Niobium-substituted octahedral molecular sieve (OMS-2) materials in selective oxidation of methanol to dimethoxymethane

Article abstract of DOI:10.1039/c9ra04804a

Octahedral molecular sieve (OMS-2) refers to a one-dimensional 2 × 2 framework of octahedral manganese oxo units based on the cryptomelane-type framework. Herein, we describe a niobium (Nb) substituted mixed metal oxide of Nb and Mn where the cryptomelane-type framework is retained. These materials are hydrothermally synthesized from the reaction of potassium permanganate, manganese sulfate, and homogeneous niobium(v) precursors. Niobium incorporation up to 31 mol% can be achieved without destroying the one dimensional 2 × 2 framework. The yields of the materials vary between 70 and 90%. These materials are analyzed by powder XRD, BET isotherm, TEM, SEM, XRF, and XPS studies. The synthesized materials show promising activity in selective oxidation of methanol to dimethoxymethane (DMM) at 200 °C. Normalized activity correlations followed the trend 21% Nb-OMS-2 > 15% Nb-OMS-2 > 31% Nb-OMS-2 > 68% Nb-OMS-2 > K-OMS-2. A fluctuation in methanol conversion was observed around 125-150 °C in most samples, suggesting this to be a catalytically important temperature regime when forming active sites for DMM production.

Full text of DOI:10.1039/c9ra04804a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Niobium-substituted octahedral molecular sieve
(OMS-2) materials in selective oxidation of
methanol to dimethoxymethane†
Cite this: RSC Adv., 2019, 9, 32665
Niluka D. Wasalathanthri,a Curtis Guild,a Quddus A. Nizami,c Shanka L. Dissanayake,a
Junkai He,b Peter Kerns,a Jared Fee,a Laura Achola,a Dinithi Rathnayake,a
ab
c
Chandima Weerakkody,a Steven L. Suib
*
*
and Partha Nandi
Octahedral molecular sieve (OMS-2) refers to a one-dimensional 2 ꢀ 2 framework of octahedral
manganese oxo units based on the cryptomelane-type framework. Herein, we describe a niobium (Nb)
substituted mixed metal oxide of Nb and Mn where the cryptomelane-type framework is retained. These
materials are hydrothermally synthesized from the reaction of potassium permanganate, manganese
sulfate, and homogeneous niobium(V) precursors. Niobium incorporation up to 31 mol% can be achieved
without destroying the one dimensional 2 ꢀ 2 framework. The yields of the materials vary between 70
and 90%. These materials are analyzed by powder XRD, BET isotherm, TEM, SEM, XRF, and XPS studies.
The synthesized materials show promising activity in selective oxidation of methanol to
dimethoxymethane (DMM) at 200 ꢁC. Normalized activity correlations followed the trend 21% Nb-OMS-
2 > 15% Nb-OMS-2 > 31% Nb-OMS-2 > 68% Nb-OMS-2 > K-OMS-2. A fluctuation in methanol
conversion was observed around 125–150 ꢁC in most samples, suggesting this to be a catalytically
important temperature regime when forming active sites for DMM production.
Received 25th June 2019
Accepted 28th September 2019
DOI: 10.1039/c9ra04804a
rsc.li/rsc-advances
templates.10,11 We studied whether it is possible to substitute
Mn with relatively larger in size Nb in the octahedral molecular
Introduction
sieve framework, and if so to what extent in order to maintain
the one dimensional 2 ꢀ 2 framework. In the present manu-
script, we report Nb substituted OMS-2 materials where the
extent of Mn substitutions by Nb can go up to 31%. Beyond this
high level of substitution, these materials do not retain the
cryptomelane structure and an amorphous phase of Nb2O5
starts to appear.
Dimethoxymethane (DMM; CH3OCH2OCH3) is an important
chemical used in many industries as a building block in organic
synthesis, a gasoline additive, and as precursor for concentrated
formaldehyde (HCHO) streams.12–15 Furthermore, oxymethylene
dimethyl ethers (OME) such as DMM are demonstrated as
potential candidates for synthetic fuels in the recent times.16
These compounds can be used as substitutes to diesel or mixed
into it. DMM is benign to the environment due to its high
oxygen content and the absence of C–C bonds.17 As a result, the
issue of diesel exhaust which has adverse effects on human
health and ecosystem can be addressed.18,19
Octahedral molecular sieve (K-OMS-2) based on the cryptome-
lane framework1–5 has emerged as an important class of redox
active heterogeneous catalysts for a multitude of catalytic
partial oxidations.6–9 Prior work has shown that a fraction of Mn
in these frameworks can be substituted with other metals and
the degree of substitutions can vary quite signicantly.1–5 Metal
substitution in the K-OMS-2 framework impacts physical and
chemical properties such as: aspect ratio of the rod (L/D), degree
of aggregation, thermal stability, redox properties (e.g., ratio of
Mn3+/Mn4+), conductivity, and catalytic activity. In most cases
the extent of Mn substitution is less than 5 wt%.1 However, in
the case of V and Co the extent of substitution can be more than
10% where the OMS-2 structure is still retained. Unlike with V,
the mixed metal oxides with Nb and Mn are known to exist in
molecular crystals as [MnNb9O28]
8ꢂ or [H2MnNb10O32]8ꢂ made
from homogeneous precursors involving syntheses with organic
aDepartment of Chemistry, University of Connecticut, Storrs, Connecticut 06269, USA.
