The research interests of my group span a wide range, from organic and organometallic synthesis, through catalysis and the surface chemistry of amorphous materials, to solid-state NMR spectroscopy. Over the years, our research has expanded especially into the following different areas:
(1) IMMOBILIZED CATALYSTS
(2) THE SURFACE CHEMISTRY OF OXIDE MATERIALS
(3) SOLID-STATE NMR SPECTROSCOPY
(1) The immobilization of catalysts is of growing interest because the advantages of heterogeneous catalysts can, in principle, be combined with those of homogeneous catalysts. Immobilized or surface-bound homogeneous catalysts are easy to recycle, and they can be highly active and selective. Furthermore, they are amenable to systematic design. We find highly interesting results when heterobimetallic systems, such as the Sonogashira Pd/Cu catalyst, are involved.
(2) Especially in this area the immobilization step requires a thorough understanding of the surface chemistry of the oxide support materials. Therefore, we investigate not only the reactivity of metal complexes and linkers, but also their mobility on the surfaces.
(3) The most powerful analytical tool for investigating amorphous materials is solid-state NMR spectroscopy. We optimized this method especially for surface-bound species, and in the meantime we can also study reactions on surfaces, or analyze the nature of our anchored linkers or catalysts.
1. IMMOBILIZED CATALYSTS
The easiest way to bind homogeneous catalysts to an oxide support is using a bifunctional linker,1 as in the following example:2
Since most transition metal complexes form stable phosphine adducts, functionalized phosphines such as Ph2P(C6H4)Si(OEt)3 or Ph2P(CH2)3Si(OEt)3 can be used. Chelating phosphines, exemplified by our new unsymmetrical versions Ph2P(CH2)xPPh(CH2)ySi(OEt)3 (x, y = 2, 3), also lead to firmly bound, well-defined catalysts. Since phosphines with anchoring functionalities, such as ethoxysilane groups, are not abundant in the literature, a substantial part of our efforts goes into developing convenient new syntheses of various linkers. For example, recently we demonstrated that dppm-type phosphines with ethoxysilane groups can be obtained in a versatile one-pot synthesis (SCHEME 1).3
SCHEME 1. Convenient one-pot synthesis of chelate phosphines with ethoxysilane groups.3
The immobilized systems studied have included the Sonogashira catalyst,2 as well as carbonylnickel and Wilkinson-type complexes,4 as shown in SCHEME 2. Their catalytic activity with respect to olefin hydrogenation has been investigated under different reaction conditions. As compared to their homogeneous analogs they all showed equal activity and selectivity. However, they can easily be separated from the reaction mixture, because they settle down within minutes, leaving a clean supernatant solution of the products. Furthermore, they can be efficiently recycled, for example 13 times in a batchwise manner for the rhodium complex with x = 3 and y = 2 and dodecene as the substrate, as shown in SCHEME 2.4
SCHEME 2. An example of a rhodium catalyst that can be recycled many times.4
In the course of our studies, we found that some simple measures can have far-reaching positive effects. For example, the lifetimes of all the immobilized catalysts, independent of the metal moiety or linker, can be enhanced substantially by diluting them on the surface. This is because dimerization of the activated catalysts leads to their deactivation. Further improvements are achieved by the proper choice of the support, pre-drying measures, and a favorable average pore diameter.4
We have also had ongoing interests in immobilized heterobimetallic systems. For example, the Sonogashira reaction involves the coupling of aryl acetylenes with aryl halides through the combined action of Pd and Cu complexes,2 (see also NMR section 3 below). We developed silica-bound Sonogashira catalysts that could be recycled 18 times without major loss of activity.2
2. SURFACE CHEMISTRY OF OXIDE MATERIALS
The success of an immobilized catalyst is crucially dependent on the choice of the proper support material, and an understanding of its reactivity. Over the years we could demonstrate that silica is the optimal support for complexes attached to the surface by bifunctional phosphines, such as Ph2P(CH2)3Si(OEt)3. Other oxide supports, for example titania or alumina, lead to leaching of the linker. Furthermore, some linkers are vulnerable with respect to moisture on oxide surfaces. For example, the tripod-type linker in SCHEME 3 loses one Ph2PCH2 group on wet silica, while it stays intact on a support rigorously dried in vacuo at 600 °C.1
SCHEME 3. Silicas varying in their degree of dryness lead to different immobilized linkers.1
The fact that oxide surfaces are not always innocent can, however, be turned into an exploitable advantage. For example, we have demonstrated that phosphines can be quarternized by the combined action of the silica surface and ethoxysilane groups, forming ethylphosphonium salts (see for example SCHEME 4).5
This finding led us to a new type of "linker kit" (SCHEME 4):5 Ethoxysilane groups quarternize phosphines even if they are not bound to the same molecule. In other words, providing the right phosphine means that the synthesis of phosphines with intramolecular ethoxysilane groups is no longer necessary. This has led us to develop a new generation of linkers, starting from tetraphosphines with rigid backbones, that can prevent the interaction of the catalysts with the reactive oxide surface, and thus prolong their lifetimes. The new linkers have the unprecedented feature that one or two phosphine groups selectively, depending on the reaction conditions, are bound via ionic bonds to the surface, while two or three phosphine moieties remain for later metal complexation, as shown in SCHEME 4.5
SCHEME 4. Immobilized linker system with a tetraphenylelement backbone.5
Not only linkers show surprising reactivity with oxide supports, but also metal complexes on their own. In lucky cases the reaction of a metal complex with an oxide surface leads to an active catalyst. This is for example the case with the Union Carbide (UC) ethylene polymerization catalyst,6 which is the reaction product of chromocene and silica (SCHEME 5). We could demonstrate that the UC catalyst is composed of several mono- and dinuclear chromium surface species. Depending on the immobilization conditions, especially the degree of dryness of the support and the solvent used, the one or other product is favored. The most active species for polymerization is a mononuclear surface-bound Cr(III) halfsandwich compound, as shown in SCHEME 5.
SCHEME 5. The active component of the Union Carbide catalyst.6
3. SOLID-STATE NMR SPECTROSCOPY
Since many years one of our main research interests is solid-state NMR spectroscopy, especially with respect to surface-bound linkers and catalysts. Besides the "classical" solid-state NMR (CP/MAS) with fast rotation of a dry sample, we have pioneered the art of measuring suspensions of materials (HRMAS).1,2,7 The latter has the advantage that the linewidths obtainable are very small, and therefore HRMAS allows even two-dimensional NMR. This is for example demonstrated in SCHEME 6. The 31P suspension HRMAS spectrum of the surface-bound copper complex gives very narrow lines even at the low spinning rate of 2 kHz, in contrast to the 31P CP/MAS of the dry material. This allows the application of the two-dimensional NOESY experiment (bottom), analogous to that commonly conducted with liquid samples. In this case, the NOESY cross-peak proves the close proximity of the different phosphorus nuclei, establishing the structure of the catalyst.2
SCHEME 6. 31P spectra of the depicted Cu complex. Top: 31P CP/MAS (4 kHz rotational speed), middle: 31P HRMAS (2 kHz).
Bottom: Two-dimensional 31P,31P NOESY spectrum of the Cu complex.2
Selected References
| [1] | J. Blümel, "Molecular Catalysts Immobilized of Oxide Supports: New Insights by Solid-State NMR Spectroscopy", Coord. Chem. Rev. 252 (2008) 2410. pdf |
| [2] | T. Posset, J. Blümel, "New Mechanistic Insights Regarding Pd/Cu Catalysts for the Sonogashira Reaction: HRMAS NMR Studies of Silica-Immobilized Systems", J. Am. Chem. Soc. 128 (2006) 8394. pdf |
| [3] | F. Piestert, R. Fetouaki, M. Bogza, T. Oeser, J. Blümel, "Easy one-pot synthesis of new dppm-type linkers for immobilizations", Chem. Commun. (2005) 1481. pdf |
| [4] | C. Merckle, J. Blümel, "Silica- and Alumina-Immobilized Rhodium Catalysts for Olefin Hydrogenation: Enhancement of Activities, Stabilities, and Lifetime", Topics in Catalysis 34 (2005) 5. pdf |
| [5] | Y. Yang, B. Beele, J. Blümel, "Easily Immobilized Di-and Tetraphosphine Linkers: Rigid Scaffolds that Prevent Interactions of Metal Complexes with Oxide Supports", J. Am. Chem. Soc. 130 (2008) 3771. pdf |
| [6] | M. Schnellbach, J. Blümel, F. H. Köhler, "The Union Carbide Catalyst (Cp2Cr + SiO2), Studied by Solid-State NMR", J. Organomet. Chem. 520 (1996) 227. pdf |
| [7] | S. Brenna, T. Posset, J. Furrer, J. Blümel, "Two-Dimensional 1H,13C HRMAS, and 14N NMR Spectroscopy of Ionic Liquids Immobilized on Silica", Chem. Eur. J. 12 (2006) 2880. pdf |