CuCr2O4 derived by the sol-gel method as a highly active and selective catalyst for the conversion of glycerol to 2,6-dimethylpyrazine: A benign and eco-friendly process

Article abstract of DOI:10.1039/c7cy00769h

Vapour phase dehydrocyclization of crude glycerol in conjunction with 1,2-propanediamine (1,2-PDA) was examined over CuCr₂O₄ obtained by different preparation methods. A high proportion of copper species interacted with Cr₂O₃ in CuCr₂O₄ derived from the sol-gel route with a low ratio of Cu²⁺/Cu⁰ demonstrating higher dehydrocyclization activity and 2,6-dimethylpyrazine (2,6-DMP) selectivity. X-ray photoelectron spectroscopy analysis of the reduced CuCr₂O₄ revealed a lower fraction of ionic Cu and a high percentage of metallic Cu in the near surface region. The HCOOH and pyridine adsorbed DRIFT spectra of CuCr₂O₄ revealed that strong basic and moderate Lewis acid sites are responsible for the selective formation of 2,6-dimethylpyrazine which is consistent with the catalyst poisoning studies on CuCr₂O₄ co-feeding with pyridine as both Br?nsted and Lewis acid site blocker and 2,6-lutidine as a selective Br?nsted acid site blocker during the dehydrocyclization reaction. The presence of isolated CuO and Cr₂O₃ species led to a high selectivity for 2,6-dimethylpiperazine. The high intrinsic activity of CuCr_sol-gel was also concomitant with the Cu metal surface areas of the catalysts. The fresh, reduced and some of the used catalysts are characterized by BET-surface area, powder XRD, FTIR, XPS, TEM, H₂-TPR, TPD of NH₃, pyridine, 2,6-dimethylpyridine and HCOOH adsorbed DRIFT spectroscopy.

Full text of DOI:10.1039/c7cy00769h

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: V. Akula, K.
Vankudoth, N. Gutta, V. K. Velisoju, S. mutyala and P. H. Aytam, Catal. Sci. Technol., 2017, DOI:
10.1039/C7CY00769H.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Volume 6 Number 1 7 January 2016 Pages 1–308
Catalysis
Science &
Technology
www.rsc.org/catalysis
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2044-4753
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
rsc.li/catalysis

Page 1 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
CuCr2O4 derived by solꢀgel method as highly active and selective catalyst for the conversion
of glycerol to 2,6ꢀdimethylpyrazine: A benign and ecoꢀfriendly process
Krishna Vankudotha,b, Naresh Guttab, Vijay Kumar Velisojub, Suresh Mutyalab, Hari
Padmasri Aytamc and Venugopal Akulaa,b,*
a Academy of Scientific and Innovative Research, India.
b Catalysis Laboratory, Inorganic and Physical Chemistry Division, CSIRꢀIndian Institute of
Chemical Technology, Hyderabad, Telangana – 500 007, India.
c Department of Chemistry, University College for Women, Osmania University, Koti,
Hyderabad ꢀ 500 095, Telangana, India.
*Corresponding author, Email: akula@iict.res.in Tel. +91ꢀ40ꢀ27193165;
Fax: +91ꢀ40ꢀ27160921
Abstract
Vapour phase dehydrocyclization of crude glycerol in conjunction with 1,2ꢀpropanediamine
(1,2ꢀPDA) was examined over CuCr2O4 obtained by different preparation methods. High
proportion of copper species interacted with Cr2O3 in CuCr2O4 derived from solꢀgel route
with a low ratio of Cu2+/Cu0 demonstrated higher dehydrocyclization activity and 2,6ꢀ
dimethylpyrazine (2,6ꢀDMP) selectivity. Xꢀray photoelectron spectroscopy analysis of the
reduced CuCr2O4 revealed lower fraction of ionic Cu and high percentage of metallic Cu in
the near surface region. The HCOOH and pyridine adsorbed DRIFT spectra of CuCr2O4
revealed strong basic and moderate Lewis acid sites are responsible for the selective
formation of 2,6ꢀdimethylpyrazine which is in consistent with the catalyst poisoning studies
on CuCr2O4 coꢀfeeding pyridine as both Brønsted and Lewis acid site blocker and 2,6ꢀlutidine
as a selective Brønsted acid site blocker during the dehydrocyclization reaction. Presence of
isolated CuO and Cr2O3 species led to a high selectivity of 2,6ꢀdimethylpiperazine. The high
1

Catalysis Science & Technology
Page 2 of 28
View Article Online
DOI: 10.1039/C7CY00769H
intrinsic activity of CuCrsolꢀgel was also concomitant with Cu metal surface areas of the
catalysts. The fresh, reduced and some of the used catalysts are characterized by BET –
surface area, powder XRD, FTꢀIR, XPS, TEM, H2ꢀTPR, TPD of NH3, pyridine, 2,6ꢀ
dimethylpyridine and HCOOH adsorbed DRIFT spectroscopy.
1.0 Introduction
Conversion of crude glycerol obtained from bioꢀdiesel plants for the synthesis of fine
chemicals and fuel additives by catalytic routes is very important from costꢀeffective and
environmental perspectives.1 It has also been found that glycerol as byꢀproduct during
biomass conversion to valuable products, such as hydrogenolysis of sugars to propylene
glycol.2 Plenty of literature available during the past one decade on the crude glycerol in to
useful chemicals as it is a rich molecule with three –OH functional groups.3 Solid acidꢀbase
catalysts are identified as activity and selective dehydrocyclization catalysts in the conversion
of crude glycerol and alkylꢀdiamines to industrially important alkylpyrazines under moderate
reaction conditions with minimized waste and ease of handling.4ꢀ15 Among numerous
alkylpyrazines, the 2,6ꢀdimethypyrazine (2,6ꢀDMP) is one of the valuable compounds for the
production of various agroꢀchemicals, food flavoring agent and as a ligand for catalyst
synthesis.16
The concept of crude glycerol conversion to 2ꢀmethylpyrazine is reported by us and in
the similar fashion synthesis of 2,6ꢀDMP was first reported by Li et al over Zn/Al2O3
catalyst.7,17 The mixed oxides of metal chromites were examined for the dehydrocyclization
of crude glycerol and 1,2ꢀpropanediamine (1,2ꢀPDA) for the synthesis of 2,6ꢀDMP.11
Selective synthesis of alkylpyrazines is challenging task due to multiple reactions are quite
possible during the course of dehydrocyclization. The metal chromite catalysts are known to
be active for the hydrogenation and dehydrocyclization processes particularly in the
conversion of biomass derived compounds to highly useful chemicals and fuel additives.18,19
2

