Introduction:

The development of environmentally more benign and efficient synthetic methods has stimulated the evolution of new strategies and technologies for the synthesis of useful products in a safe, compact, and energy efficient manner. In this respect our preliminary focus to investigate typically efficient and complementary methodology with palladium-catalyzed reductive carbonylation of nitro arenes employing CO gas. CO represents the most important C1 building block molecule to introduce a carbonyl group into parent molecules.

In the last fifty years, the reductive carbonylation of organic nitro compounds has been the subject of intense research due to the fact that industrially important chemicals can be obtained in a single step. Among these, the most relevant are isocyanates, ureas and carbamates but also a number of heterocycles and other non-cyclic compounds.

Isocyanates are commodity chemicals mostly employed in polyurethane synthesis but also intermediates in the production of carbamates and ureas. Polyurethanes are widely applied in almost every part of modern life in the form of plastic foams, coatings, adhesives, sealants and elastomers and binders. The most widely employed aromatic isocyanates are especially toluenediisocyanate (TDI) and 4,4′-methylene diphenyl diisocyanate (MDI) that account for more than seven million metric tons per year.

Synthesis of Isocyanetes:

Currently, industrial synthesis of isocyanates is carried out with two step reaction process using nitroarenes with very high yield and selectivity. Nitroarenes is usually reduced to corresponding aniline using heterogeneous transition metal catalyst, and subsequently the amine is reacted with phosgene to give corresponding isocyanate along with hydrochloric acid.

Phosgene based synthetic route seemed to be effective and well established technology employed.

Major drawbacks of the phosgene based synthesis:

There are essentially four major drawbacks with the synthesis of MDI and TDI via phosgene route. The first and most prominent observation are its extreme toxicity and flammability of phosgene and isocyanates, which make these chemicals extremely difficult to handle in bulk synthesis. Phosgene was used as a chemical weapon during the World War I, and around million people were injured and got killed by the use of poisonous gases.

The second major drawback is the production of corrosive hydrochloric acid, rendering the medium very aggressive with time, thus allowing other side reaction to occur and to result in reactor degradation.

The third limiting factor is the dilution of reaction medium as the high dilution is required to avoid recycling and concentration costs. The final drawback is the isolation of pure isocyanate from reaction mixture. The chloride containing side products are difficult-to-remove from the final product leading to detrimental to the further processing of the isocyanate. Considering above drawbacks, any process design incorporating phosgene will get extra costs to ensure a safe environment.

Industrial requirements for an alternative isocyanate synthesis:

In order to replace the phosgene route, a number of requirements can be made in the ideal scenario. First of all, readily accessible chemicals should be used and second, they should be as harmless as possible. Finally, a one-step synthetic procedure will be the route par excellence. A high overall yield, purity and selectivity, a temperature close to about 25 ^oC and the absence of over and/or under pressures in the plant should be preserved ideally.

In principle, most of the requirements could be met by an efficient catalytic system.The additional requirements such as Turn Over Frequency (TOF) in the order of 10⁴ h^-1 or higher, Turn Over Number (TON) in the order of 10⁶ or above and easy recycling of catalyst would be maintained.

On this account, the necessity of environmentally acceptable but still economically competitive phosgene-free route to isocyanates synthesis is most demanding in near future.

Alternative routes to TDI and MDI:

Various synthetic pathways to isocyanates:

Reductive carbonylation with Palladium metal catalyzed system has proved to be an effective transition metal catalytic system due to its ability to be oxidized or reduced easily during the reaction and high tendency to form complexes with carbamoyl groups.^A palladium strongly prefers the oxidation states 0 and +2, which are separated by a relatively narrow energy gap, making palladium an excellent catalyst for both oxidation and reduction reactions. Secondly, the moderately large van der Waals radius of palladium together with the high number of delectrons (favorable d10 and d8 complexes) means that the organometal is classified as rather “soft”, with a high tendency for concerted reaction as well as a high affinity to “soft” – and -donors, leading to useful chemoselectivity. Finally, Pd is relatively electronegative, resulting in a rather nonpolar Pd-C bond, suppressing the reactivity towards polar functional groups.

In our previous report, nitrobenzene was typically used as a model substrate, and the use of palladium proved to result in the most effective catalytic systems. The carbonylation of nitrobenzene is generally performed in methanol using homogeneous palladium complexes supported by variable bidentate N- or P- donor ligands. Methyl phenyl carbamate (MPC) was synthesized including other side products.

Isocyanates and catalysis:

Within the vast variety of synthetic strategy for isocyanates, one of the approaches is to catalytically convert a nitro or amine compound to the corresponding isocyanate (see Figure 4). One of the approach using oxidative carbonylation and carboalkoxylation of aniline has been studied with various catalytic systems. However, aniline have to be synthesized by hydrogenation of nitrobenzene, thus considering industrial application, the most attractive strategy would be direct synthesis of isocyanates from nitro compounds which is also thermodynamically favorable.

Reactivity of isocyanates and carbamates:

Phosphorus and Nitrogen as donor atoms:

Both phosphorus and nitrogen ligands of the YR₃ (Y = P, N) type (called phosphines and amines respectively) can be described as sp³ hybrids in a (close to) tetrahedral geometry, having a lone pair on the central atom, capable of donating its electron density to an empty (transition) metal d-orbital. Amines are more electronegative than their phosphine analogues, so it could bind strongly with metal centre. However, unlike amines, phosphines can act as a π acid with their σ^* orbitals, so they can be involved in π backbonding (providing the metal has available d-electrons), rendering the overall bond strength larger than would be expected intuitively (see Figure 6). So, the overall bond strength is determined by an interplay of σ donation and π backbonding, the first having an increasing contribution when electropositive/donating substituents are employed, the latter when electronegative/withdrawing substituents are used.

In 1990, E. Drent et. al. reported the palladium catalyze reductive carbonylation of nitroarene introducing “soft-base” ligands like diphosphines in combination with strongly coordinating anions, or “hard-base” electron-donating ligands such as phenanthroline, combined with non- or weakly-coordinating anions, resulted in relatively active and selective catalytic systems (maximal TOF’s (h^-1)/ carbamate selectivities (%) = 150/80 and 1600/98 for diphosphines and phenanthroline respectively).

General remarks on P and N based systems:

In general, the statement made by Drent already in 1990 that:

‘Pd with chelating “hard base” electron donating ligands such as phenanthroline, combined with  non or weakly coordinating anions, can result in relatively active and selective catalytic systems.’

Has been generally accepted and indeed thoroughly studied in the past decades by several people.

On the other hand, the statement that:

‘Pd with chelating “soft-base”’ ligands like diphosphines in combination with strongly  coordinating anions , can result in respectively active and selective catalytic systems.’

Has not yet been the subject of intense academic studies, most likely due to the poor results that were obtained initially, when compared to the N-donor systems. However, in principle, there is no reason why N-donor systems should be superior to P-donor systems, except that phosphine ligands are known to be easily oxidized, thus troubling both the preparation and use of such systems.

On the basis of recent observations on chelating N-donor systems like 1,10-phenanthroline with electron donating substituents (R) in combination with Pd(II) and a weakly or non-coordinating anions (Y), comprises the most active systems to date (i.e. [Pd(R_xphen)₂][Y]₂). Since, 1, 10- phenanthroline (unfettered of substituents ) is relatively cheap and readily available, this ligand ligand is most frequently used. Furthermore, the performed catalyst is more active and selective than its in situ formed analogues and in almost all cases a slight excess of free ligand is added. The addition of a BrOnsted acid (with bulky, non-coordinating anion) as a co catalyst , as well as the addition of a substrate related aniline is known to improve both reactivity and selectivity. The experiment is carried out in inert atmosphere and addition of reactive drying agent is also known to be beneficial. The temperatures are mostly around 120 – 170 ^oC, and both the concentration of all components and the reaction times are differed considerably. Methanol is used as a typical solvent , which is partially consumed to yield the carbamate, but occasionally toluene/methanol or other alcohols are used. Finally, the influence of varying CO pressure results in different reactivity. Most reactions were performed at p_CO = 40 – 80 bar, the best results were obtained at p_CO = 100 bar, and the elevated pressures are believed to enhance the reactivity even further.

The Mechanism

The palladium catalyzed reductive carbonylation of nitrobenzene in methanol has been investigated in great extent. All catalytic reactions were performed using both catalyst precursor and pre-formed complex resulted carbamate as well as side products. There are frequently reported (side-) products of this reaction are shown in Figure. : azobenzene(Azo), azoxybenzene (Azoxy), aniline and N.N’-diphenylurea (DPU). Azo and Azoxy are resulted as coupling product of nitrobenzene. Aniline and DPU are hydrogenation products which are indicating the presence of moisture in the reaction mixture as well as methanol could be source of H-atom for hydrogenation reaction. DPU is the carbamate analogue of isocyanate reacting with aniline which is the better nucleophile than methanol.

The mechanism of reductive carbonylation of nitro compounds into carbamates has been extensively studied with palladium, (substituted) phenanthroline ligands, MeOH as solvent and an acid cocatalyst.

The mechanism proposed by Mooibroek et al. seems to be the most complete as it also explains the formation of all side products Scheme: . The palladium-imido species L₂Pd=NPh (C3) and the palladacycle L₂PdC(O)N(Ph)OC(O) (C2) were considered as possible carbonylation product-releasing species for both (substituted) phenanthroline and diphosphine ligated catalytic system. The result of catalytic experiments , supported by spectroscopic (ESI-MS and NMR) compound C2 is not the major product- releasing intermediate in reactions performed in the absence of acid. In the absence of acid, Pd-imido complex C3 is the proposed reaction intermediate releasing PhN-containing (Azoxy, MPC, PhNH₂) products. On the other hand, in the presence of acid the palladacycle complex C2 becomes the major product-releasing intermediate, resulting the nitrobenzene carbonylation product MPC with high selectivity.

Different side products:

In all catalytic reactions performed using both catalyst precursor and pre-formed complex resulted carbamate as well as side products. The side product are

Conclusion and future prospects:

In conclusion, the incentive research output from the last decade facilitated us to understand the catalytic reductive carbonylation reaction of nitroarenes to produce relevant carbamates. The reactivity of Pd^II compounds supported by 1, 10-phenanthroline (phen) or the bidentate diaryl phosphane has been studied in the reaction of nitrobenzene with CO in methanol. The nitrobenzene reduction chemistry in the Pd/phen/CH₃OH/H⁺ system resulted higher in selectivity but lower in activity wherein Pd/diphospane/CH₃OH/H⁺ system shows higher in activity but lower in selectivity. Based on our continuing interest in reductive carbonylations of nitro-aromatic compounds and considering the importance towards sustainable synthesis of isocyanates, herein our further approach to develop a ultimate catalytic composition of reactants which can produce highly selective product with high yield having tiny or no side-products. However, despite rationalized molecular mechanism developed by Tiddo et. al. for above catalytic system is still not efficient enough, nor is it exactly clear how the catalyst works. In recent years, Great progress has been made in extending the scope of palladium-catalyzed synthetic organic reactions introducing in-situ reductive carbonylation reaction with CO generated from the solvent molecules. Manirul et. al. have reported polymer anchored ruthenium based catalyst for reductive carbonylation. Raquel et. al. have investigated that gold nanoparticles are efficient in N-carbamoylation of aromatic amines. In this respect, further elucidation of the overall mechanism and research in new directions is urgent requirement to produce effective palladium based reductive carbonylation of nitro-aromatic compounds considering industrial background. Thus, the current research project will address several issues which are envisaged to be worthwhile to explore. Firstly, the modification of ligand system, secondly the designing of catalyst system, thirdly the mechanism, and finally the use of other additives in the catalyst system to be fine-tuned.

Modification of ligand system:

The active species in the catalytic cycle as Pd(0) are seem to be as stable as possible to prevent inactive metallic Pd(black) formation. Our previous research work were illustrated the sharp observation of the bidentate P or N ligand based Pd-catalyzed system to achieve maximum activity and selectivity. Our further observations are envisaged based on tripodal P and N system to stabilize Pd(0) species as well as Pd(II) species during catalytic transformations. The concept of monolegated active species stabilizing Pd(0) by steric bulk ligand is rationalized to enhance the reactivity. On the basis of the palladium catalyzed coupling reaction of aryl halides system employed with electron rich N-heterocycle carbene ligand which exhibits flexible steric bulk environment, Several factors will be assumed to be execute in these system: 1) electron-rich nature enhances the rate of oxidative addition, 2) the ligands coordinate tightly to the Pd to prevent the formation of Pd black, and 3) their steric bulk favors a 12 electron, monocarbene–Pd [L-Pd] species and increases the rate of reductive elimination.

Variation in carbamate synthesis:

The formation of carbamates in the catalytic system utilizing Pd^II, phenantroline, acid and various nucleophile has been studied extensively to find an alternative for MeOH as nucleophile. The objective is to yield a carbamate or urea which can be pyrolised at lower temperatures to save energy and prevent degradation of the formed isocyanate. The changing the nucleophile is not straightforward, as the use of another nucleophile as a solvent may be too expensive or not possible (if the nucleophile is a solid), hence use of a solvent may be necessary. Therefore, non –nucleophilic solvents have to be tested in combination with potential nucleophiles in the catalytic synthesis of carbamates and ureas. Despite the use of common known compound used as a nucleophile, still there are broad spectrum of nucleophiles to be tested with various combinations.

Isolation of Palladium-imido complex: