Dehydrogenation of cyclohexanol over Cu/Al2O3 catalysts prepared with different precipitating agents

Article abstract of DOI:10.1016/j.apcata.2013.07.063

Dehydrogenation of cyclohexanol over Cu/Al₂O₃ catalysts (molar Cu:Al = 1:1) prepared by reduction of mixed oxide precursors synthesized using different precipitating agents viz. potassium carbonate, tetraalkyl ammonium hydroxides (TAAOHs) and urea was investigated. In order to assess the efficacy of TAAOH further, the chain length of tetraalkyl ammonium cations was also varied and the resulted catalysts were evaluated for their catalytic performance. The catalysts were characterized by powder X-ray diffraction, low temperature nitrogen adsorption, temperature programmed desorption of ammonia and UV-visible diffuse reflectance spectroscopy. The dependence of the catalyst performance on the precipitating agent employed during its synthesis has been clearly demonstrated. The use of TAAOH as precipitating agent led to the formation of a catalyst with better catalytic activity than those prepared using potassium carbonate and urea. Further improvement in the catalytic performance was observed when TAAOH with longer alkyl chain ammonium cation was used. The optimum catalyst prepared by reduction of mixed oxide precursor synthesized using tetrapropyl ammonium hydroxide as precipitating agent, showed highest cyclohexanol conversion (81.5%) and cyclohexanone selectivity (79.6%) at 250 C on account of higher Cu ⁺/Cu⁰ ratio, well dispersed copper, higher surface area and lower total acidity with higher contribution of sites with moderate strength.

Full text of DOI:10.1016/j.apcata.2013.07.063

Applied Catalysis A: General 467 (2013) 421–429
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Dehydrogenation of cyclohexanol over Cu/Al O catalysts prepared
2
3
with different precipitating agents
∗
Nilesh P. Tangale, Prashant S. Niphadkar, Shilpa S. Deshpande, Praphulla N. Joshi
Catalysis & Inorganic Chemistry Division, CSIR-National Chemical Laboratory, Dr. Homi Bhabha Road, Pashan, Pune 411008, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 April 2013
Received in revised form 21 July 2013
Accepted 31 July 2013
Available online 18 August 2013
2 3
Dehydrogenation of cyclohexanol over Cu/Al O catalysts (molar Cu:Al = 1:1) prepared by reduction
of mixed oxide precursors synthesized using different precipitating agents viz. potassium carbonate,
tetraalkyl ammonium hydroxides (TAAOHs) and urea was investigated. In order to assess the efficacy
of TAAOH further, the chain length of tetraalkyl ammonium cations was also varied and the resulted
catalysts were evaluated for their catalytic performance. The catalysts were characterized by powder
X-ray diffraction, low temperature nitrogen adsorption, temperature programmed desorption of ammo-
nia and UV–visible diffuse reflectance spectroscopy. The dependence of the catalyst performance on the
precipitating agent employed during its synthesis has been clearly demonstrated. The use of TAAOH as
precipitating agent led to the formation of a catalyst with better catalytic activity than those prepared
using potassium carbonate and urea. Further improvement in the catalytic performance was observed
when TAAOH with longer alkyl chain ammonium cation was used. The optimum catalyst prepared by
reduction of mixed oxide precursor synthesized using tetrapropyl ammonium hydroxide as precipitating
Keywords:
Cyclohexanol dehydrogenation
Cu/Al2O3
Precipitating agent
Tetraalkyl ammonium hydroxide
Characterization
◦
agent, showed highest cyclohexanol conversion (81.5%) and cyclohexanone selectivity (79.6%) at 250 C
+
0
on account of higher Cu /Cu ratio, well dispersed copper, higher surface area and lower total acidity
with higher contribution of sites with moderate strength.
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
Cu-ZnO-Al O3 [8], Rh-Cu/Al O3 [9], Cu/SiO2 [10–12], Cu/Fe/SiO2
2 2
[
13], Cu-Zn/SiO2 [14], Co-␥Al O [15], V O5-MoO -M O (M = Na,
2 3 2 3 2
Cyclohexanone [cyclohexyl ketone, C H10O] has gained a great
K, Cs) [16], Cu-Mg/Zn-Al [17,18], CuNi-CeO2 [19], ␥-Al O /␣-
6
2 3
deal of commercial importance due to its high solvent power
and reactivity. By virtue of its peculiar properties, it has occu-
pied a special place in various fields such as an industrial solvent
for paints and dyes, in pesticides, an activator in oxidation reac-
tions, an intermediate for pharmaceuticals, films, coating, soaps
and more importantly, as a raw material for the production of cyclo-
hexanoneoxime, cyclohexanone resins, adipic acid, ␧-caprolactam,
nylon-6, etc. The dehydrogenation of cyclohexanol is an indus-
trially important process for the production of cyclohexanone
which is endothermic (ꢀH = 65 kJ/mol) and thermodynamically
reversible [1,2]. The combined effect of the endothermic nature
and the necessity of minimizing the energy usage has stimu-
lated the search for the promising catalyst that can catalyze
Al O [20], Fe-TiMCM-41 [21,22], Cu-Cr-Mg-Al [23], etc. have been
2 3
reported as an active catalysts. Depending upon the catalyst char-
acteristics and reaction conditions, besides the formation of the
desired cyclohexanone product, other by-products such as cyclo-
hexene, phenol, cracked and condensed products are also formed
due to dehydrogenation, dehydration and aromatization of cyclo-
hexanol and other side reactions [1,14]. As a result of encouraging
results exhibited by Cu-based catalysts at low temperature, one of
the several attempts made to enhance the cyclohexanol conversion
and cyclohexanone selectivity was the use of catalysts with higher
Cu content. Among three active commercial catalysts with Cu = 24,
26 and 65 wt%, the catalyst with 26 wt% Cu has shown higher
activity on account of higher dispersion and smaller Cu crystallite
◦
◦
the reaction at lower temperature (150–300 C) rather than at
size, whereas at 290 C, the catalyst containing 65 wt% Cu showed
◦
higher temperature (350–450 C) producing cyclohexanone within
about 70% cyclohexanone yield with lower WHSV [1]. These cat-
alysts showed a high stability even at very high time on stream
[2]. The possible cause/s for the drop in the initial activity were
identified as poisoning, coking and sintering, solid phase transfor-
mation, etc. [2]. The combined results of rate measurements and
IR spectroscopic studies on Cu-Zn-Al (Cu loading 15 at.%) and Cu-
Mg (Cu loading 52 at.%) catalysts showed that, the dehydrogenation
of cyclohexanol over Cu+ proceeds through nondissociative cyclo-
hexanol adsorption with subsequent formation of cyclohexanol
acceptable range. At high reaction temperature, catalyst such as
the ZnO/CaO, has shown 70% conversion and 99% cyclohexanone
selectivity [3]. Under milder conditions, various binary and ternary
metal oxide catalyst systems such as CuCr O4 [1], Cu-Al O3 [4–7],
2
2
∗ Corresponding author. Tel.: +91 20 25902457; fax: +91 20 25902633.
