Iron(II) disulfides as precursors of highly selective catalysts for hydrodeoxygenation of dibenzyl ether into toluene

Article abstract of DOI:10.1002/cctc.201500041

In this report, we show that nanocrystalline pyrite and marcasite (FeS₂), supported on SBA-15, aerosil SiO₂, activated carbon or Al₂O₃, are precursors of highly active catalysts for the hydrodeoxygenation of dibenzyl ether into toluene. High yields of toluene (up to 100 %) were achieved in experiments performed at 250 C under initial H₂ pressure of 100 bar for 2 h. In the recycling experiments, results from XRD and XPS analyses indicate that a fresh surface, formed upon the chemical transformation of FeS₂ into Fe_(1-x)S, is responsible for the high activity and high selectivity achieved in the conversion of dibenzyl ether into toluene.

Full text of DOI:10.1002/cctc.201500041

DOI: 10.1002/cctc.201500041
Full Papers
Iron(II) Disulfides as Precursors of Highly Selective
Catalysts for Hydrodeoxygenation of Dibenzyl Ether into
Toluene
[
a]
Na Ji, Xingyu Wang, Claudia Weidenthaler, Bernd Spliethoff, and Roberto Rinaldi*
In this report, we show that nanocrystalline pyrite and marca-
site (FeS ), supported on SBA-15, aerosil SiO , activated carbon
cycling experiments, results from XRD and XPS analyses indi-
cate that a fresh surface, formed upon the chemical transfor-
2
2
or Al O , are precursors of highly active catalysts for the hydro-
mation of FeS into Fe S, is responsible for the high activity
2 (1ꢀx)
2
3
deoxygenation of dibenzyl ether into toluene. High yields of
toluene (up to 100%) were achieved in experiments performed
and high selectivity achieved in the conversion of dibenzyl
ether into toluene.
at 2508C under initial H pressure of 100 bar for 2 h. In the re-
2
Introduction
Iron(II) disulfides (i.e. pyrite or marcasite) are well known to en-
hance the reactivity of coals towards liquefaction under elevat-
ed H2 pressure (100–300 bar) at high temperatures (350–
[
1,2]
4508C).
In addition, there are also a few reports to suggest
that sulfided iron catalysts facilitate the liquefaction of techni-
[
3–6]
cal lignins under high severity conditions.
Analyzing the key
studies published over the last 50 years, one can easily identify
two main classes of FeꢀS catalysts that have been explored for
coal and lignin liquefaction. The first class comprises bulk
2
ꢀ1
iron(II) disulfides of low surface area (<2 m g ), which have
[1,2,6]
been almost exclusively applied to direct coal liquefaction.
The second class encompasses sulfided iron species formed
in situ by reactions of an iron compound [e.g. ferrocene,
Fe(NO ) ] and a sulfur source (e.g. CS , S ). These FeꢀS materi-
3
3
2
8
[
4,5]
als were explored for lignin liquefaction.
Owing to the low
surface area of bulk iron(II) disulfides, and the undefined chem-
ical nature of sulfided iron catalysts formed in situ, the factors
accounting for the catalytic activity of both classes of FeꢀS ma-
terials remain unclear. Surprisingly, despite evidence demon-
strating the potential of these materials for liquefaction of
lignin and coal, the relationships between structure and cataly-
sis of (bulk) iron(II) disulfides have been attracted little atten-
tion. Indeed, most knowledge in this field originates from re-
Figure 1. Sketch of the crystalline structure of a) pyrite, b) marcasite, and
c) pyrrhotite (based on the NiAs unit cell).
[
2,7,8]
ports dated back to the 1980s.
More recently, some in-
depth computational studies of pyrite surface and structure
[
9–15]
were reported.
From a catalysis perspective, therefore, re-
visiting this research field with modern techniques for prepara-
tion and characterization of highly dispersed iron(II) disulfide
phases on supports exhibiting high surface area could contrib-
ute insights as to whether these materials can be useful as cat-
alysts or catalyst precursors in the conversion of biomass-de-
rived aromatic streams.
Pyrite is a polymorph of iron(II) disulfide (Figure 1a) crystal-
[16]
2ꢀ
lized in a cubic crystal structure. It comprises discrete S2
units with an SꢀS distance (2.195 ꢀ) similar to that expected
[
16,17]
for an SꢀS single bond (2.00 to 2.11 ꢀ).
Thus, pyrite is
[
a] Dr. N. Ji, X. Wang, Dr. C. Weidenthaler, B. Spliethoff, Dr. R. Rinaldi
Max-Planck-Institut fꢀr Kohlenforschung
Kaiser-Wilhelm-Platz 1, Mꢀlheim an der Ruhr (Germany)
Fax: (+49)208-306-2995
2+ 2ꢀ [17,18]
properly described as Fe
S
.
