The Reasons for Nonlinear Phenomena in Oxidation of Methane over Nickel

Article abstract of DOI:10.1134/S0023158418060149

Abstract: The catalytic oxidation of methane over nickel foil is studied. It is shown that, under the oxygen-lean conditions in the regime of a flow-type reactor, nonlinear phenomena can appear in the form of self-sustained oscillations of the reaction ra

Full text of DOI:10.1134/S0023158418060149

ISSN 0023-1584, Kinetics and Catalysis, 2018, Vol. 59, No. 6, pp. 810–819. © Pleiades Publishing, Ltd., 2018.
Original Russian Text © A.A. Saraev, Z.S. Vinokurov, A.N. Shmakov, V.V. Kaichev, V.I. Bukhtiyarov, 2018, published in Kinetika i Kataliz, 2018, Vol. 59, No. 6, pp. 776–786.
The Reasons for Nonlinear Phenomena
in Oxidation of Methane over Nickel
A. A. Saraev , Z. S. Vinokurov , A. N. Shmakov , V. V. Kaichev *, and V. I. Bukhtiyarova, b
a, b
a, c
a, b, c
a, b,
a
Boreskov Institute of Catalysis, Siberian Branch, Russian Academy of Sciences, Novosibirsk, 630090 Russia
b
Novosibirsk State University, Novosibirsk, 630090 Russia
c
Budker Institute of Nuclear Physics, Siberian Branch, Russian Academy of Sciences, Novosibirsk, 630090 Russia
*
e-mail: vvk@catalysis.ru
Received April 20, 2018
Abstract—The catalytic oxidation of methane over nickel foil is studied. It is shown that, under the oxygen-
lean conditions in the regime of a flow-type reactor, nonlinear phenomena can appear in the form of self-
sustained oscillations of the reaction rate and the catalyst temperature. To determine the reasons for self-sus-
tained oscillations, X-ray diffraction and mass spectrometry in the operandо mode were used. It was found
that the appearance of oscillations in the methane oxidation is due to periodical oxidation–reduction of the
surface layer of nickel foil; metallic nickel has a higher catalytic activity than NiO. The oscillations of the cat-
alyst temperature are determined by the occurrence of exothermic and endothermic processes associated with
the reduction of nickel oxide and methane oxidation on the surface of metallic nickel.
Keywords: methane oxidation, nickel, heterogeneous catalysis, self-oscillations, reaction mechanism
DOI: 10.1134/S0023158418060149
INTRODUCTION
troscopy (XPS) and mass spectrometry of self-sus-
tained oscillations in the oxidation of propane over
nickel [8, 11, 14]. It was shown that oscillations in pro-
pane oxidation are caused by periodic oxidation and
reduction of nickel. In the state of high activity, nickel
is metallic; transition to a low-active state is accompa-
nied by the formation of a nickel oxide layer. The
activity of NiO is an order of magnitude lower than the
activity of metallic nickel.
Nonlinear phenomena in heterogeneous catalysis
have been actively studied over the several past
decades [1–3]. A distinctive feature of these phenom-
ena is that under certain conditions catalytic processes
are described by nonlinear equations. Nonlinearity is
determined by the interaction of species at the
atomic–molecular level, as well as by the influence of
temperature and mass and heat transfer processes on
the rate of a chemical reaction. The most typical
In this work, we present the results of our studies of
examples of nonlinear phenomena are temperature nonlinear processes observed in the catalytic oxidation
hysteresis [4, 5], regular and damped self-sustained of methane over nickel. As was shown earlier, under
oscillations [6–11], chemical waves [12–14] and oxygen-lean conditions at ~700°C in a flow-type
deterministic chaos [15]. In all of these cases, unusual reactor, self-sustained oscillations in methane oxida-
dependences of the main parameter of a catalytic sys- tion may occur, which are accompanied by oscillations
tem—the reaction rate—on time and/or temperature of the catalyst temperature [6, 16]. The oscillations of
are observed.
the partial pressure of products and reactants at the
reactor outlet are relaxational; methane conversion
ranges from 3 to 50%; and oxygen conversion may reach
The study of nonlinear phenomena is of interest
not only from the fundamental, but also from the
practical point of view. Particular attention is given to
the study of relaxation self-sustained oscillations,
since it is possible to determine the chemical and
9
5%. The oscillations of the catalyst temperature are
complex, and their amplitude reaches 80°C [16].