E-mail address: pn.joshi@ncl.res.in (P.N. Joshi).
0
926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.07.063

4
22
N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
0
alcoholate, whereas cyclohexanol adsorption on Cu is dissociative
and proceeds via formation of cyclohexanol alcoholate and pheno-
late species [18]. Cu/SiO2 catalyst with copper content as high as
(20%) solutions of tetramethyl ammonium hydroxide (TMAOH),
tetraethyl ammonium hydroxide (TEAOH), tetrapropyl ammonium
hydroxide (TPAOH) were purchased from V.P. Chemicals, Pune,
India.
5
0 wt%, prepared by precipitation at pH > 9 using sodium carbonate
as precipitating agent showed 74.7% cyclohexanone yield at 300 C
◦
[
11]. The selectivity to cyclohexanone may also be governed by the
2.2. Catalyst preparation
acidity of support metal oxide [10,11,24].
The chemical kinetics and mechanism of the reaction showed
that, the two kinds of copper active sites are present on the cat-
alyst viz. monovalent copper and metallic copper. The metallic
copper is responsible for the phenol formation and selectivity of
cyclohexanone while monovalent copper is responsible for for-
mation of cyclohexanone only [17]. At low temperature, catalytic
performance of copper-based catalysts primarily depends on the
presence of metallic copper and monovalent active sites [17]; the
acid-base nature of catalyst [5]; and the size, population and dis-
persion of the Cu species [1]. The activity and stability of the
Cu containing catalyst mostly depend on the preparation method
and the parameters employed in the particular method [7]. Sev-
eral methods such as, chemisorption hydrolysis [5], precipitation
Cu/Al O3 catalysts (molar Cu:Al = 1:1) were prepared by
2
reduction of mixed oxide precursors. These precursors were
synthesized by simultaneous co-precipitation method similar to
described elsewhere [33]. Aqueous solutions of potassium car-
bonate, TMAOH, TEAOH and TPAOH were used separately as
precipitating agents. Typically, solution A was prepared by dis-
solving 9.1 g of Cu(NO ) ·3H O in 250 ml deionized water followed
3
2
2
by the addition of 14.07 g of Al(NO ) ·9H O. The solution B was
3
3
2
prepared by dissolving 20.73 g of anhydrous K CO in 250 ml deion-
2
3
ized water. Solutions A and B were simultaneously added with
constant rate at room temperature under vigorous stirring. The
final pH of solution was adjusted to 8.5. The stirring was further
continued for 5 h at room temperature. The precipitate was then
recovered by filtration, washed thoroughly till effluent showed no
[
7,8,11,17,18], ion exchange [12], impregnation [12–14], electro-
less plating [6], sol–gel [25], etc. were reported for preparation
of copper-based catalysts. Among these methods, co-precipitation
method is the better choice for the controlled preparation of
binary and ternary metal oxide catalyst systems from simplic-
ity and scale up point of view. By this method, one can also
minimize the deposition of non-uniform layered copper which
are very susceptible to sintering and hence in turn catalyst life
and activity. In simultaneous co-precipitation method, it is very
important to select precipitating agent judiciously as it plays a
crucial role in the controlling the size and dispersion of the cop-
per. Various precipitating agents such as aqueous solutions of
Na CO , K CO , (NH ) CO , NaOH, KOH, NH OH and urea were
◦
potassium and then dried at 100 C for 12 h. The dried powder was
◦
ground to fine powder and then calcined at 400 C for 3 h. The sam-
ple was designated as CAR. By following the identical procedure,
mixed oxide precursors were prepared using different precipitating
agents viz. aqueous solutions of TMAOH, TEAOH and TPAOH. The
resultant samples were designated as TMA, TEA and TPA respec-
tively.
For catalytic testing purpose, a catalyst designated as Alumina
was prepared by following the identical procedure as described
above except use of Cu (NO ) ·3H O.
3
2
2
Another mixed oxide precursor was prepared by following
homogeneous precipitation method using urea similar to described
elsewhere [4]. In a typical procedure, a solution was prepared by
dissolving 9.1 g of Cu(NO ) ·3H O in 250 ml deionized water fol-
2
3
2
3
4
2
3
4
commonly used to prepare the Cu/Al O3 catalysts. Although, the
2
tetraalkylammonium hydroxides (TAAOH) were used in synthe-
sis of zeolite or zeolite-like materials [26–28], stabilization of
metal colloids [29], preparation of nanocrystalline yttrium oxide
3
2
2
lowed by the addition of 14.07 g of Al(NO ) ·9H O. To this solution,
3
3
2
an aqueous solution containing required amount of urea was added
[
[
30], spinel cobalt ferrite [31] and nanosized titania powder
32], etc. until now, as per our knowledge, there are no reports
with constant stirring. This homogeneous solution was heated at
◦
9
2 ± 3 C till pH becomes 7.5. The precipitate was recovered by fil-
on use of TAAOHs as precipitating agents for the preparation
of binary/ternary metal oxide catalyst systems with a view to
maximize the yield of cyclohexanone by cyclohexanol dehydro-
genation.
