Liquid phase conversion of Glycerol to Propanediol over highly active Copper/Magnesia catalysts

Article abstract of DOI:10.1007/s12039-015-0838-6

Abstract In this work, a series of Cu/MgO catalysts with different copper metal loading were prepared by the precipitation-deposition method. Their catalytic behaviour was investigated for glycerol hydrogenolysis to 1,2-propanediol (1,2-PDO). The physico-chemical properties of the catalysts were characterized by various techniques such as BET surface area, X-ray diffraction (XRD), temperature programmed reduction (TPR), NH₃-temperature programmed desorption (NH₃-TPD) and scanning electron microscopy (SEM) methods. The characterization results showed that the copper metal was well-dispersed over MgO support and a new phase Cu-MgO was also identified from XRD results after calcination. The 25Cu/MgO (Cu:25 wt%) catalyst exhibited the highest glycerol conversion of 88.7% and 1,2-PDO selectivity of 91.7% at 210°C, 4.5 MPa of hydrogen pressure after 12 h. The high glycerol conversion was mainly due to the Cu dispersion on MgO support and high acidic strength. Further, the effects of temperature, hydrogen pressure, catalyst loading and glycerol concentration were studied over 25Cu/MgO catalyst for optimization of reaction parameters. Kinetic study over highly active 25Cu/MgO catalyst showed that the reaction followed the pseudo second order rate with respect to glycerol and the apparent activation energy was found to be 28.7 ± 0.8 kcal/mol.

Full text of DOI:10.1007/s12039-015-0838-6

c
J. Chem. Sci. Vol. 127, No. 5, May 2015, pp. 833–842. ꢀ Indian Academy of Sciences.
DOI 10.1007/s12039-015-0838-6
Liquid phase conversion of Glycerol to Propanediol over highly active
Copper/Magnesia catalysts
SATYANARAYANA MURTY PUDI∗, ABDUL ZOEB, PRAKASH BISWAS
and SHASHI KUMAR
Department of Chemical Engineering, Indian Institute of Technology Roorkee,
Roorkee 247667, Uttarakhand, India
e-mail: psnmurthy333@gmail.com; satyadch@iitr.ac.in
MS received 29 June 2014; revised 3 December 2014; accepted 3 December 2014
Abstract. In this work, a series of Cu/MgO catalysts with different copper metal loading were prepared by
the precipitation-deposition method. Their catalytic behaviour was investigated for glycerol hydrogenolysis
to 1,2-propanediol (1,2-PDO). The physico-chemical properties of the catalysts were characterized by vari-
ous techniques such as BET surface area, X-ray diffraction (XRD), temperature programmed reduction (TPR),
NH3-temperature programmed desorption (NH3-TPD) and scanning electron microscopy (SEM) methods. The
characterization results showed that the copper metal was well-dispersed over MgO support and a new phase
Cu-MgO was also identified from XRD results after calcination. The 25Cu/MgO (Cu:25 wt%) catalyst exhib-
ited the highest glycerol conversion of 88.7% and 1,2-PDO selectivity of 91.7% at 210◦C, 4.5 MPa of hydrogen
pressure after 12 h. The high glycerol conversion was mainly due to the Cu dispersion on MgO support and
high acidic strength. Further, the effects of temperature, hydrogen pressure, catalyst loading and glycerol con-
centration were studied over 25Cu/MgO catalyst for optimization of reaction parameters. Kinetic study over
highly active 25Cu/MgO catalyst showed that the reaction followed the pseudo second order rate with respect
to glycerol and the apparent activation energy was found to be 28.7 0.8 kcal/mol.
Keywords. Glycerol hydrogenolysis; 1,2-propanediol; Cu/MgO catalysts; kinetic study.
1. Introduction
several processes. Among these processes, catalytic
hydrogenolysis of glycerol to 1,2-propanediol (1,2-
PDO) and 1,3-propanediol (1,3-PDO) is a promising
process. Some of the uses of 1,2-PDO are in the produc-
tion of resins, unsaturated polyesters, cosmetics, anti-
freeze agent, tobacco humectants, pharmaceuticals and
paints. 1,3-propanediol (1,3-PDO) can also be used as
the monomer with terephthalic acid for the production
of polyesters.3,4
Various heterogeneous catalysts such as supported
Ru,5–9 Pt,6,7,10 Pd6,9 and Ir-Rh oxide11 catalysts have
been considered as highly active for the hydrogenolysis
of glycerol. However, the selectivity towards 1,2-PDO
is poor due to the propagation of C-C bond cleavage,
which leads to the degradation products. Cu-based
catalysts12–19 have been widely studied for glycerol
hydrogenolysis to propanediols, mainly because of their
high selectivity towards C-O bond hydro-dehydrogenation
and poor activity towards C-C bond cleavage. Copper
chromite, a conventional hydrogenation catalyst, showed
good performance in the glycerol hydrogenolysis to
1,2-PDO (65% glycerol conversion with 90% 1,2-PDO
selectivity)12 Chaminand et al.13 reported 100% selec-
tivity for 1,2-propanediol in the presence of CuO/ZnO
catalysts under the conditions of 180◦C and 8 MPa
There is a need to search for better sustainable and
renewable energy resources to meet the energy demand
for mankind. At present, biodiesel has proven to be
one of the promising alternatives as a non-toxic and
biodegradable energy resource. It is essentially a mix-
ture of methyl esters of short chain alcohols such as
methanol and ethanol derived from fatty acids obtained
from renewable lipid sources via trans-esterification
process. Crude glycerol is the principal by-product
(10% in weight) of the biodiesel production process.1
This increased production of biodiesel would lead to
simultaneous increased production of glycerol which is
expected to exceed the demand in near future.2 Utiliza-
tion of glycerol into value added products can help in
reducing the overall cost of biodiesel production.
