Probing the Effect of Six-Membered N-Heterocyclic Carbene—6-Mes—on the Synthesis, Structure and Reactivity of Me₂MOR(NHC) (M = Ga, In) Complexes

Abstract

The investigation of the reactivity of six membered N-heterocyclic carbene 1,3-bis(2,4,6-trimethylphenyl)-3,4,5,6-tetrahydropyrimidin-1-ylidene (6-Mes) towards dialkylgallium and dialkylindium alkoxides/aryloxides has shown that both steric hindrances and donor properties of 6-Mes significantly influence the strength of M–C_6-Mes bond, as well as the formation, structure and reactivity of Me₂MOR(6-Mes) (M = Ga, In) complexes. While the reactions of simple dimethylgallium alkoxides with 6-Mes lead to the formation of stable monomeric Me₂Ga(OCH₂CH₂OMe)(6-Mes) (1) and Me₂GaOMe(6-Mes) complexes, the analogous Me₂InOR(6-Mes) are unstable and disproportionate to methylindium alkoxides and Me₃In(6-Mes) (2). The use of bulky alkoxide ligand—OCPh₂Me or aryloxide ligand—OC₆H₄OMe allowed for the synthesis of stable Me₂M(OCPh₂Me)(6-Mes) (M = Ga (3) and In (4)) as well as Me₂M(OC₆H₄OMe)(6-Mes) (M = Ga (5) and In (6)). The structures of 1–6 have been determined using both spectroscopic methods in solution and X-ray diffraction studies, which confirmed the effect of both steric hindrances and donor properties of 6-Mes on their structure and catalytic properties in the ring-opening polymerization (ROP) of rac-lactide.

1. Introduction

Over recent years, we have reported Me₂MOR(NHC) (M = Ga [1,2], In [3]; NHC = N-heterocyclic carbene) complexes, which constitute the first examples of gallium and indium, as well as group 13 metal alkoxides, with NHCs [4]. Moreover, they represent isoselective catalysts for the ring-opening polymerization (ROP) of rac-lactide (rac-LA), which are additionally highly active at low temperatures. Therefore, they still constitute rare examples of isoselective catalysts operating at mild conditions, both among group 13 metal complexes [5,6,7,8,9], and beyond [10,11,12,13]. Importantly, both the structure and reactivity of Me₂MOR(NHC) (M = Ga, In; NHC = SIMes, IMes) were found to be strongly dependent on the M–C_NHC bond. While in the case of corresponding gallium complexes, a relatively strong Ga–C_NHC bond led to stable complexes [1,2], in the case of mostly unstable dialkylindium alkoxides with NHCs, a significantly weaker In–C_NHC bond was observed [3]. Consequently, in the ring-opening polymerization with Me₂GaOR(NHC), Ga–C_NHC bond remained intact, resulting in the exclusive insertion of rac-LA into Ga–O_alkoxide bond, similar to Zn [14,15,16], Al [17], as well as Ti and Zr [18,19,20], metal alkoxides with NHCs, which have been reported to polymerize rac-LA due to the insertion into M–O_alkoxide bond. On the other hand, rac-LA could insert, even at low temperature, into In–C_NHC bond of Me₂GaOR(NHC), considerably weaker in comparison with Ga–C_NHC [3]. Analogously, the insertion of lactide into the M–C_NHC bond of Mg alkoxides with alkoxide ligands possessing NHC termini [16], Me₃Al(NHC) adducts [21], or yttrium and titanium complexes [22], have been reported so far.

Similarly to the observed effect of a metal on the corresponding M–C_NHC bond, as well as the structure and ROP activity of discussed main group metal alkoxides with NHCs, we should expect a considerable effect of steric and electronic properties of NHC on both the strength and reactivity of the M–C_NHC bond. Our initial studies revealed the adverse effect of steric hindrances of NHC on the synthesis of Me₂GaOR(SIPr) as well as the formation and strength of Ga–C_SIPr bond [2], which was additionally supported by the structure of Me₃M(NHC) (M = Al, Ga, In) adducts [3,21,23]. However, the more detailed effect of both steric and electronic effect of NHC on the M–C_NHC bond has not been addressed so far for main group metal alkoxides with NHCs, including Me₂MOR(NHC) (M = Ga, In) complexes. In order to probe the effect of NHC on the latter, we chose 6-Mes—an N-heterocyclic carbene of stronger donor properties, along with the increased steric demand, in comparison with SIMes or IMes (Figure 1) [24,25]. While the increased steric hindrances of 6-Mes result from wider N–C–N angle, this should be responsible for the hybridization at carbene carbon and the higher energy of HOMO orbital of 6-Mes occupied by a lone pair of electrons, and therefore its donating properties [26].

Our studies, reported hereby, have allowed for the determination of the considerable effect of steric and electronic properties of 6-Mes, as well as the relative importance of electronic and steric properties of 6-Mes, on the synthesis, structure and stability of a series of Me₂MOR(NHC) (M = Ga, In) complexes, as well as their catalytic activity in the ROP of rac-LA.

2. Results and Discussion

In order to observe the influence of steric and electronic properties NHC on the synthesis, structure and reactivity of Me₂MOR(NHC) (M = Ga, In) complexes, we investigated the reactivity of dimethylgallium and dimethylindium alkoxides and aryloxides towards 1,3-bis(2,4,6-trimethylphenyl)-3,4,5,6-tetrahydropyrimidin-1-ylidene (6-Mes). The synthesis, structure of resulting Me₂MOR(6-Mes) (M = Ga, In) complexes, and their catalytic activity in the ROP of rac-LA was compared to the analogous dialkylgallium and dialkylindium derivatives stabilized with 1,3-bis(2,4,6-trimethylphenyl)imidazolin-2-ylidene (SIMes) or 1,3-bis(2,4,6-trimethylphenyl) imidazolin-2-ylidene (IMes) [1,2,3].

2.1. Reactivity of [Me₂M(μ-OR)]_n (M= Ga, In ; OR = OCH₂CH₂OMe, OMe ; n = 2, 3) towards 6-Mes

The reaction between dimeric gallium complex [Me₂Ga(μ-OCH₂CH₂OMe)]₂ and 6-Mes (Ga:6-Mes = 1:1, Scheme 1) led to the essentially instant formation of gallium complex Me₂Ga(OCH₂CH₂OMe)(6-Mes) (1). Compound 1 was isolated as an off-white solid in high yield. Unfortunately, the crystals of 1 decomposed during an X-ray diffraction experiment, which precluded the determination of its X-ray structure. Notably, the increased reactivity of 1 towards grease during X-ray experiment, in comparison with Me₂GaOR(NHC) (NHC = SIMes, IMes), should not be surprising in the light of its structure and reactivity in solution.