◦
tration, washed thoroughly with de-ionized water, dried at 100 C
for 12 h and finally calcined at 400 C for 3 h. The sample was des-
◦
ignated as URE.
The aim of work is to investigate the performance of Cu/Al O3
2
catalysts (molar Cu:Al = 1:1) prepared by reduction of mixed oxide
precursors synthesized using different precipitating agents viz.
potassium carbonate, TAAOH and urea in dehydrogenation of cyclo-
hexanol. In view of maximizing the cyclohexanone yield, further
assessment of the efficacy of TAAOH was examined by varying
the chain length of tetraalkyl ammonium cations. All the catalysts
were characterized by powder X-ray diffraction, low tempera-
ture nitrogen adsorption, temperature programmed desorption of
ammonia and UV–visible diffuse reflectance spectroscopy. Finally,
2.3. Characterization
The phase identification, relative crystallinity and an estimation
of the crystallite sizes (by applying Debye–Scherrer equation) were
carried out by powder X-ray diffraction (XRD) patterns recorded
on a P Analytical PXRD system (Model XPert-PRO-1712) using Ni
filtered Cu K␣ radiation (ꢁ = 0.154 nm). For elucidating the nature
of copper species in the samples, a diffuse reflectance UV–vis
(DRUV–vis) spectrum was recorded on a Perkin-Elmer Lambda 6
spectrometer (Lambda 650) using BaSO4 as a reference. BET surface
results of the cyclohexanol dehydrogenation to cyclohexanone over
◦
◦
Cu/Al O3 catalysts at reaction condition of 250 C were discussed
areas were measured by N2 adsorption measurements at −196 C
2
and presented in this paper.
using Quanta Chrome CHEMBET 3000 instrument. The acidity and
acidic strength of sites were measured by NH -temperature pro-
3
grammed desorption using Micromeritics ChemiSorb (2720, USA)
equipped with thermal conductivity detector. The samples were
2
. Experimental
◦
3
−1
degassed at 200 C in He (25 cm min ) for 1 h prior to the mea-
◦
2.1. Materials
surement. Then, the temperature was decreased to 35 C and NH3
was allowed to adsorb by exposing the gas stream containing 10%
NH3 in He for 1 h. It was then flushed with He for another 1 h.
Aluminum nitrate nonahydrate (98%), copper nitrate trihydrate
99.5%), cyclohexanol (99%) were procured from Loba Chemie,
Mumbai, India. Potassium carbonate (99.5%) and urea (99.5%) were
procured from Thomas Baker and Qualigens respectively. Aqueous
3
−1
(
The NH3 desorption was carried out in He flow (25 cm min ) by
◦
increasing the temperature up to 750 C with the heating rate of
◦
−1
.
10 C min

N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
423
URE
r-URE
r-TMA
TMA
CAR
r-CAR
1
0
20
30
40
50
60
70
1
0
20
30
40
50
60
70
2
θ/º
2θ/º
[A]
[B]
Fig. 1. X-ray diffraction patterns of (A) URE, TMA, CAR and (B) r-URE, r-TMA, r-CAR.
2
.4. Catalytic activity
and full width at half maximum (FWHM) of these peaks were found
to be weaker and broader in case of CAR compared to TMA. More-
over, the most intense and narrow peaks were observed in case of
URE. The degree of crystallinity and the magnitude of crystallite
size of CuO were found to follow the trend as: CAR < TMA < URE.
The crystallite size and intensity of the CuO peaks can be taken as
an indirect measure, at least qualitatively, to analyze the copper
dispersion [1,36,37]. Thus, considering the trend observed in the
degree of crystallinity and the crystallite size, the copper disper-
sion in samples followed the trend as: CAR > TMA > URE. This can
be partly explained on the basis of the type of the anionic species
that were made available by the precipitating agent during precip-
itation process. In case of potassium carbonate precipitating agent,
the carbonate species generated during precipitation process are
likely to form molecular precursors such as amorphous/crystalline
forms of Cu (OH) CO , along with Cu(OH) and CuO phases [41].
The dehydrogenation of cyclohexanol over Cu/Al O3 catalysts
2
(
molar Cu:Al = 1:1) prepared by reduction of mixed oxide precur-
sors synthesized using different precipitating agents was carried
out in a continuous, down-flow, fixed bed reactor (SS316, 40 cm
length, 2 cm I.D.) at atmospheric pressure. In a reactor, the reduced
catalyst (3.0 g, self-bound granules with 20–40 mesh size) was
sandwiched at the center by inert ceramic beads. The reactor
was heated to the required temperature with an electrical tubu-
lar furnace having digital temperature controller cum indicator.
The mixed oxide precursor was reduced in situ in a H flow of
5
reactor with weight hour space velocity (WHSV) ranging from of
3
lected after 1 h and analyzed by gas chromatography (Shimadzu,
HP 5890A) using a flame ionization detector (FID) and carbowax
column (30 m length, 0.53 mm I.D. and 1 ␮m film thickness). Quan-
tification of various products was accomplished using response
factors of typical standard mixtures. The liquid mass balance was
2
3
−1
◦
5 cm min
at 250 C for 8 h. The cyclohexanol was fed to the
− −1
1
3
.0 to 10.0 h with nitrogen (50 cm min ). The product was col-
2
2
3
2
Whereas, in case of other two precipitating agents, formation of
copper hydroxides and/or Cu (OH) NO [4] which formed the basis
2
3
3
for the formation of more crystalline CuO phase after calcination.