To find out the reasons for the oscillatory behavior,
phase composition of the catalyst in the operando we carried out an operandо study using X-ray diffrac-
mode in the transition between states with high and tion (XRD) and mass spectrometry. It should be noted
low activities, and thus study the processes of catalyst that, unlike XPS, which makes it possible to determine
activation and deactivation and identify the reasons for the chemical composition of the surface of the func-
the high activity of catalysts and predict their behavior tioning catalyst at a pressure below 10 mbar [17], XRD
in industrial processes. For instance, recently we have can be used in in situ study at atmospheric pressure
performed a study using X-ray photoelectron spec- [18, 19]. The use of laboratory sources of X-ray radia-
810

THE REASONS FOR NONLINEAR PHENOMENA
811
tion does not allow one to carry out studies with a high monochromator and a radiation collimation system, a
temporal resolution. For operandо XRD studies of radiation detection system diffracted by the sample, an
self-sustained oscillations, it makes sense to use syn- X-ray chamber (in which a sample is placed and vari-
chrotron radiation, the intensity of which is much ous conditions can be created), as well as auxiliary
higher than the intensity of X-ray radiation of labora- equipment. The station is assembled on channel #6 of
tory sources. This makes it possible to increase the sig- the synchrotron radiation output of the VEPP-3 elec-
nal/noise ratio, decrease the time of signal acquisi- tron storage ring (Institute of Nuclear Physics, Sibe-
tion, and thus improve the temporal resolution of the rian Branch, Russian Academy of Sciences, Novosi-
experiment. In the course of experiments, we studied birsk). Three types of crystals are used at the station:
the phase composition of the catalyst by XRD with Ge(111), Si(111), and Si(220). The choice of working
online analysis of the gas phase using mass spectrom- energy was determined by the experimental condi-
etry. The period of oscillations, which was ~3 min, tions, the possible values of energies (wavelengths) of
made it possible to record the diffraction patterns of the radiation were 7.162 keV (0.1731 nm) for Ge(111),
the catalyst in the states of low and high activity. Spe- 7.551 keV (0.1642 nm) for Si(111) and 12.273 keV
cial attention was paid to the processes occurring (0.1010 nm) for Si(220). In this work, a Ge(111) single
during the induction period, which preceded the crystal was used.
appearance of regular self-sustained oscillations, and
Figure 1 shows the schematic of the station. To
to the analysis of catalyst temperature oscillations. As
obtain the diffraction patterns, a one-coordinate, par-
a result, we have found the reasons for the appearance
allax-free position-sensitive detector OD-3M-350 was
of oscillations in methane oxidation over nickel and
used (Institute of Nuclear Physics, Siberian Branch,
shown that the activity of nickel in the metallic state is
Russian Academy of Sciences, Novosibirsk, Russia)
several times higher than that of NiO.
[
20], which is designed for rapid detection of X-rays in
experiments with a high temporal resolution. The
detector makes it possible to obtain diffraction pat-
terns at ~30° with a step of ~1/100, and the minimal
time for obtaining one diffraction pattern is 1 s. The
detector was mounted on a HUBER-480 precision
goniometer (HUBER Diffraktions technik, Ger-
many) and can be positioned against the 2θ angle as
required by the specific experimental conditions.
EXPERIMENTAL
Kinetic experiments for the study of the transition
of the system from the steady-state methane oxidation
to the oscillatory mode were carried out in an experi-
mental setup consisting of a flow-type reactor, a gas
supply system, and a SRS UGA-100 quadrupole mass
spectrometer (Stanford Research Systems Inc., US).
The detailed description of the setup was given else-
where [16]. The reaction mixture consisted of argon,
methane, and oxygen. The flow rate and the composi-
tion of the reaction mixture were set using a Horiba
SEC-Z500 mass flow controller (Horiba Ltd., Japan).
The pressure of the gas mixture in all the experiments
The diffractometer is equipped with a gas supply
system based on Smart-Trak-50 gas flow controllers
(
Sierra Instruments Inc., US), an XRK-900 high tem-
perature reaction chamber (Anton Paar GmbH,
Austria), and a SRS UGA-100 mass spectrometer for
analyzing the composition of the gas phase. Figure 2
shows the schematic of a setup for in situ/operandо
XRD experiments. In in situ experiments, nickel foil
2
was 1 bar. Nickel foil with a size of 8 × 4 mm and a
thickness of 0.125 mm was used as a catalyst (Advent
Research Materials Ltd., UK, purity 99.9%). The cat-
alyst temperature was measured by a chromel–alumel
thermocouple spot-welded to the foil. The composi-
tion of a gas phase was analyzed using a mass spec-
trometer; the gas phase was sampled at the reactor out-
let by means of a heated stainless-steel capillary with a
length of 1 m and an inner diameter of 175 μm.
2
with a size of 10 × 4 mm was used. The gas flows
(CH , O , and N ) and, correspondingly, the compo-
4 2 2
sition of the reaction mixture in the chamber/reactor
were set using calibrated gas flow controllers. The
pressure inside the reaction chamber in the course of
experiments was 1 bar. The catalyst temperature was
measured with a chromel–alumel thermocouple spot-
welded directly to the foil. Diffraction patterns were
analyzed using Fityk 1.2.9 software package [21].
The operandо experiments on the study of the
phase composition of nickel foil in the process of
methane oxidation in the oscillation regime was car-
ried out at the Siberian Center for Synchrotron and
Terahertz Radiation at the Precision Diffractometry
Station. This station is designed to study the structure
RESULTS AND DISCUSSION
It has been shown earlier that, in the oxidation of
and structural transformations in a solid matter under methane over nickel under oxygen-lean conditions in
the influence of various external conditions using the temperature range 650–900°C, self-sustained
X-ray diffraction methods on monochromatized syn- oscillations of the relaxation type are observed [16,
chrotron radiation. Because of the polarization fea- 22–24]. The period and shape of these oscillations
tures of synchrotron radiation, the diffraction plane is depend on the temperature, the molar ratio of reac-
chosen perpendicular to the polarization plane of the tants, the form of the catalyst, and on the reactor
radiation, that is, vertical. The station includes a design. It was found that self-sustained oscillations of
KINETICS AND CATALYSIS Vol. 59 No. 6 2018

8
12
SARAEV et al.
