Sustainable Catalysis: Movement Away from Precious Metals in Hydrosilylation

WRITTEN BY JOSEF D. R. DANIELS

ILLUSTRATED BY ANDREA ARMENDI

November 3, 2024 | | 12 min read

Catalysis defines the scope of a broad range of modern synthetic techniques in molecular construction. Although each specific process differs mechanistically, catalysis is productive through its ability to lower the reaction activation energy. As a result, this effectively ‘speeds up’ the reaction and makes transformations possible under mild conditions, sometimes even allowing access to previously impossible functionalizations. 

Figure 1. Reaction coordinate pathway comparison of catalyzed and uncatalyzed reactions.

Functionally, catalysts work by directly interacting with a starting molecule (the substrate) that they then chemically modulate, progressing the reaction. This often involves several electronic changes and rearrangements that change a massive leap into a series of small (and feasible) steps. Significantly, the catalyst itself is not consumed in the process and does not constitute the product, though over the timescale of the reaction the catalyst may degrade.

Every day, catalysts are unknowingly observed: from the catalytic converters that reduce harmful car exhaust to amino acid active sites in DNA polymerase that encode the human genome. As a massively powerful method, applications of catalysis are integral to our everyday lives, ranging from medicine and drug development to polymer synthesis and greenhouse gas activation. Applying this forward to the global scale, especially in industrial settings where synthesis occurs on the scale of multiple kilograms, loading amounts and efficiency are crucial.

Such molecular catalysts can belong to largely any class: an organic molecule (e.g. diethyl ether in monoiodination of hydrosilanes)1, main group salt (e.g. lithium chloride in cyanosilylation of carbonyls)2, organometallic (e.g. organoboron acids in catalytic cycloadditions)3, or coordination complexes (e.g. Wilkinson’s catalyst/RuCl(PPh3)3 in olefin hydrogenation).4 Other types of catalysis such as electrochemical, materials, or enzymatic also have wide applications but will not be focused on here. Notably, transition metals found in the d-block of the periodic table—groups 4 through 12—provide unique chemical characteristics from their d-orbital interactions that lend them as forefront catalysts.

Platinum has long stood as the central metal in transition metal catalysis, along with its group 10 congener—chemically related element—palladium (Pd) and neighbors of rhodium, iridium, ruthenium, and osmium, that make up the vast majority of metal used in historical catalysis for organic transformations.5-7 Catalysts of this nature often possess both immense turnover capacity and activity. Turnover capacity, or how many times the catalyst can interact with substrates to form products before some form of decomposition, is used as a benchmark of catalytic efficiency.9 Likewise, unmatched activity, defined as how readily or fast they convert reactants to products, is a common trend. 

Despite these benefits, precious metal catalysts offer several disadvantages.8, 10 Most notably, these transition metals are only found in very small abundance, lending sustainability concerns from their sourcing and increased expenses in practice. This is compounded by many platinum (Pt) catalysts being unrecoverable, catalyzing side reactions, or forming a degradation byproduct of ‘platinum black’ that can contaminate products.10

To mend this gap, molecular catalysis has motioned a shift towards more abundant transition metals, notably non-noble metal complexes of manganese, cobalt, iron, nickel, and copper. These have become increasingly attractive alternatives, not only for their high Earth abundance but also for low cost, low toxicity, and unique chemical characteristics.10  Comparatively, group 10 metals Ni, Pt, and Pd have relative abundances of 0.0084%, 1.5 x 10-6%, and 5 x 10-7%. Their most common corresponding precursors used in catalyst synthesis cost $1.26/g, $149/g, and $176/g – a stark difference in the cost jump from non-precious to precious metal.11-13, 16

 Figure 2. Abundance of transition metals commonly used in catalysis.

In turn, the differences between periods and related atomic radius in the 3d as opposed to 4d and 5d valency, implore the need for judicious ligand and catalyst design to provide activity, selectivity, and reactivity towards the desired substrates. Indeed, geometry constraints can also be a factor as the 4d and 5d metals generally hold a low-spin electron configuration due to inherent high ligand field splitting energy. Therefore, they may occupy a different preferable geometry that is more electronically favorable. Recalling nickel and platinum in group 10, this becomes especially apparent for four-coordinate compounds: Ni can be found in tetrahedral, octahedral, and square planar organizations whereas Pt particularly favors a square planar geometry that again, minimizes electrons being placed in a high-energy orbital.

Focusing on hydrosilylation, recent developments have offered a large scope of sustainable alternatives that broadly provide the essential groundwork for many commercial syntheses. 

Hydrosilylation typically refers to the addition of a Si-H bond across carbon-carbon multiple bonds to form organosilicon compounds, such as R3C-SiR’3. As Earth’s second most abundant element (28.2%), silicon-derived compounds can be found everywhere in modern society. Not surprisingly, it is central to advancing more sustainable and greener chemistry.5-6 Organosilicon compounds themselves feature low to no toxicity, low environmental hazards, high stability, and offer easy access to a variety of reactions,  making them synthetically incredibly useful. In practice, these are often seen in everyday applications such as adhesives, rubbers, resins, and polymers.5

Most often, the Si-H bond is added across a pi-bond in an alkene, alkyne, or carbonyl. Historically, Speier’s and Karstedt’s catalysts, H2PtCl6 and Pt2[(CH3SiCH=CH2)2O]3 respectively, have presented themselves as the industry standard in olefin and alkyne hydrosilylation. Karstedt’s catalyst notably offers markedly high selectivity and turnover frequency for many unsaturated substrates.14-15 Although highly successful, regioselectivity and competing side reactions can become a concern, especially for Speier’s catalyst, where hydrogenation and redistribution side reactions are pronounced for certain silanes and functionalized substrates.10 Since the induction of these two catalysts in the 1950s and 1970s, many alternative examples have been provided.

Selected Recent Examples:

In 2018, Deng et al. presented a Cobalt N-heterocyclic carbene (NHC)-catalyzed hydrosilylation of aliphatic alkenes with secondary silanes.17 This system significantly provides flexible selectivity preference of either Markovnikov or anti-Markovnikov addition based on steric factors of the ligand scaffold that control the system.5

Table 1. Example Hydrosilylation of Olefins using Co NHC Catalysts.

Figure 3. Structure of Co Catalyst NHC ligands.

Optimal reactivity was observed with monosubstituted terminal aryl alkenes in both systems, with both internal alkenes and disubstituted (α, α) olefins showing no reactivity for the system.17 As both a double-edged sword, this specificity contracts more intent over product formation but is less broad than previously noted Pt catalysts that can hydrosilylate internal and disubstituted alkenes. Although, many other abundant transition metal catalysts have been designed to tackle those substrates.

In the reach of alkynes, hydrosilylation to vinylsilanes—a powerful synthetic intermediate for olefin functionalization—has also seen unique advances. In 2020, Hu et al. demonstrated a revolutionary iron system for this purpose, with tunable selectivity for either the a or b product by changing the position of substituents on the flanking aryl rings of a 2,9-diaryl-1,10-phenanthroline ligand scaffold.18 Here, modulation of the steric interaction upon the iron center plays a significant role in controlling both the activity and regioselectivity of the catalyst and corresponding vinylsilane respectively. This is proposed to be due to the larger ligands decreasing the orbital overlap of the alkyne and iron center, increasing the transition state energy, and providing selectivity towards products.18

Table 2. Example Hydrosilylation of Alkynes using Fe DArPHEN Catalysts.

Figure 4. Structure of Fe Catalysts.

The highest activity and selectivity was found using ligands with a 2,4,6-triisopropylphenyl (Tripp) in either the 2,9- position of phenanthroline or 3,5- position of an additional aryl linker.18 These gave the best “deflection of the alkyne” to provide the corresponding selectivity.18 A EtMgBr Grignard was essential as well to activate the catalyst in situ to an Fe(0) species that then is suggested to follow the classic Chalk-Harrod-type catalytic cycle.18 Here, the alkyne substrate first coordinates to the iron center before the addition of the silane. This then accommodates a shift of the hydrogen to the alkyne before elimination to form the product. Notably, this is the first known iron catalyst to show the capacity for high α selectivity.

Figure 5. Primary Catalytic Steps of Fe(DArPHEN) Hydrosilylation.

As a final example, hydrosilylation can be used as a tandem reaction to achieve reduction of ketones to alcohols through very mild conditions, serving as an alternative to traditional synthetic organic use of reagents such as sodium borohydride or lithium aluminum hydride.19 Here, a recent study by Li et al. in 2022 demonstrated a Copper (I) catalyst with PxNy-type ligands that efficiently hydrosilylated the oxygen in a ketone.20

Table 3. Example Hydrosilylation of Ketones using Cu PxNy-type Catalysts.

Figure 6. Structure of Cu Catalyst PxNy-type ligands.

Both aromatic and alkyl ketones were very successful under these conditions with tolerance for a variety of substituents.20 High steric hindrance on substrates was found to lower reaction yields for ketones such as 3-heptanone, and no significant electronic trend was found for both substitution or position of aromatic substituents.20

Beyond just hydrosilylation, other areas of classic catalysis are seeing such advances similarly investigated. In hydroboration, C-H and C-C activation, olefin metathesis, and cross-coupling reactions, such transitions open up new opportunities for synthetic design!

While the significance of traditional catalysts will likely be preserved, the shift overall represents a constructive capacity towards further sustainability in the field. Using these abundant transition metal catalysts can provide comparable activity without the limitations of cost, waste, toxicity, and efficiency. Much in line with movements toward green chemistry, advancements away from precious metals will not only better provide a higher degree of safety and minimize hazards for human and aquatic life, but also make commercially valuable products more accessible.

At the same time, precious metal catalysts do not need to become obsolete for these discoveries to be a success. Having the synthetic options and availability of a scope of catalysts that can be scaled and feasibly applied forward—without expense or safety limitations—holds immense value to the field.

Acknowledgements:

Special thanks to Sean Dunphy and the Figueroa lab for mentorship on catalysis and hydrosilylation.

References
  1. Keinan, E.; Perez, D. Diiodosilane. 1. A novel reagent for deoxygenation of alcohols and ethers. J. Org. Chem. 1987, 52 (22), 4846-4851. DOI: 10.1021/jo00231a004
  2. Kurono, N.; Yamaguchi, M.; Suzuki, K.; Ohkuma, T. Cyanosilylation of Aldehydes and Ketones.  J. Org. Chem. 2005, 70 (16), 6530-6532. DOI: 10.1021/jo050791t
  3. Graham, B.; Raines, R. Emergent Organoboron Acid Catalysts. J. Org. Chem. 2024, 89 (4), 2069-2089. DOI: 10.1021/acs.joc.2c01695
  4. a. Bennet, M.; Longstaff, P. Complexes of Rhodium (I) with Triphenylphosphine. Chem. Ind. 1965, 846. b. Young, T.; Osborne, J.; Jardine, F.; Wilkinson, G. Homogeneous Hydrogenation and Hydroformylation using Ruthenium Complexes. Nature 1965, 208, 1203–1204. DOI: 10.1038/2081203b0
  5. Almeida, L.; Wang, H.; Junge, K.; Cui, X.; Beller, M. Recent Advances in Catalytic Hydrosilylations: Developments beyond Traditional Platinum Catalysts. Angew. Chem. Int. Ed. 2021, 60, 550-565. DOI: 10.1002/anie.202008729
  6. Naganawa, Y.; Inomata, K.; Sato, K.; Nakajima, Y. Hydrosilylation reactions of functionalized alkenes. Tetrahedron Lett. 2020, 11 (12), 151513. DOI: 10.1016/j.tetlet.2019.151513
  7. Su, B.; Cao, Z.; Shi, Z. Exploration of Earth-Abundant Transition Metals (Fe, Co, and Ni) as Catalysts in Unreactive Chemical Bond Activations. Acc. Chem. Res. 2015, 48 (3), 886-896. DOI: 10.1021/ar500345f
  8. Marciniec, B.; Maciejewski, H. Perspectives on Hydrosilylation Reactions; Springer Press, 2023.
  9. a. Kozuch, S.; Martin, J. “Turning Over” Definitions in Catalytic Cycles. ACS Catak. 2012, 2, 2787-2794. DOI: 10.1021/cs3005264 b. Afruzi, F. Turn over number (TON) and turn over frequency (TOF) studies for heterogenous micro and nanocomposite catalysts. MNT 2022, 223-232. DOI: 10.1016/B978-0-12-824527-9.00004-6
  10. Nakajima, Y.; Shimada, S. Hydrosilylation reactions of olefins: recent advances and perspectives. RCS Adv, 2015, 5, 20603-20616. DOI: 10.1039/C4RA17281G
  11. MilliporeSigma Nickel(II) Chloride. https://www.sigmaaldrich.com/US/en/product/aldrich/339350 (accessed 2024-03-23).
  12. MilliporeSigma Palladium(II) Chloride. https://www.sigmaaldrich.com/US/en/product/aldrich/520659 (accessed 2024-03-23).
  13. MilliporeSigma Platnium(II) Chloride. https://www.sigmaaldrich.com/US/en/product/aldrich/206091 (accessed 2024-03-23).
  14. Speier, J.; Webster, J.; Barnes, G. The Addition of Silicon Hydrides to Olefinic Double Bonds. Part II. J. Am. Chem. Soc. 1957, 4, 974-979. DOI: 10.1021/ja01561a054
  15. Karstedt, B. PLATINUM COMPLEXES OF UNSATURATED SILOXANES AND PLATINUM CONTAINING ORGANOPOLYSILOXANES. US3775452A, 1973.
  16. ABUNDANCE OF ELEMENTS IN THE EARTH’S CRUST AND IN THE SEA, CRC Handbook of Chemistry and Physics, 2016, 97, 14-17. DOI: 10.1201/9781315380476
  17. Gao, Y.; Wang, L.; Deng, L. Distinct Catalytic Performance of Cobalt(I)−N-Heterocyclic Carbene Complexes in Promoting the Reaction of Alkene with Diphenylsilane: Selective 2,1-Hydrosilylation, 1,2-Hydrosilylation,and Hydrogenation of Alkene. ACS Catal. 2018, 8, 9637-9646. DOI: 10.1021/acscatal.8b02513
  18. Hu, M.Y.; He, P.; Qiao, T.Z.; Sun, W.; Li, W.T.; Lian, J.; Li, J.H.; Zhu, S.F. Iron-Catalyzed Regiodivergent Alkyne Hydrosilylation. J. Am. Chem. Soc. 2020, 142, 16894-16902. DOI: 10.1021/jacs.0c09083
  19. Issenhuth, J.T.; Notter, F.P.; Dagorne, S.; Dedieu, A.; Bellemin-Laponnaz, S. Mechanistic Studies on the Copper-Catalyzed Hydrosilylation of Ketones. Eur. J. Inorg. Chem. 2010, 4, 29-541. DOI: 10.1002/ejic.200900961
  20. Li, Z.W.; An, D.L.; Wei, Z.B.; Li, Y.Y.; Gao, J.X. Hydrosilylation of ketones catalyzed by novel four-coordinate copper(I) complexes under mild conditions. Tetrahedron Lett., 2022, 97, 153798. DOI: 10.1016/j.tetlet.2022.153798