Coordination Chemistry Reviews 272 (2014) 145–165

Contents lists available at ScienceDirect

Coordination Chemistry Reviews journal homepage: www.elsevier.com/locate/ccr

Review

Catalysis by 1,2,3-triazole- and related transition-metal complexes Deshun Huang a , Pengxiang Zhao a,∗ , Didier Astruc b,∗∗ a b

Science and Technology on Surface Physics and Chemistry Laboratory, P.O. Box 718-35, Mianyang 621907, Sichuan, China ISM, Univ. Bordeaux, 351 Cours de la Libération, 33405 Talence Cedex, France

Contents 1. 2. 3.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Coordination modes of triazole and triazolyl ligands with transition metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mn-triazole complexes for catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Mn-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fe-triazole complexes in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. 4.1. Fe-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Ni-triazole complexes in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Ni-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Cu-triazole complexes in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Cu-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Cu-carbene complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Ru-triazole complexes in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1. Ru-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2. Ru-carbenes complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. Rh/Ir-triazole complexes in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1. Rh/Ir-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2. Rh-nitrogen complexes with anionic triazoles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9. Pd-triazole complexes in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1. Pd-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2. Pd-carbenes complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3. Other types of molecular Pd-triazole catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4. Pd-nanoparticle catalysts stabilized by 1,2,3-triazole-containing macromolecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10. Au-triazole complexes in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1. Au-nitrogen coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2. Au-carbenes complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3. Au-nitrogen complexes with anionic triazoles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

a r t i c l e

i n f o

Article history: Received 10 February 2014 Accepted 7 April 2014 Available online 18 April 2014 Keywords: Catalysis Transition metal Click chemistry Triazole ligand Carbene

a b s t r a c t A short overview of the multiple coordination modes of 1,2,3-triazole- and related transition-metal complexes are provided, then the implication of and catalysis with transition-metal-1,2,3-triazole complexes are detailed with Mn, Fe, Ni, Cu, Ru, Rh, Ir, Pd, and Au catalysts including various ligand coordination modes and mechanistic features. © 2014 Elsevier B.V. All rights reserved.

∗ Corresponding author. Tel.: +86 8163369780. ∗∗ Corresponding author. E-mail addresses: [email protected] (P. Zhao), [email protected] (D. Astruc). http://dx.doi.org/10.1016/j.ccr.2014.04.006 0010-8545/© 2014 Elsevier B.V. All rights reserved.

146 146 147 147 148 148 149 149 149 149 150 150 150 151 153 153 155 155 156 158 160 160 161 162 162 163 163 163 164

146

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

1. Introduction The 1,2,3-triazole heterocycle, known since the end of the 19th century, is now a common heterocyclic ligand in chemistry and biology [1]. Relatively few studies have been reported before the year 2000 due to the limited availability of functional triazole derivatives when the non-selective Huisgens reaction was used for their synthesis [1b,1c]. The breakthrough in triazole chemistry came in the early 2000s with a novel concept, that of “click” chemistry, that was first fully presented by Sharpless’ group [2,3]. The “click” reactions described chemistry tailored to quickly and reliably generate substances by linking small units together under “green” conditions. This has proved to be a powerful concept allowing molecule fragments to assemble. Indeed, the most popular reaction representing the “click” chemistry concept is the Cucatalyzed alkyne–azide (CuAAC) reaction with the regioselective formation of 1,4-disubstituted 1,2,3-triazoles [4]. Besides, the Rucatalyzed alkyne–azide (RuAAC) reaction was later disclosed to also regioselectively form 1,2,3-triazoles, but at this time with 1,5disubstitution [5]. Thanks to these modular, facile and high-yield methods for the generation of a large number of 1,2,3-triazoles and their derivatives, 1,2,3-triazole heterocyclic chemistry now appears as a new area with potential applications of 1,2,3-triazolemetal complexes in optics, redox sensing, biomedicine and catalysis [6]. Transition-metal triazole and triazolyl complexes have recently present catalytic activity for a number of organic reactions, and the purpose of this review is to survey these properties and catalytic reactions. 2. Coordination modes of triazole and triazolyl ligands with transition metals 1,2,3-Triazoles bearing several donor sites are potentially versatile ligands for metal coordination [7]. Generally, there are mainly three modes with which triazole ligands combine with transition metals (Figs. 1, 3 and 4). The first mode is through nitrogen coordination of neutral simple triazoles and chelating triazoles (Fig. 1). DFT calculations have shown that N3 is a better donor compared to N2 [8]. The triazole ligand coordinates to a metal through the N3 nitrogen atom either as a monodentate ligand (type A) or as part of a bi- or poly-dentate chelator (type B), when there are other

Fig. 1. Simple triazoles and chelating triazoles coordinate to transition metals.

Fig. 3. Deprotonated triazolium ligands (NHCs) transition metal complexes.

Fig. 4. Deprotonated NH 4,5-disubstituted triazolates as anionic ligands in transition metal complexes.

donor sites nearby. When the additional donor site is adjacent to N1, coordination through N2 is possible to form a bi- or polydentate chelator (type C) [9]. Thus, for the metal chelators, five- or six-membered cycles are usually formed. Besides, bridging coordination modes with two metals coordinating to two of the nitrogen atoms are possible (types D and E). The second mode is C5 coordination with deprotonated triazoliums to form N-heterocyclic carbenes (NHCs, Fig. 3). NHCs are a class of well-known, very useful ligands resulting from the deprotonation of imidazolium salts, but members of the family are also obtained by deprotonation of triazolium salts. NHCs are stronger neutral electron donors (␴ donors), have a better oxidation stability and undergo easier modification than tertiary phosphines. Therefore, they have been widely used as ligands with success in transition metal catalysis [10]. Imidazolium salts are the most frequently used carbene precursors with metal bounded at the C2 position. Subsequently, imidazole-based carbenes with the metal bonded at the C4(5) position were also first reported by Crabtree and co-workers (Fig. 2) [11]. These carbenes are called “abnormal” N-heterocyclic carbenes (aNHCs), and they are even stronger ␴ donors than C2-bound “normal” N-heterocyclic carbenes (nNHCs) [12]. The difference between these two classes of carbenes is that free nNHCs have a resonance form with all-neutral formal charges, while the free aNHC are mesoionic (Fig. 2). In 2008, Albrecht and co-workers used 1,3,4-substituted 1,2,3-triazolium salts as precursors for the synthesis of new aNHCs with various transition metals [13]. These abnormal triazolylidene complexes (type F) are expected to have a great potential for the development of new catalysts with unprecedented reactivities. Recently, an example

Fig. 2. Imidazole-based nNHCs and aNHCs.

