During the past two decades a variety of transition-metal catalyzed reactions have been introducedin synthetic organic chemistry. Among the most popular examples are asymmetric oxidations and reductionsand a variety of cross-coupling reactions to form C − C and C-heteroatom bonds. Examples are the Heck reaction, Suzuki–Kumadaand Sonagashira coupling reactions and the aryl-aminations introduced by Buchwald and Hartwig.
These reactions are homogeneously catalyzed using a metal complex containing expensive metalsand ligands many of which are difficult to synthesize. In most cases the catalysts are very efficient,therefore quite often only millimolar amounts or less are applied. This portfolio of new reactions wasintroduced into lab-scale synthesis within a few years of their discovery and is today frequently usedfor first syntheses of active pharmaceutical ingredients and other high-value fine chemicals. Protocolsfor the use of these reactions in very small-scale combinatorial synthesis have been developed and moreand more hits and later on development products resulted from these efforts. Pharmaceutical products need,because of the intensive and time-consuming clinical development, up to ten years until the first industrial-scaleproduction has to be scheduled. As part of the process development a method to fully separate thecatalyst components from the products after the reaction has to be worked out. Only extremely small residuesof metals or of ligands are tolerated in active pharmaceutical ingredients (APIs). The concentration ofmetals such as Pt, Pd, Ir, Rh, Ru or Os has been limited to 5 ppm by the recommendations of the EuropeanAgency for the Evaluation of Medicinal Products (EMEA) [1].A variety of methods can be applied to fulfil these requirements (as exemplified for palladium in [2] ).
Besides environmental and health criteria, cost considerations also motivate the development of processeswhich enable the separation and reuse of the catalyst complex after the reaction. Chiral ligands or ligandsfrequently used for cross-coupling reactions are difficult to synthesize and only some ten's of kilogramsare needed even for the production of many tons of a final product. Both the demanding synthesis andthe production in kg-labs or small pilot plants makes these ligands very expensive. Prices ranging from2000 to 100000 $/kg and more are quite common [3].Both ecological and economical constraints force chemical and pharmaceutical industries to establish processesfor the separation of the catalyst after the reaction to achieve a multiple use of the catalysts.
Phase separation during or after a reaction is a proven method in chemical industry forthe extraction of a chemical compound. It has also been successfully introduced into large-scale catalysisprocesses [4, 5]mainly using systems with one aqueous and one organic phase. Typical catalysts for enantioselective orcross-coupling reactions normally are soluble in organic solvents only, just like the products of the reaction.Therefore, these proven systems cannot be applied in these cases. All these reasons have led to an increasinginterest in new solvent systems for homogeneously catalyzed reactions: Supercritical fluids [6], ionic liquids [7], thermomorphicsolvents [8, 9]and fluorous phases [10] are the most widely studied new“green” solvents that open a door for a wide variety of applications in organic chemistry.Each one seems to have a promising potential for industrial use. What is missing is the experiencewith their use, a better knowledge about scope and limitation of the systems and in particular a breakthroughin large-scale application. In the same way as a new catalyst system normally needs many years oflab experience until a first technical use arises, new solvents will also need this time period priorto their first use at the ton or multi-ton scale.
The ConNeCat (ConNeCat is the German Competence Network Catalysis; http://www.connecat.de) lighthouseproject “Regulated Systems for Muliphase Catalysis—Smart Solvents/Smart Ligands” whichwas funded by the German Ministry of Research and Education (BMBF) provided the opportunity to gain importantscientific and technical experience with the new solvent systems. Seven academic groups and four industrialpartners got the chance to modify and validate these new solvent systems within core processes of the fourcompanies. Basic aspects and new concepts could be worked out at the universities and research institutes.The industrial groups compared the new technologies with the state of the art and pointed towards the large-scaleapplicability. Several examples for the use of the new multiphase systems in homogeneous catalysis havebeen worked out. Each of the four systems investigated shows promise for industrial use, but —ascould be expected—every single system has different applications and limitations. This review summarizessome of the main results of this research network together with selected parallel developments from theliterature taking an industrial point of view.