Chemistry of Compounds Based on 1,2,3-Triazolylidene-Type Mesoionic Carbenes

Abstract

Mesoionic carbenes (MICs) of the 1,2,3-triazolylidene type have established themselves as a popular class of compounds over the past decade. Primary reasons for this popularity are their modular synthesis and their strong donor properties. While such MICs have mostly been used in combination with transition metals, the past few years have also seen their utility together with main group elements. In this paper, we present an overview of the recent developments on this class of compounds that include, among others, (i) cationic and anionic MIC ligands, (ii) the donor/acceptor properties of these ligands with a focus on the several methods that are known for estimating such donor/acceptor properties, (iii) a detailed overview of 3d metal complexes and main group compounds with these MIC ligands, (iv) results on the redox and photophysical properties of compounds based on MIC ligands, and (v) an overview on electrocatalysis, redox-switchable catalysis, and small-molecule activation to highlight the applications of compounds based on MIC ligands in contemporary chemistry. By discussing several aspects from the synthetic, spectroscopic, and application point of view of these classes of compounds, we highlight the state of the art of compounds containing MICs and present a perspective for future research in this field.

1. Introduction

The thermal cycloaddition reaction between alkynes and azides was first reported in the late 19th century.¹ Huisgen and his co-workers systematically investigated this thermal reaction that generates a mixture of the 1,4- and 1,5-regioisomers of the 1,2,3-triazoles (Scheme 1).^2,3 In 2002, two groups independently reported the Cu(I)-catalyzed version of this reaction that exclusively delivers the 1,4-substituted 1,2,3-triazole.^4,5 A few years later, synthetic methods based on Ru-catalyzed reactions as well as metal-free catalysis were reported for the exclusive generation of the 1,5-isomers.⁶⁻¹⁰ These two reactions that selectively generate one or the other regioisomers of the 1,2,3-triaozles are significant, as these reactions often provide access to a pure compound without the need for cumbersome purification processes. Furthermore, many of these reactions work under mild and ambient conditions and can be performed in environmentally benign solvents. Because of the aforementioned conditions, these reactions have often been called examples of the so-called “click” reaction.¹ Even though there are certain limitations in terms of substrate scopes, particularly for the synthesis of the 1,5-substituted 1,2,3-triazoles, these synthetic methods are by and large modular. This aspect has led to the huge popularity of these substance classes since the discovery of the catalyzed versions of the azide–alkyne cycloaddition reaction.^11,12

Scheme 1. Thermal and Catalytic [3 + 2] Azide–Alkyne Cycloaddition Reaction.

In 2008, it was reported that the 1,4-substituted 1,2,3-triazoles can be selectively methylated at the N3 atom to generate the corresponding triazolium salts in near quantitative yield.¹³ These triazolium salts can then be deprotonated to generate 1,2,3-triazol-5-ylidenes, a compound class that has been called abnormal carbenes or mesoionic carbenes (MICs) because it is not possible to formulate a localized structure of these compounds without charge separation while following the octet rule (Figure 1). A few years later, synthesis of MICs derived from 1,5-substituted 1,2,3-triazoles was reported, as well.¹⁴ These two types of MICs together with the other MIC, imidazol-4-ylidenes, have gained immense popularity in the past years in both transition metal and main group chemistry. One prominent reason for this popularity is the stronger donor properties of MICs in comparison to those of most of the classical N-heterocyclic carbenes (NHCs, Figure 1). Even though MICs of the 1,2,3-triazol-5-ylidene types are often derived from the corresponding 1,2,3-triazoles, as mentioned above, there is an important alternative route for synthesizing such compounds that start from 1,3-diaza-2-azoniaallene salts and alkynes (Scheme 2).¹⁵

Figure 1.

Normal and abnormal or mesoionic carbenes.

Scheme 2. General Synthesis of 1,2,3-Triazoles, 1,2,3-Triazolylidenes, and Their Metal Complexes.

Even though the term MIC has been coined very recently, compounds containing such properties were already reported in 1993.¹⁶ However, at that time, they were not called so. Following the first documented report on 1,2,3-triazol-5-ylidene types of MICs in 2008, the use of these compounds, particularly in combination with transition metals, has been hugely popular.¹³ Accordingly, a few review articles in the field have appeared in recent years.¹⁷⁻²⁰ In this paper, we tackle this topic with a particular focus on the use of these MICs together with first-row transition metals and with main group elements. We will describe MICs that are neutral and contain various numbers of MICs within the same compound. Additionally, we will also report on MICs that are either anionic or cationic, examples of both of which remain extremely rare in the literature. General synthetic routes for generating metal complexes from MICs will be presented. The donor/acceptor properties of these MICs will be analyzed with the help of the Tolman electronic parameter (TEP, IR spectroscopy) but also with ¹H, ¹³C, and ⁷⁷Se NMR spectroscopy. The redox properties and the photophysical properties of several MIC-based compounds will be showcased. Finally, we will present an overview on electrocatalysis, redox-switchable catalysis, and small-molecule activation to highlight the applications of compounds based on MIC ligands in contemporary chemistry. The topic of general thermal homogeneous catalysis will not be specifically addressed here, as this aspect was reviewed several times in the recent years.¹⁸⁻²³

2. Synthesis of MICs, Stability of Free MICs, and General Synthesis of MIC-Containing Compounds

As mentioned above, 1,2,3-triazol-5-ylidenes are often synthesized from the corresponding triazolium salts, which in turn are obtained by the alkylation (or sometimes arylation) of the 1,2,3-triazoles (Scheme 2). The alkylation at the N3 position of the triazoles is accomplished using alkyl halide or Meerwein salt or a triflate owing to the most nucleophilic nature of the N3 site. The N3-aryl substitution is facilitated via copper-catalyzed reaction with an aryl iodonium salt.^24,25

The synthesis and characterization of the first free 1,2,3-triazol-5-ylidene was demonstrated by Bertrand and co-workers in 2010 (Scheme 3).²⁶ The free triazolylidene with a methyl substituent at the N3 position is stable for a few days at low temperature (−30 °C) under the exclusion of air. The observed stability for the free carbene increases with bulkier substituents at the N3 position on the triazolylidene ring; therefore, the free triazolylidene with an ⁱPr substituent at N3 is significantly more stable compared to the N3-methyl-substituted triazolylidene in the solid state.²⁶ The free carbene 2 decomposes in benzene solution when heated at 50 °C for 12 h to yield triazole 6 as one of the decomposition products (Scheme 3). It is assumed that a nucleophilic attack of the carbene lone pair of 2 on the methyl group of a second molecule of the same leads to the formation of heterocycles 4 and 5, which react together to give triazole 6. Installing an aryl substituent at N3 increases the stability of these free MICs even further. Even though free triazolylidenes have been isolated and characterized by single-crystal X-ray diffraction (XRD), such examples remain rather rare.

Scheme 3. Synthesis and Reactivity of Free 1,2,3-Triazolylidene.

The synthesis of metal complexes possessing triazolylidene donor ligands is achieved by different synthetic routes depending on the nature of the metal centers. The majority of these metal complexes have been synthesized from the corresponding triazolium salts via the in situ generated triazolylidene followed by the metalation process.¹³ As there are other reviews mentioning different synthetic routes,¹⁷⁻²³ this paper just schematically (Scheme 4–8) shows the most frequently adapted methods for the synthesis of metal complexes such as (a) transmetalation from silver (Scheme 4), (b) direct metalation using basic metal precursors (Scheme 5), (c) base-mediated proton abstraction followed by metalation (Scheme 6), (d) coordination to a free triazolylidene (Scheme 7), and (e) transmetalation from copper (Scheme 8).

Scheme 4. General Synthesis of Triazolylidene Complexes via Transmetalation from Silver²⁷⁻³⁴.

Scheme 8. Synthesis of Metal Complexes via Copper–Triazolylidene Complexes⁴⁹⁻⁵¹.

Scheme 5. General Synthesis of Triazolylidene Complexes Using Basic Metal Precursors³⁵⁻⁴².

Scheme 6. Base-Mediated Synthesis of Triazolylidene Complexes⁴³⁻⁴⁵.

Scheme 7. Synthesis of Metal Complexes from Free Triazolylidenes⁴⁶⁻⁴⁸.

Even though a large number of 1,2,3-triazolylidene-containing compounds have been reported in the literature, several synthetic challenges still remain. The most pressing challenge is perhaps finding ways to increase the stability of such MICs. The positive aspect of such MICs is that the stability is not dependent on the steric bulk of the substituents on the N1 and C4 atoms of the 1,2,3-triazol-5-ylidene ring. This aspect should make the isolation of many different MICs possible, due to the modular synthetic route. The major drawback is the restriction placed on the nature of the substituent on the N3 atom. A huge majority of MICs that have been reported contain a methyl substituent at the N3 atom. The stability of such MICs is very limited as has been stated above, and it will not be possible to increase the stability of these N3-methyl-containing MICs. As robust synthetic routes are now available for placing really diverse aryl substituents at the N3 position, focusing on such MICs seems to be the way forward for increasing their stability. The alternative synthetic route starting from 1,3-diaza-2-azoniaallene (Scheme 2) for synthesizing these MICs is another important strategy to increase their stability. So far, as metal complexes of such MICs are concerned, reliable synthetic routes do exist for 4d and 5d metal complexes. The synthesis of 3d metal complexes with these MICs is still an emerging field. For such metal complexes, the Ag-mediated transmetalation route often fails. The most reliable synthetic route for generating 3d metal complexes is either the use of an internal base in the metal precursor or the use of a strong base for generating the free carbene in situ before metalation. Both of these routes have their limitations. The former one is restricted by the very limited number of available metal precursors that fulfill the requisite criteria. The latter is often based on trial and error, as the conjugate acids of several bases are known to undergo unwanted side reactions with the free carbene. The Cu-mediated transmetalation route has not yet been tested for many cases. This is a synthetic route that might bring many more positive results for the synthesis of 3d metal complexes. The best solution to the synthetic problems related to metal complexes would, of course, be the generation of free MICs with a higher stability than is currently possible.

3. Cationic and Anionic MICs

Despite the huge development of MIC donor ligands dominated by neutral MICs, examples of cationic and anionic MICs remain extremely rare.^52,53 One class of cationic MIC donor ligands (29 and 30) possesses ferrocenyl substituents on the triazolylidene ring, which allows one-electron and two-electron oxidized ferroceniumyl MICs (Figure 2). The donor properties of these MICs are strongly influenced by the oxidation steps and have also been compared with the conventional phosphine donor ligands using the TEP.^54,55 The metal complexes bearing the electron-poor oxidized MIC donor ligands also show advantages over the neutral MIC donor ligands in various organic catalytic transformations.⁵⁴ Another class of cationic MICs contains cobaltoceniumyl substituents on the central ring (31 and 32). The dicationic dicobaltoceniumyl-containing MIC 32 is an extremely electron-poor ligand and has the highest TEP (2109 cm^–1; see below) value reported to date for any carbene ligand. This TEP value is comparable to PF₃. Such ligands are highly useful in catalysis requiring a Lewis acidic metal center.⁵⁶

Figure 2.

Cationic triazolylidene donor ligands employed in metal complexes.

Smith and co-workers reported on the anionic MIC-based tris(carbene)borate ligand 33 (Figure 3), which has been synthesized from the 1,5-regioisomer of the corresponding triazole.^14,57 This ligand has a donor strength lower than that of the corresponding imidazol-2-ylidene-based tris(carbene)–borate ligands, as examined by the IR spectroscopic measurements with manganese and nickel derivatives.⁵⁷ This fact is perhaps related to the all-aryl substituents on the triazolylidene rings in this ligand. Anionic borate-based bis-MIC donor ligands 34 and 35 have recently been described by Sarkar and co-workers (Figure 3). These ligands undergo C–N isomerization reactions, and the isomerized compounds are excellent ligands for Co^II centers.⁵³ An anionic 1,2,3-triazole-4,5-diylidene ligand 36 was described by Bertrand and co-workers for the synthesis of 1,2-dimetallic complexes.⁵⁸ Monoanionic CNC-pincer ligand 37 bearing bis-MIC donors was also reported.⁵⁹ The potassium adduct of the anionic ligand was isolated and characterized by XRD. The related mono- and dianionic bistriazolylidene donor ligands 38 and 39 have been used for the synthesis of their main group adducts.^60,61 A 3,5-di-tert-butylphenolate-based dianionic triazolylidene donor ligand 40 has been employed for the synthesis of the rare version of their Ti^IV complexes.⁶²

Figure 3.

