вход по аккаунту


Directed Evolution of Noble-Metal-Free Catalysts for the Oxidation of CO at Room Temperature.

код для вставкиСкачать
Heterogeneous Catalysis
Directed Evolution of Noble-Metal-Free Catalysts
for the Oxidation of CO at Room Temperature**
Jens W. Saalfrank and Wilhelm F. Maier*
Dedicated to Professor Manfred T. Reetz
on the occasion of his 60th birthday
The development and application of high-throughput methods for the rapid discovery and optimization of solid-state
catalysts has led to promising results in the last few years.[1]
The possibility of the systematic exploration of large parameters spaces make combinatorial chemistry an interesting tool
in heterogeneous catalysis. As a model reaction the oxidation
of carbon monoxide on noble-metal-containing catalysts has
been studied frequently.[2] Recently, a photo-acoustic setup
for parallel real-time detection of reaction products was
presented in the search for new noble-metal-free low-temperature catalysts for the oxidation of CO.[3]
The catalytic oxidation of CO with O2 belongs to the most
intensively studied reactions in chemistry, especially in the
area of surface science. Despite this huge effort there is still a
large demand for better practically useful and affordable, that
is, noble-metal-free catalysts. Especially the oxidation at
room temperature with its practical application in breathing
protection is very important. Hopcalite (“CuMn2O4”) is the
commercial catalyst still used in gas masks despite its rapid
deactivation in moist air.[4] Further examples of low-temperature catalysts for the oxidation of CO are the gold catalysts
discovered by Haruta et al.,[5] ruthenium dioxide hydrate[6]
and Co3O4-based systems.[7]
The goal of our study was to discover new noble-metalfree catalysts for the oxidation of CO at low temperature by
using the established technologies of combinatorial heterogeneous catalysis. It is known in literature that the basic
oxides CoOx, MnOx, and NiOx have a detectable catalytic
activity for the oxidation of CO with air. Our search for new
catalysts started with the systematic doping of these base
oxides with 56 different metal centers. The automated
preparation of such mixed oxides based on Co, Mn, and Ni
oxides was carried out with the help of a pipetting robot
following previously developed sol–gel recipes. These recipes,
which have been especially developed for the preparation of
mixed oxides by pipetting robots, are characterized by the
[*] Dipl.-Chem. J. W. Saalfrank, Prof. Dr. W. F. Maier
Universitt des Saarlandes
Lehrstuhl f#r Technische Chemie
P.O.Box 151150, 66041 Saarbr#cken (Germany)
Fax: (+ 49) 681-3022343
[**] This project has been supported by the BMBF, project no. FKZ
03C0311. We thank Dr. Tesche and B. Spliethoff from the MPI f#r
Kohlenforschung for valuable TEM/EDX investigations and A.
M#ller, UdS, for XRD measurements. We thank Drger Safety AG &
Co. KgaA and Dr. Ammann for supplying the reference catalyst
Hopcalite and for valuable discussions.
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200351935
Angew. Chem. Int. Ed. 2004, 43, 2028 –2031
formation of a homogeneous gel over the whole range of
compositions of interest (no precipitation or separation). This
is a prerequisite for the formation of a true amorphous mixed
oxide, at least prior to the calcination step. By the formation
of clear sols precipitation and associated undesired formation
of domains is avoided. The samples are identified by the
central elements with the expected mol % given in subscripts
as expected from the composition of the starting sol. The
mixed oxide Al1Mn6.7Co92.3 for example has been generated
from 1 mol % Al, 6.7 mol % Mn, and 92.3 mol % Co of the
associated molar precursors by hydrolysis–polycondensation.
Since the oxidation state of the mixed oxides as obtained after
calcination in the high-throughput experiment has not been
determined, the true oxygen content and the associated
complete composition of the mixed oxides prepared in this
way cannot be given. The true oxidation state of all prepared
mixed oxides is not of importance, since only the catalytically
most active materials will survive the screening steps.
The design of the materials libraries was accomplished by
using the software “Plattenbau”.[8] This software calculates on
the basis of a parameterized recipe the volumes of the
different solutions of starting materials, as required for the
preparation of the individual samples and generates an
optimized pipetting list, which can be transferred directly to
the pipetting robot.
