You are on page 1of 14

Applied Catalysis A: General 439440 (2012) 111124

Contents lists available at SciVerse ScienceDirect


Applied Catalysis A: General
j our nal homepage: www. el sevi er . com/ l ocat e/ apcat a
Hydrodeoxygenation of guaiacol over carbon-supported molybdenum nitride
catalysts: Effects of nitriding methods and support properties
I. Tyrone Ghampson
a,b
, Catherine Seplveda
c
, Rafael Garcia
c
, Ljubisa R. Radovic
d,e
, J.L. Garca Fierro
f
,
William J. DeSisto
a,g,
, Nestor Escalona
c,
a
Department of Chemical and Biological Engineering, University of Maine, Orono, ME 04469, United States
b
Unidad de Desarrollo Tecnolgico, Universidad de Concepcicn, Casilla 4051, Concepcicn, Chile
c
Universidad de Concepcion, Facultad de Ciencias Quimicas, Casilla 160c, Concepcicn, Chile
d
Penn State University, University Park, PA 16802, United States
e
Universidad de Concepcicn, Facultad de Ingenieria, Dept. Ing. Quimica, Concepcicn, Chile
f
Instituto de Catalisis y Petroquimica, CSIC, Cantoblanco, 28049 Madrid, Spain
g
Forest Bioproducts Research Institute, University of Maine, Orono, ME 04469, United States
a r t i c l e i n f o
Article history:
Received 25 April 2012
Received in revised form25 June 2012
Accepted 28 June 2012
Available online 5 July 2012
Keywords:
Hydrodeoxygenation
Guaiacol
Activated carbon
Mo
2
N catalysts
a b s t r a c t
Molybdenum nitride catalysts supported onactivated carbon materials with different textural and chem-
ical properties were synthesized by nitriding supported Mo oxide precursors with gaseous NH
3
or N
2
/H
2
mixtures using a temperature-programmed reaction. The supports and catalysts were characterized by
N
2
physisorption, XRD, chemical analysis, TPD, FT-IR and XPS. Guaiacol (2-methoxyphenol) hydrodeoxy-
genation (HDO) activities at 5MPa and 300

C were evaluated in a batch autoclave reactor. Molybdenum


nitrides prepared using a N
2
/H
2
mixture resulted in more highly dispersed catalysts, and consequently
more active catalysts, relative to those prepared using ammonolysis. The HDO activity was also related to
pore size distribution and the concentration of oxygen-containing surface groups of the different carbon
supports. Increased mesoporosity is argued to have facilitated the access to active sites while increased
surface acidity enhanced their catalytic activity through modication of their electronic properties. The
highest activity was thus attributed to the highest dispersion of the unsaturated catalyst species and the
highest support mesoporosity. Surprisingly, addition of Co did not improve the HDO activity.
2012 Elsevier B.V. All rights reserved.
1. Introduction
Due tolong-termeconomic andenvironmental concerns, bio-oil
derived from pyrolysis of woody biomass has received consider-
able attention as an alternative renewable feedstock to crude oil
for the production of fuels and value-added chemicals [1]. Its uti-
lization as fuel is limited, however, by high viscosity, low heating
value, incomplete volatility and thermal instability, which stem
from the relative abundance of oxygenated organic compounds
[2]. Catalytic hydrodeoxygenation (HDO) reactions are typically
performed to rene bio-oil and increase its quality as transporta-
tion fuel. There are two signicant challenges in this process: (i)

Corresponding author at: Department of Chemical and Biological Engineering,


University of Maine, Orono, ME 04469, United States. Tel.: +1 207 581 2291;
fax: +1 207 581 2323.

Corresponding author at: Tel.: +56 41 2207236; fax: +56 41 2245374.


E-mail addresses: WDeSisto@umche.maine.edu (W.J. DeSisto),
nescalona@udec.cl (N. Escalona).
prevention of coke formation/catalyst deactivation and (ii) selec-
tive removal of oxygen without excessive hydrogenation of
aromatic and olenic compounds [2,3].
Model compounds have been used to mimic HDO studies of
bio-oil components in an effort to understand the role and fate of
different functional groups present in the feed, as well as provide
additional insight into the development of improved catalysts and
processes [3]. Guaiacol (2-methoxyphenol) is commonly used as a
model compound for HDO studies to represent the large number
of mono- and dimethoxy phenols present in bio-oil [4]; it is known
to be a precursor to catechol but it also subsequently forms coke
[5,6]. Also, guaiacol possesses two different oxygenated functional
groups ( OCH
3
and OH) which make it challenging to achieve
complete deoxygenation [7].
Heterogeneous catalysts commonly studied for HDO of guaia-
col (and many other model compounds) are conventional sulded
Co(Ni)Mo/-Al
2
O
3
[5,8] and supported noble metal catalysts such
as Ru, Rh and Pd [9,10]. The initial interest in the metal suldes
was driven by high cost and lack of selective HDO activity of
the noble metal catalysts. Despite the high catalytic activity for
0926-860X/$ see front matter 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.06.047
112 I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124
guaiacol conversion, there are some drawbacks associated with
these sulde catalysts: (i) The alumina support can be unstable in
water at processing conditions. (ii) The sulde catalyst can oxi-
dize under processing conditions, requiring in situ regeneration
with a sulding agent to prolong catalyst activity; this regenera-
tion can contaminate products [5,6,11,12]. (iii) The acidic nature
of the alumina support is known to be the cause of substantial
coke deposition and rapid catalyst deactivation [13]. These draw-
backs prompted interest in less acidic materials such as silica [14],
zirconia [9,15] and activated carbon [16,17] as catalyst supports.
Centeno et al. [14] reported that despite the lower activity of
metal suldes supported on silica and carbon compared with con-
ventional alumina-supported counterparts, the use of alternative
supports led to negligible coke formation. Furthermore, studies
involvingcarbon- andzirconia-supportedcatalysts indicateddirect
elimination of the methoxy group which favored direct production
of phenol fromguaiacol [14,18]. Particularly, based on their supe-
rior performance in hydrodesulfurization processes, carbons are
known to be promising supports for the HDO of bio-oil [1921].
Interest in carbon supports has increased mainly due to its
remarkable exibility and the ability to recover active metal after
catalyst deactivation [20]. For HDO reactions, such deactivation in
thepresenceof water couldbelimitedduetothehydrophobic char-
acter of the carbon surface [22]. On the other hand, the weaker
interaction between the support surface and the active metal may
result in a lower dispersion of the sulde phase [14,20]. The car-
bon surface can be decorated with oxygen functionalities; this
improves catalytic activity by facilitating a higher dispersion of the
active phase [23,24]. For example, oxidative treatments with HNO
3
modiedthecarbonsurfacechemistryandpromotedtheformation
of small, well-dispersed crystals of the molybdenumprecursor on
the support [21,25] although this led to lower phenol yields during
guaiacol HDO [23]. Additional studies further conrmed that HDO
chemistry can be controlled by modifying the surface chemistry of
the carbon support and consequently the dispersion of the catalyst
[16,22]. This adds to the potential use of carbon-supported systems
in rational catalyst design [17,19].
A wide variety of active phases metals [4,9,10,26], transition
metal phosphides [7] and transitionmetal nitrides [27] have been
employedfor HDOreactions inorder to obviate the needto addsul-
fur to the feed. In particular, transition metal nitrides show great
potential as catalysts due to their ceramic-like physical properties
coupledwithchemical properties resemblingplatinum-groupmet-
als [28]. These materials are also responsible for unique catalytic
pathways, leading to desirable product selectivities [28,29]. Con-
sequently, they offer a cheaper and more selective alternative to
noble-metal catalysts such as Ru, Pd and Pt. Our previous study
showed high activities and a high phenol/catechol ratio for bulk
molybdenum nitride catalysts in the HDO of guaiacol [30]. How-
ever, supported catalysts are preferred in commercial applications
for mechanical and morphological stability. The addition of Co as a
promoter is known to improve the activity of bulk and supported
Mo
2
N catalysts for HDS and HDN reactions [31,32]. Furthermore,
Co-promoted MoS
2
catalysts have been reported to exhibit sig-
nicantly higher HDO activity compared to non-promoted MoS
2
catalysts for HDOof guaiacol [8]. Although addition of Co improved
the yield of deoxygenated products, the overall activity was not
enhanced compared to the monometallic nitride [30]. This moti-
vated us to investigate also the effect of Co promoters on the
catalytic properties of supported nitrides in HDO reactions.
Here we report on the behavior of molybdenum nitrides
dispersed on four different activated carbon supports. The
supports were both microporous/mesoporous and meso-
porous/macroporous carbons. The catalysts were synthesized
by impregnation of an aqueous salt and its subsequent conver-
sion to the nitride; thermal conversion was achieved by either
ammonolysis or reduction/nitridation using a N
2
/H
2
mixture.
The effects of the synthesis procedure, support properties and the
additionof Co as a promoter onthe HDOof guaiacol were examined
in terms of catalytic activity and phenol/catechol selectivity.
2. Experimental
2.1. Preparation of catalysts
Four commercial activated carbons (Norit Americas, Inc.) were
used as supports: NORIT GAC 1240 Plus (0.422.00mm particle
size, S
BET
=976m
2
g
1
, total pore volume =0.56cm
3
g
1
), NORIT
GCA 1240 Plus (0.421.70mm, 1132m
2
g
1
, 0.51cm
3
g
1
), Darco
MRX (0.602.00mm, 613m
2
g
1
, 0.62cm
3
g
1
), and NORIT CGran
(0.501.70mm, 1402m
2
g
1
, 1.15cm
3
g
1
). Prior to their use, the
activated carbon materials were treated with 1M HNO
3
at 90