Page 3 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
It has been observed that dehydrocyclization activity is dependent upon the strength and type
of acidꢀbase (both Bronsted and Lewis) sites present on the catalyst surface.20,21 The focal
point of this work is to examine the dehydrocyclization activities under catalyst poisoning
conditions to emphasize the role of surface Lewis acid sites and basic sites which were
responsible for the selective formation of 2,6ꢀdimethylpyrazine.
The objectives of this study are fourfold: firstly to examine the activity of bulk CuO
and Cr2O3 for the dehydrocyclization of crude glycerol and 1,2ꢀPDA and comparison of
activities with CuCr2O4 under kinetic regime. Secondly, correlate copper metal surface area
(measured by N2O titration) in relation to dehydrocyclization activity and product selectivity.
Thirdly, comparative studies between the dehydrocyclization activities (of CuCr2O4 obtained
by various preparation techniques) with the nature, type and strength of Bronsted and Lewis
acidꢀbase characteristics which are examined by diffuse reflectance infrared Fourier
transform spectroscopy (DRIFTS). Finally, pyridine and 2,6ꢀdimethylpyridine are used as
probe molecules for self poisoning the CuCr2O4 catalyst surface during the
dehydrocyclization reaction. The formic acid adsorbed DRIFT spectroscopic data and the
N2O pulse chemisorption results demonstrated that a high ratio of surface basic sites on
CuCr2O4 surface (derived from solꢀgel technique) than on CuCr2O4 that is obtained by a coꢀ
precipitation method.
2.0 Materials and methods
2.1 Preparation of catalysts
The experimental conditions pertaining to the preparation of CuCr2O4 (by different
methods) and the bulk CuO and Cr2O3 are detailed in the electronic supplementary
information. The CuCr2O4 spinel was synthesized by sol–gel method using required amounts
of Cu(NO3)2.3H2O and Cr(NO3)3.9H2O (at a desired mole ratio of Cu/Cr = 0.5).22 The
CuCr2O4 sample was also prepared by coꢀprecipitation method using Cu(NO3)2.3H2O and
3

Catalysis Science & Technology
Page 4 of 28
View Article Online
DOI: 10.1039/C7CY00769H
Cr(NO3)3.9H2O with Cu : Cr = 1: 2 (mole ratio) at a constant pH ∼ 8 denoted as CuCrcp. For
comparison purpose CuCr2O4 was synthesized by mechanical mixing of CuO and Cr2O3 solid
powders with a Cu:Cr = 1:2 mole ratio using ethanol for wetting which is designated as
CuCrmm and copper impregnated on Cr2O3 prepared by simple incipient wet impregnation
method which was designated as CuCrimp (ESI).
2.2 Characterisation of catalysts
The experimental conditions adopted for these adsorption and spectroscopic methods
were put in to the Electronic Supplementary Information (ESI).
2.3 Dehydrocyclization activity measurements
The vapour phase dehydrocyclization of aqueous glycerol and 1,2ꢀPDA was carried
out in a fixed bed down flow quartz reactor at atmospheric pressure in the temperature 375
oC, under kinetic regime. The reactor dimensions and the complete details on experimental
conditions maintained for the activity studies were put in to the Electronic Supplementary
Information.
3.0 Results and Discussion
3.1 BET-surface area and powder XRD analysis of CuCr2O4 prepared by different
preparation methods
The CuCr2O4 obtained by solꢀgel technique; showed higher BET surface area than the
CuCr2O4 prepared by coꢀprecipitation (CuCrcp) and mechanical mixing (CuCrmm) methods
(Table 1). The calcined CuCrsolꢀgel, CuCrcp and CuCrmm samples exhibited strong diffraction
peaks assigned to the copper chromite spinel and (JCPDS # 72ꢀ1212) copper impregnated on
chromium oxide (CuCrimp) displayed diffraction lines due to isolated CuO (JCPDS # 80ꢀ
1268) and Cr2O3 (JCPDS # 85ꢀ0730) phases (Figure 1A and 1B). The copper chromite spinel
phase is absent in CuCrimp sample (Figure 1A). The XRD patterns of reduced (in 5%H2/Ar at
450 °C/3 h) CuCrsolꢀgel, CuCrcp, and CuCrmm samples (Figure 1B) showed sharp diffraction
4

Page 5 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
peaks attributed metallic copper and weak signals of CuCr2O4 spinel phase. Presence of
CuCr2O4 exemplifies that the CuCr2O4 spinel undergone partial reduction at 450 °C
temperature. The reduced CuCrimp sample showed isolated metallic copper (JCPDS # 04ꢀ
0836) and Cr2O3 phases. Figure S1 shows the TEM images of calcined CuCrsolꢀgel, CuCrcp,
CuCrmm and CuCrimp samples. The CuOꢀCr2O3 mixed oxides aggregate together and form
agglomerates with an average particle size of 17 nm, 25.9 nm and 30.7 nm in CuCrsolꢀgel
,
CuCrcp and CuCrmm samples respectively (Figure S1aꢀc). The CuCrimp sample (Figure S1d)
showed a particle size of 41.9 nm could be attributed to aggregates of CuO and Cr2O3
together with low dispersion.
3.2 H2 - temperature programmed reduction (H2-TPR) analysis
The reducibility of CuCr2O4 prepared by different preparation techniques are
examined by H2ꢀTPR (Figure 2) and the H2 uptakes are reported in Table 1. All the samples
undergo single stage reduction, except CuCrimp. The Tmax of 385, 426 and 438 °C are due to
partial reduction of CuCr2O4 phase in CuCrsolꢀgel, CuCrcp, and CuCrmm respectively.23 The
CuCrimp catalyst showed two distinct peaks at 263 °C and 376 °C; the low temperature peak is
attributed to reduction of copper oxide to metallic copper (Cu2+ → Cu0) and the high
temperature peak may be either contributed from CuO interacted with chromium oxide or the
Crx+ in higher oxidation state of 3< and <6 in the form of chromate species.24ꢀ26 These results
are in good correlation with XPS results (section 3.4) in the transformation of chromium with
high oxidation to lower oxidation (Figure S4 and 4b). When compared the reduction profiles;
CuCrimp sample began to undergo reduction at lower temperature, suggesting a weak or no
interaction between copper and chromium species. In contrast, the reduction of CuCrsolꢀgel
,
CuCrcp and CuCrmm samples occurred in a single stage indicating that the reducible species
are present in a single phase structure, which is consistent with the XRD data (Figure 1A).
Differences in the reducibility of CuCr2O4 emphasize the variations in surface and bulk
5