Marcasite is another poly-
2
morph of iron(II) disulfide with an orthorhombic crystal struc-
[16,17]
ture (Figure 1b).
Under H pressure at temperatures above
2
E-mail: rinaldi@kofo.mpg.de
2008C, pyrite releases H S and converts into a class of non-stoi-
2
ChemCatChem 2015, 7, 960 – 966
960
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
III
chiometric iron(II) sulfides, called pyrrhotites (Fe(1ꢀx)S where 0ꢁ
xꢁ0.125).
Pyrrhotites possess a NiAs-type structure (Figure 1c), which
2 h in a quartz tube. For the Fe /AC material, thermal treat-
[
2,8]
ment was performed under argon flow. In sequence, the air or
argon atmosphere was switched by an H S atmosphere. The
2
[17,19,20]
can exhibit monoclinic or hexagonal crystal symmetry.
material was sulfided at 4008C for an additional 2.5 h. This pro-
III
II
One can describe the monoclinic pyrrhotite by subtracting
one-eighth iron atoms from the hexagonal FeS structure, so
that the resulting structure contains alternating layers of full Fe
cedure led to the reduction of Fe to Fe as evidenced by the
formation of elemental sulfur, which sublimated and deposited
as a pale yellow layer on the cold exit of the quartz tube. Im-
portantly, the formation of iron(II) disulfides was also con-
firmed by X-ray powder diffraction (XRD). XRD patterns in
Figure 2 indicate that the sulfidation procedure renders pyrite
[
19,20]
sites and layers of Fe sites with vacancies.
The monoclinic
structure can be converted to hexagonal at temperatures
II
above 3008C. The fast diffusion of Fe vacancies at these tem-
[19,20]
peratures creates a random vacancy distribution.
As
a result, the phase transition is reversible. However, at temper-
atures lower than 3008C, the diffusion of vacancies is inhibited,
thereby, metastable hexagonal pyrrhotite structures can be
[
28]
locked by a quenching process. Troilite, the iron-rich end-
member of the pyrrhotites, is an ideal or nearly stoichiometric
[
19,20]
iron(II) sulfide (Fe( S where 0ꢁxꢁ0.05).
It shows a NiAs
1ꢀx)
derived-structure with hexagonal symmetry, which is stable at
room temperature (RT).
The role of each iron(II) sulfide and iron(II) disulfide in the
coal or lignin liquefaction processes is only partially under-
[
2,7,8]
stood.
Although pyrrhotite (prepared in advance) shows
a lower activity than pyrite in coal liquefaction, pyrrhotite
formed in situ was demonstrated to possess ‘very good’ cata-
[
8]
lytic activity. Kodaira et al. surmised that the formation of the
fresh surface with the chemical change from pyrite to pyrrho-
[
8]
tite is the cause of the high activity of pyrite. In turn, Monta-
no et al. demonstrated that troilite shows lowest activity
[
2]
among Fe( S structures. Moreover, Baldwin and Vinciguerra
1ꢀx)
Figure 2. XRD patterns of FeS
bulky FeS (pyrite).
2 2
supported on SBA-15, SiO2, AC, Al O3, and
suggested that H S, formed upon the reduction of FeS2 to
2
2
Fe( S, could be an active homogeneous catalyst for coal liq-
1ꢀx)
uefaction.
[
7]
In this report, we present the synthesis, characterization, and
catalytic performance of supported nanocrystalline iron(II) di-
(major product) and marcasite (minor product) irrespective of
the support.
sulfides supported on SBA-15, aerosil SiO , activated carbon
In previous studies on liquefaction of coal in the presence of
bulk pyrite and other iron(II) sulfides, no information was pro-
vided about the material history concerning its handling under
2
(
AC), or g-alumina (Al O ). The purpose of this work is to deter-
2 3
mine whether FeS materials can act as catalysts or catalyst
2
[2,7,8]
precursors for the hydrodeoxygenation (HDO) of dibenzyl
ether into toluene under conditions of lower severity (100 MPa
air.
Nonetheless, it is known that finely divided iron(II) sul-
fide powders may oxidize to iron oxides and (thio)sulfates
[17,21–23]
H , 2508C) than those applied for coal or lignin liquefaction
2
when exposed to humid air.
To verify whether FeS /AC is
2
(
100–350 MPa H , 350–4508C). This report is organized in three
an air-sensitive material, the fresh sulfided material was ex-
posed to air for 24 h. Figure 3 shows that the XRD reflections
characteristic to pyrite and marcasite disappeared from the
2
sections. First, the synthesis and characterization of the struc-
tural features of the supported nanocrystalline FeS materials
2
are presented. Second, the catalytic properties of the materials
in the HDO of dibenzyl ether are examined under varying con-
ditions. Third, in the recycling experiments, insight into the
FeꢀS species involved in the catalysis of HDO of dibenzyl ether
is provided by XRD and XPS analyses.
