Improved Synthesis and Coordination Behavior of 1 H -1,2,3-Triazole-4,5-dithiolates (tazdt 2 − ) with Ni II , Pd II , Pt II and Co III

: A new synthetic route to 1 H -1,2,3-triazole-4,5-dithiols (tazdtH 2 ) as ligands for the co-ordination of Ni II , Pd II , Pt II and Co III via the dithiolate unit is presented. Different N-protective groups were introduced with the corresponding azide via a click-like copper-catalyzed azide-alkyne [3 + 2] cycloaddition (CuAAC) and fully characterized by NMR spectroscopy. Possible isomers were isolated and an alternative synthetic route was investigated and discussed. After removal of the benzyl protective groups on sulfur by in situ-generated sodium naphthalide, complexes at the [(dppe)M] (M = Ni, Pd, Pt), [(PPh 3 ) 2 Pt] and [( η 5 -C 5 H 5 )Co] moieties were prepared and structurally characterized by XRD analysis. In this process, the by-products 11 and 12 as monothiolate derivatives were isolated and structurally characterized as well. With regioselective coordination via the dithiolate unit, the electronic inﬂuence of different metals or protective groups at N was investigated and compared spectroscopically by means of UV/Vis spectroscopy and cyclic voltammetry. Complex [( η 5 -C 5 H 5 )Co( 5c )] ( 10 ), is subject to a dimerization equilibrium, which was investigated by temperature-dependent NMR and UV/Vis spectroscopy (solution and solid-state). The thermodynamic parameters of the monomer/dimer equilibrium were derived.


Introduction
The award of the Nobel Prize to Sharpless, Meldal and Bertozzi in 2022 represents an accolade for click chemistry as a powerful synthetic method [1]. The concept of click chemistry was established as early as 2001 and describes a rapid and precise synthesis of molecules following the example of nature. The advantages of the method are high atomic efficiency, very few by-products and high yields while only the use of cheap and simple chemicals and short reaction time are needed [2]. Classically, click chemistry often includes Diels-Alder reactions, addition reactions on carbon-carbon double bonds, and especially copper-catalyzed Huisgen cycloaddition, which can be used for the synthesis of 1H-1,2,3-triazoles [3][4][5]. Sharpless and coworkers presented first protocols for the [3 + 2] cycloaddition of azides with terminal alkynes under Cu-catalyzed reaction conditions [5]. A [3 + 2] cycloaddition between azides and acetylenes are not regioselective [6][7][8]. Two regioisomeres with a substituent in 4-or 5-position are formed. Only in the case of electrophilically activated acetylenes is high regioselectivity possible [5,9,10]. The copper-catalyzed azide-alkyne [3 + 2] cycloaddition (CuAAC) opens a way for the regioselective synthesis of triazoles. In addition to various alkyl and aryl substituents, donors such as phosphanes, amines, sulfur and seleniums could be introduced into the 1H-1,2,3-triazole system as well [11][12][13][14][15][16]. Introduction of thiol groups at both 4-and 5-position of the triazole would result in a new ligand with five potential coordination sites in the form of the dithiolene unit and the N atoms. Both through the aromatic properties of the 1H-1,2,3-triazole ring and through the specific electronic situation of the dithiolene unit, the 1H-1,2,3-triazole-4,5-dithiolate (tazdt 2− ) could serve as a versatile bridging ligand between several metal centers. In particular, the electronic properties appear potentially interesting due to the non-innocent character of the dithiolene unit [17][18][19]. In contrast to many other triazoles, a synthesis of 1H-1,2,3-triazole-4,5-dithiols by means of a click-like copper-catalyzed azide-alkyne [3 + 2] cycloaddition is not known to the best of our knowledge. So far, synthesis of 1H-1,2,3-triazole-4,5-disulfides was reported in a Ru-catalyzed [3 + 2] cycloaddition of an azide and a bis(alkylsulfanyl) acetylene at high temperatures under inert gas atmosphere [16,20]. Alternatively, this synthesis can be carried out with [(NHC)CuI] (NHC = 1-benzyl-3-n-butyl-1H-benz[d] imidazolylidene) as catalyst. The latter is easier to use, but the yields are lower compared with the Ru-based catalyst. In addition, 1H-1,2,3-triazoles have been synthesized in an Ir-catalyzed [3 + 2] cycloaddition of internal mono(alkylsulfanyl)alkynes with an azide [21]. Herein, we present a substantially improved synthesis of 1H-1,2,3-triazole-4,5-disulfides under CuAAC click conditions using the terminal benzylsulfanylacetylene. Pitfalls of the reductive removal of S-protective benzyl groups are identified by isolation of respective thiolato complexes. Finally, we describe coordination of the corresponding dithiols to group 10 metals and Co III . The influence of the metal and the N-protective groups at the triazole on the electronic properties will be discussed.

Chemical Reagents and Instruments
Materials, details on physical measurements, X-ray determination data, original NMR and IR spectra of all products and preparative procedures as well as spectroscopic data of the only organic products (1)(2)(3)(4) are provided in the ESI.

Synthetic Protocols 2.2.1. General Synthesis of 5
A solution of 2a-c (1 mmol) in THF (50 mL) was treated with sodium (5 mmol) and naphthalene (2.5 mmol). The red-brown suspension was stirred overnight, then cooled to 0 • C. MeOH (10 mL) was added and the mixture was stirred until gas evolution ceased. For purification, the solution was dried in vacuo, taken up in H 2 O (40 mL) and washed three times with Et 2 O (10 mL aliquots). The aqueous fraction was filtered over celite in a G3 frit and subsequently acidified with aqueous HCl (pH = 3-4), leading to the formation of a beige precipitate. The suspension was extracted four times with CH 2 Cl 2 (aliquots of 10 mL). The organic fraction was dried over Na 2 SO 4 , filtered and dried in vacuo to isolate 5 as crude products. According to NMR, the samples are not analytically but sufficiently pure for successful complex synthesis. Potential by-products were characterized in the form of stable complexes as well (see compounds 11 and 12).

General Synthesis of the Metal Complexes 6 and 7
In a 50 mL flask 1 equivalent [(dppe)MCl 2 ] (M = Ni, Pd) was suspended in 15 mL H 2 O. A solution of 1.1 equivalents 5b in CH 2 Cl 2 (25 mL) and 3 equivalents KOH were subsequently added. In the two-phase system, a color change from red to green (Ni) or colorless to violet (Pd) was observed in the lower phase. The reaction system was stirred for 3 days at room temperature. To purify the product, the aqueous phase was removed and the organic fraction washed four times with H 2 O (15 mL), dried over Na 2 SO 4 and filtered and the solvent was removed in vacuo. A column chromatographic purification was carried out with a CH 2 Cl 2 /MeOH solvent mixture (20/1) as a mobile phase. Suitable crystals for X-ray structure analysis were obtained from a CH 2 Cl 2 solution by slow diffusion of n-pentane.
After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 • C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1).
In the 1 H NMR spectra, 2 new signals were observed at a chemical shift between 3.55 ppm and 3.79 ppm for the CH 2 protons of the introduced benzyl group, while the triazole proton of 1a-g between 7.05 ppm and 7.68 ppm had disappeared ( Figures S32-S46). In the case of compound 1g, the introduction of sulfur at 5-position failed. Chemistry 2023, 5, FOR PEER REVIEW 6 Scheme 1. CuAAC reaction to build 4-benzylsulfanyl-1H-1,2,3-triazole, subsequent introduction of a second sulfide group and reductive removal of the S-benzyl groups to form the free dithiol derivatives. After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1). Scheme 1. CuAAC reaction to build 4-benzylsulfanyl-1H-1,2,3-triazole, subsequent introduction of a second sulfide group and reductive removal of the S-benzyl groups to form the free dithiol derivatives. Table 1. List of N-protective groups and respective yields with regard to Scheme 1 (The letters in column 1 refer to the different N-R triazole derivatives in Scheme 1).

R 1 2
a Chemistry 2023, 5, FOR PEER REVIEW 6 Scheme 1. CuAAC reaction to build 4-benzylsulfanyl-1H-1,2,3-triazole, subsequent introduction of a second sulfide group and reductive removal of the S-benzyl groups to form the free dithiol derivatives. After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1).
Chemistry 2023, 5, FOR PEER REVIEW 6 Scheme 1. CuAAC reaction to build 4-benzylsulfanyl-1H-1,2,3-triazole, subsequent introduction of a second sulfide group and reductive removal of the S-benzyl groups to form the free dithiol derivatives. After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1).
c Chemistry 2023, 5, FOR PEER REVIEW 6 Scheme 1. CuAAC reaction to build 4-benzylsulfanyl-1H-1,2,3-triazole, subsequent introduction of a second sulfide group and reductive removal of the S-benzyl groups to form the free dithiol derivatives. After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1). After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1).

Xy
49% 38% e Chemistry 2023, 5, FOR PEER REVIEW 6 Scheme 1. CuAAC reaction to build 4-benzylsulfanyl-1H-1,2,3-triazole, subsequent introduction of a second sulfide group and reductive removal of the S-benzyl groups to form the free dithiol derivatives. After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1).
Bn 80% 65% f Chemistry 2023, 5, FOR PEER REVIEW 6 Scheme 1. CuAAC reaction to build 4-benzylsulfanyl-1H-1,2,3-triazole, subsequent introduction of a second sulfide group and reductive removal of the S-benzyl groups to form the free dithiol derivatives. After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1).  After purification by column chromatography, the N-protected 1H-1,2,3-triazole-4monosulfides were isolated in yields of 36% to 97% (Table 1) and were characterized by NMR spectroscopy. It should be noted that the regioselective cyclization led exclusively to the 4-sulfido derivative, which is in accord with observations of Meldal and Sharpless. [5,24] The introduction of the second sulfur substituent is carried out analogously to synthesis of bis(benzylsulfanyl)acetylene described in the literature. [25] For this purpose, the corresponding triazoles 1a-g were deprotonated with n-butyllithium at −78 °C, reacted with elemental sulfur and subsequently trapped with benzyl bromide (Scheme 1). After purification by column chromatography, the corresponding triazoles 2a-e were isolated in yields between 38% and 89% (Table 1).

2-Pic
91% -Due to the electron-withdrawing pyridine substituent in the 2-picolyl protective group, the acidity of the methylene proton is higher than that of the triazole proton. Accordingly, deprotonation and subsequent methylation with MeI occurs at the N-2-picolyl group to give 3, as can be observed by the doublet 1 H NMR signal at 1.88 ppm for the methyl group attached to the N-protective group ( Figure S49). Also in a [3 + 2] cycloaddition of bis(benzylsulfanyl)acetylene and 2-picolyl azide with CuSO 4 /NaAsc as catalyst 2a was not isolated. A terminal acetylene is necessary for an end-on coordination of the Cu I to catalyze the [3 + 2] cycloaddition [10].
Nevertheless, this new two-step method for the generation of a disulfide unit on the 1H-1,2,3-triazole shows clear advantages in comparison with the synthesis described in the literature. Thus, sensitive and expensive catalyst systems [(NHC)CuI] and [(η 5 -C 5 Me 5 )(cod)RuCl] are dispensable [16]. Moreover, anaerobic and anhydrous conditions are not necessary in the first reaction steps and the overall yields are higher. While Schallenberg et al. achieved a yield of 39% with the benzyl group, a yield of 65% was realized with the new route [16]. Accordingly, it was also investigated whether the disulfide unit can be introduced stepwise into a 1,2,3-triazole by the direct method. For this purpose, the unsubstituted 1-(4-methoxybenzyl)-1H-1,2,3-triazole was deprotonated with n-butyllithium and subsequently reacted with elemental sulfur and benzyl bromide for alkylation (Scheme 1). After chromatographic purification, the 1 H NMR spectrum of the isolated product 4 revealed a methylene singlet at 3.67 ppm and a triazole proton at 7.48 ppm, indicating introduction of the sulfur in 5-instead of 4-position ( Figure S52). Interestingly, a preference for the 5-substituted derivatives was also observed by Fokin et al. by rutheniumcatalyzed [3 + 2] cycloadditions of terminal alkynes with azides [26][27][28]. The regioselective deprotonation can be explained by the greater stabilization of the carbanion in 5-position due to resonance ( Figure 1). Consistently, a subsequent introduction of the second sulfur substituent at 4-position by the same procedure proved unsuccessful. Respective attempts always led to the recovery of the starting material, which can be attributed to a lack of resonance stabilization in the carbanion.
group, the acidity of the methylene proton is higher than that of the triazole proton. Accordingly, deprotonation and subsequent methylation with MeI occurs at the N-2picolyl group to give 3, as can be observed by the doublet 1 H NMR signal at 1.88 ppm for the methyl group attached to the N-protective group ( Figure S49). Also in a [3 + 2] cycloaddition of bis(benzylsulfanyl)acetylene and 2-picolyl azide with CuSO4/NaAsc as catalyst 2a was not isolated. A terminal acetylene is necessary for an end-on coordination of the Cu I to catalyze the [3 + 2] cycloaddition [10].
Nevertheless, this new two-step method for the generation of a disulfide unit on the 1H-1,2,3-triazole shows clear advantages in comparison with the synthesis described in the literature. Thus, sensitive and expensive catalyst systems [(NHC)CuI] and [(η 5 -C5Me5)(cod)RuCl] are dispensable [16]. Moreover, anaerobic and anhydrous conditions are not necessary in the first reaction steps and the overall yields are higher. While Schallenberg et al. achieved a yield of 39% with the benzyl group, a yield of 65% was realized with the new route [16]. Accordingly, it was also investigated whether the disulfide unit can be introduced stepwise into a 1,2,3-triazole by the direct method. For this purpose, the unsubstituted 1-(4-methoxybenzyl)-1H-1,2,3-triazole was deprotonated with n-butyllithium and subsequently reacted with elemental sulfur and benzyl bromide for alkylation (Scheme 1). After chromatographic purification, the 1 H NMR spectrum of the isolated product 4 revealed a methylene singlet at 3.67 ppm and a triazole proton at 7.48 ppm, indicating introduction of the sulfur in 5-instead of 4-position ( Figure S52). Interestingly, a preference for the 5-substituted derivatives was also observed by Fokin et al. by ruthenium-catalyzed [3 + 2] cycloadditions of terminal alkynes with azides [26][27][28]. The regioselective deprotonation can be explained by the greater stabilization of the carbanion in 5-position due to resonance ( Figure 1). Consistently, a subsequent introduction of the second sulfur substituent at 4-position by the same procedure proved unsuccessful. Respective attempts always led to the recovery of the starting material, which can be attributed to a lack of resonance stabilization in the carbanion. To enable coordination via dithiolene unit, the benzyl protective groups on sulfur must be removed. Due to having the best yields, compounds 2a-c were used for coordination experiments. As we previously reported, this could readily be achieved by reductive removal with elemental sodium in presence of naphthalene in THF [16]. After an acidic work-up, the corresponding dithiols 5a-c were isolated as yellow oils in reasonable yields (Scheme 1). The samples are not analytically but sufficiently pure for coordination experiments (vide infra).

Synthesis of Metal Complexes
Coordination experiments with 1H-1,2,3-triazoles-4,5-dithiols were performed with particular attention to the regioselective dithiolate over N-coordination. The dithiols H2-5a-c were reacted with the first-row and group-10 transition metals Co III , Ni II , Pd II and Pt II . The Co III complex 10 was synthesized by reacting the ligand H2-5c with [(η 5 -C5H5)Co(CO)I2] in THF in presence of NEt3 (Scheme 2). The reaction progress could be observed by a decrease of the CO band in IR spectroscopy and the reaction solution turning blue. To enable coordination via dithiolene unit, the benzyl protective groups on sulfur must be removed. Due to having the best yields, compounds 2a-c were used for coordination experiments. As we previously reported, this could readily be achieved by reductive removal with elemental sodium in presence of naphthalene in THF [16]. After an acidic work-up, the corresponding dithiols 5a-c were isolated as yellow oils in reasonable yields (Scheme 1). The samples are not analytically but sufficiently pure for coordination experiments (vide infra).

Synthesis of Metal Complexes
Coordination experiments with 1H-1,2,3-triazoles-4,5-dithiols were performed with particular attention to the regioselective dithiolate over N-coordination. The dithiols H 2 -5a-c were reacted with the first-row and group-10 transition metals Co III , Ni II , Pd II and Pt II . The Co III complex 10 was synthesized by reacting the ligand H 2 -5c with [(η 5 -C 5 H 5 )Co(CO)I 2 ] in THF in presence of NEt 3 (Scheme 2). The reaction progress could be observed by a decrease of the CO band in IR spectroscopy and the reaction solution turning blue.
In contrast to the free dithiol H 2 -5c, the corresponding complex could be purified by flash chromatography, such that a dark purple compound was isolated and identified as the Co-complex 10. Further, the dppe-complexes 6 and 7 with group-10 metals were obtained either by reaction of H 2 -5b in a two-phase system (CH 2 Cl 2 /H 2 O) with KOH and the precursors [(dppe)MCl 2 ] {M = Ni, Pd; dppe = 1,2-bis(diphenylphosphino)ethane} or with [(dppe)PtCl 2 ] and [(PPh 3 ) 2 PtCl 2 ], respectively, in MeOH using NaOMe as a base. After aqueous work-up and chromatographic purification, a green Ni compound (6), a reddish Pd compound (7) and yellow Pt compounds (8 and 9a-c) were isolated.
In contrast to the free dithiol H2-5c, the corresponding complex could be purified by flash chromatography, such that a dark purple compound was isolated and identified as the Co-complex 10. Further, the dppe-complexes 6 and 7 with group-10 metals were obtained either by reaction of H2-5b in a two-phase system (CH2Cl2/H2O) with KOH and the precursors [(dppe)MCl2] {M = Ni, Pd; dppe = 1,2-bis(diphenylphosphino)ethane} or with [(dppe)PtCl2] and [(PPh3)2PtCl2], respectively, in MeOH using NaOMe as a base. After aqueous work-up and chromatographic purification, a green Ni compound (6), a reddish Pd compound (7) and yellow Pt compounds (8 and 9a-c) were isolated.
In contrast to the free dithiol H2-5c, the corresponding complex could be purified by flash chromatography, such that a dark purple compound was isolated and identified as the Co-complex 10. Further, the dppe-complexes 6 and 7 with group-10 metals were obtained either by reaction of H2-5b in a two-phase system (CH2Cl2/H2O) with KOH and the precursors [(dppe)MCl2] {M = Ni, Pd; dppe = 1,2-bis(diphenylphosphino)ethane} or with [(dppe)PtCl2] and [(PPh3)2PtCl2], respectively, in MeOH using NaOMe as a base. After aqueous work-up and chromatographic purification, a green Ni compound (6), a reddish Pd compound (7) and yellow Pt compounds (8 and 9a-c) were isolated.
In addition to the main products, by-products were surprisingly isolated from the reaction mixtures with the crude dithiol H2-5a and corresponding metal precursors (Scheme 3). From the reaction with [(η 5 -C5H5)Co(CO)I2], a tetranuclear complex 11 and from the reaction with [(PPh3)2PtCl2] the by-product 12 were isolated and crystallized. Scheme 3. Coordination to by-products 11 and 12. Scheme 3. Coordination to by-products 11 and 12.

Molecular Structure of the Complexes
The molecular structures of all complexes 6-12 were determined by single-crystal XRD analysis (Figures 2, 3, S4 and S5). With the exception of the complexes 11 and 12, which are by-products, all complexes exhibited an exclusive dithiolato coordination. The molecular structures of the group-10 metals showed the expected square planar geometry, including a planar dithiolate unit. The deviation from the SCCS planarity fell between 1.0(5) • and 3.1(3) • , which is very much in accordance with the values described in the literature [29]. Table 2 lists selected bond lengths and angles. In comparison to classical dithiolene complexes, a larger obtuse bite angle and, related to that, somewhat longer metal-sulfur bonds are evident [30][31][32][33]. The former follows the geometric requirements of a five-membered backbone ring, in which a regular internal angle leads to a formal C-C-S angle of 126 • . In addition, comparison of the metric parameters in compounds 9a and 9b does not show any influence by the protective group on nitrogen in the bonding situation at the dithiolate unit. literature [29]. Table 2 lists selected bond lengths and angles. In comparison to classical dithiolene complexes, a larger obtuse bite angle and, related to that, somewhat longer metal-sulfur bonds are evident [30][31][32][33]. The former follows the geometric requirements of a five-membered backbone ring, in which a regular internal angle leads to a formal C-C-S angle of 126°. In addition, comparison of the metric parameters in compounds 9a and 9b does not show any influence by the protective group on nitrogen in the bonding situation at the dithiolate unit.     Moreover, when replacing the metal center from Ni II (6) to Pd II (7) or Pt II (8), the dithiolate moiety does not show significant differences in the bond lengths C1-C2 with 1.368(6) Å to 1.381(5) Å or C1-S1 and C2-S2, which are between 1.725(5) Å and 1.748(6) Å. On the other hand, the M-S bond lengths show a distinct elongation by going from Ni II to Pd II and Pt II , which is essentially related to the increasing size of the metal atom. However, the bond lengths Pd-S in 7 {2.354(2) Å and 2.334(1) Å} and Pt-S in 8 {2.349(1) Å and 2.335(1) Å} are virtually equal. This effect is well-known and is attributed to the relativistic effect of the Pt atom and the resulting shrinking of the d orbitals [34].
The molecular structure of 10 in the solid state reveals a dimerization, in which not only is the Co III center coordinated by one dithiolate unit, but a third sulfur atom of a neighboring dithiolate moiety is bound to cobalt and vice versa. The observed dimerization to (10) 2 can be rationalized by fulfilling the 18 valence electron rule. On the other hand, the monomer constitutes a 16 valence electron complex, which is less stable but more readily solvated due to the free coordination site. Such dimerization equilibria are regularly observed in related [(η 5 -C 5 H 5 )Co(dithiolene)] complexes [32,33,[37][38][39][40].
A by-product of the reaction of H 2 -5a with [(η 5 -C 5 H 5 )Co(CO)I 2 ] was isolated after chromatography and crystallization. The crystal structure of 11 undisclosed an unexpected tetranuclear complex, in which the Co III ions are linked in a cyclic fashion by N-4-methoxybenzyl-1,2,3-tiazole-5-thiolate ligands ( Figure 2). Herein, each Co III is coordinated by a thiolate of one triazole and by a nitrogen atom in the third position of another. The coordination sphere of each Co III center is saturated by one iodide and one η 5 -C 5 H 5 ligand. This structural motif uncovered the loss of one thiolate substituent at 4-position of the 1,2,3-triazole ligand.
Likewise, the triazole ligands in the by-product 12 do not contain a dithiolate unit. Instead, the two triazole ligands in 12, next to two trans-standing triphenylphosphine ligands, are coordinated via one remaining thiolate in 4-position in a quadratic planar geometry around a Pt II center. A comparison of complex 12 with 9a with respect to the influence of cis/trans configuration is interesting, because the ligands are highly similar. The trans arrangement leads to longer Pt-P1/P1* bonds (2.3220(8) Å) in 12 compared to the cis arrangement in 9a with Pt-P1/P2: 2.2853(7) Å/2.2944(7) Å, which reflects some symbiotic π-bonding effect in 9a. The successful isolation of low-yield by-products 11 and 12 indicate limitation of side reactions in the reductive removal of the thiol protective groups. Remarkably, the cleavage of the whole benzylthiolate is possible both at 4-and 5-position.

NMR Spectroscopy of Metal Complexes
The phosphine ligands in the complexes 6-8 and 9a-c are valuable probes for the electronic situation of the metal, which can be investigated by 31 P NMR spectroscopy. The Ni complex 6 as well as the Pd compound 7 show two doublets at chemical shifts of 58.7/60.5 ppm, and 56.1/58.4 ppm, respectively. The observed doublets result from the C 1 symmetry and the related chemical non-equivalence of the phosphorus atoms. Consistently, a slightly smaller coordination chemical shift ∆δ of the Pd-dppe signals is combined with a lower 31 P/ 31 P coupling constant of 18.0 Hz. The Ni-dppe complex 6 shows a substantially larger coupling constant of 47.9 Hz. The doublet signals for the corresponding Pt II compound 8 were detected at 45.4 ppm and 45.7 ppm, with a coupling constant of 10.5 Hz confirming the trend J P,P (Ni) > J P,P (Pd) > J P,P (Pt) and δ(Ni) > δ(Pd) > δ(Pt). Related observations were already reported for [(dppe)M(mnt)] (mnt = maleonitriledithiolate) serving as a selected example [29].
With the change of the ligand dppe to PPh 3 in compounds 9a-c, two doublets are observed at the chemical shift between 16.7 ppm and 17.7 ppm. In addition to the 31 P/ 31 P coupling (J P,P = 21.0 Hz), 31 P/ 195 Pt coupling constants between 2861 Hz and 2998 Hz are observed (Table 3), which are in good agreement with other dithiolene-Pt compounds [31,42,43]. Here, the PPh 3 is particularly well-suited for observing changes in the electronic situation of the complex by means of 31 P-NMR spectroscopy [42]. The individual Nprotective group in 9a-c exerts only a minor influence on the 31 P/ 195 Pt coupling constant. However, the slightly differing trans effect of the asymmetric dithiolate on the phosphines is reflected in the variance of the 31 P/ 193 Pt coupling constant, spanning ∆J range from 28 Hz (9a) to 136 Hz (9b).

Electronic Structure Elucidation
The different electronic situation in compounds 6-8 is revealed by UV/Vis spectroscopy and cyclic voltammetry. Figure 4 shows the UV/Vis spectra of compounds 6, 7 and 8. In the visible range between 400 and 700 nm characteristic absorption bands at 409 nm (8), 523 nm (7) and 602 nm (6) are observed, which are responsible for the characteristic color of the compounds: green (6), red (7) and yellow (8). According to TD-DFT calculations, the underlying excitation can be assigned to a dithiolate-π to metal-d transition. Hence, the trend 6 > 7 > 8 in λ reflect the increasing ligand field splitting in the order Ni, Pd, Pt. Consistently, in cyclic voltammetry, a reduction process requires lower potentials for heavier metals. The Ni compound 6 shows a reversible Ni II /Ni I reduction with a half-step potential of −1.79 V, while an irreversible signal at a potentials of −2.14 V and −2.60 V, respectively, are observed for complexes 7 and 8. DFT calculations on related Ni and Pd dppe complexes of N-2,6-dimethylphenyltriazole-4,5-dithiolate and the corresponding anions resulted that the reversible reduction Ni II , I is based on a substantial distortion to tetrahedral, which is not relevant for Pd and Pt. Accordingly, the calculated ∆G value for the reduction are higher for Pd and Pt compared with Ni. [16].
The different electronic situation in compounds 6-8 is revealed by UV/Vis spectroscopy and cyclic voltammetry. Figure 4 shows the UV/Vis spectra of compounds 6, 7 and 8. In the visible range between 400 and 700 nm characteristic absorption bands at 409 nm (8), 523 nm (7) and 602 nm (6) are observed, which are responsible for the characteristic color of the compounds: green (6), red (7) and yellow (8). According to TD-DFT calculations, the underlying excitation can be assigned to a dithiolate-π to metal-d transition. Hence, the trend 6 > 7 > 8 in λ reflect the increasing ligand field splitting in the order Ni, Pd, Pt. Consistently, in cyclic voltammetry, a reduction process requires lower potentials for heavier metals. The Ni compound 6 shows a reversible Ni II /Ni I reduction with a halfstep potential of -1.79 V, while an irreversible signal at a potentials of -2.14 V and -2.60 V, respectively, are observed for complexes 7 and 8. DFT calculations on related Ni and Pd dppe complexes of N-2,6-dimethylphenyltriazole-4,5-dithiolate and the corresponding anions resulted that the reversible reduction Ni II , I is based on a substantial distortion to tetrahedral, which is not relevant for Pd and Pt. Accordingly, the calculated ∆G value for the reduction are higher for Pd and Pt compared with Ni. [16].

Investigation of Dimerization
The dimerization of complex 10 to form (10)2 found in the solid state could be of great interest for the assembly of coordination polymers on multiple N-coordinated triazole ligands at one metal ion. Therefore, the dimerization equilibrium in solution was investigated by 1 H NMR and UV/Vis spectrometry as well as cyclic voltammetry. Variable temperature 1 H NMR demonstrated that at concentrations of about 0.02 mol/L, the dimer at 4.79 ppm prevails ( Figure 5, right), while the monomer is detected at 5.48 ppm. A dimerization constant KD of 290 L/mol was determined at 25 °C and a Van't-Hoff plot of KD at decreasing temperatures resulted a ΔH value of -10.63 kcal/mol and ΔS of -23.6 cal/mol·K ( Figure S89). In contrast, in UV/Vis spectroscopy at about 2·10 −4 mol/L in CH2Cl2 the monomer is dominant. The violet crystals yielded a dark blue solution. Two absorption bands, at 485 nm and 619 nm, respectively, were observed in the visible range. For the solid state, reflectance UV/Vis spectroscopy was carried out (Figure 5, left). The absorption bands at 351 nm and 510 nm apparently belong to the dimer (10)2. Accordingly, the strongest

Investigation of Dimerization
The dimerization of complex 10 to form (10) 2 found in the solid state could be of great interest for the assembly of coordination polymers on multiple N-coordinated triazole ligands at one metal ion. Therefore, the dimerization equilibrium in solution was investigated by 1 H NMR and UV/Vis spectrometry as well as cyclic voltammetry. Variable temperature 1 H NMR demonstrated that at concentrations of about 0.02 mol/L, the dimer at 4.79 ppm prevails ( Figure 5, right), while the monomer is detected at 5.48 ppm. A dimerization constant K D of 290 L/mol was determined at 25 • C and a Van't-Hoff plot of K D at decreasing temperatures resulted a ∆H value of −10.63 kcal/mol and ∆S of −23.6 cal/mol·K ( Figure S89). In contrast, in UV/Vis spectroscopy at about 2 × 10 −4 mol/L in CH 2 Cl 2 the monomer is dominant. The violet crystals yielded a dark blue solution. Two absorption bands, at 485 nm and 619 nm, respectively, were observed in the visible range. For the solid state, reflectance UV/Vis spectroscopy was carried out ( Figure 5, left). The absorption bands at 351 nm and 510 nm apparently belong to the dimer (10) 2 . Accordingly, the strongest absorption band at 619 nm is assigned to a dithiolate-π to Co III charge transfer in the monomer 10. Compared to the complex [(η 5 -C 5 H 5 )Co(bdt)] (λ = 566 nm), the band is bathochromicly shifted by 1500 cm −1 . [44] This difference can be attributed to the stronger dithiolate character in 1H-1,2,3-triazole-4,5-dithiolate ligands compared with the benzene-1,2-dithiolate, which shows a stronger conjugation to the aromatic system due to better electronegativity matching. Comparable charge transfer bands were reported for many other semi-sandwich complexes with a cobalt dithiolene ligand. [45][46][47] As expected, the equilibrium between the monomer and the dimer can be influenced by changing the temperature between 0 • C and 40 • C. An increased temperature results in an increased concentration of the monomer at 619 nm.
The cyclic voltammograms of 10 were measured at a concentration range, at which the dimer (10) 2 is the main species ( Figure 6). The signal at a potential E 1/2 of −0.99 V for the Co III /Co II redox couple exhibits quasi-reversible features. The peak difference increases from 370 mV at a scan rate of 100 mV/s to 520 mV at 300 mV/s, which supports a weakly coupled two-electron process for (10) 2 . In addition, irreversible oxidation at about +0.8 V causes the appearance of a new signal at slightly higher potential compared with the original Co II /Co III couple. This can reasonably be assigned to the monomer, because, being easier to reduce, the 16 valence electron monomer 10 should exhibit a higher potential. Apparently, one-electron oxidation leads to a release of the monomer 10.
absorption band at 619 nm is assigned to a dithiolate-π to Co III charge transfer in the monomer 10. Compared to the complex [(η 5 -C5H5)Co(bdt)] (λ = 566 nm), the band is bathochromicly shifted by 1500 cm -1 . [44] This difference can be attributed to the stronger dithiolate character in 1H-1,2,3-triazole-4,5-dithiolate ligands compared with the benzene-1,2dithiolate, which shows a stronger conjugation to the aromatic system due to better electronegativity matching. Comparable charge transfer bands were reported for many other semi-sandwich complexes with a cobalt dithiolene ligand. [45][46][47] As expected, the equilibrium between the monomer and the dimer can be influenced by changing the temperature between 0 °C and 40 °C. An increased temperature results in an increased concentration of the monomer at 619 nm. The cyclic voltammograms of 10 were measured at a concentration range, at which the dimer (10)2 is the main species ( Figure 6). The signal at a potential E1/2 of -0.99 V for the Co III /Co II redox couple exhibits quasi-reversible features. The peak difference increases from 370 mV at a scan rate of 100 mV/s to 520 mV at 300 mV/s, which supports a weakly coupled two-electron process for (10)2. In addition, irreversible oxidation at about +0.8 V causes the appearance of a new signal at slightly higher potential compared with the original Co II /Co III couple. This can reasonably be assigned to the monomer, because, being easier to reduce, the 16 valence electron monomer 10 should exhibit a higher potential. Apparently, one-electron oxidation leads to a release of the monomer 10.   The cyclic voltammograms of 10 were measured at a concentration range, at which the dimer (10)2 is the main species ( Figure 6). The signal at a potential E1/2 of -0.99 V for the Co III /Co II redox couple exhibits quasi-reversible features. The peak difference increases from 370 mV at a scan rate of 100 mV/s to 520 mV at 300 mV/s, which supports a weakly coupled two-electron process for (10)2. In addition, irreversible oxidation at about +0.8 V causes the appearance of a new signal at slightly higher potential compared with the original Co II /Co III couple. This can reasonably be assigned to the monomer, because, being easier to reduce, the 16 valence electron monomer 10 should exhibit a higher potential. Apparently, one-electron oxidation leads to a release of the monomer 10.

Conclusions
In this publication, a new synthetic route for the assembly of 1H-1,2,3-triazole-4,5dithiolenes was presented, which made use of click chemistry. Instead of complicated, expensive and sensitive catalysts, very high yields of the mono-substituted triazole sulfides 1 could be achieved using CuSO 4 in CuAAC. The second sulfur substituent could be introduced by facile deprotonation of the triazole ring and subsequent reaction with sulfur and benzyl bromide, yielding the triazole disulfides 2. Nevertheless, this new synthetic method for the generation of a dithiolene unit at the 1H-1,2,3-triazole shows clear advantages in comparison with the synthesis described in the literature. [16] In addition, all attempts at a direct introduction of both sulfide substituents into the prototype 1H-1,2,3-triazole led exclusively to the monosulfide isomers 4. Subsequent reductive removal of the S-protective groups with sodium in THF in the presence of naphthalene yielded the desired dithiol derivatives. However, by-products indicating a competing removal of the whole benzyl thiolate at either 4-or 5-position, respectively, were isolated in form of Co III and Pt II complexes (11 and 12). In coordination experiments with the dithiols, several complexes with Ni II , Pd II , Pt II and Co III could be isolated and fully characterized. It was shown that dithiolate coordination dominates the coordination behavior. Neither the coordinated metal (6, 7, 8) nor the protective group at the nitrogen atom of the triazole (9a-c) have a strong effect on the electronic situation at the dithiolate unit. With coordination of the [(η 5 -C 5 H 5 )Co] moiety, a 16 valence electron Co III center could be introduced at the dithiolate unit giving complex 10. Instead of a conceivable coordination of a triazole N atom, this complex showed a dimerization via dual µ-sulfur coordination in the solid state. By means of a temperature-dependent NMR and UV/Vis spectroscopic measurements completed by cyclic voltammetry, the thermodynamic parameters of the monomer-dimer equilibrium were determined.