The extent of interdiffusion of copper and aluminum during the
calcination may also be responsible for the variation in the crys-
tallinity of CuO with variation in type of the precipitating agent
used. Compared to other two samples, the larger crystallite size of
CuO in URE may be associated with higher rate of nucleation of pri-
mary particles and enhanced acceleration in the growth of primary
particles probably due to the strong interaction of copper with sup-
port oxide [30]. Further, the reduction in crystallite size of CuO in
TMA might be associated with the suppression of the growth of
precipitate particle size by promoting the nucleation of new par-
ticle during precipitation [30]. All the three samples showed no
peaks due to alumina indicating the presence of amorphous alu-
mina. Moreover, no XRD visible contribution of phase/s such as
9
8 ± 2%. Based on the product distribution pattern, the cyclo-
hexanol conversion and the selectivity to the products such as,
cyclohexanone, phenol, and cyclohexene were calculated.
3
. Results and discussion
3.1. Characterization of CAR, URE and TMA
Fig. 1A shows the powder X-ray diffraction patterns of CAR, URE
and TMA. All the samples have shown the presence of the charac-
teristic peaks CuO phase (JCPDS File no. 65-2309) [33–35] of which
most prominent peaks appeared at 35.64 and 38.85 . The intensity
◦
◦

4
24
N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
Table 1
+
0
BET surface area, crystallite size, Cu /Cu ratio and acidic site distribution of catalysts prepared by reduction of mixed oxide precursors synthesized using different precipitating
agents.
BET specific surface area (m2
g
−1
)
Crystallite size (nm)
+
Cu /Cu0
−1
Total acidity (mmol g )
Acid site distribution (%)
Catalyst
Cu2O (d = 0.24635)
Cu0 (d = 0.20868)
Weak
Moderate
Strong
r-CAR
r-URE
r-TMA
r-TEA
r-TPA
70
53
102
104
139
14.3
8.8
11.7
11.4
9.4
11.9
26.9
17.6
15.9
13.3
0.51
0.55
0.42
0.45
0.56
0.184
0.173
0.163
0.161
0.158
20.5
21.9
20.0
19.3
17.1
64.3
71.7
73.4
76.1
78.4
15.2
6.4
6.6
4.6
4.5
Cu O, CuAl O , CuAlO , CuAl O7, etc. was observed in all the three
samples.
found to follow the trend as: r-URE > r-CAR > r-TMA. Assuming that,
the differences in the chemical environment and coordination of
copper species present in pre-reduced samples may be the one of
the reasons for showing such trend, further probing experiments
were conducted using DRUV–vis spectroscopy.
2
2
4
2
4
After reducing CAR, TMA and URE samples in a H flow
2
3
−1 ◦
(
55 cm min ) at 250 C for 8 h, they were designated as r-CAR, r-
TMA and r-URE, respectively. The XRD patterns of all these reduced
samples are shown in Fig. 1B. All the samples have exhibited char-
The deconvoluted DRUV–vis spectra of CAR, URE and TMA are
shown in Fig. 2. Upon deconvolution, three peaks were emerged
and their area percentages are provided in Table 2. As can be seen
from the deconvoluted spectra of all the three samples that, (1) the
first band maximum was observed at about 280 nm, (2) the sec-
ond maximum was observed in the range of 350–450 nm, and (3)
the third maximum observed at about 650 nm. These bands can be
acteristic peaks due to Cu O phase (JCPDS File no.78-2076) and
2
metallic copper (JCPDS File no. 85-1326) [37] at the cost of CuO
phase. In all the samples, there is complete disappearance of CuO
phase after reduction. The most prominent peaks of Cu O phase
2
◦
◦
◦
were appeared at 36.45 and 43.32 whereas of metallic copper
were indexed at 43.32 and 50.44 [33]. The trend observed in
the degree of crystallinity and the magnitude of crystallite size of
◦
2+
ascribed to the charge transfer between mononuclear Cu ion and
oxygen, the charge transfer between Cu2+ cluster and oxygen in
[Cu-O-Cu]n cluster-like species and the d–d transition of Cu with
an octahedral environment in CuO [37,40,41], respectively. These
assignments and the corresponding values in Table 2 indicated that,
the precipitating agents play a prominent role in the distribution
of copper species as isolated Cu2+cations, oligonuclear [Cu-O-Cu]n
species and crystalline CuO. The heterogeneous distribution and
population of readily reducible copper ions and copper cluster
species might be responsible for governing the ratio of metallic
copper (non-selective) to the monovalent copper (selective) sites
after their reduction.
Cu O and metallic copper were found to follow the similar trend
2
as CuO phase as: CAR < TMA < URE. Although, the intensity of the
peaks of Cu O and metallic copper were reported to increase with
2
increase in the copper content [38,39], in present studies, it may
be attributed to the variation in the precipitating agent as there is
no variation made in the copper content. Irrespective of the pre-
cipitating agent used, it is likely that, during reduction process, the
sequential reduction of CuO to Cu O cubical phase occurs first as
2
stable intermediate and then to metallic copper [33]. The crystallite
size of metallic copper was estimated by Debye–Scherrer equation
and the values were summarized in Table 1.
In cyclohexanol dehydrogenation to cyclohexanone reaction,
monovalent copper and metallic copper are the two kinds of active
sites which are revealed as selective and non-selective sites respec-
tively [17]. Therefore, an assessment of XRD patterns of r-CAR,
r-TMA and r-URE was carried out to estimate the relative propor-
The nature of acidic sites in r-URE, r-CAR and r-TMA were
evaluated by temperature programmed desorption of ammonia
measurements. The profiles which were obtained in the present
studies are shown in Fig. 3 and acid site distribution depending
on their strengths is tabulated in Table 1. It can be seen that,
three prominent desorption zones appeared at lower temperature
tions of XRD visible Cu O and metallic copper phases. The ratio
2
◦
◦
between the integrated area under the peaks associated with the
Cu O and the integrated area under the peaks associated with the
2
(<200 C), in the temperature range (200–500 C), in the higher
◦
temperature (500–750 C) regions. The origin of these zones can
metallic copper was taken as a measure for the qualitative analysis
be assigned to presence of weak, moderate and strong acidic sites,
respectively [13]. The trend in % contribution due to weak acidic
sites was found to follow the trend as: r-URE > r-CAR > r-TMA.
