The Low-Temperature Oxidation of Propane by using H2O2 and Fe/ZSM-5 Catalysts: Insights into the Active Site and Enhancement of Catalytic Turnover Frequencies

Article abstract of DOI:10.1002/cctc.201601241

Fe-containing ZSM-5 catalysts are reported to be efficient catalysts for the partial oxidation of propane to oxygenated products at reaction temperatures as low as 50 °C in an aqueous phase reaction when using the green oxidant H₂O₂. It was previously proposed that extra framework Fe species at the exchange sites of the zeolite are responsible for activation of both the alkane and hydrogen peroxide. Through a systematic study of the influence of framework topology and exchange properties, it is now shown that this high catalytic activity is specific to the MFI-type Br?nsted acidic zeolite ZSM-5. Furthermore, through a simple aqueous acid washing treatment, leaching of approximately 77 % of iron present within a Fe/ZSM-5 catalyst only caused the relative propane conversion to decrease by 17 %; implying that most of the initially loaded Fe does not actually contribute to the catalytic activity. This small change in conversion after ‘excess’ Fe removal, amounts to a three-fold increase in turnover frequency (TOF) (Fe) from 66 h^?1 to 232 h^?1 compared with the parent Fe/ZSM-5 catalyst. By comparing these samples, it is shown by NH₃ temperature-programmed desorption, ²⁷Al magic angle spinning NMR spectroscopy, X-ray photoelectron spectroscopy and TEM analysis that surface iron oxide species are effectively spectators in the oxidation of propane with H₂O₂. This provides further insight as to the location and true nature of the catalytically active Fe species.

Full text of DOI:10.1002/cctc.201601241

Accepted Article
Title: The low temperature oxidation of propane using H2O2 and Fe/
ZSM-5 catalysts; insights into the active site and enhancement of
catalytic turnover frequencies
Authors: Virginie Peneau, Robert Armstrong, Greg Shaw, Jun Xu,
Rpbert Jenkins, Nikolaos Dimitratos, Stuart Taylor, Horst
Zanthoff, Stefan Pietz, Guido Stochniol, Qian He, Christopher
Kiely, and Graham John Hutchings
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201601241
Link to VoR: http://dx.doi.org/10.1002/cctc.201601241
A Journal of
www.chemcatchem.org

10.1002/cctc.201601241
ChemCatChem
FULL
PAPER
The low temperature oxidation of propane using H2O2 and
Fe/ZSM-5 catalysts; insights into the active site and
enhancement of catalytic turnover frequencies
Virginie Peneaua, Robert D. Armstronga, Greg Shawa, Jun Xua, Robert L. Jenkinsa, David. J.
Morgana, Nikolaos Dimitratosa, Stuart H. Taylora, Horst W. Zanthoffb, Stefan Peitzc, Guido
Stochniolc, Qian Hea, Christopher J. Kielyd and Graham J. Hutchingsa*
Abstract: Fe- containing ZSM-5 catalysts are reported to be efficient
catalysts for the partial oxidation of propane to oxygenated products
at reaction temperatures as low as 50 oC in an aqueous phase
reaction using the green oxidant H2O2. It was previously proposed that
extra framework Fe species at the zeolite’s exchange sites are
responsible for activation of both the alkane and hydrogen peroxide.
Through a systematic study of the influence of framework topology
and exchange properties it is now shown that this high catalytic
activity is specific to the MFI-type Brønsted acidic zeolite ZSM-5.
Furthermore, through a simple aqueous acid washing treatment,
leaching ca. 77 % of iron present within a Fe/ZSM-5 catalyst only
caused the relative propane conversion to decrease by 17 %; implying
that most of the initially loaded Fe does not actually contribute to the
catalytic activity. This small change in conversion after ‘excess’ Fe
removal, amounts to a three-fold increase in TOF (Fe) from 66 h-1to
232 h-1 compared with the parent Fe/ZSM-5 catalyst. By comparing
these samples, it is shown by NH3-TPD, 27Al MAS NMR, XPS and
TEM analysis, that surface iron oxide species are effectively
spectators in the oxidation of propane with H2O2. This provides further
insight as to the location and true nature of the catalytically active Fe
species.
push to find alternative, non-crude oil dependent routes to bulk
chemical synthesis. Whilst not the most abundant of these
resources, the direct oxidation of propane is still of great interest,
owing to it possessing both primary and secondary carbon atoms.
Unfortunately, the realisation of propane as a feedstock for direct
oxidation processes is stymied by relatively high C-H bond
enthalpies of 422.2 and 409.2 kJ mol-1 at the primary and
secondary positions respectively [1, 2]. To overcome this kinetic
inertness and low reactivity, current industrial practices are often
energy intensive. Indeed, aside from its primary use as a
combustible fuel, propane is currently a feedstock for the
production of acrolein, acrylic acid, isopropanol and acetone; the
first step in all these synthesis routes being steam cracking or
dehydrogenation to propene followed by functionalisation to
oxygenated products [3-6]. Such indirect processes have high
energy / economic demands and there is considerable interest in
developing direct routes to convert propane, which is available in
large quantities, to value-added oxygenated products.
Unfortunately, direct routes to C3 oxygenated products suffer
competing, undesirable deep oxidation and scission pathways
due to the higher reactivity of primary products relative to
propane. Therefore, high conversion and reaction selectivity
towards C3 oxygenated products must be realised if direct routes
are to compete with current processes which, though energy
intensive, afford high C3 product yields.
Introduction
Direct catalytic routes for the synthesis of isopropanol, acrylic
acid, acrolein and acetone have been reported [5, 7-15], with gas
phase operation favoured due to the physical properties of
propane. The low temperature oxidation of n-alkanes using ZSM-
5 catalysts and H2O2 has been studied previously [16-22], with
activation of the oxidant and alkane substrate being attributed to
extra framework dimeric µ-oxo- hydroxo iron sites [19, 20]. These
were shown to form upon high temperature activation of ZSM-5,
which promotes migration of residual iron, endemic in commercial
zeolites, from tetrahedral framework sites to extra framework
A key challenge within the field of catalysis is the selective partial
oxidation of aliphatic hydrocarbons such as methane, ethane and
propane. Valorisation of these highly abundant, inexpensive
constituents of natural gas, is an important element in the global
[a]
Cardiff Catalysis Institute
School of Chemistry, Cardiff University
Park Place, Cardiff, UK, CF10 1AQ
Corresponding E-mail: Hutch@Cardiff.ac.uk
Evonik Technology and Infrastructure GmbH
Paul-Baumann Str. 1, 45764 Marl, Germany
Evonik Performance Materials GmbH
Paul-Baumann Str. 1, 45764 Marl, Germany
Department of Materials Science and Engineering,
Lehigh University, 5 East Packer Avenue, 18015-3195, Bethlehem,
Pennsylvania, USA
-
[b]
[c]
[d]
cation exchange sites (AlO4 ) [19]. Furthermore, it has recently
been reported that the MFI zeolite H-ZSM-5 is also intrinsically
active for the partial oxidation of propane using the green oxidant
H2O2 in the aqueous phase [23]. Commercial H-ZSM-5 showed
rates of 2.7 mol propane converted kgcat-1h-1. Concordant with
previous studies into the oxidation of methane and ethane under
equally mild reaction conditions [16-20], this activity was ascribed
to Fe residues (0.014 wt %). This afforded unprecedented
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
catalytic turnover frequencies of 1064 mol molFe-1h-1 at 50 oC [23].
