Enantiomerically Pure ansa-η⁵-Complexes of Transition Metals as an Effective Tool for Chirality Transfer

Abstract

Chiral ansa-η⁵-complexes of transition metals have shown remarkable efficacy in organometallic synthesis and catalysis. Additionally, enantiomerically pure ansa-complexes hold promise for the development of novel chiral materials and pharmaceuticals. The discovery and synthesis of a diverse range of group IVB and IIIB metal complexes represents a significant milestone in the advancement of stereoselective catalytic methods for constructing metal-C, C-C, C-H, and C-heteroatom bonds. The synthesis of enantiomerically pure metallocenes can be accomplished through several strategies: utilizing optically active precursors of η⁵-ligands, separation of diastereomers of complexes with enantiomerically pure agents, and synthesis via the stereocontrolled reactions of enantiomerically pure σ-complexes with prochiral anions of η⁵-ligands. This review focuses on the analysis of various nuances of the synthesis of enantiomerically pure ansa-η⁵-complexes of titanium and lanthanum families. Their applicability as effective catalysts in asymmetric carbomagnesiation, carbo- and cycloalumination, oligo- and polymerization, Diels–Alder cycloaddition, reactions of zirconaaziridines, cyclization, hydrosilylation, hydrogenation, hydroamination, and other processes are highlighted as well.

1. Introduction

The discovery of metallocene compounds in the mid-20th century stimulated the development of several significant areas within organometallic chemistry [1,2]. Metallocenes have exhibited outstanding catalytic properties in the formation of metal–carbon (M-C), carbon–carbon (C-C), and carbon–hydrogen (C-H) bonds. Among these, single-site alkene polymerization using catalytic systems composed of IV subgroup metal η⁵-complexes and organoaluminum or -boron compounds has emerged as a particularly promising avenue [3,4,5,6,7,8,9,10,11]. Furthermore, various catalytic methods of alkene and acetylene transformations, such as di- and oligomerization [12,13,14,15,16,17,18,19], hydrometalation [20,21,22,23], carbometalation [24,25,26,27], and cyclometalation [21,24,28,29,30], are well known. Particular interest is drawn to the application of metallocenes as biosensors, antitumor agents, and other pharmaceuticals in medicine [31,32,33,34,35].

The high activity, chemo-, and stereoselectivity of metallocene complexes can be attributed to their unique structural characteristics. These compounds effectively ensure the stability of the electronic and steric environment of the transition metal center due to the high bond energy of the metal–ligand interactions. Additionally, they allow for variations in electrophilicity and geometry of the catalytically active sites through extensive structural modifications of the π-ligands. The introduction of bridging groups has led to the development of ansa-complexes with distinct properties. The global literature describes the synthesis of numerous ansa-metallocene complexes [36,37,38,39,40], which have found diverse applications in organic and organometallic chemistry. Notably, the binding of η⁵-ligands increases electrophilicity and accessibility of the catalytically active center by altering the coordination geometries of the π-ligands on the metal, while also fixing the geometry of the complex. This, combined with planar chirality, endows these compounds with significant potential for catalytic stereo- and enantioselective construction of C-C, C-H, C-N, and C-O bonds.

The generation of stereoisomers of ansa-complexes is facilitated by the presence of both enantiotopic and diastereotopic faces in the bridging ligands, whose reaction with metal salts or complexes is accompanied by a stereodifferentiating effect [39,41]. Stereoisomers can be synthesized through various pathways. First, when using nonchiral metal salts or complexes, additional separation of the racemic mixture of the resulting ansa-derivatives with an enantiomerically pure reagent is necessary (Scheme 1, path A). Second, both diastereoselective and enantioselective synthesis of ansa-complexes can be achieved from enantiomerically pure ligands that contain stereogenic centers in both the substituents at the η⁵-ligands and the bridging group (Scheme 1, path B). Third, the use of metal complexes with chiral elements in σ-substituents can determine stereoselectivity at the π-coordination stage of prochiral ligands (Scheme 1, path C).

Enantiomerically pure ansa-complexes of III and IV subgroup metals, characterized by fixed geometries of η⁵-ligands, have demonstrated high efficiency in generating stereogenic centers on carbon atoms in prochiral substrates such as alkenes, ketones, imines, etc. Several reviews in the literature have focused on the synthesis of optically active metallocenes and their applications in organic chemistry. Reviews [39,41] detail methods for synthesizing chiral metallocenes, including those with ansa-bound η⁵-ligands. Review [36] discusses the synthesis of enantiopure planar chiral organometallic complexes via diastereoselective complexation methodologies. Additionally, the mini-review [40] presents various approaches for the synthesis of chiral ansa-metallocene complexes of Ti, Zr, and Hf that incorporate carbon bridging groups. Insights into the use of chiral C₂-symmetric complexes of Zr and Ti as catalysts for the asymmetric construction of C-C and C-heteroatom bonds are provided in review articles [25,42].

This review considers and systematizes information on the approaches to the synthesis of enantiomerically pure ansa-η⁵-complexes of group IIIB and IVB transition metals, as well as various aspects of their application for asymmetric induction in the construction of C-C, M-C, C-H, and C-heteroatom bonds.

2. Synthesis of Enantiomerically Pure ansa-bis-η⁵-Complexes of Group III and IV Transition Metals

2.1. Path A: Resolution of Racemic Complexes into Enantiomers via Derivatization

The primary method for achieving enantiomerically enriched ansa-metallocenes is the kinetic resolution of rac-isomers of complexes through the use of derivatizing reagents (Scheme 1, path A).

In 1979, Brintzinger et al. first synthesized the optical isomer p-S, p-S-propandienyl-bis-(tert-butylcyclopentadienyl)titanium binaphtholate (S,p-S,p-S-3a) by the reaction of dichloride rac-complex 2a with an equivalent amount of S-binaphthol (S-BINOL) in the presence of sodium amide in toluene (Scheme 2). The diastereomerically pure binaphtholate complex 3a was isolated by column chromatography [43] and subsequently converted to the corresponding dichloride. Later, Collins et al. successfully resolved rac-ethylene-bis-tert-butylcyclopentadienyltitanium dichloride 2b into its enantiomers using S-binaphthol [44].

A combination of photoisomerization and treatment with lithium R-binaphtholate was employed to obtain enantiomerically pure Zr ansa-cyclopentadienyl complexes featuring Si-bound ligands [45] (Scheme 3). This approach yielded compounds 7 and 8 with yields reaching up to 88% and enantiomeric purity exceeding 95%, as determined by ¹H NMR spectroscopy of the carboxylate complex R,R,p-S-10.

The enantiomers of the bis-tetrahydroindenyl complex rac-11 were derivatized with S-binaphthol [46] (Scheme 4). This reaction produced two enantiomers, S,p-S,p-S-12 and p-R,p-R-11, with yields of 77% and 70%, respectively, while the enantiomeric purity of p-R,p-R-11 was 75%. Following isolation, the binaphtholate complex S,p-S,p-S-12 was converted into a dimethyl analog, which was treated with hydrochloric acid to yield a red precipitate of titanocene dichloride p-S,p-S-11.

Optically active acids present an alternative to enantiomerically pure binaphthols for the resolution of chiral ansa-metallocenes [47]. The corresponding carboxylate complexes were synthesized through various methods: (i) converting metallocene dichloride complexes into dimethyl analogs followed by their reaction with optically active carboxylic acids; (ii) reacting metallocene dichlorides with lithium or sodium salts of acids; and (iii) treating metallocene dichlorides with acids in the presence of triethylamine. It was demonstrated that metallocenes react almost completely with optically active acids, such as R-acetylmandelic acid in the presence of Et₃N (Scheme 5), yielding diastereomeric carboxylate complexes 14 and 15. These complexes can be efficiently separated by fractional crystallization, achieving yields of 77–88% and minimizing the loss of the desired diastereomerically pure complexes. The conversion of carboxylates back to dichloride complexes p-S,p-S-11 and p-S,p-S-13 was accomplished via dimethyl derivatives followed by treatment with hydrochloric acid in ether (yield > 50%).

The derivatization method for ansa-complexes described above remains the most widely utilized approach. Subsequent research has focused on refining this method to increase the yield and enantiomeric purity of the target metal complexes.

To minimize the loss of enantiomers of complex 11, the racemic mixture was subjected to reflux with sodium and S-binaphthol (Scheme 6) [48,49]. This process increased the yield of enantiomerically pure complex S,p-S,p-S-12 to 78%, while dichloride p-R,p-R-11 was isolated at a yield of 40%. Following treatment of S,p-S,p-S-12 with a solution of MeLi in hexane and subsequently HCl in ether, p-S,p-S-11 was obtained with a yield of 68% and an enantiomeric purity of 93%, as confirmed by ¹H NMR spectroscopy of the carboxylate complex R,R,p-S,p-S-14.

The enantiomerically pure Zr complex p-S,p-S-17 was synthesized through several stages, beginning with racemic 13, which underwent a reaction with dilithium binaphtholate, followed by treatment with aluminum powder in dichloromethane (Scheme 7), resulting in the formation of complex S,p-S,p-S-16. Further alkylation of S,p-S,p-S-16 with MeLi yielded the dimethyl complex p-S,p-S-17 with a yield of 63% and an enantiomeric purity of 98%ee [50]. This approach also led to the synthesis of p-S,p-S-18 with a yield of 14% [51].

The resolution of racemic ansa-cyclopentadienyl Ti (20a) and Zr (21a) complexes containing biphenyl bridging groups was successfully achieved using R-binaphtholate, while acetylmandelic acid was successfully applied only to ansa-L₂TiCl₂ (Scheme 8) [52,53]. This process resulted in the acquisition of enantiomerically pure complexes R,R,R-20b and R,R,R-21b. The reaction of titanocene 20a with R-acetylmandelic acid produced a mixture of diastereomers 24 (the yield was determined by ¹H NMR of the reaction mixture), which could not be separated via column chromatography.

Complexes rac-11 (Ti) and rac-13 (Zr) were derivatized using a mixture of R-binaphthol and p-aminobenzoic acid in toluene under reflux with an excess of triethylamine (Scheme 9) [54,55]. This reaction yielded binaphtholate complexes R,p-R,p-R-12 (34%) and R,p-R,p-R-16 (37%), along with aminobenzoic acid derivatives p-S,p-S-25 and p-S,p-S-26, which were separated based on their differing solubilities in non-polar solvents. The titanocene binaphtholates were converted into enantiomerically pure dichloride complexes p-R,p-R-11 (97%) and p-S,p-S-11 (33%) through treatment with hydrochloric acid in dichloromethane, while the Zr p-aminobenzoate was transformed into biphenolate S,p-S,p-S-27 (27%, 97%ee) via reaction with 2,2′-biphenol.

Bidentate phenolates of metallocenes rac-13 (Zr) and rac-28 (Hf) (Scheme 10) were quantitatively resolved into enantiomers using HPLC on cellulose (3,5-dimethylphenylcarbamate) as the chiral stationary phase, with a hexane/ethanol (9:1 v/v) mixture serving as the eluent [56]. The specific rotation exhibited a positive value for the first peak (S) and a negative value for the second (R). However, the dichlorides 13a and 13b were not separated into enantiomers under these conditions. The method was further extended by employing new commercially available bidentate ligands (c–h) for the derivatization of rac-13 [57]. In all cases, the (+)-isomers were eluted first, with the highest separation factor observed for complex rac-13d. Optically active biphenolate hafnocenes R,p-R,p-R-28d (α²⁵₄₃₅ = −1100°) and S,p-S,p-S-28d (α²⁵₄₃₅ = +1100°) were also isolated.

Enantiomerically pure triflate derivatives of ansa-metallocenes Ti (28) and Zr (30) were synthesized starting from titanocene p-S,p-S-11 and zirconocene binaphtholate S,p-S,p-S-16, respectively, as depicted in Scheme 11 [58,59,60]. Additionally, the cationic Zr compound p-S,p-S-29 was obtained.

2.2. Path B: Stereoselective Synthesis of ansa-Metallocenes via Metalation of Chiral η⁵-Ligand Precursors

Optically active C₂- and C₁-symmetric ansa-complexes can be synthesized from enantiomerically pure η⁵-ligand precursors (path B). This method typically results in high diastereo- and enantioselectivity in the synthesis of metallocenes.

For example, optically active C₂-symmetric titanium (p-S,p-S-34) and zirconium (p-S,p-S-35) complexes were synthesized via the metalation of ligand S-33, which features a biphenyl bridge, achieving yields of 46% and 20%, respectively (Scheme 12) [61]. The ligand was obtained by separating the dicarboxylic acid R,S-31 into enantiomers R-32 and S-32 using (+)-brucine. The reaction of enantiomer S-32 with a dimagnesium Grignard reagent, followed by protonolysis, yielded enantiomerically pure ligand S-33 with a 60% yield.

A similar approach was employed for the synthesis of a C₂-symmetric optically active titanocene 38, which contains a homotopic bis-tetrahydroindenyl ligand. This complex was synthesized from enantiomerically pure Br-substituted binaphthalene S-(+)-36, yielding only one stereoisomer of complex 38 with a yield of 68% (Scheme 13) [62]. The configuration of the binaphthyl fragment controls the structure of the complex, thereby excluding the formation of diastereomeric pairs during transmetalation.

Halterman and co-workers proposed a method for the synthesis of enantiomerically enriched complexes 41, 42, and 44 (Scheme 14) based on optically active ligands obtained from Br-substituted binaphthyl structures 39 [62,63]. The ligand R-(+)-39 was converted into (R)-(+)-2,2′-bis(1-indenylmethyl)-1,1′-binaphthyl (40), which upon coordination with titanium (41) and zirconium (42) formed only one C₂-symmetric stereoisomer (Scheme 14). This method was later adapted for substituted indene [64]. For example, the reaction of 4,7-dimethylindenyllithium with S-(−)-39 produced the ligand S-(−)-43 with a 69% yield. The subsequent reaction of the anion of ligand 43 with TiCl₃ produced complex 44 with a 24% yield. The developed method provided hard-to-reach enantiomerically pure bis-indenyl complexes; in this process, the substitution of Ti with Zr or introducing methyl groups into the indenyl ligand significantly reduced the yield of the target metallocenes.

Enantiomerically pure bis-indenyl ligand S,S-46 was synthesized from the mesylate of (R,R)-2,3-butanediol 45 (Scheme 15) [65]. Deprotonation of S,S-46 followed by transmetalation to a transition metal and hydrogenation resulted in a mixture of diastereomeric complexes, with the content of isomer C increasing when Ti was replaced with Zr. The optically active Ti complexes 47 were subjected to photolysis, after which the mixture reacted with R-(+)-binaphthol and was separated on silanized silica gel. This process yielded enantiomerically pure titanocenes p-R,p-R-47A (45%) and p-R,p-S-47C (14%). For zirconium, the yield of the target metallocenes was significantly lower, at 9%, for the isomers p-R,p-R-48A and p-S,p-S-48C, with a ratio of 5.9:1.

The reaction of the bis(methanesulfonate) ester of 2,4-diisopropyl-1,4-cyclohexanediol (50) with IndLi in THF produced ligand (+)-(1R,2R,4R,5R)-51, which upon metalation with BuLi and TiCl₃, followed by oxidative workup (HCl, air, chloroform), selectively yielded enantiomerically pure titanocene 52 with a high yield (Scheme 16) [66]. The corresponding bis(indenyl)zirconium dichloride was not obtained from this route. Complex 52 was hydrogenated over PtO₂ to give 1R,2R,4R,5R-53 in good yield.

High stereoselectivity in the synthesis of titanocene complex p-S,p-S-58 was achieved when the optically active ligand (1R,2R)-trans-1,2-bis-(1-indenylmethyl)cyclopentane 55, obtained from di-(1R,2S,5R)-menthyl-(1S,2S)-trans-1,2-cyclopentanedicarboxyl 54, was reacted with BuLi and TiCl₄·2THF (Scheme 17) [67]. Subsequent hydrogenation of p-S,p-S-58 yielded p-S,p-S-59.

Using the less conformationally rigid C₄-bridged (4S,5S)-trans-4,5-bis(1H-inden-3-ylmethyl)-2,2-dimethyl-1,3-dioxolane ligand 60, obtained from enantiomerically pure tosylate 59 as a chiral substrate, resulted in a loss of racemo-selectivity and the formation of optically active meso-isomers of complexes 62a–c and 63a–c (Scheme 18) [68].

The formation of the meso-isomer (66) was also observed in the case of (−)-dimethyl-bis-[3-(neomenthyl)cyclopentadienyl]silane 65, derived from neomenthol 1S,2S,5R-64. In the ¹H NMR spectra of the reaction mixture, signals for three possible isomers were observed with a ratio of 5:2:3. However, after recrystallization from pentane, only one optically active metallocene meso-66 was isolated with a yield of up to 34% (Scheme 19) [69].

(+)-(1S,2R,5S)-Menthol (1S,2R,5S-64) was introduced as a chiral element in the bridging group of the bis-indenyl ligand, leading to the synthesis of enantiomerically pure di[(1′S,2′R,5′S)-menthoxy]silylene-bis[η⁵-1-(R,R)-(+)-indenyl] zirconium dichloride (p-R,p-R-68) with a yield of up to 19% (Scheme 20) [70].

Various isomers of ethylene-bridged cyclopentadienyl titanium complexes 73–75 were synthesized from the sterically hindered enantiomerically pure (−)-8,10-dimethyltricyclo[5.2.2.0²,⁶]-2,5-undecadiene 69 (Scheme 21) [71]. It was observed that similar complexes with silicon or methylene bridges could not be synthesized. The metalation stage of ligand 70 occurred with varying selectivity, as this precursor represented a difficult-to-separate mixture of isomers differing by the position of the bridge. Consequently, complexes 73–75 were obtained with yields of up to 84%, with a ratio of 73/74/75 = 71:7:22.

To restrict the conformational mobility of the cyclohexane fragment of the tetrahydroindenyl ligand and to stabilize the exo-, exo-configuration (Scheme 22), ligand 80 based on (R)-myrtenal 76 was synthesized. Transmetalation of the ligand with Zr(NMe₂)₄ produced a single C₂-symmetric stereoisomer of complex 81 with an exo-, exo-configuration of the cyclohexyl rings, achieving a yield of 35% [72].

The reaction of optically active bis-amide R,R-82 with cyclohexenylmagnesium bromide lithium chloride yielded the corresponding bis-enone 83 without loss of enantiomeric purity (>99.9%ee) (Scheme 23) [73]. Through the stages of methylenation of 83 and bis-dibromocyclopropanation of compound 84, the ligand precursor 85 was obtained. The rearrangement into a chiral C₂-symmetric ligand was performed using MeLi, followed by deprotonation with sodium bis(trimethylsilyl)amide and reaction with TiCl₃·3THF without isolation. Three isomers—(S,S,R,R)-47B, the pseudoenantiomeric complex (R,R,R,R)-47A, and the pseudomeso-complex (R,S,R,R)-47C—were isolated in a ratio of B:A:C = 3.6:1:2.8. This method allows for the rapid synthesis of bis-tetrahydroindenyl complexes with good rac-selectivity from alkenyl Grignard reagents and bis-amides.

