Ruthenium-Catalyzed Azide Alkyne Cycloaddition Reaction: Scope, Mechanism, and Applications.

The ruthenium-catalyzed azide alkyne cycloaddition (RuAAC) affords 1,5-disubstituted 1,2,3-triazoles in one step and complements the more established copper-catalyzed reaction providing the 1,4-isomer. The RuAAC reaction has quickly found its way into the organic chemistry toolbox and found applications in many different areas, such as medicinal chemistry, polymer synthesis, organocatalysis, supramolecular chemistry, and the construction of electronic devices. This Review discusses the mechanism, scope, and applications of the RuAAC reaction, covering the literature from the last 10 years.


INTRODUCTION
Nitrogen-containing heterocycles are ubiquitous in nature, being components of many important biomolecules, such as RNA and DNA, peptides and proteins, and vitamins and cofactors, and found in natural products such as alkaloids. Because of the diversity of properties displayed by nitrogen heterocycles, members of this class such as triazoles are widely applied in the pharmaceutical industry, 1,2 and synthetic methods for these building blocks are thus of importance. This Review will focus on 1,2,3-triazoles. While earlier methods for preparing these molecules often required harsh reaction conditions, 3 this changed in 2002 when Meldal and co-workers and Fokin, Sharpless, and co-workers independently reported mild and direct methods for accessing 1,4-disubstituted 1,2,3-triazoles in one step from an organic azide and an alkyne, using Cu(I) catalysis. 4,5 This copper-catalyzed azide alkyne cycloaddition (CuAAC) has since found widespread use even outside the chemistry community. 6−11 However, a method to form the corresponding 1,5-disubstituted 1,2,3-triazole isomer was also needed. In 2005 this problem was solved by Fokin, Jia, and coworkers, 12 who showed that, by exchanging copper for ruthenium, this class of isomers could also be accessed. While this ruthenium-catalyzed azide alkyne cycloaddition (RuAAC) has as yet not found as prevalent use as the CuAAC reaction, reports of its application are increasing rapidly. Although the RuAAC reaction has been briefly mentioned in several reviews on triazole formation 13−23 and metal-catalyzed reactions, 24−26 a comprehensive survey focused solely on RuAAC is now essential. This Review covers the initial reports of the ruthenium-catalyzed azide alkyne cycloaddition from 2005 12 and 2008, 27 as well as the synthetic development, mechanistic studies, and applications of the RuAAC reaction up until September 2016.

BACKGROUND
The 1,3-cycloaddition of organic azides to alkynes to form 1,2,3triazoles is generally referred to as the Huisgen cycloaddition, due to Huisgen's extensive studies of the mechanism and kinetics of dipolar cycloadditions in the 1960s. 28−30 The first report of such a transformation dates back even further, to 1893, when Michael described the reaction of phenyl azide with dimethyl acetylenedicarboxylate. 31 However, many cycloadditions involving azides are impractically slow at ambient temperature, 32 and mixtures of 1,4-and 1,5-disubstituted triazoles are formed in the reaction with alkynes (Scheme 1). 28 In 2002, both Meldal and co-workers at the Carlsberg Laboratory in Copenhagen 4 and Fokin, Sharpless, and coworkers at the Scripps Research Institute in La Jolla 5 independently reported that Cu(I) can catalyze the 1,3cycloaddition of alkynes to organic azides to selectively form 1,4-disubstituted 1,2,3-triazoles under mild reaction conditions. Meldal and co-workers applied this reaction toward the solidphase synthesis of triazole-based peptidomimetics, using copper-(I) iodide as the catalyst. 4 The report from Scripps, meanwhile, provides a procedure for the reaction in solution, employing convenient in situ reduction of copper(II) sulfate with ascorbic acid to form the necessary Cu(I) species. 5 This Cu-catalyzed azide alkyne cycloaddition (CuAAC) has since been applied to a great extent, [9][10][11]34,35 not only in chemistry but also in related areas such as biology 36−38 and new materials, 39 mainly due to the experimental simplicity of the reaction, as well as its robustness and the high yields obtained. The reaction is tolerant of almost every functional group; can be performed in a wide range of solvents, including water; and has a very favorable atom economy 40 as all ingoing components are, at least in theory, incorporated into the product. Triazoles are stable compounds with interesting properties, 1 but an important application of this reaction has also been as a method for connecting or "clicking together" two molecules with each other, providing an alternative to amide bond formation, which has traditionally filled this role. Although "click chemistry", in the original definition, 41 includes a wider range of transformations, 42,43 CuAAC is normally the first reaction that springs to mind when this expression is employed. 44 As mentioned, different isomers can be obtained in the cycloaddition of alkyl and aryl azides with alkynes. While the original Huisgen reaction afforded mixtures of 1,4-and 1,5disubstituted 1,2,3-triazoles 32 and CuAAC selectively produced the 1,4-isomer, 4,5 there was a lack of robust methods for the exclusive formation of 1,5-disubstituted triazoles. Such compounds can be of interest in their own right, as the positioning of the two substituents on adjacent atoms can be advantageous in cyclization reactions, and the rigid cyclic structures of these products can be exploited in materials science applications. However, in many of the studies discussed in this Review, the main focus has been to compare the properties of 1,4-and 1,5disubstituted 1,2,3-triazoles in different contexts, and this is especially true in the area of medicinal chemistry and studies of bioactive compounds.
In looking at methods for the preparation of 1,5-triazole isomers, selected metal acetylides (Sn, Ge, Si, and Na) had early on been reported to produce 4-metalated 1,5-disubstituted triazoles, 45,46 and Akimova et al. studied the reactions of lithium and magnesium acetylides in detail in the 1960s. 47−49 These latter reactions are proposed to involve nucleophilic attack of the lithium or magnesium acetylide on the azide followed by ring closure to form metallotriazole 1 (Scheme 2). Quenching of the reaction then afforded the 1,5-disubstituted 1,2,3-triazole 2. However, the yields were low and the reaction was accompanied by the formation of byproducts. Scheme 1. Thermal Cycloaddition of Organic Azides with Alkynes Affords Isomeric Mixtures 28 Krasinśki et al., together with Fokin, Sharpless, and coworkers, returned to these early reports by Akimova et al. to see if the magnesium-mediated reaction could be improved. 50 They found that the original procedure described by Akimova et al. provided 1,5-disubstituted 1,2,3-triazole 2 in substantially higher yields than earlier reported (Scheme 3). In addition, the scope of substrates applied in the reaction was increased. The 4-metalated triazole intermediate could also be trapped with other electrophiles than protons, providing access to 1,4,5-trisubstituted 1,2,3triazoles 3.
However, the use of magnesium acetylides requires stoichiometric amounts of metals, and such reagents are incompatible with a number of functional groups. A catalytic alternative to CuAAC that provides access to 1,5-disubstituted 1,2,3-triazoles would be preferable. As Ru(II) is known to catalyze reactions involving alkynes, 51 Fokin, Jia, and co-workers reasoned that a ruthenium(II) complex could perhaps be applied toward the synthesis of 1,2,3-triazoles via a cycloaddition reaction using an alkyne in a similar manner to the CuAAC reaction. 12 Initial catalyst screening, using the reaction of benzyl azide with phenylacetylene as a test system, showed that triazoles were indeed formed when heating all components in benzene at 80°C for 4 h. Complexes such as Ru(OAc) 2 (PPh 3 ) 2 , RuCl 2 (PPh 3 ) 3 , and RuHCl(CO)(PPh 3 ) 3 afforded the 1,4-disubstituted 1,2,3triazole (4, Scheme 4) with varying degrees of conversion, and this type of selectivity is discussed in more detail in section 4.5.2.
However, a complex containing a cyclopentadienyl ligand, CpRuCl(PPh 3 ) 2 , did encouragingly produce the 1,5-isomer 5, albeit accompanied by some of the 1,4-disubstituted derivative. By switching to the more sterically hindered pentamethylcyclopentadienyl derivative Cp*RuCl(PPh 3 ) 2 , complete selectivity for the 1,5-disubstituted triazole was obtained. Several other Ru(II) catalysts containing the [Cp*RuCl] unit were also useful in effecting this transformation, i.e., Cp*RuCl(COD) (COD = cyclooctadiene), Cp*RuCl(NBD) (NBD = norbornadiene), and [Cp*RuCl 2 ] 2 . A full investigation of the scope and limitations of this reaction was then carried out as part of this initial report 12 as well as in an ensuing publication a few years later, where also RuH 2 (CO)(PPh 3 ) 3 was reported to afford the 1,4-regioisomer. 27 Certain aspects of these initial studies, concerning the substrate scope as well as the catalysts that have been evaluated, are covered in more detail in other sections below, but a summary of these two seminal papers will be given here.
The initial conditions employed 5 mol % catalyst, but this could be reduced to 1 mol % in most cases. Heating to 60−80°C did provide a more rapid reaction for Cp*RuCl(PPh 3 ) 2 , but ambient temperature could be applied by running the reaction for a longer time at higher catalyst loading. For the more-reactive complex Cp*RuCl(COD), the reaction gave high yields at room temperature even after 30 min. Both polar and nonpolar solvents could be employed, including benzene, toluene, dioxane, tetrahydrofuran, dimethylformamide, and 1,2-dichloroethane. Yields were substantially lower if the reaction was performed in protic solvents such as water and alcohols, but trace amounts of water in nonprotic solvents did not affect the reaction. As oxygen may react with the Cp*RuCl(COD) catalyst, it is recommended to perform the reaction under an inert atmosphere in this case to avoid degradation of the complex. 52 However, the Cp*RuCl-(PPh 3 ) 2 catalyst did not appear to be very sensitive in this system, and excellent conversions were obtained even if oxygen was not excluded. 27 A striking feature of the reaction is the high tolerance toward many functional groups, such as alkyl and aryl chlorides, boronic ester, alkenes, and polyalcohols, in both the azide and the alkyne reaction partners. Figure 1 shows some of the products formed using either Cp*RuCl(PPh 3 ) 2 or Cp*RuCl(COD) as the catalyst.
In contrast to the CuAAC reaction, which is limited to terminal alkynes, the RuAAC reaction was found to tolerate internal alkynes, providing access also to 1,4,5-trisubstituted 1,2,3-triazoles. 12,27 The Cp*RuCl(COD) catalyst was found to be well-adapted to this task, allowing reactions to be performed at ambient temperature in order to access a wide range of different trisubstituted triazoles. This aspect of the reaction, as well as the observed regiochemistry for this class of alkyne substrates, is discussed in more detail in section 4.4.2. The reactivity of internal alkynes is also an indication that the reaction does not proceed via a ruthenium acetylenide, in contrast to the CuAAC reaction that involves an intermediate copper acetylenide. 53 Mechanistic proposals for this reaction are summarized in the next section.