Additionally, through post synthesis deposition of iron (2.5 wt %)
onto H-ZSM-5 using a novel chemical vapour impregnation (CVI)
technique, catalyst productivity was shown to be further enhanced
to 23.5 mol propane convertedkgcat-1h-1, which correspondedto a
decrease in TOF to 66 h-1.
supported catalyst showed a far higher activity level (23.5 mol
-1
propane converted kgcat h-1), followed by 2.5 wt% Fe/SiO2 (1.1
mol propane converted kgcat-1 h-1). It is clear from these data that
the high catalytic activity shown by ZSM-5 (30) catalysts cannot
be solely attributed to Brønsted acidic AlO4- exchange sites as
Zeolites Y (SiO2/Al2O3 = 30) and Beta (SiO2/Al2O3 = 25), possess
comparable acid site densities, yet showed relatively low rates of
propane conversion. Rather, the data in Figure 1 suggests that
catalytic activity might be derived from the MFI framework
topology of ZSM-5 catalysts combined with the presence of
additional Fe species. The TS-1 and Silicalite-1 are MFI zeolites
which are isomorphous to ZSM-5 but lack cation exchange sites
Herein we report a systematic study of this catalyst system with
an aim of determining the role which the zeolite framework
topology and exchange properties play in catalyst performance.
Through better understanding the active species, more active
catalysts for the direct oxidation of propane may be designed.
-
(AlO4 ). These variants showed relatively low rates of propane and
H2O2 conversion when compared with ZSM-5 (30) (Figure 2),
particularly following Fe impregnation, which had no discernible
benefit upon the activity of the alumina-free zeolites.
Results and Discussion
To determine the contribution of physical / chemical properties,
(i.e., microporosity, Brønsted acidity and framework topology)
towards the high intrinsic activity of H-ZSM-5 (30) for propane
oxidation, a systematic series of amorphous and crystalline
alumina/-silicate samples were prepared and assessed. Given
that our previous studies of the same catalyst system showed the
catalytic productivity of H-ZSM-5 (30) to increase by a factor of 10
following impregnation with 2.5 wt% Fe, solid catalysts were also
modified via the CVI method such that the nominal deposited Fe
content was 2.5 wt% [23]. The catalyst productivities for
unmodified and Fe-impregnated silica/alumina catalysts are
shown in Figure 1.
25
20
15
10
5
25
20
15
10
5
0
TS-1
Silicalite-1
ZSM-5
Figure 2 – Comparison of the total productivity of MFI-framework zeolite
catalysts for propane oxidation with H2O2. Reaction conditions: P (C3H8) = 4 bar
(4000 μmol), Ptotal(C3H8 / He) = 20 bar, [H2O2] = 0.5 M (5000 μmol), 27 mg
catalyst, 50 ºC, 0.5 h, 1500 rpm.
 Bare support (where applicable, zeolites in H-form).  2.5 wt. % Fe
impregnated by CVI
Owing to the fact that our previous studies into low temperature
oxidation of short chain alkanes over ZSM-5 catalysts attributed
the intrinsic activity of unmodified ZSM-5 to contaminant Fe
species, the unmodified supports from Figures 1 and 2 were
analysed for Fe content. Rates of H2O2 and propane conversion
(as a function of both Fe content and mass productivity) are
shown in Table 1.
0
SiO2
Al2O3
SiO2 / Al2O3 Zeolite Y Zeolite 
ZSM-5
Figure 1 – Comparison of the total productivity of aluminosilicate catalysts for
propane oxidation with H2O2. Reaction conditions: P (C3H8) = 4 bar (4000 μmol),
Ptotal(C3H8/ He) = 20 bar, [H2O2] = 0.5 M (5000 μmol), 27 mg catalyst, 50 ºC, 0.5
h, 1500 rpm. Catalysts calcined 3 h at 550 oC (20 oC min-1, static air).
 Bare support (where applicable, zeolites in H-form).  2.5 wt. % Fe
impregnated by CVI
When rates were normalised to Fe content, the highest propane
turnover frequency was shown for H-ZSM-5 (30) (1064 mol
-1
propane converted molFe h-1). This is compared with a rate of
7588 mol H2O2 converted molFe-1 h-1 over the same catalyst. The
disparity between observed rates of propane and H2O2
conversion is an important consideration when comparing
catalyst efficiency and the atom efficiency of the process.
Appreciable rates of propane conversion were also noted for the
SiO2 and SiO2/Al2O3 materials (198 and 312 mol propane
Of the unmodified supports tested, H-ZSM-5 (SiO2/Al2O3 = 30)
showed the highest intrinsic activity (2.7 mol propane converted
-1
kgcat h-1). Zeolite-Y (SiO2/Al2O3 = 30), the second most active,
showed relatively low activity (0.5 mol propane converted kgcat-1 h-
1). Following impregnation with Fe (2.5 wt %), the ZSM-5 (30) -
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
Table 1 – Physical properties and catalytic performance of unmodified solid catalysts.
SiO2/ Al2O3
BET surface
area (m2/ g)[b]
430.7
TOFH2O2/
Catalyst
wt % Fe[a]
Rate (H2O2)[c]
TOF (H2O2)[d]
Rate (C3H8)[c]
TOF (C3H8)[d]
(molar)
TOFC3H8
[bdl]
[bdl]
[bdl]
[bdl]
SiO2
Al2O3
-
-
0.007
0.029
0.008
0.026
0.038
0.014
0.031
0.009
0.26
0.25
0.45
0.53
0.26
2.7
198.4
49.0
-
-
6.6
SiO2/ Al2O3
H+ Zeolite Y
H+ Zeolite β
H-ZSM-5
TS-1
10
30
25
30
-
527.3
26.5
16.2
52.0
18272.8
3475.8
312.1
113.2
37.8
58.5
30.7
202.1
7.1
-
716.1
706.7
7639.77
413
19.0
[bdl]
7587.8
[bdl]
1063.7
36.7
361.1
0.21
0.14
[bdl]
[bdl]
Silicalite-1
-
474.7
85.8
-
[a] as determined by ICP, [b] Surface area determined from nitrogen adsorption using the BET equation, [c] mol converted kgcat-1h-1, d] mol converted molFe-1h-1,
Catalysts calcined prior to testing (3 h, 550 oC, 20 oC min-1 in static air). [bdl] H2O2 conversion below detection limit.
-1
converted molFe h-1 respectively); however, when normalised to
mass, productivities were still low relative to H-ZSM-5 (30). A
comparison of the H2O2 and propane conversion rates observed
over aluminosilicate catalysts is given in Table 1. Stoichiometric
respectively). The homogeneous catalyst also afforded increased
CO2 selectivity and a higher rate of H2O2 decomposition, but no
overall benefit to the rate of alkane oxidation. Remarkably, the
Fenton’s system afforded high acetone selectivity (53%) with
higher overall selectivity towards C3 products (64.0 %) than the
supported Fe/ZSM-5 catalyst (34.2 %). High primary product
selectivity when utilising homogeneous FeIII / H2O2 is consistent
with the partial ethane and methane oxidation systems reported
by Shul Pin et al [23, 25]. Catalytic activity was attributed to
generation of a ferryl (FeIV=O)2+ ion, with appreciable TOFs (2.33
H2O2 utilisation would afford a (TOFH2O2 molFe h-1)/ (TOFC3H8
-1
molFe-1 h-1) ratio equal to 2. Of the aluminosilicates tested, H-ZSM-
5 (30) shows the greatest atom efficiency, with a ratio of 7.1.
