Oxidative dehydrogenation of ethane on diluted or promoted nickel oxide catalysts: Influence of the promoter/diluter

Article abstract of DOI:10.1016/j.cattod.2019.06.063

Ti- and Nb- containing NiO catalysts have been synthesized by two different preparation methods: i) by precipitation (Me-Ni-O oxides, Me = Nb or Ti), in order to prepare promoted NiO catalysts; and ii) by wet impregnation on TiO₂ or NbO_x supports, in order to prepare diluted/supported NiO catalysts. The catalysts have been also characterized and tested in the oxidative dehydrogenation of ethane. The catalytic performance of Ti- and Nb-promoted catalysts strongly depends on the composition, although in both cases the optimal one is found at similar Ti or Nb loadings (ca. 90 wt% NiO), showing similar ethylene selectivity in the ODH of ethane (ca. 90% at 10–20% ethane conversion). However, in the case of diluted catalysts, the catalytic behavior of Ti- and Nb-containing catalysts is drastically different. Then, over TiO₂-diluted NiO catalyst, the highest selectivity to ethylene (ca. 90% selectivity) is achieved at NiO loading of 20 wt.%. However, over Nb₂O₅-diluted NiO catalyst, selectivity to ethylene was lower than 70%. A discussion on the characteristics of selective catalysts is done. In this case, the best catalysts must present a low concentration of free NiO and TiO₂ or Nb₂O₅ phases, maximizing the Ni-O-Ti or Ni-O-Nb interaction. Interestingly, this takes place at different NiO loading depending on the preparation method and the nature of promoted/diluter. The low selectivity to ethylene achieved by NiO diluted with Nb₂O₅ has been related to the low interaction of NiO with the surface of Nb₂O₅, which hinders the elimination of unselective electrophilic O species.

Full text of DOI:10.1016/j.cattod.2019.06.063

CatalysisTodayxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Oxidative dehydrogenation of ethane on diluted or promoted nickel oxide
catalysts: Influence of the promoter/diluter
D. Delgadoa, B. Solsonab,⁎, R. Sanchisb, E. Rodríguez-Castellónc, J.M. López Nietoa,⁎
a Instituto de Tecnología Química, Universitat Politècnica de València-Consejo Superior de Investigaciones Científicas, Avenida de los Naranjos s/n, 46022, Valencia, Spain
b Departament d´Enginyeria Química, Universitat de València, C/ Dr. Moliner 50, 46100, Burjassot, Valencia, Spain
c Departamento de Química Inorgánica, Facultad de Ciencias, Universidad de Málaga, 29071, Málaga, Spain
A R T I C L E I N F O
A B S T R A C T
Keywords:
ODH of ethane
Supported nickel oxide
Promoted nickel oxide
Niobium oxide
Ti- and Nb- containing NiO catalysts have been synthesized by two different preparation methods: i) by pre-
cipitation (Me-Ni-O oxides, Me = Nb or Ti), in order to prepare promoted NiO catalysts; and ii) by wet im-
pregnation on TiO2 or NbOx supports, in order to prepare diluted/supported NiO catalysts. The catalysts have
been also characterized and tested in the oxidative dehydrogenation of ethane. The catalytic performance of Ti-
and Nb-promoted catalysts strongly depends on the composition, although in both cases the optimal one is found
at similar Ti or Nb loadings (ca. 90 wt% NiO), showing similar ethylene selectivity in the ODH of ethane (ca. 90%
at 10–20% ethane conversion). However, in the case of diluted catalysts, the catalytic behavior of Ti- and Nb-
containing catalysts is drastically different. Then, over TiO2-diluted NiO catalyst, the highest selectivity to
ethylene (ca. 90% selectivity) is achieved at NiO loading of 20 wt.%. However, over Nb2O5-diluted NiO catalyst,
selectivity to ethylene was lower than 70%. A discussion on the characteristics of selective catalysts is done. In
this case, the best catalysts must present a low concentration of free NiO and TiO2 or Nb2O5 phases, maximizing
the Ni-O-Ti or Ni-O-Nb interaction. Interestingly, this takes place at different NiO loading depending on the
preparation method and the nature of promoted/diluter. The low selectivity to ethylene achieved by NiO diluted
with Nb2O5 has been related to the low interaction of NiO with the surface of Nb2O5, which hinders the elim-
ination of unselective electrophilic O species.
Titanium oxide
1. Introduction
still not high enough to displace the current non-catalytic process, but
also for the risks associated to any technology shift, since steam-
The use of oil derivatives as fuels, especially in the automotive
sector, is expected to decrease in Europe. In fact, this decrease is fore-
seen to be drastic from 2040, as several countries, such as Spain, UK,
Denmark and Norway, have planned to ban vehicles fed with petrol and
diesel. Therefore, the interest of oil and natural gas, especially as raw
materials to be transformed into useful non-fuel products, is im-
portantly increasing. Nowadays, ethylene, with a worldwide production
of 150 million tons in 2015, can be considered as the most important
feedstock in Petrochemistry [1]. Ethylene is mainly produced by steam
cracking, a highly energy consuming and non-catalytic process [2,3].
The oxidative dehydrogenation (ODH) of ethane to ethylene using ac-
tive and selective catalysts can be considered as an interesting alter-
native to the steam cracking process, due to the exothermic character of
the reaction, the in situ reactivation of the catalysts and the relatively
low reaction temperatures required [4,5]. Unfortunately, this process is
not industrially developed, due to the fact that yields to ethylene are
cracking is well-established and optimized process.
Modified NiO materials are, together with multicomponent MoV
(Te,Sb)Nb oxides [6,7], the most promising catalysts for the ODH of
ethane. Patents of Symix [8] and the pioneer works of Lemonidou et al.
[9,10] with Ni-Nb-O catalysts firstly showed the high potential of this
type of materials. In fact, ethylene yields over 45% have been reported.
Pure NiO easily activates ethane at low reaction temperatures
(< 400 °C). Unfortunately, most of ethane is transformed into CO2 and
only a small fraction is converted into ethylene. Interestingly, the ad-
dition of several promoters to NiO has demonstrated to have a positive
effect in the ethylene formation. We can mention metal oxide pro-
moters as in the case of WO3-NiO [11,12], CeO2-NiO [13], ZrO2-NiO
[14], and Ta2O5-NiO [15] catalysts. The use of Sn [16,17] and, espe-
cially, Nb [18–22] as promoters of NiO leads to the best catalytic per-
formance, achieving high selectivity to ethylene at moderate and high
ethane conversions. In these promoted catalysts, mainly prepared by
⁎ Corresponding author.