MECHANISTIC AND THEORETICAL STUDIES
Regarding the mechanism of triazole formation, numerous studies have appeared addressing the reaction steps throughout the cycloaddition. These rely on both experimental techniques, employing, e.g., mass spectrometry, NMR spectroscopy, and single-crystal X-ray diffraction, as well as on quantum mechanical (QM) calculations using mainly density functional theory (DFT). We here mostly focus on the Ru-catalyzed azide alkyne cycloaddition; for studies on the Cu-catalyzed azide/alkyne 1,3cycloaddition reaction (CuAAC), the reader is directed to papers by Rostovtsev et al. and Himo et al. 5,53 Note that other transition metal complexes can be used in the chemistry of organic azides; these reactions are covered in a detailed review by Cenini et al. 54 The atomic-level insight by computations into the chemical reorganization provides an important contribution in understanding the steric and energetic aspects of the reaction. However, calculations quite often provide several alternative routes, where it is not trivial to select the most likely reaction pathway. Furthermore, the accuracy of energetic descriptions can often be undermined due to solvent effects, temperature, and other parameters. Consequently, capturing key intermediates and determining conversion rates by experimental methods provide important contributions to selecting the most probable reaction mechanism. Accordingly, in this section we concentrate on the steric and energetic details of the Ru-catalyzed cycloaddition provided by theoretical methods, but we reflect also on their correlation with experimental investigations.
The general mechanism of the azide alkyne 1,3-cycloaddition reaction (AAC) was described in an early study by Himo et al. 53 The authors employed the B3LYP hybrid functional and considered the cycloaddition for both 1,4-and 1,5regioisomers. 53 The reaction without any catalyst has a rather high barrier for both 1,4-and 1,5-regioisomers, with relative energies of 25.7 and 26.0 kcal/mol, respectively. Both reaction pathways were reported to be highly exothermic with >60 kcal/ mol, not accounting for entropy effects. 27,53 The general mechanism was further investigated by Jones and Houk, who addressed substituent effects in detail and also the reversibility of the 1,3-dipolar cycloadditions involving azides and either alkenes or alkynes. 55 Jones and Houk employed complete basis set (CBS) calculations, which allow energetic predictions with very high accuracy. 56 Their results indicated that azide 1,3-dipolar cycloadditions with alkenes and alkynes have similar reaction barriers, but for the latter the products are more exothermic by ∼30−40 kcal/mol. They also concluded that azide cycloadditions with alkynes and most alkenes are irreversible. Note that the calculations also provided the single, so far only theoretical, prediction that the reaction of methanesulfonylazide with N,N′-dimethylvinylamine may be reversible at micromolar reactant conditions. Focusing more on the ruthenium catalysts, Boren et al. employed DFT calculations using methyl azide and propyne as model reactants with [CpRuCl]. 27 These calculations demonstrated that in principle four different relative orientations are possible for the methyl azide and propyne, with rather small relative energy differences between them ( Figure 2). 27,57 Although two of these alternative orientations would eventually lead to 1,4-regioisomers, three pathways were ruled out due to either much higher barrier heights or steric repulsions in the product states. The lowest-energy pathway had its highest barrier at 13 kcal/mol, much lower than the barrier height for the same path without a catalyst. This shows that Ru is an efficient catalyst, reducing the rate of the reaction by several orders of magnitude. The calculated results ( Figure 3) fit well to the experimental observations, explaining the 1,5-regioselectivity and also providing a reasonable energetic explanation for the reaction rate. The reaction was later revisited in detail by using alternative substrates including Cp*, which confirmed the reaction paths determined earlier but further explained the effect of different substituents on the reaction rate and the product ratio of 1,4-and 1,5-regioisomers. 58 On the basis of these results,   Boren et al. proposed that the mechanism involves an irreversible oxidative coupling, which is a nucleophilic attack of the coordinated alkyne on the terminal electrophilic nitrogen of the azide, and this is followed by a reductive elimination.
The mechanism was also addressed by Hou et al., 59 who again confirmed the irreversibility of the reaction and further investigated the molecular orbitals of the C−N bond-forming transition state using DFT calculations and Bader's atoms-inmolecules theory. 60,61 The latter analyzes interactions of atoms by topological mapping of the surrounding electron density. They found that, when using larger substrates, the transition state involves the electron donation from the alkynyl carbon to the π* orbital of the NN group from the azide, opening the way to formation of a transient cyclic structure where the π electrons delocalize from the alkynyl to the azido group. It is worth noting that, for internal alkynes, both Boz and Tuzun 58 and Hou et al. 59 showed a higher-energy transition state for the C−N bondforming step than observed initially for small substrates and terminal alkynes.
Besides 1,5-substituted 1,2,3-triazoles, selective synthesis of 1,4-substituted compounds can also be achieved with high yield using Ru complexes lacking Cp ligands (see section 4.5.2). Liu et al. have demonstrated that, among other catalysts, RuH(η 2 -BH 4 )(CO)(PCy 3 ) 2 and Ru(CCPh) 2 (CO)(PCy 3 ) 2 were the most active. 62 In addition to mass spectrometry, X-ray diffraction, and NMR spectroscopy, the authors have also used DFT calculations to elucidate the mechanism of the reaction. The presence of the triazolide intermediate (structure provided by X-ray, Figure 4) indicates that the Ru center actively takes part in the reaction, forming even a Ru···H−C interaction with one hydrogen of the CMe 3 group in the triazolyl ligand. 62 In their DFT studies using methyl azide and propyne as model substrates, the authors have determined that the reaction starts with a Ru complex that coordinates the internal nitrogen in the azide. The triazolide intermediate is then reached via a metathesis step, which is also the rate-determining step at 11.5 kcal/mol. The pathway for these complexes lacking Cp ligands to 1,5disubstituted products has a much higher barrier, 21.5 kcal/mol, which explains their selectivity toward 1,4-disubstitution. The structural explanation is that coordination of the internal (−N− Me) nitrogen from the azide on the Ru complex results in an intermediate and a transition state (TS) with conjugated bond systems; thus, the barrier of the TS for the rate-determining step lowers significantly. In contrast to Cp-containing Ru catalysts, the final Ru complex with the formed triazole product has a much lower energy gain with a −38.6 kcal/mol relative energy. Overall, the lowest-energy reaction pathway for the Ru complexes lacking Cp ligands is rather similar to those with a Cu catalyst.
To provide a systematic overview on alternative Ru complexes, Nolan and co-workers have investigated the effect of substituents in Ru complexes that result in unsaturated 16-electron configurations in the general form of Cp*Ru(L)Cl. 63 Here L was a sterically demanding ligand, either phosphine or an Nheterocyclic carbene (NHC), where the nitrogens contained adamantyl-(IAd), diisopropyl-(IPr), or cyclohexyl groups (ICy). 63 They chose sterically large substrates to model the reaction, using benzylazide and phenylacetylene to afford the 1,5disubstituted 1,2,3-triazole compound. Mechanistic insights were gained by NMR experiments combined with DFT calculations. Catalysts with phosphines, Cp*Ru(PiPr 3 )Cl and Cp*Ru(PCy 3 )-Cl, showed the best catalytic activities, having 89% and 87% conversion rates, respectively, while the best catalyst with an NHC-based ligand was IAd, showing a somewhat lower conversion of 67%. In terms of mechanism, this clearly hints that the behavior of the Ru complexes cannot be directly correlated to the steric size of the L ligands. The authors have also demonstrated that the anionic Cl ligand makes an important contribution to the conversion rate. 63 While Cp*Ru(PiPr 3 )Cl had a rate of 89%, the alkoxo complex Cp*Ru(PiPr 3 )-(OCH 2 CF 3 ) produced only traces of the product. They have also found that addition of the alkyne prior to addition of the azide lowered the conversion by 6-fold, whereas adding the azide first and the alkyne second did not show any significant difference relative to adding the mixture of the two substrates. In line with the above, the calculations have shown that excess phenylacetylene results in a complex with two alkyne ligands on the Ru complex, leading to the production of a 1,4-disubstituted metallacycle as predicted earlier. 64 A similar scenario was also investigated using excess phenyl azide, which could potentially lead to a stable tetraazametallacyclopentene ring. However, the highest barriers are rather low for the former, below 10 kcal/mol, while several barriers are well above 20 kcal/mol in the latter case. These give an explanation for the experimental behavior of the studied Ru complex in excess of one of the substrates. In contrast to the species based on [Cp*RuCl], here the extrusion of ligands to create the coordinative unsaturation at the Ru center is of importance. The calculations have demonstrated that, out of the activated catalysts having the original Cp*Ru(PiPr 3 )Cl and either phenylacetylene or the internal-or terminal Ncoordinated benzylazide, only the Cp*RuCl(η 2 -HCCPh)(PiPr 3 ) complex favors phosphine dissociation over that of the bound substrate. This is also in agreement with the NMR studies that observed this intermediate complex at low temperatures. These results suggest that activation of the catalyst should proceed through initial coordination of an alkyne substrate, and thus alkyne binding precedes azide coordination. Starting from the activated catalyst, the reaction for the investigated complex then progresses on the same pathway as observed by Boren et al. 27

METHODOLOGY, SUBSTRATES, AND CATALYSTS
This section will look at the general conditions that can be applied for the RuAAC reaction, summarize the catalysts that have been employed in different studies, and report on more specialized studies where the focus has been on the reaction itself rather than on the applications of the target molecules.

General Methodology
A typical RuAAC procedure involves the reaction of an alkyne with an organic azide in the presence of catalytic amounts of a ruthenium(II) complex containing a [Cp*RuCl] unit in a nonprotic solvent. A variety of solvents can be used for the reaction, where the most commonly used are aromatic solvents, like benzene or toluene, or ethers such as tetrahydrofuran (THF) and dioxane. Certain polar solvents can also be applied, i.e., dimethylformamide (DMF) and dimethylacetamide (DMA), while reactions using dimethyl sulfoxide (DMSO) have been reported to be problematic, 65 which is most likely related to the ability of DMSO to act as ligand to ruthenium. 66 Protic solvents are not suitable, giving low yields and a high degree of byproduct formation. 27,52 Most reported RuAAC reactions employ either Cp*RuCl(PPh 3 ) 2 or Cp*RuCl(COD) as the catalyst, using between 1 and 5 mol % catalyst, and both of these complexes are commercially available. For full details of the catalyst scope, see section 4.5. Although heating is generally employed to shorten reaction times, reactions at ambient temperature are also possible, especially when using a catalyst with high reactivity such as Cp*RuCl(COD). 27 A typical procedure for the RuAAC reaction has been described by Oakdale and Fokin in Organic Synthesis and involves the reaction of benzyl azide with phenyl acetylene in dichloroethane at 45°C, using 1 mol % Cp*RuCl(COD) as the catalyst. 52 The high tolerance to a wide range of functional groups has already been mentioned in section 2; these include alcohols and polyols, amines and pyridines, alkyl and aryl halides, ethers, aldehydes and ketones, esters, amides, sulfonamides, carbamates, and boronic esters. 12,27 However, acidic functionalities such as carboxylic acids, boronic acids, and HCl/HBr salts of amines can be problematic. 65 This broad application scope is also one of the reasons why the RuAAC reaction has been extensively applied in the area of medicinal chemistry, where highly functionalized substrates, often including H-bond donors and acceptors, are used. Such applications are covered in more detail in section 5.
The use of microwave heating has been applied in several instances in the CuAAC reaction to accelerate triazole formation 67,68 but is of even more relevance for the RuAAC reaction, where heat is in many cases required to achieve a high conversion in a short time. Fokin and co-workers applied this technology in the RuAAC reaction when investigating aryl azides as substrates (see section 4.3). 69 The authors reportedly opted for a microwave-mediated reaction to facilitate optimization of the reaction conditions but also found that direct comparison of a microwave-assisted reaction with conventional oil bath heating was in favor of the former (Scheme 5). Not only was the yield higher when microwave heating was used, but the conventional method was also fraught with byproduct formation. Looking at other comparisons between conventional and microwave heating, Carvalho and co-workers also report a higher yield by employing microwave-assisted RuAAC in their synthesis of benznidazole analogues, 70 of interest for the treatment of Chagas disease. However, Agrofoglio and colleagues obtained similar results in terms of yield and selectivity when comparing oil bath heating with microwave heating in the synthesis of ribavirin analogues via RuAAC. 71 However, the reaction time was significantly shortened.
A sequential microwave-assisted, one-pot method to produce triazoles directly from alkyl halides, more easily available than alkyl azides, was described by Johansson et al. 65 Although direct one-pot methods for CuAAC have been reported, 72 75,76 However, by first heating the alkyl halide with sodium azide for 10 min and then adding the catalyst and the alkyne (Scheme 6), this problem was solved and 14 different triazoles were prepared in moderate to excellent yields. Alkyl chlorides as well as iodides could also be used, while acidic functionalities in the alkyne or alkyl halide were not tolerated.

Structural Assignment
Once the triazole is formed, suitable analytical methods are needed to determine the regioselectivity in the reaction and to confirm that the 1,5-disubstituted isomer is indeed the one formed, as the 1,4-and 1,5-isomers are difficult to differentiate between using simple 1 H NMR. In the original report by Fokin, Jia, and co-workers, the structure and 1,5-regioselectivity of selected products were verified by X-ray crystallography (see Figure 5). 12 1D or 2D NMR techniques based on the nuclear Overhauser effect (NOE), providing information on interactions through space rather than through bonds, are also convenient tools to differentiate between the 1,4-and 1,5-isomers and have been applied in several instances. 65,77 Creary et al. sought to develop a simpler means of distinguishing between the 1,4-and 1,5-isomers of a 1,2,3-triazole using 13 C NMR. 78 Nineteen pairs of triazole isomers bearing a variety of different substituents were investigated, focusing on the signal displayed by the unsubstituted carbon of the triazole ring. As a general trend, the C5 carbon of the 1,4-disubstituted triazole displays a signal around 120 ± 3 ppm, while the corresponding C4 carbon of the 1,5-disubstituted isomer instead appears around 133 ± 3 ppm ( Figure 6). For triazoles carrying a carbonyl group, the signals from the unsubstituted carbons are shifted slightly but nevertheless show a clear difference between the two isomers. Ethoxy-substituted isomers also deviated from the norm, with carbon signals for the unsubstituted carbons appearing at 106 ppm for the 1,4-isomer and 114 ppm for the 1,5-isomer. Both these discrepancies are in line with predictions obtained via supportive theoretical studies. The authors employed DFT calculations and used gauge-independent atomic orbitals (GIAOs) to calculate the shielding tensors for the investigated compounds. The theoretical calculations showed the same relative differences for C4 and C5 as in the experiments, although having a consistent 6 ppm upfield shift compared to the experiments.
The heterocyclic ring of triazoles contains three nitrogen atoms that could potentially also be analyzed using NMR. The low natural abundance and sensitivity of 15 N has earlier been a limitation in applications of 15 N NMR spectroscopy, although probe development and new pulse techniques have greatly facilitated the use of this technology. 79 Alfonso and co-workers 80 instead applied an indirect method, i.e., a gradient-enhanced indirect detection pulse sequence, 81 that makes use of the 1 H nucleus to obtain information about 15 N NMR shifts in order to differentiate between 1,4-and 1,5-disubstituted 1,2,3-triazole isomers, as well as to distinguish these from the less-common 2,4disubstituted isomer. The substituted nitrogen appears at a much lower shift (in the range of −120 to −140 ppm) in comparison to the other two nitrogen atoms for all three isomers, allowing the 2,4-substituted isomer to be easily identified. The 1 H/ 15 N correlation pattern can then be employed to distinguish the 1,4and 1,5-disubstituted isomers. To aid interpretation of the experimental results, the GIAO method mentioned above was used with DFT calculations. To reach a consistent data set, the authors have calculated the shielding value for all obtained conformers and used a Boltzmann weighing based on the relative energies of these different conformers. These data showed reasonable agreement with the NMR results and helped in differentiating between the different observed isomers.

