Dysprosium-doped zinc tungstate nanospheres as highly efficient heterogeneous catalysts in green oxidation of terpenic alcohols with hydrogen peroxide

Article abstract of DOI:10.1039/d0nj05623e

A green route to oxidize terpenic alcohols (nerol and geraniol) with H₂O₂over a solid catalyst was developed. The Dy-doped ZnWO₄catalyst was synthesized by coprecipitation and microwave-assisted hydrothermal heating, containing different dysprosium loads. All the catalysts were characterized through infrared spectroscopy, powder X-ray diffraction, surface area and porosimetry, transmission electronic microscopy image, andn-butylamine potentiometric titration analyses. The influence of main reaction parameters such as temperature, the stoichiometry of reactants, loads, and catalyst nature was assessed. ZnWO₄2.0 mol% Dy was the most active catalyst achieving the highest conversion (98%) and epoxide selectivity (78%) in nerol oxidation. The reaction scope was extended to other terpenic alcohols (i.e., geraniol, borneol, and α-terpineol). The highest activity of ZnWO₄2.0 mol% Dy was assigned to the lower crystallite size, higher surface area and pore volume, higher acidity strength and the greatest dysprosium load.

Full text of DOI:10.1039/d0nj05623e

NJC
View Article Online
View Journal | View Issue
PAPER
Dysprosium-doped zinc tungstate nanospheres as
highly efficient heterogeneous catalysts in green
oxidation of terpenic alcohols with hydrogen
peroxide†
Cite this: New J. Chem., 2021,
4
5, 6661
a
b
Daniel Carreira Batalha, Kellen Cristina Mesquita Borges,
b
b
Rosana de Fatima Gonçalves, Murillo Henrique de Matos Rodrigues,
´
b
c
Mario Junior Godinho, Humberto Vieira Fajardo,
´
´
d
a
Carlos Giovani de Oliveira Bruziquesi and Marcio Jose da Silva
*
´
´
A green route to oxidize terpenic alcohols (nerol and geraniol) with H
2 2
O over a solid catalyst was
developed. The Dy-doped ZnWO catalyst was synthesized by coprecipitation and microwave-assisted
4
hydrothermal heating, containing different dysprosium loads. All the catalysts were characterized
through infrared spectroscopy, powder X-ray diffraction, surface area and porosimetry, transmission
electronic microscopy image, and n-butylamine potentiometric titration analyses. The influence of main
reaction parameters such as temperature, the stoichiometry of reactants, loads, and catalyst nature was
Received 17th November 2020,
Accepted 4th March 2021
assessed. ZnWO
epoxide selectivity (78%) in nerol oxidation. The reaction scope was extended to other terpenic alcohols
i.e., geraniol, borneol, and a-terpineol). The highest activity of ZnWO 2.0 mol% Dy was assigned to the
4
2.0 mol% Dy was the most active catalyst achieving the highest conversion (98%) and
(
4
DOI: 10.1039/d0nj05623e
lower crystallite size, higher surface area and pore volume, higher acidity strength and the greatest
dysprosium load.
rsc.li/njc
processes even more desirable and tuned with the principles
of green chemistry. In this sense, hydrogen peroxide has
Introduction
1
1
Terpenic alcohols are abundant and renewable feedstock that great advantages such as the generation of water as the only
1
–3
has risen great interest for application in fine chemistry.
by-product, low cost, non-flammability, and easy handling. Its
In particular, their oxidation and epoxidation reactions led to efficiency as an oxidant is even more prominent due to its
attractive products for perfume, fragrance, and pharmaceutical ability to bond to the surface of a solid catalyst giving reactive
4
–7
industries.
hydroxyl group and the double bond) make these substrates reactions.
sources of carbonylic products, and chiral compounds, which Among the various catalysts used in oxidations of terpenic
are both potential drug intermediates, thereby being valuable alcohols with hydrogen peroxide, Keggin polyoxometalates
The presence of two oxidizable sites (i.e., the oxygen species that will directly participate in oxidation
1
2,13
8
–10
compounds to the fine chemical industry.
should be highlighted due to their high structural versatility
Besides the use of a renewable raw material, the application that allows their use as homogeneous and heterogeneous
1
4–16
of an environmentally benign oxidant makes the oxidative catalysts.
Various metal-exchanged lacunar heteropolyacid
salts (i.e., cesium, potassium, or sodium) have been demon-
strated to be highly efficient in the oxidation of terpenic
alcohols such as geraniol, nerol, borneol, and linalool.
a
Chemistry Department, Federal University of Viçosa, Campus Universit a´ rio,
Avenue P. H. Rolfs, s/n, Viçosa, Minas Gerais, CEP 36570-900, Brazil.
E-mail: silvamj2003@ufv.br; Tel: +55 31 3899-3071
17–19
In this sense, solid catalysts play an essential role in making
the oxidation processes greener by generating fewer effluents
and residues than homogeneous counterparts, reducing their
environmental impact. In addition, solid catalysts can be easily
recovered and reused. The doping of solid catalysts with
different elements can trigger a synergism that significantly
improves their efficiency in oxidation processes. For instance,
b
c
Chemistry Institute, Federal University of Catal a˜ o, Catal a˜ o, Goi a´ s
CEP 75.704-020, Brazil
Chemistry Department, Federal University of Ouro Preto, Ouro Preto
CEP 35400-000, Minas Gerais, Brazil
d
Chemistry Department, Federal University of Minas Gerais, Belo Horizonte,
Minas Gerais, Zip code 31279-201, Brazil
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0nj05623e
4
zinc tungstate (ZnWO ) has been used in photocatalysis, due to
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6661–6670 | 6661

View Article Online
Paper
NJC
2
0–22
its electronic properties, good stability, and low cost.
distribution of average diameter, and pore volume of the
Nonetheless, their doping with rare earth elements can bring catalysts, respectively. The porosimetry analyses were performed
not only structural benefits, in terms of particle size and surface on the NOVA 1200 Quantachrome equipment, using the isotherms
area, but also electronic properties, consequently increasing of physical adsorption/desorption of N , at 77 K. All catalytic
2
2
3
their catalytic activity. Some of these properties can be materials were characterized by X-ray diffraction (XRD) using
interesting in oxidation reactions, such as the presence of a D/Max-2500PC diffractometer (Rigaku, Japan), (Cu-Ka radiation
2
4–26
Lewis acid sites, and high stability.
