Carbon monoxide hydrogenation on potassium promoted Mo2N catalysts

Article abstract of DOI:10.1016/j.apcata.2016.12.015

The catalytic performance of Mo₂N and K-Mo₂N were assessed in hydrogenation of carbon monoxide in the reaction temperature range of 275–325 °C, 7 MPa and 60,000 h^?1. The nitrides were synthesized via temperature-programmed

Full text of DOI:10.1016/j.apcata.2016.12.015

Accepted Manuscript
Title: Carbon monoxide hydrogenation on potassium
promoted Mo N catalysts
2
Author: Sharif F. Zaman Nagaraju Pasupulety Abdulrahim A.
Al-Zahrani Muhammad A. Daous Saad S. Al-Shahrani
Hafedh Driss Lachezar A. Petrov Kevin J. Smith
PII:
S0926-860X(16)30620-2
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2016.12.015
APCATA 16100
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
12-10-2016
12-12-2016
19-12-2016
Please cite this article as: Sharif F.Zaman, Nagaraju Pasupulety, Abdulrahim A.Al-
Zahrani, Muhammad A.Daous, Saad S.Al-Shahrani, Hafedh Driss, Lachezar A.Petrov,
Kevin J.Smith, Carbon monoxide hydrogenation on potassium promoted Mo2N
catalysts, Applied Catalysis A, General http://dx.doi.org/10.1016/j.apcata.2016.12.015
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

K2MoO4
MoO
2
MoO
2
K-O
3
.16 Å,
K-O
Chemical interaction between
K-O located on MoO
2
in 1.3K-Mo N
2
MoO
2
and K-O in 3K-Mo N
2
K MoO formation in 6.2K-Mo N
2
4
2
2
2
1
1
5
0
5
0
5
0
1
.3K-Mo N
2
3
K-Mo N
2
0
.45K-Mo N
2
6
.2K-Mo N
2

-Mo N
2
0
.00
0.03
0.06
0.09
0.12
0.15
0.18
1.9
K/(Mo+2 + Mo+3 ) ratio
1

Highlights
2
CO hydrogenation to oxygenates was studied on Mo N for the first time.
CO conversion was decreased on K-Mo
2
N compared to unpromoted Mo
2
N.
Unpromoted Mo
2
N showed higher hydrocarbon selectivity.
+
Improved C
2
oxygenates selectivity was observed on 1.3K-Mo
2
N catalyst.
Well distributed K in the matrix of γ-Mo
2
N improved the molecular CO adsorption.
2

Carbon monoxide hydrogenation on potassium promoted Mo N catalysts
2
a, b,
b,
a,b
†
∗
Sharif F. Zaman
, Nagaraju Pasupulety , Abdulrahim A. Al-Zahrani , Muhammad A.
Daous , Saad S. Al-Shahrani , Hafedh Driss , Lachezar A. Petrov and Kevin J. Smith
a,b
a
a
b
c
a
Chemical and Materials Engineering Department, Faculty of Engineering, King Abdulaziz University,
P.O. Box 80204, Jeddah 21589, Saudi Arabia
Center of Excellence in Catalysis, King Abdulaziz University, P.O. Box 80204, Jeddah 21589, Saudi
b
Arabia
c
Department of Chemical and Biological Engineering, University of British Columbia, Vancouver, BC,
Canada
Abstract
2 2
The catalytic performance of Mo N and K-Mo N were assessed in hydrogenation of
o
-1
carbon monoxide in the reaction temperature range of 275-325 C, 7MPa and 60,000 h . The
nitrides were synthesized via temperature-programmed treatment of ammonium heptamolybdate
(AHM) and K-AHM precursors under continuous NH
3
flow. The influence of potassium loading
(0.45 to 6.2 wt.%) on the Mo nitride phase and its resultant effect on the catalytic properties were
been investigated. The nitride catalysts were characterized in terms of BET/pore size, powder X-
ray diffraction, DRIFTs, CO-TPD-mass, HR-TEM and XPS measurements. The formation of
cubic γ-Mo
2
N and monoclinic K
2
MoO
4
phases were confirmed by XRD and TEM. Platelet
N. CO-TPD-mass
morphology with particle size in the range of 5-10 nm was observed for γ-Mo
2
results mainly revealed molecular CO sorption on Mo
2
N and K-Mo
2
N surface. The highest total
o
δ+
oxygenate selectivity (44%, 300 C) was observed on 1.3K-Mo
2
N with a K/Mo surface ratio of
0
.06. This is associated with well distributed K in the matrix of γ-Mo
reduced due to K MoO formation at 3 and 6.2 wt.% K loading. DRIFTs results clearly
demonstrates greater hydrocarbon formation on unpromoted Mo N due to (i) CO dissociative
hydrogenation (ii) water-gas shift reaction. Further, addition of K to Mo N significantly
2
N. Distribution of K was
2
4
2
2
improved the oxygenate selectivity by promoting the molecular CO insertion into –CHx
intermediate.
2
Key words: Mo N catalysts, alkali metal, CO hydrogenation and higher oxygenates.
†
Corresponding author. Tel.: +966 563063594; Fax : 966-2-6952257
E-mail address: zfsharif@gmail.com; sfzaman@kau.edu.sa
∗
Co-corresponding author. Tel.: +966 531596412
E-mail address: pasupulety@gmail.com; nsampathrao@kau.edu.sa
3

1
. Introduction
Mixed alcohol synthesis from thermochemical conversion of syngas (synthesis gas:
CO+H ) is an important process for the production of oxygenate fuels and fuel additives.
2
Majorly syngas is produced from the gasification of feedstocks such as coal, natural gas,
biomass, shale oil and tar sands. Most important commercialized processes utilizing syngas are
the production of synthetic gasoline via Fischer-Tropsch synthesis (FTs) reaction using iron (Fe)
or cobalt (Co) based catalysts (eq. 1) [1-3], methanol synthesis using copper-zinc oxide (Cu-
ZnO) based catalyst (eq. 2) [1,4]. Heterogeneous catalysts employed for the synthesis of ethanol
and higher alcohols can be classified into four categories: (i) Modified methanol synthesis
catalyst, (a) High temperature, Zn-Cr based catalyst (b) Low temperature, Cu-Zn based catalyst
(
ii) Modified FTs based catalyst (iii) Rhodium based catalyst (iv) Molybdenum based catalyst.
Recently, Tan et al. [5] published isobutanol synthesis from syngas using alkali metal modified
Cr/ZnO catalysts. Rh based catalysts show high selectivity to ethanol derived from syngas [6].
Cost and availability of Rh are the main barriers to the adaptation of this catalyst for commercial
operation. Among other alternative higher alcohol synthesis catalysts, modified methanol
synthesis catalysts selectively produce methanol and isobutanol, with low selectivity to ethanol
[
7]. Modified FTs based, Cu-Co catalysts show high yield to ethanol and higher alcohols, but
they produce considerable amounts of hydrocarbons [8]. Among the molybdenum based
catalysts K-MoS shows very promising result for ethanol synthesis (40 C atom% selective to
ethanol on CO
2
2
free basis) [9]. Recently Zaman and Smith [1, 10, 11] reported MoP based
+
catalysts for higher alcohol synthesis, which also shows very interesting characteristics for C
2
oxygenate selectivity [43 C atom%]. β-molybdenum carbide catalyst also showed high activity
for this reaction but suffers from high hydrocarbon selectivity [12].
1
2
. FT type reaction
CO+2H  -CH
CO + 3H  CH
CO + H O  CO
-
1
1
2
2
- + H
+H
+ H
2
O
ΔH298K = -165 kJ mol
ΔH298K = -206 kJ mol
eq. (1)
eq. (2)
-
2
4
2
O
-1
2
2
2
ΔH298K = -41.1 kJ mol
. CO insertion reaction
CO + 2H
CH OH + CO + 2H
OH + CO + 2H
2
-
1
2
 CH
 C
 C
3
OH
ΔH298K = -90.5 kJ mol
O ΔH298K = -165.6 kJ mol
O ΔH298K = -155.0 kJ mol
-
-
1
1
3
2
2
H
H
5
OH + H
OH + H
2
C
2
H
5
2
7
2
4