Catalysis Science & Technology
Page 6 of 28
View Article Online
DOI: 10.1039/C7CY00769H
properties of the spinels during the synthesis e.g. particle size of the metal oxide clusters.
Reduction of finely dispersed metal oxide particles occurs at low temperatures.27 The high
temperature reduction of Cu2+ species is due to the strong interaction between Cu2+ and Cr3+
species in a CuCr2O4 spinel. On the other hand the segregated Cu2+ and Cr3+ species in
CuCrimp are reduced at lower temperatures. The hydrogen uptakes are in the following order:
CuCrsolꢀgel > CuCrcp > CuCrmm. The low temperature reduction signal in CuCrsolꢀgel sample can
be explained by the presence of small size CuCr2O4 particles compared to CuCr2O4 derived
from coꢀprecipitation and mechanical mixing techniques. Reduction of trivalent chromium to
zero valent chromium is improbable under the TPR experimental conditions.28 These results
are in good agreement with the XRD data of the reduced samples that showed the presence of
metallic copper and CuCr2O4 phase.
3.3 Fourier transform infrared spectroscopy (FT-IR)
The FTIR spectra (Figure 3) showed bands at 614 cmꢀ1 and 513 cmꢀ1 in CuCrsolꢀgel
,
CuCrcp, CuCrmm, are originated from CrꢀO and CrꢀOꢀCu bands in the CuCr2O4 spinel (Figure
3A).29ꢀ31 These bands are very sharp in CuCrsolꢀgel compared to other samples. The vibrational
bands found at 630 cmꢀ1 and 570 cmꢀ1 in CuCrimp sample is originated from Cr–O in
crystalline Cr2O3.32 Additionally the sharp band appeared at 415 cmꢀ1 corresponds to CuO.33
Absence of the bands at 614 cmꢀ1 and 513 cmꢀ1 in CuCrimp clearly indicates that copper
chromite spinel is not formed when copper impregnated on Cr2O3 (Figure 3A). The FTIR
spectra demonstrated the isolated copper (415 cmꢀ1) and chromium oxide (630 cmꢀ1) in
CuCrimp sample which is in good agreement with XRD analysis (Figure 1). While compared
to fresh; the reduced form of the samples did not show the fundamental band at 934 cmꢀ1
region (Figure 3B) ascribed to the reduction of ionic chromium species in oxidation state
greater than 3 and less than 6 in the form of chromates.31 However, the CuCr2O4 spinel phase
(513 and 614 cmꢀ1) was retained to some extent even after reduction in all the samples except
6

Page 7 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
CuCrimp. From these results it can be concluded that the spinel structure is partially reduced
under the reduction conditions applied.
3.4 X-ray photoelectron spectroscopy
XPS analyses of reduced CuCr2O4 samples are reported in Figure 4 and the relative
ratios of Cu0 and Cu2+ are summarized in Table 2. The XP spectra of Cu 2p show BE at 932.4
± 0.3 eV and 935.2 ± 0.4 eV which are corresponding to Cu0 and Cu2+ species respectively.34
The Cu0 signal is dominant in their spectra, indicating ~70ꢀ80% of surface copper is in
reduced state except in CuCrimp that showed 100% Cu0 which otherwise did not form the
spinel structure. However, Williams and coꢀworkers comprehensively studied the influence
of reduction temperature on CuCr2O4 using Auger electron spectroscopy which revealed the
formation of Cu0 and Cu+ species.35,36
The shoulder peak around 935 eV is originated from Cu2+ in CuCr2O4 spinel.37
Absence of Cu2+ signal in the reduced CuCrimp exemplify that the surface Cu2+ are
completely reduced to Cu0. The copper and chromium species seem to be strongly bonded in
spinel structure as certain portion of CuCr2O4 even after the reduction process. The CuCrmm
demonstrated a strong shift in Cu2p3/2 towards higher binding energies indicating a high
proportion of ionic Cu on the catalyst surface compared to CuCrsolꢀgel and CuCrcp samples
(Table 2). These results are in exact accordance with the N2O titration data (Table 1) since
the N2O decomposition occurs on metallic Cu species present on the catalyst surface.
Furthermore, the relative ratio of Cu2+/Cu0 is lower on CuCrsolꢀgel (0.27) when compared to
CuCrcp (0.39) and CuCrmm (0.47) samples and is almost nil in CuCrimp in which the CuCr2O4
spinel was absent; emphasizing a large extent of Cu2+ reduction in CuCrsolꢀgel under the
reduction conditions adopted. This finding is in agreement with H2 uptakes measured by TPR
analysis (Table 1). Though, TPR is a bulk technique, a similar surface structure may be
expected at the near surface region analysed by XPS. From the bulk (XRD, H2ꢀTPR) and
7

