Synthesis of dimethyl ether from CO and H2

Article abstract of DOI:10.1134/S0036024411070168

Thermodynamic conditions for synthesizing dimethyl ether from synthesis gas are determined. The optimum conditions of the process are as follow: T ～ 300°C at p = 3 MPa for two catalysts loaded into the reactor: methanol synthesis catalyst (Katalco-58) and catalyst of methanol dehydration to dimethyl ether (γ-Al₂O₃). The changes that occur with the catalysts during this process are demonstrated by electron scanning microscopy.

Full text of DOI:10.1134/S0036024411070168

ISSN 0036ꢀ0244, Russian Journal of Physical Chemistry A, 2011, Vol. 85, No. 7, pp. 1140–1144. © Pleiades Publishing, Ltd., 2011.
Original Russian Text © N.I. Kosova, L.N. Kurina, L.P. Shilyaeva, 2011, published in Zhurnal Fizicheskoi Khimii, 2011, Vol. 85, No. 7, pp. 1246–1250.
CHEMICAL KINETICS
AND CATALYSIS
Synthesis of Dimethyl Ether from CO and H2
N. I. Kosova, L. N. Kurina, and L. P. Shilyaeva
Tomsk State University, Tomsk, 634050 Russia
eꢀmail: kosova@mail.tsu.ru
Received August 5, 2010
Abstract—Thermodynamic conditions for synthesizing dimethyl ether from synthesis gas are determined.
The optimum conditions of the process are as follow:
T ~ 300°C at p = 3 MPa for two catalysts loaded into
the reactor: methanol synthesis catalyst (Katalcoꢀ58) and catalyst of methanol dehydration to dimethyl ether
ꢀAl O3). The changes that occur with the catalysts during this process are demonstrated by electron scanꢀ
(γ
2
ning microscopy.
Keywords: synthesis of dimethyl ether, thermodynamic conditions, catalysis.
DOI: 10.1134/S0036024411070168
INTRODUCTION
EXPERIMENTAL
Commercial catalyst Katalcoꢀ58 (Johnson Mattley
Catalysts), referred to below as Rꢀ1, was used for
Ecological problems and the depletion of oil
reserves are forcing researchers to seek alternative
energy resources. The synthesis of dimethyl ether
methanol synthesis, and ꢀAl O3 (Institute of Catalyꢀ
γ
2
sis, Siberian Branch, Russian Academy of Sciences,
Novosibirsk) was used as a catalyst for methanol dehyꢀ
sible way of replacing oil stock. DME has been widely dration. The phase composition of the samples was
used as a refrigerant, extractant, propellant, and interꢀ determined on a Shimadzu XRD 6000 Xꢀray diffracꢀ
(
DME) from COꢀcontaining gas is considered as posꢀ
mediate product in organic synthesis. Its low boiling tometer with a copper anode (Сu
temperature (–25
°С) allows us to accelerate the proꢀ
К ). The obtained
data were compared with the tabulated values accordꢀ
ing to the ASTM card file using the Match software
α
cesses of mixture formation and combustion, reducing
the ignition delay period and ensuring good engine
start at low ambient temperatures. The high oxygen
content in DME (35%) ensures the smokeless comꢀ
bustion of fuel without formation of solid particles,
preventing the atmosphere from being polluted with
exhaust gases [1].
(
Phase Identification from Powder Diffraction, Verꢀ
sion 2.1). Rꢀ1 was shown to have a composition of
CuO/ZnO/Al О3; alumina was used as the form.
γ
2
The porous structure and specific surface area of
the samples prior to and after conducting catalytic
studies were determined by BET using a TriStar II
(
3020) automated gas adsorption analyzer. The distriꢀ
bution of acid sites was obtained by ammonium desꢀ
orption (table).
There are two technologically possible ways of synꢀ
thesizing DME from synthesis gas: (1) in two series
rectors for methanol synthesis and its dehydration; (2)
in one reactor over a bifunctional catalyst, due to the
simultaneous occurrence of both methanol synthesis
reaction and its dehydration to DME. The catalysts
Prior to the experiment, the catalyst was reduced in
a hydrogen–nitrogen mixture (10 vol % H + N2) at
2
300°С for 3 h. The recovery procedure was selected
used are an oxide catalyst for methanol synthesis and Characteristics of Rꢀ1 and ꢀAl O catalysts prior to and afꢀ
γ
2 3
ter catalysis
Al O or zeolites for dehydration of the resultant
2
3
methanol to dimethyl ether. Even though hybrid cataꢀ
lysts have been designed, research continues into comꢀ
mercial methanol synthesis catalysts, such as SNMꢀ1
2
Catalyst
prior to
after
sspec, m /g
V
, cm/g
R
,
Å
β
Rꢀ1
110
87
0.29
0.27
0.71
0.68
81
0.064
–
125
77
[
2] and ICI, KTM, SüdChemi [3] as components for
γ
ꢀAl O3 prior to
250
311
0.254
–
methanol synthesis followed by dehydration into
DME on an acid catalyst. In this work, we study the
process of producing DME from synthesis gas from a
2
after
87
Note: Sspec is specific surface area, V is pore volume, R is pore
radius, and β is catalyst acidity, mmol NH3/g.
mixture of methanol catalyst Katalcoꢀ58 and
γ
ꢀAl O3
2
.
1140

SYNTHESIS OF DIMETHYL ETHER FROM CO AND H2
1141
p
T
ture flow (10% H2 in Ar); the linear heating rate was
0 K/min to 350 . The signal was measured using a
−
Δ
G
,
kJ/mol
1
°С
thermal conductivity detector. Water vapor was frozen
out in a trap; this eliminated the error contributed to
the measurement of the mixture’s thermal conductivꢀ
ity from the water formed in the reaction and released
from the sample.
0
4
0
0
1
2
3
4
5
A setup enabling us to perform the process under
elevated pressure was used for our catalytic studies.
The volume of the loaded catalysts (1–0.5 mm fracꢀ
8
3
tion) was 5 cm . Hydrogen and carbon monoxide were
fed from gas bottles. The purity of the outgoing gases
and the composition of the resultant products were
determined chromatographically. The investigation of
morphological features of the samples prior and after
the catalytic experiments was performed by scanning
electron microscopy (SEM) on a VEGA II LMU elecꢀ
tron microscope with a primary electron beam energy
of 20 keV.
1
20
3
00
400
500
600
, K
T
p
Fig. 1. The calculated Gibbs energy values (ΔGT ) of reacꢀ
tion (3) over the 300–600 K temperature range at ( ) 1,
) 2, ( ) 3, ( ) 5, and ( ) 6 MPa.
1
(2
3
4
5
from the results of temperatureꢀprogrammed reducꢀ
tion (TPR) of the original catalyst for methanol synꢀ
thesis, which showed reduction peaks in the 150–
RESULTS AND DISCUSSION
3
00
°
C
temperature range. TPR spectra were obtained
Thermodynamics. DME synthesis from the CO+ Н2
on a Chemisorb 2750 chemisorption analyzer. A mixture comprises the following key reactions: synꢀ
weighed portion of the sample was reduced in a mixꢀ thesis of methanol, its dehydration, direct DME synꢀ
1
2
259
1
2
a
'
'
235
1
3
2
1
3
b
30
40
50
2θ, deg
208
a
200
217
b
0
100
200
300
400
T, °C
Fig. 2. The TPR spectrum of the initial (
h, and ( ) after a catalytic experiment. The XRD patterns of (
and ( ) Cu.
а
) catalyst Rꢀ1 reduced in a hydrogen–argon mixture (17 vol % H2 + Ar) at 300
°С for
3
b
a
'
) the initial Rꢀ1 sample and ( ) after catalysis: ( ) ZnO, ( ) CuO,
b'
1
2
3
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 85
No. 7 2011