Recently, Hargreaves et al. [13, 14] reported the usage of metal nitrides for
heterogeneous catalysis applications. The use of transition metal nitrides as hydrotreating
catalysts are also reported [15-17]. The group VI transition metal nitrides exhibit comparable
activity to that obtained with Pd and Pt based precious metal catalysts [16, 18] in hydrogen
mediated reactions because of the contraction of the Mo d-band and modification of electron
density due to the interstitial incorporation of nitrogen in the metal lattice [19]. A recent DFT
study by Zaman [20] for CO adsorption and dissociation over γ-Mo
CO dissociation has an activation barrier close to that of MoS (1001) plane, and concluded that it
may show similar activity as of MoS catalyst for syngas conversion.
In this study, we have established the feasibility of Mo N and K-Mo
2
N(111) plane reveals that
2
2
2
2
N as an active
catalytic material for the CO hydrogenation to higher hydrocarbons and oxygenates. Until this
moment, we were unable to find any study in the open literature dealing with the hydrogenation
of CO to oxygenates and or to higher alcohols on potassium promoted molybdenum nitride. In
this report, for the first time, we also present the drifts results on Mo
for CO hydrogenation reactions.
2 2
N and 1.3K-Mo N catalysts
2
2
. Experimental
.1. Catalyst preparation
Materials: Ammonium heptamolybdate ((NH₄)₆Mo₇O₂₄·4H₂O, Fluka ≥99.9%), anhydrous
citricacid (C ≥99%, Techno Pharmchem, Haryana, India) and potassium nitrate (KNO
99%, Sigma-Aldrich) were obtained from their designated sources and used without any further
purification.
6
H
8
O
7
3
≥
For example, for the preparation of sample containing molybdenum nitride with 5 wt. %
of nominal potassium loading requires 5.2 g of ammonium heptamolybdate, 5.7g of citric acid
3
(CA) and 0.39g of KNO
3
, which were dissolved in 100 cm of deionized water under continuous
o
stirring. The temperature of this solution maintained at 50 C for 2 h for homogenization. Then
the water was evaporated quickly by means of hot plate. The obtained pale yellow gel was
o
cooked for 24 h using a water bath with temperature at 90 C. The resultant foam like solid
material was crushed to powder. Identical synthesis procedure was followed for unpromoted (no
potassium containing sample), as well as for samples containing 1, 10 and 15 wt. % nominal
potassium loading. For all the synthesized samples, molybdenum to citric acid mole ratio was
5

o
maintained at 1:1. All powdered samples were calcined at 500 C for 4h under static air
conditions. The calcined samples are named as Mo-CA, 1K-Mo-CA, 5K-Mo-CA, 10K-Mo-CA
and 15K-Mo-CA. Here, the prefix 1, 5, 10 and 15 represents the nominal K loading wt.%.
All the calcined samples were subjected to stepwise temperature programmed ammonia
treatment. One gram of the calcined sample was placed in a quartz reactor fitted with quartz frit.
The nitridation procedure is as follows:
o
(
i) In the first step, catalyst was kept at 120 C for 30 min. under continuous He flow.
(ii) In the second step, gaseous ammonia was introduced to the catalyst bed with an approximate
3
-1
o
flow rate of 400 cm min and the reactor temperature was increased from 120 to 350 C with a
o
-1
ramping rate of 1 C.min .
iii) In the third step, the catalyst temperature was increased from 350 to 700 C with a ramping
o
(
o
-1
rate of 0.5 C.min and aged at this temperature for 2 h under continuous flow of ammonia.
(iv) In the fourth step, the reactor was cooled down to room temperature under He flow rate of
3
-1
3
0 cm min .
2
(v) In the fifth step, the sample was passivated with flow of He containing 1 vol. % O at a flow
3
-1
rate of 30 cm min for 1 h.
Finally, the catalyst was preserved in a glass bottle and unless otherwise stated, used for
characterization studies and for catalytic activity tests.
Potassium content in passivated K-Mo
suggest that potassium loss took place during the nitridation process (Table 1). The passivated
catalysts are denoted as Mo N, 0.45K-Mo N, 1.3K-Mo N, 3K-Mo N and 6.2K-Mo N. Here, the
prefix 0.45, 1.3, 3 and 6.2 represents the actual weight percentage of potassium presented in
Mo N.
2
N samples were analyzed by XRF and the results
2
2
2
2
2
2
2
.2. Catalyst Characterization
The BET surface area and pore size analysis (using DFT method) of nitride samples were
determined by Nova Station Quantachrome (USA) adsorption equipment at liquid nitrogen
o
temperature after out gassing the samples at 200 C under vacuum for 2 h. FTIR-ATR of the
powdered samples were analyzed by using a Tensor-II spectrometer with ATR facility and
OPUS software.
6

The X-ray diffractograms of the nitride samples were obtained on a EQUINOX 1000 Inel
XRD instrument using Co K =1.7902Å with X-ray source generator settings at 40kV and 30mA
α
with real time acquisition over 2θ of 110 degree. X-ray fluorescence (XRF) was performed using
an Amptek XRF kits constituted of X-ray generator (mini-x) and X-ray detector (X-123SDD).
The system was fitted with a silver source and operated at 50 kV and 30 mA. The data analyzed
using the software XRF-TFR V1.33.
DRIFTs experiments were performed using a Bruker Tensor-II FTIR spectrometer
equipped with a DLATGS detector at room temperature, Harrick praying mantis, high
temperature and high pressure (with 4mm thick ZnS windows (≥ 6 MPa)) reaction chamber.
o
Approximately, 0.1g of catalyst was loaded and reduced at 450 C for 4h under the flow of H
2
-1
o
(
20 ml min ). Subsequently, the catalyst temperature was brought back to 300 C under the flow
of He and flush the reaction chamber for 1h. Then the reaction mixture (CO/H 1:1) gas was
2
introduced into the chamber with an applied pressure of 5 MPa and collected the sample data.
Transmission electron microscopy (TEM), High resolution TEM images and energy
2 2
dispersive x-ray analysis (EDS) of Mo N and K-Mo N samples were collected on a Technai 200
kV D1234 Super Twin microscope with camera length of 97 cm. Selected area electron
diffractogram (SAED) analysis was performed using Diffraction ring profiler 1.7.
2 2
XPS results of Mo N and K containing Mo N samples was collected on SPECS GmbH
high vacuum multi-technique surface analysis system equipped with Mg Kα 1253.6 eV X-ray
source. The reported binding energy values are based on C 1s 284.8 eV.
Temperature programmed desorption (TPD) analysis of prepared samples was performed
by adsorption-desorption of CO using Quantachrome Pulsar automated chemisorption
instrument. The certified gas mixture of 14.95 vol. % CO in He and pure H
2
of 99.99% were
supplied by Abdulla Hashim Gas (AHG), Saudi Arabia. In a typical experiment, approximately
3
−1
o
0
.1 g of nitride sample was pretreated under helium gas (10 cm min ) at 200 C for 1 h
o
followed by hydrogen gas treatment at 450 C for 4 h. Subsequently the catalyst was brought to
room temperature and saturated with probe gas CO. Desorption of probe gas was performed in a
o
o
-1
temperature range of 30-400 C at a ramping rate of 10 C min after 1 h of helium flushing.
Mass analysis was performed using ThermoStarTM GSD 320 quad core mass spectrometer
(Pfeiffer, Germany). The typical mass fragment inputs are as follows: m/z = 28 (CO) and m/z =
7