Another strategy for obtaining optically active ansa-metallocenes via Path B involves the synthesis of complexes with nonequivalent ligands (C₁-symmetric complexes). In this scenario, the chirality of the complex is introduced by incorporating a substituent with stereogenic centers into one of the η⁵-ligands.

For example, a method for synthesizing optically active ansa-metallocenes 88–90a,b from Si-bridged bis-cyclopentadienyl ligands 86a and 86b, where one of the Cp fragments contains an enantiomerically pure menthyl or neomenthyl substituent, was developed (Scheme 24) [74,75]. The reaction with TiCl₃ yielded a mixture of diastereomers R- and S-88a with a ratio of 3:1. The titanium complex R-88a was crystallized from hot hexane, while the S-enantiomer 88a was obtained after UV irradiation of a mixture of complexes in toluene, followed by passing the solution through silylated silica gel and crystallization from hot pentane. Zr (89a,b) and Hf (90b) complexes were synthesized similarly. The content of the R-enantiomers of ansa-metallocenes increased with the growing atomic radius of the transition metal when using the menthyl-substituted ligand 86a. In all cases, replacing the menthyl group with a neomenthyl group led to a loss of diastereoselectivity.

Following a similar approach to bis-cyclopentadienyl complexes, optically active mixed menthyl-substituted cyclopentadienyl-octahydrofluorenyl ansa-complexes 93 and 94 were synthesized (Scheme 25) [76]. For this purpose, ligand 92 was metalated via reaction with Zr(NMe₂)₄ (route A, Scheme 25) in toluene at 120 °C. Subsequent treatment with Me₂NH·HCl produced a mixture of diastereomers of complex 93 with a yield of 82% and a ratio S/R = 85:15. Deprotonation of 92 with butyllithium (route B, Scheme 25) and transmetalation with ZrCl₄ yielded stereoisomers of complex 93 with a low yield.

Neomenthyl-substituted ansa-complexes with an isopropylidene bridge (98) were synthesized (Scheme 26) [77]. In the initial stages of synthesis, the menthyl-substituted fulvene 96 reacted with lithium fluorenide to yield precursor 97. The metalation of ligand 97 resulted in a mixture of diastereomeric complexes 98 in a ratio of 60:40. Thus, introducing the neomenthyl group and isopropylidene bridge into the ligand decreased the stereoselectivity of complex synthesis.

High diastereoselectivity was achieved in the synthesis of the enantiomerically pure (>99%ee) indenyl-fluorenyl ansa-zirconocene with a naphthyl bridge p-S-103 (Scheme 27) [78]. The ligand 102 used in this synthesis was obtained by the reaction of (1R)-menthyl-(S)-1-(p-tolylsulfinyl)naphthalene-2-carboxylate (100) with 1-methylfluorenyllithium at 0 °C. The resulting ester reacted with a Grignard dimagnesium reagent to afford the R-enantiomer of the ansa-ligand 102, which upon metalation selectively produced complex p-S-103.

The reaction of R-(−)-epoxystyrene 104 with fluorenyllithium and subsequent transformations through various stages provided enantiomerically pure ansa-ligands S-107 and S-108 with a stereogenic center in the bridge (Scheme 28) [79,80]. Deprotonation of S-107 with ZrCl₄ yielded diastereomers R,p-R-109 and R,p-S-109 in a 1:3 ratio. Hydrogenation of 109 with H₂/PtO₂ produced corresponding R,p-R-110 and R,p-S-110 in a 1:1 ratio. The metalation of bis-fluorenyl ligand 108 afforded Zr and Hf complexes R-111a,b with yields of 28% and 63%, respectively. Octahydrofluorenyl complexes R-112a,b were obtained following the hydrogenation of R-111a,b.

Enantiomerically pure C₁-symmetric ansa-complexes S-116a–c were synthesized from optically active cyclopentadiene with a branched hydrocarbon substituent (115) (Scheme 29) [81]. To enhance the rigidity of the complex molecule, the ligands were linked by two Me₂Si-bridges.

The presence of two different substituents in one of the cyclopentadienyl fragments of ligand 121 led to the formation of two isomers of complexes S-122 in a 1:1 ratio when using ZrCl₄ (Scheme 30) [81]. Utilizing Zr(NMe₂)₄ as a metalation agent significantly increased the content of one of the possible diastereomers of complex S-123.

The literature also provides examples of preparing enantiomerically pure complexes of group IIIB metals [82]. The synthetic approaches to chiral organolanthanide ansa-η⁵-complexes have been discussed, with most work focusing on the synthesis of C₁-symmetric compounds with stereogenic substituents in one of the cyclopentadienyl rings (

Table 1. The synthesis of C₁-symmetric enantiomerically pure ansa-complexes of lanthanides [84].

Entry	Complex	R*	Solvent for Synthesis (Crystallization)	Yield, %	Ratio of Epimers R:S (Solvent)
1	127a (La)	(+)-neomenthyl	THF (Et2O)	47	75:25 (THF)
2	128a (Nd)	(+)-neomenthyl	THF (Et2O)	65	74:26 (THF)
3	129a (Sm)	(+)-neomenthyl	THF (Et2O)	47	74:26 (THF)
4	129a (Sm)	(+)-neomenthyl	THF (Et2O)	58	34:66 (Et2O)
5	130a (Y)	(+)-neomenthyl	THF (Et2O)	63	74:26 (THF)
					46:54 (Et2O)
6	131a (Lu)	(+)-neomenthyl	THF (Et2O)	57	74:26 (THF)
					51:49 (Et2O)
7	129b (Sm)	(−)-menthyl	THF (Et2O)	60	>95:5 (Et2O)
					20:80 (THF)
9	130b (Y)	(−)-menthyl	THF (Et2O)	64	20:80 (THF)
					90:10 (Et2O)
10	131b (Lu)	(−)-menthyl	THF (Et2O)	70	20:80 (THF)
					85:15 (Et2O)
11	130c (Y)	(−)-phenylmenthyl	THF (Et2O)	48	6:94 (THF)
					83:17 (Et2O)
12	129b, S (Sm)	(−)-menthyl	THF, DME (Et2O)	88
13	131b, S (Lu)	(−)-menthyl	THF, DME (Et2O)	89
14	129a, R (Sm)	(+)-neomenthyl	THF, DME (Et2O)	54

) [83,84]. A series of enantiomerically pure La, Nd, Sm, Y, and Lu complexes 127–131a–c were obtained via metalation of optically active cyclopentadienyl ligands with neomenthyl, menthyl, and phenylmenthyl substituents 124a–c (Scheme 31). The study also demonstrated the potential for the epimerization of chloro lanthanide complexes to enhance the yield and isolation of their diastereomeric antipodes. It was shown that the degree of epimerization depends on the nature of the donor solvent, the lanthanoid ion, the structure of the substituent R*, and the temperature (Table 1). The epimerization of chloride complexes 127–131a–c can be completely inhibited by adding ligands like TMEDA or 12-crown-4. To obtain stable lanthanide complexes effective in asymmetric catalysis, which do not undergo epimerization in solution, complexes 127–131a–c were converted into amides 132–135a–c and alkyls 136–139a–c (Scheme 31).

Chiral C₁-symmetric organolanthanide complexes S-140a–c (M = Sm, Y, Lu), with ansa-bridged η⁵-octahydrofluorenyl and cyclopentadienyl ligands substituted with a menthyl group, were synthesized analogously to complexes 127–131a–c (Scheme 32) [85]. The highest diastereoselectivity, reaching a ratio of 9:1, was attained when the reaction was performed in dimethoxyethane (DME). The reaction of the diastereomerically enriched complexes S-140a–c with KN(TMS)₂ proceeded to furnish S-141a–c in quantitative yield. Remarkably, at ambient temperature, the process preserved the diastereomeric ratio almost entirely, mirroring that of the initial chloro complexes. The resulting amido complexes exhibited notable thermal stability and were isolated via extraction with boiling hexane. A single recrystallization from hot hexane afforded exclusively the diastereomerically pure amido complexes S-141a–c.

Bercaw and colleagues [86] synthesized enantiomerically pure C₂-symmetric yttrocenes (R,S)-144, characterized by stereogenic centers within the bridging moiety. This was accomplished through the preliminary preparation of enantiomerically pure ansa-ligands 143 via the reaction of R- or S-binaphthol with tert-butyl-substituted bis(cyclopentadienyl)chlorosilane (142) (Scheme 33). Consequently, coordination to yttrium proceeded with complete diastereoselectivity: the ligand derived from R-binaphthol yielded (R,S)-144, whereas the ligand obtained from S-binaphthol directed the reaction exclusively toward the formation of (S,R)-144.

2.3. Path C: Stereoselective Synthesis of ansa-Metallocenes via Enantiomerically Pure σ-Complexes

Enantiomerically pure σ-complexes can serve as effective stereodifferentiating precursors. For example, the synthesis of the R,R-enantiomer of amide complex 148, which achieved a yield of 84%, was carried out starting from R,R-diphenylpropanediamine (147) (Scheme 34) [87]. Subsequent reactions of amide 148 with lithium salts of ansa-ligands containing ethanediyl or Me₂Si bridges provided enantiomerically pure amide complexes 149 and 150 with yields of 97% and 95%, respectively. Compound 149 was transformed into the bis-indenyl complex p-S,p-S-151, attaining a yield of 91% and enantioselectivity exceeding 99%ee.

The diastereo- and enantioselective synthesis of the ethylene-bis-tetrahydroindenyl zirconium complex 13 was proposed via the association of ZrCl₄ (R- or S-153) with enantiomerically pure sterically hindered S- or R-binaphthol ethers (Scheme 35) [88]. A higher enantioselectivity (95–96%ee) was observed in the reaction of binaphthol dimethyl ether complexes with the ansa-ligand dianion.

The growing interest in enantiomerically pure metallocenes is primarily motivated by their potential application as catalysts in various asymmetric syntheses. Another significant aspect is the investigation of stereoinduction and chiral recognition mechanisms, which facilitate the understanding of the structural features and dynamics of catalytically active centers, as well as the processes of substrate coordination and transformation by metal centers. In this regard, the subsequent chapters of this review will explore the applications of enantiomerically pure group IVB and IIIB transition metal ansa-η⁵-complexes, along with some mechanistic aspects of their asymmetric action across various types of reactions.

3. Application of Enantiomerically Pure ansa-η⁵-Complexes in Asymmetric Catalysis

3.1. Carbomagnesation of Unsaturated Compounds

The complexes of group IVB transition metals are widely acknowledged as effective catalysts for the formation of metal–carbon, carbon–carbon, and carbon–hydrogen bonds. The synthesis of Brintzinger’s ansa-titanocene and zirconocene complexes subsequently facilitated the development of highly active and stereoselective catalytic systems for alkene polymerization [3,4,5,6,7,8,9,10,11,89,90,91,92,93,94,95,96,97,98,99,100,101,102]. These complexes are also well known for their capability to catalyze di- and oligomerization reactions, as well as carbometalation and cyclometalation of alkenes and acetylenes [12,13,14,15,16,17,18,19,20,21,24,25,26,27,28,29,30,42,103,104,105,106,107,108]. A pivotal step in these reactions is the carbometalation of the unsaturated substrate, which generates a highly reactive metal–carbon bond. Consequently, the introduction of enantioselective catalysts into these reactions has enabled the development of efficient methodologies for the functionalization of alkenes, dienes, and acetylenes, yielding a wide range of optically active synthons that are in demand in organic synthesis [24,25,26,27,42,107,108,109,110,111,112].

The first report detailing the application of the enantiomerically pure zirconocene complex p-R,p-R-13 as a stereoselective catalyst for the carbomagnesiation of alkenes was presented in 1993 [113]. Ethylmagnesiation of the allylic R- and S-enantiomers of alcohols 154a,c and esters 154b,d,e (Scheme 36) exhibited varying diastereoselectivity contingent upon the type of starting enantiomer. It was demonstrated that high diastereoselectivity could be attained when employing S-enantiomers of allylic alcohols (syn:anti = 92:8) and R-enantiomers of esters (syn:anti = 7:93). The reaction with R-norbornene alcohol 154f proceeded with high regioselectivity, favoring the formation of the 2R,5S-enantiomer of the ethylmagnesiation product 155f.

In further developments of the method, the reaction of alkyl magnesium halides with readily available heterocyclic alkenes, catalyzed by the enantiomerically pure complexes p-R,p-R-13 and R,p-R,p-R-16, was investigated (Scheme 37) [114,115]. Specifically, the reaction of cyclic olefins 156a–d with EtMgCl or n-PrMgCl in the presence of p-R,p-R-13 yielded carbomagnesation products 157a–g with good yields (40–77%) and excellent enantioselectivity (over 90%ee). In the carbomagnesation reaction of cyclic amines 156b,e,j,i, the carbometalation proceeded with product yields of 54–81% and enantioselectivity exceeding 95%ee.

Utilizing rac-5,6-dihydropyrans 158a–i as substrates, carbomagnesation products 159a–i were obtained with high enantiomeric purity. Moreover, enantiomerically enriched 158a–i were isolated as a result of the kinetic resolution of the racemic mixture of the starting compounds, due to the participation of only the S-enantiomer in the reaction (Scheme 38) [115,116]. Thus, asymmetric catalytic carbomagnesation proves to be an effective method for the kinetic resolution of enantiomers of unsaturated pyrans, which are important synthons for medicinal chemistry [116].

To investigate the kinetic effect of p-R,p-R-13 in the ethylmagnesiation reaction, enantiomerically pure substituted pyrans were employed. It was found that the reaction was most favorable for S-enantiomers of olefins (S-158d,k), which subsequently provided R-enantiomers of 160d, 159k with good yields and enantioselectivity (Scheme 39) [115].

Substituted dihydrofurans 162a–d also undergo kinetic resolution in the carbomagnesation reaction [117]. Unlike pyrans, both enantiomers of substituted furans fully participate in the reaction with the Grignard reagent (Scheme 40). In this case, the reaction proceeded with high regio-, enantio- (>98%ee), and diastereoselectivity (95%de) concerning both enantiomers of the substrates, resulting in the formation of different products: the S-isomer yielded only primary alcohol S-162a–d, while the R-enantiomer produced the secondary alcohol R,S-163a–d.

The application of the binaphtholate complex p-R,p-R-16 provided a high degree of enantiomer separation for seven-membered cycloolefins 164a–i, predominantly resulting in the isolation of R-enantiomers 165 (Scheme 41) [115,118]. The use of the biphenol complex p-R,p-R-27 yielded nearly identical results. A decrease in enantioselectivity to 70%ee was observed for the cyclohexyl-substituted substrate 165c.

During the carbomagnesiation reaction catalyzed by R,p-R,p-R-16 or S,p-S,p-S-16, cyclic alkoxy-substituted olefins 166a–v were also subjected to kinetic resolution, leading to the enrichment of the mixture with S-enantiomers exhibiting purities ranging from 50% to 99%ee (Scheme 42) [119,120,121]. In this case, the highest degree of separation was achieved in reactions involving seven-membered cyclic substrates 166k,l and 166q,r,u,v.

Thus, cyclic olefins in the presence of enantiomerically pure ansa-tetrahydroindenyl Zr complexes and Grignard reagents undergo kinetic resolution, during which one of the enantiomers can participate in an ethylmagnesiation reaction with ring opening. The structure of the starting olefins plays a significant role in these processes. For example, the enantioselectivity of the reaction is influenced by the size of the olefinic cycle (5-, 6-, 7-, or 8-membered) and the structure of the alkyl group in the R-O moiety. It is noted that increased steric hindrance of the substrate, both from the larger cycle and the substituent, leads to a loss of enantioselectivity in the process.

The ethylmagnesiation of inactivated alkenes by MgEt₂ in the presence of the rigid ansa-complex R,p-R,p-R-16 proceeded with the formation of products 168a–c with low yield and enantioselectivity (Scheme 43) [122,123].

Optically active titanocene complexes p-S,p-S-4b [44] and p-S,p-S-11 [48] were evaluated as catalysts for enantioselective reactions aimed at producing homoallylic alcohols [124,125]. The utilization of various metal complexes in the alkylation of ketones and aldehydes allows for the modulation of stereoselectivity in this process. For example, the reaction of crotyl titanocenes with different substituents in the cyclopentadienyl ring (R = H, Me, i-Pr) with primary, secondary, and tertiary aldehydes was described (Scheme 44) [125]. The products of the reaction, homoallylic alcohols 169, were formed with good to excellent selectivity, consistently favoring the anti-diastereomer. The degree of diastereoselectivity correlates with the size of the substituent in the cyclopentadienyl ring of the Ti complex and in the starting aldehyde. Specifically, for aliphatic and aromatic aldehydes, an increase in diastereoselectivity was observed with greater steric hindrance at the metal center.

The reaction of allyl-(171) and crotyltitanocenes (172), derived from ansa-titanocene p-S,p-S-11, with aldehydes afforded homoallylic alcohols 170a–h with excellent yields exceeding 80% and a diastereoselectivity of anti/syn = (2.5–40)/1 (Scheme 45). It was suggested that allyltitanocene additions proceed via six-membered chair- or boat-like transition states, which regulate the reaction stereoselectivity [48,59,125].

The substitution of the tetrahydroindenyl catalyst p-S,p-S-11 with ansa-ethylene-bis-(tert-butyl-cyclopentadienyl)titanium dichloride p-S,p-S-4b resulted in a change in absolute configuration from R- to S- in the reaction products—anti-homoallylic alcohols (Scheme 46) [44]. The diastereoselectivity of the process increased with the size of the substituent in the aldehyde (anti:syn = >99:1). In the reaction with pivaldehyde, the predominance of the anti-R enantiomer 169d was observed. Furthermore, the syn-R-stereoisomer of homoallylic alcohol 169a was obtained with greater enantioselectivity than the anti-isomer.

Titanocenes 52, 58, and 59 reacted with 2-alkyl-1,3-butadienes 174a,b and i-PrMgCl, yielding allylic complexes 175, which were then subjected to interaction with excess CO₂ or R’CHO, resulting in the formation of S-enantiomers of carboxylic acids 176a,b or S-threo/erythro-homoallylic alcohols 177, respectively (Scheme 47) [126]. The highest enantiomeric excess of >91%ee was observed in the reaction, catalyzed by complex 52, with a predominance of S-threo isomers of the alcohols. In this case, the yield and enantiomeric excess of erythro- isomers of the alcohols were low—12–44%ee.

Enantiomerically pure titanocene p-S,p-S-11, when employed as a reagent in the Barbier-type carbonyl compound propargylation or allylation, yielded homo-propargylic (179) or allylic (180) alcohols, respectively, demonstrating the fundamental possibility of conducting this reaction in an asymmetric manner (Scheme 48) [127,128,129].

3.2. Carbo- and Cycloalumination of Unsaturated Compounds

The asymmetric Zr-catalyzed enantioselective carboalumination of alkenes was first demonstrated by Negishi in 1995 [130,131,132], exemplified by the asymmetric carboalumination of alkenes using AlR₃ (R=Me, Et) in the presence of the conformationally labile complex p-S,p-S-bis-neomenthylindenylzirconium dichloride (p-S,p-S-186), achieving enantioselectivities of up to 85%ee for AlMe₃ and 92%ee for AlEt₃, respectively.

Studies on the enantioselectivity of ansa-complexes p-S,p-S-13, p-S,p-S-16, p-S,p-S-18, p-S,p-S-19 [51], p-R,p-R-13, p-R,p-R-16 [130], p-S,p-S-151, p-S,p-S-187a–c, p-S,p-S-188 [132], and p-R,p-S-189 [133] indicated the dependence of the stereodifferentiating effect on the structure of the substrate and the catalyst (

Table 2. Carbo-(182) and cycloalumination (184) of alkenes, catalyzed with enantiomerically pure metallocenes.

Catalyst	AlR3	Product	R′	ee%	Ref.
p-S,p-S-187	AlMe3	182	n-C6H13, i-Bu, Cy, Bn, Ph, C4H8OH, C3H6NEt2	65–85, R	[130]
		182	n-C4H9, n-C6H13	48–73, R	[134]
	AlEt3	182	R′ = n-C6H13	92, R	[135]
		182	n-C4H9, n-C5H11, n-C6H13, n-C7H15, n-C8H17, i-Bu, Cy, Cy8, Ph, Bn	33–68, S	[136]
		184	n-C6H13	33	[135]
		184	n-C4H9, Bn	5–11, S	[136]
			Cy, Cy8	45–57, S
			Ph	31–40, R
p-R,p-R-13	AlMe3	182	n-C8H17	6, R	[130]
R,p-R,p-R-16	AlMe3	182	n-C6H13	8, R	[130]
p-S,p-S-13	AlEt3	182	n-C6H13	51, R	[51]
		184	n-C6H13	11, S
S,p-S,p-S-16	AlEt3	182	n-C6H13	15, R	[51]
		184	n-C6H13	26, S
p-S,p-S-18	AlMe3	182	n-C6H13	58, R	[51]
	AlEt3	182	n-C6H13	50, R
		184	n-C6H13	12, S
S,p-S,p-S-19	AlEt3	182	n-C6H13	15, R	[51]
		184	n-C6H13	20, S
p-S,p-S-149	AlMe3	182	Bn	29	[132]
p-S,p-S-188a	AlMe3	182	Bn	25	[132]
p-S,p-S-188b	AlMe3	182	Bn	29	[132]
p-S,p-S-188c	AlMe3	182	Bn	33	[132]
p-S,p-S-189	AlMe3	182	Ph	80	[132]
p-R,S-190	AlMe3	182	Ph,	19	[133]
			Bn	28

, Scheme 49).

It has been demonstrated [132] that the carboalumination of styrene and allylbenzene by AlMe₃ in the presence of 10–13 mol% of p-S,p-S-151 or p-S,p-S-187a–c proceeded with conversions of the starting alkenes in the range of 72–89% over 3–6 h. The highest enantioselectivity of 80%ee was achieved in the reaction with styrene, then the content of the p-S,p-S-188 complex, activated by [Ph₃C][B(C₆F₅)₄], was increased to 26 mol%. The methylalumination of allylbenzene, followed by oxidation of the reaction mixture, afforded 2-methyl-3-phenyl-1-propanol with an enantiomeric excess of 25–33%ee. The transition from an ethylene bridge to a dimethylsilylene bridge in the p-S,p-S-151 and p-S,p-S-187a complexes, along with the introduction of methyl groups at the 4,7- or 3- positions of the indene ligands in the p-S,p-S-187b and p-S,p-S-187c complexes, did not lead to significant changes in the enantioselectivity of the reaction.

To elucidate this phenomenon, the mechanism of alkene methylalumination in the presence of enantiomerically pure ansa-zirconocene complexes was proposed (Scheme 50) [132]. According to Scheme 50, the starting ansa-complex in the presence of MAO provides an activated state—the cationic complex S,S-1a.1. Subsequent coordination and insertion of the olefin yield a new alkylzirconium cation S,S-1a.3, whose association with an AlMe₃ molecule leads to the heterobimetallic complex S,S-1a.4. The target OAC is formed as a result of alkyl exchange in complex S,S-1a.4 and its dissociation. It is suggested that the higher enantioselectivity of the reaction is due to the coordination stage of the olefin to the activated complex in a “side-on” manner, while lower selectivity is associated with a “head-on” approach. It is noted that, apparently, in complexes p-S,p-S-187b and p-S,p-S-187c, the methyl substituents do not possess sufficient volume to obstruct the “head-on” direction. The increase in enantioselectivity of the methylalumination when utilizing styrene is attributed to the enhanced interaction between the ligand framework of the complex and the phenyl fragment of styrene.

The reaction of allylbenzene or styrene with AlEt₃ catalyzed with (+)-[η⁵:η¹-indene dimethylsilyl(α-methylbenzyl)amido]titanium dichloride (p-R,S-189) yielded the cycloalumination product 184 with moderate yields of 40–60% and an enantiomeric excess ranging from 19 to 28%ee [133]. Remarkably high enantioselectivities of 96–99%ee were observed in the reaction of 2,5-dihydrofurans with AlEt₃, catalyzed by the complexes p-R,p-R-13 or R,p-R,p-R-16 (Scheme 51) [137].

The methylalumination of terminal alkenes with AlMe₃ catalyzed by the binaphtholate complex bearing Si-bridged ligands p-S,p-S-18, activated by MAO, afforded the corresponding products with a yield of 66% and an enantiomeric excess of 65%ee (R) [51]. The dichloride complexes p-S,p-S-13 and p-S,p-S-18 exhibited the highest enantioselectivities in carboalumination, achieving 50–51%ee (R). The enantiomeric purity of cyclic organoaluminum compounds OACs (185) formed in reactions catalyzed by complexes S,p-S-13, p-S,p-S-18, S,p-S,p-S-16, and S,p-S,p-S-19 ranged between 12 and 26%ee (S). It is noteworthy that the fixation of the geometry of the zirconocene complex provided methyl- and ethylalumination products with identical configurations at the β-stereogenic center but with diminished stereoselectivity compared to the conformationally flexible complex p-S,p-S-bis-neomenthylindenylzirconium dichloride (p-S,p-S-186) [135,136].

3.3. Asymmetric Oligo- and Polymerization of Alkenes and Dienes

The potential of chiral metallocenes of group IVB metals as catalysts for the stereoselective synthesis of oligomers and polymers from terminal alkenes has been convincingly demonstrated [99,100]. For example, the polymerization of propylene utilizing a catalytic system based on p-R,p-R-17 and methylaluminoxane (MAO) afforded a mixture comprising hydrogenated isotactic polypropylenes 194 and liquid hydrogenated oligomers 195 (Scheme 52) [138,139].

Infrared and ¹³C NMR spectroscopic analyses of the oligomer fractions revealed the presence of terminal n-butyl, n-propyl, and isobutyl groups, with a predominance of n-propyl moieties. Notably, the polymers were devoid of double bonds, thus characterizing the process as “hydrooligomerization.” The hydrooligomers exhibited isotactic configurations, with the first asymmetric carbon atom in the growing oligomer chain possessing an S-configuration. Comparison of mass-spectrometry data, as well as ¹H and ¹³C NMR spectra of the obtained low-molecular-weight hydrooligomers (isolated by distillation of 4,6-dimethyldecane and 2,4,6,8-tetramethylnonane), with the spectra of rac-4,6-dimethyldecane and meso- and rac-2,4,6,8-tetramethylnonane, led to the conclusion that absolute configurations of the stereogenic centers at positions 4 and 6 differ. As a result, 4S,6R-4,6-dimethyldecane and 4R,6S-2,4,6-trimethylnonane were identified.

The polymerization of 1-pentene in the presence of R,p-R,p-R-16 or p-R,p-R-17 and MAO, followed by treatment with D₂, yielded optically active deuterated pentanes of general formula C₅H_12−xD_x and deuterated oligomers (Scheme 53) [140]. Fractional distillation and successive solvent extractions (methanol, acetone, ethyl acetate) facilitated the isolation of these oligomers. It was proposed that the elongation of the oligomer chain enhances the stereoselectivity of the process, as evidenced by the superior optical purity of the dimers relative to the deuteropentanes. This observation aligns with the hypothesis that stereoselectivity in α-olefin polymerization is governed both by the chirality of the catalytic center and the spatial orientation of the last monomeric unit of the growing chain towards the metal atom [138].

The enantiomerically pure acetylmandelate complex R,R,p-S,p-S-15, combined with MAO, was evaluated for propene and 1-butene oligomerization (Scheme 54) [141]. Under increased catalyst concentration and elevated monomer content, isotactic oligomers bearing terminal double bonds were obtained, indicative of an accelerated chain termination rate relative to polymer chain propagation.

Initially, methylaluminoxane mediates the substitution of the mandelate ligand with a methyl group in the complex, generating a Zr–Me bond that facilitates substrate incorporation (Scheme 54). Subsequent olefin coordination promotes chain elongation. β-Hydride transfer to the metal center terminates the chain, yielding a complex bearing a free Zr–H bond capable of coordinating another alkene to regenerate the metal alkyl species. Gas chromatographic analysis revealed the presence of oligomers with degrees of polymerization (n) ranging from 2 to 7. Under analogous conditions, propylene predominantly formed pentamers, whereas 1-butene yielded trimers and tetramers. The higher oligomers exhibited isomeric mixtures primarily composed of configurational isomers. For example, the tetramer fraction contained a diastereomeric mixture of (R,R)/(S,S)- and (R,S)/(S,R)-2,4,6-trimethyldecene in a 5:4 ratio, while the pentamer fraction comprised four pairs of enantiomers of 2,4,6,8-tetraethyl-1-dodecene. Gas chromatographic separation of the enantiomers of the 1-butene trimer mixture was achieved via modified cyclodextrin chromatography [142,143], with an enantiomeric purity of 27%ee. Consistent with the hydrooligomerization of propylene, the reaction favored the formation of R-enantiomers. The stereoselectivity of the second alkene insertion step was notably enhanced, as the stereogenic centers in the growing oligomer chain enhanced the stereodifferentiating effect (growing chain control).

The catalytic system p-S,p-S-13-MAO exhibited a high degree of isotacticity in the polymerization of propylene, 1-butene, 1-pentene, and 1-hexene [144]. The elongation of the alkyl substituent exerted minimal influence on isotacticity, which remained between 95 and 99%mmmm at temperatures below 20 °C. The optical rotation of higher oligomers diminished with increasing reaction temperature, whereas the enantiomeric excess of the oligomers increased from 10% to 90% as the temperature decreased from 60 °C to 0 °C.

Comparative studies of propene polymerization stereoselectivity in the presence of R,R,p-S,p-S-15 and a mixture of diastereomers R/S-15 revealed that the exchange of polymer chains between metal centers of opposing chirality significantly undermines stereochemical control [145]. In most experiments, the precursor was activated with an optimal quantity of AlBuⁱ₃ and stoichiometric amounts of [Ph₃C][B(C₆F₅)₄]. Both racemic and enantiopure catalysts displayed comparable activities; however, the racemic catalyst exhibited diminished stereo- and regioselectivity. The stereoselectivity of R/S-15 improved markedly upon lowering its concentration, reducing the reaction temperature, or immobilization on silica supports. It was concluded that the racemate forms two catalytically active centers with different configurations capable of exchanging polymer alkyl chains, thereby reducing stereoselectivity and increasing regioerrors.

Furthermore, the catalytic activity and stereoselectivity of chiral homogeneous and heterogeneous catalytic systems were compared [146]. The optically active heterogeneous catalyst, prepared by depositing TiCl₃ on MgCl₂ in the presence of di[(S)-2-methylbutyl]phthalate as an electron donor, exhibited moderate activity in racemic 4-methyl-1-hexene (4MH) polymerization (2.6·10³ g polymer/(mol Ti·mol 4MH·h)). The (4S)-4-methylhexene enantiomer was incorporated more favorably than the (R)-enantiomer, with a polymerization rate ratio R(S)/R(R) of 1.02 and a polymer rotation angle [α]_D²⁵ of +4.8°. Consequently, the residual monomer was enriched in the R-enantiomer. In contrast, the homogeneous catalytic system R,R,p-S,p-S-15-MAO exhibited a higher rate ratio of 1.4 and a polymer rotation angle of +17 to +19°, with an activity of 112·10³ g polymer/(mol Zr·mol 4MH·h). The stereoregularity index (SI) was lower in heterogeneous catalysis (maximum 85%) than in the homogeneous system (maximum 97%). Fractionation of polymer in various solvents (acetone, ethyl acetate, diethyl ether, and cyclohexane) yielded fractions with distinct rotation angles. For instance, fractions obtained in the system R,R,p-S,p-S-15-MAO displayed [α]_D²⁵ values of –7.8° (acetone), +22° (ethyl acetate), and +31° (cyclohexane), whereas the TiCl₃-MgCl₂-Et₃Al-di[(S)-2-methylbutyl]phthalate system yielded positive rotations of +5.4°, +10°, +15°, and +9.9° in the respective solvents. This disparity, including the inversion of rotation sign, was attributed to stereoerrors, i.e., the incorporation of both S- and R-enantiomers under specific conformations of the catalytically active centers in homogeneous systems—a phenomenon more prevalent at elevated temperatures due to increased conformational flexibility. Conversely, heterogeneous catalysis features fixed geometries of active centers, albeit with greater diversity in Lewis acidity and steric environments.

Cyclopolymerization of 1,5-hexadiene catalyzed by achiral precatalysts Cp₂MX₂ (M = Ti, Zr; X = Cl, Me) and MAO at ambient temperature afforded poly(methylene-1,3-cyclopentane) (PMCP) 196, which can exhibit four possible microstructural variants (Scheme 55) [147]. Among these, only the trans-isotactic microstructure is chiral. According to ¹³C NMR data, approximately 80% of PMCP rings adopted the trans-configuration. Lowering the polymerization temperature enhanced trans-selectivity, reaching 90% trans rings at –78 °C. In contrast, sterically hindered complexes (CpMe₅)₂MX₂ (Cp* = pentamethylcyclopentadienyl) predominantly yielded cis-PMCP, with 86% cis content at –25 °C. Cyclooligomerization of 1,5-hexadiene catalyzed by chiral catalysts rac-13 and rac-151 afforded PMCP with 70% trans content (Scheme 55). Here, the cis/trans stereochemistry was less sensitive to polymerization temperature compared to achiral metallocenes. Enantiomerically pure precatalysts R,p-R,p-R-16 and S,p-S,p-S-16 produced optically active trans-isotactic PMCP 196 with molar optical rotations [Φ]₄₀₅²⁸ = +51.0° and [Φ]₄₀₅²⁸ = −51.2°, respectively.

Enantiomerically enriched styrene hydrooligomers were synthesized via catalysis by complex p-R,p-R-16, activated with methylaluminoxane under a hydrogen atmosphere [148]. The oligomers included R-1,3-diphenylbutane (197) with an enantiomeric excess of 77%ee (Scheme 56). Subsequent functionalization without loss of chirality underscores the potential of this catalytic system for preparing optically active synthons for organic synthesis.

Polymerization of racemic α-olefins bearing bulky substituents 199a–f catalyzed by enantiomerically pure S-116a–d in the presence of MAO achieved kinetic resolution, yielding the R-enantiomers of 199a–f with enantiomeric excesses of 7–59%ee and chiral, highly isotactic polymers 200a–f enriched in the S-enantiomer (Scheme 57,

Table 3. The polymerization of racemic α-olefins with bulky substituents 199a–f, catalyzed with enantiomerically pure S-116a–d in the presence of MAO.

Entry	Alkene-1	Complex	Time, h	TOF, h−1	Conversion, %	s	%ee
1	199a	S-116a	18	72	24	2.8	13.3
2		S-116a	47	47	38.3	2.4	20.3
3		S-116a		60		2.6
4		S-116b		266		2.3
5		S-116c		374		2.5
6		S-116d		280		3.2
7	199b	S-116a	13.5	551	75	1.8	40
8		S-116b		299		1.8
9		S-116c		706		1.9
10	199c	S-116a	40.5	45.6	32.5	17.6	40.6
11		S-116a	69	34	42.4	15.9	58.6
12		S-116a		40		16.8
13		S-116b		109		12.5
14		S-116c		450		15
15	199e	S-116a	16.5	77.6	64.4	1.1	7.6
16		S-116a	16.5	73.7	58.7	1.1	4.6
17		S-116a		75		1.1
18		S-116b		111		1.2
19		S-116c		633		1.2
20	199f	S-116a		18		7.6
21		S-116d		16		3.2
22	199d	S-116a	22.7	55.8	37.8	2.0	16.2
23		S-116a	43	37	56.1	2.1	30.3
24		S-116a		37		2.1
25		S-116b		49		6.4
26		S-116c		111		1.6
27		S-116d		988		8.5

) [81,86,149,150,151]. The selectivity factor s (rate of the fast-reacting enantiomer)/(rate of the slow-reacting enantiomer) for most olefins ranged from approximately 1.1 to 3.2, indicating partial kinetic resolution; however, polymerization of 3,4-dimethyl-1-pentene (199c) exhibited pronounced resolution (s > 15) [150,151]. To rationalize isotactic polymer formation, the authors proposed epimerization of the catalytically active center, whereby the polymer chain alternates sides relative to the metallocene cation after each monomer insertion. A high rate of site epimerization relative to olefin coordination ensures that the olefin encounters a consistent chiral pocket, accounting for the high stereoselectivity. The stereogenic centers in the growing polymer chain may further influence the stereoselectivity of the process.

In the presence of catalysts 116a and 116d, the stereoselectivity of propene polymerization was markedly influenced by substrate concentration ([mmmm] = 2.8–62.9%, [rrrr] = 6.8–49.6%) (Scheme 58) [150]. Elevated isotacticity was attained at diminished monomer concentrations ([mmmm] = 62%), whereas conducting the reaction in neat propylene yielded a syndiotactic polymer ([rrrr] = 44–49%). At intermediate concentrations, catalyst 116d favored a higher proportion of isotactic polymer relative to 116a. Moreover, 116d exhibited enhanced selectivity during the kinetic resolution of 3-substituted racemic α-olefins.

To rationalize these observations, it is noted that the polymerization of racemic α-olefins likely proceeds via the involvement of a pair of diastereomeric pairs at each catalytic site (even more if chain end control is taken into account), instead of involving just a pair of diastereomers as observed in simple α-olefin polymerizations [150]. Consequently, the selectivity in racemic polymerization is governed by matching and mismatching pairs in a double diastereoselective manner. The proposed enantiomorphic site control model (Scheme 58) incorporates a double diastereoselective transition state—for instance, the (S)-metallocene/si-olefin face/(S)-olefin diastereomer is favored over the (S)/si/(R) diastereomer. Nonetheless, the scenario is more complex, especially when considering potential insertions from both faces of the zirconocene, chain end control, and/or poor enantiofacial selectivity. Additionally, the rate of anion reorganization during racemic α-olefin polymerization may also contribute significantly.

Subsequently, a one-pot method for the synthesis of the deoxypropionate scaffold was developed based on the stereocontrolled oligomerization of propylene (Scheme 59) [152]. It was demonstrated that the reaction of propylene with ZnEt₂ in the presence of optically active zirconocene p-S,p-S-13 and MAO afforded oligomers. Under identical conditions, the employment of AlMe₃ failed to yield the target products. Following oxidation and hydrolysis, branched primary alcohols 201 were obtained with high enantiomeric purity (99%ee), albeit with modest yields due to the formation of polymerization side products. The C_S-symmetric zirconocene 202 catalyzed syndiospecific oligomerization under analogous conditions, delivering rel-(2R,4S,6R,8S)-2,4,6,8-tetramethylundecanol-1 [153], a constituent of Antitrogus parvulus (Scheme 59).

Oligomerization of 1-hexene catalyzed by enantiomerically pure Zr ansa-complexes p-S,p-S-13, p-S,p-S-18, S,p-S,p-S-16, and S,p-S,p-S-19 in the presence of AlR₃ (R = Me, Et) and MAO yielded optically active, functionally substituted oligomers 204a,b with degrees of oligomerization up to six and yields ranging from 10 to 54% (Scheme 60) [51].

In developing this methodology, stereoselective oligomerization of 1-hexene was achieved by employing p-R,p-R-13, AlMe₃, and activators such as MMAO-12 or [Ph₃C][B(C₆F₅)₄]. This approach yielded organometallic intermediates 203a bearing an initial methyl group, which upon oxidation and hydrolysis furnished alcohols 204a (Scheme 60) [154]. In the presence of MMAO-12, the oligomer fraction (204a, n = 2–6) constituted 63–81%, whereas the use of [Ph₃C][B(C₆F₅)₄] as an activator enhanced oligomer yields to 94–98%. The functionally substituted oligomers 204a exhibited enantiomeric enrichment and diastereomeric purity (99%dr). An increase in oligomer chain length correlated with augmented enantioselectivity, irrespective of activator type. Nevertheless, the nature of the activator significantly influenced enantioselectivity. Specifically, reactions with AlMe₃ and MMAO-12 proceeded with moderate enantioselectivity—60–64%ee for oligomers containing two monomeric units, and 78–84%ee for oligomers with n = 3,4—reaching 93%ee for n = 5, 6. Substitution of MMAO-12 with [Ph₃C][B(C₆F₅)₄] diminished enantioselectivity to an average of 10–20%ee. Therefore, stereoselectivity in these catalytic systems is governed by both the metal center and the growing polymer chain, which is consistent with the Corradini model [155]. The experimental data further underscore the pivotal role of counterion effects during alkene insertion into catalytically active centers of the type [L₂Zr⁺Alk…SLA] (SLA—strong Lewis acid).

Despite substantial progress in stereoselective olefin polymerization, enantioselective oligomerization and polymerization of alkenes remain comparatively underexplored. Significant advancements have been realized employing enantiomerically pure ansa-zirconocenes. Mechanistic descriptions frequently invoke multiple conformations of catalytically active centers during both alkene coordination and chain propagation, implying that reaction stereoselectivity is largely dictated by the ligand environment surrounding the transition metal and the growing polymer chain. Recently, an increasing imperative has emerged to scrutinize the influence of activator structure on the organization of catalytically active centers, and consequently, on the stereochemical outcome of these transformations.

3.4. The Diels–Alder Cycloaddition

Enantiomerically pure zirconium bis-alkoxy or triflate ansa-complexes serve as catalysts for Diels–Alder cycloadditions between cyclopentadiene and various dienophiles, exemplified by 205a–g (Scheme 61) [58]. Reactions involving dienophiles bearing ethenyl substituents (205a,c) preferentially yield the endo-product. The endo/exo product ratio depends on the substituent’s nature at the double bond. In all cases, enantioselectivity did not exceed 50%ee. The authors attribute the low enantioselectivity of the reactions to both the steric factors of the substrate (endo- and exoselectivity) and the steric environment of the transition metal center.

The cycloaddition of enones 205e–g with cyclopentadiene catalyzed by chiral triflate complexes Ti (p-S,p-S-28) and Zr (p-S,p-S-30) predominantly afforded R-endo isomers 206e–g, with enantiomeric excess heavily influenced by solvent polarity (Scheme 61) [60]. In dichloromethane, enantioselectivity was below 69%ee (R), whereas in nitromethane or 2-nitropropane, stereoselectivity reached 90–95%ee (R, endo/exo = (6–15)/1) [59]. The degree of asymmetric induction in this reaction was ascribed to the conformations of the intermediates—coordination products of the ansa-complex and the starting oxazolidinone substrate 205e–g (Scheme 61) [60]. Spectroscopic (EXSY, NOESY) and kinetic analyses revealed the formation of two isomeric complexes upon substrate–catalyst binding. The relative reactivities of these intermediates determine the diastereo- and enantioselectivity of the reaction. Thus, solvent-dependent stereoselectivity likely arises from the differential stability of complexes 207A and 207B. It is plausible that the minor isomer 207B reacts with cyclopentadiene more rapidly and selectively, favoring the formation of endo enantiomers 206e–g.

3.5. Reactions of Imine ansa-Zirconocenes—Zirconaaziridines

Metal–carbon bonds exhibit pronounced reactivity toward organic substrates and serve as pivotal intermediates in many catalytic processes. Since 1991, asymmetric catalysis employing imine complexes of ansa-zirconocenes—termed zirconaaziridines—has been reported. The synthesis of zirconaaziridines is typically conducted in situ (Scheme 62). For example, the reaction of the dimethyl complex p-S,p-S-17 with trifluoromethanesulfonic acid and lithium amide provided zirconocene methyl amide 209. Subsequent methane elimination at 80 °C yielded zirconaaziridine 210 [50,156].

A diverse array of heteroatom-containing compounds has been synthesized via reactions of zirconaaziridines with unsaturated substrates. Notably, asymmetric induction in the reaction of amines with alkynes, alkenes, or aldehydes in the presence of zirconaaziridines 210 has been demonstrated, yielding allylic amines 212a–i and phenylalkyl amines 212j,k with respectable yields (3872%) and high enantioselectivity (Scheme 63) [50].

Reactions of complexes 214 with various alkenes furnished metallocycles 215a–d in yields ranging from 67% to 94% (Scheme 64) [157]. Diastereoselectivity in reactions with aliphatic alkenes was enhanced using the bis-indenyl catalyst rac-213, achieving up to 83%de, whereas the tetrahydroindenyl analog rac-17 exhibited diminished stereoselectivity (66%de). Employing styrene or vinyltrimethylsilane as substrates altered regioselectivity and augmented diastereoselectivity to 98%de, independent of catalyst structure. Reaction of 1-alkenes with excess picoline in the presence of optically active dimethylzirconocene p-S,p-S-17 and [HNBu₃][BPh₄] afforded substituted pyridine 216 with 58%ee (R) (Scheme 64).

Reactions of alkyl- or aryl-substituted zirconaziridine 210 with ethylene carbonate or isocyanates produced amino acid esters 218a–h with enantiomeric purities up to 98%ee, and phenylglycinamides 220a–d with 80-99%ee, respectively (Scheme 65) [156]. In these transformations, the ansa-complex functions both as catalyst and substrate, necessitating stoichiometric amounts of the initial complex.

In discussing the plausible mechanism of stereoinduction, a hypothesis was proposed correlating enantioselectivity with the equilibrium between epimers of zirconaziridine 210. This ratio can be modulated by varying reagent quantities (ethylene carbonate or isocyanate), reaction parameters (temperature, solvent), or substrate substituents (Scheme 66).

Furthermore, it was demonstrated that the reaction of enantiomerically enriched imido-complex p-S,p-S-222 with allenes proceeds via kinetic resolution, yielding a single R-enantiomer of allene 225 with 50–61% yield and 7898%ee (Scheme 67) [158]. The more reactive allene enantiomer was subsequently regenerated from the metallacycle by treatment with the parent allene, 1,2-propadiene. Thus, the kinetically favored S-allene enantiomer was recovered from metallacycle R,p-S,p-S-223 via reaction with 1,2-propadiene.

3.6. Cyclization Reactions

Complexes of Group IVB metals, similar to cobalt complexes, facilitate the Pauson–Khand cyclo-condensation of enynes with carbon monoxide, yielding cyclopentenones. The first examples of this reaction being carried out in an asymmetric manner are presented in Refs. [159,160] (Scheme 68). It was found that p-S,p-S-226 serves as a selective catalyst for transforming various 1,6-enynes 227a–h in the presence of CO into the corresponding bicyclic cyclopentenones 228a–h, achieving yields of 70–94% and enantioselectivity of 72–96%ee.

Subsequently, a method for the asymmetric cyclocarbonylation of nitrogen-containing enynes was developed (Scheme 69) [161]. Studies of the effects of substituents at the nitrogen atom and catalyst concentration on the enantioselectivity of this type of cyclization revealed that enynes 229a–j undergo cyclization in the presence of CO and catalytic amounts of p-S,p-S-226, producing azabicyclo[3.3.0]oct-5-en-7-ones 230a–j with yields exceeding 62% and enantioselectivity up to 95%ee. It was demonstrated that reactions involving enynes with N-octyl, N-allyl, or N-benzyl groups exhibited high enantioselectivity. In contrast, the presence of an anilino, amide, or sulfonamide moiety in the starting molecules resulted in products with lower ee values. When nitrogen was part of a carbamate group, moderate enantioselectivity was observed. It is suggested that substrates with a relatively electron-rich nitrogen center cyclize to enones with higher enantioselectivity compared to those with an electron-deficient nitrogen atom.

The intramolecular asymmetric reductive coupling of ketones with nitriles, catalyzed by the optically active titanocene p-R,p-R-11, was developed. This method yields substituted α-hydroxyketones (acyloins)—structural fragments of numerous natural compounds. Cyclic hydroxy-substituted cyclopentanones 232a–o were synthesized with commendable yields ranging from 47% to 94%, with enantioselectivity exceeding 65%ee (Scheme 70) [162].

