Conversion of aliphatic C1–C2 alcohols on In– Nb– Mo-doped complex lithium phosphates and HZr2(PO4)3 with NASICON-type structure

Article abstract of DOI:10.1016/j.jallcom.2018.03.099

In– Nb– Mo-doped lithium complex phosphates and HZr₂(PO₄)₃ with NASICON-type structure were synthesized in this paper. Particle size distribution lies between 50 and 300 nm. The obtained samples were characterized by X-ray diffraction analysis, scanning electron microscopy and X-ray microanalysis. Investigation of the catalytic properties of synthesized compounds in the C1–C2 alcohols conversions showed that heterovalent doping has a determining effect on the obtained catalysts’ activity and selectivity. It is shown that the thermodynamic factors and the dopant ability to change the degree of oxidation and acid function of the catalysts play a key role in methanol and ethanol conversion. A number of catalysts show the high activity and selectivity of the formation of dimethyl and diethyl ethers and ethylene. High selectivity for C4 hydrocarbons is achieved by LiZr₂(PO₄)₃ and Li_0.5Zr₂P_2.5Mo_0.5O₁₂ catalysts (64 and 49%, respectively) in the case of ethanol conversion.

Full text of DOI:10.1016/j.jallcom.2018.03.099

Accepted Manuscript
Conversion of aliphatic C1-C2 alcohols on In-, Nb-, Mo-doped complex lithium
phosphates and HZr (PO ) with NASICON-type structure
2
4 3
A.B. Ilin, M.M. Ermilova, N.V. Orekhova, M. Cretin, A.B. Yaroslavtsev
PII:
S0925-8388(18)30966-6
10.1016/j.jallcom.2018.03.099
JALCOM 45329
DOI:
Reference:
To appear in: Journal of Alloys and Compounds
Received Date: 12 November 2017
Revised Date: 7 March 2018
Accepted Date: 9 March 2018
Please cite this article as: A.B. Ilin, M.M. Ermilova, N.V. Orekhova, M. Cretin, A.B. Yaroslavtsev,
Conversion of aliphatic C1-C2 alcohols on In-, Nb-, Mo-doped complex lithium phosphates and
HZr (PO ) with NASICON-type structure, Journal of Alloys and Compounds (2018), doi: 10.1016/
2
4 3
j.jallcom.2018.03.099.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Conversion of aliphatic C1-C2 alcohols on In-, Nb-, Mo-doped complex
lithium phosphates and HZr2(PO4)3 with NASICON-type structure
A. B. Ilin*, **, M. M. Ermilova*, N. V. Orekhova*, M. Cretin***, A. B. Yaroslavtsev*, **
*A.V. Topchiev Institute of Petrochemical Synthesis of the Russian Academy of Sciences, 119991
Moscow 29 Leninsky prospect, Russia
**N.S. Kurnakov Institute of General and Inorganic Chemistry of the Russian Academy of
Sciences, 119991 Moscow 31 Leninsky prospect, Russia
***Institut Européen des Membranes, IEM UMR-5635, Université de Montpellier, ENSCM,
CNRS Place Eugène Bataillon 34095 Montpellier cedex 5, France
e-mail: yaroslav@igic.ras.ru
Abstract
In-, Nb-, Mo-doped lithium complex phosphates and HZr2(PO4)3 with
NASICON-type structure were synthesized in this paper. Particle size distribution
lies between 50 and 300 nm. The obtained samples were characterized by X-ray
diffraction analysis, scanning electron microscopy and X-ray microanalysis.
Investigation of the catalytic properties of synthesized compounds in the C1-C2
alcohols conversions showed that heterovalent doping has a determining effect on
the obtained catalysts’ activity and selectivity.
It is shown that the thermodynamic factors and the dopant ability to change
the degree of oxidation and acid function of the catalysts play a key role in
methanol and ethanol conversion. A number of catalysts show the high activity and
selectivity of the formation of dimethyl and diethyl ethers and ethylene. High
selectivity for C4 hydrocarbons is achieved by LiZr2(PO4)3 and
Li0.5Zr2P2.5Mo0.5O12 catalysts (64 and 49%, respectively) in the case of ethanol
conversion.
Key words: Nasicon, heterogeneous catalysis, methanol conversion, ethanol
conversion.
1

ACCEPTED MANUSCRIPT
Introduction
The main goal of green chemistry now is the search for environmentally
friendly, renewable energy sources. Biomass is considered as a promising option
and the bioalcohols appear while it`s processing [1]. The main application of the
obtained alcohols is reduced to their use as a fuel, or additives to it [2, 3]. But in
the mass production of such alcohols, it will be expedient to use them also for the
production of industrially important products, such as hydrocarbons, ethers,
aldehydes, ketones, hydrogen, etc. In this regard, it is important to find suitable
catalysts capable of effectively converting alcohols into these compounds.
According to the literature, plenty types of catalysts have been investigated
in ethanol and methanol conversion processes, such as Zeolites [4], clays [5],
mixed metal oxides [6, 7, 8], noble and non-noble metal catalysts on different
supports [9, 10]. In the case of catalyst by metals the main process is steam
reforming, hydrocarbons, olefin and ether formation is typical in the case of zeolite
and ceramic catalysts. Bio-diesel production should be mentioned also in the case
of the use of sulfonic acid and it`s derivatives [11].
The NASICON-type structure compounds (NA+ Super Ionic CONductor)
were chosen as catalysts in this work. Materials belonging to this structural type
are compounds with the general formula AxB2(ZO4)3, where A is usually alkali or
alkaline earth element, B is a polyvalent element (Zr, Ti, Sc etc.), Z is a
phosphorus or silicon, or both immediately. Their structure consists of joined
vertices of BO6 octahedrons and ZO4 tetrahedrons. A-cations are located in the
cavities of this structure [12, 13].
NASICON-type compounds were first pointed be M. Sljukic and co. in 1967
[14]. The interest to these compounds has increased dramatically after Hong and
other scientists’ detection of high ionic conductivity in them [15-17]. This
conductivity is achieved due to the presence of conduction channels in their
structure, in which the A-ions move [13]. Among other advantages of NASICON-
type structure compounds, it is worth to mention their high chemical and thermal
stability, radiation resistance and structure variability, which makes it possible to
2