4
4 (CO
2
). Desorption stream was analyzed simultaneously using a TCD and a mass detector by
means of an automated split valve.
2
.3. Catalytic activity tests
Catalytic activity tests were performed using a PID Eng and Tech system (Spain)
equipped with a flow micro reactor, temperature, pressure and mass flow controllers. The
-1
2
mixture of CO: H = 1:1 volume ratio was selected as model feed at 60,000 h GHSV. Catalytic
o
o
o
activity tests were carried out in the temperature range of 275 - 325 C (±1 C) with 25 C step
size at an applied pressure of 7 MPa (± 0.2 MPa). For example, 100 mg of nitride catalyst was
loaded in to a copper lined stainless steel tube reactor (inside diameter 6.35mm and length
3
-
9
1
.53mm) diluted with equal amount of quartz beads was treated with H
2
flow rate of 50 cm min
and reactor temperature was
2
= 1:1 volume ratio was introduced with
o
at 450 C for 4h. Subsequently, the H
2
flow was replaced by N
2
o
brought back to 275 C. The reactant gas mixture CO: H
3
-1
a flow rate of 100 cm min through an activated carbon trap (traps iron carbonyls formed inside
the gas tank) into the reactor to attain the required pressure. After attaining the steady state (2h)
at a set temperature and pressure of the reaction the reactor exit stream directed through a heated
o
line at 160 C to Agilent 7890A GC equipped with TCD, connected with Restek Plot Q column
2
for CO and CO separation, and to FID, fitted with HP Plot Q column for separation of
hydrocarbons (methane, ethane, ethylene, propane, propylene and butane compounds) and
oxygenates (methanol, ethanol, propanol, butanol, acetone, acetic acid, acetaldehyde and ethyl
acetate) estimation. Same catalyst load was used for all the three temperatures. At each
temperature five data points were collected. The empty copper lined reactor showed negligible
CO conversion (≤0.1%) under identical reaction conditions. The reported results are within the
error range of ± 0.1%. The syngas conversion and selectivity of the products [10] were
calculated according to the following equations:
CO initial moles
CO final moles
CO conversion (mass %)
Selectivity (mass %)
(3)
=
X 100
(4)
CO initial moles
i i
n C product
X 100
=
Σ ni
C
i
8

th
where, n
i
is the number of carbon atoms contained in the molecule of the i component and C
i
is
th
the mole fraction of i component in the reaction mixture at the reactor exit.
3
3
. Results and discussion
.1. X-ray powder diffraction analysis
o
Powder X-ray diffraction patterns of calcined (550 C, 4 h under static air) samples are
presented in Fig. 1a. The calcined samples principally shows X-ray diffraction patterns at 2θ
equal to 15.0, 27.4, 30.0, 32.0, 34.6, 38.8, 39.6, 45.6, 46.5, 53.6, 54.2, 57.9, 58.8, 62.3, 63.9,
6
5.0, 66.7, 68.0 and 69.5°. These peaks are consistent with the presence of orthorhombic MoO
phase [PDF 00-005-0508]. Identical X-ray signals are pronounced due to the presence
orthorhombic MoO up to 5K-Mo-CA predominantly. Polycrystalline phases such as, K
PDF 01-086-243] and K 33 [21] are observed in 10K-Mo-CA and 15K-Mo-CA samples
Fig. 1a, green and magenta lines) which are resulted due to the chemical interaction between K
phase in 15K-Mo-CA is reduced significantly compared to other calcined
3
3
3 22
Mo14O
[
(
6
Mo10O
and Mo-O. The MoO
3
samples. Further, the unindexed X-ray reflections at 2θ equal to 39.0, 39.5, 40.0, 44.0 and 50.0°
in 10K-Mo-CA and 15K-Mo-CA samples are probably due to the presence of other
polycrystalline K-Mo-O phases.
Powder X-ray diffraction patterns of passivated Mo
2
N and K-Mo
2
N samples are
presented in Fig.1b. The -NH
x
species which are formed during the ammonolysis process above
o
5
00 C, are adsorbed on the coordinatively unsaturated sites existing on the surface of Mo
2
N
[
22]. A rapid autothermal oxidation of adsorbed -NH
To prevent the spontaneous reaction a careful passivation by oxygen provides a protective layer
on Mo N surface by eliminating most of the bonded -NH species [22]. The diffraction pattern of
Fig. 1b, black line) unpromoted Mo N showed X-ray reflections at 2θ equal to 43.7, 50.8, 75.0,
0.8 and 96.0° that can be associated, respectively, with (111), (200), (220), (311) and (222)
N crystallographic phase [PDF 00-025-1366]. Identical X-ray patterns are
observed up to 1.3K-Mo N due to the presence of cubic γ- Mo N phase. Along with the cubic γ-
Mo N phase, unpromoted Mo N sample showed X-ray patterns at 2θ equal to 30.0, 62.5, 71.5
and 79.0° with respective planes (-111), (211), (310) and (-402), which are consistent with
monoclinic MoO phase [PDF 00-078-1070]. The formation of the partially reduced
molybdenum oxide, such as MoO during the reduction-nitridation process was also reported in
reference [23, 24]. Nagai et al. [25] also recorded the formation of MoO
x
species take place upon exposure to air.
2
x
(
2
9
planes of cubic γ-Mo
2
2
2
2
2
2
2
2
, γ-Mo
2
N, β-Mo
2
N and
9

o
metallic Mo during the treatment of MoO
phases related to potassium oxide is detected in the prepared nitrides for K concentration below
.3 wt.% (1.3K-Mo N). This observation might be associated with the presence of highly
dispersed potassium crystallites, sized less than 4nm, which are below the detection limit of X-
ray diffraction method. The 3K-Mo N sample shows the presence of cubic γ-Mo N, hexagonal δ-
MoN and monoclinic K MoO phases. At 6.2 wt. % of potassium content the sample primarily
displays δ-MoN, K MoO phases along with minor portion of cubic γ-Mo N. It is, however,
worth noting that with the increase of K loading in Mo N, the amount of the MoO phase was
3 3
with NH at temperatures up to 900 C. No crystallite
1
2
2
2
2
4
2
4
2
2
2
decreased due to the chemical interaction between K and Mo-O, resulted in formation of the K-
Mo-O crystals. The observed crystallographic changes from the calcined to nitridation-
passivation samples can be summarized as follows:
(
i) Orthorhombic MoO
phase.
ii) Hexagonal δ-MoN is an intermediate phase to produce cubic γ- Mo
from MoO nitridation-passivation [13]. In the present work, molybdenum chemically interacts
with K at 6.2 wt.% and eventually yields K MoO . Due to this, hexagonal δ-MoN to cubic γ-
Mo N phase transformation was limited at this loading.
iii) Prominent K MoO phase appeared due to decomposition of the polycrystalline K-Mo-O
precursors and/or potassium oxide interaction with MoO
iv) Molybdenum nitride structure was intact even after K addition.
3 2
precursor upon nitridation-passivation primarily yields cubic γ- Mo N
(
2 3
N from MoO and or
2
2
4
2
(
2
4
2
.
(
3
.2. BET surface area and pore size analysis
All registered isotherms, shown in Fig.2a, are type IV with H4 hysteresis loop. The type
H4 loop is often associated with narrow slit-like pores [26]. Fig.2b shows the tri-modal pore size
distribution for Mo N. The micropores have an average pore diameter below 20 Å while the
mesopores have an average pore diameter between 22 and 55 Å. The mesopores of Mo
themselves have bimodal distribution in the range of 22-35 Å and 35-55 Å. Upon the
incorporation of K into Mo N the number of micropores was significantly reduced and new
mesopores were formed with an average pore diameter in the range of 22-40 Å. It is noteworthy
that, no micropores were found in 6.2K-Mo N sample. The pore volume and BET surface area
2
2
N
2
2
1
0