C
for 6h. The solution was then ltered and extensively washed with
distilled water to bring the pH to ca. 7. The samples were dried
overnight under vacuum at 120

C. The supported molybdenum


oxide precursors were preparedbyincipient wetness impregnation
using aqueous solutions of ammonium heptamolybdate (Fischer,
AHM, (NH
4
)
6
Mo
7
O
24
4H
2
O, A.C.S. grade). After the impregnation,
the samples were kept at room temperature for 24h, followed
by drying overnight at 110

C. The bimetallic oxide precursors


were prepared by sequential impregnation: Mo-loaded samples
were rst prepared using the same dryingcalcination procedure
described above; these samples were thenimpregnated withaque-
ous solution of cobalt (II) nitrate hexahydrate (Acros Organics,
Co(NO
3
)
2
6H
2
O, 99%) and kept overnight at roomtemperature fol-
lowed by drying overnight at 110

C. The supported oxides were


prepared to obtain a nominal loading of 10wt% Mo for monometal-
lic samples and 10wt% Mo plus 2.4wt% Co for the bimetallic
samples. All oxide precursors were sieved to obtain 180450m
particle size range.
Molybdenum nitrides were prepared by loading a 10mm i.d.
quartz reactor tube with 2.5g of the oxidic precursor, while pass-
ing ammonia (Matheson Tri-Gas, NH
3
, 99.99%) or a N
2
/H
2
mixture
(N
2
: BOC Gases, Grade 5; H
2
: Matheson Tri-Gas, 99.99%) over the
sample [33,34]. The reactor was initially purged with nitrogen for
30min and switched to NH
3
(300mL min
1
) or a N
2
/H
2
mixture
(300mL min
1
, N
2
/H
2
=5/1 (v/v)). The temperature was linearly
increased from ambient temperature to 300

C within 30min
(9.33

Cmin
1
), then from 300 to 500

C at 0.6

Cmin
1
, and from
500 to 700

C at 2

Cmin
1
. The temperature was maintained at
700

Cfor 2h. The nitrides preparedusing NH


3
were cooledto room
temperature using the same ow rate of NH
3
while the nitrides
prepared using the N
2
/H
2
mixture were cooled in 300mL min
1
of
N
2
. The materials were then passivated in 1% O
2
/N
2
(BOC Gases,
UHP grade) for 12h to avoid oxidation upon exposure to air. The
preparation condition was adopted after careful review of the lit-
erature in addition to previous results to ascertain the optimal
condition for unsupported molybdenumnitrides [30]. Preparation
of nitride catalysts using NH
3
and N
2
/H
2
is referred to as method
1 and method 2, respectively. For notation, Mo nitrides prepared
usingmethod1have sufxA, while sufxNH implies method2:
e.g., MoN/Darco-A and MoN/Darco-NHare Darco activated carbon-
supported Mo nitride catalysts prepared using NH
3
and N
2
/H
2
,
respectively.
Molybdenum, cobalt andnitrogencontents inthe catalysts were
performed by Galbraith Laboratories using ICP-AES for the metals
and a combustion method for nitrogen.
2.2. Nitrogen porosimetry
Nitrogen adsorption/desorption isotherms were obtained at
77K using a Micromeritics ASAP-2020 instrument to evaluate the
I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124 113
porous structure of the support and catalyst samples. Prior to
the measurements, the samples were outgassed under vacuum
at 200

C for 12h. The isotherms were collected within a broad


relative pressure range, 10
6
<P/P
0
<0.995, and using a low pres-
sure incremental dosing (3cm
3
g
1
STP) in order to obtain an
adequate characterization of the micropore region. The isotherms
were used to calculate the BET specic surface area (S
BET
), total
pore volume (TPV), average pore diameter (d
pore
), and micropore
volume (V

). S
BET
was obtained from the adsorption branch in
the range 0.04P/P
0
0.14 and TPV was recorded at P/P
0
=0.995.
Average pore diameters were calculated fromthe equation d
pore
=
2 TPV/S
BET
, assuming slit-shaped pores. The pore size distribu-
tions (PSD) for 0.4100nm were determined from the adsorption
branchof the isothermusing the nonlocal density functional theory
(NLDFT) method [35,36]. Micropore volume was calculated from
the NLDFT cumulative volume of pores whose size was below2nm.
2.3. X-ray diffraction
Wide-angle X-ray diffraction (XRD) patterns of powdered sam-
ples were obtained using a PANalytical XPert Pro diffractometer
equipped with a graphite monochromator and Cu K radiation
(45kV, 40mA) in a parallel beam optical geometry. The standard
scan parameters were 1585

2 with a step size of 0.02

and a
counting time of 10s per step. Identication of the phases was
achieved by reference to the JCPDS data les.
2.4. X-ray photoelectron spectroscopy (XPS)
X-ray photoelectron spectra of reduced catalysts were obtained
on a VG Escalab 200R electron spectrometer using a Mg K
(1253.6eV) photon source. The samples were pre-reduced ex situ
with H
2
at 450

C for 6h. After reduction, the samples were cooled


to room temperature, ushed with nitrogen and stored in asks
containing isooctane (Merck, 99.8%), then transferred to the pre-
treatment chamber of the spectrometer. The binding energies (BE)
were referenced to the C 1s level of the carbon support at 284.9eV.
Anestimatederror of 0.1eVcanbeassumedfor all measurements.
Intensities of the peaks were calculated from the respective peak
areas after background subtraction and spectrumtting by a com-
bination of Gaussian/Lorentzian functions. Relative surface atomic
ratios were calculated from
(Mo/C)
atomic ratio
=
(S
Mo3d
/f
Mo3d
)
(S
C1s
/f
C1s
)
; (Co/C)
atomic ratio
=
(S
Co2p
/f
Co2p
)
(S
C1s
/f
C1s
)
; (N/C)
atomic ratio
=
(S
N1s
/f
N1s
)
(S
C1s
/f
C1s
)
where S is the corresponding peak areas and f is the respective
tabulated sensitivity factor [37] and the factor for the device. The
atomic ratios were calculated at a precision of 7%.
2.5. Temperature-programmed decompositionmass
spectroscopy (TPDMS)
TPD analyses of the carbon supports were carried out in an
in-house apparatus which consisted of a U-shaped quartz tube
micro-reactor, placed inside a programmable electrical furnace.
The TPD proles of CO, CO
2
and H
2
O were obtained from room
temperature to 1040

C, at 10

Cmin
1
under helium (AGA Chile,
99.995%) ow of 50mL min
1
. Evolution of desorbed gases was
monitored by a thermal conductivity detector (TCD). In addition,
to quantify the gases produced during thermal decomposition of
the surface functionalities, TPD was coupled with mass spectrom-
etry (Altamira AMI-200 R-HP instrument equipped with a SRS
RGA-300 mass spectrometer). About 0.2g of the activated carbon
sample was rst pretreated at 100

C for 4h in He (50mL min


1
) to
remove most of the weakly adsorbed water, and cooled to room
temperature in He. The pretreated sample was then heated in
a ow of He (50mL min
1
) from room temperature to 800

C at
10

Cmin
1
.
2.6. Fourier transform-infrared transmission (FT-IR) spectroscopy
FT-IR analyses of the supports were performed on a Nicolet
Nexus FTIR in the wavenumber range (4000400cm
1
) and with a
scanof 64. Thesamples werepreparedusinga1:100mgof activated
carbon and KBr support.
2.7. Acidity measurements
Acid site concentrations and acid strength measurements of the
supports and selected catalysts were determined using a potentio-
metric method [38], whereby a suspension in acetonitrile (Merck,
99.9%) was titrated with n-butylamine (Merck, 99%). The varia-
tion in electric potential was registered on a Denver Instrument
UltraBasic pH/mV meter.
2.8. Reaction characterization
Reactivity studies were performed in a 300mL stirred-batch
autoclave set-up (Parr Model 4841) at 300