It is clear that the high productivity and TOFs shown by ZSM-5
catalysts are due to a combination of (i) MFI framework topology,
-
(ii) Fe content and (iii) the availability of AlO4 cation exchange
-1
sites.
mol methane converted molFe h-1 and 22.67 mol ethane
converted molFe-1 h-1) in the absence of promoters [25]. As shown
in Table 2, addition of a hydroxyl radical scavenger (NaSO3) to
Fe/ZSM-5 (30) catalysed propane oxidation resulted in a 37 %
decrease in conversion, which fell from 7.9 % to 5.0 %. This is
consistent with the initial activation of propane proceeding at least
in part through a free radical-process. As such processes have
been previously reported for homogeneous Fe systems [25], the
heterogeneity of the Fe/ZSM-5 catalyst system was assessed
through hot filtration studies (Table 3). In the absence of Fe/ZSM-
5, no further propane conversion occurred. However, undesirable
C-C scission reactions continue, with increased yields of acetic
and formic acids, suggesting potential OH radical-driven side
reactions (further details are shown in ESI, Table 1). Results
Extra-framework dimeric µ- oxo-hydroxo iron sites, situated at the
AlO4 exchange sites in ZSM-5 have been proposed as
-
catalytically active for methane and ethane oxidation reactions
[16, 24]. Indeed, there is precedent within the literature for
Fenton’s- type decomposition of H2O2 to form hydroxyl radicals,
which can propagate radical-based mechanisms through
abstraction of R-CH2-H. Reactions using Fenton’s type reagents
were therefore carried out, and the activities compared with those
of our Fe supported catalysts in Table 2. The homogeneous FeIII
Fenton’s- type reagent, Fe(NO3)39H2O, showed comparable
propane turnover rates to Fe/ZSM-5 (30) prepared by the CVI
-1
method (47.7 and 63.5 mol propane converted molFe h-1
Table 2 – The activity of Fenton’s-type reagents and effect of adding a ˙OH scavenger (NaSO3) upon propane oxidation reactions.
Product Selectivities / %
C2 Products
χ
χ
H2O2
[b]TOF
(h-1)
C3 Products
C1 Products
MeO Formic
Catalyst
Propane H2O2 Used
/ %
/ % / %
i-
PrO
H
n-
PrOH
Propanoi
c Acid
Acetic
Acid
Acetone
C3H6 EtOH
C2H4 C2H6
CH4 CO2
H
Acid
H-ZSM-5
0.9
7.9
5.2
29
18
43
9.9
6.4
5.8
5.7
1.1
23.3
8.8
12.9
11.4
9.5
2.8
2
6.6 1.5 0.7
14.1 11 0.6
5.8
1.5
0.5
0.6 27.9 1.26 0.9 1063.7
1.3 34.6 0.9 1.4 63.5
0.2 12.2 0.1 4.3 47.7
2.1% Fe/ZSM-5
Fe(NO3)39H2O
7.18 52.5 12.0
53.0
0.4
0.0
0.2 18.2 0.1
8.4 15.6 0.2
[a]
2.1% Fe/ZSM-5
(30) + NaSO3
5.0
32.5 19.2
12.9
3.9
7.6
12.3
1.1
2.7
1.2 29.9 1.2 3.1 39.4
[b]
Reaction conditions: P (C3H8) = 4 bar (4000 μmol), Ptotal(C3H8/ He) = 20 bar, [H2O2] = 0.5 M (5000 μmol), 27 mg supported catalyst where applicable,
50 ºC, 0.5 h, 1500 rpm. [a] 1.2 x 10-5 mol Fe – based upon ICP analysis, [b] [NaSO3] = 0.053 M (530 μmol)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
Table 3 – Assessment of reaction heterogeneity through hot filtration
Product Selectivities / %
Total
carbon / H2O2 Used
µmol / % / %
χ
H2O2
C3 Products
C2 Products
Acetic
C1 Products
Reaction
i-
PrO
H
n-
PrOH
Propanoi
c Acid
MeO Formic
C2H4 C2H6 CH4 CO2
Acetone
C3H6 EtOH
Acid
39.2 127.2
37.5 141.2 0.1
H
Acid
Catalysed
1240.9 28.4 29.0
1296.7 0.64 30.0 [b]
50.8
50.0
10.6
7.1
13.8
10.1
42.0
42.3
-
-
-
14.3 541.8
-
-
Hot Filtration
0.0
1.8 14.0 581.1 0.5 11.5
Reaction conditions: P (C3H8) = 4 bar (4000 μmol), Ptotal(C3H8/ He) = 20 bar, [H2O2] = 0.5 M (5000 μmol), 27 mg supported catalyst where
applicable, 50 ºC, 0.5 h, 1500 rpm. [a] where 3 μmolcarbon corresponds to 1 μmolpropane [b] % H2O2 utilisation over reactions 1 and 2.
.
therefore indicate that the supported Fe/ZSM-5 catalyst is
behaving as a heterogeneous Fenton’s – type catalyst for the
propane oxidation reaction. This is in agreement with the data
from Figures 1 and 2, which suggest a cationic Fe active site
exists within the ZSM-5 catalysts. This is also consistent with our
previous studies into ethane oxidation over the same catalyst
system, in which it was observed that near-complete removal of
surface iron oxides caused only a limited decrease in reaction rate
and consequently led to significant increases in TOFs [26].
5) effected further Fe- removal, leading to increased TOFs, with
reaction selectivities comparable to the non-acid washed parent
2.06 % Fe/ZSM-5 (30) catalyst. Indeed, all Fe- modified ZSM-5
catalysts represented in Table 4 showed ca. 34 % selectivity
towards C3 reaction products. This is a key requirement for direct
propane oxidation, however we previously showed that primary
products undergo cracking/ further oxidation over the same
catalysts to yield the C2 and C1 products shown in Table 4 [23].
Another key consideration in this system is the selectivity with
which H2O2 is converted, as such synthetic oxidants incur a high
cost relative to molecular oxygen. Acid treatment effected a
decrease in the % H2O2 used in products, from 43 % for the parent
Fe/ZSM-5 catalyst to 31 % following a 2 h treatment at
isoconversion of H2O2. Increases in either the temperature or
concentration of the HNO3 (aq) solution used, showed no further
beneficial effect on the catalyst activity (see ESI, Table 2). In
order to determine the effect of acid washing upon the zeolite
support, the location/nature of removed Fe- species and thereby
hopefully elucidate details of the catalytically active site in
Fe/ZSM-5, this series of catalysts were characterised using a
range of techniques.
To further study the nature and location of the active site for
propane oxidation, we adopted the following approach; a catalyst
containing 2.5 wt% Fe (nominal)/ ZSM-5 (30), (2.06 wt %
determined by ICP), was treated in HNO3 (aq)(2.4 M) at 50 oC for
varying time periods. The resulting catalysts were analysed for
iron content, and their activity for propane oxidation was re-
o
assessed. As shown in Table 4, a 15 min. treatment at 50 C
removed ca. 73 % of total Fe with only a slight decrease in
propane conversion being observed (down from 7.9 % to
6.8 %)(Table 4, Entry 3). This equated to a 210.5 % increase in
TOF, from 63.5 to 204.6 mol propane converted molFe-1 h-1. It was
also found that longer acid treatment times (Table 4, Entries 4 &
Table 4 – Propane oxidation using a series of untreated and acid-treated ZSM-5 catalysts.
Product Selectivities / %
C2 Products
[a]
χ
χ
H2O2
[b]TOF
(h-1)
C3 Products
C1 Products
Formic
Entry Propane H2O2 Used
wt.%
Fe
/ %
/ %
/ %
18
i-
PrO
H
n-
PrOH
Propanoi
c Acid
Acetic
Acid
Acetone
C3H6 EtOH
C2H4 C2H6 MeOH
CH4 CO2
Acid
1
2
3
4
5
0.9
7.9
6.8
6.8
6.6
5.2
9.9
6.4
4.6
4.6
4.4
5.8
5.7
5.8
6.2
5.7
23.3
8.8
12.9
11.4
8.8
2.8
2
6.6
1.5
0.7
0.6
5.8
1.5
1.7
1.8
2.8
0.6
1.3
1.3
1.3
1.2
27.9 1.26 0.9 0.014 1063.7
34.6 0.9 1.4 2.06 63.5
33.2 1.0 1.1 0.55 204.6
30.5 1.1 1.1 0.5 225.0
35.5 1.0 1.4 0.47 232.4
29.0 43
27.3 34
23.4 37
28.3 31
14.1 11
10.6
12.1
10.1
2.7
2.9
2.8
17.2 11.4 0.4
18 11.1 0.5
15.7 12.7 1.7
8.8
7.2
Reaction conditions: P (C3H8) = 4 bar (4000 μmol), Ptotal(C3H8/ He) = 20 bar, [H2O2] = 0.5 M (5000 μmol), 27 mg catalyst, 50 ºC, 0.5 h, 1500 rpm. [a]
Determined by ICP, [b] mol propane converted molFe-1 h-1.
Entry 1: H-ZSM-5 (30), Entry 2: 2.5%Fe/ZSM-5 (30), Entry 3: 2.5 wt. % Fe/ ZSM-5 (30)AT 0.25, Entry 4: 2.5 wt. % Fe/ ZSM-5 (30)AT 1.0, Entry 5: 2.5 wt. %
Fe/ ZSM-5 (30)AT 2.0
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
Table 5 - XPS analysis of acid treated 2.5 wt. % Fe/ ZSM-5 (30) catalysts.
Surface Content (atomic %)
Binding Energy (eV)[a]
Catalyst
wt% Fe[b]
Al
Fe 2p
709.9
O 1s
Si 2p
101.9
Al 2p
73.9
Fe
O
Si
2.5 wt. % Fe/ ZSM-5 (30)
2.5 wt. % Fe/ ZSM-5 (30)AT 0.25
2.5 wt. % Fe/ ZSM-5 (30)AT 1.0
2.5 wt. % Fe/ ZSM-5 (30)AT 2.0
530.9
5.46
68.72
25.49
0.33
0.26
0.36
0.28
2.06
0.55
0.5
711.6
711.6
711.6
531.8
531.6
531.6
102.8
102.6
102.6
73.8
74.6
73.6
0.33
0.30
0.18
75.15
73.37
71.27
24.25
25.96
28.28
0.47
[a] All binding energies referenced to C 1s = 284.7 eV, [b] Determined by ICP with HF digestion.
The 2.5% Fe/ZSM-5 (30) material prepared by the CVI method
has been fully characterised previously [16, 17]. HR-TEM showed
that Fe is deposited as a patchy iron oxide film on the zeolite
surface and also within its pores [16, 17]. Meanwhile UV-vis
spectroscopy indicated broad speciation of iron as; isolated iron
clusters, oligomeric iron species, and cationic species at
exchange sites within the zeolite pores [16, 17].
The spectra in Figure 3 show impregnation of ZSM-5 with Fe to
be unselective, giving rise to broad absorbance across the whole
bandwidth in Figure 3(a). Following acid treatment a significant
decrease in the absorbance feature at λ > 350 nm suggests
removal of surface oxide species. The retention of low wavelength
absorbance bands (200 – 350 nm) implies that framework iron
sites and those species sited within the zeolite channels are at
least partially retained following acid treatment, independent of
the treatment duration.
(a)
(b)
(c)
(d)
(e)
(a)
(b)
(c)
(d)
200
300
400
500
600
700
800
Wavelength / nm
750
740
730
720
710
700
690
B.E / eV
Figure 3 – Diffuse reflectance UV-vis spectra of (a) untreated 2.5 wt. % Fe/
ZSM-5 (30), (b) 2.5 wt. % Fe/ ZSM-5 (30)AT 0.25, (c) 2.5 wt. % Fe/ ZSM-5 (30)AT
1.0, (d) 2.5 wt. % Fe/ ZSM-5 (30)AT 2.0 and (e) H-ZSM-5 (30).
Figure 4 – XPS spectra showing the Fe 2p3/2 binding energy region for (a)
untreated 2.5 wt. % Fe/ ZSM-5 (30), (b) 2.5 wt. % Fe/ ZSM-5 (30)AT 0.25, (c) 2.5
wt. % Fe/ ZSM-5 (30)AT 1.0 and (d) 2.5 wt. % Fe/ ZSM-5 (30)AT 2.0
Our characterisation studies showed that the concentration of all
these Fe species increased with Fe loading, suggesting that Fe-
deposition via CVI was relatively non-selective. Additionally, XPS
surface analysis showed the existence of Fe3+, which was due to
the presence of Fe2O3 particles [17].
Quantification of surface Fe from XPS spectra presented in Figure
4 is in agreement with DR UV-vis spectroscopic data. Following
acid treatment, a significant decrease in Fe enrichment of the
zeolite surface is observed, falling from 5.46 to 0.33 mol % after
15 min. acid treatment. ICP analysis of the same materials
showed Fe contents of 2.06 and 0.55 wt % Fe respectively. Based
on XPS and DR UV-vis analysis it can be concluded that CVI
impregnation affords significant Fe speciation, with surface
enrichment exceeding the theoretical Fe loading. Meanwhile the
disparity between XPS and ICP measurements on acid treated
Diffuse reflectance UV-vis spectra of the acid-washed series of
2.5 % Fe/ZSM-5 catalysts are shown in Figure 3. Speciation of Fe
in ZSM-5 catalysts is expected to give rise to four UV-active
species; isolated Fe3+ in framework sites (200-250 nm), isolated
or oligomeric extra framework Fe species in zeolite channels
(250-350 nm), iron oxide clusters (350-450 nm) and large surface
oxide species (> 450 nm)[27, 28].
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
catalysts implies that the acid treatment is selective, favouring
leaching of surface iron oxides.
NH3-TPD plots for the iron free ZSM-5 catalysts show two key
desorption features occurring at 266 oC and 446 oC. The low
temperature desorption is assigned to NH3 adsorbed at weak acid
sites, either Lewis or Brønsted acidic in nature, whilst the higher
temperature desorption is accepted to represent NH3
chemisorbed at strongly acidic Brønsted acid sites [29-35].
Following impregnation of Fe by CVI, a significant decrease in the
high temperature desorption feature is observed (Figure 6d),
which is consistent with occupation or blocking of Brønsted acid
sites by Fe cations and oxides respectively. Following acid
washing, the amount of NH3 desorbed at this high temperature is
seen to increase again. This suggests either (a) removal of
exchanged Fe cations and/or (b) the removal of site-blocking Fe
oxides.