E-mail addresses: benjamin.solsona@uv.es (B. Solsona), jmlopez@itq.upv.es (J.M. López Nieto).
https://doi.org/10.1016/j.cattod.2019.06.063
Received 6 February 2019; Received in revised form 28 May 2019; Accepted 15 June 2019
0920-5861/©2019PublishedbyElsevierB.V.
Pleasecitethisarticleas:D.Delgado,etal.,CatalysisToday,https://doi.org/10.1016/j.cattod.2019.06.063

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
mixing solutions of the corresponding nickel and promoter salts, the
formation of solid solutions and/or the partial incorporation of the
promoter into the NiO lattice is favored. High valence of the foreign
cation is suitable for achieving good performance, whereas the addition
of alkalis and alkaline earths does not improve, or even worsen, the
catalytic behavior compared to pure nickel oxide.
Although most of scientific papers deal with promoted NiO cata-
lysts, supported/diluted NiO catalysts have also demonstrated to lead to
high ethylene formation. Particularly, supports such as Al2O3 [23],
porous clays [24], TiO2 [25] or complex catalysts such as NiO-Al2O3/
Ni-foam [26] have reached similar performance than optimal promoted
NiO catalysts. In contrast, the use of silica, which allows only a weak
NiO-support interaction, hardly modifies the poor behavior of pure NiO
[24].
The reason for the drastic improvement observed in supported/di-
luted or promoted NiO catalysts compared to undoped NiO is not fully
understood. Nevertheless, it is known that an excess of high valence
Ni3+ species, the presence of non-stoichiometric oxygen species favors
the undesired ethane deep oxidation [18,27,28]. The concentration of
defects, mainly Ni and O vacancies, has also demonstrated to play an
important role on the catalytic performance [12,29]. Overall the in-
teraction between nickel and the promoter or diluter must be max-
imized whereas the amount of unmodified NiO sites must be mini-
mized.
2.2. Catalytic tests in the ODH of ethane
Ethane oxidation tests were conducted in a tubular isothermal flow
reactor, mainly in the 350–450 °C temperature range. The feed con-
sisted of a mixture of C2H6/O2/He with a molar ratio of 3/1/26. The
contact times were varied by modifying the catalyst weight or the total
flow in order to obtain the desired conversions at a given reaction
temperature. Catalysts were introduced in the reactor together with
silicon carbide in order to reach a constant volume in the catalytic bed.
Reactants and reaction products have been analyzed by gas chroma-
tography. Two packed columns were necessary to carry out the ana-
lyses: (i) molecular sieve 5 Å (2.5 m); and (ii) Porapak Q (3 m). Blank
runs in the absence of catalyst showed no conversion in the range of
reaction temperatures studied. Further details of the reaction system
are included in Supporting Information.
2.3. Characterization techniques
N2-adsorption isotherms were collected in a Micromeritics ASAP
2000. Approximately 300 mg of sample were degassed in vacuum at
400 °C prior to nitrogen adsorption. Surface areas were calculated by
BET method.
X-ray diffraction patterns were measured in Bragg-Brentano geo-
metry in a PANalytical X´Pert PRO diffractometer with an X´Celerator
detector. Diffractograms were collected using Cu-Kα radiation.
Raman spectra were obtained in an inVia Renishaw spectrometer
equipped with a Renishaw HPNIR laser, at an excitation wavelength of
514 nm. Power on the samples was of ca. 15 mW.
Nb5+ is probably the best promoter for NiO reported up to date,
whereas TiO2 can be considered among the best NiO supports for the
ODH of ethane. Thus, in the present article we have followed two
synthetic approaches, by which NiO has been promoted with Ti4+ or
Nb5+(in promoted NiO catalysts) or diluted with the corresponding
oxides (i.e. TiO2 or Nb2O5, in diluted catalysts), with the aim of un-
derstanding the role and main effects of promoters and diluters in the
chemical nature of NiO. The results are presented taking into con-
sideration changes in the chemical nature of NiO and its role in the
catalytic behavior in the ODH of ethane.
XPS studies were carried out on
a Physical Electronics PHI
VersaProbe II spectrometer using monochromatic Al-Kα radiation
(49.1 W, 15 kV and 1486.6 eV) for analyzing the core-level signals of
the elements of interest with a hemispherical multichannel analyzer.
The energy scale of the spectrometer was calibrated using Cu 2p3/2, Ag
3d5/2 and Au 4f7/2 photoelectron lines at 932.7, 368.2 and 84.0 eV,
respectively. Under a constant pass energy mode at 23.5 eV condition,
the Au 4f7/2 line was recorded with 0.73 eV FWHM at a binding energy
(BE) of 84.0 eV. The X-ray photoelectron spectra obtained were ana-
lyzed using PHI SmartSoft software and processed using MultiPak 9.3
package. The binding energy values were referenced to adventitious
carbon C 1s signal (284.8 eV). Shirley-type background and Gauss-
Lorentz curves were used to determine the binding energies.
2. Experimental
2.1. Catalyst synthesis
Diluted NiO/TiO2 or NiO/Nb2O5 catalysts were prepared through
the evaporation at 60 °C of
a stirred ethanolic solution of Ni
X-ray absorption measurements in the Ni K-edge were carried out at
CLAESS line at ALBA synchrotron (Barcelona, Spain). Spectra were
collected from 8200 to 9175 eV. The optimum mass amount of each
catalyst (i.e. the one to maximize signal-to-noise ratio; Ln(Io/I1) ≈ 1)
was diluted in boron nitride and pressed into wafers. Spectra normal-
ization was carried out in Athena software.
(NO3)2•6H2O (from Sigma-Aldrich) and oxalic acid (oxalic acid/Ni
molar ratio of 3) to which the corresponding titanium or niobium oxide
was added. The solids obtained were dried overnight at 120 °C and fi-
nally calcined in static air at 500 °C for 2 h. The catalysts have been
named as xNiO/Y, in which x is the theoretical NiO wt% loading and Y
the diluter employed (TiO2 or Nb2O5).
Temperature-programmed reduction in H2 was performed in a
Micromeritics Autochem 2910 device, which was equipped with a TCD
detector. A mixture 10% H2 in Ar was used to perform the reduction
(total flow of 50 mL min−1). Samples were heated up to 800 °C at a
The TiO2 support employed (Degussa P25) mainly consists of ana-
tase (low proportion of rutile) and presents a surface area of 55 m2 g−1
.