Currently propanediols are produced from petroleum
derivatives via chemical catalytic routes. Their produc-
tion from glycerol via catalytic hydrogenolysis could
potentially offer great environmental benefits. Glycerol
can be transformed into a number of value-added
products namely propanediols, lactic acid, acrolein by
∗For correspondence
833

834
Satyanarayana Murty Pudi et al.
hydrogen pressure but the conversion was very low The numbers indicate the weight % of Cu on the MgO
(17%) at longer reaction time of 90 h. Huang et al.14
reported 29% glycerol conversion and 99% 1,2-PDO
selectivity over Cu/SiO2 catalyst at 180◦C and 9
MPa hydrogen pressure. Balaraju et al.15 achieved
49.3% conversion over Cu-MgO catalyst with 92%
of selectivity for 1,2-PDO after 8 h at moderate
reaction conditions (4 MPa 200◦C). Guo et al.16
described Cu/Al2O3 catalyst showed 96.8% selectivity
to propanediols with a glycerol conversion of 49.6%
at 220◦C, 1.5 MPa initial H2 pressure after 10 h. More
recently, mixed metal oxides such as Cu/Zn-Mg-Al,17
support.
2.2 Catalyst characterization
BET surface areas of prepared catalysts were determined
by the Micromeritics Accelerated Surface Area and
Porosimetry (ASAP-2020) system, employing nitrogen
physisorption at liquid nitrogen temperature. Prior to
the analysis, the catalyst samples were degassed at
300◦C for 6 h. X-ray diffraction (XRD) of prepared cata-
lysts were obtained in a Bruker diffractometer (Model
AXS D8, Germany) using Ni filtered Cu Kα radiation
(λ = 1.5406 Å) with a scanning angle (2θ) of 10◦–80◦,
a scanning speed of 0.02◦ with 3 s collection time. The
identification of compounds in powder samples was
attained by comparison of the peaks with data files pro-
vided by Joint Committee on Powder Diffraction Stan-
dards (JCPDS) using PCPDFWIN software. The average
crystallite size was determined using the Scherrer equa-
tion, D = 0.90λ/B cosθ, where θ is the diffraction
angle, B is the full width at half-maximum (FWHM).
Temperature programmed reduction experiments with
hydrogen (H2-TPR) were studied using Micromerit-
ics Chemsorb 2720 pulse chemisorption system. For
each measurement, 25 mg of catalyst sample was placed
in a quartz U-tube reactor. Before reduction, sam-
ples were pre-treated at 150◦C in flowing argon for
2 h which helps to remove the physically adsorbed
moisture. Then, 10% H2/Ar flow of 20 ml/min was
passed over the sample under heating at a rate of
10◦C/min until temperature was raised up to 700◦C.
The hydrogen consumption was monitored by a thermal
conductivity detector (TCD) by means of analyzed
effluent gas with argon as reference. Temperature
programmed desorption experiments with ammonia
(NH3-TPD) were carried out using same apparatus as
above used for H2-TPR. Prior to the measurements,
Cu/ZnO/ZnAl2O4 and Cu-Zn-Cr-Zr19 were reported
18
as highly active and achieved 80–100% glycerol con-
versions. Regardless of several investigations, the het-
erogeneous catalytic hydrogenolysis of glycerol is not
fully commercialized because of the high reaction tem-
peratures (250–350◦C), high hydrogen pressures (6–
30 MPa), poor catalyst stability and lower glycerol
activity.
In the present study, highly active and selective MgO-
supported copper metal catalysts have been employed
for glycerol hydrogenolysis in aqueous phase, since
alkaline supports are known to influence the dehy-
dration of glycerol during hydrogenolysis reaction.
Cu/MgO catalysts having different copper loadings
were prepared and characterized. The effect of pro-
cess parameters such as catalyst loading, reaction tem-
perature, hydrogen pressure and glycerol concentration
on the conversion of glycerol and product selectivity
were examined. Also, for a better understanding of the
reaction mechanism, kinetic study was undertaken.
2. Experimental
2.1 Catalyst synthesis
A metal supported composite catalyst was prepared about 80 mg of catalyst samples were degassed in
by precipitation-deposition method. Firstly, calculated flowing helium at 150◦C. Then samples were cooled
amount of Cu(NO3)2.3H2O (Himedia chemicals, India, to room temperature and ammonia adsorption carried
AR) and MgO (Thomas Baker, India, AR) were taken out by switching gas to 27% NH3/He (30 mL/min)
and precipitated with dropwise addition of 1 M aque- for 1 h. After saturated adsorption, samples were
ous sodium hydrogen carbonate solution until the pH purged with helium (20 mL/min) for 1 h to remove any
of mixed solution reached to 8.0–8.25. The result- physisorbed ammonia on the surface. The ammonia
ing slurry was stirred for 6 h at ambient temperature desorption measurements were conducted in flowing
and aged for 12 h. The filtrate obtained was separated helium (20 mL/min) from 30◦C to 800◦C at a heating
by vacuum filtration and washed thoroughly with dis- rate of 15◦C/min. The morphology of catalysts and
tilled water to eliminate traces of any sodium ions. The metal distribution of catalysts were examined by scan-
obtained precipitate was dried overnight at 110◦C in ning electron microscopy (SEM) on a microscope
oven followed by calcination in air at 550◦C for 4 h. equipped with energy-dispersive X-ray analysis EDX
The prepared catalysts are designated as 5Cu/MgO, used for estimation of local elemental composition.
