Cu-Zr-Zn catalysts for methanol synthesis in a fluidized bed reactor

Article abstract of DOI:10.1016/j.apcata.2011.01.010

Ten Cu-Zr-Zn catalysts prepared by reverse coprecipitation were evaluated for use for methanol synthesis at 230-270 °C and 4 MPa. The catalysts were characterized by N₂O-chemisorption, X-ray diffraction, scanning electron microscopy, and their attrition resistance and particle size distribution. The data showed that many of the catalysts had suitable activity, shape, size and mechanic strength. Catalyst stability during reaction at 260 °C for 120 h was also studied. The catalyst with the composition of Cu:Zr:Zn = 4.5:3:1.5 was optimal for methanol synthesis in a fluidized bed reactor. Crown Copyright

Full text of DOI:10.1016/j.apcata.2011.01.010

Applied Catalysis A: General 394 (2011) 281–286
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Cu-Zr-Zn catalysts for methanol synthesis in a fluidized bed reactor
Guogao Wang, Yizan Zuo, Minghan Han∗, Jinfu Wang
Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Ten Cu-Zr-Zn catalysts prepared by reverse coprecipitation were evaluated for use for methanol synthesis
at 230–270 ◦C and 4 MPa. The catalysts were characterized by N2O-chemisorption, X-ray diffraction, scan-
ning electron microscopy, and their attrition resistance and particle size distribution. The data showed
that many of the catalysts had suitable activity, shape, size and mechanic strength. Catalyst stability dur-
ing reaction at 260 ◦C for 120 h was also studied. The catalyst with the composition of Cu:Zr:Zn = 4.5:3:1.5
was optimal for methanol synthesis in a fluidized bed reactor.
Received 13 October 2010
Received in revised form 2 January 2011
Accepted 6 January 2011
Available online 12 January 2011
Keywords:
Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved.
Methanol synthesis
Cu-Zr-Zn catalyst
Fluidized bed reactor
1. Introduction
Although there were been investigations on the reactor and its
design, studies on the catalyst used in a fluidized bed have been
Methanol is a fundamental product in the chemical industry
and its production capability has been increasing due to its wide
use as a chemical feedstock, especially for clean energy resources.
Since the 1970s, fixed bed reactors have been used with Cu/ZnO-
based catalysts for methanol synthesis [1]. However, the reactions
in methanol synthesis from syngas (CO + 2H2 = CH3OH + 90.4 kJ,
CO2 + 3H2 = CH3OH + H2O + 49.33 kJ, CO + H2O = CO2 + H2 + 40.96 kJ)
are highly exothermic and effective heat removal is required.
Unfortunately, it is not easy to do this with fixed bed reactor. A
fluidized bed reactor is much more efficient in heat transfer.
Although most methanol synthesis works have been focused
on enhancing the per pass conversion of carbon monoxide by the
development of more active catalysts [2], some researchers have
explored the use of fluidized bed reactor. Compared with fixed
bed reactor, fluidized bed is not only much better at heat transfer,
but it is also less expensive to manufacture. More importantly, the
temperature uniformity in the whole reactor and the avoidance of
diffusion limitation by the use of much smaller catalyst particles
have greatly improve methanol synthesis production efficiency.
Wagialla and Elnashaie [3] carried out a theoretical investigation
on methanol synthesis in a fluidized bed reactor and discussed the
difference between methanol synthesis in fixed bed and fluidized
bed reactors, and showed that a 30% improvement can be obtained
with a fluidized bed. Tabis [4] and Rahimpour et al. [5,6] have done
a lot of work on the optimization of fluidized bed reactor design for
methanol synthesis.
few. Unlike the catalysts in a fixed bed, the catalysts in a fluidized
bed reactor are subject to attrition from collisions between catalyst
particles and with the reactor wall [7]. In the fluidized bed methanol
synthesis investigations mentioned above, researchers used fixed
bed catalyst that was ground to small directly. The catalyst used
for methanol synthesis in a fixed bed is usually a Cu-Zn-Al catalyst
based on the catalyst invented by ICI (Imperial Chemistry Industry)
researchers. This kind of catalyst is poor in attrition resistant per-
formance and cannot be used for long-term methanol synthesis in
a fluidized bed reactor.
In the present work, a series of Cu-Zr-Zn catalysts were prepared
and evaluated. They were characterized by N2O-chemisorption, X-
ray diffraction (XRD), scanning electronic microscopy (SEM), and
their attrition resistance and particle size distribution (PSD). Some
of these catalysts performed well and their attrition resistance and
catalytic stability reached the basic standard for a fluidized bed.
2. Experimental
2.1. Catalyst preparation
Ten catalysts were prepared by the sodium carbonate (Bei-
jing Chemical Corporation) coprecipitation method. Three metal
salt solutions (1.2 mol/L Cu(NO3)2, 0.6 mol/L ZrOCl2 and 1.2 mol/L
Zn(NO3)2) were premixed in the specific catalyst compositions and
added dropwise to a sodium carbonate solution (1.2 mol/L, pre-
heated to 70 ◦C) that was stirred vigorously. The addition of the
mixed solution was stopped when the pH reached 7. Then the prod-
uct slurry was maintained at 70 ◦C for 1 h, filtered and washed in
deionized water several times. A gel-like catalyst precursor was
∗
Corresponding author. Tel.: +86 10 62781469; fax: +86 10 62772051.
E-mail addresses: hanmh@tsinghua.edu.cn, wgg1215@sina.com (M. Han).
0926-860X/$ – see front matter. Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2011.01.010

282
G. Wang et al. / Applied Catalysis A: General 394 (2011) 281–286
The X-ray pattern was recorded in the 2ꢁ range between 10◦ and
90◦. The morphology of the catalysts before reaction was charac-
terized by a JSM 7401F scanning electron microscope operated at
3 kV and 20 mA.
Table 1
Molar composition and metallic copper specific area of the ten catalysts prepared.
Catalyst
Composition
Cu
Metallic copper
specific area (m2/g)
Zn
Zr
The mechanical strength of the catalysts was measured using
ASTM-D-5757-00, which is a standard method for the determina-
tion of attrition and abrasion of powdered catalysts by air jets. In
this standard, the catalyst mechanical strength was evaluated by
the Air Jet Index (AJI), which is the percentage of fine particles
smaller than 10 ␮m carried out of the fluidized bed by a high veloc-
ity gas in a 5 h period. The apparatus used was self-constructed in
accordance with the standard.
The particle size distributions (PSDs) of the catalysts were
measured by a Mastersizer (Malvern Corp.). Before measurement,
catalyst powder was added to deionized water and ultrasonically
dispersed for 2 min. Then the PSD was determined by the laser
method.
A-1
A-2
A-3
A-4a
A-5
B-1
B-2
B-3
B-4a
B-5
1.5
3
4
4.5
5
4.5
4.5
4.5
4.5
4.5
4.5
3
2
1.5
1
1.5
1.5
1.5
1.5
1.5
3
3
3
3
3
1.5
2
2.5
3
4
16.7
18.2
38.4
39.2
32.4
53.2
48.7
42.6
39.5
28.6
a
A-4 and B-4 were two same-composed catalysts prepared for under the same
conditions.
obtained that was kneaded in a kneader with 1 wt% polyvinyl alco-
hol solution, which was used as an organic binder for catalyst
mechanic strength improvement. Next, the precursor was spray
dried in a spray dryer. In the spray dried process, the inlet tempera-
ture was 200 ◦C and the outlet temperature was 90 ◦C. The obtained
near-spherical catalysts were calcined in a muffle oven at 400 ◦C
for 2 h. These catalysts were named according to their composi-
tions and divided into two groups. Group A, with the ratio of Cu/Zn
varied, comprised five catalysts with (Cu + Zn):Zr = 6:3 named A-
1 to A-5. Group B comprised the other five catalysts named B-1
to B-5 with the composition Cu:Zn = 4.5:1.5 and Zr content varied
(see Table 1). A-4 and B-4 have the same composition, but were
prepared separately using the same experimental conditions.