In solution, both ¹H NMR, and ¹³C NMR were indicative of the formation of 1. Although two sets of signals were revealed by ¹H NMR of 1 in toluene-d₈, the main set of signals, including a singlet corresponding to Ga–Me protons (−1.06 ppm), which was significantly shifted to higher field, similarly to Me₂Ga(OCH₂CH₂OMe)(SIMes) [2], was indicative of the formation of Me₂Ga(OCH₂CH₂OMe)(6-Mes). The additional minor set of signals corresponded to [Me₂Ga(OCH₂CH₂OMe)]₂ (16%), which suggested the presence of an equilibrium, presented in Scheme 1. Such an equilibrium was further evidenced by a 2D ROESY (rotating frame Overhauser effect spectroscopy) experiment (Figure S3). Moreover, essentially the same ratio of Me₂Ga(OCH₂CH₂OMe)(6-Mes): [Me₂Ga(OCH₂CH₂OMe)]₂, for 1, and the reaction mixture of [Me₂Ga(μ-OCH₂CH₂OMe)]₂ and 6-Mes (Ga:6-Mes = 1:1), was in line with the presence of indicated equilibrium. Most probably, the latter was also responsible for the instant reaction of 1 with CD₂Cl₂ resulting in the formation of [Me₂Ga(μ-OCH₂CH₂OMe)]₂ and [6-Mes-D]⁺, which were evidenced by NMR spectroscopy (see the Supplementary Materials). Interestingly, similar reactivity was observed for the reaction between free SIMes and methylene chloride. On the contrary, more acidic chloroform was required in order to react instantly with N-heterocyclic carbene of Me₂GaOR(SIMes), while the latter remained essentially stable in CD₂Cl₂ during the NMR experiment [1]. While the presence of the equilibrium presented in Scheme 1 is not surprising in the light of similar equilibrium observed in the case of Me₂In(OCH(Me)CO₂Me)(NHC) (NHC = SIMes, IMes) [3], the one observed for 1 represents the first such equilibrium for Me₂GaOR(NHC) complexes, which is in line with the weaker Ga–C_6-Mes bond in comparison with Me₂GaOR(NHC) (NHC = SIMes, IMes). On the contrary, the shift between free 6-Mes (Δ = 244.9 ppm) and coordinated 6-Mes in the case of 1 (197.1 ppm, Δ = −47.8 ppm) was considerably larger in comparison with Me₂MOR(NHC) (NHC = SIMes (Δ = −43.5–(−45.2) ppm), IMes (Δ = −43.1–(−43.6) ppm) [1,2]. Although the latter could indicate the presence of stronger Ga–C_6-Mes bond in solution [25], it rather reflects the stronger donor properties of 6-Mes vs SIMes/IMes in the light of both the structure of 1, as well as the weaker Ga–C_6-Mes evidenced by the X-ray analysis of Me₂Ga(OCPh₂Me)(6-Mes) (3) (see below). While the carbene carbon signal in ¹³C NMR could not be observed for Me₂GaOMe(6-Mes), which was synthesized analogously to 1, the ¹H NMR was in line with the presence of the equilibrium presented in Scheme 1 (see the Supplementary Materials), and therefore, similarly to 1, indicated weaker Ga–C_6-Mes bond in comparison with Ga–C_NHC of Me₂GaOMe(NHC) (NHC = SIMes, IMes).

Despite the adverse effect of steric hindrances of 6-Mes on the strength of Ga–C_6-Mes bond, the stronger donor properties of 6-Mes in comparison with SIMes/IMes, prompted us to investigate its effect on the synthesis and structure of Me₂InOR(NHC) complexes. The main question was, whether the larger radius of indium, in comparison with gallium, could allow for the formation of a stronger In–C_6-Mes bond and the stabilization of Me₂InOR(NHC), which had been previously shown by us to be unstable and prone to ligand disproportionation [3]. Although the reaction between [Me₂In(μ-OCH₂CH₂OMe)]₂ and 6-Mes led to the formation of Me₂In(OCH₂CH₂OMe)(6-Mes), the latter complex disproportioned readily to Me₃In(6-Mes) and Mitsubishi type complex In{Me₂In(μ-OCH₂CH₂OMe)}₃ (Scheme 2) (see the Supplementary Materials), analogously to Me₂In(OCH₂CH₂OMe)(NHC) (NHC = SIMes, IMes) [3]. The formation of both Me₃In(6-Mes) and In{Me₂In(μ-OCH₂CH₂OMe)}₃ was evidenced by NMR spectroscopy, while, in contrast to our previous studies, In{Me₂In(μ-OCH₂CH₂OMe)}₃ could be isolated as a white crystalline solid, similar to the only two other examples of Mitsubishi type indium In{Me₂In(µ-OR)}₃ complexes [27,28]. Unfortunately, we did not succeed in obtaining crystals of In{Me₂In(μ-OCH₂CH₂OMe)₂}₃ suitable for X-ray analysis, and its structure has been confirmed using ¹H and ¹³C NMR spectroscopy (see the Supplementary Materials).

Although the reaction of 6-Mes with [Me₂In(μ-OMe)]₃ led to the formation of Me₂InOMe(6-Mes) complex (see the Supplementary Materials), which was more stable in comparison with Me₂In(OCH₂CH₂OMe)(6-Mes), it seemed to disproportionate more readily in comparison with Me₂InOMe(NHC) (NHC = SIMes, IMes), and as result could not be isolated in contrast to the latter [3]. While we could not estimate the strength of In-C_6-Mes in the case of Me₂InOR(6-Mes), due to the tendency of the latter to disproportionate , it could be analyzed for the Me₃In(6-Mes) (2). The crystals of 2 suitable for X-ray analysis could be obtained either by the decomposition of Me₂In(OCH₂CH₂OMe)(6-Mes) or quantitative reaction between 6-Mes and Me₃In. Compound 2 was synthesized and isolated in bulk using only the second method, and was used to confirm the formation of 2 in solution due to the decomposition of Me₂In(OCH₂CH₂OMe)(6-Mes). X-ray diffraction analysis of 2 revealed, similarly to Me₃In(NHC) (NHC = SIMes, IMes) monomeric structure, with the coordination sphere adopting a distorted-tetrahedral geometry (Figure 2). The In–C_6-Mes bond (2.367(2) Å, 0.51 valence units (vu) [29]) in 2 turned out to be noticeably longer, and therefore weaker, than In–C_SIMes (2.334(6) Å, 0.55 vu [3]; 2.316(8) Å [23]), In–C_IMes (2.307(2) Å, 0.60 vu [3]; 2.304(7) Å [23], 2.292(6) Å [21]), In–C_SIPr (2.342(2) Å) [23] and In–C_IPr (2.309(2) Å) [23] in Me₃In(SIMes), Me₃In(IMes), Me₃In(SIPr) and Me₃In(IPr) adducts, respectively. It showed a more significant effect of steric hindrances of 6-Mes, in comparison with SIMes, IMes, and even SIPr and IPr, on the strength of In–C_6-Mes bond. Interestingly, the latter was in agreement with smaller steric hindrances of SIMes and IMes, and in contrast with larger steric demand of SIPr and IPr, in comparison with 6-Mes, as represented by the buried volume of discussed NHCs [25]. The analysis of the strength of In–C_6-Mes bond of 2 in solution, in comparison with In–C_NHC bonds of already characterized Me₃In(NHC) adducts [3,21,23], was in sharp contrast with the solid state result. In solution, the shift between free and coordinated NHCs in ¹³C NMR, indicated stronger In–C_6-Mes bond (δ = 207.5 ppm, Δ = −37.4 ppm) in comparison with In-C_SIMes (Δ = −35.2 ppm) and In–C_IMes (Δ = −35.7 ppm) of Me₃In(SIMes) and Me₃In(IMes) respectively. While the relationship between the distance of essentially the same bonds and their strength should be considered a tenet in the case of interpretations of molecular structures [30], the analysis of the strength of a M–C_NHC bond using the ¹³C NMR spectroscopy could be influenced by other factors affecting the distribution of electron density at carbene carbon. However, it must be noticed that the shift between free and coordinated NHCs in solution, using ¹³C NMR spectroscopy, reflected stronger donor properties of 6-Mes in comparison with SIMes and IMes. Notably, analogous results concerning the strength of M–C_NHC bonds were obtained for alkoxide derivatives Me₂M(OCPh₂Me)(6-Mes) (M = Ga, In).

