Insights into the direct selective oxidation of methane to methanol over ZSM-5 zeolytes in aqueous hydrogen peroxide

Article abstract of DOI:10.1016/j.cattod.2018.03.031

The direct selective oxidation of methane by aqueous hydrogen peroxide over ZSM-5(50) has been investigated. Methyl hydroperoxide was confirmed as the initial product of methane oxidation. The decomposition of methyl hydroperoxide to formaldehyde is established as a key intermediate in an oxidation pathway to formic acid and ultimately CO₂. Hydrogen was detected during the oxidation of methane over ZSM-5 zeolites. The hydrogen was evaluated in terms of its origins and possible role in the production of MeOH. The addition of a copper salt to the ZSM-5(50) zeolite was found to decrease the overall productivity of all methane oxygenates. The use of copper also promoted the selectivity for methyl hydroperoxide whilst removing formic acid. The role of hydroxymethyl-methyl hydroperoxide is considered as an intermediate in the decomposition of methyl hydroperoxide to methanol. A tentative proposal into the nature of the iron species responsible for the catalytic species in ZSM-5(50) is made.

Full text of DOI:10.1016/j.cattod.2018.03.031

Accepted Manuscript
Title: Insights into the direct selective oxidation of methane to
methanol over ZSM-5 zeolytes in aqueous hydrogen peroxide
Authors: Saeed Al-Shihri, Christian J. Richard, Hamid
Al-Megren, David Chadwick
PII:
S0920-5861(18)30261-X
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2018.03.031
CATTOD 11314
To appear in:
Catalysis Today
Received date:
Revised date:
Accepted date:
8-11-2017
9-3-2018
15-3-2018
Please cite this article as: Al-Shihri S, Richard CJ, Al-Megren H, Chadwick
D, Insights into the direct selective oxidation of methane to methanol
over ZSM-5 zeolytes in aqueous hydrogen peroxide, Catalysis Today (2010),
https://doi.org/10.1016/j.cattod.2018.03.031
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Insights into the Direct Selective Oxidation of Methane to Methanol over ZSM-5
Zeolytes in Aqueous Hydrogen Peroxide
Saeed Al-Shihria*, Christian J Richardc, Hamid Al-Megrena, David Chadwickb
a
National Petrochemical Technology Center, Materials Science Research Institute, King
Abdulaziz City for Science and Technology (KACST), P. O.Box6086, Riyadh11442, Saudi
Arabia
b Department of Chemical Engineering, Imperial College London, London, UK
c Mass Spectrometry Facility, King’s College London, London SE1 9NH.
*Corresponding author: alshihri@kacst.edu.sa,
Graphical abstract
RESEARCH HIGHLIGHTS:
 Hydrogen evolution during the oxidation of methane with ZSM-5 and
hydrogen peroxide.
 Formaldehyde is an intermediate in the oxidation reaction pathway.
 Total productivity of liquid oxygenated products is higher.
 Adding copper to ZSM-5(50) resulted in overall decrease in oxygenates
productivity and increase in MeOOH selectivity.
Abstract
The direct selective oxidation of methane by aqueous hydrogen peroxide over ZSM-
5(50) has been investigated. Methyl hydroperoxide was confirmed as the initial product of
methane oxidation. The decomposition of methyl hydroperoxide to formaldehyde is
1

established as a key intermediate in an oxidation pathway to formic acid and ultimately CO2.
Hydrogen was detected during the oxidation of methane over ZSM-5 zeolites. The hydrogen
was evaluated in terms of its origins and possible role in the production of MeOH. The addition
of a copper salt to the ZSM-5(50) zeolite was found to decrease the overall productivity of all
methane oxygenates. The use of copper also promoted the selectivity for methyl
hydroperoxide whilst removing formic acid. The role of hydroxymethyl-methyl hydroperoxide
is considered as an intermediate in the decomposition of methyl hydroperoxide to methanol.
A tentative proposal into the nature of the iron species responsible for the catalytic species in
ZSM-5(50) is made.
Key words: methane oxidation, ZSM-5, catalysis, methanol, hydrogen peroxide,
formaldehyde, copper.
1.0 Introduction
Methane is the main constituent of natural gas, which according to an estimate made
in 2005, occupies a global volume of 190 trillion cubic meters [1]. Accordingly methane is
abundant, cheap and a widely used carbon based feedstock for the production of platform
chemicals such as methanol(MeOH), formaldehyde(HCOOH), and formic acid(HCOOH).
Although MeOH has also been deemed as a viable “clean” fuel, the current technologies
required to produce it are infrastructure and energy intensive and need multistep gas phase
reactions over heterogeneous Cu-ZnO catalysts [2,3]. In view of this, a current goal in research
is to identity cost effective “direct” oxidation pathways for the convert methane to MeOH, in a
one step process [2,4].
The first aqueous system to convert methane to MeOH with 100% selectivity was
2
reported by Shilov who investigated the homogenous catalyst system PtCl42-/PtCl6 , at 130°C.
In this system the terminal oxidant was a Pt(IV) complex [5]. However cheaper metals such
as iron and copper also oxidise methane to MeOH if hydrogen peroxide is the oxidant; though
the methanol selectivities are low [6]. These metals operate via Fenton reactions which are
highly pH dependant. For example, high pH favours the copper catalysed oxidation reactions
as opposed to low pH which favours the iron catalysed oxidations [6]. In nature methane is
oxidised by the iron containing soluble (sMMO) and copper containing particulate (pMMO)
2

methane monoxygenases in methanotrophic bacteria [7,8]. Attempts to prepare biomimetic
equivalents have been partially successful. For example a copper Cu(II)–O–Cu(II)
configuration on a Cu-ZSM-5 zeolite catalysed a heterogeneous methane oxidisation reaction
with high selectivity under mild conditions [9]. An important feature of copper is that it has
accessible Cu(0), Cu(I), Cu(II), and Cu(III) oxidation states making one-electron or two-
electron processes relatively facile. As a result, both radical and two-electron bond-forming
pathways via organometallic intermediates are possible, similar to those of palladium. In
addition, the different oxidation states of copper allow it to bond with a number of different
functional groups via Lewis acid interactions or π-coordination. Not surprisingly a number of
zeolite systems with copper are catalytic, some being of relevance to MeOH production [10].
In a land mark study of especial significance to this work, Hammond et al studied the
oxidation of methane with hydrogen peroxide over a range of heterogeneous ZSM-5 zeolites.
ZSM-5(30) was found to be the most active zeolite on account of its inherent levels of trace
iron [11] [12]. According to DFT calculations and EXAF studies, some of this iron content
formed highly cationic and highly catalytic clusters which were postulated to be: [Fe2(μ2-
OH)2(OH)2(H2O)2]2+ (abbreviated to [Fe2O2]2+ in this text). Both iron centres in [Fe2O2]2+ were
antiferromagnetically coupled as high-spin Fe3+ and also stabilized by the Al3+ cation-
exchange sites in the zeolite framework. These Al3+ cation-exchange sites also prevented the
aggregation of [Fe2O2]2+ into inactive clusters in the zeolyte microspores [12]. Regarding
catalysis, the most favorable Fe content of given clusters was thought to range from 2 to 3 Fe
atoms [13, 14]. In any case the catalytic [Fe2O2]2+ clusters were deemed responsible for both
the activation of methane to give methyl hydroperoxide (MeOOH) as a primary product and
also its decomposition. Although homogeneous occurred , the decomposition of MeOOH over
ZSM-5(30) was proposed to result in the loss of an •OH radical from {[Fe2O2]n+}-MeOOH which
then led to an iron-bonded methoxy group: {[Fe2O2]n+}-OMe. Following hydrolysis the catalyst
was regenerated and MeOH was produced. It was noted that the methanol was also prone to
over-oxidation by adventitious •OH radicals over iron oxides [11] to give HCOOH and CO2 by
the sequential decomposition of MeOOH. This was according to steps, B and C in Scheme
3

