NASICON Catalysts with Composition Na(Cs)1 – 2xMxZr2(PO4)3 for Transformations of Aliphatic Alcohols

Article abstract of DOI:10.1134/S0965544120100096

Abstract: The activity of solid electrolytes from the family of Na(Cs)–M_x–Zr phosphates (NZP and CZP) with dopant ions M⁺² = Co⁺² and Ni⁺² synthesized by the sol–gel method is studied in the gas-phase transformations of isopropanol and isobutanol. It is found that the amount of M⁺² and its nature, as well as the direction of a change in the catalyst temperature (heating or cooling in the range of 473?693 K), affects the activity of M-NZP and M-CZP. In the transformation of isopropanol, hysteresis of the acetone yield, counter-clockwise (type I) for Co-NZP and clockwise (type II) for Ni-NZP, is observed in the heating–cooling cycles. An increase in the activity of Co-NZP in the cooling mode (type I) is related to a rise in the apparent activation energy of alcohol dehydrogenation under conditions of the formation of new catalytically active sites in the form of a Co⁺²?Zr⁺⁴ → Co⁺³?Zr⁺³ ion pair with the oxidized form of M and the reduced form of Zr. Type I hysteresis of the total alcohol conversion is observed in the transformations of isobutanol to olefin and aldehyde over Co(Ni)-CZP. The hysteresis is associated with a decrease in the activation energy of alcohol dehydrogenation. The main reaction over the Co(Ni)-CZP catalysts is the dehydration of alcohol with an increase in the activation energy of the reaction upon cooling, for example, by 53 kJ/mol for Ni_0.25-CZP.

Full text of DOI:10.1134/S0965544120100096

ISSN 0965-5441, Petroleum Chemistry, 2020, Vol. 60, No. 10, pp. 1176–1183. © Pleiades Publishing, Ltd., 2020.
Russian Text © The Author(s), 2020, published in Nanogeterogennyi Kataliz, 2020, Vol. 5, No. 2, pp. 162–169.
NASICON Catalysts with Composition Na(Cs)1 – 2xMxZr2(PO4)3
for Transformations of Aliphatic Alcohols
I. I. Mikhalenkoa, *, A. I. Pylininaa, and E. I. Knyazevaa
aPeoples’ Friendship University of Russia, Moscow, 117198 Russia
*e-mail: mikhalenko_ii@pfur.ru
Received July 22, 2019; revised February 7, 2020; accepted June 11, 2020
Abstract—The activity of solid electrolytes from the family of Na(Cs)–Mx–Zr phosphates (NZP and CZP)
with dopant ions M+2 = Co+2 and Ni+2 synthesized by the sol–gel method is studied in the gas-phase trans-
formations of isopropanol and isobutanol. It is found that the amount of M+2 and its nature, as well as the
direction of a change in the catalyst temperature (heating or cooling in the range of 473−693 K), affects the
activity of M-NZP and M-CZP. In the transformation of isopropanol, hysteresis of the acetone yield,
counter-clockwise (type I) for Co-NZP and clockwise (type II) for Ni-NZP, is observed in the heating–
cooling cycles. An increase in the activity of Co-NZP in the cooling mode (type I) is related to a rise in the
apparent activation energy of alcohol dehydrogenation under conditions of the formation of new catalytically
active sites in the form of a Co+2−Zr+4 → Co+3−Zr+3 ion pair with the oxidized form of M and the reduced form
of Zr. Type I hysteresis of the total alcohol conversion is observed in the transformations of isobutanol to olefin
and aldehyde over Co(Ni)-CZP. The hysteresis is associated with a decrease in the activation energy of alcohol
dehydrogenation. The main reaction over the Co(Ni)-CZP catalysts is the dehydration of alcohol with an
increase in the activation energy of the reaction upon cooling, for example, by 53 kJ/mol for Ni0.25-CZP.
Keywords: sol–gel synthesis, complex zirconium orthophosphates, dopant cations, nickel, cobalt, isopropa-
nol, isobutanol, dehydrogenation and dehydration of alcohol, catalytic activity, thermal hysteresis
DOI: 10.1134/S0965544120100096
Catalytic dehydrogenation of low-molecular- plex orthophosphates were considered in [1–4], and
weight aliphatic alcohols is a simple method for the data on the catalytic activity of such materials in
obtaining carbonyl compounds and hydrogen. It is vapor-phase reactions involving alcohols were
easy to vary the direction of transformation (dehydro- reported in [5–11].
genating and dehydrating functions of the catalyst) by
The aim of this work is to investigate effect of the
changing the composition of the catalyst.
partial replacement of conducting sodium or cesium
Complex zirconium orthophosphates with the
framework structure, which belong to solid cationic
electrolytes with the NASICON structure of the gen-
eral formula MA2(PO4)3, where M are the conducting
ions of alkaline metals and A are the ions of 3d-transi-
tion metals, attract considerable attention of catalytic
chemists. This is associated with the fact that modifi-
cation of the catalyst aimed at improvement of its cat-
alytic activity and change in the selectivity of the pro-
cess may involve introduction of various cations into
the phosphate composition through replacement of
both the framework ion A and the conducting ions M
without violation of the framework structure. Rhom-
cations with nickel and cobalt cations on the activity of
Na(Cs) zirconium phosphate synthesized by the sol–
gel method in the transformations of isopropanol and
isobutanol under temperature increase–decrease
cycles in the temperature range of 473−693 K. This
work continues the research topics of [9–11].
EXPERIMENTAL
The Sol–Gel Synthesis of Ni(Co)-NZP
The initial substances were ZrO2 (Acros Organics),
H3PO4, NaCl, NiCl2 ⋅ 2H2O, and CoCl2 ⋅ 6H2O (all of
bohedral, rhombic, monoclinic, and cubic structures reagent grade). The stoichiometric amounts of NaCl
are possible for NASICON electrolytes. For basic and ZrO2 aqueous solutions of predissolved in HCl to
orthophosphate NaZr2(PO4)3 the main structure is
avoid hydrolysis were poured together under continu-
rhombohedral, which represents PO4 tetrahedra and ous stirring at room temperature, and NiCl2 or CoCl2
ZrO6 octahedra with the vertex contacts between the was slowly added dropwise a H3PO4 solution accord-
polyhedra united into a framework. The preparation, ing to the stoichiometry. The gel formed upon heating
crystallochemistry features, and application of com- was dried at 80 and 110°C and subjected to thermal
1176

NASICON CATALYSTS WITH COMPOSITION Na(Cs)1 – 2xMxZr2(PO4)3
1177
Table 1. Texture, elemental composition of the surface, and structural characteristics of the initial samples of Na1 − 2xMxZr2(PO4)3
with Ni and Co
Dispersity
specific surface area, m /g pore volume, cm /g
Unit cell parameters and volume
Mx
V, Å3
2
3
pore diameter, Å a = b, Å
c, Å
Ni0.125-NZP
Ni0.25-NZP
Co0.125-NZP
Co0.25-NZP
2.4
1.5
1.2
1.4
0.0252
0.0359
0.0309
0.0205
8.801
8.835
8.798
8.788
22.791
22.83
22.75
22.82
267
276
214
256
547
393
476
526
at % (X-ray photoelectron spectroscopy data)
Unit cell parameters and volume
Mx
V, Å3
Cs
M
Zr
a = b, Å
c, Å
Ni0.15-CZP
Ni0.25-CZP
Co0.15-CZP
Co0.25-CZP
8.569
8.569
8.575
8.571
24.938
24.938
24.958
24.948
1585.9
1585.8
1589.3
1587.2
4.2 (2.8)
4.2 (2.8)
3.7 (1.4)
5.4 (1.4)
12.1 (11.3)
12.1 (11.3)
Bracketed figures in italics refer to the stoichiometric concentration of the element.
treatment at 600 and 800°C for 8 h accompanied by a carrier gas, a flame ionization detector, and column
grinding at each stage.
packed with a Porapak-Q sorbent). The feed rate of a
helium + alcohol mixture to the microreactor was
1.2 L h−1. The weight of the catalyst with a 1–2-mm-
thick catalyst bed on a porous filter was 30 mg. The
sample was pre-exposed in a helium flow at 673 K for
The Sol–Gel Synthesis of Ni(Co)-CZP
The initial substances were ZrOCl2 ⋅ 8H2O (Acros
Organics), H3PO4, CsBr, NiCl2, and CoCl2 (all of 1 h followed by a slow decrease in the temperature of
the reactor to 373 K. The temperature dependences of
the yield of the products were obtained in the modes of
increasing and decreasing temperature provided that
the stationary activity of the catalyst was reached at
each temperature, which took 15–20 min. For the
temperature range, in which alcohol conversion did
not exceed 50%, the experimental activation energies
reagent grade). H3PO4 was added dropwise in a mix-
ture of the weighed portions of solid salts according to
stoichiometry. The resulting gel was dried for three
days with regular grinding, the ground samples were
placed into a drying oven at 300°C for 4 h, after which
they were subjected to calcination at 600°C for 6 h.
The obtained samples were polycrystalline pow-
ders with characteristic pale green (Ni) and lilac (Co)
colors:
C=O
C=С
of dehydrogenation
and dehydration
of
)
(Ea
)
(Ea
alcohol were calculated from the Arrhenius depen-
dences of the yield of reaction product N
M-NZP
M-CZP
Na0.75Ni0.125Zr2(PO4)3
Na0.75Co0.125Zr2(PO4)3
Na0.5Ni0.25Zr2(PO4)3
Na0.5Ni 0.25Zr2(PO4)3
Cs0.7Ni0.15Zr2(PO4)3
Cs0.7Co0.15Zr2(PO4)3
Cs0.5Ni0.25Zr2(PO4)3
Cs0.5Co0.25Zr2(PO4)3
RESULTS AND DISCUSSION
The data on the porosity, specific surface area, and
crystal structure of M-NZP for Na0.75M0.125Zr2(PO4)3
and Na0.5M0.25Zr2(PO4)3 with M = Co and Ni are pre-
sented in Table 1. The rhombohedral modification of
the NASICON structure, which is characteristic for
the family of sodium zirconium phosphates (NZP),
was confirmed by fact that the positions and relative
intensity of the maxima of the diffraction patterns
coincided with the published data.
The phase and chemical compositions, specific
surface area, and porosity of M-NZP and M-CZP
were studied by X-ray powder diffraction analysis
(Rigaku Ultim IV, DRON-7), scanning ion–electron
spectroscopy (Quanta 200 3D), low-temperature
nitrogen adsorption (TriStar 3000, ASAP 2020-MP
Micromeritics), and X-ray photoelectron spectros-
copy (XPS) (Axis Ultra, Kratos Analytical).
The catalytic activity of complex zirconium phos- of pore sizes (210−280 Å) NZP belong to mesoporous
phates in the transformations of isopropanol and systems. The conductance channels filled with Na,
isobutanol was tested on a flow-type unit followed by Cs, Ni, and Co cations have a size of 8–10 Å, and,
the chromatographic analysis of substances (helium as according to the published data, such pores are appar-
Ceramic materials have a small specific surface
area Ssp, and the sample with Ni0.125 is characterized by
the highest surface area in the M-NZP series. In terms
PETROLEUM CHEMISTRY Vol. 60 No. 10 2020

1178
MIKHALENKO et al.
(a)
(b)
0
–1.8
–3.5
(c)
0.7
0.4
0
0.7
Co
Ni
0.4
0
1/T*
620
Т, К
700
620
Т, К
700
1.6
1.8
540
540
1.4
103/Т, К–1
Fig. 1. (a, b) Temperature dependences of the yield of acetone over the Na
M
Zr (PO ) catalysts in the mode of ( solid lines)
0.75 0.125 2 4 3
increasing and (dashed lines) decreasing temperature and (c) the Arrhenius dependence for Ni-NZP.
ently inaccessible to the sorption/desorption of nitro- amount of the formed acetone for the catalysts with
gen. The unit cell volume V decreases by 28% with an the concentration of cobalt and nickel x = 0.125 in the
increase in the concentration of Ni, while Co-NZP it heating–cooling cycle is presented in Figs. 1a and 1b.
increases by ~10%, which is within the accuracy of
determination of V. The X-ray powder diffraction
analysis showed that all the Ni(Co)-NZP samples
contain small amounts of ZrO2 and ZrP2O7.
Only in the case of the Co-containing sample
activity was higher upon decrease in temperature com-
pared with the mode of increasing temperature; i.e., a
counter-clockwise thermal hysteresis (type I hystere-
sis) was observed. A small decrease in the values of
NC=O with decreasing temperature (type II hysteresis)
was recorded for the Ni-containing sample.
A similar activity testing was performed for
the number of samples with x = 0.25. In the case of
Co0.25-NZP, the type of hysteresis changed while for
Ni0.25-NZP the clockwise hysteresis was more pro-
nounced.
A distinctive feature of sodium zirconium phos-
phates is that the energetically favorable position of the
ion in the conductance channel changes at 573–
593 K. This factor and its role in the dehydrogenating
ability of M-NZP were discussed in [10]. In the tem-
perature range of T* ~ 593–613 K in the Arrhenius
dependences of the yield of acetone over the Ni-NZP
catalysts a break is observed with a decrease in the
slope of straight lines ln NC=O−T−1 (Fig. 1c). This is
evidence that the catalytic reaction is sensitive to the
transition of cation in the NZP conductance channel
from position M1 to position M2. The break in the
dependences ln NC=O−T−1 is repeated under the mode
of decreasing temperature of the catalyst, which indi-
cates the reversibility of the M1–M2 transition with
low (T > T*) and high (T < T*) experimental activa-
tion energies of isopropanol dehydrogenation Ea being
preserved. Co0.125-NZP does not have such a charac-
teristic feature, while for Co0.25-NZP it is slightly pro-
nounced, which suggests that cobalt occurs in the spe-
cial state.
According to X-ray analysis, Cs0.7M0.15Zr2(PO4)3
and Cs0.5M0.25Zr2(PO4)3 have a rhombohedral crystal
lattice like CsZr2(PO4)3. The crystal lattice parameters
of triple phosphates (Table 1) are close to the lattice
parameters of the double phosphate (a = 8.581
0.001 Å, c = 24.973 0.005 Å, V = 1592.3 0.4 Å3)
but with decreased values of a, c, and V. A linear
change in the unit cell parameters of Co-CZP with an
increase in the concentration of the dopant ion was
discussed in [11]. For triple phosphates the specific
surface area and the pore volume are 4.1 m2/g and
0.0002 cm3/g for Cs0.7M0.15Zr2(PO4)3 and 1.5 m2/g and
0.002 cm3/g for CsZr2(PO4)3, respectively.
The values of the binding energy of all elements
obtained from the XPS spectra of the samples from the
Ni(Co)-CZP series corresponded to the phosphate
formula. The atomic ratio of oxygen to phosphorus
O/P was overestimated against the stoichiometric
ratio; it was 6 and 5.6 for Co-CZP and Ni-CZP,
respectively. This finding suggests an excessive con-
centration of the oxygen of the phosphate groups on
the surface. The surface concentration of Cs is
increased 1.5-fold and the surface concentration of
dopant ions is increased 2.6-fold (Ni) and 3.8-fold
(Co) in comparison with the stoichiometry of the
phosphate (Table 1). The concentration of Zr corre-
sponds to the stoichiometric concentration.
Over Na–Co(Ni) zirconium phosphates (Co-NZP
and Ni-NZP) only the dehydrogenation of isopropa-
nol proceeds, and the double phosphate NaZr2(PO4)3
The yield of acetone normalized to the unit surface
demonstrates no catalytic activity. A change in the of M-NZP samples, that is, the specific activity of the
PETROLEUM CHEMISTRY
Vol. 60
No. 10
2020

NASICON CATALYSTS WITH COMPOSITION Na(Cs)1 – 2xMxZr2(PO4)3
1179
(a)
W673 K, %
(b)
60
N6C7=3 OК, µmol/(m2 h)
0.5
50
40
30
20
10
x = 0.25
x = 0.125
0.3
0
0
Co Ni
Co Ni
Co
0.125
Co
0.25
Ni
0.125
Ni
0.25
Fig. 2. (a) Yield of acetone normalized to the unit surface (the heating mode) and (b) degree of transformation of isopropanol
over the Na
umns).
M Zr (PO ) catalysts, where M = Co and Ni, in the mode of heating (dark columns) and cooling (light col-
x 2 4 3
(1 − 2x)
catalysts, does not depend on the concentration of M;
Let us analyze a change in the Arrhenius parame-
it is higher in the case of Co-NZP (Fig. 2a). The depth ters of dependences ln NC=O−T−1 for the dehydroge-
nation reaction of isopropanol (the values of ΔEa and
Δln N0). The yield of the product is the rate of reaction
of alcohol dehydrogenation, which is equal or directly
proportional to the rate constant of the reaction K
in the absence of diffusion limitations; therefore,
N↓/N↑ ≈ K↓/K↑. Taking the logarithm of N↓/N↑, we
obtain the expression
of the transformation (conversion) of isopropanol W
increases in the mode of decreasing temperature for
Co0.125-NZP (Fig. 2b), while for Co0.25-NZP this
value decreases. The conversion of alcohol
over the Co0.25-NZP catalyst is lower than that over
Co0.125-NZP. This feature distinguishes the samples
with Co from the samples with Ni. Upon cooling of
the Ni-NZP catalysts conversion decreased.
ln N ↓ N ↑ = ΔlnN − RT −1 ΔЕа ,
(
)
(1)
(
)
0
Table 2 presents the yields of the product N at a
temperature of 593 K, which is in the region of the
M1–M2 transition and the values of Ea. It follows
from the analysis of ΔN that a large counter-clockwise
hysteresis of the acetone yield (the “plus” sign in front
of ∆N = N↓ − N↑ in Тable 2) is characteristic for the
Co0.125-NZP catalyst; for the corresponding nickel
catalyst, clockwise hysteresis (the “minus” sign).
from which it follows that the thermal hysteresis may
be related to a change in both the apparent activation
energy of the reaction ∆Ea = Ea↓ − Ea↑ and the loga-
rithm of the preexponential factor ∆lnN0 = lnN0↓ −
lnN0↑ (Table 1). The accuracy of determination of Ea
is 5–7 kJ/mol; therefore, the values of ∆Ea are neg-
ligible within 10 kJ/mol. Expression (1) can be written
as the equation
Table 2. Hysteresis parameters for the Na(1 − 2x)MxZr2(PO4)3 catalysts: the change in the yield of acetone in the cooling–
heating cycle (ΔN, %) at T ∼ T*, apparent activation energy of formation of acetone (Ea) in the mode of heating (numerator)
and cooling (denominator), and changes in the activation energy and logarithm of the preexponential factor
Ea↑/Ea↓
∆Ea, kJ/mol
T < T* T > T*
∆lnN0
Mx
∆N593 K %
T < T*
T > T*
T < T*
T > T*
Co0.125
+90
−30
−57
0
70/96
+ 5.5
+26
Co0.25
Ni0.125
Ni0.25
93/60
72/91
67/67
36/61
44/64
0
−5.5
+3.3
+6.2
+1
−33
+19
+31
+25
+20
+4.4
+3.7
83/114
PETROLEUM CHEMISTRY Vol. 60 No. 10 2020

1180
MIKHALENKO et al.
−1
which is of interest for the polymer industry. It was
(
)
(2)
(3)
ln N ↓ N ↑ = ΔlnN − RT
αQ↑ − α'Q↓ ,
(
)
(
)
0
found that the activity of triple CZP, like that of triple
NZP, is higher than the activity of the double phos-
phate. Under our experimental conditions the total
conversion of alcohol over the CsZr2(PO4)3 catalyst is
below 25%; however, the isobutylene selectivity
is 100%.
because according to the linearity ratio
Еа = Еa0 − αQ,
0
where
is the true activation energy of the reaction,
Q is the heat of adsorption of alcohol, and α is the con-
Еa
The conversion of isobutanol increases upon tran-
sition to the cooling mode; as a result, olefin selectivity
decreases except for Co0.25 at 593 K and Ni0.25 at 653 K
(Fig. 3). This means that an increase in the total con-
version is associated with the formation of aldehyde,
that is, intensification of the dehydrogenating func-
tion of the catalyst with sites involved in the redox pro-
cess. The dehydrating function predominates in the
Ni(Co)-CZP catalysts—the conversion of alcohol to
isobutylene is substantially higher than the conversion
of alcohol to isobutanol. This effect is especially pro-
nounced in the samples with x = 0.15 (Figs. 3, 4). As
opposed to Ni(Co)-NZP, there is no clear break
(manifestation of the M1 → M2 transition) in the
dependences ln N−T−1 in the transformations of
isobutanol over the Ni(Co)-CZP catalysts (Fig. 4c),
possibly, because of a larger radius of the Cs+ ion
(rCs+ = 0.166 nm versus rNa+ = 0.098 nm [13]).
tribution of heat to Ea.
The growth in activity upon cooling the catalyst—
the counter-clockwise hysteresis—may be associated
with both a decrease in Ea (∆Ea < 0) and an increase in
the preexponential factor (∆ln N0 > 0). Both factors
act simultaneously with the predominance of one over
another. The opposite type of hysteresis is also possi-
ble with a similar-in-sign change in Ea and ln N0, like
in the case of Co0.125-NZP and Ni0.125-NZP catalysts.