data for prepared samples are presented in Table 1. The BET surface area decreased significantly
2
-1
2
-1
2
from 110 m g to 10 m g after the incorporation of 0.45wt. % of K into Mo N.
Citric acid was used as a chelating component for promoting the BET surface area of
tungsten phosphide (WP) [27] and molybdenum phosphide (MoP) [28] catalysts. It is playing a
role of chelating/bridging agent between the metal ions. In this study, the obtained high surface
2
-1
area of 110 m g for Mo
formation of molybdenum citrate [29]. However, the BET surface area decreased with K
addition to citric acid induced synthesis of Mo N catalysts. Lower BET surface area was
reported for potassium promoted MoS with mesopore texture compared to the parent MoS
having micro and mesopore texture. The authors suggested that the decrease in the BET surface
area for K-Mo S was probably a result of the intercalated K-water species in between the MoS
slabs [30]. Lower BET surface area was also reported for 10 wt.% K supported on Mo C and it
was attributed to Mo C pore blockage by K CO [31].
2
N, in the presence of cheating agent citric acid, was associated with the
2
2
2
2
2
2
2
2
3
In this work, the significant decrease in the BET surface area for K promoted samples
might be associated with melting of K and or K/Mo salts to some extent during ammonolysis at
3
elevated temperatures, which might led to pore blockage. Further, MoO -potassium oxide binary
o
phase diagram shows, at around 450-600 C, the existence of liquid K-phase composed of
different potassium oxides KO
.3. XPS analysis
K 2p, Mo 3d and N KLL (Auger) orbital XPS analysis of K-Mo
2 2 2 2
, K O and K O [32].
3
2
N passivated samples are
presented in Fig. 3A, 3B and 3C respectively. The corresponding surface composition data are
summarized in Table 2. K 2p region is composed of a doublet with contributions at 292.5 and
2
95.5 eV which are, respectively, assigned to K 2p3/2 and K 2p1/2 lines [33]. These XPS signals
are exhibited due to the presence of K O or K MoO species on the surface of K-Mo N samples.
2
2
4
2
However, at 6.2wt. % K loading, the sample shows an additional shoulder at 291 eV, which
might be associated with the presence of impurity of carbonates coming from the potassium
nitrate precursor. Similar XPS results were reported by Galvez et al. [34] for Fe-K/Al
samples.
2 3
O
The deconvoluted Mo 3d XPS result (Fig. 3B) shows several oxidation states ranging from
6
+
2+
2
Mo to Mo . After deconvolution of unpromoted Mo N, Mo 3d5/2 and 3d3/2 transition pairs
appeared at binding energies of 228.4 and 231.6, 229.2 and 232.3, 230.2 and 233.7, 232.5 and
1
1

2+
3+
4+
2
35.4 eV. The Mo 3d 5/2 signal attributed to Mo at 228.6 eV, Mo at 229.2 eV, Mo at 230.2
6
+
eV and Mo at (232.5 eV) species [29]. Identical signals of Mo 3d5/2 were pronounced for
samples containing up to 1.3 wt.% K due to the presence of similar Mo species. The above
signals marginally shifted to lower binding energies by 0.4 eV for 3K-Mo
2
N sample. Which
N, Mo 3d
0
suggests Mo 2+ and 3+ reduction towards Mo started at this loading. For 6.2K-Mo
2
XPS pattern was completely changed compared to unpromoted Mo
2
2
N and other K-Mo N
4
+
6+
2+
samples. Interestingly, Mo and Mo species were dominating over Mo species and a new
shoulder appeared at 227.0 eV might be associated with the presence of metallic Mo species [35]
3
+
for 6.2K-Mo
Hence, at higher K loading metallic Mo formation suggests the reduction of Mo and or Mo
species. This might be associated with ease of reduction of low valence Mo species (δ-MoN)
under nitridation conditions. Among the studied K-Mo N samples, Mo 3d distributions (Table 2)
2
N sample. It is noteworthy that, no Mo species were detected at this K loading.
3
+
2+
2
4
+
varied significantly. In the present study, the Mo valence species are related to MoO
2
and or
phase
6+
K
2
MoO
3
phase and Mo valence species might be associated with MoO
3
and or K
2
MoO
4
in agreement with XRD data.
2
XPS induced Auger spectra of the passivated K-Mo N samples are presented in Fig.3C.
The major Auger signal appeared at 383 eV with a shoulder at 389 eV. However, these signals
shifted to higher kinetic energy and the major signal appeared at 386 eV with a shoulder at 391
eV for 6.2K-Mo
molybdenum nitride samples. The transition at 385 was attributed to KL2,3
nitrogen atom in the nitride and the later transition at 392 eV was due to L level of nitrogen to
4,5 level of molybdenum. Hence, the higher kinetic energy transitions are likely due to the
nitrogen chemical environment of hexagonal δ-MoN and the lower energy transitions related to
nitrogen chemical environment of cubic γ-Mo N phase.
The surface Mo species distribution and surface ratio of K:(Mo + Mo ) and N: (Mo +
2
N. Kawai et al. [36] reported Auger transitions at 385 and 392 eV for
L2,3 level of the
N
2
2
+
3+
2+
3+
3
+
2+
Mo ) are summarized in Table 2. Among the K promoted catalysts the surface Mo + Mo
δ+
species (denoted as Mo in the literature related to γ-Mo
2
N) had the highest concentration at 1.3
4
+
6+
wt. % K loading. On the other hand Mo and Mo species were decreased in samples
containing up to 1.3 wt. % K loading and increased in samples with higher K loadings. The
4
+
6+
2
results indicate that the K-O mainly dispersed on Mo and Mo species up to 1.3K-Mo N
1
2

4
+
6+
sample (Fig. 4a, TEM image). Whereas, the surface abundance of Mo and Mo species for
3
2 2 3 2 4
K- and 6.2K-Mo N suggests K-Mo-O (K MoO , K MoO ) phases.
3
.4. TEM/HRTEM/SAED measurements
The passivated K-Mo N samples were examined by TEM/HRTEM/SAED techniques and
the representative images are shown in Fig. 4. The TEM image of 1.3K-Mo N catalyst (Fig. 4a)
showed bar like morphology for MoO2, which might be characterized by Mo-O-Mo links. The
potassium species were found with spherical shape located on the MoO bar like structures of
.3K-Mo N. The high-resolution image of the corresponding slot on 1.3K-Mo N (Fig. 4b)
showed an average d-spacing of 3.71 Å, consistent with the (111) plane of dipotassium oxide
O) with an average particle size of 2-3 nm. Further, the SAED (inset, Fig.4b) analysis yielded
d- spacing of 2.26, 1.85 and 1.44 Å, consistent respectively, with the (202), (222) and (402)
planes of K O [ICSD 96-101-0879]. The chemical interaction between K O and MoO has been
clearly established in Fig. 4c for the sample containing 3wt. % of potassium in Mo N. Fig. 4d for
.2K-Mo N catalyst revealed the d spacing of 3.16 Å, consistent with (310) plane of K MoO
PDF 029-1021]. Plate like morphology was observed for γ-Mo N in potassium promoted
samples (example image of 3K-Mo N presented as Fig. 4e). Further, homogeneous distribution
of these platelets with particle size range of 5-10 nm (Table 1) illustrated in supplementary
image S1a. The inset of Fig. 4e shows SAED of γ-Mo N phase. Within B1-structured nitrides,
2
2
2
1
2
2
(K
2
2
2
2
2
6
2
2
4
[
2
2
2
one fcc lattice is defined by the transition metal atoms, and the N atoms are presumed to occupy
interstitial octahedral sites to form the second interpenetrating fcc lattice. Vacancies on the anion
sublattice result in non-stoichiometric nitrides such as Mo
particles might be associated with octahedral N atoms within the fcc lattice of Mo atoms [37].
The careful analysis of 6.2K-Mo N by HRTEM/SAED (Fig. 4f) yields significant information
about the structure of hexagonal δ-MoN phase where these particles are also of plate type and
located on the large plates of γ-Mo N (S1c). The SAED reveals average d-spacing of 2.48 Å
2
N. Hence, the observed platelet type
2
2
consistent with (200) plane of δ-MoN. Similar SAED results was observed by Ganin et al. [38]
for MoN thin films and attributed to distorted NiAs type structure with triangular clusters of Mo
atoms and a centered N atom.
In conclusion, TEM/HRTEM/SAED results revealed the chemical interaction between K
2 3 2 4
and Mo-O and have form the K-Mo-O species (K MoO and or K MoO ) in agreement with
1
3

