By Valentinos Mouarrawis and Sander Kluwer
Catalysis is key for modernizing the chemical industry by making more efficient use of resources and minimizing waste production. The development of efficient catalytic processes is of crucial importance for sustainable-oriented applications, positioning catalysis at the heart of our quality of life. In the fine chemical industry, complex transformations are typically done by transition metal catalysts. As a result, there is a high demand for more efficient technologies with the aim of sustainable and economically valuable chemical processes.
Palladium (Pd) often referred to as the chameleon of catalysis, is commonly found as a catalyst in the synthesis of fine chemicals, pharmaceutical intermediates, and active pharmaceutical ingredients (APIs). The extensive research devoted to Pd-catalysis made a toolbox available to the synthetic chemist for both achiral and chiral chemical transformations (i.e., hydrogenation/dehydrogenation, C−C and C−heteroatom bond-formation, and C−H bond functionalizations) and provided access to more complex structures with less waste and fewer steps. However, palladium is a scarce metal with a high price ($62,081.45 per kg), and therefore different strategies involving more abundant metals are crucial for a sustainable future.
Figure 1. Nickel vs palladium, which metal to choose.
Nickel (Ni) would be a good alternative for Pd catalysts as the heterogeneous Ni catalysts are effective in numerous organic transformations on very large scales (i.e., large-scale hydrogenation of benzene to cyclohexane and production of sorbitol from glucose.). When we look at the industry and the use of homogeneous Ni-catalysts, it is clear that the products obtained (i.e., ethylene carbonylation to propionic acid, oligomerization of alkenes, hydrocyanation ) are not as diverse as those described for the Pd-counterparts. The lower cost of Ni ($22.46 per kg) allows the use of a higher catalyst loading and in turn, can benefit the scalability of a process. Although nickel is cheaper, palladium has been the workhorse in the fine chemical industry as Pd-catalysts display an excellent tolerance toward functional groups and hence, it is a favorite catalyst to do complex transformations. The question can be asked whether Pd catalysts are inherently more active and selective, or are the homogeneous Ni catalysts just the new kids on the block? Answering this question requires an examination of the key similarities and differences between Pd and Ni in catalysis. The catalytic properties of Pd and Ni are influenced by their behavior in oxidation/reduction processes. For instance, Ni has lower electronegativity, ionization energy, and oxidation/ reduction potentials than Ni, and zerovalent Ni(0) species are more prone to oxidation and less stable than Pd(0) species. In addition, Pd(II) precursors are often easily reduced to active Pd(0), whereas divalent nickel (Ni(II)) requires additional reductants to generate Ni(0) or Ni(I). Therefore, using Ni(0) complexes, such as Ni(cod)2 (cod = cyclooctandiene), adds extra cost to the process.
The research devoted to homogeneous palladium catalysis is far more extensive, leading to the development of a wide range of ligands under optimal conditions, resulting in great catalytic performance. As mentioned earlier, it is obvious that ligands play a crucial role in stabilizing active species in Ni catalysis, providing specific electronic and steric interactions in the transition state of the catalyzed reaction. This strongly affects reaction rate and selectivity. As a result, there is a greater demand for more research to identifyoff-reaction pathways, identify deactivation mechanisms, understand metal−ligand interactions, develop general principles for ligand design, and formulate trends in Ni catalysis.
In typical 2-electron (Ni0/Ni2+) catalysis, especially with reluctant electrophiles, strong σ-donating ligands such as bulky phosphines and NHCs are applied for both Pd and Ni systems. Ni-based bidentate and tridentate ligands such as bipyridines, bisoxazolines, terpyridine, diimine, and aromatic amine pincer ligands are usually preferred for Ni-catalyzed radical-type reactions. Many nitrogen (N-) ligands are redox-active and lead to the stabilization of the open-shell paramagnetic nickel intermediates. Although N-based ligands in Ni catalysis play an important role in several Ni-catalyzed transformations, this blog will focus on the current insights and trends for monodentate and bidentate phosphine ligands. Phosphines bind strongly to metals and together with the ease of their preparation make them very powerful ligands. This leads to the easy tuning of their steric and electronic properties and has led to wide-ranging catalytic applications.
Monodentate Phosphines as ligands in Nickel catalysis
Monodentate phosphines are common and have a predictable binding with the metal. Although many privileged monodentate phosphines are catalytically relevant, we will focus on Buchwald-type phosphines, which have been applied in several cases in Ni-catalyzed cross-coupling reactions. A recent study published by Abigail G. Doyle and coworkers utilized the parameter “percent buried volume” (%Vbur), which is defined as the percentage of a sphere (r = 3.5 Å) around the metal center that is occupied by a given ligand. Using the (%Vbur) of Buchwald-type phosphines, the authors identified the structure−reactivity relationship in Ni-catalyzed cross-coupling chemistry.
Figure 2. Schematic representation of monodentate phosphines bound to Ni (Left), general structure of the monodentate Buchwald ligand (Right). The A-ring and B-ring are highlighted.
The authors used the Boltzmann-weighted average value of %Vbur (Boltz) to elucidate the structure−reactivity relationships of Buchwald-type ligands in four representative Ni-catalyzed cross-coupling reactions. They showed how Buchwald-type ligands interact with the metal center, both from a structural and catalytic perspective. For this purpose, they investigated four Ni-catalyzed cross-coupling datasets as case studies:
Figure 3. Ni-catalyzed cross-coupling reaction case studies.
Interestingly, the authors investigated the two intermediates in the proposed catalytic cycle of the Ni-catalyzed reactions. Similar to the Pd case, the Ni(0) is stabilized by the η2-Carene interactions formed with the B-ring of Buchwald-type ligands. More specifically, [PR3]2Ni complexes can be prepared in situ from Ni(cod)2 with ligands that have minimal B-ring substitution (i.e., alkyl groups smaller than t-Bu). These complexes can undergo oxidative addition with aryl halide electrophiles. In these 16 e− complexes, one of the two phosphines adopts a high % Vbur conformation with so-called pseudobidentate binding (η2-Carene -type binding), and the other adopts a low % Vbur conformation with monodentate binding, and due to this arrangement, the ligands must be small enough not to overcrowd the complex (Figure 4).
Most Buchwald-type ligands, including those that have B-ring substitution and/or t-Bu groups, readily form the mono-ligated complex [PR3]Ni(substrate) complexes in the presence of π-accepting substrates like aldehydes that can react with Ni(cod)2 on their own. In these cases, the ligand adopts a high %Vbur conformation, with stabilizing η2-Carene interactions observed between the π-basic Ni center and the B-ring, resulting in 16 e− Ni(0). However, ligands with 2,6-substitution and t-Bu alkyl groups are too bulky to form stable complexes.
For divalent [PR3]Ni(II) complexes (PR3 being a Buchwald ligand), stabilizing interactions between the Ni and B-ring arene are weak, thus making the pseudobidentate (η2-Carene) binding insufficient to stabilize coordinatively unsaturated oxidative addition complexes. To stabilize the 14 e− complexes and to facilitate catalytic turnover, a σ-donating substrate must bind to the [PR3]Ni(II) oxidative adduct. If σ-donors are absent or the steric profile of the phosphine prevents binding, decomposition to Ni(I) or cyclonickelation occurs, giving off-cycle Ni species (Figure 4). These insights are helpful for selecting the “right” ligand for your Ni-catalyzed process.
Figure 4. Ni(0) and Ni(II) complexes (Top). Off-cycle catalytically inactive species formed due to decomposition of the oxidate addition adduct (Bottom).
Bidentate Phosphines as ligands in Nickel catalysis
Although monodentate phosphines are useful, there are many situations where they are not an ideal choice. For instance, the dynamic bonding of metals and phosphines can result in off-pathway reactions that lower yields and selectivities. Sometimes, monodentate chiral phosphines can be less effective at asymmetric catalysis because they provide a limited chiral pocket for chiral catalysis to occur with high enantioselectivity. Additionally, side reactions, such as b-hydride elimination, are more likely to occur with monodentate phosphines due to their propensity to dissociate and leave an open coordination site. Bidentate phosphines used with Ni have found several catalytic applications, including, but not limited to, cross-coupling, additions, cycloadditions, C−H functionalization, polymerization, hydrogenation, etc. The active species in such catalytic cycle is the zerovalent Ni(0), which is generated after the formation/reductive elimination step. It is interesting that bidentate phosphines ligands with a larger bite angle (>100°) often lead to destabilization of intermediates and stabilization of products, resulting in increased reactivity.
Figure 5. Schematic representation of bidentate phosphines bound to nickel.
The most common bidentate ligands used in nickel catalysis are DPPE, DCPE, DPPP, DPPB, DPPF, BINAP, and Xantphos. The most successful overall ligands are DCPE, DPPB, BINAP, and DPPF. It is interesting to note that DCPE and DPPF have a high success rate and they form only (bisphosphine)Ni(cod) complexes with Ni(cod)2. DPPB is commonly used in cycloadditions and it has been observed to form (dppb)Ni(cod) in a 99:1 ratio with (dppb)2Ni when reacted with Ni(cod)2. Although BINAP predominantly forms (BINAP)2Ni, the success rate is still high. These observations suggest that the catalytic performance is strongly influenced by the ratio of (bisphosphine)2Ni to (bisphosphine)Ni(cod) formed during the reaction. The effect is ligand-specific and in most examples, higher activity is observed when (bisphosphine)Ni(cod) is predominantly formed. However, a discrepancy in reactivity is observed when comparing cross-coupling and cycloaddition reactions. The ligands DPPE and DPPP have been shown to be effective for many high-yielding cross-coupling reactions, while the same ligands have been reported to have much lower efficacy for many cycloaddition reactions. This suggests that (phosphine)2Ni may be activated by a step in the catalytic cycle of cross-coupling reactions.
Chiral phosphines are an important group of privilege bidentate ligands. A look at the literature, however, reveals that except for Me-Duphos, DIOP, (R)- and (S)-BINAP, powerful chiral phosphines (i.e., Josiphos) have not been studied for their reactivity with Ni(cod)2 or any N(II) complex.
Figure 6. Most used achiral and chiral bidentate phosphine ligands.
Zerovalent Ni(0) species are more prone to oxidation and less stable than Pd(0) under the same conditions. In addition, Pd(II) precatalysts can usually be easily reduced to active Pd(0) species, whereas Ni(II) precatalysts often require the use of special reductants. The general problem of the reluctant activation of Ni(II) often entails using more expensive and unstable Ni(0) complexes, such as Ni(cod)2, that cannot persist outside a freezer or in air, so a more practical Ni(0) source is needed.
In this direction, Ni(cod)(DQ) (DQ = duroquinone) is a thermal and air-stable alternative that can be used as a precatalyst to form Ni(0) phosphines. Ni(stb)3 (stb = stilbene) can also be used as a stable Ni(0) precursor that can, similarly, release its monoalkenes to afford catalytically active species. Although Ni(stb)3 is slightly more reactive, Ni(cod)(DQ) has the advantage of being stable to air and moisture at room temperature. Because they are air-stable, precatalysts can be handled outside an inert-atmosphere glovebox, which means one can rapidly screen ligands, substrates and reaction conditions. The application of these precursors makes Ni catalysis more accessible for cost-effective process development and opens up opportunities for an industrial application.
Catalytic reactions are becoming more common with homogeneous nickel catalysts. The rapid growth does not seem to be slowing down, particularly as the scientific community continues to devote more research and recourses to the use of more abundant materials. As homogeneous nickel catalysis becomes more prevalent, understanding its properties and reactivity will allow catalytic reactions to be more efficient. A key goal in Ni-catalysis is to increase the amount of active metal in the solution, as this will enable the use of lower catalyst loadings. Understanding the structure-activity relationships, combined with detailed knowledge of the formation of off-cycle species is essential. Current developments in the field of nickel catalysis are directed to further map these issues to uncover the true potential of homogeneous nickel catalysis.
By: Samuel H. Newman-Stonebraker, Jason Y. Wang, Philip D. Jeffrey, and Abigail G. Doyle
Article 2: “Nickel and Palladium Catalysis: Stronger Demand than Ever“
By: Victor M. Chernyshev and Valentine P. Ananikov