2.2. Reactivity of [Me₂M(μ-OCPh₂Me)]₂ (M = Ga, In) towards 6-Mes

While the reaction between [Me₂Ga(μ-OCPh₂Me)]₂ and 6-Mes (Ga:6-Mes = 1:1, Scheme 3) resulted in the formation of Me₂Ga(OCPh₂Me)(6-Mes) (3), the prolonged time of 24 h was required to reach c.a. 87% of conversion. Monitoring of the reaction progress using ¹H NMR revealed significantly slower reaction in comparison with the essentially instant reaction of [Me₂Ga(μ-OCPh₂Me)]₂ with less sterically demanding carbenes (i.e. SIMes, IMes) [3], therefore indicating the strong adverse effects of 6-Mes steric hindrances on its reactivity towards dimeric [Me₂Ga(μ-OCPh₂Me)]₂ and formation of monomeric Me₂Ga(OCPh₂Me)(6-Mes) (3) (Scheme 3). The analogous effect was not observed in the case of the reaction of [Me₂In(μ-OCPh₂Me)]₂ with 6-Mes (In:6-Mes = 1:1, Scheme 3), leading to the formation of the indium complex Me₂In(OCPh₂Me)(6-Mes) (4), which can be explained by the significantly larger radius of indium in comparison with gallium. Complexes 3 and 4 were isolated in high yields as colorless crystals.

For complexes 3 and 4 X-ray diffraction analysis revealed the presence of four-coordinated gallium/indium species with the coordination sphere adopting a distorted-tetrahedral geometry. Interestingly, in contrast to essentially symmetrical gallium complex 3 (Figure 3), due to the orientation of the 6-Mes ligand with respect to In–C_Me bonds, the presence of essentially asymmetrical indium complex 4 (Figure 4) was observed. These observations, reflected by O(1)–Ga(1)–C(1)–N(1) (–81.98(15)°) (Figure 3) and O(1)–In(1)–C(1)–N(1) (–139.0(8)°) (Figure 4) torsion angles, were in line with the previously characterized essentially symmetric Me₂Ga(OCPh₂Me)(IMes) and asymmetric Me₂In(OCPh₂Me)(IMes) complexes [3]. While for 3 and 4 the symmetrical/asymmetrical orientation of NHC in the solid state, with respect to M–C_Me (M = Ga, In) bonds, could be tentatively explained by the presence of different weak CH···π interactions, leading to the formation of chains in the solid state (see the Supplementary Materials), the essentially analogous CH···π interactions had been previously observed for Me₂Me(OCPh₂Me)(IMes) (M = Ga, In) [3]. Although the symmetrical/asymmetrical orientation of NHC with respect to M–C_Me bonds could influence the catalytic properties, including stereoselectivity, of Me₂MOR(NHC), and is of our current interest, the factors controlling the arrangement of NHC in the case of 3 and 4 could not be unequivocally indicated. However, it was the effect of M–C_6-Mes bond on the structure and reactivity of Me₂MeOR(6-Mes), which focused our attention. In the case of complex 3, the Ga–C_6-Mes bond (2.139(2) Å, 0.57 vu) was significantly weaker in comparison with the Ga–C_IMes bond of Me₂Ga(OCPh₂Me)IMes (2.096(4) Å, 0.64 vu) [3], although similarly small pitch angles (out of NHC plane tilting) and yaw angles (in plane tilting) were observed for both gallium complexes. Moreover, the longest Ga–C_6-Mes bond of 3, among other crystallographically characterized Me₂GaOR(NHC) complexes [1,2,3], should be expected to result from steric hindrances of 6-Mes. Although the In–C_6-Mes bond of 4 (2.329(10) Å, 0.56 vu) was also weaker in comparison with the In–C_IMes bond of Me₂In(OCPh₂Me)IMes (2.301(8) Å, 0.61 vu), the smaller effect of the steric hindrances of 6-Mes on the In–C_6-Mes bond should be explained by a larger radius of indium in comparison with gallium. Interestingly, the latter resulted in only slightly weaker In–C_6-Mes bond of 4 (2.329(10) Å, 0.56 vu) in comparison with Ga–C_6-Mes bond (2.139(2) Å, 0.57 vu) of 3.

While the major set of signals revealed by the ¹H NMR of 3 and 4 was in line with the formation of Me₂M(OCPh₂Me)(6-Mes) (M = Ga, In), the presence of minor signal could be ascribed to the [Me₂Ga(μ-OCPh₂Me)]₂, therefore suggesting the presence of equilibrium, analogously to 1. However, due to relatively slow reaction rate between [Me₂M(μ-OCPh₂Me)]₂ and 6-Mes (see above), the presence of minor set of signals could not be unequivocally associated with the equilibrium. The different ratios of Me₂M(OCPh₂Me)(6-Mes):[Me₂Ga(μ-OCPh₂Me)]₂ both in the reaction mixture, as well as in the spectrum of 3 or 4 were not clearly indicative of the presence of such equilibrium. In the ¹³C NMR of both 3 and 4, significant shifts of carbene carbon signals between free and coordinated 6-Mes were observed. In the case of gallium complex 3, the observed shift (Δ = −47.5 ppm) was comparable with Me₂Ga(OCH₂CH₂OMe)(6-Mes) (Δ = −47.8 ppm), and slightly smaller than for Me₂Ga((S)-OCH(Me)CO₂Me)(6-Mes) (Δ = −48.4 ppm) (see below), but was larger in comparison with Me₂GaOR(NHC) (NHC = SIMes, IMes) complexes (Δ_max = −45.2 ppm) [2]. Similarly, for 4 the observed shift (Δ = −41.7 ppm) was larger in comparison with Me₂InOR(NHC) complexes (NHC = SIMes, IMes) (Δ_max = −38.9 ppm) [3]. However, in the light of X-ray analysis of 3 and 4, which revealed weaker M–C_6-Mes bonds (M = Ga, In) in comparison with M–C_NHC of Me₂M(OCPh₂Me)(NHC) (M = Ga, In; NHC = SIMes, IMes), the shift of carbene carbon observed in ¹³C NMR was indicative, similarly to 1 and 2, of the stronger donor properties of 6-Mes in comparison with SIMes or IMes, rather than stronger M–C_6-Mes bonds in comparison with M–C_NHC (NHC = SIMes, IMes). While the detailed discussion on the factors affecting the carbene carbon shift in ¹³C NMR, and therefore the analysis of the strength of the M–C_NHC bond using the ¹³C NMR spectroscopy, cannot be done at the current stage, our present research in this area focuses on the effect of the character of the M–C_NHC bond, rather than only its strength, on the synthesis, structure and reactivity of group 13 metal Me₂MOR(NHC) complexes.

2.3. Reactivity of [Me₂M(μ-OC₆H₄OMe)]₂ (M = Ga, In) towards 6-Mes

Although the investigation of the synthesis and structure of Me₂M(OC₆H₄OMe)(6-Mes) (M = Ga (5), In (6)) (Scheme 4) was important prior to the studies on their catalytic activity in the ring-opening polymerization (ROP) of rac-lactide (rac-LA) (see below), they were also interesting with regard to our studies on Me₂Ga(O,C_NHC) complexes possessing chelate alkoxide and aryloxide ligands with NHC functionality [33]. The latter showed that aryloxide (O_Ar,C_NHC) ligands, in comparison with alkoxide chelate (O,C_NHC) ligands, resulted in stronger Ga–C_NHC bond, which could be crucial for the synthesis and structure of 5 and 6 in comparison with their alkoxide analogues Me₂MOR(6-Mes) (M = Ga, In). The reactions of 6-Mes with [Me₂M(μ-OC₆H₄OMe)]₂ (M:6-Mes = 1:1, M = Ga, In) led to the instant formation of 5 and 6, which were isolated in high yields as white crystalline solids. However, the crystals suitable for X-ray analysis could not be isolated. Therefore, the structure of 5 and 6 was determined in solution using NMR spectroscopy.

Importantly, ¹H NMR spectra of 5 and 6, which were essentially identical to the corresponding reaction mixtures of [Me₂M(μ-OC₆H₄OMe)]₂ (M = Ga, In) and 2 equivalents of 6-Mes, revealed only one set of signals of Me₂M(OC₆H₄OMe)(6-Mes). The lack of any signals corresponding to [Me₂M(μ-OC₆H₄OMe)]₂, indicating the presence of the equilibrium analogues to that of 1, could be associated in this case both with the significantly weaker M₂O₂ central bridges of dialkylgallium and dialkylindium aryloxides, in comparison with respective alkoxide derivatives, and stronger M–C_6-Mes bonds in the case of aryloxide complexes 5 and 6. These observations were in line with the shift between free and coordinated carbene carbon signal in ¹³C NMR for 5 (196.1 ppm, Δ = –48.8 pm), which was only slightly different from 1 (Δ = −47.8 ppm) and 3 (Δ = −47.5 ppm), and considerably larger in comparison with Me₂GaOR(NHC) (NHC = SIMes, IMes; Δ_max = −44.8 ppm) complexes, as well as Me₂Ga(O_Ar)(SIMes) (Δ = −45.2 ppm) [34]. In the case of 6, the carbene carbon of 6-Mes in ¹³C NMR (204.3 ppm, −40.6 ppm) was only slightly shifted to the higher field in comparison with Me₂InOR(NHC) (NHC = SIMes, IMes; Δ_max = −38.9 ppm), and even slightly shifted to the lower field in contrast to 4 (Δ = −41.7 ppm). However, the observed carbene carbon shifts could not be directly related to the stronger M–C_6-Mes bond in 5 and 6, in comparison with alkoxide-Me₂M(OR)(NHC), and aryloxide-Me₂Ga(O_Ar)(NHC) complexes [1,2,3], in the light of the structure of 1–4.