Position-sensitive
detector
OD-3M-350
Sample
Four-channel scheme
in the high-resolution
mode
Input collimator
Scintillation
detectors
Crystals/
analyzers
Ge(111) monochromator
Position-sensitive
ionization camera
e
Fig. 1. Schematic of the station “Precision Diffractometry” on the channel #6 output of the electron storage ring VEPP-3 with a
position-sensitive detector.
Heating unit
Cooling
system
Absolute pressure
sensor
Temperature
sensor
XRK900
FMC
FMC
FMC
Sample
CHx
Capillary
heating
O2
Mass
spectrometer
Pump
Ar/He
Fig. 2. Schematic of the experimental setup based on the reaction chamber XRK-900 for carrying in situ/operandо studies of the
phase composition of catalysts in different reaction media. GFC is the gas flow controller.
the methane oxidation rate occur in the ranges of temperature of ~680°C. Figure 3a shows the results of
molar ratios CH : O from (2 : 1) to (19 : 1) [16]. This the study of system transition from methane steady-
4
2
process is accompanied by the oscillations of the cata- state oxidation to oscillations. In this experiment,
lyst temperature, the amplitude of which may reach
fresh nickel foil was placed into the flow-type reactor,
80°С. The transition of the system into an oscillatory
to which the reaction mixture Ar : CH : O = 9 : 10 : 1
4
2
mode is preceded by an induction period, the duration
of which is from several minutes to several hours
depending on the catalyst temperature, the molar ratio
of reactants and the flow rate of the reaction mixture
was supplied. Then, the reactor was gradually heated
to a temperature of ~720°C. The catalyst temperature
was ~690°C. On the plots of changes in mass spectro-
metric signals corresponding to the products of total
and partial oxidation of methane (H O, CO , H , and
[
16, 22].
2
2
2
The induction period is observed only when fresh
CO), one can see that an increase in the catalyst tem-
nickel foil is used as a catalyst. On the “activated”
nickel foil (which was previously used as a catalyst for
methane oxidation in the oscillatory mode), the oscil-
perature above 420°C leads to a gradual growth of H O
2
and CO signals, which corresponds to the total oxida-
2
lations occur immediately after heating the system to a tion of methane. Upon a further increase in the cata-
KINETICS AND CATALYSIS
Vol. 59
No. 6
2018

THE REASONS FOR NONLINEAR PHENOMENA
а)
813
(
(b)
T, °C
50
T, °C
10
T, °C
7
740
7
600
450
300
720
700
680
660
700
690
680
150
0
670
Mass spectrometric signals, arb. units
00
Mass spectrometric signals, arb. units
00
1
H2/10 + 80
2
CH4 + 130
9
0
0
0
0
0
0
0
0
8
7
6
5
4
3
2
150
2
O + 130
4
4 min
CO/2 + 45
H2/10 + 90
CO + 50
100
CO2 + 10
H2O
50
CO2 + 20
H2O
10
0
500 1000 3250
3500
3750
4000
Time, s
0
500 1000 1500 2000 2500
Time, s
Fig. 3. Oscillations of mass-spectrometric signals of reaction products and reactants and oscillations of the catalyst temperature
observed in the course of methane oxidation on nickel: (a) The process of system transfer from the steady-state methane oxidation
to the oscillatory mode at the following molar ratio of reactants: CH : O = 10 : 1. (b) Sustainable self-oscillations at the following
4
2
molar ratio CH : O = 3 : 1. Mass spectrometric signals of reactants and products of the reaction are vertically shifted to the values
4
2
designated by numbers.
lyst temperature to 690°C, the conversion of oxygen and this pointed to the irreversible change in the sur-
reached 20%, and the conversion of methane was 2–3%. face morphology or chemical and phase composition
of the catalyst. We did not detect any change in the
The first peak of the products of total and partial
composition of the catalyst by XRD and XPS during
oxidation of methane (as well as a significant change
the induction period. We showed in [16] using scan-
ning electron microscopy that, during the oxidation of
methane in the oscillatory mode, the morphology of
the surface layer of nickel foil substantially changed. A
porous structure with a developed surface area is
formed on the surface of the nickel foil under reaction
in the temperature of the catalyst) appeared after
~
44 min. Then, during ~20 min, irregular oscillations
with different amplitudes were observed, and after the
end of the induction period, stable, regular oscillations
were set in the system.
The observed distribution of methane oxidation conditions. A similar change in the morphology of the
products indicates that, during the induction period, surface layer of the catalyst was observed in the oxida-
the main reaction on the surface of the catalyst is the tion of propane over nickel foil, also taking place in an
total oxidation of methane. After the transition of the oscillatory mode [11]. It is obvious that the appear-
system to the oscillatory mode, in the high-active state ance of the porous structure is a consequence of peri-
the total and partial oxidation of methane occurs on odic oxidation and reduction of nickel (the molar vol-
the catalyst surface, and in the low-active state only umes of NiO and metallic nickel are 11.2 and
the total oxidation of methane takes place. In the high-
active state, the oxygen conversion reaches 97%, while
in the low-active state it reaches ~50%. As noted
3
6
.6 cm /mol, respectively).