ACCEPTED MANUSCRIPT
carry out both iso- and heterovalent doping without destroying the original cavity
structure [18, 19]. These features determine a wide range of possible applications
of these compounds as cathode materials in lithium current sources [20-23], ionic
conductors in high-temperature fuel cells [24], in sensors [25-27], in radionuclide
purification [28], in various types of heterogeneous catalysis [29-34], including
membrane catalysis [35, 36] and photocatalysis [37].
The ability to heterovalent substitution is particularly important for their use
as catalysts, since this allows to vary the number and strength of acidic (according
to Lewis and Brønsted) and oxidation-reduction centers on the surface. Chemical
and thermal resistance allow the use of catalysts based on these compounds under
such conditions where metal-based catalysts are poisoned (SO2 presence) or fused.
The aim of this work was to obtain heterovalently doped compounds with
NASICON structure based on LiZr2(PO4)3 and to investigate their activity in the
catalytic conversions of C1-C2 alcohols. It was assumed that the partial
substitution of phosphorus by molybdenum may allow to improve their activity in
redox reactions (dehydrogenation reactions) due to a change in the oxidation state
of molybdenum ions. Partial substitution of zirconium for niobium or phosphorus
for molybdenum can lead to a change in the Lewis acidity of these catalysts, as
well as to a change their ability to form coordination bonds with oxygen atoms of
alcohols. The increase in Brønsted acidity can be achieved due to the substitution
of lithium by protons (HZr2(PO4)3). Varying the composition can allow to change
the catalytic activity in the dehydration and dehydrogenation of the alcohols.
3

ACCEPTED MANUSCRIPT
Material and Methods
In this work, we synthesized compounds of the composition LiZr2(PO4)3,
Li1±0.5Zr1.5M0.5(PO4)3 (M = In or Nb), Li0.5Zr2P2.5Mo0.5O12, and HZr2(PO4)3. The
synthesis of LiZr2(PO4)3, Li1±0.5Zr1.5M0.5(PO4)3 (M = In or Nb) and
Li0.5Zr2P2.5Mo0.5O12 was carried out by the Pechini method [38, 39]. The
stoichiometric sample weight of ZrOCl2*8H2O, citric acid (acid:M = 4:1, where M
= Zr + In or Nb), Li2CO3 and NH4H2PO4 were sequentially dissolved in an alumina
crucible in a mixture of ethylene glycol (2 ml) and deionized water (10 ml). To
obtain Li1.5Zr1.5In0.5(PO4)3, the exact sample weight of In2O3 calculated from the
stoichiometry of the final product was dissolved in a minimum amount of hot
concentrated nitric acid, ethylene glycol and water were added to the resulting
solution, and then the remaining reagents in the above sequence. NbCl5 dissolved
in a minimal amount of concentrated hydrochloric acid was used for
Li0.5Zr1.5Nb0.5(PO4)3 syntheses. The resulting solution was mixed with
ZrOCl2*8H2O solution. A mixture of NH4H2PO4 with (NH4)6Mo7O24 were used
instead of NH4H2PO4 for the Li0.5Zr2P2.5Mo0.5O12 synthesis. After addition of
NH4H2PO4, the pH of the solution was quickly adjusted to 5.5 by adding a
concentrated ammonia solution to prevent precipitation of the zirconium
phosphate. The resulting solution was kept for 24 hours, first at 95°C, 24 hours at
150°C, then 4 hours at 350°C and, after thorough grinding, subjected to final
annealing at 750°C for 10 hours.
HZr2(PO4)3 was synthesized by the hydrothermal method. 1M solutions of
ZrOCl2*8H2O and NH4H2PO4 in a ratio of 1:2 were mixed in an autoclave with
Teflon insert. The autoclave was sealed and kept at 250°C for 24 hours. After that
obtained precipitate was washed from the dissolved impurities and annealed at
500°C for 72 hours.
X-ray diffraction analysis (XRD) was carried out using a Rigaku D/Max-
2200 X-ray diffractometer (CuKα1 − radiation) with thermal attachment. Heating
speed for obtaining high temperature X-ray patterns was 10 K/min. Samples were
held at a predetermined temperature for 5 minutes before measurement. Rigaku
4

ACCEPTED MANUSCRIPT
Application Data Processing software was used to process the diagrams and
qualitative analysis. The Rietveld method refinement was provided by FullProf
Suite software package.
The Scherrer equation was used to estimate the coherent scattering region
(CSR) from the X-ray diffraction lines broadening:
k
λ
d =
;
(1)
B2 −b2 cos
θ
where k = 0.89 is the Scherrer constant, λ = 1.5406 Å is the wavelength of the
radiation used, B is the half-width at the half-height of the peak (in 2θ), b is the
instrumental broadening (in 2θ), and θ is the angle of the peak position. The
lanthanum hexaboride powder LaB6 (Standard Reference Material® 660a) was
used as a standard for the determination of instrumental broadening.
The specific surface area was determined by the BET method using the
Micromeritics ASAP 2020. The samples were heated to a temperature of 350°C
and held for 1 hour for degassing. The analysis was carried out in the region of
relative pressures 0.01 ≤ p/po ≤ 0.99 in nitrogen atmosphere.
The micrographs of the samples were obtained by a scanning electron
microscope (SEM) Carl Zeiss NVision 40 with an attachment for X-ray
microanalysis. The accelerating voltage was 1 kV.
The FTIR spectra were recorded by Nicolet Protégé 380 FTIR spectrometer
at 4 cm-1 optical resolution. 20 mg of sample powder were pressed in a tablet and
placed in a heating cell, attached to the spectrometer and vacuum line. Activation
was performed in the vacuum at 400oC for 4 hours. The adsorption of pyridine was
performed at room temperature, partial pressure of pyridine was 2 torr. Spectra
were recorded after treatment of the pyridine exposed samples at 150oC in the
vacuum to remove weakly sorbed pyridine. All spectra are presented after
subtraction of the spectrum of activated sample. The subtraction of spectra was
performed using OMNIC 7.3 software package.
The catalytic properties of the catalysts synthesized were studied in a
conventional flow-type quartz reactor in a helium or argon streamline with an
5

ACCEPTED MANUSCRIPT
admixture of alcohol vapors. The reaction products were analyzed qualitatively and
quantitatively on a Crystallux 4000M chromatograph with a thermal conductivity
detector on HayeSep T 60/80 mesh (3 m, 150°C, 30 ml/min, He), SKT-6 (3 m,
150°C, 30 ml/min, He), Mole Seive 5 A (3 m, 25°C, 10 ml/min, Ar) columns. All
chromatograms were processed by using the NetChrom software package. The
error of chromatographing method was ± 5%.
For the experiment, a 0.3 g catalyst was mixed with ground quartz (d = 250-
500 ꢀm) and placed in a quartz reactor tube with a length of 25 cm with an internal
diameter of 6 mm so that the length of the catalytic layer was 10 cm. Methyl and
ethyl alcohols were used as reagents. The carrier gas was passed through a
thermostatic bubbler with an alcohol at a flow rate of 20 ml/min to create the
desired concentration of alcohol vapors in the feed gas mixture. The bubbler
temperature was 0 and 11°C for methyl and ethyl alcohols, respectively.
The conversion was calculated using the following equation:
ϕini
−ϕcurr
X =100×
;
(2)
ϕini
where φini/curr – the initial and current volume fraction of alcohol vapors in the feed
mixture, respectively.
Selectivity was calculated as following:
ϕ prod nprod
S = 100×
;
(3)
(
ϕini −ϕcurr ) nalc
where φprod – product volume fraction; nprod and nalc – the number of carbon atoms
in the product and alcohol molecule, respectively (in order to take in account the
carbon balance).
To calculate the productivity (mmol/h*g), the following equation was used:
X S ϕini
v
A =
;
(4)
10000 m Vm
6

ACCEPTED MANUSCRIPT
where v – volumetric rate of the feeding alcohol-containing mixture (cm3/h); m is
the sample weight of the catalyst (grams); Vm – molar volume (cm3/mmol); 10000
– is the coefficient that converts volume percentages into volume parts for X and S.
Discussion of the results
Characterization of the obtained samples
All obtained samples are single-phase. The samples LiZr2(PO4)3,
Li1.5Zr1.5In0.5(PO4)3, Li0.5Zr1.5Nb0.5(PO4)3 and Li0.5Zr2P2.5Mo0.5O12 obtained by the
Pechini method are monoclinic, the sample HZr2(PO4)3 obtained by hydrothermal
synthesis is rhombohedral (Fig. 1). The X-Ray microanalysis data correspond to
the expected composition within the error ranges. Refinement by Rietveld method
gives satisfactory description of X-ray pattern of monoclinic lattice. The example
of such description for LiZr2(PO4)3 with Bragg R-factor = 4.56 is given on Fig. 2.
Figure 1 is here
Figure 2 is here
X-ray pattern of LiZr2(PO4)3 obtained at high temperatures differ from those,
obtained at room temperature. It`s caused by phase transition from monoclinic
structure into orthorhombic. This phase transition occurs in the temperature range
290 - 350oC (Fig. 3).
Figure 3 is here
According to the scanning electron microscopy all the obtained samples
contain particles with the average size from 50 to 200 nm (Fig. 4).
The size of the CSR of all the samples obtained from XR-data, is somewhat
smaller than the particle size estimated from the SEM data (Table 1), due to the
agglomeration. All these samples have approximately the same specific surface
7

ACCEPTED MANUSCRIPT
area about 20 m2/g (Table 1). HZr2(PO4)3 obtained hydrothermally consists of
particles whose shape is close to cubic with a clear facet (Fig. 4b).
Figure 4 is here
Table 1 is here
Infrared study of the prepared samples indicated vibration bands at ~ 1610,
1545 and 1450 cm-1, which are correspond to v8a v(CC) Lewis, v19b v(CN) Brønsted
and v19b v(CN) Lewis vibration bands [40] (Fig. 5). LiZr2(PO4)3 and
Li1.5Zr1.5In0.5(PO4)3 possess only Lewis acid sites, Li0.5Zr1.5Nb0.5(PO4)3,
Li0.5Zr2P2.5Mo0.5O12 and HZr2(PO4)3 showed both Lewis and Brønsted acid sites. In
case of HZr2(PO4)3 the intensity of vibration bands is small, but this could be
caused by lower specific surface area of this sample (10 m2/g vs 17-21 m2/g).
Figure 5 is here
By measuring the shift of the 19b Lewis vibration band we can estimate the
strength of the Lewis acid sites. It forms the following row: LiZr2(PO4)3 (1446 cm-
1) < Li0.5Zr1.5Nb0.5(PO4)3 (1448 cm-1) ≈ Li0.5Zr2P2.5Mo0.5O12 (1448 cm-1) ≈
HZr2(PO4)3 (1448 cm-1) < Li1.5Zr1.5In0.5(PO4)3 (1451 cm-1). 19b Brønsted to 19b
Lewis peak area ratio for Li0.5Zr1.5Nb0.5(PO4)3, Li0.5Zr2P2.5Mo0.5O12, HZr2(PO4)3 are
following: 1:2.6, 1:6.7, 1:0.84 (In sample HZr2(PO4)3, the v19b v(CN) Brønsted line
is very wide). The indium-doped sample has the stronger Lewis acid sites than the
other samples and might be active in dehydrogenation processes. HZr2(PO4)3 has
the lowest amount of acid sites but the Brønsted to Lewis peak area ratio for this
sample is highest, so it might show activity in the dehydration processes.
Li0.5Zr1.5Nb0.5(PO4)3 and Li0.5Zr2P2.5Mo0.5O12 might be active in both
dehydrogenation and dehydration processes because of significant amount of
Lewis and Brønsted acid sites, but for Li0.5Zr2P2.5Mo0.5O12 the Brønsted to Lewis
8

ACCEPTED MANUSCRIPT
peak area ratio is the lowest, so the activity in dehydrogenation processes for this
sample might be more pronounced.
Catalytic properties
All synthesized compounds showed activity in the alcohol’s conversion
(Fig. 6). In the case of ethanol conversion, the catalytic activities of all samples
were really similar, only Li0.5Zr2P2.5Mo0.5O12 has some advantages, the use of
which allows to achieve the same conversion values at temperatures 30-50oC lower
than for other catalysts. In methanol conversion the difference in activity is much
distinctly.
Li0.5Zr2P2.5Mo0.5O12 catalyst has the highest activity in methanol conversion.
Li0.5Zr1.5Nb0.5(PO4)3 and HZr2(PO4)3 slightly inferior to Li0.5Zr2P2.5Mo0,5O12,
whereas Li1.5Zr1.5In0.5(PO4)3 and especially LiZr2(PO4)3 are significantly less
active.
Figure 6 is here
Methanol conversion
Since the methanol molecule is fairly simple, its transformations can start
with two main reactions:
2CH3OH ↔ CH3OCH3 + H2O
CH3OH ↔ CH2O + H2
ꢁH0298 K = −23.9 kJ/mole
ꢁH0298 K = 85.1 kJ/mole
(5)
(6)
Comparing the thermodynamics of these processes, we can conclude that the
most probable process at low temperatures is the dehydration of methanol with the
formation of dimethyl ether. The dehydrogenation of methanol with the formation
of formaldehyde also becomes possible at elevated temperatures (>500оС).
Accordingly, the main product is dimethyl ether for all catalysts at low
temperatures (Fig. 7). However, only in the case of HZr2(PO4)3, the selectivity of
dehydration process remains close to 100% in the entire temperature range (Fig. 7
e). The same high selectivity at lower temperatures is also characteristic for
9

ACCEPTED MANUSCRIPT
Li0.5Zr1.5Nb0.5(PO4)3 and Li0.5Zr2P2.5Mo0.5O12 catalysts. Methanol conversion
reaches values close to 82-85% (Fig. 6 a), which correspond to equilibrium yields
and, therefore, are the highest for reaction (5). It should be noted that the same
conversion values are also reached for HZr2(PO4)3, for which this reaction
dominates over the entire temperature range. However, above 420 and 350°C,
significant amounts of methane and CO are formed on these catalysts (Fig. 7 c, d),
and conversion reaches 100% for Li0.5Zr1.5Nb0.5(PO4)3 and Li0.5Zr2P2.5Mo0.5O12.
This corresponds to the fact that at high temperatures dehydrogenation processes
take place, which become not thermodynamically banned.
Figure 7 is here
The most effective in dimethyl ether formation are catalysts with
Li0.5Zr2P2.5Mo0.5O12, Li0.5Zr1.5Nb0.5(PO4)3 and HZr2(PO4)3 composition. The
maximum productivity of this process was about 2.5 mmol/h*g for
Li0.5Zr1.5Nb0.5(PO4)3 and HZr2(PO4)3 (Fig. 8).
Figure 8 is here
The significant contribution is made by secondary reactions associated with
the conversion of dimethyl ether for LiZr2(PO4)3 and Li1.5Zr1.5In0.5(PO4)3 even at
low temperatures. These processes can proceed with the formation of
formaldehyde, methane, hydrogen, carbon and carbon oxides (Fig. 7 a, b), for
example, according to the following equations:
CH3OCH3 ↔ CH4 + CO + H2
CH3OCH3 ↔ CH4 + CH2O
CH3OCH3 ↔ CH4 + C + H2O
ꢁH0298 K = -1.3 kJ/mole
ꢁH0298 K = -6,7 kJ/mole
ꢁH0298 K = -132.6 kJ/mole
(7)
(8)
(9)
The contribution of these processes is significant for such active catalysts as
Li0.5Zr1.5Nb0.5(PO4)3 and Li0.5Zr2P2.5Mo0.5O12 at elevated temperatures. Methanol
10

ACCEPTED MANUSCRIPT
conversion in these systems increases with temperature and approaches 100% (Fig.
7 c, d).
One could assume that the high activity of Li0.5Zr2P2.5Mo0.5O12 and
Li0.5Zr1.5Nb0.5(PO4)3 is associated with the possibility of changing the degree of
niobium and molybdenum oxidation. This effect should be less pronounced in the
case of Li1.5Zr1.5In0.5(PO4)3, what is actually observed in the experiment. In
addition, reaction (5) is dominant even for Li0.5Zr2P2.5Mo0.5O12 and
Li0.5Zr1.5Nb0.5(PO4)3 at low temperatures, which can occur, for example, due to
proton binding by negatively charged lithium vacancies. These vacancies are
formed by the intercalation of niobium and molybdenum cations having a higher
charge in comparison with substituted zirconium and phosphorus. An alternative
mechanism may be performed due to the adsorption of methanol molecules by
these cations because of their lower coordination numbers in oxygen (6 and 4
respectively), which are far from limit [41]. In this case, the catalyst acts as a
Lewis acid. In our opinion, this process is most probable. HZr2(PO4)3 is an
essentially more active catalyst in comparison with LiZr2(PO4)3 (Fig. 6 a). The
increased Brønsted acidity is the only difference (Fig. 5). At the same time, since
the activity of Li0.5Zr2P2.5Mo0.5O12 and Li0.5Zr1.5Nb0.5(PO4)3 is even higher, it is
precisely the Lewis acidity that is preferred for the dehydration reactions, the
presence of which in samples doped with niobium is shown in [42]. Thus, it can be
concluded that the main process in the first stage of dehydration is the adsorption
of oxygen ions of methanol on polycharging cations having an unsaturated
coordination.
It can also be noted that when methanol is converted on Li1.5Zr1.5In0.5(PO4)3
(Fig. 7 b) up to 34% of formaldehyde is formed according to the reaction (6). The
activity of this catalyst in dehydration processes is significantly lower due to the
Lewis acidity suppression when zirconium is replaced by an ion with a lower
valence. Still lower activity in the dehydration of undoped LiZr2(PO4)3 (Fig. 6 a)
seems to indicate that the reaction is possible, although less efficiently with basic
catalysis.
11

ACCEPTED MANUSCRIPT
The reactions of the alcohols conversion on these catalysts should not be
considered as process for the hydrogen production. Its production is low and
substantially inferior to those for the processes of steam reforming. The highest
activity has Li1.5Zr1.5In0.5(PO4)3 in the hydrogen formation by methanol conversion.
But even for this catalyst the activity at such high temperature as 510°C reaches
only 0.83 mmol/h*g.
Ethanol conversion
More diverse products are characteristic for the ethyl alcohol conversion,
during which the following reactions are possible:
2C2H5OH →C2H5OC2H5 + H2O
C2H5OH → C2H4 + H2O
∆H0298 = -24,4 kJ/mole
∆H0298 = + 45,3 kJ/mole
∆H0298 = + 68,8 kJ/mole
∆H0298 = + 49,4 kJ/mole
∆H0298 = + 255,6 kJ/mole
∆H0298 = + 173,2 kJ/mole
(10)
(11)
(12)
(13)
(14)
(15)
C2H5OH → CH3CHO + H2
C2H5OH → CH4 + CO + H2
C2H5OH + H2O → 2CO + 4H2
C2H5OH + 3H2O → 2CO2 + 6H2
It is noteworthy that, from the point of view of thermodynamics, all these
processes can occur at relatively low temperatures, but the most possible process
must be a dehydration which leads to the diethyl ether formation (10). Minor
amounts of methane, carbon dioxide and carbon monoxide were also observed. In
addition, the formation of heavier hydrocarbons is possible:
xC2H5OH → yCnHm + xH2O + H2, n≥2
(16),
and products decomposition processes, accompanied by the catalyst coking, for
example, according to the following reaction:
CH3CHO → 2C + H2 + H2O ∆H0298 = - 75,8 kJ/mole
(17)
12

ACCEPTED MANUSCRIPT
The main products during the ethyl alcohol conversion were: diethyl ether,
acetaldehyde, hydrogen, C2- and C4-hydrocarbons. The dehydration of ethanol with
the formation of diethyl ether at low temperatures was the dominant process on
LiZr2(PO4)3 and Li0.5Zr1.5Nb0.5(PO4)3 catalysts (Fig. 9 a, c). The selectivity of this
process reaches 90-91% at temperatures of about 300-330°C. The significant
number of secondary processes take place on all catalysts at elevated temperatures.
Figure 9 is here
It is noteworthy that the heterovalent doping effect is most clearly
manifested in the ethyl alcohol conversion. Thus, partial substitution of zirconium
by indium or phosphorus for molybdenum significantly increases the
dehydrogenating properties of the catalyst, and at low temperatures acetaldehyde
formation predominates (Fig. 9 b, d). Obviously, in the case of Li0.5Zr2P2.5Mo0.5O12
this is explained by the easier change of molybdenum degree of oxidation. Indium
can also change its oxidation state on 1+, although this is less characteristic for it.
The indium ion has a smaller positive charge compared to zirconium, so it will be
negatively charged relative to the residual crystal structure. Therefore, such a
defect will have tendency to adsorb of the alkyl side of the alcohol rather than it`s
hydroxyl group. This, in turn, should facilitate the processes, associated with
abstraction of the proton and intensification of the dehydrogenation processes. At
the same time, at elevated temperatures this process proceeds easily. In addition, in
the case of indium, as noted above, the ability of the material to adsorb alcohols
oxygen is suppressed, which complicates the dehydration processes course. It
should be noted that in the case of Li0.5Zr1.5Nb0.5(PO4)3 the deeper dehydration
reaction becomes dominant at the temperatures above 390°C, which leads to the
ethylene formation (reaction (11), Fig. 9 c). It is interesting that this process is also
typical for HZr2(PO4)3. In this case it is dominant for all temperatures (Fig. 9 d).
Apparently, this is due to the high Brønsted acidity of this material.
13

ACCEPTED MANUSCRIPT
Finally, it is worth noting that the possibility of obtaining C4-hydrocarbons
seems very attractive. This process proceeds on LiZr2(PO4)3 and
Li0.5Zr2P2.5Mo0.5O12, and the selectivity in this case reaches 73% and 49%,
respectively (Fig. 9 a, d). Even less selectivity - no more than 35% is typical for
Li1.5Zr1.5In0.5(PO4)3 (Fig. 9 b). Equation (16) shows that both dehydrogenation and
dehydration should take place in the case of CnHm (n> 2) formation. The relatively
high activity of LiZr2(PO4)3 suggests that zirconium itself can stimulate oxidation-
reduction transformations at elevated temperatures. The continuation of this
transformations chain is the catalysts coking, which most actively occurs on
Li1.5Zr1.5In0.5(PO4)3 and Li0.5Zr2P2.5Mo0.5O12, which leads to a decrease in the total
selectivity (Fig. 9 b, d).
The Li1.5Zr1.5In0.5(PO4)3 sample has the highest activity in hydrogen
formation and this fact is in accordance with the assumption described above. It
should be noted that the amount of hydrogen formed on it exceeds the amount of
acetaldehyde formed by reaction (12) (Fig. 10). This may be explained by
additional generation of hydrogen by secondary reactions, accompanied, as noted
above, by catalyst coking.
Figure 10 is here
Conclusions
The synthesis of materials with the NASICON structure doped by
heterovalent elements ((LiZr2(PO4)3, Li1.5Zr1.5In0.5(PO4)3, Li0.5Zr1.5Nb0.5(PO4)3,
Li0.5Zr2P2.5Mo0.5O12, HZr2(PO4)3) were carried out by using Pechini and
hydrothermal methods. The obtained samples were characterized by X-ray phase
analysis, and transmission electron microscopy. The characteristic particle size of
the obtained compounds was 50-300 nm. The specific surface area for Pechini
samples
(LiZr2(PO4)3,
Li1.5Zr1.5In0.5(PO4)3,
Li0.5Zr1.5Nb0.5(PO4)3,
Li0.5Zr2P2.5Mo0.5O12) was about 20 m2/g, and for the hydrothermal sample
(HZr2(PO4)3) was 10 m2/g.
14

ACCEPTED MANUSCRIPT
All synthesized samples were catalytically active in the C1-C2 alcohols
conversion. The main products in the methanol conversion were dimethyl ether,
methane, formaldehyde, CO and hydrogen. For ethanol conversion were C2-C4
hydrocarbons, acetaldehyde, diethyl ether and hydrogen.
It should be noted the high activity and selectivity of a number of catalysts
in the dimethyl and diethyl ethers formation, as well as the ethylene formation. The
high selectivity for C4 hydrocarbons is achieved for LiZr2(PO4)3 and
Li0.5Zr2P2.5Mo0.5O12 at elevated temperatures, reaching 64 and 49% respectively.
It is shown that the main directions of the process of methanol and ethanol
conversion is determined by thermodynamic factors, as well as the heterovalent
doping and ability of the doped polyvalent elements to change its degree of
oxidation and their adsorption of alcohols through the oxygen atom with the Lewis
acidity manifestation.
Acknowledgment
This investigation was realized in the frame of a joint French-Russian PHC
Kolmogorov 2017 project of the French-Russian International Associated
Laboratory on “Ion-exchange membranes and related processes” with the financial
support of Minobrnauki (Ref. N°RFMEFI58617X0053) Russia and CNRS, France
(project N° 38200SF).
The author is also grateful to the Institute of General and Inorganic
Chemistry for photomicrography shooting of synthesized samples.
15

ACCEPTED MANUSCRIPT
References
[1] K. Waldron (Ed.), Bioalcohol production: Biochemical conversion of
lignocellulosic biomass, Woodhead Publishing Limited and CRC Press LLC, 2010,
p. 496.
[2] B. Rajesh Kumar, S. Saravanan, Use of higher alcohol biofuels in diesel
engines: A review. Renew. and Sust. Energy Rev. 60 (2016) 84-154.
[3] X. Zhen, Y. Wang, An overview of methanol as an combustion engine fuel.
Renew. and Sust. Energy Rev. 52 (2015) 477-493.
[4] T.K. Phung, R. Radikapratama, G. Garbarino, A. Lagazzo, P. Riani, G. Busca,
Turning of product selectivity in the conversion of ethanol to hydrocarbons over
H-ZSM-5 based zeolite catalysts. Fuel Process. Technol. 137 (2015) 290-297.
[5] J.J. Lovón-Quintana, J.K. Rodriguez-Guerrero, P.G. Valença, Carbonate
hydroxyapatite as a catalyst for ethanol conversion to hydrocarbon fuels. Appl.
Catal., A. 542 (2017) 136-145.
[6] F. Wang, W. Xia, X. Mu, K. Chen, H. Si, Z. Li, A combined experimental and
theoretical study on ethanol conversion to propylene over Y/ZrO2 catalyst. Appl.
Surf. Sci. 439 (2018) 405-412.
[7] W. Xia, F. Wang, X. Mu, K. Chen, Remarkably enhanced selectivity for
conversion of ethanol to propylene over ZrO2 catalysts. Fuel Process. Technol. 166
(2017) 140-145.
[8] A.S. Rocha, A.M. da S. Forrester, E.R. Lachter, E.F. Sousa-Aguiar, A.C. Faro
Jr., Niobia-modified aluminas prepared by impregnation with niobium peroxo
complexes for dimethyl ether production. Cat. Тоday. 118 (2012) 104-111.
[9] R. Baruah, M. Dixit, P. Basarkar, D. Parikh, A. Bhargav, Advances in ethanol
autothermal reforming. Renew. and Sust. Energy Rev. 51 (2015) 1345-1353.
[10] Y.C. Sharma, A. Kumar, R. Prasad, S.N. Upadhyay, Ethanol steam reforming
for hydrogen production: Latest and effective catalyst modification strategies to
minimize carbonaceous deactivation. Renew. and Sust. Energy Rev. 74 (2017) 89-
103.
[11] N. Mansir, Y.H. Taufiq-Yap, U. Rashid, I.M. Lokman, Investigation of
heterogeneous solid acid catalyst performance on low grade feedstocks for
biodiesel production: A review. Energy Convers. Manage. 141 (2017) 171-182.
16

