Selective hydrogenolysis of glycerol to 1,2-propanediol: Comparison of batch and continuous process operations

Article abstract of DOI:10.1021/op1001897

The screening of copper chromite catalysts with various promoters such as Al, Zn, and Ba for glycerol hydrogenolysis to 1,2-propanediol (1,2-PDO) in a batch reaction showed that Cu-Cr (Ba) catalyst gave the highest conversion of 34% and selectivity of 84% to 1,2-PDO. In a continuous operation (23 g catalyst bed) the same catalyst showed higher conversion of glycerol and selectivity for 1,2-PDO of 65% and >90%, respectively, with an on-stream activity of ～800 h. Better performance in a continuous operation could be due to the in situ activation of the catalyst, suppression of glycerol cracking to ethylene glycol, as well as excessive hydrogenation of 1,2-PDO to 2-propanol due to lower contact time of 1.3 h as compared to that in a batch operation (5 h). Effects of various process parameters on conversion and selectivity also have been compared for batch and continuous operations.

Full text of DOI:10.1021/op1001897

Organic Process Research & Development 2010, 14, 1385–1392
Selective Hydrogenolysis of Glycerol to 1,2-Propanediol: Comparison of Batch and
Continuous Process Operations†
C. V. Rode,* A. A. Ghalwadkar, R. B. Mane, A. M. Hengne, S. T. Jadkar, and N. S. Biradar
Chemical Engineering and Process DeVelopment, National Chemical Laboratory, Pune - 411008, India
Abstract:
supported on active carbon, SiO2, or Al2O3.3 Among noble
metals, ruthenium on carbon in the presence of amberlyst resin
gave glycerol conversion of 79% with the highest selectivity
of 82% to 1,2- propanediol at 393 K and 80 bar initial H2
pressure for 10 h.4 However, higher reaction pressure, longer
reaction time, and the formation of byproducts such as ethylene
glycol due to C-C bond cleavage favored by Ru make this
catalyst system unviable from a process point of view. Although
sulfur addition to Ru/C catalyst increased the selectivity to 1,2-
PDO, the glycerol conversion was limited to only 45%.4 In the
case of Pt catalyst, in spite of its higher activity for the glycerol
hydrogenolysis, it gave higher selectivity to lactate.5 Since,
copper is responsible for selective cleavage of the C-O bond
without affecting the C-C bond and also it is much cheaper
than Ru or Pt, more attention has been focused on copper-based
catalysts for glycerol hydrogenolysis. Glycerol hydrogenolysis
was first reported using copper and zinc catalysts under very
high pressure and temperature conditions (150 bar; 513-545
K) with 75-85% selectivity to 1,2-PDO while >25% were
cracked products such as EG, ethanol, methanol and lactic acid.6
Chaminand et al.7 reported glycerol conversion of 20% with
>90% selectivity to 1,2-PDO, using Cu-ZnO in the presence
of tungstic acid in 92 h. Bifunctional Cu-ZnO catalysts were
also proposed to obtain 84% selectivity to 1,2-PDO with 23%
glycerol conversion in 12 h.8 Sato et al.9 showed that hydrogen
pretreated Cu/Al2O3 materials could effectively catalyze the
dehydration of glycerol to produce hydroxyacetone where 1,2-
propanediol was formed as a byproduct. Schuster and Eggers-
dorfer, and Cameron et al.10 reported a catalyst system
comprising cobalt, manganese, molybdenum, along with copper
and an inorganic polyacid for achieving a 95% yield of
The screening of copper chromite catalysts with various promoters
such as Al, Zn, and Ba for glycerol hydrogenolysis to 1,2-
propanediol (1,2-PDO) in a batch reaction showed that Cu-Cr
(Ba) catalyst gave the highest conversion of 34% and selectivity
of 84% to 1,2-PDO. In a continuous operation (23 g catalyst bed)
the same catalyst showed higher conversion of glycerol and
selectivity for 1,2-PDO of 65% and >90%, respectively, with an
on-stream activity of ∼800 h. Better performance in a continuous
operation could be due to the in situ activation of the catalyst,
suppression of glycerol cracking to ethylene glycol, as well as
excessive hydrogenation of 1,2-PDO to 2-propanol due to lower
contact time of 1.3 h as compared to that in a batch operation (5
h). Effects of various process parameters on conversion and
selectivity also have been compared for batch and continuous
operations.
Introduction
Catalytic production of propylene glycols is one of the most
attractive downstream applications of glycerol which is a main
byproduct of biodiesel formation.1 Propylene glycol is an
industrially important chemical which is used in unsaturated
polyester resins, functional fluids (antifreeze, de-icing, and heat
transfer), foods, cosmetics, pharmaceuticals, liquid detergents,
flavors and fragrances, tobacco humectants, paints, etc. The use
of propylene glycol in functional fluids is growing because of
the toxicity associated with ethylene glycol-based products to
human and animals. The commercial route for propylene glycol
involves the hydration of petroleum-based propylene oxide
derived by either chlorohydrin process or the hydroperoxide
process.2 Hence, catalytic hydrogenolysis of glycerol to 1,2-
PDO is a sustainable process utilizing a renewable feedstock
as well as substantially improving the process economics of
the biodiesel manufacture.
(3) (a) Furikado, I.; Miyazawa, T.; Koso, S.; Shimao, A.; Kunimori, K.;
Tomishige, K. Green Chem. 2007, 9, 582. (b) Ma, L.; He, D.; Li, Z.
Catal. Commun. 2008, 9, 2489. (c) Lahr, D. G.; Shanks, B. H. J. Catal.
2005, 232, 386.
Consequently, glycerol hydrogenolysis has been studied
using various catalyst systems involving both noble metals such
as Rh, Ru, Pt, Pd as well as other transition metals, mainly
copper, chromium, and zinc in combination with each other or
(4) (a) Miyazawa, T.; Kusunoki, Y.; Kunimori, K.; Tomishige, K. J. Catal.
2006, 240, 213. (b) Miyazawa, T.; Koso, S.; Kunimori, K.; Tomishige,
K. Appl. Catal., A 2007, 318, 244. (c) Miyazawa, T.; Koso, S.;
Kunimori, K.; Tomishige, K. Appl. Catal., A 2007, 329, 30. (d)
Furikado, I.; Miyazawa, T.; Koso, S.; Shimao, A.; Kunimori, K.;
Tomishige, K. Green Chem. 2007, 9, 582.
† This paper is dedicated to the memory of my father Vasant V. Rode who
passed away on 28th August 2010.
(5) Casale, B.; Gomez, A. M. U.S. Pat. 5,214,219, 1993.
(6) (a) Casale, B.; Gomez, A. M. U.S. Pat. 5,276,181, 1994. (b) Casale,
B.; Gomez, A. M. U.S. Pat. 5,214,219, 1993.
* Author to whom correspondence may be sent. E-mail: cv.rode@ncl.res.in.
(1) (a) Corma, A.; Iborra, S.; Velty, A. Chem. ReV. 2007, 107, 2411. (b)
Behr, A.; Eilting, J.; Irawadi, K.; Leschinski, J.; Lindner, F. Green
Chem. 2008, 10, 13.
(7) Chaminand, J.; Djakovitch, L.; Gallezot, P.; Marion, P.; Pinel, C.;
Rosier, C. Green Chem. 2004, 6, 359.
(8) Wang, S.; Liu, H. C. Catal. Lett. 2007, 117, 62.
(2) (a) Martin, A. E.; Murphy, F. H.; Kirk-Othmer Encyclopedia of
Chemical Technology, 4th ed.; Wiley-Interscience: New York, 1994;
Vol. 17, p 715. (b) Trent, D. T. Kirk-Othmer Encyclopedia of Chemical
Technology, 4th ed.; Wiley-Interscience: New York, 1996; Vol. 20, p
271.
(9) Sato, S.; Akiyama, M.; Takahashi, R.; Hara, T.; Inui, K.; Yokota, M.
Appl. Catal., A 2008, 347, 186.
(10) (a) Schuster, L.; Eggersdorfer, M. U.S. Pat. 5,616,817,1997. (b)
Cameron, D. C.; Altaras, N. E.; Hoffman, M. L.; Shaw, A. J.
Biotechnol. Prog. 1998, 14, 116.
10.1021/op1001897  2010 American Chemical Society
Published on Web 09/14/2010
Vol. 14, No. 6, 2010 / Organic Process Research & Development
•
1385

propanediol under severe conditions of 250 bar pressure and
523 K for glycerol hydrogenolysis.
and stability up to 800 h for continuous hydrogenolysis of
glycerol in a fixed bed reactor.
