Study on the new hydrogenation catalyst and processes for hydrogen peroxide

through anthraquinone route

Hui Shang
a,
*, Hongjun Zhou
a,
**, Zehua Zhu
b
, Wenhui Zhang
a
a
State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Beijing 102249, PR China
b
SINOPEC CORP. Research Institute of Petroleum Processing, Beijing 100083, PR China
1. Introduction
Hydrogen peroxide (H
2
O
2
) is an important green chemical
product because its utilization does not cause any secondary
pollution. It is widely used in the military industry, textile industry,
papermaking, chemical synthesis, environmental protection, food
processing, medical sterilization and other elds.
H
2
O
2
is commercially produced using the anthraquinone
process [13], where 2-ethylanthraquinone (eAQ) in an organic
solvent is rst hydrogenated in the presence of a supported Pd
catalyst to yield 2-ethylanthrahydroquinone (eAQH
2
) under mild
reaction conditions (temperature 4080 8C, pressure: 0.3 to several
MPa). The eAQH
2
is then oxidized by molecular oxygen to produce
hydrogen peroxide regenerating eAQ in the process at the
temperature of 4060 8C. Fig. 1 details the reactions.
The reactions for H
4
eAQ are similar, with hydrogenation to
H
4
eAQH
2
rst and then oxidized to H
4
eAQ with a molecule of H
2
O
2
production simultaneously.
In working solution, eAQH
2
, eAQ and H
2
O
2
are produced and
separated subsequently via water extraction. The H
2
O
2
is
evaporated and distilled to produce nal hydrogen peroxide
products (27.570% in concentration) [4]. The working solution
from the extraction tower is puried by alkali and treated in a post-
treatment step. Subsequently it is circulated back to the
hydrogenation reactor. Therefore, the working solution contains
active AQs, organic solvent as well as by-products formed during
the hydrogenated and oxidized step.
The regenerated eAQ is again hydrogenated in the next cycle to
obtain another molecule of H
2
O
2
. Although starting with eAQ the
main reaction is as shown above, some eAQH
2
on further
hydrogenation is converted to tetrahydro-2 ethylanthrahydroqui-
none (H
4
eAQH
2
). When H
4
eAQH
2
is oxidized it produces one
molecule of H
2
O
2
and one molecule of tetrahydroanthraquinone
(H
4
eAQ). Like eAQ, this H
4
eAQ is also used again and again to
produce H
2
O
2
. During the circulation, eAQ and H
4
eAQH
2
can also be
degraded by deep hydrogenation of aromatic ring, the hydrolysis of
carbonyl group in inactive AQs, and further deep oxidation into by-
products, which lose the ability to generate H
2
O
2
.
Santacesaria et al. reported that hydrogenation of anthraqui-
none derivatives on Pd catalyst is a fast reaction, and mass transfer
resistance controls the reaction rate in conventional reactors [5].
The hydrogenation reaction was proved to be a zero-order
kinetics for hydrogen, and an rst order kinetic law for eAQ and
H
4
eAQ [58]. Studies by Drelinkiewicz and Waksmundzka [9] used
Pd/SiAl (Pd 2%) revealed that the activation energy was as low as
12.520.9 kJ/mol with the catalysts diameter of 0.060.2 mm and
the activation energy of 29.333.5 kJ/mol with the catalysts
diameter less than 0.06, indicating that the former tests were
inuenced by internal diffusion, whilst the latter tests with smaller
catalyst particles eliminated such inuence.
Journal of Industrial and Engineering Chemistry 18 (2012) 18511857