C and under a hydrogen


pressure of 5MPa. Prior to catalyst testing, the passivated samples
were activated ex situ under H
2
(AGA Chile, 99.99%) at a ow rate
of 60mL min
1
and 450

C for 6h, conditions that were shown in


previous studies to maximize HDOconversion [27]. Approximately
0.25g of freshly pre-treated catalyst was added to the reactor
charged with 80mL of decalin (Merck, 99%), 2.53mL of guaiacol
(0.232mol L
1
, Merck, 99.5%), and 700L of hexadecane (Merck,
99%). Hexadecane was used as an internal standard for quanti-
tative GC analysis. The sealed reactor was ushed with nitrogen
(AGA Chile, Grade 5) for 30min to evacuate air from the system.
While continuously stirring the mixture, the reactor was heated to
300

C under N
2
. Once the reaction temperature was attained, N
2
was replaced with H
2
and then pressurized to 5MPa. This pressure
was maintainedfor the entire durationof the experiment by adding
H
2
to the reactor whenever necessary. Liquid samples were peri-
odically withdrawn during the course of the reaction after purging
the sampling line by withdrawing a small amount of the reactant
mixture. The samples were analyzed by a Perkin Elmer (Clarus
400) gas chromatograph equipped with a ame ionization detector
(FID) and a CP-Sil 5 CB column (Agilent, 30m0.53mm1.0m
lm thickness). The injector and FID were held at 275 and 180

C,
respectively. (The GC oven programconsisted of an initial isother-
mal operation at 30

C for 6min, followed by heating to 70

C at
30

Cmin
1
with an isotherm of 22min, and a subsequent ramp
to 275

C at 30

Cmin
1
.) The product distributions were identied
by their column retention time in comparison with available stan-
dards. The initial concentration of guaiacol was taken as 100% in
order to ignore slight conversion before isothermal condition was
achieved. The catalytic activity was expressed as the initial reaction
rate, calculated from the slope of the conversion vs. time plot, as
well as by the intrinsic activity (in moles of guaiacol consumed per
mole of Mo per second). Anumber of repeated runs under the same
conditions were performedtoensure satisfactoryreproducibilityof
the data. The uncertainty in the calculation of reaction rates from
the GC peaks is 3%. The phenol/catechol ratios were determined at
10% conversion of guaiacol.
The stability of selected nitride catalysts during the HDO reac-
tion was compared to that of a commercial reference catalyst
in a stainless steel continuous-ow micro-reactor. In a typi-
cal experiment, approximately 0.2g of catalyst was diluted 1:1
114 I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124
Fig. 1. Pore size distribution of HNO
3
treated activated carbon supports.
with SiC (Soviquim, Chile) and loaded into the reactor tube,
while the remaining reactor space was packed with SiC. Prior to
the reaction, the nitride catalyst was reduced in situ using the
same pretreatment conditions employed for the batch reaction,
while the commercial NiMo/Al
2
O
3
catalyst (Procatalyse, HR 346,
S
BET
=256m
2
g
1
) was sulded in situ using a 10vol.% H
2
S (AGA
Chile, 99.99%) in H
2
, at a ow rate of 67.5mL min
1
and 350

C for
3h. The liquid reactant mixture and hydrogen were connected to
the reactor inlet where they owed downward through the cata-
lyst bed. The conditions for HDO reactions were as follows: 300

C,
3MPa, 5.4g/h of liquid feed corresponding to liquid hourly space
velocity (LHSV) of 27h
1
, H
2
gas hourly space velocity (GHSV)
of 3600h
1
, H
2
/guaiacol molar ratio of 23. Fresh samples were
collected at an hourly interval for 89h with regular ushing pre-
ceding each collection. The liquid products were then analyzed by
GCFID.
3. Results and discussion
3.1. Textural properties
The pore size distributions (PSD) for the activated carbons, cal-
culatedfromtheadsorptionbranchof theisothermusingtheNLDFT
method, are shown in Fig. 1. All the support materials have a wide
range of pore sizes, including micropores in the range 0.42.0nm.
However, the results for CGran and Darco supports reveal a pre-
dominance of larger mesopores (upto100nm). Incontrast, the GCA
and GAC supports are more microporous; the latter also possesses
an appreciable amount of larger mesopores (3100nm).
The BET surface areas, total and micropore volumes of supports
and nitride catalysts are presented in Table 1. Signicant differ-
ences in these textural parameters among the four supports are
obvious. The CGranandGCAcarbons bothhave higher surface areas
but the latter has a larger fraction of micropores (76 vs. 32%). The
lowest S
BET
displayed by the Darco support is consistent with this
material having the lowest microporosity (19%) of all the supports.
The porosity of the GAC support is more evenly distributed: 57%
mesoporosity and 43% microporosity.
Oxidation pretreatment of the supports with HNO
3
produced
varying changes in the textural properties of the original samples.
In particular, a signicant loss of surface area (28%), total pore vol-
ume (25%) and micropore volume (24%) was found for the CGran
activated carbon; this is attributed to an increase in the quantity of
oxygen-containing surface groups on the pore walls and entrances,
making them inaccessible to N
2
molecules at 77K [39]. The pre-
treatment resulted in signicantly less micropore volume decrease
inthe GACcarbon(5%), corresponding toa 4%loss of surface area; in
contrast, there was an increase in surface area and pore volume for
the Darco and GCA activated carbons, possibly due to the removal
of impurities fromthe pores. The pore size distributions of the car-
bons were not signicantly modied after the treatment, however.
Results summarized in Table 1 also showthat impregnation of the
supports with Mo and Mo/Co, followed by thermal conversion to
the nitrides, ledto a general decrease insurface area, as well as total
and micropore volume, especially in the case of the CGran support.
These results suggest that nitride formation occurred inside the
porous structure. For the Darco support in particular, they suggest
preferential nitrideformationinsidethemesopores; conversely, for
I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124 115
Table 1
Nitrogen porosimetry results for supports and catalysts.
Sample SBET (m
2
g
1
) dpore (nm) Pore volume (cm
3
g
1
)
TPV V
Darco
as-received
612 2.0 0.62 0.12
Darco
pretreated
664 2.0 0.68 0.14
MoN/Darco-A 561 2.1 0.58 0.13
MoN/Darco-NH 560 2.1 0.60 0.14
CoMoN/Darco-A 475 2.3 0.54 0.10
CGran
as-received
1402 1.6 1.15 0.37
CGran
pretreated
1014 1.7 0.86 0.28
MoN/CGran-A 566 1.6 0.46 0.15
MoN/CGran-NH 571 1.6 0.47 0.15
CoMoN/CGran-A 461 1.9 0.44 0.11
GAC
as-received
976 1.1 0.56 0.32
GAC
pretreated
942 1.2 0.55 0.31
MoN/GAC-A 775 1.2 0.46 0.25
MoN/GAC-NH 752 1.2 0.45 0.24
CoMoN/GAC-A 706 1.5 0.52 0.21
GCA
as-received
1132 0.9 0.51 0.39
GCA
pretreated
1202 0.9 0.55 0.42
MoN/GCA-A 995 0.9 0.45 0.35
MoN/GCA-NH 1066 0.9 0.49 0.35
CoMoN/GCA-A 950 0.9 0.44 0.33
the CGran support they suggest preferential deposition in microp-
ores or at the entrances to such pores.
3.2. Surface chemical properties of the support
The chemical nature of oxygen functionalities present on the
activated carbon surface after HNO
3
treatment was determined
from TPD/MS measurements, as shown in Fig. 2 and summarized
in Table 2: lactonic (190650

C) [4042], carboxylic (200300

C)
[41,42], phenolic (600700

C) [40,42], carbonyl (800980

C)
[40,42], and quinone groups (7001000

C) [42,43]. Decomposi-
tion of groups whose carbon atomis bonded to two oxygen atoms
(carboxylic acids, lactones, and carboxylic anhydrides) releases
CO
2
, indicative of the presence of strong acidic sites [21,24]. All
the supports exhibited pronounced peaks at low temperatures
(250400

C) and thus possessed large amount of acidic CO


2
-
desorbing groups, as expected [24]. In the high temperature region,
the GCA, GAC and Darco supports presented broader 5151000

C
shoulders, which is indicative of the prevalence of phenolic and
quinone groups. The CGran support exhibited a 415730