10Cu/MgO, 15Cu/MgO, 20Cu/MgO and 25Cu/MgO. SEM images were collected using Quanta scanning

Hydrogenolysis of glycerol on Cu/MgO
835
electron microscope (Model 200 FEG, USA) equipped nitrogen gas (0.5 MPa) for few times, further reactor
was heated to reaction temperature of 210◦C and pres-
surized the reactor by H2 gas up to desired pressure of
with EDX.
4.5 MPa. The reaction was allowed to proceed under
these conditions for 12 h at 700 rpm. After the reaction,
2.3 Catalytic activity
liquid phase products were centrifuged (Heraeus
Glycerol hydrogenolysis was carried out in a Teflon-
Biofuge Stratos, Thermo Scientific) to separate any cat-
lined stainless steel autoclave (250 mL) (Amar Equip-
alyst particles present in the sample. The reactant and
ments, India). The reactor was equipped with pro-
products involved in this reaction were analyzed using
grammable temperature controller (PID), pressure
a gas chromatograph (Model no: GC 6800 Newchrom
gauge up to 10 MPa, stirrer (1–1450 rpm), a vent and a
Technologies, India) equipped with flame ionization
sample port for liquid sampling. Prior to the each run,
detector (FID) and Chromosorb-101 packed column
catalyst samples were reduced in a tubular reactor under
(1.52 m × 3.1 mm OD × 2 mm ID). n-butanol was used
hydrogen flow of 50 mL/min at 350◦C for 3 h. In a typ-
as an internal standard in analysis for the quantification
ical run, 100 mL of 20 wt% aqueous glycerol solution
and 1.6 g of reduced Cu/MgO catalyst were fed into the
reactor. Prior to reaction, the reactor was flushed with
of glycerol derived products. The following equations
were used to calculate glycerol conversion and product
selectivity:
Moles of glycerol in feed − Moles of glycerol in product
Conversion (%) =
× 100
Moles of glycerol in feed
ꢀ
ꢁ
C-based moles of product X
Selectivity of product X (%) =
× 100
Total C-based moles of all products
value of 36.9◦, 74.6◦ and 78.6◦ corresponds to (111),
(311) and (222) planes.
The error in reproducibility of experiments was within
3% and carbon balance was 100 5%.
The diffraction peaks corresponding to very high
intensity CuO were not detected in any of the catalysts
from 5Cu/MgO to 20Cu/MgO, whereas new diffraction
appeared at 18.6◦ and 38.7◦. This indicates the forma-
tion of new phase CuO-MgO between precursors. Pre-
viously, it was reported that the Cu-MgO phase can
be easily generated on the interface of active Cu and
MgO by the synergetic effect of Cu and MgO after
the application of high temperature treatment to both
mixtures.20 While for 25Cu/MgO, high intense reflec-
tions are observed for monoclinic CuO (JCPDS: 80–
1917) at 2θ = 35.6◦ and 38.9◦ correspond to (111) and
(111) planes. For all Cu loaded catalysts, a low intensity
diffraction peak is obtained at 2θ = 58.8◦ correspond-
ing to CuO (202) plane. The low intensity or absence
of CuO phases in some catalysts suggesting high dis-
persion of copper on the MgO surface.21 Previous
studies also reported similar kind of Cu based catalysts
with high dispersion on support and good metal-support
interactions.22,23 The crystallite size of Cu/MgO were
calculated based on the diffraction angle and X-ray
line broadening using Scherrer equation are shown in
3. Results and Discussion
3.1 Catalyst characterization
The BET surface area and pore volume of Cu/MgO
catalysts are summarized in table 1. The nitrogen
adsorption-desorption isotherms represents typical me-
soporous nature of type IV isotherms. The BET surface
area was calculated from the relative pressure data
(P/Po) obtained in the range of 0.05–0.3. The surface
area decreased sharply with increasing Cu loading from
70.3 m2/g (for 5Cu/MgO) to 24.5 m2/g (for 25Cu/MgO)
probably due to the formation of CuO clusters on
the MgO support which might be blocking the pores
of MgO.15 Barret-Joyner-Halenda (BJH) method is
applied to desorption branch of isotherms to calculate
the pore volume. The pore volume of all catalysts is
observed to be in the range of 0.13–0.03 (cm3/g).
XRD patterns of Cu/MgO catalysts having different
metal loading (5 wt% to 25 wt %) is shown in figure 1.
For pure MgO, the most intense peaks are observed at table 1. The average crystallite size of MgO was cal-
2θ value of 42.9◦ and 62.2◦ corresponding to (200) and
(220) planes, respectively (JCPDS: 78–0430). Three
low intensity peaks are observed for MgO at the 2θ
culated from the line widths of characteristic peaks at
2θ = 42.9◦, 62.2◦ and 78.6◦ corresponding to (200),
(220) and (222) planes and the average crystallite size

836
Satyanarayana Murty Pudi et al.
Table 1. Physico-chemical properties of Cu/MgO catalysts.
BET Surface Pore volume
Acidity
Average crystallite size (nm)
Catalysts
area (m2/g)
(cm3/g)
(mmol NH3/g catalyst)
2θ
CuO
2θ
MgO
19.6
5Cu/MgO
10Cu/MgO
15Cu/MgO
20Cu/MgO
25Cu/MgO
70.3
60.8
31.6
27.5
24.5
0.13
0.08
0.04
0.03
0.03
2.21
2.81
3.47
4.31
4.89
58.8◦
58.8◦
16.7
42.9◦, 62.2◦
18.0 42.9◦, 62.2◦, 78.6◦ 24.6
16.1 42.9◦, 62.2◦, 78.6◦ 15.1
19.1 42.9◦, 62.2◦, 78.6◦ 19.2
58.8◦
35.6◦, 58.8◦
35.6◦, 38.9◦, 58.8◦ 26.0 42.9◦, 62.2◦, 78.6◦ 24.1
(200)
♦
♦
MgO (JCPDS:78-0430)
CuO (JCPDS:80-1917 )
CuO-MgO
∗
∗
Δ
(220)
♦
(222)
(311)
(111)
♦
MgO
♦
♦
(111)
(111)
♦
(202)
∗
♦
∗
∗
25Cu/MgO
♦
♦
♦
♦
♦
20Cu/MgO
15Cu/MgO
Δ
♦
♦
♦
∗
∗
∗
∗
∗
♦
♦
T
Δ
reduction
Δ
25Cu/MgO
Δ
Δ
Δ
Δ
10Cu/MgO
5Cu/MgO
♦
♦
100
150
200
250
300
350
400
450
500
10
20
30
40
50
60
70
80
o
Temperature ( C)
2theta (Degree)
Figure 2. Temperature programmed reduction profile of 25
Cu/MgO catalyst.
Figure 1. X-ray diffraction patterns of calcined Cu/MgO
catalysts on different loadings.
of CuO was determined from the line width of XRD
peaks at 2θ = 35.6◦, 38.9◦ and 58.8◦ representing
(111), (111) and (202) planes. The average crystallite
size of CuO is observed in the range of 16.1–26 nm,
whereas for MgO crystallite sizes are in the range of
15.1–24.6 nm.
TPR analysis was carried out to study the redox prop-
erties of catalysts. TPR profile for 25Cu/MgO cata-
lyst is shown in figure 2. For 25Cu/MgO catalyst one
high intensity broad reduction peak centered at 288◦C
is observed. Intense hydrogen consumption is noticed
between 200–350◦C which may be attributed to the
reduction of larger CuO particles to Cu. The broad
reduction peak is might be due to assimilation of reduc-
tion peaks associated with dispersed and CuO cluster
sites for high Cu metal loadings.15,21
Moderate acidic sites
Weak acidic sites
Strong acidic sites
25Cu/MgO
20Cu/MgO
15Cu/MgO
10Cu/MgO
5Cu/MgO
100
200
300
400
500
600
700
o
Temperature ( C)
NH3-TPD experiments were carried out to study the
acidic characteristics of Cu/MgO catalysts. The quanti-
tative estimation of acidic site distribution at different
regions according to the desorbed amount of ammonia
Figure 3. NH3-Temperature programmed desorption pro-
files of Cu/MgO catalysts with different Cu loadings.
are summarized in table 1. For all the catalyst samples, be attributed to ammonia desorption from weak acidic
the acidic sites distributed in three different regions sites, the second region refers to moderate strength
at 80–250◦C, 250–500◦C and 500–750◦C (figure 3). acidic sites and the third region represents desorption of
According to previous literature, the first region should ammonia from strong acidic sites.24,25

Hydrogenolysis of glycerol on Cu/MgO
837
(a)
(b)
(c)
(d)
(e)
Figure 4. SEM images of Cu/MgO catalysts. (a) 5 Cu/MgO, (b) 10 Cu/MgO,
(c) 15 Cu/MgO, (d) 20 Cu/MgO, (e) 25 Cu/MgO.
The acidic strength of Cu/MgO catalysts increased The particle agglomeration maybe the reason for further
from 2.21 to 4.89 mmol NH3/g catalyst as the metal decrease in surface area.
loading increases from 5% to 25%. Initially the weak
and moderate acidic sites are more in case of lower
metal loaded catalysts (5Cu/MgO and 10Cu/MgO). As
the metal loading increased, the strength of weak acidic
sites gets decreased and becomes moderate, and strong
acidic sites become dominant in the catalysts (figure 3).
SEM images were used in order to observe the mor-
phology of prepared catalysts. Figure 4 represents SEM
images of Cu catalysts supported on MgO. All the cat-
alysts supported on MgO are found as amorphous in
nature. The particle morphology of all catalysts show
similar to quasi spherical nanoparticles of 10–40 nm
size. As the Cu loading increased, the coverage of the
MgO surface by Cu particles is observed. At 5–25 wt%
Cu loading, the agglomeration of the catalyst particles
is noticed, which results in formation of big lumps.
3.2 Catalytic activity measurements
3.2a Influence of Cu loading on MgO on glycerol
hydrogenolysis: The glycerol conversion and 1,2-PDO
selectivity obtained over different catalysts is tabulated
in table 2. The results show that 39.4% conversion
of glycerol, and 94.2% selectivity towards 1,2-PDO is
achieved over 5Cu/MgO catalyst.
As the Cu metal loading increases from 10 to 25 wt%,
the conversion of glycerol increased significantly from
39.4% to 88.7% and selectivity to 1, 2-PDO found to
be in range of 90.5–94.2%. From the above results, it
is observed that high Cu loading (25 wt%) on MgO
showed high glycerol hydrogenolysis activity. Previous

838
Satyanarayana Murty Pudi et al.
Table 2. Conversion and selectivity of the products obtained for
liquid phase hydrogenolysis.
Selectivity (%)
Catalyst
Conversion (%) 1,2-PDO EGa Acetol Othersb
5Cu/MgO
10Cu/MgO
15Cu/MgO
20Cu/MgO
25Cu/MgO
39.4
69.2
78.6
85.0
88.7
94.2
92.0
90.5
92.9
91.7
4.4
5.0
4.5
4.6
5.5
0.4
0.7
2.0
0.5
0.5
0.9
2.2
2.8
1.9
2.2
a EG: Ethylene glycol, b Others: 1-propanol, 2-propanol, traces of ethanol
and methanol
Reaction condition: 20 wt% glycerol concentration, Temperature =
210◦C; Hydrogen pressure = 4.5 MPa, catalyst weight = 1.6 g, reaction
time = 12 h.
literature has discussed about the development of nation of Ru/C with solid acid (Amberlyst) enhanced
various catalysts system for glycerol hydrogenolysis. the turn over frequency (TOF) of 1,2-PDO formation.
However, the information regarding the Cu/MgO Gandarias et al.10 suggested that Pt/ASA catalysts hav-
catalysts is limited. It is also reported in literature that ing acid sites are responsible for achieving higher glyc-
high glycerol conversion was achieved over Cu-MgO erol conversions with maximum 1,2-PDO selectivity.
catalysts.15,26 Yuan et al.26 reported 72% glycerol Cu based catalysts such as Cu/Al2O3 and Cu-Zn-
16
conversion and 97.6% 1,2-PDO selectivity over Cu- Cr-Zr19 also suggested that increasing acidic strength
15/MgO catalyst at 13.3% catalyst loading, 180◦C, of catalysts enhanced the rate of glycerol conversion.
3.0 MPa H2 pressure after 20 h of reaction and when This indicates that dehydration of glycerol to acetol
small amount of NaOH was added to the reaction mix- catalyzes by the acidic sites and subsequent hydro-
ture, the glycerol conversion and 1,2-PDO selectivity genation on the metal catalyst results 1,2-PDO. Inter-
reached to 82% and 97.6%. Balaraju et al.15 observed estingly, Xia et al.17 reported that both acidic and/or
49.3% glycerol conversion and 92.3% 1,2-PDO selec- basic sites are responsible for maximum activity over
tivity over 20 CuMgO catalyst at 6% catalyst loading, Cu0.4/Zn0.6Mg5.0Al2O8.6 layered double oxide catalysts.
200◦C, 4.0 MPa H2 pressure after 8 h of reaction time. Earlier studies8,10,16,19 and the preceding discussion
Yue et al.20 achieved 36.2% glycerol conversion and resolved that the activity of Cu/MgO catalysts reported
85% 1,2-PDO selectivity over Raney Cu/MgO catalyst in this study increased obviously with their acidity
at 2.5% catalyst loading, 180◦C, 1.0 MPa H2 pressure (table 1). The glycerol conversion of 88.7% with 81.3%
after 24 h. In the present study, maximum glycerol con- yield to 1,2-PDO over 25Cu/MgO can be attributed to
version of 88.7% was obtained over 25Cu/MgO cata- its stronger acidity (4.89 mmol NH3/g catalyst). The
lyst with 91.7% selectivity to 1,2-PDO at 8% catalyst activity and yield of the present study are comparable
loading at 210◦C, 4.5 MPa H2 pressure. Comparison of with published literature.8,10,16,19 Similar conclusions
the result reported in this study with the previous litera- can be drawn from the XRD results, NH3-TPD data and
ture, it is clearly seen that the high glycerol conversion small particles of Cu observed from crystallite size and
and 1,2-PDO selectivity are achieved at the mild reac- SEM that high dispersion of Cu and acidic strength of
tion condition which is mainly due to the acidity of cat- Cu/MgO catalysts improved the hydrogenolysis activity
alysts and bifunctional nature of Cu-MgO. As the acidic and 1,2-PDO yield.
strength increases from 2.21 to 4.89 mmol NH3/g cat-
alyst, the glycerol conversion increased from 39.4% to
88.7%.
3.2b Effect of reaction temperature: The effect of
reaction temperature on the catalytic performance over
25Cu/MgO catalyst is shown in figure 5. The conver-
sion of glycerol increased from 72% to 88% as the tem-
perature increases from 190◦C to 210◦C and the selec-
tivity of 1,2-PDO also increased from 88.7% to 91.7%.
Further increase in temperature from 210◦C to 225◦C
the conversion of glycerol slightly increased to 91.8%
whereas the selectivity of 1,2-PDO is decreased from
In published works, the role of acidic sites of cata-
lysts was explained in different ways. Balaraju et al.8
found that solid acid (Niobia) as a co-catalyst with
Ru/C could improve the glycerol activity and increas-
ing the amount of Ru/C + solid acid led to a signifi-
cant increase in both glycerol conversion and 1,2-PDO
selectivity. Miyazawa et al.6 reported that the combi-

Hydrogenolysis of glycerol on Cu/MgO
839
100
90
80
70
60
50
40
30
20
10
0
100
the conversion of glycerol is increased from 49% to
88%. Further increase in pressure to 6 MPa, the conver-
sion decreased to 81% indicating that at higher pres-
sures hydrogen atoms compete with the readsorption
of an intermediate product (acetol) for active catalyst
sites resulting the decrease in glycerol conversion.15,27
However the selectivity of 1,2-PDO is decreased from
91% to 86% at higher pressure range of 4.5–6 MPa.
The decrease in selectivity is due to formation EG. The
selectivity of EG is found to be in range of 3–6%.
90
80
70
60
50
40
30
20
10
0
Glycerol conversion
1,2-PDO selectivity
EG
3.2d Effect of catalyst weight: The effect of catalyst
weight on the catalytic performance over 25Cu/MgO
catalyst is shown in figure 7. Catalyst weight was var-
ied from 0.8 g (catalyst/glycerol = 4%) to 2 g (cata-
lyst/glycerol = 10%). Results show that, glycerol con-
version increased with increase in the catalyst amount.
Glycerol conversion is found to be 56.7% at 0.8 g of cat-
alyst loading. Further increase in catalyst loading up to
2 g, glycerol conversion is reached to 95%. The selec-
tivity of 1,2-PDO obtained is in the range of 86–91%
and the selectivity of EG is in the range of 5–7%. The
increase in glycerol conversion with catalyst loading is
due to the more availability of active catalytic sites for
selective hydrogenolysis of glycerol.12,27
190
195
200
205
210
215
220
225
Reaction temperarture ( )
oC
Figure 5. Effect of reaction temperature on hydrogenoly-
sis of glycerol over 25 Cu/MgO catalyst. Reaction condi-
tions: 20 wt% glycerol solution, 1.6 g catalyst, H2 pressure:
4.5 MPa, time: 12 h.
91% to 83% at high temperatures. EG selectivity is
obtained in the range of 5–7%. This result indicates
that at high reaction temperatures (>210◦C), excessive
hydrogenolysis of glycerol and 1,2-PDO is taking place
which further yields byproducts like EG, propanol,
methanol, ethanol and other gaseous products.12 Hence
reaction temperature of 210◦C is identified as optimum
temperature for hydrogenolysis of glycerol.
3.2e Effect of glycerol concentration: The effect of
glycerol concentration on the catalytic performance
over 25Cu/MgO catalyst is shown in figure 8. It is
observed that glycerol conversion decreased from 88%
to 38% with increase in glycerol concentration from
10 wt% to 60 wt% which is likely due to the limited
number of available Cu sites with respect to glycerol as
3.2c Effect of reaction pressure: The effect of
reaction pressure on the catalytic performance over
25Cu/MgO catalyst is shown in figure 6. As the hydro-
gen pressure in reaction increases from 1.5 to 4.5 MPa,
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
Glycerol conversion
1,2-PDO selectivity
EG
glycerol concersion
1,2-PDO selectivity
EG
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Reaction pressure (MPa)
Catalyst weight (g)
Figure 6. Effect of reaction pressure on hydrogenolysis
of glycerol over 25 Cu/MgO catalyst. Reaction conditions:
20 wt% glycerol solution, 1.6 g catalyst, reaction tempera-
ture: 210◦C, time: 12 h.
Figure 7. Effect of catalyst weight on hydrogenolysis of
glycerol over 25 Cu/MgO catalyst. Reaction conditions:
20 wt% glycerol solution, H2 pressure: 4.5 MPa, reaction
temperature: 210◦C, time: 12 h.

840
Satyanarayana Murty Pudi et al.
namics at low temperature,31 it is considered that glyc-
100
100
90
80
70
60
50
40
30
20
10
0
90
80
70
60
50
40
30
20
10
0
erol dehydration is resulted by the combined effect of
hydrogen spill over of Cu and acid-base property of
MgO in Cu/MgO catalysts. Previously Sato et al.28 pro-
posed an analysis based on transition state of hydro-
gen effect on Cu surface and glycerol dehydrogenation,
the secondary hydroxyl groups are removed under the
effect of Mg2+ on Cu/MgO surface in free radical form
under synergetic catalysis. Otherwise, it will produce
aldehydes via dehydrogenation, followed by dehydra-
tion and hydrogenation to acetol. None of the acetiliza-
tion products (cyclic compounds) are detected after
reaction and enol is produced as intermediate by dehy-
dration. Unstable enol is further converted to acetol by
tautomerization followed by acetol to 1,2-PDO through
catalytic hydrogenation on Cu surface (scheme 1). This
result has been supported by other Cu based catalysts on
hydrogenolysis of glycerol.20,32 Propanols are obtained
by the over hydrogenolysis of 1,2-PDO and other prod-
ucts such as ethylene glycol, ethanol and methanol are
generated by the other catalytic reactions. Nevertheless,
further research and in-depth investigation are needed
to affirm this hypothesis.
Glycerol conversion
1,2-PDO selectivity
EG
10
20
30
40
50
60
Glycerol concentration (%)
Figure 8. Effect of glycerol concentration on hydrogenoly-
sis of glycerol over 25Cu/MgO catalyst. Reaction conditions:
1.6 g catalyst, H2 pressure: 4.5 MPa, reaction temperature:
210◦C, time: 12 h.
the catalyst weight is constant. The selectivity towards
1, 2-PDO decreased from 91% to 84% as the glycerol
concentration increases from 10 wt% to 60 wt%, which
is due to the formation of degradation reaction prod-
ucts. EG selectivity was obtained in the range of 3–5%.
These results indicate that glycerol conversion depends
3.3 Kinetics of glycerol hydrogenolysis
on the concentration of glycerol in the reaction mixture. Figure 9 shows the effect of time on glycerol
The previous literature indicated that the glycerol hydrogenolysis over 25Cu/MgO catalyst at 190, 200,
activity depends upon Cu metal dispersion and basic- 210 and 225◦C. The conversion of glycerol increased
ity obtained from CO2-TPD in Cu-MgO catalysts pre- linearly at lower temperatures (190 and 200◦C). The
pared via co-precipitation method.15 It was reported that glycerol conversion is increased from 11.4% to 48.8%
Raney Cu catalyst promotes the transformation of glyc- as the time increases from 1 h to 5 h at 190◦C. At
erol into hydroxyacetone (intermediate) by free radi- 220◦C, the conversion of glycerol obtained from 20%
cal process by removing the primary hydroxyl group in to 67% with increase in time range of 1–5 h. The selec-
glycerol.28 Thus, Cu is co-operated with MgO catalyzed tivity of 1, 2-PDO was found in the range of 86–88%.
glycerol to improve the selectivity to 1,2-PDO. Mean- On the other hand at 210◦C, the conversion of glyc-
while, based on acid-base properties of MgO,29 very erol increased quickly from 41.5% (at 1 h) to 81.2% (at
good catalytic dehydrogenation property of Cu-MgO,30 5 h). The selectivity of 1,2-PDO was found in the range
and tendency to produce metastable acetol in thermody- of 85–87%. Similarly for 225◦C, the conversion of
Scheme 1. Reaction pathway scheme for glycerol hydrogenolysis to 1,2-PDO.

Hydrogenolysis of glycerol on Cu/MgO
841
100
90
80
70
60
50
40
30
20
10
It is observed from the results of time courses at
different temperatures, the relation between (XG/1-XG)
and reaction time is found to be passing through ori-
gin (figure 10a), which confirms that the reaction is
pseudo second order with respect to glycerol. The cal-
culated reaction rate constants are 7.55, 17.53, 40.03
and 67.75 L mol−1h−1 at 190, 200, 210 and 225◦C,
respectively. Arrhenius plot was made by plotting ln (k/
L mol−1h−1) vs. (1/T) (figure 10b). The calculated acti-
vation energy of glycerol hydrogenolysis over Cu/MgO
catalyst in temperature range of 190–225◦C is found
190 oC
200 oC
210 oC
225 oC
to be 28.7
0.8 kcal/mol, a value in agreement with
that reported by Sharma et al.19 (31.5 kcal/mol) and
Yadav et al.33 (27.5 kcal/mol). This activation energy
value also indicates that the hydrogenolysis reaction is
intrinsically kinetically controlled.33
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Time (h)
Figure 9. Effect of temperature with time on hydrogenol-
ysis of glycerol over 25 Cu/MgO catalyst. Reaction condi-
tions: 20 wt% glycerol solution, 1.6 g catalyst, H2 pressure:
4.5 MPa, time: 12 h.
From the current study it is observed that the
MgO supported Cu catalysts increase the activity for
hydrogenolysis of glycerol to produce 1,2-PDO in good
yield. The increase in activity is due to the combined
effects of metal dispersion over support and acidic
strength which is confirmed by characterization results.
glycerol increased quickly to 68% in two hours and then
increased slowly from 68% (after 2 h) to 87% (after
5 h).
7
4. Conclusion
y= 1.4701x
R2 = 0.9894
190 oC
6
200 oC
210 oC
5
A series of Cu catalysts supported on MgO with differ-
ent Cu metal loading were synthesized by precipitation-
deposition method. The catalysts were characterized
by various techniques and successfully evaluated for
selective hydrogenolysis of glycerol.
225 oC
y= 0.8688 x
R2 = 0.9986
4
3
y= 0.3804 x
R2 = 0.9909
2
The glycerol conversion was found to increase with
increase in Cu metal loading and achieved maximum
glycerol conversion of 88.7% over 25Cu/MgO catalyst
with 91.7% selectivity to 1,2-PDO at 210◦C, 4.5 MPa
of hydrogen pressure after 12 h. Further, the effect of
different reaction parameters such as temperature, pres-
sure, catalyst amount and glycerol concentration was
evaluated over 25Cu/MgO catalyst to maximize the
glycerol conversion and 1,2-PDO selectivity. Effect of
temperature study revealed that the glycerol conversion
was increased from 72% to 92% with a rise in tempera-
ture from 190 to 225◦C. The activity of glycerol conver-
sion was tested at different catalyst weight and it was
found that even at lower catalyst loading of 6% and 4%,
the glycerol conversion was 67.5% and 56.7%, respec-
tively and the 1,2-PDO selectivity was ∼91% under
same reaction condition.
y= 0.1639 x
1
R2 = 0.9728
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
(a)
Time (h)
4.4
4.0
3.6
3.2
2.8
2.4
2.0
y= -14475 x+ 33.399
R2 = 0.9857
0.00201 0.00204 0.00207 0.00210 0.00213 0.00216
(1/T)/ K-1
The catalyst characterization results revealed that the
combined effects of Cu metal dispersion over MgO
support and acidic strength played a significant role
to enhance the catalytic activity. Product distribution
obtained over 25Cu/MgO catalyst suggested that the
(b)
Figure 10. (a) Pseudo second order plot of (XG/1-XG) vs.