Azides
The highly energetic nature of azides and their toxicity are aspects that need to be considered when working with 1,3cycloaddition reactions of azides to alkynes, and we recommend that new users of the CuAAC and RuAAC reactions familiarize themselves with the properties of azides before embarking upon experimental work. A treatise by Keicher and Lobbecke on the safety of azides provides some general guidelines. 82 As a rule of thumb, azides with a (C + O)/N ratio of <3 are considered to be highly explosive.
In the initial report of the RuAAC reaction, Fokin, Jia, and coworkers reported that some limitations were found in terms of the azide component. 12 Although primary aliphatic azides functioned well in the reaction, both tertiary azides (tert-butyl and adamantyl) and aromatic azides afforded low yields. Reactions of the latter were also accompanied by the formation  . 13 C NMR as a tool for distinguishing between 1,4-and 1,5disubstituted 1,2,3-triazole isomers. 78 of byproducts. An exception was seen in the reaction of 1-azido-4-methoxybenzene with a tertiary propargylic alcohol, forming the desired triazole in 94% yield. Reactions involving sterically hindered azides could be improved by increasing the catalyst loading to 5% and extending the reaction time. Seeking to find a solution also to the problematic aryl azides, Fokin and coworkers investigated other ruthenium catalysts to see if a more efficient conversion to 1-aryl-substituted 1,2,3-isomers could be attained. 69 During the initial development of the RuAAC reaction, the authors had found that the ruthenium tetramer [Cp*RuCl] 4 performed well in RuAAC reactions of aliphatic azides with alkynes using dimethylformamide as the solvent. 69 This catalyst, prepared by reduction of [Cp*RuCl 2 ] n with LiBHEt 3 , was thus evaluated with aryl azides and was indeed found to effect the desired transformation efficiently. Microwave irradiation was employed to facilitate optimization of the reaction, and suitable conditions involved the use of 10 mol % catalyst in DMF, heating for 20 min at 110°C (Scheme 5). Longer reaction times did not provide a higher yield, indicating that catalyst deactivation is a competing reaction. Both electronrich and electron-poor aryl azides could be employed (Scheme 7), although the authors report that strongly electron-withdrawing groups as well as diortho substituents inhibited the reaction.
The remainder of this section will look at two reports where the focus has been on the azide component. β-Tosylethylazide can be employed as a synthon for hydrazoic acid, HN 3 , in the RuAAC reaction to access 5-substituted 1H-1,2,3-triazoles after deprotection (Scheme 8), as shown by Yap and Weinreb. 83 Phenylacetylene was heated with an excess of β-tosylethylazide in the presence of Cp*RuCl(PPh 3 ) 2 , using benzene as the solvent, affording selectively the 1,5-disubstituted triazole 6 in 77% yield. Deprotection of the β-tosylethyl group with potassium tert-butoxide in THF at low temperature produced the free 1H-1,2,3triazole 7 via a retro-Michael reaction, providing convenient access to this class of substrates. Internal alkynes were also applicable in this reaction. The β-tosylethylazide reagent was prepared in one step from commercially available p-tolylvinyl sulfone and sodium azide, and could be stored at low temperature for an indefinite period of time.
Fused tetrazoles can be applied as azide surrogates in CuAAC reactions with alkynes to form 1,4-disubstituted triazoles appended with a heterocyclic functionality such as a 2-pyridyl moiety, as demonstrated by Gevorgyan and co-workers. 84 The reaction is based on the equilibrium existing between the closed form of the tetrazole and the corresponding open azide form (Scheme 9a), where the position of this equilibrium can depend on a variety of factors, including the substitution pattern and the reaction conditions. Attempted cyclization of phenylacetylene with tetrazole 8 using Cp*RuCl(PPh 3 ) 2 was unsuccessful, however (Scheme 9b). The authors speculated that the nitrogen atom of the pyridine could be involved in chelation with the ruthenium complex, resulting in catalyst deactivation. A reaction using 4-azidopyridine, where such chelation is not possible, afforded mainly the 1,5-disubstituted product, indicating that it is indeed the positioning of the 2-azide functionality relative to the pyridine nitrogen that is the problem. Unfortunately, this convenient azide surrogate methodology is not applicable in RuAAC.

Alkynes
A wide scope of alkynes are compatible with the RuAAC reaction, also including internal triple bonds and alkynes directly connected to a heteroatom. Many of the terminal alkynes employed in RuAAC belong to the category of functionalized terminal alkynes, and in this Review the majority of these alkynes have been classified according to the potential applications of the triazoles produced. In this section, we will discuss various functionalized alkynes where method development has been the focal point of the study.  85 Ketal 9 as well as ketone 10 were selected as substrates and reacted with benzyl azide in the presence of CuSO 4 to form triazoles 11 and 12 via CuAAC (Scheme 10). Disappointingly, however, treatment of ketal 9 with Cp*RuCl(PPh 3 ) 2 afforded only recovered starting materials upon workup. Ketone substrate 10 did react but produced the aromatic product 13 instead of a triazole, which the authors propose is formed via an ionic cyclotrimerization reaction. The same group also investigated the debenzylation of N-benzylated 1,4-and 1,5-disubstituted 1,2,3triazoles under hydrogenation conditions, showing that both isomers afford the same 1H-1,2,3-triazole via tautomerization of the product. 86 Piperidines are common motifs in natural products as well as in pharmaceuticals. Haug and colleagues prepared a set of 3hydroxypiperidine triazoles using both CuAAC and RuAAC. 87 1,5-Disubstituted triazoles 14 and 15 were successfully synthesized from alkynes 16 using 1 mol % of Cp*RuCl(PPh 3 ) 2 as the catalyst (Scheme 11).

4.4.2.
Internal Alkynes. One advantage of RuAAC compared to CuAAC is that the former allows the use of internal alkynes, while the latter reaction involves the formation of a copper acetylide 53 and thus requires a terminal alkyne. In the original report of the RuAAC reaction, Fokin, Jia, and co-workers included one example of a symmetrical internal alkyne with benzyl azide, showing that, while the ruthenium-catalyzed reaction afforded a 1,4,5-substituted triazole in 80% yield after 2 h of reflux in benzene, the corresponding uncatalyzed reaction showed only trace amounts of product even after 24 h (Scheme 12).
Inspired by this study, Majireck and Weinreb decided to apply unsymmetrical internal alkynes in RuAAC to investigate the regiochemical outcome when different substitution patterns on the alkyne were used. 77 Standard reaction conditions were applied, involving the use of benzyl azide as the azide component and Cp*RuCl(PPh 3 ) 2 as the catalyst in refluxing benzene for 2 h, using an excess of the alkyne relative to the azide. Isomeric ratios were measured by 1 H NMR on the crude products, and the regiochemistry of each isolated product was then verified by 1 H NMR NOE. A selection of triazoles derived from internal alkynes was also prepared by Fokin, Jia, and co-workers using Cp*RuCl(COD) as the catalyst in toluene at room temperature. 27 Similar trends were seen by both groups (Figure 7), allowing some insight into factors that might govern the regioselectivity. Mixed alkyl-/aryl-substituted alkynes gave

Chemical Reviews
Review mixtures of regioisomers 17 and 18, and the same result was seen for alkynes bearing two different alkyl groups. However, an exception to this was seen for the reaction of benzyl azide with 4,4-dimethylpent-2-yne, bearing a methyl group on one of the acetylenic carbons and a tert-butyl group on the other, where triazole 19 was obtained with complete regioselectivity, albeit in a rather low yield. Triple bonds with an electron-withdrawing group afforded triazoles with this group in the 4-position with complete selectivity (20 and 21). An alcohol or amine in the propargylic position of the alkyne instead resulted in the selective formation of triazole isomers where these substituents are in the 5-position of the triazole (compounds 22 and 23), explained by hydrogen bonding of −OH or −NH to the chloride ligand of the ruthenium complex. 27 However, a homopropargylic alkyne was found to be unselective, 77 producing a mixture of the regioisomers, indicating that the hydrogen-bond donor needs to be placed in the propargylic position and not further away. Poor regioselectivity when using homopropargylic alcohols has also been noted by Pelly and co-workers. 88 Majireck and Weinreb also found that alkynes containing heteroatoms connected directly to the triple bond also favored the isomer where this heteroatom substituent ends up in the 5-position in the product, as seen in triazole 24. The regiochemistry can thus be summarized as follows: propargylic hydrogen-bond donors, bulky substituents, and heteroatoms will be found on C5 in the triazole, while electron-deficient groups will be directed to C4.
A more specialized class of internal alkynes was studied by Qing and colleagues, 89 who focused their work on propargylic alcohols with a trifluoromethyl group connected to the alkyne. The precursor alkynes were made by treating 2-bromo-3,3,3trifluoropropene with LDA followed by the addition of a ketone or aldehyde, to provide four different trifluoromethylated propargylic alcohols to evaluate in RuAAC, using either benzyl azide or n-octyl azide, and [Cp*RuCl 2 ] n as the catalyst. Complete selectivity for isomers such as 25 (Scheme 13), with the CF 3 group in the 4-position, was obtained, and the authors propose a polarity effect to explain the regiochemical outcome. A limited study was also performed using other reaction conditions, i.e., a thermal cyclization (80°C) performed in water and a Pdcatalyzed reaction carried out in refluxing benzene. Mixtures of regioisomers were obtained in both these cases, demonstrating the utility of RuAAC in this context.
C3-Symmetric tripodal triazole-containing ligands for copper-(I) complexation have been described by Toth and colleagues. 90 Propargylic alcohol 26 (Scheme 14) was reacted with benzyl azide in the presence of Cp*RuCl(COD), forming a 3:1 mixture of the two possible triazole regioisomers of 27. Following separation by chromatography, the major isomer was subsequently coupled via esterification to a central cyclohexane scaffold, affording tripodal ligand 28, which was found to be an efficient ligand for the CuAAC reaction.
Halogenated internal alkynes can be employed to form 5-halo-1,2,3-triazoles and will be discussed in the next section. 91 An internal bisbithiophenyl alkyne also has been applied in RuAAC with good results. 92 4.4.3. Heteroatom-Substituted Alkynes. Heteroatomsubstituted alkynes have been applied in RuAAC, although there are few examples as yet in the literature. One of the early reports came from Cintrat and colleagues, 93 who applied three different ynamides (29−31) in RuAAC, providing access to 5-amido-1,2,3-triazoles (Scheme 15). The authors had earlier reported the corresponding CuAAC reaction 94 and now set out to develop methodology providing access to the other triazole isomer. An initial study using benzyl azide and N-benzyl-N-tosyl ynamide (29)  in toluene afforded solely the 1,5-disubstituted triazole isomer; encouraged by these results, the authors screened a wide variety of functionalized azides in reactions with ynamides 29−31. Complete regioselectivity and good-to-excellent yields were obtained in all cases, although certain sterically demanding substrates needed longer reaction times and heating to drive the reaction to completion. One example of an internal alkyne was also included, affording selectively the isomer with the amido substituent in the 5-position of the triazole, following the same trend as for the terminal alkynes.
Another example of the use of nitrogen-substituted alkynes comes from Taddei and colleagues, in this case involving N-Bocprotected ynamides with a methyl ester functionality at the other end of the alkyne. 95 The purpose was to prepare suitable building blocks for incorporating into triazole peptidomimetics, and these applications are covered in more detail in section 5.1. The authors wished to position the necessary carboxyl and amino functionalities closer to the triazole ring than is normally the case and thus employed an alkyne where these two groups, in protected form, are directly appended to the triple bond. N-Bocynamides 32 and 33 were prepared in two steps from TIPSprotected bromoacetylene and were subsequently converted to a wide range of functionalized N-Boc-5-aminotriazoles in high yields, employing [Cp*RuCl] 4 as the catalyst (Scheme 16). An ynamide-derived 1,5-disubstituted triazole has also been reported by Diaz et al. 96 Halogenated internal alkynes have been investigated by Fokin and co-workers, affording 5-halo-1,2,3-triazoles that can be further derivatized using palladium-catalyzed cross-coupling reactions. 91 Cp*RuCl(COD) is commonly employed for room-temperature RuAAC reactions 27 but was found to be inefficient for reactions involving halogen-substituted internal alkynes. However, replacing Cp* with the less-bulky Cp-ligand was found to solve this problem. A variety of alkyl azides reacted with bromo-, chloro-, and iodo-substituted internal alkynes (Scheme 17). Aryl azides were inefficient in the reaction, but apart from this limitation, the reaction was found to tolerate a wide scope of substituents, including even a second heteroatom on the alkyne in the form of a phosphonate as in 34. Acetonitrile was found to be the most versatile solvent, and many of the reactions could be performed at room temperature, while a few more functionalized or sterically hindered substrates required heating to 50°C. In general, the highest yields were obtained for chloro-substituted alkynes. The authors also showed that triazoles such as 35 could be employed in palladium-catalyzed transformations such as Sonogashira, Stille, Heck, and Suzuki coupling (Scheme 18). Likewise, alkyne precursors substituted with a Weinreb amide as in 36 afforded triazoles that could be converted into aldehydes or ketones (Scheme 19). This possibility for further elaboration of the primary products into more complex structures makes this a very versatile methodology indeed. The halogenated alkynes were also employed in conjunction with nitrile oxides to form halooxazoles, as well as in cyclotrimerization reactions affording tribrominated aromatics.
Sulfur-substituted aryl alkynes were employed by Chen, Guo, and co-workers to prepare (trifluoromethyl)thio-substituted triazoles, potentially useful substrates for medicinal chemistry purposes. 97 Using 10 mol % Cp*RuCl(COD) in benzene, the reaction could be performed at room temperature, and three 1,4,5-trisubstituted triazoles were prepared in 68−85% yields ( Figure 8). HMBC (heteronuclear multiple-bond correlation spectroscopy), which can detect heteronuclear coupling over several bonds, 81 was employed to confirm the location of the −SCF 3 group at the 5-position of the triazole. Reactions involving 2,2,2-trifluoroethyl-substituted alkynes also afforded the desired triazoles in high yields with the same regioselectivity.