Catalysis Science & Technology
Page 8 of 28
View Article Online
DOI: 10.1039/C7CY00769H
surface (XPS) analyses it is inferred that the extent of CuCr2O4 reduction is dramatically
changed by the method of preparation and varied cluster size of CuCr2O4 are formed (Figures
1ꢀ3) with different Cu2+/Cu0 ratios. In the comparative analysis the surface metallic copper is
found to be CuCrsolꢀgel > CuCrcp > CuCrmm; which is in good correlation with N2O titration
data. Binding energy of the Cr 2p3/2 is observed at 576.6 eV confirming that Cr3+ species are
present on the catalyst surface (Figure 4b).8,10,38,39 The O 1s spectra (Figure 4c) shows broad
signal at a BE of 530.2 eV in all the CuCr samples except the CuCrimp. The deconvoluted O
1s spectra exhibited a shoulder at high energy region indicating oxygen in two different
environments, in the form of Cr2O3 and CuO.
3.5 Evaluation of surface acidic and basic sites by DRIFTS studies
The NH3ꢀTPD patterns of the reduced CuCr2O4 samples prepared by different routes
are reported in Figure 5. The surface acidity of CuCr samples is mainly contributed from
chromium species present on the surface.40 In comparison, the CuCrsolꢀgel showed higher
acidity (5.6 ꢁmol gꢀ1) than other three samples prepared by different techniques (Table 1).
The CuCrimp catalyst exhibited lower number of acid sites than the other samples. A lower
acidity on CuCrimp probably due to surface coverage of Cr2O3 by copper particles, as the
acidity of the catalysts is mainly contributed from surface chromium. Overall the CuCrsolꢀgel
demonstrated higher number of moderate and strong acid sites when compared to others.
The FTIR spectra of formic acid adsorbed on CuCrsolꢀgel, CuCrcp, CuCrmm and CuCrimp
samples are illustrated in Figure 6. Due to broad IR signals the spectra were deconvoluted
using Gaussian fitting. The DRIFT spectra show IR absorption band at 1677 (shoulder),
1620, 1584 and 1352 cmꢀ1, which are due to the ν(CH), νas(COO), δ(CH) and νs(COO)
modes of surface formate groups.41 The 1752 cmꢀ1 and 1620 cmꢀ1 bands are assigned to
associative adsorption of formic acid on the weak and strong basic sites present on the
catalyst surface respectively (Figure 6).41,42 The vibrational band centred at 1284 cmꢀ1 is
8

Page 9 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
assigned to formic acid adsorption on CuCr2O4 surface.24 The CuCrsolꢀgel sample showed a
very strong band (at 1620 cmꢀ1) when compared to other samples even though an equimolar
dose of formic acid is injected during the DRIFTS analysis. The basic sites are expected from
the reduced copper species upon exposure to steam at and above 300 °C.43 The copper metal
surface areas estimated by N2O titration revealed a higher N2O uptake on CuCrsolꢀgel
compared to others (Table 1). A rough estimation of the normalized DRIFT spectra showed
the peak intensity of the band at 1620 cmꢀ1 is proportional to the Cu metal surface areas of the
catalysts.
3.6 Dehydrocyclization activity measurements
Table 3 shows the catalytic activity and product selectivity on CuCrsolꢀgel, CuCrcp,
CuCrmm and CuCrimp catalysts in the dehydrocyclization of aqueous glycerol and 1,2ꢀPDA.
Among all, the CuCrsolꢀgel catalyst demonstrated considerably better 2,6ꢀDMP yield than the
other CuCr2O4 catalysts; while the bulk CuO showed a lower 2,6ꢀDMP rate of an order of
magnitude and ~20 times lower on Cr2O3 (Table 3).
The bulk chromium oxide showed high selectivity towards 2,6ꢀdimethylpiperazine
(2,6ꢀDMPip) (Table 3). High ratio of 2,6ꢀDMPip on bulk Cr2O3 can be explained by the
primary reaction step i.e. dehydration of glycerol to hydroxyacetone (HA) on Cr2O3 and
subsequent condensation of HA with 1,2ꢀPDA to produce 2,6ꢀDMPip (Scheme 1). It appears
that the rate of dehydrogenation of 2,6ꢀDMPip to generate 2,6ꢀDMP may be low on CuCrimp
because of weak basicity (Figure 6) as a result of it slightly higher 2,6ꢀDMpip selectivity is
observed compared to other samples which showed CuCr2O4 phase. A small fraction of side
products were also found over all the catalysts mainly composed of 2,5ꢀdimethylpyrazine
(2,5ꢀDMP) and 5ꢀhydroxymethylpyrazine (5ꢀHMP). Using a simulated crude glycerol
mixture (expected during the bioꢀdiesel production); the CuCrsolꢀgel has shown promising
activity than the other CuCr samples (Table 3). Formation of 2,6ꢀDMP is occurred by
9

Catalysis Science & Technology
Page 10 of 28
View Article Online
DOI: 10.1039/C7CY00769H
dehydration of glycerol to hydroxyacetone (HA) followed by cyclocondensation of HA with
1,2ꢀPDA (Scheme 1).
The effect of gas hourly space velocity (GHSV) on conversion of glycerol and 1,2ꢀ
PDA is studied in the range of 11.04 to 35. 24 mL gcatꢀ1 sꢀ1, at 375 °C and the results are
reported in Table 4. Upon increasing the space velocity both the glycerol and 1.2ꢀPDA
conversions are decreased which is probably due to reduction in residence time of substrates.
At lower GHSV, catalyst particles were partially wetted; therefore, there would be direct
contact of reactant and catalyst surface as result an increase the reaction rate is observed.
Subsequently at higher GHSV, the increased wetted fractions of catalyst surface would slow
down the reaction rate. However, the selectivity of 2,6ꢀDMP is maintained around 85% even
at very high space velocity.
The time on stream analysis data over CuCrsolꢀgel sample showed a consistent activity
(Figure S2) with a marginal decrease in activity after 10 h on stream. The decrease of
catalytic activity may probably be due to coke deposition on the catalyst surface over a period
of time. However, the carbon deposition on the used catalyst estimated by CHNS analysis
showed less than 3wt% of carbon after 10 h of continuous operation (Table S4). The
difference in catalytic activity is explained based on the nature of active sites present on the
catalyst surface. The XRD, FTꢀIR and XPS data showed segregated copper and chromium
oxides in the CuCrimp and only CuCr2O4 spinel in the remaining samples.
The selectivity of 2,6ꢀDMP is higher on sample containing copper chromite spinel
phase compared to the isolated CuO and Cr2O3 species on CuCrimp sample. For comparative
analysis, we have collected DRIFT spectra of CuCrsolꢀgel and CuCrcp samples and both these
samples exhibited Lewis (LAS) and Brønsted acid sites (BAS) (Figure 7a, 7b). The ratios of
the normalized relative peak intensities of Lewis (1445 cmꢀ1) and Brønsted acid sites (1558
cmꢀ1) are found to be BAS:LAS = 1.3 and 2.1 over CuCrsolꢀgel CuCrcp catalysts respectively.
10