Figure 3b presents typical oscillations of mass spec-
above, when using nickel foil as a catalyst “activated” trometric signals corresponding to the partial pressure
as described above, there was no any induction period, of products and reactants at the reactor outlet,
KINETICS AND CATALYSIS Vol. 59 No. 6 2018

8
14
SARAEV et al.
Thus, it can be assumed that a sharp increase in the
observed in the oxidation of methane over the catalyst
with developed surface morphology. It is seen that catalyst temperature and a similar sharp decrease in its
self-sustained oscillations take the form of relaxation transition from the low-active state to a high-activity
oscillations, and that part of the time the catalyst is in state is determined by the reduction of highly defective
the state of low activity and periodically, sponta- NiO and then stoichiometric NiO. The subsequent
neously passes into the state of high activity. At a molar slow growth of the catalyst temperature is determined
ratio Ar : CH : O of 16 : 3 : 1, the conversion of meth- by the occurrence of exothermic reactions of complete
4
2
ane, for instance, oscillates in the range from 5 to 40%, (III) and partial (IV) oxidation of methane.
and the conversion of oxygen oscillates in the range
CH + 2O → CO + 2H O,
4
2
2
2
from 12 to 95%. In this state of high activity, the con-
version of oxygen is 70–95% with the formation of
products of the total and partial oxidation of methane
(
III)
0
ΔH1 = −800.6 kJ mol ,
000
[
16]. Simultaneously with the oscillations of the par-
CH
4
+ 0.5O
2
→ CO + 2H
2
,
(
IV)
tial pressures of the reactants and products, the tem-
perature of the catalyst also oscillates (Fig. 3b). Since
the heating element of the flow-type reactor operates
0
ΔH1 = −21.8 kJ mol.
000
As can be seen, a gradual decrease in the conver-
in a constant power mode, the temperature oscilla- sion of methane and oxygen is observed at this stage.
tions cannot be due to uneven heating. Moreover, the Such an effect can be caused by the gradual oxidation
amplitude of temperature oscillations is substantial of nickel with the formation of NiO clusters on the cat-
and may vary in a range of 0–80°C [16].
alyst surface. Recently, a similar mechanism of cata-
lyst deactivation in the oxidation of propylene on pal-
ladium was described by Kaichev et al. [5]. It should
be noted that nickel oxidation (V) is also exothermic.
The entire surface of nickel foil is gradually covered
with a NiO layer, and the catalyst passes into the low-
activity state. As a result, the catalyst temperature falls
exponentially, and this is accompanied by an increase
in the oxygen partial pressure.
As can be seen from Fig. 3b, the profile of tempera-
ture oscillations of the catalyst has a complex shape,
which points to the occurrence of both exothermic
and endothermic processes. Transition from the low-
active state to the high-active state is accompanied by
a sharp increase in the catalyst temperature and the
subsequent equally sharp decrease in this temperature.
Such a behavior indicates that an exothermic process
first occurs and is followed by an endothermic pro-
cess. Earlier, it has been assumed that the increase in
the catalyst temperature was determined by the com-
bustion of carbon deposits during nickel transition
from the oxidized state to the metallic state [16]. How-
ever, the formation of carbon in a noticeable amount
on the oxidized surface is unlikely. Moreover, no car-
bon-containing fragments were detected by XPS in
the nickel surface during the self-sustained oscillations
in propane oxidation [11]. We believe that a sharp
decrease in the catalyst temperature together with an
N i + 0.5O → N iO,
2
(
V)
0
ΔH1
= −235.1 kJ m ol .
000
It should be noted that routine ex situ studies do not
allow us to directly identify the reasons for self-sus-
tained oscillations. Under the conditions of oscillatory
process, the catalytic system is far from thermody-
namic equilibrium, and any change in the catalyst
conditions will lead to a change in its state. Thus, for
example, if oxygen is removed from the reaction mix-
ture, the catalyst surface will be carbonized, and if
methane is removed, nickel will undoubtedly turn into
the oxidized state. Both of these states have low activ-
ity in hydrocarbon oxidation reactions. In an ex situ
study, it is necessary to stop the reaction, withdraw the
catalyst from the reactor, and transfer the catalyst to an
analytical instrument. Consequently, upon cooling in
the reaction mixture, the catalyst will turn into one of
I) the above states depending on the ratio of the reac-
increase in the partial pressures of CO and CO are
2
due to the reduction of nickel oxide. Indeed, accord-
ing to thermodynamic calculations, the reduction of
stoichiometric nickel oxide by methane to form the
products of partial (I) and total (II) oxidation is endo-
thermic:
NiO + CH → Ni + CO + 2H ,
4
2
(
0
ΔH1
000
= +213.3 kJ mol,
tants. On the contrary, with increasing temperature,
the probability of reduction of nickel oxide and com-
4
NiO + CH → 4Ni + CO + 2H O,
4 2 2
bustion of carbon deposits increases, but this can be
(
II)
0
ΔH1 = +139.7 kJ mol.
detected only in in situ studies.
000
To obtain a definitive answer to the question of the
reason for a periodic change in the catalyst activity, an
operando XRD study was carried out and consisted of
two parts:
However, if nickel oxide contains a lot of defects,
then the reduction processes become endothermic.