Nb2O5 support was prepared by hydrothermal synthesis. An aqueous
solution of ammonium niobate(V) oxalate hydrate (Sigma Aldrich) was
heat treated at 80 °C for 10 min, and subsequently introduced in a
Teflon-lined stainless steel autoclave, which was heat-treated at 175 °C
for 48 h. Finally, the resulting solid was filtered, washed with distilled
water, dried (16 h at 100 °C) and heat-treated under N2 flow for 2 h at
550 °C (66 m2 g−1).
heating rate of 10 °C min−1
.
Scanning Electron Microscopy (SEM) images were collected in a
JEOL 6300 Microscope, equipped with an Oxford LINK ISIS system to
perform compositional analysis by energy-dispersive X-ray
Spectroscopy (EDX)
Promoted Ni-Ti-O and Ni-Nb-O catalysts were prepared by evapor-
ating an ethanolic mixture of nickel nitrate Ni(NO3)2•6H2O (Sigma-
Aldrich), niobium oxalate monooxalate adduct C10H5NbO20 (ABCR) or
titanium ethoxide C8H20O4Ti (ACROS) and oxalic acid (oxalic acid/Ni
molar ratio of 3). The pastes obtained were dried overnight in a furnace
at 120 °C and then calcined in static air at 500 °C for 2 h. The catalysts
have been named as xNi-Ti-O or xNi-Nb-O, in which x is the theoretical
NiO wt% loading.
3. Results and discussion
3.1. Catalytic performance in the ODH of ethane
As Nb-promoted NiO samples are possibly the most efficient nickel
oxide catalysts for the ODH of ethane to ethylene [9], we wanted to
know if supporting/diluting NiO on Nb2O5 by a wet impregnation
method could have similar effect on the catalytic properties than
standard Nb-promoted NiO catalysts. Thus, a set of NiO catalysts
Detailed synthesis procedures for both promoted and supported NiO
catalysts are included in Supporting Information.
2

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Table 1
catalysts (Fig. 1B). In both cases a high selectivity to ethylene can be
reached, either by promoting with Ti (92Ni-Ti-O, ca. 90% selectivity to
ethylene) (Table 1), or by diluting nickel oxide with TiO2 (20NiO/TiO2,
ca. 90% selectivity to ethylene) (Table 2). The variation of the se-
lectivity to ethylene with NiO-loading in Ti-promoted series follows a
similar trend than that observed in Nb-promoted materials, i.e. the
selectivity progressively increases with NiO in the catalysts, attaining
its maximum value at NiO loading of 92 wt% (Table 1) (Fig. 1B). On the
contrary TiO2-diluted nickel oxide materials displayed a different trend,
in which the selectivity to ethylene progressively increases, reaching a
maximum at NiO-loadings of 20 wt% (Fig. 1B). In this way, 20NiO/TiO2
sample shows a selectivity comparable to the one obtained with the best
Ti-promoted catalysts (92Ni-Ti-O). Then, both the synthesis procedures
(i.e. evaporation or impregnation) and NiO-loading have an important
effect on the nature of active sites available in each series of catalysts.
One of the key aspects in the development of effective catalysts for
the ODH of ethane is to study the evolution of the selectivity to the
olefin when the conversion of ethane increases. This is a crucial point
when the aim is achieving high yields to ethylene in ethane ODH, since
ethylene tends to react on non-selective sites to give carbon oxides in
most of the catalytic systems reported in literature.
Catalytic results in the ODH of ethane obtained by supported/diluted NiO/TiO2
and NiO/Nb2O5 catalysts prepared by wet impregnationa.
Catalyst
Surface
area
NiO
(wt.
%)
Ethane
conversion
(%)
Selectivity
to ethylene
(%)
Ethylene
productivityc
(m2 g−1
)
NiO
15.4
40.2
51.4
46.6
50.4
52.2
55.4
100
92
80
50
20
5
7.5
33.3
64.2
75.5
87.0
89.3
84.7
55.4
58.6
58.8
63.4
59.7
59.2
48.3
0
339
92NiO/TiO2
80NiO/TiO2
50NiO/TiO2
20NiO/TiO2
5NiO/TiO2
TiO2
98NiO/Nb2O5 n.a.
92NiO/Nb2O5 39.9
80NiO/Nb2O5 n.a.
50NiO/Nb2O5 61.2
20NiO/Nb2O5 65.0
15.5
16.3
15.9b
11.9b
6.4b
1.9b
20.5
21.1
18.7
13.4
10.1
3.2
1360
1683
946
726
371
72.0
1642
1696
1621
1094
817
0
98
82
80
50
20
5
5NiO/Nb2O5
Nb2O5
n.a
66.4
211
0
0
0.6b
a
−1
At 450 °C and a contact time, W/F, of 2 gcat h molC2
At 450 °C and a contact time, W/F, of 4 gcat h molC2
Formation rate of ethylene as gC2H4/kgcat h.
.
.
b
c
−1
Fig. 2 shows the evolution of the selectivity to ethylene with the
ethane conversion (at 450 °C) for representative diluted and promoted
NiO catalysts (Fig. 2A and B, respectively). Unmodified NiO shows a
very low selectivity to ethylene (ca. 35%), being CO2 the main reaction
product. The non-detection of CO in the output stream has been re-
ported to be due to the relatively low reactivity of ethylene over these
catalysts (i.e. a low tendency of NiO-based materials to ethylene deep
oxidation) [30]. The fact that ethylene selectivity does not substantially
decrease when increasing ethane conversion (even on this non-selective
unmodified NiO catalyst) would be in line with this assumption.
However, nickel oxide has been also reported to be a highly efficient
catalyst in the oxidation of CO to CO2, which is able to perform at much
lower reaction temperatures than those applied in the present study
[31,32].
supported on Nb2O5 (NiO/Nb2O5 series) was synthesized with different
NiO loadings and compared with Nb-promoted NiO catalysts (Ni-Nb-O
series). On the other hand, NiO/TiO2 catalysts have shown one of the
most promising results in the ODH of ethane among all supported NiO
systems [25]. Accordingly, series of promoted (Ni-Ti-O) and diluted
(NiO/TiO2) nickel oxide catalysts with different NiO loadings have been
prepared and tested in the oxidative dehydrogenation of ethane.
The catalytic results in the ODH of ethane for diluted and promoted
catalysts with different NiO contents are summarized in Tables 1 and 2,
respectively. Fig. 1 shows the variation of the selectivity to ethylene
(measured at an ethane conversion of ca. 10%) as a function of NiO-
loading for Nb- and Ti-containing catalysts, prepared by evaporation
(promoted series) or by impregnation (diluted series).