Page 11 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
The existence of high proportion LAS on CuCrsolꢀgel than on CuCrcp sample is observed. The
weak to moderate acid sites in the temperature region 300ꢀ450 °C are found to be higher on
CuCrsolꢀgel compared to CuCrcp sample (Figure 5). We believe that these moderate acid sites
are attributed to Lewis and the high temperature NH3 desorption peaks may probably be due
to Bronsted acid sites on the CuCr2O4 surface. Quite contrast to this the CrCrimp sample upon
reduction showed exclusively metallic Cu however; the active surface area of Cu is very low
compared to others. Hence the CuCrimp sample exhibited very weak basic sites compared to
other samples (Figure 6).
The peak intensities assigned to Bronsted (1558 cmꢀ1) and Lewis (1445 cmꢀ1) acid
sites on CuCrsolꢀgel (Figure 7a) and CuCrsolꢀgel (Figure 7c) are varied to a great extent. The
relative ratios of the band intensities due to 1445 cmꢀ1 and 1558 cmꢀ1 revealed a high surface
coverage of 2,6ꢀdimethylpyridine adsorption on Bronsted acid sites (~80%) indicating a self
consistency in our analysis with respect to the pyridine and 2,6ꢀdimethylpyridine adsorption
on CuꢀCrsolꢀgel sample.40 In order to assess the role of Bronsted and Lewis acid sites on the
dehydrocyclization activity; an investigation is carried out using two different probe
molecules while coꢀfeeding along with the aqueous glycerol and 1,2ꢀPDA such as pyridine as
both Brønsted and Lewis acid site blocker and 2,6ꢀdimethylpyridine (as a selective Brønsted
acid site blocker) (Figure 7). An equimolar concentration of both pyridine (19.8 mmol) and
2,6ꢀdimethylpyridine (19.8 mmol) was introduced in to the reaction stream in (8 equal)
successive pulses in separate experiments. The activity data obtained after the addition of
pyridine and/or 2,6ꢀdimethylpyridine is reported in Table 5. It shows that the conversion of
glycerol is drastically decreased to 21.4% from 72.5% when pyridine is coꢀfed along with
aqueous glycerol and 1,2ꢀPDA stream, but the 2,6ꢀDMP selectivity is not effected (S2,6ꢀDMP
=
90.1%) (Table 5). On the other hand using 2,6ꢀdimethylpyridine the glycerol conversion is
marginally decreased (Xglycerol = 57.6%). A drastic decrease in conversion by pyridine
11

Catalysis Science & Technology
Page 12 of 28
View Article Online
DOI: 10.1039/C7CY00769H
poisoning results pointed out that Lewis acid sites are responsible for dehydration of glycerol
in the first step (Scheme 1). These results clearly indicate that there is no influence of
Brønsted acid sites on conversion of glycerol and 2,6ꢀDMP selectivity, because 2,6ꢀ
dimethylpyridine would selectively block the surface Brønsted acid sites (Figure 7 ).
Dehydrogenation reactions occur on basic sites present on the catalyst surface. The
surface basic sites on these CuCr samples (prepared by different methods) are analysed by
formic acid adsorbed DRIFT spectroscopy (Figure 6). The relative ratios pertaining to the
formic acid adsorbed IR bands appeared at 1284 cmꢀ1 attributed to formic acid adsorption on
CuCr2O4 and the band at 1620 cmꢀ1 (dissociative formic acid adsorption band) indicated that a
high proportion of CuCr2O4 species (I1284/I1620 = 0.3) on the CuCrsolꢀgel sample compared to
CuCrcp (I1284/I1620 = 0.11) and CuCrmm (I1284/I1620 = 0.095). It is further confirmed the absence
of CuCr2O4 species in the formic acid adsorbed IR spectra of CuCrimp at 1284 cmꢀ1 (Figure
6d). Hence it can be concluded that selective synthesis of alkyl hetero aromatics by vapour
phase dehydrocyclization of glycerol and 1,2ꢀPDA requires a pair of strong basic sites and
moderate Lewis acid sites for the selective formation of 2,6ꢀDMP. Finally, surface
modification of CuCr2O4 by varying the acidꢀbase strengths, one can fine tune the product
distribution in order to obtain various dialkylꢀpyrazines in the dehydrocyclization of aqueous
glycerol and 1,2ꢀPDA e.g. 2,6ꢀdimethylpiperazine.
4.0 Conclusions
The present study is aimed to examine the CuCr2O4 obtained by different preparation
techniques and the catalytic activities were rationalized by surface acidꢀbase characteristics
using DRIFTS studies in conjunction with copper metal surface areas estimated by N2O
titration technique. All the samples showed CuCr2O4 phase except in the impregnated sample,
which was confirmed by XRD, FTꢀIR, XPS, TEM analysis. The N2O decomposition studies
revealed a higher copper metal surface area on CuCr2O4 obtained from solꢀgel technique
12