According to quantum chemical calculations, the
energy of formation of a point defect of the Ni vacancy
type in the NiO lattice is 236 kJ/mol [25]. The forma-
(1) operandо XRD study of the phase composition
tion of defects in the structure of NiO during self- of the catalyst surface during system transition from
oscillations was observed in propane oxidation [11].
the low-active state to the high-active state and
KINETICS AND CATALYSIS
Vol. 59
No. 6
2018

THE REASONS FOR NONLINEAR PHENOMENA
2) operandо XRD study of the phase composition
T, °C
815
(
of the catalyst surface during the methane oxidation
over nickel foil in the oscillatory mode.
6
6
50
40
High
activity
state
Most researchers agree that the low activity corre-
sponds to the oxidized state of nickel and high activity
corresponds to its metallic state. To confirm this
assumption, we carried out operandо XRD studies of
the process of system transition from the low-active
state to the high-active state. For the oxidation of
nickel foil, a mixture of oxygen and argon with flow
630
620
Low
activity
state
6
10
00
90
Reduction
of surface
oxide
6
5
Mass spectrometric signals, arb. units
3
rates of 20 and 40 cm /min, respectively, were fed into
9
0
an X-ray diffraction chamber. Fresh nickel foil was
used in this experiment, and the oxidation was carried
out at a temperature of ~600°C. Upon the total oxida-
tion of the foil surface (as confirmed by diffraction
peaks in the diffraction pattern corresponding to
NiO), the oxygen flow was discontinued. The residual
amounts of oxygen were removed from the chamber by
purging with argon. Then, a mixture of reactants with
80
CH
4
2
+ 50
+ 60
70
O
6
0
0
5
2
H /10 + 30
a molar ratio CH : O of 2 : 1 was supplied to the
4
2
40
30
20
chamber. The flow rates of methane, oxygen, and
CO + 15
3
argon were 20, 10, and 40 cm /min, respectively.
Figure 4 shows the plots of changes in mass spec-
trometric signals corresponding to the products of the
total and partial oxidation of methane (CO , H , and
CO
2
10
2
2
CO) and the catalyst temperature. It is seen that at a
catalyst temperature of 595°C, the system is in the
low-active state, and the formation of the products of
the total and partial oxidation of methane in small
amounts is observed. Then, the system was gradually
heated to ~640°C, and a gradual increase in the con-
versions of oxygen and methane with an increase in
0
2
50 500 750 1000 1250 1500 1750
Time, s
Fig. 4. Changes in the temperature of nickel foil and mass
spectrometric signals of reactants and products obtained in
the process of system transition from the low-activity state
to the high-activity state at a molar ratio of reactants of
the CO were observed. After a while, the catalyst
2
turned into the high-active state, as was evident from a
р(СН ) : р(О ) = 2 : 1. Mass spectrometric signals of reac-
4
2
sharp increase in the conversions of oxygen and meth-
tants and products of the reaction are vertically shifted to
the values designated by numbers.
ane and an increase in the signals from H and CO.
2
The CO signal decreased, which pointed to the fact
2
that the main reaction in the high-active state is the
partial oxidation of methane. A stepwise change of the
catalyst temperature that preceded system transition to
the high-active state is related to the reduction of
nickel oxide. Indeed, diffraction patterns obtained in
the low-active state and in the process of heating the
system to 640°C, points to the fact that the surface of
nickel foil is covered by nickel oxide. However, diffrac-
tion patterns obtained during the stepwise change of
catalyst temperature, point to the reduction of nickel
oxide.
Figure 5 shows the diffraction patterns of nickel foil
surface obtained in the process of system transition
from the low-active state to the high-active state.
Before the stepwise change of the catalyst tempera-
ture, four diffraction peaks are present on the diffrac-
tion pattern. The peaks at 41.4° and 48.2° refer to the
and (2, 0, 0). During system transition from the low-
active state to the high-active state, the intensity of dif-
fraction peaks corresponding to nickel oxide
decreases. At the same time, the intensity of the dif-
fraction peaks corresponding to metallic nickel
increases. Therefore, the system transition to a high-
active state is associated with the reduction of NiO.
The second part of the operandо XRD experiments
was devoted to the study of the phase composition of
the catalyst during the oscillatory methane oxidation.
To shorten the duration of experiments, preliminarily
activated nickel foil was used. It is important that,
despite the differences in the design of the XRD dif-
fraction chamber and the flow-type reactor used to
study the kinetics of methane oxidation, self-sustained
oscillations in the methane oxidation were observed in
the chamber as well. In the operandо XRD experi-
ments, we used the reaction mixture with N : CH : O
diffraction pattern of NiO with
2, 0, 0). The peaks at 49.5° and 57.9° refer to the dif-
h,k,l
fraction pattern of metallic nickel with = (1, 1, 1) 100 cm /min. The oscillations of the partial pressures
( )
h,k,l
= (1, 1, 1) and
2
4
2
(
ratios of 29 : 10 : 1. The overall flow of gases was
3
(
)
KINETICS AND CATALYSIS Vol. 59 No. 6 2018

8
16
SARAEV et al.
Intensity, arb. units
NiO(111)
Ni(111)
700
600
500
400
300
200
Ni(200)
NiO(200)
100
0
2
0
4
0
6
0
8
0
1
00
20
40
1
1
4
0 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
2θ, deg
Fig. 5. Diffraction patterns of the surface of nickel foil obtained in the process of system transition from the low low-activity state
to the high-activity state.
of reactants and products of the oxidation of methane metallic state occurs. The results obtained indicate
appeared at a catalyst temperature of ~715°C with a that the oscillations in the rate of methane oxidation
catalyst temperature oscillation amplitude of ~20°C. over nickel are determined by periodic oxidation and
The period of oscillations was ~170 s, which made it reduction of the catalyst surface. Taking into account
possible to obtain diffraction patterns of the surface at the results of our previous studies [11, 16], it can be
different moments of time with an acceptable sig- stated that the rate oscillations in methane and pro-
nal/noise ratio. The time for obtaining one diffraction pane oxidation are determined by the periodic transi-
patterns at 2θ from 38° to 48° was 4 s.
tion of nickel from oxidized to metallic state and back.
The high-active state of nickel is metallic, and the
transition to the low-active state is accompanied by
the formation of a layer of nickel oxide on the catalyst
surface. Despite the significant difference in the pres-
sures of the reaction mixture (atmospheric pressure in
methane oxidation and ~1 mbar in propane oxida-
tion), in both cases nickel is most active in the metallic
state (at least under oxygen-lean conditions). The
main reaction route on the metallic surface of nickel is
the partial oxidation of methane and propane to form
Figure 6 shows diffraction patterns of the surface
obtained at different moments of time in the course of
self-sustained oscillations in methane oxidation. The
diffraction patterns show reflexes, corresponding to
metallic nickel and NiO. The inset also includes the
plots of catalyst temperature changes and mass spec-
trometric signals of the reactants and reaction prod-
ucts obtained simultaneously with the recording of
diffraction patterns. The intensity of reflexes corre-
sponding to metallic nickel and nickel oxide oscillates
simultaneously with the partial pressures of reactants
and reaction products and with the catalyst tempera-
ture. The change in the intensity of the NiO(111) and
NiO(200) reflexes occurs in antiphase with respect to
a change in the intensity of the Ni(111) reflex.
CO and Н .
2
Similar conclusions have been drawn in the works
devoted to the study of self-sustained oscillations in
methane oxidation over palladium catalysts. For
example, in the recent work [26] devoted to methane
oxidation on a Pd/Al O supported palladium catalyst,
Figure 7 shows characteristic diffraction patterns of
2
3
the surface of nickel foil obtained in the state of low it was shown by in situ methods of thermogravimetry
and high activity of the catalyst. The intense NiO(111) and mass spectrometry that metallic palladium has the
and NiO(200) reflexes are observed when the system highest activity in the total oxidation of methane. The
in the low-active state. The transition to the high- oxidation of metallic palladium leads to a sharp
active state is accompanied by an increase in the inten- decrease in the catalytic activity. Similar conclusions
sity of the Ni(111) reflex and a simultaneous decrease have been drawn in [27] where self-sustained oscilla-
in the intensity of NiO(111) and NiO(200) reflexes to tions were studied in methane oxidation over a palla-
zero. That is, complete reduction of nickel oxide to dium catalyst using operandо X-ray absorption spec-
KINETICS AND CATALYSIS
Vol. 59
No. 6
2018

THE REASONS FOR NONLINEAR PHENOMENA
а)
817
T, °C
20
10
00
(
7
7
7
Mass spectrometric signals,
In situ XRD
arb. units
CH4 – 20
1
8
1
4
O2 + 10
1
0
H2/10 + 6
6
2
CO2 + 3
H2O
Ni(111)
0
100 200 300 400 500 600 700 800 9001000
Time, s
(b)
NiO(200)
NiO(111)
Time, s
550
650
750
850
38
39
40
41
42
43
44
45
46
47
48
2θ, deg
Fig. 6. The plots of catalyst temperature changes and mass spectrometric signals of reactants and products of the reaction (a), and
the corresponding diffraction patterns of the catalyst surface (b) in the course of methane oxidation reaction. Mass spectrometric
signals of reactants and products of the reaction are vertically shifted to the values designated by numbers.
troscopy with online mass spectrometry. The authors coordinatively unsaturated atoms on the facets of
observed hydrogen formation and oxygen absorption nickel oxide formed under the reaction conditions.
in the gas phase when palladium was predominantly in
the metallic state. On the contrary, when palladium
was oxidized to PdO, the partial pressure of hydrogen
was close to zero.
The indicated contradictions require new approaches
to the study of the mechanisms of catalytic reactions.
The operandо method that combines catalyst charac-
terization (XPS, X-ray absorption spectroscopy,
These results contradict some works in the field of XRD) and the measurement of the catalytic activity
oxidative catalysis. Platinum group metal oxides, such using mass spectrometry or gas chromatography
as RuO , PdO, and IrO , are believed to be more active applied to the oscillating reactions makes it possible to
2
2
in the catalytic oxidation of hydrocarbons than the determine the state of a catalyst when it shows high
corresponding metals [28]. For example, the oxidation
of methane over Pd(100) was studied in [29] using
XPS at a pressure of the reaction mixture of several
millibars. It turned out that the greatest activity was
observed when a two-layer oxide film PdO(101) was
formed on the palladium surface. Using quantum
chemical calculations Martin and co-workers [29]
showed that the presence of oxygen above a coordina-
tively unsaturated palladium in the two-layer structure
of PdO(101) film is crucial for efficient methane dis-
sociation.
