Activity of Molybdate Intercalated Layered Double Hydroxides in the Oxidation of Styrene with Air

Activity of Molybdate Intercalated Layered Double Hydroxides in the Oxidation of Styrene with Air tài liệu, giáo án, bài...

Activity of Molybdate-Intercalated Layered Double Hydroxides

in the Oxidation of Styrene with Air

Nguyen Tien Thao1• Nguyen Duc Trung1•Dang Van Long1

Received: 2 February 2016 / Accepted: 3 February 2016

 Springer Science+Business Media New York 2016

Abstract Molybdate anions were intercalated into the

interlayer spacings of (Mg, Al) like-hydrotalcite

com-pounds as interlayer compensating anions The synthesized

samples have been characterized by XRD, FT-IR, Raman,

EDS, UV–vis, BET, and XPS The solids possess lamellar

structure and uniform platelet particles There is mainly

Mo(VI) present in both tetrahedral and octahedral

config-uration in the samples All the synthesized catalysts have

been tested for the liquid oxidation of styrene at mild

conditions Under reported conditions, styrene conversion

varies with the total amount of molybdate ions

Ben-zaldehyde and styrene oxide were two major components

in the product mixture The selectivity to styrene oxide was

found to be associated with the nature of oxidants and the

amount of tetrahedrally-coordinated Mo species in layered

double hydroxides while that to benzaldehyde is related to

the overall amount of MoO4-anions in the sample

Graphical Abstract Product distribution obtained on

Mg–Al–Molybdate like hydrotalcite catalysts in the liquid

oxidation of styrene with air

80 90 100 110

4 hours

4 hours

24 hours

24 hours

0 10 20 30 40 50 60 70 80 90 100

Temperature (

oC)

Keywords Molybdate Epoxidation  Styrene  Oxidation Hydrotalcite  LDH

1 Introduction

Olefin oxidation is an important catalytic process in the chemical industry This reaction produces two valuable products, aldehyde and epoxide, which are important chemical feedstock for the production of a wide variety of fine chemicals and pharmaceuticals In tradition, this reaction has been, in general, carried out over both homogeneous and heterogeneous catalyst systems [1 3] However, the use of homogeneous catalysts has recently become less attractive because of the difficulty in the

Electronic supplementary material The online version of this

article (doi: 10.1007/s10562-016-1710-0 ) contains supplementary

material, which is available to authorized users.

& Nguyen Tien Thao

ntthao@vnu.edu.vn

1 Faculty of Chemistry, Vietnam National University, Hanoi,

19 Le Thanh Tong ST, Hanoi 10000, Vietnam

DOI 10.1007/s10562-016-1710-0

separation of products and catalysts Thus, heterogeneous

or immobilized catalysts have been currently paid much

attention Among huge amounts of heterogeneous catalysts

used for this oxidation reaction, firstly it should be taken

account of titanium-based catalysts, VIIIB-based metals or

oxides, noble metals and their derivatives, molecular sieves

[2 10] These catalysts are reported to have a good activity

in the oxidation of olefinic hydrocarbons Indeed, titana

was known as an excellent photocatalyst for the oxidation

of organic compounds [2,4 6] This oxide can be used as

supported oxide catalysts (e.g TiO2/SiO2, TiO2/MCM-41)

[2,11,12], mixed oxides (e.g TiO2–SiO2, TiO2–ZrO2) [13,

14], framework-substituted molecular sieves (e.g TS-1,

TS-1, Ti–MCM-41, Ti–MCM-48, Ti-b) [2,4 6,15] in the

oxidation of unsaturated hydrocarbons In practice, TiO2

-based photocatalysts are active for the oxidation of olefins,

but rapidly deactivated in the presence of water

Mean-while, titanium-substituted molecular sieves can perform

the oxidation reaction in aqueous solution because the

intra-lattice titanium can make hydroperoxo complexes by

hydrolysis of Ti–O–Si connectivities Thus, TS-1 was very

selective for the epoxidation of styrene and its catalytic

activity is remarkably related to the amount of Ti(IV) in the

crystalline framework, catalyst morphology, oxidant,

sol-vent, preparation method [2, 4 6] For instance, Yeung

et al [5] observed a high selectivity to phenylacetaldehyde

in the oxidation of styrene over intra-framework titanium

TS-1 catalyst However, this catalyst possesses small pore

size which only allows small reactants accessibly to the

active sites Thus, other catalysts developed from the group

VIIIB-based metals have been recently reported for the

oxidation of olefins [7,9,16,17] The catalytic activity is

governed as the appropriate oxidant was used and the

oxidation state of transition metal ions is controlled [7 9,

16] In practice, aside from Co-based catalysts, other

VIIIB-metal based catalysts seldom catalyze the

epoxida-tion of olefin when oxygen/air is used as oxidant [2,7 10,

16, 17] Fe–Ni—containing catalyst exhibits a good

activity in the conversion of styrene with H2O2or peracids

only [1,2,7 9] Co–Ni–Cu nanoxides are recently reported

as active catalysts for the oxidation of styrene with

tert-butyl hydroperoxide (TBHP) [7] Co-based catalysts can

oxidize styrene into styrene oxide although a small amount

of benzaldehyde is always detected in the product mixture

as air is used as oxidant [8,9] The other catalysts may be

designed from transition metals such as copper, silver,

vanadium, tungsten, molybdenum, manganese… although

the catalytic activity of these catalysts is still modest [2,10,

18–22] In general, the group VIB–VIIB metals are usually

incorporated into complex-derived compounds or to make

metal immobilized catalysts [2,21–24] In latter cases, the

active components are present as transition metal

oxo-complexes which undergo the selective oxidation of

alkenes [22,25–30] Following this trend, we have incor-porated some oxoanions into the interlamellar spaces of hydrotalcites for the purpose of the preparation of oxida-tion/reduction catalysts It was well known that hydrotal-cites are common anionic clays described by the empirical formula A2þ1yB3þy ðOHÞ2þy

 Xz

y=z nH2O

; where

A2?and B3?are the metal cations, water and exchangeable inorganic or organic charge-compensating anions (Xz-) are present in the interlayer galleries [31, 32] This composi-tion leads to the preparacomposi-tion of numerous hydrotalcite derivatives by substitution of A2? (or B3?) with another transition metal ion or intercalation of foreign anions in the interlayer domains This allows a great flexibility in mixing

A2? and B3? cations to obtain several desired reduction/ oxidation properties due to the brucite sheets to accom-modate cations of various sizes and valences [19,27–30]

In other ways, substitution of charge-compensating ions with transition metal oxoanions also gives rise to novel redox catalyst systems In reality, a series of layered double hydroxides intercalated with molybdate [21, 23, 33], tungstate [21], chromate [32], and manganate [31] have been prepared and used as catalysts for the oxidation reaction In overall, the conversion and product selectivity were reported to be associated with the position of octa-hedral sites, nature of interlayer Xz- anion, Mg/Al ratio, basicity of the LDH catalyst [9,30,31,34,35]

It was known that some Mo-containing hydrotalcite like compounds were proved to be active catalysts for different oxidation reactions such as selective olefin oxidation [23,

33] or oxidative dehydrogenation of propane [36] In the present work, a series of Mg–Al molybdate oxoanion-in-tercalated layered double hydroxides was prepared and expected to have a good reduction–oxidation activity in the tailored oxidation of C=C bond The synthesized catalysts have been tested for the liquid oxidation of styrene with air and the effects of molybdate contents, reaction variables are reported

2 Experimental Section

2.1 Catalyst Preparation

Molybdate-intercalated layered double hydroxides were prepared through a modified conventional co-precipitation method Solution A was prepared by dissolving Mg2?and

Al3? metal nitrate salts with different Mg2?/Al3? molar ratio into 150 mL distillated water (Table 1S in Supple-mentary Materials) Solution B was prepared by adding desired amounts of ammonium heptamolybdate ((NH4)

6-Mo7O24) and NaOH into 100 mL of distillated water Two

solutions (A and B) were added simultaneously with a

constant flow rate of 1 mL/min under magnetic stirring for

3 h while the pH was adjusted at 9.0 The weighted

amounts of starting chemicals are reported in Table 1S

(Supplementary Materials) The resultant was then

sub-mitted to an aging treatment at 65C for 24 h, followed by

filtration, washing with hot distilled water, and drying at

70C for 24 h The obtained solid was ground into

pow-der In the case of preparation of the Mg/Al intercalated

carbonate anion in the free gallery, designated as MAC-00

(Magnesium, Aluminum, Carbon), ammonium

hepta-molybdate was replaced by sodium carbonate

For a mixed oxide (reference) sample with Mg/Al/Mo

molar ratio of 6/4/2, three salts, ((NH4)6Mo7O24,

Al(NO3)39H2O, Mg(NO3)26H2O were blended together

prior to calcinate at 450C for 3 h The reference sample is

designated as MiOx

2.2 Catalyst Characterization

Powder X-ray diffraction (XRD) patterns were recorded on

a D8 Advance-Brucker instrument using CuKa radiation

(k = 0.1549 nm) Fourier transform infrared (FT-IR)

spectra were obtained in 4000–400 cm-1range on a FT/IR

spectrometer (DX-Perkin Elmer, USA) The Raman spectra

of samples were analyzed by a LabRAM HR800

spec-troscopy (HORIBA, French) The recorded spectral range

was 100–2500 cm-1 and scanned three times with the

wavelength of laser beam of 632 nm UV–vis spectra were

collected with UV–Visible spectrophotometer, JASCO

V-670 BaSO4 was used as a reference material The

spectra were recorded at room temperature in the

wave-length range of 200–800 nm The XPS analysis was made

on a photoelectron spectrometer (KRATOS Axis 165,

Shimadzu, Japan) with Mg Ka radiation (1253.6 eV)

Deconvolution of the experimental photopeaks was carried

out using a Lorentzian peak fit procedure The scanning

electron microscopy (SEM) images were obtained with a

JEOS JSM-5410 LV Energy-dispersive spectroscopy

(EDS) data were obtained from Varian Vista Ax X-ray

energy-dispersive spectroscope

2.3 Oxidation of Styrene

The catalytic oxidation of styrene in N,N0-dimethylformide

(DMF) solvent was carried out in a 100 mL three-neck

glass flask fitted with a reflux condenser For a typical run,

17.4 mmol of styrene, 7.0 mL of solvent and 0.2 g of

catalyst were loaded into the flask After the reaction

mixture was magnetically stirred and heated to the desired

temperature, then t-butyl hydrogen peroxide (TBHP, 70 %,

Sigma Aldrich) or hydrogen peroxide solution (H2O2,

30 %) was dropped into the flask As air was used, the flow

of air (5 mL/min) was conducted into stirred reaction mixture and the reaction time starts recorded After reac-tion, the mixture was cooled down to room temperature and then catalyst was filtered off The reaction product mixture was then analyzed by gas chromatography and GC–MS (HP-6890 Plus, capillary column HP-5 MS cross-linked PH 5 % PE Siloxane, 30 m 9 1 lm 9 0.32 lm)

3 Results

3.1 Characteristics of the Catalysts

The nomenclatures of samples prepared in this study and some typical characteristics are presented in Table1 Powder-X ray patterns of all samples in Fig.1 display the main reflection planes which are typically characteristics for layered double hydroxide material of sample MAC-00

to MAM-20 [8,26,37] Indeed, two sharp and symmetric peaks at low 2-theta of 11.20, 22.49 are essentially assigned to the reflections by the basal planes of (003), (006), respectively The other broad and asymmetric peaks

at 2-theta of 34.25, 38.24, 45.53, 60.23, 61.37 correspond respectively to the reflections by the basal planes of (012), (015), (018), (110), and (113), confirming the formation of

a crystallized layered double hydroxide structure [23, 33,

34, 36–39] For sample MAM-30, some reflection lines characterizing for Al(OH)3phase appear in Fig.1[JCPDS File 01-072-0623], representing the formation of a mixture

of products instead of single LDH phase Furthermore, X-ray reflection signals of the molybdate-intercalated samples are somewhat broader and noisier than that of Mg– Al–carbonate hydrotalcite (MAC-00) as seen in Fig.1 [9,

10, 38] The c parameter of the Mo-containing LHD is higher than that of the hydrotalcite reference MAC-00, reflecting a successful intercalation of MoO42-anions into the interlayer spaces between (Mg, Al) brucite-like sheets [23,33,36,37,39,40] The value of c parameter slightly decreases from sample MAM-10 to MAM-20 due to a stronger electrostatic interaction between brucite-like sheets and molybdate anions and a decreased molar ratio of Mg/Al [23,28,33,34]

In order to investigate the nature of molybdate anions in the synthesized samples, FT-IR, Raman, and UV–vis spectra were recorded FT-IR and Raman spectra of the catalysts are displayed in Fig.2 The FT-IR spectra of MAM-10 and MAM-20 are resembled As shown in Fig.2a, FT-IR spectra of the Mo-containing catalyst show broad adsorption bands suggesting a high disorder of the molecules in the interlayer galleries The broad band is found around 3450 cm-1due to the stretching mode of the hydroxyl groups present in the metal hydroxide layers, Mo-oxoanions, and water molecules in the interlayer domain

[34, 37, 40] A broaden band appeared at 920 cm-1 is

assigned to the vibrations of Mo=O in polymolybdate

Mo7O246- A band at 670 cm-1 with a shoulder at

856 cm-1 is typically characteristics for Mo–O–Mo stretching vibration of MoO42- in the interlayer region The lower wave number band at 546 cm-1and is assigned

Table 1 Catalyst

characteristics of all samples Batch # Expected formula a (A˚ )* c (A˚)* Mg (wt%) Al (wt%) Mo (wt%)

MAC -00 [Mg0.7Al0.3(OH)2](CO3)0.15mH 2 O 3.047 22.805 24.70 12.39 – MAM-10 [Mg0.8Al0.2(OH)2](MoO4)0.10mH 2 O 3.078 23.783 23.47 7.10 2.73 MAM-15 [Mg0.7Al0.3(OH)2](MoO4)0.15mH 2 O 3.074 23.755 20.41 8.94 4.02 MAM-20 [Mg0.6Al0.4(OH)2](MoO4)0.20mH 2 O 3.044 23.091 18.45 10.57 5.45 MAM-20R [Mg0.6Al0.4(OH)2](MoO4)0.20mH 2 O 3.036 22.725 18.48 10.32 4.97 MAM-30 [Mg0.4Al0.6(OH)2](MoO4)0.30mH 2 O – – 21.17 8.05 6.27

* a = 2 9 d110and c = 3/2(d003? 2 9 d006) [ 26 ] (MAM-20R: reused sample MAM-20 after a cycle)

2-thetra

MAC-00

MAM-10

MAM-15 MAM-20 MAM-30

MixO

Fig 1 Powder-XRD patterns of

catalyst samples

400 700 1000 1300 1600 1900 2200 2500 2800 3100 3400 3700 4000

Wavenumber (cm-1)

MAM-10

MAM-20

MAM-30

3480

990

1370

1658

798

670

856

926

894

546

451

Raman Shift (cm-1)

MAM-10 MAM-15

MAM-30 MixO

1000 826

673 385

479 343 296 251

Fig 2 FT-IR (a) and Raman spectra (b) of catalyst samples

to the lattice vibrations of Mg–O, Al–O bond The band

around 450 cm-1 has been ascribed to a condensed

[AlO6]3-group or as single Al–O bonds [34,35,39,40]

