Applications of transition metal catalyzed P-radical for synthesis of organophosphorus compounds

Introduction

Organophosphorus compounds are widely present in nature, and due to their unique chemical properties, play a fundamental role in many important fields. Among the more prominent features that demonstrate their status as unique and versatile compounds, such as multivalency, asymmetry, variable oxidation states, and metal-binding properties. Their presence in bioactive natural products, small molecule therapeutic agents, endogenous biomolecules and pro-drugs substantiates their role in medicinal chemistry and biochemistry [1], [2], [3]. Considering these diverse applications, the development of simple and efficient protocols for the synthesis of organophosphorus compounds has become an intensive topic of discussion in organic synthetic chemistry.

Over the past decades, there are numerous general strategies that have been developed for the synthesis of organophosphorus compounds, the most important of which focused on the C–P bond formation strategy [4]. Meanwhile, radical additions of phosphorus-centered radicals to unsaturated compounds have been recognized as a powerful method for rapidly constructing organophosphorus compounds and attracted much attention [5], [6], [7]. Generally, the green-stable P(O)–H compound can generate phosphorus-radical under the action of various free radical initiators. Initially, highly toxic triethylborane is used as the P-radical initiator, but relatively dangerous [8]. Since then, chemists have also developed quantities of trivalent manganese salts and peroxo compounds as phosphorus free radical initiators [9], [10]. These methods are limited, however, by narrower substrates and applications as well as environmentally unfriendly. As copper and silver salts have the advantages of mild chemical properties, low cost and low toxicity. Therefore, the use of copper and silver salts as P-radical initiators has become one of the most popular research directions [11], [12], [13]. Recently, the groups of Zhao et al. as well as others have made hard work and exciting contributions in organophosphorus chemistry [14], [15], [16], [17], [18], [19], [20], [21], [22], [23]. Over the past several years, we are devoted to developing simple and efficient protocols for the synthesis of new-style organophosphorus compounds. Herein, in this account, we summarize our efforts in this field through transition metal catalyzed P-radical promoted difunctionalization for synthesis of some important phosphorylated heterocyclic, alkenyl and allyl compounds, such as oxindoles, indolines, phenanthridines, alkenylphosphine, allylphosphine oxide compounds and imidoylphosphonate as well as others (Scheme 1).

Scheme 1:

Transition metal catalyzed P-radical promoted functionalization.

Within the field of organic chemistry, indoles [24] are commonly occurring structural motifs in natural products and it is very important to realize the synthesis of phosphonoindoles by efficient and simple route. In 2013, under our efforts, we reported the first difunctionalization reaction between phosphorus-centered radicals and N-arylacrylamide to afford the corresponding phosphorylated oxindoles (Scheme 2) [11]. Under preliminary mechanistic studies including radical trapping and mass spectrometry experiments as well as other several control experiments, we discovered silver salts can work with Ph₂P(O)H to form the corresponding active [Ph₂P(O)Ag] intermediate 2A. Then, intermediate 2A occurred homolysis to produce the P-radical which was added to alkene to form the alkyl radical intermediate 2B. Finally, the resulting alkyl radical 2B participated in an intramolecular radical cyclization reaction to afford the desired products via C–H functionalization. To further explore the scope with respect to the substrates, various N-arylacrylamides with functional groups such as benzyl, trifluoromethyl, phthalimide, and others were tested. Fortunately, a wide range of substrates could react through inexpensive and nontoxic silver catalyzed difunctionalization. Notably, similar catalytic systems have also made outstanding contributions in the synthesis of other organophosphine compounds, representative work come from Zhu et al. as well as others [25].

Scheme 2:

Silver-catalyzed C–P difunctionalization of N-arylacrylamide.

After the initial discovery of Ag-catalyzed C–P difunctionalization of N-arylacrylamide through a carbon phosphonation and C–H functionalization cascade process, we needed to determine whether this Ag-catalyzed C–P difunctionalization could be extended to other types of reactions. Aiming to construct synthetically useful structural motifs, we wished to expand this methodology to C–P difunctionalization of 1,6-dienes which are widespread in nature. After systematic studies, we found that the Ag₂CO₃/Mg(NO₃)₂·6H₂O catalytic system was the best choice in DCE at 40°C under an Ar atmosphere for 6 h. With the optimized reaction conditions, a series of substrated 1,6-dienes and different phosphate sources were examined and afforded the corresponding products in good yields (Scheme 3) [26]. With a unique electronic structure of isonitrile, it is also very susceptible to attack by P-radicals, which was reported by several groups [27], [28]. We have also developed a simple and efficient method to synthesize a series of 6-phosphorylated phenanthridines by employing AgNO₃ as an oxidant for C–P difunctionalization of isonitrile (Scheme 4) [29].