When nickel oxide is dispersed using TiO2 and Nb2O5 as diluters, the
selectivity to ethylene in the ODH of ethane increases drastically, up to
60–70% in the less selective catalysts, and up to 90% selectivity in the
most selective ones (i.e. 20 NiO/TiO2) (Fig. 2A). More importantly, the
selectivity to ethylene remains almost constant in all cases when ethane
conversion increases. We can tentatively ascribe this effect to the ab-
sence of those non-selective sites that deeply oxidize ethylene to CO2.
This behavior is also observed in Nb-and Ti-promoted catalysts with
high NiO-loadings (92Ni-Nb-O and 92Ni-Ti-O samples), which are the
most selective catalysts within promoted series. On the contrary, a
decrease in the selectivity to ethylene when ethane conversion in-
creases is observed on those promoted catalysts with low NiO-loadings.
In this case, the decreasing trend indicates ethylene decomposition, due
to the consecutive transformation of the olefin to carbon oxides.
The formation rate of ethylene (STY) has also been determined on
these catalysts (Figure S1). The variation of STY with NiO-loading fol-
lows the same trend regardless of the synthetic procedure and the
promoter/diluter used. Thus, the productivity increases drastically even
at very low amounts of diluter/promoter, likely due to the observed
increase of the selectivity to ethylene (even at NiO loadings near 100 wt
%). According to the different physicochemical features observed along
NiO-based catalyst series, the fact that the catalytic activity does not
vary proportionally with NiO content in the materials can be explained
in terms of the different surface areas and NiO particle size, which
would lead to different amounts of exposed active sites, as it will be
discussed below. Overall, it has been observed that the catalytic per-
formance strongly depends on both the preparation method (diluted or
promoted catalysts) and on the nickel content.
Considering Nb-containing materials, the catalytic behavior differs
for Nb-promoted and Nb2O5-diluted catalysts (Fig. 1A). In the case of
promoted series, the selectivity to ethylene gradually increases with
NiO-loading, reaching a maximum of 90% at NiO-loadings of 92 wt%.
On the other hand, Nb2O5-diluted materials present much lower ethy-
lene selectivity (in the 48–64% range) regardless of the amount of NiO
loaded. Nevertheless, the selectivity to ethylene in the ODH of ethane
notably increases by both synthetic approaches, compared to un-
modified NiO (Sethylene = 33.3%).
A different trend is observed for Ti-promoted and TiO2-diluted
Table 2
Catalytic results in the ODH of ethane obtained over Ti- and Nb-promoted NiO
catalystsa.
Catalyst
Surface
area
NiO
wt.%
Ethane
conversion
(%)
Selectivity to
ethylene (%)
Ethylene
productivityc
(m2 g−1
)
NiO
15.4
n.d
122
n.d
129
58.9
100
98
92
80
50
20
97
92
80
50
20
7.5
33.3
68.4
88.1
84.3
79.9
83.7
74.3
86.2
86.3
87.5
90.8
339
804
1686
2028
955
98Ni-Ti-O
92Ni-Ti-O
80Ni-Ti-O
50Ni-Ti-O
20Ni-Ti-O
97Ni-Nb-O n.a.
92Ni-Nb-O 157
80Ni-Nb-O n.a.
50Ni-Nb-O 107
20Ni-Nb-O 31.6
17.2b
14.0
17.6
17.5b
6.2b
21.4
18.5
13.3
4.1
354
2174
2180
1569
491
0.9
112
a
−1
High selectivity to ethylene can be obtained using Ni-Ti catalysts
prepared either by impregnation (diluted NiO/TiO2 catalysts) or by
evaporation of solutions (promoted Ni-Ti-O). However, while achieving
At 450 °C and a contact time, W/F, of 2 gcat h molC2
At 450 °C and a contact time, W/F, of 4 gcat h molC2
Formation rate of ethylene as gC2H4/kgcat h.
.
.
b
c
−1
3

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 1. Selectivity to ethylene in the ODH of ethane as a function of NiO-loading for promoted/diluted NiO catalysts: A) xNb-Ni-O and xNiO/Nb2O5-series; B) xTi-Ni-
O and xNiO/TiO2-series. Reaction conditions in text.
Fig. 2. Variation of the selectivity to ethylene with ethane conversion during the ODH of ethane for selected diluted (A) and promoted (B) NiO catalysts. Reaction
conditions in text. Temperature = 450 °C.
similar selectivity to ethylene, different NiO-loadings are required de-
pending on the synthetic approach chosen. Thus, much lower NiO
loadings are needed in the case of diluted catalysts (20 wt% NiO) than
in the case of Ti-promoted NiO (92 wt% NiO). It must be noted that
NiO-loading must not be too high, in order to avoid ethane total oxi-
dation, but neither too low, in order to prevent the ethylene decom-
position into CO2. As similar selectivity vs conversion profiles are ob-
served in the optimal catalysts of each series, similar or equivalent Ni-Ti
interactions must be taking place. The optimal Ni-loading is different in
each case (for promoted and diluted catalysts), likely because an im-
portant part of TiO2 support cannot interact with nickel oxide (i.e. the
presence of non-accessible bulk Ti sites).
In the case of Ni-Nb materials, the catalytic behavior highly differs
depending on the preparation method. Thus, when using promoted Ni-
Nb-O catalysts with the appropriate composition, a high selectivity to
ethylene can be attained. On the contrary, the selectivity to ethylene on
NiO/Nb2O5 catalysts hardly exceeds 60%. Then it seems that the in-
teraction between nickel oxide and niobium is different in diluted and
promoted catalysts. The reason for the poor catalytic performance of
diluted NiO/Nb2O5 mainly arises from the low initial selectivity to
ethylene, due to the direct decomposition of ethane into carbon di-
oxide. In promoted Ni-Nb-O catalysts the selectivity achieved is high,
but the amount of nickel must be controlled in order to avoid both
ethylene decomposition (if nickel content is too low) and ethane total
oxidation (if nickel content is too high).
3.2. Physicochemical characterization of promoted and diluted NiO
catalysts
Surface areas of both diluted and promoted catalysts were de-
termined by BET method from N2-adsorption isotherms (Table 1). Un-
modified NiO presents low surface area (15 m2/g). In the case of diluted
catalysts, two supports/diluters with similar surface area were used to
deposit NiO (i.e. TiO2 and Nb2O5, with surface area of 55 and 66 m2/g,
respectively). When nickel is added by impregnation to both metal
oxides, a progressive decrease in the surface area is observed as the
nickel content increases (Table 1) (Fig. S2).