1[11][15]. In addition, CH4 was also oxidized by an uncharacterized parallel non-sequential
pathways which also gave CO2 directly [11].
In contrast to the methane oxidation products summarized in Scheme 1 [11, 12 ,15],Table 1
shows that formaldehyde (HCHO) and hydrogen (H2) were also produced when the methane
oxidations were carried out over ZSM-5(50), as described further below.
In Table 1 it can be seen that the distribution of the main liquid oxygenate products was very
similar to those reported by the Hutching’s group [11][12]. For example the MeOOH selectivity
at similar reaction times in the present results was 27.5 % for ZSM-5(50), and 23% for ZSM-
5(30), Supporting Information Table S2, compared to 23% reported earlier for ZSM-5(30) [11].
A slightly higher value for MeOOH (43%) was reported for ZSM-5(30) by Kalamaras et al [17].
Therefore, ZSM-5(50) was the main zeolite used in this work because of its high selectivity for
MeOOH. However, HCHO only made a small contribution to the product yield as measured
by 1H NMR. The presence of HCHO was not reported in any of the earlier studies [11, 12, 15,
18]. Moreover neither was the gaseous hydrogen product [11, 12, 15, 16, 18].
The effect of sample filtration was examined in all our standard methane oxidation
experiments over ZSM-5(50). Each the sample was cold-filtered through Celite® S to remove
any zeolite colloids or particulate matter before NMR analysis. As expected, the solutions
1
became clear. Filtration removed the dissolved methane signal in the H NMR. The product
analysis over ZSM-5(50) for the filtered sample is also given in Table 1. Furthermore the effect
of filtration on the liquid oxygenate product distribution was marginal as shown by comparing
the calculated TOF values. In general, the MeOOH yield was found to be slightly higher in the
filtered samples probably due to the removal of colloidal catalyst entities which were found to
4

continue to slowly catalyse reactions in the NMR tube producing increasing amounts of
HCOOH.
In addition to the liquid products, samples of the head space gas were taken and analysed by
GC/TCD using He and Ar as carrier gas, and by MS. Hydrogen was found to be the main
gaseous product as shown in Table 1. CO2 was barely detectable in the gas phase. Taking
10ppm as a conservative minimum sensitivity, we estimate the amount of CO2 in the gas
phase is about 0.3µmol. Based on our measured CO2 solubility in dilute H2O2 (0.33 mol.l-
1bar1), we estimated that the total CO2 produced was 2µmol. This level of CO2 is was in
reasonable agreement with previous studies on ZSM-5 zeolites [12, 15, 18], under the
experimental conditions used. It should be noted that the presence of CO, ethylene, or other
carbon based gases were not detected.
We propose that the ZSM-5(50) zeolite used by us contained a different subset of iron species
that that these catalysed the decomposition of MeOOH to HCHO which is depicted in
Scheme2:
Significantly, the occurrence of HCHO was not detected by Hammond et al who assumed that
it was either rapidly oxidised to formic acid or not formed at all [11]. We cannot therefore
confirm that the HCHO was also produced in the ZSM-5(30) reported system [11] but only
speculate on its presence. A noteworthy aspect of this work was the higher total liquid
oxygenate productivity compared to the values reported earlier [12,18]. Considering just the
productivity of MeOOH, MeOH, and HCOOH in the ZSM-5(50) catalysed reaction, we obtain
24.4 and 20.5 mol kgcat-1h-1, compared to 2.7 and 5.7 mol kgcat-1h-1 for ZSM-5(30) respectively
which was reported earlier [12, 18], and 2.33 mol kgcat-1h-1 reported for ZSM-5(30) by
Kalamaras et al [15]. This is despite the lower methane pressure (10 bar) used in the present
work. A possible explanation is that the batches of ZSM-5 used in the present study had a
higher Fe content (e.g: ZSM-5(50), 342 ppm), compared to the zeolite samples used in the
previous studies: for example ZSM-5(50) = 210 ppm and ZSM- 5(30) = 140 ppm, [11,12] and
ZSM-5(30) = 123 ppm [17]. Moreover, there were also significant differences in the
5

experimental procedures and analytical methodologies used such as the analysis of unfiltered
samples by NMR in the present work. Interestingly, the total amounts of the liquid oxygenate
products (MeOOH, MeOH, and HCOOH) were not dissimilar from the previous works [12,18].
For example 138.7 µmol for ZSM-5(50), compared to 77.1 µmol for ZSM-5(30) and 162.1
compared to 36.5 µmol [15]. Admittedly the value for ZSM-5(50) was much higher, but our
batch had a much higher Fe content as noted above. Therefore the differences in productivity
for similar ZSM-5 zeolites may simply reflect the lower amount of catalyst used in the present
study, thus implying that physical factors may begin to influence the productivity at higher
catalyst mass. The methane conversions in the present work are estimated to be around 1%,
Table 1 and Supporting Information Table S2, but it is important to note that methane
conversion is also dependent on the ratio of head space to liquid volume for the particular
reactor used.
The following is an extension to a previous parametric study [16] on the direct oxidation
of methane in aqueous hydrogen peroxide over purchased ZSM-5 zeolites [16]. However the
main focus of this study was on the zeolite ZSM-5(50). The chemical reactivity of HCHO is
evaluated further and the role of an added copper salt to ZSM-5(50) is examined with respect
to the selectivity of MeOOH and MeOH. A novel reaction scheme for the production of
methanol is proposed in which H3COOCH2OH is a plausible precursor to MeOH and HCHO.
Speculations into the nature of catalytically active Fe sites is made.
2.0 Experimental
Zeolite catalyst ZSM-5 (SiO2/Al2O3 mole ratio 23, 30, 50 and 80), was studied. It was activated
by calcination (550ᴼC for 3hr) as in previous methane oxidations.[16] However the main
catalyst studied in this work was ZSM-5(50) which was found to be optimal for this work. All
reactions were magnetically stirred in a stainless-steel autoclave (Parr reactor) containing a
non-proprietary glass liner (50 ml) and glass coated stir bar to eliminate the possibility of metal
leaching into solution from reactor components. A stainless steel thermocouple was in contact
with the reaction liquid to measure the reaction temperature. It was established in a blank run
6

that this did not contribute to any conversions by leeched metals. The glass reactor liners were
always cleaned in concentrated nitric acid to remove adventitious metals. The optimum
concentration for methane oxidation experiments was 0.5 moles/L H2O2.[11]. The starting
H2O2 was standardised against permanganate. In a typical experiment, de-ionized water and
catalyst (13.5 mg) were added to the autoclave and maintained at 50ᴼC by the immersion of
the reactor body in a temperature-controlled liquid bath. After the desired reaction time, the
reaction products were cooled externally in ice(during which the ZSM-5 zeolite settled). A
sample of the reaction mixture was prepared immediately for analysis by proton NMR. The
H2O2 (0.5 M, 10 ml) added at the reaction temperature (50ᴼC) to avoid its decomposition
during the warm up period. The addition of H2O2 generally caused the reaction mixture to
effervesce; presumably releasing oxygen. The sealed reactor was then pressurised with
Argon (99.99%, BOC) as a further measure to purge air, and then re-pressurised with methane
(10 bar), (25% CH4 in Argon, BOC). After 30 min, with vigorous stirring, the autoclave was
cooled externally to 5ᴼC in an ice bath to minimise any chemical processes and reduce the
loss of volatile products. A gas bag was used to collect gaseous products which were analysed
by a Shimadzu GC equipped with a TCD detector and a packed column (length10 Ft, 1/8 in,
ID 2 mm, He carrier gas) to determine gas phase CH4 and CO2 which was quantified against
the appropriate calibration curves. The pH of the reaction mixtures was generally in the region
of 6-7. In the following text the above experimental protocol is referred to as a Standard
Methane Oxidation(StMO) experiment whether run in the presence of methane or zeolite or
not. Liquid phase products were analysed by both GC-FID (capillary column, length 30 m, ID
0.32 mm, Film 0.25 µm, 7 inch cage, He carrier gas) and 1H NMR. The NMR measurements
were made at 298K using a Bruker AVANCE III HD 500 spectrometer running TopSpin 3.2
and equipped with a z-gradient 5mm tuneable SmartProbe and a GRASP II gradient
spectroscopy accessory providing a maximum gradient output of 53.5G/cm (5.35G/cmA). The
spectra were collected at a frequency of 500.13MHz using the Bruker pulse program zgesgp
[19] with a spectral width of 44642.86Hz (89.2ppm) and 262144 data points. A relaxation delay
of 1s was employed and the centre of the spectrum was placed on the water signal (~4.7ppm).
7

Water suppression was achieved using an 180o selective square shaped pulse
(Squa100.1000) of 2ms duration and smoothed square shaped gradients (SMSQ10.100) of
1ms duration and strengths of 31% and 11%. Typically, 512 transients were collected.The
data was processed using an exponential function with a line broadening of 0.3Hz and 262144
data points to give a 1st order zero filling.
Hydrogen peroxide concentrations were measured using a reported procedure using
Ti(IV)SO4, because it does not react with formaldehyde or hydroperoxides [20]. The amounts
of products: MeOOH, MeOH, HCHO and HCOOH were quantified using externally prepared
calibration curves [16]. This approach obviated the need to analyse each sample in the
presence of an internal standard since it was found to be unstable in residual H2O2 present in
the samples, and in any case overlapped with the signal from dissolved methane. The initial
calibration curves were updated after three months and found to have remained virtually
identical confirming that the NMR spectrometer had remained correctly tuned.
2.1 Catalyst characterisation
Commercial ZSM-5 zeolites with nominal SiO2/Al2O3 were used in this work. The properties of
the commercial ZSM-5 used are given in Table S1. The XRD of the zeolites are given in
Figure S1 and are consistent with the ZSM-5 structure.
2.3 Analytical Calculations
Definitions: Partial oxygenates; MeOOH(aq), MeOH(aq) and HCOOH(aq)
.
Conversion = (moles produced)/ (initial moles methane) x 100
Product Selectivity = moles (product)/moles (total liquid produced) x 100;
Total liquid productivity = moles (products)/weight (catalyst)/time;
Turnover Frequency (TOF) = moles(partial oxygenates species formed) mol-1 (Fe) h-1
The conversions, selectivities and catalyst productivities were estimated and calculated
according to a reported method [14]. To obtain the exact values of these calculations, an
8

exhaustive analysis of the methane conversion would be needed. However, the products from
the above ZSM-5 reactions [11,12,14] have been extensively characterised and are believed
to account for most of the methane conversion.
3.0 Results and discussion.
According to Hammond et al neither hydrogen nor HCHO were detected in the closed
ZSM-5(30) catalytic methane oxidation system summarised in Scheme 1 [11]. In this work
however, the formation of HCHO was detected and is shown in Scheme 2. The homogeneous
decomposition of MeOOH was assumed to produce HCHO; presumably promoted by OH•
radicals from the decomposition of aqueous H2O2. CH3OOH is a potent oxidising agent [21]
and relatively stable at room temperature, but less reactive than H2O2.[21] The promotional
role of H2O2 in this respect has been previously reported by Suss Fink et al [22]. Moreover,
gas phase studies have shown that •OH radicals led to the homolytic cleavage of the O-O
bond in CH3OOH, to give HCHO and water [21]; a reaction which is also catalysed
heterogeneously by contact with the walls of reactors[23]. However, in another study using
GC-MS equipped for chemical ionisation methods, protonated CH3OOH2+ was observed to
loose water to give a formaldehyde radical ion HCHO•+[19]. In view of the protonic sites in
+
ZSM-5 zeolites the protonation of MeOOH to CH3OOH2 might be possible [38,48].
In the ZSM-5(50) catalysed oxidation system (Scheme 2) hydrogen built up gradually
in the gas phase, indicating it was formed throughout given experimental runs, the results of
three of them being listed in Table 2. The results are consistent with the source of hydrogen
being related to one of the early intermediates in the methane oxidation reaction sequence.
The concentration of CO2 in the gas phase was very low and barely measurable after a 30 min
reaction time as noted above.
Since hydrogen must have been produced from the decomposition of MeOOH and /or its liquid
phase products, we carried out an independent study of HCHO, MeOH and HCOOH oxidation
at the typical StMO conditions over ZSM-5(50), to establish the presence or absence of
hydrogen. The results are given in Table 3
9

From Table 3 it is obvious that MeOH was oxidised by H2O2 to HCOOH (22.5% conversion)
and that no H2 was detected. Formic acid was relatively stable over ZSM-5(50) in the
presence of H2O2 at 50oC (19% conversion) yielding only CO2 and, importantly, no hydrogen
(Eq 1). The oxidation of formic acid and methanol by H2O2 is well known[24][25] and occurs
according to Eq 1:
HCOOH + H2O2
2CO2 + H2O
Eq 1
The promotional role of Fenton oxidations in the production of adventitious .OH radicals from
H2O2 has been commented on above [11]. Therefore Fe may have catalysed Equation 1.
From the TOF values in Table 3 it is clear that the oxidation of HCHO was at least 3 times
faster than the oxidation of MeOH or HCOOH. Only the oxidation of HCHO produced
hydrogen.
3.1 Evaluations on the Origins of Hydrogen from Formaldehyde
Oxidation of HCHO was carried out using a 37wt% solution of HCHO stabilised by 15
wt% methanol. It should be noted that the net amount of MeOH produced was estimated by
subtracting the amount of MeOH present as a stabilizer in the purchased commercial HCHO
used from the overall amount of MeOH in the reaction products. The HCHO was rapidly
oxidised by H2O2 over ZSM-5(50) (Table 3 Entry 2: 75% conversion). Therefore, no account
was taken of the amount of methanol converted to HCOOH as this was slow relative to the
extent of oxidation of HCHO..In contrast to the oxidation of methanol or formic acid, the
oxidation of formaldehyde resulted in H2 being evolved in significant amounts, Table 3 (Entry
2). Furthermore, it was confirmed that H2 was not produced at StMO conditions from H2O2 and
catalyst in the absence of methane or HCHO. This demonstrates that ZSM-5(50) does not
decompose H2O2 to H2 and O2 and underlines that the source of H2 is from the oxidation of
HCHO by H2O2. The evolution of hydrogen can be rationalised from the works of Fry and
Payne. In the 1930’s [24,25] they investigated the uncatalysed reaction chemistry of HCHO in
acidified aqueous solutions of H2O2. These reactions form a basis for understanding the
production of hydrogen over ZSM-5. The oxidation of HCHO is summarised by Equations 2-
3 below.
10