XRD pattern of the FeS /AC sample exposed to air. This result
2
demonstrates that the nanocrystalline FeS phases are air-sen-
2
sitive. As will be presented in the next section, this material
also shows a poor catalytic performance, compared with FeS2/
Results and Discussion
Synthesis and Characterization of FeS Materials
2
To synthesize FeS supported on SBA-15, SiO , AC, or Al O , in-
2
2
2
3
cipient impregnation of the supports with an iron(III) nitrate
solution was performed in order to provide a nominal loading
of iron(III) of 10 wt%. The materials were calcined at 4008C for
2
Figure 3. XRD patterns of FeS /AC ‘as sulfided’ and exposed to air for 24 h.
ChemCatChem 2015, 7, 960 – 966
www.chemcatchem.org
961
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
of surface area, as well as pore
volume, about 25 to 35% small-
er than the respective supports.
Catalytic properties of support-
ed FeS materials
2
To find the conditions of lowest
severity needed for the catalytic
conversion of dibenzyl ether, ex-
periments (Figure 5) with FeS2/
AC varying a) temperature (150–
2508C), b) initial pressure of H2
(
15–100 bar), and c) reaction
time (0–4 h) were performed.
Summarized in Figure 5 are the
results of the optimization of the
reaction parameters in the cata-
lytic conversion of dibenzyl
ether with FeS /AC suspended in
2
cyclohexane (solvent). Figure 5
shows that full conversion of di-
benzyl ether into toluene is ach-
ieved at a temperature of 2508C
2 2 2 2 2 2
Figure 4. TEM images of a) FeS /SBA-15, b) FeS /SiO , c) FeS /AC, and d) FeS /Al O3. The insets provide a thumbnail
view for representative material particles. Scale bars=100 nm (inset=200 nm).
under a H pressure of 100 bar
2
in 2 h. Accordingly, we chose
AC handled under argon. Accordingly, material transfer and
other procedures described here were performed under argon.
Table 1. Textural properties of the supports and FeS
materials.
2
supported
Displayed in Figure 4 are the TEM images of FeS /SBA-15,
2
Entry
Material
S
BET
2
Pore volume
Pore diameter
[nm]
ꢀ1
3
ꢀ1
FeS /SiO , FeS /AC, and FeS /Al O . The average size of individ-
[m g
]
[cm g
]
2
2
2
2
2
3
ual FeS particles is in the range of 5–20 nm for FeS /SBA-15,
2
2
1
2
3
4
5
6
7
8
SBA-15
814
791
424
334
626
480
122
109
0.96
0.96
0.76
0.57
0.20
0.13
0.25
0.17
5.2
4.1
5.3
5.3
3.7
3.5
6.2
5.3
5
–10 nm for FeS /SiO , 10–20 nm for FeS /AC, and 5–35 nm for
FeS /SBA-15
2
2
2
2
FeS /Al O . In general, the TEM images show that FeS particles
SiO
2
2
2
3
2
FeS
AC
2
/SiO
2
are quite well-dispersed on the supports. However, for FeS2/
SBA-15 (Figure 4a), it is apparent that FeS is poorly dispersed.
2
FeS
2
/AC
Moreover, EDX analyses indicate that the Fe:S atom ratio is
nearly 1:2 in FeS /SBA-15, FeS /SiO and FeS /Al O , but 1:3 in
2
Al O
3
2 2 3
FeS /Al O
2
2
2
2
2
3
FeS /AC. The sulfidation procedure resulted in the formation of
2
S . However, AC itself is reported to oxidize H S to S at high
8
2
8
[
24,25]
temperatures.
A part of the S content stays entrapped in
these conditions to further evaluate the catalytic performance
of the other materials.
8
the pore structure of AC, as determined by XPS analysis (Fig-
ure 7c).
Listed in Table 2 are the results obtained from blank experi-
ments and catalyst screening. Regarding the blank experi-
ments, Entry 1 shows that a 1% conversion of dibenzyl ether
was achieved in a blank experiment in absence of catalyst. This
result indicates that the substrate is thermally stable at 2508C.