Copper along with chromium forms a copper chromite
system which is also well studied for hydrogenation as well as
for the decomposition or dehydration of alcohols.11,12 After the
first report of copper chromite catalyst by Fleckenstein et al.13
for glycerol hydrogenolysis at very high pressure (250 bar),
several reports have appeared using various types of copper
chromite catalysts.14-16 Commercial as well as proprietary
copper chromite systems are also reported for which the reaction
conditions could be much milder, such as 473 K and 14 bar H2
pressure.13,17 Another possibility with this catalyst system is that,
by altering the composition and reaction conditions, very high
selectivity to intermediate acetol (80%) with 86% glycerol
conversion can be achieved. Recently, a nonchromium Cu/Al
nano-structured catalyst system has been reported for the
hydrogenolysis of glycerol under relatively milder conditions,
giving >47% glycerol conversion in 5 h.18
However, it is evident from the literature that most of the
work on various catalyst systems including promising ones such
as copper chromite has been carried out using a batch reactor
operation. Although few reports describe the continuous dehy-
dration/hydrogenolysis of glycerol, reaction conditions are very
harsh, and/or the major product formed was acetol along with
considerable formation of unidentified side products. Continuous
process for the production of a commodity product such as 1,2-
PDO will be highly desirable from a commercialization point
of view. However, performance of the same catalyst would
differ considerably in a continuous operation from that in a batch
mode. Therefore, the main objectives of our work were: (i) to
study the product distribution and a systematic comparison of
batch and continuous process for glycerol hydrogenolysis over
copper chromite catalyst of a specific composition prepared in
our own laboratory; (ii) to optimize the process conditions for
a continuous operation; (iii) to study the catalyst life. For this
purpose, copper chromite catalysts of different compositions
were prepared with promoters such as Al, Zn, and Ba by the
coprecipitation method. These catalysts were then screened for
glycerol hydrogenolysis in a batch reaction. Among these
catalysts, Cu-Cr with barium as a promoter showed the highest
glycerol conversion of 34% with 84% selectivity to 1, 2
propanediol. Hence this catalyst was used to carry out further
studies on effect of reaction parameters in both batch and
continuous operations. Our catalyst showed an excellent per-
formance (65% conversion with selectivity >90% to 1,2-PDO)
Experimental Section
Materials. Glycerol (99.9%) was supplied by Merck Spe-
cialties, Mumbai, India, and acetol, 1,2-propanediol, and 2-pro-
panol were obtained from Sigma-Aldrich, Bangalore, India.
Copper nitrate, ammonium dichromate, aluminium nitrate, zinc
nitrate, and barium nitrate were purchased from Loba Chemie,
Mumbai, India. Hydrogen and nitrogen of high purity (>99.99%)
were obtained from Inox India and were used directly from
the cylinders.
Catalyst Preparation. Copper chromite catalyst was pre-
pared by a coprecipitation method. The required amounts of
each of Cu (NO3)2 ·3H2O and nitrate precursors of respective
promoters, Al, Zn, or Ba were dissolved in deionized water.
To this solution the aqueous solution of ammonium chromate
was added, which was already prepared by dropwise addition
of 30% aqueous ammonia to an aqueous solution of ammonium
dichromate. The brown precipitate formed was then separated
by filtration and washed with deionized water. The precipitate
thus obtained was dried in a static air oven at 373 K for 8 h
and calcined at 673 K for 4 h.
Procedure. Batch Operation. Batch reactions were carried
out in a Parr autoclave of 300 mL capacity. Typical hydro-
genolysis conditions were the following: temperature, 493 K;
glycerol concentration, 20 wt %; catalyst loading, 1 g; and
hydrogen pressure, 35-70 bar. The prepared catalysts were pre-
reduced under H2 at 473 K for 12 h.
Continuous Operation. Continuous hydrogenolysis of glyc-
erol was carried out in a bench scale, high-pressure, fixed-bed
reactor supplied by M/s Geomechanique, France. A schematic
of the reactor setup is shown in Figure 1. This reactor set up
consisted of a stainless steel single tube of 0.34 m length and
1.5 × 10-2 m inner diameter. The reactor was heated by two
tubular furnaces whose zones (TIC1 and TIC2) were indepen-
dently controlled at the desired bed temperature. The reactor
was provided with mass flow controllers, pressure indicator,
and controller (PIC) devices and two thermocouples to measure
the temperature at two different points. A storage tank was
connected to the HPLC pump through a volumetric buret to
measure the liquid flow rate. The pump had a maximum
capacity of 3 × 10-4 m3/h under a pressure of 100 bar. The
gas-liquid separator was connected to other end of the reactor
through a condenser.
The powdered catalyst was pelletized in the form of pellets
of 1 × 10-2 m diameter and cut into 4 pieces each having 2.5
× 10-3 m diameter. Twenty-three g of the pelletized catalyst
was charged in to the reactor. The section of 7 × 10-2 m above
and 0.16 m below the catalyst bed was packed with carborun-
dum as an inert packing, thus providing the catalyst bed of
0.11 m. Before starting the actual experiment the reactor was
flushed thoroughly, first with N2 and then with H2 at room
temperature. Then the reactor was pressurized with H2 after
attaining the desired temperature. The liquid feed was “switched
on” after the reactor reached the operating pressure and was
kept at that value for 1 h to obtain the constant liquid flow rate.