Ni(200)
Ni(111)
NiO(200)
45
NiO(111)
2
From this point of view, it can be assumed that the
reason for the low activity of nickel oxide is the
absence of coordinatively unsaturated nickel atoms in
the surface layer of NiO. However, the surface struc-
ture of nickel oxide formed under reaction conditions
has not yet been determined, which does not allow us
to say unambiguously that this is the reason for the
lower activity of nickel oxide compared to metallic
nickel. A possible reason may be deficiency of the
1
3
5
40
50
55
60
2θ, deg
Fig. 7. Diffraction patterns of the catalyst surface obtained
in the states of (1) low and (2) high activity.
KINETICS AND CATALYSIS Vol. 59 No. 6 2018

8
18
SARAEV et al.
and low activity, to determine the reasons for the tran- Shared Use Center Siberian Center for Synchrotron and
sitions between these states, and thus determine the Terahertz Radiation at Budker Institute of Nuclear
mechanism of the reaction under study.
Physics, Siberian Branch, Russian Academy of Sci-
ences, Novosibirsk.
Taking into account the results of the operando
XRD experiments, as well as the results of our previous
studies, it can be stated that a nickel-based catalyst
undergoes reversible changes under the reaction con-
ditions that affect the catalytic activity and can cause
the transition to an oscillatory mode. These changes
are determined by a change in the chemistry and mor-
phology of the surface layers of the catalyst. Oscilla-
tions in the reaction rate are due to periodic transitions
between metallic nickel and nickel oxide, which can
occur via the formation of intermediate states such as
a solid solution of oxygen in metallic nickel or highly
defective nickel oxide [11]. The rates of methane [26]
and propane [8, 11] oxidation over metallic nickel are
REFERENCES
1. Ertl, G., Adv. Catal., 1990, vol. 37, p. 213.
2. Slinko, M.M. and Jaeger, N.I., Oscillating Heteroge-
neous Catalytic Systems, Elsevier, 1994, vol. 86, p. 1.
3. Imbihl, R. and Ertl, G., Chem. Rev., 1995, vol. 95,
p. 697.
4. Matveev, A.V., Kaichev, V.V., Saraev, A.A., Goro-
detskii, V.V., Knop-Gericke, A., Bukhtiyarov, V.I., and
Nieuwenhuys, B.E., Catal. Today, 2015, vol. 244, p. 29.
5. Kaichev, V.V., Saraev, A.A., Matveev, A.V., Dubinin, Y.V.,
Knop-Gericke, A., and Bukhtiyarov, V.I., J. Phys.
Chem. C, 2018, vol. 122, p. 4315.
1
0–20 times higher than on NiO. On the metallic sur-
face, the oxidation occurs according to the Lang-
muir–Hinshelwood mechanism, and on the oxide
surface, oxidation occurs according to the Mars–van
Krevelen redox mechanism. In the first case, the reac-
tant molecules are adsorbed and dissociate to form
adsorbed atoms C, H, and O, which further react to
form the reaction products. The dissolution of oxygen
atoms in the surface layers of nickel is possible, and
this can lead to a decrease in activity and selectivity to
6. Zhang, X., Lee, C., Hayward, D., and Mingos, D.,
Catal. Today, 2005, vol. 105, p., 283.
7. Hendriksen, B.L.M., Ackermann, M.D., van Rijn, R.,
Stoltz, D., Popa, I., Balmes, O., Resta, A., Wermeille, D.,
Felici, R., Ferrer, S., and Frenken, J.W.M., Nat.
Chem., 2010, vol. 2, p. 730.
8. Kaichev, V.V., Gladky, A.Y., Prosvirin, I.P., Saraev, A.A.,
Hävecker, M., Knop-Gericke, A., Schlögl, R., and
Bukhtiyarov, V.I., Surf. Sci., 2013, vol. 609, p. 113.
9. Vendelbo, S.B., Elkjær, C.F., Falsig, H., Puspitasari, I.,
Dona, P., Mele, L., Morana, B., Nelissen, B.J., van
Rijn, R., Creemer, J.F., Kooyman, P.J., and Helveg, S.,
Nat. Mater., 2014, vol. 13, p. 884.
CO and H (Fig. 1b). When a certain concentration of
2
dissolved oxygen is reached, the rapid formation of
nickel oxide is possible, which leads to the transition of
the catalyst to the low-active state.
1
0. Bychkov, V.Y., Tyulenin, Y.P., Gorenberg, A.Y.,
Sokolov, S., and Korchak, V.N., Appl. Catal., A, 2014,
vol. 485, p. 1.
CONCLUSIONS
1
1. Kaichev, V.V., Teschner, D., Saraev, A.A., Kosolobov, S.S.,
Gladky, A.Y., Prosvirin, I.P., Rudina, N.A., Ayupov, A.B.,
Blume, R., Hävecker, M., Knop-Gericke, A., Schlögl, R.,
Latyshev, A.V., and Bukhtiyarov, V.I., J. Catal., 2016,
vol. 334, p. 23.
This example demonstrates the capabilities of a
study of nonlinear phenomena in heterogeneous cata-
lytic systems by X-ray diffraction using synchrotron
radiation. Simultaneous use of X-ray diffraction and
mass spectrometry (the operandо mode) makes it pos-
sible to study the state of a catalyst and its activity in
the reaction in one experiment. In the study of self-
sustained oscillations, the advantage of synchrotron
radiation is the high brightness of radiation, which
makes it possible to determine the state of the catalyst
at various instants of oscillations, i.e. determine the
phase composition of the catalyst in the states of both
high and low activity. In the case of methane oxidation
over nickel, it was possible to show that metallic nickel
is more active than NiO. Self-sustained oscillations in
the rate of methane oxidation occur as a result of peri-
odic oxidation and reduction of nickel.
1
2. Gorodetskii, V., Lauterbach, J., Rotermund, H.H.,
Block, J.H., and Ertl, G., Nature, 1994, vol. 370, p. 276.
1
3. Kim, M., Bertram, M., Pollmann, M., Oertzen, A.V.,
Mikhailov, A.S., Rotermund, H.H., and Ertl, G., Sci-
ence, 2001, vol. 292, p. 1357.
1
4. Kaichev, V.V., Saraev, A.A., Gladky, A.Y., Prosvirin, I.P.,
Blume, R., Teschner, D., Hävecker, M., Knop-Ger-
icke, A., Schlögl, R., and Bukhtiyarov, V.I., Phys. Rev.
Lett., 2017, vol. 119, p. 026001.
15. Zhdanov, V.P., Phys. Rev. E, 1999, vol. 59, p. 6292.
1
6. Saraev, A.A., Kaichev, V.V., Bukhtiyarov, V.I., and
Kosolobov, S.S., Kinet. Catal., 2015, vol. 56, vol. 5,
p. 598.
1
7. Knop-Gericke, A., Kleimenov, E., Hävecker, M.,
Blume, R., Teschner, D., Zafeiratos, S., Schlögl, R.,
Bukhtiyarov, V.I., Kaichev, V.V., Prosvirin, I.P.,
Nizovskii, A.I., Bluhm, H., Barinov, A., Dudin, P.,
and Kiskinova, M., Adv. Catal., 2009, vol. 52, p. 213.
ACKNOWLEDGMENTS
This work was performed within the framework of
the budget project (АААА-А17-117041710078-1) for
Boreskov Institute of Catalysis, Siberian Branch, Rus-
sian Academy of Sciences. In situ/operandо XRD
studied were carried out using the equipment of the
1
8. Bulavchenko, O.A., Vinokurov, Z.S., Afonasenko, T.N.,
Tsyrul’nikov, P.G., Tsybulya, S.V., Saraev, A.A., and
Kaichev, V.V., Dalton Trans., 2015, vol. 44, p. 15499.
KINETICS AND CATALYSIS
Vol. 59
No. 6
2018

THE REASONS FOR NONLINEAR PHENOMENA
819
1
9. Bulavchenko, O.A., Venediktova, O.S., Afonasenko, T.N., 25. Yu, J., Rosso, K.M., and Bruemmer, S.M., J. Phys.
Tsyrul’nikov, P.G., Saraev, A.A., Kaichev, V.V., and
Chem. C, 2012, vol. 116, p. 1948.
Tsybulya, S.V., RSC Adv., 2018, vol. 8, p. 11598.
2
6. Bychkov, V.Y., Tulenin, Y.P., Slinko, M.M., Khu-
dorozhkov, A.K., Bukhtiyarov, V.I., Sokolov, S., and
Korchak, V.N., Appl. Catal., A, 2016, vol. 522, p. 40.
20. Aulchenko, V.M., Evdokov, O.V., Kutovenko, V.D.,
Pirogov, B.Y., Sharafutdinov, M.R., Titov, V.M.,
Tolochko, B.P., Vasiljev, A.V., Zhogin, I.A., and Zhu- 27. Stötzel, J., Frahm, R., Kimmerle, B., Nachtegaal, M.,
lanov, V.V., Nucl. Instrum. Methods Phys. Res., Sect. A.,
009, vol. 603, p. 76.
and Grunwaldt, J.-D., J. Phys. Chem. C, 2012, vol. 116,
p. 599.
2
2
8. Weaver, J.F., Chem. Rev., 2013, vol. 113, p. 4164.
2
1. Fytyk 1.2.9. http://www.fityk.nieto.pl.
2
9. Martin, N.M., Van den Bossche, M., Hellman, A.,
Grönbeck, H., Hakanoglu, C., Gustafson, J.,
Blomberg, S., Johansson, N., Liu, Z., Axnanda, S.,
Weaver, J.F., and Lundgren, E., ACS Catal., 2014,
vol. 4, p. 3330.
2
2
2
2. Zhang, X., Hayward, D.O., and Mingos, D.M.P.,
Catal. Lett., 2002, vol. 83, p. 149.
3. Zhang, X., Hayward, D.O., and Mingos, D.M.P.,
Catal. Lett., 2003, vol. 86, p. 235.
4. Bychkov, V.Y., Tyulenin, Y.P., Korchak, V.N., and
Aptekar, E.L., Appl. Catal., A, 2006, vol. 304, p., 21.
Translated by Andrey Zeigarnik
KINETICS AND CATALYSIS Vol. 59 No. 6 2018