A significant growth in the activity of Co0.125-NZP in
the cooling mode is related to the increase in Ea and
ln N0 due to the involvement of new sites with a
decreased heat of adsorption Q in the reaction, that is,
to the strength of the bond of the alcohol with the cat-
alyst. There is no break in the ln NC=O−T−1 depen-
dences for Co0.125-NZP. For the Ni0.125-NZP catalyst
with the clockwise hysteresis, which is sensitive to the
M1–M2 transition, the result was similar to that for
Co0.125-NZP: ΔEa > 0 and Δln N0 > 0 at T < T* and
T > T*. The absence or a weak thermal hysteresis is
possible in the case of the same contribution from the
first and second summands of Eq. (1). An example of
such a compensation is the dehydrogenation of iso-
propanol over Ni0.25-NZP with the same character of
change in Ea and ln N0.
The activity in the dehydration reaction of alcohol
decreases upon transition to the cooling mode, and
the activation energy increases (Table 3). Differences
in the activation energies of olefin formation ∆EC=C
increase with the concentration of Mx due to a
decrease in Ea↑ because the values of Ea↓ are close and
fall within 86–99 kJ/mol. An opposite character of
changes in Ea and ln N0 is observed in the case of the
dehydrogenation reaction: the active state of the sites
in the mode of decreasing temperature is character-
C=O
The charge states of nickel and cobalt can change in
the reducing medium of the alcohol dehydrogenation
reaction. In terms of the oxidation–reduction mecha-
nism of the reaction, in which the stage of reoxidation
of M is treated as the limiting stage [5], it is difficult to
explain the activity of Ni+2 and Co+2 by the negative
ized by a decrease in
with a small divergence of
Ea
∆EC=O = 55 kJ/mol in the sample with Co0.15
.
A counter-clockwise thermal hysteresis (type I)
was observed in the reaction of isobutanol dehydroge-
nation, and a clockwise thermal hysteresis (type II)
was observed in the reaction of isobutanol dehydra-
tion; i.e., the catalytic processes differing in mecha-
nisms and products have different types of hysteresis
over the same catalyst. A change in the Arrhenius
parameters in the case of hysteresis is affected by the
amount of the dopant ion and its nature. The positive
value of the reduction potential1: −0.257 V (Ni) and
−0.27 V (Co) for the reaction M+2 + 2e = M0.
0
The reduction potential is positive
= +1.39 V
Er
only for Co+3 → Co+2. The appearance of the oxidized
form of Co+3 in the composition of the catalytically
active site of the reaction can be explained by forma-
tion of a bimetallic pair Co+2−Zr+4 → Co+3−Zr+3
(Co+2 is the donor of electrons for Zr+4), in which
Co+3 stabilizes Zr+3. The presence of Zr+3 paramag-
netic ions in the NZP catalysts was noted in [12].
values of ∆EC=C increase in the order Co0.15 < Co0.25
<
Ni0.15 < Ni0.25. The negative values of ∆EC=O increase
in the same order. Little attention is paid in the litera-
ture to the discussion of the thermal hysteresis of cat-
alytic reactions within the framework of variability of
the chemical composition and structure of the active
site of a catalyst in the heating–cooling cycle due to
the complexity of recording such changes. The ther-
mal hysteresis was described for the oxidation reac-
tions of CO and alkanes over oxide and supported cat-
alysts [14–16]. In [14], the theoretical analysis of the
The formation of two products, isobutylene (the
main product) and isobutanal, is observed in the
transformations of isobutanol over the Ni(Co)-CZP
catalysts, which indicates two routes for alcohol trans-
formation. Isobutylene is the most valuable product,
1
J. Emsley, The Elements, 2nd ed. (Oxford Univ. Press, Oxford,
1991; Mir, Moscow, 1993).
PETROLEUM CHEMISTRY
Vol. 60
No. 10
2020

NASICON CATALYSTS WITH COMPOSITION Na(Cs)1 – 2xMxZr2(PO4)3
(a) (b)
C=C, %
1181
W, %
S
100
100
90
80
70
60
50
80
60
40
20
0
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
Co 0.15
Co 0.25
Ni 0.25
Co 0.15
Co 0.25
Ni 0.25
Ni 0.15
Ni 0.15
Fig. 3. (a) Total conversion of isobutanol and (b) isobutylene selectivity over the Cs
M Zr (PO ) catalysts, where M = Co
x 2 4 3
(1 − 2x)
and Ni, at (1) 593 and (2) 653 K in the mode of heating (dark columns) and cooling (light columns).