Page 13 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
compared to coꢀprecipitation and mechanical mixing methods. The Cu impregnated on Cr2O3
exhibited segregated CuO and Cr2O3 species that showed a high selectivity of 2,6ꢀ
dimethylpiperazine. The solꢀgel derived CuCr2O4 catalyst was stable when subjected to a
crude simulated glycerol obtained from the bioꢀdiesel process. The XPS analysis of reduced
samples demonstrated that CuCr2O4 undergone partial reduction indicating the formation of
both metallic and ionic copper species. The TPD of NH3 results showed both moderate and
strong acid sites present on the CuCrsolꢀgel catalyst surface. The self poisoning studies
emphasized the importance of Lewis acid sites and the formic acid DRIFT spectroscopy
demonstrated the role of strong basic sites in the selective formation of 2,6ꢀDMP. Finally it
can be concluded that fine tuning of surface acidꢀbase sites would give us the desired
compounds in the dehydrocyclization of aqueous glycerol and 1,2ꢀPDA.
Acknowledgements: The authors AV and MS sincerely thank DST New Delhi for funding
through IndiaꢀAustralia (AISRF program). One of the authors VK thanks the CSIR New
Delhi for the award of fellowship.
References
1 C. H. Zhou, J. N. Beltramini, Y. X. Fan and G. Q. Lu, Chem. Soc. Rev., 2008, 37, 527–549.
2 A. Corma, G. W. Huber, L. Sauvanaud and P. O’Connor, J. Catal., 2008, 257, 163–171.
3 A. D. Mohanprasad, P. Kiatsimkul, W. R. Sutterlin and G. J. Suppes, Appl. Catal. A: Gen.,
2005, 281, 225–231.
4 P. S. Kong, M. K. Aroua and W. M. A. W. Daud, Renewable and Sustainable Energy
Reviews, 2016, 63, 533–555.
5 U. Apparao, S. Vijayalakshmi and J. Ranjitha, International Journal of ChemTech
Research, 2016, 9(4), 576ꢀ586.
13

Catalysis Science & Technology
Page 14 of 28
View Article Online
DOI: 10.1039/C7CY00769H
6
L. Forni, J. Catal., 1988, 111, 199ꢀ209.
7 R. Sarkari, Ch. Anjaneyulu, V. Krishna, R. Kishore, M. Sudhakar and A. Venugopal,
Catal. Commun., 2011, 12, 1067–1070.
8 A. Venugopal, R. Sarkari, Ch. Anjaneyulu, V. Krishna, M. Kotesh Kumar, Appl.
Catal. A: Gen., 2012, 441– 442, 108–118.
9 A. Venugopal, R. Sarkari, Ch. Anjaneyulu, V. Krishna, M. Kotesh Kumar, N. Narender
and A. H. Padmasri, Appl. Catal. A: Gen., 2014, 469, 398– 409.
10 A. Venugopal, R. Sarkari, S. Naveen Kumar, M. Kotesh Kumar, S. Syed John, J. Krishna
Reddy and A. H. Padmasri, J. Chem. Sci., 2014,126, 387–393.
11 V. Krishna, S. Naveen Kumar, S. Reema, A. Hari Padmasri, K. V. R. Chary and A.
Venugopal, Appl. Catal. A: Gen., 2014, 488, 275–284.
12 G. W. H. Cheesman and E. S. G. Werstiuk, Adv. Heterocyc. Chem., 1972, 14, 99‒209.
13 T. Rizk, E. J. F. Bilodeau and A. M. Beauchemin, Angew. Chem. Int. Ed., 2009, 48,
8325‒8327.
14 G. Buchi and J. Galindo, J. Org. Chem., 1990, 56, 2605‒2606.
15 V. Krishna, G. Naresh, S. Naveen Kumar, J. Beltramini, A. H. Padmasri and A.
Venugopal, Current Catalysis 2017, 6, 135ꢀ143.
16 J. J. Brophy and G. W. K. Cavill, Heterocycles, 1980, 14, 477‒504.
17 X. Li, C. Xu, C. Liu, N. Zhang, J. You and Q. Wu, J. Mol. Catal. A, 2013, 371,104‒110.
18 D. Liu, D. Zemlyanov, T. Wu, R. J. L. Lapidus, J. A. Dumisic, J. T. Miller and C. L.
Marshall, J. Catal., 2013, 299, 336ꢀ345.
19 K. L. Deutsch and B.H. Shanks, J. Catal., 2012, 285, 235ꢀ241.
20 L. Li, T. I. Koranyi, B. F. Sels and P. P. Pescarmona, Green Chem., 2012, 14, 611‒1619.
21 V. Krishna, G. Naresh, V. Vijay Kumar, R. Sarkari, A. Hari Padmasri and A. Venugopal,
Appl. Catal. B: Environ., 2016, 193, 58–66.
22 Z. Ma, Z. Xiao, J. A. van Bokhove and C. Liang, J. Mater. Chem., 2010, 20, 755–760.
14

Page 15 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
23 A.Venugopal, J. Palgunadi, J. K. Deog, O. S. Joo and C. H. Shin, Catal. Lett., 2008, 123,
142‒149.
24 Y. J. Tu, Y. W. Chen and L. Chiuping, J. Mol. Catal., 1994, 89, 179–190.
25 O. V. Shutkina, O. A. Ponomareva, P. A. Kots and I. I. Ivanova, Catal. Today, 2013,
218‒219, 30‒34.
26 D. Li, Q. Yu, S. S. Li, H. Q. Wan, L. J. Liu, L. Qi, B. Liu, F. Gao, L. Dong and Y. Chen,
Chem. Eur. J., 2011, 17, 5668‒5679.
27 Y. S. Xia, H. X. Dai, L. Zhang, J. G. Deng, H. He and C. T. Au, Appl. Catal. B: Environ.,
2010, 100, 229‒237.
28 A. Wollner, F. Lange, H. Schmelz and H. Knozinger, Appl. Catal. A: Gen., 1993, 94,
181‒203.
29 B. Dragoi, A. Ungureanu, A. Chirieac, V. Hulea, S. Royer and E. Dumitriu, Catal. Sci.
Technol., 2013, 3, 2319‒2329.
30 R.D. Waldron, Phy. Rev., 1955, 99, 1727‒1735.
31 W.B. White and B. A. Deangelis, Spectrochim. Acta, 1967, 25, 985‒993.
32 R. Bajaj, M. Sharma, D. Bahadur, Dalton Trans., 2013, 42, 6736‒6744.
33 M. A. Khafri and M. H. Kakaei Lafdani, Powder Technol., 2012, 222, 152‒159.
34 G. F. Zou, H. Li, D. W. Zhang, K. Xiong, C. Dong and Y.T. Qian, J. Phys. Chem. B,
2006, 110, 1632‒1637.
35 Z. Xiao, X. Wang, J. Xiu, Y. Wang, C.T. Williams and C. Liang, Catal. Today 2014, 234,
200ꢀ207.
36 Z. Xiao, C. Li, J. Xiu, X. Wang, C.T. Williams and C. Liang, J. Mol. Catal. 2012, 365,
24ꢀ31.
37 R. B. Mane and C. V. Rode, Green Chem., 2012, 14, 2780‒2789.
38 N. D. Kim, J. R. Park, D. S. Park, B. K. Kwak and J. Yi, Green Chem., 2012, 14,
15