Indeed, following acid treatment, the total Fe content exceeds the
surface enrichment for all catalysts, suggesting that a greater
proportion of Fe sites are located within the pores of the zeolite
material. This supposition is also evidenced by electron
microscopy studies. Representative TEM bright field (BF) images
(a)
(b)
of 2.5% Fe/ZSM-5 (30) catalysts prepared by CVI before and after
a 2 h acid treatment are shown in Figures 5 (a) and (b),
respectively. Particles around 2 nm in size (highlighted by white
arrows) were found on ZSM-5 particles in the un-treated catalyst.
Those particles are also visible in the Z-contrast STEM high angle
annular dark field (HAADF) images (ESI, Figure S4), displaying a
significantly higher contrast than the surrounding zeolite support,
confirming that the particles contain iron.
Table 6 - N2 BET surface area and porosimetry analysis of ZSM-5 catalysts.
(b)
(a)
BET surface area VMicropores
/
VMesopores
(cm3 g-1)
/
Catalyst
(m2/ g)[a]
(cm3 g-1)
H-ZSM-5 (30)
413
363
0.147
0.123
0.148
0.098
2.5 wt. % Fe/ ZSM-5 (30)
2.5 wt. % Fe/ ZSM-5 (30)AT 0.25
2.5 wt. % Fe/ ZSM-5 (30)AT 1.0
2.5 wt. % Fe/ ZSM-5 (30)AT 2.0
353
348
257
0.122
0.116
0.086
0.093
0.094
0.094
Figure 5 – Representative TEM bright field (BF) images of the (a) the untreated
catalyst - 2.5 wt. % Fe/ ZSM-5 (30) and (b) the catalyst after 2 h of acid wash -
2.5 wt. % Fe/ ZSM-5 (30)AT 2.0. Iron species of about 2 nm in size are clearly
visible (e.g., highlighted by white arrow) in the un-treated catalyst. Those are no
longer visible after acid treatment. The scale bars represent 20 nm.
[a] Surface area determined from nitrogen adsorption measurement using the
BET equation.
After the acid treatment, these iron containing particles can no
longer be found (Figure 5(b)), confirming that they are being
removed effectively by the acid wash treatment.
The effect of acid treatment upon the zeolite support’s
physicochemical properties was also studied through NH3-TPD,
N2 physisorption, 27Al MAS NMR, DRIFTS and XRD studies.
N2 adsorption studies show a decrease in surface area following
Fe impregnation (from 413 m2g-1 to 363 m2g-1) and
a
corresponding decrease in Vmicropore, from 0.147 to 0.123 cm3g-1
as shown in Table 6. Both surface area and Vmicropore are shown
to decrease, reaching 257 m2g-1and 0.086 cm3g-1 respectively
following a 2 h acid treatment, whilst no significant change in V
mesopore was observed following acid treatment. This suggests that
the acid washing is also effecting a structural change within the
zeolite pores, which is consistent with our previous studies [26].
In the 2 h acid washed 2.5% Fe/ZSM-5 (30)AT2.0 catalyst, some
agglomerates of 10-20 nm nanocrystals can be occasionally
found (ESI, Figure S5(a)). These are much smaller than the
typical parent ZSM-5 particles and X-ray energy dispersive
spectroscopy (XEDS) analysis (ESI, Figure S5(b)) suggests that
they are deficient in Al compared to ZSM-5. It is likely that these
are silicate material, resulting from some minor ZSM-5
dealumination during the acid wash, which might be expected
under the conditions used [36, 37]. Stabilisation of H2O2 in the
presence of a Brønsted acid is well known. The decreased
selectivity in H2O2 conversion observed in Table 4 following acid
treatment of Fe/ZSM-5 might therefore be attributed to a loss of
Brønsted acidic aluminum sites. However, the majority of the
zeolite structure remains intact as evidenced by XRD (ESI, Figure
S3 and ESI, Table S3). There is no apparent change in the zeolite
unit cell following either impregnation with Fe or acid treatments,
(e)
(d)
(c)
(b)
(a)
200
400
600
800
Temperature / oC
Figure 6 - NH3-TPD analyses of (a) H-ZSM-5 (30), (b) untreated 2.5 wt. % Fe/
ZSM-5 (30), (c) 2.5 wt. % Fe/ ZSM-5 (30)AT 0.25, (d) 2.5 wt. % Fe/ ZSM-5 (30)AT
1.0 and (e) 2.5 wt. % Fe/ ZSM-5 (30)AT 2.0
.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
with dimensions remaining consistent with those of the parent H-
ZSM-5. Therefore, in order to determine whether the observed
changes in surface area and pore distribution were due to
significant dealumination of the zeolite framework, the catalysts
were further analysed using 27Al MAS NMR and DRIFT
spectroscopies.
observance of DRUV-Vis absorbance bands at ca. 200-350 nm
following acid treatment indicated that some iron oxide species,
are retained within the zeolite’s pores. Acid washing for extended
periods of time (> 0.5 h) was found to be unnecessary, indeed this
is found to induce a structural change in the zeolite support
through dealumination. The Fe/ ZSM-5 catalysts were shown to
be strictly heterogeneous and addition of a hydroxyl radical
scavenger led to decreased propane conversion at isoconversion
of H2O2. It is clear therefore that this propane oxidation process
proceeds at least in part via a free- radical mechanism.
Interestingly, Fe (III) nitrate is found to be an efficient catalyst for
propane oxidation with H2O2 at 50 oC, yielding acetone as a major
reaction product. These studies suggest that Fe sites exchanged
within ZSM-5 act as a heterogeneous Fenton’s – type catalyst and
it is clear that acid sites (Lewis and/ or Brønsted) in the
hydrophobic pores of ZSM-5 play a key role in the observed
cracking of C3 products. Potential routes towards promoting the
mass transport of primary products away from the catalytically
active sites in the pores of ZSM-5 are currently under
investigation, with an aim to increase C3 selectivities.
Integrated areas corresponding to 27Al resonances in 27Al MAS
NMR spectra of these catalysts are shown in the ESI, Table S4.
Impregnation of ZSM-5 with Fe (2.06 wt%) caused a decrease in
the Td/Oh ratio which is consistent with either (i) exchange of
cationic Fe species at AlO4- sites or (ii) migration of Td Al to extra
framework Oh sites. Indeed, a decrease in the intensity of the Td
27Al resonance at 55 ppm is consistent with the presence of
paramagnetic Fe species at ion exchange positions [38-40],
which is consistent with the proposed dimeric µ- oxo-hydroxo iron
active site. An increase in the Td/Oh Al ratio following acid
treatment (0.25 – 1 h) is consistent with a portion of the removed
Fe species having occupied cation exchange sites. This is in
agreement with the decrease in catalyst productivity shown in
Table 4 following acid treatment. DRIFTS spectra (ESI, Figure S2)
show a clear decrease in the peak corresponding to O-H groups
coordinated to tetrahedral framework Al3+ (3605 cm-1) following
Fe impregnation [38]. After acid treatment however, no significant
change is observed for the peaks corresponding to either O-H
groups coordinated to tetrahedral framework Al3+ (3605 cm-1)[41]
or O-H coordinated to extra framework tetrahedrally co-ordinated
Al atoms (3658 cm-1) [42]. However, a clear decrease in intensity
of the peak assigned to terminal Si-O-H entity (3737 cm-1) is
observed following Fe impregnation. This is observed to increase
in intensity following acid treatment, which suggests that
supported iron oxides/ clusters interact with terminal silanol
groups on the zeolite surface.
Acknowledgements
The authors would like to thank Evonik Industries for financial
support.
Experimental Section
Catalyst Preparation
Metal impregnation using chemical vapour impregnation
(CVI)
Conclusions
NH4-ZSM-5 (SiO2/Al2O3 molar ratio = 30) was obtained from
ZSM-5 catalysts are active for the partial oxidation of propane
under mild reaction conditions using the environmentally benign
oxidant H2O2. These catalysts have been shown to derive their
intrinsic activity from a combination of synergistic physical
properties unique to ZSM-5, specifically the MFI framework
o
Zeolyst and activated in flowing air prior to use (550 C, 3 h, 20
oC min-1). Fe was deposited at the desired loading by the CVI
technique using Fe (acac)3. This preparation technique has been
described in detail previously, and was shown to afford strict
control of actual metal loadings[17].
-
topology, Brønsted acidic AlO4 exchange sites and trace iron
residues. The productivity afforded by H-ZSM-5 is significantly
enhanced through deposition of Fe through CVI. Much of this
added iron has been shown to be inactive for the catalytic
reactions involved; H2O2 conversion and propane oxidation.
Indeed, when 77 % of supported Fe was leached using an
aqueous acidic solution, relative conversion only decreased by
17 % meaning that the TOF (Fe) increased by over 200 %. This
strongly indicates that it is exchanged cationic Fe sites within the
zeolite pores that are active for the studied reaction. This was
further supported through characterisation of the synthesised and
acid washed catalysts, which show near complete removal of
surface oxides following acid treatment. The continued
The procedure for preparation of 2.5% Fe/ZSM-5 (30) is as
follows; H-ZSM-5 (30) (2.50 g) was dried under vacuum (160 ºC,
3 h, 10-3 mbar) prior to sieving (40 mesh). Dried H-ZSM-5 (30)
(1.95 g) and Fe(acac)3 (0.316 g, 0.895 mmol) were then physically
mixed and transferred to a Schlenk flask. The flask was then
evacuated (10-3 mbar) and heated under vacuum for 2 h at 150
ºC. The resulting solid was then calcined in static air (550ºC, 3 h,
ramp rate 20 ºC min-1).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
Other supports used in this study include; Zeolite β (Zeolyst, SiO2/
Al2O3 = 25), Zeolite Y (Zeolyst, SiO2/ Al2O3 = 30), TS-1 (ACS
Reagents, SiO2/ TiO2 = 30), SiO2 (Silica Nanopowder, Sigma
Aldrich), aluminium oxide (Puriss, 98 %, Sigma Aldrich),
amorphous silica-alumina (SiO2/ Al2O3 = 10, for preparation
procedure see Supplementary Information Section 1.1) and
Silicalite-1 (for preparation procedure see Supplementary
Information Section 1.2). Where applicable, the zeolitic materials
aliquots of the final solution against Ce(SO4)2 using Ferroin
indicator.
Catalyst Characterisation
Powder X-ray diffraction was performed using a PANalytical
X’PertPRO X-ray diffractometer, with a CuΚα radiation source (40
kV and 40 mA) and Ni attenuator. Diffraction patterns were
recorded over a 2θ angular range of 5-75o employing a.0167 o
step size (time / step = 150 sec).
o
o
were activated in flowing air (3h, 550 C, 20 C min-1) prior to
testing or modification through metal impregnation.
NH3 – TPD was carried out using a CHEMBET TPR/ TPD
chemisorption analyser, Quantachrome Industries fitted with a
TCD. Samples were pre-treated for 1 h at 130 oC (15 oC min-1) in
a flow of He (80 ml min-1). NH3 was adsorbed at room
temperature, with physisorbed NH3 removed in a flow of He (80
Acid treatment methodology
The procedure for acid washing of a catalyst was as follows; the
sample (typically 1 g) was stirred in an aqueous solution of nitric
acid (50 mL, typically 10 v/v % HNO3, 2.4 M) for the desired time
(typically 0.25 h) at a temperature of 50 ºC. The sample was then
recovered by filtration and washed with deionised water until the
washings reached pH 7. The catalyst were then dried for 4 h at
110 ºC. Where applicable, the iron content of catalysts was
determined via ICP with HF digestion (Exeter Analytical
Services). The standard nomenclature employed hereafter is 2.5
wt. % Fe/ ZSM-5 (30) AT n, where ‘’AT’’ denotes acid treatment and
n = the duration of the acid treatment in hours.
o
o
ml min-1, 110 C, 1 h, 15 C min-1). Chemisorbed NH3 was then
desorbed by heating to a Tmax of 900 ºC (15 ºC min-1) in a flow of
He (80 ml min-1) during which period desorbed NH3 was
monitored using a TCD, (current 180 mV, attenuation 1).
27Al MAS NMR experiments were carried out on a 400 MHz
Varian CMX infinity spectrometer, equipped with 4.0 mm probe
1
with resonance frequencies of 400.1 and 100.4 MHz for H and
27Al respectively. Single-pulse 27Al experiments were performed
with a pulse length of 1 μs and a pulse delay of 1 s. The magic
angle spinning rate was set to be 8 kHz.
Catalyst Assessment
Catalyst testing for the oxidation of propane with added H2O2 was
carried out in a 50 ml stainless-steel Parr autoclave fitted with a
Teflon liner and a total workable volume of 35 ml. In a typical
experiment, the vessel was charged with an aqueous solution of
H2O2 (10 ml, 0.5 M, 5000 µmol) and the desired amount of catalyst
(typically 27 mg). After purging with helium, the system was
charged with propane (4 bar, 4000 µmol) and then the total
pressure was increased to 20 bar with He as diluent. The
autoclave was then heated to the desired reaction temperature
(typically 50 oC) with vigorous stirring (1500 rpm) and maintained
at a constant temperature for the desired reaction time (typically
0.5 h). After completion of the reaction the vessel was cooled in
BET analysis was conducted on an Autosorb-1 Quantachrome
instruments system at 77 K. A Monte Carlo based model was
used in determining pore volumes. Points in the range of 0.06 to
0.35 were used for the BET multi-point surface area
quantification.
UV-Vis spectra were collected using a Varian 4000 UV-Vis
spectrophotometer. Scans were collected across the wavelength
range 200-800 nm, at a scan rate of 150 nm/ min, with a UV-Vis
changeover wavelength of 260 nm.
o
ice to ca. 12 C and gas phase vented into a gas sampling bag.
Following this the liquid phase was recovered and filtered prior to
analysis.
Transmission electron microscopy (TEM) analysis was performed
using a JEOL 2100 microscope, operating at 200kV with a LaB6
electron source. The microscope was equipped with a Gatan
1000XP CCD camera for TEM work, a JEOL dark field and bright
field detectors for scanning transmission electron microscopy
(STEM) work, and an Oxford instruments X-MaxN 80mm2 Silicon
Drift Detector (SDD) for X-ray analysis. Specimens were
prepared by dispersing dry catalyst powders onto 400 mesh lacey
carbon TEM Cu grids (TAAB).
1
Liquid products were analysed by H NMR spectroscopy on a
Bruker 500 MHz Ultra-Shield NMR spectrometer and quantified
against a 1 vol% TMS/CDCl3 internal standard calibrated against
commercial standards. Gaseous phase products were analysed
using a Varian 450-GC fitted with a CP-Sil 5CB capillary column
(50 m length, 0.33 mm ID), a methaniser and both TCD and FID
detectors. The H2O2 conversion was quantified by titration of
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
IR spectra were collected on a Bruker Tensor 27 spectrometer
fitted with a liquid nitrogen-cooled mercury cadmium telluride
(MCT) detector. Samples were housed within a Praying Mantis
high temperature diffuse reflection environmental reaction
chamber (HVC-DRP-4) fitted with zinc selenide windows.
Samples were pre-treated prior to acquisition by heating the cell
to 200 ºC (10 ºC min-1) under continuous vacuum (10-3 mbar) and
maintained at this temperature for 2 h to ensure removal of
residual water. Multiple scans (64) were collected across the 4000
cm-1 to 1500 cm-1 wavenumber range, at 4 cm-1intervals.
[9] S. Hernández-Morejudo, A. Massó, E. García-González, P.
Concepción, J.M. López Nieto, Preparation, characterization and
catalytic behavior for propane partial oxidation of Ga-promoted
MoVTeO catalysts, Appl. Catal. A: General. 2015, 504, 51-61.
[10] W. Li, K. Oshihara, W. Ueda, Catalytic performance for
propane selective oxidation and surface properties of 12-
molybdophosphoric acid treated with pyridine, Appl. Catal. A.
1999, 182, 357-363.
[11] M.M. Lin, Selective oxidation of propane to acrylic acid with
molecular oxygen, Appl. Catal. A: General. 2001, 207, 1-16.
X-ray photoelectron spectra (XPS) were collected using a Kratos
Axis Ultra DLD system having a monochromatic Al Κα X-ray
source operating at 120 W. Data was collected in the hybrid mode
of operation, using a combination of magnetic and electrostatic
lenses, and at pass energies of 40 and 160 eV for high resolution
and survey spectra respectively. Magnetically confined charge
compensation was used to minimize sample charging and the
resulting spectra were calibrated to the Si (2p) line at 103.2 eV.
[12] E.K. Novakova, J.C. Védrine, E.G. Derouane, Propane
Oxidation on Mo–V–Sb–Nb Mixed-Oxide Catalysts: 1. Kinetic
and Mechanistic Studies, J. Catal. 2002, 211, 226-234.
[13] M. Sun, J. Zhang, P. Putaj, V. Caps, F. Lefebvre, J.
Pelletier, J.-M. Basset, Catalytic Oxidation of Light Alkanes (C1–
C4) by Heteropoly Compounds, Chem. Rev. 2014, 114, 981-
1019.
Keywords: Propane oxidation, zeolite catalysts, selective
oxidation, natural gas
[14] T. Mazari, C.R. Marchal, S. Hocine, N. Salhi, C. Rabia,
Oxidation of propane over substituted Keggin
[1] S.J. Blanksby, G.B. Ellison, Bond dissociation energies of
phosphomolybdate salts, J. Nat. Gas Chem. 2009, 18, 319-324.
organic molecules, Acc. Chem. Res. 2003, 36, 255-263.
[15] N.A. Mashayekhi, M.C. Kung, H.H. Kung, Selective
oxidation of hydrocarbons on supported Au catalysts, Catal.
Today. 2014, 238, 74-79.
[2] Y.R. Luo, Handbook of bond dissociation energies in organic
compounds, in, CRC Press, 2002.
[3] R.K. Grasselli, J.D. Burrington, Oxidation of Low-Molecular-
Weight Hydrocarbons, in: Handbook of Heterogeneous
Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, 2008.
[16] M.M. Forde, R.D. Armstrong, C. Hammond, Q. He, R.L.
Jenkins, S.A. Kondrat, N. Dimitratos, J.A. Lopez-Sanchez, S.H.
Taylor, D. Willock, C.J. Kiely, G.J. Hutchings, Partial Oxidation of
Ethane to Oxygenates Using Fe- and Cu-Containing ZSM‑5, J.
Am. Chem. Soc. 2013, 135, 11087-11099.
[4] N. Rahimi, R. Karimzadeh, Catalytic cracking of
hydrocarbons over modified ZSM-5 zeolites to produce light
olefins: A review, Appl. Catal. A: General. 2011, 398, 1-17.
[17] M.M. Forde, R.D. Armstrong, R. McVicker, P.P. Wells, N.
Dimitratos, Q. He, L. Lu, R.L. Jenkins, C. Hammond, J.A. Lopez-
Sanchez, C.J. Kiely, G.J. Hutchings, Light alkane oxidation
using catalysts prepared by chemical vapour impregnation:
tuning alcohol selectivity through catalyst pre-treatment, Chem.
Sci. 2014, 5, 3603-3616.
[5] R. Raja, C.R. Jacob, P. Ratnasamy, Direct oxidation of
propane to isopropanol, Catal. Today. 1999, 49, 171-175.
[6] S. Sifniades, A.B. Levy, H. Bahl, Acetone, in: Ullmann's
Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH
& Co. KGaA, 2000.
[18] C. Hammond, N. Dimitratos, J.A. Lopez-Sanchez, R.L.
Jenkins, G. Whiting, S.A. Kondrat, M.H.a. Rahim, M.M. Forde, A.
Thetford, H. Hagen, E.E. Stangland, J.M. Moulijn, S.H. Taylor,
D.J. Willock, G.J. Hutchings, Aqueous-Phase Methane
[7] F. Frusteri, L. Spadaro, C. Espro, A. Parmaliana, F. Arena,
Liquid-phase selective oxidation of propane on silica-supported
Nafion catalysts, J. Nat. Gas Chem. 2002, 11, 180-185.
Oxidation over Fe-MFI Zeolites; Promotion through Isomorphous
Framework Substitution, ACS Catal. 2013, 3, 1835-1844.
[8] E. Balcells, F. Borgmeier, I. Grißtede, H.G. Lintz, Partial
oxidation of propane and propene to acrylic acid over a Mo-V-
Te-Nb oxide catalyst, Catal. Letters. 2003, 87, 195-199.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
[19] C. Hammond, M.M. Forde, M.H.a. Rahim, A. Thetford, Q.
He, R.L. Jenkins, N. Dimitratos, J.A. Lopez-Sanchez, N.F.
Dummer, D.M. Murphy, A.F. Carley, S.H. Taylor, D.J. Willock,
Eric E. Stangland, J. Kang, H. Hagen, C.J. Kiely, G.J. Hutchings,
Direct Catalytic Conversion of Methane to Methanol in an
Aqueous Medium by using Copper-Promoted Fe-ZSM-5,
Angew. Chem. Int. Ed. 2012, 51, 5129-5133.
silicalite and Fe-SBA-15 with the same iron constitution, Appl.
Catal. B. 2006, 62, 244-254.
[28] J. Perez-Ramirez, J.C. Groen, A. Brueckner, M.S. Kumar,
U. Bentrup, M.N. Debbagh, L.A. Villaescusa, Evolution of
isomorphously substituted iron zeolites during activation:
comparison of Fe-β and Fe-ZSM-5, J. Catal. 2005, 232, 318-
334.
[20] C. Hammond, I. Hermans, N. Dimitratos, Biomimetic
Oxidation with Fe-ZSM-5 and H2O2? Identification of an Active,
Extra-Framework Binuclear Core and an FeIII OOH Intermediate
with Resonance- Enhanced Raman Spectroscopy,
ChemCatChem. 2015, 7, 434-440.
[29] B. Hunger, J. Hoffmann, O. Heitzsch, M. Hunger,
Temperature- programmed desorption (TPD) of ammonia from
HZSM-5 zeolites, J. Therm. Anal. 1990, 36, 1379-1391.
[30] K.H. Schnabel, C. Peuker, B. Parlitz, E. Loffler, U.
Kurschner, H. Kriegsmann, IR- spectroscopic investigation of the
adsorption of NH3 and H2O to Na-ZSM-5 and H-ZSM-5,
Zeitschrift Fur Physikalische Chemie-Leipzig. 1987, 268, 225-
234.
[21] A.K.M.L. Rahman, R. Indo, H. Hagiwara, T. Ishihara, Direct
conversion of ethane to acetic acid over H-ZSM-5 using H2O2 in
aqueous phase, Appl. Catal. A: General. 2013, 456, 82-87.
[22] A.K.M.L. Rahman, M. Kumashiro, T. Ishihara, Direct
synthesis of formic acid by partial oxidation of methane on H-
ZSM-5 solid acid catalyst, Catal. Commun. 2011, 12, 1198-
1200.
[31] H. Dang Lanh, D. Thi Thuy Hanh, J. Engeldinger, M.
Schneider, J. Radnik, M. Richter, A. Martin, TPR investigations
on the reducibility of Cu supported on Al2O3, zeolite Y and
SAPO-5, J. Solid State Chem. 2011, 184, 1915-1923.
[23] V. Peneau, G. Shaw, R.D. Armstrong, R.L. Jenkins, N.
Dimitratos, S.H. Taylor, H.W. Zanthoff, S. Peitz, G. Stochniol,
G.J. Hutchings, The partial oxidation of propane under mild
aqueous conditions with H2O2 and ZSM-5 catalysts, Cat. Sci.
Tech. 2016, 20, 7521-7531.
[32] C. Ding, X. Wang, X. Guo, S. Zhang, Characterization and
catalytic alkylation of hydrothermally dealuminated nanoscale
ZSM-5 zeolite catalyst, Catal. Commun. 2007, 9, 487-493.
[33] R.Q. Long, R.T. Yang, Temperature-programmed
desorption/surface reaction (TPD/TPSR) study of Fe-exchanged
ZSM-5 for selective catalytic reduction of nitric oxide by
ammonia, J. Catal. 2001, 198, 20-28.
[24] C. Hammond, R.L. Jenkins, N. Dimitratos, J.A. Lopez-
Sanchez, M.H.a. Rahim, M.M. Forde, A. Thetford, D.M. Murphy,
H. Hagen, E.E. Stangland, J.M. Moulijn, S.H.T.J. Willock, G.J.
Hutchings, Catalytic and mechanistic insights of the low-
temperature selective oxidation of methane over Cu-promoted
Fe-ZSM-5, Chem. Eur. J. 2012, 18, 15735-15745.
[34] M. Iwasaki, K. Yamazaki, K. Banno, H. Shinjoh,
Characterization of Fe/ZSM-5 DeNOx catalysts prepared by
different methods: Relationships between active Fe sites and
NH3-SCR performance, J. Catal. 2008, 260, 205-216.
[25] G.B. Shul'pin, G.V. Nizova, Y.N. Kozlov, L. Gonzalez
Cuervo, G. Süss-Fink, Hydrogen Peroxide Oxygenation of
Alkanes Including Methane and Ethane Catalyzed by Iron
Complexes in Acetonitrile, Adv. Synth. Catal. 2004, 346, 317-
332.
[35] B. Dou, G. Lv, C. Wang, Q. Hao, K. Hui, Cerium doped
copper/ZSM-5 catalysts used for the selective catalytic reduction
of nitrogen oxide with ammonia, Chemical Engineering Journal.
2015, 270, 549-556.
[26] R.D. Armstrong, S.J. Freakley, M.M. Forde, V. Peneau, R.L.
Jenkins, S.H. Taylor, J.A. Moulijn, D.J. Morgan, G.J. Hutchings,
Low temperature catalytic partial oxidation of ethane to
oxygenates by Fe– and Cu–ZSM-5 in a continuous flow reactor,
J. Catal. 2015, 330, 84-92.
[36] P.J. Kooyman, P. van der Waal, H. van Bekkum, Acid
dealumination of ZSM-5, Zeolites. 1997, 18, 50-53.
[37] Y. Wang, T. Yokoi, S. Namba, T. Tatsumi, Effects of
dealumination and desilication of beta zeolite on catalytic
performance in n-Hexane cracking, Catalysts. 2016, 6, 8.
[27] M.S. Kumar, J. Perez-Ramirez, M.N. Debbagh, B. Smarsly,
U. Bentrup, A. Brueckner, Evidence of the vital role of the pore
network on various catalytic conversions of N2O over Fe-
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
[38] C.A. Fyfe, J.M. Thomas, J. Klinowski, G.C. Gobbi, Magic-
Angle-Spinning NMR (MAS-NMR) Spectroscopy and the
Structure of Zeolites, Angew. Chem. Int. Ed. 1983, 22, 259-275.
[41] F.W. Schuetze, F. Roessner, J. Meusinger, H. Papp,
Hydrogen/deuterium exchange on dealuminated H-ZSM-5
zeolites studied by time resolved FTIR spectroscopy, Stud. Surf.
Sci. Catal. 1997, 112. 127-134.
[39] P. Marturano, L. Drozdová, A. Kogelbauer, R. Prins,
Fe/ZSM-5 prepared by sublimation of FeCl3: the structure of the
Fe species as determined by IR, 27Al MAS NMR, and EXAFS
spectroscopy, J. Catal. 2000, 192, 236-247.
[42] S. Bordiga, E. Escalona Platero, C. Otero Arean, C.
Lamberti, A. Zecchina, Low temperature carbon monoxide
adsorption on Na-ZSM-5 zeolites: an FTIR investigation, J.
Catal. 1992, 137, 179-185.
[40] G.A. Webb, The effects of paramagnetism on NMR spectra
of nuclei other than protons, in, Wiley, 1974, 127-142.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601241
ChemCatChem
FULL PAPER
Layout 1:
FULL PAPER
A framework for change:
Virginie Peneau, Robert D. Armstrong,
Greg Shaw, Jun Xu, Robert L. Jenkins,
David. J. Morgan, Nikolaos Dimitratos,
Stuart H. Taylor, Horst W. Zanthoff,
Stefan Peitz, Guido Stochniol, Qian
He, Christopher J. Kiely and Graham
J. Hutchings*
Extraframework cationic Fe species
in Fe/ZSM-5 zeolite catalysts are the
active site for the low temperature
partial oxidation of propane with
H2O2. Activity is intrinsic and unique
to this MFI- type zeolite framework,
with unprecedented TOFs of up to
1063 (h-1) observed. Surface oxide
species are found to be effective
spectators within the reaction.
Page No. – Page No.
The low temperature oxidation of
propane using H2O2 and Fe/ZSM-5
catalysts; elucidation of the active
site and enhancement of catalytic
This article is protected by copyright. All rights reserved.