This observation agrees with previous studies on the kinetics
of thermolysis of dibenzyl ether, which showed that the reac-
tion takes place at appreciable rates at temperatures higher
Summarized in Table 1 are the textural properties of the sul-
fided materials and their parent supports. FeS /SBA-15 and
2
FeS /Al O display values of surface area 3 to 11% smaller than
2
2
3
the respective supports. The small decrease in the surface area
is in line with the TEM visualization of the FeS particles partial-
2
ly filling the pore system of SBA-15 and Al O supports (Figure-
2
3
s 4a and 4d) in an inhomogeneous manner. This observation
[26]
is particularly notable for FeS /SBA-15 in which only some
than 3508C. However, entry 2 shows that a 29% conversion
of dibenzyl ether into toluene (carbon selectivity, 38%), benzyl
alcohol (23%) and benzaldehyde (41%) was reached in a blank
experiment conducted in the presence of unloaded AC. The
presence of low levels of benzaldehyde seems to be related to
the presence of (Brønsted) acid sites in AC, as benzaldehyde is
2
channels of SBA-15 are loaded with FeS (Figure 4a). This fact
2
seems to be the reason for no change in the pore volume of
SBA-15 materials to be detected by N physisorption analysis.
2
In contrast, FeS particles are better dispersed on AC and SiO2
2
(
Figures 4b and 4c). Hence, FeS /AC and FeS /SiO have values
2 2 2
ChemCatChem 2015, 7, 960 – 966
www.chemcatchem.org
962
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
transfer should take place from phenylmethanolate to phenyl-
methylium, resulting in the formation of benzaldehyde and tol-
uene.
Interestingly, the experiments performed with FeS support-
2
ed on different materials achieved high conversion of dibenzyl
ether into arenes (entries 3–8). On the one hand, FeS /AC and
2
FeS /Al O were both highly active and highly selective cata-
2
2
3
lysts, leading to very high to full conversion of dibenzyl ether
into toluene (entries 3 and 6). In contrast, FeS /SBA-15 and
2
FeS /SiO were also highly active catalysts, but less selective to
2
2
toluene than FeS /AC and FeS /Al O . Indeed, in the presence
2
2
2
3
of FeS /SBA-15 or FeS /SiO , the conversion of dibenzyl ether
2
2
2
also led to 1,2-diphenylethane, which was formed in carbon
yields close to 15% (entries 4 and 5).
Figure 5. Optimization of the reaction parameters in the conversion of
dibenzyl ether in the presence of FeS /AC. a) Reaction temperature (initial
pressure of H : 100 bar, reaction time: 2 h); b) initial pressure of H (reaction
temperature: 2508C; reaction time: 2 h); c) reaction time (initial pressure of
: 100 bar, reaction temperature: 2508C) on the HDO of dibenzyl ether.
Cyclohexane was used as solvent. The reaction time was recorded upon
2
As previously discussed, the supported nanocrystalline FeS2
phases are air-sensitive. These materials decompose, forming
an amorphous material (Figure 3). The experiment performed
2
2
H
2
in the presence of a FeS /AC sample exposed to air for 24 h
2
ꢀ
1
reaching the aforementioned temperatures (heating rate 108Cmin ).
(
FeS /AC-Air, entry 7) resulted in a 53% conversion of dibenzyl
2
ether to a mixture comprising toluene (carbon selectivity,
0%), benzyl alcohol (7%) and benzaldehyde (23%). These re-
7
Table 2. Conversion of dibenzyl ether in the absence and presence of
different iron catalysts.
sults compare to those obtained from the experiment with
Fe O /AC (Table 2, entry 8). Nonetheless, there are two features
[
a]
2
3
distinguishing FeS /AC-Air from Fe O /AC: 1) the lower carbon
2
2
3
Entry Material Conv.
Carbon selectivity [%]
selectivity to toluene (70% with FeS /AC-Air vs. 78% with
2
Fe O /AC), and 2) the higher carbon selectivity to benzalde-
2
3
[
%]
1
hyde (23% with FeS /AC-Air vs. 10% with Fe O /AC).
2
2
3
1
2
3
4
5
6
7
8
None
AC
0.7
38
0.5
23
0
0
0
0
7
12
0
0
0
16
12
0
1.4
41
0
0
0
0
Recyclability of the FeS /AC catalyst
2
29
98
FeS
FeS
FeS
FeS
FeS
2
2
2
2
2
/AC
100
80
87
98
70
78
Displayed in Table 3 are the results obtained from the experi-
ments of catalyst recycling over three runs, and after the cata-
lyst resulfidation (RSD). The conversion of dibenzyl ether stead-
ily decreased (from 98 to 72%) throughout the recycling ex-
periments (entries 1–3). In addition, the selectivity to toluene
slightly reduced (from 100 to 97%). After RSD of the spent cat-
/SBA-15 98
/SiO
/Al
2
98
95
2
O
3
/AC-Air 53
/AC 50
0
0
23
10
Fe
2
O
3
[
a] Reaction conditions: dibenzyl ether (2.5 mmol), n-decane (0.7 mmol,
internal standard for GC), catalyst (0.15 g) and cyclohexane (15 mL, sol-
vent) under 100 bar H (measured at RT), 2508C for 2 h.
alysts with H S at 4008C for 2 h, an 89% conversion of dibenzyl
2
2
ether into toluene was achieved. This result shows that the cat-
alytic performance can be partially recovered by RSD.
