC which indicates
possible interactions exist between surface Mo and AC
functional groups. (ii) In particular, relatively intensi-
fied -C-O and -C=O bands with respect to Mo loading
3+
◦
of 150–200 C with different CO desorption capacities.
28
According to Miura et al., molecular CO desorption
can be enhanced by the surface modification of Mo
species. The CO-TPD result reveals CO sorption on
+
δ
Mo sites present on AC in Mo-N/AC samples.
indicates the presence of the interaction between the 3.4b H -TPD studies: H -TPD results of Mo-N/AC
2
2
surface Mo atoms and these functional groups through samples are presented in Figure 2b. A single desorption
delocalization of electrons.
signal characterized by several humps were observed
◦
O 1s XPS pattern of AC and Mo-N/AC samples in the temperature range of 150–350 C in all Mo-
are presented in Figure S3. The Shirley deconvolution N/AC samples. Hydrogen desorption from Mo (100)
of O 1s yields four different components in the bind- surface was reported to happen at room temperature.27
ing energy range of 528-540 eV. The peak centered at It was accepted that the hump-like profile was the result
5
32.8 eV is attributed to carbonyl oxygen (C=O). The from either reversible adsorption of H and/or simul-
2
peaks registered at 534.0 and 535.2 eV are ascribed taneous H desorption from the Mo/AC surface. The
2
to hydroxyl oxygen (C-OH) and carboxylate/carboxyl maximum H desorption of 60 μmol/g was observed
2
oxygen, respectively.19 The peak centered at 530.8 eV at 10Mo-N/AC compared to 5- and 15Mo-N/AC sam-
suggests the presence of lattice oxygen in O-Mo species. ples. Hence, the decreasing order of H desorption is
2
Furthermore, FigureS4(SupplementaryInformation) as follows: 10Mo-N/AC (60 μmol/g) > 15Mo-N/AC
shows that the Mo 3d peaks are pronounced at 232.6 and (26 μmol/g) > 5Mo-N/AC (17μ mol/g). The result

5
94
Sharif F Zaman et al.
trostatic attractions between anionic part of one unit and
(a)
30
counter cationic part of the other unit. Mo content by
EDS analysis for 5-, 10- and 15 Mo-N/AC samples were
found to be 4.8, 9.5 and 14.0 wt%, respectively, which
agree with the nominal loading of the Mo metal at the
impregnation step.
The high resolution TEM images of 10Mo-N/AC
sample are presented in Figure S6 (Supplementary
Information). Well-distributed Mo particles (marked
with circles in Figure S6(a)) are observed in the wrin-
kled channels of AC. The Mo particles are sized in the
range of 1-2 nm. The d-spacing line profile of Mo par-
ticle is presented in Figure S6 (b). The d-spacing of
channels (0.144 nm) is consistent with the plane (062)
of orthorhombic molybdenum trioxide [00-005-0508].
These results are in agreement with XPS data where
1
5Mo-N/AC
1
0Mo-N/AC
5
Mo-N/AC
AC
1
00
150
200
250
300
350
Temperature °C
6+
Mo valence state is observed.
(b)
In present work, ammonia treated sample charac-
terization exhibited oxide form of Mo. This might be
associated with quick conversion of nanosized Mo-N
particles (≤4 nm) to oxide under X-ray or electron
beam exposure and or due to catalyst exposure to atmo-
spheric air during the characterization. For this reason,
we performed the catalytic activity test on ammonia-
5Mo-N/AC
gas-untreated MoO /AC catalyst for this reaction. In
3
the following section, activity results reveal that the cat-
alytic behavior of Mo-N/AC is different in fair amount,
in both conversion and selectivity, when compared with
10Mo-N/AC
those of MoO /AC catalyst.