2.4. Reactivity of [Me₂M(μ-(S)-OCH(Me)CO₂Me)]₂ (M = Ga, In) towards 6-Mes

The synthesis and structure of Me₂M((S)-OCH(Me)CO₂Me)(6-Mes) (Me₂Ga((S)-melac)(6-Mes)) (M = Ga, In), in which the methyl lactate (melac) ligand mimics a growing polylactide (PLA) chain [1,2,35,36,37], were investigated prior to the studies on the activity of Me₂MOR(6-Mes) in the ROP of rac-LA (see below). Although both gallium and indium complexes could not be isolated in this case, among others due to the reactivity of 6-Mes with (S)-melac ligand, they were indicative of the reactivity of sterically hindered NHCs towards methyl lactate ligands, as well as indicative of the structure and reactivity of Me₂M(O(PLA))(NHC) propagating species in the ROP of rac-LA. The reaction of [Me₂Ga(μ-(S)-melac)]₂ with 6-Mes (Ga:6-Mes = 1:1) resulted in the formation of Me₂Ga((S)-melac)(6-Mes), which was evidenced both by ¹H NMR and ¹³C NMR spectroscopy (see the Supplementary Materials). The presence of two singlets at −1.10 and −1.14 ppm, considerably shifted to higher field in comparison with [Me₂Ga(μ-(S)-melac)]₂, could be unequivocally ascribed to Ga–Me protons of Me₂Ga((S)-melac)(6-Mes) on the basis of our earlier studies on Me₂Ga((S)-melac)(NHC) (NHC = SIMes, IMes) [1,2]. Additionally, the considerable shift of carbene carbon signal in the ¹³C NMR spectra (196.5 ppm, Δ = −48.4 ppm) was in line with the formation of Ga–C_6-Mes bond. However, the presence of additional signals in the ¹H NMR spectra, including broad signal at −0.03 ppm indicated the presence of other alkylgallium species, most probably without 6-Mes coordinated to gallium. The presence of the latter signal could be associated, similarly to the synthesis of 3, with Me₂Ga protons of unreacted [Me₂Ga(μ-(S)-melac)]₂. However, further monitoring of the reaction in time, which revealed an exclusive presence of a signal at −0.02 ppm after 4 days, suggested that initially observed broad signal at −0.03 ppm could rather result from the transformation of (Me₂Ga((S)-melac)(6-Mes)) to new alkylgallium species. While the instability of (Me₂Ga((S)-melac)(6-Mes)) was in contrast with the strong Ga–C_6-Mes bond, the reactivity of 6-Mes towards methyl lactate ligand could be expected, based on the reactivity of SIPr towards (S)-melac ligand of [Me₂Ga(μ-(S)-melac)]₂, which led to the formation of among others [Me₂Ga(OCH(Me)CO₂)][SIPr-H] [2]. However, the possible equilibrium between Me₂M(O-(S)-CH(Me)CO₂Me)(6-Mes) and [Me₂Ga(μ-(S)-melac)]₂/6-Mes mixture, as revealed by the structure of 1, could result in the complete reaction of 6-Mes with (S)-melac, leading to the formation of a complex mixture of products, as shown in Scheme 5. Although no alkylgallium species could be isolated from the reaction mixture of [Me₂Ga(μ-(S)-melac)]₂ with 6-Mes (Ga:6-Mes = 1:1) the obtained results were in line with the reactivity of 6-Mes towards (S)-melac) of Me₂M((S)-melac)(6-Mes) presented in Scheme 5, which was suggested on the basis of the reaction of [Me₂Ga(μ-(S)-melac)]₂ with SIPr (Ga:SIPr = 1:1) [2], as well as the reactivity of [MeIn(μ-(S)-melac)]₂ with 6-Mes (In:6-Mes = 1:1). Contrary to the reactivity of [MeGa(μ-(S)-melac)]₂ with 6-Mes, the formation of monomeric Me₂In((S)-melac)(6-Mes) was not even indicated by ¹H NMR in the analogous reaction between [MeIn(μ-(S)-melac)]₂ and 6-Mes (In:6-Mes = 1:1) (see the Supplementary Materials). Moreover, in contrast to the formation of unstable Me₂In((S)-melac)(NHC) (NHC = SIMes, IMes), which showed the tendency for the ligand disproportionation and the formation of Me₃In(NHC) and alkylindium alkoxides [3], the reactivity of 6-Mes towards (S)-melac) was confirmed by the isolation of (6-Mes)=CH₂, which was evidenced by X-ray analysis (Figure 5). Interestingly, the formation of the latter confirmed the possibility of the abstraction of Me⁺ from (S)-melac ligand with NHC, which was only suggested in the case of our previous studies by the formation of [Me₂Ga(OCH(Me)CO₂)]⁻. Moreover, the isolation of (6-Mes)=CH₂ strongly indicated the conversion of [(6-Mes)-Me]⁺ to (6-Mes)=CH₂ with the evolution of H⁺. The latter could be responsible for the formation of [(6-Mes)-H]⁺, leading to the complex mixture of products. Moreover, it could also explain the formation of SIPr-H⁺ in our previous reaction of [Me₂Ga(μ-(S)-melac)]₂ with SIPr (Ga:SIPr = 1:1) (Scheme 5), instead of the possible abstraction of H⁺ directly from (S)-melac ligand, which was tentatively suggested in the latter case [2].

2.5. Activity of Me₂Ga(OCPh₂Me)(6-Mes) (3 and 4), Me₂M(OC₆H₄OMe)(6-Mes) (5 and 6) (M = Ga, In) and Me₂Ga(OCH₂CH₂OMe)(6-Mes) (1) in the ROP of rac-Lactide

In order to investigate the effect of 6-Mes on the catalytic properties of Me₂MOR(NHC), we examined the activity of 1, 3, 4, 5 and 6 complexes in the ring-opening polymerization (ROP) of rac-LA. With regard to the latter, we focused on the effect of M–C_6-Mes bond on the catalytic properties of selected complexes. While the M–C_6-Mes bond affects the structure of all selected complexes, it should be expected to influence their reactivity, as well the structure and reactivity of Me₂MO(PLA)(6-Mes), where O(PLA) represents growing polylactide chain, and therefore the microstructure of resulting polylactide (PLA). While our previous results showed that both gallium and indium Me₂MOR(NHC) complexes were highly active already at −20 °C [1,2,3], in order to compare our results with the latter, we investigated the ROP of rac-LA with 1, 3, 4, 5 and 6 at identical conditions. Gallium complexes: alkoxide derivative Me₂M(OCPh₂Me)(6-Mes) (3) and aryloxide derivative Me₂Ga(OC₆H₄OMe)(6-Mes) (5), were inactive in the polymerization of rac-LA at −20 °C. While in the case of complexes 3 and 5 the bulky alkoxide or aryloxide group did not allow for the insertion of rac-LA into Ga–O bond, most importantly, the strong Ga–C_6-Mes bond precluded, the initiation of rac-LA polymerization by N-heterocyclic carbene, similarily to Me₂Ga(OCPh₂Me)(NHC) (NHC = SIMes, IMes) [3]. Contrary to the latter, a much weaker In–C_NHC bond in Me₂In(OCPh₂Me)(NHC) (NHC = SIMes, IMes) resulted in the initiation of rac-LA by N-heterocyclic carbenes [3], and similar reactivity was observed in the case of Me₂In(OCPh₂Me)(6-Mes) (4) and Me₂In(OC₆H₄OMe)(6-Mes) (6), which led to the formation of cyclic PLA (see the Supplementary Materials). However, the observed reactivity of 6-Mes of indium complexes 4 and 6 towards lactide, as well as the lack of activity of gallium derivatives 3 and 5, may also indicate the considerable effect of a character of the M–C_NHC bond on the reactivity of investigated complexes. As we were unable to isolate any stable alkoxide Me₂InOR(6-Mes) complex, we focused on the activity of 1 in the ROP of rac-LA. Due to the reactivity of 1 towards CH₂Cl₂, which led to the almost quantitative and instant formation of [Me₂Ga(μ-OCH₂CH₂OMe)]₂ [9,35], no activity of 1 in the ROP of rac-LA in CH₂Cl₂, at −20 °C was observed. On the other hand, the polymerization of rac-LA with 1 (25:1) in toluene led to essentially full conversion after 12 h (−20 °C) or 10 min (room temperature). For the PLA obtained at both temperatures the MALDI-TOF analysis revealed the presence of an end group of 76 Da, which was in agreement with the insertion of rac-LA into Ga–OCH₂CH₂OMe bond (see the Supplementary Materials). Additionally, for PLA obtained at both −20 °C and r.t. the extensive intermolecular transesterification was evidenced, while minor distributions referring to cyclic PLA could be assigned to the presence of an intermolecular transesterification or additional initiation of rac-LA ROP by 6-Mes. Although the initiation of the lactide polymerization by 6-Mes of 1, especially at −20 °C, was unlikely in the light of the lack of activity of 3 and 5, it could be assumed for Me₂GaO(PLA)(6-Mes) propagating species. However, as we were not able to synthesize Me₂Ga((S)-melac)(6-Mes) (see above), which mimics propagating species Me₂GaO(PLA)(6-Mes), we synthesized the latter in the reaction between 1 and 10 equiv of rac-LA both at −20 °C and r.t.. Although the ¹H NMR spectra after 36 h, both at −20 °C and r.t. were essentially the same, as the spectrum registered after mixing of 1 and 10 equiv of rac-LA at room temperature, they were inconclusive for the presence of Me₂GaO(PLA)(6-Mes) with 6-Mes coordinated to gallium. In this case, the presence of complex Ga–Me signals in ¹H NMR in the range between −0.31 and 0.02 was in contrast to Me₂Ga((S)-melac)(SIMes) (−0.80, −0.82 ppm) or Me₂Ga(OCH(Me)C(O))₂O(CH₂)₂OMe(IMes) (−0.79, −0.87 ppm) [2], and could not strongly support the presence of Me₂GaO(PLA)(6-Mes). Moreover, the lack of signal corresponding to coordinated/uncoordinated carbene carbon in ¹³C NMR was also inconclusive for the formation of the latter. Therefore, we could not unequivocally conclude whether the initiation of the ROP of rac-LA by 6-Mes was possible, or the presence of discussed transesterification reactions simply resulted from the catalytic properties of Me₂GaO(PLA)(6-Mes) propagating species. However, irrespective of the origin, the presence of transesterification in the case of the ROP of rac-LA with 1 at –20 °C, which is in contrast with the essentially no transesterification observed for the ROP of rac-LA with Me₂GaOR(NHC) (NHC = SIMes, IMes) [1,2], led to a considerable decrease of isoselectivity (P_m (probability of meso linkages in PLA) = 0.66–0.69 at −20 °C, P_m = 0.63–0.66 at r.t.) in comparison with the ROP of rac-LA with Me₂GaOR(NHC) (P_m = 0.76–0.78 at −20 °C, P_m = 0.65–0.68 at r.t.). Importantly, the ROP of rac-LA with 1 showed the considerable effect of the structure of NHC on the catalytic properties of Me₂GaOR(NHC), which should be of importance and is of our current interest as Me₂GaOR(NHC) are still rare examples of highly active metal alkoxides for the isoselective ROP of rac-LA [5,6,7,8,9,10,11,12,13].

3. Materials and Methods

3.1. General Procedures

All operations were carried out under dry argon using standard Schlenk techniques. Solvents and reagents were purified and dried prior to use. Solvents were purified using MBRAUN Solvent Purification Systems (MB-SPS-800) (MBRAUN, Garching, Germany) and stored over molecular sieves. rac-Lactide was purchased from Aldrich (Poland) and further purified by crystallization from anhydrous toluene and then sublimated. (S)-Methyl lactate, 2-methoxyethanol, and methanol were purchased from Aldrich, dried over molecular sieves, and distilled under argon. 1,1-diphenylethanol was purchased from Aldrich and used as received. Me₃In and Me₃Ga were purchased from STREM Chemicals, Inc. (Bischheim, France) and used as received. 6-Mes was synthesized according to the literature [38]. Indium and gallium complexes: [Me₂M(μ-OCH₂CH₂OMe)]₂, [Me₂M(μ-OMe)]₃, [Me₂M(μ-OCH(CH₃)COOMe)]₂, (Me₂M(μ-OCPh₂Me)]₂, Me₂M(OC₆H₄OMe)]₂ were synthesized as described by us previously [1,2,3]. ¹H and ¹³C NMR spectra were recorded on an Agilent 400-MR DD2 400 MHz spectrometer (Agilent, Palo Alto, CA, USA) with shifts given in ppm according to the deuterated solvent shift. MALDI-TOF spectra were recorded on a Bruker Model ultrafleXtreme instrument (Bruker Daltonics, Bremen, Germany). Elemental analysis was performed on a Vario EL III instrument (ELEMENTAR Analysensysteme, Hanau, Germany).

3.2. Synthesis of Indium and Gallium Complexes

Synthesis of 1. To a solution of [Me₂Ga(μ-OCH₂CH₂OMe)]₂ (76.4 mg, 0.44 mmol) in toluene (2 mL) was added 2 mL of a toluene solution of 6-Mes (140.0 mg, 0.44 mmol) at room temperature, and the resulting solution was stirred for 2 h. Then toluene was removed under vacuum to give a faint-yellow oil, which turned to an off-white solid after 45 min of drying. The solid was subsequently crystallized from toluene (0.3 mL), Et₂O (4 mL) and hexane (3 mL) at −40°C to give off-white crystals, which were dried under vacuum (1, 158 mg, 73%). Anal. Calcd. for C₂₇H₄₁GaN₂O₂: C, 65.47; H, 8.34; N, 5.66. Found: C, 65.32; H, 8.34; N, 5.60. ¹H NMR (toluene-d₈, 400 MHz): −1.08 (s, 6H, GaCH₃), 1.49 (q, 2H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 2.11 (s, 6H, CH₃), 2.29 (s, 12H, CH₃), 2.60 (t, 4H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 3.27 (s, 3H, OCH₃), 3.45 (t, 2H, ³J (H,H) = 6.5 Hz, CH₂CH₂), 3.69 (t, 2H, ³J (H,H) = 6.5 Hz, CH₂CH₂), 6.75 (s, 4H, CH_Ar). ¹³C{1H} NMR (toluene-d₈, 100 MHz): −5.6 (GaMe₂), 18.2, 20.9, 21.0, 42.0, 47.2, 58.5, 64.0, 77.8, 129.3, 129.8, 135.0, 135.6, 137.7, 141.8, 197.1 (carbene).

Synthesis of 2. A stirred solution of Me₃In (116 mg, 0.73 mmol) in toluene (5 mL) was cooled to −20 °C, and a toluene solution (2 mL) of 6-Mes (232 mg, 0.73 mmol) was added dropwise. After addition the reaction mixture was warmed to room temperature and stirred for 2 h. Solvent and volatile residues were then removed under vacuum to give a white solid. The solid was subsequently recrystallized from a toluene/hexane solution (2/6 mL) at −20 °C to give white crystals, which were washed twice with 3 mL of hexane and dried under vacuum to give Me₃In(6-Mes) (nr; 311 mg, 92%). Anal. Calcd. for C₂₅H₃₇InN₂: C, 62.50; H, 7.76; N, 5.83. Found: C, 62.36; H, 7.88; N, 5.76. ¹H NMR (toluene-d₈, 400 MHz): −0.88 (s, 9H, InCH₃), 1.50 (m, 2H, CH₂), 2.11 (s, 6H, CH₃), 2.19 (s, 12H, CH₃), 2.58 (m, 4H, CH₂CH₂), 6.77 (s, 4H, CH_Ar). ¹³C{1H} NMR (toluene-d₈, 100 MHz): −8.0 (InMe₃), 18.1, 20.9, 46.3, 130.1, 135.1, 137.4, 137.7, 142.1, 207.5 (carbene).

Synthesis of 3 and 4. To a solution of [Me₂In(μ-OCPh₂Me)]₂ (110 mg, 0.32 mmol) or [Me₂Ga(μ-OCPh₂Me)]₂ (96 mg, 0.32 mmol) in toluene (6 mL) was added 2 mL of a toluene solution of 6-Mes (103 mg, 0.32 mmol) at room temperature, and the resulting solution was stirred for 24 h. Then, toluene was removed under vacuum to give a light yellow solid. The solid was subsequently crystallized from toluene (2 mL) after addition of hexane (6 mL) at −20 °C to give white crystals, which were dried under vacuum (nr In, 187 mg, 88%; nr Ga, 158 mg, 80%). Data for 4 are as follows. Anal. Calcd. for C₃₈H₄₇InN₂O: C, 68.88; H, 7.15; N, 4.23. Found: C, 66.92; H, 7.25; N, 4.17. ¹H NMR (toluene-d₈, 400 MHz): −1.30 (s, 6H, InCH₃), 1.42 (q, 2H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 1.78 (s, 3H, CCH₃), 2.14 (s, 6H, CH₃), 2.26 (s, 12H, CH₃), 2.54 (t, 4H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 6.77 (br s, 4H, CH_Ar), 7.03–7.07 (m, 2H, CH_Ar), 7.16–7.21 (m, 4H, CH_Ar), 7.51 (m, 4H, CH_Ar). ¹³C{1H} NMR (toluene-d₈, 100 MHz): −5.0 (InMe₂), 14.3, 18.4, 21.0, 23.1, 32.0, 33.7, 46.4, 77.6, 124.9, 126.9, 127.2, 127.6, 128.4, 129.2, 130.2, 135.6, 138.0, 141.9, 155.7, 203.2 (carbene). Data for 3 are as follows. Anal. Calcd. for C₃₈H₄₇GaN₂O·C₆H₈(toluene): C, 76.16; H, 7.81; N, 3.95. Found: C, 76.30; H, 8.14; N, 3.80. ¹H NMR (toluene-d₈, 400 MHz): −1.36 (s, 6H, GaCH₃), 1.45 (q, 2H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 1.78 (s, 3H, CCH₃), 2.14 (s, 6H, CH₃), 2.25 (s, 12H, CH₃), 2.58 (t, 4H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 6.74 (s, 4H, CH_Ar), 7.04 (m, 2H, CH_Ar), 7.17 (m, 4H, CH_Ar), 7.46 (m, 4H, CH_Ar); ¹³C{1H} NMR (toluene-d₈, 100 MHz): −1.2 (GaMe₂), 17.3, 18.5, 20.9, 21.0, 21.3, 32.6, 47.5, 77.7, 125.0, 125.6, 127.1, 127.8, 128.4, 129.2, 129.9, 135.5, 137.7, 142.0, 154.8, 156.3, 197.4 (carbene).

Synthesis of 5 and 6. To a solution of [Me₂In(μ-OC₆H₄OMe)]₂ (143 mg, 0.53 mmol) or [Me₂Ga(μ-OC₆H₄OMe)]₂ (118 mg, 0.53 mmol) in toluene (6 mL) was added 2 mL of a toluene solution of 6-Mes (171 mg, 0.53 mmol) at room temperature, and the resulting solution was stirred for 2 h. Then toluene was removed under vacuum to give a pale white solid. The solid was subsequently crystallized from toluene (1.5 mL) after addition of hexane (5 mL) at −20 °C to give white crystals, which were dried under vacuum (nr In, 283 mg, 90%; nr Ga, 249 mg, 86%). Data for 6 are as follows. Anal. Calcd. for C₃₁H₄₁InN₂O₂: C, 63.27; H, 7.02; N, 4.76. Found: C, 63.09; H, 7.20; N, 4.80. ¹H NMR (toluene-d8, 400 MHz): −1.00 (s, InCH₃), 1.56 (q, 2H, ³J (H,H) = 5.8 Hz, CH₂CH₂CH₂), 2.07 (s, 6H, CH₃), 2.31 (s, 12H, CH₃), 2.63 (t, 4H, ³J (H,H) = 5.8 Hz, CH₂CH₂CH₂), 3.29 (s, 3H, OCH₃), 6.49 (m, 2H, CH_Ar), 6.67 (br s, 4H, CH_Ar), 6.86 (m, 1H, CH_Ar), 6.96 (m, 1H, CH_Ar); ¹³C{1H} NMR (toluene-d₈, 100 MHz): −6.7 (InMe₂), 18.0, 20.9, 46.5, 54.1, 110.9, 112.7, 118.9, 121.8, 130.2, 135.6, 137.9, 141.8, 150.8, 157.3, 204.3 (carbene). Data for 5 are as follows. Anal. Calcd. for C₃₁H₄₁GaN₂O₂: C, 68.52; H, 7.61; N, 5.16. Found: C, 68.22; H, 7.56; N, 5.16. ¹H NMR (toluene-d₈, 400 MHz): −1.00 (s, GaCH₃), 1.55 (q, 2H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 2.06 (s, 6H, CH₃), 2.29 (s, 12H, CH₃), 2.65 (t, 4H, ³J (H,H) = 5.9 Hz, CH₂CH₂CH₂), 3.55 (s, 3H, OCH₃), 6.54 (m, 1H, CH_Ar), 6.62 (m, 1H, CH_Ar), 6.67 (br s, 4H, CH_Ar), 6.70 (m, 1H, CH_Ar), 6.80 (m, 1H, CH_Ar); ¹³C{1H} NMR (toluene-d₈, 100 MHz): −3.9 (GaMe₂), 18.1, 20.9, 47.4, 55.7, 113.3, 114.4, 119.8, 121.7, 129.9, 135.6, 137.8, 141.5, 152.1, 155.9, 196.1 (carbene).

3.3. General Procedure for the ROP of rac-LA with 1 and 3–6

To a rac-LA (334.8 mg, 2.32 mmol, 25 eq) suspension in toluene (10 mL) cooled to −20 °C was added a toluene solution (1 mL) of the catalyst (0.09 mmol, 1 eq), and the reaction was stirred overnight at −20 °C until the rac-LA fully dissolved. In the case of polymerization at room temperature, rac-LA dissolved fully after 10 min. Each polymerization was quenched by the addition of a HCl solution (5%, 50 mL). The organic phase was separated, washed twice with water (50 mL), dried over anhydrous MgSO₄ and dried under vacuum to give PLA as a white solid. ¹H NMR (CDCl₃, 400 MHz): (a) PLA signals, 1.46–1.55 (m, 3H, CHCH₃), 5.10–5.23 (m, 1H, CHCH₃); (b) end groups for PLA obtained with 1 3.36 (s, 3H, OCH₃), 3.57 (s, 2H, CH₂), 4.26 (s, 2H, CH₂). PLA (~0.2 g) was also precipitated from methylene chloride solution (0.5 mL) with 20 mL of MeOH, filtered off, and dried under vacuum to yield PLA with approximately 70% yield. The precipitated PLA was further analyzed by ¹H NMR, homodec NMR and ¹³C NMR (see Supplementary Materials).