W, %
70
(a)
(b)
(c)
W, %
70
lnN
3.0
2.4
1.8
1.2
0.6
0
35
35
0
0
563
613
663
563 613
T, К
663
1.6
1.7
1/T, 10–3
1.8
1.9
513
513
1.5
T, К
Fig. 4. (a, b) Temperature dependences of isobutanol conversion to isobutene (solid lines) and isobutanal (dashed lines) over
M -CZP samples with (a) Co and (b) Ni for x = 0.15 (squares) and 0.25 (circles) and (c) the Arrhenius dependence of isobutylene
x
yield for x = 0.25.
Co
hysteresis of the partial oxidation reaction of methane type reactor was analyzed in [16]. Using the new in situ
CH4 + ½O2 = CO + 2H2 (ΔH = −36 kJ/mol) was per- obtained data the authors replaced the earlier explana-
tion of the hysteresis by the removal of adsorbed oxy-
gen from the Pt sites and their local warmup with
another one—the formation of active carbon sites
upon the dissociative adsorption of methane.
formed using the Reynolds and Damköhler numbers
and taking into account the indirect mechanism,
including three more reactions: CH4 + 2O2 = CO2 +
2H2O (ΔH = −803 kJ/mol), CH4 + H2O ↔ CO + 3H2
(ΔH = +206 kJ/mol), and CH4 + CO2 ↔ 2CO + 2H2
(ΔH = +247 kJ/mol). The authors [15] observed the
thermal hysteresis in the co-oxidation reaction of CO
and C2H6 over a Pt/Al2O3 catalyst at 300–600 K,
which, in their opinion, differs from the thermal hys-
teresis of stationary activity. The thermal counter-
clockwise hysteresis in the catalytic combustion (com-
plete oxidation) of methane over Pt/Al2O3 in a flow-
Thus, the often-observed thermal hysteresis of the
activity of heterogeneous catalysts has not been unam-
biguously explained yet. Thermal hysteresis also takes
place in the transformations of isopropanol and isobu-
tanol over Co(Ni)-NZP and Co(Ni)-CZP complex
phosphate catalysts with the NASICON structure.
The counter-clockwise (type I) or clockwise (type II)
type of hysteresis with a change in the activation
PETROLEUM CHEMISTRY Vol. 60 No. 10 2020

1182
MIKHALENKO et al.
Table 3. Characteristics of isobutanol transformations over Cs(1 − 2x)MxZr2(PO4)3: the experimental activation energy for
C=С
C=O
the formation of isobutylene
and isobutanal
in the mode of heating (numerator) and cooling (denominator)
Ea
Ea
and changes in the activation energy
Dehydration of alcohol
Dehydrogenation of alcohol
Mx
ΔlnN0C=C
ΔlnN0C=O
ΔEaC=С
,
ΔEaC=O
,
kJ/mol
Ea↑/Ea↓
Ea↑/Ea↓
kJ/mol
Co0.15
87/90
61/91
57/99
33/86
+3
+30
+22
+53
+1
+6
112/57
74/47
67/48
69/48
−55
−27
−19
−21
−10
−6
−3
−4
Co0.25
Ni0.15
Ni0.25
+9
+11
A.I. Pylinina, ORCID: https://orcid.org/0000-0002-
8190-6167
E.I. Knyazeva, ORCID: https://orcid.org/0000-0001-
6254-0959
energy and the preexponential factor depends on the
composition of NZP (the conducting ion and dopant
ion M+2 = Co, Ni), reaction type (alcohol dehydroge-
nation or dehydration), and the range of reaction tem-
peratures because M can occupy position M1 or M2 in
the conductance channel of the solid electrolyte
depending on temperature.
REFERENCES
1. J. Alamo, Solid State Ionics 63–65, 547 (1993).
https://doi.org/10.1016/0167-2738(93)90158-Y
ACKNOWLEDGMENTS
2. B. E. Scheetz, D. K. Agrawal, E. Breval, and R. Roy,
Waste Manag. 14 (6), 489 (1994).
We are grateful to the Shared Research Center Diagnos-
tics of the Structure and Properties of Nanomaterials of the
Belgorod State National Research University and person-
ally to Dr. Sci. (Chem.) O.E. Lebedeva for her help in the
study of Ni(Co)-NZP by X-ray diffractometry, scanning
ion-electron spectroscopy, and nitrogen thermal desorp-
tion; to the Shared Research Center of the Moscow State
University and personally to Cand. Sci. (Chem.) S.V. Savi-
lov for his help in measuring the surface area, pore size, and
X-ray photoelectron spectra of Ni(Co)-CZP samples; and
to the Shared Research Center Physicochemical Studies of
New Materials and Catalyst Systems of the Peoples’ Friend-
ship University of Russia and personally to Cand. Sci.