To assess whether structural changes are occurring in the
catalyst on its reuse, the samples of the spent and the resulfid-
ed catalyst were analyzed by XRD. Figure 6 shows the XRD pat-
not a typical product of metal-catalyzed hydrogenolysis of di-
[
27–29]
benzyl ether,
the ether bond.
but a product of acid-catalyzed cleavage of
terns, and reveals that the FeS phases were converted into
2
[
30]
Fe( S upon the first reaction run. No considerable structural
1ꢀx)
Shown in Scheme 1 is a proposed pathway for the formation
of benzaldehyde and toluene by acid catalysis (“A” stands for
an acid site). The proposed pathway involves the formation of
phenylmethylium, upon cleavage of the ether bond. A hydride
changes could be detected in the XRD pattern of the catalyst
nd
rd
samples of the 2 and 3 runs. The resulfidation of the spent
catalyst could regenerate the FeS phases, but only part of the
2
catalytic performance (Table 3, entry 4).
In the presence of H2 at temperatures above
2
008C, pyrite is converted into pyrrhotites (Fe
S
1ꢀx)
(
[2,7,8]
where 0ꢁxꢁ0.125) upon releasing of H S.
In the
2
conversion of biomass using metal sulfide catalysts,
the addition of a sulfidizing agent to the stream is
a common practice in order to keep the high level of
[31,32]
catalytic performance.
Hypothetically, the addi-
tion of H S could shift the equilibrium towards pyrite.
Scheme 1. Proposed reaction pathway for the formation of benzaldehyde and toluene
2
from dibenzyl ether. “A” stands for a Brønsted or Lewis acid site.
Should FeS (pyrite) or a sulfur-richer member of the
2
ChemCatChem 2015, 7, 960 – 966
www.chemcatchem.org
963
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
2
Table 3. Performance of the FeS /AC catalyst in the recycling experi-
[
a]
ments conducted on dibenzyl ether.
Entry
Reaction
Conv. [%]
Carbon selectivity [%]
1
2
3
4
1st
2nd
3rd
RSD/4th
98
84
72
89
100
99
97
0
1
3
0
100
[
a] Reaction conditions: dibenzyl ether (2.5 mmol), n-decane (0.7 mmol,
internal standard for GC), FeS /AC (0.15 g) and cyclohexane (15 mL, sol-
vent) under 100 bar H (measured at RT), 2508C for 2 h.
2
2
pyrrhotites be the active phase for HDO of dibenzyl ether, the
addition of H S would be favorable to the reaction per-
2
formance. Moreover, should H S be a homogeneous catalyst
2
Figure 6. XRD patterns of FeS /AC catalyst (as prepared), spent catalysts, and
2
for the HDO of dibenzyl ether, the addition of H S could also
the resulfided catalyst.
2
improve the overall reaction performance.
To verify these hypotheses, an experiment with FeS /AC
2
under a partial pressure of 1 bar H S and 100 bar H was per-
least four different sulfur species. The main line at 162.7 eV is
2
2
2
ꢀ
formed. We found that the presence of H S has no inhibitory
characteristic of S2 bulk species (pyrite or marcasite, FeS ).
2
2
2ꢀ
effect on the formation of Fe( S. However, the catalytic re-
The peak at 161.6 eV is consistent with S species that can
1ꢀx)
2
ꢀ
[22,33,34]
sults showed a decrease (from 98 to 81%) in the conversion of
belong either to bulk sulfides or to S surface sites.
The
[
22,33,34]
dibenzyl ether. Furthermore, the selectivity to toluene dropped
signal at 164.2 eV is assigned to S8,
suggesting that part
(
from 100 to 72%), while the selectivity to benzyl alcohol
of the elemental sulfur content remained in the pore structure
of AC. Finally, there is a peak at around 168 eV, which indicates
reached 22%.
2
ꢀ
II
[22,33,34]
In an attempt to characterize the transformation of the cata-
lyst surface after the HDO of dibenzyl ether, the Fe2p and S2p
the presence of SO4 species associated with Fe ions.