HCHO + H2O2
2HCHO + H2O2
HCOOH + H2O
2HCOOH + H2
Eq 2
Eq 3
Figure 1 shows a proposed reaction mechanism for the homogeneously evolution of
hydrogen from a bridged peroxy-methanediolate intermediate [26].
High concentrations of H2O2 oxidation favoured formic acid via Eq 2. Fry and Payne also
reported that the production of hydrogen (Eq 3) was favoured at lower concentrations of H2O2
[24, 25] typical of the “standard” oxidation conditions used and represented in Table 1. In our
methane oxidation experiments, the progressive decrease in the concentration of H2O2
caused by its consumption in the oxidation of methane and its decomposition over ZSM-5(50)
should therefore have contributed to the promotion of H2 production from HCHO.
It should be noted that the oxidation of HCHO at “standard” StMO conditions in the absence
of catalyst resulted in only a very small conversion (<5%) to formic acid, which is consistent
with the homogeneous reaction studies of Fry and Payne [24] done at 60oC. Overall, this
shows that the reactions are mainly catalysed by the ZSM-5.
3.2 Speculations the Occurrence of HCHO and MeOH Over ZSM-5(50)
The presence of HCHO and hydrogen in the ZSM-5(50) suggests that the nature of the
catalytic species that were responsible for the decomposition of MeOOH in Scheme 1(Path
A) were different. According to Hammond et al,[11][12] the Fe catalytic centres which were
both responsible for the conversion of methane to MeOOH (Equations 4 - 7) were also
responsible for the subsequent decomposition of MeOOH to give MeOH(Equations 8 - 10).
The following Equations 9-10 outline the reported catalytic cycle for the production and
conversion of MeOOH to MeOH [11][12].
[Fe2(µ2-OH)2(OH)2(H2O)2]2+ (I)+ H2O2
[Fe2(µ2-OH)2(OH)2(HOOH)( H2O)] 2+ + H2O
Eq 4
Eq 5
[Fe2(µ2-OH)2(OH)2(HOOH)( H2O)] 2+
[Fe2(µ2-OH)2(OH)(OOH)( H2O)2]2+
11

[Fe2(µ2-OH)2(OH)(OOH)( H2O)2]2+ + H2O2
[Fe2(µ2-OH)2(=O)(OH)(OOH)( H2O)]2+ + 2H2O
Eq 6
Eq 7
[Fe2(µ2-OH)2(=O)(OH)(OOH)( H2O)]2+ + CH4(rate limiting)
[Fe2(µ2-OH)2(OH)2(CH3OOH)( H2O)]2+ + H2O2
[Fe2(µ2-OH)2(OH)2(CH3OOH)( H2O)]2+
Fe2(µ2-OH)2(OH)2(HOOH)( H2O)] 2+ + CH3OOH Eq 8
Decomposition of MeOOH at [Fe] to HCHO:
[Fe2(µ2-OH)2(OH)2(CH3OOH)( H2O)]2+
[Fe2(µ2-OH)2(OH)2(CH3O)( H2O)]2+ + H2O
[Fe2(µ2-OH)2(OH)2(CH3O)( H2O)]2+ + •OH
Eq 9
[Fe2(µ2-OH)2(OH)2(HO)( H2O)]2+ + MeOH
Eq 10
Addition of H2O2 to [Fe2(µ2-OH)2(OH)2(HO)( H2O)]2+ to regenerate catalyst:
[Fe2(µ2-OH)2(OH)2(HO)( H2O)]2+ + H2O2
[Fe2(µ2-OH)2(OH)2(HOOH)( H2O)]2+
[Fe2(µ2-OH)2(OH)2(HOOH)( H2O)]2+ + •OH
Eq 11
[Fe2(µ2-OH)2(=O)(OH)(OOH)( H2O)]2+ + 2H2O
Eq 6-7
Equations 9-11 formed a closed system in which H2O2 was conserved by the recombination
of two •OH radicals. Regarding Equations 7 Hammond et al suggested that a non-diffusive
•CH3 radical was formed at adjacent Fe(IV)=O and [Fe]-OOH sites which were generated
when H2O2 reacted with [Fe2O2]2+ [11, 12]. Thus, the proposed hydrogen abstraction at
Fe(IV)=O to give •CH3 was followed by the instantaneous incorporation of •CH3 into the
hydroperoxy moiety: [Fe]-OOH. Therefore, the activation of the C-H bond was not by a
standard heterolytic mechanism at Brönsted/Lewis and acid/base sites, since Silicalite-1(an
iron free zeolite analogue) was inactive [12, 27, 28]. Overall Equations 9 - 11 define the
decomposition of MeOOH by the loss of an •OH radical to give {[Fe2O2]n+}-OMe, and hydrolysis
to yield MeOH and a new catalytic cycle. However, taking the listed catalytic steps in
Equations 8-11 as a reference benchmark, the detection of hydrogen and HCHO in the ZSM-
5(50) system in Scheme 2, might imply that the catalytic species defined in Equations 9 – 11
were operating in a different mode to produce HCHO instead of MeOH. In effect, the catalysed
decomposition of MeOOH might have occurred at completely different [Fe] sites; oligomeric
or isolated. Alternatively the MeOOH may have been prematurely released (Equations 8) to
12

homogeneously form HCHO and HCOOH in aqueous solution [22]. Whether, the MeOOH
decomposed to HCHO at Fe sites or over the acid sites of the zeolite is not clear as both
processes are in principle possible and indeed may take place in parallel. In an attempt to
rationalise the origins of MeOH in our ZSM-5(50) catalysed system, it was speculated that
MeOOH could have become reduced by hydride anion produced by ZM-5(50). In fact MeOOH
can be reduced to MeOH by NaBH4 [22]. To test this hypothesis, methane was oxidised under
our StMO conditions for a short time (5 - 15 min) (since the selectivity to MeOOH is in the
range 60-70%,). After these times, the methane was purged rapidly and replaced with 5 bar
of hydrogen for a further 5, 15 and 30 min. The results of three separate experiment are shown
in Table 4. The NMR results for a five minute reaction run without hydrogen is shown in the
supporting information, S2.2.
Comparing the overall reactions run for 10 and 30 mins run with and without hydrogen Table
4 shows that the MeOOH selectivity dropped to a low value and the main product was HCOOH
with little evidence of a large increase in methanol selectivity, possibly because MeOOH
decomposed to so HCHO rapidly that it reacted faster to give HCOOH than MeOH. Moreover,
no CO or CO2 was detected. Overall the results suggest that under methane selective
oxidation conditions, MeOOH decomposed to MeOH and HCHO; the latter becoming rapidly
oxidised to HCOOH. On the whole it was deemed that HCHO was an intermediate Path A and
in the “uncharacterised non sequential” production of HCOOH shown in Scheme 1 [11, 15],
which is discussed below. Notwithstanding, the possibility of hydrogenation cannot be
completely excluded although there is no evidence that either HCHO or MeOOH is
hydrogenated to methanol under StMO reaction conditions. The role of hydrogen on the
decomposition of MeOOH is unexplained, but may have been initiated by its protonation at
Brönsted sites in the zeolite micropores.
3.3 Oxidation of Methane in the Presence of a Copper Salt
One of the most notable findings reported by Hammond et al [11] was that the addition of the
appropriate concentration of copper to ZSM-5 switched the selectivity of HCOOH to 0% giving
13