1142
KOSOVA et al.
system starting from 3 MPa. At pressures below
MPa, the occurrence of the reaction is possible
%
2
below 550–575 K.
1
6
4
2
0
0
0
Temperatureꢀprogrammed reduction (TPR). Figure 2
shows the results from TPR of the methanol catalyst
Rꢀ1 prior to and after catalysis. As can be seen, the
original sample is reduced over the 150–300
Since the sample contains copper, the reduction peaks
with = 208, 235, and 259 (Fig. 2, curve ) are
probably associated with the reduction of copper from
°С range.
Т
°С
а
2
3
2+
+
0
the oxidized to the metallic state: Cu –Cu –Cu [6].
The amount of hydrogen absorbed was 0.003533 mol/g.
A considerably smaller amount of hydrogen
(
0.000562 mol/g) was required for the reduction of
the sample after catalysis; two peaks with = 200 and
17 were observed (Fig. 2, curve ).
It is likely that a small amount of copper remained
0
4
Т
2
°
С
b
60
500
540
580
, K
T
in the oxidized state within the catalyst during catalyꢀ
sis. The XRD data presented in Fig. 2 provide support
for this assumption; according to Fig. 2, the initial catꢀ
Fig. 3. The effect of temperature on (
) DME yield. Experimental conditions:
H /CO = 2, = 6 s; ( ) concentration of CO2, vol %.
1
) CO conversion and
(
2
2
p
= 3 MPa,
τ
3
cont
alyst Rꢀ1 (Fig. 2, curve
and zinc oxides, while two new peaks characteristic of
metallic copper (Fig. 2, curve ) and negligible peaks
а') consists mainly of copper
b'
thesis from synthesis gas, and a steam conversion reacꢀ
tion:
of copper oxide are detected on the XRD pattern after
the catalytic experiments.
CO + 2H2
→
CH OH,
3
Δ
H
= –90.7 kJ/mol, (1)
Catalysis. The catalytic studies were performed
with a hydrogen : carbon (II) oxide ratio of two. In
order to ensure the complete reaction of the initial
mixture with the catalyst surface, the following
method for catalyst loading was used:
First bed: copperꢀcontaining catalyst Rꢀ1 for
methanol synthesis;
2CH OH
3
→
CH OCH + H O,
2
3
3
(2)
(3)
(4)
Δ
H = –23.4 kJ/mol,
2CO + 4H2
→
CH OCH + H O,
3
3
2
Δ
H
= –207.4 kJ/mol,
Second bed: a mixture of methanol catalyst Rꢀ1
and
of the nonreacted CO and Н2
Third layer: alumina ꢀAl O3 for methanol dehyꢀ
γ
ꢀAl O3 for methanol dehydration and conversion
H O + CO → CO + H ,
2
2
2
2
;
Δ
H = –41.2 kJ/mol.
γ
2
dration into dimethyl ether.
The production of dimethyl ether directly from synꢀ
thesis gas (3) is accompanied by heat release and a
Figure 3 shows the effect of temperature on the
reduction in the number of moles; this process is yield of dimethyl ether, CO conversion, and formation
therefore likely to be efficient at low temperatures and of byꢀproduct СО2 under pressure of 3 MPa, Н /СО
2
high pressures.
ratio = 2, contact time of 6 s. These conditions were
selected in the course of the experiment as the optiꢀ
mum conditions. As can be seen, CO conversion and
the yield of dimethyl ether attain their maximum valꢀ
ues (68 and 18%, respectively) upon a single pass of the
p
The changes in the Gibbs energy (ΔGT ) for the
direct production of dimethyl ether from synthesis gas
were calculated on the basis of Eq. (5) [4]:
p
T
0
T
CO–H mixture through the reactor at 300°С. The
main contribution to the formation of carbon dioxide
as a byꢀproduct is made by the reaction of steam conꢀ
2
ΔG = ΔG + ΔnRT ln(p p )
,
(5)
0
p
where ΔGT is the Gibbs energy at specified pressure
p
version (4), since H O under these conditions interꢀ
2
0
(
Pa) and temperature Т (K), kJ/mol; ΔGT is the acts with CO, yielding hydrogen and carbon dioxide.
change in the Gibbs energy calculated using the The accumulation of СО2 in the system can be
Temkin–Schwartzman equation [5] at specified temꢀ regarded as a partial loss of carbon as well [7].
perature, kJ/mol; Δn is the change in the number of
moles in the course of the reaction; and p0 is the stanꢀ
dard pressure (1.013
The catalytic data obtained coincide with the
results from thermodynamic calculations; the reaction
10 Pa). As can be seen from range lies below 600 K. With increasing temperature,
5
×
Fig. 1, the change in the Gibbs energy becomes more however, the yield of DME and CO conversion under
negative with increasing pressure; i.e., reaction (3) is constant pressure increase, unlike the data presented
allowed over the 300–600 K range, the pressure in the in Fig. 1. In this case, reaction (3) occurs via a path
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 85
No. 7 2011

SYNTHESIS OF DIMETHYL ETHER FROM CO AND H2
1143
20
µ
m
20 µm
(a)
(b)
Fig. 4. Microimages of Rꢀ1 (a) prior to and (b) after catalytic experiments lasting 70 h.
20
µ
m
20 µm
(a)
(b)
Fig. 5. Microimages of γꢀAl O (a) prior to and (b) after catalytic experiments lasting 70 h.
2 3
that does not obey the laws of thermodynamics. This lyst was ~70 h, no carbon deposits were observed on
could be due to the complexity of the process, since
reactions (1)–(4) proceed simultaneously.
its surface, as was supported by the data from differꢀ
ential thermal analysis.
Surface morphology of the catalysts. Figure 4 preꢀ
sents microimages of the Rꢀ1 sample prior to and after
the catalytic experiments. As can be seen, the surꢀ
face of the initial methanol catalyst (Fig. 4a) is hetꢀ
Figure 5 shows microimages of γꢀAl O3 prior to
and after the catalytic experiments. While the surꢀ
face of the initial sample (Fig. 5a) is homogeneous,
2
erogeneous, uneven, and covered with fractures the surface of the dead catalyst (Fig. 5b) is covered
without the formation of any isolated particles. After with craters with an average diameter of 2
µ
m and
the catalytic experiments with temperature and
individual particles with dimensions of ~1
µ
m. The
pressure varied from 200 to 300°С and from 1 to
formation of such holes is probably due to changes
in the alumina volume under pressure; i.e., the crysꢀ
talline form is likely to undergo a transformation
due to the rearrangement of the electron shell of
5
MPa, respectively, there was a change in the cataꢀ
lyst morphology (Fig. 4b): the surface coalesced
with the formation of individual dark cleavages. The
average size of these formations varies from 1 to
4
0 m. Even though the operating time of the cataꢀ alumina [8]. The pore size increases simultaneously,
µ
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 85
No. 7 2011

1144
KOSOVA et al.
(
table), leading to the formation of holes on the surꢀ
REFERENCES
face.
1. C. Arcoumanis, C. Bae, R. Crookes, and E. Kinoshita,
Fuel 87, 1014 (2008).
2
. A. Ya. Rozovskii and G. I. Liin, Theoretical Basis of
Methanol Synthesis (Khimiya, Moscow, 1990) [in Rusꢀ
sian].
CONCLUSIONS
When synthesis gas passes once over sequentially
loaded Rꢀ1 and ꢀA O catalysts at 300 , 3 MPa,
γ
°С
3. G. R. Moradi, H. Ghaneia, and F. Yaripour, React.
2
3
Н /СО = 2, and a contact time of 6 s, the DME yield
2
Kinet. Catal. Lett. 92, 137 (2007).
is 18%, with CO conversion of 68%. The composition
and structure of the catalysts for DME synthesis
undergo changes under the influence of pressure and
temperature. Copper oxide within the methanol cataꢀ
lyst Rꢀ1 is reduced in the course of catalysis to the
metallic state, the surface coalesces, and the pore size
of the alumina increases.
4. Yu. M. Zhorov, Thermodynamics of Chemical Processes.
Petrochemical Synthesis, Reprocessing of Coal, and Natꢀ
ural Gas (Khimiya, Moscow, 1985) [in Russian].
5. A. G. Stromberg and D. P. Semchenko, Physical Chemꢀ
istry (Vyssh. Shkola, Moscow, 2006) [in Russian].
6. D. S. Wang, Y. S. Tang, Y. S. Han, and N. Tsubaki, Fuel
Chem. Technol. 36, 171 (2008).
7. O. V. Krylov, Heterogeneous Catalysis (Akademkniga,
Moscow, 2004) [in Russian].
ACKNOWLEDGMENTS
8
. G. G. Mark, Chemical Equilibrium and Reaction Rates
under High Pressure (RISO AN SSSR, Moscow, 1960)
[in Russian].
This work was supported by the Federal Target Proꢀ
gram, state contract no. 02.523.12.3023.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 85
No. 7 2011