3.4. X-Ray Structure Determination

Single crystals of 2, 3, 4 and (6-Mes)=CH₂ were grown from toluene/hexane solutions. Suitable single crystals were selected under a polarizing microscope and glued to a glass capillary. The diffraction data were collected on an Oxford Diffraction Gemini A ultra diffractometer at room temperature. Data collection and reduction were performed using the CrysAlis^PRO software developed by Rigaku Oxford Diffraction [39]. The structures were solved with the ShelXT structure solution program using Intrinsic Phasing and refined with the ShelXL refinement package using Least Squares minimization [40]. These programs were invoked from within Olex2 suite which was also used for the production of Figure 2, Figure 3, Figure 4 and Figure 5 [41].

Crystal Data for 2, C₂₅H₃₇InN₂ (M = 480.38 g/mol): monoclinic, space group P2₁/n (no. 14), a = 9.5680(3) Å, b = 16.9878(4) Å, c = 16.0384(5) Å, β = 106.868(3)°, V = 2494.73(13) ų, Z = 4, T = 293(2) K, μ(MoKα) = 0.959 mm⁻¹, D_calc = 1.279 g/cm³, 22528 reflections measured (6.5° ≤ 2Θ ≤ 52.0°), 4884 unique (R_int = 0.0219, R_sigma = 0.0162) which were used in all calculations. The final R₁ was 0.0245 (I > 2σ(I)) and wR₂ was 0.0646 (all data).

Crystal Data for 3, C₃₈H₄₇GaN₂O (M = 617.49 g/mol): monoclinic, space group P2₁/c (no. 14), a = 11.5312(3) Å, b = 18.7807(5) Å, c = 19.0639(5) Å, β = 100.916(3)°, V = 4053.87(19) ų, Z = 4, T = 293(2) K, μ(MoKα) = 0.704 mm⁻¹, D_calc = 1.012 g/cm³, 65065 reflections measured (6.7° ≤ 2Θ ≤ 53.0°), 8392 unique (R_int = 0.0504, R_sigma = 0.0305) which were used in all calculations. The final R₁ was 0.0373 (I > 2σ(I)) and wR₂ was 0.0947 (all data).

Crystal Data for 4, C₃₈H₄₇InN₂O (M =662.59 g/mol): tetragonal, space group I4₁cd (no. 110), a = 28.4159(12) Å, c = 19.2051(10) Å, V = 15507.4(15) ų, Z = 16, T = 293(2) K, μ(MoKα) = 0.636 mm^–1, D_calc = 1.135 g/cm³, 21599 reflections measured (7.1° ≤ 2Θ ≤ 50.0°), 6490 unique (R_int = 0.0763, R_sigma = 0.0914) which were used in all calculations. The final R₁ was 0.0505 (I > 2σ(I)) and wR₂ was 0.1532 (all data).

Crystal Data for (6-Mes)=CH₂, C₂₃H₃₀N₂ (M =334.49 g/mol): orthorhombic, space group Pbcn (no. 60), a = 15.7258(15) Å, b = 7.9668(8) Å, c = 16.0813(14) Å, V = 2014.7(3) ų, Z = 4, T = 293(2) K, μ(MoKα) = 0.064 mm⁻¹, D_calc = 1.103 g/cm³, 7777 reflections measured (7.2° ≤ 2Θ ≤ 53.0°), 2080 unique (R_int = 0.0267, R_sigma = 0.0224) which were used in all calculations. The final R₁ was 0.0552 (I > 2σ(I)) and wR₂ was 0.1570 (all data).

4. Conclusions

In summary, we have investigated the influence of steric and electronic properties of 1,3-bis(2,4,6-trimethylphenyl)-3,4,5,6-tetrahydropyrimidin-1-ylidene (6-Mes) on the synthesis, structure and reactivity of Me₂MOR(NHC). Particularly, we have focused on the M–C_6-Mes bond in Me₂MOR(6-Mes), how it is influenced by the structure of 6-Mes in comparison with other NHCs, and how it can influence the structure of Me₂MOR(6-Mes) as well as their catalytic activity in the ring-opening polymerization (ROP) of racemic lactide (rac-LA). The considerable effect of steric hindrances of 6-Mes was reflected both by significantly longer, and therefore weaker M–C_6-Mes in the solid state in comparison to M–C_SIMes and M–C_IMes, which was revealed by X-ray analysis and the presence of equilibrium between Me₂MOR(NHC) and [Me₂M(µ-OR)]₂/NHC in solution. The stronger donor properties of 6-Mes, in comparison with other NHCs, were revealed in solution by significant shift of carbene carbon in ¹³C NMR upon coordination of 6-Mes to Ga and In of Me₂MOR(6-Mes) (M = Ga, In) and Me₃In(6-Mes). However, in light of other results, including X-ray analysis, they could not indicate the presence of a stronger M–C_6-Mes bond for investigated complexes in comparison with M–C_SIMes, M–C_IMes, and M–C_SIPr of their Me₂MOR(NHC) (M = Ga, In; NHC = SIMes, IMes) and Me₃In(NHC) (NHC = SIMes, IMes, SIPr) analogues. The formation of In–C_6-Mes bond in the case of dimethylindiumalkoxides was not sufficient in order to stabilize Me₂InOR(NHC) complexes. Finally, the investigation of the reactivity of Me₂M(OCH(Me)CO₂Me) towards 6-Mes, with regard to the structure of propagating Me₂M(OPLA)(6-Mes) species in the ROP of rac-LA, allowed to clarify the mechanism of the reaction of Me₂InOR(NHC) towards sterically hindered NHCs. With regard to the catalytic activity of Me₂MOR(NHC), the effect of Ga–C_6-Mes and a resulting structure of Me₂GaOR(NHC) influenced mostly stereoselectivity, as well as the extent of transesterification reactions in the polymerization of rac-LA, which demonstrated the crucial role of NHC on the catalytic properties of Me₂GaOR(NHC). The mechanism of the ROP of rac-LA was in line with the insertion of rac-LA into Ga–O_alkoxide bond, although the insertion into Ga–C_6-Mes could not be excluded. As a result of even weaker In–C_6-Mes bond in comparison to other In–C_NHCs in Me₂InOR(NHC) the insertion of rac-LA into In–C_6-Mes occurred and was not surprising in the light of the polymerization of rac-LA with previously described by us Me₂InOR(NHC) (NHC = SIMes, IMes). Importantly, the reported studies, concerning both the structure and catalytic activity of Me₂MOR(6-Mes), revealed the effect of the character of M–C_NHC, rather than only its strength, which is the subject of our current research.