rise to a concomitant increase in the methanol selectivity [11]. The overall oxygenate
productivity was however unchanged [11,12]. The copper was effective when added
homogeneously to give an uncharacterised mechanical mix of zeolite and copper nitrate, or
already present as a heterogeneous additive in ZSM-5(30). The most effective system was
Cu/Fe3+-ZSM-5(30) (2.5% loading of Fe3+) which gave a MeOH selectivity of about 80%. The
role of copper was postulated to prevent the non-selective over-oxidations of methanol by
Fenton type •OH radicals, derived from H2O2. Significantly, Cu2+ salts are potent •OH
scavengers. In this respect it was suggested that the copper modified the electronic properties
of the [Fe2O2]2+ centres, giving a unique Cu/[Fe] configuration which removed •OH radicals
derived from hydrogen peroxide (H2O2), without affecting the catalytic performance of the
primary methane oxidation step. Although the nature of Cu/[Fe] configuration was not
explained by Hammond et al [11], the suppression •OH radicals was deemed unusual since
copper it is known to promote selective oxidation in Fenton systems [11, 30, 31].
To examine the effects of copper on MeOH selectivity, copper was added to ZSM-5(50) and
methane oxidised at our standard StMO conditions. Following a similar procedure to that
adopted by Hammond et al [11], a solution of Cu(NO3)2(aq) was added to ZSM-5(50) in
concentrations which ranged from 0.5μmol to 2μmol. The mechanical mix of Cu(NO3)2(aq) and
ZSM-5(50) produced a marked change in the reaction solution which turned orange, possibly
due to colloidal copper. Figures 2 and 3 show the 1H NMR spectra of unfiltered products run
without and with the addition of Cu(NO3)2(aq) in typical StMO experiments.
The presence of dissolved methane, MeOOH, hydrated HCHO, and MeOH are clearly
identified in Figures 2 and 3. However formic acid is only present in Figure 2. The chemical
shifts of all methane oxygenates were validated against purchased compounds and published
values [12, 22, 33] and in the case of methyl hydroperoxide also by synthesis of the compound
according to a published procedure [34]. The amounts of products MeOOH, MeOH, HCHO
and HCOOH were quantified using externally prepared calibration curves as previously
14

reported [16]. The 1H NMR spectrum in Figure 2 also has two notable peaks at 1.95 and 2.1
ppm. We initially attributed these signals to water-soluble polyoxomethylenes (POM) which
have a molecular formula of HO(CH2O)nH [16]. However these peaks might be due to an
unknown impurity which appeared erratically in samples.
Figures 4(a-b) show that the productivity of HCOOH fell rapidly with increasing
Cu(NO3)2 as previously reported by Hammond et al [11, 12]. However the selectivity of
MeOOH rose; an unexpected result. In contrast to the reported results [11] the MeOH
selectivity never exceeded the MeOOH selectivity [16].
In the presence of approximately 1.0 μmol of Cu(NO3)2, the selectivity of MeOOH dominant,
presumably because the •OH radicals from H2O2 or MeOOH were suppressed, making
MeOOH less prone to decomposition by H2O2. At concentrations of Cu(NO3)2 above 1.0 μmol,
the concentration of MeOOH remained almost constant, only gradually decomposing to give
a corresponding increase in HCHO selectivity. Overall, Figure 4a shows that copper led to an
overall reduction of oxygenate productivity which was not apparent in the Hammond system
[11]. Since transition metals decompose H2O2, the preservation of MeOOH and HCHO was
attributed to the ability of Cu(NO3)2 to decompose H2O2. To establish the effectiveness of
Cu/ZSM-5(50) at decomposing H2O2 the listed StMO experiments shown in Table 5 were run,
to measure the decomposition of H2O2. In the presence of ZSM-5(50) alone, the H2O2
decreased ca 14 times within 30 mins; whereas in the combined presence of Cu(NO3)2, the
concentration of H2O2 decreased from its initial value by ca 40 - 100 times in the same time.
It should be noted that the addition of Cu(NO3)2 produced an immediate change in the reaction
solution which turned orange, possibly due to colloidal Cu2O or metallic copper. This was the
case in all the reactions studied with copper. Interestingly, the addition of H2O2 to Cu(NO3)2 is
a known reaction which gives a CuO2H hydroperoxide under mildly acidic conditions [35], the
formation of which would also remove available H2O2. Additionally, an independent
experiment showed that Cu(NO3)2 did not react with HCHO with ZSM-5(50), though it has
been reported to do so under alkaline conditions [36].
15

In Scheme 1, the addition of Copper Nitrate to ZSM-5(30) system in Scheme 1 inhibited the
adventitious oxidation of MeOH to HCOOH in path B [11]. Since the production of HCOOH
fell drastically in the presence of copper (Fig 4 a-b), StMO experiments undertaken to
establish the nature of the HCOOH decomposition by Cu/ZSM-5(50). To do this samples of
the head space decomposition gases were taken and analysed by GC. Both Hydrogen and
CO2 were the main gaseous products as tabulated in Table 6.
The results In Table 6 (Entry 1) show that Cu(NO3)2 and ZSM-5(50) decarboxylated HCOOH
to H2 and CO2. However, the addition of H2O2 led to the highest HCOOH conversion and H2
selectivity (Table 6, Entry 2). On the other hand Table 6 (Entries 3 and 4) show that no H2
was detected in the absence of ZSM-5(50), though HCOOH was oxidised to CO2 (and water)
by H2O2, Therefore HCOOH is a potent source of hydrogen in the presence of copper and
ZSM-5(50). The production of hydrogen was not reported in the Hammond Cu/ZSM-5(30)
system, despite the substantial amount of CO2 evolved from the “uncharacterised” oxidations
of CH4; presumably via formic acid [11, 12]. Interestingly, In Figure 4 b and S4, the addition
of 1.0µmol of copper to ZSM-5(50) produced a corresponding increase in MeOOH selectivity
by 29%, and a corresponding decrease in the concentration of H2O2 by 11.2%. This may have
been due to the much higher [Fe] content in the ZSM-5(50) used by us, which contained iron
species that promoted hydrogen peroxide decomposition.
ZSM-5 is a multifunctional material possessing both Brønsted and Lewis acidity, which is due
to associated with Al3+ [14, 38]. The catalytic activity of ZSM-5 with respect to methane
oxidation was shown not to depend on the presence of Brønsted acid sites [14]. This was
even in the case of H-ZSM-5 [14]. However, non-catalytic framework Al3+ was considered to
16