Anionic triazolylidene donor ligands employed in metal complexes.

Both the cationic and anionic MICs are unique ligand classes that are completely underdeveloped until now. The cationic MICs are excellent ligands in catalysis that require a combination of strong metal–ligand bonds with attenuated donor properties of the ligands. That ligands such as 29–32 foot that bill have already been demonstrated by catalysis using their gold complexes (see below). The redox-switchable nature of the substituents is another advantage of these cationic MICs. This aspect will potentially be useful in developing orthogonal catalysis with metal complexes of such ligands. The disadvantage of these MICs is the fleeting nature of 29 and 30 because of a combination of the limited stability of the MICs themselves and of ferrocenium. The unstable nature of MICs 31 and 32 in their reduced form make their use in redox-switchable catalysis impossible. While there is still a plethora of new and exciting chemistry to be explored and developed with 29–32, synthetic access to more stable cationic MICs will also be necessary for the future development of the field. Almost all of the anionic MICs shown in Figure 3, with the exception of 37, have only found sporadic use until now (Figure 3). For 34 and 35, the relatively weak nature of the B–N bonds and the resulting C–N isomerization reactions in these ligands provide new challenges as well as opportunities. The elucidation and control of these isomerization reactions will give access to two complementary classes of ligands (C-bound and N-bound). These isomerization reactions will likely be operative in 33, as well, but this aspect has not been explored until now. 33–35 are also valuable synthons for generating platforms for small-molecule activation and catalysis with small molecules. This feature has not been tested until now. Ligand 36 can be converted into a very intriguing platform for cooperative catalysis (with very small metal–metal distance) by introducing additional donor atoms at the substituents on N1 and N3 for additional stability of the resulting metal complex. Such variations have not been carried out on 36 as yet. 38, 39, and 40 are relatively new additions to the field of MICs. 38 is related to the famous porphyrin class of molecules but is not conjugated and possesses two different kinds of donor atoms. This ligand platform is thus expected to give rise to chemistry that is distinct from the porphyrin systems. First examples have already been published (see below). 39 will likely be an important platform for generating several emissive compounds. The flexibility around the triazolylidene rings in this compound is rather limited, thus placing severe restrictions on nonradiative decay in compounds containing 39. 40 is and will be a useful ligand for generating high oxidation state metal complexes in combination with early transition metals. In general, as mentioned for the neutral and cationic MICs, the chemistry of anionic MICs will also greatly benefit from the development of strategies to increase the stability of the free anionic MICs.

4. Electronic Properties (Donor/Acceptor) of MICs

Except for very few exceptions, the triazolylidene-based MIC ligands are proposed as spectator ligands in catalytically active metal complexes.^63,64 The optimization of electronic properties of the ligand plays a major role in enhancing the catalytic activities and photochemical/photophysical properties of metal complexes.⁶⁵ The TEP is an extremely useful tool to determine the electronic properties of MICs, despite its certain shortcomings. The TEP was initially introduced by Tolman in 1970 as a parameter to interpret the donor–acceptor properties of phosphorus ligands.⁶⁶ The method originally described the determination of electron donor–acceptor performance of particularly phosphorus ligands based on the A₁ stretching frequency of the carbonyl ligands in Ni(CO)₃PR₃. A strong net electron donor ligand would make the complex more electron-rich, which therefore increases the amount of back-donation to CO and leads to further weakening of the CO bond, and subsequently, a smaller wavenumber is observed. This method has also successfully been extended to NHC ligands.⁶⁷ However, only a limited number of NHC ligands have been evaluated experimentally using nickel carbonyl complexes due to extreme toxicity and the low boiling point associated with [Ni(CO)₄]. The square planar cis-[IrCl(CO)₂(NHC)] and cis-[RhCl(CO)₂(NHC)] complexes have generally been employed as alternatives to the toxic nickel carbonyl complexes to determine the TEP for the NHC donor ligands.^27,68 The average CO stretching frequencies of the corresponding iridium^68,69 and rhodium^68,70 carbonyl complexes in CH₂Cl₂ can be converted to TEP values with the help of regression coefficients.⁷¹ A wide range of complexes possessing triazolylidene-based MIC ligands have been investigated by this method (Figure 4). It has been observed that the general TEP values of 1,2,3-triazolylidene ligands fall in the range between 2039 and 2051 cm^–1 (Figure 4).⁶⁸ This donor strength of the triazolylidene ligands measures between classical imidazolylidene donor (TEP: 2049–2052 cm^–1) and mesoionic imidazolylidene-type donor ligands (TEP: 2003–2015 cm^–1), which are even more electron-donating than triazolylidene MIC ligands.⁶⁸ The observed high catalytic activity of 1,2,3-triazolylidene ligands containing metal complexes is often attributed to their strong overall electron-donating properties.

Figure 4.

Overview of experimental and density functional theory determined TEP values of triazolylidene-based MICs.⁶⁸

The trend of the overall donor properties within the triazolylidene-type ligands generally follows the chemical intuition: the triazolylidene possessing an electron-deficient 4-iodo substitution shows the lowest donor capacity based on the experimentally determined TEP value (2051 cm^–1),⁷² followed by a 4-ethoxy-substituted MIC.¹⁵ Therefore, the observed TEP value shows that both 4-iodo-1,2,3-triazolylidene and imidazolylidenes possess similar electronic properties. 1,4-Diarylated triazolylidenes show TEP values slightly higher than those of the 1,4-diphenyl-substituted MIC.^15,54,73 The triazolylidenes with alkyl substituents at the 1- and/or 4-position show TEP values lower than those of their 1,4-diarylated counterparts.^13,35 These values might be due to the stronger +I effect of the alkyl substituents. Interestingly, a tritopic tris-MIC ligand with a TEP of 2043 cm^–174 appeared as a stronger overall electron-donating ligand than the corresponding mono-MICs (2047 cm^–1). The aryl core and/or cooperative effects seem to influence the donor capacity of the ligand.

One feature of triazolylidene-based MICs is the straightforward incorporation of metallocene units in the triazolylidene moiety through the corresponding ferrocenyl and/or cobaltoceniumyl alkyne and/or azide. The metallocene substituents can incorporate special electronic properties into MICs. The calculated TEP values for mono- and dicationic colbaltoceniumyl-substituted MICs suggested the extremely weak electron donor properties of these ligands.⁵⁶ The calculated value for the dicationic MIC (2109 cm^–1, Figure 4) is close to the TEP of the π-acidic PF₃ ligand.⁷⁵ The complexes containing ferrocenyl-substituted MIC ligands also show intriguing electrochemical and chemical reversibility upon oxidation. It has been demonstrated experimentally in various cases that the ferrocenyl-center-based electrochemical or chemical oxidation of a ferrocenyl-substituted triazolylidene significantly decreases the donor strength of the corresponding MIC ligand (Figure 5).^54,55 The TEP values obtained from analysis of the iridium carbonyl complexes showed that the neutral ferrocenyl MIC ligands are stronger donors than the imidazolylidene-based carbenes. The donor strength of one-electron-oxidized ferrocenyl MICs is in the range of that for the tricyclohexyl phosphines, and the electron-poor two-electron-oxidized forms lie in the range of triphenyl phosphines.⁵⁴ In addition, these redox processes are chemically reversible in nature. These advantages have also been successfully incorporated for the synthesis of gold(I) complexes bearing electron-poor ferrocenyl-based MIC donor ligands. The gold(I) complexes featuring oxidized MIC ligands appeared to be a better catalyst compared to their neutral counterparts by almost 10 times in the synthesis of oxazolines. These results indicate the rare version of redox-induced and redox-switchable catalysis with ferrocenyl-substituted MIC gold complexes.^47,54,55

Figure 5.

Experimentally determined TEP values of the ferrocenyl-based MIC donor ligands.^47,54,55

The electronic parameters of triazolylidene-based MIC ligands have also been studied experimentally and ranked on a unified ¹³C NMR scale using easily accessible trans-[PdBr₂(ⁱPr₂-bimy)L] (ⁱPr₂-bimy = 1,3-diisopropylbenzimidazolin-2-ylidene; L = ligand in interest)-type complexes as spectroscopic probes.⁷⁶ This method to determine the electronic parameter was introduced by Huynh and co-workers.^68,76 The Huynh’s electronic parameter (HEP) methodology is based on the sensitivity of the ⁱPr₂-bimy carbene signal to the donor strengths of the varying coligand L. The method offers the detection of backbone and substituent effects of a ligand (mostly σ-donor strength) in a more accurate manner than the previously described carbonyl-based systems, which can only detect the net donating ability of a ligand that consists of both σ-donor and π-acceptor contributions. The noticeable advantage of this method includes the easy and carbonyl-free one-pot preparation of the palladium(II) complexes featuring novel heterobis(carbene) donor ligands. A square planar palladium complex of type 41 or a linear cationic gold complex of type 42 containing the ligand under investigation and a 1,3-diisopropylbenzimidazolin-2-ylidene ligand trans to the investigated ligand needs to be synthesized (Figure 6).^77,78

Figure 6.

Model complexes to determine the HEP.⁶⁸

The chemical shift of the benzimidazolylidene carbene atom in the ¹³C NMR spectra is used as the probe for σ-donor strength of the NHC under investigation. For the gold complexes, a linear regression formula converts the chemical shift into the actual HEP.⁶⁸ It is assumed that the metal centers of the model palladium(II) and gold(I) complexes are relatively Lewis acidic, and consequently, the π-back-donation to the ligand is assumed to be insignificant.⁷⁷ Until now, the HEP scale has covered only a small selection of 1,2,3-triazolylidenes ranging from 181.2 to 177.9 ppm (Figure 7).^68,72 Similar to the TEP values, the HEPs of the 1,2,3-triazolylidenes range between the values of classical imidazolylidene-type NHCs and mesoionic imidazolyidenes.⁶⁸ Even though the triazolylidene HEPs cover a small range on the HEP scale, the resolution of HEP is expected to be higher compared to the TEP. The ranking of the MIC within the scale follows the chemical intuitions as for the TEP. The N1-isopropylated triazolylidene displays the better σ-donor capacity compared to the N1-Ph- or N1-Bn-substituted triazolylidenes owing to their increasing +I effect of their N1 substituents in the order Ph < Bn < ⁱPr. The para-nitrophenyl-substituted MIC is less electron-donating according to the HEP.

Figure 7.

HEP values of triazolylidene based MIC donor ligands.^68,72

The MIC with the smallest HEP (Figure 7), the 4-iodo-substituted MIC (177.9 ppm), was reported to be tunable by adding electron donors to the NMR spectroscopically analyzed solution.⁷² The addition of iodide ions to the gold(I) complex bearing 4-iodo-substituted MIC donor ligand showed a significant shift of the HEP to 179.6 ppm, which is attributed to the halogen bonding between the gold complex and iodide donor. A shift of 1 ppm on the HEP scale should display a significant increase in σ-donor capacity of the MIC. Halogen bonding of the iodide ion with the gold complex has also been established by single-crystal XRD analysis.⁷²

Ganter and co-workers used the ⁷⁷Se chemical shift for the assessment of the π-acceptor strength of NHC donor ligands with the easily available selenium carbene adducts. They observed a downfield shift of the ⁷⁷Se signals with increasing π-acidity of the NHC ligands.⁷⁹ Cavallo, Nolan, and co-workers have analyzed this method with quantum mechanical calculations.⁸⁰ The results confirmed a correlation between the ⁷⁷Se NMR shift of the selenium adduct and the π-accepting properties of the corresponding NHC. In a recent contribution, Sarkar and co-workers investigated the π-accepting properties of a series of 1,2,3-triazolylidene-based MIC by probing the ⁷⁷Se NMR shift of the corresponding selenium educt.⁸¹ They observed a broad spread for ⁷⁷Se NMR shifts indicating the π-accepting properties of the MICs can be tuned as a function of substituents on the triazolylidene. The bidentate donor MIC ligands seem to have enhanced π-accepting ability compared to the monodentate MICs. The chemical shift data from ⁷⁷Se NMR of the triazoline selones have also been compared with the ¹J_C–H coupling constants of the corresponding triazolium salts and the TEP of the corresponding Ir–CO complexes.⁸¹

Another method for the determination of electronic properties of carbenes has been introduced by Ganter and co-workers, suggesting the analysis of the ¹J_C–Se or ¹J_C–H coupling constants of the corresponding adducts in NMR spectroscopy to gauge the π-accepting and σ-donor abilities of carbenes.⁸² Sarkar and co-workers have recently applied the ¹J_C–H coupling constants of the triazolium salts for probing the electronic properties of these MICs. They observed really marginal differences in σ-donor abilities of the investigated MICs.⁸¹ The trend of the determined values for the analyzed NHCs follows chemical intuition. The ¹J_C–H coupling constant values indicate a σ-donor ability of the benzyl-substituted triazolylidene higher than that of the phenyl-substituted triazolylidene. This is in agreement with a stronger +I effect of the benzyl substituent compared to that with phenyl.⁸¹ Nevertheless, further research work for the validity of the method might be beneficial. The analysis of the donor abilities via this method looks more straightforward than the analysis of the HEP model complexes. The inexpensive and easy to achieve the azolium salts, which are the precursor for the free carbenes, can be investigated via time effective ¹H NMR spectroscopy.

Multidentate ligands exhibiting triazolylidenes have also been analyzed in terms of their electronic properties. As an example, rhodium carbonyl complexes bearing bistriazolylidene, mixed triazolylidene/imidazolylidene, and bisimidazolylidene ligands have been synthesized to check their electronic properties using IR spectroscopy (Figure 8).^46,83,84 No standardized procedure/model complex has been introduced to date to analyze the electronic properties of these ligand types. Comparable carbonyl stretching frequency values in IR spectroscopy have also been observed for various multidentate ligands under similar conditions. A tripodal tritriazolylidene borate ligand with an additional aryl substituent was compared to the corresponding trisimidazolylidene borate ligands featuring aryl and alkyl substituents. The nickel and manganese carbonyl complexes of these ligands were analyzed by IR spectroscopy. The average CO-stretching frequency of the triazolylidene-based ligands ranges between average frequencies of the arylated and alkylated imidazolylidene-based ligands (see discussion on anionic MICs above, Figure 3). These observations are also comparable with the electronic properties observed for monodentate triazolylidenes and imidazolylidenes showing similar TEP (∼2047 cm^–1) values for both imidazolylidene with an electron-donating alkyl substituent and arylated monotriazolylidenes.⁸⁵

Figure 8.

Rh(I) complexes possessing bidentate MIC ligands analyzed by IR spectroscopy.^46,83,84

Detailed spectroscopic and electrochemical investigations of rhenium carbonyl complexes featuring mixed triazole/triazolylidene/pyridyl ligands have been performed by Sarkar and co-workers.⁸⁶ Using IR spectroscopy, it was observed that the overall donor strength of a combined action of mixed donor ligands behaves additively. It was also demonstrated that the triazolylidenes are stronger donors than pyridine and triazoles.⁸⁶

In addition to common spectroscopic studies like TEP, HEP, and related methods in organometallic chemistry, electrochemical investigations have also been carried out using the ligand electrochemical parameter E_L to analyze the electronic properties of ligands.^68,87 The Lever electronic parameter (LEP), based on electrochemical analyses, was introduced by Alfred Beverley Philip Lever in 1990.⁸⁷ The oxidation potentials of metal complexes bearing the ligand of interest are taken as a measure for the net donor ability. Unlike the spectroscopically investigated local probes, the redox potentials are used as global probes. The oxidation must be mainly located at the metal center of the investigated complex to yield meaningful results for electron donor properties of the ligand. It is important to have reversible or at least quasi-reversible behavior of the oxidation for the accurate determination of the electrochemical potential. The mentioned study was mainly based on the redox potentials of most common type Ru^II/III complexes;⁸⁸⁻⁹⁰ however, other redox couples have also been incorporated for the mentioned purpose.⁹¹ The method was also supported with quantum mechanical calculations, and the results have also been compared with the results obtained from IR spectroscopic investigations.⁹² Furthermore, noninnocent ligands which interfere in the metal-centered redox processes are difficult to study. Therefore, E_L determination for NHCs remains very rare owing to these drawbacks.

Albrecht and co-workers demonstrated the E_L value for generic unsaturated imidazolin-2-ylidenes, using piano-stool-type carbene complexes [Fe(Cp)(CO)₂(NHC)]⁺ and [Fe(Cp)(CO)(di-NHC)]⁺, assuming insignificant N-substituent effects (Figure 9).⁹² The E_L = 0.29 was reported for imidazolin-2-ylidenes 48, which is surprisingly similar to the E_L (E_L = 0.25) value observed for pyridine. The similar E_L values for both ligands (unsaturated imidazolin-2-ylidenes and pyridine) were explained on the basis of a significant π-back-donation from the elcctron-rich iron(II) center to NHC ligands (Figure 9).⁹² Electrochemical responses recorded for the iron(II) complexes, possessing imidazolylidene and triazolylidene donor ligands, indicate a significant difference in the electron density at the metal center. The iron(II) complex 46 bearing triazolylidene donor ligand (LEP = 0.12 V, Figure 9) is oxidized at a lower potential compared to the analogous iron(II) complex 47 possessing an imidazolylidene donor ligand (LEP = 0.18 V) and, therefore, indicates an electron density at the metal center contributed by the triazolylidene donor ligand higher than that of the imidazolylidene donor (Figure 9).^93,94 Similar trends in electrochemical studies have also been documented with [Ru(bpy)₂(aNHC/NHC∧Py)]²⁺ type complexes.^95,96 The rhenium carbonyl complexes (52 and 53, Figure 9) featuring mixed triazole/triazolylidene/pyridyl donor ligands showed chelating ligand-centered reduction. The reduction potentials in these complexes have also been used to gauge the π-acceptor capacity of these ligands. Based on the available electrochemical investigations, the triazolylidene is found to be a weaker π-acceptor than a pyridyl ligand but a stronger π-acceptor than triazole.⁸⁶

Figure 9.

Electrochemical responses of imidazolylidene and triazolylidene complexes.

All of these spectroscopic and electrochemical analyses, therefore, indicate the strong donor capabilities of 1,2,3-triazolylidenes compared to Arduengo-type imidazolylidenes and CAACs-type donor ligands. However, the triazolylidenes appear as weaker donors than mesoionic imidazolylidenes. These donor properties are also consistent with the impact of heteroatoms next to the carbene center, and it is also well-established that heteroatoms generally reduce the electron density in aromatic systems. The difference in donor strength of triazolylidenes compared to Arduengo-type NHCs is estimated quantitatively using different methods, such as Tolman electronic parameter (ΔTEP = ∼7 cm^–1), the Lever electronic parameter (ΔLEP = ∼0.1 V), and by HEP. The amount of electrochemical data available related to E_L is insufficient for a detailed comparison of NHC donor ligands. The redox potential of specific metal complexes might be rather applied as a qualitative comparison of the overall donor or π-accepting properties of structurally similar ligands. Overall, detailed analysis of the electrochemical process is crucial to verify the validity of electrochemical investigations for determination of ligand electronic properties. MICs appear to be somewhat weak π-acceptors (according to the ⁷⁷Se NMR scale) than CAACs, even though the π-acceptor properties seem to be tunable. A detailed analysis using a high level theoretical study will be extremely useful in putting the π-acceptor properties of MICs with respect to NHCs or CAACs into context.

5. 3d Metal Complexes

Even though many metal complexes of 1,2,3-triazolylidenes were reported with late 4d and 5d transition metals, their use together with 3d metals have been rather sporadic and limited. Rare examples of titanium(IV) complexes bearing a 3,5-di-tert-butylphenolate-based triazolylidene donor ligand was synthesized by Hohloch and co-workers.⁶² They have also reported the first example of imido complexes bearing the 1,2,3-triazolylidene donor ligand (Scheme 9). All titanium(IV) complexes possessing MIC donor ligands have been synthesized starting from a triazolium chloride salt. The reaction of the triazolium chloride salt 54 with NEt₃ and subsequent addition of this mixture to a solution of [Ti(N^tBu)Cl₂Py₃] resulted in the formation of a mixture of the homoleptic complex 56 (main product) and the dimeric complex 55 (minor product). The homoleptic complex 56 has also been synthesized independently from the same triazolium salt at low temperatures using [TiCl₄(THF)₂] and triethylamine as a base. Switching the base from triethylamine to lithium diisopropylamide (LDA) and performing the reaction at very low temperature resulted in the clean formation of the imido complex 57 bearing the triazolylidene donor ligand.⁶² The scandium complex with related dianionic bistriazolylidene donor ligand has been demonstrated.⁶⁰

Scheme 9. Synthesis of Ti^IV Complexes Possessing 1,2,3-Triazolylidene Donor Ligands⁶².

The chromium(0) complex bearing the 1,2,3-triazolylidene donor ligand was reported by Sarkar and co-workers. The irradiation of Cr(CO)₆ with UV light in THF resulted in the formation of a solvato complex, which was further reacted with the pyridine-appended triazolium salt 58 in the presence of NEt₃ to yield the chromium(0) complex 59 (Scheme 10).⁹⁷

Scheme 10. Synthesis of Cr⁰ Complexes Possessing 1,2,3-Triazol-5-ylidene Ligands.

A manganese(I) complex 61 bearing a tris(carbene)borate donor ligand has also been reported in the literature. The complex was prepared from the corresponding tris(triazolium)borate salt 60 by deprotonation with LDA followed by subsequent addition of [Mn(CO)₃(CN^tBu)₂Br]₂ (Scheme 11). This complex is also the first example of a manganese complex containing mesoionic 1,2,3-triazolylidene donor ligand.⁵⁷

Scheme 11. Synthesis of Manganese Complexes Possessing Tris(carbene)borate Ligands.

New manganese(0) and manganese(I) complexes containing ditriazolylidene (ditrz) donor ligands have been recently described by Royo and co-workers.⁹⁸ Two different coordination modes of the ditrz donor ligands were observed based on the base-dependent metalation (Scheme 12). Manganese(I) complexes of type fac-[Mn(ditrz^R)(CO)₃Br] (63–65) bearing a chelating ditrz ligand have been synthesized via the Ag-mediated transmetalation route, whereas bimetallic manganese(0) complexes 67 and 68 of type [Mn₂(CO)₈(μ-ditrz^R)] with a bridging ditrz ligand were obtained by the in situ deprotonation of the triazolium salts with KO^tBu (Scheme 12).^98,99 A cationic fac-[Mn(ditrz^Et)(CO)₂(PPh₃)₂]Br complex 66 has been synthesized by the same research group as a rare example of a Mn^I complex bearing a mixed dicarbonyl/NHC/PPh₃ donor set. This complex was obtained from complex 64 and PPh₃ upon irradiation with visible light.⁹⁹ The electrochemical studies with the manganese(I) complexes will be discussed below in the section related to redox properties.

Scheme 12. Synthesis of Mn⁰ and Mn^I Complexes Possessing 1,2,3-Triazol-5-ylidene Ligands.

Iron is much more abundant, less costly, and forms less toxic compounds compared to heavier transition metals that are often applied in catalysis.¹⁰⁰ The application of iron complexes might improve the production and the recycling/disposal of such catalysts and photoactive materials.

The first example of an in situ generated iron complex was reported by Sarkar, Plietker, and co-workers.¹⁰¹ They reacted Bu₄N[Fe(CO)₃(NO)] (TBAFe) with a triazolium salt in the presence of a base. The in situ formed iron complexes were directly used in catalysis. No attempts were made in that work to isolate a well-defined iron complex bearing a 1,2,3-triazolylidene-type MIC donor ligands. A series of iron(II) complexes (69–73, Scheme 13) with a piano stool geometry around the metal center was synthesized by Albrecht and co-workers.⁹³ In these complexes, the iron(II) center is bound to monodentate MIC donor ligands containing different aryl and alkyl substituents. In addition, examples of iron(II) complexes 74 and 75 bearing C∧N-chelating pyridine-substituted triazolylidene donor ligands were also reported (Scheme 13).⁹³ The electrochemical, infrared spectroscopic, and X-ray diffraction analysis were used to assess the electronic and steric effect resulting from the wingtip modification.⁹³

Scheme 13. Synthesis of Iron(II) Complexes Possessing 1,2,3-Triazol-5-ylidene Ligands.