Catalysis Science & Technology
Page 16 of 28
View Article Online
DOI: 10.1039/C7CY00769H
2638‒2646.
39 J. J. BravoꢀSuarez, B. Subramaniam and R. V. Chaudhari, J. Phy. Chem., 2012, 116,
18207‒18221.
40 J. Li, X. Liang, S. Xu and J. Hao, Appl. Catal. B: Environ., 2009, 90, 307‒312.
41 C. Martin, I. Martin and V. Rives, J. Mol. Catal., 1992, 73, 51‒63.
42 J. C. Lavalley, Catal. Today, 1996, 27, 377‒401.
43 M. Zhang, G. Li, H. Jiang and J. Zhang, Catal. Lett., 2011, 141, 1104–1110.
List of Table and Figure captions
Table 1. Physicochemical characteristics of the CuCr2O4 spinels prepared by different
methods.
Table 2. Relative ratios of Cu0 and Cu2+ (measured from XPS analysis) over reduced CuCr
samples.
Table 3. Dehydrocyclization of aqueous glycerol and 1,2ꢀPDA over reduced CuCr catalysts;
catalyst weight ~0.1g; reaction temperature = 375 °C; GHSV at 11.0 mL gcatꢀ1 sꢀ1.
Table 4. Effect of GHSV on dehydrocyclization activity of CuCrsolꢀgel catalyst at a reaction
temperature = 375 °C; catalyst weight 0.1g.
Table 5. Influence of probe (pyridine and 2,6ꢀdimethylpyridine) addition on product
distribution. Reaction conditions: 10 wt% aq. glycerol and 1,2ꢀPDA; reaction temperature:
375 °C; catalyst wt: 0.1g, carrier N2 = 30 cc minꢀ1, GHSV at 11.04 mL gcatꢀ1 sꢀ1.
Figure 1. XRD patterns of (A) calcined and (B) reduced samples.
Figure 2. TPR patterns of CuCr2O4 obtained by different preparation techniques.
Figure 3. FTIR pattern of (A) calcined and (B) reduced CuCr2O4 samples.
Figure 4. XPS analysis of reduced CuCr2O4 samples.
Figure 5. NH3ꢀTPD patterns of reduced CuCr2O4 samples.
16

Page 17 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
Figure 6. Formic acid adsorbed DRIFT spectra of (a) CuCrsolꢀgel (b) CuCrcp (c) CuCrmm and
(d) CuCrimp (e) Cr2O3 and (f) CuO samples.
Figure 7. Pyridine adsorbed on (a) CuCrsolꢀgel (b) CuCrcp and (c) 2,6ꢀdimethylpyridine
adsorbed DRIFT spectra of CuCrsolꢀgel samples.
Scheme 1. Proposed reaction mechanism for the formation of 2,6ꢀdimethylpyrazine during
dehydrocyclization of glycerol and 1,2ꢀpropanediamine over CuCr2O4 catalyst.
17

Catalysis Science & Technology
Page 18 of 28
View Article Online
DOI: 10.1039/C7CY00769H
Table 1. Physicochemical characteristic of the CuCr2O4 spinels prepared by different
methods.
Samples aBETꢀSA
bSCu
Crystallite size (nm)
cCuCr2O4
dParticle Uptake (µmol gꢀ1)
size (nm)
N2O uptake
ꢀ1
(m2 g−1)
(m2 gꢀ1)
(ꢁmol gcat
49.2
)
eNH3
5.6
fH2
CuCrsolꢀgel
CuCrcp
38.7
21.3
17.4
8.2
4.10
3.55
2.92
0.90
12.3
17.3
253.4
15.9
17.4
nf
25.9
30.7
41.9
3.7
2.7
1.6
227.1
209.5
203.2
43.0
35.4
11.6
CuCrmm
CuCrimp
a BET–surface areas of the fresh calcined catalysts. b Copper metal surface area measured by N2O decomposition
studies. c Crystallite size measured from XRD patterns of calcined CuꢀCr samples using Scherrer formula. d Average
particle size measured from TEM images. e NH3 uptakes on reduced CuCr samples measured from NH3ꢀTPD. f H2
uptakes measured by TPR analysis calibrated with a standard TPR of Ag2O. nf: not found.
Table 2. Relative ratios of Cu0 and Cu2+ (measured from XPS analysis) over reduced CuCr
samples.
Samples
Binding energy (eV) of 2p3/2
XPS peak area (%)
Cu2+/Cu0
0.277
Cu0
Cu2+
Cr3+
Cu0
Cu2+
CuCrsolꢀgel
CuCrcp
932.20
935.36
576.30
78.30
21.72
576.27
576.24
576.24
932.34
932.47
932.17
935.41
935.62
ꢀ
71.52
68.06
100
28.48
31.94
0.00
0.398
0.469
0.00
CuCrmm
CuCrimp
Table 3. Dehydrocyclization of aqueous glycerol and 1,2ꢀPDA over reduced CuCr catalysts;
catalyst weight ~0.1 g; reaction temperature = 375 °C; GHSV = 11.0 mL gcatꢀ1 sꢀ1.
br2,6ꢀDMP /10ꢀ8 mol
Catalyst
Conversion (%)
Xglycerol X1,2ꢀPDA
Selectivity (%)
S2,6ꢀDMP S6ꢀHMP S2,6ꢀDMPip aSothers
(gcat)ꢀ1 sꢀ1
CuCrsolꢀgel
CuCrcp
CuCrmm
CuCrimp
cCuCrsolꢀgel
Cr2O3
74.3
48.3
23.2
10.4
9.2
5.2
7.0
55.2
27.3
9.2
76.5
50.4
24.7
12.3
10.1
6.1
90.3
81.2
78.2
60.4
82.7
28.4
90.2
80.3
69.5
63.8
51.4
4.7
6.2
7.0
15.6
9.3
20.4
2.6
10.7
17.3
21.4
24.7
2.6
5.6
8.5
18.6
6.5
47.6
1.8
1.4
5.1
9.2
2.4
6.0
6.3
5.4
1.5
3.6
5.4
7.6
8.1
5.6
10.5
760
444
206
71
86
39
CuO
8.4
71
dCuCrsolꢀgel
dCuCrcp
dCuCrmm
57.3
28.7
9.8
500
215
66
dCuCrimp
3.1
3.9
13.4
18
a
b
Others include 2,5ꢀDimethylpyrazine and 5ꢀHydroxyꢀ2ꢀmethylpyrazine. Rate of 2,6ꢀDMP measured with
c
respect to glycerol conversion and 2,6ꢀ Dimethylpyrazine selectivity. Activity on unreduced CuCrsolꢀgel
sample. d Activity data using crude glycerol with a composition of 10 mmol glycerol +10 mmol 1,2ꢀPDA + 0.2
µmol KOH+ 0.3 mmol CH3OH + 200 mmol H2O.