A R T I C L E I N F O
Article history:
Received 13 October 2011
Accepted 22 April 2012
Available online 27 April 2012
Keywords:
Hydrogen peroxide
Anthraquinone
Hydrogenation catalyst
Trickle bed reactor
A B S T R A C T
The main process for industrial synthesis of hydrogen peroxide is to subject 2-ethylanthraquinone to
cyclic hydrogenation and oxidation. Hydrogenation catalyst is one of the key techniques of
anthraquinone route. A new egg-shell palladium catalyst with racshig-ring alumina supported was
developed for meeting the mechanism and improving mass transformation based on the kinetics and
mass transfer theory, the effects of materials feeding methods together with operating conditions on the
hydrogenation efciency were detailed. It was concluded that the egg-shell palladium catalyst
developed by this research has higher hydrogenation efciency than the commercial contrast catalysts.
2012 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights
reserved.
Abbreviations: AHQ, anthrahydroquinone; AQ, anthraquinones; eAQ, 2-ethylan-
thraquinone; eAQH
2
, 2-ethylanthrahydroquinone; H
4
EAQ, 2-ethyltetrahydroan-
thraquinone; H
4
EAQH
2
, tetrahydro-2-ethylanthrahydroquinone; H
2
O
2
, hydrogen
peroxide; LHSV, liquid hourly space velocity.
* Corresponding author. Tel.: +86 10 89731120; fax: +86 10 89731120.
** Corresponding author. Tel.: +86 10 89731979; fax: +86 10 89731979.
E-mail addresses: [email protected] (H. Shang), [email protected] (H. Zhou).
Contents lists available at SciVerse ScienceDirect
Journal of Industrial and Engineering Chemistry
j ou r n al h o mepag e: w ww. el sevi er . co m / l ocat e/ j i ec
1226-086X/$ see front matter 2012 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jiec.2012.04.017
The activity and selectivity of hydrogenation catalyst determine
the device productive capacity. Side reactions can also be avoided
by improving the catalysts high activity and selectivity. Based on
research results and plant operation experience, a catalyst with a
high surface area and a structure conductive to reactants diffusion
in the bed as well as a process to the benet of external diffusion
are required, both of which are the technology development trend
in anthraquinone hydrogenation.
This paper discusses the most important issues encountered
during the development of industrial AQ process with the
focus on the key step hydrogenation of active anthraquinones.
An innovative catalyst was developed according to the
hydrogenation kinetics within this research and its performance
was investigated.
2. Experimental
The hydrogenation of working solution in a trickle bed reactor
containing very small particles of Pd/Al
2
O
3
catalyst was investi-
gated in this paper. The working solution was a mixture of eAQ,
H
4
eAQ, trioctyl phosphate and aromatics of C9C10, there
contained 22.1 g/L eAQ, and 116.5 g/L H
4
eAQ, the volume ratio
of C9C10 aromatics to trioctyl phosphate is 3:1. The character-
istics of the working solution at 50 8C are: density: 897.76 kg/m
3
,
surface tension: 0.349 N/s, viscosity: 1.0237 10
3
Pa s.
2.1. Catalysts preparation
Raschig ring alumina was selected as a support with a size of
w3 w2 5 (mm), the support was impregnated with PdCl
2
solution in a pH value of 34, after 2 h the sample was washed with
water only until Cl

ions were removed and then dried at room

temperature. The prepared catalyst named as GXH-1 was dried for
16 h at 105 8C in an oven and calcined at 550 8C in a furnace, whose
properties are shown in Table 1.
2.2. Hydrogenation experiments and analytical procedure
The activity test was carried out in a stainless steel reactor with
an internal diameter of 50 mm and height of 620 mm. The
experimental setup is schematically shown in Fig. 2.
The catalyst was reduced by hydrogen at a temperature of 60 8C
and a pressure of 0.2 MPa for 12 h in the reactor, and then cooled
down to a desired temperature. The reactor was rst ushed by
aromatics and then packed by 510 g GXH-1 catalyst with the
height of 600 mm. Nitrogen was used for pressure and leakage
tests. Hydrogen gas employed in this study was obtained from a
gas cylinder with an ultra high purity grade (99.99%). The eAQ
solution was pumped at a pre-determined ow rate. Hydrogen gas
owed from a cylinder through a ow meter. The hydrogen and
working solution were mixed in the mixer and then downstream
or upstream into the reactor packed with GXH-1 catalyst. The
product mixture was collected and the un-reacted hydrogen was
vented to the atmosphere. The liquid product was collected for
analysis.
Hydrogenation efciency was dened as the grams of hydro
peroxide produced by 1 L working solution.
3. Results and discussion
3.1. GXH-1s hydrogenation efciency
The performances of GXH-1 catalyst together with two
commercial catalysts used in China for hydrogen peroxide
Symbols used
B hydrogenation efciency (g/L)
C concentration (mol/L)
C
eAQ
concentration of 2-ethyl-anthraquinone in work-
ing solution (mol/L)
C
0
eAQ
concentration of 2-ethyl-anthraquinone in work-
ing solution (mol/L)
C
eAQH
2
mole concentration of anthraquinone in hydro-
genation solution (mol/L)
D diffusion coefcient (m
2
/s)
D
b
intrinsic diffusion coefcient (m
2
/s)
E activation energy (kJ/mol)
H Henry constant of hydrogen (MPa/(cm
3
mol))
I
0
zero-order Bessel function
I
1
rst-order Bessel function
k constant reaction rate (m
3
/(s kg))
k
g
a
l
gasgas mass transfer coefcient (s
1
)
k
l
a
l
gasliquid mass transfer coefcient (s
1
)
k
s
a
s
liquidsolid mass transfer coefcient (s
1
)
l reactor bed height (m)
m mass (g)
m
Pd
Pd mass (g)
N mass transfer rate (kmol/(m
2
s))
P pressure (MPa)
R constant, R = 8.314 J/(mol K) (J/(mol K))
R
eAQH
2
eAQH
2
s production rate (mol/(gPd L min))
r micro-chemical reaction rate (mol/(m
3
s))
s
0
selectivity
T temperature (8C)
t reaction time (s)
V consumed volume of potassium permanganate by
titration (mL)
V
p
particle volume (m
3
)
A
p
particle external surface area (m
2
)
e the voidage of catalyst bed
h effective factor
w Thiele modulus
t Tortuosity factor
Subscripts
H
2
hydrogen
g gas phase
l liquid phase
s solid phase
Table 1
Characteristics of catalysts GXH-1, industrial contrast agents A and B.
Catalyst A B GXH-1
Shape Sphere Four-leaf clover Raschig ring
Pd (%) 0.30 0.02 0.30 0.02 0.28 0.02
Active layers structure Egg-shell Egg-shell
Pd penetration (mm) 0.3 0.3 0.2
Bulk density (kg/m
3
) 0.78 0.76 0.51
BET surface area (m
2
/g) 100 99 120
Voidage of bed (%) 33.04 34.00 44.14
Strength (N/particle) 50 50 55
Average pore diameter (nm) 10.26 18.24 18.21
Pore volume (cm
3
g
1
) 0.46 0.68 0.63
H. Shang et al. / Journal of Industrial and Engineering Chemistry 18 (2012) 18511857 1852
production were investigated at temperature of 40100 8C,
pressure of 0.2 MPa, hydrogen space rate of 600 h
1
, and liquid
space rate of 12 h
1
. Results are shown in Fig. 3.
Fig. 3 shows that the activities of GXH-1 are much higher than
the other two commercial used catalysts through the temperature
range of 40100 8C. A and B are industrial catalysts used for
contrast agents in this research, which are used in Yueyang
petrochemical plant, their operating conditions are as the
followings: temperature of 4075 8C, pressure of 0.10.35 MPa,
and the LHSV of 712 h
1
.
Fig. 1. Production routes for hydrogen peroxide.
Fig. 2. Laboratory schematic setup for eAQ hydrogenation. 1 hydrogen cylinder; 2, 16 pressure gage; 3 hydrogen ow meter; 4, 9, 10, 13, 14, 18 valve; 5 working
solution container; 6 metering pump; 7 working solution ow meter; 8 mixer; 11 hydrogenation reactor; 12 thermometer; 15 gasliquid separator; 17 release
valve; 19 hydrogenated solution container.
H. Shang et al. / Journal of Industrial and Engineering Chemistry 18 (2012) 18511857 1853
Intrinsic kinetics for the hydrogenation of eAQ and H
4
eAQ are
very important toward the development of AO process, and it was
reported by a few papers that the hydrogenation reaction rate is
rst order with respect to hydrogen partial pressure and zero order
with respect to eAQ and H
4
eAQ concentration [10,11]. Therefore
eAQ and H
4
eAQ can be taken as one single substance due to the
identical hydrogenation kinetics. The reaction rate constant for the
hydrogenation of eAQ in a stirring slurry reactor excluding any
mass transfer resistance are shown in Eq. (1) [12], and eAQs
production rate can be written as Eq. (2):
K
eAQ
K
0
eAQ
exp
E=RT
(1)
R
eAQ

dC
eAQ
dt
hm
Pd
K
eAQ
P
H
2
(2)
The hydrogenation process is thought to act at isothermal
temperature due to no obvious thermal effect of eAQ hydrogena-
tion. The reactants gas and liquid mass balance and momentum
balance can be calculated using axial dispersion model based on
the two-lm theory, and the axial dispersion reactor model is thus
achieved as the followings [13,14].
Mass balance:
Gas phase:
D
H
2
g
d
2
C
H
2
g
dl
2
u
0g
dC
H
2
g
dl
N
H
2
g
0 (3)
Liquid phase:
D
H
2
l
d
2
C
H
2
l
dl
2
u
0l
dC
H
2
l
dl
s N
H
2
l
N
H
2
s
0 (4)
D
eAQl
d
2
C
eAQl
dl
2
u
0l
dC
eAQl
dl
N
eAQs
0 (5)
Solid phase:
N
H
2
s
N
eAQs
R
eAQs
(6)
Momentum balance:
d p
dl

Dp
Dl

lg
(7)
Total effective factor h will be achieved using Ramachandran
and Chaudaris method [7]. If the reaction is totally controlled by
external dispersion, then:
N
H
2
g
k
g
a
l
C
H
2
g
C

H
2
g
(8)
N
H
2
l
k
l
a
l
C

H
2
l
C
H
2
l
(9)
N
H
2
s
k
s
a
s
C
H
2
l
C
H
2
s
(10)
When the gasliquid mixture reaches their balance at the phase
interface, following equation can be obtained according to Henrys
law:
C

H
2
g
HC

H
2
l
(11)
From literature, eAQ and H
4
eAQs hydrogenation reaction rates
are in rst order, the reaction rate can be written as (12) when
internal dispersion is considered:
R
eAQ

dC
eAQ
dt
hm
Pd
K

eAQ
C
H
2
s
(12)
According to the intrinsic reaction kinetics, (2) = (12), then:
K

eAQ
C
H
2
s
K
eAQ
P
H
2
(13)
Taken hydrogen acting as ideal gas and there is no external
dispersion resistance, the hydrogen concentration at catalyst
surface and eAQs reaction rate constant can be written as Eqs. (14)
and (15).
C
H
2
s

P
H
2
RT
(14)
K

eAQ
K
eAQ
RT (15)
The reactions rate should be equal for each step once the
process reaches their steady-state, Eqs. (8)(12) can be combined
and the following equations would be obtained:
R
eAQ

dC
eAQ
dt
K
0
C
H
2
g
(16)
1
K
0
a
s

1
k
g
a
l

H
k
l
a
l

H
k
s
a
s

H
hm
Pd
K

eAQ
(17)
According to Eqs. (15) and (17):
1
K
0
a
s

1
k
g
a
l

H
k
l
a
l

H
k
s
a
s

H
hm
Pd
K
eAQ
RT
(18)
K
0
is the macroreaction rate constant, whose reciprocal is the
process resistance. From Eq. (18), the process resistance is
composed by ve parts, four of which are mass transfer resistance,
and the other one is chemical reaction resistance.
1=k
g
a
l
is the gas-lm resistance, H=k
l
a
l
is the liquid-lm
resistance, H=k
s
a
s
is the liquid-lm resistance at liquidsolid
interface, H=hm
Pd
K
eAQ
RT is the chemical resistance and internal
dispersion resistance.
From the view of new catalysts development, the chemical
resistances contribution from Eq. (18) is really important. The
racshig-ring egg-shell catalyst can reduce the internal diffusion
resistance to a great extent, due to the signicant small catalyst
particles, especially when the particle size reduces to microake
which covers on the surface of the support, the Pds penetration
depth is only 0.2 mm for GXH-1.
Catalysts effective coefcient can be achieved by (19) for the
spherical particles, (20) for at ake particles and (21) for long
cylinder particles:
h
1
tanh3f

1
3f

(19)
Fig. 3. Comparison of catalysts activity.
H. Shang et al. / Journal of Industrial and Engineering Chemistry 18 (2012) 18511857 1854
h
tanh f
f
(20)
h
I
1
2f
fI
0
2f
(21)
f is the Thiele Module, which is dened as:
f V
p
=A
p

eAQ
=D
H
2
e

q
, the value of V
p
/A
p
represents the
equivalent length of catalyst particle. For different particle shapes,
the value of V
p
/A
p
is different, for example, for at plates, the value
is half thickness; for cylinder with ends sealed and sphere, the
value is the radius, whilst for long cylinder, the value is half radius.
The effective coefcient (h) for these three shape catalysts can be
found in Table 2 [15].
If taken the four-leaf clover catalyst as the cylinder, and the
Raschig ring as the at-plate particle, Table 2 clearly shows that the
shape of at-plate contributes to higher h than cylinder and sphere
catalysts; it is certied that GXH-1 catalyst developed by this work
can efciently reduce the internal diffusion resistance.
Due to the fact that the internal diffusion resistance is mainly
originated by particle pore distribution, the pore size distribution
of catalysts A, B and GXH-1 were analyzed through BET. Fig. 4
illustrates the pore size distributions based on BET desorption data.
From Fig. 4 and other BET results, most of the pores in catalyst A
have smaller diameter than those in catalysts B and GXH-1, since
large pore helps to reduce the mass transfer resistance.
Although catalyst B has the similar pore size distribution to
GXH-1, based on the Fick Law [16], the effective diffusion
coefcient D
H
2
e
can be expressed as (22):
D
H
2
e

e
t
D
b
(22)
From Table 1 catalyst GXH-1 has higher voidage than the other
two commercial catalysts, resulting higher effective diffusion
coefcient. Furthermore, the raschig ring catalyst particle has an
internal and external surface, which can signicantly enhance the
contact surface area for chemical reactions, thus reduce the
external diffusion.
From the above discussions, together with Table 1, GXH-1 has
several signicant advantages over the other two commercial
catalysts: 20% higher in surface area, about 30% higher in porosity,
3040% lighter in bulk density, and the Pd penetration depth was
smaller than the commercial catalyst, making its corresponding
hydrogenation efciency better and competitive in industrializing
the AO process.
3.2. Effects of operating parameters on the hydrogenation efciency
Material feeding method such as gasliquid up-ow and down-
ow would inuence the materials residence time in the reactor
and thus affect the reaction. Downstream of working solution and
hydrogen and upstream together with operating conditions were
fully investigated in this section. Two different concentrations
working solution of active anthraquinones were employed for this
research, both of which were sampled from the industry
equipment sideline. One named No. 1 working solution has eAQ
of 239.6 mol/m
3
and H
4
eAQ 366.7 mol/m
3
, the other named No. 2
consists eAQ of 93.4 mol/m
3
and H
4
eAQ of 484.9 mol/m
3
.
3.2.1. Effects of temperature on the hydrogenation efciency
Inuence of temperature together with feeding method on
GXH-1 hydrogenation efciency is shown in Fig. 5, the experi-
ments were carried out with hydrogen pressure of 0.2 MPa, LHSV
of 15 L/h and hydrogen ow rate of 120 L/h.
Fig. 5 indicates that the hydrogenation efciency increases with
temperature increment for both No. 1 and No. 2 solutions,
especially at above 80 8C. The inuence of temperature on the
hydrogenation efciency of No.1 sample is higher than that of No.
2, indicating that the hydrogenations of eAQ and H
4
eAQ occur at
the same time, only because the active energy of H
4
eAQ is lower
than eAQ, the conversion at low temperature is higher, whereas,
eAQs hydrogenation gives the most contribution at higher
temperature. Fig. 4 also clearly indicates that at the same
temperature the hydrogenation efciency from upstream method
is signicant higher than downstream method.
Table 2
Effective coefcient for various shapes of catalysts.
f Internal effective coefcient (h)
Flat-plate Cylinder Sphere
0.1 0.997 0.995 0.994
0.2 0.987 0.981 0.977
0.5 0.924 0.892 0.876
1 0.762 0.698 0.672
2 0.482 0.432 0.416
5 0.200 0.197 0.187
10 0.100 0.100 0.097
Fig. 4. Pore size distributions of catalysts A, B and GXH-1.
Fig. 5. Effects of temperature on the hydrogenation efciency for down and up
stream feeding methods.
H. Shang et al. / Journal of Industrial and Engineering Chemistry 18 (2012) 18511857 1855
3.2.2. Effects of pressure on hydrogenation efciency
The inuence of pressure is shown in Fig. 6. Experiments were
carried out at a temperature of 60 8C, LHSV of 15 L/h and hydrogen
ow rate of 120 L/h using catalyst GXH-1 and No. 2 sample. Fig. 6
shows that hydrogenation efciency goes higher with the pressure
increasing, which is thought due to the fact that high pressure
benets to the gas molecular numbers reduced reaction, therefore
the equilibriums shift to the production direction.
3.2.3. Effects of LHSV on the hydrogenation efciency
The effects of LHSV on the hydrogenation efciency and output
capacity are shown in Fig. 7. The tests were employed at a
temperature of 60 8C, pressure of 0.4 MPa, and hydrogen ow rate
of 2 L/min. It can be seen from Fig. 7 that the hydrogenation
efciency slightly decreases with LHSV increasing, which means
that the contact time may be not enough for eAQs hydrogenation.
Both Figs. 5 and 6 clearly show that at the same temperature the
hydrogenation efciency from upstream method is signicant
higher than downstream method, which may be complained by
the following theory.
Based on Eq. (18), it is obvious that the research to reduce
1=k
g
a
l
, H=k
l
a
l
and H=k
s
a
s
is the key matter.
For downstream process:
The surface area of gasliquid interface increases with gas and
liquid ow rate, a
l
is much higher than the catalysts surface area a
s
.
For trickle area, a
l
can be calculated as (23):
a
l
a
s
0:81
Dp
Dl