3. Results and discussion
3.1. Catalyst activity and methanol selectivity
For methanol synthesis, our research group has reported a
fibrous catalyst that is much more active than the commercial
industrial catalysts generally used [2]. This catalyst is so active that
the carbon monoxide per pass conversion can almost reach as high
as the thermodynamic equilibrium [11]. However, this catalyst can-
not be used in a fluidized bed reactor because of its poor mechanical
strength. To develop a new catalyst for methanol synthesis in a flu-
idized bed, we carried out a variety of experiments and found that
a Cu-Zn-Zr catalyst obtained from a gel-like precipitate possesses
sufficient activity and mechanical strength.
2.2. Catalytic activity evaluation
The catalytic properties of the catalysts were evaluated in a
laboratory fixed bed reactor. Before reaction, the catalyst was
pretreated in 4% (v/v) H2/N2 (GHSV = 2500 mL gcat−1 h−1) at atmo-
spheric pressure by raising the temperature from ambient to 230 ◦C
over 10 h. Then the inlet gas was switched to the reaction gas
(GHSV = 5000 mL gcat−1 h−1) composed of 5.3 vol% carbon dioxide,
26.3 vol% carbon monoxide and 68.4 vol% hydrogen. The reaction
was carried out at 4 MPa and temperatures from 230 to 270 ◦C.
The first effluent sample was taken for analysis after the system
had attained steady state. Carbon monoxide conversion (Xco) was
defined as
Copper crystallites are regarded as the basic active sites on Cu-
based catalysts, and many investigations have related their activity
with copper dispersion or something similar [12,13]. Zinc oxide,
on the other side, is regarded as a promoter that serves as atomic
hydrogen supplier [14]. Zirconia was also found to enhance the
activity of Cu-based methanol catalysts [15,16], but the detailed
mechanism is still unknown. For the purpose of screening to get
an active and attrition resistant catalyst, a series of catalysts of dif-
ferent compositions were prepared, evaluated and characterized.
Group A catalysts were designed to determine the optimal Cu/Zn
ratio, which is believed to be important for methanol synthesis
activity. Group B catalysts were designed to determine the opti-
mal (Cu + Zn)/Zr ratio, which is important for both catalytic activity
and attrition resistance.
Fmethanol,out
Xco =
(1)
Fco,in
where Fmethanol,out and Fco,in are the molar flow of methanol and CO,
respectively. Subscripts out and in denote the flow out of or into the
reactor.
Fig. 1 shows the relation between the Cu/Zn ratio of the cata-
lysts and their activities. As the Cu/Zn ratio increased, the activity
of the Group A catalysts first increased and then decreased. It is
well known that too low or too high a Cu/Zn ratio is unfavorable
for high carbon monoxide conversion: a low Cu content gives less
methanol synthesis active sites and a low ZnO content gives inade-
quate promoter for the catalytic site. The best ratio of Cu/Zn in our
system was 4.5:1.5.
The catalytic activities of the Group B catalysts are shown in
Fig. 2. When the Cu/Zn ratio was kept at 4.5:1.5, the activity
decreased monotonically as the ZrO2 content increased. This can
be explained by that in the range of the ZrO2 content investigated,
the increase in ZrO2 content decreased the amount of copper active
site per mass catalyst. However, this result also does not contradict
the promotional effect of zirconia because the amounts of ZrO2 used
here were so large that they exceeded the content range where ZrO2
promotion can be seen. The ZrO2 content was selected here to be
in this range used to determine the optimal ZrO2 amount for good
mechanical strength.
2.3. Catalyst characterization
N2O-chemisorption is the accepted routine for metallic copper
specific surface area measurement [8,9]. In the present work, the
metallic copper surface area was determined by the method of
Vandergrift et al. [8]. This method is also termed N2O titration.
It consists of three main reduction or oxidation procedures, with
the amount of hydrogen consumption during the reduction pro-
cess before and after N2O-chemisorption, and the ratio of surface
copper to bulk copper obtained. The metallic copper surface area
was calculated by using the molar stoichiometry of N2O/Cu(s) = 0.5
and an average value of 1.46 × 1019 copper atoms/m2 [10].
The catalysts were characterized on a Bruker D8 advance X-ray
diffractometer. Nickel-filtered Cu K␣ radiation (ꢀ = 1.54 × 10−10 m)
was used as the X-ray source, which was used at 40 kV and 20 mA.

G. Wang et al. / Applied Catalysis A: General 394 (2011) 281–286
283
Cu/Zn= 1.5:4.5
Cu/Zn= 3:3
Cu/Zn= 4:2
Cu/Zn= 4.5:1.5
Cu/Zn= 5:1
0.42
0.36
0.30
0.24
0.18
0.12
0.99
A-1
A-2
A-3
A-4
A-5
B-1
B-2
B-3
B-4
B-5
0.96
0.93
0.90
230
240
250
260
270
230
240
250
260
270
Temperature/oC
Temperature/oC
Fig. 1. Activities of Group A catalysts with different Cu/Zn ratios.
Fig. 3. Methanol selectivity of all Cu-Zr-Zn catalysts prepared in this work.
at 230 ◦C is presented in Fig. 4. All the points fell near the diagonal,
which suggested a linear relationship between catalytic activity
and copper surface area. This result was in good agreement with
many other studies [17,18], which reproved that metallic copper
was the active site center of methanol synthesis. Too much ZrO2
have no more promotion effect. It was inferred that ZrO2 might
be coated on some copper oxide particles, leading to the decreas-
ing of metallic copper specific area. Here, superfluous ZrO2 were
added purely to improve catalysts’ attrition resistance at the price
of activity sacrifice.
The methanol selectivities of the ten catalysts were almost the
same. The data are shown in Fig. 3. The selectivity to methanol
decreased gradually as the reaction temperature increased. When
the reaction temperature was lower than 250 ◦C, the selectivity was
nearly 100%. When the reaction temperature was 260 ◦C or 270 ◦C,
some carbon monoxide was converted to methane, leading to a
decrease in methanol selectivity.
3.2. N2O-chemisorption
To better understand catalysts’ active site center in the catalysts,
catalysts’ metallic copper specific area (Table 1) were measured
and correlated with their methanol synthesis activity. The relation
of the metallic copper surface area and methanol synthesis activity
3.3. XRD
The XRD patterns of the ten catalysts are shown in Fig. 5. These
catalysts had the same crystalline phases. All the diffraction peaks
were due to the crystalline phases of CuO or ZnO. No clear peak
due to the ZrO2 phase was seen due to the relatively low calci-
nation temperature (400 ◦C). According to the literature [19,20],
the ZrO2 phase appears after calcination above 500 ◦C. In all the
XRD patterns, the rather featureless peaks implied an amorphous
phase of the catalyst component or small crystallites. Our previ-
ous work found that a smaller crystallite size gave a higher activity
0.56
Zr-1.5
Zr-2
Zr-2.5
Zr-3
Zr-4
0.48
0.40
0.32
0.24
50
40
30
20
10
230
240
250
260
270
0.1
0.2
0.3
0.4
Temperature/oC
Xco/%
Fig. 2. Activities of Group B catalysts with different (Cu + Zn)/Zr ratios.
Fig. 4. Metallic copper surface area versus CO conversion.

284
G. Wang et al. / Applied Catalysis A: General 394 (2011) 281–286
12
ZnO
CuO
PSD of A-4
A-1
A-2
9
6
3
0
A-3
A-4
A-5
B-1
B-2
1
10
Size/μm
100
B-3
B-4
Fig. 7. Particle size distribution of A-4 catalyst.
B-5
3.5. PSD
The catalysts obtained after spray drying had similar PSDs (all
the catalysts were spray dried under the same conditions. There-
fore, there should be no evident difference in their PSDs). As an
example, the PSD of A-4 is shown in Fig. 7. Most of the catalyst
particles were in the range 10–200 ␮m and the volumetric average
size was 62.4 ␮m. The bulk densities of the ten spray dried cata-
lysts were about 0.65 g cm−3. These particles should be classified as
Geldart-A particles. So, the catalysts should be easily fluidized [22].
0
20
40
2 theta/
60
80
o
Fig. 5. XRD patterns of the ten Cu-Zr-Zn catalysts.
for methanol synthesis [21]. A comparison of the diffraction peak
intensities showed that these roughly reflected the compositions
of the catalysts listed in Table 1.
3.6. SEM
Four SEM images of A-4 catalyst are shown in Fig. 8. It is
shown in Fig. 8(1 and 2) that most of catalyst particles were near-
spherical. Pham et al. [23] reported that spherical particles were
more attrition resistant than non-spherical particles. The particle
shape indicated the prepared catalysts would have high attrition
resistance. In accordance with the PSD measured by the Master-
sizer, most of the particles were ten to a hundred micrometer.
Fig. 8(3 and 4) shows microscopic views of the catalyst pow-
der sphere. The A-4 catalyst was found to comprise wedge-like
particles. It can be inferred that this kind of microstructure can
provide more effective active sites for methanol synthesis, and
maybe hindered copper crystallites from sintering during a long-
term methanol synthesis reaction.
3.4. Attrition resistance
The AJI data of the Group A and B catalysts are shown in Fig. 6.
The AJI of the Group B catalysts showed a drastic decrease from
B-1 to B-5, indicting that the AJI decreased as the ZrO2 content
increased. This meant that ZrO2 played an important role in cata-
lyst mechanical strength enhancement. For the Group A catalysts,
the AJI gradually increased from A-1 to A-5. In the Group A cata-
lysts, the ZrO2 content remain unchanged, which then implied that
more Cu and less Zn is unfavorable for catalyst mechanical strength,
but the influence is limited. Generally, a fluidized catalyst should
have an AJI below 0.05 to be acceptable. From considering activity
and attrition resistance, A-4 catalyst is the optimal catalyst for flu-
idized bed methanol synthesis. In the following experiments, A-4
was used as the catalyst.
3.7. Catalyst stability
High activity, high selectivity and long life time are the three
essential requirements foran excellentcatalyst. UnlikeFCC catalyst,
a Cu-based catalyst used for methanol synthesis cannot be regen-
erated online. Thus, a long life time is of special significance for
fluidized bed methanol synthesis. In the preparation of the fluidized
bed catalysts, many binders were also used for the improvement
of attrition resistance. Unfortunately, it was found in our experi-
ments that all the binders such as Al2O3, SiO2 and kaolin caused
the catalytic activity to drop significantly as the reaction was run
for a longer time. The present series of catalysts does not use any
such binders.
It is shown in Fig. 9 that the prepared Cu-Zr-Zn catalyst showed
high stability for methanol synthesis. To investigate the catalyst
stability, 260 ◦C was used as the reaction temperature for the
acceleration of deactivation (230 ◦C is usually used for methanol
synthesis) [24]. For a reaction time as long as 120 h, no obvious
deactivation was observed. This proved the high catalyst stability
of the Cu-Zr-Zn catalyst (A-4).
Group A
0.15
Group B
0.12
0.09
0.06
0.03
1
2
3
4
5
Catalysts(A or B)
Fig. 6. AJI values of Group A and B catalysts.

G. Wang et al. / Applied Catalysis A: General 394 (2011) 281–286
285
Fig. 8. SEM images of A-4 catalyst with different magnifications.
this work showed good catalytic stability during methanol synthe-
sis in a fluidized bed reactor.
0.45
0.36
0.27
0.18
Acknowledgements
The authors gratefully acknowledge the financial support of
this study by the National Basic Research Program of China (grant
no 2009CB219901). Many thanks are also given to Prof. Wang for
English revision.
References
[1] K.C. Waugh, Catal. Today 15 (1992) 51–75.
0
20 40 60 80 100 120
[2] X. An, F. Ren, J.L. Li, J.F. Wang, Chin. J. Catal. 26 (2005) 729–730.
[3] K.M. Wagialla, S. Elnashaie, Ind. Eng. Chem. Res. 30 (1991) 2298–2308.
[4] B. Tabis, Chem. Eng. J. 83 (2001) 191–200.
time-on-stream/h
[5] M.R. Rahimpour, K. Alizadehhesari, Chem. Eng. Technol. 31 (2008) 1775–1789.
[6] M.R. Rahimpour, M. Bayat, F. Rahmani, Chem. Eng. Sci. 65 (2010) 4239–4249.
[7] Y. Petukhov, H. Kalman, Part. Part. Syst. Charact. 20 (2003) 267–275.
[8] C.J.G. Vandergrift, A.F.H. Wielers, B.P.J. Joghi, J. Vanbeijnum, M. Deboer, M. Ver-
sluijshelder, J.W. Geus, J. Catal. 131 (1991) 178–189.
Fig. 9. Long-term activity of A-4 catalyst at the reaction temperature of 260 ◦C (up
to 120 h).
[9] B. Denise, R.P.A. Sneeden, B. Beguin, O. Cherifi, Appl. Catal. 30 (1987) 353–363.
[10] J. Agrell, H. Birgersson, M. Boutonnet, I. Melian-Cabrera, R.M. Navarro, J.L.G.
Fierro, J. Catal. 219 (2003) 389–403.
[11] X. An, Y.Z. Zuo, Q. Zhang, J.F. Wang, Chin. J. Chem. Eng. 17 (2009) 88–94.
[12] L.S. Kau, K.O. Hodgson, E.I. Solomon, J. Am. Chem. Soc. 111 (1989) 7103–7109.
[13] B.S. Clausen, G. Steffensen, B. Fabius, J. Villadsen, R. Feidenhansl, H. Topsoe, J.
Catal. 132 (1991) 524–535.
4. Conclusions
The methanol synthesis activity of Cu-Zr-Zn catalysts first
increased and then decreased as the Cu/Zn ratio increased, but
monotonically decreased as the (Cu + Zn)/Zr ratio decreased. The
attrition resistance increased as the ZrO2 content increased. Tak-
ing activity and attrition resistance into consideration, the optimal
composition was Cu:Zr:Zn = 4.5:3:1.5. PSD and SEM characteriza-
tion indicated that the catalyst particles were suitable for use in a
fluidized bed reactor. More importantly, the catalysts prepared in
[14] M.S. Spencer, Catal. Lett. 50 (1998) 37–40.
[15] Y. Nitta, T. Fujimatsu, Y. Okamoto, T. Imanaka, Catal. Lett. 17 (1993) 157–165.
[16] Y. Nitta, O. Suwata, Y. Ikeda, Y. Okamoto, T. Imanaka, Catal. Lett. 26 (1994)
345–354.
[17] T.H. Fleisch, R.L. Mieville, J. Catal. 90 (1984) 165–172.
[18] W.X. Pan, R. Cao, D.L. Roberts, G.L. Griffin, J. Catal. 114 (1988) 440–446.

286
G. Wang et al. / Applied Catalysis A: General 394 (2011) 281–286
[19] L.C. Wang, Q. Liu, M. Chen, Y.M. Liu, Y. Cao, H.Y. He, K.N. Fan, J. Phys. Chem. C
111 (2007) 16549–16557.
[22] Y.J. Chen, H.Q. Zhou, J. Zhu, Q. Zhang, Y. Wang, D.Z. Wang, F. Wei, Catal. Lett.
124 (2008) 297–303.
[20] J. Sloczynski, R. Grabowski, A. Kozlowska, P. Olszewski, J. Stoch, J. Skrzypek, M.
Lachowska, Appl. Catal. A 278 (2004) 11–23.
[23] H.N. Pham, A. Viergutz, R.J. Gormley, A.K. Datye, Powder Technol. 110 (2000)
196–203.
[21] G.G. Wang, Y.Z. Zuo, M.H. Han, J.F. Wang, React. Kinet. Mech. Catal. 101 (2010)
443–454.
[24] Y.Z. Zuo, Q. Zhang, M.H. Han, J.F. Wang, T.F. Wang, D.Z. Wang, Y. Jin, Chin. J.
Catal. 30 (2009) 624–630.