Moreover, the trend observed in % contribution due to intermedi-
ate acidic sites was as: r-TMA > r-URE > r-CAR. Further, the trend
observed in % contribution due to strong acidic sites was as: r-
CAR > r-TMA > r-URE. The higher acidity of the catalyst plays an
important role in favoring the selectivity of cyclohexene by dehy-
dration of cyclohexanol [10,11,24]. It can be seen from Table 1 that,
r-CAR and r-URE catalysts possess higher total acidity as compared
to the catalyst r-TMA.
+
0
and these values are provided in Table 1 as Cu /Cu . The BET sur-
face areas of r-URE, r-TMA and r-CAR are also included in Table 1. It
0
is clearly evident that, the trend observed in Cu crystallite size
was as: r-CAR < r-TMA < r-URE whereas in the BET surface area,
it was as: r-URE < r-CAR < r-TMA. Interestingly, the r-URE sample
has shown lower surface area in spite of having larger crystallite
size. Similarly, r-TMA has shown higher surface area than r-CAR
despite the fact that it has larger crystallite size compared to r-
CAR. Such mismatch may be attributed to the differences in the
origin of the target material. The estimation of crystallite sizes was
done using metallic copper phase whereas the target material for
evaluation of BET surface area was the Cu/Al O composite phase.
2
3
0
Probably, larger crystallite size of Cu in r-URE may also be respon-
sible for the formation of bigger particles via aggregation leading to
decrease in BET surface area. Although the samples were reduced
under the identical conditions, Cu /Cu were found to vary with
the variation in precipitating agent (Table 1). Thus, there exists
a dependence of relative proportions of XRD visible selective and
non-selective catalytic active sites on the precipitating agent. In this
regard, compared to XRD visible metallic copper (non-selective)
sites, the population of monovalent copper (selective) sites was
Table 2
2+
Variations in percentagesof isolated Cu cations, oligonuclear [Cu–O–Cu]n species
and crystalline CuO as a function of precipitating agents.
+
0
Isolated Cu2+
Catalyst
[Cu–O–Cu]n
CuO
URE
CAR
TMA
TEA
TPA
21
27
23
18
17
60
57
59
62
64
19
16
18
20
19

N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
425
1
0
8
6
4
2
0
1
0
8
6
4
2
0
URE
CAR
200
300
400
500
600
700
800
2
00
300
400
500
600
700
800
Wawelength nm
Wawelength nm
8
6
4
2
0
TMA
200
300
400
500
600
700
800
Wawelength nm
Fig. 2. DRUV–vis spectra of URE, TMA and CAR.
3
.2. Characterization and comparison of TEA and TPA with TMA
Fig. 4A. The intensity of CuO peaks were found to increase with
the decrease in chain length of tetraalkyl ammonium cation. The
lower intensities and higher FWHM of peaks for CuO in TPA sample
indicates high dispersion of smaller CuO species than TMA and TEA
samples [1,36,37]. Similarly, the powder X-ray diffraction patterns
of r-TEA and r-TPA were shown along with r-TMA for compari-
son purpose in Fig. 4B. The intensity and width of the diffraction
lines of metallic copper were found to increase with the decrease
in chain length of tetraalkyl ammonium cation. The ratio between
In order to maximize the conversion of cyclohexanol in dehy-
dration of cyclohexanol reaction, attempts were made to prepare
mixed oxide precursors using TAAOH having longer alkyl chain
ammonium cations. The powder X-ray diffraction patterns of TEA
and TPA were shown along with TMA for comparison purpose in
the integrated area under the peaks associated to the Cu O and the
2
integrated area under the peaks associated to the metallic copper
+
0
(
Cu /Cu ) were estimated and provided in Table 1. The BET surface
areas of r-TEA and r-TPA are also included in Table 1. It is clearly
evident from the Table 1 that, longer the alkyl chain ammonium
cation, smaller the crystallite size of the catalyst. The reverse trend
was observed in case of BET surface area. The r-TPA has shown
the higher BET surface area and smaller crystallite size among all
the three samples. This can be explained partly on the basis of
longer chain length of tetrapropyl ammonium cation which might
be responsible for the reduction in collision rate during precipi-
tation and suppression of particle growth rate [32]. Longer alkyl
chain ammonium cation seems to be operative in the increasing the
relative distance between precipitated copper species-hydroxide
entity which results in the increase in the nucleation growth. The
extent of suppression in the copper agglomeration was found to
follow the trend as: r-TMA > r-TEA> > r-TPA.
TMA
URE
CAR
Although, the samples were reduced under the identical condi-
0
100
200
300
400
500
600
700
800
+
0
o
tions, Cu /Cu were found to vary with the variation in chain length
Temperature ( C)
+
0
of tetraalkyl ammonium cation. The Cu /Cu ratio (Table 1) was
found to increase with increase in the chain length of tetraalkyl
Fig. 3. NH3-TPD profiles r-URE, r-TMA and r-CAR.

4
26
N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
r-TPA
r-TEA
TPA
TEA
TMA
r-TMA
10
20
30
40
50
60
70
10
20
30
40
50
60
70
2
θ/º
2θ/º
[
A]
[B]
Fig. 4. X-ray diffraction patterns of (A) TMA, TEA, TPA and (B) r-TMA, r-TEA, r-TPA.
ammonium cation and followed the trend as: r-TMA < r-TEA < r-
acidity decreases with increase in the copper content in Cu/Al O3
2
TPA. Thus, longer the chain length of tetraalkyl ammonium cation,
higher is the population of Cu as compared to Cu that are XRD
visible.
catalyst, in present studies, it may be attributed to the variation
in the precipitating agent as there is no variation made in copper
content.
+
0
DRUV–vis absorption spectra of TMA, TEA and TPA were
recorded to understand the nature of Cu species in catalyst as
shown in Fig. 5. Similar to TMA, TEA and TPA have exhibited the
similar patterns except the changes in the area percentages of
deconvoluted peaks. The distribution of copper species of all the
catalysts is summarized in Table 2. It can be clearly seen from the
table that, as the chain length of the tetraalkyl ammonium cation
increases, there is a decrease in the percentage of mononuclear Cu2
species accompanied by the increase in the oligomeric [Cu-O-Cu]n
species. However, the amount of CuO species in all the samples
remained unaltered irrespective of the chain length of organoca-
tion.
3
3
.3. Catalytic performance
.3.1. Influence of precipitating agents
The catalytic performance of r-CAR, r-URE and r-TMA in the
dehydrogenation of cyclohexanol to cyclohexanone is depicted in
Fig. 7. In the absence of catalyst, there was almost no cyclohexanol
conversion at 250 C. However, under the identical set of reac-
tion conditions, Alumina showed10% cyclohexanol conversion with
nearly 83.5% cyclohexene selectivity. The higher selectivity may be
attributed to the acidity of alumina which is capable of dehydrating
the cyclohexanol leading to the selective formation of cyclohexene
+
◦
Further attempts were made to study the effect of chain length
in tetraalkyl ammonium cation on the acidity of catalysts by estab-
[
9–11,24].
Among r-CAR, r-URE and r-TMA, r-TMA showed the highest
lishing the NH -temperature programmed desorption of ammonia
3
cyclohexanol conversion (69.6%) with 79.4% cyclohexanone selec-
tivity. Probably, the higher conversion of cyclohexanol might be
due to the high amount of readily reducible copper ions and cop-
per cluster like species [Cu-O-Cu]n [5]. The improved selectivity
of cyclohexanone over r-TMA may be associated with combined
effect of high BET surface area, low total acidity, small crystallite
size and well dispersed copper. Probing further into the role of Cu
and Cu species in controlling the selectivity of cyclohexanone, the
catalytic activity of all the samples were compared as a function
of Cu /Cu ratio (Table 1). Despite of lower Cu /Cu ratio, r-TMA
exhibited highest cyclohexanone selectivity. This suggests that, not
profiles. The profiles of r-TMA, r-TEA and r-TPA (Fig. 6) showed
◦
prominent peaks assigned to weak acidic sites (<200 C), moder-
◦
◦
ate acidic sites (200–500 C) and strong acidic sites (500–700 C).
The total acidity expressed as millimoles of ammonia desorbed per
g of catalyst and % distribution of the acid sites are provided in
Table 1. The total acidity was found to decrease with the increase
in the chain length in tetraalkyl ammonium cations. The con-
tribution due to weak and strong acid sites in all the catalysts
was found to decrease with the increase in the chain length of
tetraalkyl ammonium cations. Interestingly, the trend observed
with respect to moderate acid sites was reverse and followed the
trend as: TMA < TEA < TPA. Although, it is reported that [24], the
+
0
+
0
+
0
+
only XRD visible Cu species are responsible for higher selectivity of
cyclohexanone, but also smaller crystallite site, lower total acidity,

N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
427
8
TEA
8
6
4
2
0
TMA
6
4
2
0
2
00
300
400
500
600
700
2
00
300
400
500
600
700
8
wawelength (nm)
wawelength (nm)
8
6
4
2
0
TPA
200
300
400
500
600
700
800
wawelength (nm)
Fig. 5. DRUV–vis spectra of TMA, TEA and TPA.
higher surface area and possibly XRD invisible Cu+ species play an
important role in enhancing the cyclohexanone selectivity. How-
ever higher selectivity of phenol over r-TMA might be due to the
sites and total acidity. Moreover, it showed lower phenol selectiv-
ity due to higher XRD visible Cu /Cu ratio confirming lower Cu0
species [17]. Although the copper crystallite size of r-CAR was low-
est among all, it showed 61.3% cyclohexanol conversion with 78.6%
cyclohexanol and 19.3% cyclohexene selectivity which is in good
agreement with reported literature [24]. Because of higher copper
crystallite size and lower BET surface area, r-URE showed lower
cyclohexanol conversion (42.1%) with 75.2% cyclohexanone selec-
tivity. Thus, the overall performance of the catalyst in cyclohexanol
dehydrogenation reaction is controlled by cumulative effect of its
BET surface area, crystallite size, total acidity, distribution of acid
+
0
0
higher population of Cu species. In comparison with r-TMA, the
r-URE showed lower cyclohexanol conversion on account of higher
+
0
XRD visible Cu /Cu ratio, lowest surface area and higher cop-
per crystallite size. The highest cyclohexene selectivity exhibited
by r-CAR can be attributed to higher contribution of strong acidic
0.35
0.30
0.25
0.20
0.15
0.10
0.05
r-TMA
r-TEA
r-TPA
+
0
sites, copper dispersion and bulk Cu /Cu ratio which, in turn,
governed by judicious choice of the precipitating agent. In view
of further improvement in the catalytic performance and to gain
insight, catalyst prepared using TAAOH with different chain length
of tetraalkyl ammonium cations were examined.
3.4. Influence of chain length of tetraalkyl ammonium cation
The catalytic conversion of cyclohexanol and the product selec-
2
tivity to cyclohexanone, cyclohexene and phenol over Cu/Al O3
catalyst prepared by using different tetraalkyl ammonium hydrox-
ides as precipitating agent are shown in Fig. 8. In present study,
r-TPA showed maximum conversion of cyclohexanol (81.5%) with
79.6% cyclohexanone selectivity. The catalyst, r-TEA, showed 78.9%
cyclohexanol conversion with 79.2% cyclohexanone selectivity.
As compared to both these catalysts, r-TMA showed the least
cyclohexanone yield. Thus, the Cu/Al2O3 catalyst prepared using
0
100
200
300
400
500
600
700
800
o
Temperature ( C)
Fig. 6. NH3-TPD profiles of r-TMA, r-TEA and r-TPA.

4
28
N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
1
00
conversion of cyclohexanol
selectivity to cyclohexanone
selectivity to cyclohexene
selectivity to phenol
100
90
8
7
6
5
4
3
0
0
0
0
0
0
80
60
40
20
0
Conversion of cyclohexanol
Selectivity of cyclohexanone
Selectivity of cyclohexene
Selectivity of phenol
20
10
0
Alumina
r-CAR
r-URE
r-TMA
2
3
4
5
6
7
8
9
10
11
-1
WHSV hr
Fig. 7. Catalytic performance of Alumina and Cu/Al2O3 catalysts prepared using dif-
ferent precipitating agents. (Reaction conditions: catalyst loading: 3 g, temperature:
2
50 ◦C, reaction time: 1 h, WHSV: 5.0 h−1, N2 flow: 50 cm3 min−1.)
Fig. 9. Effect of contact time of r-TPA on conversion of cyclohexanol and selectivity
to cyclohexanone, phenol, cyclohexene. (Reaction conditions: TPA catalyst loading:
◦
3
−
1
3
g, temperature: 250 C, reaction time: 1 h, N2 flow: 50 cm min .)
conversion of cyclohexanol
1
00
selectivity of cyclohexanone
selectivity of cyclohexene
selectivity of phenol
3.5. Influence of weight hourly space velocity (WHSV)
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
The effect of WHSV on the cyclohexanol conversion and the
product selectivities over optimum catalyst (r-TPA) is illustrated
in Fig. 9. It can be seen that, WHSV has an influence on the extent
of formation of cyclohexanone, cyclohexene and phenol. The con-
version of cyclohexanol to cyclohexanone was found to increase
considerably from 65.0 to 81.5% without considerable change in
the cyclohexanone selectivity when WHSV was decreased from 10
−
1
to 3 h
.
4. Conclusions
Cu/Al O3 catalysts (molar Cu:Al = 1:1) were prepared by reduc-
2
tion of mixed oxide precursors synthesized using different
precipitating agents viz. potassium carbonate, tetraalkyl ammo-
nium hydroxides (TAAOHs) and urea. The degree of crystallinity
and the magnitude of crystallite size of CuO in mixed oxide precur-
sor were found to follow the trend as: CAR < TMA < URE. Compared
to other two samples, the larger crystallite size of CuO in URE
may be associated with higher rate of nucleation of primary parti-
cles and enhanced acceleration in the growth of primary particles.
Although the samples were reduced under the identical conditions,
r-TMA
r-TEA
r-TPA
Fig. 8. Catalytic activity and product distribution over Cu/Al2O3 catalysts prepared
using different tetraalkyl ammonium hydroxides as precipitant. (Reaction condi-
tions: catalyst loading: 3 g, temperature: 250 C, reaction time: 1 h, WHSV: 5.0 h
◦
−1
,
N2 flow: 50 cm3 min .)
−1
+
0
Cu /Cu were found to vary with the variation in precipitating
agent. The r-TMA catalyst prepared using tetramethyl ammonium
hydroxide as precipitating agent was found to be potential cata-
lyst for cyclohexanol dehydrogenation at lower temperature. The
overall performance of the catalyst in cyclohexanol dehydrogena-
tion reaction is controlled by cumulative effect of its BET surface
area, crystallite size, total acidity, distribution of acid sites, cop-
longer chain length in tetraalkyl ammonium cation has shown
enhanced catalytic performance. An increase in the chain length of
tetraalkyl ammonium cation resulted in the improved catalytic per-
formance in the order: r-TMA < r-TEA < r-TPA. Further the increase
in chain length of organocation corresponded to the increased
in the intermediate acidity per sq. meter of the catalysts. The
enhancement of chain length in tetraalkyl ammonium cation may
increase the degree of nucleation and the growth of new parti-
cle leads to increase in dispersion of copper which in the order
TMA < TEA < TPA. The establishment of increasing trend of reducible
copper cluster species [Cu-O-Cu]n facilitate the cyclohexanol con-
version [5]. The optimum catalyst prepared by reduction of mixed
oxide precursor synthesized using tetrapropyl ammonium hydrox-
+
0
per dispersion and bulk Cu /Cu ratio which, in turn, governed
by judicious choice of the precipitating agent. The increase in the
chain length of tetraalkyl ammonium cation was found to improve
further the catalytic performance in terms of cyclohexanol conver-
sion in the order of r-TMA < r-TEA < r-TPA with similar selectivity
◦
to cyclohexanone. In present studies, at temperature 250 C, after
reduction of mixed oxide precursor synthesized using tetrapropyl
ammonium hydroxide as precipitating agent, showed 81.5% cyclo-
hexanol conversion and 79.6% selectivity. It can be concluded that,
use of TAAOHs as precipitating agents resulted in the formation of
ide as precipitating agent, showed highest cyclohexanol conversion
◦
(
81.5%) and cyclohexanone selectivity (79.6%) at 250 C on account
+
0
of higher Cu /Cu ratio, well dispersed copper, higher surface area
and higher population of moderate acidic sites. It has shown 1.7%
phenol selectivity.
+
0
catalyst with higher Cu /Cu ratio, well dispersed copper, higher
surface area and lower total acidity with higher population of

N.P. Tangale et al. / Applied Catalysis A: General 467 (2013) 421–429
429
moderate acidic sites which helped to maximize the yield of cyclo-
hexanone by cyclohexanol dehydrogenation.
[19] G. Ranga Rao, S.K. Meher, B.G. Mishra, P. Hari, K. Charan, Catal. Today 198 (2012)
40–147.
1
[
[
20] X. Li, B. Liang, J. Taiwan, J. Taiwan Inst. Chem. Eng. 43 (2012) 339–346.
21] A. Szegedi, M. Popova, K. Lazar, React. Kinet. Mech. Catal. 104 (2011)
291–301.
Acknowledgements
[
[
22] M. Popova, A. Szegedi, K. Lazar, A. Dimitrova, Catal. Lett. 9 (2011) 1288–1296.
23] G. Bai, H. Wang, H. Ning, F. He, G. Chen, React. Kinet. Catal. Lett. 94 (2008)
The authors express their gratitude to Dr C.V. Rode and Dr. V.V.
Bokade for helpful discussions and guidance. The assistance of Miss
Sharda Kondawar in catalyst characterization, in particular to BET
surface area and NH3 TPD measurements is gratefully acknowl-
edged.
375–383.
[24] C.H. Sivaraj, S.T. Srinivas, V.N. Rao, P.K. Rao, J. Mol. Catal. 60 (1990) L23–L28.
[
[
[
25] S. Nishimura, T. Shishido, K. Ebitani, K. Teramura, T. Tanaka, Appl. Catal. A: Gen.
78 (2010) 185–194.
26] M.W. Kasture, P.S. Niphadkar, N. Sharanappa, S.P. Mirajkar, V.V. Bokade, P.N.
Joshi, J. Catal. 227 (2004) 375–383.
3
27] P.S. Niphadkar, P.N. Joshi, S.S. Deshpande, V.V. Bokade, Appl. Catal. A: Gen. 401
(
2011) 236–240.
References
[
[
28] S.L. Burkett, M.E. Devis, Chem. Mater. 7 (1995) 920.
29] H. Bönnemann, G. Braun, W. Brijoux, R. Brinkmann, A.S. Tilling, K. Seevogel, K.
Siepen, J. Organomet. Chem. 520 (1996) 143.
[30] M.D. Fokema, E. Chiu, J.Y. Ying, Langmuir 16 (2000) 3154–3159.
[31] V.V. Paike, P.S. Niphadkar, V.V. Bokade, P.N. Joshi, J. Am. Ceram. Soc. 90 (2007)
3009–3012.
[32] J. Yang, S. Mei, J.M.F. Ferreira, J. Am. Ceram. Soc. 84 (2001) 1696–1702.
[33] R.B. Mane, A.M. Hengne, A.A. Ghalwadkar, S. Vijayanand, P.H. Mohite, H.S. Pot-
dar, C.V. Rode, Catal. Lett. 135 (2010) 141–147.
[34] P.A. Kumar, M.P. Reddy, L.K. Ju, B.H. Sook, H.H. Phil, J. Mol. Catal. A: Chem. 291
(2008) 66–74.
[35] A. Bialas, P. Niebryzydowska, B. Dudek, Z. Piwowarska, L. Chmielarz, M. Micha-
lik, M. Kozak, P. Kustrowski, Catal. Today 176 (2011) 413–416.
[36] R.O. Idem, N. Bakshi, Ind. Eng. Chem. Res. 33 (1994) 2047.
[37] M. Yin, C.K. Wu, Y. Lou, C. Burda, J.T. Koberstein, Y. Zhu, S. O’Brien, J. Am. Chem.
Soc. 127 (2005) 9506–9511.
[38] Y. Feng, H. Yin, L. Shen, A. Wang, Y. Shen, T. Jiang, Chem. Eng. Technol. 36 (2013)
73–82.
[39] L. Chen, T. Horiuchi, T. Osaki, T. Mori, Appl. Catal. B: Environ. 23 (1999) 259–269.
[40] M.C. Marion, E. Garbowski, M. Primet, J. Chem. Soc. Faraday Trans. 86 (1990)
3027–3032.
[
[
[
[
[
1] A. Romero, A. Santos, D. Escrig, E. Simon, Appl. Catal. A: Gen. 392 (2011) 19–27.
2] E. Simon, J.M. Rosas, A. Santos, A. Romero, Catal. Today 187 (2012) 150–158.
3] F.J. Brocker, M., Hesse, R. Markl, US Patent no. 6162758 (2000).
4] C.H. Sivaraj, P.K. Rao, Appl. Catal. 45 (1988) 103–114.
5] M. Popova, M. Dimitrov, V. Dal Santo, N. Ravasio, N. Scotti, Catal. Commun. 17
(
2012) 150–153.
6] H.F. Chang, M.A. Saleque, W. Su Hsu, W.H. Lin, J. Mol. Catal. A: Chem. 109 (1996)
49–260.
[
2
[
[
[
7] H.F. Chang, M.A. Saleque, Appl. Catal. A: Gen. 103 (1993) 233–242.
8] C.H. Sivaraj, B.M. Reddy, P.K. Rao, Appl. Catal. 45 (1988) L11–L14.
9] F.M.T. Mendes, M. Schmal, Appl. Catal. A: Gen. 151 (1997) 393–408.
[
[
[
[
[
[
[
10] G.S. Jeon, G. Seo, J.S. Chung, Korean J. Chem. Eng. 13 (1996) 642–646.
11] G.S. Jeon, G. Seo, J.S. Chung, Korean J. Chem. Eng. 12 (1995) 132–133.
12] G.S. Jeon, G. Seo, J.S. Chung, Korean J. Chem. Eng. 13 (1996) 412–414.
13] G.S. Jeon, J.S. Chung, Korean J. Chem. Eng. 14 (1997) 49–58.
14] D. Ji, W. Zhu, Z. Wang, G. Wang, Catal. Commun. 8 (2007) 1891–1895.
15] G.K. Reddy, P.K. Rao, Catal. Lett. 45 (1997) 93–96.
16] Y.K. Vyawahare, V.R. Chumbhale, S.A. Pardhy, V. Samuel, A.S. Avsar, Indian J.
Chem. Technol. 17 (2010) 43–49.
[
[
17] V.Z. Fridman, A.A. Davydov, J. Catal. 195 (2000) 20–30.
18] V.Z. Fridman, A.A. Davydov, K. Titievesky, J. Catal. 222 (2004) 545–557.
[41] H. Praliaud, S. Mikhailenko, Z. Chajar, M. Primet, Appl. Catal. B: Environ. 16
(1998) 359–374.