On the other hand, promoted catalysts (prepared by evaporation of
solutions) show higher surface areas (over 100 m2/g) than diluted
catalysts (Table 2) (Fig. S1). BET surface areas increase ten-fold even at
low promoter contents (8 wt%, i.e. 92 wt% NiO loading), and decreases
with the amount of Ni loaded (Table 2) (Fig. S2).
XRD patterns of diluted and promoted NiO-based catalysts are de-
picted in Figure S3 and Figure S4, respectively (see Supporting
Information). In the case of NiO/TiO2 (Fig. S3, A) and NiO/Nb2O5 (Fig.
S3, B) series, all the materials show Bragg reflections corresponding to
NiO cubic phase (Fm3m, ICSD No: 184626), together with the char-
acteristics diffraction signals of the corresponding diluter oxide used in
4

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 3. Raman spectra of Nb- (A) and Ti- (B) promoted NiO catalysts.
each case, i.e. TiO2-P25 (showing a mixture of anatase and rutile-type
phases) [33], and a highly distorted niobium oxide [34] (Fig. S3). On
the other hand, promoted Ni-Nb-O and Ni-Ti-O catalysts present dif-
ferent structural features in comparison with diluted series, showing a
comparable trend regardless of whether Nb or Ti are incorporated as
promoters (Fig. S4, A and Fig. S4, B; respectively). At low promoter
contents (NiO contents in the 0–80 wt% range), either for Ti-promoted
or Nb-promoted series, the catalysts display the only presence of dif-
fraction lines corresponding to NiO (Fig. S4, A; patterns a to d). At
higher promoter concentrations (NiO wt% of 50 and 20%) additional
features appear in the diffraction patterns. In the case of Nb-promoted
catalysts, broad diffraction peaks are found at ca. 2θ = 20–40°, which
suggest the presence of amorphous niobium oxide [35] (Fig. S4, A;
patterns d and f). For Ni-Ti-O series, TiO2-anatase diffraction signals
(JCPDS: 84-1286) are found at these promoter contents (NiO wt% of 50
and 20%) (Fig. S4, B; patterns e and f). All these observations suggest a
limit in the incorporation of both Ti and Nb within nickel oxide fra-
mework, after which simple titanium and niobium oxides are found as
secondary phases. Nevertheless, according to broad features observed
in the diffraction patterns, the presence of other Ni-promoter mixed
oxide phases cannot be ruled out, as it will be discussed below.
(d = 12.4–15.1 nm). Also, a slight increase in NiO a lattice parameter is
observed in all modified NiO materials (Table S1). Again, the effect is
more significant in Ni-Nb-O and Ni-Ti-O promoted series. Thus, the use
of promoters seems to modify the original crystal framework and
morphological characteristics of NiO in a higher extent than diluters do.
To get further insights into promoter/diluter-NiO interactions, se-
lected samples were analyzed by Raman spectroscopy. Fig. 3 displays
Raman spectra of promoted Ni-Ti-O and Ni-Nb-O catalysts (Fig. 3A and
B, respectively). Unpromoted/undiluted NiO shows a main Raman band
centered at 493 cm−1, associated to Ni-O deformation modes [36]
(Fig. 3, spectrum a). The original position of this band is shifted to
higher frequencies (up to ca. 520 cm−1) when Nb and Ti are added as
promoters, even at low promoter concentrations (Fig. 3, spectra b and
c). At higher promoter contents, Raman signals associated to the pre-
sence of Nb2O5 (band centered at 703 cm−1) [37] and TiO2-anatase
(bands at 395, 516 and 638 cm−1) [38] are also observed for Ni-Nb-O
and Ni-Ti-O catalysts, respectively (Fig. 3, spectra d and e). Interest-
ingly, additional features appear in the 600–900 cm−1 range in both Nb
and Ti-promoted materials. In the case of Nb-promoted series, the
spectra of the catalysts display bands at 782 and 849 cm−1, ascribed to
the presence of Ni-Nb-O mixed phases [39] and a signal at 908 cm−1
,
Focusing on the possible effects of the promoter/diluter on the
structural characteristics of NiO, we have calculated both average
crystallite size (by using Scherrer equation) and a-parameter (by profile
fitting). The results for both promoted and diluted catalysts are sum-
marized in Table S1. A decrease in the crystallite size of NiO (dNiO
=25 nm) is observed in all cases. This fact can be deduced from NiO
line broadening found for all promoted and diluted materials.
attributed to Nb = O stretching vibrations [40] (Fig. 3A). In the same
way, Ti-promoted catalysts present Raman bands centered at 702 and
770 cm−1, which can be also assigned to mixed Ni-Ti-O oxide, speci-
fically to an ilmenite NiTiO3-type phase [41] (Fig. 3B). Moreover, ap-
plying a higher heat-treatment temperature (i.e. 700 °C), this ilmenite-
type phase is clearly identified by X-ray diffraction (JCPDS: 33-0960)
(Fig. S5).
Interestingly, promoted catalysts show
(d = 6.5–9.9 nm), i.e. broader diffraction peaks, than diluted catalyst
a
smaller particle size
In the case of diluted-NiO catalysts, both NiO/TiO2 and NiO/Nb2O5
present common features in the Raman spectra, which are different
5

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 4. Ni 2p3/2 core level spectra of promoted and diluted NiO catalysts. A) Ni-Nb-O series. B) Ni-Ti-O series. C) NiO/Nb2O5 series. D) NiO/TiO2 series.
than those found in promoted NiO catalysts (see Fig. S6 in Supporting
Information). In this sense, both series show always Raman band cor-
responding to NiO (ca. 493 cm−1), which is not shifted regardless the
diluter content used. In addition, they displayed characteristic Raman
diluted NiO catalysts (Fig. 4). This effect is more evident in Ni-Nb-O and
Ni-Ti-O materials than in the case of diluted NiO/Nb2O5 and NiO/TiO2,
suggesting that promoters are able to modify the chemical nature of
surface nickel species in a higher degree than diluters.
Fig. 5 displays the corresponding O 1s core level spectra of pro-
moted and diluted NiO catalysts. The spectra of unpromoted/undiluted
NiO presents two contributions, appearing at binding energy values of
529.4 and 531.1 eV, which can be assigned to nucleophilic and elec-
trophilic surface oxygen species, respectively (Fig. S7) [46].