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

147

Scheme 1. Syntheses of abnormal triazolylidene complexes (the most common R3 X used so far in the literature is MeI).

of normal 1,2,3-triazolylidene carbene with a 1,2,4-substitution pattern was also reported (type G) [14]. Interestingly, owing to its unprecedented substitution pattern, the normal triazolylidenes exhibit even higher donor strength than the abnormal 1,3,4substituted triazolylidenes [14a]. Similarly, metal centers can also chelate to both the carbene carbons and the other adjacent donor sites to form more complicate complexes. Additionally, the relatively acidic C H bond on the 5-position can be directly inserted by transition metal to form a carbon–metal bond (see Section 9.3). The third coordination mode results from the combination of deprotonated NH 4,5-di-substituted triazolates as anionic ligands in metal complexes (Fig. 4). The acidic N H protons can be used to generate anionic ligands. Under basic conditions, benzotriazoles bind to metal via N1 (type H) while 4,5-disubstituted-NH-1,2,3triazoles bind to metal via N2 (type I). Moreover, additional metal centers can coordinate to the free neutral nitrogen atoms of the metal triazolates forming bridged complexes (types J, K, and L) [3]. The nitrogen-coordinated triazole complexes are easily formed by just combining the triazole with metal complex precursors, resulting in ligand substitution. Thus the C5-coordinated abnormal triazolylidene complexes have been generated by using 1,2,3triazolium precursors as follows (Scheme 1) [13]. The neutral triazole heterocycles 1 are readily available by “click” reactions of the corresponding alkynes and azides. Selective nitrogen alkylation of the resulting triazoles 1 at the N3-position yields triazolium salts 2 that are precursors of abnormal triazolylidenes. Thus palladium (3) and silver (4) abnormal carbene complexes are obtained by

metallation of the triazolium salts 2 via C H bond activation using Pd(OAc)2 or Ag2 O. Other transition metal carbene complexes 5 are also obtained by transmetallation of the resulting silver carbene complexes. The generation of normal 1,2,4-substituted 1,2,3-triazolylidenes complexes is described in Scheme 2 [8]. Different from the synthesis of 1,3,4-substituted 1,2,3-triazoles using “click” chemistry, the precursors 1,2,4-substituted 1,2,3-triazolium chlorides 6 are synthesized via a ring closing procedure based on hydrazonoyl chlorides and isocyanides. Then the metal carbene complexes 9 are generated by either reacting the transition metals with the ammonia adducts 7 [14a] or transmetallation of the corresponding silver carbene complexes 8 [14b]. 3. Mn-triazole complexes for catalysis 3.1. Mn-nitrogen coordination Following the order of elements in periodic table, Mn is the first mentioned transition metal for which transition metal catalysts have been disclosed. For instance, N4 tetradentate ligands have been recently widely used as metal ligands for many important and challenging catalytic reactions. However, most of them are based on a pyridyl framework, which limited the possibilities of their modification. Recently, an efficient construction method using “click” chemistry has been reported, forming a series of triazole-based N4 tetradenate ligands (Scheme 3) [15]. Their Mn(II) complexes

Scheme 2. Syntheses of normal triazolylidene complexes. (The difference between “normal” 9 and “abnormal” 5 is that they have different carbene precursors (Fig. 3). The precursor of “normal” 9 is neutral, while the precursor of “abnormal” 5 is mesoionic.)

148

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

Scheme 5. Mn(II) complex-catalyzed epoxidation of cyclooctene.

decade, iron complexes have been examined in catalytic procedures that originally required expansive precious transition metals such as palladium, ruthenium or rhodium. Obvious reasons are the abundance of iron on earth, possibly lower toxicity than that of other transition metals, and low cost [16b,16c]. Viewing the rapid development of iron catalysts in the last decade, various efficient iron-catalyzed processes have now become recognized. Scheme 3. Synthesis of triazole-based N4 tetradenate ligands and their Mn(II) complexes.

Scheme 4. Mn(II) complex-catalyzed epoxidation of terminal olefins.

were used in the epoxidation of various terminal aliphatic olefins with peracetic acid as oxidant, and showed good catalytic activities with low catalyst loading and short reaction time (Scheme 4). As an example of epoxidation of internal olefin, the Mn(II) catalyst also showed good catalytic performance (Scheme 5) [15]. 4. Fe-triazole complexes in catalysis Iron complexes have long been known and widely used in coordination chemistry and organometallic chemistry [16]. In the last

4.1. Fe-nitrogen coordination A remarkable property of iron catalysts is their dual reactivity, where Fen+ serves either as a Lewis acid or as a redox center through single-electron-transfer processes [17]. With their combination of ␴-donor (nitrogen lone-pair electrons) and ␲-receptor properties, the N-heterocycles are considered to be very useful ligands for Fen+ catalysis as witnessed in the recent literature [18]. However, up to now, reports concerning the 1,2,3-triazole-Fe catalysts are very limited, and little is known concerning the comparison of these ligands with other N-heterocyclic ligands. Shi’s research has addressed this aspect, however [19]. As shown in Fig. 5, among other heteroaromatic ligands including pyridine, imidazole, tetrazole, and differently substituted triazoles with the same binding patterns, only the 1,2,3-triazole ligands provided good yields of enyne in the propargyl alcohol dehydration reaction. This property has been attributed to the required electronic effect of 1,2,3-triazole in adjusting the Lewis acidity of Fe3+ in order to achieve the needed chemoselectivity for this transformation [19]. This finding shows an example of the

Fig. 5. Conversions and yields obtained with iron complexes of various ligands for propargyl alcohol dehydration.

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

149

6. Cu-triazole complexes in catalysis

Scheme 6. Intermolecular nickel-catalyzed coupling of ␲ components in the presence of a main-group organometallic complex M R.

advantage of 1,2,3-triazoles over other N-heterocycles and should attract attention for catalysts using 1,2,3-triazole-Fe complexes in future research. 5. Ni-triazole complexes in catalysis Nickel-catalyzed coupling and cyclization reactions, including oligomerizations, olefin polymerizations, dimerizations, and hydrometallations have been applied for more than 50 years [20]. The key role of nickel in these reaction is reductive coupling of two ␲ components with a main-group organometallic reagent or metal hydride (Scheme 6) [20b].

Copper(I) complexes are frequently used catalysts in organic synthesis. Cu(I) triazole complexes have essentially been used with triazoles as nitrogen donors for the popular Cu-catalyzed azide–alkyne cycloaddition (CuAAC) reaction, the most current socalled “click” reaction. The CuAAC was first introduced using Cu(II) sulfate with sodium ascorbate as a reducing agent in situ to form air-sensitive Cu(I) species, then with a large variety of copper salts, in particular with nitrogen ligands and carbenes that activate the Cu(I) catalyst toward the CuAAC reaction [2,3]. The search for highly active Cu(I) catalysts was indeed driven by the need to avoid product contamination that is unacceptable in electronics, or can lead to toxicity in biomedical applications. To overcome this problem, various copper catalysts have been applied to the CuAAC reactions, including nano-sized solid catalysts, dendritic copper catalysts and Cu(I) complexes of organic ligands [24–29]. Thus, as a member of both nitrogen donating ligands and ␲-acceptor carbenes, Cu(I) triazole complexes are very active toward CuAAC reactions [30–34] and other organic reactions [35] in recent progress.

5.1. Ni-nitrogen coordination 6.1. Cu-nitrogen coordination Among all the above nickel-catalyzed reactions, olefin polymerization is the most important one due to its broad applications in both academic and industrial areas. The use of catalysts based on nickel complexes (and other late transition metals) containing Schiff-base ligands has been a major advance in the development of this reaction because of the unique N ␴-donor properties of these ligands [21]. Therefore the related 1,2,3-triazole complexes [22,23] have also been investigated in this context during the last few years, and they have exhibited a comparable or even better catalytic activity than those of Schiff-base complexes. Fig. 6 shows the triazolyl-functionalized Schiff base bis(imino)acenaphthene (BIAN) as a ligand. When this ligand was coordinated to Ni(II), and using the classic co-catalyst methylaluminoxane (MAO), it showed better activity for ethylene, norbornene and styrene polymerization than the BIAN-Ni(II)/MAO system. In this case, the triazolyl groups were considered as labile moieties of chelating ligands, which were substituted by the norbornene and styrene monomers, thus generating 16-electron active species that accelerated the polymerization. Ni(II)-triazole catalysis has only been reported after 2010, and thus this finding may now encourage the development of Ni(II) catalysts.

Very efficient N-coordinated triazole-Cu(I) catalysts for CuAAC reactions were reported in 2004. In their seminal study, the authors used tris(benzyltriazolylmethyl)amine (TBTA) as a ligand to stabilize Cu(I) for the “click” reaction between phenylacetylene and benzyl azide (Scheme 7). The tetradentate nature of TBTA allows completely encapsulation of the Cu(I) center, with the central basic tertiary amine supposed to be permanently coordinated during catalysis, while the pendant triazole groups temporarily dissociate from the metal center to allow formation of a Cu(I)-acetylide intermediate that is crucial in the catalytic cycle. The tetradentate TBTA ligand protects the Cu(I) center from oxidation and disproportionation, while enhancing its catalytic activity [30b]. Therefore it has been considered as much more efficient than mono- or bis-triazolyl ligands. Later, Williams et al. reported that Cu(II) also coordinated TBTA, and in the presence of the proper reducing agent, for example sodium ascorbate, the Cu(II) complex converted to an active Cu(I) catalyst. In addition, with the classic copper precursor [Cu(CH3 CN)4 ][BF4 ], the binuclear dicationic Cu(I) complex represented in Fig. 7 formed and is also a very efficient catalyst for the CuAAC reaction [32]. In addition, covalently immobilized catalysts for CuAAC reactions have been frequently used in recent years [36], but in these cases the reactions involve either over-stoichiometric amounts of catalytic resin or high copper loadings [37]. Therefore, due to their unique nature, the tris(triazolyl) ligands with apical functional groups were considered as appropriate ligands for CuAAC reactions

R1

+

R2 N3

Cu(I) (0.25-1 mol%) TBTA (1 mol%) t

R1 N

N R2 N >99%

BuOH:H2O, rt

Bn N N N TBTA =

N N N

NBn

N BnN N Fig. 6. The structure of BIAN and triazolyl-functionalized BIAN, useful ligands in Ni(II) polymerization catalysis.

Scheme 7. The TBTA-Cu(I)-catalyzed CuAAC reaction (R1 = Ph, R2 = CH2 Ph).

150

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

Scheme 8. Synthesis of a copper(I) catalyst on polymer-supported tris(triazolyl)methanol ligand [33b].

and further investigated for the design of polymer-immobilized copper catalysts. As shown in Scheme 8, the tris(triazolyl) methanol ligand was firstly linked to the Merrifield resin by alkylation, then CuCl was added to obtain the immobilized catalyst. This catalyst exhibited a very efficient “click” activity toward a series of substrates in aqueous or methanol/water system, and tolerated variations of reaction parameters. In addition it showed an excellent recyclability and was easy to separate from the product by filtration [33a]. Later on, Astruc’s group improved this strategy by anchoring this catalyst onto SiO2 -coated ␥-Fe2 O3 nanoparticles as supports rather than on polymers. Due to their insoluble, paramagnetic and nanosized nature, they display better stability and reusability, more efficient catalytic activity, lower preparation costs, and lower toxicities in comparison with other materials that are used as supported heterogeneous catalysts. Moreover, this catalyst can be successfully extended to various organic azides and alkynes, and this procedure is easy to operate, economical, and environmentally friendly if the azides are not isolated but just intermediates before “click” reactions with alkynes in situ [33b]. 6.2. Cu-carbene complexes Although the carbene-metal bond is very popular in transition metal catalysis, Cu-carbene complexes have rarely been reported in the last decade for the CuAAC reaction. This might be due to the formation of the Cu-carbene bond between the product and the copper catalyst during the CuAAC reaction, competing with the Cu-carbene bond of the catalyst itself. However, there is one example showing the possibility of a Cu-carbene catalyst from triazole for the CuAAC reaction in which the authors used copper with 1,4-diphenyl, 1,4-dimesityl, and 1-(2,6-diisopropylphenyl)4-(3,5-xylyl)-1,2,3-triazol-5-ylidene (aNHC) to prepare the new CuCl(aNHC) catalyst (Fig. 8). This catalyst efficiently catalyzed

Fig. 8. Triazole-based Cu-carbene catalysts.

CuAAC reactions of azides with alkynes to give 1,4-substituted 1,2,3-triazoles in excellent yields at room temperature with short reaction times [34]. Besides, in very recent results from Sarkar’s group, the copper(I) complexes containing two triazolylidene ligands (see Fig. 9) was proven to be capable of catalyzing the click reaction between bulky azides (a class of substrates considered as difficult to transform using the original click recipe) and electronically diverse alkynes under mild conditions [38]. 7. Ru-triazole complexes in catalysis Ruthenium (Ru) is an important transition metal under various oxidation states from 0 to 6 with a variety of ligands for catalysis of a large number of homogeneous reactions [39]. However, among all the oxidation states, 1,2,3-triazole Ru catalysts have essentially been synthesized in the low oxidation state Ru(II). The Ru(II) complexes show high efficiency in many reactions such as hydrogenation, reduction via hydrogen transfer, alkyne transformations via ␩2 -alkyne, ␩1 -vinylidene or allenylidene intermediates and oxidative coupling processes, cyclopropanation, and olefin metathesis [39b]. In spite of the application of Ru(II), its limited stability and the low activity toward the substrates are still sometimes drawbacks. Therefore, the proper ligands for stabilizing and activating Ru(II) are called for [39e], and 1,2,3-triazoles allowed to improve catalytic activity as NHC ligands [40–48]. 7.1. Ru-nitrogen coordination Ru complexes with N-coordinated triazole ligands are active catalysts for alcohols oxidation. The synthetic route to these halfsandwich complexes is shown in Scheme 9, and the X-ray crystal

Fig. 7. ORTEP picture, with ellipsoids at the 20% probability level, of the dicationic Cu(I) catalyst. Reprinted with permission from Ref. [32] (White’s group). Copyright 2008 Royal Chemical Society.

Fig. 9. The two triazolylidene ligand-copper complexes.

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

151

Scheme 9. Syntheses of 1,2,3-triazole ligands and chelated N-triazole Ru catalysts.

Fig. 10. ORTEP diagram of catalyst 11 (a) and 13 (b) with ellipsoids given at the 30% probability level. H atoms and the PF6 − anion have been omitted for clarity. Reprinted with permission from Ref. [40] (Singh’s group). Copyright 2013 American Chemical Society.

structures (Fig. 10) of these Ru catalysts were determined. In the presence of N-methylmorpholine N-oxide (NMO) these complexes exhibited a high activity toward the oxidation of a series of primary and secondary alcohols, and these complexes were also demonstrated to be catalytically efficient in transfer hydrogenation of ketones to alcohols [40]. In both reactions the species in which N2 is bonded to Ru (catalyst 13 and 14) are more efficient than those involving N3 (catalysts 11 and 12). Cenini and co-workers studied the influence of a triazole ligand (triazoline) on the catalytic aziridination of olefins [41]. As shown in Fig. 11, they used Ru(TPP)CO (TPP = dianion of tetraphenylporphyrin) with 1-(p-nitrophenyl)-5-methyl-5phenyl-1,2,3-triazoline to yield a 2 -1,2,3-triazoline Ru(II) porphyrin complex that is responsible for the catalyst deactivation in the aziridination reaction of ␣-methylstyrene by p-nitrophenyl azide.

7.2. Ru-carbenes complexes The complexes (␩6 -arene)Ru-(NHC) are among the best catalysts in the formation of amides by direct coupling between alcohols and amines [42]. An illustration was provided by Albrecht and co-workers who introduced Ru(II) complexes comprizing a 1,2,3-triazolium-derived carbene complex as catalyst for the selective oxidation of benzylic alcohols and amines and for the oxidative coupling of alcohols and amines to form amides [43]. As shown in Fig. 12, the triazolylidene complexes (top) are more effective than the imidazolylidene systems (bottom) in the base-free oxidation of alcohols, while the opposite is found for the oxidative homocoupling of amines and the coupling of amines and alcohols to form amides. This result illustrates how subtle changes in the

Fig. 11. The triazole ligand for the Ru(II)-catalyzed aziridination of olefins.

Fig. 12. Triazolylidene and imidazolylidene ligands in Ru catalysts.

152

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

Scheme 10. Postulated activation pathway with formation of intermediates.

Scheme 11. Synthesis of 1,3-diaryl-substituted MICs based ruthenium olefin metathesis catalysts.

electron-donating properties of ancillary ligands may affect the catalytic activity of Ru(␩6 -arene)(NHC) scaffolds. Later, the same group expanded studies to triazolylidene complexes as catalysts for the oxidation of benzylic alcohols under base-free conditions [44], and the very best activity was disclosed for R = n-Bu and R = n-Bu, or R = n-Hex, R = n-Hex (Fig. 13). Additionally, other primary and secondary benzylic alcohols provided the corresponding aldehydes and ketones in good to excellent yields when these catalysts were used. The postulated catalytic activation pathway is represented in Scheme 10. The absence of base and oxidant is appealing in terms of atom economy and should allow wide functional group tolerance [44,45]. Grubbs and co-workers reported the preparation of a variety of 1,3-diaryl-substituted MICs by cycloaddition between 1,3-diaza2-azoniaallene salts and alkynes, followed by deprotonation with alkoxide bases. The highly stable N-arylated MICs were then transferred to ruthenium olefin metathesis catalysts 16 by ligand substitution (Scheme 11). The MICs-bearing ruthenium complexes

Fig. 13. Imidazolylidene-Ru complexes for the oxidation of benzylic alcohols.

Scheme 12. ROMP of cyclic olefins and RC olefin metathesis reactions.

16 are highly active in catalyzing the ring-opening metathesis polymerization (ROMP) of cyclic olefins and ring-closing (RC) olefin metathesis reactions (Scheme 12) [46]. A ruthenium complex containing an N-heterocylic carbene (NHC) and an unhindered mesoionic carbene (MIC) is extremely active for alkene metathesis upon protonation of the Ru–MIC bond by reaction with a Brønsted acid (Scheme 13) [47]. A bis-Ru(II) complex supported by a pyrrole-containing 1,2,3triazolylidene framework is an active catalyst for the ROMP of

Scheme 13. Protonation of the Ru–MIC bond to generate a metathesis-active species.

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

153

Fig. 14. Rhodium(I) and iridium(I) complexes with pyrazolyl–triazolyl bidentate ligands.

norbornene when activated with (trimethylsilyl)-diazomethane (Scheme 14) [48]. 8. Rh/Ir-triazole complexes in catalysis Rh and Ir complexes are well-known catalysts, particularly for alkene hydrogenation, hydroformylation and hydroamination, and phosphines and NHCs are frequently used as donor ligands in these reactions. Recently, the versatile 1,2,3-triazoles ligands were also combined with Rh or Ir to promote these reactions, and the corresponding complexes were good catalysts [49]. 8.1. Rh/Ir-nitrogen coordination Messerle and co-workers synthesized pyrazolyl–1,2,3-triazolyl bidentate ligands via “click” chemistry and a series of cationic rhodium and iridium complexes herewith (Fig. 14) [50a]. Singlecrystal X-ray diffraction (Fig. 15) showed that the metal center

coordinates to the N3 atom of the triazolyl moiety and the N2 atom of the pyrazole moiety, forming six-membered metallacycles. The triazolyl donating capacity is stronger than that of the pyrazolyl donor, as illustrated by the slightly shorter M–N(triazole) bonds compared with the M–N(pyrazole) bonds. These new metal complexes are efficient catalysts for the intramolecular hydroamination of a series of alkynamines and alkenamines (Scheme 15). The iridium complexes were generally more active for the intramolecular hydroamination reaction of 4-pentyn-1-amine than their rhodium analogs, while the rhodium complexes were more active catalysts than their iridium counterparts for the cyclization of alkenamines. Complexes containing the BArF 4 − (BArF 4 = tetrakis[3,5bis(trifluoromethyl)phenyl]borate) counteranion and dicarbonyl co-ligands are superior catalysts than the analogous complexes with the BPh4 − counteranion and the COD co-ligand. The pyrazolyl–triazolyl rhodium and iridium complexes showed better efficiency and selectivity than previously reported late transition metal catalysts for the same reactions [50a].

Fig. 15. ORTEP depictions of the cationic fragments of the solid-state structures of (a) [Rh(PyT)(COD)]BPh4 (19, n = 1) and (b) [Ir(PyT)(COD)]BPh4 (20, n = 1) at 40% thermal ellipsoids for the non-hydrogen atoms. Reprinted with permission from Ref. [50a] (Messerle’s group). Copyright 2012 American Chemical Society.

Scheme 14. A bis-Ru(II)-pyrrolyl-1,2,3-triazolylidene catalyst for the ROMP of norbornene.

154

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

NH2 n

(Fig. 16) [51]. However, these complexes showed limited catalytic efficiency for the intramolecular hydroamination of 4-pentyn-1amine to form 2-methylpyrroline. These results appear to confirm the general trend according which complexes with ligands that strongly bind the metal center decrease the catalytic activity for the hydroamination reaction. Asymmetric reduction of ketones was obtained by combining [RhCl2 Cp*]2 with a series of l-amino acid thioamide ligands functionalized with 1,2,3-triazoles via an asymmetric transfer hydrogenation (ATH) process [52]. The active reduction catalyst was formed in situ from [RhCl2 Cp*]2 and the amino acid-triazole ligands in the presence of sodium isopropoxide and lithium chloride. A series of aryl alkyl ketones underwent ATH in the presence of the Rh complex to obtain secondary alcohol products with high conversions and moderate to good enantiomeric excesses (Scheme 17). Generally, these triazole-functionalized catalysts showed higher activity but inferior enantioselectivity in comparison with other amino acid-based catalyst systems used in the ATH of acetophenone. The combination of phosphine and “clicked” 1,2,3-triazole ligands have also received much attention in Rh catalysis. The first new chiral 1,2,3-triazole ferrocenyl-based P,P- and P,N-ligands (ClickFerrophos) (Fig. 17, 28 and 29) were prepared by Fukuzawa et al. The Rh complexes of ClickFerrophos 29 were effective catalysts for the hydrogenation of alkenes affording products with up to 99.7% ee. Catalytic asymmetric hydrogenation of ketones with up to 98% ee was achieved by Ru complexes of Click Ferrophos 29, and the palladium complex with ClickFerrophos 28 performed well in asymmetric allylic alkylation with 79% ee [53]. The asymmetric hydrosilylation of ketones is also a versatile method to synthesize enantiomerically enriched alcohols.

N

[Ir]/[Rh]

n

R1

R1 n = 0, 1; R1 = H, Ph

R2 R2

H N [Ir]/[Rh]

NH2 H

R2 R2 R2 = Me, -(CH 2) 5-, Ph

Scheme 15. Rh and Ir catalyzed intramolecular hydroamination of alkynamines and alkenamines.

Later on, the same group further reported the synthesis of a series of Rh(III)/Ir(III) and Rh(I)/Ir(I) complexes with pyrazolyl–1,2,3-triazolyl bidentate ligands [50b]. Increasing the electron-withdrawing strength of the phenyl substituent on the triazolyl ring led to poorer donating capacity of the triazolyl donor toward the metal center. Both the Rh(I) and Ir(III) complexes are effective catalysts for C N bond formation of 2-(hydroxyalk1-ynyl)-anilines yielding the corresponding indoles, the Rh(I) complexes being the most efficient catalysts. Interestingly, the Ir(III) complexes showed efficient catalytic activity for the synthesis of tricyclic indoles by tandem C N/C C bond formation (Scheme 16) [50b]. A series of Rh(I), Rh(III) and Ir(III) complexes with bidentate NHC–1,2,3-triazolyl ligands in combination with a strong imidazolium NHC-C donor and a weak 1,2,3-triazolyl-N donor were reported by Messerle and co-workers and were expected to be desirable catalysts presenting both good stability and selectivity

BArF4 N Ir Cl N N N N 23 5 mol% BArF4

R

NO2

= tetrakis[(3,5-trifluoromethyl)phenyl]borate

n

R

R

n

HO

HO

C-N Bond Formation

n

C-C Bond Formation t

NH 2

n = 1; R = H, Me n = 2; R = H

BuOK

N H

Scheme 16. Ir(III) complex-catalyzed tricyclic indoles synthesis by tandem C N/C C bond formation.

Fig. 16. Rh(I), Rh(III) and Ir(III) complexes with bidentate imidazolium–1,2,3-triazole ligands.

N H

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

155

Scheme 17. Rhodium-catalyzed asymmetric transfer hydrogenation of aryl alkyl ketones.

Scheme 18. Rhodium-catalyzed hydroformylation of 1-octene.

8.2. Rh-nitrogen complexes with anionic triazoles

Fig. 17. Chiral triazole-based P,P- and P,N ligands.

Several Rh complexes with chiral glycosyl-triazole-based P,N ligands (Fig. 17, 30) were synthesized and applied to the catalytic asymmetric hydrosilylation of substituted acetophenones to obtain optical active alcohols in good conversion with moderate enantiomeric excesses (up to 72% ee) [54]. In some cases, ligands with strong ␲-accepting P-O-groups are more efficient in catalysis than phosphines. Takacs and co-workers prepared a series of chiral diphosphites using “clicked” triazoles to construct ligand scaffolds [55]. The triazole diphosphites were coordinated with Rh to form a 16-membered P,P-macrocyclic Rh(I) chelate (Fig. 18), which showed high enantioselectivity (up to 97% ee) in rhodium-catalyzed asymmetric hydrogenation of enamide. Mono- and bidentate phosphite ligands based on 1,2,3-triazole backbone (Fig. 19) have been prepared and applied to the Rh-catalyzed hydroformylation of 1-octene, leading to high conversions and up to 87% n-regioselectivities (Scheme 18) [56].

N N N

O P

O O

O O P O

Rh

In 2008, Shi and co-workers reported an efficient synthesis of 4,5-disubstituted-NH-1,2,3-triazoles (TRIA) through a catalytic cascade three-component condensation (Scheme 19) [57]. In comparison with generated “click” N-substituted triazoles, the NH-triazole possesses acidic N H protons and can potentially be applied as an anionic ligand. As expected, the NH-triazoles were applied to coordinate cationic Rh(I) under basic conditions, forming a new class of triazole-bridged [Rh(COD)(TRIA)]2 complexes as confirmed by X-ray diffraction (Fig. 20) [58]. Prior to this, all reported efforts regarding triazole anion binding had focused on benzotriazoles. The [Rh(COD)(TRIA)]2 complexes showed great stability toward air and moisture and exhibited effective catalytic properties in Pauson–Khand reactions (Scheme 20). 9. Pd-triazole complexes in catalysis Palladium complexes have received enormous attention from academic and industrial aspects in the last decades as exceptional catalysts for a variety of reactions including oxidation, substitution, allylic alkylation and cross-coupling reactions [59,60], and their reactivity is greatly dependent on the supporting ligands. During the last decades of last century, phosphine-based ligands have been used as the most common ligands in palladium-catalyzed reactions. However, these ligands often have drawbacks such as cost, air sensitivity, toxicity and instability at high temperatures. Thus, the search of efficient phosphine-free ligands has become active during the last 15 years, and among them N-heterocyclic carbenes

31

H R1

NO 2

R2

O

NaN3, L-Proline

+ R

2

Ar

H

1

R

N NH

DMSO, rt Ar

Fig. 18. Triazole diphosphites chelated 16-membered P,P-macrocyclic Rh(I) complex.

N

Scheme 19. Catalytic cascade syntheses of 4,5-disubstituted-NH-1,2,3-triazoles (TRIA).

156

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

Fig. 19. 1,2,3-Triazole based mono- and bidentate phosphite ligands.

KOH, [Rh(COD)Cl2 ]

N

Ph

NH

Ph Ar

Ph O

N Ph N N N N Ar 35a : Ar = Ph 35b: Ar = p-NO2 Ph

N

MeOH, rt

N

Ar

Rh

Rh

Ph

9.1. Pd-nitrogen coordination

5% 35, 6% dppp CO, xylene, 130 oC

O

O

yield: 80% (35a), 88% (35b) Scheme 20. Synthesis of Pauson–Khand reactions.

corresponding normal NHC-Pd complexes in the Miyaura–Suzuki and Mizoroki–Heck reactions, probably due to the stronger ␴donor ability of aNHC compared to NHC [62]. The versatile synthesis of 1,2,3-triazoles has made them appropriate aNHC ligands, and the 1,2,3-triazol-5-ylidene palladium complexes have showed great potential for catalytic activity in organic reactions. Furthermore, the flexibility of the nitrogen coordination of 1,2,3-triazoles provides additional opportunities in palladium catalysis.

[Rh(COD)(TRIA)]2

complexes

and

catalysis

of

have emerged as the most versatile ligands for palladium catalyzed cross-coupling reactions [61]. The majority of NHC ligands are based on imidazol-2-ylidenes and 1,2,4-triazol-5-ylidenes as normal NHCs. The Pd-aNHC complex were more effective than the

NHC ligands generally demonstrate higher donating capability but less labile properties than P- and N-donor ligands. Thus, mixed tridentate donor ligands that comprise the strong carbene ligand and relatively labile nitrogen donor atoms are effective in achieving both high catalytic activity and stability. A novel ligand that features a carbene and a labile triazolyl donor has been synthesized via a copper catalyzed click reaction by Chen and co-workers [63], and silver, palladium, and platinum complexes of this ligand have been synthesized and characterized. For instance, the pincer palladium complex represented in Fig. 21 shows high activity

Fig. 20. Crystal structure of 35a. Reprinted with permission from Ref. [58] (Shi’s group). Copyright 2009 American Chemical Society.

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

157

Fig. 21. A pincer palladium complex with a labile triazolyl donor.

N

N N N

R

Pd

AgBF4 Pd Cl

-AgCl

N

N N

R BF 4-

N R = Ph, 36a R = n Pr, 36b

Scheme 21. Syntheses of triazole chelated cationic palladium allyl complexes.

for Suzuki–Miyaura cross coupling reactions of aryl bromides and 1,1-dibromo-1-alkenes in neat water under air. Two triazole chelated cationic palladium allyl complexes have been synthesized by Scrivanti and co-workers (Scheme 21) [64], and single-crystal X-ray diffraction analysis of the solid-state structure revealed that the triazole ligand chelates the palladium through N2 nitrogen on the triazole heterocycle and N4 nitrogen on the pyridine substituent (Fig. 22). These complexes have also are active catalysts in the Suzuki–Miyaura coupling of phenyl boronic acid with aryl bromides (Scheme 22). Very recently, Wang et al. reported the use of 2-pyridyl-1,2,3triazole (pyta) as a bidentate ligand to coordinate with palladium via N3 of 1,2,3-triazole and the nitrogen atom of the pyridinyl attached to the C4 of the 1,2,3-triazole. Various mono- and polymetallic palladium complexes containing a 2-pyridyl-1,2,3-triazole

Fig. 22. ORTEP view of the cation of 36a. The hydrogen atoms and the tetrafluoroborate anion have been omitted for clarity. Thermal ellipsoids drawn at the 40% probability level. The position of the C(16A) atom with refined site occupancy = 0.41 has been drawn with white bonds. Reprinted with permission from Ref. [64a] (Scrivanti’s group). Copyright 2011 Elsevier B.V.

R

Br + (HO)2 B

36a or 36b 0.01-0.1 mol% K2 CO3, DMF/H 2 O

R

R = CH 3C(O) 36a, 68% conv; 36b, 98% conv 36a, 28% conv; 36b, 22% conv R = CH 3 R = CH 3CO 36a, 40% conv; 36b, 10% conv

Scheme 22. Catalytic activities of complexes 36 in Miyaura–Suzuki coupling reactions.

ligand and a nonabranch-derived ligand have been synthesized (Scheme 23) and are excellent homogeneous or heterogeneous catalysts for Suzuki–Miyaura, Sonogashira and Heck reactions [65].

Scheme 23. Synthesis of mono palladium complexes containing 2-pyridyl-1,2,3-triazole ligands (37) and polymetallic palladium complexes containing a nonabranch-derived ligand (38).

158

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

Scheme 24. The applications of benzotriazole palladium complex in various coupling reactions.

Fig. 23. A fluoroalkylated triazole ligand.

Verma et al. have designed a robust N,N-type of bidentate ligand based on benzotriazole with N3 being substituted by 2-pyridyl, for the palladium-catalyzed C C (including Miyaura–Suzuki, Heck, Fujiwara–Moritani, and Sonogashira), C N and C S coupling reactions (Scheme 24) [66]. The bidentate ability of the ligand is considered to be enhanced by the donor ability of the N N bond of the benzotriazole ring and the lone pair of electrons on the nitrogen atom of the pyridyl ring. The ligand is inexpensive, thermally stable, and easy to synthesize. Along with the tolerance of a variety of functionalized reactants, this ligand is expected to find great applications in organic synthesis. A recyclable fluoroalkylated 1,2,3-triazole (Fig. 23) was prepared as an efficient ligand for the palladium-catalyzed Suzuki–Miyaura and Mizoroki–Heck reactions [67]. The ligand

could be recovered by fluorous solid-phase extraction with only slightly decreased activity. In recent years, the combination of phosphine and “clicked” 1,2,3-triazole have emerged and acted as ligands in catalysis (Fig. 24). Zhang and co-workers have prepared triazole-based monophosphine ligands (ClickPhos) (39), and their palladium complexes have been demonstrated to be effective catalysts in the amination and Miyaura–Suzuki coupling reactions of unactivated aryl chlorides [68]. ClickPhine ligands (40) have also been synthesized by click cyclization of propynyldiphenylphosphine as reported by van Maarseveen and co-workers. These ligands were efficient in the Pd-catalyzed allylic alkylation reaction [69]. 9.2. Pd-carbenes complexes Since the first synthesis and characterization of 1,2,3-triazol5-ylidene-metal (Pd, Rh, and Ir) complexes by Albrecht in 2008, their catalytic properties have subsequently been investigated. In 2009, Sankararaman and co-workers [70] reported the first chiral palladium aNHC complex 41 and the first palladium bis-aNHC chelated pincer complex 42, both of which derived from 1,2,3triazolylidene (Fig. 25). The palladium complexes were obtained by

Fig. 24. ClickPhos ligands 39 and ClickPhine ligands 40.

Fig. 25. Several palladium aNHC complexes developed by Sankararaman.

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

COOEt

43 (0.5 mol%) +

COOE t

159

COOEt

TFA, CH 2Cl2

+

44%

COOEt 13%

Scheme 25. Hydroarylation of alkynes.

Mes

Me

Mes Cl N N N Pd N N N Cl Me Mes Mes 44

Me

Dipp Cl N N Pd N

Dipp Ph Dipp = 2,6-diisopropylphenyl 45

Fig. 27. Palladium aNHC complexes prepared by Fukuzawa.

Fig. 26. ORTEP drawing of 43 with thermal ellipsoids at the 30% probability level. Reprinted with permission from Ref. [71] (Sankararaman’s group). Copyright 2011 American Chemical Society.

transmetallation of the corresponding silver-carbene complexes. Both complexes are effective in the catalysis of the Suzuki coupling reaction for the synthesis of biphenyl derivatives. However, they failed to catalyze the formation of binaphthyl derivatives, instead leading to a deboronation reaction. Later on, the same group synthesized a 1,2,3-triazolylidene-based binuclear palladacycle complex 43 with bridging acetate ligands (Figs. 25 and 26) [71]. Both palladium atoms are attached to both the carbene carbon and the carbon atom located at the ortho position of the phenyl ring on the nitrogen atom of the triazole ring. The complex showed moderate catalytic activity in the hydroarylation reactions of alkynes in the presence of TFA (Scheme 25). Several electron-rich arenes underwent hydroarylation to form the corresponding vinyl derivatives in a stereoselective manner. Fukuzawa and co-workers [72a,72b] prepared a palladium aNHC complex 44 derived from 1,4-dimesityl-1,2,3-tirazole (Fig. 27). This catalyst was successfully used in C C cross-coupling reactions such as the Miyaura–Suzuki, Mizoroki–Heck, and Sonogashira reactions. It was demonstrated that this bis-TMes-Pd complex was even more efficient than the bis-IMes-Pd analog complex that is known as an effective imidazole carbene complex due to electronically-rich and sterically-hindered mesityl groups. A mono-triazolylidene aNHC-palladium complex 45 (Fig. 27) with a cinnamyl ligand was also synthesized and is a highly active catalyst for the room-temperature Suzuki–Miyaura coupling reaction

R Cl Cl N N Pd N Me N Cl R' R = Me, Et, nBu, Mes R' = Ph, nBu, Mes 46

[72c]. The reactions proceeded with aryl chlorides, regardless of the electronic and steric properties of the substituents, as well as with sterically crowded arylboronic acids. The remarkably superior catalytic activity of these triazolylidene NHC complexes may be attributed to the stronger donor properties of the ligands than those of the imidazole-derived NHCs. In 2012, Albrecht and co-workers [73] reported the synthesis of a series of 1,2,3-triazolylidene-derived pyridine enhanced precatalyst preparation stabilization and initiation (PEPPSI) palladium complexes 46 with the 3-chloropyridine ligand as an easily cleavable ligand (Fig. 28). The activity of these complexes in Suzuki–Miyaura cross-coupling can be adjusted by varying the substituents on the triazolylidene ring. In contrast to imidazol-2ylidene, less bulky substituents induce better catalytic activity than the bulkier, sterically congested mesityl substituents. Experimental evidence indicates that palladium nanoparticles are generated as the resting state of the catalyst in a heterogeneous manner, and palladium atoms are leached from the nanoparticles as active species to catalyze the reaction under relatively mild conditions. However, these complexes appear to be less effective than the analogous NHC derivatives reported by Organ [74]. Nearly in the same time, Crudden and co-workers [75] reported similar mono- and bimetallic 1,2,3-triazol-5-ylidene abnormal carbene complexes of palladium with pyridine as ligand (Fig. 28, 47 and 48). These new PEPPSI complexes were tested in the Mizoroki–Heck reaction, and high conversion was observed with methyl acrylate in the case of aryl iodides and electron-deficient bromides. Consistent with Albrecht’s observation, the reaction likely proceeds via palladium nanoparticles as suggested by the mercury-poisoning test. A series of hetero-bis(carbene) complexes bearing i Pr2 -bimy and mesoionic 1,2,3-triazolin-5-ylidenes with various substituents have been reported by Huynh and co-worker (Fig. 29) [76a]. A study of the 13 C NMR spectra shows that the 1,2,3-triazolin-5-ylidenes are generally stronger donors than normal NHCs but weaker than some abnormal NHCs such as pyrazolin-3-ylidenes and mesoionic imidazolin-4-ylidenes. The trifluoroacetato analogs have been

Dipp N N Cl N Cl Pd Cl N Pd Cl Pd N Cl N Me Cl N Dipp Dipp N Ph N N N N Dipp = 2,6-diisopropylphenyl Me Me 47 48 Fig. 28. PEPPSI palladium complexes.

160

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

Scheme 26. Direct arylation of pentafluorobenzene catalyzed by a trifluoroacetato complex.

Scheme 27. Synthesis of a normal 1,2,3-triazolylidene carbene Pd complex and its X-ray structure. Reprinted with permission from Ref. [14b] (Kuhn’s group). Copyright 2013 American Chemical Society.

[14b]. During the reaction, formation of a precipitate was observed, which indicated that complex 51 may be not stable under oxidative conditions. 9.3. Other types of molecular Pd-triazole catalysts

Fig. 29. Hetero-bis(carbene) palladium complexes.

synthesized through salt metathesis of the bromides with AgO2 CCF3 . Both of them were used as catalysts in the direct arylation of pentafluorobenzene, and trifluoroacetato complexes showed better reactivity than their bromo analogs (Scheme 26). Generally, complexes bearing less donating triazole ligands perform better in this catalysis. Although the majority of 1,2,3-triazolylidenes form abnormal carbenes exhibiting a 1,3,4-substitution pattern, an example of normal 1,2,3-triazolylidene carbene with a 1,2,4-substitution pattern was reported very recently [14a]. The Pd complex shown in Scheme 27 was applied to Suzuki–Miyaura coupling of aryl bromides and chlorides. Mercury-poisoning experiments suggested that activity of the catalytic species was at least partly due to Pd nanoparticles. Moderate performances with aryl chloride substrates were observed probably due to the instability of the catalysts [14b]. The Mo complex was used to catalyze the epoxidation of olefins and showed moderate catalytic activity (Scheme 28)

Beside the above modes of triazole bonding with palladium, the relatively acidic C H bond on the 5-position can be inserted by palladium to form a carbon–metal bond. The use of pincer complexes as catalysts has been extensively investigated because of their unique coordination environment and adjustable steric and electronic properties. However, the synthesis of various related pincer ligands still remains a challenge. Gandelman and co-workers [76b] successfully synthesized a hetero-tridentate triazole-based pincer complex in which palladium is bonded to the ligand through two donor groups in the 1,4-positions and the carbanion in the 5position by directed insertion of the relatively acidic C H bond (Scheme 29). And the structure was confirmed by X-ray structure analysis (Fig. 30). The resulting triazole-based pincer complex exhibits extremely high catalytic efficiency in the Heck reaction. The application of “clicked” triazole rings in ligand synthesis should allow access to a broad range of tailor-made pincer ligands. 9.4. Pd-nanoparticle catalysts stabilized by 1,2,3-triazole-containing macromolecules In recent years, interest in nanoparticle (NP) catalysts has considerably increased because of their high reaction efficiency and environmentally benign conditions [77]. Ornelas and co-workers have synthesized Pd nanoparticles stabilized by 1,2,3-triazole-

N N N D1

N N N

[PdCl2 (tmeda)] D2

Et 3N, DMF, 70 oC

D1

Pd Cl

D2

D 1 = PPh2 , SPh D 2 = pyridine, PPh2 , P(o-MeOC 6H 4) 2 Scheme 28. Mo catalyst for the epoxidation of cyclooctene.

Scheme 29. Synthesis of hetero-tridentate triazole-based pincer complexes.

D. Huang et al. / Coordination Chemistry Reviews 272 (2014) 145–165

161

Fig. 30. X-ray structure of a pincer complex (D1 = D2 = PPh2 ) in Scheme 29. Reprinted with permission from Ref. [76b] (Gandelman’s group). Copyright 2008 Wiley-VCH.

containing dendrimers with ferrocenyl and other termini (Fig. 31) and related polymers. These PdNPs showed good catalytic activity in Miyaura–Suzuki reactions and hydrogenation of olefins [77a–77d]. Water-soluble sulphonated 1,2,3-triazole-containing dendrimer-stabilized palladium nanoparticles also demonstrated to be highly effective to catalyze allylic alcohol hydrogenation and Suzuki coupling reactions in aqueous media under ambient conditions [77e]. Clicked dendrimers that are terminated by triethylene glycol (TEG) termini are stabilizers of palladium nanoparticles (PdNPs) that show excellent catalytic activity for Miyaura–Suzuki reactions of bromoaromatics at 80 ◦ C with Pd amounts down to sub-ppm level. These PdNPs are stable in air for months and retain their catalytic activity in the case of small “clicked” zeroth-generation arene-cored dendrimers surrounding 1.4-nm-sized PdNPs [77f]. With related polymers containing both TEG and triazole polymers, the same reactions are faster with a few ppm Pd, but the PdNPs are less stable [77g]. The parameters are consistent with an “homeopathic” and leaching mechanism, i.e. the catalytic reactions work

all the better as the Pd concentration is lower, because the leached active Pd atoms are less quenched by the mother PdNPs as the Pd concentration decreases down to the ppm level [78a–78c].

10. Au-triazole complexes in catalysis For a long time, Au was considered to be inert in catalysis. The pioneer of Au-catalyzed chemistry was Haruta who discovered in the 1980s that small oxide-supported Au nanoparticles (