Time (h) to calculate rate constant. (b) Arrhenius plot to
calculate energy of activation of glycerol hydrogenolysis to
1,2-PDO.

842
Satyanarayana Murty Pudi et al.
12. Dasari M A, Kiatsimkul P P, Sutterlin W R and Suppes
G J 2005 Appl. Catal. A: Gen. 281 225
13. Chaminand J, Djakovitch L, Gallezot P, Marion P, Pinel
C and Rosier C 2004 Green Chem. 6 359
14. Huang Z W, Cui F, Kang H, Chen J, Zhang X and Xia
C 2008 Chem. Mater. 20 5090
15. Balaraju M, Jagadeeswaraiah K, Sai Prasad P S and
Lingaiah N 2012 Catal. Sci. Technol. 2 1967
16. Guo L, Zhou J, Mao J, Guo X and Shuguang Z 2009
Appl. Catal. A: Gen. 367 93
hydrogenolysis reaction followed the tautomerization
of unstable enol to acetol, followed by catalytic hydro-
genation to 1,2-PDO. Kinetic results showed that the
reaction is pseudo second order with respect to glycerol
and the apparent activation energy was found to be 28.7
0.8 kcal/mol.
Acknowledgements
17. Xia S, Nie R, Lu X, Wang L, Chen P and Hou Z 2012 J.
Catal. 296 1
18. Tan H, Hedhil M N, Wang Y, Zhang J, Li K, Sioud S,
Al-Tall Z, Amad M H, Zhan T, Tal O E and Han Y 2013
Catal. Sci. Technol. 3 3360
19. Sharma R V, Kumar P and Dalai A K 2014 Appl. Catal.
A: Gen. 477 147
The authors thank the Department of Science and
Technology (DST), Govt. of India for financial sup-
port under the FTYS scheme (SR/FTP/ETA-0032/2011,
DATED 21.2.2012). Satyanarayana Murty Pudi and
Abdul Zoeb thank MHRD (Govt. of India) for the award
of fellowship.
20. Yue C J, Gu L P, Su Y and Zhu S P 2014 Reac. Kinet.
Mech. Cat. 111 633
21. Xu S, Huang C, Zhang J and Chen B 2009 Korean J.
Chem. Eng. 26 1568
22. Rao G R, Sahu H R and Mishra B G 2003 Colloids Surf.
A: Physicochem. Eng. Aspects 220 261
23. Rao G R, Meher S K, Mishra B G and Charan P H K
2012 Catal. Today 198 140
References
1. Karinen R S and Krause A O I 2006 Appl. Catal. A: Gen.
306 128
2. Meher L C, Gopinath R, Naik S N and Dalai A K 2009
Ind. Eng. Chem. Res. 48 1840
3. Pagliaro M, Ciriminna R, Kimura H, Rossi M and Pina
C D 2007 Angew. Chem. Int. Ed. 46 4434
4. Corma A, Iborra S and Velty A 2007 Chem. Rev. 107
2411
5. Ma L, He D and Li Z 2008 Catal. Commun. 9 2489
6. Miyazawa T, Kusunoki Y, Kunimori K and Tomishige
K 2006 J. Catal. 240 213
7. Maris E P, Ketchie W C, Murayama M and Davis R J
2007 J. Catal. 251 281
8. Balaraju M, Rekha V, Sai Prasad P S, Prabhavathi Devi
B L A, Prasad R B N and Lingaiah N 2009 Appl. Catal.
A: Gen. 354 82
9. Kusunoki Y, Miyazawa T, Kunimori K and Tomishige
K 2005 Catal. Commun. 6 645
10. Gandarias I, Arias P L, Requies J, Guemez M B
and Fierro J L G 2010 Appl. Catal. B: Environ. 97
248
11. Nakagawa Y, Shinmi Y, Koso S and Tomishige K 2010
J. Catal. 272 191
24. Dumitriu E and Hulea V 2003 J. Catal. 218 249
25. Khandan N, Kazemeini M and Aghaziarati M 2008
Appl. Catal. A: Gen. 349 6
26. Yuan Z, Wang J, Wang L, Xie W, Chen P, Hou Z and
Zheng X 2010 Biores. Technol. 101 7088
27. Rode C V, Ghalwadkar A A, Mane R B, Hengne A M,
Jadkar S T and Biradar N S 2010 Org. Process Res. Dev.
14 1385
28. Sato S, Akiyama M, Takahashi R, Hara T, Inui K and
Yokota M 2008 Appl. Catal. A: Gen. 347 186
29. Rekoske J E and Barteau M A 1997 J. Catal. 165 57
30. Shi R, Wang F, Mu X, Li Y, Huang X and Shen W 2009
Catal. Commun. 11 306
31. Huang L, Zhu Y L, Zheng H Y, Ding G Q and Li Y W
2009 Catal. Lett. 131 312
32. Akiyama M, Sato S, Takahashi R, Inui K and Yokota M
2009 Appl. Catal. A: Gen. 371 60
33. Yadav G D, Chandan P A and Devendra P T 2012 Ind.
Eng. Chem. Res. 51 1549