C peak
which is suggestive of the presence of mostly phenolic groups;
decomposition of these groups, which leads to desorption of CO,
indicates the presence of weakly acidic, neutral and basic groups
whose carbon atomis bonded to one oxygen atom[21,24]. In addi-
tion, integration of the evolved gas TPD proles shows that CGran
Fig. 2. TPD proles of the activated carbon supports.
contained the greatest amount of CO
2
- and CO-desorbing groups
among the activated carbon supports. Table 2 also shows that the
GCA and Darco carbons had similar quantities of CO
2
- and CO-
releasing functional groups. (The H
2
O proles at lowtemperature,
indicatingweaklybounddesorbedmolecules, were not quantied.)
Furthermore, comparison of TPD/MS results of pretreated and as-
received activated carbon supports (not shown) indicates greater
amounts of CO
2
- and CO-desorbing groups in the former. This is
consistent with the well documented effectiveness of oxidative
HNO
3
treatment [21]. It should be claried, however, that if advan-
tage is not taken of the surface oxygen anchoring sites by making
sure, during catalyst preparation, that there is attraction between
the positively charged support surface and the negatively charged
catalyst precursor, or vice versa then the only benet of surface
oxidation can be to render the support more hydrophilic (unless
textural properties are signicantly modied).
Fig. 3 shows the FT-IR spectra of the activated carbon sup-
ports. The supports displayed bands in the 35003300cm
1
,
16001500cm
1
and 15001000cm
1
regions which can be
assigned to anhydrides, quinonic, carboxylic or ether groups,
respectively [44]. There is no apparent difference between the
GCA, GAC and Darco spectra, indicating, in agreement with TPD
results, that they have similar surface oxygenated species. For the
Fig. 3. FT-IR spectra of the activated carbon supports.
116 I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124
Table 2
Surface chemical and acidic properties of oxidized supports.
Support TPD (arbitrary units) Acidity measurements
CO
2
CO Acid strength (mV) Total acidity (mequiv. m
2
)
Darco 12 8 127 2.3
CGran 26 29 290 1.5
GAC 16 16 119 1.6
GCA 12 8 61 1.2
CGran support, a strong band appears at 1691cm
1
which can
be assigned to lactonic groups [44], in agreement with the cor-
responding TPD prole; another two bands were detected in the
31243037cm
1
and 29042844cm
1
regions, assigned to aro-
matic and aliphatic groups [44]. By comparing the spectra in Fig. 3,
CGran support displayed the greatest intensity of carboxylic or
ether groups suggesting that it contains the greatest quantity of
surface acidic oxygen groups. This result is consistent with the
interpretation fromTPD proles.
The surface acidity of the supports was estimated from poten-
tiometric titrationcurves withn-butylamineas theprobemolecule.
The results include the maximum acid strength of the surface
sites (derived fromthe initial electrode potential, E
0
) and the total
number of acid sites normalized by the surface area (acid site
density). The acid strength can be determined according to the cri-
terion proposed by Cid and Pecchi [38]: E
0
>100mV, very strong
sites; 0<E
0
<100mV, strong sites; 100<E
0
<0mV, weak sites;
E
0
<100mV, very weak sites. Accordingly, the results conrmthat
the HNO
3
-treatment produced strong acid sites on the activated
carbon supports. Table 2 shows that the CGran, Darco and GAC sup-
ports displayed very strong acid sites with E
0
>100mV, whereas
the GCA carbon displayed strong acid sites with 0<E
0
<100mV.
As mentioned, the Darco carbon had the highest acid site density,
while the GAC and CGran carbons had similar densities. The lowest
density of acid sites was measured for the GCA carbon.
3.3. Bulk and surface composition of nitrided catalysts
Bulk molybdenum, cobalt and nitrogen contents of passivated
supported catalysts are listed in Table 3. The nitrogen content is
due to nitride formation and the creation of pyridine- and pyrrole-
like functions during ammonolysis and N
2
/H
2
treatment [43]. To
distinguish the nitrogen content due to the molybdenum nitride
from those associated with the carbon support, we nitrided blank
support in a manner similar to that used to prepare the sup-
ported Mo and CoMo nitride catalysts, and performed chemical
nitrogen analysis: the results indicate that between 52 and 73% of
the total nitrogen content of the monometallic method 1 samples
were due to contributions from the supports nitrogen-functional
groups; by comparison, the nitrogen species of the carbon support
accounted for between 27 and 45% of the total nitrogen content
of the monometallic method 2 catalyst; the corresponding sup-
port nitrogen groups for the bimetallic samples ranged 4290%.
This clearly shows that for the same support the nitrogen contents
were consistently higher for samples prepared via method 1 (A-
series). Ammonolysis was thus more effective thanwithN
2
/H
2
. The
measured metal content values, in most part, correspond to theo-
retical values (10wt% for Mo and 2.4wt% for Co). However, some
of the catalysts had metal contents lower than the nominal values
which could be attributed to the possible loss of volatile molyb-
dates and cobaltates during the decomposition steps of catalyst
preparation such as drying and nitridation, a phenomenon which
has also been observed by other authors [22,45]. The bulk N/Mo
atomic ratios (due to the nitrogen structure of the molybdenum
nitride and denoted (N/Mo)
nitride
in Table 3) for the catalysts were
generally higher than the stoichiometric N/Mo ratio for -Mo
2
N
(i.e. 0.5) and -Mo
2
N
0.78
(i.e. 0.39) which suggests nitrogen enrich-
ment, possibly residing in interstitial sites. Table 3 also shows that
the N/Mo ratio was higher for the method 1 samples than method 2
samples; this suggests a higher degree of nitridation of the former.
The surface composition of the reduced and passivated nitride
catalysts was determined by XPS and the results are summarized in
Table 4. The C 1s binding energies consisted of four peaks between
284.8 and 289.2eV. The peak at 284.8eV was assigned to C C
and/or C C bonds of aromatic (and aliphatic) carbon [42,46], while
at 286.3eV it is indicative of C Obonds in phenolic or ether groups
[47,48], or may be due to the presence of C Nbonds [49]. The peak
at 287.7eV is consistent with quinone-type groups or C N species
[49,50], and at 289.3eV with carboxyl groups and esters [48]. The
relative peak intensities (shown in parentheses) indicate the pre-
dominance of aromatic carbon on the surface of all the catalysts.
The XRDpatterns of the GAC-, GCA- and Darco-supported metal
nitrides (see Supplementary data) revealed only peaks due to the
original carbon supports. The absence of Mo nitride diffraction
peaks suggests that the catalysts contained small crystallites of
Mo nitride below the detection limit, or that the broad 002 and
10 bands from the support may have masked the low-intensity
Mo nitride peaks. The XRD results of CGran supported nitrides are
summarized in Fig. 4. The MoN/CGran-NH catalyst displayed the
characteristic peaks for -Mo
2
N
0.78
(2 =37.61

, 62.53

, 75.53

)
together with broad features (2 =26

and 43

) associated with the


carbon supports; a barely discernible peak at 2 =37.41, consistent
with the -Mo
2
N (111) phase, was observed in the MoN/CGran-
A catalyst. Also shown in Fig. 4 is the diffraction pattern for
CoMoN/CGran-A, which indicated the presence of Co
3
Mo
3
N crys-
tallites (2 =40.09

, 42.59

, 46.59

).
XPS results of Mo 3d
5/2
, N 1s and Co 2p
3/2
species are summa-
rized in Table 4. The binding energies of Mo 3d and N 1s of some
selected catalysts are shown in Fig. 5. The Mo 3d spectra, shown in
Fig. 5(a), presented two spectral lines centered at 232.60.3eV
and 235.70.3eV corresponding to the Mo 3d
5/2
and Mo 3d
3/2
spin orbits respectively, assigned to Mo(VI) species [51]. This indi-
cates that after the reduction treatment only Mo oxynitrides was
present on the catalyst surface. In the CoMo nitride catalysts the
Fig. 4. XRD patterns of CGran carbon-supported nitrides.
I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124 117
Table 3
Chemical analysis of metal nitride catalysts.
Catalyst Elemental composition (wt%) Atomic ratio
Mo Co N (N/Mo)
total
(N/Mo)
nitride
MoN/Darco-A 10.20 3.70 2.5 0.8
MoN/Darco-NH 9.57 1.59 1.1 0.7
CoMoN/Darco-A 8.53 2.05 4.10 3.3 1.2
MoN/CGran-A 9.62 - 5.58 4.0 1.1
MoN/CGran-NH 10.3 1.26 0.8 0.5
CoMoN/CGran-A 10.1 2.87 4.55 3.1 0.3
MoN/GAC-A 7.6 2.14 1.9 0.9
MoN/GAC-NH 9.97 1.58 1.1 0.8
CoMoN/GAC-A 9.48 1.97 2.35 1.7 0.9
MoN/GCA-A 9.93 2.09 1.4 0.7
MoN/GCA-NH 8.79 1.16 0.9 0.6
CoMoN/GCA-A 8.44 2.03 2.54 2.1 1.2
Fig. 5. The XPS spectra of the (a) Mo 3d doublet and (b) N 1s in selected catalysts.
118 I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124
Table 4
XPS results of reduced-passivated nitride catalysts.
Catalyst Binding energy (eV) (distribution/%) Surface atomic ratio (at.%)
C 1s Mo 3d
5/2
N 1s Co 2p
3/2
Mo/C Co/C (N/C)T (N/C)
nitride
a
N/Mo
b
MoN/Darco-A
284.8 (72)
232.7

0.0140

0.0506 0.0096 0.7
286.2 (16) 396.5 (19)
287.7 (6) 398.4 (55)
289.3 (6) 400.1 (26)
MoN/Darco-NH
284.8 (73)
232.7

0.0159

0.0400 0.0084 0.5
286.2 (16) 396.9 (21)
287.7 (6) 398.5 (59)
289.3 (5) 400.1 (20)
CoMoN/Darco-A
284.8 (72)
232.6 781.4 0.0102 0.0061 0.0468 0.0089 0.9
286.2 (16) 396.5 (19)
287.7 (6) 398.4 (54)
289.3 (6) 400.1 (27)
MoN/CGran-A
284.8 (77)
232.8

0.0088

0.0531 0.0037 0.4
286.3 (14) 396.8 (7)
287.7 (5) 398.4 (63)
289.3 (4) 400.1 (30)
MoN/CGran-NH
284.8 (79)
233.0

0.0176

0.0529 0.0064 0.4
286.3 (13) 396.9 (12)
287.7 (4) 398.6 (55)
289.3 (4) 400.1 (33)
CoMoN/CGran-A
284.8 (80)
232.7 781.5 0.0182 0.0095 0.0088 0.0015 0.8
286.3 (12) 396.6 (17)
287.7 (4) 398.5 (57)
289.2 (4) 400.2 (26)
MoN/GAC-A
284.8 (76)
232.5

0.0128

0.0380 0.0057 0.5
286.3 (16) 396.3 (15)
287.7 (5) 398.3 (56)
289.2 (3) 399.6 (29)
MoN/GAC-NH
284.8 (76)
232.5

0.0157

0.0410 0.0078 0.5
286.3 (15) 396.8 (19)
287.7 (5) 398.4 (62)
289.2 (4) 400.1 (19)
CoMoN/GAC-A
284.8 (73)
232.5 781.3 0.0143 0.0075 0.0485 0.0184 1.3
286.2 (16) 396.3 (38)
287.7 (6) 398.3 (40)
289.3 (5) 399.6 (22)
MoN/GCA-A
284.8 (77)
232.4

0.0139

0.0436 0.0065 0.5
286.2 (15) 396.3 (15)
287.7 (5) 398.3 (56)
289.2 (3) 399.6 (29)
MoN/GCA-NH
284.8 (76)
232.5

0.0146

0.0379 0.0072 0.5
286.3 (15) 396.8 (19)
287.7 (5) 398.4 (62)
289.2 (4) 400.1 (19)
CoMoN/GCA-A
284.8 (72)
232.5 781.5 0.0104 0.0045 0.0421 0.0160 1.5
286.2 (17) 396.3 (38)
287.7 (6) 398.3 (40)
289.2 (5) 399.6 (22)
a
Calculated by multiplying the distribution of N 1s in the region of 396.3396.9eV to the total N/C atomic ratios.
b
N/Mo = (N/C)
nitride
/(Mo/C).
Co 2p
3/2
binding energy was 781.4eV; BE characteristic of Co(II)
species is normally observed at 781.8eV [52], showing a decrease
in positive Co
2+
charge, indicating partial replacement of oxygen
with a less electronegative nitrogen forming oxynitrides [53]. The
result conrms that despite reduction of the passivated catalyst at
450