XRD/XPS data. Hexagonal δ-MoN plates characterized by thin in nature exists on the big plates
of γ-Mo N. The distributed K content in to the Mo N matrix estimated by parallel beam EDS
showed in Table 1. The greater K content observed at 1.3wt.% K loading suggests greater K
2
2
distribution in the Mo
hydrogenation.
2
N matrix which played a vital role in oxygenates formation during CO
3
.5. CO-TPD-mass measurements
Fig. 4 shows the CO-TPD-mass results on reduced Mo
distinct desorption peaks were observed for Mo
2
N and K-Mo
2
N samples. Two
o
2
N with peak maxima at 150 and 300 C for
m/z=28 (Fig.4a). The earlier CO desorption peak was assigned to the linearly adsorbed CO on
the surface molybdenum atom of 2+ and 3+ valence state (Mo-C≡O). The later CO desorption
peak assigned to NCO species, resulted from CO adsorbed on the surface N sites [39]. Similar,
mass signals were pronounced due to Mo-CO and N-CO species up to 1.3 wt.% potassium
loading. However, Mo-CO sorption was considerably reduced for 3K- and 6.2K- loadings
δ+
compared to 1.3K-Mo
2
N. The results are in agreement with XPS data where surface Mo
species was found to be low for 3K- and 6.2K-Mo
2
N. In addition, at 6.2K- loading an extended
o
CO desorption signal was observed in the temperature range of 225-400 C with signal maxima
o
at 320 C. This is associated with metallic Mo sites, where CO dissociates to produce carbon and
o
recombination of adsorbed C and O desorb as CO below 340 C on Mo
2
N surface [40]. Fig. 4b
presents m/z=44 (CO
2
) sorption results for Mo
was observed over the Mo N and up to 3 wt.% K-Mo
desorbed at 325 C on 6.2K-Mo N. This is likely due to the presence of metallic Mo sites
[41]. Thompson et al.
N. Yang et al. [39] reported that majority of
surface. Our CO-TPD-mass results principally
showed the molecular CO sorption, which was also reported by Zaman [20] through DFT study
over γ -Mo N (111) plane.
Influence of K addition to Mo
2
N and K-Mo
2
N samples. A negligible amount of
CO
2
2
2
2
N. However, considerable amount of
o
CO
2
(
as evidenced by XPS analysis) which dissociates CO and produce CO
40] found that CO partially dissociated on bulk Mo
N/Al
2
[
2
CO is adsorbed molecularly on Mo
2
2 3
O
2
2
N on the CO sorption is as follows:
i) Principally molecular CO sorption was observed on Mo N and K-Mo
ii) CO formation indicates CO dissociation on these catalysts.
(
(
2
2
N catalysts.
2
1
4

δ+
2
(iii) Well distributed K in Mo S slabs improved the back donation of Mo and stabilizes Mo
sites [42]. In the present work, the good distribution of K observed in 1.3K-Mo
greater molecular CO sorption.
2
N and exhibited
(
iv) K
2
MoO
4
formation reduces the K distribution in to Mo
N.
2
N, hence, lowers molecular CO
sorption for 3K- and 6.2K-Mo
2
3
.6. FTIR-DRIFTs analysis
The insitu FTIR-DRIFTs analysis on Mo
2
N and 1.3K-Mo
feed 1:1 volume ratio and the results are presented as Fig. 6a &b. The
N showed the formation of hydrocarbons, mainly methane and CO from the
beginning of the reaction at 3015 and 2355 cm respectively. The hydrocarbons and CO
formation was increased with the increase in reaction time indicates greater hydrogenation
tendency of unprompted Mo N. The CO formation mainly associated with water-gas shift
reaction. CO dissociative hydrogenation reported on the surface of Mo N to produce methane as
first hydrocarbon and water as a byproduct [40]. The improved relative intensity of methane
2
N catalysts was performed at
o
3
00 C, 5 MPa with CO:H
2
unprompted Mo
2
2
-1
2
2
2
2
-1
(
3015 cm ) in Fig. 6a, time resolved spectra suggest CO dissociative hydrogenation under
reported reaction conditions. Further, the relative intensity of CO was also increased and the
separation between CO and CO was decreased with time. The result indicates that CO
concentration is almost equal to CO at reactor zone after 3h of reaction. On the other hand,
2
2
2
-1
-1
hydrocarbons in the stretching frequency range of 2850-3000 cm and 3050-3200 cm also
increased with reaction time, which reveals the chain propagation under the hydrogen rich
environment was created due to water-gas shift reaction. After 30 min. of reaction, a new
-1
-1
contribution appeared in the frequency range of 950-1080 cm centered at 1025 cm is
attributed to the formation of alcohols (methanol and ethanol). After 1h of reaction, bands
-
1
appeared between 1480 and 1680 cm suggests the formation of other oxygenates such as esters.
o
Time resolved DRIFTs analysis on 1.3K-Mo
2
N catalyst at 300 C, 5MPa is presented as
Fig.6b. Potassium promoted catalyst significantly reduced the formation of hydrocarbons
particularly methane and CO at the initial stages of reaction (up to 45 min.) compared to
unpromoted Mo N. However, after 1h of reaction the relative formation of hydrocarbons and
CO were increased. The results indicate CO dissociative hydrogenation and water-gas shift
2
2
2
-
1
reactions occurred at slower pace on this catalyst. The alcohol formation at 1025cm increased
from 1h to 3h of reaction time with consistent hydrocarbon formation. It suggests molecular CO
1
5

insertion to –CHx intermediate to form –CxHyO (acyl) intermediate to produce alcohols on this
catalyst surface. Hence the CO activation is different on K distributed Mo N compared to
unpromoted Mo N. Similar results were reported by Santos et al. [42] on K-MoS catalyst for
2
2
2
alcohol synthesis from syngas. The authors suggested that K occupies the edges and corners of
δ+
MoS
chemical environment, which decreases the hydrogenation ability of K-MoS
or methoxy (–CHxO) species on the K-Mo N surface desorb in the form of methanol. The
2
by removal of sulfur and creates coordinative unsaturated sites (Mo ) with different
2
. Stabilized alkoxy
2
presence of K, in close vicinity to an adsorbed methyl group, will favor CO insertion, stabilizing
acyl intermediates that are further hydrogenated to ethoxy and finally to ethanol or higher
alcohols.
3
.7. Catalytic activity results
The CO
2
selectivity was obtained in the range of 50-60% in the temperature range of
N. On the other hand, observed CO selectivity was 40-50% on
, due to the water gas shift reaction, was reported on
N [40].
Fig. 7a shows the influence of K loading on CO hydrogenation product distribution in the
o
2
75-325 C on unpromoted Mo
2
2
K-Mo
2
N catalysts. High selectivity to CO
2
Mo based catalysts such as Mo
2
C and Mo
2
o
-1
temperature range of 275-325 C, 7 MPa and 60,000 h . The CO conversion increased with
2
increase in the reaction temperature and reached to a maximum value of 10% on γ-Mo N at 325
o
C. The K promotion up to 1.3wt.% decreases the conversion to 4.5% and slightly improves the
o
conversion to 7 % after this loading at 325 C. The decrease in the conversion at 1.3 wt. % of K
2
loading might be associated with partial blockage of Mo sites by K O (TEM Fig. 4a), which
contributes to the CO hydrogenation to some extent. After this loading (3 and 6.2 wt.%), the
surface present metallic Mo sites slightly improved the CO conversion and mostly contributed to
hydrocarbons. In addition, at 3 and 6.2 wt.% K loading, N-CO sorption (CO-TPD results) was
2
slightly improved compared to 1.3K-Mo N. The results indicate that surface N sites (Table 2)
can also influence the CO conversion. Further, selectivity to hydrocarbons was increased with
the increase in reaction temperature (Fig. 7a). Andersson et al. [30] also noted increased
2
hydrocarbon selectivity with the increase in the reaction temperature on K-Ni-MoS catalysts.
The authors hypothesized that alcohols can undergo hydrogenation/dehydration to form
hydrocarbons at higher reaction temperatures. Higher temperatures (upon reaching the
1
6