ACCEPTED MANUSCRIPT
IV
[12] L.O. Hagman and P. Kierkegaard, The Crystal Structure of NaMe2 (PO4)3;
MeIV = Ge, Ti, Zr, Acta Chem. Scand., 22 (1968) 1822.
[13] H. Kohler, H. Schulz, NASICON solid electrolytes part I: The Na+-diffusion
path and its relation to the structure, Mat. Res. Bull. 20 № 12 (1985) 1461–1471.
[14] M. Sljukic, B. Matkovic, B. Prodic, S. Scavnicar, Preparation and
crystallographic data of phosphates with common formula MIMIV (PO4)3 (MI=Li,
2
Na, K, Rb, Cs; MIV=Zr, Hf). Croat. Chem. Acta. 39 (1967) 145-148.
[15] H.Y.-P. Hong, Crystal structures and crystal chemistry in the system
Na1+xZr2SixP3-xO12, Mater. Res. Bull. 11 № 2 (1976) 173–182.
[16] P. Agaskar, R. Grasselli, D. Buttrey, B. White, Structural and Catalytic
Aspects of Some NASICON - Based Mixed Metal Phosphates, Stud. Surf. Sci.
Catal. 110 (1997) 219.
[17] H.Y.-P. Hong, J.A. Kafalas, M.L. Bayard, High Na+-ion conductivity in
Na5YSi4O12, Mat. Res. Bull. 13 № 8 (1978) 757–761.
[18] I.A. Stenina, A.B. Yaroslavtsev, Complex phosphates with NASICON
structure (AXB2(РO4)3), Z. Neorg. Khim. Suppl. 51 (2006) 95–114.
[19] V.I. Pet’kov, Complex phosphates formed by metal cations in oxidation states
I and IV, Russ. Chem. Rev. 81 № 7 (2012) 606-637.
[20] K.S. Nanjundaswamy, A.K. Padhi, J.B. Goodenough, S. Okada, H. Ohtsuka,
H. Arai, J. Yamaki, Synthesis, redox potential evalution and electrochemical
characteristics of NASICON-related-3D framework compounds, Solid State Ionics.
92 №1-2 (1996) 1-10.
[21] G.-A. Nazri, G. Pistoia, Lithium Batteries: Science and Technology, Springer,
2004 p. 708.
[22] D. Morgan, G. Ceder, M.Y. Saidi, J. Barker, J. Swoyer, H. Huang,
G. Adamson, Experimental and computational study of the structure and
electrochemical properties of monoclinic LixM2(PO4)3 compounds, J. Pow. Sour.
119-121 (2003) 755-759.
17

ACCEPTED MANUSCRIPT
[23] M.Y. Saidi, J. Baker, H. Huang, J.L. Swoyer, G. Adamson, Performance
characteristics of lithium vanadium phosphate as a cathode material for lithium-ion
batteries, J. Pow. Sour. 119-121 (2003) 266-272.
[24] A.D. Robertsona, A.R. Westa, A.G. Ritchieb, Review of crystalline lithium-
ion conductors suitable for high temperature battery applications, Solid State
Ionics. 104 №1-2 (1997) 1-11.
[25] M. Cretin, P. Fabry, Comparative study of lithium ion conductors in the
system Li1+xAlxA2-xIV(PO4)3 with AIV = Ti or Ge and 0≤x≤0.7 for use as Li+
sensitive membranes, J. Europ. Ceram. Soc. 19 №16 (1999) 2931-2940.
[26] M. Cretin, H. Khireddine, P. Fabry, NASICON structure for alkaline ion
recognition, Sensors and Actuators. 43 №1-3 (1997) 224-229.
[27] T. Kida, H. Kawate, K. Shimanoe, N. Miura, N. Yamazoe, Interfacial
structure of NASICON-based sensor attached with Li2CO3 – CaCO3 auxiliary
phase for detection of CO2, Solid State Ionics. 136-137 (2000) 647–653.
[28] R.D. Ambashta, M.E.T. Sillanpää, Membrane purification in radioactive
waste management: a short review, Journal of Environmental Radioactivity. 105
(2012) 76-84.
[29] P.C. Yao, D.J. Fray, The preparation and properties of the solid state ionic
conductor, CuZr2(РО4)3, Solid State Ionics. 8 № 1 (1983) 35-42.
[30] G. Le Pollès, A. Еl Jazouli, R. Olazcuaga, J.M. Dance, G. Le F1em,
P. Hagenmuller, On а new hydrogenated Сu(I) Nasicon-type phosphate
Н0.5СuI0.5Zr2(РО4)3, Маt. Res. Вull. 22 № 9 (1987) 1171-1177.
[31] P.A. Agaskar, R.K. Grasselli, D.J. Вuttrey, B. White, Structural and Catalytic
Aspects of Some NASICON - Based Mixed Metal Phosphates, Stud. Surf. Sci.
Catal. 110 (1997) 219-226.
[32] A. Serghini, R. Brochu, M. Ziyad, M. Loukah, J. Vedrine, Behaviour of
Copper-Zirconium Nasicon-type Phosphate, CuIZr2(РО4)3, in the Decomposition of
Isopropyl Alcohol, J. Chem. Soc. Faraday Trans. 87 № 15 (1991) 2487-2491.
18

ACCEPTED MANUSCRIPT
[33] A. Serghini, R. Brochu, M. Ziyad, J. Vedrine, Synthesis, characterization and
catalytic behaviour of Сu0.5М2(РО4)3 (М = Zr, Sn, Ti), J. Аlloy Compd. 188 (1992)
60-64.
[34] Y. Brik, M. Kacimi, F. Воzоn-Verduraz, M. Ziyad, Characterization of active
sites on AgHf2(РО4)3 in butan-2-ol conversion, Microporous and Mesoporous
Materials. 43 №1 (2001) 103-112.
[35] G. Tereshchenko, N. Orekhova, M. Ermilova, A. Malygin, A. Orlova,
Nanostructured phosphorus-oxide-containing composite membrane catalysts, Cat.
Тоday. 118 №1-2 (2006) 85-89.
[36] M.M. Ermilova, N.V. Orekhova, G.F. Tereshchenko, A.A. Malygin,
A.A. Malkov, A. Basile, F. Gallucci, Methanol oxidative dehydrogenation on
nanostructured vanadium-containing composite membranes, J. Membr. Sci. 317
№1-2 (2008) 88-95.
[37] J. Fu, Photocatalytic activity of glass ceramics containing Nasicon-type
crystals, Маt. Res. Bull. 48 № 1 (2013) 70-73.
[38] M.P. Pechini, US Patent 3,330,697 A (1967).
[39] M. Kakihana, M. Yoshimura, Synthesis and characterization of complex
multicomponent oxides prepared by polymer complex method, Bull. Chem. Soc.
Jpn. 72 (1999) 1427–1443.
[40] M. Che, J.C. Vedrine. Characterization of Solid Materials and Heterogeneous
Catalysts. Vol. 1, WILEY-VCH Verlag & Co. KGaA, Weinheim, 2012, pp. 24-26.
[41] R.D. Shannon, Revised effective ionic radii and systematic studies of
interatomic distances in halides and chalcogenides, Acta Cryst. A32 (1976) 751-
767.
[42] A.B. Ilin, N.V. Orekhova, M.M. Ermilova, A.B. Yaroslavtsev, Catalytic
activity of LiZr2(PO4)3 nasicon-type phosphates in ethanol conversion process in
conventional and membrane reactors, Catal. Today. 268 (2016) 29-36.
19

ACCEPTED MANUSCRIPT
Table 1. Specific surface area and average particle size of the synthesized samples.
Fig. 1. X-ray patterns of samples obtained by the hydrothermal (3) and
Pechini (4-7) methods. Line-diagram for monoclinic LiZr2(PO4)3 Card №70-5819
(1), line-diagram for rhombohedral HZr2(PO4)3 Card №38-0004 (2), X-ray patterns
for HZr2(PO4)3 (3), LiZr2(PO4)3 (4), Li1.5Zr1.5In0.5(PO4)3 (5), Li0.5Zr1.5Nb0.5(PO4)3
(6), Li0.5Zr2P2.5Mo0.5O12 (7).
Fig. 2. Rietveld method Refinement of the Pechini synthesized LiZr2(PO4)3
X-ray pattern.
Fig. 3. X-ray patterns obtained at different temperatures for LiZr2(PO4)3
sample.
Fig. 4. Typical micrographs of samples, obtained by Pechini (a) and
hydrothermal (b) methods.
Fig. 5. FTIR spectra of pyridine on the LiZr2(PO4)3 (1), Li1.5Zr1.5In0.5(PO4)3
(2), Li0.5Zr1.5Nb0.5(PO4)3 (3), Li0.5Zr2P2.5Mo0.5O12 (4), HZr2(PO4)3 (5) samples.
Positions of the v8a v(CC) Lewis, v19b v(CN) Brønsted and v19b v(CN) Lewis
vibration bands are presented (a). Enlarged spectra for 3-5 samples (b).
Fig. 6. The conversion temperature dependence of methanol (a) and ethanol
(b) for LiZr2(PO4)3 (1), Li1.5Zr1.5In0.5(PO4)3 (2), Li0.5Zr1.5Nb0.5(PO4)3 (3),
Li0.5Zr2P2.5Mo0.5O12 (4), HZr2(PO4)3 (5).
Fig. 7. Temperature dependencies of methanol conversion selectivity for
LiZr2(PO4)3
(a),
Li1.5Zr1.5In0.5(PO4)3
(b),
Li0.5Zr1.5Nb0.5(PO4)3
(c),
Li0.5Zr2P2.5Mo0.5O12 (d), HZr2(PO4)3 (e) with the formation of CH4, CO, CH2O and
CH3OCH3.
20

ACCEPTED MANUSCRIPT
Fig. 8. The temperature dependence of dimethyl ether productivity for
LiZr2(PO4)3 (1),
Li1.5Zr1.5In0.5(PO4)3
(2),
Li0.5Zr1.5Nb0.5(PO4)3
(3),
Li0.5Zr2P2.5Mo0.5O12 (4), HZr2(PO4)3 (5).
Fig. 9. The selectivity temperature dependence during the ethanol
conversion for LiZr2(PO4)3 (a), Li1.5Zr1.5In0.5(PO4)3 (b), Li0.5Zr1.5Nb0.5(PO4)3 (c),
Li0.5Zr2P2.5Mo0.5O12 (d), HZr2(PO4)3 (e) with the formation of C2- hydrocarbons,
C3- hydrocarbons, C4- hydrocarbons, CH3CHO and C2H5OC2H5.
Fig. 10. The temperature dependence of the hydrogen productivity for
LiZr2(PO4)3 (1),
Li1.5Zr1.5In0.5(PO4)3
(2),
Li0.5Zr1.5Nb0.5(PO4)3
(3),
Li0.5Zr2P2.5Mo0.5O12 (4), HZr2(PO4)3 (5).
21

ACCEPTED MANUSCRIPT
Composition
Specific surface area,
Particle size
m2/g
SEM, nm
CSR, nm
56
LiZr2(PO4)3
Li1.5Zr1.5In0.5(PO4)3
Li0.5Zr1.5Nb0.5(PO4)3
Li0.5Zr2P2.5Mo0.5O12
HZr2(PO4)3
22 ± 1
20 ± 1
21 ± 1
17 ± 1
10 ± 1
138
204
149
261
215
93
67
46
>200

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT