Epoxidation of cyclohexene with H2O2 and acetonitrile catalyzed by Mg-Al hydrotalcite and cobalt modified hydrotalcites

Article abstract of DOI:10.1007/s10562-009-0238-y

The paper presents a comparison between the results obtained for cyclohexene epoxidation in the presence of base catalysts: Mg _0/75Al_0.25-hydrotalcite (HT), the corresponding mixed oxide (CHT), the reconstructed hydrotalcite (RHT), a hydrotalcite containing cobalt in the brucite-type layer ((Co-Mg_0.75)Al_0.25)) and one containing cobalt complex-species supported on HT carrier (Co-complex/HT). The selective conversion of cyclohexene to the corresponding epoxide is closely related to the base strength of the catalysts.

Full text of DOI:10.1007/s10562-009-0238-y

Catal Lett (2010) 134:309–317
DOI 10.1007/s10562-009-0238-y
Epoxidation of Cyclohexene With H2O2 and Acetonitrile
Catalyzed by Mg–Al Hydrotalcite and Cobalt Modified
Hydrotalcites
•
Rodica Ionescu Octavian Dumitru Pavel
•
•
Ruxandra Bırjega Rodica Zavoianu
•
ˆ
Emilian Angelescu
˘
Received: 12 July 2009 / Accepted: 28 November 2009 / Published online: 16 December 2009
Ó Springer Science+Business Media, LLC 2009
Abstract The paper presents a comparison between the
results obtained for cyclohexene epoxidation in the pres-
ence of base catalysts: Mg0/75Al0.25-hydrotalcite (HT), the
corresponding mixed oxide (CHT), the reconstructed
hydrotalcite (RHT), a hydrotalcite containing cobalt in
the brucite-type layer ((Co-Mg0.75)Al0.25)) and one con-
taining cobalt complex-species supported on HT carrier
(Co-complex/HT). The selective conversion of cyclohex-
ene to the corresponding epoxide is closely related to the
base strength of the catalysts.
since by ring opening reactions, epoxides are directly
transformed into a wide array of compounds. Among var-
ious oxidation agents the use of hydrogen peroxide
deserves a special attention as an easy to handle and clean
oxidation agent which can be reduced only to water by the
end of the process. According to Payne’s works [1, 2] by
combining an oxidant such as H2O2 and a nitrile in the
presence of a base catalyst, the nitrile is converted to amide
via a peroxycarboximidic acid intermediate which serves
as a terminal oxygen donor.
Recent studies performed by Kirm et al. [3] showed that
solid bases such as mixed oxides obtained from Mg/Al
hydrotalcite-like precursors can act as heterogeneous cat-
alysts for the epoxidation of styrene, using a combined
oxidant of hydrogen peroxide and acetonitrile in the pres-
ence of acetone and water as solvents. The authors con-
cluded that several factors such as the Mg/Al ratio, the
presence of the pure hydrotalcite phase in the mixed oxide
samples, the reconstruction rate of the hydrotalcite-like
phase during the reaction, and the addition of water, play
an important role in the selective epoxidation of styrene.
Therefore, one of the goals of this work has been to
provide new insights on the role played by the hydrotalcite-
like phase and the impact of the reconstruction through the
,,memory effect’’ presented by hydrotalcite-like materials
on the selectivity of the transformation of cyclohexene to
cyclohexane epoxide. To do so, the catalytic activities of a
parent hydrotalcite (HT), have been compared with those
of the corresponding mixed oxide (CHT) and the recon-
structed hydrotalcite obtained through rehydration of the
mixed oxide (RHT) under identical reaction conditions. In
order to select the operational parameters the effects
induced by the solvents and the presence of the acetonitrile
on the catalytic performances of the parent hydrotalcite
have been investigated.
Keywords Cyclohexene epoxidation ꢀ Mg–Al–HTlc ꢀ
MgAl mixed oxides ꢀ Reconstructed hydrotalcite ꢀ
Co-HTlc
1 Introduction
Selective epoxidation of olefinic compounds is one of the
important steps in organic synthesis of fine chemicals,
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-009-0238-y) contains supplementary
material, which is available to authorized users.
˘
R. Ionescu ꢀ O. D. Pavel ꢀ R. Zavoianu (&) ꢀ E. Angelescu
Faculty of Chemistry, Department of Chemical Technology and
Catalysis, University of Bucharest, Bd. Regina Elisabeta,
No. 4-12, S3, 030018 Bucharest, Romania
e-mail: rodicazavoianu@gmail.com
ˆ
R. Bırjega
National Institute for Lasers, Plasma and Radiation Physics,
˘
P. O. BOX Mg-27, Magurele, Bucharest, Romania
123

310
R. Ionescu et al.
Another aspect considered in the elaboration of this
The second cobalt-modified hydrotalcite, containing the
same amount of cobalt (3.5 9 10-4 mol Coꢀg-1) has been
prepared using as cobalt source an admixture of bis-tri-
phenylphosphine-dichlorocobalt (II) Co[P(C6H5)3]2Cl2 (A)
and benzyltriphenyl phosphonium—tetrachlorocobaltate
[C6H5CH2P(C6H5)3]?2 [CoCl4]2- (B) which has been
obtained according to the reaction:
work is that a significant number of researches concern
the partial oxidation of the organic substrates with H2O2,
[4–9], alkylhydroperoxides or peroxyacids [10–12] using
transition metal complexes as catalysts. The use of these
catalysts in homogeneous catalytic conditions is inconve-
nient because the catalyst cannot be recovered and reused.
2ꢂ
Â
Ã
Â
Ãþ
CoCl2 þ 2PðC6H5Þ3þC6H5CH2Cl ! 0:5Co PðC6H5Þ3 Cl2 þ 0:5 C6H5CH2PðC6H5Þ3 ½CoCl4ꢁ
ð1Þ
2
2
ðAÞ
ðBÞ
The current trend is to immobilise the complex on a host
support such as zeolites [13–16], macroporous carbon [17],
polymers [12, 18, 19], or hydrotalcites [20–22].
The reaction (1) was performed under nitrogen at 60 °C
during 6 h using a mixture of 0.02moles triphenylphosphine
(P(C6H5)3) with 0.01moles benzylchloride C6H5CH2Cl and
0.01 moles CoCl2ꢀ6H2O solved in n-butanol. Then the
solvent was evacuated under vacuum and the solid bright
blue crystals were separated under nitrogen and recrystal-
lized from n-butanol. The product is a mixture of A and B
cobalt complexes since in the case of chlorine ionic com-
plexes the compound benzyltriphenyl phosphonium-tri-
phenyl phosphine-trichlorocobaltate [C6H5CH2P C6H5)3]?
[(C6H5)3PCoCl3]- could not be obtained by this procedure
[25].
Based on this state of the art, the second goal of this work
was to highlight an eventual synergetic effect of the transition
element and the base function of the HT solid for the selective
epoxidation of cyclohexene. Therefore, the performances of
cobalt modified hydrotalcites with cobalt introduced in the
brucite-type layer or simply supported as complex species on
a hydrotalcite carrier have also been investigated.
2 Experimental
An acetone solution of the cobalt precursor (Co con-
centration 6.5 9 10-3mol.L-1) was contacted with the HT
support previously dried under vacuum during 24 h at
25 °C. The resulting solid was filtered, washed three times
with acetone, then with benzene and finally dried under
vacuum (10-2 torr) at 25 °C. The obtained catalyst was
stored under nitrogen.
2.1 Catalysts Preparation
The Mg0.75Al0.25 hydrotalcite (HT) was synthesized at pH
10 under low supersaturation conditions, using Fluka ana-
lytical grade Mg(NO3)2ꢀ6H2O, Al(NO3)3ꢀ9H2O, Na2CO3
and NaOH as raw materials, according to the procedure
described in detail earlier [23, 24].
2.2 Catalysts Characterization
The CHT mixed oxide derived from HT was obtained
through the calcination of the dried HT at 460 °C during
18 h under nitrogen flow.
The metal content in all catalysts has been determined by
AAS spectroscopy.
The reconstructed hydrotalcite structure (RHT) was
prepared by rehydration of CHT which was immersed in
decarbonated water for 24 h at 25 °C under nitrogen. Then
the sample was dried at 25 °C under vacuum (10-2 torr). All
samples were kept under nitrogen blanket before being used.
One cobalt modified hydrotalcite (Co-Mg)0.75Al0.25 was
obtained by coprecipitation using the method previously
describedfor HTsynthesisinwhicha partofMg(NO3)2ꢀ6H2O
was substituted by Co(NO3)2ꢀ6H2O in order to correspond at
(Mg2? ? Co2?)/Al3? = 3 molar ratio and a concentration of
3.5 9 10-4 mol Coꢀg-1 in the final catalyst.
The XRD patterns were collected on a PANalytical
X’Pert MPD theta-theta system in continuous scan mode
(counting 2 s per 0.02 2h) ranging from 5o to 72o 2h, using
a Ni filter (k = 0.15418 nm), and a curved graphite
monochromator. Data were analyzed using PANalytical
X’Pert HighScore Plus software package.
The as-synthesized complex and the catalysts were
characterized by DR–UV–Vis in the range of 210–915 nm.
The spectra have been recorded with a SPECORD 80 UV–
VIS spectrometer with an integration sphere coated with
MgO taken as reference.
123

Epoxidation of Cyclohexene With H2O2 and Acetonitrile
311
A Micromeritics ASAP 2020 apparatus was used in
order to evaluate the specific surface area of the samples by
nitrogen adsorption after degassing the samples in situ at
70 °C for 24 h.
and a capillary column of 30 m length with DB5 stationary
phase. The oxidation products were identified by compari-
son with a standard sample (retention time in GC). The
reaction products were identified also by mass spectrometer-
coupled chromatography, using a GC/MS/MS VARIAN
SATURN 2100 T equipped with a CP-SIL 8 CB Low Bleed/
MS column of 30 m length and 0.25 mm diameter.
The surface base sites of the catalysts were determined
using a method similar to the one developed by Parida and
Das [26] on the basis of the irreversible adsorption of
organic acids with different pKa, e.g. acrylic acid,
pKa = 4.2 and phenol pKa = 9.9. The amount of adsorbed
phenol is related to the number of strong surface base sites
while the amount of adsorbed acrylic acid is related to the
total number of base sites. The number of weak base sites is
given by the difference between the amount of adsorbed
acrylic acid and the adsorbed phenol. Before base sites
measurement, all catalysts were thoroughly degassed under
vacuum at room temperature. Catalyst samples (0.05 g)
were introduced in brown sealable bottles and 10 mL of
freshly prepared solution containing the organic acid (e.g.
phenol and acrylic acid, respectively) in methylcyclohexane
was added. Afterwards, the bottles were sealed and were
shaken for 2 h at 30 °C. It was assumed that the interaction
of the solids with atmospheric CO2 and water was negli-
gible since their exposure to atmosphere was limited to the
weighing period. The time required for the system to reach
the equilibrium at constant temperature was checked for
each solid and it was never longer than 1 h.
3 Results and Discussion
3.1 Characterization
3.1.1 XRD and Textural Analyses
The XRD patterns of the parent hydrotalcite HT and
reconstructed hydrotalcite RHT (Fig. 1a) exhibit the typi-
cal pattern of a layered double hydroxide material (JCPDS
file 70-2151), indexed in a hexagonal lattice with a R3m
rhombohedral symmetry, while the pattern of the parent
CHT support (Fig. 1b) is that of a mixed oxide Mg(Al)O
with MgO-periclase structure (JCPDS file 45-0946,
indexed in a cubic symmetry. The XRD patterns of the
cobalt containing samples are similar to those of the Co-
free samples with no extra peaks originating from crys-
talline cobalt species, indicating a high dispersion of cobalt
on the supports. The data in Table 1 show that no sub-
stantial modification of the a and c parameters were
observed for Co-complex/HT which suggests that, more
likely, the cobalt complexes covered preferentially the
outer surface and in a very low amount the interlayer space,
adopting flat-lying position.
After 2 h the concentration of the acid remaining in
solution was determined spectrophotometrically at kmax
(e.g. 276.5 nm for phenol, and 231.5 nm for acrylic acid)
using the calibration curves absorption versus acid con-
centration in the range where the Lambert–Beer law is
fitted. The amount of adsorbed phenol and adsorbed acrylic
acid, respectively, is given by the difference between the
amount of the respective acid introduced in the test and the
amount that remained in solution after 2 h. The reproduc-
ibility of the spectrophotometric method was checked
experimentally, and the measurements were repeated 2–
3 times. The relative error was approximately 5%.
In the case of (Co-Mg)0.75Al0.25 the isomorphous sub-
3?
stitution of Mg2?/Al
by Co2?/Co3? was not to evi-
denced by any variation of the a parameter value due to the
low amount of Co as well as to the very good match of the
ionic radii of Co2? and/or Co3? with Mg2? and/or Al3?
.
The results of the textural analysis presented in Table 1
showed that the hydrotalcite-like materials have similar
specific surface areas in the range 125–138 m2/g. The
modification with cobalt leads to a small decrease of the
pore size compared to the parent HT. Both the surface area
and the pore size of the reconstructed hydrotalcite were
sensibly lower than those of the parent HT.
2.3 Catalytic Tests
The activity and selectivity of the catalysts were tested in
the oxidation of cyclohexene with hydrogen peroxide using
acetonitrile as reductant. The reactions were performed in a
stirred flask (500 mL) at 60 °C, for 5 h reaction time. In a
typical experiment cyclohexene (4 mmol) and acetonitrile
(32 mmol) were dissolved in 20 mL of solvent. (two types
of solvents have been used: (a) methanol and (b) equivol-
umic admixture of water and acetone). All reagents were
freshly distilled. In all reactions the catalyst concentration
was 0.3% in the reaction admixture (w/w).
3.1.2 Characterization by DR–UV–Vis
The DR–UVVis spectra of cobalt containing catalysts as
well as those of the neat complexes are presented in Fig. 2.
The spectrum of (Co-Mg)0.75Al0.25 catalyst presents only
the features characteristic for Co cation species in octahe-
dral symmetry, consistent with the structure of the brucite-
type layer [27, 28]. The presence of the bands at 419 and
The reaction was monitored hourly using a GC K072320
Thermo Quest Chromatograph equipped with a FID detector
123

312
R. Ionescu et al.
Fig. 1 XRD patterns of the
hydrotalcite-like catalysts (a),
the mixed oxide CHT and the
cobalt complex (b)
(A)
(B)
Co-complex/HT
CHT
(Co-Mg)0.75Al0.25
RHT
HT
Co-complex free
0
20
40
60
80
10
20
30
40
50
60
70
80
2
θ
2θ
Table 1 Structural and textural properties of the catalysts
Catalysts
XRD analysis
Texture analysis
Unit cell parameters
Crystallite
sizes (nm)
BET surface
area (m2/g)
Pore
size (nm)
Vol. ads. (cm3/g)
a (nm)
c (nm)
HT
0.3056
0.3055
0.3056
0.3055
0.4181
2.3185
2.3232
2.3154
2.3181
9.1
9.3
131
138
125
3.2
18.9
15.5
16.7
6.7
0.6220
0.5351
0.5562
0.0054
0.8990
Co-complex/HT
(Co-Mg)0.75Al0.25
RHT
8.7
16.3
3.2
CHTa
250
14.4
a
Cubic structure
517 nm indicates that some of Co (II) ions were oxidized to
Co(III). This partial oxidation of Co(OH)2 to Co(OH)3 (or
CoOOH) in basic media was also reported by Leroux and
Zheng.[28, 29] who showed that the oxidation is thermo-
dynamically very favorable at low oxygen pressure which is
reached during the aging period of the precipitate.
The spectrum of the solid admixture of Co-complexes
shows the characteristics specific for distorted tetrahedral
symmetry of Co (II) species with very intense bands in the
region of d-d transitions [25, 27, 30, 31]. Meanwhile for the
complex solved in acetone (that was used as catalyst in
homogeneous conditions), the intense bands from the
region of charge transfer up to 300 nm are absent and only
the bands corresponding to d-d transitions of the complex
can be noticed.
The DR–UV–Vis spectrum of Co-complex/HT, reveals
that upon impregnation the structures of the cobalt pre-
cursor complexes are altered, and the cobalt species are
found mainly in octahedral symmetry. The presence of the
traces of tetrahedral coordinated Co(II) species is indicated
by the shoulder of the band at 705 nm (m2), while the band
at 664 nm indicates the presence of Co(II) in octahedral
symmetry. Some of the bands corresponding to ligand to
metal charge transition (LMCT) are overlapped by the
bands characteristic for the tetrahydrocarbylphosphonium
cation around 300 nm [32]. The bands at 521 and 610 nm
indicate the presence of some traces of Co (III) species
with octahedral coordination. The oxidation of Co(II) to
Co(III) may occur due to the base character of the support
during the impregnation.
123

Epoxidation of Cyclohexene With H2O2 and Acetonitrile
313
Co-complex/HT
have been used for this purpose. This type of method
presents the advantage that it may be used to determine the
base sites of solids that cannot undergo calcinations with-
out altering their structure.
Co-complex in acetone
(Co-Mg)0.75Al0.25
1.1
1
(Co-Mg)0.75Al0.25
Co-complex free
2 per. Mov. Avg. (complex AB)
Even if the results obtained using this method should not
be considered as absolute values since the number of base
sites titrated depends also on the accessibility of the
organic acid to the base sites (which is related to the
dimensions of the pores in the solid) the method allows a
relative classification of the catalysts as a function of the
number of accessible base sites.
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
It has been assumed that the results obtained using the
titration method may be correlated to the results of the
catalytic activity tests since the molecular dimensions of
phenol are comparable to those of the largest molecular
dimensions of the participating molecules (e.g. cyclohex-
ene and cyclohexane epoxide) in the investigated catalytic
oxidation process.
The results of these investigations as presented in Table 2
show that the catalyst CHT has the highest number of total
base sites. This fact is rather surprising since by consider-
ation of their theoretical formulae one would expect that the
calcined solid has a lower number of base sites compared to
HT and RHT. When comparing the results obtained by
titration with the theoretical values of the base sites numbers
(e.g. moles anions per gram of catalyst) calculated using
their chemical formulae it seems that the titration method
allows the determination of ca 15–22% of the total of base
sites, probably due to the fact that only the base sites on the
external surface of the catalyst are titrated. Therefore it
seems that the larger number of total base sites for CHT is a
consequence of its larger surface area.
220
320
420
520
620
720
820
Wavelength (nm)
Fig. 2 DR–UV–Vis spectra of Co-containing catalysts
3.1.3 Determination of the Surface Base Sites
Taking into account that the determination of the base sites
using TPD of CO2 can be applied only for the calcined
hydrotalcite sample (e.g. CHT) and cannot be applied to
the dried hydrotalcite-type samples, it has been preferred to
apply the method based on the titration of the surface base
sites at 30 °C in order to compare the basicity of all the
catalysts investigated in this study. Acids with different
pKa, such as acrylic acid, pKa = 4.2 and phenol pKa = 9.9
The values of the total number of base sites for RHT and
HT vary in the order indicated by the theoretical calcula-
tions, while respecting the above statement that only the
accessible base sites are titrated. The hydroalcite containg
Co in the brucite type layer, (Co-Mg)0.75Al0.25, has a
slightly lower basicity expressed as mmoles/g compared to
HT. This fact should be expected since the molar weight of
(Co-Mg)0.75Al0.25 is slightly higher than that of
Table 2 Distribution of base sites for the investigated catalysts
Catalysts
Total number of base sites
(mmol acrylic acid/g catalyst)
Distribution of base sites
Strong base sites
(mmol phenol/g catalyst)
Weak and medium strength base sites
(mmol/g catalyst)a
HT
6.73
4.46
6.58
7.21
8.36
0.26
0.23
0.24
0.34
0.38
6.47
4.23
6.34
6.87
7.98
Co-complex/HT
(Co-Mg)0.75Al0.25
RHT
CHT
a
Weak and medium strength base sites = total number of base sites - strong base sites)
123

314
R. Ionescu et al.
Mg0.75Al0.25, and consequently the number of mmoles per
gram of catalyst will be lower.
high toxicity of methanol it would be preferable to use the
admixture water–acetone as solvent since it implies a lower
amount of organic solvent and it is more eco-friendly.
In the reactions performed in the presence of HT cata-
lyst while using CH3CN reductant the best results are
obtained after 5 h reaction time in acetone–water solvent at
a molar ratio cyclohexene:H2O2 = 1:8.
The lowest number of base sites is obtained for Co-
complex/HT and it may be a consequence of a screening of
the base sites by the Co-complex.
3.2 Results of Catalytic Tests
The catalysts have been tested for the oxidation of cyclo-
hexene with hydrogen peroxide in the presence of aceto-
nitrile as reductant agent using either an admixture water–
acetone (1:1 volume ratio) or methanol as solvent. The
reaction path adapted from the work of Kirm et al. [3] is
presented in Scheme 1.
(A)100
90
80
70
60
50
40
30
20
10
0
The results presented in Fig. 3 show that the presence of
both the acetonitrile and the solid base catalyst are abso-
lutely necessary for the reaction.
When both the acetonitrile and the catalyst are absent
from the reaction mixture or when the acetonitrile was
added to the reaction mixture in the absence of the catalyst,
the conversions of cyclohexene are limited to 20% and the
values of the yield to cyclohexane epoxide are low (less
than 5%) (Fig. 3a, b).
The reaction performed with cyclohexene and H2O2 in
the presence of the hydrotalcite-like catalyst without add-
ing acetonitrile, leads to high conversions of cyclohexene
(e.g. 88%). However the process has a poor selectivity and
allows a wide range of byproducts to be obtained besides
the cycohexane epoxide for which the yield is only 10%.
In the reactions performed under identical conditions in
the presence of the catalyst HT, it has been noticed that the
values of the conversion and selectivity to cyclohexane
epoxide depend on the molar ratio cyclohexene:H2O2 as
well as on the nature of the solvent employed.
0
1
2
3
4
5
6
Reaction time (h)
100
90
80
70
60
50
40
30
20
10
0
(B)
The results show that an excess of hydrogen peroxide is
favorable to obtain cyclohexene conversions over 90%
with a yield to epoxide over 80% after 5 h reaction time.