equilibrium) are also favorable for the formation CO
the results reported for K/Mo C catalysts [31]. In addition, increased CO
temperatures can also alter the CO/H ratio in the reactor zone and further influence the CO
insertion process followed by higher alcohol formation [43-45]. Our DRIFTs data clearly
demonstrates enhanced CO formation via water gas shift reaction on unpromoted Mo N.
Hydrocarbon formation, particularly methane formation in CO hydrogenation on Mo based
2
via water gas shift reaction, consistent with
2
2
concentration at higher
2
2
2
0
δ+
4+
6+
catalysts are reported as Mo metal (Mo ) > Mo
2
N (Mo ) > MoO
2
(Mo ) > MoO
3
(Mo ) [46].
δ+
Therefore, in the present work the hydrocarbon formation is mainly associated with Mo and
0
Mo sites. Further, K addition to γ-Mo
2
N at 1.3wt.% loading significantly reduces the
hydrocarbon formation due to improved molecular CO insertion in to hydrocarbon intermediate
δ+
via K-Mo sites (Scheme 1). Tavasoli et al. [47] reported reduced hydrocarbon formation during
2 4
CO hydrogenation on K MoO phase. Based on these reports Mo in +6 valence state was found
to be less active for hydrocarbon formation. In FTs, α value indicates the formation of long-
chained hydrocarbons. Methane will always be the largest single product as long as α is less than
0
.5, however, by increasing α close to one, increases the formation of long-chained hydrocarbons
48]. In the present study, α value obtained as 0.40 for unpromoted Mo N. On the other hand,
.39, 0.38, 0.40 and 0.42 (α value) was observed for 0.45K-, 1.3K-, 3K- and 6.2K-Mo N samples
hydrocarbons
[
2
0
2
o
+
respectively, at 300 C. The results suggest that at higher K loadings greater C
2
formation over lower K loadings.
2
Among the studied catalysts 1.3K-Mo N showed maximum total oxygenates selectivity
o
of 50, 44 and 34% at 275, 300 and 325 C respectively. DRIFTs results showed molecular CO
insertion into hydrocarbon (–CHx) intermediate and further hydrogenation of resultant
intermediate yielded oxygenates at this loading. This was likely due to well distribution of K into
δ+
Mo
2
N matrix, stabilizes Mo state by back donation of electrons. At higher loadings (3 and 6.2
wt.%) most of the K chemically interacted with molybdenum oxide and produced K
2 4
MoO . The
distribution of K into the Mo N matrix was effected by this phenomenon. On the other hand, K
2
loading was inadequate at 0.45 wt.% to produce oxygenates. The results in terms of oxygenate
selectivity indicates the optimum potassium loading was at 1.3 wt.%.
3
.8. Alcohol distribution
The distribution of alcohols on γ-Mo
The butanol selectivity was found to be <1 % at all K loadings and was included in other liquid
2
2
N and K-Mo N catalysts are illustrated in Fig. 7b.
1
7

2
products selectivity. Unpromoted Mo N showed only 4% of methanol selectivity followed by
ethanol (1 %) and other liquid products (butanol, acetone and ethyl acetate) selectivity (1 %) at
o
3
00 C. On the other hand, methanol selectivity was increased considerably to 22 % at 1.3wt. %
2
of K promotion in γ-Mo N. At the same time ethanol, propanol and other liquid product
selectivity improved to 6, 7 and 9 % respectively. At higher K loadings (3 and 6.2 wt. %)
o
methanol, ethanol, propanol and other liquid product selectivity was decreased at 300 C. This is
associated with reduced capacity for molecular CO adsorption followed by molecular CO
o
insertion. Increase in the reaction temperature to 325 C decreases the alcohol selectivity for all
the studied catalysts. This is associated with improved hydrogenation tendency of these catalysts
via water gas shift reaction [31, 43-45]. One of the conventional aspects, for the formation of
higher alcohols from syngas is that the catalyst should be able to adsorb CO molecularly as well
as dissociatively. Molecular CO adsorption leads to methanol formation as a first alcohol
product, while dissociate CO adsorption give rise the –C
hydrocarbons. The existence of both sites on the catalyst surface leads to higher alcohol
formation via insertion of molecular CO in to the –C intermediate [48, 49]. Christensen et al.
50] recently reported the possibility of alcohol coupling, which intern led to condensation type
of reactions for the formation of n-propanol and n-butanol on alkali promoted MoS catalysts.
x y
H intermediate and eventually to
x
H
y
[
2
In the present study, among the alcohol distributions, methanol was the major product
and the results are in line with results of the CO-TPD-mass studies where the CO principally
adsorbed molecularly. Jiang et al. [51] reported K-Mo-S species formation at lower K loadings
and K aggregation at higher loadings, which lead to blockage of the co-ordinatively unsaturated
sites favoring hydrocarbon formation. We found K distribution in the γ-Mo
2
N matrix by
TEM/EDS analysis (Table 1, supplementary figure S1a & b). The ATR-FTIR results of K-Mo
2
N
-1
samples (supplementary figure S3) shows FTIR band for Mo-N vibration at 665 cm . Further,
-1
Mo-O stretching vibrations in the finger print region of 700-975 and 600-650 cm was observed
for K-Mo N samples. These vibrations are attributed to polycrystallites of -MoO and -Mo
phases of K. However, a new shoulder appeared at 680 cm for K-Mo
2
4
4 13
O
-1
2
N samples was likely due
o
to K-Mo-N species. The decreasing order of total alcohol formation at 300 C is as follows:
1
.3K-Mo
The overall result (EDS, FTIR and activity) suggests that, K distribution was good at 1.3 wt.% of
K loading in to Mo N. Aggregation of K as K MoO or K Mo 13 started at 3 wt.% and
2 2 2 2 2
N (35%) > 3K-Mo N (27%) > 0.45K-Mo N (23%) > 6.2K-Mo N (21%) > Mo N (5%).
2
2
4
2
4
O
1
8

δ +
intensified at 6.2 wt% K loading. Further, the distributed K helps to stabilize the Mo species by
δ +
back donation of electrons and improves the molecular CO insertion (K- Mo ).
Further, olefins (mainly ethylene and propylene) selectivity was increased from Mo
2
N (5
%
) to 1.3K-Mo N (8%) and decreased slightly on 3K- (7%) and 6.2K-Mo N (6.5%). This is
2
2
associated with transformation of corresponding alcohol and or alcohol intermediate dehydration
to olefins under the reaction conditions.
3
.8.1. Influence of space velocity
o
2
The product distribution as a function of GHSV on 1.3K-Mo N at 300 C, 7MPa is
-1
presented in Fig. 8a. At higher space velocity (60,000 h ) approximately, 4.5 % of CO was
converted into hydrocarbons and oxygenates with selectivity of 50 and 44 %, respectively. At
-1
lower space velocity (15000 h ), CO conversion was increased to 28 % with 76 % of
+
hydrocarbon and 16% of total oxygenate selectivity. Further, the C
2
hydrocarbons selectivity
was increased with decrease in the space velocity of the reaction mixture. In addition, methanol
+
selectivity was reduced (difference between total oxygenates and C
2
oxygenate selectivity in
Fig. 8a) at lower space velocity. The results indicate that hydrocarbon chain propagation
+
followed by CO insertion is favorable to form C
2
oxygenates at lower space velocity. It is
noteworthy that, unknown product selectivity (100-(total hydrocarbons + total oxygenates)) is
also increased with decrease in the space velocity.
δ+
+
3
2
.8.2. Correlation between K/Mo surface ratio and C oxygenates
+
o
δ+
The selectivity to C
2
oxygenates at 300 C is plotted as a function of K: Mo surface
+
ratio of K-Mo
2
2 2
N catalysts in Fig.8b. Unpromoted cubic γ-Mo N showed only 2.2 % of C
+
oxygenate selectivity. Whereas, the C
2
oxygenate selectivity was improved to 14 % by addition
+
of small amount of K (0.45 wt. %). Further, C
value of 22 % at 1.3wt.% of K loading in Mo
selectivity was decreased to 15 and 9 % respectively. In this study, the highest C
2
oxygenate selectivity reached to a maximum
+
2
2
N. At 3 & 6.2 wt.% K loadings the C
oxygenate
oxygenate
+
2
2
selectivity on 1.3K-Mo N is associated with Molecular CO adsorption followed by efficient
insertion into –CHx.
3
.9. Time on stream study
1
9

The time on stream results on unpromoted Mo
2 2
N and 1.3K-Mo N are presented in Fig.9. It
is obvious from the results that the catalytic activities are quite consistent for 10h of CO
o
-1
hydrogenation at 300 C, 7Mpa and 60,000 h . Unpromoted Mo
hydrocarbons and 5% of alcohols selectivity at 5% of CO conversion. On the other hand, K
promoted Mo N exhibited around 50% of hydrocarbons and 35% of alcohols selectivity at 2.5%
2
N showed around 92% of
2
of CO conversion. The results suggest that the deactivation of molybdenum nitride catalyst is
slow under the studied reaction conditions.
Fig.10 shows the XRD patterns for 1.3K-Mo
2
N at different stages of its treatment such as
3
-1
nitridation-passivation at room temperature for 1h under 30 cm min gas flow rate of 1% O
2
-He,
and spent in the CO
hydrogenation reaction at 325 C, 7MPa and 60,000 h GHSV. It is obvious from the results that
passivated and spent catalysts principally showed cubic γ-Mo N with small portion of MoO
formation slightly improved for the spent catalyst indicates that partial
o
3
-1
reduced at 450 C, for 4h under 50 cm min gas flow rate of 99.99% H
2
o
-1
2
2
phase. However, MoO
2
δ+
oxidation of Mo species during the reaction. Further, reduced catalyst principally showed
MoO
2
phase with minor cubic γ-Mo
2
N phase. The MoO
2
phase formed due to the exposure of
o
catalyst to air during the XRD analysis. The results indicate that after reduction at 450 C,
δ+
catalyst surface was enriched in Mo species and oxidized to MoO
2
up on exposure to air. In
conclusion, fresh to reduce to spent Mo
2
N (S2) and 1.3K-Mo
2
N (Fig.10) catalysts XRD results
clearly demonstrate the intact molybdenum nitride structure after the addition of K as well as
after 10 h of reaction.
4
. Conclusions
Unpromoted Mo
catalysts. This is associated with CO dissociative hydrogenation followed by water gas shift
2
2
N showed higher hydrocarbon selectivity than K promoted Mo N
+
δ
reaction on Mo sites. The structure of molybdenum nitride was intact even after the
incorporation of K in to it. K promotion to Mo N led to (i) improved molecular CO sorption (ii)
improved molecular CO insertion into –CHx intermediate. Among the K-Mo N catalysts 1.3K-
Mo N showed greater selectivity to total oxygenates during CO hydrogenation. This is
associated with well distribution of K into Mo
the molecular CO insertion. At higher potassium loadings (3 and 6.2 wt.%), K
has reduced the K distribution in to Mo N matrix, which effects the molecular CO insertion.
2
2
2
+δ
2
N matrix led to K-Mo species, which improves
2
MoO formation
4
2
2
0

5
. Acknowledgements
The project funded by the National Plan for Science, Technology and Innovation
MAARIFAH)-King Abdulaziz City for Science and Technology - the Kingdom of Saudi
(
Arabia, award number (12-PET2720-03). The authors also, acknowledge with thanks Science
and Technology Unit, King Abdulaziz University for technical support.
6
. References
[
[
[
[
1] S. Zaman, K.J. Smith, Cat. Rev. -Sci. Eng. 54 (1) (2012) 41–132.
2] D. Leckel, Ind. Eng. Chem. Res. 23 (2009) 2342–2358.
3] A.Y. Khodakov, W. Chu, P. Fongarland, Chem. Rev. 107 (2007) 1692–1744.
4] W. S. Cheng, H.H. Kung, (Eds.), Methanol Production and Use; Marcel Dekker Inc.: New
York, 1994.
5] L. Tan, G. Yang, Y. Yoneyama, Y. Kou, Y. Tan, T. Vitidsant, N. Tsubaki, Appl Catal. A:
Gen. 505 (2015) 141–149.
[
[
[
[
6] V. Abdelsayed, D. Shekhawat, J. A. Poston Jr., J. J. Spivey, Catal. Today 207 (2013) 65–73.
7] R.G. Herman, Cat. Today 55 (2000) 233–245.
8] P. Chaumette, Ph. Courty, A. Kiennemann, R. Kieffer, S. Boujana, G. A. Martin, J. A.
Dalmon, P. Meriaudeau, C. Mirodatos, Ind. Eng. Chem. Res. 33 (1994) 1460–1467.
[
[
[
[
[
[
(
[
[
[
[
[
9] Q. J. Quarderer, G. A. Cochram, US patent WO84/03696, 1984.
10] S. F. Zaman, K. J. Smith, Appl Catal. A: Gen. 378 (2010) 59–68.
11] S. F. Zaman, K. J. Smith, Catal. Today 171 (2011) 266–74.
12] M. Xiang, D. Li, H. Xiao, J. Zhang, H. Qi, W. Li, Fuel 87 (2008) 599–603.
13] J.S.J. Hargreaves, Coord. Chem. Rev. 257 (2013) 2015–2031.
14] D. Mckay, J.S.J. Hargreaves, J.L. Rico, J.L. Rivera, X.-L. Sun, J. Solid State Chem. 181
2008) 325–333.
15] A. M. Alexander, J. S. J. Hargreaves, Chem. Soc. Rev. 39 (2010) 4388–4401.
16] S. T. Oyama, Catal. Today 15 (1992) 179–200.
17] M. Nagai, Appl. Catal. A: Gen. 322 (2007)178–190.
18] J.G. Chen, Chem. Rev. 96 (1996) 1477–1498.
19] M. E. Eberhart, J. M. MacLaren “The chemistry of transition metal carbides and nitrides”
S.T. Oyama Ed., Champman & Hall, New York, 1996, p 107.
[20] S. F. Zaman, Bul. Chem. Commun. 47 (2015) 125–132.
[21] N. Arumugam, thesis titled “New ternary alkali oxides and quaternary alkali oxy-nitrides of
molybdenum and tungsten” Chapter 5, p 88.
[
[
[
[
(
[
22] Z. Wei, Q. Xin, P. Grange, B. Delmon, J. Catal. 168 (1997) 176–182.
23] R. Kojima, K.-I. Aika, Appl. Catal. A: Gen. 219 (2001) 141–147.
24] C.H. Jaggers, J. N. Michaels, A. M. Stacy, Chem. Mater. 2 (1990) 150–157.
25] M. Nagai, Y. Goto, A. Miyata, M. Kiyoshi, K. Hada, K. Oshikawa, S. Omi, J. Catal. 182
1999) 292–301.
26] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T.
Siemieniewska, Pure Appl. Chem. 57 (1985) 603–619.
[27] B. Ibeh, S. Zhang, J. M. Hill, Appl. Catal. A: Gen. 368 (2009) 127–131.
[28] V. M. L. Whiffen, K. J. Smith, S. K. Straus, Appl. Catal. A: Gen. 419–420 (2012) 111–125.
2
1

[29] S. Podila, S. F. Zaman, H. Driss, Y. A. Alhamed, A. Al-Zahrani, L. A. Petrov, Catal. Sci.
Technol. 6 (2016) 1496–1506.
[
[
[
[
30] R. Andersson, M. Boutonnet, S. Jaras, Appl. Catal. A: Gen. 417–418 (2012) 119–128.
31] E.T. Liakakou, E. Heracleous, Catal. Sci Technol. 6 (2016) 1106–1119.
32] J. Sangster, J. Phase Equilib. Diff. 34 (2013) 43–55.
33] J. M. Moggia, V. G. Milt, M. A. Ulla, L. M. Cornaglia, Surf. Interface Anal. 35 (2003) 216–
2
[
25.
34] M.E. Gálvez, S. Ascaso, P. Stelmachowski, P. Legutko, A. Kotarba, R. Moliner, M.J.
Lázaro, Appl. Catal. B: Environ. 152–153 (2014) 88–98.
[35] K. Hada , M. Nagai , S. Omi, J. Phys. Chem. B 105 (2001) 4084–4093.
[36] T. Kawai, K. Kunimori, T. Kondow, T. Onishi, K. Tamaru, Phy. Rev. Lett. 33 (1974) 533–
5
[
36.
37] C. L. Bulla, T. Kawashima, P. F. McMillan, D. Machon , O. Shebanova , D. Daisenberger,
E. Soignard, E. T-. Muromachi, L. C. Chapon, J. Solid State Chem. 179 (2006) 1762–1767.
38] A.Y. Ganin, L. Kienle, G.V. Vajenine, Electro Chemical Society Proceedings, Vol 2005-09,
p 452–453.
[
[
[
[
[
39] S. Yang, C. Li, J. Xu, Q. Xin, J. Phys. Chem. B 102 (1998) 6986–6993.
40] J. A. Schaidle, L. T. Thompson, J. Catal. 329 (2015) 325–334.
41] J. S. Lee, K. H. Lee, J.Y. Lee, J. Phys. Chem. 96 (1992) 362–366.
42] V. P. Santos, B. van der Linden, A. Chojecki, G. Budroni, S. Corthals, H. Shibata, G. R.
Meima, F. Kapteijn, M. Makkee, J. Gascon, ACS Catal. 3 (2013) 1634–1637.
43] J. M. Christensen, L. D. L. Duchstein, J. B. Wagner, P. A. Jensen, B. Temel, A. D. Jensen,
Ind. Eng. Chem. Res. 51 (2012) 4161–4172.
[
[
[
[
[
(
[
44] X. Xiaoding, E. B. M. Doesburg J. J. F. Scholten, Catal. Today 2 (1987) 125–170.
45] N. T-.Thao, H. M. Z-. Niaki, H. Alamdari, S. Kaliaguine, J. Catal. 245 (2007) 348–357.
46] M. Saito, R.B. Anderson, J. Catal. 63 (1980) 438–446.
47] T. Ahmad, K. Saba, Z. Zahra, T. Somayeh, A. Hamideh, B. Mokhtar, Iran. J. Chem. Eng 32
2013) 21-29.
48] P.L. Spath, D.C. Dayton. "Preliminary Screening -Technical and Economic Assessment of
Synthesis Gas to Fuels and Chemicals with Emphasis on the Potential for Biomass-Derived
Syngas", NREL/TP510-34929,December, 2003, p. 95.
[49] V. R. Surisetty, A. K. Dalai, J. Kozinski, Appl. Catal. A: Gen. 404 (2011) 1–11.
[50] J. M. Christensen, P. A. Jensen, N. C. Schiødt, A. D. Jensen, ChemCatChem. 2 (2010) 523–
5
26.
[51] M. Jiang, G. -Z. Bian, Y. -L. Fu, J. Catal. 146 (1994) 144–154.
2
2