Scheme 3:

Silver-catalyzed C–P difunctionalization of 1,6-dienes.

Scheme 4:

Silver-promoted C–P difunctionalization of isonitrile.

In light of the above results of Ag-catalyzed C–P difunctionalization of alkene and isonitrile, we hypothesized that copper salts could also catalyze generation of P-radicals under an appropriate oxidant to react with olefins. Coincidentally, copper-catalyzed phosphonation reaction has become one of the most popular research directions, and significantly advanced the field from Ji et al. groups [13], [25], [30], [31]. With this design principle in mind, we chose the N-Ts-2-allylaniline and HP(O)Ph₂ as the model substrates, after conditional optimization, we were able to get the desired product in 74% yield under Cu(ClO₄)₂·6H₂O (20 mol%) as the catalyst, K₂S₂O₈ (3.0 equiv.) as the oxidant in CH₃CN at 35°C for 6 h [32]. With the optimized conditions, various P-radical precursors, N-protected group, and alkenes were tested and afforded the corresponding products in good yields (Scheme 5). Preliminary mechanism studies have shown that the oxidized P-radical may initiate the reaction, then undergo oxidative addition to form Cu(III), and cyclization to afford the desired product finally.

Scheme 5:

Copper-catalyzed N–P difunctionalization of N-protected 2-allylarylamine.

Since the alkenylphosphine and allylphosphine oxide compounds are central to many broadly used synthetic modifications in organic synthesis, pharmaceutical chemicals, agrochemicals, as well as in materials science [33], we next turned our attention to the P-radical promoted C–H phosphonation reaction of olefins. It is well known that HP(O)R¹R² is a good nucleophile, and it is easy to react with the unsaturated bonds to effect hydrophosphination frequently [34]. Thus, can we use our developed P-radical promoted C–H phosphonation reaction to suppress hydrophosphination and achieve high regioselectivity alkenylphosphine and allylphosphine products with cheap and readily available alkene? Initially, 1,1-diphenylethylene and HP(O)Ph₂ were chose as modle substrates [35]. After a series of screening, the alkenylphosphine oxides could be obtained through CuCl₂·H₂O catalyzed and DCP oxidated C–H phosphonation reaction. As shown (Scheme 6) [35], a wide range of olefin and phosphorus sources were surveyed, and the corresponding products could be obtained in moderate to good yields. However, the scope was limited to only 1,1-disubstituted arylenes; styrene was unreactive. Based on our former work, allyl C(sp³)–H bonds phosphonation were explored via a C–H phosphonation reaction [36]. At present, allyl C–H bond activation construct C–C bond has been studied extensively; some examples also involved in C–N, C–O, C–F and C–Si bond formations [37], [38]. But the example disclosed the C–P bond formation has not reported. As we previously reported, the mechanism was similar, copper salts were still the best catalysts. And silver salt was used as an oxidant in order to form a phosphonium silver intermediate (Ag-P(O)R¹R²) and in turn to produce P-radicals. Both aliphatic and aromatic olefins with useful functional groups were compatible in this transformation. The reactions with the biologically active estrone derivative and natural product D(+)-Carvone proved its application in natural products and drugs (Scheme 7).

Scheme 6:

Cu-catalyzed oxidative C–H phosphonation reactions of olefins.

Scheme 7:

Copper-catalyzed allylic C–H phosphonation.

In addition, for several years, oxidative cross-coupling and catalytic dehydrogenative cross-coupling reactions have attracted great attention due to their excellent atom economy and an environmentally friendly approach [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50]. Herein, we first examined the CDP reaction with α-amino ketones [51]. Unlike hexahydropyridine and pyrrolidine derivatives [52], α-amino ketones possess a more universal appeal given their extensive applications. What’s more, the expected products imidoylphosphonate play a pivotal role in organic synthesis and metal-ligand design [53]. When Cu(OAc)·H₂O was used as catalyst and TBHP as oxidant, we obtained the optimum reaction conditions. As diverse α-amino ketones can be employed as the substrates, this reaction allows access to imidoylphosphonate with different electronic and steric properties (Scheme 8).

Scheme 8:

Cu(II)-catalyzed oxidative CDP reactions of α-amino ketones.

Conclusions

In summary, from the above-mentioned advances in the application of transition metal catalyzed P-radical promoted functionalization for the construction of organophosphorus compounds. Some an important phosphorylated heterocyclic, alkenyl and allyl compounds, such as oxindoles, indolines, phenanthridines, alkenylphosphine allylphosphine oxide compounds and imidoylphosphonate as well as others have been obtained. This strategies not only make great progress on the reactivity, chemoselectivity and regioselectivity, but also have more extensive substrate applicability and step economy than traditional methods. With these new strategies, a variety of challenging phosphine-containing organic molecules has been obtained with different electronic properties and spatial effect under different reaction conditions.