Chabera et al. reported the synthesis of an iron carbene complex 77 starting from a bis(1,2,3-triazol-5-ylidene) dibromide salt 76.¹⁰² The deprotonation of the triazolium salt 76 with potassium tert-butoxide followed by the coordination of ferrous bromide resulted in the formation of an intermediate bromide complex, which upon treatment with an aqueous solution of NH₄PF₆ yielded the iron complex 77 with iron in the formal oxidation state +III (Scheme 14). The reduced complex [Fe(btz)₃](PF₆)₂ (78) was obtained using sodium dithionite as a reducing agent in acetonitrile. A heteroleptic ferrous complex 79 containing a mixed bpy/bis(1,2,3-triazol-5-ylidene) ligand has been reported by Wärnmark, Sundström, and co-workers.¹⁰³ The complex was synthesized from the corresponding triazolium salt 76 and [Fe(bpy)Cl₂] in the presence of KO^tBu as base at −78 °C (Scheme 14).

Scheme 14. Synthesis of Iron Complexes Possessing 1,2,3-Triazol-5-ylidene Ligands.

The synthesis of iron(II) complex 23 bearing two units of pyridylene-bridged biscarbene donor ligands from the corresponding free carbene was mentioned previously (Scheme 7).⁴⁸ Similar homoleptic iron(II) complexes of type 81 possessing 4-bromophenylacetylene-substituted pyridylene-bridged di-MIC donor ligands have also been disclosed very recently (Scheme 15). These complexes have been synthesized from the corresponding triazolium salts of type 80 via deprotonation using KO^tBu in the presence of iron(II) bromide.¹⁰⁴

Scheme 15. Synthesis of Fe^II Complexes Possessing di-MIC Donor Ligands.

Melle et al. recently reported a novel diiron hydrogenase mimic complex 84 featuring a 1,2,3-triazol-5-ylidene donor ligand. The complex was synthesized by reacting the hexacarbonyl diiron complex [Fe₂(CO)₆(μ-pdt)] (pdt = propanedithiolate) with a triazolium salt 83 in the presence of potassium tert-butoxide as a base (Scheme 16).¹⁰⁵ The complex was also structurally characterized by single-crystal X-ray diffraction. The redox and potential electrocatalytic behavior of the complex were examined by cyclic voltammetry. These experiments showed that no efficient catalytic proton reduction is possible with such a complex. This observation is also similar to other monosubstituted complexes possessing imidazole-derived carbenes.¹⁰⁵

Scheme 16. Synthesis of Dinuclear Iron Complexes Possessing 1,2,3-Triazol-5-ylidene Ligands.

A rare example of a cobalt complex bearing a MIC donor ligand was described by Siewert, Sarkar, and co-workers. The cobalt(III) complex 86 with a mesoionic pyridylcarbene donor ligand was synthesized by mixing the corresponding triazolium salt 85, silver(I) oxide, and potassium chloride in dry acetonitrile followed by the addition of the chlorobridged dimeric metal precursor [Cp*CoCl₂]₂ (Scheme 17).¹⁰⁶ This complex is the first known cobalt complex with a 1,2,3-triazolylidene ligand. This Co^III complex has been used as an excellent electrocatalyst for the reduction of protons to dihydrogen in organic medium.¹⁰⁶ Recently, tetrahedral cobalt(II) complexes containing two MIC ligands and two halides were investigated for their single-ion magnet properties.¹⁰⁷

Scheme 17. Synthesis of a Co^III Complex Possessing 1,2,3-Triazol-5-ylidene Ligands.

The borate-based cobalt(II) complex 88 bearing a mesoionic 1,2,3-triazol-4-ylidene donor ligand has been demonstrated by Sarkar and co-workers very recently. The complex was synthesized from the corresponding borate-based triazolium salt 87 using the deprotonation with LDA followed by the addition of CoCl₂ (Scheme 18).⁵³ The 1,2,3-triazol-4-ylidene moieties showed a C–N isomerization reaction during the deprotonation and metalation process, and the isomerized compounds appeared as excellent ligands for Co^II centers (88 and 89). A cobalt(II) complex 89 possessing completely N∧N donor modes of the ligands was also obtained through double C–N isomerization reactions on each ligand backbone. In the ligand containing cyclohexyl substituents on the borate backbone, the presence of strong agostic interactions with the “C–H” groups of the cyclohexyl moieties resulted in the unusual low-spin square planar cobalt(II) complex.⁵³ A related complex 91 with phenyl substituents on the borate backbone displays a pseudotetrahedral coordination geometry at the Co(II) center, and this complex reacts with external substrates and can activate molecular O₂. These complexes are thus rare examples in which the geometry, the spin state, and the reactivity at the metal center are controlled through weak agnostic interactions in the secondary coordination sphere of the metal center.

Scheme 18. Synthesis of Co^II Complexes Containing Borate-Based 1,2,3-Triazol-4-ylidene Ligands.

Cobalt(III) and cobalt(II) complexes 93 and 94 bearing tripodal tris[4-(1,2,3-triazol-5-ylidene)methyl]amine MIC donor ligands have been synthesized and structurally characterized. These complexes were prepared from the corresponding triazolium salts 92 via the silver-mediated transmetalation reaction (Scheme 19).¹⁰⁸

Scheme 19. Synthesis of Co^II and Co^III Complexes with Tripodal Tris-MIC Donor Ligands.

The tris(carbene)borate-based {NiNO}^I/0 complex 96 was prepared starting from the borate-derived tris(triazolium) salt 95. The triazolium salt was initially deprotonated using 3 equiv of LDA, and the deprotonated ligand was transferred in situ to [Ni(NO)(PPh₃)₂Br] to yield the final nickel complex 96 (Scheme 20).⁵⁷ The three-fold symmetric structure of the nickel complex was established with single-crystal X-ray diffraction.⁵⁷

Scheme 20. Synthesis of a Nickel Complex with 1,2,3-Triazol-5-ylidene-Based Tris(carbene)borate Ligands.

Albrecht and co-workers synthesized and spectroscopically characterized the mononuclear nickel(II) complexes bearing triazolylidene-type MIC donor ligands (Scheme 21).⁴² The [NiCp(X)(MIC)]-type complexes (97–99) were prepared from the triazolium salt of type 7 by direct metalation with NiCp₂. Although short reaction time induced the formation of mono-MIC complexes, a long reaction time resulted in the transformation of the mono-MIC complexes to di-MIC complexes (100 and 101). Similar transformation was also observed for complex [NiCpI(MIC)], which was transformed to complexes of type [NiI₂(MIC)₂] (105 and 106) and NiCp₂ via a disproportionation reaction (Scheme 21).¹⁰⁹ Both complexes [Ni(Cp)(MIC)₂]⁺ (100 and 101) and [NiCpX(MIC)] (97–99) are efficient catalyst precursors for the Suzuki–Miyaura cross-coupling reactions between aryl bromides and phenylboronic acid. The nickel(II) complexes 103 and 104 bearing chelating pyridyl∧triazolylidene-based donor ligands have also been synthesized by Albrecht and co-workers (Scheme 21).¹¹⁰ These complexes showed appreciable catalytic performance in the hydrosilylation of aldehydes.

Scheme 21. Synthesis of Nickel Complexes with Mono- and Di-MIC Ligands.

Nickel(II)-hydride complex 109 possessing a bis(mesoionic carbene)amido pincer-type ligand was reported by Bezuidenhout, Bertrand, and co-workers (Scheme 22).⁵⁹ The complex was synthesized from the corresponding dicationic triazolium salt precursor 108. The reaction of the carbazole-based dicationic salt and nickel(II) dichlorodimethoxyethane adduct ([Ni(DME)Cl₂]) in the presence of 5 equiv of potassium hexamethyldisilazide (KHMDS) as the external base resulted in the formation of the neutral Ni-carbene-hydride complex 109 (Scheme 22).⁵⁹ The intermediate potassium salt was also synthesized and characterized independently. The distorted square planar geometry around the nickel center in the Ni-hydride complex 109 was also established by single-crystal XRD.

Scheme 22. Synthesis of a Nickel Complex with Di-MIC Ligands.

A trinuclear nickel(II) complex 111 containing a 1,3,5-triphenylbenzene-derived tris-MIC donor ligand was reported by Guisado-Barrios, Peris, and co-workers (Scheme 23).⁷⁴ The complex has been prepared from the corresponding tris(triazolium) hexafluorophosphate salt 110 by the reaction with NiCp₂ in 1,4-dioxane under heating (Scheme 23). The TEP for the tris-MIC ligand was compared with the TEP value for a related 1,3,5-triphenylbenzene-tris-NHC ligand. The 1,3,5-triphenylbenzene-derived tris-MIC ligand appears to be a stronger electron donor than the corresponding NHC-based ligands. The trinuclear nickel(II) complex was also tested as a catalyst for the addition reaction of arylboronic acids to α,β-unsaturated ketones.⁷⁴

Scheme 23. Synthesis of Trinuclear Nickel(II) Complex.

Copper complexes probably form the largest class of 3d metal complexes that have been synthesized with 1,2,3-triazolylidene ligands. The copper complexes containing 1,2,3-triazolylidene-based MIC ligands were synthesized via transmetalation, base-mediated proton abstraction, and direct metalation. The synthetic routes are also similar to those for the synthesis of gold and silver MIC complexes.

Sarkar and co-workers have synthesized [(MIC)CuI]-type complexes 112 and 115 starting from the corresponding triazolium salts of type 10. The reaction of the triazolium salt and CuI followed by the addition of KO^tBu as an external base yielded the [CuI(MIC)]-type complexes in good yields. Neutral [Cu(MIC)₂I]-type complex 114 and cationic complex 115 of type [Cu(MIC)₂](BF₄) were also reported by the same research group (Scheme 24).⁴³⁻⁴⁵ The halide-free cationic complex 115 showed that the copper center is linearly coordinated with the carbon atoms of two triazolylidene rings. The copper(I) complexes also appeared to be highly efficient catalysts for the [3 + 2] cycloaddition between azides and alkynes. Similar copper complexes 117–119 with 1,4-diphenyl, 1,4-dimesityl, and 1-(2,6-diisopropylphenyl)-4-(3,5-xylyl)-1,2,3-triazol-5-ylidene ligands were synthesized by Fukuzawa and co-workers (Scheme 24).^111,112 These complexes were prepared by the reaction of the corresponding triazolium salts with Ag₂O followed by the addition of copper chloride. The [CuCl(MIC)] complexes have also been used as efficient catalysts for click reactions of azides with alkynes to yield 1,4-substituted 1,2,3-triazoles in excellent yields. Albrecht and co-workers have also synthesized the [CuCl(MIC)] type complexes bearing 1,2,3-triazole-5-ylidene ligands by following the transmetalation route.¹¹³ The copper(I) complex 117 was also used to prepare the mesoionic compounds possessing an exocyclic oxygen.¹¹³

Scheme 24. Synthesis of Copper(I) Complexes Containing 1,2,3-Triazole-5-ylidene Ligands.

The copper(I) complexes of type 121 have also been synthesized with a H substituent at the 4-position of the 1,2,3,-triazol-5-ylidenes (Scheme 25).⁵⁸ Similar to the coordination chemistry of pyrazolate complexes, deprotonation of the 4-position of the 1,2,3,-triazol-5-ylidenes resulted in the formation of copper complex 122 possessing an anionic 1,2-dicarbene donor ligand.

Scheme 25. Synthesis of Mono- and Dicarbene Copper(I) Complexes.

Copper(I) and copper(II) complexes 123 and 124 with a monoanionic CNC-pincer ligand featuring two mesoionic carbenes have been reported by Bezuidenhout, Bertrand, and co-workers.⁵⁹ Both the copper complexes were prepared from the corresponding triazolium salt 108 by the deprotonation reaction with potassium hexamethyldisilazide (KHMDS) followed by the addition of copper(I) halide (Scheme 26).⁵⁹ The isolated unusual copper(II) MIC complex displayed a seesaw geometry owing to the peculiar electronic and steric properties of the MIC ligand. This is also a rare example of a Cu(II) center bound to a carbene-type ligand.

Scheme 26. Synthesis of Copper(I) and Copper(II) MIC Complexes.

The mononuclear copper(I) complex 25 bearing a 1,2,3-triazol-5-ylidene ligand has also been prepared by Cazin and co-workers via an inexpensive synthetic pathway, using readily available copper oxide and triazolium salt 24 (Scheme 8).⁴⁹ The complex showed outstanding catalytic activity in the [3 + 2] cycloaddition reaction of alkynes and azides.

Dinuclear copper(I) complexes 126–128 with short Cu···Cu distances (∼2.8 Å) have also been reported by Sarkar and co-workers (Scheme 27).¹¹⁴ These well-defined dicopper(I) complexes containing dimesoionic carbene donor ligands were prepared from the corresponding bistriazolium salts 125 via the transmetalation route. These complexes displayed excellent activity for the azide–alkyne cycloaddition (click) reaction owing to the cooperative action of the copper centers during catalysis. The dicationic dicopper(I) complex 130 containing a di-MIC ligand with similar Cu···Cu distance (2.801 Å) has also been synthesized by the same research group.¹¹⁵ Both Ag-mediated transmetalation and direct deprotonation methods have been successfully applied for the synthesis of the dicopper(I) complex 130 (Scheme 27).¹¹⁵ The complex was also used for the azide halo-alkyne (click) cycloaddition reaction. The calatytic activity of the dicopper complexes was also compared with the mononuclear copper(I) complexes bearing MIC donor ligands, and the dicopper(I) complex appeared to be twice as active compared to the monuclear complex.¹¹⁵ The cationic complexes showed activities superior to those of the neutral copper(I) complexes.¹¹⁵

Scheme 27. Synthesis of Dinuclear Copper(I) Complexes.

Dinuclear copper(I) complex 132 containing a pyridine-bridged bis(triazolylidene) ligand has been presented by Matsubara and co-workers (Scheme 27).¹¹⁶ The complex was synthesized via the transmetalation route. The complex was also employed for the catalytic hydroboration of styrene derivatives with bis(pinacolato)diborane for the formation of the corresponding β-selective alkylboronate esters.

As was already mentioned in the section related to the general synthesis of MICs and metal complexes, the development of the synthetic chemistry of 3d metals with MICs has been rather slow compared to those of 4d and 5d metal complexes. The most number of metal complexes with a 3d metal has been reported with Cu(I). This aspect is likely related to the fact that Cu(I)–NHC complexes were already known to be very stable, and it was rather logical to extend that chemistry to MICs. For all other 3d metals, the number of complexes known with MICs are rather limited. Several early transition 3d metals still remain unexplored. As 3d metal complexes are known to be more labile than their 4d and 5d counterparts, the development of multidentate MICs is likely to accelerate discoveries in this field. In this context, ligands such as 33–35, 94, and 108 are likely to play very important roles in the future. Additionally, MICs that are more stable in the free form will make an important contribution to the development of the synthetic chemistry of 3d metal complexes with these ligands.

6. Main Group Educts of MICs

While NHC main group adducts have been explored extensively over the past few decades, the field of triazolylidene MIC main group adducts is still in its infancy. The NHC adducts have been used as valuable ligands in transition metal chemistry.¹¹⁷ Additionally, NHCs are also known to stabilize low valent and radical main-group-element-based species.¹¹⁸ Pertinent examples include the isolation of a Ge analogue of vinylidene¹¹⁹ and the detection of transient boryl radicals.¹²⁰ Herein, the already existing research in the field of 1,2,3-triazolylidene main group adducts is summarized.

6.1. Groups 1–2 Adducts

Triazolylidene complexes with groups 1 and 2 element species are rare. A potassium adduct of a carbazole-derived monoanionic CNC-pincer ligand 133 featuring two mesoionic carbenes and a lithium adduct of a MIC containing porphyrinoide 135 have been isolated and characterized.^59,60 The potassium adduct of a CNC-pincer-type MIC ligand 133 was prepared via the triple deprotonation of the triazolium salt 108 using excess (5 equiv) potassium hexamethyldisilazide (KHMDS) to a diethyl ether suspension of the triazolium salt (Scheme 28).⁵⁹ Moreoever, a MIC Grignard adduct is proposed to be formed in situ and to catalyze Grignard allylic substitution reactions.¹²¹

Scheme 28. Synthesis of Lithium and Potassium MIC Complexes.

Lithium and magnesium complexes 137 and 138 containing carbazole-derived CNC-pincer-type bistriazolylidene ligands have also been reported recently in literature by Hohloch, Guldi, Munz, and co-workers (Scheme 29).⁶¹ The lithium complex 137 was obtained by triple deprotonation of the corresponding carbazole-based triazolium salt by LiHMDS. A transmetalation of the lithium complex with MgBr₂ led to the formation of the magnesium(II) complex 138. These complexes, possessing the trideprotonated form of the ligand, showed bright luminescence, which was investigated by absorption and luminescence spectroscopy.

Scheme 29. Synthesis of Li and Mg MIC Complexes.

6.2. Group 13 Adducts

Crudden and co-workers introduced 1,2,3-triazolylidene borane adducts of type 139 (Figure 10).¹²² The bench-stable species were synthesized by generating the free carbene and reacting it further with a dimethylsulfide adduct of trihydroborane. These adducts showed higher reactivity in the reduction reaction of carbonyl groups with an acidic additive in comparison to classical NHC-boranes.

Figure 10.

Reported triazolylidene boranes.^123−125

Sterically more encumbered 9-BBN MIC adducts 140 and chiral MIC-boranes have been synthesized.^124,125 Deprotonation of the 9-BBN MIC adducts yielded the MIC-stabilized borenium salt 141.¹²⁴ The frustrated Lewis pairs 141 (X^– = sterically encumbered borate) are highly active in hydrogenation catalysis at room temperature (Figure 10). The “unprotected” 5-hydrogen-substituted borenium salts (141, R = H) significantly exceeded the 5-arylated triazolylidenes and an imidazolylidene adduct in activity. The steric properties seem to be the determining factors here.

Dihaloborane triazolylidene adducts of type 142 were reported by Braunschweig and co-workers in analogy to the corresponding NHC-borane adducts.¹²³ The first MIC-supported diborane has also been synthesized by the same research group. In contrast, upon reduction of an MIC adduct, intramolecular C–H or C–C activation in the compound was observed instead (144 and 145, Figure 10).¹²³ Density functional theory (DFT) calculations account for borylene species as intermediates. Less sterically demanding MICs might enable the formation of triazolylidene-stabilized diborenes. Reacting in situ generated triazolylidene with dibromodimesitylborane generated the mixed sp²–sp³ diborane 143.

Recently, Sarkar and co-workers extended the scope of these MIC-borane adducts to di-MIC-diborane compounds of type 146. Suprisingly, in the case of a methylene-bridged ditriazolium salt 129, the ligand was seen to decompose, leading to the formation of a new type of triazole borane 148 containing a N–B bond (Scheme 30).¹²⁶

Scheme 30. Decomposition of MIC-Borane Adducts.

6.3. Group 14 Adducts

Although N-heterocyclic olefins (NHOs) are widely explored and have received enormous popularity as organocatalysts and ligands,^127,128 the corresponding triazolylidene-based NHOs have just recently been introduced by Hansmann and co-workers.¹²⁹ These mesoionic NHOs (mNHOs) were synthesized via deprotonation of 5-methylated triazolium salts. The pinkish/red mNHOs are highly sensitive to air but stable for several days under inert conditions in the solid state and in solution. The structure of the MIC carbene adduct can be pictured in the olefinic and the yilidic form (Scheme 31). The reactivity of the triazole-derived mesoionic N-heterocyclic olefin with various small molecules such as N₂O (154), CO, and CS₂ has also been demonstrated by the same research group.^130,131

Scheme 31. Triazolylidene-Based Mesoionic NHOs Represented in the Olefinic and the Ylidic Resonance Form.

Quantum mechanical calculations on mNHOs proposed a higher ylidic character, higher proton affinity, and stronger electron-donating abilities compared to classical NHOs but weak donor abilities compared to imidazole-5-ylidene-based NHOs (Figure 11).^129,132 These trends were confirmed through competitive experiments and determined TEP values of the various NHO types. The shift of the TEP from the NHC to the corresponding NHO falls in the range of 15–20 cm^–1 for each NHO type. This might be interpreted as the transfer of the nucleophilicity of the carbene species to the exocyclic moiety in the adduct. As the NHOs are very weak π-electron acceptors, they are stronger overall electron donors. In addition to the Rh complex for TEP determination, a borane adduct has been synthesized for the mentioned purpose. The mNHOs also appeared as active catalysts in hydroboration.¹³²

Figure 11.

Net electron donor abilities of NHCs vs mNHOs.¹³³

Extensive research on the reactivity of 5-hydrogen-substituted “unprotected” triazolylidene-based mNHOs has been undertaken by Song and co-workers.¹³⁴ Tautomerization between mNHO possessing a doubly substituted exocyclic olefinic position and the corresponding free carbene has also been reported (Scheme 32).

Scheme 32. Tautomerism of “Unprotected” mNHO into the Mesoionic Carbene Form and Their Reactivities¹³⁴.

The reactivity of a 5-hydrogen mNHO with various Lewis acids was investigated. At low temperature, the mNHO adducts are formed, which transform to the MIC adducts with increasing temperature (Scheme 33). Various MIC adducts such as the first 1,2,3-triazolylidene carbon dioxide adduct 163, a trimethylaluminum adduct 164, and a further borane adduct 165 (Scheme 33) were isolated by following the aforementioned method.¹³⁴

Scheme 33. MIC Main Group Adducts Obtained after Tautomerization of the Olefin Adduct¹³⁴.

The mNHOs are not the only reported MIC-based carbon adducts. The MIC allenylidene form has also been reported in the literature.¹³⁵ Even though this adduct could not be isolated in free form, Chen and co-workers succeeded in isolating the gold(I) complex possessing this ligand (Scheme 34).¹³⁶

Scheme 34. MIC Allenylidene and the Corresponding Gold(I) Complexes.

6.4. Group 15 Adducts

In the 1970s, 1,2,3-triazolimines were reported, which might be interpreted as the amino adducts of triazolylidenes.¹³⁵ Severin and co-workers have mainly contributed to the modern research of 1,2,3-triazolylidene adducts with group 15 elements.^137,138 They observed that the free MIC under a nitrous oxide atmosphere yielded the nitrous oxide adduct 171. The adduct was further reacted with suitable arenes in the presence of AlCl₃ to give triazolium-based azo dyes 172–174 (Scheme 35).

Scheme 35. Synthesis of MIC-Based Azo Dyes via the MIC-N₂O Adduct¹³⁸.

The dimerization of the triazolium-based azo dye 172 at the para-methyl group of the mesityl substituent was observed in the presence of KHMDS as a strong base, and diradical 175 was formed in congruence to the reactivity of imidazolium-based azo dye (Scheme 36).¹³⁷ Notably, not only the dimerization occurred but at the same time demethylation at the triazolium moiety and the subsiquent η²-coordination of potassium ion also took place. The existence of the diradical was validated through ¹H NMR and ESR spectroscopic investigations. However, no such dimerization reaction was noted for the corresponding neutral triazole azo compound. The compounds containing 1,2,3-triazole-derived fluorescent scaffolds were also synthesized by Anak, Cisnetti, and co-workers.¹³⁹ These compounds, which might be described as an intramolecular triazolylidene nitrene adducts, have been prepared via mesoionic carbene–nitrene cyclization reactions.¹³⁹

Scheme 36. Reduction of a MIC-Based Azo Dye¹³⁷.

A PS⁺ cation stabilized by 1,2,3-triazol-5-ylidene 179 was described by Yan and co-workers (Scheme 37).¹⁴⁰ The molecular structure of the carbene-stabilized PS⁺ has also been established by single-crystal X-ray diffraction analysis. Computational investigations revealed a multiple-bond character of the P–S bond.¹⁴⁰

Scheme 37. Synthesis of Carbene-Stabilized PS⁺.

6.5. Group 16

A mesoionic oxide 180 was described as possibly the first triazolylidene main group adduct by Begtrup and Pedersen (Figure 12) in 1965.¹⁴¹ It was assumed that the heterocycle was N1,N2-disubstituted,¹⁴¹ though in a follow up report, the authors corrected this hypothesis as more likely N1,N3-disubstitution.¹⁴²

Figure 12.

Chalcogen adducts of 1,2,3-triazolylidenes.