C, its surface is mainly an oxynitride rather than a nitride. The


absence of nitride species on the external surface is attributed to a
higher concentration of oxygenated functionalities on the support
surface (as a result of the HNO
3
-pretreatment): negatively charged
heptamolybdate units interact more strongly with the positively
charged carbon surfaces which may have inhibited the formation
of fully nitrided groups during nitridation. In contrast to these
results, XRD analyses showed detectable nitride amounts on the
CGran support, suggesting that nitride forms inside the pores of
the support rather than on the external surface of the supported
catalyst particles. This could be related to the heterogeneity in the
distribution of oxygenated functionalities of the carbon support
during HNO
3
-treatment. Due to diffusional effects the amount of
oxygen groups inside the pores is expected to be less than those
at the exterior [25]; consequently, the inner pores of the sup-
port are less positively charged and thus adsorbed less strongly to
the heptamolybdate anion, leading to relatively easier decomposi-
tion and transformation of the precursor to nitride. This proposal
goes further by demonstrating that the decrease in support textu-
ral parameters after impregnation and thermal conversion shows
that the XRD-detected Mo
2
N, Mo
2
N
0.78
and Co
3
Mo
3
Nspecies were
I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124 119
present on internal surfaces of the particles. Thus, the results imply
difference in composition between internal and external surfaces
of the carbon support.
The XPS peaks in the N 1s region shown in Fig. 5(b) and pre-
sented in Table 4, 398.30.1, 399.6 and 400.1eV, are ascribed to
pyridine, amide and pyrrolic groups created at the edges of the
graphenelayers bynitridation[47,54]. Thecomponent at 396.5eV
is attributedtoN1s intheMo Nbond[55], conrmingthepresence
of molybdenumoxynitrides. The relative abundance of each nitro-
gen species of the catalysts indicates that monometallic catalysts
prepared via method 2 contained more N species associated with
the Mo N bonds than catalysts prepared via method 1; the result
suggests that the effectiveness of surface molybdenumoxynitride
formation was related to the nitridation method. At the same time
higher total abundance of nitrogen-functional groups of the carbon
support was obtained for the method 1 samples, compared to the
method 2 samples. This result qualitatively agrees with the bulk
nitrogen content analysis in Table 3, conrming that ammonia is
more effective at nitriding the carbon support.
The atomic Mo/C, N/C and Co/C ratios are also listed in Table 4.
The XPS atomic ratios, estimated from intensity ratios of particle
related peaks and support related peaks, have a strong dependence
on catalyst dispersion, and can qualitatively be used to compare
the dispersion of supported catalysts [56]. The differences in Mo/C
ratios areindicativeof dispersiondifferences whenthecatalysts are
prepared under different conditions. Thus, for example, the cata-
lysts preparedvia method2(NHseries) displayedhigher Mo/Cratio
than those prepared via method 1 (A-series). It can be concluded
from Table 4 that the thermal conversion of the oxide catalyst
precursor using N
2
/H
2
mixture led to more highly dispersed Mo
oxynitride particles. We can make a hypothesis on the basis of the
information from the bulk and surface concentration of nitrogen
moieties on the carbon support. Nitridation reduces the quantity of
surface oxygen groups and simultaneously increases the nitrogen
functional groups; it has been conrmed that ammonolysis is more
effective than N
2
/H
2
mixture for the nitridation of carbon supports.
Consequently, the greater quantity of negatively charged surface
nitrogen species in the method 1 samples would induce a greater
degree of partial distribution of molybdenumoxynitrides towards
the interior of the catalyst. The N/C atomic ratio, which expresses
thedispersionof thetotal nitrogenspecies (includingnitrogenfrom
oxynitride and nitrogen from organic species) on the carbon sup-
port surface, did not followany observable trend; these results are
less informativeinrelationtothedegreeof dispersionof themolyb-
denumoxynitride phase due tocontributions fromsurface nitrogen
species on the carbon support. However, the N/C atomic ratio per-
taining to the molybdenum oxynitride species can be elucidated
by multiplying the total N/C atomic ratio by the percent distribu-
tion, denoted as (N/C)
nitride
and shown in Table 4. Except for the
samples supported on Darco carbon, (N/C)
nitride
atomic ratios for
monometallic samples are higher for method 2 samples than for
method 1 samples. This result indicates the nitridation method is
anessential factor inthe control of the relative dispersionof molyb-
denumoxynitride species for carbon supported nitrides. The Co/C
ratio did not followany observable trend.
The surface N/Mo atomic ratio was evaluatedfromthe XPS Mo/C
and (N/C)
nitride
atomic ratios and is summarized in Table 4. The
calculated values were close to the stoichiometric N/Mo value of -
Mo
2
Nand -Mo
2
N
0.78
. However, because XPS results indicated the
presence of only molybdenumoxynitride, this observation is prob-
ably due to excess nitrogen residing in defect sites like irregular
grainboundary surfaces. Moreover, the N/Moatomic ratiowas sim-
ilar to each other suggesting that the method of nitridation and the
surface chemistry of the support had no observable control on the
surface Mo oxynitride species formed. This infers that the nature
of the nitrogen decient sites was the same for all the catalysts.
Fig. 6. Surface N/Mo atomic ratio (from XPS) vs. bulk N/Mo atomic ratio (from
chemical analysis) of the carbon-supported catalysts.
A plot of the surface N/Mo atomic ratio (calculated from XPS) vs.
bulk N/Mo atomic ratio (calculated from chemical analysis) after
subtracting nitrogen content on the carbon support is shown in
Fig. 6, and clearly indicates that for monometallic catalysts the bulk
remained enriched with nitrogen relative to the surface compo-
sition. This result suggests that molybdenum nitrides/oxynitrides
with higher nitrogen deciencies were located in the interior of the
catalysts, in line with the interpretation derived fromXRDand XPS.
In contrast, the CoMo nitride catalysts, except for CoMoN/CGran-A
catalyst, had higher nitrogen concentration on the external surface
than in the bulk; this result is not yet clear.
The textural and chemical surface properties of the support
could also inuence Mo dispersion as demonstrated abundantly in
the literature [24]. Indeed, the highest dispersion displayed by the
MoN/CGran-NH catalyst is attributed to the fact that this support
has the most abundant oxygen surface groups (most hydrophilic
character) and the highest mesoporosity; this facilitates access of
the aqueous solution to the internal pore structure and allows a
homogenous radial distribution of the metal precursor within the
particles of the support [23]. Conversely, the lowest dispersion of
the MoN/GCA-NH catalyst is attributed to the lowest concentra-
tionof oxygensurface groups (most hydrophobic character) as well
as the predominance of micropores in the support. Furthermore,
the intermediate dispersion of the MoN/GAC-NH catalyst corre-
lates with this solid possessing intermediate surface chemical and
textural property. It is important to remark that the presence of
oxygenfunctional groups canrender the support surface negatively
charged over a wide range of pH condition, causing electrostatic
repulsions between the support and the heptamolybdate anion
[25]. This can result in metal particles aggregation, leading to less
surface Mo atoms as evident by the low Mo/C atomic ratio of the
MoN/CGran-A catalyst.
3.4. Catalytic activity
The conversion of guaiacol and the evolution of reaction prod-
ucts are illustrated in Figs. 7 and 8. Periodic samplings of the
liquid mixture in the reactor were analyzed by GC, fromwhich the
concentration of the reactant and the product yields were deter-
mined relative to the hexadecane standard. The observed products
were similar for all the catalysts and include phenol, catechol,
120 I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124
Fig. 7. Variation of the transformation of guaiacol and the yield of products with time with (a) MoN/GCA-A, (b) MoN/GCA-NH, (c) CoMoN/GCA-A, (d) MoN/GAC-A (e)
MoN/GAC-NH, and (f) CoMoN/GAC-A catalysts.
deoxygenated compounds (such as cyclohexene, cyclohexane and
benzene), and heavy compounds (such as mono- to tetra-methyl
phenols and dimethyl catechols). Other possible products, such as
anisole, m- and o-cresol were not observed. Furthermore, CGran-
and GAC-supported catalysts produced very small quantities of
methylcatechol which were attributed to the higher total acidi-
ties of these catalysts. Fig. 7 illustrates similarities in the product
distribution for the GCA-supported Mo nitride catalysts (Fig. 7(a)
and (b)) and the GAC-supported Mo nitride catalysts (Fig. 7(d)
and (e)): guaiacol was mainly converted to phenol, while cate-
chol and the deoxygenated compounds were observed in smaller
quantities. Heavy compounds were formed in signicant amount
and are attributed to the methylation of the aromatic ring [57].
The production of heavy compounds was even more prominent in
the CoMo nitride catalysts supported on GCA and GCA, as seen in
Fig. 7(c) and (d), respectively. By comparison, the CGran-supported
catalysts exhibited some slight differences in the products distri-
bution at higher conversion, as shown in Fig. 8(a)(c): phenol was
found to be the dominant product over the MoN/CGran-NH cata-
lyst while heavy products and phenol were the main competing
products formed in almost equal amounts with the MoN/CGran-A
and CoMoN/CGran-A catalysts. As seen in Fig. 8(d)(f), the Darco-
supported catalysts showed similar behavior in the evolution of
products. The product evolution observed indicates that guaiacol
HDO followed the reaction scheme proposed by Bui et al. [57]
shown in Fig. 9. The two general pathways are: (i) initial demethy-
lation(DME) toformcatechol, followedby dehydroxylationtoform
phenol; or (ii) direct demethoxylation(DMO) toformphenol. (Light
products such as methane and methanol could not be separated
by the column used under batch conditions, but they are expected
I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124 121