Even if at short reaction time in water–acetone solvent
the conversion and the yield to epoxide are lower than
those obtained in methanol solvent, after 5 h reaction time
they reach almost the same value. Taking into account the
0
1
2
3
4
5
6
OH
|
OH-
NH
||
C
Reaction time (h)
__Al__
H2O2
CH3CN
O
Fig. 3 Temporal variation of the conversion of cyclohexene (a) and
yield to epoxide (b) during the oxidation of cyclohexene with H2O2 at
60 °C, solvent water–acetone 1:1 vol./vol
square molar ratio cyclohexene:H2O2 1:8; catalyst HT, reductant
CH3CN filled circle molar ratio cyclohexene:H2O2 1:4 catalyst HT,
reductant CH3CN triangle molar ratio cyclohexene:H2O2 1:8 catalyst
HT, without reductant open circle molar ratio cyclohexene:H2O2 1:8
without catalyst, without reductant, multiplication symbol solvent
water–acetone 1:1 vol./vol; molar ratio cyclohexene:H2O2 1:8 without
catalyst, with reductant CH3CN
__,
solvent methanol—
___Al+___
HOO-
H3C OOH
H2O
NH2
||
C
N-
||
C
H2O
H3C
O
H3C OOH
Scheme 1 Peroxidation of cyclohexene with hydrogen peroxide on
base catalysts in the presence of CH3CN
123

Epoxidation of Cyclohexene With H2O2 and Acetonitrile
315
As it may be seen from the results displayed in Fig. 4, in
the reactions performed in the presence of HT catalyst
while using CH3CN reductant in acetone–water solvent
(molar ratio cyclohexene:H2O2 = 1:8), the conversions
and yields to cyclohexane epoxide obtained while using
HT, the corresponding mixed oxides CHT, or the recon-
structed hydrotalcite RHT as catalysts are high and their
values are comparable.
100
(A)
The above-mentioned catalysts are solids with base
character, their base strength varying in the following
order: CHT [ RHT [ HT as it was shown in several pre-
vious publications [33, 34] and confirmed by our results of
the base sites determinations as presented in Table 2.
Corma et al. showed that pure MgO possesses strong
base sites consisting of O2- while CHT contains surface
sites with different base strength such as OH- groups
(low), M–O pair (medium) and O2- (strong) [33].
The rehydrated sample RHT contains OH- ions in the
90
80
70
60
50
40
30
20
10
0
¨
interlayer which act as Bronsted base sites with a stronger
2-
basicity than CO3 ions located inside the parent hydro-
talcite structure [35].
These three catalyst samples differ neatly not only by
their base strength but also by their specific surface areas
and porous structures, as it may be seen from the data
presented in Table 1. In some of our previous publications
[34] as well as in Takehira’s work [36] it has been showed
that HT and RHT, respectively, have different morphology
even if both samples exhibit a hydrotalcite-like structure.
The change of morphology in the RHT sample appears
both at the edge of crystals where worm-like structures are
formed as well as defects on the plate that were evidenced
by TEM [34]. Since the worm-like structures formed at the
edge of the crystals block the access to the interlayer there
is a dramatic decrease of the BET surface area of RHT
compared to the HT sample. Taking into account this
aspect it may be assumed that the base sites are preferen-
tially located on the external surface of the solid.
0
2
4
6
Reaction time
100
90
80
70
60
50
40
30
20
10
0
(B)
To what concerns the catalytic activity of CHT which one
would expect to be much higher due to its larger surface area
and number of base sites, the presence of water used as
solvent in the reaction mixture may induce the partial
reconstruction of the hydrotalcite structure when contacted
to CHT leading to a solid more similar to RHT or HT. This
fact may explain the similarity of the results obtained when
using RHT, HT and CHT catalysts even if the fresh CHT has
a much higher surface area. Therefore, the small differences
of activities and selectivities obtained in the reactions per-
formed with these three catalysts, suggest that in the case of
the investigated reaction, the most important role is played
by their acid–base character and not particularly by their
porous structure or by the morphology of the catalyst.
The experiments aimed at highlighting an eventual
synergetic effect of cobalt and the base function of the HT
solid for the investigated reaction have been performed
using the Co modified hydrotalcite-like compounds as
catalysts for the oxidation of cyclohexene with hydrogen
0
2
4
6
Reaction time
Fig. 4 Conversion of cyclohexene (a) and selectivity to cyclohexane
epoxide (b) after oxidation with H2O2 during 5 h reaction time at
60 °C molar ratio cyclohexene: H2O2 = 1:8; : solvent : water–acetone
1:1 vol./vol, filled diamond HT; Open diamond (Co-Mg)0.75Al0.25
;
filled circle Co-complex/HT; multiplication symbol Co-complex; open
circle CHT; filled triangle RHT
123

316
R. Ionescu et al.
Table 3 Distribution of the reaction by-products at cyclohexene oxidation with hydrogen peroxide using CH3CN reductant, water–acetone 1:1
vol/vol solvent after 5 h reaction time at 60 °C for the investigated catalysts
Catalyst
Distribution of by-products (%)a
Cyclohexenol
Cyclohexane-1,2-diol
Cyclohexanone
Cyclohexenone
Adipic acid
Others
HT
–
–
1.1
–
4.5
–
5.2
20.4
11.2
0.2
Co-complex homogeneous catalysis
4.7
2.0
23.5
7.9
41.8
7.8
Co-complex/HT
–
3.0
0.6
1.0
1.8
CHT
0.3
0.5
1.2
RHT
0.3
(Co-Mg)0.75Al0.25
0.9
a
The balance up to 100% is cyclohexane epoxide
peroxide under the selected above-mentioned reaction
conditions. Their catalytic performances have been com-
pared to those of the unmodified hydrotalcite as well as to
those of a cobalt complex catalyst used under homoge-
neous catalysis conditions.
hydrotalcite are affected by the interaction with the cobalt
complex species. This fact is also confirmed by the data
presented in Table 2, which show the lower number of
weak base sites in Co-complex/HT. Moreover the proper-
ties which determine the selective catalytic activity of the
cobalt species, namely their composition, structure and
symmetry are altered by the influence of the hydrogen
peroxide as it is the case under homogeneous catalysis
conditions and it can be seen in the DR–UV–Vis spectra.
The analysis of the by-products obtained during the
reactions performed with cobalt complex supported on HT
as well as the modest yield to cyclohexane epoxide com-
plete the finding that similar to the test under homogeneous
catalysis conditions using only cobalt complexes as cata-
lyst without the hydrotalcite, the reaction is non-selective
with respect to cyclohexane epoxide formation.
The test reaction under homogeneous catalysis condi-
tions has been performed using cyclohexene, H2O2, ace-
tonitrile and the cobalt complex catalyst solved in the
acetone–water solvent without adding the hydrotalcite. The
results are also displayed in Fig. 4.
Under homogeneous catalysis conditions, both the
conversion and the selectivity to cyclohexane epoxide after
5 h reaction time has low values e.g. 40 and 10.56%
respectively as it may be seen in Fig. 4. Under the influ-
ence of the powerful oxidizing agent the color of the
solution owed to the presence of the solved cobalt complex
turns from blue to green due to the oxidation of the cobalt
complexes by the hydrogen peroxide.
In the latter case the absence of the base solid in the
reaction mixture leads the chemical transformation mainly
to form 1,2-cyclohexanediol and adipic acid (the selectiv-
ities towards these products being 24 and 42%, respectively
as it may be seen in Table 3).
From Fig. 4 it may also be seen that compared to the
activity of the parent HT, after 5 h reaction time the activity
of (Co-Mg)0.75Al0.25 is practically unmodified. Thus it may
be concluded that the effect of the Co2? ions partially
substituting Mg2? in the brucite-type layer is not significant.
This fact is also confirmed by the determination of structural
and textural data (Table 1) and the number of base sites
presented in Table 2. However, it is interesting to notice that
at the beginning of the catalytic test (Co-Mg)0.75Al0.25 is
more active than HT. This proves its bifunctional character
due to the presence of both base sites related to the hydro-
talcite structure and redox sites related to the presence of Co.
The hydrotalcite-like catalyst modified by impregnation
with cobalt complex leads to a conversion that is 13% lower
and a yield to cyclohexane epoxide lower by 31% compared
to those obtained with the un-modified hydrotalcite. The
ability of Co2? to modify both its oxidation state to Co3? as
well as its coordination number from 4 to 6, increases the
possibility to catalyze competitive parallel reactions leading
to the formation of by-products. These results suggest that
following the impregnation, some of the base sites of the
In the reaction catalysed by the cobalt complex sup-
ported on HT, the selectivities to these by-products are
lower probably due to the screening of some base sites of
the support.
The use of the cobalt-free hydrotalcites HT, CHT and
RHT for this reaction leads to the selective formation of
cyclohexeneperoxide by the epoxidation of cyclohexene
with high conversions. The by-products cyclohexanone,
cyclohexenone and unidentified by-products are formed to
low extent while the presence of the base sites in the hy-
drotalcite hinders strongly the formation of 1,2 cyclohex-
anediol by opening of the oxirane ring.
4 Conclusions
Hydrotalcite catalysts, reconstructed by memory effect and
their corresponding mixed oxides have been studied in the
123

Epoxidation of Cyclohexene With H2O2 and Acetonitrile
317
4. Sheldon RA (2002) In: Cornils B, Herrmann WA (eds) Applied
homogeneous catalysis with organo-metallic compounds, vol 1,
2nd edn. VCH, Weinheim, Germany, p 412
5. Sheldon RA, Kochi JK (1981) Metal-catalyzed oxidation of
organic compounds. Academic Press, New York
6. Holm RH (1987) Chem Rev 87:1401
selective oxidation of cyclohexene with H2O2 in the pres-
ence of acetonitrile as reductant agent. The activity and
selectivity of the catalysts depends on the molar ratio
cyclohexene/H2O2 as well as on the nature of the solvent
used.
7. Ono Y (2003) J Catal 216:406
Correlating the results of the determination of surface
base sites and those obtained in the catalytic tests leads to
the conclusion that the highest catalytic activity is obtained
for the samples exhibiting the highest number of weak and
medium strength base sites. Therefore the base strength of
the catalysts has been proven to be essential for the
selective conversion of cyclohexene to the corresponding
epoxide.