The use of dysprosium l = 1.5406 Å) in the 2y range from 101 to 701 at a scanning rate of
ꢂ1
as a dopant has been demonstrated to be an efficient option in 0.021 min . Crystallite sizes were calculated using the Scherrer
2
7
various oxidative processes.
equation:
In this work, we describe the synthesis of efficient zinc tungstate
catalysts doped with dysprosium by a simple coprecipitation
methodology, followed by microwave-assisted hydrothermal
synthesis. The catalysts were properly characterized by different
methodologies and evaluated in oxidation reactions with
hydrogen peroxide of various terpenic alcohols, which are
renewable origin feedstock of interest for the synthesis of
fragrances, fine chemicals, and drug intermediates. Nerol
was the model molecule. The influence of the main reaction
parameters and the reuse of the solid catalyst were assessed.
K ꢀ l
L ¼
(1)
b ꢀ cos y
where l is the X-ray wavelength (nm), b is the peak width of the
diffraction peak profile at half maximum height resulting from
small crystallite size in radians, and K is a constant related to the
crystallite shape, normally taken as 0.89.
Fourier transform infrared spectroscopy (FT-IR) spectra were
recorded on a PerkinElmer Spectrometer, model Frontier. The
shape, size, and form of the nanocrystals were analyzed by
high-resolution transmission electron microscopy (HR-TEM)
with a Tecnai G2-F20 microscope (200 kV). The strength of
the acidic sites of Dy-doped tungstate catalysts was determined
Experimental
Synthesis
3
by ammonia temperature-programmed desorption (NH -TPD)
The catalysts were synthesized through the coprecipitation performed on a CHEMBET-3000 TPR/TPD chemisorption
method at room temperature followed by the microwave hydro- analyzer (Quantachrome instrument). As such, 0.2 g was
thermal method (MWH – in a microwave system of 2.45 GHz, inserted in a U-shaped quartz tube and degassed for 1 hour
ꢂ1
with a maximum power of 800 W). Four catalysts with different under a He gas flux of 80 mL min (99.99%) at 393 K. After
Dy loads: ZnWO – pure; ZnWO – 0.5% Dy; ZnWO – 1.0% Dy pre-treatment, the sample was cooled to 303 K, and it was kept
4
4
4
ꢂ1
and ZnWO – 2.0% Dy, were synthesized. Sodium tungstate(VI) in contact with the NH gas flux of 40 mL min (99.9%) for
4
3
dihydrate (Na
2
WO
4
ꢀ2H
2
O; 99% purity, Sigma-Aldrich), zinc 10 minutes, to saturate the acidic sites through the chemical
ꢂ1
nitrate hexahydrate (Zn(NO
and dysprosium(III) nitrate hexahydrate (Dy(NO
3
)
2
ꢀ6H
2
O, 99.5% purity, Aldrich), adsorption of the probe molecule. Then, a He flux of 40 L min
ꢀ6H
9.999% purity, Aldrich) were used as synthesis precursors. eventual physisorbed ammonia quantity, and immediately
3
)
3
2
O; was set up for 10 minutes without heating, to remove an
9
ꢂ3
ꢂ1
Firstly, Na
2
WO
4
ꢀ2H
2
O (ca. 5 ꢁ 10
mol) was dissolved in after, the material was heated up at a rate of 283 K min to
deionized water (50 mL) with constant stirring. Separately, 1173 K. Additionally, the strength of the acidic sites of
ꢂ3
Zn(NO ) ꢀ6H O (ca. 5 ꢁ 10 mol) was dissolved in deionized Dy-doped tungstate catalysts also was determined by potentio-
3
2
2
water (50 mL) at pH = 6. Afterward, the Dy(NO ) ꢀ6H O dopant metric titration with n-butylamine solution, using a BEL
3
3
2
(ca. 0.5, 1.0, 2.0 mol%) was added to the aqueous solutions potentiometer (model W3B) with a glass electrode. To do this,
2
+
containing Zn ions, and subsequently, this solution was each catalyst was suspended in acetonitrile and magnetically
added to the WO
2
ꢂ
4
solution, heated at 363 K, and stirred for stirred for 3 h. In the next step, the suspension was titrated with
ꢂ1
30 min. Immediately, a white suspension was formed and n-butylamine (0.025 mol L ) solution in acetonitrile.
placed in a Teflon autoclave. Finally, the Teflon autoclave
containing the solution was heated in a microwave system at Catalytic tests
4
2
13 K for 1 h. The heating rate in this system was fixed at
5 K min , and the pressure inside the autoclave was stabilized
The catalytic runs were carried out in the liquid phase with
magnetic stirring, in a glass reactor coupled to the reflux
condenser, heated in a water batch at atmospheric pressure.
ꢂ1
at 245 kPa. The formed suspension was washed with deionized
water to neutralize the solution. White precipitates were
collected and dried at 333 K.
Typically, H
2
O2(aq) (ca. 34 wt%) and nerol (ca. 1.375 mmol) were
dissolved in CH
3
CN (ca. 10 mL) and heated to 333 K. After
adding the solid catalyst, the reaction was started. The reaction
progress was followed for 8 h, periodically collecting aliquots
Characterization
To assist the interpretation of the results of the catalytic tests and to that were analyzed in a Shimadzu GC-2010 Plus equipped with
verify the effects of structural properties, the catalysts were analyzed an AOC-20i auto-injector and a flame ionization detector (FID).
by several techniques of physical–chemical characterization.
All the products were duly identified and it was confirmed
Brunauer–Emmett–Teller (BET) and Density Functional that spectrometry mass analysis was performed in Shimadzu
Theory (DFT) methods were used to determine the surface area, GCMS-QP2010 Ultra equipment.
6
662
|
New J. Chem., 2021, 45, 6661–6670
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
Results and discussion
Characterization of the catalysts
The main structural properties of the catalysts were determined
aiming to support the results of the catalytic activity of the
synthesized samples. Table 1 shows the results of porosimetry
2
analysis obtained from the adsorption/desorption of N .
The doping level of Dy on the zinc tungstate salts increased
the surface area of these materials. The higher the Dy load, the
greater the observed increase (Table 1). This effect was more
noticeable when 2.0 mol% of Dy was loaded into zinc tungstate.