The reagents required for the synthesis of the mixed
oxides were pitpetted from the robot from alcoholic solutions
into 2-mL sampling flasks positioned in arrays of 50 flasks.
The sols were mixed thoroughly in an orbital mixer, then
dried and calcined, and manually transferred into the 207
hexagonally positioned wells (13.5 mm) in the library plate
made of slate (199 mm, Figure 1 a). Compared to direct
Figure 1. a) A slate plate with 182 catalysts (catalyst library). b) Reactor for IR-thermography with catalyst library, detectable under the sapphire window.
synthesis in the wells of the library plate, the applied
procedure has the advantage that all catalysts are positioned
as powders with comparable filling heights in the wells. This
reduces potential errors due to significantly different mass
transport properties in catalytic gas-phase reactions. Additionally, it allowed a powdered sample of the reference
catalyst Hopcalite to be positioned in one of the wells of every
library. The libraries were then studied with emissivitycorrected IR-thermography[9] at 50 8C (1 vol % CO in synthetic air, volume flow 50 mL min 1). This method, which
visualizes relative heats of reaction, allowed us to rapidly
Angew. Chem. Int. Ed. 2004, 43, 2028 –2031
sample the relative catalytic activity of potential catalysts for
the oxidation of CO at low temperatures in a parallel
approach. Figure 1 b shows the IR-reactor containing the
catalyst library.
For catalyst selection and library design none of the new
methods for experimental design were applied.[10] Our search
strategy was based on the concept of evolution (= variation
and selection). Variation was obtained by doping and
composition spread; for selection, catalytic activity, determined through heat of reaction, was chosen. In the first
generation of catalysts binary mixed oxides of the three base
oxides with 56 selected elements (E) in two different
concentrations were examined (Table 1). The higher degree
Table 1: Composition and number of mixed oxides studied sorted by
generation. Hits are the active catalysts of one generation, which were
used as base materials for the catalysts studied in the following
Sample composition[a]
Sample number
MnxCo100 x[b]
AlxMn6.7Co100 6.7
[a] E = Ag, Al, Au, B, Ca, Cd, Ce, Co, Cr, Cs, Cu, Dy, Er, Eu, Fe, Ga, Gd, Ge,
Hf, Ho, In, Ir, La, Li, Lu, Mg, Mn, Mo, Na, Nb, Nd, Ni, Pd, Pr, Pt, Rb, Re,
Rh, Ru, Sb, Sc, Se, Si, Sm, Sn, Ta, Tb, Te, Th, Ti, Tm, V, W, Y, Yb, Zn, Zr.
[b] x = 1.7, 3.4, 5.1, 6.7, 8.3, 9.9, 11.5, 13.1, 14.6, 16.1. [c] x = 1, 1.9, 2.9,
3.9, 4.9, 5.8, 6.8, 7.8, 8.8, 9.7.
of doping was not applied to noble metals and selected
elements of the Group 1 and 2 of the periodic table. The hit
selection was based on highest activity. The activity of these
most active samples was further optimized by compositional
variation and doping.
In the first catalyst generation (two libraries) Mn3Co97 and
Mn10Co90 were identified by the IR-thermography as the
catalytically most active samples. In the second generation the
effect of Mn content in cobalt oxide was studied with the help
of a binary composition spread. Within the range of 1.7 to
16.1 mol % Mn, Mn6.7Co93.3 showed the highest relative
catalytic activity. This binary mixed oxide was chosen as
lead composition for the third generation of these ternary
mixed oxides by doping it with the above-mentioned remaining 55 elements. Among several active materials
Al1Mn6.7Co92.3 and Al2.9Mn6.7Co90.4 showed the highest activity.
In the fourth generation the effect of the Al content on
Mn6.7Co93.3 was examined in the range of 1 to 9.7 mol %, and
Al1Mn6.7Co92.3 (sample 3) was confirmed as the most promising candidate for the oxidation of CO at low temperatures
(Figure 2). The relative catalytic activity of Al1Mn6.7Co92.3 was
found to be comparable to that of the reference catalyst
Hopcalite (samples 1 and 2). While higher aluminum contents
lead to deactivation, a fine-tuning in the range of 0.5 to
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
in the gas-phase flow reactor in comparison to the
reference catalyst Hopcalite. It shows that the catalytic activity of the basic oxides was continuously
improved by the combinatorial experiments. Just
doping Mn6.7Co93.3 with 1 mol % aluminum results in
an improvement of the conversion of CO by about
30 %. The NiOx prepared by the sol–gel procedure
here shows no activity (deactivation) under our test
conditions after one hour. The good reproducibility of
the combinatorial results under conventional conditions confirms that the development of catalyst with
high-throughput methods as used here can be carried
out reliably. Under conventional reaction conditions
Al1Mn6.7Co92.3 shows a very high catalytic activity for
Figure 2. a) Emissivity-corrected IR-thermographic image of a catalyst library during
the oxidation of CO in synthetic air at 25 8C (95 %
oxidation of CO at 52.4 8C. b) Composition and relative catalytic activity of the mixed
oxides on the catalyst library.
For the characterization of the best catalyst for
the oxidation of CO (Al1Mn6.7Co92.3), its specific
surface area was determined by physisorption measurements
1.5 mol % of Al might be useful but could not be carried out
and its phase composition by X-ray diffraction. The BET
due to insufficient reliability of the screening method IRsurface of Al1Mn6.7Co92.3 (54 m2 g 1) is significantly smaller
After the high-throughput screening the hit detected was
than that of the reference catalyst Hopcalite (180 m2 g 1). The
confirmed by conventional studies. The three basic oxides and
X-ray diffraction pattern shows reflections at 2 V = 31.03,
the two hits of the second and fourth generation (Mn6.7Co93.3
36.625, 44.61, 59.06, and 65.1958. These reflections can be
assigned to the spinels Co3O4, Co2AlO4, CoAl2O4, and
and Al1Mn6.7Co92.3) were prepared in bulk in the laboratory
and were tested in a gas-phase flow reactor. For the convenCoMnAlO4. Which and to what extent these crystalline
tional preparation of the discovered materials the scale of the
phases are present in the material could not be determined
combinatorial synthesis was increased by a factor of 25. The
unequivocally. The reflections are weak and broad indicating
sol–gel syntheses were carried out in 20-mL flasks. The
small particles in an amorphous matrix. Additional TEM/
difference was that the sols were not mixed by an orbital
EDX studies with a Hitachi HF-2000 show that the compomixer but with the help of a magnetic stirrer. All other
sition of the catalyst is not homogeneous. The average
synthesis conditions (reagents, drying time, calcination temcomposition is in complete agreement with the composition
perature) were identical. For the conventional tests 200 mg
expected from the synthesis, however, larger deviations are
(particle size 100–200 mm) of the catalysts were used. The
observed at nanometer resolution. Areas of almost pure
reaction temperature was reduced to 25 8C, but all other
cobalt oxide as well as areas of highly enriched manganese
reaction conditions (1 vol % CO in synthetic air, flow rate
oxide were identified. Owing to its low content the aluminum
50 mL min 1) were identical to those employed in the highcontent could not always be determined, but areas of enriched
aluminum content were not found. This indicates a homogethroughput experiment.
neous distribution of the aluminum in the material. The
Figure 3 shows the catalytic activity of the materials after
results from XRD measurements were confirmed by TEM/
a reaction time of 1 h under conventional reaction conditions
EDX studies. The catalyst is an inhomogeneous mixture of
amorphous material and crystalline particles. Elemental
analysis has shown that the catalyst contains small amounts
of carbon (0.25 wt %), which results from the relatively low
calcination temperature of 300 8C. The heterogeneous structure of the catalyst and its low surface area indicate additional
potential for further optimization.
This study has shown that with the combinatorial methods
applied the development of heterogeneous catalysts can
be rapid and effective. In this study 439 catalyst samples
were prepared and tested. Within four generations a directed
evolution of heterogeneous catalysts was accomplished.
The best noble-metal-free catalyst Al1Mn6.7Co92.3 showed
under the chosen ideal laboratory conditions better
Figure 3. Catalytic activity of selected catalysts for the oxidation of CO with
catalytic properties for the oxidation of CO at room tempersynthetic air at 25 8C in the conventional gas-phase flow reactor. MnOx and
ature than the reference catalyst Hopcalite. The further
CoOx = basic oxides for the combinatorial development, Mn6.7Co93.3 = most
development of the catalysts for practical applications
active catalyst of the second generation. Al1Mn6.7Co92.3 = most active catalyst
(moisture-resistant, long-time stability) is currently under
of the third and fourth generation. XCO = conversion of CO at 25 8C after
60 min.
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2004, 43, 2028 –2031
Experimental Section
High-throughput syntheses of catalyst libraries: The automated
syntheses of catalysts were carried out with a commercial pipetting
robot (Zinsser Analytic, Lissy) on the basis of sol–gel recipes. The
precursors used were positioned as alcoholic solutions in 5-mL vials.
For the basic oxide 1m solutions of cobalt(ii) propionate in ethanol,
manganese(ii) propionate in methanol, and nickel(ii) propionate in
methanol were used. The oxides were doped with 0.1m solutions of 54
elements in the form of nitrates or alkoxides in 2-propanol. The
individual sols were prepared by pipetting in 2-mL vials, which were
positioned in a rack of 50 vials. The preparation of Al1Mn6.7Co92.3 for
example was carried out by pipetting the following volumes of single
solutions: cobalt(ii) propionate (184.6 mL, 184.6 mmol), manganese(ii)
propionate (13.4 mL, 13.4 mmol), aluminum tri(sec-butoxide) (20 mL,
2 mmol), 4-hydroxy-4-methylpentan-2-one (74.9 mL, 600 mmol), and
ethanol (365.1 mL, 6.26 mmol). After the pipetting process of an
entire rack was completed, this rack was placed on an orbital shaker
(Heidolph, Titramax 100) for 1 h and dried for five days at room
temperature. For calcination the samples were placed in an oven,
heated to 300 8C at a heating rate of 1 K min 1, and kept there for 5 h.
The catalyst powders obtained were manually transferred into the
wells of the slate library plate.
High-throughput screening of catalyst libraries: For the parallel
detection of catalytic activity the catalyst libraries were investigated
with emissivity-corrected IR-thermography (camera: AIM, Aegais,
PtSi-FPA-Detector 256 J 256 pixels). The library was positioned in a
gas-tight gas-phase flow reactor under the IR-transparent sapphire
window (Figure 1 b), which allowed the direct recording of temperature changes with the IR-camera. Prior to a measurement, the
library emissivity was determined at six different temperatures in a
range of 5 8C around the desired reaction temperature for temperature correction of the different emissivities of the samples. Before
the start of the actual measurement an additional IR image of the
library was recorded and substracted as background from all
following IR images. As standard reaction conditions 50 8C and
1 vol % CO in synthetic air at a flow rate of 50 mL min 1 were used.
The temperature increase of each sample in the IR image was
integrated with MatLab-based software[11] and standardized by
relating it to the sample of highest temperature increase on the
plate. This provided relative catalytic activities of the active catalysts
on the plate.
Conventional catalyst syntheses: 4-Hydroxy-4-methylpentan-2one (1.872 mL, 15 mmol) and ethanol (9.129 mL, 156.5 mmol) were
placed in a 20-mL vial. Subsequently 0.335 mL of a 1m maganese(ii)
propionate solution in methanol, 4.165 mL of a 1m cobalt(ii)
propionate solution in ethanol, and 0.5 mL of a 0.1m aluminum(iii)
tri(sec-butoxide) solution in 2-propanol were added while stirring.
The reaction mixture was stirred for 3 h and then dried for five days at
room temperature before calcination at 300 8C for five hours (heating
rate 1 K min 1).
Conventional testing of the catalysts: The conventional catalytic
studies were carried out in a gas-phase flow reactor with 200 mg of the
catalyst (100–200 mm). The standard reaction conditions were 25 8C
and 1 vol % CO in synthetic air at a flow rate of 50 mL min 1
(GHSV = 15 000 h 1). Prior to the actual measurement the catalysts
were treated for 60 min in synthetic air followed by 30 min in nitrogen
at 200 8C. The concentration of CO, CO2, and O2 in the product gas
was continuously monitored with gas sensors (GfG-mbH, special
production). The sensors were calibrated to allow us to correct for
cross sensitivity of the CO-sensor for CO2. From time to time the
reliability of the gas-phase analysis was controlled by GC measurements (Micro-GC model CP 4900, Varian, micro-channel mol sieve
5 K (10 m) column).
Keywords: CO oxidation · combinatorial chemistry ·
heterogeneous catalysis · high-throughput screening ·
[1] a) S. Senkan, Angew. Chem. 2001, 113, 322 – 341; Angew. Chem.
Int. Ed. 2001, 40, 312 – 329; b) J. Scheidtmann, P. A. Weiß, W. F.
Maier, Appl. Catal. A 2001, 222, 79 – 89; c) F. SchMth, O. Busch,
C. Hoffmann, T. Johann, C. Kiener, D. Demuth, J. Klein, S.
Schunk, W. Strehlau, T. Zech, Top. Catal. 2002, 21, 55 – 66; d) B.
Jandeleit, H. W. Turner, T. Uno, J. A. M. van Beek, W. H.
Weinberg, CATTECH 1998, 2, 101 – 123.
[2] a) P. Cong, R. D. Doolen, Q. Fan, D. M. Giaquinta, S. Guan,
E. W. McFarland, D. M. Poojary, K. Self, H. W. Turner, W. H.
Weinberg, Angew. Chem. 1999, 111, 508 – 512; Angew. Chem. Int.
Ed. 1999, 38, 484-;b) C. Hoffmann, A. Wolf, F. SchMth, Angew.
Chem. 1999, 111, 2971 – 2975; Angew. Chem. Int. Ed. 1999, 38,
2800 – 2803.
[3] T. Johann, A. Brenner, M. Schwickardi, O. Busch, F. Marlow, S.
Schunk, F. SchMth, Angew. Chem. 2002, 114, 3096 – 3100; Angew.
Chem. Int. Ed. 2002, 41, 2966 – 2968.
[4] a) A. B. Lamb, W. C. Bray, J. C. W. Frazer, J. Ind. Eng. Chem.
1920, 12, 213 – 221; b) G.-M. Schwab, G. Drikos, Z. Phys. Chem.
1940, 185, 405 – 425.
[5] a) M. Haruta, S. Tsubota, T. Kobayashi, H. Kageyama, M. J.
Genet, B. Delmon, J. Catal. 1993, 144, 175 – 192; b) M. Haruta,
M. DatP, Appl. Catal. A 2001, 222, 427 – 437.
[6] L. Zang, H. Kisch, Angew. Chem. 2000, 112, 4075 – 4076; Angew.
Chem. Int. Ed. 2000, 39, 3921 – 3922.
[7] a) J. Jansson, A. E. C. Palmqvist, E. Fridell, M. Skoglundh, L.
Qsterlund, P. ThormRhlen, V. Langer, J. Catal. 2002, 211, 387 –
397; b) J. Jansson, J. Catal. 2000, 194, 55 – 60.
[8] J. Scheidtmann, J. Saalfrank, W. F. Maier, Stud. Surf. Sci. Catal.
2003, 145, 13 – 20.
[9] a) A. Holzwarth, H.-W. Schmidt, W. F. Maier, Angew. Chem.
1998, 110, 2788 – 2792; Angew. Chem. Int. Ed. 1998, 37, 2644 –
2647; b) A. Holzwarth, W. F. Maier, Platinum Met. Rev. 2000 44,
16 – 21.
[10] J. N. Cawse in Experimental Design for Combinatorial and High
Throughput Materials Development (Ed.: J. Cawse), Wiley, New
Jersey, 2003.
[11] G. Kirsten, PhD thesis, UniversitRt des Saarlandes, 2003.
Received: May 20, 2003
Revised: October 23, 2003 [Z51935]
Angew. Chem. Int. Ed. 2004, 43, 2028 –2031
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Без категории
Размер файла
127 Кб
oxidation, free, metali, temperature, evolution, room, directed, catalyst, noble
Пожаловаться на содержимое документа