Fig. 8. Variation of the transformation of guaiacol and the yield of products with time with (a) MoN/CGran-A, (b) MoN/CGran-NH, (c) CoMoN/CGran-A, (d) MoN/Darco-A, (e)
MoN/Darco-NH, and (f) CoMoN/Darco-A catalysts.
byproducts of DME and DMO, respectively.) Further deoxygenation
of phenol produces benzene, cyclohexene and cyclohexane, while
methyl-substitution of catechol forms methylcatechol. Continuous
production of catechol at longer reaction times indicated that the
conversion of catechol to phenol was not prominent although both
demethylation and direct demethoxylation occurred over these
catalysts. The same tendency was observed for bulk metal nitrides
[30].
Table 5 summarizes the catalytic activities. The initial reaction
rates calculated fromthe slopes of the guaiacol conversion curves
are given in Fig. 10. Negligible conversions were obtained using
the bare supports (not shown), indicating that the activities were
associated with the Mo oxynitride and not the support. Thus, the
surface oxygen functional groups of the activated carbon did not
participate in the conversion of guaiacol; however, the interaction
Table 5
Catalytic activities of carbon-supported Mo nitride catalysts.
Catalyst Activity
(10
6
mol g
1
catalyst
s
1
)
Intrinsic activity
(10
4
molec. Mo at
1
s
1
)
MoN/Darco-A 6.4 60.5
MoN/Darco-NH 6.5 65.1
CoMoN/Darco-A 6.2 69.6
MoN/CGran-A 4.6 46.3
MoN/CGran-NH 9.0 83.6
CoMoN/CGran-A 5.3 49.9
MoN/GAC-A 5.9 74.1
MoN/GAC-NH 7.9 75.6
CoMoN/GAC-A 2.9 33.1
MoN/GCA-A 6.8 65.8
MoN/GCA-NH 7.5 81.8
CoMoN/GCA-A 5.4 61.2
122 I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124
Fig. 9. Hydrodeoxygenation pathway of guaiacol.
Adapted fromBui et al. [57].
between the metal precursor compound and the surface groups of
the support promotes good dispersion, leading to enhanced activ-
ities. This is supported by the observed higher activities compared
to our previous results for non-modied carbon supported Mo
2
N
catalysts [27]. The reactionrate was not related to the surface N/Mo
atomic ratio, as a consequence of the catalysts possessing the same
nature of surface nitrogen decient sites discussed previously in
regard to the reported XPS N/Mo ratio data. However, the reported
activities were affected by the method of nitridation: the cata-
lysts prepared by method 2 (NH series) had consistently higher
activities. This is related to their higher Mo oxynitride dispersion
(illustratedinFig. 11), conrmedbytheMo/Cand(N/C)
nitride
atomic
ratio in Table 4. The method 2 catalysts contained a greater number
of active sites associated with molybdenumoxynitride, leading to
higher activities. Correlation of HDO activity with dispersion has
similarly been reported for sulded CoMo/carbon and reduced
Fig. 10. Reaction rates of carbon-supported Mo nitride catalysts.
NiW/carbon catalysts [22,23]. The higher HDOactivity for method
2 catalysts coupled with the reported advantages in the large-scale
synthesis of Mo nitride using a N
2
/H
2
mixture as a reactant over the
NH
3
synthesis makes method 2 attractive for potential industrial
application [34]; thus, for example, the N
2
and H
2
reactants can be
economically recycled by drying, and their use simplies handling
procedures as well as eliminate heat transfer problems associated
with endothermic decomposition of NH
3
[34].
Other trends can be observed in Fig. 10 when the catalysts
prepared using the same method but dispersed on different sup-
ports are considered. The specic activity of the method 2 catalysts
(NH series) decreased in the order: MoN/CGran-NH>MoN/GAC-
NH>MoN/GCA-NH>MoN/Darco-NH. Thus HDO activity appears
to be favored by a combination of higher Mo dispersion and
higher support mesoporosity (MoN/Darco-NH being the excep-
tion): the highest Mo oxynitride dispersion and the easiest
reactant accessibility to the supports mesoporous structure led
Fig. 11. Reaction rates vs. XPS Mo/C atomic ratio.
I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124 123
to the highest HDO activity of the MoN/CGran-NH catalyst.
The reaction rates of the method 1 catalysts (A-series) corre-
late with Mo oxynitride dispersion and decrease in the order:
MoN/GCA-A>MoN/GAC-A>MoN/Darco>MoN/CGran-A. Some of
theDarco-supportedcatalysts exhibitedinferior activity, compared
to GCA-supported catalysts, despite their higher dispersion and
higher mesoporosity. This surprising behavior could be due to an
overestimation of the Mo signal obtained by XPS. Considering that
the Darco support possessed the lowest surface area of all the sup-
ports, andthat all the catalysts were impregnatedwitha similar Mo
content, this catalyst should contain the largest Mo nitride particle
sizes. However, their measured Mo/C atomic surface ratios were
higher probably due to the inability of X-ray photons to penetrate
larger Mo nitride particles, leading to a higher intensity of the Mo
3d XPS signal and thus an overestimation of the Mo/C atomic sur-
face ratios. Similar behavior was previously observed by Lagos et al.
[58]. Therefore, the low activity of Mo nitride catalysts supported
onDarcocarbonwas probably due to the loss of active sites through
the formation of agglomerates.
The intrinsic activities based on the Mo content are also given
in Table 5. The trends were similar to the total reaction rates.
Figs. 10and11alsoshowthat the additionof Codidnot increase the
activity of the catalysts. In fact, GCA- and GAC-supported CoMoN
catalysts were not nearly as active as their MoN counterparts. In
our previous, it was shown that addition of Co did not enhance the
activityof unsupportedMo nitride catalyst [30]. This was attributed
to incomplete formation of the bimetallic nitride, Co
3
Mo
3
N, phase.
The same interpretation applies to supported catalysts as well.
Unlike bimetallic suldes and oxides, the formationof single-phase
bimetallic nitride is not trivial and often exists in multiple phases.
Synthesis of pure-phase bimetallic nitride catalysts (Co
3
Mo
3
N and
Ni
2
Mo
3
N) and its HDO catalysis are warranted. Moreover, the sur-
face Mo/Co atomic ratio (fromXPS analysis) is higher than the bulk
Mo/Co atomic ratio (from metal content analysis), indicating that
cobalt is preferentially deposited in the interior of the support and
are possibly trappedunderneathmolybdenum. This canbe takento
suggest that during impregnation cobalt migrated into the interior
of the catalyst and caused the partial migration of molybdenum
towards the exterior, in good agreement with the results obtained
byFerrari et al. [17]. This inhomogeneityinthedistributionof cobalt
could explain the diminishing inuence on activity after Co incor-
poration: a higher concentration of surface Co species could create
new (or modify) active sites which could enhance activity; con-
versely, a large proportion of Co in the pores will not only hinder
their accessibility but may also limit reactant diffusion into other
catalytic active sites located in the internal surfaces of the support
which will negatively affect reaction rates. However, a more exten-
sive analysis of promoter effect (utilizing different Co precursors
and preparative methods such as sequential vs. co-impregnation)
on activity requires further investigation.
The selectivity results are summarized in Fig. 12. In con-
trast to metal hydrogenation catalysts like Ru [4], metal sulde
and metal nitride hydrotreating catalysts have a higher selectiv-
ity for HDO reactions relative to hydrogenation of aromatic and
olenic compounds [8,30]. During conversion of guaiacol over
Mo nitride/carbon catalysts, both demethylation and demethoxy-
lation reactions take place only at the active sites situated on
the metal nitrides due to the inertness of the carbon support.
Blank experiments with all the supports resulted in negligi-
ble conversion, suggesting that the physicochemical differences
of the carbon supports had no direct inuence on the product
selectivity. The dual-pathway behavior of the Mo
2
N/carbon cat-
alytic system provides evidence that two kinds of active sites
are present on these catalysts. Theoretical studies reveal that the
surface of fresh bulk Mo
2
N consists of coordinately unsaturated
Mo and N atoms, and 4-fold type vacancies [59]. The surface
Fig. 12. Phenol/catechol ratio calculated at 10% guaiacol conversion.
nature of reduced passivated supported catalysts is inevitably
different due tothe eliminationof the weakly adsorbedNH
x
species
by passivation; thus, we infer that the surface of the carbon-
supported Mo nitride catalysts expose coordinately unsaturated
Mo atoms and 4-fold type nitrogen decient sites. Hypothesis can
be made that coordinately unsaturated Mo atoms are responsible
for demethoxylation (C
aromatic
OCH
3
bond cleavage) and dehy-
droxylationwhile demethylation(C
methyl
Obond cleavage) occurs
at nitrogen decient sites. This hypothesis is analogous to the
distinction between sites for sulde catalysts reported by Fer-
rari et al. [60]. All the catalysts displayed high phenol production
and this cannot be related directly to the surface acidic proper-
ties of the carbon supports; recently, Sepulveda et al. [61] showed
that strong acid sites favor catechol formation. The preference
for the demethoxylation pathway (C
aromatic
OCH
3
bond cleavage)
suggests that unsaturated Mo sites were predominant in the cata-
lyst. However, these high phenol/catechol ratios were signicantly
lower than those observed for unsupported nitride catalysts [30],
suggesting that the active sites on nitrides and/or oxynitrides were
modied by the support, rendering them less selective for the
demethoxylation route compared to the unsupported catalysts.
This conrms that the generation of oxygen groups on the carbon
surface has an indirect contribution on phenol/catechol selectivity
through the precursor/support interaction, modifying the nature of
the active sites and their selectivity. For example, N atoms present
on the surface of bulk Mo nitride catalysts may not be there on sup-
ported catalysts due to the metal nitridesupport interaction and
this can have an overall effect on the phenol/catechol ratio and the
amount of deoxygenated products. Fig. 12 does not reveal any clear
tendencies in the phenol/catechol ratio, suggesting that the extent
and manner of the effect of the surface chemistry of the support,
dictated by the surface oxygen groups, is unclear. Therefore, this
warrants additional research.
Fig. 13. Time-on-stream behavior of the MoN/GAC-NH catalysts in terms of total
conversion for HDOof guaiacol at 300