Liquid samples were withdrawn at regular intervals of time.
Liquid samples were analyzed by GC (Varian 3600) equipped
(11) (a) Thomas, C. L. Catalytic Processes and ProVen Catalysts; Academic
Press: New York, 1970. (b) Zhenga, H.; Yanga, J.; Zhub, Y.; Zhaob,
G. React. Kinet. Catal. Lett. 2004, 82, 263. (c) Jaganathan, R.;
Chaudhari, S. T.; Rode, C. V.; Chaudhari, R. V. Ind. Eng. Chem. Res.
1998, 37, 2099.
(12) Rao, R.; Dandekar, A.; Baker, R. T. K.; Vannice, M. A. J. Catal.
1997, 171, 406.
(13) Fleckenstein, T., Glebel, G.; Carduck, F. J. Ger. Pat. DE4302464, 1994.
(14) Dasari, M. A.; Kiatsimkul, P. P.; Sutterlin, W. R.; Suppes, G. J. Appl.
Catal., A 2005, 281, 225.
(15) Chiu, C. W.; Dasari, M. A.; Suppes, G. J. AIChE J. 2006, 52, 10-
3543.
(16) Chiu, C.; Tekeei, A.; Sutterlin, W. R.; Ronco, J. M.; Suppes, G. J.
AIChE J. 2008, 54, 9-2456.
(17) Suppes, G.; Sutterlin, W.; Dasari, M. A. WO/2005/095536 A2, 2005.
(18) Mane, R. B.; Hengne, A. M.; Ghalwadkar, A. A.; Vijayanand, S.;
Mohite, P. H.; Potdar, H. S.; Rode, C. V. Catal. Lett. 2010, 135, 141.
1386
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development

Figure 1. A schematic of the fixed-bed reactor setup.
Table 1. Specifications of copper chromite catalyst
Table 2. Range of operating conditions
Range
continuous
20-60
2-propanol 2-propanol
type
Cu/Cr
parameter
batch
Form
size, m
pellets
0.005 × 0.002
54.24
1.35
Al, Ba, and Zn
Initial concentration of glycerol, wt% 20-60
surface area, m2/g
bulk density, g/mL
promoters, 30%
solvent
temperature, K
H2 pressure, bar
453-513
453-513
20-60
23 g
0.78-2.34
434-1304
0.11
35-70
catalyst wt, g
0.8 - 2 g
liquid velocity, Ul LHSV, h-1
gas velocity, Ug GHSV, h-1
catalyst packing length, m
particle diameter, dp, m
density of the catalyst, kg/m3
with a flame ionization detector and a capillary column (HP-
FFAP 30 m, 0.53 mm, 1 µm).
Following this procedure, the experiments were carried out
at different inlet conditions of liquid and gas flow rates. The
reactor was operated in the temperature and pressure ranges of
453-513 K and 20-60 bar, respectively. Steady-state perfor-
mance of the reactor was observed by analysis of the reactant
and products in the exit stream.
0.004
1685
Table 3. Catalyst screening in a batch reactora
% selectivity
%
1, 2-
catalyst components conversion PDO acetol EG 2-propanol methanol
NMT 002 Cu, Cr (Al)
NMT 005 Cu, Cr (Ba)
NMT 006 Cu, Cr
31
34
16
29
81
84
82
77
5
3
6
3
4
3
1
9
6
6
3
2
3
3
Results and Discussion
Major objectives of this work were (i) screening of several
Cu-Cr systems and studying the product distribution for
glycerol hydrogenolysis in a batch reactor; (ii) evaluating time
of on-stream activity and product distribution of the best catalyst
for continuous glycerol hydrogenolysis; (iii) optimizing reaction
conditions for the best catalyst in both batch and continuous
hydrogenolysis of glycerol and to compare their performances.
For this purpose, Cu-Cr catalysts were prepared with various
promoters and the specifications of these catalysts are given in
Table 1. The effects of temperature, H2 pressure, liquid and
gas flow rates and glycerol concentration on the conversion of
glycerol and the product selectivities were also investigated.
The range of reaction conditions used for this study is given in
Table 2. In all the experiments, concentrations of the reactant
and products in the reactor effluent were determined quantita-
tively by GC on the basis of which the conversion of glycerol
and product selectivities were calculated. Reproducibility of the
rate, conversion, and selectivity measurements were found to
be within 3-4% error as indicated by a few repeated experiments.
NMT 011 Cu, Cr (Zn)
19
-
a Reaction conditions: temperature, 220°C; 20 wt % glycerol; pH2, 52 bar;
solvent, 2-propanol; catalyst, 0.01g/mL; reaction time, 5 h.
The results on catalyst screening in a batch reactor are
summarized in Table 3, in which various promoters used in
Cu-Cr catalysts are shown in the parentheses. As can be seen
from this Table, Cu-Cr catalyst without any promoter gave
the lowest glycerol conversion of 16%, while Cu-Cr catalysts
with Al and Zn promoters gave almost similar glycerol
conversions of 31 and 29% respectively. Al and Zn promoters
were chosen mainly due to their Lewis acid characteristics that
were expected to play an important role in the glycerol
hydrogenolysis, since it involves the first step of dehydration
of glycerol to acetol which is an acid-catalyzed reaction.
However, Cu-Cr with a barium promoter was found to give
the highest glycerol conversion of 34% with >84% selectivity
to 1,2-propanediol. Probably, the stability of Cu-Cr catalyst
Vol. 14, No. 6, 2010 / Organic Process Research & Development
•
1387

Figure 2. Concentration vs time profile of glycerol hydro-
genolysis in a batch reaction Reaction conditions: temperature,
493 K; 20 wt % glycerol; pH2, 52 bar; solvent, 2-propanol;
catalyst, 0.01 g/mL; reaction time, 5 h.
Figure 3. Concentration vs time profile of glycerol hydro-
genolysis in batch reaction using n-butanol solvent. Reaction
conditions: temperature, 493 K; 20 wt % glycerol; pH2, 52 bar;
solvent, 1-butanol; catalyst, 0.01 g/mL; reaction time, 5 h.
against sintering imparted by barium seems to be more
important than the acidity in enhancing the activity of these
catalysts in hydrogenolysis of glycerol.19 This was also evi-
denced by the consistent activity of Cu-Cr (Ba) catalyst for
about 800 h for continuous glycerol hydrogenolysis. Therefore,
all the further studies for both batch and continuous glycerol
hydrogenolysis were carried out using Cu-Cr (Ba) catalyst.
Batch Reactor Performance. Product Distribution. Figure
2 shows the conversion, selectivity vs time profile for liquid-
phase glycerol hydrogenolysis over Cu-Cr (Ba) catalyst in
2-propanol at 493 K, pH2 of 52 bar for 5 h. Glycerol conversion
increased from 10 to 34% with increase in reaction time from
1 to 5 h, beyond which no effect of reaction time was observed
on the conversion levels. For Cu-Cr catalysts, usually very
long reaction times of >12 h are reported;14 however, our
catalyst gave a maximum conversion of 34% within a very short
reaction time of 5 h. Another novelty of our catalyst was that
the highest selectivity of >84% to 1,2-PDO was achieved, and
from the beginning of the reaction it was ∼>4 times higher
than that of the intermediate dehydration product, acetol,
indicating that the hydrogenation of acetol to 1,2-PDO is a very
fast reaction. The highest selectivity of acetol formed was ∼15%
in one h, after which it started decreasing and was stable at
∼5%. The only side products formed under the conditions of
the present work were ethylene glycol, methanol, and 2-pro-
panol. Two separate hydrogenolysis experiments starting with
1,2-PDO and glycerol were carried out using n-butanol as a
solvent, in order to confirm the pathway of formation of
2-propanol and ethylene glycol. In case of hydrogenolysis of
1,2-PDO, the only product formed was 2-propanol, confirming
that 2-propoanol was a product of further hydrogenolysis of
primary -OH, and no cleavage products were formed from
1,2-PDO. As shown in Figure 3, glycerol hydrogenolysis gave
2-propanol, ethylene glycol, and methanol along with acetol
(minor) and 1,2-PDO as a major product. It was interesting to
note (Figure 3) that the selectivity to 1,2-PDO reached a
maximum value of 86% in about 2 h, after which it marginally
decreased to 84% due to the onset of formation of 2-propanol.
This indicates that after a certain critical concentration of 1,2-
PDO was reached, with a longer contact time its further
hydrogenation to 2-propanol started. Also the formation of
ethylene glycol was due to cleavage of C-C bond directly from
glycerol and not from the 1,2-PDO, while 2-propanol was
formed at the cost of excessive hydrogenolysis of 1,2-PDO.20
As expected, methanol formation was also observed due to the
hydrogenation of formaldehyde formed during C-C bond
cleavage to give ethylene glycol. These results as well as no
formation of the gaseous product, viz. methane, justify the
reaction pathway for the hydrogenolysis of glycerol shown in
Scheme 1.
Effect of Temperature. The effect of temperature on both
glycerol conversion and 1,2-PDO selectivity was studied in the
temperature range of 453-513 K for Cu-Cr (Ba) catalyst, and
the results are shown in Figure 4. Glycerol conversion increased
from 18 to 43% with increase in temperature from 453 to 513
K. As expected, the selectivity to 1,2-PDO decreased from 86
to 78% due to the excessive hydrogenation of 1,2-PDO favored
at higher temperature to give 2-propanol. Also the selectivities
to ethylene glycol and methanol increased from 0 to 6%, and
0 to 4%, respectively, indicating that the degradation due to
C-C cleavage of glycerol was more facile at high temperature.
An Arrhenius plot was also obtained as shown in Figure 5, from
which the activation energy was estimated to be 147.5 kJ mol-1.
Effect of Pressure. The results of the effect of hydrogen
pressure on glycerol hydrogenolysis studied at a constant
temperature of 493 K are shown in Figure 6. Glycerol
conversion decreased marginally from 34% to 28% with an
increase in H2 pressure from 35 to 67 bar, indicating that the
hydrogen atoms at a higher pressure compete with the read-
sorption of an intermediate, acetol for the active catalyst sites
causing the decrease in conversion. Increase in hydrogen
pressure in the higher range from 50 to 67 bar resulted in
(19) Bianchi, C. L.; Cattania, M. G.; Ragaini, V. Surf. Interface Anal. 1992,
19, 533.
(20) Kusunoki, Y.; Miyazawa, T.; Kunimori, K.; Tomishige, K. Catal.
Comm. 2005, 6, 6645.
1388
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development

Scheme 1. Reaction pathway for hydrogenolysis of glycerol
decreasing the selectivity to 1,2-PDO from 84 to 77% due to
enhanced further hydrogenation of 1,2-PDO to 2-propanol. It
is interesting to note that an increase in H2 pressure did not
affect the selectivity to the degradation product EG because
copper is responsible for selective cleavage of C-O bond
without affecting C-C bond. This was also confirmed by the
study of catalyst loading effect discussed below.
Effect of Catalyst Loading. The effect of catalyst loading
on the hydrogenolysis reaction was studied in the range of
0.8-2.0 g at 493 K and 52 bar of hydrogen pressure, and the
results are presented in Figure 7. It was observed that the
conversion of glycerol increased from 31 to 47% with increase
in catalyst weight from 0.8 to 2 g. Selectivity to 1,2-PDO
initially increased from 81 to 84% for an increase in catalyst
weight from 0.8 to 1 g, while at the same time acetol selectivity
decreased, indicating that the rate of hydrogenation of acetol
to 1,2-PDO increased at higher catalyst loading. The selectivity
to the degradation product EG remained constant, irrespective
of the catalyst loading, confirming that a copper-based catalyst
does not favor C-C bond cleavage.
Effect of Glycerol Loading. The effect of glycerol con-
centrations on the conversion of glycerol and selectivity pattern
was also studied in the range of 20-60 wt % glycerol, and the
results are shown in Figure 8. The conversion of glycerol
decreased drastically from 34 to 11%, while the selectivity to
1,2-PDO marginally decreased from 84 to 82% with increase
in glycerol loading from 20-60 wt %. Interestingly, ethylene
glycol selectivity increased by twice (from 3 to 6%), while that
Figure 6. Effect of H2 pressure. Reaction conditions: glycerol,
20 wt %; temperature, 493 K; solvent, 2-propanol; catalyst,
0.01 g/mL; reaction time, 5 h.
Figure 4. Effect of temperature on conversion and selectivity
Reaction conditions: glycerol, 20 wt %; pH2, 52 bar; solvent,
2-propanol; catalyst, 0.01 g/mL; reaction time, 5 h.
Figure 5. Plot of ln R vs 1/T. Reaction conditions: glycerol, 20
wt %; pH2, 52 bar; solvent, 2- propanol; catalyst, 0.01 g/mL;
reaction time, 5 h.
Figure 7. Effect of catalyst loading. Reaction conditions:
glycerol, 20 wt %; temperature, 493 K; pH2,52 bar; solvent,
2-propanol; reaction time, 5 h.
Vol. 14, No. 6, 2010 / Organic Process Research & Development
•
1389

Figure 8. Effect of glycerol loading. Reaction conditions:
temperature, 493 K; pH2, 52 bar; catalyst, 0.01 g/mL; solvent,
2-propanol; reaction time, 5 h.
Figure 10. Concentration vs time profile for continuous glycerol
hydrogenolysis. Reaction conditions: catalyst wt, 23 g; solvent,
2-propanol; 20 wt % glycerol; feed flow rate, 30 mL/h; H2 flow
rate, 10 NL/h; pressure, 30 bar; temperature, 493 K.
Figure 9. Time on stream activity for continuous glycerol
hydrogenolysis. Reaction conditions: catalyst wt, 23 g; solvent,
2-propanol; 20 wt % glycerol; feed flow rate, 30 mL/h; H2 flow
rate, 10 NL/h; pressure, 40 bar; temperature, 493 K.
of 2-propanol decreased substantially from 7 to 3% as the
concentration of glycerol increased from 20-60 wt %. Hence,
20 wt % of glycerol was found to be the optimum glycerol
concentration. The substantial decrease in glycerol conversion
could be due to the limiting number of catalytic sites at higher
glycerol concentration. Miyazawa et al. also reported a high
glycerol conversion at lower glycerol concentration of 10%.4a
Continuous Reactor Performance. As mentioned above,
our Cu-Cr (Ba) catalyst was found to be the best catalyst during
screening in a batch operation; hence, its time on stream activity
was also evaluated for the continuous hydrogenolysis of
glycerol. Figure 9 shows that this catalyst gave an excellent
performance for 800 h for continuous operation in 2-propanol
under 493 K, and 40 bar hydrogen pressure conditions, with
an average glycerol conversion of ∼65% and >91% selectivity
to 1,2-PDO. The product distribution pattern studied in 2-pro-
panol solvent is shown in Figure 10. The consistent performance
of the catalyst was without any deactivation-reactivation cycle
indicating that the activity of the catalyst was being maintained
by in situ activation in a continuous flow of hydrogen and also
resulted in higher activity (>65% conversion) as compared to
that in a batch operation (34% conversion). As shown by a
concentration vs time profile in Figure 11 formation of
2-propanol was again confirmed by a separate glycerol hydro-
genolysis experiment in butanol solvent. Interestingly, much
Figure 11. Concentration vs time profile for continuous glycerol
hydrogenolysis in n-butanol Reaction conditions: catalyst wt,
20 g; solvent, n-butanol; 20 wt % glycerol; feed flow rate, 30
mL/h; H2 flow rate, 10 NL/h; pressure, 30 bar; temperature,
493 K.
less (<2%) formation of EG was observed in the continuous
glycerol hydrogenolysis due the lower contact time (1.3 h) as
compared to that in the case of a batch operation (5 h). The
lower EG formation also led to a very negligible (<0.8%)
formation of methanol in the continuous operation and hence
could not be shown in concentration vs time profiles. The effects
of various reaction parameters on glycerol conversion and the
selectivity pattern are discussed below.
Effect of Liquid Flow Rate. The effect of liquid flow rate
on conversion of glycerol was studied in the range of 18-54
mL/h, under constant temperature, H2 pressure, and gas flow
rate conditions, and the results are shown in Figure 12. With
increase in liquid flow rate from 18 to 54 mL/h, glycerol
conversion decreased proportionately. The reason for this
decrease in conversion is the reduction in residence time of
glycerol in the reactor with an increase in the liquid flow rate.
Therefore, less time was available for the intimate contact of
H2 with liquid glycerol and the catalyst pellets, thus inhibiting
the substrate diffusion. At lower liquid flow rate, catalyst
1390
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development

Figure 14. Effect of hydrogen flow rate. Reaction conditions:
catalyst wt, 23 g; solvent 2-propanol; 20 wt % glycerol; feed
flow rate, 30 mL/h; pressure, 40 bar; temperature, 493 K.
Figure 12. Effect of liquid flow rate. Reaction conditions:
catalyst wt, 23 g; solvent, 2-propanol; 20 wt % glycerol; H2
flow rate, 10 NL/h; pressure, 40 bar; temperature, 493 K.
Figure 15. Effect of temperature. Reaction conditions: catalyst
wt, 23 g; solvent 2-propanol; 20 wt % glycerol; feed flow rate,
30 mL/h; H2 flow rate, 10 NL/h; H2 pressure, 40 bar.
Figure 13. Effect of hydrogen pressure. Reaction conditions:
catalyst wt, 23 g; solvent, 2-propanol; 20 wt % glycerol; feed
flow rate, 30 mL/h; H2 flow rate, 10 NL/h; temperature,
493 K.
NL/h under constant temperature and pressure conditions. From
Figure 14, it was observed that there was not any effect of gas
flow rate on glycerol conversion. The product selectivities also
remained constant since the change in gas flow rate did not
affect the reaction rate to a considerable extent.
Effect of Temperature. The effect of temperature was
studied over the range of 453 to 513 K by keeping other
parameters constant. It was observed from Figure 15 that
reaction temperature had a considerable effect on the conversion
of glycerol. However, as compared to that of the batch
operation, the effect of temperature on the glycerol conversion
was highly predominant in a continuous operation as evidenced
by about more than 6 times increase in glycerol conversion
(from 13 to 87%) as temperature increased from 453-513 K,
while the selectivity to 1,2-propanediol decreased marginally
from 95 to 86%, correspondingly increasing the selectivity to
the cracking product such as ethylene glycol.
Effect of Glycerol Concentration. Effect of glycerol
concentration was also studied by varying the wt % of glycerol
from 20 to 60% in liquid feed, and the results are shown in
Figure 16. Although similar to that of the batch operation where
glycerol conversion initially decreased from 68 to 40% with
an increase in concentration from 20 to 40 wt %, it remained
constant for further increase in glycerol loading up to 60 wt %.
particles were partially wetted; therefore, there would be direct
contact of gas-phase reactant and catalyst surface which
increases the reaction rate. Subsequently at higher liquid flow
rate, the increased wetted fractions of catalyst surface would
slow down the reaction rate. A similar observation was reported
for continuous hydrogenation of 2-butyne-1,4-diol by Rode
et al.21
Effect of H2 Pressure. The effect of H2 pressure was studied
at constant temperature of 493 K in the range of 20-60 bar H2
pressure, and the results are shown in Figure 13. Unlike the
results in a batch operation, (Figure 6), glycerol conversion
increased from 64 to 74% with an increase in H2 partial pressure
from 20 to 60 bar, in a continuous operation. At higher H2
pressure, in situ reactivation of the catalyst could be more
efficient, leading to higher glycerol conversion. 1, 2-Propanediol
selectivity initially increased from 82 to 91%, while the
selectivity to acetol decreased from 10 to 3% with an increase
in H2 pressure from 20 to 40 bar, indicating a higher hydro-
genation rate at higher H2 pressure.
Effect of Gas Flow Rate. Effect of hydrogen gas flow rate
on glycerol hydrogenolysis was studied in the range of 10 -30
(21) Rode, C. V.; Tayade, P. R.; Nadgeri, J. M.; Jaganathan, R.; Chaudhari,
R. V. Org. Process Res. DeV. 2006, 10, 278.
Vol. 14, No. 6, 2010 / Organic Process Research & Development
•
1391

Conclusions
Glycerol hydrogenolysis over Cu-Cr (Ba) catalyst in a
continuous operation was carried out for a total period of 800 h
without any deactivation-activation cycle, which also showed
better performance than that in a batch operation. Product
distribution studies in a batch operation confirmed that 2-pro-
panol was formed due to excessive hydrogenation of 1,2-PDO,
while the C-C bond cleavage in glycerol led to the formation
of ethylene glycol and methanol. Selectivity of >90% to 1,2-
PDO was achieved in a continuous operation higher than that
in a batch operation (84%) due to elimination of the formation
of ethylene glycol. A study of the effect of process conditions
in a continuous operation revealed that glycerol conversion and
selectivity to 1,2-PDO increased from 65 to 74% and from 82
to 91%, respectively, with an increase in H2 partial pressure
from 20 to 60 bar, while the selectivity to acetol decreased from
10 to 3%, indicating a higher hydrogenation rate at higher H2
pressure.
Figure 16. Effect of glycerol loading. Reaction conditions:
catalyst wt, 23 g; solvent, 2-propanol; feed flow rate, 30 mL/h;
H2 flow rate, 10 NL/h; H2 pressure, 40 bar; temperature,
493 K.
Thus, the effect of glycerol loading on the catalyst activity was
much milder in a continuous operation, which could be due to
the fact that the continuous flow of hydrogen leads to the catalyst
surface renewal inhibiting the strong adsorption of the substrate.
Another observation different from that in the batch operation
was that the selectivity to 1,2-PDO remained constant at 90%
with initial increase in glycerol loading from 20 to 40% and
then decreased to 85% with further increase in glycerol loading
up to 60%. At this glycerol loading, corresponding formation
(∼10%) of other side products was observed.
Acknowledgment
We gratefully acknowledge the Council of Scientific and
Industrial Research, New Delhi, for its financial support to this
work under NMITLI program (TLP 002326).
Received for review July 3, 2010.
OP1001897
1392
•
Vol. 14, No. 6, 2010 / Organic Process Research & Development