Temperature-controlled catalyst recycling in homogeneous transition-metal catalysis: Minimization of catalyst leaching

Article abstract of DOI:10.1002/anie.201208667

Reduce-reuse-recycle! One of the challenges in applied homogeneous catalysis is the efficient recycling of the valuable metal catalyst. The catalyst recycling concept of temperature-controlled multicomponent solvent systems was successfully applied to the hydroformylation of long-chain alkenes. The factors that signficantly influence catalyst leaching and how it can be minimized effectively were systematically investigated for the first time. Copyright

Full text of DOI:10.1002/anie.201208667

.
Angewandte
Communications
DOI: 10.1002/anie.201208667
Catalyst Recycling
Temperature-Controlled Catalyst Recycling in Homogeneous
Transition-Metal Catalysis: Minimization of Catalyst Leaching**
Yvonne Brunsch and Arno Behr*
Catalysis is a key technology in the chemical industry and has
a significant impact on value creation. However, less than
10% of all chemical processes are based on homogeneously
catalyzed reactions where the catalyst and the product are in
the same phase after reaction.[1] The advantages of the
homogeneous operation mode are often mild reaction con-
ditions and high selectivities. However, the technological
challenge of homogeneous catalysis is the efficient separation
of the catalyst from the product as well as the recycling of the
metal catalyst.[2]
thermomorphic) multicomponent solvent systems (TMS
systems), which was developed in our group in 1999
(Figure 1).[4]
The principle of TMS systems is based on temperature-
dependent phase behavior. During reaction a TMS system is
single phase, thus homogeneous, so that no mass transport
The first industrial-scale application of homogeneous
transition-metal catalysts in the chemical industry was hydro-
formylation. More than 8 million tons of aldehydes and
alcohols are produced worldwide each year by the hydro-
formylation of alkenes. These intermediates are used for the
production of detergents and plasticizers. The most important
technical hydroformylation is the conversation of the short-
chain alkene propene, which is catalyzed by rhodium. In one
important process alternative, the Ruhrchemie/Rhꢀne–Pou-
lenc process, catalyst recycling is accomplished by aqueous–
organic liquid–liquid two-phase extraction.[3]
Figure 1. The TMS catalyst recycling concept.
In the hydroformylation of long-chain alkenes this two-
phase extraction technique is not appropriate for catalyst
recycling. Long-chain alkenes are only poorly soluble in the
aqueous catalyst phase; consequently, mass transfer is limited,
reaction times are long, and conversions are low. In the
chemical industry the hydroformylation of long-chain alkenes
to aldehydes with subsequent hydrogenation to alcohols is
possible by the thermal separation of the long-chain products
under reduced pressure. However, owing to the high boiling
points of the long-chain products and thereby thermal stress
for the temperature-sensitive catalyst system, the catalyst
slowly decomposes under these conditions.[3]
limitation occurs. After the temperature is reduced the TMS
system possesses a miscibility gap and separates into two
liquid phases, a polar and apolar phase. The catalyst is
dissolved in the polar phase, while the apolar phase contains
the organic product. The different solubility properties of the
catalyst and the product in the phases make their separation
possible. After reaction and phase separation the apolar
product phase can be purified in a down-streaming process,
while the polar catalyst phase can be recycled. If the catalyst is
dissolved in the product phase, a loss of the catalyst (catalyst
leaching) takes place.[5]
The model reaction in our work was the homogeneously
catalyzed hydroformylation of the long-chain 1-dodecene to
the linear n-tridecanal with the highly selective catalyst
system [Rh(acac)(CO)2]/Biphephos (Scheme 1, acac =
acetylacetonato). Our results concerning the hydroformyla-
tion in TMS systems are contributing to an interdisciplinary
process development project.[6] During hydroformylation
isomeric alkenes and dodecane are formed. According to
the literature, the use of the bidentate phosphite ligand
Biphephos provides the aldehydes with very high n/iso ratios
of up to 97:3.[7]
Consequently an effective catalyst recycling concept is
required for the hydroformylation of long-chain alkenes with
the highly active but expensive rhodium catalyst. Therefore
we applied the concept of the temperature-controlled (or
[*] M. Sc. Y. Brunsch, Prof. Dr. A. Behr
Department of Biochemical and Chemical Engineering
Technical Chemistry A (Chemical Process Development)
Technical University Dortmund
Emil-Figge-Strasse 66, 44227 Dortmund (Germany)
E-mail: arno.behr@bci.tu-dortmund.de
Homepage: http://www.tca.bci.tu-dortmund.de
[**] This project was funded by the Deutsche Forschungsgemeinschaft
(Collaborative Research Centre InPROMPT Transregio 63). We
thank Umicore for the rhodium catalyst and I. Henkel for ICP-OES
measurements.
Previous studies of the rhodium-catalyzed hydroformyla-
tion of 1-dodecene predominantly dealt with the use of the
liquid–liquid two-phase technique for the development of
a catalyst recycling method.[8] Furthermore, the application of
microemulsions and micellar solvent systems (MLS) have also
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201208667.
1586
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1586 –1589

Angewandte
Chemie
hyde yields higher than 80% were
obtained. Thus the product concentration
increases along with the substrate con-
centration. At higher substrate or product
concentrations the catalyst leaching also
increases. Up to
a concentration of
25 wt% substrate, phase separation
could be achieved at a separation temper-
ature of 258C. At 30 wt% a lower sepa-
ration temperature of 108C was necessary
Scheme 1. The hydroformylation of 1-dodecene.
been described.[9] The TMS catalyst recycling concept has
already been applied successfully for many homogeneously
catalyzed reactions.[10] The use of thermomorphic solvent
systems has also been described for the hydroformylation of
long-chain alkenes.[11] The hydroformylation of 1-dodecene
has been studied in the organic two-phase system poly(ethyl-
ene glycol)-400/1,4-dioxane/n-heptane, which is heterogene-
ous under reaction conditions. Under optimal reaction
conditions a conversion of 96% was achieved with a moderate
n/iso ratio of 1:1. Catalyst recycling was accomplished in 23
recycling cycles with an average catalyst leaching of 1%.[12] In
our interdisciplinary studies the TMS system dimethylforma-
mide (DMF)/decane was examined.[13] The thermodynamic
basics and the temperature-dependent phase behavior of the
pure substances were determined. For the hydroformylation
of 1-dodecene, catalyst recycling with eight recycling cycles
was already demonstrated with the TMS system DMF/
decane.[13a]
Now for the first time we have focused on the systematic
investigation of the catalyst separation in the context of
integral process development. Nearly complete recovery of
the catalyst is absolutely necessary for an economically viable
process with expensive rhodium catalysts. Therefore knowing
the factors that influence catalyst leaching is essential for the
chemical process development of homogeneous catalyzed
reactions. This work deals with the question of how catalyst
leaching can be minimized effectively. To evaluate the
efficiency of catalyst recycling, the separation of the product
and catalyst was investigated and the catalyst leaching
determined.
Table 1: Hydroformylation results and catalyst leaching upon variation of
the TMS system.[a]
Entry
1
Polar TMS
component
X [%][b]
Y [%][c]
n/iso
Rh [%][d]
35
P [%][e]
36
propylene
carbonate
acetonitrile
dimethyl-
99
79
99:1
2
3
99
99
79
80
99:1
99:1
43
7
34
9
formamide
[a] T=1008C, p=20 bar, 30 mL solvent: polar TMS solvent component/
decane 50:50 (wt%), t=1 h. [b] Conversion X of 1-dodecene. [c] Yield of
aldehyde Y. [d] Rhodium leaching. [e] Ligand leaching.
Table 2: Hydroformylation results and catalyst leaching upon variation
[a]
of the amount of substrate and the separation temperature TSep
.
Entry C[b] TSep [8C] X [%][c] Y [%][d] n/iso Rh [%][e] P [%][f]
1
2
3
4
5
6
7
8
10
15
20
25
30
15
15
15
15
15
15
25
25
25
25
10
35
25
15
5
99
99
99
99
99
99
99
99
99
99
99
80
80
81
81
81
81
81
81
81
81
81
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
3
7
5
9
19
45
61
10
5
5
4
1
0
21
46
61
13
6
4
3
2
2
9
10
11
0
À9
[a] T=1008C, p=20 bar, 30 mL solvent: DMF/decane 50:50 (wt%),
t=1 h. [b] Mass fraction of the substrate 1-dodecene in wt% (entries 1–
5) or separation temperature TSep in 8C (entries 6–11). [c] Conversion X of
1-dodecene. [d] Yield of aldehyde Y. [e] Rhodium leaching. [f] Ligand
leaching.
First, three different TMS systems were tested for catalyst
recycling. In the TMS system propylene carbonate/decane
(entry 1 in Table 1) as well as in the TMS systems acetonitrile/
decane (entry 2) and DMF/decane (entry 3) high conversions
of 99% were achieved with n/iso ratios of 99:1. However, the
catalyst leaching with the propylene carbonate/decane system
into the product phase (35%) was five times higher than that
with the DMF/decane system. Nearly half of the rhodium
catalyst was dissolved in the “wrong” phase in the acetoni-
trile/decane system. Therefore the TMS system DMF/decane,
which has the lowest catalyst leaching, was considered for
ongoing investigations.
for phase separation. Furthermore, we found that the
solubilizing product influences the catalyst leaching: it is
favorable to run the hydroformylation reaction with lower
product concentrations for effective separation of the catalyst
and therefore minimization of the catalyst leaching.[13a] We
also found that the leaching of the rhodium catalyst is
equimolar to the leaching of the phosphite ligand, which
indicates that a rhodium–ligand complex is dissolved in the
product phase.
We then determined what parameters influence catalyst
leaching and how it can be minimized effectively. Based on
the investigation of the phase behavior, the substrate
concentration was varied between 10–30 wt% (Table 2,
entries 1–5).[13a] In all hydroformylation experiments alde-
We also investigated the influence of the separation
temperature Tsep on the catalyst leaching (Table 2, entries 6–
11). We found that catalyst leaching into the product phase
can be minimized effectively by reducing the separation
temperature. At the same time the solubility of the polar
Angew. Chem. Int. Ed. 2013, 52, 1586 –1589
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1587

.
Angewandte
Communications
solvent in the product phase is reduced at lower temperatures.
If the loss of polar solvent is reduced, then the catalyst
leaching is reduced. At a separation temperature of 358C the
catalyst leaching is 10%, while no catalyst loss can be
detected at a separation temperature of À98C with ICP-OES
analysis.
To check whether the catalyst concentration has a signifi-
cant influence on catalyst leaching, the substrate/metal molar
ratio was varied between 100:1 and 5000:1 (Table 3). With
the catalyst leaching could be reduced effectively. When
hexane (C6) was used, 25% of the rhodium was carried into
the product phase, while with hexadecane (C16) the rhodium
leaching was only 2%.
We also checked the effect of the chain length of the
substrate and thus the chain length of the product on the
catalyst leaching (Table 4, entries 7–13). With long-chain
substrates and products, up to 13% of the rhodium catalyst
was dissolved in the product phase. This is because the
solubility of the polar DMF in the product phase increases
with increasing chain length and therefore also catalyst
leaching. With increasing chain lengths, both substrates and
products were better dissolved in the product phase. The
shorter the chain length of the substrate or product, the lower
the undesirable leaching of the rhodium catalyst.
The efficiency of the optimized TMS concept was verified
in recycling experiments, in which the catalyst phase was
separated after reaction and recycled for the next experiment
(Figure 2). The hydroformylation of 1-dodecene was carried
out in the TMS system DMF/decane and the phase separation
Table 3: Hydroformylation results and catalyst leaching upon variation
of the substrate/metal ratio.[a]
Entry Substrate/metal X [%][b] Y [%][c] n/iso Rh [%][d] P [%][e]
1
2
3
100:1
1000:1
5000:1[f]
99
99
97
89
81
82
98:2
99:1
98:2
11
6
1
14
7
3
[a] 0.026 mol substrate, T=1008C, p=20 bar, 30 mL solvent: DMF/
decane 50:50 (wt%), t=1 h. [b] Conversion X of 1-dodecene. [c] Yield of
aldehyde Y. [d] Rhodium leaching. [e] Ligand leaching. [f] Reaction time
t=5 h.
increasing catalyst concentration, the aldehyde yield
increases to 89%. However, the greater the catalyst loading,
the higher the level of catalyst leaching. At a substrate/metal
ratio of 5000:1 only 1% of the rhodium catalyst is dissolved in
the product phase. Our results demonstrate that a low catalyst
concentration is preferable to minimize the catalyst leaching.
An important factor in the selection of solvents for a TMS
system is the polarity. Therefore the length of the carbon
chain of the apolar TMS component was varied (Table 4,
entries 1–6). With increasing chain length of the apolar
alkane, the polarity difference to the polar solvent DMF is
raised. The chain length of the apolar solvent component had
no influence on the hydroformylation reaction. In all experi-
ments high conversions, yields, and n/iso ratios were obtained.
By increasing the chain length of the apolar TMS component
Figure 2. Catalyst recycling in the hydroformylation of 1-dodecene in
30 consecutive experiments.
was conducted at À108C after each reaction. For the first
reaction cycle a substrate/metal ratio of 1000:1 was selected
and a reaction time of only one hour was sufficient for
complete conversion.[13a]
Catalyst recycling was carried out for a total of 30
consecutive experiments. Aldehyde yields of nearly 80%
were obtained as well as n/iso ratios of 99:1. In recycling
experiment 13 it turned out that the reaction time of one hour
was not sufficient for complete conversion; this was apparent
from the decreasing aldehyde yields and increasing alkene
yields. By extending the reaction time the conversion could be
increased. It is supposed that a small amount of the catalyst is
removed with the product phase in each recycling cycle. For
this reason we investigated the product phase from recycling
experiment 30, which showed a rhodium content of 1.7 ppm
(1%) and a phosphorus ligand content of 14 ppm (1%). The
reduction of the catalyst concentration resulted in longer
reaction times. This hypothesis was additionally proved by
recording the reaction progress of recycling experiment 30
(see the Supporting Information). After 6 h a high aldehyde
yield of nearly 80% was obtained with an n/iso ratio of 99:1.
Our work shows that the TMS concept can be a valuable
tool for the recycling of homogeneous catalysts under mild
conditions and therefore it provides economic potential for
Table 4: Hydroformylation results and catalyst leaching upon variation
of the apolar TMS component and the chain length of the 1-alkene.[a]
Entry
U[b]
V[c]
X [%][d]
Y [%][e]
n/iso
Rh [%][f]
P [%][g]
1
2
3
4
5
6
7
8
C6
C8
C12
C12
C12
C12
C12
C12
C6
99
99
99
99
99
99
97
99
99
99
99
99
99
81
81
81
81
81
82
75
82
81
81
81
81
79
99:1
99:1
99:1
99:1
99:1
99:1
99:1
98:2
99:1
99:1
98:2
97:3
97:3
25
21
6
6
4
2
1
2
4
29
24
7
7
4
4
2
4
6
2
4
7
5
C10
C12
C14
C16
C10
C10
C10
C10
C10
C10
C10
C8
9
C10
C12
C14
C16
C18
10
11
12
13
5
11
10
13
[a] T=1008C, p=20 bar, 30 mL solvent: DMF/decane 50:50 (wt%),
t=1 h. [b] Chain length of the apolar TMS component. [c] Chain length
of the 1-alkene. [d] Conversion X of 1-dodecene. [e] Yield of aldehyde Y.
[f] Rhodium leaching. [g] Ligand leaching.
1588
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1586 –1589

Angewandte
Chemie
[4] A. Behr, N. Toslu, Chem. Ing. Tech. 1999, 71, 490 – 493.
the technical application of homogeneous transition-metal
catalysis. This was the first systematic investigation that
focused on whether catalyst leaching can be influenced and
which influencing factors are significant. It was shown that the
product has an unfavorable influence on catalyst leaching.
Furthermore we demonstrated that catalyst leaching can be
minimized effectively by reducing the separation temperature
and the catalyst concentration and by increasing of the
polarity difference between the TMS solvent components.
The TMS catalyst recycling concept offers the possibility for
the development and realization of industrial processes of
homogeneous catalyzed reactions.
[5] a) B. Cornils, W. A. Herrmann, Aqueous-Phase Organometallic
Catalysis: Concepts and Applications, Wiley-VCH, Weinheim,
2004, pp. 327 – 329; b) A. Behr, R. Roll, Chem. Ing. Tech. 2005,
77, 748 – 752; c) A. Behr, G. Henze, D. Obst, B. Turkowski,
Green Chem. 2005, 7, 645 – 649; d) A. Behr, G. Henze, R.
Schomꢃcker, Adv. Synth. Catal. 2006, 348, 1485 – 1495; e) A.
Behr, B. Turkowski, R. Roll, R. Schçbel, G. Henze, Top.
Organomet. Chem. 2008, 23, 19 – 25; f) A. Behr, G. Henze, L.
Johnen, C. Awungacha, J. Mol. Catal. A 2008, 285, 20 – 28; g) A.
Behr, A. Wintzer, Chem. Ing. Tech. 2011, 83, 1356 – 1370.
[6] http://www.inprompt.tu-berlin.de/.
[7] a) C. Vogl, E. Pꢃtzold, C. Fischer, U. Kragl, J. Mol. Catal. A 2005,
232, 41 – 44; b) E. Billig, A. G. Abatjoglou, D. R. Bryant,
EP0214622B1, 1992; c) A. Behr, D. Obst, C. Schulte, T. Schosser,
J. Mol. Catal. A 2003, 206, 179 – 184; d) X. Y. Huang, R. X. Li, H.
Chen, X. J. Li, Chin. Chem. Lett. 2003, 198, 1 – 7; e) M.
Haumann, M. Jakuttis, R. Franke, A. Schçnweiz, P. Wasser-
scheid, ChemCatChem 2011, 3, 1822 – 1827; f) M. Jakuttis, A.
Schçnweiz, S. Werner, R. Franke, K. D. Wiese, M. Haumann, P.
Wasserscheid, Angew. Chem. 2011, 123, 4584 – 4588; Angew.
Chem. Int. Ed. 2011, 50, 4492 – 4495.
[8] a) H. Chen, Y. Z. Li, R. X. Li, X. J. Li, J. Mol. Catal. A 2003, 232,
41 – 44; b) W. J. Tic, Sep. Sci. Technol. 2009, 44, 2609 – 2625; c) C.
Yang, X. Y. Bi, Z. S. Mao, J. Mol. Catal. A 2002, 187, 35 – 46;
d) M. Yuan, H. Chen, R. X. Li, Y. Z. Li, X. J. Li, Catal. Lett.
2004, 94, 15 – 16; e) Y. Zhang, Z. S. Mao, J. Chen, Ind. Eng.
Chem. Res. 2001, 40, 4496 – 4505; f) Y. Q. Zhang, Z. S. Mao, T. Y.
Chen, Catal. Today 2002, 74, 23 – 25; g) E. Pꢃtzold, G. Oehme, C.
Fischer, M. Frank, J. Mol. Catal. A 2003, 200, 95 – 103; h) H.
Klein, R. Jackstell, M. Beller, Chem. Commun. 2005, 2283 –
2285; i) Z. J. Li, J. Y. Jiang, Y. H. Wang, Chin. Chem. Lett.
2010, 21, 515 – 518.
[9] a) M. Haumann, H. Koch, P. Hugo, R. Schomꢃcker, Appl. Catal.
A 2002, 225, 239 – 249; b) M. Li, B. Xu, H. J. Zheng, X. Y. Huang,
Y. Z. Li, X. J. Li, Chin. Chem. Lett. 2004, 15, 1022 – 1024.
[10] a) A. Behr, C. Fꢃngewisch, J. Mol. Catal. A 2003, 197, 115 – 126;
b) A. Behr, Q. Miao, J. Mol. Catal. A 2004, 222, 127 – 132; c) A.
Behr, R. Roll, J. Mol. Catal. A 2005, 239, 180 – 184; d) A. Behr,
L. Johnen, A. J. Vorholt, ChemCatChem 2010, 2, 1271 – 1277;
e) A. Behr, L. Johnen, N. Rentmeister, Adv. Synth. Catal. 2010,
352, 2062 – 2072.
Experimental Section
In a typical hydroformylation reaction the catalyst precursor [Rh-
(acac)(CO)2] and the ligand Biphephos (ratio 1:5) were dissolved in
the solvents (30 mL DMF/decane 50:50 (wt%)). After the addition of
1-alkene (0.026 mol) the reaction mixture was transferred under
argon atmosphere into a 300 mL steel Parr autoclave. The autoclave
was adjusted with 20 bar of synthesis gas (CO/H2 1:1). The autoclave
was heated to 1008C and the reaction time of 1 h started at the onset
of heating. Samples were removed with a capillary and the reaction
progress was recorded. The reaction was stopped by cooling with an
ice-water bath. After removal of the synthesis gas, the reaction
mixture was tempered to the required separation temperature in
a double-walled 50 mL separating funnel with a cooling circulation
thermostat (HAAKE K40, Thermo Electron Corporation HAAKE
DC50, internal temperature regulation, tempering medium: ethylene
glycol/water 1:1). After phase separation samples from both the
product and catalyst phase were analyzed by gas chromatography and
emission spectrometry.
Received: October 29, 2012
Published online: December 12, 2012
Keywords: catalyst recycling · homogeneous catalysis ·
.
hydroformylation · rhodium · thermomorphic solvent mixtures
[11] a) A. Behr, D. Obst, C. Schulte, Chem. Ing. Tech. 2004, 76, 904 –
910; b) A. Behr, D. Obst, B. Turkowski, J. Mol. Catal. A 2005,
226, 215 – 219; c) J. Tijani, B. ElAli, Appl. Catal. A 2006, 303,
158 – 165; d) M. S. Shaharun, H. Mukhtar, B. K. Dutta, Chem.
Eng. Sci. 2008, 63, 3024 – 3035; e) M. S. Shaharun, H. Mukhtar,
B. K. Dutta, Chem. Eng. Sci. 2010, 65, 3903 – 3905; f) M. S.
Shaharun, B. K. Dutta, H. Mukhtar, S. Maitra, Chem. Eng. Sci.
2010, 65, 273 – 281; g) M. S. Shaharun, J. Appl. Sci. 2011, 11,
1157 – 1163; h) A. Behr, A. J. Vorholt, Top. Organomet. Chem.
2012, 39, 103 – 128.
[12] Y. C. Yang, J. Y. Jiang, Y. H. Wang, C. Liu, Z. L. Jin, J. Mol.
Catal. A 2007, 261, 288 – 292.
[13] a) E. Schꢃfer, Y. Brunsch, G. Sadowski, A. Behr, Ind. Eng.
Chem. Res. 2012, 51, 10296 – 10306; b) A. Behr, Y. Brunsch, A.
Lux, Tetrahedron Lett. 2012, 53, 2680 – 2683.
[1] F. Schꢁth, Chem. Unserer Zeit 2006, 40, 92 – 103.
[2] a) A. Behr, Angewandte Homogene Katalyse, Wiley-VCH,
Weinheim, 2008, pp. 1 – 28; b) A. Behr, P. Neubert, Applied
Homogeneous Catalysis, Wiley-VCH, Weinheim, 2012, pp. 1 –
32; c) D. J. Cole-Hamilton, R. P. Tooze, Catalyst Separation,
Recovery and Recycling: Chemistry and Process Design,
Springer, Dordrecht, 2006, pp. 1 – 36.
[3] a) M. Baerns, A. Behr, A. Brehm, J. Gmehling, H. Hofmann, U.
Onken, A. Renken, Technische Chemie, Wiley-VCH, Weinheim,
2006, pp. 574 – 576; b) P. W. N. M. van Leeuwen, C. Claver,
Rhodium catalyzed Hydroformylation, Kluwer Academic Pub-
lishers, Dordrecht, 2000, pp. 1 – 280; c) K. Weissermehl, H. J.
Arpe, Industrial Organic Chemistry, Wiley-VCH, Weinheim,
2003, pp. 127 – 140; d) B. Cornils, W. A. Herrmann, I. T.
Horvꢂth, W. Leitner, S. Mecking, H. Olivier-Bourbigou, D.
Vogt, Multiphase Homogeneous Catalysis, Vol. 1 and 2, Wiley-
VCH, Weinheim, 2008.
Angew. Chem. Int. Ed. 2013, 52, 1586 –1589
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1589