promote the dispersion of potentially catalytic site-specific [Fe], present as oligonuclear or
isolated species [14]. Moreover, the dispersion reduced unwanted [Fe] clustering to inactive
species and bulk Fe oxide. These oxides decomposed H2O2 and led to the competitive over-
oxidation processes illustrated in Scheme 1 (Paths B, C). Therefore the zeolite Si/Al ratio is
important for catalytic activity [17]. In fact, the presence of α-Fe2O3 leads to large increases in
H2O2 decomposition and COx selectivity [14, 39]. Decreases in H2O2 concentration have been
previously linked to decreases in methane oxidation[11]; this fact is especially pertinent to use
of ZSM-5(50) in this work which had a much higher iron content than the ZSM-5(30) reported
by Hammond et al [11]. In view of the potential diversity of active Fe sites relevant to the
selective oxidation of methane, Hammond et al proposed that “the number, geometry, valence
state, and nuclearity of the Fe sites at given Fe loadings merited an extensive study [14].
Hence in view of the above results it is possible that the ZSM-5(50) batches used in this work
may have contained active sites, which selectivity catalysed the decomposition of MeOOH to
HCHO. Such a reaction could have then run in parallel to the MeOOH decomposition pathway
depicted in Scheme 1, but perhaps to a larger extent.
There is absolutely no evidence to invalidate that the proposed formation of MeOOH occurs
according to Equations 4-8. It is however questionable whether the same [Fe] species
responsible for the decomposition of MeOOH to MeOH over ZSM-5(30),[11, 12] were
electronically similar to the ones responsible for the decomposition of HCHO for the ZSM-
5(50) system [16].
In an attempt to rationalise the production of HCHO in Scheme 2, we reformulated Equations
9 and 11, to give Equations 12 and 13 which are tentatively presented,
[*Fe2(µ2-OH)2(OH)2(CH3OOH)( H2O)]2+
[*Fe2(µ2-OH)2(OH)2(CH2O)(H2O)]2+ + H2O
Eq 12
Eq 13
[*Fe2(µ2-OH)2(OH)2(CH2O)( H2O)]2+ + H2O2
[*Fe2(µ2-OH)2(OH)2(HOOH)( H2O)]2+ + HCHO
In the case of Equation 12 the decomposition of Fe-MeOOH to give the internally produced
•OH radicals leads to the coordination complex *[Fe]-OCH2 which then gives HCHO. This
reaction differs from Equation 9 in which [Fe]-OMe is formed [11,12]. It is noteworthy that the
17

formation of HCHO and regeneration of catalyst by Equations 12 and 13 results in a net loss
of •OH in the overall catalytic system. Therefore, H2O2 is consumed, possibly invalidating the
reported first-order relationship between the catalytic rate and H2O2 concentration, applicable
to Scheme 1 [11,12]. As mentioned above, H2O2 is conserved in methane oxidations over
ZSM-5(30) (Equations 9-11,).
In Equations 11 and 12, the iron species are labelled as *[Fe] to indicate an analogous role
to the [Fe] species represented in Equations 9 and 11 above [12]. This is in order to account
for any electronic differences in the coordination sphere of the catalytic extra-framework [Fe]
centres in the zeolytic micropores.
The addition of copper to ZSM-5(50) promoted the decomposition of H2O2 by 40-100
times as shown in Table 5. This result would be of significance to the potential supply of ●OH
in the ZSM-5(50) system studied by us. In the case of the Hammond system depicted in
Scheme 1, Equations 4-11 show that the H2O2 is not directly involved with the rate limiting
activation of methane in (Equation 7) or in the decomposition of MeOOH to give MeOH
(Equations 9-11). On the other hand, this is not the case for the ZSM-5(50) system depicted
●
in Scheme 2. In this case Equations 12-13 show that the evolution of OH from H2O2
decomposition is a necessary requirement for the catalytic decomposition of the primary
product MeOOH which leads to HCHO, and not MeOH. This facts explains the observed
productivity and selectivity trends in Figures 4 a-b. In Figure 4a the marked decreases in
oxygenate productivities can be linked to H2O2 removal because H2O2 is needed to promote
the decomposition of *[Fe]-MeOOH and also any homogenous MeOOH. Therefore, the
addition of copper promoted the production of MeOOH by premature release from the catalyst
and and also inhibited MeOOH to its decomposition products: HCHO, MeOH and HCOOH.
Since the removal of •OH increased the selectivity of MeOOH and to a lesser extent those,
HCHO (Figure 4 b), MeOH could not have been produced by the sequential process given in
Scheme 1. In other words the role of copper to the ZSM-5(50) system does not switch the
selectivity of HCOOH to MeOH as proposed for the ZSM-5(30) Hammond system [11]. This
implies that the inhibition of the adventitious •OH radicals and MeOH over-oxidation in
18

Scheme 1 (Paths B and C) does not explain the decreased MeOH selectivity in the ZSM-
5(50) depicted in Figure 4 b. Conceivably the production of MeOH via HCHO might be related
to the set of uncharacterised “non sequential” reactions highlighted in Scheme 1.
Interestingly, the formation of HCHO is also catalysed by FeCl2 salts in H2O2 promoted
oxidations of methane [41]. However no leached Fe2+(aq) was detected in solution under StMO
conditions.
Clearly, it was necessary to determine the origins of MeOH in the ZSM-5(50) catalysed
oxidation of methane and how it might be formed from HCHO. As the addition of hydrogen to
an StMO reaction proved unsuccessful, the Cannizzzaro reaction was considered as it
involves the disproportionation of HCHO to give MeOH and HCOOH; catalysed either
homogeneously or heterogeneously by metal oxides [42, 43]. Although the Cannizzaro
reaction is normally run under alkaline conditions it can be run under acidic conditions to give
MeOH and HCOOH [44]. In the presence of Cu2O and alkali the Cannizzaro reaction also
produces hydrogen and metallic copper [45].
Alternatively, another pathway to the formation of MeOH from HCHO is presented in
Equations 14-17. These were formulated by Rieche for the decomposition of homogeneous
MeOOH [37]:
2CH3OOH
2HCHO + H2O
Eq 14
Eq 15
2CH3OOH + 2HCHO
2CH3OOCH2OH
2CH3OOCH2OH
2CH3O• +2HCOOH + 2H• Eq 16
2CH3O•
CH3OH + HCHO
Eq 17
The presence of CH3OOCH2OH in the ZSM-5(50) system studied by us is speculative as it
was not detected by NMR. In this respect it is noteworthy that peroxides are very vulnerable
to decomposition in the presence of [Fe] [46] which, taking into consideration the [Fe] content
in ZSM-5, is significant [11, 12]. In view of the above, it is possible that the formation of MeOH
in the ZSM-5(50) oxidation system may have been due to the formation and decomposition of
19

homogeneous or heterogeneous CH3OOCH2OH present as a short lived species. In either
case the formation of HCHO in Equation 15 is analogous to the MeOOH decomposition in
Equations 12-13. An advantage of invoking the prevalence of CH3OOCH2OH decomposition
is that it also accounts for all of the observed products in Table 1 above.
The addition of a copper salt had a drastic effect on HCOOH selectivity. In Scheme 1
the role of copper was to block path B[11] so as to prevent the formation of HCOOH so that
MeOH could be preserved. Conversely the results in Figure 4 a-b and Table S4 show that
MeOOH was preserved instead of MeOH. Moreover the addition of copper to ZSM-5(50) not
only decreased the HCOOH selectivity but also that of MeOH as well. Therefore, since the
addition of copper quenches •OH the deficit of •OH radicals explains the deficit of HCHO (via
MeOOH decomposition) and hence MeOH according to Equations 14-15. Therefore, it is
unlikely that the MeOOH decomposed according to Scheme 1 in which no HCHO was
detected. Presumably, the iron species responsible for the differences in the ZSM-5 systems
must have been partially or wholly different. The fact that different iron species may have been
operative in the ZSM-5(50) system studied by us is evident from the way copper reacted with
HCOOH. The decomposition of HCOOH can can occur in the two reaction routes summarised
by Beller in Equations 18–19: [47]
HCOOH = CO2 + H2;
HCOOH = CO + H2O;
mol-1 = -32.9 (dehydrogenation/decarboxylation)] Eq 18
mol-1 = -12.4 (dehydration/decarbonylation)]
Eq 19
From Table 6(Entry 1) it is clear Cu/ZSM-5(50) only removed HCOOH acid according
Equation 18, since although hydrogen was produced no CO was detected. Interestingly, the
conversion of HCOOH almost doubled when H2O2 was added Table 3(Entry 2). This suggests
that a terminal oxidant may have been needed to regenerate Cu from CuO at catalytic centres
in Cu/ZSM-5(50). This was suggested for another previously studied reaction, which was
evaluated for its potential towards the production of fuel cell hydrogen [36]. The catalysed
decomposition of HCOOH by Equation 18 using Cu/ZSM-5(50) maybe a convenient source
of fuel cell hydrogen but such considerations are beyond the scope of this present work .
20