Ruthenium Catalysts
The most commonly applied catalysts for RuAAC are Cp*RuCl-(PPh 3 ) 2 and Cp*RuCl(COD), and a number of examples of their use have already been shown in earlier parts of this Review. Table 1 provides a summary of catalysts for RuAAC, together with selected references to their application, while the remainder of this section deals with more specialized Ru catalysts as well as Ru catalysts that afford the 1,4-disubstituted 1,2,3-triazole isomer.
4.5.1. Ruthenium Catalysts Affording the 1,5-Disubstituted 1,2,3-Triazole. Catalysts employed for the RuAAC reaction are generally 18-electron complexes such as the abovementioned Cp*RuCl(PPh 3 ) 2 and Cp*RuCl(COD). Nolan and co-workers investigated the effect of using 16-electron ruthenium complexes of the general structure Cp*RuLX in RuAAC, where L was a phosphine or carbene ligand and X was either chloride or alkoxide. 63 The mechanistic aspects of this study have already been described in detail in section 3 and will be mentioned only briefly here. Using the reaction between phenyl acetylene and benzyl azide as a test system, seven different ruthenium complexes were evaluated as catalysts for the cycloaddition at room temperature in dichloromethane, with a reaction time of 40 min. The best results were obtained when L was a phosphine and X was a chloride ligand ( Figure 9, 37 and 38). Replacing chloride with −OCH 2 CF 3 was detrimental to the reaction, and complex 39 afforded only 2% product. Among complexes with carbene     (40), with adamantyl substituents on the carbene, was the most successful, albeit affording a lower yield than that for the phosphine-substituted complexes. Lo and colleagues prepared several ruthenium azido complexes bearing Tp (trispyrazolylborate) ligands and found that complex 41 ( Figure 10) was an efficient catalyst for the RuAAC reaction between benzyl azide and various terminal acetylenes. 98,99 Interestingly, the catalyst gave similar results in toluene and in water.
A practical approach to RuAAC was reported by Astruc and co-workers, who prepared ruthenium complex 42 ( Figure 11) for immobilization onto magnetic nanoparticles, allowing for facile recycling of the catalyst. 100 A triarylphosphine linker with a pendant trimethoxysilyl group was employed to tether the pentamethylcyclopentadienyl ruthenium(II) catalyst to the surface of a silica-coated γ-Fe 2 O 3 nanoparticle. The catalyst was highly selective for the 1,5-disubstituted 1,2,3-triazole isomer and afforded good yields for a range of alkynes and azides, also including an internal alkyne. Facile catalyst recovery was effected using a magnetic field, and the catalyst could be recycled up to five times without substantial loss in selectivity and yield.
4.5.2. Ruthenium Catalysts Affording the 1,4-Disubstituted 1,2,3-Triazole. In the original reports by Jia, Fokin, and co-workers, it was noted that ruthenium catalysts lacking a cyclopentadienyl ligand afforded the 1,4-disubstituted isomer instead of the 1,5-disubstituted 1,2,3-triazole. 12,27 Although such compounds can be accessed via CuAAC, it was nevertheless of interest to evaluate the range of ruthenium catalysts that were applicable in this variant of the cycloaddition as well as to investigate the scope of substrates that could be used. Liu, Jia, and co-workers focused on the most active of the earlier studied catalysts, i.e., RuH 2 (CO)(PPh 3 ) 3 . 101 Using 5 mol % catalyst in THF at 80°C, a wide variety of different 1,4-disubstituted triazole products could be produced. Of interest is the synthesis of podophyllotoxin derivative 43 (Scheme 20); derivatization of this natural product via RuAAC also has been reported by Tron and co-workers (see section 5.5), but in this case the triazole replaced the lactone functionality. A one-pot cycloaddition/ transfer hydrogenation, using the same catalyst for both reactions, was also described. Liu and co-workers have also shown that the ruthenium-catalyzed cycloaddition can be performed in water using the same catalyst, with a lower catalyst loading (0.2%) than when using an organic solvent. 102 A convenient one-pot method for directly accessing the 1,4disubstituted triazoles via in situ formation of the azide from the corresponding bromide was also included in this study (Scheme 21).
In a later paper by Jia, together with Liu, Fokin, and coworkers, a larger range of ruthenium catalysts was instead explored. 62 Two new catalysts in particular, i.e., RuH(η 2 -BH 4 )(CO)(PCy 3 ) 2 and Ru(CCPh) 2 (CO)(PCy 3 ) 2 , were found to give both high yields and fast reactions. A wide range of alkynes and three different azides were explored in RuAAC using catalyst RuH(η 2 -BH 4 )(CO)(PCy 3 ) 2 . Although tert-butylacetylene as well as an aryl azide were difficult substrates, all other azide/alkyne combinations afforded triazoles in high yields (71−95% yield). Mechanistic studies, also involving DFT calculations, indicate a ruthenium acetylide as a key intermediate.
Porphyrin derivatization is an example of where Ru catalysts that form the 1,4-disubstituted isomer can be put to good use, as the Cu catalyst used for CuAAC can result in copper inserting into the porphyrin. If this is not desired, blocking the central coordination site with Zn(II) is necessary, but this involves extra reaction steps. 103 As part of a study on porphyrin−vitamin B 12 conjugates, Gryko and co-workers employed RuH 2 (CO)(PPh 3 ) 3 to append triazoles onto a central porphyrin scaffold functionalized with an azide (44,Scheme 22) or an alkyne. 104 However, although the desired 1,4-disubstituted structures were obtained, this catalyst afforded low yields and substantial amounts of byproduct formation. Thus, this strategy was abandoned, and the authors instead returned to CuAAC, finding after some optimization that [Cu(phen)(PPh 3 ) 2 ]NO 3 could provide the 1,4-isomers without copper insertion. A set of 1,5-isomers were   also successfully prepared as part of this study, again with precursor 44 but this time using Cp*RuCl(COD) as the catalyst (Scheme 22). As a point of interest, Gallo and co-workers explored ruthenium porphyrin complexes as catalysts for RuAAC but found that reaction of an aryl azide with an aryl alkyne produced indoles rather than the expected triazoles. 105 Lee and co-workers employed ruthenium nanoparticles in ruthenium-catalyzed 1,3-cycloadditions reaction, where the catalyst was prepared by calcination of ruthenium hydroxide at 500°C under a hydrogen atmosphere. 106 The 1,4-triazole isomers were obtained in this case, and of particular interest is the formation of tristriazoles 45 and 46 in 60% and 64% yields, respectively, using this catalyst system ( Figure 12). The ruthenium nanoparticles could be recycled three times without any significant change in morphology or catalytic activity.
Another example of a supported catalyst has been described by Tuhina, Islam, and co-workers, 107 in this case employing a polymeric support in contrast to the nanoparticles employed by Astruc and co-workers (section 4.5.1) 100 and Lee and coworkers. 106 The polymer-bound ruthenium(III) catalyst 47 (Scheme 23) was prepared by reaction of RuCl 3 with a polystyrene-tethered β-alanine ligand, affording a supported catalyst containing 8.5 wt % ruthenium (determined by AAS). To test the catalyst in RuAAC, phenyl acetylene was mixed with sodium azide and benzyl bromide in the presence of varying amounts of the supported catalyst. By evaluation of the reaction conditions, including exploration of a wide range of organic solvents, the authors concluded that water provided the best medium for this transformation. Employing 2 mol % catalyst and a reaction time of 3 h at 40°C produced the 1,4-disubstituted 1,2,3-triazole in quantitative yield, after a simple workup involving only filtration, washing with ethanol, and concentration. The substitution pattern of the product is not surprising considering that the catalyst contains neither a Cp nor a Cp* ligand, and thus, this result is in accord with earlier studies by Jia, Fokin, and co-workers. 12,27 The scope of the reaction was further investigated by introducing substituents on the aromatic groups of the precursors, with product yields in the range of 83−94% (see Scheme 23 for an example). The recyclability was investigated and the supported catalyst could be employed up to six times without noticeable reduction in catalytic activity. The polymer-bound catalyst was also investigated in transfer hydrogenation of aromatic and aliphatic ketones with good results.
4.5.3. Ruthenium Catalysts Affording Other Substitution Patterns. An interesting ruthenium-catalyzed tandem reaction has been described by Bhattacharjee and co-workers. 108 In the presence of silver, the ruthenium compound [Ru-(dppp) 2 (CH 3 CN)Cl][BPh 4 ], prepared from either RuCl 2 (PPh 3 ) 3 or RuCl 3 ·nH 2 O, could effect the homocoupling of alkynes via the formation of silver acetylenides. After heating the reaction for 30 min, sodium azide was added, and after an additional 6−8 h, a product was formed where one of the two triple bonds had participated in a cycloaddition reaction to form a 4,5-disubstituted 1,2,3-triazole (Scheme 24). Five different substrates were isolated in good yields. Attempts to replace NaN 3 with an alkyl azide did not afford any triazole product.

Solid-Phase RuAAC Reactions
The use of solid-phase synthesis (SPS) can sometimes be advantageous, especially in solid-phase peptide synthesis (SPPS). The RuAAC reaction has been successfully applied using SPS methodology by several groups, 109−113 where SPS RuAAC works under similar conditions as the solution-phase reaction in terms of catalysts, temperature, and solvent. An increased amount of catalyst and slightly prolonged reaction time seems, as expected, to be beneficial due to the slower reaction kinetics on the solid  Table 2.

MEDICINAL AND BIOLOGICAL APPLICATIONS
Since catalysts suitable for RuAAC became commercially available, numerous applications in the fields of medicinal chemistry, biochemistry, and drug discovery have been reported. This section aims to give an overview of different strategies that have been applied in the various fields, with the properties of 1,5triazoles in mind.

Peptidomimetics
Compounds that mimic peptides in terms of their secondary structure and/or biological functions, but which can exhibit other beneficial properties such as improved proteolytic stability or improved affinity to a desired target, are known as peptidomimetics. 114 A broad range of different peptidomimetics have been developed, and their use has become a standard tool in medicinal chemistry and drug research today. 115 The field can be broadly divided into three categories, involving scaffold-based peptidomimetics as well as local and global modifications of the parent peptide. The aim of peptidomimetic design is to either maintain or improve biological activity, increase selectivity, and, most importantly, increase proteolytic stability. In this section we will discuss the properties and applications of the RuAAC reaction as a tool for peptidomimetic applications.
5.1.1. cis-Amide-Bond Mimetics. In natural peptides and proteins, most of the amide bonds in the peptide backbone are populated in their low-energy trans-conformation, which is used in the construction of α-helices and β-sheets. However, the ability to form a high-energy cis-conformation, especially for cisprolyl peptide bonds (∼6% of all Xaa-Pro peptide bonds in nature), may influence the secondary structure, as well as many processes such as the rate of protein folding. 116 In proteins, the cis-conformation can be stabilized by additional intramolecular hydrogen bonds and/or disulfide bonds, which is normally not the case for small peptides. The ability to mimic cis-amide bonds governs the possibility to study such biological systems and provides an opportunity for drug design and research. The 1,5disubstituted 1,2,3-triazole exhibits similar properties to a cisamide bond, particularly concerning the distance between the two adjacent α-carbons (3.2 Å in the 1,5-triazole versus 3.0 Å in a cis-amide bond) ( Figure 13) and thus constitutes a good mimic for a cis-amide bond, despite lacking hydrogen-bond donor abilities.
Appella and co-workers were among the first to demonstrate the synthesis of a 1,5-disubstituted 1,2,3-triazole amino acid (48, Scheme 25) for use as a cis-amide bond mimic, albeit not using RuAAC, in order to induce turn formation in an unnatural peptide in aqueous solution. 118 1,5-Triazole amino acid 48 was synthesized via a noncatalyzed thermal Huisgen cycloaddition in a five-step route with an overall yield of 31%. The synthesis of such derivatives was later improved using the RuAAC reaction, affording the corresponding triazole amino acid protected with Boc instead of Fmoc, in 75% yield over two steps. 119 Insertion of triazole 48 into an unnatural peptide sequence using solid-phase peptide synthesis (SPPS) leads to peptoid 49, (Scheme 25), which adapts a hairpin-like turn structure as the major conformer in 10 mM sodium phosphate buffer (pH 7.0), as revealed by detailed NMR experiments.
The synthesis and use of 1,5-disubstituted 1,2,3-triazoles as cisprolyl peptide bond mimics has been studied by Raines and coworkers, 120 who found that the Xaa-1,5-triazole-Ala unit mimics an Xaa-cis-Pro dipeptide remarkably well. By using the RuAAC reaction, a number of protected Xaa-1,5-triazole-Ala building blocks such as 50 and 51 were prepared in moderate-to-high yields (Scheme 26).
Bovine pancreatic ribonuclease A (RNase A), consisting of 124 residues, was used as a model protein to test the concept of 1,5triazoles as cis-peptide bond mimic. The region Gly112-Asn113-Pro114-Tyr115 in the protein sequence forms a Type VIb reverse turn where Asn113-Pro114 has a cis-peptide bond. This    dipeptide segment can be mimicked rather well by Xaa-1,5triazole-Ala. Triazoles 50 and 51 were incorporated in a peptide sequence corresponding to residues 95−124 by using standard solid-phase peptide synthesis. No epimerization was observed during the basic Fmoc-deprotection steps of the peptide synthesis. Residues 1−94 were created by recombinant DNA methods. After ligation of the two sequence parts, the protein was folded and purified, and the enzymatic activity was compared with the native ones. The two 1,5-triazole-containing enzymes obtained showed catalytic activities comparable to the wild-type analogues. It can be concluded that the Xaa-1,5-triazole-Ala building blocks are efficient mimics of the cis-peptide bond conformation in Xaa-Pro residues, as the catalytic activities are closely related to the formed protein tertiary structure. 121 A similar cis-prolyl bond mimetic has also been described by Tietze et al. and applied as a LeuPro cis-peptide bond mimetic in the synthesis of a mimic of the nickel superoxide dismutase (NiSOD) metallopeptide. 122 The Fmoc-protected 1,5-triazole amino acid 52 was synthesized using 4% Cp*RuCl(COD) as the catalyst, followed by deprotection of the tert-butyl ester, and was then inserted into heptapeptide 53 using standard SPPS. (Scheme 27) Metallopeptide 54 was obtained by treating triazole peptide 53 with 1 equiv of aqueous NiCl 2 solution and titration to pH 7.8.
Ferrini et al. have prepared a slightly different 1,5-triazole as a peptide backbone mimetic that exhibits the ability to mimic βturns. 95 As discussed in section 4.4.3, the 5-aminotriazole was obtained exclusively when using N-Boc-aminopropiolates in the RuAAC reaction. These 5-aminotriazoles are appealing as peptide backbone mimetics, as they are equipped with three functional groups that can be used for peptide synthesis or functionalization. Conformational analysis by computational methods (MM, QM, and MD) revealed that triazole 55 ( Figure  14) has some populated conformers satisfying reverse turn requirements, and one of the two most stable conformers shows the CO−HN hydrogen bond that is typical for β-turns.
Tischler et al. have also employed a 1,5-triazole as a cis-peptide mimic. 112 They targeted the Ile-cis-Pro amide bond in the highly potent sunflower trypsin inhibitor 1 (SFTI-1). 123 In the native form, its 14 amino acid backbone GRCTKSIPPICFPD is cysteine bridged and head-to-tail cyclized, and the Ile-cis-Pro amide bond is fundamental for bioactivity. By applying RuAAC in standard Fmoc SPPS (see section 4.6), 1,5-triazole cis-peptide bond mimics could be prepared, and the disulfide bridge was obtained by DMSO or air-mediated oxidation on the crude peptide. The trypsin-inhibition activity of the mimics was tested, and the most-potent compound, with a K i of 34 nM, was found to be the triazole-containing 56 ( Figure 15).

Peptide Side-Chain
Mimetics. In addition to mimetics of the peptide bond and peptide backbone, there are also side-chain mimetics, where the role of such mimetics can be to stabilize interactions or activated species made by the sidechain in the native peptide. Peptide side-chain mimetics are also frequently used to constrain peptide conformations and to stabilize secondary structures such as helices, commonly termed stapled peptides. 124 In this section the use of 1,5-triazoles as sidechain mimetics will be discussed.
Interesting stable phosphohistidine (pHis) mimetics and phosphoryltriazolylalanines (pTza) have been developed by Kee et al. 129 A cycloaddition of diethyl ethynylphosphonate with protected azidoalanine under both CuAAC and RuAAC conditions gave the desired pTza derivatives 58 and 59 in 72% and 68% yields, respectively (Scheme 29). According to molecular modeling, these compounds display favorable pHismimetic properties in terms of geometry and electronics. Particularly the hydrolytic liability of the original phosphohistidine moiety is overcome by the replacement of the labile N−P bond with a stable C−P bond.
With the two stable pHis isomers pTza 58 and 59 at hand, the possibility to study the function of pHis in biological systems was opened. It is known that histone H4 histidine kinase (HHK) activity is upregulated by 200-fold in human hepatocellular carcinomas compared to normal liver tissue. 130 Stabile pHis analogues could thus be useful tools, enabling the study of histone histidine phosphorylation. Peptides 60 and 61 (Scheme 30) were designed and synthesized by SPPS. These compounds correspond to the N-terminal tail of human histone H4 protein (residues 14−23), containing either the 1-pHis or the 3-pHis analogue instead of His18. 131 Triggered by the possibility that the pTza peptides 60 and 61 could be used in the generation of antibodies that can selectively recognize pHis in phosphoproteins, peptide 60 was employed as an immunogen to obtain rabbit polyclonal antibodies such as Ab-3pHis. Ab-3pHis is an example of an antibody that selectively recognizes a pHis-containing protein. Accordingly, Ab-3pHis was formed and its specificity was determined by peptide dot blot assays with a series of histone H4 tail peptides, harboring the two pTza isomers (60 and 61) as well as the His, pTyr, pSer, and pThr analogues. As Ab-3pHis only recognizes 60 and not 61 or the other phosphorylated analogues, no cross-reactivity toward His or other phosphoamino acids was demonstrated. Western blot analysis showed that Ab-3pHis selectively recognizes the histidine-phosphorylated histone H4 but not the nonphosphorylated counterpart.
Buysse et al. have investigated amino triazolo diazepines (Ata, Figure 16) as constrained histidine mimics and demonstrated their potential as histidine mimics by replacement of the His-Pro dipeptide in angiotensin IV and subsequent evaluation in two enzyme inhibition assays. 132 Two different strategies for the synthesis of such Ata-Xaa dipeptidomimetics were explored as seen in Scheme 31. Route A, involving the initial amide coupling of Boc-β-azido-Ala 62 with N-propargyl amino acid methyl ester 63 followed by a thermal Huisgen cycloaddition to form 64, was found to be unsatisfactory, due to not only low yields but also a considerable degree of epimerization. Route B, employing an intermolecular RuAAC reaction followed by intramolecular lactamization, proved superior both in terms of yield and suppressed epimerization after optimization of the reaction conditions. The Ata-Gly dipeptidomimetic 64 was applied to the angiotensin IV sequence, Val-Tyr-Ile-His-Pro-Phe, using standard SPPS. The activity of this modified sequence was determined by measuring the inhibition of insulin-regulated aminopeptidase (IRAP) and aminopeptidase-N (AP-N). Inhibition activitity of the peptidomimetic was in principle the same as that measured for the original peptide (pK i IRAP = 7.09 vs 7.14), concluding that Ata functions as a potent mimic for His.
Keller, Pyne, and co-workers have reported the synthesis of mono-, di-, and tripeptidomimetics, introducing a C-terminal triazole as an ester isostere, using both CuAAC and RuAAC. 133 The antibacterial activity of the compounds was evaluated, with a particular focus on the gastric anaerobe Clostridium dif f icile, where two dipeptides with 1,5-and 1,4,5-substitution patterns were among the most active.

α-Helix
Mimetics. Among the tremendous number of biological processes that occur in all living cells, a large portion of these involve protein−protein interactions (PPIs). The ability to control and modulate PPIs is of great significance for increased understanding of these processes, as well as for the development of new molecular therapeutics. The discovery of cell-permeable small-molecular inhibitors of PPIs is still a big challenge, partly due to the absence of natural small-molecular ligands, the inherent conformational flexibility of proteins, and also the lack of simple, suitable binding assays. Helical secondary structure motifs are the most abundant forms in proteins and correspond to over 30% of the structures. For that reason, it is rational to imagine that a significant population of PPIs should involve αhelices and that a mimic of a helix recognition face could act as a small-molecular competitive inhibitor. 134−137 α-Helix mimetics can be divided into three different categories. Type I mimetics are short oligomers that strive to replicate the local topography of an α-helix. These often involve peptide backbone mimetics with the aim to stabilize the helical conformation and to improve proteolytic stability; examples are stapled peptides and foldamers (foldamers are discussed in section 5.1.4). Type II mimetics are small nonpeptidic molecules that bind to a peptide receptor but do not necessarily mimic the original helix structure and rather serve as functional mimetics. 134 Type III mimetics are nonpeptidic scaffolds that mimic the surface formed by nonsequential hot-spot residues of the αhelix. 138−140 Attempts to use 1,5-triazoles in such type III α-helix mimetics have recently been reported by the groups of Konig 111,141 and Grøtli. 142 Both groups have designed and synthesized mimics bearing three side chains, which by calculated lowest-energy conformation mimic the orientation of the side chains in an α-helix at position i, i+3, and i+7. The Konig group used Cp*RuCl(PPh 3 ) 2 as catalyst in dioxane at 70°C to obtain the bistriazole scaffold 65 in good yield ( Figure 15). Grøtli and colleagues synthesized the 8-(1,5-triazolyl)-purine scaffold 66 under microwave conditions at 120°C in DMF, (Figure 17) aiming at fluoroscent PPI inhibitors of the MDM2/p53 interaction. Disappointingly, however, these compounds showed no activity in the fluorescence polarization (FP) assay. Thus, a biologically active 1,5-triazole-based α-helix mimetic still remains to be discovered.

Foldamers.
In the area of peptidomimetics, nonnatural peptidic oligomers having the capacity to mimic structural and folding properties of natural peptides are called "peptidic foldamers" or "foldamers". The field was originally initiated by two parallel papers published by Seebach et al. 143 and Gellman and co-workers in 1996. 144 By now, foldamers demonstrate diverse secondary structures and also very exotic side-chain chemistry for the non-natural amino acid-like building units. 145,146 Peptidic foldamers are useful tools in target validation and can also provide additional modalities when designing drugs for challenging target classes like PPIs. 147 As already mentioned above, because helical and turn structures are easily achieved by employing backbones involving cyclic insertions, 145,146 both the 1,4-and 1,5-substituted 1,2,3-triazole ring coupled with N-and C-terminal groups of amino acids have accordingly come up as novel "triazole amino acids" with an application potential as non-natural insertions into natural amino acid sequences. 14,148 The important abilities, such as stability against degrading enzymes or more controlled secondary structures, also hold for the 1,5-disubstituted 1,2,3-triazole amino acid insertions, which promotes their potential use for creating foldamer oligomers. By applying 4 mol % [Cp*RuCl 2 ] n in the RuAAC reaction, Johansson et al. 119 obtained the 1,5disubstituted 1,2,3-triazole amino acid 67 in excellent yield (Scheme 32). Accordingly, this amino acid was also oligomerized into triazole foldamers, resulting in short δ-peptide trimer and tetramer (68) derivatives showing diverse structural properties and having several conformations present simultaneously in solution phase. 119 The relatively rich set of stable conformers for the 1,5-disubstituted triazole amino acids, the on-purpose extended nature of their 1,4-regioisomers, 149 and their applicability in both standard peptide synthesis and expressed protein ligation 120 have already provided some initial studies toward the "foldamer direction". 119,150,151 The latter examples together also foreshadow a rapid increase in the number of new promising triazole foldamer constructs in the near future.

Macrocycles
In the past decade, the interest for macrocycles as drug leads has grown, 152,153 especially in the case of nonpeptidic macrocycles, due to their generally more-favorable properties, i.e., higher metabolic stability and permeability, which facilitate their use as oral drug candidates. Macrocycles populate a different chemical space than the space traditionally seen as druggable according to Lipinski's rule of five. 154 The turnlike structure of 1,5disubstituted 1,2,3-triazoles makes them excellent motifs for incorporation into macrocycles. 155 Two different strategies can be employed for the introduction of a 1,5-disubstituted triazole in a macrocycle. Either the triazole formation can be used in the macrocyclization step or the 1,5-triazole moiety can be introduced prior to the cyclization step. Both strategies have been utilized by several groups and will be discussed in this section.
Horne et al. have employed the 1,5-triazole as a cis-peptide bond mimetic in a conformational study of apicidin, a naturally occurring cyclic tetrapeptide inhibitor of histone deacetylases (HDACs). 109 They initially tried to apply the triazole formation as the macrocyclization step in order to access both the 1,4-and 1,5-triazoles (69 and 70) via the same linear intermediate 71, as shown in Scheme 33. A 2:1 mixture of 1,5-and 1,4-triazole was obtained when using thermal Huisgen cycloaddition conditions, but the 1,5-triazole 70 was isolated in only 8% yield. When RuAAC was instead applied to introduce the 1,5-triazole in the middle of the sequence, using solid-phase methodology, subsequent macrolactamization in solution after cleavage from the resin afforded the macrocycles 72−74 in 9−10% total yield (based on resin loading) over 8 steps and with >95% yield in the macrocyclization step (Scheme 34). The macrocyclization in this case is greatly favored by the turn conformation of the linear precursor 75 containing the 1,5-triazole, which brings the ends closer together.
Extensive NMR studies of the macrocycles revealed one single conformation in solution for 69, 72, and 73, but 73 formed the c-t-c-t conformation. The NMR study, together with enzyme activity data, suggests that the less-populated (15%) c-t-t-t conformation of apicidin is the more bioactive one.
The first report of a successful method for using the RuAAC as the macrocyclization step was published by the Marcaurelle group. 156 They desired a method that could produce nonpeptidic macrocycles, of ring size both n and n+1, from a single starting material by using different catalysts. After some optimization they found that [RuClCp*] 4 was more advantageous compared to the classic Cp*RuCl(PPh 3 ) 2 in terms of monomer/dimer ratio. Dimer formation is usually the most common issue in all macrocyclizations in solution. Dilution and increased temperature was also beneficial for the monomer/dimer ratio. The RuAAC conditions were optimized on azido alkyne 76, and with the optimized conditions in hand, a number of 11-, 12-, and 13membered macrocycles were synthesized in good yields. Some selected examples are shown in Scheme 35.
The same group then utilized this method in a large diversityoriented synthesis (DOS) library in their search for new macrocyclic histone deacetylase inhibitors (HDACs), 157 using a three-phase DOS strategy called build/couple/pair (B/C/P), a method described in more detail in section 5.7. 158 In the pair phase, different macrocyclic structures of a linear intermediate were subjected to different reactions, including RuAAC, to generate a total of 48 macrocyclic scaffolds ranging from 8-to 14membered ring systems. These 48 scaffolds were then finally further diversified on solid phase by utilizing SynPhase Lantern technology 159 to produce a library of more than 30 000 compounds. The RuAAC macrocyclization reactions were reliable and were routinely performed on 5−10 g scale in 50− 80% yield. An example is shown in Scheme 36.
Zhang et al. studied two different strategies for making triazole macrocycles in their synthesis of macrocyclic vancomycin mimics. 160,161 Macrocyclization by RuAAC was successful but with some difficulty for the smallest ring 77 (n = 1), due to the rigidity of the corresponding linear azido alkyne 78 (n = 1), giving rise to both dimer and trimer byproducts (Scheme 37).
However, when the 1,5-triazole was introduced in the linear sequence using 79, and the ring was closed via lactamization instead, the yield of the macrocyclization step was enhanced from 14% to 47% for the smallest ring size. The amount of dimer formation was equal, 28% vs 31%, for the two methods, but the trimer byproduct was not formed during the macrolactamization.
The same group also utilized a RuAAC double-macrocyclization in the synthesis of a 1,5-triazole-bridged vancomycin bicyclic mimic. 161   Disulfide bridges are very common and important in oligopeptides and proteins as they contribute to defining tertiary folding and conformational stability of proteins. One special group of naturally occurring peptides are called cysteine knot miniproteins 162 or cyclotides. 163−166 These are cyclic peptides with three cross-linked disulfide bridges. The disulfide bridges provide the cyclotide with extraordinary thermal stability and resistance against protolytic degradation. 167 The in vitro generation of disulfides bridges is usually achieved postsynthetically and is not always straightforward, especially for controlled regiospecific formation of such linkages.
Empting et al. have demonstrated successful mimicking of the disulfide bridge by introducing a 1,5-triazole into an analogue of the sunflower trypsin inhibitor-I (SFTI-1) 82 (Scheme 39). 110 Macrocyclization of the linear azido alkyne 83, using RuAAC on solid phase, yielded the triazole macrocycle 84 after deprotection and cleavage from the resin. Overlay of energy-minimized models of 84 with a previous solution structure of 82 (PDB code: 1JBN) 123 showed a comparable distance (4.00 vs 4.18 Å) between the two C α atoms of residue 3 and 11. The inhibition of trypsin-catalyzed proteolysis was also tested, and interestingly no significant drop in activity was seen. This methodology could be useful in the future in controlling folding of peptides where several disulfide formations are possible.
Isidro-Llobet et al. have also utilized the RuAAC reaction as a macrocyclization strategy in a diversity-oriented synthesis of macrocyclic peptidomimetics, 168 applying the three-phase DOS build/couple/pair strategy mentioned earlier. 158 In the pair phase, the RuAAC macrocyclization method developed by the Marcaurelle group 156,157   Fang et al. used RuAAC for macrocyclization to obtain 11-and 12-membered macrocycles, 171 using previously reported unoptimized conditions 156 but with a slightly higher temperature (Scheme 42). Macrocycles 96 and 97 were obtained in moderate-to-good yield, but it remains uncertain to what degree the yield was reduced by any dimer formation.
To conclude this section on 1,5-triazole-containing macrocycles, one can easily realize the utility of the 1,5-triazole as a building unit due to its natural turn conformation. As discussed in the beginning of this section, the two strategies to introduce the 1,5-triazole either prior to or during macrocyclization are both fruitful, as shown by the many examples discussed. The optimal strategy depends heavily on the nature of the linear precursor for a given macrocycle. In general, introduction of the 1,5-triazole prior to the macrocyclization increases the yield of the macrocyclization step due to its conformational preorganization. However, the use of RuAAC as a macrocyclization reaction can also be effective, but it is very much dependent on the flexibility of the linear precursor and also the size of the ring to be formed.

Nucleoside and Nucleotide Analogues
Nucleosides and nucleotides both contain a nucleobase (also called nitrogenous base) connected to a ribose or deoxyribose unit, and in the case of nucleotides also linked to one or more phosphate groups (Figure 18), providing many potential sites for derivatization. Synthetic compounds of this type are of interest, for instance, in the development of antivirals and anticancer agents. 172−174 A triazole unit can be employed to either derivatize or entirely replace the nucleobase moiety, or it may be appended to the base or to the sugar unit, to afford a nucleoside analogue. Subsequent phosphorylation also provides access to nucleotides. As for many other areas of drug discovery, reported applications of triazole incorporation mainly employ CuAAC, 175 while RuAAC has been less investigated in this context.  in the treatment of hepatitis, but also used to combat the respiratory syncytial virus (RSV). 176 Agrofoglio and co-workers prepared a set of ribavirin-like compounds by replacing the 1,3disubstituted 1,2,4-triazole ring in ribavirin with a 1,4-or 1,5disubstituted 1,2,3-triazole. 71 A benzoyl-protected β-azido ribose was subjected to RuAAC conditions using Cp*RuCl(PPh 3 ) 2 as the catalyst in THF (Scheme 43). Microwave heating was found to be more efficient than thermal heating in this case, and the desired derivatives were obtained in good-to-excellent yields using a reaction time of only 5 min. Minor amounts of the 1,4disubstituted isomer accompanied the formation of the desired 1,5-isomer, but this byproduct could be removed by chromatography. Unprotected azido ribose could also be applied in the reaction with excellent results. The prepared compounds were not found to display any activity toward the hepatitis C virus (HCV) upon screening, however. 176 The same group prepared nucleoside derivatives with a triazole appended to the C5 position of a 2′-deoxyuridine structure using the microwave conditions for RuAAC described earlier. 177 Compounds 98−101 were obtained in good yields (Scheme 44) and could be deacetylated using ammonia in methanol. While the corresponding 1,4-disubstituted derivatives, prepared using CuAAC, displayed some antiviral and cytostatic activity, the 1,5-isomers were inactive. Similar triazoles were also reported by Nielsen and co-workers for use in DNA−RNA duplexes. 178 5.3.2. Derivatization of the Ribose or Deoxyribose Unit via RuAAC. The attachment of a triazole to the sugar moiety instead of the nucleobase in the nucleoside structure has been reported by several groups, and the prepared structures are summarized in Figure 19. 27,93,179−182 The most extensive study in this area has been reported by Wang and co-workers, with the purpose of investigating the antiviral properties of nucleoside analogues. 180−182 3′-Azidothymidine (AZT, Figure 20) was one of the earliest drugs employed toward the human immunodeficiency virus (HIV), 183 but treatment with this drug can be accompanied by undesired side effects. As the AZT structure contains an azido functionality already at the outset, a natural strategy for derivatization would be the formation of triazoles via CuAAC or RuAAC. Previous attempts to investigate this pathway toward analogues in general did not afford compounds with antiviral activity, but the explored triazole side chains were in most cases aliphatic. However, arylsubstituted triazole-AZT derivatives were reported to show potent thymidine kinase 2 (TK-2) inhibition. 184 Inspired by this,  Wang and colleagues prepared a number of both 1,4-and 1,5disubstituted triazole-AZT derivatives containing aromatic substituents (see Figure 20 for two RuAAC derivatives). 180 Standard conditions (Cp*RuCl(PPh 3 ) 2 , THF, 60°C) were employed for the RuAAC reaction, but extended reaction times (1−2 days) were required to obtain the products in sufficient amounts. These compounds were then screened toward HIV-1, employing a cytopathic effect (CPE)-based assay. Many of the prepared compounds displayed antiviral activity. Especially 1,5substituted triazole compounds with bulky substituents, such as the 1-naphthyl derivative 102 (Figure 20), showed promising properties. Triphosphates of selected compounds were also prepared. Interestingly, silylation of the free hydroxyl group in derivatives such as 102 and 103 with tert-butyldimethylsilyl chloride afforded compounds displaying antiviral activity toward the Western Nile and Dengue viruses. 182 Having successfully derivatized the 3′-azidothymidine structure, Wang and colleagues then turned to the corresponding 4′azidothymidine (ADRT, Scheme 45), but then ran into problems. 181 While ADRT could be converted into various compounds using CuAAC, albeit involving long reaction times and heating to 60°C, the corresponding RuAAC reaction did not proceed at all despite extensive efforts to vary the reaction conditions. The azide in ADRT is tertiary as compared to the secondary azide in 3′-azidothymidine (Figure 18), and steric hindrance is believed to be the cause of this low reactivity.
Etheve-Quelquejeu and co-workers focused on the cycloaddition of 3′-azidoadenosine with chiral alkynes to form analogues of aminoacyl-tRNA. 179 Benzoyl-protected 3′-azidoadenosine 104 (Scheme 46) was reacted with two different chiral propargylic amine derivatives to form triazoles 105 and 106 in 37% and 39% yields, respectively, using standard conditions for the RuAAC reaction. Compound 105 was then coupled with commercially available deoxycytidine phosphoramidite to form dinucleotide 107 after deprotection. Ligation of 107 to an RNA microhelix completed the synthesis to form the desired aminoacyl-tRNA analogue.

Glycomimetics
Carbohydrates are important structural starting motifs in drug design as they are involved in inter-and intracellular communication in many organisms. 185 Moreover, carbohydrate analogues and mimetics are commonly exploited as transitionstate analogues for enzyme inhibition and as tools for understanding enzyme-inhibition mechanisms. 186 The use of CuAAC in carbohydrate chemistry has been recently reviewed. 187 The RuAAC reaction has been used as a method to couple carbohydrate moieties to each other as well as to connect saccharides to aromatics or other functionalized units, forming new carbohydrate derivatives. Several of the resulting compounds possess properties that make them promising as future therapeutic lead structures.
The first application utilizing the RuAAC reaction in a carbohydrate context was reported by Cintrat and co-workers, 93  The RuAAC reaction also can be used to couple two carbohydrate units to each other. The Crich group was the first to exploit the triazole motif as a linker in this manner, to provide a compound resembling a trisaccharide. 188 [Cp*RuCl] 4 was employed as the catalyst in DMF under microwave irradiation at 110°C to afford an azidomethyl glycoconjugate linked by a 1,5-disubstituted triazole in 75% yield (Scheme 48). The corresponding 1,4-substituted isomer was also synthesized using CuAAC.
A similar strategy has been applied by Opatz and co-workers, who in their search for metabolically stable glycomimetics demonstrated that pivaloylated and acetylated galactosylazides can be coupled with fucosylacetylene selectively, forming the corresponding 1,5-diglycosylated 1,2,3-triazoles (Scheme 49). 189 Cp*RuCl(COD) was employed as the catalyst under microwave irradiation at 100°C in DMA. Yields were 76% for the acetylated and 60% for the pivaloylated galactosylazides, respectively. The corresponding 1,4-isomers were obtained using the CuAAC reaction.
Multivalent glycoclusters can be synthesized by employing the RuAAC reaction as a key step. Uhrig, Kovensky, and co-workers have proved that a variety of multivalent thiogalactosides and thiolactosides can be formed using various azidosugars and carbohydrate-containing alkyne linkers. 190 With the aid of Cp*RuCl(COD) as the catalyst at room temperature in dioxane, a number of different glycoclusters, such as tetravalent compound 110 and even up to octavalent derivatives, were synthesized and evaluated as ligands for peanut lectin (Scheme 50). The yields varied depending on the substrate, stretching from 35% up to 82%.
The azide can be located in alternate positions of the saccharide ring. Besra and co-workers prepared triazole derivatives from 6″-azido-6″-deoxy-alpha-galactosyl ceramide (112) and phenyl acetylenes in 72−78% yield, employing 5 mol % Cp*RuCl(PPh 3 ) 2 as the catalyst (Scheme 52). 192 Both a terminal and a disubstituted internal alkyne were employed, with slightly lower yields for the product from the internal alkyne substrate.
2,3,6-Trideoxysugar triazole hybrids are yet another class of carbohydrate-like compounds that have been examined as potential drugs, more specifically as new broad-spectrum antimicrobial agents. The Shaw group synthesized a library of sugar−1,4-triazole conjugates for this purpose 193  (PPh 3 ) 2 as the catalyst. The same group also synthesized a class of bicyclic iminosugar hybrids, exploiting an intramolecular version of the RuAAC reaction as the key reaction step. 194 Both thermal and ruthenium-catalyzed conditions were examined, where the bicyclic triazole product 113 was produced in both higher yield and shorter reaction time using the Ru-catalyzed reaction (Scheme 53).

Natural Product Analogues
Various natural products have been employed as core structures for triazole-containing analogues, generally with the purpose of evaluating the biological properties of these new target compounds for comparison with the original molecule. Several groups have studied potentially cytotoxic derivatives, based on lignin, 195 alkaloid 196 and natural phenolic 197,198 scaffolds. Tron and co-workers envisaged that the same two building blocks, i.e., piperonyl alcohol and 3,4,5-trimethoxybenzyl alcohol, could be employed to prepare triazole derivatives of the cytotoxic lignans steganacin and podophyllotoxin. 195 RuAAC involving 114 and 115, using Cp*RuCl(PPh 3 ) 2 as the catalyst, afforded a precursor that could be oxidatively coupled to form derivative 116 ( Figure  22). The same strategy using 117 and 118 instead provided analogue 119 in a good overall yield. This methodology was not directly applicable toward the podophyllotoxin analogues, however. A modified reaction sequence involving a Friedel− Crafts reaction to effect the final ring closure allowed the formation of 120, albeit in a rather low yield, and the remaining compound 121 could not be prepared using either of these strategies. Compounds 116, 119, and 120 were all found to display some degree of cytotoxic activity upon screening.
Ferrocene-substituted chalcones have been connected to cinchona alkaloids using both CuAAC and RuAAC as reported by Csaḿpai and co-workers. 196 The ruthenium-catalyzed cyclization of 122 and 123, using Cp*RuCl(COD) as the catalyst, was accompanied by Ru-catalyzed epimerization at the C9 position, affording 124 in 48% yield, together with a byproduct where the azide had been reduced to an amino group (Scheme 54). Both 124 and the corresponding 1,4-isomer displayed an improved in vivo cytostatic activity toward HepC2 hepatoma and HT-29 colorectal adenocarcinoma human tumor cell lines in comparison to the reference compound Tamoxifen, with the 1,5-isomer 124 being the more active of the two.
Combrestatin A-4 is a cytotoxic natural product isolated from the bark of the tree Combretum caff rum, found in South Africa. 199 Hansen and co-workers had earlier prepared analogues of combrestatin, where the cis double bond, necessary for activity, had been replaced by a triazole ring in order to lock the two aromatic rings into a cis relationship. Both 1,4-and 1,5disubstituted triazoles were included in this previous study, 197 but the 1,5-compounds were prepared using magnesium acetylides. In a more recent study, seeking to investigate the effect of introducing a methylene group between the triazole nitrogen and the trimethoxy-substituted aromatic ring, compounds 125 and 126 were prepared in excellent yields using Cp*RuCl(COD) in refluxing benzene (Scheme 55). 198 Although the 1,5-disubstituted triazole derivatives were found to be more cytotoxic than their 1,4-counterparts, the introduction of the extra methylene carbon was not found to be beneficial for the biological activity.
Appendino, Di Marzo, and co-workers employed vanillin as a core unit in the preparation of synthetic capsaicinoids, where the amide unit of capsaicin (Scheme 56) was replaced by a triazole moiety to improve stability toward hydrolysis. 200 Vanillin was converted into protected vanillazides 127 and subsequently reacted with 1-decyne under Cu(I) or Ru(II) catalysis. While the copper-catalyzed conversion proceeded with excellent regioselectivity for the 1,4-isomer in both cases, the corresponding ruthenium-catalyzed reaction somewhat surprisingly afforded mixtures of the 1,4-and 1,5-disubstituted isomers of 128 and 129. The isomers could be separated by preparative highperformance liquid chromatography (HPLC), however, and the activity of the prepared compounds toward vanilloid and cannabinoid receptors was compared to that of capsaicin itself as well as other synthetic derivatives.
The fatty acid side chain of capsaicinoids may contain double bonds, and the presence of at least one such olefin unit in these compounds is believed to be important both for passive transport across cell membranes and for interaction with the vanilloid receptor. Triazoles can act as mimics for the geometrical constraints imposed by double bonds (see section 5.1.1); to investigate this feature, triazole analogues of olvanil, where the double bond was replaced by a triazole, were prepared via CuAAC and RuAAC (Scheme 56, only RuAAC reaction shown), and their vanilloid activity was evaluated. Both the 1,4-and 1,5isomers were found to be less active than olvanil itself. However, the 1,5-isomer 130 (Scheme 56), which more closely mimics the cis isomer of olvanil, was found to be more active than the corresponding 1,4-isomer, and this is in line with the activity of the two double-bond isomers of olvanil, where the cis isomer is more active than the trans isomer.
Analogues of naamine A, a natural product isolated from calcareous sponges, were prepared by Blache and co-workers, with the purpose of evaluating the ability of these compounds to control the formation of bacterial biofilms. 201 A set of compounds with variation of both the substitution pattern of the aromatic rings as well as the chain length between the triazole and the substituents were prepared from the corresponding azides and alkynes via RuAAC, using Cp*RuCl(PPh 3 ) 2 as the catalyst. Compound 131 (Figure 23) showed the most interesting antibiofilm activity, while the hydroxylated analogues were found to be inactive. 1,5-Disubstituted triazole derivatives of the triterpenoid oleanoilic acid have been prepared by Ben Jannet and coworkers using Cp*RuCl(PPh 3 ) 2 as the catalyst under microwave conditions. 202 The anticancer and anti-inflammatory properties of the compounds were evaluated, and one of the prepared 1,5isomers was found to show anticancer activity against murine breast and human colon cancer cells.

Target-Oriented Medicinal Chemistry
Over the last 10 years, the utilization of the RuAAC reaction in drug discovery has been noted as a growing trend in the literature. A diverse set of 1,5-triazole products can be found in  these reports, which cover a wide range of targets and target classes including various enzyme inhibitors, 203− 213 kinases, 214−217 proteases, 110,112 antivirals 218 (see also section 5.3), G-protein coupled receptors (GPCRs), 219 ion channels, 220−222 heat shock proteins, 95 and tRNA ligands. 223 1,5-Triazole derivatives have shown activity against numerous cancer cell lines, 205,215,224−227 as well as against the parasites Trypanosoma cruzi 70,228 and Plasmodium falciparum. 205 They have also been employed in the search for new antimicrobiotics 229 and antifungal agents. 230 Some selected examples are shown in Figure 24 to demonstrate the broad diversity of 1,5triazoles in target-oriented medicinal chemistry. Many of these 1,5-triazoles have been made without any rational design but rather to obtain structural diversity or by scaffold hopping.  Figure 25). 211 Overlay of the relevant lower-energy conformation of triazole 134 with an X-ray for the bound structure of 133 displayed potential for a good replacement, although a 3−4-fold drop in inhibition against HCV NS5B was seen. Linking different binders together into conjugates with dual or polypharmacological properties is another emerging area of interest for the pharmaceutical industry, especially in targeting strategies for drug delivery.
Two exciting examples using RuAAC as a linkage strategy have been reported. In the first example by Chardon et al., 225,231 some inspiring results were obtained when the RuAAC reaction was employed to link an ER ligand with a cytotoxic Pt II complex for potential targeting of hormone-dependent diseases (e.g., breast, colon, and ovarian cancers). A 1,5-triazole (135, Figure 24) was obtained in 41% yield using 8 mol % Cp*RuCl(PPh 3 ) 2 in THF at 70°C overnight. A CuAAC reaction of similar precursors failed to deliver the 1,4-triazole product, most likely because dihalocomplexes of palladium (and platinum) are known to react with terminal alkynes in the presence of a copper(I) catalyst and amine base as a method for preparing acetylide complexes with Sonogashira-type reactivity. 232,233 A 1,5-triazole displayed in vitro cytotoxic activity against several cancer cell lines; however, its selectivity toward positive estrogen receptor cancer cells has not yet been evaluated in vivo. The second example of a 1,5-triazole-linked dual inhibitor is triazole 136 reported by Ko et al. (Figure 26). 215 In this case, the selective c-Src kinase kinase (a key signaling kinase in cancer) inhibitor 137, reported by Brandvold et al., 216 was combined with the histone deacetylase (HDAC) inhibitor 138 to afford the chimeric Src/HDAC inhibitor 136, which is the first-in-class dual c-Src/HDAC inhibitor. Interestingly, both the c-Src and HDAC binding affinities of 136 increased compared to each corresponding smaller inhibitor. Chimera 136 also showed similar or better growth inhibition of several cancer cell lines in the National Cancer Institute's panel screen (NCI-60 panel) compared to vorinostat and dasatinib. 215 Wang et al. has recently published the use of chloromethyl triazoles (CMTs) as a new warhead for covalent inhibitor. 226 A range of CMTs were synthesized using RuAAC, and their activity against O 6 -alkylguanine DNA methyltransferase (MGMT) was determined, with 139 ( Figure  24) being identified as a potent and selective covalent inhibitor of MGMT. Labeling experiments of the C145A MGMT mutant and wild-type (WT) protein with a probe CMT also showed that these compounds bind exclusively to the active-site cysteine 145.

Diversity-Oriented Medicinal Chemistry
CuAAC and RuAAC provide convenient methods for elaborating a central molecular scaffold to increase chemical complexity, as use of either an azide or an alkyne as the point of derivatization, in combination with the possibility to vary the metal catalyst, can provide access to four different types of 1,2,3-triazole units appended onto the same scaffold.
The build/couple/pair (B/C/P) strategy, 158 already mentioned in section 5.2, involves the synthesis of molecular scaffolds (build phase), subsequent coupling of different combinations of these scaffolds using the same coupling reaction (couple phase), followed by the pair phase that instead employs different reactions that will define the final structures of the molecules prepared. Spring and co-workers have taken this method one step further in the formation of DOS-derived macrocycles, by employing different reactions in the couple phase, thus allowing for a higher degree of diversification when combining two scaffolds. 234 Macrocycles of a nonpeptidic nature are often difficult to prepare and are thus not well represented in compound collections employed for pharmaceutical-screening purposes. Azido building blocks, of different geometries and containing a fluorous tag to facilitate purification, 235 were constructed and then coupled, not using cycloadditions, but via aza-Wittig-type reactions. The RuAAC and CuAAC reactions were instead reserved for the final macrocyclization step, constituting the pair phase. A total of 73 macrocycles were prepared, and two examples, formed via RuAAC at high dilution in the final step, can be seen in Figure 27. In a later paper, 236 RuAAC was applied in an example of multidimensional DOS, where a functional group introduced during the build/couple/

Chemical Reviews
Review pair phases is reacted further, thus providing an additional possibility for diversification (i.e., build/couple/pair/modify).
Looking more at classical DOS methods involving RuAAC, Park and co-workers have described the synthesis of a benzopyran library, 237 also using solution-phase fluorous-tag chemistry. Described as privileged-substructure-based DOS (pDOS), with the central benzopyran scaffold being considered a privileged structure in drug discovery, 238,239 Park and colleagues prepared a set of 32 benzopyranyl triazoles with a 1,5-substitution pattern in high yields and purities, using Cp*RuCl(PPh 3 ) 2 as the catalyst (see Figure 28 for selected examples). Aryl azides were excluded from the study, as initial experiments showed that these substrates were problematic in RuAAC, in line with earlier reports by Lin, Jia, Fokin, and coworkers 27 (see section 4.3).
Another report on the use of RuAAC for DOS purposes comes from the group of Zondlo, 113 who employed this reaction in the context of proline edition, i.e., modification of a proline residue within a peptide chain with the purpose of studying the steric and stereoelectronic effects of this variation on the peptide conformation. A total of 123 different peptides were prepared by functionalization of a 4-hydroxyproline moiety within a peptide attached to a solid support. Diversification was effected using a wide range of both organic and organometallic reactions and included the formation of two diastereomeric 1,5disubstituted triazoles following conversion of the 4-hydroxyl unit of prolinol to an azide.

ORGANOCATALYSTS
Chiral pyrrolidines are often employed as organocatalysts, with their extensive application triggered by the original discovery by List and co-workers, showing that L-proline itself is an efficient catalyst for asymmetric aldol reactions. 240 Luo, Cheng, and coworkers investigated the organocatalytic asymmetric Michael addition of ketones and aldehydes to nitroolefins. 241 L-proline itself, while a viable catalyst, is known to afford low enantioselectivities in this reaction; 242 there were indications that substituted derivatives such as L-prolinol could be more successful. Seeking a convenient method for derivatizing a proline scaffold, the authors employed CuAAC to produce a library of 1,4-disubstituted triazoles to be used as organocatalysts for screening purposes. Two 1,5-disubstituted triazoles were also prepared (140 and 141, Scheme 57), but as this study was published only shortly after the initial report on the RuAAC reaction, the authors here employed the more classical magnesium acetylides (see section 2) to attain the 1,5-isomeric structure instead. Catalysts 140 and 141 both performed well in the reaction between cyclohexanone and 2-nitrovinylbenzene (Scheme 57), but only marginal differences were seen between 140 and the corresponding 1,4-substituted isomer 142, indicating that the space-shielding effect of the triazole ring itself was more important than the positioning of the substituents. Thiazolidine-based scaffolds were also explored but found to be inefficient as catalysts in the reaction.
Thioureas are also privileged structures in organocatalysis, and the bifunctional organocatalyst 143 (Figure 27), containing this moiety, was originally designed by Takemoto and co-workers for the enantioselective Michael addition of malonates to nitroolefins. 243 The catalyst contains both acidic and basic units and was constructed on the principle that these two sites will activate one reactant each, while simultaneously bringing them into close proximity to facilitate reaction. Seeking to expand this concept toward a trifunctional organocatalyst, Takasu, Azuma, and Takemoto constructed a set of triazole-substituted thioureas employing both RuAAC and CuAAC. 244 Although the RuAAC reaction is compatible with a wide spectrum of functionalities, the sulfur atom in a thiourea can ligate to the ruthenium catalyst, causing deactivation. A synthetic strategy where the thiourea unit was introduced in the last step was thus envisaged, and catalysts of the type 144 ( Figure 29) were prepared using a microwavemediated RuAAC reaction with [Cp*RuCl] 4 as the catalyst, followed by coupling reactions to introduce the thiourea in the final step. Screening of three of the catalysts in the enantioselective conjugate addition of cyclohexanone to transβ-nitrostyrene afforded up to 92% ee. In a later study, the scope of aryl-substituted nitroolefins was expanded, with a 2-furylsubstituted nitroolefin affording up to 98% ee using these organocatalysts. 245 NANOCHEMISTRY, AND ELECTRONIC DEVICES Triazole formation provides a convenient method for elaborating supramolecular structures, which often contain a complex central scaffold and require a simple and high-yielding reaction for the final step. CuAAC has been extensively used for this purpose, 246,247 while examples using RuAAC are more scarce. The aromatic properties of the triazole also make these units well-suited for inclusion into molecules intended for electronic devices or nanochemistry applications, where an extended conjugated system is often desired.
De Cola and colleagues have investigated the effect of introducing triazoles into iridium complexes bearing difluorophenylpyridine (F2ppy) ligands. 248 Metal complexes can be employed in light-emitting diodes (LEDs), acting as phosphorescent dopants and facilitating intersystem crossing from the singlet to the triplet excited state. The colors red, green, and blue are all necessary for applications in color displays, and De Cola and co-workers found that a complex prepared via RuAAC, using Cp*RuCl(PPh 3 ) 2 as the catalyst, displayed blue emission with high quantum yields ( Figure 30).
Theranostic agents combine both diagnostic and therapeutic properties into one entity, and seeking to develop contrast agents for optoacoustic imaging, Franchini and co-workers employed both copper-and ruthenium-catalyzed cycloadditions to link silver nanoparticles (NPs), derivatized with an alkyne unit, to lipophilic azide-functionalized gold nanorods (GNRs). 249 The cycloadducts were subsequently incorporated into PEG-based copolymers to produce polymeric nanoparticles (PNPs). These new constructs were found to display similar electroacoustic and optical behavior to GNR-PNPs without silver NPs, indicating that click-attachment of silver NPs does not disturb the desired properties.
Xie and co-workers employed both CuAAC and RuAAC to derivatize benzothiadiazoles (BTDs) intended for fluorescent chemosensors. 250 Comparing structure 145 with its 1,4disubstituted counterpart (Scheme 58), the authors found significant differences in the spectral properties of the molecules, where blue-shifts for 145 relative to the corresponding 1,4-isomer in the absorption and emission spectra were attributed to a lesser degree of conjugation in 145, due to a larger dihedral angle between the triazoles and the central BTD unit. Binding studies to various metal ions were also carried out. Sensors for metal binding have also been reported by Quayle and colleagues, who employed both CuAAC and RuAAC for clicking together a reporter unit (azo dye) with a metal-recognizing moiety (macrocycle) and a carrier molecule (carbohydrate). 251 Ruthenium(II) polypyridyl complexes, where one or two pyridyl ligands were replaced by 1,5-disubstituted triazoles, were synthesized by Schubert and co-workers, 252 and alkylation of the triazoles with methyl iodide also gave access to ruthenium triazolylidene complexes. The electrochemical and photochemical properties of 146−148 ( Figure 31) were investigated and complemented by computational studies using DFT. These compounds emit red light and could be of interest in electroluminescence devices, although evaluation in a dyesensitized solar cell gave modest efficiency in terms of power conversion. Red emissive triazole luminogens have also been prepared by Qin, Tang, and co-workers, using CuAAC and  RuAAC to adorn a pyrazine structure. 253 The compounds were shown to have good luminescence properties in the aggregate and solid state, which can be of importance in the development of optoelectronic devices for instance.
Tetrathiafulvalenes (TTFs) are frequently employed for the construction of molecular materials due to their electroactive properties. In a series of reports, Avarvari and co-workers have investigated the structures and properties of a TTF scaffold functionalized with 1,5-disubstituted triazole units. 254−256 By employing a tetrathiafulvalene with one or two appended alkyne units, compounds 149−154 ( Figure 32) were prepared via RuAAC, using Cp*RuCl(PPh 3 ) 2 as the catalyst in THF at 65°C . 254 Cyclic voltammetry and DFT calculations showed 149 and 150 to be more electron-deficient and less prone to oxidation than their 1,4-disubstituted counterparts. Cu(II) complexes of these ligands were prepared and characterized by X-ray crystallography, where the triazole N3 atom was found to bind to copper. These metal complexes are predicted to be useful starting materials for paramagnetic molecular conductors. In a second study, the solid-state structure of 149 and analogue 153 were compared. Interestingly, despite their similarities, these compounds crystallize in different fashions, where 150 forms head-to-tail dimers while 153 is arranged in columns with each unit slightly shifted in comparison to the molecule above and below so that the TTF moiety stacks with a triazole unit directly above. The effect of protonation and alkylation of the triazole ring in 149 was also studied, 256 and a triazolium salt was formed via treatment with trimethyloxonium tetrafluoroborate ((Me 3 O)BF 4 ). A third report focused on chelating pyridine triazole structures, i.e., 152 and 154, and their ability to act as ligands for metals such as cobalt and cadmium. 255 Lumpi et al. have also applied the alkyne−azide cycloaddition as a means to prepare new molecular materials, in this case nonlinear optic materials (NLOs) derived from a central enyne scaffold. 257 The majority of structures were formed using CuAAC, but one compound was prepared via microwaveassisted RuAAC using [Cp*RuCl] 4 as the catalyst (Scheme 59).
X-ray diffraction showed that this compound crystallized in an enantiomorphic space group and displayed slightly improved nonlinear susceptibility in comparison to the corresponding 1,4disubstituted isomer.
Dihydrazulenes (DHAs) can act as molecular photoswitches, where irradiation triggers a ring-opening to form a vinylheptafulvalene, which can then thermally return to the original structure. This feature, for instance, could be exploited in molecular electronics. Nielsen and co-workers studied the effect that appending a triazole onto the benzene ring of the dihydroazulene had on this isomerization process. 258 Compound 155 as well as its meta isomer were prepared via microwaveassisted RuAAC (Scheme 60). Despite attempts to optimize the reaction, the yields were rather low. The rate of ring-closure of these reactions was then measured and employed to determine the Hammett substituent constants for the meta-and para-tolyl isomers, as well as for the corresponding 1,4-disubstituted triazoles.

POLYMERS
CuAAC has found many applications in the area of polymer chemistry and generally has been employed for either "clicking" a molecule onto a polymeric scaffold derivatized with an alkyne or azide or employing the 1,3-cycloaddition for constructing the polymer itself. 259 In comparison with the field of medicinal chemistry, RuAAC as yet has been rather sparingly used in a polymer context, but examples include the synthesis of triazolecontaining monomers, polymers, and dendrimers using RuAAC, as well as polymer functionalization.

Triazole Monomers for Polymerization
Nulwala, Hawker, and co-workers compared the behavior of 1,4and 1,5-disubstituted vinyl triazoles 156 and 157 in living free radical polymerization (Scheme 61). 260 Monomers 156 were prepared in good yields by applying CuAAC to homopropargylic alcohol, followed by mesylation and elimination to introduce the vinyl functionality. While Fokin and co-workers have reported successful RuAAC reactions with hindered propargylic alcohols, 12,27 Nulwala and co-workers found homopropargylic alcohol to be a problematic substrate in this ruthenium-catalyzed reaction, and low yields of the triazole were obtained. Catalyst deactivation by the substrate was stated as the likely cause. By mesylating the alcohol before the RuAAC reaction, this problem  was solved and vinyl triazoles 157 were prepared in reasonable yields. Both monomer types were subsequently polymerized using two different methods, i.e., reversible addition−fragmentation chain transfer (RAFT) polymerization, affording homopolymers, and nitroxide-mediated polymerization (NMP), producing block polymers. As a general trend, the 1,5-isomer produced polymers with a higher glass-transition temperature and lower solubility than their 1,4-isomeric counterparts, indicating a more rigid structure. In a later study, the copolymerization of some of these monomers with styrene was also explored, with the RuAAC-derived monomers showing a higher reactivity than the 1,4-isomers in this case. 261 A sequential one-pot click-elimination procedure to access the same class of monomers has been reported by Blache and coworkers, starting from 4-bromobutyne and using sodium hydroxide in water to effect the elimination to a vinyl functionality in the final step. 262 Su and Hua have also reported the synthesis of vinyl 1,4-disubstituted triazoles, this time under acidic conditions. 263 Using a bimetallic catalyst system consisting of CuCl and RuCl 3 ·H 2 O, propargylic alcohols were reacted with alkyl, aryl, and benzyl azides and subsequently dehydrated in situ by trifluoroacetic acid present in the reaction mixture. Each of the two catalysts could be used separately, but the bimetallic catalyst increased the yield substantially. Bielawski and co-workers constructed polymer initiators containing 1,4-and 1,5-triazole isomers and subsequently grew poly(methyl acrylate) chains from these structures in order to study the mechanical properties of the formed polymers. 264 Ultrasound irradiation of these structures in acetonitrile resulted in a cycloreversion to the starting azide and alkyne, with a faster reaction rate for the 1,5isomer in comparison to the 1,4-derivative. A theoretical study of the cycloreversion of 1,4-and 1,5-disubstituted 1,2,3-triazoles has also been reported by Blank and co-workers. 265

Polymerization via RuAAC
The azide alkyne cycloaddition reaction can be employed as a method of polymerization itself, by linking monomers functionalized with azide and alkyne units. While this technique has been extensively applied using CuAAC, 259 there are as yet few examples of this type of application employing RuAAC. 266 Storey and co-workers compared the regioselectivity in the reaction of a diazide-functionalized derivative of bisphenol A (158) with a dialkyne-substituted tetraethylene glycol derivative (159) using thermal cycloadditions as well as CuAAC and RuAAC (Scheme 62). 267 While the copper-catalyzed click reaction could be performed under neat conditions with only the two monomers present, the solubility of Cp*RuCl(COD) in this bulk system at 70°C was not sufficient and a thermal, nonregioselective reaction was observed instead. However, solution-phase polymerization with the same catalyst in refluxing THF afforded the desired polymer with a 94:6 selectivity for the 1,5disubstituted triazole compared to the 1,4-isomer.
In another example, Brady et al. employed RuAAC to construct polymers containing Mo−Mo bonds, where the sensitivity of these metal−metal bonds constitutes a problem when conventional polymerization techniques are employed. 268 Building block 160 was first reacted with benzyl azide and with 1,4-bis(azidomethyl)benzene (161) in model reactions ( Figure  33). Employing Cp*RuCl(PPh 3 ) 2 as the catalyst, complete reaction was attained after 4 h. Reactions with diazidefunctionalized polystyrene and polyethylene glycol oligomers (162 and 163) were then pursued with good results, affording polymeric structures with the Mo−Mo bond still intact according to analysis of the CO bond in IR.

Polymer Functionalization
Fokin and co-workers have prepared halo-substituted triazoles using ruthenium catalysis, described in detail in section 4 Included in this study also was an example of RuAAC involving a soluble non-cross-linked functionalized polystyrene resin. Polymer 164, with 7% of the backbone carrying a pendant azide unit, was reacted with an excess of bromoalkyne 165, affording the bromotriazole-derivatized resin 166 (Scheme 63). IR analysis of the final polymer verified that the azide stretch had indeed disappeared, and 1 H NMR confirmed the presence of the amide methyl groups.
Pola et al. have explored the attachment of peptides to polymeric drug carriers using RuAAC. 269,270 Azide-terminated peptides were reacted with an aminopropargyl-functionalized copolymer of N-(2-hydroxypropyl)methacrylamide (HPMA) bearing the anticancer drug doxorubicin. The choice of catalyst was found to be important here. In a test reaction between 5azidopentanoic acid and the alkyne-functionalized polymer, using Cp*RuCl(PPh 3 ) 2 as catalyst at 60°C afforded only 25% of the polymer conjugate after 12 h. By switching to Cp*RuCl-(COD), which can be employed at ambient temperature, 27 full conversion was attained after 6 h, however, and this catalyst was subsequently used for the conjugation reactions.

Hyperbranched Structures
Tang and co-workers 271 and Brady and Tyler 272 have both targeted dendritic polytriazole structures via RuAAC. Tang and co-workers employed amine-centered triyne 167, combining this with diazides of different chain lengths, while Brady and Tyler selected 1,3,5-triethynylbenzene (168) as the central core for appending metal-containing building blocks, similar in structure to those described in section 8.2 ( Figure 34). Tang and coworkers found that Cp*RuCl(PPh 3 ) 2 , and also the Ru(III) oligomer [Cp*RuCl 2 ] n , could be employed in the coupling of 167 with diazides to produce the desired hyperbranched triazoles in good yields. The degree of branching was determined using 1 H NMR, and the ability of the polymers to form films and the photophysical properties of these films were studied also. Brady's attempted coupling of 168 with 169 was unsuccessful, however, affording instead a polymer devoid of molybdenum. The authors propose that a competing cyclotrimerization reaction, not involving 169, is taking place instead.
9. OTHER METHODS FOR THE FORMATION OF 1,5-DISUBSTITUTED 1,2,3-TRIAZOLES The use of other metals than Ru and Cu in cycloaddition reactions to form triazoles reaction has been reviewed recently by Astruc and co-workers 21 and will be mentioned only very briefly here. Iridium can catalyze the reactions of bromoalkynes with organic azides to 1,5-isomers with a 4-bromo substituent, 273 as well as can effect the cyclization of internal thioalkynes. 274 Wang et al. reported that cerium could be applied to access 1,5disubstituted 1,2,3-triazoles, in this case employing a nitroolefin instead of an alkyne together with benzyl or phenyl azide. 275 Ytterbium has been applied in one-pot cascade reactions, affording fused 1,4,5-substituted triazoles, 276 and samariumcatalyzed reactions have been reported to cyclize terminal alkynes with azides to provide the 1,5-isomer. 277 Iron can effect the cyclization of chalcones with sodium azide but affords the 2,4,5-trisubstituted 1,2,3-triazole product. 278 Koguchi and Izawa employed a copper catalyst in ionic liquids in a method that initially affords the 1,4-isomer with a protecting group in the 4position. 279 This triazole is subsequently alkylated on the 5position and deprotected to finally afford the 1,5-isomer. Lautens and co-workers used a combination of copper and palladium in a tandem reaction to form 1,4,5-trisubstituted triazoles, 280 while Smith and Greaney reported a zinc-mediated synthesis of 1,5and 1,4,5-isomers. 281 1,5-Disubstituted 1,2,3-triazoles can also be formed without the use of a metal. These methods have the advantage that an azide is not necessarily involved, and azide-free methods for 1,2,3-triazole formation have been highlighted recently. 282 A feature article by Dehaen and co-workers summarizes the use of organocatalysis for the formation of substituted 1,2,3-triazoles, and only a few examples will be mentioned here. 283 Azide-free multicomponent reactions have been applied; for instance, Wan et al. describe the use of enaminones with primary amines and tosylhydrazine in the presence of I 2 . 284 Heating to 110°C in DMSO afforded triazoles functionalized with an aldehyde or ketone in the 5-position. Another three-component method, albeit not azide-free, has been reported by Dey and Pathak and uses vinyl sulfones in conjunction with alkyl halides and sodium azide. 285 The same group also reports the use of selenium to access 1,4,5-trisubstituted 1,2,3-triazoles. 286 Fokin and coworkers also have reported a robust method for the reaction of aryl azides with terminal aryl alkynes in the presence of catalytic amounts of tetramethylammonium hydroxide at room temperature, affording 1,5-diaryl-substituted 1,2,3-triazoles in good yields. 287 For a full survey of triazole synthesis without employing ruthenium, we refer to the aforementioned reviews.

CONCLUSIONS AND FUTURE OUTLOOK
The ruthenium-catalyzed azide alkyne cycloaddition (RuAAC) provides direct access to 1,5-disubstituted 1,2,3-triazoles and acts as a counterpart to the copper-catalyzed "click" reaction where the 1,4-isomer of the triazole is formed. Many Ru(II) complexes catalyze the reaction, but a pentamethylcyclopentadienyl (Cp*) ligand as well as a chloride bonded to ruthenium are generally required to attain good regioselectivity. Mechanistic studies indicate a different pathway for RuAAC compared to CuAAC, and this is further substantiated by the fact that internal alkynes can be employed in the Ru-catalyzed reaction, which is not the case for CuAAC. Further advantages of RuAAC include the 1,5disubstitution pattern that can be exploited, for instance, in the formation of macrocycles, where the regiochemistry in the RuAAC reaction is better suited for smaller rings than the 1,4pattern obtained in CuAAC. The CuAAC reaction, however, has the benefit of using a less-expensive catalyst and can generally be performed at a lower temperature than RuAAC. Both reactions are compatible with a wide range of functional groups, although RuAAC can be problematic when using substrates containing acidic functional groups.
One can easily realize the potential of 1,5-triazoles in various fields of medicinal chemistry and biochemistry, both as tools as well as inhibitors, and the RuAAC reaction has found more widespread use among medicinal chemists than in any other area of research. However, a derivative prepared via RuAAC for clinical purposes has yet to emerge in a drug on the market. Evaluation of the 1,5-disubstituted triazoles as peptide bond or disulfide mimetics, or as α-helix mimetics, is also of interest, and several studies in this area have been reported. Other areas of chemistry are surprisingly underdeveloped in terms of applying the RuAAC reaction, and there is ample scope for further work in many fields, including the development of new materials and electronic devices, as well as the use of 1,5-disubstituted triazoles as ligands in coordination chemistry and catalysis. A commercially available polymer-supported catalyst for RuAAC also could also be of interest. We envisage many new examples of RuAAC applications in the future.

Notes
The authors declare no competing financial interest.