E-mail: steven.suib@uconn.edu
bInstitute of Material Science, University of Connecticut, Storrs, Connecticut 06269,
DMM synthesis can be achieved by partial oxidation of
methanol to formaldehyde followed by nucleophilic addition of
methanol to formaldehyde (twice).12 The overall balanced
chemical equation for the reaction is given in eqn (E 1).12,20
USA
cCorporate Strategic Research, ExxonMobil Research and Engineering Company,
Annandale, New Jersey 08801, USA. E-mail: partha.nandi@exxonmobil.com
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c9ra04804a
6CH3OHðlÞ þ O2ðgÞ/2CH3OCH2OCH3ðlÞ þ 4H2OðlÞ;
(E 1)
DHfꢁð298
¼ ꢂ384:4 kJ molꢂ1
KÞ
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 32665–32673 | 32665

View Article Online
RSC Advances
Paper
According to the literature, methanol oxidation may produce expenditure.14 The redox catalyzed pathways lead to a sequence
various products [formaldehyde (CH2O), dimethyl ether of oxidized species whereas the acid-catalyzed pathways yield
(CH3OCH3), methyl formate (HCOOCH3), dimethoxymethane dehydration products. In this respect, the motivation for the
(CH3OCH2OCH3), and carbon oxides (CO & CO2)] depending on work reported here is to develop a catalyst which oxidizes
various parameters such as catalyst, reaction temperature, methanol to DMM with high selectivity. An appropriate system
reactant partial pressure, residence time, and methanol to would be a bifunctional catalyst which contains adequate
oxygen ratio.20 The following reactions show different possible amounts of both acid and redox sites.13,14 Thus, incorporating
pathways for methanol oxidation.
redox innocent but acidic Nb into redox active K-OMS-2 would
create a bifunctional mixed metal oxide catalytic systems for
aerobically oxidizing methanol into DMM at lower temperature
for potential energy and CO2 savings.
2CH3OH(g) / CH3OCH3(g) + H2O(g) (dehydration) (E 2)
1
2
CH3OH(g) + O2(g) / CH2O(g) + H2O(g) (oxidation) (E 3)
Experimental
Materials synthesis
CH2O(g) + 2CH3OH(g) /
CH3OCH2OCH3(g) + H2O(g) (dehydration) (E 4)
Nb incorporation in OMS-2 framework was achieved by reacting
soluble Nb precursors such as Nb(OEt)5 and Nb(C2O4H)5$XH2O
with KMnO4 and MnSO4 in acidic medium. Other Nb sources such
as niobic acid, niobium oxides, and LiNbO2 did not incorporate Nb
in the OMS-2 framework in our experience. The syntheses of
various Nb incorporated OMS-2 materials is described below:
15 mol% Nb. In a 300 mL round bottom ask, a solution of
MnSO4$H2O (Sigma Aldrich 99+%), was prepared by mixing
solid manganese sulfate (25.35 g, 150 mmol); deionized water,
90 mL; and HNO3 (J.T. Baker 69–70%), 9 mL. In a beaker,
a solution of KMnO4 (Sigma Aldrich 99+%), was prepared by
mixing solid potassium permanganate (17.54 g, 111 mmol) and
deionized water, 300 mL. Potassium permanganate solution
was added dropwise to manganese sulfate solution while stir-
ring the solution. Nb(OEt)5 (Sigma Aldrich 99.95%) (3.85 g, 12
mmol) was added to the solution. The mass of metal dopant was
varied depending on the desired ratios of niobium to manga-
nese. The resulting solution aer being mixed was reuxed at
CH2O(g) + 12O2(g) / HCOOH*(g) + CH3OH(g) /
HCOOCH3(g) + H2O(g) (oxidation + dehydration) (E 5)
1
2
CH2O(g) + O2(g) / HCOOH*(g) /
CO(g) + H2O(g) (oxidation + dehydration) (E 6)
1
2
HCOOH*(g) + O2(g) / CO2(g) + H2O(g) (oxidation) (E 7)
The partial oxidation of methanol to formaldehyde is a large-
scale industrial process carried out over silver or copper cata-
lysts.21 Due to its economic value, this reaction has been studied
extensively in order to improve the efficiency of the process and
to understand the reaction mechanism. Mixed and doped metal
oxides have shown better activity towards the methanol partial
oxidation reaction due to the variation in acidity and redox
activity of the material.22 Acidic sites of the catalyst could drive
the dehydration reaction to completely yield dimethyl ether. On
the other hand, increase in the basic character of the catalyst
will give formate species, which further decompose to carbon
dioxide. For example, metal oxides with Lewis acidic sites,
niobia, and alumina have been used to produce dimethyl ether
from methanol. Metal oxides with redox sites such as vanadia,
tin oxide, and molybdena have been studied as catalysts for
methanol partial oxidation to produce formaldehyde and
methyl formate.
Notably, vanadium-based systems show exceptional activity
for the selective oxidation of methanol with most of the cata-
lysts being supported catalysts and for some nanostructuring
was key to the reactivity.23,24 The doping of SO42ꢂ and PO43ꢂ ions
further improved the conversion and selectivity of the catalyst
for the reaction. It has been suggested that acidic sites favor the
condensation of formaldehyde to DMM, thus the addition of
acidic additives has been vastly employed to improve the DMM
yield.25,26 Others have co-doping vanadium with sulfur or
phosphorus and still observe high activity.12,27 Gornay et al. re-
ported a mixed metal oxide catalyst, FeMo (MoO3–Fe2(MoO4)3),
with a high yield of DMM under high methanol partial pres-
sures.28 Industrial scale production of DMM is currently carried
out through a two-step process. First, oxidation of gas phase
methanol to formaldehyde followed by a liquid phase acetyla-
tion step. However, synthesizing DMM via a direct single-step
process is more economical due to minimized capital
ꢁ
100–110 C overnight. Upon cooling, the solution was ltered
and washed with deionized water, and then dried to 120 ꢁC
overnight. Aer drying, the yield of the material was 12.73 g
using between 25.35 and 17.54 g of manganese reactant.
21 mol% Nb. In a 300 mL round bottom ask, a solution of
MnSO4$H2O (Sigma Aldrich 99+%), was prepared by mixing
solid manganese sulfate (25.35 g, 150 mmol); deionized water,
90 mL; and HNO3 (J.T. Baker 69–70%), 9 mL. In a beaker,
a solution of KMnO4 (Sigma Aldrich 99+%), was prepared by
mixing solid potassium permanganate (17.54 g, 111 mmol) and
deionized water, 300 mL. Potassium permanganate solution
was added dropwise to manganese sulfate solution while stir-
ring the solution. Nb(OEt)5 (Sigma Aldrich 99.95%) (3.85 g, 12
mmol) was added to the solution. The mass of metal dopant was
varied depending on the desired ratios of niobium to manga-
nese. The resulting solution aer being mixed was reuxed at
ꢁ
100–110 C overnight. Upon cooling, the solution was ltered
and washed with deionized water, and then dried at 120 ꢁC
overnight. Aer drying, the yield of the material was 14.74 g
using between 25.35 and 17.54 g of manganese reactant.
31 mol% Nb. In a 300 mL round bottom ask, a solution of
MnSO4$H2O (Sigma Aldrich 99+ %), was prepared by mixing
solid manganese sulfate (25.35 g, 150 mmol); deionized water,
90 mL; and HNO3 (J.T. Baker 69–70%), 9 mL. In a beaker,
32666 | RSC Adv., 2019, 9, 32665–32673
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
a solution of KMnO4 (Sigma Aldrich 99+%), was prepared by a Rigaku ZSX Primus IV wavelength-dispersive XRF spectrom-
mixing solid potassium permanganate (17.54 g, 111 mmol) and eter with a tube above geometry and an ultra-thin end window
deionized water, 300 mL. Potassium permanganate solution having an X-ray tube with a 4 kW Rh anode.
was added dropwise to manganese sulfate solution while stir-
Table 1 above shows the distribution of Mn, Nb and K by
ring the solution. Nb(OEt)5 (Sigma Aldrich 99.95%) (11.45 g, 36 weight. Mn/K ratio gradually decreases with increasing Nb
mmol) was added to the solution. The mass of metal dopant was loading conrming the removal of potassium ions from the
varied depending on the desired ratios of niobium to manga- tunnel structure.29,30 For all the materials up to 31% Nb loading,
nese. The resulting solution aer being mixed was reuxed at the Nb/K ratio has increased further conrming the partial
ꢁ
100–110 C overnight. Upon cooling, the solution was ltered substitution of potassium ions. However, upon 68% loading of
and washed with deionized water, and then dried to 120 ꢁC Nb, mass percentages of both Mn and K has drastically dropped
overnight. Aer drying, the yield of the material was 10.26 g compared to the materials with lower Nb loading. The results
using between 25.35 and 17.54 g of manganese reactant. XRF further proves that aer 31% loading of Nb, the cryptomelane
analysis conrmed the loading of Nb.
structure has collapsed and a material with dominating amount
68 mol% Nb. In a 300 mL round bottom ask, a solution of of Nb has formed.
MnSO4$H2O (Sigma Aldrich 99+%), was prepared by mixing
Powder XRD analysis. Powder X-ray diffraction (PXRD) was
solid manganese sulfate (17.6 g, 104 mmol); deionized water, 90 used to determine bulk sample identication and phase purity
mL; and HNO3 (J.T. Baker 69–70%), 6 mL. In a beaker, a solu- of the product, and to determine the various loadings of
tion of KMnO4 (Sigma Aldrich 99+%), was prepared by mixing niobium. The PXRD studies were performed with a Rigaku
solid potassium permanganate (9.5 g, 60 mmol) and deionized Ultima IV diffractometer using Cu Ka (l ¼ 0.15406 nm) radia-
water, 200 mL. Potassium permanganate solution was added tion. A beam voltage of 40 kV and a beam current of 44 mA were
dropwise to manganese sulfate solution while stirring the used. The wide-angle PXRD patterns were collected at 0.02ꢁ
solution. Nb(OEt)5 (Sigma Aldrich 99.95%) (15.26 g, 48 mmol) steps over a 2q range of 10ꢂ70ꢁ with a continuous scan rate of
was added to the solution. The mass of metal dopant was varied 2.0ꢁ minꢂ1. The crystallographic phases were identied by
depending on the desired ratios of niobium to manganese. The comparing the wide-angle PXRD pattern with the Joint
resulting solution aer being mixed was reuxed at 100–110 ꢁC Committee on Powder Diffraction Society (JCPDS) database.
overnight. Upon cooling, the solution was ltered and washed The cryptomelane structure was retained upon 15% Nb incor-
with deionized water, and then dried to 120 ꢁC overnight. Aer poration (Fig. 1). Peak broadening was observed with increasing
drying, the yield of the material was 10.22 g using between 17.6 Nb loading indicating smaller crystallite-size, consistent with
and 9.5 g of manganese reactant.
the TEM data. For materials, higher than 31% Nb content, an
amorphous powder diffraction pattern was obtained. Since the
reactions were carried under a range of temperature, it is
important to study the structural changes of the materials with
temperature. Thus, in situ PXRD data were collected using an in
situ PANalytical X-ray diffractometer under nitrogen and oxygen
Catalytic activity
Methanol partial oxidation reactions were carried out in a xed-
bed quartz reactor (i.d. 6 mm) containing 0.100 g of catalyst
supported by quartz wool, operating at atmospheric pressure. A
mixture of gas containing oxygen and helium were bubbled
through methanol at room temperature (RT). The outlet gas was
monitored by an on-line GC equipped with two columns (Hay-
esep and 13X siloxane) connected in parallel, and two detectors
(thermal conductivity detector (TCD) and Helium ionization
detector (HID)). The temperature at which maximum conver-
sion of desired partial oxidation product (dimethoxy methane,
DMM), was rst investigated by feeding the system with 5% O2
balanced with He through a methanol bubbler at room
temperature (RT). The ow rate was 50 standard cubic centi-
meters (sccm). Final methanol/oxygen/helium composition of
the feed gas was ꢁapproximately 13/4/83 mol%. Temperature was
˚
ow equipped with Cu Ka (l ¼ 1.5405 A) radiation within
a range of 5ꢁ # 2q # 70ꢁ (step size: 0.02ꢁ, counting time: 1 s per
step, 14 angle incident slit, 45 kV, and a current of 40 mA). 15%
Nb-OMS-2 and 21% Nb-OMS-2 materials were able to withstand
up to 400 ꢁC under the oxidizing environment. At temperatures
above 400 ꢁC signicant particle growth and sintering was
observed for both the materials (see ESI Fig. S1†). The observed
peak sharpening was consistent with BET data in Fig. 2. For
31% Nb-OMS-2 and 68% Nb-OMS-2 materials the extents of
particle growth and agglomeration was greater compared to the
materials with lower Nb content. Amorphous powder diffrac-
tion pattern was obtained for both 31% Nb-OMS-2 and 68% Nb-
ꢁ
increased by 25 C increments from (100–250) C. Three injec-
tions were done at each temperature and the average activity is
reported.
Table 1 XRF composition of elements (by weight)
Material description
Mn (wt%)
Nb (wt%)
K (wt%)
Results and discussion
Characterization
15% Nb-OMS-2
21% Nb-OMS-2
31% Nb-OMS-2
68% Nb-OMS-2
75
71
48
16
16
19
43
80
9
10
9
Elemental analyses (wt%). The elemental composition of the
materials was performed by X-ray Fluorescence technique. The
elemental composition of the materials was estimated using
4
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 32665–32673 | 32667

View Article Online
RSC Advances
Paper
Fig. 1 Powder XRD stack plot of Nb-OMS-2.
OMS-2 and produced Nb2O5 crystallites upon heating above OMS-2 to 68% Nb-OMS-2 samples. As far as the pore volumes
ꢁ
500 C (see ESI Fig. S1†).
from QSDFT method on the adsorption branch is considered, it
BET surface area. Nitrogen (N2) physisorption analysis was decreases from 0.536 cm3 gꢂ1 to 0.182 cm3 gꢂ1 for 15% Nb-OMS-
performed in a Quantachrome Autosorb iQ instrument to study 2 to 31% Nb-OMS-2. Furthermore, pore volume increases in
surface area of undoped K-OMS-2 and Nb doped manganese 68% Nb-OMS-2 to 0.285 cm3 gꢂ1. The pore diameter decreases
oxide materials. All the samples were degassed at 150 ꢁC for 6 h from 29.0 nm to 1.0 nm for 15% Nb-OMS-2 to 68% Nb-OMS-2
under vacuum prior to measurement. The nitrogen adsorption samples when QSDFT method is utilized. Therefore, both
experiments were performed at ꢂ196 ꢁC. Twenty-six-point methods indicate the same trend in the change of pore volumes
adsorption and twenty-point desorption were performed with and pore diameters of materials. The material with 21 mol% Nb
P/P0 ranging from 0.002 to 0.995. Points from 0.03 to 0.25 were loading which shows the highest catalytic activity has a mono-
selected for the BET method giving a R value > 0.999 for all modal pore size distribution around 20 nm. 15 mol% Nb
samples for surface area calculation. Surface areas of four sample primarily contains macropores and the pore volume is
different Nb loadings 15 mol%, 21 mol%, 31 mol%, and very high. However, for the samples with 31 mol% and 68 mol%
68 mol% were compared to undoped K-OMS-2. Surface areas of Nb loading, the pore size ranges in a wide range and this may
all the materials were decreased upon increasing calcination due to the collapsed cryptomelane structure as proven from the
temperature. Allꢁof these materials turn into very dense phases XRD data.