The effect of nucleophilic and electrophilic oxygen sites constitutes
one of the central points in selective oxidation [47]. In this way, nu-
cleophilic lattice oxygen (i.e. O2−) has been reported to be responsible
for the partial oxidation of the substrate, meanwhile electrophilic
oxygen species (likely O- or adsorbed O2-) are prone to deep oxidation
[48]. In this case, there exists a general trend among all analyzed di-
luted and promoted catalysts. Thereby, the XPS signal at B.E
=531.1 eV, corresponding to electrophilic oxygen species, decreases
for the most selective catalysts (Fig. 5) (Table S2 and Figs. S8 and S9 in
Supporting Information). Conversely, diluted NiO/Nb2O5 catalysts
show the highest relative intensity of this signal, and the lowest se-
lectivity to ethylene from all the series.
signals of each diluter added, i.e. a highly distorted Nb2O5 (700 cm−1
)
(Fig. S6, A) and TiO2-P25 (395, 515 and 638 cm−1) (Fig. S6, B).
To evaluate possible modifications in the nature of nickel and
oxygen surface species, selected catalysts were characterized by XPS.
Fig. 4 shows the Ni 2p3/2 core level spectra of diluted and promoted
catalysts. All the spectra show the typical NiO features, displaying a
main peak in the 853–854 eV range, which shows line-broadening,
which is generally considered a new peak (ca. 1.5 eV over the main
signal), known as satellite I (Sat I). This Sat I has been reported to
appear due to the presence of a wide variety of defects, such as Ni3+
species [42], Ni2+-OH sites or Ni2+ vacancies [43]. In addition to Sat I,
another broad shake-up satellite (Sat 2) appears at ca. 7 eV over the
main peak, which is generally attributed to ligand-metal charge transfer
[43–45].
According to the complex nature of the NiO system, derived prin-
cipally from its non-stoichiometric nature, it is difficult to discriminate
among all the species that actually contribute to the XPS spectra in each
case. Nevertheless, a decrease in the relative intensity of the main peak
with respect to Sat I is observed along all the series of promoted and
Surface Ni content for diluted and supported materials was calcu-
lated from XPS spectra (Table S2). For promoted catalysts and TiO2-
6

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 5. O 1s core level spectra of promoted and diluted NiO catalysts. A) Ni-Nb-O series. B) Ni-Ti-O series. C) NiO/Nb2O5 series. D) NiO/TiO2 series.
supported series, surface Ni concentrations increases proportionally
with Ni-loading in the samples. However, Nb2O5-supported samples
show high surface Ni contents, regardless of the amount of Ni loaded in
the synthesis. We analyzed selected samples from NiO/TiO2 and NiO/
Nb2O5 series by SEM-EDX (Figs. S10 and S11). While TiO2-supported
NiO catalysts show a homogeneous Ni-content in all cases, Nb2O5-
supported catalysts present a highly heterogeneous Ni-concentration all
along the samples.
Hence, both diluters and promoters would act as modifiers of the
O2− sub-lattice, by eliminating the more electrophilic species, and
improving the selectivity to ethylene in the ODH of ethane. The extent
of this interaction will mainly depend on the nature of the support
/diluter. Nb5+ and Ti4+ can be considered as high valence dopants
within the pristine NiO framework, then pushing Ni to its lower oxi-
dation state, i.e. Ni2+, thus decreasing the number of electrophilic
oxygen species [28]. On the other hand, when using diluters instead of
promoters, this will depend fundamentally on the ability of the diluter
(TiO2-P25 and Nb2O5) to interact with NiO.
With the aim of gaining additional information about the chemical
nature of nickel species, selected catalysts were analyzed by X-ray ab-
sorption spectroscopy (XAS). Specifically, we have focused our atten-
Fig. 6. XANES region of XAS spectra in the Ni K-edge of the most selective
tion into XANES features in XAS spectra, which are highly dependent on
(sample 92Ni-Nb-O sample) and diluted (20NiO/TiO2 sample) catalysts. For
the electronic properties of the absorbing atom. Fig. 6 displays XANES
comparison, unmodified NiO and metallic nickel are also presented.
region of the XAS spectra in the Ni K-edge of unmodified NiO (i.e un-
selective in the ODH of ethane) and 20NiO/TiO2 and 92Ni-Nb-O
7

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 7. TPR-H2 profiles of promoted (blue) and diluted (black) NiO-based catalysts (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article).
samples (i.e. the most selective catalysts in the ODH of ethane). For
comparison, the XAS spectrum corresponding to metallic Ni is also in-
cluded in this figure. A decrease in the “white line” intensity is observed
for both 20NiO/TiO2 and 92Ni-Nb-O samples, with respect to un-
selective NiO catalyst, likely due to a decrease in the average oxidation
state of Ni atoms in modified catalysts [49]. More interestingly, it can
be observed that the drop of the intensity is more drastic for promoted
92Ni-Nb-O than in the case of supported 20NiO/TiO2. This goes in line
with the previous observations, in which the modification of Ni species
in NiO highly depends on the synthetic procedure chosen, although
both diluters and promoters are able to improve the catalytic perfor-
mance of NiO in the ODH of ethane.
According to these results, we can conclude that: i) promoters
modify surface Ni sites in a higher extent than diluters, but; ii) both
promoters and diluters are capable of eliminating the electrophilic
oxygen species, which are responsible for total oxidation to carbon
oxides. The main difference among diluted and promoted NiO-based
catalysts is the amount of diluter/promoter necessary to achieve the
proper interaction with the active phase, which leads to highly efficient
materials in the ODH of ethane. Thus, especially due to the specific
synthesis conditions, in which a more homogeneous distribution of
Nb5+ and Ti4+ species can be assumed, the use of promoters requires
much lower promoter contents (ca. 10 wt.%, i.e. a NiO concentration in
the catalyst of ca. 90 wt.%), with respect to diluted catalysts, in which
the amount of diluter must be much higher (ca. 80wt.%, i.e. a NiO
content of ca. 20 wt. %) to achieve a comparable catalytic behavior in
the ODH of ethane (i.e. selectivity to ethylene of about 90%).
The effect of both promoters and diluters on the reducibility of NiO-
based catalysts was studied by means of temperature programmed re-
duction in H2 (TPR-H2). Fig. 7 compares TPR-H2 profiles of selected
promoted and diluted NiO catalysts. The samples show generally two
clear reduction peaks, which can be associated to the two-step reduc-
tion of Ni2+ sites: Ni2+ → Niδ+→ Ni0 [50]. It can be seen that both Nb-
and Ti-promoted catalysts substantially shift the maximum consump-
tion of hydrogen to higher temperatures, with respect to unmodified
NiO (Fig. 7, blue profiles).
Then, the reducibility of the catalysts seems to be related to the
nature of nickel rather than to the type of oxygen species. According to
XPS and XAS results, nickel species in diluted NiO catalysts seem to be
more similar to those in pure NiO. This way, the reducibility of those
sites will be equivalent in both cases. However, the elimination of
electrophilic oxygen species takes place by both synthetic strategies, i.e.
by promotion with high valence dopants (i.e. Nb5+ and Ti4+), or by the
use of the corresponding diluter oxides (i.e. TiO2-P25 or Nb2O5).
4. General remarks
During this work we have tried to shed some light into NiO-pro-
moter and NiO-support interactions in NiO catalysts for the ODH of
ethane. To cope with the problem, we have followed two synthetic
approaches, by which nickel oxide has been promoted with Ti and Nb,
or supported on TiO2 and Nb2O5.
A similar NiO-promoter interaction has been observed in the case of
Nb- and Ti-promoted catalysts. In this sense, XRD and Raman analyses
suggest the isomorphic substitution of the promoter for Ni in nickel
oxide framework, together with the formation of Ni-promoter mixed
oxide phases, like ilmenite-type structure in the case of Ti-promoted
NiO. A high interaction between nickel and both promoters can be also
deduced from XPS and XAS studies, which suggest a decrease in the
average oxidation state of nickel species and the elimination of elec-
trophilic oxygen species (responsible for total oxidation of ethane). This
leads to a decrease of the reducibility of the catalysts (as observed by
TPR-H2) and a drastic increase of the selectivity to ethylene (up to 90%)
in the ODH of ethane at low promoter contents (92 wt. % of NiO).
On the other hand, TiO2- and Nb2O5-diluted NiO materials show
differences concerning the active phase-diluter interactions.
Considering NiO/TiO2 catalysts, they show the maximum selectivity to
ethylene at low NiO-loading (i.e. high diluter contents). Then, a lower
interaction with TiO2 can be deduced with respect to Ti-promoted
materials, although an outstanding performance in the ODH of ethane
is achieved in both cases (ca. 90% selectivity to ethylene). This has been
confirmed by XAS and XPS experiments. In this way, TiO2 does not
substantially modify the chemical nature of Ni sublattice, but it is able
to eliminate the most electrophilic surface oxygen sites (non-selective
sites). Consequently, the reducibility of diluted materials does not de-
crease so drastically with respect to pure NiO as it has been observed in
Ti-promoted series.
This indicates a decrease in the reducibility of nickel species present
in promoted catalysts. This is considerably different in the case of TiO2-
P25 and Nb2O5-diluted materials, in which the reducibility does not
vary significantly, according to the small variations in the TPR-H2
profiles with respect to pure NiO (Fig. 7, black profiles).
8

D. Delgado, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Interestingly, Nb-diluted catalysts present low selectivity to ethy-
[12] H. Zhu, D.C. Rosenfeld, M. Harb, D.H. Anjum, M.N. Hedhili, S. Ould-Chikh, J.-
M. Basset, ACS Catal. 6 (2016) 2852–2866.
[13] B. Solsona, P. Concepción, S. Hernández, B. Demicol, J.M. López Nieto, Catal.
Today 180 (2012) 51–58.
[14] H. Zhu, H. Dong, P. Laveille, Y. Saih, V. Caps, J.-M. Basset, Catal. Today 228 (2014)
58–64.
[15] H. Zhu, D.C. Rosenfeld, D.H. Anjum, S.S. Sangaru, Y. Saih, S. Ould-Chikh, J. Catal.
329 (2015) 291–306.
lene in the ODH of ethane (48–68%). This behavior is observed re-
gardless of the amount of NiO loaded. Their catalytic performance can
be explained taking into consideration the low diluter-nickel interaction
achieved. XPS measurements show that Nb2O5 is not able to eliminate
non-selective electrophilic oxygen surface sites, associated to deep
oxidation reaction path. In addition, the presence of Nb2O5 does not
seem to modify the nature of surface Ni species.
[16] B. Solsona, P. Concepción, B. Demicol, S. Hernández, J.J. Delgado, J.J. Calvino,
J.M. López Nieto, J. Catal. 295 (2012) 104–114.
[17] D. Delgado, B. Solsona, A. Ykrelef, A. Rodríguez-Gómez, A. Caballero, E. Rodríguez-
Aguado, E. Rodríguez-Castellón, J.M. López Nieto, J. Phys. Chem. C 121 (2017)
25132–25142.
5. Conclusions
[18] Z. Skoufa, E. Heracleous, A.A. Lemonidou, J. Catal. 322 (2015) 118–129.
[19] B. Savova, S. Loridant, D. Filkova, J.M.M. Millet, Appl. Catal. A Gen. 390 (2010)
148–157.
[20] Z. Skoufa, E. Heracleous, A.A. Lemonidou, Catal. Today 192 (2012) 169–176.
[21] H. Zhu, S. Ould-Chikh, D.H. Anjum, M. Sun, G. Biausque, J.-M. Basset, V. Caps, J.
Catal. 285 (2012) 292–303.
[22] Z. Zhang, G. Zhao, R. Chai, J. Zhu, Y. Liu, Y. Lu, Catal. Sci. Technol. 8 (2018)
4383–4389.
[23] E. Heracleous, A.F. Lee, K. Wilson, A.A. Lemonidou, J. Catal. 231 (2005) 159–171.
[24] B. Solsona, P. Concepción, J.M. López Nieto, A. Dejoz, J.A. Cecilia, S. Agouram,
M.D. Soriano, V. Torres, J. Jiménez-Jiménez, E. Rodríguez Castellón, Catal. Sci.
Technol. 6 (2016) 3419–3429.
[25] R. Sanchis, D. Delgado, S. Agouram, M.D. Soriano, M.I. Vázquez, E. Rodríguez-
Castellón, B. Solsona, J.M. López Nieto, Appl. Catal. A Gen. 536 (2017) 18–26.
[26] Z. Zhang, J. Ding, R. Chai, G. Zhao, Y. Liu, Y. Lu, Appl. Catal. A Gen. 550 (2018)
151–159.
[27] E. Heracleous, A.A. Lemonidou, J. Catal. 270 (2010) 67–75.
[28] J.M. López Nieto, B. Solsona, R.K. Grasselli, P. Concepción, Top. Catal. 57 (2014)
1248–1255.
[29] D. Delgado, R. Sanchís, J.A. Cecilia, E. Rodríguez-Castellón, A. Caballero,
B. Solsona, J.M. López Nieto, Catal. Today 333 (2019) 10–16.
[30] C.A. Gärtner, A.C. van Veen, J.A. Lercher, ChemCatChem 5 (2013) 3196–3217.
[31] S.W. Han, D.H. Kim, M.-G. Jeong, K.J. Park, Y.D. Kim, Chem. Eng. J. 283 (2016)
992–998.
Highly selective NiO-based catalysts for the ODH of ethane have
been synthesized either by promoting NiO with Ti- or Nb- or by diluting
NiO with TiO2. However, Nb2O5-diluted NiO catalysts have not reached
a satisfactory catalytic performance.
The catalytic behavior of promoted NiO catalysts seems to be de-
pendent on the ability of the high valence dopants to be incorporated in
the pristine NiO lattice. Thus, just low amounts of promoters are needed
to eliminate non-selective active sites. On the other hand, in the case of
diluted NiO catalysts, the modification of NiO mainly depends on the
ability of the support to interact with nickel oxide particles. In this case,
TiO2 gives rise to a proper diluter-NiO interaction, although high TiO2
contents are required to achieve an optimal catalytic performance.
Conversely, Nb2O5 has not shown good properties as a NiO diluter,
being unable to eliminate a large proportion of non-selective sites.
Acknowledgements
The authors would like to acknowledge the DGICYT in Spain
(CRTl2018-099668-B-C21 and RTl2018-099668-B-C22 and MAT2017-
84118-C2-1-R projects). Authors from ITQ also thank Project SEV-2016-
0683 for supporting this research. D.D. thanks MINECO and Severo
Ochoa Excellence Program for his fellowship (SVP-2014-068669).
[32] M.-G. Jeong, I.H. Kim, S.W. Han, D.H. Kim, Y.D. Kim, J. Mol. Catal. A Chem. 414
(2016) 87–93.
[33] B. Ohtani, O.O. Prieto-Mahaney, D. Li, R. Abe, J. Photochem. Photobiol. A 216
(2010) 179–182.
[34] T. Murayama, J. Chen, J. Hirata, K. Matsumoto, W. Ueda, Catal. Sci. Technol. 4
(2014) 4250–4257.
[35] A. Fernández-Arroyo, D. Delgado, M.E. Domine, J.M. López Nieto, Catal. Sci.
Technol. 7 (2017) 5495–5499.
[36] R.E. Dietz, G.I. Parisot, A.E. Meixner, Phys. Rev. B 4 (1971) 2302–2310.
[37] J.M. Jehng, I.E. Wachs, Chem. Mater. 3 (1991) 100–107.
[38] X. Wang, J. Shen, Q. Pan, J. Raman Spectrosc. 42 (2011) 1578–1582.
[39] E. Rojas, J.J. Delgado, M.O. Guerrero-Pérez, M.A. Bañares, Catal. Sci. Technol. 3
(2013) 3173–3182.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at https://doi.org/10.1016/j.cattod.2019.06.063.
[40] L.J. Burcham, J. Datka, I.E. Wachs, J. Phys. Chem. B 103 (1999) 6015–6024.
[41] M.A. Ruiz Preciado, A. Kassiba, A. Morales-Acevedo, M. Makowska-Janusik, RSC
Adv. 5 (2015) 17396–17404.
References
[42] P. Salagre, J.L.G. Fierro, F. Medina, J.E. Sueiras, J. Mol. Catal. A Chem. 106 (1996)
125–134.
[1] A. Boulamanti, J.A. Moya, Renew. Sustain. Energy Rev. 68 (2017) 1205.
[2] T. Ren, M. Patel, K. Blok, Energy 31 (2006) 425–451.
[3] T. Ren, M.K. Patel, K. Blok, Energy 33 (2008) 817–833.
[4] F. Cavani, N. Ballarini, A. Cericola, Catal. Today 127 (2007) 113–131.
[5] J.M. López Nieto, B. Solsona, Gas phase heterogeneous partial oxidation reactions,
in: J.C. Védrine (Ed.), Metal Oxides in Heterogeneous Catalysis, Elsevier, 2018, pp.
211–286.
[43] J.C. Vedrine, G. Hollinger, D. Tran Minh, J. Phys. Chem. 82 (1978) 1515–1520.
[44] V. Biju, M. Abdul Khadar, J. Nanopart. Res. 4 (2002) 247–253.
[45] M.A. van Veenendaal, G.A. Sawatzky, Phys. Rev. Lett. 70 (1993) 2459–2462.
[46] V.V. Kaichev, V.I. Bukhtiyarov, M. Hävecker, A. Knop-Gercke, R.W. Mayer,
R. Schlögl, Kinet. Catal. 44 (2003) 432–440.
[47] R.K. Grasselli, Fundamental principles of selective heterogeneous oxidation cata-
lysis, Top. Catal. 21 (2002) 79–88.
[48] J. Haber, Mechanism of heterogeneous catalytic oxidation, in: R.A. Sheldon,
R.A. van Santen (Eds.), Catalytic Oxidation: Principles and Applications, World
scientific, 1995, pp. 17–51.
[6] J.M. Lopez Nieto, P. Botella, M.I. Vázquez, A. Dejoz, Chem. Commun. (2002)
1906–1907.
[7] T.T. Nguyen, B. Deniau, M. Baca, J.-M.M. Millet, Top. Catal. 59 (2016) 1496–1505.
[8] Y. Liu, US Patent US6355854 B1 (2001), assigned to Symyx Technologies Inc.
[9] E. Heracleous, A.A. Lemonidou, J. Catal. 237 (2006) 162–174.
[10] E. Heracleous, A.A. Lemonidou, J. Catal. 237 (2006) 175–189.
[11] B. Solsona, J.M. López Nieto, P. Concepción, A. Dejoz, F. Ivars, M.I. Vázquez, J.
Catal. 280 (2011) 28–39.
[49] J. Rabeah, J. Radnik, V. Briois, D. Maschmeyer, G. Stochniol, S. Peitz, H. Reeker,
C. La Fontaine, A. Brückner, ACS Catal. 6 (2016) 8224–8228.
[50] W. Shan, M. Luo, P. Ying, W. Shen, C. Li, Appl. Catal. A Gen. 246 (2003) 1–9.
9