C, 3MPa H
2
pressure, H
2
/guaiacol ratio of 23.
124 I.T. Ghampson et al. / Applied Catalysis A: General 439440 (2012) 111124
Developing a robust HDO catalyst for pyrolysis oil upgrading
is a considerable challenge. Possible reasons for catalyst deacti-
vation during HDO include coking, poisoning and loss of active
sites through surface chemistry changes [11,12]. A preliminary
investigation of time-on-streambehavior of our nitrided catalysts,
compared to a reference commercial sulded NiMo/Al
2
O
3
catalyst,
was conducted in a continuous ow reactor. It is summarized in
Fig. 13 and agrees with the results of Monnier et al. [62]. The liquid
ow rate was chosen to obtain low conversion. The nitrided cata-
lysts displayed higher stability than the sulded catalyst after 4h
on streamunder continuous operation. The gradual deactivation of
the sulded NiMo/Al
2
O
3
catalyst could be due to loss of the sulde
phase during HDO reaction.
4. Conclusions
Use of activated carbon materials with different textural and
chemical surface properties to prepare supported Mo nitride cat-
alysts resulted in their different activities in HDO of guaiacol,
demonstrating rapid production of signicant amounts of phe-
nol. This indicated that the transformation of guaiacol proceeded
mostly through the direct demethoxylation route, bypassing the
formation of catechol. The higher specic activity of carbon-
supported catalysts prepared by using a N
2
/H
2
mixture, compared
to similarly supported Mo nitrides prepared by ammonolysis, was
attributed to a higher dispersion of Mo oxynitride. The dispersion
was related to the surface chemistry and the textural properties
of the support: high concentration of oxygen surface groups on the
support and high mesoporosity of the support promotes better dis-
persion. The most active HDO catalyst (MoN/CGran-NH) was the
one that contained highly exposed Mo species on a highly meso-
porous support. Surprisingly, a generally diminishing inuence on
activity was observed after incorporation of Co to prepare bimetal-
lic nitrided catalysts.
Acknowledgements
The authors acknowledge the nancial support of DOE Epscor
Grant #DE-FG02-07ER46373 and the nancial support fromCON-
ICYT Chile, projects PFB-27, PIA-ACT-130 and FONDECYT No.
1100512 grants. I. Tyrone Ghampson is indebted to NSF Career
Award 0547103 for sponsoring a trip to the University of Concep-
cin. The authors also gratefully acknowledge valuable discussions
on N
2
sorption with Rachel Pollock, and the technical assistance of
Nick Hill and Manuel Veliz.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.apcata.2012.06.047.
References
[1] A.V. Bridgwater, D. Meier, D. Radlein, Org. Geochem. 30 (1999) 14791493.
[2] D.C. Elliott, Energy Fuels 21 (2007) 17921815.
[3] E. Furimsky, Appl. Catal. A 199 (2000) 147190.
[4] D.C. Elliott, T.R. Hart, Energy Fuels 23 (2008) 631637.
[5] E. Laurent, B. Delmon, Appl. Catal. A 109 (1994) 7796.
[6] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, Chem. Rev. 110
(2010) 35523599.
[7] H.Y. Zhao, D. Li, P. Bui, S.T. Oyama, Appl. Catal. A 391 (2011) 305310.
[8] V.N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Appl. Catal. B101(2011) 239245.
[9] A. Gutierrez, R.K. Kaila, M.L. Honkela, R. Slioor, A.O.I. Krause, Catal. Today 147
(2009) 239246.
[10] Y.-C. Lin, C.-L. Li, H.-P. Wan, H.-T. Lee, C.-F. Liu, Energy Fuels 25 (2011) 890896.
[11] E. Laurent, B. Delmon, Appl. Catal. A 109 (1994) 97115.
[12] E. Laurent, B. Delmon, J. Catal. 146 (1994) 281291.
[13] A. Popov, E. Kondratieva, J.M. Goupil, L. Mariey, P. Bazin, J.-P. Gilson, A. Travert,
F. Mauge, J. Phys. Chem. C 114 (2010) 1566115670.
[14] A. Centeno, E. Laurent, B. Delmon, J. Catal. 154 (1995) 288298.
[15] P.E. Ruiz, K. Leiva, R. Garcia, P. Reyes, J.L.G. Fierro, N. Escalona, Appl. Catal. A
384 (2010) 7883.
[16] M. Ferrari, R. Maggi, B. Delmon, P. Grange, J. Catal. 198 (2001) 4755.
[17] M. Ferrari, B. Delmon, P. Grange, Carbon 40 (2002) 497511.
[18] V.N. Bui, D. Laurenti, P. Delichre, C. Geantet, Appl. Catal. B101 (2011) 246255.
[19] A. Centeno, V. CH, R. Maggi, B. Delmon, in: A.V. Bridgwater (Ed.), Developments
in Thermochemical Biomass Conversion, Blackie Academic &Professional, Lon-
don, 1997, pp. 602610.
[20] A. Centeno, O. David, C. Vanbellinghen, R. Maggi, B. Delmon, in: A.V. Bridgwater
(Ed.), Developments inThermochemical Biomass Conversion, BlackieAcademic
& Professional, London, 1997, pp. 589601.
[21] G. de la Puente, A. Centeno, A. Gil, P. Grange, J. Colloid Interface Sci. 202 (1998)
155166.
[22] S. Echeandia, P.L. Arias, V.L. Barrio, B. Pawelec, J.L.G. Fierro, Appl. Catal. B 101
(2010) 112.
[23] G. de la Puente, A. Gil, J.J. Pis, P. Grange, Langmuir 15 (1999) 58005806.
[24] L.R. Radovic, F. Rodriguez-Reinoso, in: P.A. Thrower (Ed.), Chem. Phys. Carbon,
Marcel Dekker, 1997, pp. 243358.
[25] G. de la Puente, J.A. Menendez, Solid State Ionics 112 (1998) 103111.
[26] T. Nimmanwudipong, R. Runnebaum, D. Block, B. Gates, Catal. Lett. 141 (2011)
779783.
[27] C. Seplveda, K. Leiva, R. Garca, L.R. Radovic, I.T. Ghampson, W.J. DeSisto, J.L.G.
Fierro, N. Escalona, Catal. Today 172 (2011) 232239.
[28] S.T. Oyama, Catal. Today 15 (1992) 179200.
[29] J.G. Chen, Chem. Rev. 96 (1996) 14771498.
[30] I.T. Ghampson, C. Sepulveda, R. Garcia, B.G. Frederick, M.C. Wheeler, N. Escalona,
W.J. DeSisto, Appl. Catal. A 413414 (2012) 7884.
[31] Y. Li, Y. Zhang, R. Raval, C. Li, R. Zhai, Q. Xin, Catal. Lett. 48 (1997) 239245.
[32] J.W. Logan, J.L. Heiser, K.R. McCrea, B.D. Gates, M.E. Bussell, Catal. Lett. 56(1998)
165171.
[33] L. Volpe, M. Boudart, J. Solid State Chem. 59 (1985) 332347.
[34] R.S. Wise, E.J. Markel, J. Catal. 145 (1994) 344355.
[35] J.P. Olivier, Carbon 36 (1998) 14691472.
[36] M. Kruk, M. Jaroniec, K.P. Gadkaree, J. Colloid Interface Sci. 192 (1997) 250256.
[37] C.D. Wagner, L.E. Davis, M.V. Zeller, J.A. Taylor, R.H. Raymond, L.H. Gale, Surf.
Interface Anal. 3 (1981) 211225.
[38] R. Cid, G. Pecchi, Appl. Catal. 14 (1985) 1521.
[39] H.P. Boehm, Carbon 32 (1994) 759769.
[40] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. rfo, Carbon 37 (1999)
13791389.
[41] Y. Otake, R.G. Jenkins, Carbon 31 (1993) 109121.
[42] U. Zielke, K.J. Httinger, W.P. Hoffman, Carbon 34 (1996) 983998.
[43] J.M. Calo, D. Cazorla-Amors, A. Linares-Solano, M.C. Romn-Martnez, C.S.-M.
De Lecea, Carbon 35 (1997) 543554.
[44] P.E. Fanning, M.A. Vannice, Carbon 31 (1993) 721730.
[45] G.M. Dolce, P.E. Savage, L.T. Thompson, Energy Fuels 11 (1997) 668675.
[46] J.P.R. Vissers, S.M.A.M. Bouwens, V.H.J. de Beer, R. Prins, Carbon 25 (1987)
485493.
[47] S. Biniak, G. Szymanski, J. Siedlewski, A. Swiatkowski, Carbon 35 (1997)
17991810.
[48] S.D. Gardner, C.S.K. Singamsetty, G.L. Booth, G.-R. He, C.U. Pittman, Carbon 33
(1995) 587595.
[49] E. Riedo, F. Comin, J. Chevrier, F. Schmithusen, S. Decossas, M. Sancrotti, Surf.
Coat. Technol. 125 (2000) 124128.
[50] E. DAnna, M.L. De Giorgi, A. Luches, M. Martino, A. Perrone, A. Zocco, Thin Solid
Films 347 (1999) 7277.
[51] S.W. Yang, C. Li, J. Xu, Q. Xin, J. Phys. Chem. B 102 (1998) 69866993.
[52] J.S. Girardon, E. Quinet, A. Griboval-Constant, P.A. Chernavskii, L. Gengembre,
A.Y. Khodakov, J. Catal. 248 (2007) 143157.
[53] M.L. Kaliya, S.B. Kogan, Catal. Today 106 (2005) 9598.
[54] R.J.J. Jansen, H. van Bekkum, Carbon 33 (1995) 10211027.
[55] K. Hada, M. Nagai, S. Omi, J. Phys. Chem. B 105 (2001) 40844093.
[56] A.M. Venezia, Catal. Today 77 (2003) 359370.
[57] V.N. Bui, G. Toussaint, D. Laurenti, C. Mirodatos, C. Geantet, Catal. Today 143
(2009) 172178.
[58] G. Lagos, R. Garca, A.L. Agudo, M. Yates, J.L.G. Fierro, F.J. Gil-Llambas, N.
Escalona, Appl. Catal. A 358 (2009) 2631.
[59] G. Frapper, M. Pelissier, J. Hafner, J. Phys. Chem. B 104 (2000) 1197211976.
[60] M. Ferrari, S. Bosmans, R. Maggi, B. Delmon, P. Grange, Catal. Today 65 (2001)
257264.
[61] C. Seplveda, N. Escalona, R. Garcia, D. Laurenti, M. Vrinat, First International
Congress on Catalysis for Bioreneries (CatBior), Torremolinos-Mlaga, Spain,
2011.
[62] J. Monnier, H. Sulimma, A. Dalai, G. Caravaggio, Appl. Catal. A 382 (2010)
176180.