going from 400 C to 600 ꢁC. Data in Table S1† and in the cor-
TEM, STEM and HAADF imaging. Fig. 3 below shows TEM
responding Fig. 2 below show that increased Nb loadings from images of Nb-doped OMS-2 materials. The scanning trans-
0 to 15 mol% signicantly reduced particle growth and sinter- mission electron microscopy (STEM) and transmission electron
ing from the as made state to materials calcined at 350 ꢁC. microscopy (TEM) images with different Nb concentrations are
While all K-OMS-2 with no Nb showed ꢃ70% surface area loss shown above. The TEM images of samples 15% Nb-OMS-2, 21%
in this heat treatment and retention of the OMS-2 framework Nb-OMS-2, and 31% Nb-OMS-2 show typical OMS-2 nanorod
structure. The corresponding surface area loss for the 15% Nb-
OMS-2 material was only around 20%. Although the extent of
sintering and particle growth slightly increased for 21% Nb-
OMS-2, the 31% Nb-OMS-2 material and the amorphous mate-
rial derived from 68 mol% Nb showed similar trends. In
materials that had higher than 31% Nb content the OMS-2
framework was no longer retained and hence interpreting the
particle growth and sintering mechanism in them became
challenging.
Both Quench Solid Density Functional Theory (QSDFT) and
Barrett–Joyner–Halenda (BJH) methods were employed to
calculate the pore volumes of the materials (Table S2†). When
BJH pore volumes from desorption branch is considered
(Fig. S2†), it decreases from 0.674 cm3 gꢂ1 to 0.216 cm3 gꢂ1 for
15% Nb-OMS-2 to 31% Nb-OMS-2. Furthermore, pore volume
increases in 68% Nb-OMS-2 to 0.344 cm3 gꢂ1. The pore diameter
decreases from 17.6 nm to 1.6 nm when considering 15% Nb-
Fig. 2 Particle growth and sintering of K-OMS-2 and Nb-OMS-2
materials.
32668 | RSC Adv., 2019, 9, 32665–32673
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
Fig. 3 (i) TEM images of (a) 15%-Nb-OMS-2 (b) 21% Nb-OMS-2 (c) 31% Nb-OMS-2 are shown in 20 nm resolution, lower resolution (100 nm)
images of corresponding materials are shown in a1, b1, c1 respectively, (ii) a2, b2 and c2 shows selective area diffraction and corresponding
d spacings, (iii) HAADF images of (a), (b), (c) are shown by a3, b3 and c3 respectively, (iv) elemental mapping for K, Mn and Nb are shown by (a4, a5,
a6), (b4, b5, b6) and (c4, c5, c6) respectively.
morphologies, with average lengths of 121.2 nm, 133.5 nm, and crystallinity similar to amorphous Mn oxide. The crystal clusters
51.3 nm, and average diameters of 14.2 nm, 11.2 nm, and were observed for high resolution TEM, with a d spacing value
5.4 nm and average aspect ratios of 11.2, 9.7, and 9.6, respec- of 0.31 nm, attributed to (310) planes of OMS-2. The STEM
tively (statistical measurement results for TEM for randomly mappings show uniform distributions of the elements K, Mn,
selected 60 nanorods). Meanwhile, there is detrital Nb species and Nb.
seen in the 31% Nb-OMS-2 material. The increasing Nb loading
Elemental mapping of K, Mn and Nb conrmed that while K
in OMS-2 leads to smaller crystallite size and the rod like and Mn track the rod morphology, a small amount of Nb clus-
morphology of the OMS-2 materials gradually decrease in terms tering is seen for 31%-Nb-OMS-2. In other materials the Nb
of aspect ratio. The crystal lattice spacing (d) of ꢃ0.5 nm was distribution appeared to be more uniform.
observed for the 21 mol% sample but the spacing for the
X-ray photoelectron spectroscopy (XPS) analysis. X-ray
31 mol% sample appeared to be slightly increased to ꢃ0.7 nm. photoelectron spectra (XPS) characterization of the synthe-
Aer increasing the Nb concentration to 73 mol%, the nanorods sized materials were done on a PHI model Quantum 2000
completely disappeared. Instead, randomly distributed amor- spectrometer with scanning ESCA multiprobe (F Physical
phous mixed Mn, Nb oxides were seen. The selected area elec- Electronics Industries Inc.), using Al Ka radiation (l ¼ 1486.6
tron diffraction (SAED) patterns of both materials show broad eV) as the radiation source. The spectra were recorded in the
diffraction rings, which means the materials have low xed analyzer transmission mode with pass energies of
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 32665–32673 | 32669

View Article Online
RSC Advances
Paper
Fig. 4 XPS Spectra of Nb-OMS-2 materials: (a and b) 15% Nb-OMS-2, (c and d) 21% Nb-OMS-2, (e and f) 31% Nb-OMS-2, (g and h) 68% Nb-
OMS-2.