Almost 50 years later, this substance class was investigated more precisely.¹¹³ In that study, 180 was obtained from an MIC copper chloride complex by the reaction with CsOH; however, 180 could also be obtained from the triazolium salt by direct reaction with CsOH. The first MIC sulfur adducts of type 181 were reported by Begtrup and co-workers.^143,144 Later, sulfur adducts of ditriazolylidenes were attempted.¹⁴⁵ Interestingly, N-methylene-bridged triazolium salts underwent decomposition even under mild sulfurization conditions. For ditriazolium salts without a methylene bridge, the deprotonation and sulfurization were successful. The last reported triazoline chalcogen is the MIC selenium adduct of type 182 (Figure 12). This type of adduct was recently introduced by the Prabusankar and co-workers and Sarkar and co-workers.^81,145 MIC selenium and sulfur adducts were obtained with K₂CO₃ as a weak base in the presence of elemental chalcogen (Scheme 38). These adducts were used as ligands for the synthesis of zinc complexes, which proved to be active catalysts in thioetherification. The sterically controlled oxidation of the triazoline selones 183 in the presence of a copper(II) salt yielded MIC-stabilized di- and tetraselenides (Scheme 38).¹⁴⁶ The reaction product is probably dependent on the steric bulk of the MIC substituents. The MIC selenium adducts were also investigated via ⁷⁷Se NMR spectroscopy to gauge the π-acceptor properties of the corresponding triazolylidenes (see section on donor/acceptor properties above).^79,81 Furthermore, the electrochemical properties of the MIC selenium adducts were investigated, and it was shown that, with appropriate substituents, reduction of these adducts, which are triazolylidene-based, is reversible. The reduced compounds were characterized via UV–vis–NIR and EPR spectroscopy. These are the first examples of the electrochemical and spectroscopic properties of 1,2,3-triazol-5-ylidene-based radicals.¹⁴⁷

Scheme 38. Oxidation of 1,2,3-Triazolin-5-selones to Di- and Tetraselenides¹⁴⁶.

Even though the field of 1,2,3-triazolylidene main group adducts is relatively unexplored, the already existing reports displayed a promising preview on what might be revealed in future research. The adducts and their metal complexes, in some cases, showed catalytic activities superior to those of the classical imidazolylidene-based NHC main group adducts. The availability of the “unprotected” 5-hydrogenated 1,2,3-triazolylidene main group adducts, unlike for imidazolylidene-based NHCs, enables intriguing and unique reactivities. The mNHOs have already shown very high promise in terms of follow-up reactivity. Several properties of the edducts with nitrogen have not yet been explored (cis–trans isomerism of the azo compounds containing MICs). The S- and Se-MIC compounds are likely to be highly useful ligands in their own right. Several combinations of MICs with other main group components are yet to be synthetically attempted and are expected to deliver various new types of chemistry.

7. Redox, Photochemical/Photophysical Properties of Metal Complexes

The redox and photochemical/photophysical properties of metal complexes are usually dependent on their electronic structure. The right choice of the metal center in combination with tailored ligand design can provide metal complexes with intriguing redox activity and photoactivity. Nitrogen-coordinating ligands have already been well-documented to be good candidates to form redox-active and photoactive metal complexes.¹⁴⁸ In recent years, NHC ligands, and among them MICs, have been successfully established in photo- and redox-active metal complexes.¹⁴⁹ MICs can influence the electronic structure of the complex significantly, owing to their strong overall electron donor properties and a tunable π-acceptor property. The modular synthesis of the MICs leads to a wide range of choice of substituents. This enables fine-tuning of the electronic properties and the introduction of additional coordinating substituents. Therefore, this subsection aims to provide an overview on the whole range of MIC metal complexes investigated toward their redox and photochemical/photophysical properties.

7.1. Groups 1–7 Metal Complexes

There have been only a few groups 1–7 metal complexes with MIC ligands investigated toward their redox activity and photoactivity. The lithium and magnesium complexes 137 and 138 (Scheme 29) containing CNC pincer ligand featuring carbazole-based bis-MIC donors show bright luminescence.⁶¹ Their photophysical properties have been investigated by absorption and luminescence spectroscopy. The lithium and magnesium centers prevent the distortion of the ligand following excitation, which supports bright luminescence by avoiding nonradiative deactivation.⁶¹ A MIC-containing macrocyclic scandium complex exhibits fluorescence.⁶⁰ As the corresponding triazolium ligand displays similar fluorescent behavior, the excited and emissive states are likely ligand-centered. A titanium(IV) di(MIC-bisphenolate) complex 56 displayed two oxidations with the typical characteristics of a reversible process in cyclic voltammetry (Figure 13).⁶² The reversible nature of the oxidation makes the titanium complex a potential candidate for redox-switchable catalysis. A comparable complex with an imidazolylidene-based ligand displayed a similar oxidation process but with a significant cathodic shift of about 0.3 V, indicating a significant influence of the core in the tridentate ligand on the redox behavior. This observation, together with the observed inertness of titanium(IV) toward oxidation, suggests a ligand-centered oxidation.⁶²

Figure 13.

Groups 4–7 metal–triazolylidene complexes investigated for redox and photochemical properties.^62,86,97,150

Sarkar, Gerhards, and co-workers reported on chromium(0) and molybdenum(0) carbonyl complexes (59 and 187) with bidentate pyridyl-MICs (py-MICs), which were analyzed through CV, IR, UV–vis–NIR, and EPR spectroelectrochemical and time-dependent (TD)-DFT methods (Figure 13).⁹⁷ While the oxidation of the molybdenum complex is irreversible, the oxidation of the chromium complex is completely reversible under the applied electrochemical conditions. The oxidation was unambiguously assigned at the metal center, making the py-MIC chromium complex a rare example of a chromium(0) carbonyl complex, displaying a reversible oxidation. Through the strong electron-donating properties of the ligand, the formal charge at the chromium(I) center can be compensated. Both the metal carbonyl complexes (Cr, Mo) showed reversible ligand-centered reduction processes. The radical anion generated upon reduction is almost exclusively located at the py-MIC moiety. This is the first time a radical-stabilizing ability could unambiguously be attributed to a MIC-containing chelating ligand of 1,2,3-triazol-5-ylidene type.⁹⁷

In addition, photophysical studies were performed on the chromium(0) and molybdenum(0) complexes. Time-resolved Fourier transform infrared studies revealed a metastable photoproduct, which was formed upon excitation, and this product completely reverted back to the starting material via a slow dark reverse reaction.⁹⁷ In contrast to homoleptic molybdenum carbonyl complexes,^151,152 no CO dissociation was observed for the molybdenum(0) complexes bearing the Py∧MIC donor ligand. A reversible breakup of the pyridyl moiety of the MIC ligand to the metal center was suggested as a plausible mechanism for the observed reversibility.⁹⁷ More recently, the same groups have shown that the aforementioned Cr(0) and Mo(0) and the analogous W(0) complexes are dual emitters in the solid state at room temperatures.¹⁵³ These complexes emit in the visible as well as in the NIR II region. The quantum yield, particularly for the W(0) complex, is remarkably high. These are also the first examples of Cr(0), Mo(0), and W(0) complexes that emit at room temperature in the NIR II region. These examples show that MIC ligands are well-suited for generating NIR emitters with earth-abundant metals.

The rhenium carbonyl chloride complexes bearing di-MIC 52 or N_Py∧C_MIC53 ligands have been studied (spectro-)electrochemically and theoretically in a similar manner as in the aforementioned study (Figure 13).^86,150 The di-MIC ligand assisted in achieving a reversible oxidation process predominantly at the Re(I) center. A N_Py∧C_MIC rhenium complex with an N1-Dipp substituent displayed an EC mechanism upon reduction. The follow-up reaction was assigned to halide dissociation and solvent coordination. Overall, the process appeared to be chemically reversible owing to halide recoordination. The reversibility of the chemical reduction of those complexes is heavily dependent on the MIC substituents. An N1-perfluorobenzene substituent led to limited chemical reversibility, whereas an N1-benzyl substitution yielded a completely irreversible reduction process. In DMF, photoluminescence at about 550 nm was observed for two of the complexes. With an emission lifetime of τ = 56 ns, the Dipp-substituted complex displayed a lifetime longer than that of the bipyridine (bpy)-ligated counterpart. These complexes are also active in electrocatalytic CO₂ reduction with high selectivity for the formation of CO.¹⁵⁰

7.2. Group 8 Metal Complexes

The design of photoactive iron complexes is of great interest in chemical research, owing to its high abundance, low costs, and comparatively low toxicity.¹⁰⁰ The application of iron complexes would improve the production and the recycling/disposal of such photoactive materials and catalysts. The main challenge is the generation of iron complexes with long-lived excited states. The first excited states are typically MLCT-based for iron, and the ligand field splitting is relatively weak as for other 3d metal complexes. Accordingly, energetically low-lying metal-centered states usually lead to a fast, nonradiative deactivation to the ground state and very short lifetimes of the excited state. For example, the exited-state lifetime of [Fe^II(bpy)₃]²⁺ is τ = 130 fs. Therefore, the report by Wärnmark and co-workers of an Fe(II) complex 79 with two di-MIC ligands and one 2,2′-bpy, which showed an excited state lifetime of τ = 13 ps, marked a remarkable improvement in this field (Figure 14).¹⁰³ Both photo and redox behavior were investigated in detail by electrochemical and spectroscopic methods and quantum mechanical calculations. Cyclic voltammetric investigations of that complex displayed an anodic shift of the metal-centered reversible oxidation and an irreversible ligand-centered reduction compared to [Fe^II(bpy)₃]²⁺. As the anodic shift for the reduction is lower than that for the oxidation process, overall lower MLCT transition energy is proposed. The comparatively long excited-state lifetime of the complex could be assigned to a destabilization of the ³MC and ⁵MC states, therefore preventing fast deactivation to the ground state. A related [Fe(btz)₃]³⁺ (where btz is 3,3-dimethyl-1,1-bis(p-tolyl)-4,4-bis(1,2,3-triazol-5-ylidene)-type complex 77 exhibits an excited-state lifetime of 100 ps (Figure 14).¹⁰² This lifetime range facilitates the use of this complex as a photosensitizer. Mössbauer and DFT studies showed that the iron center is best described as Fe(III). Reversible reduction of the metal center was identified through EPR and NMR spectroscopic methods. The investigated complex was reported as the first luminescent iron complex bearing a triazolylidene donor ligand. 2D excitation–emission experiments, transient absorption spectroscopy, and time-resolved photoluminescence measurements and DFT studies indicated that emission occurs from a long-lived doublet ligand-to-metal charge-transfer (²LMCT) state instead of an expected MLCT state.¹⁰² Very recently, an Fe(II) complex 81 with two tridentate ligands, each possessing one pyridyl and two MIC donor, was investigated photophysically. In this complex, destabilized metal-centered states resulted in prolonged lifetimes of charge-transfer excited states.¹⁰⁴

Figure 14.

Emissive MIC iron complexes.^102,103

Apart from these photophysical investigations, the redox properties of different Fe(II) complexes possessing triazolylidene donor ligands have also been analyzed. An Fe(II) MIC piano stool complex exhibits reversible oxidation, as judged from cyclic voltammetry (CV).⁹³ Analysis of the iron complexes with varying MICs showed that the oxidation potential is not significantly influenced by the MIC substituents. A hydrogenase-mimicking MIC Fe(II) complex displaying similar redox behavior appeared inactive in the electrocatalytic hydrogen evolution reaction.¹⁰⁵

A wider variety of reports is available on MIC ruthenium(II) complexes and their redox activity and photoactivity.^154−156 Some of the investigated Ru(II) complexes with a pyridyl-MIC ligand are active electrocatalysts in water oxidation.¹⁵⁷ Additionally, a MIC ruthenium(0) carbonyl complex bearing a cyclopentadienone ligand displays a quasi-reversible reduction, as observed from CV.¹⁵⁸ This oxidation was proposed to be metal-centered. The complex also appeared as an active precatalyst in oxidation catalysis.¹⁵⁸

Ruthenium(II) complex 189 of type [Ru(tpy)(di-MIC-py)]²⁺ bearing a click-derived MIC pincer ligand was reported in the literature together with their photophysical investigations (Figure 15). The complex [Ru(tpy)(di-MIC-py)]²⁺ showed an excited-state lifetime of τ > 600 ns; this value is higher by orders of magnitude (2500 times longer) than the value observed for [Ru(tpy)₂]²⁺ and is also comparable to that of [Ru(bpy)₃]²⁺.¹⁵⁹ Upon excitation, the longest wavelength ¹MLCT absorption is associated with the terpyridine (tpy) ligand, whereas the ³MLCT emission originates from the MIC ligand owing to the redistribution of electron density during vibrational relaxation and intersystem crossing.¹⁵⁹ Fine tuning of ligand electronic properties such as an electron-donating substituent at the terpyridine ligand and electron-withdrawing substituents at the MIC-containing ligand led to improved lifetimes up to 7.9 μs.¹⁶⁰ The observed lifetimes reported are the highest documented lifetimes for mononuclear ruthenium(II) complexes, and these values measured 4 orders in magnitude higher than that the value observed for [Ru(tpy)₂]²⁺.¹⁶⁰ This observation is associated with the strongly σ-donating and weakly π-accepting nature of the 2,6-bis(1-mesityl-3-methyl-1,2,3-triazol-4-yl-5-idene)pyridine donor ligand, which resides adjacent to the tpy and, therefore, generates a large energy separation between the MLCT and ligand field states along with maintaining a large ³MLCT energy. Electrochemical, spectroscopic, and DFT-based methods were applied in these studies. Similar ruthenium(II) complexes containing 1,2,3-triazolide and 1,2,3-triazolylidene units featuring long alkyl chain substituents at the MIC displayed photophysical properties similar to those of previously described ruthenium(II) complexes of type 189 (Figure 15).¹⁶¹ It has to be noted that these measurements were performed under ambient conditions. The complexes were implemented into a dye-sensitized solar cell, but only weak performance was observed. The ruthenium complexes possessing bidentate cyclometalated MIC ligands exhibited less spectacular photophysical properties, which have also been reported.^95,162 Detailed redox and photophysical investigations on [Ru(di-MIC/py-MIC)(bpy)₂]²⁺ and [Os(di-MIC/py-MIC)(bpy)₂]²⁺-type complexes with the Latimer diagrams have also been reported.¹⁶³ These complexes exhibit excited-state lifetimes up to 300 ns and quantum yields of up to 1.43%. Natural transition orbital analysis of these ruthenium complexes showed that the excited state is predominantly Ru-bpy centered with a small contribution from the MIC moiety. Through the choice of the donor ligands, the MLCT states can be influenced significantly.

Figure 15.

MIC-pyridine-MIC ligand-containing ruthenium(II) complexes.^159−161

7.3. Group 9 Metal Complexes

Only one report on electrochemically investigated MIC cobalt complex (86; Scheme 17) is known.¹⁰⁶ The complex displays two reductions in CV. The first reduction is proposed to be metal-centered and led to halide dissociation. The second reduction is suggested to be ligand-centered. These cobalt complexes are found to be active in electrocatalytic dihydrogen production.¹⁰⁶

The analogous iridium MIC complexes are well-established. Iridium complexes are popular and show high performance for redox or photochemical and photophysical applications, similar to the previously discussed ruthenium complexes.^164,165 CV-based investigations of iridium(III) complexes containing MIC donor ligands displayed mainly quasi-reversible or irreversible redox processes.^74,166,167 Some of the reported MIC iridium complexes are also active as water oxidation catalysts in a photoelectrode setup or in an electrochemical cell.^168,169

The photophysical investigations on iridium(III) complexes with triazolylidene ligands were reported in 2016 by Sambri and co-workers.¹⁷⁰ They analyzed cationic Ir(III) complexes of type 193 containing two pyridyl MICs and two chloride ligands or a bistetrazole ligand in a distorted octahedral environment (Figure 16). According to spectroscopic measurements and computational calculations, the first excited state displays MLCT character. The emissive state is an LC state with quantum yields of 1.4–11.7%. Quantum chemical calculations proposed that the use of cyclometalating phenylpyridine (ppy) ligands would enhance the photochemical properties.¹⁷¹ Sambri and co-workers implemented those ancillary ligands in their MIC Ir(III) complexes. It was observed that only one pyridyl/triazole/triazolide MIC and two phenylpyridine ligands were coordinated to iridium(III).¹⁷² However, the investigated complexes did not exhibit the proposed improvement of photophysical properties, as shorter excited-state lifetimes and similar quantum yields were observed for these complexes. A possible explanation might be due to the incorporation of only one MIC ligand. The incorporation of a triazolide MIC ligand (Figure 16) increased the excited-state lifetime and quantum yield significantly. Quantum mechanical investigations were supported by electrochemical and spectroscopic measurements, which identified the emitting states as ³MLCT and ³LC states. In the neutral triazolide MIC iridium complex, the deactivation pathway over non-emissive ³MC states appears to be unavailable. It is proposed that the deactivation observed for the complexes with a triazole/py-MIC ligand is prone to dissociation of the N-coordination site, whereas the comparatively stronger C_Trz–iridium bond of the triazolide MIC complex is less prone to dissociation.¹⁷²

Figure 16.

Selected MIC Ir(III) complexes exhibiting photoluminescence.^170,172

Similar iridium complexes with a di-MIC/NHC ligand bearing an alkylene bridge complemented by two ppy ligands were introduced.¹⁷³ These complexes exhibited photoluminescent quantum yields of up to 57% in acetonitrile. First, in vitro studies indicate the potential of these complexes as luminescent probes in biological systems. Different iridium(III) complexes with less detailed studies on luminescence have also been reported.^174,175 An iridium(III) complex possessing a luminescent organic unit at the MIC was also presented.¹⁷⁶ Ferrocenyl-substituted MICs have been implemented for the redox chemical study of the corresponding rhodium complexes. Electrochemically reversible oxidation of the ferrocenyl unit and rhodium-centered irreversible or quasi-reversible oxidation were observed for these complexes.⁴⁷

7.4. Group 10 Metal Complexes

Nickel(II) complexes bearing a triazolylidene ligand have been investigated electrochemically (Figure 17). The trinuclear Ni^II complex 111 with an extended aromatic core exhibited one quasi-reversible reduction, as judged from the cyclic voltammetry experiment. Only one oxidation process was observed in a differential pulse voltammetry experiment.⁷⁴ As the oxidation was proposed to be metal-centered, the metal centers are decoupled in this case. Di- and trinuclear palladium(II) PEPPSI-type complexes 195 and 196 featuring phenylene-bridged bis- and tris-MIC ligands displayed two and three irreversible reduction processes, respectively (Figure 17).¹⁷⁷ Based on previous literature reports, these reduction processes were assigned to be metal-centered. Consequently, stepwise reduction indicated the electrochemical coupling between palladium centers.¹⁷⁷ Different palladium(II) complexes containing MIC donor ligands have also been studied via cyclic voltammetry experiments.^178,179

Figure 17.

Multimetallic nickel(II) and palladium(II) complexes studied by CV.

Photophysical analysis has been carried out with the MIC palladium complex, which revealed photoluminescence with a quantum yield of 74% in THF.¹⁸⁰ The luminescent properties are unaffected over a broad pH range. The complexes were used as fluorescent probes in living cells. Moreover, palladium(II) complexes with a MIC ligand linked to a fluorescent organic probe have also been introduced.³⁶

Strassner and co-workers extensively studied the photophysical properties of cyclometalated phenyl-based MIC Pt^II complexes with β-diketonate auxiliary ligands (Figure 18).^181−183 These complexes display photoluminescence in polymer films with quantum yields of up to 84% and excited-state lifetimes of up to 17.9 μs. Electrochemical investigations of complexes of this type exhibit a reduction wave that accounts for a reversible or quasi-reversible process. As the potential significantly depends on the MIC substituents, the reduction is therefore assigned to be ligand-centered. In a follow-up study, a substantially red-shifted emission maximum for complexes with a phenyl-substituted MIC ligand was observed and compared to the methyl-substituted MIC ligand.¹⁸² To rationalize this effect, in-depth studies were performed. DFT-based investigations of the excited triplet-state predicted coplanarity of the phenyl ring and the MIC donor planes. The resulting enlargement of the delocalized π-system is proposed to stabilize the excited state, yielding less energetic emission. To verify this hypothesis, the phenyl substituent was replaced with a bulky mesityl group to hinder the free rotation, and the resulting complex exhibited only a small bathochromic shift of the emission compared to the methyl-substituted counterpart.

Figure 18.

Pt^II complexes with triazolylidene ligands as phosphorescent emitters.^181−183

Cationic platinum(II) complexes of type 200 featuring a pyridyl-based MIC donor with various auxiliary ligands (Cl^–/CN^–/MeCN) showed a quasi-reversible oxidation process (Figure 18).¹⁸⁴ The calculated HOMO orbital dimensions depend on the ancillary ligand; therefore, for the chlorinated complex, the HOMO level was found to be delocalized over the pincer ligand, the metal, and the chloride. For the other complexes, a tridentate ligand-centered HOMO orbital was observed. Based on this, an oxidation delocalized over the whole complex is proposed for the first complex, and oxidation processes for the other complexes might be considered as predominantly ligand-centered. The complexes are luminescent with a lifetime in the microsecond range when measured in dichloromethane. TD-DFT investigations assigned excitation to a ¹MLCT state and emission from a ³MLCT state. Those complexes displayed a vapochromic response upon exposure to THF with a shift of the emission from green to sky-blue light. Luminescent platinum(IV) complexes combined with cyclometalated bi- or tridentate 2,6-diarylpyridines and a monodentate MIC or a cyclometalated MIC were synthesized and analyzed.^185,186 The DFT methods applied with these complexes indicated that they emit from a ³LC state, which involves the diarylpyridine ligand.¹⁸⁵ The incorporation of MIC ligands leads to a larger ligand field splitting. Consequently, the energy level of the deactivating LMCT states is increased, and nonradiative decay is less favored.

7.5. Group 11 Metal Complexes

A variety of redox-active MIC gold complexes have been reported by Sarkar and co-workers (Figure 19).^52,54,55 The redox-active unit is a ferrocenyl substituent at the MIC ligands. Those complexes can be oxidized reversibly. Complexes with two ferrocenyl units exhibit two separated reversible oxidation processes.⁵⁴ Investigations via spectroelectrochemical and TD-DFT methods assigned both oxidation processes to be ferrocenyl-centered. A mixed-valent complex is generated upon one-electron oxidation. The reduction of the complexes is irreversible. Measurements at low temperature (−40 °C) resulted in a small reoxidation wave.⁵⁴ Those metallocene-based MIC gold complexes were applied in redox-induced and redox-switchable catalysis with ferrocenyl as the redox-switchable unit.^52,55 Similar complexes introduced by Sarkar, Bildstein, and co-workers possessed a cobaltoceniumyl substituent instead of ferrocenyl moiety. These complexes displayed cobaltoceniumyl-centered reversible reductions.^56,187 The most recent addition to the field of redox properties of Au(I) complexes with MIC ligands is the report of a completely reversible reduction step for a complex 208 of type [(MIC)AuPh] (Figure 19).¹⁴⁷ By performing extensive electrochemical, UV–vis–NIR/EPR spectroelectrochemical and DFT studies, Sarkar and co-workers reported the first example of a metal center bound to a triazolylidene-based radical anion. Judicious choice of substituents on the MIC ligand was a necessity for achieving a reversible reduction process at room temperature. The reversible nature of this reduction process will likely open up new avenues for these complexes in the field of redox-switchable catalysis.

Figure 19.

Redox-active gold complexes containing triazolylidene ligands.^52,54,56

Furthermore, coinage metal complexes were analyzed via photophysical studies. Photoreactivity of gold complexes containing triazolylidene donor ligands was studied.¹⁸⁸ A conversion of the gold(I) complex 210 to well-defined gold(III) complex 211 was observed upon exposure of visible light via a disproportionation reaction (Scheme 39). Cationic homoleptic mono- and digold MIC complexes were investigated photophysically.¹⁸⁹ Spectroscopic analysis revealed intraligand π–π* transitions and MLCT transitions upon excitation, which was further supported by TD-DFT calculations. Two emissive states were proposed, yielding parallel phosphorescent and fluorescent emissions. Digold complexes possessing di-MIC/NHCs, featuring an alkylene bridge, exhibit luminescence with a lifetime in the microsecond range when measured in the solid state.^190,191 The observed lower emission energy was correlated with a lower Au–Au distance in the complexes. The d¹⁰–d¹⁰ interactions are also supposed to stabilize the triplet emissive state. Inspired by carbazolate coinage metal complexes, which exhibit highly potent luminescent properties,^192,193 pincer-type MIC-carbazolate-MIC coinage metal complexes have been synthesized and analyzed.¹⁹⁴ The lifetime of the excited state was found in the microsecond range even in solution state, and a photoluminescent quantum yield of 5–8% in acetonitrile was also observed. Through spectroscopic measurements and TD-DFT calculations, the emission was assigned to be a ³ILCT transition from the π-orbital of the carbazolate unit into a π*-orbital of the MIC. Significant longer excited-state lifetimes at lower temperature might be a sign for ³IL emissive states or thermally assisted delayed fluorescence. Similar excitation and emission properties were found for MIC Cu(I) and Au(I) carbazolate complexes.¹⁹⁵

Scheme 39. Photoinduced Synthesis of Au^III from Au^I Complexes.

Overall, MIC ligands are valuable ligands for the generation of photo- and redox-active metal complexes. Especially their strong electron donor properties, possibly coupled with relatively weak π-accepting properties, enhance the stability of the excited states or destabilize states that lead to nonradiative decay. The effect of MICs on influencing the redox properties (in particular, the oxidation step) is now fairly well-understood. Their ability to stabilize metal centers in otherwise “unsually” high oxidation states has been documented. The new addition of the spectroscopic characterization of a reduced MIC ligand opens up new avenues in this field. It will be interesting to see if such reduced MICs, either in the free or in the metal-bound form, can be isolated and crystallographically characterized. Furthermore, they should have the ability to act as electron reservoirs and participate in redox catalysis and redox-switchable catalysis. One of the biggest breakthroughs for MIC ligands has been in the photochemistry/photophysics of their metal complexes. The observation of emissive compounds with groups 1 and 2 metals such as Li and Mg is unique and might open up completely new aspects in the photochemistry/photophysics of these metal centers. Similarly, the generation of NIR emitters based on Cr(0), Mo(0), and W(0) carbonyl complexes is exceptional, as such complexes are expected to be photolabile owing to CO loss. Furthermore, the discovery of the emissive Fe(III) complex with a bi-MIC ligand is revolutionary because of many different reasons (first emissive Fe complex, emission from a ²LMCT state, and so on). However, all of these observations from a photochemistry/photophysics point of view are rather isolated observations (exceptions are the properties of Ru(II), Ir(III), Pt(II), and Au(I) complexes which have been systematically investigated). A structure–function correlation type of study is still missing in all of those examples. Additionally, the combination of MICs with as yet unexplored elements will likely deliver compounds with further pathbreaking photochemical/photophysical properties.

8. Small-Molecule Activation, Electrocatalysis, and Redox-Switchable Catalysis

Homogeneous catalysis is the field where compounds containing MIC ligands have found the most applications until now. As this topic has been reviewed several times in the recent past, we present here an overview on three specific cases of bond activation and catalysis, namely, small-molecule activation, electrocatalysis, and redox-switchable catalysis.

A cobalt(III) complex bearing pyridine-derived triazolylidene donor ligand 86 appeared as an excellent electrocatalyst for the reduction of protons to dihydrogen in organic media (Scheme 40). The complex showed the lowest overpotential (130 mV) reported for molecular catalysts used for dihydrogen production along with a reasonably high turnover frequency.¹⁰⁶ Cobalt complex 86 had unique stability toward acids, and the robustness of the Co–C_MIC bond and the presence of free and basic N atoms in the ligand backbone are assumed to be responsible for the high efficiency of the catalyst. Although the detailed mechanism of this process is still unclear, initial studies showed the involvement of the triazolylidene ligand as an electron reservoir,¹⁰⁶ therefore, pointing to its potentially noninnocent nature.^196−198 The corresponding complex 212 with pyridine-based triazole donor ligand was found to be inactive for the dihydrogen production under identical conditions. This observation highlights the importance of the MIC donor for generating efficient electrocatalysts for dihydrogen production.

Scheme 40. Cobalt Triazolylidene Complex for Dihydrogen Production.

Complexes 214 and 215 bearing cyclometalated MIC donor ligands, as well as the complexes type 213 and 216, appeared as active precatalysts for water oxidation in the presence of cerium ammonium nitrate (CAN) as an oxidizing agent (Figure 20).^{168,169,199−201} Evidence was presented that catalysis functions in a homogeneous fashion. It has also been observed that the substituents on the triazolylidene moiety show a strong influence on the catalytic activity of these complexes.²⁰¹ The potential ability of triazolylidene ligands as proton and electron reservoirs was found to be responsible for the observed high catalytic activity involved with such water oxidation, which required multiple proton and electron transfers.¹⁹⁹ Even though these precatalysts showed very high activity for the water oxidation reaction in the presence of CAN, electrocatalysis, while maintaining high catalytic activity, has remained elusive with these complexes.

Figure 20.

Selection of iridium(III) triazolylidene complexes.

Bimetallic iridium(III) complexes possessing ditopic triazolylidene ligands were synthesized and evaluated as water oxidation catalysts; these bimetallic showed maximum turnover frequency even using low catalyst loadings compared to the analogous mononuclear complexes (Figure 20).²⁰²

The rhodium(I) complexes of type 220 containing chelating bis(tzNHC) ligands have been employed as catalysts for the reductive formylation of amines using CO₂ and hydrosilanes (Scheme 41). These complexes were identified to be highly effective with excellent catalytic activity at a low catalyst loading under ambient temperature. The best active complex showed excellent activity over a wide range of substrates, including amines containing reducible functional groups.²⁰³ A rhodium(I) complex containing a pincer-type ligand with MIC donors was found to activate molecular O₂ and to form a stable Rh–O₂ complex. This complex was shown to act as a catalyst for the conversion of various alkyne-containing substrates.²⁰⁴

Scheme 41. Rh^I Complexes Employed in Catalysis.

Copper(I) complexes of type 221 efficiently catalyzed the direct C–H carboxylations of benzoxazole and benzothiazole in the presence of CO₂ followed by the addition of alkyl halide to give the corresponding esters (Scheme 42).¹¹²

Scheme 42. CO₂ Activation by Copper(I) and Re^I Triazolylidene Complexes.

Rhenium(I) complexes of type 224 containing a pyridyl MIC ligand was found to be an efficient electrocatalyst for CO₂ reduction (Scheme 42).¹⁵⁰ With the appropriate choice of substituent (Dipp) on the 1,2,3-triazolylidene ring, the corresponding complex was found to generate CO electrocatalytically from CO₂ in near quantitative Faradaic efficiency. This catalyst was found to perform better for CO₂ reduction compared to the Lehn-type catalyst under identical conditions. Mechanistic studies with IR spectroelectrochemistry were performed, and these studies showed the dissociation of chloride from the starting rhenium complex, as well as the formation of relevant intermediates during electrocatalysis. A comparison of catalysts with various substituents on the triazolylidene ring displayed that benzyl-type substituents lead to catalysts that are fragile under electrocatalytic conditions and thus deliver poorer conversion of CO₂ to CO.

First examples of redox-switchable Au(I) catalysis have recently been reported with Au(I) complexes (203, 205, 225, and 226) containing MIC ligands (Scheme 43).^52,54,55 In traditional Au(I) catalysis, usually a Ag(I) salt is used as an additive. The function of the additive is to abstract the chloride ligand from the Au(I) center, thus increasing the Lewis acidity at Au(I). Instead of using such an additive, an alternative strategy is to increase the Lewis acidity at the Au(I) center by performing an oxidation step on the backbone of ligands containing a redox-active unit. This strategy was successfully applied on Au(I) complexes that contain a MIC donor with ferrocenyl substituents. The donor properties of such MIC ligands can be tuned by simple redox processes. It was shown that, on performing oxidation steps on the ferrocenyl units, the TEP can be tuned over more than 10 units for each oxidation step. Thus, the electron-rich MICs can be converted to rather electron-poor donors “on demand”. As expected, the electron-poor version of the ligands imparts high Lewis acidity at the Au(I) center and make such complexes very active catalysts for the conversion of amides to oxazolines (Scheme 43). Furthermore, as the donor properties of the MIC ligands can be tuned by redox processes, the catalytic activity of the Au(I) centers can also be tuned by redox processes. It was shown that the catalytic activity of such complexes can be tuned on and off by oxidizing or reducing the ferocenyl units.

Scheme 43. Cyclization of Oxazole with Oxidized Heterobimetallic [Fc-Au-MIC]⁺-Type Complexes.

More recently, the first example of a dicationic Au(I) complex with extremely electron-poor carbene ligand based on a MIC donor with two cobaltoceniumyl substituents 227 was reported.⁵⁶ These ligands are the most electron-poor carbenes reported to date with TEP values that fall in the range of those of PF₃. The Au(I) complex 227 with this electron-poor MIC features a highly Lewis acidic Au(I) center. This complex catalyzes the cyclization of substituted amides to oxazoles at very fast rates and without the need of any additives whatsoever. Despite the very high activity of these catalysts, the systems were unfortunately not redox-switchable. Most likely, the cobaltocene substituents that are generated on reduction of the compounds are labile under the catalytic conditions, thus leading to catalyst decomposition on reduction.

Interestingly, the aforementioned Au(I) complexes could be oxidized to Au(III) using aqua regia (!) as an oxidizing agent.¹⁸⁷ The Au(III) complex was shown to deliver a different isomer of the oxazoline on its reaction with the substituted amide mentioned above. Thus, the Au(I) complex with the same ligand delivers one isomer in the catalytic reaction, whereas the Au(III) complex delivers a different isomer from the same substrate. Both the catalytic processes work under additive-free conditions. Just like with the Au(I) complex, catalysis with the Au(III) complex containing the MIC donor with the redox-active cobaltoceniumyl substituents is also not redox-switchable, likely due to the same reasons mentioned above.

Metal complexes of MICs have been successfully applied for the activation of small molecules such as H₂O, CO₂, and O₂. H₂O oxidation has primarily been carried out with Ir complexes of MICs. Even though the activity of such complexes as catalysts for water oxidation is very high, they suffer from several drawbacks. The exceptional activity in chemical water oxidation (with CAN for example) unfortunately does not translate into electrochemical water oxidation. Additionally, iridium is a very rare and a rather expensive metal. Both of the aforementioned aspects make the use of these complexes for a large-scale technological process such as water oxidation rather impractical. O₂ reduction has only been investigated rather sporadically (and never in an electrocatalytic way). This is field where complexes of MIC ligands can be potentially useful in the future. Metal complexes of MIC ligands have, in general, shown exceptional activity in reductive electrocatalysis such as CO₂ reduction and H₂ production. This is perhaps a combination of the rather robust nature of the catalysts under reductive conditions (strong M–C bonds) and the still low overpotential somewhat comparable to that of the 2,2′-bpy systems. Electrocatalysis is also a field where much work in terms of structure–activity correlation and the development of other novel MIC-based electrocatalysts still remain to be done. MIC ligands have been pioneered in the field of redox-switchable catalysis with Au(I) complexes. These concepts will likely be transferable to a host of other metal complexes and other catalytic processes. Generation of redox-switchable and orthogonal catalysis will be a very important field of interest in the future. Additionally, the use of MICs themselves as direct electron reservoirs for reductive electrocatalysis will be a field worth exploring in the near future.

9. Conclusions and Future Perspectives

In about 13 years since the first report on the metal complexes of triazolylidene-type MIC donors, the field has seen a massive growth. While late transition metal chemistry has largely dominated the field until now, compounds of MICs with first-row transition metals or with main group elements are also becoming increasingly popular. The huge advantage of these types of MICs is their easy steric and electronic tuning, which is a result of the modular synthesis of these ligands. While the strong donor properties of these MICs have been the primary reason behind their use in homogeneous catalysis, their tunable and switchable donor/acceptor properties are other positive attributes of this ligand class. The limited stability of the free MICs of this type is sometimes a disadvantage of this ligand class. However, there are a number of emerging strategies to improve their stability. These approaches will very likely lead to different types of new chemistry in the near future. As has been discussed in this article, the steric and the donor and the π-accepting properties of these MICs are largely tunable. These properties make such compounds suitable for a number of applications. The recent breakthroughs that have been achieved in photochemistry, redox-switchable catalysis, small-molecule activation, and electrocatalysis will likely inspire several new types of chemistry and applications with this compound type in the future.

The authors declare no competing financial interest.