Figure2b shows the Raman spectra for molybdate

containing samples (MAM-10, 15, 30) and that for the

mixed oxide solid (MixO), for comparison The Raman

spectrum of MixO exhibits the characteristic sharp bands

of bulk MoO3 at 1000 cm-1 (Mo=O symmetric stretch),

826 cm-1 (Mo–O–Mo asymmetric stretching vibrations),

673 cm-1 (Mo–O–Mo symmetric stretching vibrations),

343 cm-1 (bending of terminal Mo=O), 260–220 cm-1

(the bridging Mo–O–Mo deformation) The Raman

spec-trum of Al2O3possesses a band at 386 cm-1 For a series

of hydrotalcite-like compounds, the Raman spectra show

that the band at 560 cm-1is assigned to the lattice

vibra-tion of the brucite octahedral sheets (Mg–O–Al) The

strong band at 1050 cm-1 is described for the nitrate

vibrations and this peak intensity gradually vanishes with

increasing amount of molybdates, indicating the successful

replacement of nitrate ions for molybdate anions in the

interlayer region [23] Evidently, MoO42- symmetric

stretching vibrations of the molybdates in the LDH

com-pounds were described by the appearance of bands at 908

and 892 cm-1 [23,29] Indeed, the two bands were

char-acteristics for two different species of MoO42-anions, the

first one is hydrated and the other one is bonded to the

brucite-like hydroxyl surface [34,37] A broad shoulder at

823 cm-1 is the antisymmetric stretching mode of the

MoO4units A band observed at 325 cm-1 is ascribed to

the bending mode of Mo–O [23, 41] Thus, the Raman

spectra let us suggest that MoO42--intercalated

like-hy-drotalcite compounds are prepared at basic preparation

conditions of pH = 9.0, in a good agreement with the

FT-IR, UV–Vis analyses and results reported in the literature

[34,37,39,41,42] For the molybdate-rich sample

(MAM-30), the bands at 940 and 359 cm-1 are assigned to the

Mo–O symmetric stretching mode and a Mo=O bending

mode in Mo7O246-anions, respectively [33,34,41]

Figure3 presents the UV–vis spectra of three

molyb-date-containing catalysts in the region of 220–800 nm The

absorption bands in the region of 220–280 nm is described

as the O2-to Mo6?charge transfer of the isolated MoO4

2-species with Mo in tetrahedral coordination [43,44] For a

higher molybdate contents (MAM-15 and MAM-30), the

absorption bands are more broaden and the edge of the

bands shifts to a higher wavelength, which is

characteris-tics of the presence of molybdenum polyoxoanions with

octahedral coordination although the presence of these

polyoxoanions is not expected in the experimental

condi-tions used during catalyst preparation [35]

One of the most useful techniques for the investigation

of the chemical state of the atoms present in the surface

layer region of the heterogeneous catalyst is X-ray

photoelectron spectroscopy (XPS) In the present work, the sample MAM-15 was chosen to record the XPS spectra Figure4a presents the Mo 3d region which shows Mo 3d3/2–Mo 3d5/2 doublets due to the spin–orbit coupling Furthermore, the peaks are very broad, reflecting molyb-denum species having different environments For the fresh MAM-15 sample, the doublet with the main Mo 3d5/2and 3d3/2 peak at 232.6 and 235.9 eV respectively, is firmly attributed to typical Mo(VI) in MoO42- [35, 45] The additional shoulders at 231.7 and 234.9 eV are assigned to Mo(V) species due to the possibility of the charge transfer between (Mo6?? O2-) and (Mo5?? O-) [24, 34, 46] This is strongly corroborated by the analysis of the O 1s photoline of sample MAM-15 Indeed, the binding energy value of O 1s XPS signals observed at 532.4 eV for the fresh sample MAM-15 is assigned to the O2-(Fig.4b) [35,

38,45] However, it is likely composed of two overlapping photoelectron lines which can be deconvoluted into two peaks at 532.4 and 531.5 eV (Fig.4b) The BE value at 532.4 eV is attributed to the metal hydroxides while the other signal at 531.6 eV is assigned to oxygen O- in the oxomolybdenum [35,45,47]

SEM micrographs reveal the morphology of the cata-lysts SEM images of some representative samples are shown in Figs.5and 2S The morphology of these catalysts

is typical characteristics of hydrotalcite-like materials with platelet thickness ranging from 15 to 30 nm Larger LDH plate was observed on the molybdate-rich sample (MAM-15), possibly affected by pH constant during the catalyst preparation (Table 1S) [34,38,44,48–50] Indeed,

MAM-15 is mainly thin disk-like platelets about 200 nm in diameter and 30 nm in thickness Thus, it is expected to be existence void spaces between catalyst platelets In prac-tice, nitrogen adsorption/desorption measurement for Mo-containing LDH samples shows isothermal curves with a plateau from 0 to 0.5 and a hysteresis loop in the range of 0.62–0.95 (Fig 1S) The patterns are likely classified to the

Wavelength (nm)

MAM-10

MAM-15

MAM-30

Fig 3 UV-vis spectra of samples

II type and the hysteresis loops are closely to the

H3-classification, suggesting that these solids are either

mesopores or nonporous materials In the present work, the

H3-like hysteresis loop is described to the nitrogen

con-densation/evaporation phenomena in slit-shaped pores

created by the agglomeration of uniform plate-like particles

[7,36,38] The specific surface area of samples is in the

range of 5–20 m2/g

Scanning electron microscopy and energy-dispersive

X-ray spectrometry (SEM–EDS) analysis provides local

information of the concentrations of different elements in

the outermost layers of the platelet of LDH Alumina,

magnesium, molybdenum, and oxygen are clearly

identi-fied on the platelet surface of all samples as displayed in

Fig.6 No major difference in the percentage of elements

on four spots of each sample indicates a good dispersion of

elements in the LDH at the micrometer scale [32, 38]

Furthermore, molybdenum metal content is close to the

theoretical value, but observably minor changes after a

reaction cycle Analytical results of the synthesized solids

using the EDS technique are reported in Table1 The EDS

analysis clearly shows that traces of nitrogen for nitrate residue are present within the limits of the experiment Thus, it is concluded that the balance of the negative charge must be due to hydroxyl and molybdate anions, in good accordance with results of FT-IR and Raman spectra

Binding Energy (eV)

MAM-15 (Fresh)

MAM-15 (Spent)

232.3

235.4

Binding Energy (eV)

MAM-15 (Fresh)

MAM-15 (Spent)

531.6 532.3

Fig 4 XPS scan of Mo 3d (a) and O1s (b) for MAM-15 samples before and after reactions at 90 C, 4 h, DMF solvent, air oxidant for spent sample MAM-15

Fig 5 SEM images of MAM-10 (a) and MAM-15 (b)

Energy (keV)

O Mg

Al Mo

Mo

MAM-10 MAM-15

MAM-20 MAM-30 MAM-20R -Spent N

Fig 6 EDS spectra of Mg–Al–molybdate hydrotalcite-like compounds

3.2 Catalytic Studies

The catalytic activity of Mg/Al–molybdates

hydrotalcite-like catalysts in the liquid oxidation has been examined at

atmospheric pressure and air was led into the reaction

mixture without any further purification

3.2.1 Oxidation of Styrene Catalyzed by

Molybdate-Intercalated LDH Catalysts

A series of catalysts with different compositions has been

tested at the same conditions Figure7 presents the

cat-alytic activity of all samples in the styrene oxidation For

the purpose of comparison, a blank test (no catalyst) has

been also performed, giving null conversion In addition,

ammonium heptamolybdate, Mg/Al–CO3 (sample

MAC-00), and the mixed oxides (sample MixO) each has been

tested for the reaction of styrene with air for comparison,

but only traces of products were detected after 6 h-reaction

time Meanwhile, Mo-containing hydrotalcite-like

com-pounds show a good catalytic activity in the oxidation of

styrene under similar experimental conditions Styrene

conversion ranges from 7 to 12 %, substantiating that

Mo-species in the layered double hydroxides act as active

components for the selective oxidation of styrene In detail,

the styrene conversion varies with the overall amount of

molybdate anions in the catalysts as follows of

MAM-20 [ MAM-15 [ MAM-10 [ MAM-30 The highest

molybdate-containing sample shows a lowest styrene

conversion (Fig.7) This is not surprising as considering

the characterization of the sample MAM-30 since it is

constituted of oxides, hydroxides and has a low Mg/Al

ratio (Fig.1) Indeed, a low activity of sample MAM-30 is

explained by a poorer crystalline hydrotalcite-like structure

and a smaller amount of Mo species and a lower basicity (Mg/Al ratio) [9,23,42,51,52]

The two main products are benzaldehyde and styrene oxide in the product mixture With the exception of two reference samples (MAM-00 and MixO), the other cata-lysts show a high selectivity to the two main products of

99 % (Fig.7) The selectivity to styrene oxide decreases linearly from the sample MAM-15 to MAM-30 At a similar conversion of styrene, the selectivity to styrene oxide over MAM-10 is much higher than that over sample MAM-30

3.2.2 Effect of Oxidant Nature

Effect of nature of oxidants in the oxidation of styrene was studied to improve the yield of products Three oxidants including aqueous H2O2 solution (30 %), tert-butyl hydroperoxide (70 % in water), and air have been used for the oxidation of styrene over selected samples The results collected in Table2give important information Under the same reaction conditions, H2O2 was the most active oxi-dant for the oxidation of styrene over molybdate-contain-ing catalysts The conversion of styrene reaches to 90–99 % but a broad spectrum of products including benzoic acid, 1-phenylethane-1,2-diol, and some unidenti-fied polymer compounds is detected, clearly indicating that the oxidation of styrene with hydro peroxide is a less selected reaction (Table2) [26]

Once air was used, styrene conversion remains about 10–15 %, but varies observably with the amount of molybdate contents in the catalysts In this case, two desired products were obtained with the total selectivity of