The role of copper in the Cu /ZSM-5(30) mixtures was not characterised by Hammond et al
by physio chemical methods [11]. Likewise this was also the case with the mixtures of Cu
/ZSM-5(50) mixtures produced by us. However, the added copper must have interacted which
each ZSM-5 zeolite in a unique way, to give Fe species with electronically different catalytic
sites. Further work would be needed to determine which of these were responsible for the
production of hydrogen.
Conclusion
The copper deposition in the ZSM-5(50) micropores, failed to achieve the methanol
selectivities reported by Hammond et al [13]. Instead, the greatest selectivity was for MeOOH.
This was presumably because the initially high iron content in ZSM-5(50) contained a wider
range of catalytic iron clusters which interacted with the added copper; giving different
electronic enhancements in different [Fe2O2]2+ species. However the addition of copper to
ZSM-5(50), also promoted a substantial decomposition of hydrogen peroxide, and an overall
decrease in the productivity of all methane oxygenates. Formic acid was completely removed
from methane oxidation reactions by the addition of a copper salt alone. However in
combination, the copper salt and ZSM-5(50) gave hydrogen which was not reported in the
removal of HCOOH by the Cu/ZSM-5(30) in the Hammond system[11].
Overall, we propose that the high iron content in the ZSM-5(50) zeolite used in this work,
contained a total population, or partial population of catalytic [Fe] species which decomposed
MeOOH to HCHO. Following its production, the secondary product HCHO then reacted to
give methanol indirectly.
The dominance of the unadulterated [Fe] species responsible for the direct HCHO production
in ZSM-5(50) may, therefore, have masked the processes responsible for the direct production
of MeOH shown in Scheme 1. Thus in principle both processes might have been operative in
parallel. We propose that the ZSM-5(30) used by Hammond et al contained a greater
proportion of the active [Fe] species necessary for the direct production of MeOH, which
advantageously, had a lower activity towards the decomposition of H2O2.
21

Acknowledgements
The authors acknowledges the financial support of the King Abdulaziz City for Science and
Technology (KACST).
22

References :
1. V. Degirmenci, D. Uner, A. Yilmaz, Catalysis Today, Volume 106, Issues 1–4, 2005,
Pages 252-255
2. M. Ravi, M. Ranocchiari, J. A. van Bokhoven, Angew. Chem. Int. Ed. 2017, 56, 16464.
3. P. Tang, Q. Zhu, Zhaoxuan Wu and Ding Ma, Energy Environ. Sci, 2014, 7, 2580.
4. [4] C. Hammond, S. Conrad and I. Hermans, ChemSusChem, 2012, 5, 1668-1686.
5. N.F. Gol'dshleger, V.V. Es'kova, A.E. Shilov, A. Shteinman, A. Zh. Fiz. Khim. (Engl.
Trans.), 1972, 46, 785.
6. H. Lee a, H. J. Lee a, D. L. Sedlak and C. Lee a, Chemosphere, 2013, 92, 652-658.
7. D.A Kopp and S.J Lipard, 2002, Curr Opin Chem Biol, 6, 568-578.
8. R. Balasubramanian and A. Rosenzweig, 2007, Acc Chem Res, 40, 573-580.
9. R. A Himes and K. D Karlin, PNAS, 2009, 106, 18877-18878.
10. (a)P. Vanelderen, J. Vancauwenbergh, B. F. Sels, R. A. Schoonheydt, Coordination
Chemistry Reviews, 2013, 257, 483-494.(b) P. J. Smeets, M. H. Groothaert, R. A.
Schoonheydt, Catal. Today 2005, 110, 303–309.(c)
11. C. Hammond, R. L. Jenkins, N. Dimitratos, J. Antonio Lopez-Sanchez, M. Hasbi ab
Rahim, M. M. Forde, A. Thetford, D. M. Murphy, H. Hagen, E. E. Stangland, J. M.
Moulijn, S. H. Taylor, D. J. Willock, and G. J. Hutchings, Chem. Eur. J., 2012, 18,
15735-15745.
12. C. Hammond, M. M. Forde, M. Hasbi Ab Rahim, A. Thetford, Q. He, L. Jenkins, N.
Dimitratos, J. A. Lopez-Sanchez, N. F. Dummer, D. M. Murphy, A. F. Carley, S. H.
Taylor, D. J. Willock, E. E. Stangland, J. Kang, H. Hagen, C. J. Kiely, and G. J.
Hutchings, Angew. Chem. Int. Ed., 2012, 51, 5129 –5133.
13. C. Hammond, N. Dimitratos, J. A. Lopez-Sanchez, R. L. Jenkins, G. Whiting, S. A.
Kondrat, M. Hasbi ab Rahim, M. M. Forde, A. Thetford, M. M. Forde, A. Thetford, H.
Hagen, E. E. Stangland, J. M. Moulijn, S. H. Taylor, D. J. Willock and G. J. Hutchings,
Catal, 2013, 3, 1834-1844.
14. C. Hammond, N. Dimitratos, R. L. Jenkins, J. Antonio Lopez-Sanchez, S. A. Kondrat,
M. Hasbi Ab Rahim, Michael M. Forde, Adam Thetford, Stuart H. Taylor, Henk Hagen,
Eric E. Stangland, Joo H. Kang, Jacob M. Moulijn, David J. Willock, and Graham J.
Hutchings, Catal. 2013, 3, 689−699
15. J. Xu, Robert D. Armstrong, G. Shaw, N. F. Dummer, S. J. Freakley, S. H. Taylor, G.
J. Hutchings, Catalysis Today 2016, 270, 93–100.
16. S. Al-Shihri, C. J. Richard, and D. Chadwick, ChemCatChem, 2017, 9, 1276 – 1283.
17. C. Kalamaras, D. Palomas, R. Bos, A. Horton, M. Crimmin, K. Hellgardt, Catal. Lett.
2016, 146, 483–492.
23

18. M. H. Ab Rahim, M. M. Forde, R. L. Jenkins, C. Hammond, Q. He, N. Dimitratos, J. A.
Lopez-Sanchez, A. F. Carley, S. H. Taylor, D. J. Willock, D. M. Murphy, C.J. Kiely, G.J.
Hutchings, Angew. Chem. Int. Ed. 2013, 52, 1280–1284.
19. T-L. Hwang, and A.J. Shaka, J.Magn. Reson., Series A, 1995, 112, 275-279.
20. C. N. Satterfield, and A. H. Bonell, Analytical Chemistry, 1955, 27, 1174-1175.
21. C. A. Schalley, M. Dieterle, D. Schrrder, H. Schwarz, E. Uggerud, International Journal
of Mass Spectrometry and Ion Processes, 1997, 163, 101-119.
22. G. Süss-Fink, Galina V. Nizova, Sandrine Stanislas, Georgiy B. Shul’pin, Journal of
Molecular Catalysis A: Chemical, 1998, 130, 163-170; Wieland, H., and Wingler, A.,
Ann., 1923, 431, 301-322
23. A. D. Kirk, Canadian Journal of Chemistry, 1965, 43, 2236.
24. H. S. Fry, J. H. Payne, J. Am. Chem. Soc. 1931, 53, 1973.
25. H. S. Fry, J. J. Uber, J. W. Price, Recl. Trav. Chim. 1931, 50, 1060 – 1065.
26. L. J. Durham, C. F. Wurster, Jr., and H. S. Mosher, J. Am. Chem. Soc., 1958, 80, 332-
337.
27. J. Haber, and M. Witko, Journal of Catalysis, 2003, 216, 416-424.
28. [28] R. Burch, M.J. Hayes, Journal of Molecular Catalysis A: Chemical 1995, 100, 13-
33.
29. R.A Sheldon, and J.K Kochi, Adv. Catal., 1978, 25, 272-413.
30. J. K. Kim, I. S. Metcalfe, Chemosphere 2007, 69, 689 – 696.
31. Q. Zhu, Y. Lian, S. Thyagarajan, S. E. Rokita, K. D. Karlin, N. V. Blough, J. Am. Chem.
Soc. 2008, 130, 6304.
32. C. N. Satterfield, A. H. Bonnell, Anal. Chem. 1955, 27, 1174 – 1175.
33. G. L. Vaghjiani, and A. R. Ravishankara, J. Phys. Chem., 1989, 93, 1948-1959.
34. D. Martin Davies, and Michael E. Deary, J. Chem. Soc. Perkin Trans. 2, 1992, 559-
562.
35. R. S. Varma and R. Dahiya, Tetrahedron Letters, 1998, 39, 1307-1308.
36. D. Preti, S. Squarcialupi, and G. Fachinetti, Angew. Chem., 2009, 121, 4857-4860.
37. A, Rieche, "Alkylperoxyde und Ozonide,'' Theodor Steinkopff, Dresden und Leipzlg,
1931; (a) pp. 48-60; (b) pp. 16-21; (c) pp. 23-26.
38. G.L Woolery, G.H Kuehl, H.C Timken, A.W Chester, and J.C Vartuli., 1997, Zeolytes,
19, 288-296
39. M. Hermanek, R. Zboril, Ivo Medrik, J. Pechousek, and C. Gregor., J. Am. Chem. Soc,
2007, 129, 10929-10936.
24

40. G. Li, Peter Vassilev, M. Sanchez-Sanchez , J. A. Lercher , E. J.M. Hensen, E. A.
Pidko., 2016, Journal of Catalysis 338, 305–312.
41. Q. Yuang, W. Deng, Q. Zhang and Y. Wang, Adv. Synth. Catal., 2007, 349, 1199-
1209.
42. M. C. Kung, S. S.-Y. Lin, H. H. Kung, Top. Catal. 2012, 55, 108 – 115.
43. G. Busca, J. Lamotte, J.-C. Lavalley, V. Lorenzelli, J. Am. Chem. Soc. 1987, 109,
5197–5202.
44. H. Staudinger, R. Signer, H. Johner, O. Schweitzer and W. Kern., Ann, 1939, 474,
2545.
45. A. Harden., A. Proc. Chem. Soc. 1899, I5, 158-9.
46. E.G.E Hawkins., J. Chem. Soc. C, 1971, p1-22.
47. B. Loges, A. Boddien, F. Gärtner, H. Junge, M. Beller., 2010, Top Catal, 3, 902–914.
48. R. Shah, M. C. Payne, J. D. Gale, 1997, international Journal of Quantum Chemistry,
61, 393-398.
25

Figures
Figure 1 (1) Decomposition of peroxy-methanedioate to HCHO and H2.
Figure 21H NMR spectrum of methane oxidation products over ZSM-5(50) in D2O. δ/ppm; (HCOOH, 8.26),
((HCHO, 4.94(methane diol)), (MeOOH, 3.76), (MeOH, 3.25), (*,1.96, impurity), (Me, 0.07)
Figure 3 1H NMR spectrum of methane oxidation products over Cu-ZSM-5(50) in D2O.
δ/ppm; ((HCHO, 4.94(methane diol)), (MeOOH, 3.76), (MeOH, 3.25), (*,1.96, impurity), (Me, 0.07)
26

125.
100.
75.
50.
25.
0.
MeOH
MeOOH
HCOOH
HCHO
0
0.5
Cu, µmol
1
2
75.
60.
45.
30.
15.
0.
MeOH
MeOOH
HCOOH
HCHO
1
2
3
4
Cu, µmol
Figure 4(a)The effect of added Cu(NO3)2(aq) on the activity of ZSM-5(50);(b) The effect of added Cu(NO3)2(aq) on
oxygenate selectivities. StMO conditions: ZSM-5(50)+H2O2 at 50ᴼC, time = 30 min
27

Scheme
Scheme 1[11] Methane Oxidation over ZSM-5(30) in Aqueous H2O2
Scheme 2. Methane oxidation over ZSM-5(50).
28

Tables
Table 1 Product Analysis from methane oxidation over ZSM-5(50) [16].
Zeolyte
Products a
/µmol
Total liq.
Conv f
/%
TOFh
h-1
productivity e
/mol kgcat-1h-1
c
c
MeOHb
MeOOHb
HCOOHb
CH2Ob
CO2
H2
ZSM-5(50)-UF g
ZSM-5(50)-F g
19.0
14.9
44.9
53.2
98.2
2.3
1.8
<2
<2
10.7
10.9
25
0.87
0.91
5.3
5.6
102.6
26.1
Selectivityd
%
/
11.6
8.6
27.3
30.8
59.7
59.5
1.4
1.0
a
b
Standard reaction conditions: 10 ml, 30 min, 50C, PCH4 10 bar, [H2O2] 0.5 mol/l, catalyst 13.5mg, 1500 rpm.
Analyzed by 1H NMR.
No CO2 was detected by GC-TCD.
Product selectivity, calculated as moles
Total productivity, calculated as moles (products)/weight
c
d
e
(product)/moles (total liquid produced) x 100.
(catalyst)/time. f Conversion, calculated as (moles produced)/ (initial moles methane) x 100. h Turn over frequency
(TOF), calculated as moles(products)/moles(metal catalyst)/time
Table 2 The evolution of H2 with time a
Entry
Time/min
H2
1
2
3
5
2.2
5.1
10
30
10.7
Conditions: 10 ml, 50◦C, [H2O2] =0.5M; CO2 not measured, aZSM-5(50)=13 mg, PCH4 =10 bar.
Table 3 Oxidation of MeOH, HCHO and HCOOH with H2O2 over ZSM-5(50)
Entry
Reactantsa
Products/µmolb
µmol
Conve
TOFf
h-1
c
c
MeOH
MeOOH
HCOOH
11.3
30.2
-
CH2(OH)2
CO2
n/d
6
H2
%
23
75
19
1
2
3
Methanol
-
1.2
0
0
0
-
n/d
2.98
-
0.36
1.0
Formaldehyde
Formic acid
0
0
0
9
0.29
Selectiviy
d
%
4
5
6
Methanol
-
0
-
100
80.7
-
0
-
-
Formaldehyde
Formic acid
3
0
16
0
0
100
29