Catalytic Conversion of Isopropanol on a Heteropoly Acid–η-Aluminum Oxide System

Article abstract of DOI:10.1134/S0036024420010082

Abstract: The results from catalytic conversion of isopropanol on a heteropoly acid–η-aluminum oxide system are discussed. It is established that modification increases its activity and selectivity toward oxygen-containing products. Based on a comparative analysis of reaction products, it is shown that the active sites of unmodified η-aluminum oxide consist only of strong Lewis acid sites (LASes) and strong Lewis basic sites (LBSes), represented predominantly by Br?nsted basic sites (BBSes) after modification with a heteropoly acid. The synthesized systems are characterized via X-ray diffraction, scanning electron microscopy, elemental analysis, and IR spectroscopy. The set of obtained results shows that the Keggin structure decomposes as it interacts with the basic oxide to form heteropoly anions. Due to their negative charge, these interact with the electron-accepting Lewis acid sites to form a surface–ligand complex. This then begins to act as a new Br?nsted base, due to its uncompensated negative charge.

Full text of DOI:10.1134/S0036024420010082

ISSN 0036-0244, Russian Journal of Physical Chemistry A, 2020, Vol. 94, No. 2, pp. 301–309. © Pleiades Publishing, Ltd., 2020.
Russian Text © The Author(s), 2020, published in Zhurnal Fizicheskoi Khimii, 2020, Vol. 94, No. 2, pp. 220–229.
CHEMICAL KINETICS
AND CATALYSIS
Catalytic Conversion of Isopropanol on a Heteropoly
Acid–η-Aluminum Oxide System
E. A. Guseinova *, K. Yu. Adzhamov , and S. E. Yusubova
Azerbaijan State University of Oil and Industry, Baku, AZ1010 Azerbaijan
a,
a
b
a
b
Research Institute for Geotechnological Problems of Oil, Gas, and Chemistry,
Azerbaijan State University of Oil and Industry, Baku, AZ1010 Azerbaijan
*e-mail: elvira_huseynova@mail.ru
Received March 4, 2019; revised March 4, 2019; accepted May 14, 2019
Abstract—The results from catalytic conversion of isopropanol on a heteropoly acid–η-aluminum oxide sys-
tem are discussed. It is established that modification increases its activity and selectivity toward oxygen-con-
taining products. Based on a comparative analysis of reaction products, it is shown that the active sites of
unmodified η-aluminum oxide consist only of strong Lewis acid sites (LASes) and strong Lewis basic sites
(LBSes), represented predominantly by Brønsted basic sites (BBSes) after modification with a heteropoly
acid. The synthesized systems are characterized via X-ray diffraction, scanning electron microscopy, elemen-
tal analysis, and IR spectroscopy. The set of obtained results shows that the Keggin structure decomposes as
it interacts with the basic oxide to form heteropoly anions. Due to their negative charge, these interact with
the electron-accepting Lewis acid sites to form a surface–ligand complex. This then begins to act as a new
Brønsted base, due to its uncompensated negative charge.
Keywords: isopropanol, η-aluminum oxide, heteropoly acid, catalysis
DOI: 10.1134/S0036024420010082
INTRODUCTION
has found the most extensive application as an adsor-
bent, a support, and a component of the most widely
used catalysts. It is also employed in different oil refin-
ing and petrochemical processes (e.g., reforming,
hydrotreatment, and hydrocracking). The most widely
used modifications are γ-, η-, θ-, and χ-aluminum
oxides, since they have developed surfaces and differ
in their catalytic properties. Despite the numerous
studies of aluminum oxide that have been performed,
this question in many cases remains in the focus of
attention.
Due to their set of valuable properties, heteropoly
acids (HPAs) are used as heterogeneous catalysts of
oxidative and acid-catalyzed reactions, and as bifunc-
tional catalysts that have oxidative and acidic proper-
ties simultaneously [1–5]. Because of the small spe-
2
cific surface areas (1–5 m /g) of these catalysts, how-
ever, their use in heterogeneous catalysis is often
limited. There have been works in which heteropoly
acids were deposited onto different supports (SiO ,
2
Al O , ZrO , TiO ), allowing compositions of new
2
3
2
2
It is therefore of undoubted interest to study spe-
cific features of the effect modifying η-aluminum
oxide with a heteropoly acid has on its catalytic prop-
erties in the process of isopropanol conversion. The
selection of the η-Al O polymorph as an object of
catalysts to be proposed, particularly for the deep
hydrotreatment of straight-run and secondary diesel
fractions and heavy vacuum gasoil, the selective
hydrotreatment of catalytic cracking gasolines, plant
oil, and diesel fractions, the isomerization of n-
alkanes, and the synthesis of oxygen-containing com-
pounds (OCCs) [6–10].
2
3
study is due to it having the largest specific surface area
[
20], which is greatest in the η-Al O > γ-Al O >
2 3 2 3
χ + γ)-Al O > δ-Al O > θ-Al O > κ-Al O >
(
2
3
2
3
2
3
2
3
In connection with the prospects for using hetero-
poly acids in heterogeneous catalysis, the question
naturally arises of immobilizing them on the surfaces
of other usually used supports (e.g., η-aluminum
oxide) with subsequent study of their catalytic activity.
A great many original works and reviews have been
devoted to different aspects of the industrial use of alu-
minum oxide [11–19]. Being thermally stable, easily
α-Al O series of aluminum oxides and is one of the
most important factors in modifying with a heteropoly
acid.
2
3
EXPERIMENTAL
The HPA on η-Al O catalyst with a specific com-
2 3
regenerable, and highly adsorptive, aluminum oxide position was prepared by impregnating aluminum
301

3
02
GUSEINOVA et al.
The surface morphology and elemental composi-
6
5
4
3
2
1
0
0
0
0
0
0
0
tion of samples were studied via scanning electron
microscopy (SEM) on a JEOL JSM-6610 LV micro-
scope that had a lattice resolution of 0.14 nm and a
built-in EDS Genesis 4000 EDX analyzer, using a
Si(Li) detector at an accelerating voltage of 100 kV.
Samples for scanning electron microscopy were fixed
on carbon-coated copper grids. The statistical diame-
ter of the particles was measured by analyzing SEM
micrographs of each sample.
3'
3
2'
2
1
1
'
RESULTS AND DISCUSSION
Catalytic Properties
1
3
5
7
9
11
13
15
τ, h
Isopropanol was chosen as a model feedstock
because the mechanisms of its dehydration and dehy-
drogenation are thoroughly studied, and there are a
great volume of accumulated practical data, and the
possibility of testing the effect the chemical composi-
tion of the modifier had on the course of the reaction
Fig. 1. (Color online) Effect of process time τ on conver-
sion γ of isopropanol in the presence of η-Al O at tem-
2
3
peratures of (1) 150, (2) 200, and (3) 250°C. Dashed lines
are the corresponding trends at temperatures of (1′) 150,
(
2′) 200, and (3′) 250°C.
[21–24].
oxide (dried at 200°C for 3 h) with a HPA solution in
aqueous ammonia (HPA on η-Al O ). The com-
The results on the change in the conversion of iso-
2
3
propanol in the presence of η-Al O at different pro-
2
3
pounds used in the process were
cess temperatures and times are presented in Fig. 1. As
(
1) bayerite (calcined for 3 h at 300°C to form can be seen, η-Al O began to display catalytic activity
2
3
η-Al O );
2
3
at 150°C: maximum alcohol conversion was 36.2%.
Raising the temperature to 200°C led to a 7% increase
in alcohol conversion (up to 43.1%). The maximum
feedstock conversion of 51.8% was observed at a tem-
perature of 250°C.
(
2) aqueous ammonia (GOST 3760-79; ammonia
content, no less than 10 wt %; concentration, 90–
1
00 g NH /L);
3
(
3) phosphomolybdic heteropolyacid (HPA)
Н [P(Mo O ) ] · Н О; and
As can be seen from the data presented in Fig. 1, a
specific feature of the effect the process time and rest
breaks had on the parameters of the process was
observed at all three experimental temperatures. Every
time the process was started, activation of the catalyst
took 2 h. After repeated activation, the initial parame-
ters were lower than those observed earlier. These were
7
2
7 6
2
(
4) isopropanol (GOST 9805-84).
The catalytic properties of the initial and modified
η-Al O were studied in a flow reactor in the tempera-
2
3
ture range of 150–250°C at a feedstock hourly space
–1
velocity of 75 h .
The feedstock and liquid reaction products were reached again after 2 h of operation, and the catalyst
analyzed via chromatography on a CHROM 5 chro- returned to its steady-state level.
matograph equipped with a flame ionization detector
It should be noted that the catalytic activity exhib-
and a column packed with polyethylene glycol (17%)
on CELITE C-22. Gaseous reaction products were
analyzed on a CHROM 5 chromatograph equipped
with a thermal conductivity detector. The sorbent was
diethylene glycol n-butyrate supported on Inza brick
INZ-600. The chromatograms of the obtained cataly-
zate were analyzed quantitatively via absolute calibra-
tion and internal normalization.
ited by η-Al O for long periods changed in different
2
3
ways, depending on the process temperature. At a
temperature of 150°C, the conversion of alcohol fell
abruptly after 8 h of operation, but its catalytic activity
at higher temperatures remained the same for 14–15 h.
With η-Al O , the conversion of isopropanol fol-
2
3
lowed two paths (Fig. 2): dehydration to propylene via
The infrared spectra were recorded on a Thermo-
mechanism Е1 and to diisopropyl ether via the mech-
Scientific Nicolet IS 10 spectrometer in the 400 to
anism of S 2β substitution (the latter is typical of basic
N
−
1
4
000 cm range of wavenumbers at a resolution of
oxides. Their yield is due to the predominance of one
path or another of the two competing processes, and
−
1
4
.0 cm .
X-ray diffraction analysis was performed on a thus the nature of the acid sites on the catalyst’s sur-
RIGAKU SC-70 diffractometer in the 3°–60° range face. The highest yield of diisopropyl ether was
of angles at an angular scanning rate of 10 deg/min observed at a temperature of 250°C in the 3 to 9 h
using CuK radiation (40 kV, 15 mA).
period of η-Al O catalyst operation.
2 3
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 94
No. 2
2020

CATALYTIC CONVERSION OF ISOPROPANOL
303
1
0
9
8
7
6
5
4
3
2
1
0
60
5
0
0
3
3
3
2
4
3
2
2
2
30
1
1
2
1
0
0
1
0
1
1
2
3
4
5
6
7
8
9
10 11 12
13 14 15 16
τ, h
Fig. 2. (Color online) Effect of process time on the yield of diisopropyl ether (histograms) and propylene (solid lines) in the pres-
ence of η-Al O at temperatures of (1) 150, (2) 200, and (3) 250°C.
2
3
Dependences illustrating the effect process time
Also noteworthy is the stability of catalytic proper-
has on the conversion of isopropanol in the presence ties during the process, though a period of activation
similar to the one mentioned above for the unmodi-
fied analogs was also observed here. In addition, the
modified samples of 5, 7, 10, and 13% HPA exhibited
high selectivity toward oxygen-containing products
of modified HPA/η-Al O catalysts are shown in
2
3
Fig. 3. The obtained results indicate that the pres-
ence of HPA had an inconclusive effect on the cata-
lyst’s activity. The sample with the low HPA content
(
3
Table 1). The selectivity of the sample of 7% HPA is
(
5%) is characterized by low conversion of alcohol,
.5 times higher than for the unmodified sample.
but the picture is reversed abruptly when the content
of HPA in the catalyst is raised to 7–10%. The max-
imum feedstock conversion is observed with for the
sample with 10% HPA and reaches 56%. The abrupt
drop in alcohol conversion in the presence of the
sample of 13% HPA was likely due to its fast deacti-
vation.
Another characteristic feature of the HPA effect is the
presence of acetone in the reaction products. (No ace-
tone was formed in the presence of the initial sample.)
The nature of the formed products indicates that the
active sites of unmodified aluminum oxide are exclu-
sively Lewis sites: strong Lewis acid sites and strong
Lewis basic sites, which are responsible for the high con-
tent of propylene (the Е1 mechanism; in light of the low
process temperatures, the contribution from the forma-
tion of propylene from ether can be ignored) and ether
70
60
50
40
30
20
10
4
(
the S 2 mechanism), respectively. Considering the high
3
3'
N
ratio of propylene to the overall yield of oxygen-contain-
ing compounds, we may assume that strong Lewis sites
are in the majority on the η-Al O surface, which is in
2'
2
3
2
good agreement with the literature data [13, 19, 25].
To make it easier to consider the contribution from
Lewis acid sites and basic sites (both Lewis and Brøn-
sted), let us consider the ratio of reaction products that
form on them:
1
'
1
LAS
C H
3
6
1
3
5
7
9
11
13
15
τ, h
i-C H OH
3
7
Fig. 3. Effect of process time on the conversion of isopropa-
nol in the presence of HPA/η-Al O -catalysts with HPA
LBS
BBS
2
3
C H OC H
3 7 3 7
contents of (1) 5, (2) 7, (3) 10, and (4) 13%. Dashed lines are
the corresponding trends at HPA contents of (1′) 5, (2′) 7,
(3′) 10, and (4′) 13%. The process temperature is 230°C.
CH COCH
3 3
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 2 2020

3
04
GUSEINOVA et al.
4000
3000
2000
1000
0
3
2
1
000
000
000
0
(
а)
(b)
1
0
20
30
40
50
θ, deg
60
10
20
30
40
50
2θ, deg
60
2
Fig. 4. (Color online) X-ray diffraction pattern of (a) initial η-Al O and (b) modified η-Al O + 7% HPA.
2
3
2 3
A comparative analysis of the reaction products
Phase Composition
formed in the presence of the modified samples shows
that the introduction of 5% HPA results in an abrupt
increase in the yield of acetone (the maximum ace-
tone : ether ratio and one of the highest propyl-
ene/OCC ratios). In contrast, the sample of 7% HPA
with the highest yield of OCCs has the lowest propyl-
ene/OCC ratio. The active sites of the latter are thus
predominantly basic; according to the acetone : ether
According to the results of X-ray diffraction analy-
sis, the studied samples were poorly crystallized (i.e.,
predominantly X-ray amorphous) and consisted of η-
,
χ, and γ-polymorphs (Fig. 4a). This can be explained
by the background of the precursor sample (bayerite)
and the temperature of its calcination. These data are
in good agreement with the results in [8–11], accord-
ing to which the nature of initial aluminum hydroxide
ratio, these sites are Brønsted basic sites. It should be (amorphous aluminum hydroxide, pseudoboehmite,
noted that the predominance of acetone over ether is boehmite, bayerite, gibbsite) has a crucial effect on the
typical of all the modified samples and strongly differs formation of metastable aluminum oxide polymorphs
(
γ-, η-, χ-, and θ-Al O ) upon thermal treatment.
2 3
from the unmodified sample. Raising the HPA con-
tent to 10 and then to 13% was accompanied by a con-
tribution from Lewis acid sites and a drop in the con-
tribution from basic sites. Despite the apparent incon-
sistency, similar results are possible if we assume HPA
loses its structure after being sorbed on η-Al O .
It is clearly seen that the X-ray diffraction pattern
of unmodified alumina contains three strongly
smeared halos in the 2θ regions of 15°–30° and 35°–
5
0°, which correspond to the most intense reflections
2
3
at diffraction angles 2θ of 32.4° and 42.67° and can be
Table 1. Relationship between catalyst composition, process parameters, and the ratio of reaction products*
Ratio
Catalyst
Kconv, %
τ, h
SOCCs, % αacetone, vol % αether, vol %
С Н /OCCs С Н О/C H O
3
6
3
6
6
14
η-Al O (without HPA)
42
8
7
14.9
46.6
52.1
40.1
28.5
—
6.25
4.5
8.7
7.3
5.72
1.44
0.92
1.49
2.50
—
2
3
η-Al O + 5% HPA
44.6
50.1
54.3
59.2
16.3
17.4
14.5
10.2
3.62
2.0
2
3
η-Al O + 7% HPA
11
3
2
3
η-Al O + 10% HPA
1.99
1.52
2
3
η-Al O + 13% HPA
1
6.7
2
3
*
The given alcohol conversions and process times τ correspond to the highest yield of OCCs for a particular sample; Т = 200°C.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 2 2020

CATALYTIC CONVERSION OF ISOPROPANOL
305
Table 2. Effect of modification on the structural characteristics of the η-Al O samples*
2
3
Iintegr
counts deg
,
Peak no.
2θ, deg
d, Å
h, counts/s
l, deg
wintegr, deg
CSR, Å
Initial η-Al O
2
3
1
2
3
4
5
6
19.2
4.61
111
6.0 (7)
2.063
2.06 (17)
2.063
2.063
1.66 (9)
710 (130)
811.10
1309 (56)
725.33
817.70
763 (39)
6 (3)
112.1
330.8
126.2
121.7
—
32.40
37.53
39.311
42.67
45.66
2.76
2.39
2.29
2.12
1.99
457.86
408 (58)
409.091
460.956
394 (57)
1.7715
3.2 (6)
1.7730
1.7739
1.9 (4)
74.3
Modified η-Al O + 7% HPA
2
3
1
2
3
4
5
6
7
8
9
12.94
19.80
31.83
37.55 (8)
39.40
42.56 (7)
45.64 (6)
48.73
6.834
4.478
2.809
2.393 (5)
2.284
2.123 (3)
1.986 (3)
1.867
317.927
382.586
364.316
553 (68)
463.848
198 (41)
562 (68)
314.618
357.932
2.479
2.479
2.479
4.3 (3)
2.479
1.0 (3)
2.48 (17)
2.47952
2.47952
1030.71
1240.78
1182.75
4674 (595)
1507.3
376 (115)
2651 (270)
1023.94
1167.55
3.24198
3.24314
3.24648
8 (2)
3.24955
1.9 (10)
4.7 (11)
3.25455
3.26193
94.3
330.6
103.6
100.8
69.9
58.95
1.565
*
2θ is the angle characterizing the position of the peak, d is the interplanar distance, h is height, l is the full width at half maximum, I is
intensity, and w is width.
classified as reflections typical of γ- and η-aluminum
oxide polymorphs, respectively.
Surface Morphology
Electron micrographs of the initial and modified
aluminum oxides are shown in Fig. 5. As can be seen,
the surface of the unmodified sample appears to con-
sist of closely packed coarse agglomerates (Fig. 5a),
but the morphology and size of these formations
changed after the introduction of 7% HPA; i.e., they
were destroyed and then represented by two types (see
the particle size distribution curve). In the over-
whelming majority of cases, these were highly dis-
persed nearly spherical particles, their aggregates that
were more than 50 μm in size (the formation of finely
dispersed particles is in good agreement with the
results from X-ray diffraction), and a small number of
aggregates that were micrometers in size (Figs. 5b–
In the X-ray diffraction pattern of the HPA modi-
fied sample (Fig. 4b), we can see new peaks at 2θ
angles of 12.9°, 48.73°, and 58.96°, which can be
attributed to molybdenum hydr(oxide) against the
background of the remaining X-ray amorphous
phases.
No phosphorus peaks are observed in the X-dif-
fraction pattern, due probably to either its small
amount (4.65 wt %) or its penetration into aluminum
oxide. The height and integral intensity of the reflec-
tions at 19.8°, 37.55°, 39.4°, and 46.64° grow by 2–
3
.4 times as a result of an increase in the relative
5d), respectively. Their loose packing ensures a larger
η-Al O content against the background of a drop in
2
3
surface area, which is a necessary condition for the
the other aluminum oxide phases (Table 2). The rea- effective use of HPAs in heterogeneous catalysis.
son for these changes could be the conditions under
X-ray mapping performed in addition to SEM
which the modified sample was prepared, particularly
the additional stage of calcination. Also noteworthy is
(
Figs. 6a–6d) and semi-quantitative EDX analysis
Fig. 6e) showed that the contents of molybdenum
(
the change in the interplanar distances and the broad- and phosphorus after modifying aluminum oxide with
ening of reflections at 2θ angles of 19.8°, 37.55°, 39.4°, HPA were 4.65 and 0.2 wt %, and they were uniformly
and 45.6° (the characteristic width grew by 20–110%) distributed over the sample volume.
in the modified sample. The partial phase transition
IR spectroscopic studies were performed ex situ
was thus accompanied by appreciable dispersion of the (Fig. 7) for two types of aluminum oxide: unmodified
initial phase, as is confirmed by the reduction of the and HPA-modified η-Al O samples. Without excep-
2
3
coherent scattering region (CSR) (Table 2), and pos- tion, our aluminum oxide samples had two common
sibly the emergence of microstrains in this structure.
features.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 2 2020

3
06
GUSEINOVA et al.
600 μm
600 μm
(а)
(b)
0.855 μm
4.396 μm
1.641 μm
2.306 μm
0.773 μm
6.522 μm
1.699 μm
1.212 μm
2.646 μm
0.936 μm
7.952 μm
3.981 μm
1.690 μm
5
μm
2 μm
(c)
(d)
80
60
40
20
0
(
e)
<1
1−10 10−20 20−50 >50
Crystal size, μm
Fig. 5. (Color online) SEM micrographs of (a) initial η-Al O and (b–e) modified η-Al O + 7% HPA, (b) general view, (c, d)
2
3
2 3
structure of individual particles, (e) particle size distribution histogram.
(
1) The presence of a low-intensity vibrational
(1) In the initial (prior to catalytic experiments)
mode (a transverse optical phonon) with a maximum aluminum oxide (Fig. 7a), the surface OH groups were
−1
at 450 cm , testifying to the formation of O–Al bond predominantly bridging ОН– located between two
stretching vibrations typical of crystalline aluminum aluminum atoms, the coordination numbers of which
−1
oxide, according to the literature data [13, 26–29].
2) IR spectra with a solitary intense peak at
were 1 and 2, as is indicated by the bands at 3742 cm
the groups are linked to two octahedral aluminum
ions; the authors of [30] gave a proton affinity of PA =
(
(
−1
1
632 cm , which belongs to the molecular water
1
410–1440 kJ/mol, which testifies to its characteristic
O‒H bond bending vibrations δ(Н О) and is split.
2
−1
medium acidity) and 3813 cm (high-frequency OH
The intensity of the latter is higher for the modified
samples.
groups were linked to tetrahedral aluminum ions;
according to the same authors, PA = 1440–
Subsequent comparative analysis showed that the 1590 kJ/mol, so these hydroxyl groups had the lowest
four studied samples had several differences.
acidity), respectively. These bands completely disap-
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 94
No. 2
2020

CATALYTIC CONVERSION OF ISOPROPANOL
307
Summary spectrum
(e)
0
2
4
6
8
10 12 14 16
keV
Full scale 19282 counts. Cursor: 0
(
а)
OKα1
(b)
AlKα1
NaK
AlK
SiK
0.63
48.25
0.26
PK
MoL
O
0.20
4.65
45.01
In total 100.00
(
c)
PKα1
(d)
MoLα1
Fig. 6. (Color online) (a–d) Element distribution maps and (e) EDX analysis of η-Al O + 7% HPA.
2
3
(
a)
b)
(
(
c)
d)
(
(e)
4000
3500
3000
2500
2000
1500
1000
500
Fig. 7. IR spectra of η-Al O : initial (a) before and (b) after catalysis, modified η-Al O + 7% HPA (c) before and (d) after catal-
2
3
2 3
ysis, and (e) initial modifier (phosphomolybdic heteropoly acid).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 2 2020

3
08
GUSEINOVA et al.
peared for this sample after catalytic experiments, due groups (dehydration) of ABSes, with their conversion
probably to such surface transformations as the forma- into LASes (the bands of the bridging and terminal
tion and liberation of water molecules and the forma- hydroxide groups disappeared in the IR spectra of the
3
+
2–
tion of Lewis acid (Al ) and basic (О ) sites (the modified samples).
results for the catalytic conversion of isopropanol on
unmodified aluminum oxide agree with this model
when Lewis sites predominate). The modified samples
(III) Surface LASes interact with the heteropoly
anion to form surface complexes. Since the highest
electron density is concentrated in the heteropoly
anions on the oxygen atoms of corner М–О–М
groups, the formation of bonds with LASes proceeds
through the oxygen atoms of this group.
had no such bands, showing indirectly that η-Al O
2
3
contains aluminum vacancies randomly distributed
between the tetrahedral and octahedral positions.
(
2) The spectrum of the unmodified sample after
(
IV) Since η-Al O is predominantly characterized
2 3
catalytic process (Fig. 7b) was characterized by the
by weak LASes, the charge of the heteropoly anion is
only partially compensated for, and the surface com-
plex exhibits the properties of BBSes, which are accep-
−1
emergence of a transmission band at 1533.41 cm ,
which is indicative of surface carbonate groups coordi-
3+
nated to Al . It is likely that these groups, being tors of hydrogen, as has been observed in catalytic
strongly adsorbed during the process, block active sites studies.
and are responsible for the rapid drop in the catalytic
activity of this sample.
CONCLUSIONS
(
3) For all four aluminum oxide samples, the
−
1
region of 3100–3700 cm , which corresponds to the
Considering the obtained data and the results on
absorption of O–H bond stretching vibrations, con- the catalytic activity of the HPA/η-Al O system in the
2
3
tained a broad intense absorption band with a maxi- conversion of isopropanol, we may assume that the
mum at 3416 cm−1 (for the unmodified samples) and interaction between a Keggin structure and the basic
−1
bands shifted toward the 3404 and 3385 cm region of oxide destroys the latter’s structure with the formation
lower frequencies (for the modified samples; Figs. 7c of heteropoly anions. Being negatively charged, they
and 7d), due probably to the presence of the modifier. interact with surface electron-accepting centers
Also conspicuous was a drop in intensity for the mod- (LASes). According to the results in [19, 27, 30, 31],
ified samples, indicative of modifier–support interac- the concentration of weak LASes on a surface of η-
tion;
4) The bands at 462.84, 494.65, 550.82, 802.88,
Al O is several times higher than the content of strong
2 3
and medium sites (there are almost none of the latter),
so the interaction of the heteropoly anions likely
occurs with weak acid sites, which cannot completely
compensate for its negative charge. As our data indi-
cate, this results in the formation of a negatively
charged surface–ligand complex that subsequently
acts as a hydrogen acceptor.
(
−1
8
96.42, 965.65, 1060.03, 1124.55, and 1167.45 cm ,
observed when the initial HPA was used as a modifier
(
Fig. 7e), disappeared after deposition onto aluminum
oxide, testifying to the lack of a Keggin structure on
the samples’ surfaces.
The considered data indicate that the evolution of
active sites in the HPA/aluminum oxide system pro-
ceeds according to a specific mechanism.
REFERENCES
(
I) The oxide surfaces contain mainly LASes
1
. E. N. Nikitina, Heteropoly Compounds (Goskhimizdat,
3+
(
coordinatively unsaturated aluminum atoms Al )
Moscow, 1962) [in Russian].
and LBSes (coordinatively unsaturated oxygen atoms
2. M. Pope, Heteropoly and Isopoly Oxometalates (Spring-
2
−
О ); the bridging and terminal hydroxide groups,
which do not exhibit catalytic activity but were
detected by studying the results from IR studies, indi-
cate a certain number of ABSes that, according to the
Knözinger–Ratnasamy model (currently recognized
as the one most complete and with the best correspon-
dence to the current level of knowledge about alumi-
num oxide [28, 29]) have the structure
er, Berlin, Heidelberg, 1983).
3. G. M. Maksimov, Russ. Chem. Rev. 64, 445 (1995).
4. X. López, J. J. Carbó, C. Bo, and M. Josep, Chem. Soc.
Rev., No. 41, 7537 (2012).
5. Jian Zhou, Yan Liu, Yang Hu, Chuang Zhou, and
Zhuohong Yang, Ind. Crops Products 94, 424 (2016).
6. I. V. Kozhevnikov, Catalysis by Heteropoly Compounds
Znanie, Moscow, 1985) [in Russian].
7. M. R. Salaev, E. A. Guseinova, M. R. Al-Battbotti,
K. Yu. Adzhamov, F. A. Amirov, and A. A. Gasanov,
Russ. J. Appl. Chem. 91, 906 (2018).
(
H
OH
O
Al
Al
Al
8
. M. H. Timofeeva and S. Ts. Khankhasaeva, Kinet.
(
II) At the stage of modification with HPA, there is
Catal. 50, 57 (2009).
interaction (possibly multipoint dissociative HPA
adsorption) between its cations and the hydroxy
9. N. N. Tomina, P. A. Nikul’shin, and A. A. Pimerzin,
Pet. Chem. 48, 92 (2008).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 94
No. 2
2020

CATALYTIC CONVERSION OF ISOPROPANOL
309
1
0. V. D. Stytsenko, A. A. Lavrinenko, and V. A. Vi- 21. E. A. Tveritinova, Yu. N. Zhitnev, S. A. Chernyak,
nokurov, RF Patent No. 2400467, Byull. Izobret.,
No. 27 (2010).
E. A. Arkhipova, S. V. Savilov, and V. V. Lunin, Russ. J.
Phys. Chem. A 91, 448 (2017).
1
1. A. S. Ivanova, Kinet. Catal. 53, 425 (2012).
22. W. Qi and D. Su, ACS Catal. 4, 3212 (2014).
12. Z. R. Ismagilov, R. A. Shkrabina, and N. A. Koryabki- 23. D. S. Su, J. Zhang, B. Frank, et al., Chem. Sus. Chem.
na, Analytical Overview, SO RAN, No. 50, Ser. Ecolo-
gy (Inst. Katal. im. G. K. Boreskova, Novosibirsk,
3, 169 (2010).
2
2
2
4. P. Serp, M. Corrias, and P. Kalck, Appl. Catal. A: Gen.
1998).
2
53, 337 (2003).
13. G. D. Chukin, The Structure of Alumina and Hydrode-
5. D. T. Lundie, A. R. McInroy, R. Marshall, et al.,
sulfurization Catalysts. Reaction Mechanisms (Paladin,
Printa, Moscow, 2010) [in Russian].
J. Phys. Chem. B 109, 11592 (2005).
6. K. Nakamoto, Infrared and Raman Spectra of Inorganic
and Coordination Compounds (Mir, Moscow, 1991; Wi-
ley, New York, 1986).
14. A. M. Becerra, M. E. Iriarte, and A. E. Castro Luna,
React. Kinet. Catal., Lett. 79 (2003).
15. P. J. Kim, B. H. Joo, W. Kim, Y. Kim, I. K. Kim, and
2
7. A. S. Ivanova, G. S. Litvak, G. N. Kryukova, S. V. Tsybu-
J. Yi. Song, J. Catal. Lett. 104, 181 (2005).
lya, and E. A. Paukshtis, Kinet. Catal. 41, 122 (2000).
1
6. S. N. Lanin, A. A. Bannykh, A. E. Vinogradov,
N. V. Kovaleva, K. S. Lanina, and S. A. Nikolaev,
Russ. J. Phys. Chem. A 90, 1427 (2016).
2
8. E. A. Paukshtis, Infrared Spectroscopy in Heterogeneous
Acid–Base Catalysis (Nauka, Novosibirsk, 1992) [in
Russian].
17. V. A. Chumachenko and E. V. Ovchinnikova, Catal.
2
9. G. D. Chukin, S. A. Sergienko, Yu. L. Seleznev, et al.,
Ind. 8, 134 (2016).
Zh. Prikl. Spectrosk. 47, 427 (1987).
1
8. A. V. Lavrenov and V. K. Duplyakin, Kinet. Catal. 50,
35 (2009).
9. N. E. Kotlovanova, A. N. Matveeva, Sh. O. Omarov,
3
0. R. G. Romanova and E. V. Petrova, Kinet. Catal. 47,
2
138 (2006).
1
2
V. V. Sokolov, D. N. Akbaeva, and V. I. Popkov, Inorg. 31. A. A. Lamberov, R. G. Romanova, I. G. Shmelev, and
Mater. 54, 392 (2018).
V. F. Sopin, Russ. J. Appl. Chem. 75, 396 (2002).
0. D. M. Sung, Y. H. Kim, E. D. Par, and J. E. Yie, Res.
Chem. Intermediates 36, 653 (2010).
Translated by E. Glushachenkova
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 94 No. 2 2020