Page 19 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
Table 4. Effect of GHSV on dehydrocyclization activity of CuCrsolꢀgel catalyst at a reaction
temperature = 375 °C; catalyst weight 0.1g.
GHSV
Conversion (%)
Selectivity (%)
(mL (gcat)ꢀ1 sꢀ1
Xglycerol
X1,2ꢀPDA
S 2,6ꢀDMP
S6ꢀHMP
S2,6ꢀDMPip
aSothers
17.10
23.14
29.20
35.24
56.3
57.2
23.3
10.4
7.1
90.0
5.3
2.9
3.1
3.7
4.2
1.8
22.1
10.3
6.2
87.3
85.7
85.2
5.8
7.8
9.3
3.8
2.8
1.3
a Others include 2,5ꢀDimethylpyrazine and 5ꢀHydroxyꢀ2ꢀmethylpyrazine.
Table 5. Influence of probe (pyridine and 2,6ꢀdimethylpyridine) addition on product
distribution. Reaction conditions: 10 wt% aq. glycerol and 1,2ꢀPDA; reaction temperature:
375 °C; catalyst wt: 0.1 g, carrier N2 = 30 cc minꢀ1, GHSV = 11.04 mL gcatꢀ1 sꢀ1.
Reaction condition
Xglycerol X1,2ꢀPDA S2,6ꢀDMP Sothers r2,6ꢀDMP /10ꢀ8 mol rothers /10ꢀ8 mol
(gcat)ꢀ1 sꢀ1
(gcat)ꢀ1 sꢀ1
a Before pyridine addition
b After pyridine addition
72.5
21.4
70.7
71.8
57.6
69.1
73.7
23.1
72.0
72.3
60.8
70.5
90.6
90.1
89.3
90.2
87.3
89.1
9.4
9.9
10.7
9.8
744
218
715
734
570
698
77
24
86
80
82
85
c After regeneration
a′ Before 2,6ꢀdimethylpyridine addition
b′ After 2,6ꢀdimethylpyridine addition
c′ After regeneration
12.6
10.9
a
b
Activity over CuCrsolꢀgel sample. Activity over CuCrsolꢀgel sample after coꢀfeeding 19.8 x10ꢀ3 moles of pyridine along with
c
a′
b′
reaction mixture. Activity after regeneration of the catalyst. Activity over CuCrsolꢀgel sample. Activity over CuCrsolꢀgel
sample after coꢀfeeding 19.8 x10ꢀ3 moles of 2,6ꢀdimethylpyridine along with reaction mixture. Activity after regeneration of
c′
the catalyst.

Catalysis Science & Technology
Page 20 of 28
View Article Online
DOI: 10.1039/C7CY00769H
Figure 1. XRD patterns of (A) calcined and (B) reduced samples.
242x145mm (96 x 96 DPI)

Page 21 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
Figure 2. TPR patterns of CuCr2O4 obtained by different preparation techniques.
368x392mm (96 x 96 DPI)

Catalysis Science & Technology
Page 22 of 28
View Article Online
DOI: 10.1039/C7CY00769H
Figure 3. FTIR pattern of (A) calcined and (B) reduced CuCr2O4 samples.
254x138mm (96 x 96 DPI)

Page 23 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
Figure 4. XPS analysis of reduced CuCr2O4 samples.
254x106mm (96 x 96 DPI)

Catalysis Science & Technology
Page 24 of 28
View Article Online
DOI: 10.1039/C7CY00769H
Figure 5. NH3-TPD patterns of reduced CuCr2O4 samples.
457x506mm (96 x 96 DPI)

Page 25 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
Figure 6. Formic acid adsorbed DRIFT spectra of (a) CuCrsol-gel (b) CuCrcp (c) CuCrmm and (d) CuCrimp
(e) Cr2O3 and (f) CuO samples.
458x436mm (96 x 96 DPI)

Catalysis Science & Technology
Page 26 of 28
View Article Online
DOI: 10.1039/C7CY00769H
Figure 7. Pyridine adsorbed on (a) CuCrsol-gel (b) CuCrcp and (c) 2,6-dimethylpyridine adsorbed DRIFT
spectra of CuCrsol-gel samples.
423x364mm (96 x 96 DPI)

Page 27 of 28
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00769H
Scheme 1. Proposed reaction mechanism for the formation of 2,6-dimethylpyrazine during
dehydrocyclization of glycerol and 1,2-propanediamine over CuCr2O4 catalyst.
69x63mm (600 x 600 DPI)

Catalysis Science & Technology
Page 28 of 28
View Article Online
DOI: 10.1039/C7CY00769H
263x107mm (96 x 96 DPI)