Conversely, an increase in the Dy load led to lowering of both
the volume and pore size. These two opposite effects may be
closely related to the pore shape of each solid.
A plausible explanation for this increase in the surface area
3
+
2+
is that the Dy ions may be punctually replacing Zn ions in the
crystalline structure of zinc tungstate, altering their properties.
3
+
Fig. 1 (a) Isotherms of adsorption/desorption of N
2
, (b) pore diameter
It is ascribed to the greater ionic radium of Dy when compared
to Zn (ca. 102 pm and 74 pm, respectively). The effect will be
confirmed by the DRX and FT-IR analysis (Fig. 2).
distribution (b) and (c) potentiometric titration curves with n-butylamine of
2
+
21
the synthesized catalysts.
The isotherms presented in Fig. 1a help to understand this
phenomenon. All isotherms have a type V profile with H2 type
hysteresis, which best matches weak interactions between
adsorbents and adsorbates typical of micro and mesoporous
materials.
Through the data in Table 1 and the pore distribution in
Fig. 1b, it is possible to conclude that all the catalysts
predominantly have mesopores. The key to the reported
differences in surface areas may be related to the slope of the
isotherms can be important to justify the different results
achieved in nerol oxidation reactions by these catalysts.
Fig. 1c shows the potentiometric titration curves of the four
29
synthesized catalysts. According to Pizzio and Blanco (2007),
all the catalytic materials whose solutions have an initial
electrode potential between 0 and 100 mV are characterized
as catalysts with strong acidic sites. Fig. 1c shows that when
zinc tungstate catalysts were doped with Dy, the strength of
acid sites increased. Moreover, a higher Dy load gives stronger
acidity strength to the solid zinc tungstate. These results agree
desorption curves. The ZnWO
and ZnWO – 1.0 mol% Dy catalysts have the same and steeper
desorption branch compared to the ZnWO – 2.0 mol% Dy
4 4
– pure, ZnWO – 0.5 mol% Dy
4
4
30–32
with the literature.
desorption branch. It may be related to a more homogeneous
distribution of the accessible pores, as illustrated in Fig. 1b.
This characteristic can be directly linked to the Dy load present
Fig. 2a shows the XRD diffraction patterns of the synthesized
catalysts, together with the main Miller indices, which were
properly indexed to the monoclinic wolframite zinc tungstate
structure according to the ICDD Card number 00-015-0774.
Through this analysis, it is possible to notice the differences
between the XRD diffraction profiles of different Dy-doped
2
7,28
in the catalyst.
We have found that the greater the amount of Dy, the higher
the heterogeneity in the distribution of pores, that is, the number
of pores within a narrow range of diameters is also higher. The
decrease in Dy load results in a more homogeneous distribution of
the number of pores of different pore diameter ranges. However,
these structural changes do not change the fact that the material is
mesoporous, since most of the quantified pores belong to the
region of 20 to 500 Å in diameter, characteristic of mesopores.
As can be seen in the next sections, the differences
evidenced by the results of nitrogen adsorption/desorption
4
ZnWO catalysts. As the amount of Dy increases in the catalyst
structure, a loss of crystallinity is noticed. This factor was
confirmed by calculating the crystallite size using the Scherrer
equation.
The values of crystallite sizes obtained for the catalytic materials
were the following: pure (24 nm), 0.5 mol% Dy (13 nm), 1.0 mol%
Dy (9 nm), and 2.0 mol% Dy (7 nm). Therefore, an increase in the
doping of Dy led to lower size of the crystallite.
Although an enlargement of typical diffraction peaks of
Table 1 Characterization of the catalysts through the physical adsorption ZnWO
4
and a diminishment in their intensities have occurred
of nitrogen
most of them remained in the diffractograms even when the
3
+
2
ꢂ1
3
ꢂ1
highest load of Dy (ca. 2.0 mol%) was present. This behavior
Catalysts
S
BET (m g
)
V
DFT (cm g
)
DDFT (Å)
3
+
may be related to the size of the ionic radius of Dy , which is
ZnWO
ZnWO
ZnWO
ZnWO
4
4
4
4
59.8
59.1
60.6
85.2
0.212
0.161
0.171
0.187
122.3
93.3
93.3
78.0
2+
larger than the radius of Zn ions, triggering an expansion of
0.5 mol% Dy
1.0 mol% Dy
2.0 mol% Dy
the crystalline network of the material, mainly while a greater
3
2
4
amount of dysprosium is doped on the ZnWO sites. This
different crystallinity may help to explain the different catalytic
activity of these materials.
S
BET = surface area; VDFT = cumulative pore volume; DDFT = pore
diameter.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6661–6670 | 6663

View Article Online
Paper
NJC
Fig. 3 TEM image of the nanocrystals: (a) ZnWO
4
– pure; (b) ZnWO
4
–
0
.5 mol% Dy; (c) ZnWO – 1.0 mol% Dy and (d) ZnWO
4
4
– 2.0 mol% Dy.
Fig. 2 Powder XRD patterns (a) and FT-IR spectra of the catalysts (b).
was possible to obtain nanostructured materials by simple
methodologies of synthesis and with high potential in oxidation
reactions.
Fig. 4 shows high-resolution TEM images (HRTEM) obtained
for the tungstate catalysts. It was demonstrated that the pure
Fig. 2b displays the infrared spectra of the synthesized
catalysts. Absorption bands placed in the region between 400
ꢂ1
ꢂ1
and 1050 cm can be visualized. At 870 cm close wavenumbers,
a band attributed to the Zn–O–W bond vibration is noted, which
is a consequence of bending and stretching deformations. The
stretching vibrations of the W–O bond give absorption bands in
ZnWO
the (002) direction with an interplanar distance of 0.250 nm,
.256 nm, and 0.260 nm, respectively. For 1% Dy doped ZnWO
4 4
, and ZnWO doped with 0.5%, and 2% of Dy grew along
0
4
ꢂ1
the regions between 811 and 694 cm . It is possible to observe
an interplanar distance of 0.363 nm was measured relative to
the crystallographic plane (110). These results are consistent
with the XRD technique.
that, as the concentration of Dy increases, there is a displacement
ꢂ1
of the band close to 811 cm toward a higher wavenumber.
This may be related to an increase in vibration energy due to the
3
Fig. 5 displays the TPD-NH curves. Normally, it has been
3+
presence of Dy ions, which consequently led to a deformation in
the material’s structure, reinforcing that previously said for the
shown that the desorbed ammonia in the temperature ranges
of 423–573 K, 573–773 K, and above 773 K was previously
adsorbed on the weak, medium, and strong acid sites,
21,33
XRD results.
Still on the FTIR spectra, the bands referring to
the bending vibrations of the W–O bonds are noticed at 463 and
ꢂ1
6
5
64 cm wavenumbers. Finally, the absorption bands present at
ꢂ1
16 and 425 cm wavenumbers can be assigned to the symmetric
and asymmetric stretching modes of the W–O and Zn–O bonds,
21
6 6
respectively, for the ZnO and WO octahedral units.
Fig. 3a–d shows the high-resolution transmission electron
microscopy images of the catalytic materials. From them, it is
possible to verify the contrasting differences in the shapes and
particle sizes for the different solids. The profile of the pure
catalyst (Fig. 3a) shows basically the nanorod structure with
sizes varying between 11 and 59 nm. In contrast, as the
concentration of Dy increases in each material (Fig. 3b–d),
noticeable changes occur, such as an increase in the presence
of nanospheres, besides the increase in particle uniformity,
together with a decrease in size (0.5 mol% Dy: 2–37 nm;
1.0 mol% Dy: 3–19 nm and 2.0 mol% Dy: 3–7 nm).
It should be highlighted that when the catalysts are doped
with 1.0 and 2.0 mol% Dy, they basically have nanospheres,
indicating the influence of the presence of rare earth elements
Fig. 4 HRTEM image of: (a) ZnWO
4
– pure; (b) ZnWO
4
– 0.5 mol% Dy; (c)
in the shape of the catalysts. It is important to emphasize that it ZnWO
4
– 1.0 mol% Dy and (d) ZnWO
4
– 2.0 mol% Dy.
6
664
|
New J. Chem., 2021, 45, 6661–6670
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
3
Table 2 Total integrated area of the TCD signal of the NH -TPD profile
and percentage of weak, medium, and strong acidity sites (i.e., calculated
by integration of the area peaks at each specific temperature range) of the
ZnWO
4
, ZnWO
4
0.5% Dy, ZnWO
4 4
1.0% Dy and ZnWO 2.0% Dy catalysts
Catalysts
Total area/a.u. Weak/% Medium/% Strong/%
ZnWO
ZnWO
ZnWO
ZnWO
4
4
4
4
2042
0
8
10
11
84
57
75
73
16
35
15
16
0.5% Dy 2206
1.0% Dy 2548
2.0% Dy 5430
3 4 4 4
Fig. 5 NH -TPD curves of ZnWO , ZnWO 0.5% Dy, ZnWO 1.0% Dy and
4
ZnWO 2.0% Dy.
3
3–35
respectively.
The TCD signals for ZnWO , ZnWO 0.5% Dy,
4 4
ZnWO 1.0% Dy, and ZnWO 2.0% Dy were integrated to assess
the characteristic acidity strength of each catalyst, and the data
are summarized in Table 2.
4
4
Fig. 6 Conversion and selectivity of nerol oxidation reactions with H
in the presence of zinc tungstate catalysts. Reaction conditions: nerol
1.375 mmol), H (2.750 mmol), temperature (333 K), catalyst (15 mg),
CH CN (10 mL)
2 2
O
(
2 2
O
The integrated area values indicated that an increase in the
3
4
Dy load increased the acidity strength. The ZnWO 2.0% Dy
catalyst displayed the highest content of medium acidity
strength sites (highest area value, see Table 2). Nevertheless,
the percentages of each type of acidity strength were different
the highest load of Dy (Table 1). Therefore, when we correlate
only the activity of Dy doped ZnWO catalysts, we concluded that
the catalyst with the largest volume of pores (ca. 0.187 cm g
was the most active. When we verified the volume of the pores of
the undoped catalyst we realize that its pores are still bigger
4
for each material. The ZnWO
of medium acidic sites (84%), while the ZnWO
4
catalyst had the highest content
0.5% Dy had
3
ꢂ1
)
4
the highest percentage of acidic sites of strong nature (35%).
Overall, the increase in Dy content suggests an increase in
the proportion of weak acidic sites, with percentage values of 0,
3
ꢂ1
(
ca. 0.210 cm g ) than the most active catalyst. It may explain
the high activity of pure ZnWO . Another key aspect is that the
tungstate catalyst can itself catalyze oxidations with hydrogen
4
8
, 10, and 11% for the ZnWO , ZnWO 0.5% Dy, ZnWO 1.0%
4 4 4
Dy and ZnWO 2.0% Dy catalysts, respectively.
4
2
0
peroxide. Conversely, the catalyst with the lowest Dy load was
less active due to its pores with the lowest volume.
Catalytic tests
However, the Dy load plays an essential role too. When the
Initially, screening to select the most active catalyst was performed. catalysts are doped with a load higher than 0.5 mol% Dy, this
Fig. 6 shows the kinetic curves with zinc tungstate catalysts doped additional active site is present in enough amounts to compen-
with different loads of dysprosium, besides the pure zinc sate for the reduction of the pore volume of ZnWO triggered by
4
tungstate.
Clearly, an increase in Dy load had a beneficial effect on the important insights. An increase in Dy load diminished the size
catalytic activity. Nonetheless, the catalyst 0.5 mol% Dy/ZnWO of the crystallite. Applying the Scherrer equation to the most
had the worst performance compared to pure ZnWO . It can be intense diffraction peaks (Fig. 2a) the following values were
the Dy load. The crystallinity of doped catalysts provides too
4
4
assigned to the differences of porosity between the zinc obtained: 0.5 mol% Dy (13 nm), 1.0 mol% Dy (9 nm), and
tungstate and the DY doped catalysts (Table 1), to the crystal- 2.0 mol% Dy (7 nm). This effect was attributed to the greater
3
+
2+
linity and Dy load.
size of the ionic radius of Dy ions when compared to the Zn
The pure catalyst (ZnWO ) presented the highest values of cations. The doping of Dy triggers an expansion of the crystal-
4
3
2
diameter and pore volume. On the other hand, an increase in line network of the material, mainly when at higher loads.
Dy load had opposite effects in the porosity of doped catalysts; This lower crystallite size may improve the ability of materials
while the volume of pores was gradually increased with a higher for binding with H molecules, which can increase the
load, the diameter of the pores was being decreased, mainly for formation of reactive oxygen species.
2 2
O
3
0,31
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6661–6670 | 6665

View Article Online
Paper
Another important factor that may be influencing nerol
NJC
oxidation is the acidity of the catalysts. According to the results
presented by NH -TPD (Fig. 4), the total acidity is proportional
3
to the amount of Dy, thus, the ZnWO catalyst 2.0% Dy, which
4
presented the greatest number of acidic sites (Table 2), was the
most active. It is interesting to note that the distinct types of
acidic sites (i.e., weak, medium, and strong) can have different
effects on the activity of Dy catalysts in the oxidation reactions.
Among the used catalysts, those with the highest percentage in
medium acidity strength sites were those that showed the best
performance in the nerol oxidation. The presence of acidic sites
on the catalyst surface can be essential for the interaction of the
3
0
peroxide oxidant with the substrate.
Fig. 7 Conversion and selectivity of nerol oxidation reactions with H O
2
2
Scheme 1 describes a probable reaction pathway involving in the presence of different amounts of ZnWO
Reaction conditions: nerol (1.375 mmol), H (2.750 mmol), temperature
333 K), CH CN (10 mL)
4
– 2.0% Dy catalyst.
2 2
O
the peroxidation of the Dy doped-zinc tungstate catalyst (i.e.,
Dy3/2(Zn)WO ) and the allylic oxidation assisted by the hydroxyl
group.
(
3
4
3
0,31,38
In allylic alcohol oxidation reactions, the active species
commonly formed are the metal-peroxide intermediates.
The metal acts as a Lewis acid, binding to an oxygen atom of
the oxidant, diminishing the efficiency of oxidation, a fact that
occurred mainly when 30 mg of catalyst was used.
The catalytic performance of ZnWO – 2.0 mol% Dy in nerol
4
oxidation reactions with hydrogen peroxide was compared with
various metal oxide and heteropolyacid catalysts (see Fig. SM1,
H O . While an oxygen atom of the oxidant linked to the
2
2
metal peroxide the olefin double bond, the oxygen atom of
the allylic hydroxyl group is bonded to the metal. Thus, the
O–O bond cleavage of the peroxide group occurs and,
consequently, the epoxide is formed with the release of the
ESI†). It was found that the Dy/ZnWO catalyst was more active
4
36,37
and selective than other metal oxide catalysts (see ref. ESI†).
Only the heteropolyacid salt (i.e., Na PW O ) was more efficient,
7
11 39
3
1,38
catalyst.
19
however, it was totally soluble in the reaction medium.
Fig. 6 shows the conversion and selectivity of the reaction
products. Nerol has two potentially oxidizable sites; the carbonyl
group which can be oxidized to aldehyde, and a double bond,
which may be epoxidized. We have found that regardless of
catalyst, nerol epoxide was always the major product,
neraldehyde being the minor product. Oligomers, which are
not detectable by the GC analysis, were formed in all the
reactions, as evidenced by the mass balance of reaction. Once
the most active catalyst is selected, we have investigated the
effect of the amount of catalyst on the conversion and selectivity
of products. Kinetic curves and selectivity of reactions are
presented in Fig. 7.
The impact of the molar ratio of the oxidant to the substrate
was also assessed (Fig. 8). Since the reaction was highly
efficient with a low load of the oxidant, this effect was assessed
only at three levels. At an equimolar stoichiometric amount
(ca. 1 : 1), the oxidation was slightly compromised; a lower
selectivity toward the main product (i.e. nerol epoxide) was
achieved. Even an isomerization product (i.e., geraniol) was
formed under these conditions, although with a low selectivity
(ca. 4%). Besides, the highest neraldehyde selectivity was
reached with this load of the oxidant. When a 1 : 1.5 or 1 : 2
molar ratio of nerol to the oxidant was used, no significant
4
An increase in the amount of ZnWO 2.0 mol% Dy catalyst
enhanced the initial rate and the final conversion of the
reactions. This effect can be attributed to the higher number
of active sites present in the reaction when a higher mass of
catalyst is used. On the other hand, with a load higher than 15.0 mg,
the selectivity for oligomers shows a drastic increase. It is a
consequence of the noticeable decomposition underwent by
Fig. 8 Conversion and selectivity of nerol oxidation reactions with different
Scheme 1 Hydroxyl group-assisted allylic oxidation of nerol with hydro-
amounts of H
conditions: nerol (1.375 mmol), catalyst (15 mg), temperature (333 K),
CH CN (10 mL).
2 2 4
O in the presence of ZnWO 2.0 mol% Dy catalyst. Reaction
3+
gen peroxide in the presence of Dy
adapted from ref. 38).
doped zinc tungstate catalyst
(
3
6
666
|
New J. Chem., 2021, 45, 6661–6670
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
difference was observed in both conversion and reaction
selectivity.
The influence of temperature was assessed, without over-
taking the boiling point of acetonitrile (Fig. 7). The initial rate
of reaction was very sensitive to the increasing temperature, as
was also observed for the conversion of the reactions.
Conversely, varying the reaction temperature we noticed that
the selectivity of products remained almost untouchable
(Fig. 9). Neraldehyde and mainly the nerol epoxide were two
oxidation products formed (Scheme 2).
The reaction scope was extended to the other terpenic
alcohols and the main results are shown in Fig. 8. The results
of the nerol oxidation were compared to those obtained with
different alcohols; geraniol (i.e., allylic primary alcohol and
geometric isomer of the nerol), borneol (i.e., secondary bicyclic catalyst (15 mg), CH
alcohol), and a-terpineol (unsaturated tertiary alcohol) (Fig. 10).
The substrates were chosen because they have hydroxyl
Fig. 10 Conversion and selectivity of different substrate oxidation reac-
tions with H
conditions: nerol (1.375 mmol), H
CN (10 mL).
2
O
2
in the presence of ZnWO
4
2.0 mol% Dy catalyst. Reaction
2 2
O
(2.750 mmol), temperature (333 K),
3
groups and double bonds with different reactivity. The pair
(
Z) nerol and (E) geraniol are the cis and trans isomers,
respectively. Although both have been completely consumed
ca. 100%, after 8 h), and both had been converted to aldehydes
(
with the same selectivity, their double bonds presented different
reactivity, revealing that the reaction selectivity was distinct.
While the geraniol provided two epoxide products (i.e., geraniol Scheme 3 Oxidation reaction of geraniol to geranaldehyde, geraniol
epoxide and geraniol diepoxide) (Scheme 3), the nerol provided epoxide and geraniol diepoxide.
only the epoxide product (Scheme 2).
Firstly, it is important to highlight that although Tatsumi
et al. previously described that the oxidation reaction of these presence of the allylic hydroxyl groups.
2
1
Therefore, the
alcohols toward aldehydes is minimized, attributing it to the preferential formation of epoxides could be expected. Recently,
we have assessed the oxidation of terpenic alcohols over
3
8
niobium catalysts and verified the same effect.
On the other hand, the literature describes that epoxidation
3
9
of this double bond is a hydroxyl group-assisted reaction.
In oxidation reactions of the geraniol by hydrogen peroxide
over tungsten, titanium, niobium or vanadium oxide catalysts,
the coordination of metal oxide catalysts to the olefin double
bond, as well as the peroxidation of the metal are always
3
4,35
involved.
The peroxidation step of the niobium catalyst
and the interaction with the organic substrate are attributed to
the different species of active sites present in the niobium
catalyst. Lewis acid sites and Brønsted acid sites allow the
simultaneous adsorption of the oxidant and the organic sub-
4
0
strate even in the presence of water. Herein, we suppose that
both tungsten and dysprosium are active sites for this reaction.
The different chemoselectivity of the oxidation reactions of
geraniol and nerol can be assigned to the steric hindrance
effects on the double bond of nerol (i.e., cis isomer). It hampers
the epoxidation that led to diepoxide. Conversely, in the trans
isomer (i.e., geraniol), this effect is absent and the diepoxide is
Fig. 9 Conversion and selectivity of nerol oxidation reactions with H
in the presence of ZnWO – 2.0 mol% Dy catalyst at different tempera-
tures. Reaction conditions: nerol (1.375 mmol), H (2.750 mmol),
catalyst (15 mg), CH CN (10 mL).
2 2
O
4
2 2
O
3
4
1
easily formed.
Compared to the geraniol and nerol, the reactivity of a-terpineol
was significantly lower; only 35% conversion was achieved.
However, even with a tertiary alcohol, two oxidation products
were formed. In the presence of ZnWO
the double bond of terpineol was epoxidized providing the
Scheme 2 Oxidation reaction of nerol to nerol epoxide and neraldehyde. a-terpineol epoxide (Scheme 3). Even in a lower yield than that
4
2.0 mol% Dy catalyst,
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6661–6670 | 6667

View Article Online
Paper
NJC
achieved in the reactions of nerol and geraniol, an epoxidation
product was formed. The lower epoxide formation when
compared to the other two terpenic alcohols can be attributed
to the absence of the hydroxyl group in the allylic position to
the double bond, which assists this kind of epoxidation. In
addition, a rare cyclization reaction of this epoxide led to the
formation of p-mentan-2-ol, 1,8 epoxide (Scheme 4).
The formation of this bicyclic compound can be assigned to
the nucleophilic attack of the hydroxyl group on the trisubstituted
carbon of a-terpineol epoxide, which provokes the rupture of the
oxirane ring, followed by the prototropism between the hydroxyl
group and the oxygen atom of the epoxide. Previously it
was revealed that 1,2-epoxy-cyclohexanes exist in a half-chair
conformation, and the hydroxy-substituted isopropyl groups then
Fig. 11 Evaluation of the efficiency in the nerol ns in recovery and
4 2 2
recycling tests of the ZnWO 2.0 mol% Dy catalyst with H O oxidation
require this conformation for a nucleophilic attack on the reaction. Reaction conditions: nerol (1.375 mmol), H O2 (2.750 mmol),
2
4
2
epoxide.
The reaction between borneol and hydrogen peroxide in the
presence of the ZnWO – 2.0% Dy catalyst was highly selective
3
temperature (333 K), catalyst (15 mg), CH CN (10 mL).
4
third cycle it was verified that the oxidation became less
efficient, with a consequent increase in the formation of
oligomers.
toward the carbonylic product (ca. 100% selectivity, Fig. 8),
exclusively leading to camphor formation (Scheme 5).
The lower conversion achieved in borneol oxidation
Aiming to investigate the reason for the loss in activity of the
catalyst after the successive reuses, analyses of XRD patterns
and infrared spectroscopy were performed. A comparison of
reused and fresh catalyst is shown in Fig. 12 and 13.
As shown in Fig. 10, the diffractogram profile of the reused
catalyst showed some differences in relation to the fresh
(
ca. 60%) when compared to the oxidation of other alcohols
is a consequence of different reactivity of hydroxyl groups.
While geraniol and nerol have primary hydroxyl groups,
borneol has the secondary one. The total conversion of the
borneol occurred after longer reaction times (ca. 20 h).
The reusability of the ZnWO4 2.0 mol% Dy catalyst was
assessed. Kinetic curves and the main results of conversion and
selectivity are shown in Fig. 11.
Although omitted in Fig. 11, a high recovery rate of the
catalyst was achieved in all the recycles (ca. 495%), indicating
that the procedure used was very efficient. After the first cycle
of reuse, there was a gradual decrease in the conversion of
reactions, which was followed during three cycles of recovery/
reuse. In terms of selectivity, it was noticed that until the
second cycle it remained basically constant, however, in the
Fig. 12 Powder XRD patterns of a fresh and reused ZnWO
catalyst in nerol oxidation reactions with H . Reaction conditions: nerol
1.375 mmol), H (2.750 mmol), temperature (333 K), catalyst (15 mg),
CH CN (10 mL).
4
2.0 mol% Dy
2 2
O
(
2 2
O
3
Scheme 4 Oxidation reaction of a-terpineol to p-menthan-2-ol, 1,8-
epoxy and a-terpineol epoxide.
Fig. 13 Infrared spectra of the fresh and reused ZnWO
catalyst in nerol oxidation reactions with H . Reaction conditions: nerol
1.375 mmol), H (2.750 mmol), temperature (333 K), catalyst (15 mg),
CH CN (10 mL).
4
2.0 mol% Dy
2 2
O
(
2 2
O
Scheme 5 Oxidation reaction of borneol to camphor.
3
6
668
|
New J. Chem., 2021, 45, 6661–6670
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
catalyst. With the aid of calculating the crystallite size, it is with controlled morphological properties, which demonstrated
possible to evaluate the difference between the two materials. to be effective catalysts for the selective epoxidation of allylic
The fresh catalyst presents a crystallite size of approximately terpenic alcohols with hydrogen peroxide. Reactions were
7
1
nm, while after the fourth run this size was increased to carried out in the presence of a dysprosium doped zinc tung-
3 nm. Once we had previously demonstrated that the lower the state solid catalyst (ca. 0.5; 1.0, 2.0 mol% Dy), and hydrogen
crystallite size, the higher the catalytic activity, which partly peroxide, a cheap and green oxidant. ZnWO
4
2.0 mol% Dy was
explains the loss of activity of the ZnWO 2.0 mol% Dy catalyst. the most active and selective catalyst, a consequence of the
4
In previously published works, the authors found that lowest crystallite size, and the highest surface area, pore
crystallinity is a factor that can make the difference in catalytic volume, and Dy load. As described for tungsten, niobium, or
activity. It was found that in materials with less crystallinity and titanium catalysts, the epoxidation of allylic terpenic alcohols
smaller crystallite size, there is a higher ability to interact more (i.e., nerol and geraniol) with H O is a hydroxy-assisted reac-
2
2
easily with hydrogen peroxide molecules, which consequently tion. The different stereochemistry of nerol and geraniol
leads to the cleavage of O–O bonds in a greater number of revealed that only the geraniol was converted to epoxide and
reactive oxygen species that will directly contribute to the diepoxide products. The reactivity of other terpenic alcohols
1
8,30,34
oxidation process.
was also evaluated. Noticeably, the double bond of a-terpineol
On the other hand, comparing the bands referring to the was epoxidized, and after undergoing a nucleophilic attack of
ꢂ1
bending of the W–O bonds (ca. 463 cm ), and the symmetric the hydroxyl group led to the formation of a rare product
and asymmetric stretching modes of the W–O and Zn–O bonds (i.e., p-mentan-2-ol, 1,8 epoxide). Borneol was selectively
ꢂ1
(
ca. 516 and 425 cm wavenumbers), present in the infrared converted to camphor, however, since that it is a secondary
spectrum of the fresh catalyst, and the bands present in the alcohol, its total conversion occurred after longer reaction
spectrum of the used catalyst, it is possible to see that they were times (ca. 20 h). The reusability of the ZnWO 2.0 mol% Dy
4
6 6
shifted. This suggests that the ZnO and WO octahedral units, catalyst was investigated. High recovery rates (ca. 95%) were
respectively, undergone a distortion, possibly caused by the loss obtained. After recovery, the solid catalyst was reused without
of Dy ions. In addition to this effect, it is possible also that the loss of selectivity, however, there was a decrease of conversion
size of the nanoparticles has undergone an increase, due to after the third run. Infrared spectroscopy analysis and XRD
heating in the recycling process and due to the coalescence patterns of the reused catalyst showed that the crystallite size of
effect caused in the reaction medium.
the recovered catalyst was increased, which compromised their
The leaching of the catalyst was also assessed as reported in activity.
4
3–46
3+
the literature.
Due to the difficulty of detecting the Dy
ions in solution, two sets of experiments were performed (see
Fig. SM2 and SM3, ESI†). The following approaches were used
to evaluate the leaching. In the first one, two solutions with
only acetonitrile were heated for 30 minutes, one containing
the pure zinc tungstate catalyst and the other containing the
Conflicts of interest
There are no conflicts to declare.
3
+
Dy -doped zinc tungstate catalyst. Afterward, the catalysts were Acknowledgements
removed, and the reactions proceeded for 8 h. While the first
The authors thank the Federal University of Viçosa, the
reaction was almost stopped, the second one proceeded to
3
+
PPGMQ-MG, and the development agencies CNPq and
FAPEMIG. This study was financed in part by the Coordenaç ˜a o
de Aperfeiçoamento de Pessoal de N ´ı vel Superior – Brasil
achieve a conversion of 80%. It is suggestive that Dy ions
truly are being leached to the solution (Fig. 2SM and 3SM, ESI†).
Additionally, a second test was performed with the two catalysts;
they were heated for 30 minutes in a solution containing the
oxidant and substrate, after that, the catalysts were removed (i.e.,
pure, or doped zinc tungstate), and the reaction was monitored
(CAPES) – Finance Code 001 and PROCAD 2013/2998/2014.
Notes and references
3
+
for 8 h. Once, the results confirmed the Dy leaching, indicated
by the conversion of 80%. However, different than verified with
acetonitrile, there was also leaching of zinc tungstate, although
lower than the leaching of the Dy-doped catalyst. It was revealed
that the presence of the aqueous solution of the oxidant
increased the leaching of both catalysts (i.e., pure, and doped
zinc tungstate).
1 M. Kope ˇc n ´a , M. Mach ´a ˇc ek, A. Nov ´a ˇc kov ´a , G. Paraskevopoulos,
J. Roh and K. V ´a vrov ´a , Sci. Rep., 2019, 9, 1–12.
2 W. Schwab, C. Fuchs and F. Huang, Eur. J. Lipid Sci.
Technol., 2013, 115, 3–8.
3 M. J. da Silva, A. A. Julio and K. T. dos Santos, Catal. Sci.
Technol., 2015, 5, 1261–1266.
4
M. J. da Silva and J. A. Villarrea, Catal. Lett., 2017, 147,
646–1653.
1
5
C. B. Vilanculo, M. J. da Silva, M. G. Teixeira and
J. A. Villarreal, RSC Adv., 2020, 10, 7691–7697.
Conclusions
4
ZnWO –Dy catalysts were efficiently synthesized by co-precipitation
6 M. Besson, P. Gallezot and C. Pinel, Chem. Rev., 2013, 114,
followed by the microwave-assisted hydrothermal heating method,
1827–1870.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6661–6670 | 6669

View Article Online
Paper
NJC
7
8
9
L. A. S. Viana, G. R. N. da Silva and M. J. da Silva, Catal. Lett., 27 M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier,
2
018, 148, 374–382.
F. Rodriguez-Reinoso, J. Rouquerol and K. S. W. Sing, Pure
Appl. Chem., 2015, 87, 1051–1069.
28 V. G. Teixeira, F. M. B. Coutinho and A. S. Gomes, Quim.
Nova, 2001, 24, 808–818.
Z. Guo, B. Liu, Q. Zhang, W. Deng and Y. Yang, Chem. Soc.
Rev., 2014, 43, 3480–3524.
S. Serra and D. de Simeis, Catalysts, 2018, 8, 362.
1
1
1
0 K. C. Nicolaou, C. V. C. Prasad, P. K. Somers and 29 L. R. Pizzio and M. N. Blanco, Microporous Mesoporous
C. K. Hwang, J. Am. Chem. Soc., 1989, 111, 5330–5334. Mater., 2007, 103, 40–47.
1 P. T. Anastas and M. M. Kirchhoff, Acc. Chem. Res., 2002, 35, 30 M. Ziolek, I. Sobczak, P. Decyk, K. Soba n´ ska, P. Pietrzyk and
86–694. Z. Sojka, Appl. Catal., B, 2015, 164, 288–296.
2 W. M. Ventura, D. C. Batalha, H. V. Fajardo, J. G. Taylor, 31 M. Ziolek, I. Sobczak, P. Decyk, K. Soba n´ ska, P. Pietrzyk and
6
N. H. Marins, B. S. Noremberg, T. Ta n´ ski and
N. L. V. Carre n˜ o, Catal. Commun., 2017, 99, 135–140.
3 J. Piera and J. Baeckvall, Angew. Chem., Int. Ed., 2008, 47,
Z. Sojka, Appl. Catal., B, 2015, 164, 288–296.
32 D. P. Dutta and P. Raval, J. Photochem. Photobiol., A, 2018,
357, 193–200.
1
1
1
1
3
506–3523.
4 N. C. Coronel and M. J. da Silva, J. Cluster Sci., 2018, 29,
95–205.
5 M. J. da Silva and N. A. Liberto, Curr. Org. Chem., 2016, 20,
263–1283.
6 M. J. da Silva, L. C. de Andrade Leles, S. O. Ferreira, R. C. da
33 J. Lu, M. Liu, S. Zhou, X. Zhou and Y. Yang, Dyes Pigm., 2017,
136, 1–7.
34 P. Manjunathan, S. P. Maradur, A. B. Halgeri and
G. V. Shanbhag, J. Mol. Catal. A: Chem., 2015, 396, 47–54.
35 D. Liu, P. Yuan, H. Liu, J. Cai, D. Tan, H. He, J. Zhu and
T. Chen, Appl. Clay Sci., 2013, 80, 407–412.
1
1
Silva, K. de, V. Viveiros, D. M. Chaves and P. F. Pinheiro, 36 S. Thanasilp, J. W. Schwank, V. Meeyoo, S. Pengpanich and
ChemistrySelect, 2019, 4, 7665. M. Hunsom, J. Mol. Catal. A: Chem., 2013, 380, 49–56.
7 M. J. da Silva, P. H. S. Andrade, S. O. Ferreira, C. B. Vilanculo 37 M. Sun, J. Xia, H. Wang, X. Liu, Q. Xia and Y. Wang, Appl.
and C. M. Oliveira, Catal. Lett., 2018, 148, 2516–2527. Catal., B, 2018, 227, 488–498.
8 D. C. Batalha, S. O. Ferreira, R. C. da Silva and M. J. da Silva, 38 F. Zhang, H. Hu, H. Zhong, N. Yan and Q. Chen, Dalton
ChemistrySelect, 2020, 5, 1976–1986. Trans., 2014, 43, 6041–6049.
9 C. B. Vilanculo and M. J. Da Silva, New J. Chem., 2020, 44, 39 D. C. Batalha, N. H. Marins, R. M. Silva, N. L. V. Carre n˜ o,
813–2820. H. V. Fajardo and M. J. da Silva, Mol. Catal., 2020, 489, 110941.
0 X. Chai, J. Li, X. Wang, Y. Li and X. Yao, RSC Adv., 2017, 7, 40 R. Kumar, G. C. G. Pais, B. Pandey and P. Kumar, J. Chem.
0046–40052. Soc., Chem. Commun., 1995, 1315–1316.
1 R. F. Gonçalves, E. Longo, A. P. A. Marques, M. D. P. Silva, 41 I. W. C. E. Arends and R. A. Sheldon, Appl. Catal., A, 2001,
L. S. Cavalcante, I. C. Nogueira, I. M. Pinatti, P. F. S. Pereira 212, 175–187.
and M. J. Godinho, J. Mater. Sci.: Mater. Electron., 2017, 28, 42 W. Adam, A. Corma, T. I. Reddy and M. Renz, J. Org. Chem.,
5466–15479. 1997, 62, 3631–3637.
2 Z. Liu, J. Tian, D. Zeng, C. Yu, L. Zhu, W. Huang, K. Yang 43 R. M. Carman and M. T. Fletche, Aust. J. Chem., 1984, 37,
and D. Li, Mater. Res. Bull., 2017, 94, 298–306. 1117–1122.
3 A. Phuruangrat, P. Dumrongrojthanath, T. Thongtem and 44 S. R. Churipard, P. Manjunathan, P. Chandra, G. V. Shanbhag,
1
1
1
2
2
2
4
1
2
2
S. Thongtem, J. Ceram. Soc. Jpn., 2017, 125, 62–64.
R. Ravishankar, P. V. C. Rao, G. S. Ganesh, A. B. Halgeri and
2
2
4 M. Nishiura and Z. Hou, Nat. Chem., 2010, 2, 257–268.
S. P. Maradur, New J. Chem., 2017, 41, 5745–5751.
5 S. Molaei and M. Ghadermazi, Solid State Sci., 2020, 45 S. R. Churipard, K. S. Kanakikodi, J. R. Choudhuri and
01, 106148. S. P. Maradur, RSC Adv., 2020, 10, 35988–35997.
6 K. Thirumalai, M. Shanthi and M. Swaminathan, Mater. 46 K. S. Kanakikodi, S. R. Churipard, A. B. Halgeri and
1
2
Today: Proc., 2018, 5, 15342–15347.
S. P. Maradur, Sci. Rep., 2020, 10, 13103–13111.
6
670
|
New J. Chem., 2021, 45, 6661–6670
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021