2 2
Table 1: BET, XRF, TEM and CO-TPD data for Mo N and K-Mo N catalysts
Catalyst
BET
SA
m /g
Pore
volume
cc/g
Nominal
K wt.%
XRF
K wt.%
EDS
K wt.%
Average TEM
Mo N particle
size (nm)
b
2
2
Mo
2
N
110
10
8
0.071
0.015
0.013
0.012
0.013
-
-
-
7
0
1
3
6
.45K-Mo
2
N
1
0.45
-
-
a
.3K-Mo
K-Mo
2
N
5
1.3 (1.2)
0.5
0.2
0.1
8
a
2
N
7
10
15
3 (2.9)
10
6
a
.2K-Mo
2
N
6
6.2 (6.1)
a
spent catalyst
b
Approximate K content present in the cubic γ-Mo
supplementary image S1b)
2
N matrix estimated using parallel beam
(
Table 2: XPS data for Mo
Catalyst
2
N and K-Mo
2
N catalysts
Mo distribution at. %
Surface ratio
0
+2
3+
4+
6+
2+
3+
2+
3+
Mo
Mo
10
9
Mo
3
Mo
9
Mo
10
11
8
K/(Mo +Mo )
N/(Mo +Mo )
1.0
Mo
Mo
2
2
N
-
-
-
-
-
-
3
-
0
0
.95
.6
N spent
2
10
8
-
0
1
1
3
6
.45K-Mo
.3K-Mo
.3K-Mo2N spent
K-Mo
.2K-Mo
2
N
8
9
0.008
0.06
0.058
0.18
1.9
0.4
2
N
15
14
8
6
7
7
6
8
8
0.39
0.55
2
N
10
-
8
9
2
N
5
12
10
1.0
2
3

Figure captions
Fig. 1A: XRD pattern for calcined precursors with molybdenum, citric acid and potassium (a)
Mo-CA (b) 1K-Mo-CA (c) 5K-Mo-CA (d) 10K-Mo-CA (e) 15K-Mo-CA; 1B: XRD pattern for
passivated samples.
Fig. 2: (a) BET isotherms for passivated samples (b) pore size distribution of passivated samples
Fig.3A: K 2p; 3B: Mo 3d and 3C: N KLL Auger spectra of
(
a) 1.3K-Mo
Fig.4: TEM/HR-TEM, SAED images (a) 1.3K-Mo
K-Mo N TEM (d) 6.2K-Mo N HR-TEM (e) HR-TEM of Mo
TEM of Mo N & MoN matrix of 6.2K-Mo N sample
Fig.5: CO-TPD-mass analysis of passivated samples
Fig.6: DRIFTs analysis on (a) Mo N (b) 1.3K-Mo N at 300 C and 5MPa.
Fig.7: (a) CO hydrogenation activity with respect to potassium loading at reaction temperature
2 2 2
N (b) 3K-Mo N (c) 6.2K-Mo N
2
N TEM (b) 1.3K-Mo
2
N HR-TEM/SAED (c)
N (f) HR-
3
2
2
2
N matrix in 3K-Mo
2
2
2
o
2
2
o
-1
2
75-325 C, 7MPa and GHSV 60000 h (b) alcohol and other liquid product (butanol, acetone
o
and ethyl acetate) distribution with respect to potassium loading at 300 and 325 C.
o
2
Fig.8: (a) Influence of GHSV on CO hydrogenation activity of 1.3K-Mo N at 300 C and 7 MPa
δ+
+
(
b) influence of K/Mo surface ration on C
2
oxygenates selectivity at different potassium
o
loadings, 300 C and 7 MPa.
o
-
Fig.9: Time on stream results on Mo
2
N and 1.3K-Mo
2
N catalysts at 300 C, 7MPa and 60,000 h
1
.
2
Fig.10: XRD pattern of passivated, reduced, spent 1.3K-Mo N catalyst.
2
4

.

0
K Mo O
MoO (005-0508)
3
(A)
10 33
6

.
K Mo O (01-086-0243)

.
K MoO
2 4
3
14 22

0
0




0 0
0
0
0
00
0
0
(
e)
(d)







(
c)
b)
a)












(
(
1
0
20
30
40
50
60
70
2


(PDF 00-029-1021)
K MoO4
(B)
*
2

- Mo N (PDF 00-025-1366)
2


- MoN (PDF 00-025-1367)
MoO2 (PDF 00-078-1070)
^


*
*
*
*
*
*

*




*
*

6
.2K-Mo N
2
*
*
*
*
 3
K-Mo N
2
^
1.3K-Mo N
2
0
.45K-Mo N
2
^
^
^
Mo N
2
20
30
40
50
60
70
80
90
100
110
2

Fig. 1
2
5

50
40
30
20
10
0
(a)
Mo N
2
0
1
3
6
.45K-Mo N
2
.3K-Mo N
2
K-Mo N
2
.2K-Mo N
2
0
.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Relative pressure
(b)
0
0
0
0
0
.008
.006
.004
.002
.000
Mo N
2
0
1
3
6
.45K-Mo N
2
.3K-Mo N
2
K-Mo N
2
.2K-Mo N
2
10
20
30
40
50
60
70
80
90
100
Pore width (Å)
Fig. 2
2
6

(B)
Mo3d 5/2
2+
(
A)
Mo3d 3/2
Mo
K2p 3/2
(a)
(C)
(a)
N KLL (Auger)
(
a)
6+
Mo
K2p 1/2
4+
3+
Mo
Mo
(
b)
(
b)
(
b)
(
c)
(
c)
(
c)
0
Mo
Fig. 3
2
7

1
.44 Å (402)
.85 Å (222)
.26 Å (202)
(
b)
1
MoO
2
2
SAED, K O
2
ICSD 96-101-0879
K-O
3
.71 Å
(
111)
(
a)
K2MoO4
MoO
2
(d)
(
c)
K-O
3
.16 Å, (310)
δ-MoN SAED
2
.48 Å (200)
δ-MoN
γ-Mo N
2
Fig.4
γ-Mo N
2
28
(
f)
(
e)

Fig. 4
(
a)
m/z = 28, CO
6
3
.2K-Mo N
2
K-Mo N
2
1
0
.3K-Mo N
2
.45K-Mo N
2
Mo N
2
(b)
m/z = 44, CO2
6
.2K-Mo
K-Mo
2
N
3
2
N
1
.3K-Mo
.45K-Mo
Mo
2
N
0
2
N
2
N
o
Temperature C
Fig. 5
2
9