(Chem.) N.N. Lobanov for the X-ray powder diffraction
analysis of Ni(Co)-CZP samples.
https://doi.org/10.1016/0956-053X(94)90133-3
3. V. I. Pet’kov, Russ. Chem. Rev. 81 (7), 606 (2012).
https://doi.org/10.1070/RC2012v081n07ABEH00424
3
4. Sh. Zhang, B. Quan, Zh. Zhao, B. Zhao, Y. He, and
W. Chen, Mater. Lett. 58 (1–2), 226 (2004).
https://doi.org/10.1016/S0167-577X(03)00450-6
5. S. Arsalane, M. Kacimi, M. Ziyad, G. Coudurier, and
J. C. Vedrine, Appl. Catal., A 114 (2), 243 (1994).
https://doi.org/10.1016/0926-860X(94)80177-0
6. S. N. Pavlova, V. A. Sadykov, E. A. Paukshtis,
E. B. Burgina, S. P. Degtyarev, D. I. Kochubei,
A. V. Kalinkin, N. F. Saputina, R. I. Maximovskaya,
V. I. Zaikovskii, R. Roy, and D. Agrawal, in Studies in
Surface Science and Catalysis, Ed. by A. Parmaliana,
D. Sanfilippo, F. Frusteri, A. Vaccari, and F. Arena
(Elsevier Science B.V.: Amsterdam, 1998), Vol. 119,
pp. 759–764.
FUNDING
Preparation of this publication was supported by the
Program 5-100 of the Peoples’ Friendship University of
Russia.
https://doi.org/10.1016/S0167-2991(98)80523-5
7. M. M. Ermilova, M. V. Sukhanov, R. S. Borisov,
N. V. Orekhova, V. I. Pet’kov, S. A. Novikova,
A. B. Il’in, and A. B. Yaroslavtsev, Catal. Today 193
(1), 37 (2012).
CONFLICT OF INTEREST
https://doi.org/10.1016/j.cattod.2012.02.029
The authors declare that there is conflict of interest to be
disclosed in this paper.
8. A. B. Ilyin, M. M. Ermilova, N. V. Orekhova, and
A. B. Yaroslavtsev, Nanotechnol. Russ. 13 (7–8), 365
(2018).
https://doi.org/10.1134/S1995078018040067
ADDITIONAL INFORMATION
9. E. I. Povarova, A. I. Pylinina, and I. I. Mikhalenko,
Rus. J. Phys. Chem. A 86 (6), 935 (2012).
I.I. Mikhalenko, ORCID: https://orcid.org/0000-
0002-5406-1015
https://doi.org/10.1134/S0036024412060210
PETROLEUM CHEMISTRY
Vol. 60
No. 10
2020

NASICON CATALYSTS WITH COMPOSITION Na(Cs)1 – 2xMxZr2(PO4)3
1183
10. E. I. Povarova, A. I. Pylinina, and I. I. Mikhalenko, 14. W.-H. Chen, T.-W. Chiu, and C.-I. Hung, Int. J. Hy-
Prot. Met. Phys. Chem. Surf. 50 (3), 331 (2014).
https://doi.org/10.1134/S2070205114030125
drogen Energy 35 (12), 6291 (2010).
https://doi.org/10.1016/j.ijhydene.2010.03.133
11. A. I. Pylinina, M. N. Chernyshova, N. N. Lobanov,
S. V. Savilov, and E. M. Kasatkin, Theor. Exp. Chem.
53 (1), 47 (2017).
15. R. K. Dadi, D. Luss, and V. Balakotaiah, Chem. Eng.
J. 297, 325 (2016).
https://doi.org/10.1016/j.cej.2016.03.139
https://doi.org/10.1007/s11237-017-9500-3
12. S. Arsalane, M. Ziyad, G. Coudurier, and J. C. Ved-
16. L. Chen, J. P. McCann, and S. L. Tait, Appl. Catal., A
549, 19 (2018).
rine, J. Catal. 159 (1), 162 (1996).
https://doi.org/10.1006/jcat.1996.0075
https://doi.org/10.1016/j.apcata.2017.09.008
13. L. T. Bugaenko, S. M. Ryabykh, and A. L. Bugaenko,
Moscow Univ. Chem. Bull. 63 (6), 363 (2008).
https://doi.org/10.3103/S0027131408060011
Translated by E. Boltukhina
PETROLEUM CHEMISTRY Vol. 60 No. 10 2020