8. Sels BF, De Vos DE, Jacobs P (2001) Catal Rev Sci Eng 43:443
9. Banfi S, Legramandi F, Montanari F, Pozzi G, Quici S (1991)
J Chem Soc, Chem Commun 1285
10. Mukaiyama T (1996) Aldrichimica Acta 29:59
11. Fdil N, Romane A, Allaoud S, Karim A, Castanet Y, Mortreux A
(1996) J Mol Catal A Chem 108:15
12. Blaz E, Pielichowski J (2006) Molecules 11:115
13. Varkey SP, Ratnasamy C, Ratnasamy P (1998) J Mol Catal A
Chem 135:295
14. Ernst S, Traa Y, Deeg U (1994) Stud Surf Sci Catal 84:925
15. Seelan S, Sinha AK (2003) Appl Catal A Gen 238:201
16. Ernst S, Disteldorf H, Yang X (1998) Microporous Mesoporous
Mater 22:457
17. Lim SY, Kang M, Kim JM, Lee IM (2005) Bull Korean Chem
Soc 26:887
18. Kowalski G, Pielichowski J, Jasieniak M (2003) Appl Catal A
Gen 247:295
19. Wentzel BB, Leinonen SM, Thompson S, Sherrington DC,
Feiters MC, Nolte RJM (2000) J Chem Soc, Perkin Trans 1:3428
20. Chibwe M, Pinnavaia TJ (1993) J Chem Soc Chem Commun
3:278
21. Chatti I, Ghorbel A, Grange P, Colin JM (2002) Catal Today
75:113
22. Tsyganok A, Sayari A (2006) J Solid State Chem 179:1830
Modifying the hydrotalcite structure with low concen-
trations of Cobalt ions (ca. 2% w/w) localized in the bru-
cite-type layer does not change sensibly the performances
of the catalyst compared to the un-modified hydrotalcite.
The catalyst modified with the same concentration of
cobalt using cobalt complexes with triphenyphosphine and
chlorine ligands as cobalt source lead to interactions
between the complex species and some of the base sites of
the support which affect negatively their activity and
selectivity.
The differences between the performances of
(Co-Mg)0.75Al0.25 prepared by co-precipitation and Co-
complex/HT obtained by impregnation suggest that the
hydrogen peroxide alters by oxidation the Co-species grafted
to the base carrier, leading also to a leaching of the complex
during the catalytic test. The modification of the composition
and structure of these Co-complex species is reflected
by the different activity and selectivity compared to
(Co-Mg)0.75Al0.25 that has Co2? ions in the brucite-type
layer. Researches are in course using air or pure oxygen as
oxidant in order to avoid the destruction of the cobalt
complex.
ˆ
23. Angelescu E, Pavel OD, Che M, Bırjega R, Costentin G (2004)
Catal Commun 5:647
ˆ
24. Bırjega R, Pavel OD, Costentin G, Che M, Angelescu E (2005)
Appl Catal A Gen 288:185
25. Cotton FA, Faut OD, Goodgame DML, Holm RH (1961) J Am
Chem Soc 83(8):1780
26. Parida K, Das J (2000) J Mol Catal A Chem 151:185
27. Lever ABP (1986) Inorganic electronic spectroscopy, 3rd edn.
Elsevier, Amsterdam New York, pp 480–505
´
28. Leroux F, Moujahid EM, Taviot-Gueho C, Besse J-P (2001) Solid
State Sci 3:81
29. Zeng HC, Xu ZP, Qian M (1998) Chem Mater 10:2277
30. Rives V, Ulibarri MA (1999) Coord Chem Rev 181:61
˘
ˆ
31. Angelescu E, Zavoianu R, Pavel OD, Bırjega R, Angelescu A
(2007) Revista de Chimie Bucuresti 58:1104
32. Alfassi ZB, Neta P (1997) J Phys Chem A 101:2153
33. Corma A, Fornes V, Martin-Aranda MR, Rey F (1992) J Catal
134:58
Acknowledgments The authors express their gratitude for the
financial support of this work provided by ANCS-MECT through
grant 71-043/2007.
34. Angelescu E, Pavel OD, Birjega R, Florea M, Zavoianu R (2008)
Appl Catal A Gen 341:50
35. Kumbhar PS, Sanchez-Valente J, Figueras F (1998) Chem
Commun 10:1091
References
36. Takehira T, Kawabata T, Shishido T, Murakami K, Ohi T, Shoro
D, Honda M, Takaki K (2005) J Catal 231:92
1. Payne GB (1961) J Org Chem 26(3):668
2. Payne GB (1962) Tetrahedron 18:763
´
3. Kirm I, Medina F, Rodrıguez X, Cesteros Y, Salagre P, Sueiras J
(2004) Appl Catal A Gen 272:175
123