99 % This reaffirms an important role of tetrahedrally-coordinated molybdate anions in activating oxygen mole-cules to oxidize styrene into aldehyde or epoxide As air is substituted by tert-butyl hydrogen peroxide, the styrene conversion decreases to 6–7 % and the oxidation reaction leads to the formation of styrene oxide only, in good consistent with the results of oxidation of olefins over Mo(VI)-containing catalysts reported in the literature [2,

23, 27, 35] It is noteworthy that TBHP is an expensive organic reagent and always accompanies by a large amount

of organic waste in the product Thus, Table2 indicates that Mg/Al-molybdate intercalated LDHs are promising catalysts for the oxidation of styrene with air

3.2.3 Effect of Reaction Time

In order to enhance the conversion of styrene at friendly mild reaction conditions, an increased reaction time is a wise choice A series of oxidation reactions has been car-ried out for different reaction times at 90C The catalytic activity of MAM-15 is displayed in Fig.8and that of some

0

10

20

30

40

50

60

70

80

90

Catalyst Batch

Styrene conversion Benzaldehyde Sel Styrene oxide Sel.

Fig 7 Catalytic activity of molybdate-containing hydrotalcite-like

catalysts in the oxidation of styrene with air at 90 C, 4 h, DMF

solvent

other catalysts are presented in Fig 3S (Supplementary

Materials)

Figure8 shows that the conversion of styrene remains

about 12 % at 90C for 4 h, but monotonically increases

to 74 % for 20 h It is noteworthy that the styrene

con-version increases with the progress of the reaction time

while the product distribution remains almost constant

within 8 h This indicates that both benzaldehyde and

styrene oxide may be simultaneously produced under

reported reaction conditions [6, 7, 9, 25] Furthermore,

benzaldehyde becomes a major product as the reaction is

kept for longer reaction time (Fig.8) The change in

pro-duct selectivity may be associated with the instability of

oxygenate intermediate products and secondary reactions

occurring in the product mixture as the reaction was lasted

for a long period of time [9,11,16,17,20,38]

3.2.4 Effect of Reaction Temperatures

Effects of reaction temperature on catalytic activity have been investigated using the sample MAM-10 in the range of 80–120C Some additional data were provided in Fig 4S (Supplementary Materials) Figure9 shows that styrene conversion slightly increases with increasing reaction tem-perature as the reaction was kept for 4 h, but in overall fraction

of styrene converted is still modest under these conditions Figure9shows another trend in styrene conversion varied with reaction temperature as the reaction mixture is kept for

24 h Clearly, the styrene conversion approaches almost

99 % at 110C while the total selectivity to benzaldehyde and styrene oxide is very high ([90 %) Figure9b presents the effect of temperature on products selectivity The selectivity to products changes with the reaction temperature

of 80–100C This is explained by the fact that cleavage of C=C considerably happens at lower temperatures and the epoxidation competes more favorably against C=C cleavage

at higher temperatures [32,50] At a reaction temperature of

110 C, more benzaldehyde was found in the product mix-ture may result from the decomposition of styrene oxide though hydrolysis reaction [2,9,35,52]

3.2.5 Reusability of Catalysts

The catalyst was recovered from the reaction mixture by filtration, washed with ethanol, dried at room temperature prior to reuse for the oxidation reaction under the same reaction conditions Figure 10 shows the results of the recycling experiments of sample MAM-15 The styrene conversion remains about 29–33 % after two recycles and then decreases to 5 % in the fourth cycle

Table 2 Effect of oxidant nature on catalytic activity of the catalysts in the oxidation of styrene at 90 C, 4 h, DMF solvent

Oxidant Batch # Nominal formula Styrene conversion (%) Product selectivity (%)

H2O2 MAM-10 [Mg0.8Al0.2(OH)2](MoO4)0.1mH 2 O 94.0 14.2 2.3 83.5

MAM-15 [Mg0.7Al0.3(OH)2](MoO4)0.15mH 2 O 99.0 29.8 – 70.2 MAM-20 [Mg0.6Al0.4(OH)2](MoO4)0.20mH 2 O 95.4 21.2 – 78.8

Air MAM-10 [Mg0.8Al0.2(OH)2](MoO4)0.1mH 2 O 10.7 58.0 42.0 –

MAM-15 [Mg0.7Al0.3(OH)2](MoO4)0.15mH 2 O 12.7 48.6 51.4 – MAM-20 [Mg0.6Al0.4(OH)2](MoO4)0.20mH 2 O 15.0 61.5 36.5 –

TBHP MAM-10 [Mg0.8Al0.2(OH)2](MoO4)0.1mH 2 O 7.3 – 99.0 1.0

MAM-15 [Mg0.7Al0.3(OH)2](MoO4)0.15mH 2 O 6.5 – 99.0 1.0 MAM-20 * [Mg0.6Al0.4(OH)2](MoO4)0.20mH 2 O 6.0 1.0 98.0 1.0

Others: benzoic acid, 1-phenylethane-1,2-diol, phenyl acetaldehyde, unidentified compounds; (-): Traces; BA benzaldehyde, SO styrene oxide

0

20

40

60

80

100

Reaction time (h)

Benzaldehyde Sel Styrene Oxide Sel.

Other produtcs Styrene conversion

Fig 8 Effect of reaction time on the catalytic activity over MAM-15

at 90 C, DMF solvent, 0.2 g of catalyst, DMF solvent (others:

benzoic acid, styrene glycol, phenyl acetaldehyde)

It is noted that the styrene conversion changes

insignificantly while the selectivity to styrene oxide

decreases sharply after two cycles This is postulated by the

phenomenon of migration of MoO42- anions from the

interlayer spacing to the external surface Thus, overall

amount of tetrahedrally-coordinated Mo-species is almost

preserved, but the content of MoO4- in the interlamellar

spaces of hydrotalcites decreases remarkably An increased

number of cycles result in the leaching MoO4-out of the

samples, evidenced by an observably decreased Mo

com-ponent in the reused samples (Table1) Thus, both

con-version and styrene oxide selectivity decreases with the

number of cycles

4 Discussion

Molybdenum complexes are well-known catalysts for the

oxidation of styrene with alkyl peroxides In all cases, the

active species are usually associated with Mo(VI)

compounds although the reaction mechanism of Mo(VI) catalyzed oxidation of unsaturated compounds still remains

a subject of debate [22,24,25,35,53] It is known that the molybdate anion MoO42- maintains a tetrahedral config-uration in neutral and alkali solutions This anion is easily changed into polymolybdate salts as separated from solu-tion at neutral condisolu-tions [23,34] Thus, incorporation of MoO42- in the interlayer region of Mg1-xAlx(OH)2 (MoO4)x/2mH2O hydrotalcite-like compounds is to stabi-lize its tetrahedral configuration

Analysis of XRD data indicates that no additional diffraction peaks appeared as x = 0–0.2, suggesting the formation of single LHD phase with the intercalation of tetrahedrally-coordinated Mo species in the interlayer domains As a consequence, a slight shift in ‘c’ or inter-layer thickness was observed An intercalation of MoO4 2-tetrahedral ions into interlayer free gallery dramatically modifies the reduction–oxidation characteristics of layered double hydroxides [22, 23,32–35] However, the molyb-date-rich samples contain both Mo in tetrahedral and octahedral coordination The amount of octahedrally coordinated Mo-species is prevailing in the molybdate-richer samples because the interlayer spacings can only allow a certain amount of MoO4-accessible [34]

As molybdate-intercalated LDHs were used as catalysts for the oxidation of styrene with oxygen, the styrene version varies with the molybdate amount, reaction con-ditions The analysis of the product selectivity allows to shed light on the role of the molybdate locations in the production of the main products From the results in Figs 1, 7, 8 and 10, it was suggested that the product selectivity is strongly dependant on the location of tetra-hedral MoO4-anions while styrene conversion depends on the total amount of molybdate anions although the role of basic sites from hydrotalcites are not ruled out Indeed, the role of Mg/Al brucite sheets could be stabilization of

0

10

20

30

40

50

60

70

80

90

100

Reaction temperature (oC)

24 hours

4 hours

80 90 100 110

4 hours

4 hours

24 hours

24 hours

0 10 20 30 40 50 60 70 80 90 100

Temperature (

o C)

Fig 9 Effects of reaction temperature on styrene conversion (a) and product selectivity (b) over MAM-10 after 4 (a) and 24 h (b), DMF solvent, 0.2 g of catalyst

0

10

20

30

40

50

60

70

80

90

100

Number of Cycle

Other Sel. Conversion

Fig 10 Oxidation of styrene with air over reused MAM-15 sample at

90 C, 8 h, DMF solvent, 0.2 grams of catalyst

radical oxygen species generated through interaction with

basic sites [38,51] As discussed above in Sect 3.2.1, the

changes in molar ratios (Mg/Al/Mo) would vary the

basicity/acidity of the catalysts An increased Mo-content

in LDHs (from sample MAC-00 to MAM-20) leads to a

remarkable augmentation of styrene conversion, but a

decrease in the styrene oxide selectivity (Fig.7) due to an

increased the acidity of higher Mo-content samples A low

styrene conversion on sample MAM-30 and MixO

indi-cates a synergistic interaction between MoO4-anions and

the Mg–Al hydroxide layers in the oxidation of styrene

Benzaldehyde and styrene oxide are postulated to produce

parallel to each other (Figs.8 and 3S) The tetrahedral

MoO4-intercalated anions in the interlayer free gallery are

responsible for the epoxidation of styrene while

ben-zaldehyde can produce over both tetrahedral MoO4

-intercalated anions and molybdate species on the external

surface (Fig.10) [53] In addition, all MoO4-ions would

only produce styrene oxide as TBHP was used as oxidant

This may be due to the presence of TBHP to steer the

oxidation reaction mechanism into another pathway

Styr-ene oxide is suggested to be produced over molybdate ions

though the formation of a Mo(VI) butyl peroxide and

transfer of the distal oxygen atom of butyl peroxide [53]

5 Conclusions

Mg–Al–molybdate hydrotalcite-like samples with different

ratio of Mg/Al/Mo were synthesized at pH constant value

Molybdate anions were introduced in the interlayer regions

of Mg–Al hydrotalcite as interlamellar anions and present

in a number of oxoanions The catalysts possess the

hydrotalcite structure with layered structure, but have low

external surface area The crystallinity of the sample

decreased with increasing molybdate content The Mg–Al–

molybdates catalysts were found to have a good activity in

the oxidation of styrene with air The styrene conversion

depends on the nature of oxidant, total amount of

molyb-date anions; reaction variables while the product selectivity

was found to be related to the position of molybdates

between interlayer spacing or external surface of the solids

The preliminary investigation results indicated that air was

found to be a promising oxidant for the selective

conver-sion of styrene into styrene oxide and benzaldehyde Mg/

Al/MoO4-like hydrotalcite catalyst showed 74–90 %

styr-ene conversion with 70 % benzaldehyde and 26 % styrstyr-ene

epoxide selectivity at 90C in the presence of DMF

sol-vent at 90C after 20–24 h

Acknowledgments This research is funded by Vietnam National

Foundation for Science and Technology Development (NAFOSTED)

under Grant Number 104.05-2014.01.

References

1 Centi G, Cavani F, Trifiro F (2001) Selective oxidation by heterogeneous catalysis Kluwer Academic Publishers, New York

2 Dusi M, Mallat T, Baiker A (2000) Catal Rev-Sci Eng 42(1&2):213–287

3 Monnier JR (2001) Appl Catal A 221:73–91

4 Wang X, Zhang X, Wang Y, Liu H, Qiu J, Wang J, Han W, Yeung KL (2011) ACS Catal 1:437–445

5 Qiu F, Wang X, Zhang X, Liu H, Liu S (2009) King Lun Yeung Chem Eng J 147:316–322

6 Laha SC, Kumar R (2001) J Catal 204:64–70

7 Valand J, Parekh H (2013) H B Friedrich 40:153–157

8 Kishore D, Rodrigues AE (2008) Appl Catal A 345:104–111

9 Tien-Thao N, Trung HH (2014) Catal Commun 45:153–157

10 Wei X-L, Lu X-H, Ma X-T, Peng C, Jiang H-Z, Zhou D, Xia Q-H (2015) Catal Commun 61:48–52

11 Nie L, Xin KK, Li WS, Zhou XP (2007) Catal Comm 8:488–492

12 Bhattacharyya K, Varma S, Tripathi AK, Vinu A, Tyagi AK (2011) Chem A Eur J 17:12310–12325

13 Lotero E, David V, Nguyen C, Wagner J, Larsen G (1998) Chem Mater 10:3756–3764

14 Murata C, Yoshida H, Kumagai J, Hattori T (2003) J Phys Chem

B 107:4364–4373

15 Yang Q, Li C I, Yuan S, Li J, Ying P, Xin Q, Shi W (1999) J Catal 183:128–130

16 Tang Q, Zhang Q, Hongli W, Wang Y (2005) J Catal 230:384–397

17 Wang Y, Zhang Q, Shishido T, Takehira K (2002) J Catal 209:186–196

18 Gomez S, Garces LJ, Villegas J, Ghosh R, Giraldo O, Siub SL (2005) J Catal 233:60–67

19 Jana SK, Kubota Y, Tatsumi T (2007) J Catal 247:214–222

20 Adam F, Iqbal A (2010) Chem Eng J 160:742–750

21 Gardner E, Pinnavaia TJ (1998) Appl Catal A 167:65–74

22 Maurya MR, Arya A, Adao P, Pessoa JC (2008) Appl Catal A 351:239–252

23 Ciocan CE, Dumitriu E, Cacciaguerra T, Fajula F, Hule V (2012) Catal Today 198:239–245

24 Behera GC, Parida KM (2013) Appl Catal A 464–465:364– 373

25 Valente AA, Petrovski Z, Branco LC, Afonso CAM, Pillinger M, Lopes AD, Romao CC, Nunes CD, Goncalves IS (2004) J Mol Catal A 218:5–11

26 Zeng R-C, Liu Z-G, Zhang F, Li S-Q, Cui H-Z, Han E-H (2014) J Mater Chem A 2:13049–13057

27 Kuhn FE, Groarke M, Bencze E, Herdtweck E, Prazeres A, Santos AM, Calhorda MJ, Romao CC, Goncalves IS, Lopes AD, Pillinger M (2002) Chem A Eur J 8:2370–2383

28 Koch AKA, Borthakur R, Chakraborty M, De AK, Phukan A, Bez G, Lal RA (2014) J Mol Str 1063:92–101

29 Lei X, Chelamalla N (2013) Polyhedron 49:244–251

30 Mitchell S, Gomez-Aviles A, Gardner C, Jones W (2010) J Solid State Chem 183:198–207

31 Vasant R Choudhary, Jayant R Indurkar, Vijay S Narkhede, Rani Jha (2004) J Catal 227:257–261

32 Frost RL, Musumeci AW, Bostrom T, Adebajo MO, Weier ML (2005) Thermochim Acta 429:179–187

33 Zavoianu R, Bırjega R, Pavel OD, Cruceanu A, Alifanti M (2005) Appl Catal A 286:211–220

34 Rives V (2001) Layered double hydroxides: present and future Nova Science, New York

35 Barrio L (2008) Jose M Campos-Martin, M Pilar de Fru-tos and Jose L G Fierro Ind Eng Chem Res 47:8016–8024

36 Mitchell PCH, Wass SA (2002) Appl Catal A 225:153–165