187.85 eV and 29.35 eV for recording survey and high resolution analysis chamber. Binding energies (BE) were measured for Nb
spectra, respectively. The powder samples were pressed on 3d, Mn LMM, Mn 3s, Mn 2p and O 1s regions. The XPS spectra
a double-sided carbon tape mounted on an Al coupon pinned to obtained were analyzed and tted using CasaXPS soware
a sample stage with a wash and screw then placed in the (version 2.3.16). Sample charging effects were eliminated by
32670 | RSC Adv., 2019, 9, 32665–32673
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
Fig. 5 Activity profiles of Nb incorporated OMS-2: (a) methanol conversion vs. reaction temperature, (b) DMM yield vs. reaction temperature, (c)
DMM selectivity vs. reaction temperature.
ꢁ
correcting the observed spectra with the C 1 s BE value of trend of increasing activity until 125 C and then maintaining
284.8 eV.
the activity around 20–25% of methanol conversion. Average
XPS spectra of the Nb loaded sample are given in Fig. 4. The DMM yield from methanol conversion over the studied samples
binding energy of Mn 2p3/2 at 641.9 eV of the 15% Nb sample ranged from 20–25% except for 21% Nb-OMS-2 sample in which
can be ascribed to oxidation states 2+, 3+, and 4+ of manga- the DMM yield was above 40% and reached maximum yield at
nese.31 This suggests the presence of mixed valent manganese 65% at 250 ꢁC. Selectivity for DMM was typically around 90–95%
species in the material. Both 15% and 21% samples show the for all the Nb incorporated samples. Total oxidation products
binding energy of Nb 3d5/2 at 207.9 eV can be ascribed to the 5+ (CO2, H2O) were detected at the high end of the temperature
oxidation state of Nb.32 However, with higher Nb loading the range (>200 C) with methanol conversion typically decreasing
ꢁ
binding energy of Nb 3d5/2 has shied up to 209.1 eV. The XPS in the range, due to either kinetic or physical considerations
spectra for samples are summarized in Table S3.† Mn 2p3/2 (e.g. coking, quenching of active sites). Normalized activity
binding energies of manganese of all the samples could be correlations followed the trend 21% Nb-OMS-2 > 15% Nb-OMS-
ascribed to 2+, 3+, and 4+ states of manganese and the 5+ state 2 > 31% Nb-OMS-2 > 68% Nb-OMS-2 > K-OMS-2. Interestingly,
of niobium. Very similar XPS data were obtained from other Nb- a uctuation in methanol conversion was observed around 125–
OMS-2 materials as shown in Fig. 4.
150 ꢁC in most samples, suggesting this to be a catalytically
important temperature regime when forming active sites for
DMM production (Fig. 5).
Catalytic activity
Methanol conversion under given conditions was around 1%
for undoped K-OMS-2 for the entire temperature range from
room temperature (RT) up to 250 C. All the Nb incorporated
Conclusion
ꢁ
samples show in any case 10% methanol conversion. 15% Nb- In summary, Lewis acidic and redox innocent Nb(V) can be
OMS-2 sample reached the highest conversion under the incorporated into the K-OMS-2 framework by substituting Mn
given conditions of 30% at 150 ꢁC. 21% Nb-OMS-2 sample atoms. TEM, STEM, and HAADF imaging of Nb doped OMS-2
shows 12% conversion until 200 ꢁC and then the activity drops. materials showed retention of rod like morphologies with
Both 31% Nb-OMS-2 and 68% Nb-OMS-2 samples show similar subtle changes in d spacings and aspect ratios. BET data
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 32665–32673 | 32671

View Article Online
RSC Advances
Paper
suggested that upon substitution with Nb, the OMS-2 rods
became less susceptible to particle growth and sintering upon
heat treatment compared to all Mn K-OMS-2 materials. The
cryptomelane rod like morphology of K-OMS-2 is retained even
up to 31% substitution of Mn with Nb. At a higher loading of Nb
an amorphous solid solution is obtained. Including single
crystalline niobium rich Mn oxide materials and all Mn K-OMS-
2, this work maps out the structural subtleties of mixed metal
oxides of Mn and Nb. Normalized activity correlations followed
the trend 21% Nb-OMS-2 > 15% Nb-OMS-2 > 31% Nb-OMS-2 >
68% Nb-OMS-2 > K-OMS-2. Thus having large pores preferably
the same size promote the activity of the catalysts. The surface
analysis data along with the XRD data thus conrm that not
only the incorporation of Nb but also retaining the cryptome-
lane structure with larger pore volume is promising to achieve
a better performance in this study. Interestingly, a uctuation in
methanol conversion was observed around 125–150 ꢁC in most
Distortion of Molybdenum-Doped Manganese Oxide
Octahedral Molecular Sieves for Enhanced Catalytic
Performance, Inorg. Chem., 2015, 54(21), 10163–10171.
6 Y. Wang, H. Kobayashi, K. Yamaguchi and N. Mizuno,
Manganese Oxide-Catalyzed Transformation of Primary
Amines to Primary Amides through the Sequence of
Oxidative Dehydrogenation and Successive Hydration,
Chem. Commun., 2012, 48(20), 2642.
7 K. Yamaguchi, H. Kobayashi, Y. Wang, T. Oishi,
Y. Ogasawara and N. Mizuno, Green Oxidative Synthesis of
Primary Amides from Primary Alcohols or Aldehydes
Catalyzed by a Cryptomelane-Type Manganese Oxide-Based
Octahedral Molecular Sieve, OMS-2, Catal. Sci. Technol.,
2013, 3(2), 318–327.
8 K. Yamaguchi, H. Kobayashi, T. Oishi and N. Mizuno,
Heterogeneously Catalyzed Synthesis of Primary Amides
Directly from Primary Alcohols and Aqueous Ammonia,
Angew. Chem., Int. Ed., 2012, 51(2), 544–547.
samples, suggesting this to be
a catalytically important
temperature regime when forming active sites for DMM
production.
9 K. Yamaguchi, Y. Wang and N. Mizuno, Manganese Oxide-
Catalyzed Additive- and Solvent-Free Aerobic Oxidative
Synthesis of Primary Amides from Primary Amines, Chem.
Lett., 2012, 41(6), 633–635.
Conflicts of interest
10 J.-H. Son, J. Wang and W. H. Casey, Structure, Stability and
There are no conicts to declare.
Photocatalytic H Production by Cr-, Mn-, Fe-, Co-, and Ni-
2
Substituted Decaniobate Clusters, Dalton Trans., 2014,
43(48), 17928–17933.
Acknowledgements
11 J.-H. Son and W. H. Casey,
A Decatungstate-Type
This work was funded by ExxonMobil Research and Engineering
Company. We acknowledge the US Department of Energy,
Office of Basic Energy Sciences, Division of Chemical, Biological
Polyoxoniobate
with
Centered
Manganese:
[H2MnIVNb10O32]8ꢂ as a Soluble Tetramethylammonium
Salt, Dalton Trans., 2013, 42(37), 13339.
and Geological Sciences under grant DE-FG02-86ER13622.A000 12 H. Liu and E. Iglesia, Selective One-Step Synthesis of
for partial support of this research.
Dimethoxymethane via Methanol or Dimethyl Ether
Oxidation on H3+nVnMo12ꢂnPO40 Keggin Structures, J. Phys.
Chem. B, 2003, 107(510), 10840–10847.
References
13 H. Ma, H. Wang, B. Lu, J. Zhao and Q. Cai, VOx Molecular
Level Graed g-C3N4 for Highly Selective Oxidation of
Methanol to Dimethoxymethane, Mol. Catal., 2019, 469,
48–56.
14 K. Thavornprasert, M. Capron, L. Jalowiecki-duhamel,
O. Gardoll, M. Trentesaux, A. Mamede and G. Fang, Highly
Productive Iron Molybdate Mixed Oxides and Their
Relevant Catalytic Properties for Direct Synthesis of 1,1-
Dimethoxymethane from Methanol, Appl. Catal., B, 2014,
145, 126–135.
1 X. Chen, Y. F. Shen, S. L. Suib and C. L. O. Yo,
Characterization of Manganese Oxide Octahedral
Molecular Sieve (M-OMS-2) Materials with Different Metal
Cation Dopants, Chem. Mater., 2002, 14, 940–948.
2 X. Shen, A. M. Morey, J. Liu, Y. Ding, J. Cai, J. Durand,
Q. Wang, W. Wen, W. A. Hines, J. C. Hanson, et al.,
Characterization of the Fe-Doped Mixed-Valent Tunnel
Structure Manganese Oxide K-OMS-2, J. Phys. Chem. C,
2011, 115(44), 21610–21619.
3 C. Calvert, R. Joesten, K. Ngala, J. Villegas, A. Morey, X. Shen 15 Y. Meng, T. Wang, S. Chen, Y. Zhao, X. Ma and J. Gong,
and S. L. Suib, Synthesis, Characterization, and Rietveld
Renement of Tungsten-Framework-Doped Porous
Manganese Oxide (K-OMS-2) Material, Chem. Mater., 2008,
20(20), 6382–6388.
Selective Oxidation of Methanol to Dimethoxymethane on
V2O5 – MoO3/g -Al2O3 Catalysts, Appl. Catal., B, 2014, 160–
161, 161–172.
16 R. Sun, I. Delidovich and R. Palkovits, Dimethoxymethane as
a Cleaner Synthetic Fuel: Synthetic Methods, Catalysts, and
Reaction Mechanism, ACS Catal., 2019, 9(2), 1298–1318.
4 C. K. King'Ondu, N. Opembe, C. H. Chen, K. Ngala,
´
H. Huang, A. Iyer, H. F. Garces and S. L. Suib, Manganese
Oxide Octahedral Molecular Sieves (OMS-2) Multiple 17 N. Ribeiro, A. C. Pinto, C. M. Quintella, G. O. da Rocha,
Framework Substitutions: A New Route to OMS-2 Particle
Size and Morphology Control, Adv. Funct. Mater., 2011,
21(2), 312–323.
5 C. H. Chen, E. C. Njagi, S. Y. Chen, D. T. Horvath, L. Xu,
A. Morey, C. Mackin, R. Joesten and S. L. Suib, Structural
L. S. G. Teixeira, L. L. N. Guarieiro, M. d. C. Rangel,
M. C. C. Veloso, M. J. C. Rezende, R. S. da Cruz, et al., The
Role of Additives for Diesel and Diesel Blended (Ethanol or
Biodiesel) Fuels: A Review, Energy Fuels, 2007, 21(4), 2433–
2445.
32672 | RSC Adv., 2019, 9, 32665–32673
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
18 A. Omari, B. Heuser and S. Pischinger, Potential of 27 C. J. Gommes, S. Blacher, J. H. Dunsmuir and A. H. Tsou,
Oxymethylenether-Diesel Blends for Ultra-Low Emission
Engines, Fuel, 2017, 209, 232–237.
19 N. D. Wasalathanthri, T. M. SantaMaria, D. A. Kriz,
Selective Oxidation of Methanol to Dimethoxymethane
over Mesoporous Al–P–V–O Catalysts, AIChE J., 2012, 55(8),
2000–2012.
´
´
S. L. Dissanayake, C. H. Kuo, S. Biswas and S. L. Suib, 28 J. Gornay, X. Secordel, G. Tesquet, B. De Menorval, S. Cristol,
Mesoporous Manganese Oxides for NO2 assisted Catalytic
Soot Oxidation, Appl. Catal., B, 2017, 201(2), 543–551.
20 J. M. Tatibouet, Methanol Oxidation as a Catalytic Surface
Probe, 1997, vol. 148.
P. Fongarland, M. Capron, L. Duhamel, E. Payen,
J. L. Dubois, et al., Direct Conversion of Methanol into 1,1-
Dimethoxymethane: Remarkably High Productivity over an
FeMo Catalyst Placed under Unusual Conditions, Green
Chem., 2010, 12(10), 1722–1725.
21 R. J. Beuhler, R. M. Rao, J. Hrbek and M. G. White, Study of
the Partial Oxidation of Methanol to Formaldehyde on 29 S. L. Suib, L. Pahalagedara, Y. Meng, S. Dissanayake,
a Polycrystalline Ag Foil, J. Phys. Chem. B, 2001, 105(25),
5950–5956.
22 C.-T. Wang and R. J. Willey, Mechanistic Aspects of
Methanol Partial Oxidation over Supported Iron Oxide
Aerogels, J. Catal., 2001, 202(2), 211–219.
N. Wasalathanthri, M. Pahalagedara, R. J. Meyer,
C. Weerakkody, P. Nandi, D. A. Kriz, et al., Benchmarking
of Manganese Oxide Materials with CO Oxidation as
Catalysts for Low Temperature Selective Oxidation, Appl.
Catal., B, 2016, 204, 411–420.
23 Y. Fu and J. Shen, Selective Oxidation of Methanol to 30 H. C. Genuino, Y. Meng, D. T. Horvath, C.-H. Kuo,
Dimethoxymethane under Mild Conditions over V2O5/TiO2
with Enhanced Surface Acidity, Chem. Commun., 2007,
2(21), 2172–2174.
M. S. Seraji, A. M. Morey, R. L. Joesten and S. L. Suib,
Enhancement of Catalytic Activities of Octahedral
Molecular Sieve Manganese Oxide for Total and
Preferential CO Oxidation through Vanadium Ion
Framework Substitution, ChemCatChem, 2013, 5(8), 2306–
2317.
24 X. Chen, P. Wu, G. Zeng, Q. Wang, Z. Liu, Y. Sun, G. Liu,
T. Qin and R. Sima, Selective Oxidation of Methanol to
Dimethoxymethane at Low Temperatures through Size-
Controlled VTiOx Nanoparticles, ChemCatChem, 2017, 31 N. D. Wasalathanthri, A. S. Poyraz, S. Biswas, Y. Meng,
9(10), 1776–1781.
C.-H. Kuo, D. A. Kriz and S. L. Suib, High-Performance
Catalytic CH4 Oxidation at Low Temperatures: Inverse
Micelle Synthesis of Amorphous Mesoporous Manganese
Oxides and Mild Transformation to K2ꢂxMn8O16 and 3-
MnO2, J. Phys. Chem. C, 2015, 119(3), 1473–1482.
25 D. Li, D. Jiang, W. Li, Y. Sun and H. Guo, The One-Step
Oxidation of Methanol to Dimethoxymethane over
Nanostructure Vanadium-Based Catalysts, Catal. Lett.,
2010, 135(1–2), 48–56.
26 S. Chen, S. Wang, X. Ma and J. Gong, Selective Oxidation of 32 H. Liu, N. Gao, M. Liao and X. Fang, Hexagonal-like Nb2O5
Methanol to Dimethoxymethane over Bifunctional VOx/TS-1
Nanoplates-Based Photodetectors and Photocatalyst with
Catalysts, Chem. Commun., 2011, 47(33), 9345–9347.
High Performances, Sci. Rep., 2015, 5(7716), 1–9.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 32665–32673 | 32673