3
15Mo-N/AC
3.6 Catalytic activity
The CO hydrogenation activity results on the prepared
catalysts are presented in Figure 3. The catalytic activ-
Figure 2. (a) CO-TPD results of AC and Mo-N/AC sam-
ples in terms of thermal conductivity signal (a. u.). (b)
H2-TPD results.
ity tests were performed in the temperature range of
◦
2
50–325 C, 7.0 MPa and 12000/h GHSV (reactor vol-
indicates that the CO was preferentially adsorbed on ume 0.5 mL). Catalytic activity of the support was
Mo species rather than hydrogen.
also studied under similar reaction conditions. Only,
◦
<
0.5% CO conversion was observed on AC at 325 C.
On the other hand, maximum CO conversion of 18.0%
3.5 TEM studies
was observed on 10Mo-N/AC compared to 5Mo-N/AC
◦
The TEM images of AC and Mo-N/AC samples are (9.5%) and 15Mo-N/AC (13.5%) at 325 C. The result
presented in Figure S5 (Supplementary Information). suggests inadequate amount of Mo active surface at 5%
The wrinkle-folded morphology was observed for AC Mo loaded sample, and at 15% Mo loading, aggre-
3
(
Figure S5(a)), which is attributed to the carbon in sp
gation of Mo species leads to low catalytic activity.
30
hybridization in C-C and or C-O links. The mor- Hence, 10% Mo loading was found to be optimal for AC
phology of the carbon changed upon Mo loading and for CO hydrogenation reaction. Further, CO conversion
appears as nebulous type. The Mo spherical particles was increased with an increase in the reaction temper-
(
marked with circle) are distributed on the AC with the ature. Hydrocarbon selectivity was increased with an
size range of 1–3 nm. The large sized (≈4 nm) Mo increase in the reaction temperature for all the Mo-N/AC
particles on AC were observed at 15% loading (Fig- catalysts. Higher temperature is favorable for the for-
ure S5(d)) which suggest aggregation of Mo particles. mation of hydrocarbons and CO via water gas shift
2
Usually, Mo species agglomerate because of the elec- reaction. The higher hydrocarbon formation can also be

Ammonia treated Mo/AC catalysts for CO hydrogenation
595
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
Total HC selec
C2 + HC selec
Conversion
Total oxygenates selec
C2 + oxygenates selec
5
Mo-N/AC
0
2
50
275
300
325
9
8
7
6
5
4
3
2
1
0
Total HC selec
C2 + HC selec
Conversion
Total oxygenates selec
C2 + oxygenates selec
0
0
0
0
0
0
0
0
10MoO
3
/AC
1
0Mo-N/AC
%
o
0
Temperature C
250
275
300
325
9
0
0
0
0
0
0
0
0
0
Total HC selec
C2 + HC selec
Conversion
Total oxygenates selec
C2 + oxygenates selec
8
7
6
5
4
3
2
1
1
5Mo-N/AC
0
250
275
300
325
Temperature oC
◦
Figure 3. CO hydrogenation activity on Mo/AC catalysts at reaction temperature 250–325 C, 7MPa and
GHSV 12000/h (reactor volume 0.5 mL).
related to the hydrogenation of carbonaceous intermedi- respect to reaction temperature suggests higher hydro-
2
9
ates(−C H )ratherCOinsertionintoit. Furthermore, genation tendency of these catalysts.
x
Y
the increased CO concentration at higher temperatures
On the other hand, the total oxygenates selectivity
2
can also alter the CO/H ratio in the reactor zone and decreased with increase in the reaction temperature.
2
influence the CO insertion process followed by higher The studied Mo-N/AC catalysts showed maximum
alcohol formation. In the present work, increased hydro- oxygenates selectivity, ∼50%, CO -free basis, mainly
2
◦
carbon selectivity with increased CO conversion with methanol in the liquid product, at 250 C, however, with

596
Sharif F Zaman et al.
◦
a low CO conversion (∼2.0%). At 300 C, a reasonable sented in Figure 4. It is obvious from the results that
CO conversion of 5.5, 8.6 and 6.7%, with a reason- methanol was the major alcohol produced on all the
◦
able total oxygenate selectivity of 30, 35 and 28% were studied Mo-N/AC catalysts at temperatures up to 300 C.
observed on 5-, 10- and 15 Mo-N/AC catalysts, respec- Further, methanol selectivity decreased with increase in
tively.
the reaction temperature. At the same time, improved
+
Catalysts, respectively. Further, oxygenates selectiv- C alcohol selectivity indicates that methanol homolo-
2
ity of 21% with 9.5% CO conversion was obtained on gation/ CO insertion to −C H intermediate to chain
x
y
5
Mo-N/AC, whereas, around 14% of both oxygenates propagation, was the additional reaction route leading
+
selectivity and CO conversion was observed on 15Mo- to C alcohol formation.
N/AC at 325 C. Around 16.5% of total oxygenates
2
◦
◦
This phenomenon was intensified at 325 C partic-
selectivity with 18% CO conversion was obtained on ularly on the catalysts having 10 and 15 wt% Mo.
◦
1
0Mo-N/AC at 325 C. Decrease in oxygenate selectiv- Interestingly, propanol was the second highest selec-
ity was compensated by the increase in hydrocarbons, tive compound in the alcohol distribution after methanol
+
◦
especially methane. Also C2 hydrocarbon selectiv- in all the cases, except on 10Mo-N/AC at 325 C. Fur-
ity increased with reaction temperature suggesting an ther, selectivity of propanol (39%) was greater than
increase in hydrogenation capacity of Mo-N/AC cata- that of methanol (34 %) and ethanol (13.5%) on10Mo-
◦
lyst at higher temperatures, i.e., hydrogenation of liquid N/AC at 325 C. Butanol selectivity for all the cases
oxygenates to produce the corresponding hydrocarbons. was less than 3%. It is obvious from the alcohol
The catalytic activity of 10MoO /AC (without distribution results that Mo-N/AC catalyst has some
3
ammonia treated catalyst) was also studied under simi- unique characteristics for showing high selectivity to
lar reaction conditions and the results are compared with C alcohol. Getting high selectivity to propanol is
3
1
9
0Mo-N/AC catalyst. Only 4% of CO conversion, with very interesting and a new phenomenon, as accord-
6% hydrocarbon and 4% oxygenate selectivity, was ing to Anderson-Schulz-Flory (ASF) distribution, the
◦
observed on 10MoO /AC at 325 C. The 10MoO /AC chain propagation should continue to high selectivity
3
3
catalyst has exhibited considerable amount of methanol to the bulky end product, i.e., butanol. Somehow, the
with very little amounts of ethanol, propanol and other chain propagation was ceased after formation of C3
oxygenates in the liquid product distribution. On the alcohol on Mo-N/AC catalysts. Further study is under
other hand, oxygenates selectivity was significantly progress to assess the effect of alkali metals on the cat-
◦
improved to 16.5% on 10Mo-N/AC catalyst at 325 C. alyst.
However, on10MoO /ACcatalystalsoatlowerreaction
temperature (250 C), maximum oxygenates selectivity in alcohol products together at 325 C decreased in the
The selectivity (CO free) of ethanol and propanol
3
2
◦
◦
of 30% with 0.5% CO conversion was observed. These following order: 10Mo-N/AC (52.5%) > 15Mo-N/AC
+
results suggest that improved oxygenates selectivity due (43.5%) > 5Mo-N/AC (33%). The greater C alcohol
2
δ +
to ammonia treatment might be associated with Mo
selectivity on 10Mo-N/AC was associated with higher
sites of Mo-N/AC catalysts. The results are in agreement adsorption capacity for CO and H . Miura et al., also
28
2
with our CO-TPD data where, molecular CO adsorption reported that CO hydrogenation selectivity depends on
δ +
take place on Mo sites.
CO chemisorption capacity of bulk MoO and molyb-
3
31
Li et al., reported 36% CO conversion with 99% dophosphoric acids. The other oxygenates selectivity
hydrocarbon selectivity on 18Mo/AC (not treated with also increased with increase in the reaction temperature
◦
gaseous form of ammonia) catalyst at 325 C, 5 MPa on these catalysts.
3
and GHSV 6 m /h/kg. Hydrocarbon formation, particu-
larly methane formation, in CO hydrogenation using Mo
based catalysts was reported in the following order: Mo
3
.8 Influence of support properties and Mo loading
on the oxygenates formation
0
δ +
4+
metal (Mo ) > Mo N (Mo ) > MoO ( Mo ) >
2
2
6+
32
MoO (Mo ). The decrease in the CO conversion on In the present study, the AC contains small portion
3
1
0MoO /AC catalyst is associated with the bulk pres- of surface acidic groups as evidenced by XPS/FTIR
3
ence of the higher oxidation state molybdenum oxide analysis. It is well-known that acidic supports such as
sites.
.7 Alcohol distribution
Al O and ZrO suppress the formation of alcohols
2 3 2
and increase the hydrocarbon formation on pre-reduced
Mo catalysts. The present DARCO AC contains mostly
3
3
sp hybridized carbon atoms and C-O (ethers and phe-
The results of alcohol distribution in the temperature nol) functional groups with mesopores, which provide
◦
range of 250–325 C on Mo-N/AC catalysts are pre- suitable interaction conditions between Mo and AC sur-

Ammonia treated Mo/AC catalysts for CO hydrogenation
597
◦
Figure 4. Alcohol distribution on Mo/AC catalysts in the temperature range of 250–325 C, 7 MPa and GHSV 12000/h
reactor volume 0.5 mL).
(
face. FTIR analysis of the spent AC and Mo-N/AC to the mesopores of AC during the CO hydrogena-
catalysts are presented in Figure S7 (a) (Supplemen- tion. It is obvious from the spent catalyst analysis that
◦
tary Information). The sample exhibited identical bands the employed DARCO AC is stable at 325 C and 7
as fresh catalysts in the finger print region of 400– MPa.
−
1
1
1
500 cm and in the functional group region (4000–
−1
500 cm ). In addition, new bands appeared at 2850
4
. Conclusions
−1
and 2920 cm for all the spent catalysts. They are
attributed to C-H vibration of hydrocarbons present Mo-N/AC catalyst showed interesting product selectiv-
on the surface of Mo/AC catalysts. The results of ity, especially to alcohols (C −C ), for CO hydrogena-
1
3
δ +
XPS measurements of the spent 10Mo-N/AC catalyst tion reaction. This is associated with Mo sites present
shows (Figure S7 (b)) the relatively increased C-O func- on AC as revealed by CO-TPD studies. Among the Mo-
tional group of AC over its fresh counterpart, which N/AC catalysts, 10Mo-N/AC was found to be the most
reveals that Mo mobilization from C-O-Mo complex promising catalyst for CO hydrogenation to hydrocar-

598
Sharif F Zaman et al.
bons and oxygenates due to high affinity for molecular
from CO2 Contained Syngas on Supported Molybdenum
Sulfide Catalysts Appl. Catal. A 150 243
. Iranmahboob J and Hill D 2002 Alcohol Synthesis from
Syngas over K2CO3/CoS/MoS2 on Activated Carbon
Catal. Lett. 78 49
0. Li Z R, Fu Y L, Jiang M, Hu T D, Liu T and Xie Y
N 2001 Active Carbon Supported Mo–K Catalysts Used
for Alcohol Synthesis J. Catal. 199 155
1. Ma X, Lin G and Zhang H 2006 Co-Mo-K Sulfide-Based
Catalyst Promoted by Multiwalled Carbon Nanotubes
for Higher Alcohol Synthesis from Syngas Chin. J.
Catal. 27 1019
2. Surisetty V R, Dalai A K and Kozinski J 2010 Alkali-
Promoted Trimetallic Co-Rh-Mo Sulfide Catalysts for
Higher Alcohols Synthesis from Synthesis Gas: Com-
parison of MWCNT and Activated Carbon Supports Ind.
Eng. Chem. Res. 49 6956
3. Liakakou E T, Heracleous E, Triantafyllids K S and
Lemonidou A A 2015 K-promoted NiMo Catalysts
Supported on Activated Carbon for the Hydrogenation
Reaction of CO to Higher Alcohols: Effect of Support
and Active Metal Appl. Catal. B: Environ. 165 296
4. Ma C-H, Li H-Y and Lin G-D 2010 MWCNT-Supported
Ni–Mo–K Catalyst for Higher Alcohol Synthesis from
Syngas Catal. Lett. 137 171
5. Wang J-J, Xie J-R, Huang Y-H, Chen B-H, Lin G-D and
Zhang H-B 2013 An Efficient Ni–Mo–K Sulfide Catalyst
Doped with CNTs for Conversion of Syngas to Ethanol
and Higher Alcohols Appl. Catal. A: Gen. 468 44
6. Shao Y, Xu Z, Wan H, Chen H, Zheng S and Zhu D
2011 Enhanced Liquid Phase Catalytic Hydrodechlori-
nation of 2,4-dichlorophenol over Mesoporous Carbon
Supported Pd Catalysts Catal. Commun. 12 1405
7. Zhang Y, Wang A and Zhang T 2010 A New 3D Meso-
porous Carbon Replicated from Commercial Silica as a
Catalyst Support for Direct Conversion of Cellulose into
Ethylene Glycol Chem. Commun. 46 862
8. Hussain M and Ihm S-K 2009 Synthesis, Charac-
terization, and Hydrodesulfurization Activity of New
Mesoporous Carbon Supported Transition Metal Sulfide
Catalysts Ind. Eng. Chem. Res. 48 698
9. Strzemiecka B, Voelkel A, Donate-Robles J and Martin-
Martinez J M 2014 Assessment of the Surface Chemistry
of Carbon Blacks by TGA-MS, XPS and Inverse Gas
Chromatography Using Statistical Chemometric Analy-
sis Appl. Surf. Sci. 316 315
0. Sing K S W, Everett D H, Haul R A W, Moscou L,
Pierotti R A, Rouquerol J and Siemieniewska T 1985
Reporting Physorption Data for Gas/Solid Systems with
Special Reference to the Determination of Surface Area
and Porosity Pure Appl. Chem. 57 603
1. Ghampson I T, Sepulveda C, Garcia R, Radovic L
R, Fierro J L G, DeSisto W J and Escalona N
CO adsorption and insertion into −C H . High selectiv-
x
y
9
ity to propanol, 39% over 34% for methanol, in alcohol
distribution highlights the unique catalytic behavior of
◦
1
0Mo-N/AC for CO hydrogenation reaction at 325 C.
1
The 10MoO /AC catalyst, with no treatment of
3
ammonia, showed only 4% of CO conversion and
6% hydrocarbon selectivity. These results suggest
1
9
that improved oxygenates selectivity is associated with
δ +
ammonia treatment which led to Mo sites. Our group
is performing experiments to assess the effect of alkali
metal on 10Mo-N/AC catalyst for CO hydrogenation
reaction.
1
Supplementary Information (SI)
1
The XRD, FTIR, XPS, TEM, HR-TEM of fresh catalyst, Fig-
ures S1–S6 and FTIR, XPS of spent catalyst, Figures S7 are
presented in Supplementary Information, available at www.
ias.ac.in/chemsci.
1
1
Acknowledgements
This project was funded by the National Plan for Science,
Technology and Innovation (MAARIFAH)-King Abdulaziz
City for Science and Technology, the Kingdom of Saudi Ara-
bia, (Award Number 12-PET2720-03). The authors acknowl-
edge technical support with thanks Science and Technology
Unit, King Abdulaziz University.
1
1
1
1
References
1
. Zaman S and Smith K J 2012 A Review of Molybdenum
Catalysts for Synthesis Gas Conversion to Alcohols: Cat-
alysts, Mechanisms and Kinetics Catal. Rev. Sci. Eng. 54
41
2
. Paris R S, Montes V, Boutonnet M and Jaras S
015 Higher Alcohol Synthesis over Nickel-Modified
2
Alkali-doped Molybdenum Sulfide Catalysts Prepared
by Conventional Coprecipitation and Coprecipitation in
Microemulsions Catal. Today 258 294
3
. Iranmahboob J, Toghiani H and Hill D 2003 Dispersion
of Alkali on the Surface of Co-MoS2/Clay Catalyst: A
Comparison of K and Cs as a Promoter for Synthesis of
Alcohol Appl. Catal. A: Gen. 247 207
2
4
5
6
. Liakakou E T and Heracleous E 2016 Transition Metal
Promoted K/Mo2C as Efficient Catalysts for CO Hydro-
genation to Higher Alcohols Catal. Sci Technol. 6 1106
. Lee J H, Hariprasad Reddy K, Jung J S and Yang E -H
2
2
012 Hydrodeoxygenation of Guaiacol over Carbon-
2
014 Role of Support on Higher Alcohol Synthesis from
supported Molybdenum nitride Catalysts: Effects of
Nitriding Methods and Support Properties Appl. Catal.
A: Gen. 439–440 111
Syngas Appl. Catal. A: Gen. 480 128
. Surisetty V R, Dalai A K and Kozinski J 2011 Alcohols
as Alternative Fuels: An Overview Appl. Catal. A: Gen.
2
2. Sepulveda C, Leiva K, Garcia R, Radovic L R, Ghamp-
404 1
son I T, DeSisto W J, Fierro J L G and Escalona N 2011
7
. Duchet J C, Vanoers E M, Debeer V H J and Prins R 1983
Carbon-supported Sulfide Catalysts J. Catal. 80 386
. Gang L, Chengfang Z, Yanqing C, Zhibin Z, Yianhui N,
Linjun C and Fong Y 1997 Synthesis of Mixed Alcohols
Hydrodeoxygenation of 2-methoxyphenol over Mo N
2
Catalysts Supported on Activated Carbons Catal. Today
8
172 232

Ammonia treated Mo/AC catalysts for CO hydrogenation
599
2
3. de la Puente G, Centeno A, Gil A and Grange P 1998 29. Zaman S F, Pasupulety N, Al-Zahrani A A, Daous M
Interactions between Molybdenum and Activated Car-
bons on the Preparation of Activated Carbon-Supported
Molybdenum Catalysts J. Colloid Interface Sci. 202 155
4. Shaheen W M and Selim M M 2002 Thermal Behaviour
A, Al-Shahrani S S, Driss H, Petrov L A and Smith K
J 2017 Carbon Monoxide Hydrogenation on Potassium
Promoted Mo2N Catalysts Applied Catalysis A: Gen.
532 133
2
2
2
of Pure and Binary Basic Nickel Carbonate and Ammo- 30. Kumar P, Mungse H P, Cordier S, Boukherroub R, Khatri
nium Molybdate Systems Mater. Lett. 52 272
5. Vissers J P R, Bouwens S M A M, de Beer V H J and Prins
R 1987 Carbon Black-supported Molybdenum Sulfide
Catalysts Carbon 25 485
O P and Jain S L 2015 Hexamolybdenum Clusters Sup-
ported on Graphene Oxide: Visible-light Induced Pho-
tocatalytic Reduction of Carbon Dioxide into Methanol
Carbon 94 91
6. Tien H -W, Huang Y -L, Yang S -Y, Hsiao S -T, Wang 31. Li X, Feng L, Zhang L, Dadyburjour D B and Kugler
J -U, Ma C -C M 2011 Graphene Nanosheets Deposited
on Polyurethane Films by Self-assembly for Preparing
Transparent, Conductive Films J. Mater. Chem. 21 14876
E D 2003 Alcohol Synthesis over Pre-Reduced Acti-
vated Carbon-Supported Molybdenum-Based Catalysts
Molecules 8 13
2
7. Lee J S, Lee K H and Lee J Y 1992 Selective Chemisorp- 32. Saito M and Anderson R B 1980 The Activity of Several
tion of Carbon Monoxide and Hydrogen over Supported
Molybdenum Carbide Catalysts J. Phys. Chem. 96 362
8. Miura H, Osawa M, Sugiyuma K and Matsuda T 1986
The Adsorption of CO and Hydrogen on Reduced
Molybdenum Catalysts Used for CO Hydrogenation J.
Catal. 101 178
Molybdenum Compounds for the Methanation of CO J.
Catal. 63 438
2