lg
e
a
s
!
1:2
(23)
Mass transfer on the trickle bed is high due to the high surface
area of gasliquid interface, not because of the mass transfer
coefcient.
(1) Gas-lm resistance, 1=k
g
a
l
The gas-lm resistance can be ignored if the gas substitutes are
composed by pure gas and liquid components and reactant
production are of non-volatile materials. Therefore, the gas-lm
resistance can be neglected by using the high purity grade
hydrogen (99.99%).
(2) Liquid-lm resistance, H=k
l
a
l
E
t
expresses the energy consumption of liquid:
E
t
D p=Dl
lg
u
0l
. When 0.4 < E
t
< 10 kg/m
2
s:
k
l
a
l
0:0011E
t
D
H
2
l
2:4 10
9
(24)
D
H
2
l
is the molecular diffusion coefcient of hydrogen from gas
phase to liquid phase. When 5 < E
t
< 1000 kg/m
2
s:
k
l
a
l
0:017E
0:5
t
(25)
From (24) and (25) it can be seen that the liquid-lm resistance
can be efciently reduced by improving the liquid ow rate and
raising bed pressure drop, since upstream process has higher bed
pressure drop than down stream process, contributing to reduce
the liquid-lm resistance.
(3) Liquid-lm resistance on the liquidsolid interface H=k
s
a
s
The mass transfer coefcient between liquid phase and
catalysts surface is k
s
a
s
, the correlation is:
esh
0:8
h
t
Re
1=2
l
Sc
1=3
l
(26)
When Re
l
> 10, the mass transfer coefcient on liquidsolid
interface is:
sh 48Re
1=4
g
(27)
Re
l
u
l
d
b
r
l
=em
l
, Schmidt number Sc
l
m
l
=r
l
D
H
2
l
,
Sh k
s
d
b
=D
H
2
l
, the effective diameter of catalyst particle is d
b
,
d
b
ed
p
=1:51 e.
From (24)(27) it can be seen that at the same operating
conditions, gas and liquid velocity has more effects on mass
transfer in a bubble column.
Further discussion was focused on the reactants residence time
on the catalyst bed for both upstream and downstream. The tests
for this propose were conducted on a xed bed column with the
diameter of 80 mm and height of 1585 mm by using water as the
mobile phase at 60 8C. The residence time of catalysts A and GXH-1
were investigated, and the results are shown in Fig. 8. It was found
that the residence times of upstream process for both GXH-1 and
industrial catalyst A are signicantly higher than those of
downstream process, beneting to the hydrogenation perfor-
mance. The experiments also indicate that the residence time from
using catalyst GXH-1 is obviously higher than that from A, which is
Fig. 6. Effects of pressure on the hydrogenation efciency.
Fig. 7. Effects of LHSV on the hydrogenation efciency and output capacity.
H. Shang et al. / Journal of Industrial and Engineering Chemistry 18 (2012) 18511857 1856
thought due to the catalysts shape effects. Therefore, according to
the above discussions, GXH-1 catalyst developed by this research
has excellent hydrogenation performance due to its character-
istics.
4. Conclusion
From above sections, the new catalyst GXH-1, developed
according to kinetic data and mass transfer model, has higher
hydrogenation activity compared to the other two industrial
catalysts.
Temperature, pressure, LHSV as well as material feeding
method have great impact on hydrogenation efciency. The effect
of temperature on hydrogenation efciency is relative small at the
temperature below 80 8C, whereas above 80 8C hydrogenation
efciency increase rapidly with temperature rising. It was also
found that the hydrogenation efciency was enhanced by higher
pressure, the materials feeding methods have markedly inuenced
on the hydrogenation efciency; up-ow feeding method helps
reduce mass transfer resistance, enhancing reactants contact time.
At the hydrogen pressure of 0.2 MPa and the same temperature,
the hydrogenation efciency of up-ow process can be enhanced
30% than that of down-ow process.
References
[1] R. Halder, A. Lawal, Catalysis Today 125 (2007) 48.
[2] R. Aksela, J. Paloniemi, Hydrogenation of a working solution in a hydrogen
peroxide production process, EP: 1245534 A2 (2002).
[3] G. Goor, E. Staab, J. Glenneberg, Process for the preparation of hydrogen peroxide
by the anthraquinone cyclic process, USP: 7238335 (2007).
[4] Q. Chen, Journal of Cleaner Production 14 (2006) 708.
[5] E. Santacesaria, M. Di Serio, A. Russo, U. Leone, R. Velotte, Chemical Engineering
Science 54 (1999) 2799.
[6] T. Berglin, N.H. Sho o n, Industrial and Engineering Chemistry Process Design and
Development 20 (1981) 615.
[7] E. Santacesaria, P. Wilkinson, P. Babini, S. Carrii, Industrial and Engineering
Chemistry Research 27 (1988) 780.
[8] E. Santacesaria, D.M. Serio, R. Velotti, U. Leone, Industrial and Engineering
Chemistry Research 33 (1994) 277.
[9] A. Drelinkiewicz, A. Waksmundzka, Journal of Molecular Catalysis A: Chemical
258 (2006) 1.
[10] Q. Chen, Chemical Engineering and Processing 47 (2008) 787.
[11] Q. Chen, Inorganic Chemical Industry 33 (2001) 20.
[12] G.Z. Liu, Z.T. Mi, Chinese Journal of Chemical Engineering 12 (2004) 372.
[13] S.-F. Li, Chemical and Catalytic Reaction Engineering, Chemical Industry Press,
Beijing, 1986 (in Chinese).
[14] S. Authayanun, W. Pothong, D. Saebea, Y. Patcharavorachot, A. Arpornwichanop,
Journal of Industrial and Engineering Chemistry 14 (2008) 771.
[15] R. Aris, Chemical Engineering Science 50 (1995) 3899.
[16] J.J. Carberry, A. Varma, Chemical Reaction and Reactor Engineering, Marcel
Dekker Inc., New York, 1987.
Fig. 8. Residence time passing through the catalysts bed.
H. Shang et al. / Journal of Industrial and Engineering Chemistry 18 (2012) 18511857 1857