Again, after the first reaction run, the S2p XP spectrum also
changed considerably. Figure 7d displays that the signals at
core-level X-ray photoemission (XP) spectra of the ‘as prepared’
st
2ꢀ
2ꢀ
and spent FeS /AC (after 1 reaction run) were collected. Fig-
161.4 eV (S ) and at 163.9 eV (S ) increased, while S
bulk
2
8
2
ure 7a shows the Fe2p XP spectrum of the freshly sulfided cat-
alyst sample. This XP spectrum
species (from pyrite or marcasite) decreased in intensity. In ad-
displays photoemission (PE)
peaks typical of FeS compounds
2
with the binding energy peak of
[
22,33,34]
Fe2p3/2 at 707.1 eV.
De-
spite the careful handling of the
catalyst sample under argon, the
low-intensity line at ꢂ711 eV
suggests the presence of a small
II
population of Fe sites consis-
tent with hydrated iron(II) oxide
[
22,33,34]
or iron(II) sulfate.
After the
first reaction run, Fe2p XP spec-
trum changed significantly. Fig-
ure 7b shows that the narrow
signal at 707 eV markedly de-
creased in intensity. The Fe2p
XP spectrum is dominated by
the peak at about 711 eV.
Shown in Figure 7c is the S2p
XP spectrum for the freshly pre-
Figure 7. XPS spectrum of the as prepared FeS
catalyst, b) Fe2p spectrum of the used catalyst, c) S2p spectrum of the as prepared FeS /AC, d) S2p spectrum of
2
/AC and the used catalyst: a) Fe2p spectrum of the as prepared
pared catalyst FeS /AC. The
2
2
spectrum is a convolution of at the used catalyst.
ChemCatChem 2015, 7, 960 – 966
www.chemcatchem.org
964
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
2ꢀ
dition, the signal for SO4 was also more pronounced in the
XP spectrum of the spent catalyst than that found for the as-
prepared catalyst.
Table 4. Typical weight information for the synthesis of catalysts with
a Fe loading of 10 wt%.
Entry
Support
Weight of
support [g]
Weight of Fe(NO
·9H O [g]
3
)
3
Volume of
Water [mL]
Overall, the current results demonstrate the formation of
2
a fresh surface upon the conversion of FeS into Fe S to be
2
(1ꢀx)
1
2
3
4
AC
1.00
1.00
1.00
1.00
0.72
0.72
0.72
0.72
3.0
0.9
2.3
6.5
the cause of the high activity of pyrite in the HDO of dibenzyl
ether to toluene. Unfortunately, XPS results are not very specif-
ic. We found that there is a partial desulfurization of the parti-
cle surface, which creates a complex chemical environment for
Al
SiO
SBA-15
2
O
3
2
Fe- and S-species. Importantly, the addition of H S interferes in
2
the formation of the non-stoichiometric pyrrhotite surface re-
sponsible for the HDO of dibenzyl ether.
box. All procedures involving supported nanocrystalline FeS mate-
rials were performed in a glove box under argon.
2
Characterization methods
Conclusions
Powder X-ray diffraction (XRD) patterns of the samples were ob-
tained by using a STOE STADIP transmission diffractometer operat-
ed at 50 kV and 40 mA, using monochromatized Mo-Ka1 radiation
and a position sensitive detector.
We demonstrated HDO of dibenzyl ether, in the presence of
supported nanocrystalline FeS materials, to render high yields
2
of toluene. The recycling experiments showed that the catalyst
exhibited relatively good reusability. Throughout the recycling
experiments, there is a gradual loss in the conversion of diben-
zyl ether and in the selectivity to toluene. However, the spent
catalyst could be partially regenerated by resulfidation with
Transmission electron microscopy (TEM) images were taken by
using a Hitachi HF-2000 FE transmission electron microscope oper-
ating at a voltage of 200 kV. Energy-dispersive X-ray spectroscopy
(EDX) analysis was performed by a Noran System Six EDX with a Si
H S at 4008C for 2 h. Importantly, XRD and XPS results show
2
(Li) detector.
that the active phase is not pyrite (FeS ), but a freshly formed
2
N sorption experiments were performed by using an ASAP 2000
2
(
unknown) surface of Fe( S. Accordingly, there is a clear need
xꢀ1)
surface area analyzer (Micromeritics). The specific surface areas of
samples were determined using the Brunauer–Emmett–Teller (BET)
method. The pore volume and pore size distribution were derived
from the desorption profiles of the isotherms using the Barrett–
Joyner–Halanda (BJH) method.
for detailed studies on the surface science of iron(II) (di)sulfides
and their interaction with hydrogen and with oxygenates. Evi-
dent lines of investigations for the future are high-pressure
XPS studies and XRD analysis under operando conditions to
gain insight into the functionality of the surface structure of
the real ‘FeꢀS’ catalyst. Currently, studies are in progress to
evaluate the applicability of the pyrite catalyst precursor in the
X-ray photoelectron spectroscopy (XPS) measurements were per-
formed by using a Kratos HSi spectrometer with a hemispherical
analyzer. The monochromatized AlKa X-ray source (1486.6 eV) was
operated at 15 kV and 15 mA. For the narrow scans, a pass energy
of 40 eV was applied. The hybrid mode was used as lens mode.
The base pressure during the experiment in the analysis chamber
[
35]
conversion of lignin and lignin-derived phenolics.
ꢀ7
was 4ꢂ10 Pa. To account for charging effects, all spectra are re-
ferred to C1s XP emission at 284.5 eV.
Experimental Section
Chemicals
Dibenzyl ether (Aldrich, 99%), n-decane (Aldrich, +99%), cyclohex-
ane (Aldrich, +99.9%), SiO2 (Silica gel, Aldrich, product number
Hydrogenolysis of dibenzyl ether
2
36810) and Al O (Aldrich product number 544833) were used as
Dibenzyl ether (2.5 mmol), catalyst (0.15 g) and 15 mL solvent were
placed into a batch reactor (36 mL) under argon (glove box). After
2
3
purchased. SBA-15 support was synthesized by a method reported
in ref. [36]. Activated carbon (AC) was supplied by NORIT company
purging the reactor with H , the reaction vessel was loaded with
2
(
Product name: DARCO MRX). To exclude any possible influence of
100 bar H (258C). The experiments were performed at 2508C for
2
metal impurities on the commercial carbon support, the activated
carbon was treated with nitric acid solution (33 wt%) at 808C for
2 h (the reaction time was recorded upon reaching 2508C (the
time required for increasing the temperature from 25 to 2508C
was 30 min) under mechanical stirring (300 rpm). The products
were analyzed by GC-MS and GC-FID.
2
1
4 h, and then washed with deionized water until pHꢂ7, dried at
208C overnight for further use.
The carbon yield and selectivity was determined as given by Equa-
tion (1) and (2), respectively:
Preparation of the supported nanocrystalline FeS catalysts
2
nproduct ꢃ N
productcarbon
Yieldð%Þ ¼ nsubstrate
Selectivityð%Þ ¼
ꢃ 100
ꢃ 100
ð1Þ
ð2Þ
The supported nanocrystalline FeS catalysts were prepared via in-
ꢃ Nsubstratecarbon
2
cipient wetness impregnation method. The supports (AC, Al O3
2
Yieldproduct
Conversionsubstrate
SiO or SBA-15) were impregnated with an iron(III) nitrate solution
2
(
Table 4), dried at 1208C for 12 h. In sequence, the material was
ꢀ1
sulfided with H S (50 mLmin ) at 4008C for 2.5 h. Finally, the
2
sample was cooled down to RT under argon. The supported FeS2
materials (air-sensitive) were stored in closed vials inside a glove
where: nproduct is the amount of product (in mmol) determined by
GC-FID, Nproduct carbon is the number of carbon atoms in a product,
ChemCatChem 2015, 7, 960 – 966
www.chemcatchem.org
965
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
nsubstrate is the amount of remaining substrate (mmol), Nsubstrate carbon
[12] K. Ramesha, R. Seshadri, C. Ederer, T. He, M. A. Subramanian, Phys. Rev. B
2
004, 70.
is the number of carbon atoms in the substrate.
[
[
13] R. S. Sun, M. K. Y. Chan, G. Ceder, Phys. Rev. B 2011, 83.
14] C. H. Zhao, J. H. Chen, B. Z. Wu, X. H. Long, Trans. Nonferrous Met. Soc.
China 2014, 24, 491–498.
15] K. M. Rosso, Rev. Mineral. Geochem. 2001, 42, 199–271.
16] F. A. Cotton, G. Wilkinson, C. A. Murillo, M. Bochmann in Advanced inor-
ganic chemistry, Wiley, New York, 1999, pp. 507–508.
17] D. J. Vaughan, A. R. Lennie, Sci. Prog. 1991, 75, 371–388.
18] H. S. Genin, J. A. Ibers in Transition Metal Sulphides, Vol. 60 (Eds.: T.
Weber, R. Prins, R. van Santen), Springer, Netherlands, 1998, pp. 1–35.
19] H. Wang, Miner. Process. Extr. Metall. Rev. 2008, 29, 1–41.
Acknowledgements
[
[
The funds for this research have been provided by the Alexander
von Humboldt Foundation in the framework of the Sofja Kova-
levskaja Award 2010 endowed by the German Federal Ministry of
Education and Research. This work was performed as part of the
Cluster of Excellence “Tailor-Made Fuels from Biomass”, which is
funded by the Excellence Initiative of the German federal and
state governments to promote science and research at German
universities. The project “Sustainable Chemical Synthesis (Sus-
ChemSys)” is co-financed by the European Regional Development
Fund (ERDF) and the state of North Rhine-Westphalia, Germany,
under the Operational Program “Regional Competitiveness and
Employment” 2007–2013.
[
[
[
[20] H. P. Wang, I. Salveson, Phase Transitions 2005, 78, 547–567.
[
21] A. I. Vogel, G. Svehla, Vogel’s qualitative inorganic analysis, Longman,
996.
1
[
[
22] D. Brion, Appl. Surf. Sci. 1980, 5, 133–152.
23] K. M. Rosso, D. J. Vaughan, Rev. Mineral. Geochem. 2006, 61, 557–607.
[24] S. V. Mikhalovsky, Y. P. Zaitsev, Carbon 1997, 35, 1367–1374.
25] W. J. Huff, J. C. Holtz, Ind. Eng. Chem. 1930, 22, 639–645.
26] V. Y. Korobkov, E. N. Grigorieva, O. V. Senko, I. V. Kalechitz, Fuel Process.
Technol. 1988, 19, 243–252.
27] X. Wang, R. Rinaldi, Energy Environ. Sci. 2012, 5, 8244–8260.
28] X. Wang, R. Rinaldi, ChemSusChem 2012, 5, 1455–1466.
29] X. Wang, R. Rinaldi, Angew. Chem. Int. Ed. 2013, 52, 11499–11503;
Angew. Chem. 2013, 125, 11713–11717.
[
[
[
[
[
Keywords: hydrodeoxygenation · iron · lignin · lignite ·
sulfides
[30] M. R. Sturgeon, S. Kim, K. Lawrence, R. S. Paton, S. C. Chmely, M. Nimlos,
T. D. Foust, G. T. Beckham, ACS Sustainable Chem. Eng. 2014, 2, 472–
4
85.
[
[
[
1] S. Vasireddy, B. Morreale, A. Cugini, C. Song, J. J. Spivey, Energy Environ.
Sci. 2011, 4, 311–345.
2] P. A. Montano, Y. C. Lee, A. Yeyeodu, C. H. Chien, ACS Symp. Ser. 1986,
[
[
[
31] B. Pawelec, J. L. Garcia Fierro in Catalytic Hydrogenation for Biomass Val-
orization (Ed.: R. Rinaldi), The Royal Society of Chemistry, Cambridge,
2
015, Chapter 8, pp. 174–203.
301, 416–429.
32] J. Roberto Gomes, S. Bezergianni, J. L. Zotin, E. F. Sousa-Aguiar in Cata-
lytic Hydrogenation for Biomass Valorization (Ed.: R. Rinaldi), The Royal
Society of Chemistry, Cambridge, 2015, Chapter 9, pp. 204–222.
3] J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius, B. M. Weckhuysen, Chem.
Rev. 2010, 110, 3552–3599.
[
[
[
4] U. Schuchardt, O. A. Marangoni Borges, Catal. Today 1989, 5, 523–531.
5] D. W. Goheen, Adv. Chem. Ser. 1966, 59, 205–225.
6] P. Fei-Ling, C. Chin-Hua, S. Zakaria, N. Soon-Keong, L. Tze-Khong, Sains
Malays. 2011, 40, 221–226.
33] Y. Cai, Y. Pan, J. Xue, Q. Sun, G. Su, X. Li, Appl. Surf. Sci. 2009, 255, 8750–
8
760.
[
[
34] K. Laajalehto, I. Kartio, P. Nowak, Appl. Surf. Sci. 1994, 81, 11–15.
35] P. Ferrini, R. Rinaldi, Angew. Chem. Int. Ed. 2014, 53, 8634–8639; Angew.
Chem. 2014, 126, 8778–8783.
36] D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka,
G. D. Stucky, Science 1998, 279, 548–552.
[
[
7] R. M. Baldwin, S. Vinciguerra, Fuel 1983, 62, 498–501.
8] S. Yokoyama, R. Yoshida, H. Narita, K. Kodaira, Y. Maekawa, Fuel 1986,
[
65, 164–170.
[9] D. R. Alfonso, J. Phys. Chem. C 2010, 114, 8971–8980.
[
[
10] J. H. Chen, L. H. Lan, Y. Chen, Miner. Eng. 2013, 46–47, 136–143.
11] G. V. Gibbs, D. F. Cox, K. M. Rosso, N. L. Ross, R. T. Downs, M. A. Spack-
man, J. Phys. Chem. B 2007, 111, 1923–1931.
Received: January 16, 2015
Published online on February 17, 2015
ChemCatChem 2015, 7, 960 – 966
www.chemcatchem.org
966
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim