Question 1

In-Situ Spectroscopic Studies of Hydroformylation Catalysis with ReactIR

Accepted Answer

Hood, D.M., Johnson, R.A., Vinyard, D.J., Fronczek, F.R. and Stanley, G.G. (2023). Cationic Cobalt(II) Bisphosphine Hydroformylation Catalysis: In Situ Spectroscopic and Reaction Studies. J. Am. Chem. Soc., 145(36), 19715–19726. https://doi.org/10.1021/jacs.3c04866

By observing and tracking key cobalt intermediates in situ, ReactIR measurements shed light on catalyst structure, activity and stability for hydroformylations.

The authors commented that a cationic Co(II) chelating bisphosphine catalyst system, HCo(CO)ₓ(bisphosphine), x = 1–3, is very effective for hydroformylation reactions of internal branched alkenes. The catalyst is both active and stable at temperatures and pressures that are not feasible with other cobalt catalyst systems such as HCo(CO)₄. However, research by other groups postulated that the true catalyst in these cationic Co(II) bisphosphine catalyst systems is actually HCo(CO)₄. The research presented in this current paper uses a combination of spectroscopic studies including in-situ FTIR, NMR, and EPR to show that [HCo(CO)ₓ(bisphosphine)]⁺, x = 1–3, is the primary hydroformylation catalyst system.

The in-situ FTIR studies of the cationic cobalt(II) bisphosphine catalyst system were performed using a ReactIR system equipped with a high pressure silicon ATR probe. In a 101.5 hr experiment with the Co(acac)(DPPBz) catalyst precursor under 30–54 bar 1:1 H₂/CO, the effect of varying temperature was investigated. At room temperature, a Co-CO band at 1937 cm⁻¹ is observed from the cobalt 5-coordinate, 17e⁻ complex [Co(acac)(CO)(DPPBz)]⁺. At 120 °C, the catalyst precursor reacts with H₂ and a number of new carbonyl bands from 2088 to 1974 cm⁻¹ are observed while the 1937 cm⁻¹ band gradually decreases over time. The authors noted that a strong carbonyl band around 1888 cm⁻¹ is observed during the initial formation of the [HCo(CO)ₓ(DPPBz)]⁺, x = 1–3, catalyst mixture. They state that this could arise from the [Co(CO)₄]⁻ anion, which would indicate that the cationic Co(II) catalyst was decomposing to HCo(CO)₄ and [Co(CO)₄]⁻ anion. Temperatures cycling experiments showed that cooling from 120 to 140 °C to room temperature causes the 1888 cm⁻¹ band to reappear and then increasing the temperature back to 120–140 °C causes the 1888 cm⁻¹ band to disappear. A stability study showed that at 120 °C and 53 bar (1:1 H₂/CO) the 1888 cm⁻¹ band completely disappears, and the bands proposed to be due to the [HCo(CO)ₓ(DPPBz)]⁺, x = 1–3, catalyst system remain the same. Furthermore, the IR band intensities of the IR were unchanged, indicating no decomposition to cobalt metal.

At the end of the 101 hr experiment, the system was brought to ambient temperature and pressure and the strong 1888 cm⁻¹ band was again observed along with an intensified 2086 cm⁻¹ band, arising from the 19e⁻ tricarbonyl complex, [HCo(CO)₃(DPPBz)]⁺. This solution was then used in a hydroformylation reaction producing the same results as seen with a fresh Co(acac)(DPPBz) catalyst precursor.

Considering the demonstrated stability of the catalyst system shown by this extended temperature experiment, the authors propose that the 1888 cm⁻¹ band is associated with the formation of a dicationic CO-bridged cobalt(I) dimer [Co₂(μCO)₂(CO)(DPPBz)₂]²⁺ and not [Co(CO)₄]⁻ anion. If the 1888 cm⁻¹ band was a result of the catalyst precursor falling apart to HCo(CO)₄ and [Co(CO)₄]⁻ anion, one would expect to see cobalt metal forming from the further decomposition of HCo(CO)₄ and significant changes in IR band intensities. DFT calculations performed resulted in developing the proposed structure of the dicationic Co(I) dimer, [Co₂(μ-CO)₂(CO)(DPPBz)₂]²⁺.

Furthermore, the IR studies show that as the catalyst precursor reacts with H₂ to form the catalyst, the 1888 cm⁻¹ band due to dimer forms simultaneously. At the higher temperatures at which catalysis is performed, the dimer is unstable and is not likely to be an active catalyst species.

The authors conclude that the extensive EPR, in-situ NMR, in-situ FTIR and reaction studies all support the proposed cationic Co(II) bisphosphine catalyst system: HCo(CO)ₓ(bisphosphine), x = 1–3 and in depth consideration of electronic factors and structure explain the stability and excellent activity for hydroformylation.

Question 2

ReactIR Provides Mechanistic Insight into Copper(I)-Catalyzed Aerobic Alcohol Oxidation

Accepted Answer

Lagerspets, E., Valbonetti, E., Eronen, A. & Repo, T. (2021). A New Catalytic Approach for Aerobic Oxidation of Primary Alcohols Based on a Copper(I)-Thiophene Carbaldimines. Molecular Catalysis, 509, 111637. https://doi.org/10.1016/j.mcat.2021.111637

ReactIR provides insight into mechanistic differences in oxidation of alcohols and diols by novel copper(I) catalyst.

Oxidation of alcohols to their respective aldehydes is typically performed with inorganic oxidants, however this method may require harsh conditions and can generate substantial hazardous waste and by-products. Thus, there is significant effort in developing chemistry that is more benign, less costly and better selectivity. The use of copper-based catalysts with N-donor ligands such as bipyridine or phenanthroline along with the stable radical 2,2,6,6-tetramethylpiperdine-N-oxyl (TEMPO) and N-methylimidazole (NMI) as an auxiliary base is an active area of research interest in achieving oxidations in a more sustainable manner. The ideal would be to develop a catalytic system for alcohol oxidation that optimizes catalytic activity and chemoselectivity, while using air as the oxidant at ambient temperature. The authors have developed novel new copper(I) catalysts using thiophene with carbaldimine-type ligands that meet those goals.

Based on observations from earlier work, they synthesized and tested a series of six thiophene carbaldimines as supporting ligands in copper(I) based catalysts on an oxidation reaction of benzyl alcohol. One copper catalyst system, using the ligand 1-(thiophen-2-yl)-N-(4-(trifluoromethoxy)phenyl)methanimine gave the highest activity for oxidation to benzyl aldehyde, resulting from the increased electronegativity of the trifluoromethoxy group para-substituted on the phenyl. Under optimized conditions, the reactivity of the catalyst was tested on a series eight primary alcohols with π-activated substrates all which were converted to their corresponding aldehydes in exceptionally high yields.

To extend the scope of this reaction for different substrates, the oxidation of a series of diols was undertaken with this new copper catalyst system. In other copper catalyst and enzymatic catalyst systems, the formation of lactones from diols were shown to proceed through a lactol intermediate, giving very good yields. The new copper(I)-thiophene carbaldimines system displayed a different selectivity, producing lactols as the major product, rather than lactones. In the case of the diols, ReactIR measurements showed that the lactol-forming cyclisation reaction occurred rapidly and no IR band for aldehyde was observed, in stark contrast to the reaction for benzyl alcohol, wherein a prominent spectral band at 1704 cm⁻¹, arising from formation and accumulation of aldehyde, was immediately observed. Based on the experimental evidence, the authors commented that their new copper system appears to be selective for oxidation of primary alcohols to aldehydes. When the lactol forms in the cyclisation reaction, a hydroxyl groups transforms into a secondary alcohol, which does not oxidize. Interestingly, and in contrast, they discovered that for 1,4-butandiol, very high yields of the thermodynamically more stable lactone are formed.

In summary, the Copper(I)-Thiophene Carbaldimine catalyst is shown to be efficient and selective for aerobic oxidation of aliphatic, allylic and benzylic primary alcohols to their corresponding aldehydes. The catalyst is also effective for selective oxidation of a number of diols to their respective lactols.