3.7. Asymmetric Hydrosilylation of Ketones

The asymmetric hydrosilylation [163] of ketones 233a–v using polymeric methylhydrosiloxane Me₃SiO[MeSi(H)O]_nSiMe₃ in the presence of chiral titanocene binaphtholates R,p-R,p-R-12 proceeded with high enantioselectivity (82–96%ee) (Scheme 71) [164]. Among these, ketones lacking a double bond at the α-position (234s,u) exhibited poor enantioselectivity. Furthermore, it was demonstrated that employing substrates with bulky alkyl groups in the alkylaryl ketone structure enhanced enantiomeric excess. Generally, conducting the reaction with branched substrates necessitated an additional catalyst amount (up to 10 mol%) and extended reaction times.

Hydrosilylation of ketones was performed using silanes (EtO)₃SiH, Me(EtO)₂SiH, [MeSi(H)O]₄, Me₃SiO[MeSi(H)O]_nSiMe₃, and MeSiH₃, alongside catalysts prepared by alkylation of S,p-S,p-S-12 with MeLi or n-BuLi (Scheme 72) [165]. The enantiomeric purity of the acetophenone hydrosilylation product 234a, obtained in the presence of the S,p-S,p-S-12-n-BuLi system, was significantly higher (approximately 99%ee) than that achieved with the MeLi-alkylated catalyst (approximately 40–50%ee). The hydride complex [(EBTHI)TiH]₂, synthesized in the reaction of S,p-S,p-S-12 with MeLi and PhMeSiH₂, exhibited essentially the same enantioselectivity as the MeLi-based catalyst. Since the titanocene hydride demonstrated low enantioselectivity in the acetophenone hydrosilylation, the authors proposed two potential reaction mechanisms involving TiIII and TiIV species (Scheme 72). In these mechanisms, key intermediates were suggested to be species with Ti-Si bonds, whose coordination with the substrates determines both the reaction rate and stereoselectivity. The observed differences in enantiomeric excess with MeLi and n-BuLi were attributed to the rapid decomposition of L₂TiBu₂ alongside Ti reduction, in contrast to the L₂TiMe₂ complexes, as well as the direct involvement of Ti alkyl complexes as intermediates.

Optically active titanocene difluoride p-R,p-R-235 was found to be an effective catalyst for the hydrosilylation of ketones by PhSiH₃ in the presence of alcohol additives (Scheme 73) [166,167]. It has been shown that primary alcohols were more effective as additives than secondary alcohols or primary amines. For example, the gradual addition of methanol to the reaction mixture resulted in an increased conversion of the starting ketone (99%) and the enantiomeric purity of the synthesized secondary alcohols (99%ee) [166].

In the proposed mechanism illustrated in Scheme 74, the stereodifferentiating and rate-limiting steps of the reaction are distinct. The smaller the size of the alkyl group in the alcohol, the more rapid the σ-bond metathesis occurs in the rate-limiting step. Consequently, the added alcohol does not participate in the stereodifferentiating step; rather, it accelerates the reaction, thereby diminishing the probability of non-selective pathways during the hydrosilylation of ketones.

The catalytic activity of η⁵-titanium complexes S-88a, R-88a, p-S,p-S-38, and 1R,2R,4R,5R-53 has been studied in the reduction reactions of aryl-substituted ketones using phenylsilane (Scheme 75) [74,168,169]. These catalysts exhibited varying activity (0–100% substrate conversion) and relatively low enantioselectivity (with a maximum of 82%ee in the case of acetophenone). Consequently, their practical significance is considerably lower compared to, for example, complexes with ethylene bridges [164,166]. In this case, the steric bulk of the complexes increases the enantioselectivity of the process; however, it significantly reduces their catalytic activity.

3.8. Asymmetric Hydrogenation of N-Substituted Compounds

With the growing practical significance of enantiomerically pure nitrogen-containing compounds in the pharmaceutical and agrochemical industries, there is a pressing need to develop synthetic methods that exhibit high chemo-, regio-, and stereoselectivity. Asymmetric catalytic reduction of imines stands out as an effective method for the synthesis of chiral amines. Enantiomerically pure titanocene complexes R,p-R,p-R-12 and R,R,R-22 showed high activity and enantioselectivity in imine hydrogenation (Scheme 76) [53,170,171,172,173]. The catalytically active center in this reaction is presumably a titanocene (III) hydride, generated in situ through the interaction of titanocene complexes with an excess of triphenylsilane. Subsequently, phenylsilane was replaced with gaseous hydrogen, which proved to be a more effective stoichiometric reductant. The hydrogenation of imines 236a–v, catalyzed by R,p-R,p-R-12, was carried out at hydrogen pressures of 80–2000 psi for durations ranging from 8 to 48 h [170,171,172,173] (Scheme 76). As a result, both cyclic and acyclic amines 237a–v were formed with varying enantiomeric purity. High enantioselectivity (97–99%ee) was achieved with cyclic substrates, particularly ketimines, whereas acyclic compounds were reduced with moderate enantioselectivity (59–93%ee). The observed decrease in enantiomeric purity of the products is likely attributable to the fact that acyclic imines exist as mixtures of anti- and syn-isomers, which can interconvert during the reaction. Utilizing the same approach, 1-phenylpyrroline (236w) and N-(R-methylbenzylidene)benzylamine (236x) were reduced in the presence of the titanium complex R,R,R-22 [53]. Under conditions of 150 bar H₂ and 80 °C for 12 h, the corresponding secondary amines 237w and 237x were obtained with yields of 96% and enantioselectivities of 76–98%ee.

The binaphtholate Ti complex S,p-S,p-S-12 can be effectively employed as a catalyst for the hydrogenation of 1,1-disubstituted enamines [174]. This reaction yielded enantiomerically enriched tertiary amines 239a–i (>91%ee) with good yields (73–92%) (Scheme 77). Sterically hindered substrates reacted under elevated pressures (80 psi) and temperatures. Bulkier or bromine-containing substrates did not participate in the reaction.

The catalytic system based on titanocene difluoride p-S,p-S-235 demonstrated high stereoselectivity in the hydrosilylation of imines, owing to the rapid conversion of the Ti-F bond to Ti-H under the influence of PhSiH₃ or polymethylhydrosiloxane (PMHS) in a MeOH/pyrrolidine medium (Scheme 78) [175,176,177,178,179]. Consequently, the asymmetric hydrosilylation of N-aryl-substituted imines 240a–ad using PhSiH₃ or PMHS was developed, providing chiral amines 241a–ad with yields of 55–97% and substantial enantiomeric excesses.

Hydrosilylation of rac-2,5-disubstituted pyrroles 242a–f in the presence of the binaphtholate complex S,p-S,p-S-12 was accompanied by the kinetic resolution of the substrates, resulting in cis-pyrrolidines 243a–h with yields of 34–44% and enantiomeric purity exceeding 95%ee (Scheme 79) [180]. The unreacted pyrrole was isolated with yields of 33–43%, also exhibiting a high enantiomeric excess of >95%ee. The reaction catalyzed by the difluoride complex p-R,p-R-235 [181] resulted in a highly enantioselective (>94%ee) kinetic resolution of the starting substrates 242a,d,g, yielding pyrrolidines 243a,d,g, along with isomers of the starting pyrrole iso-242a,d,g.

Optically active indanones or tetralones are also of significant interest as synthons for the synthesis of biologically active compounds. The kinetic resolution of N-alkyl (R = Me, i-Pr, n-Pr) imines derived from 3-substituted indanones 244a–f and 4-substituted tetralones 244g–j can be achieved through hydrosilylation in the presence of catalytic amounts of titanocene difluoride p-R,p-R-235 or p-S,p-S-235 (Scheme 80) [182]. Following the hydrolysis of the reaction mixture, ketones 245a–j were obtained with enantiomeric purity of 68–99%ee, as well as amines 246a–j exhibiting high diastereomeric (cis > 82%) and enantiomeric purity (47–97%ee). This method has been successfully applied in the diastereo- and enantioselective synthesis of the antidepressant sertraline 1S,4S-246j.

An example of hydrosilylation of alkenes was demonstrated using phenylbutene 247a, which, in the reaction with PhSiH₃ in the presence of samarium complexes R- or S-137a, provided silane 248 with yields up to 98% and enantioselectivity of 65–68%ee (Scheme 81) [183].

3.9. Enantioselective Hydrogenation of Olefins

Asymmetric hydrogenation of olefins has been the subject of extensive research [184,185,186,187]. Enantiomerically pure ansa-complexes have also been evaluated as catalysts for this process. Pino et al. [188] (Scheme 82) demonstrated that the hydrogenation of 2-phenylbut-1-ene (247a), catalyzed by zirconocene binaphtholate R,p-R,p-R-16 and MAO, yielded 2-phenylbutane 249a with a good yield (over 89%) but low enantioselectivity (36%ee, R), while rac-13 exhibited no catalytic activity. The hydrogenation of 247a in the presence of enantiomerically pure zirconocene p-S,p-S-17 and [PhMe₂NH][Co(C₂B₉H₁₁)₂] produced phenylbutane with an enantioselectivity of 23%ee, S (Scheme 82) [189]. It is suggested that the low enantioselectivity in these systems arises from the minimal energy differences between states with different enantiofaces of the olefin at the catalytic site.

Upon activation with n-BuLi, diastereomers of complexes 88a,b catalyzed the hydrogenation of 2-phenylbut-1-ene, with a moderate enantioselectivity of 35–60%ee [168].

Enantiomerically pure C₁-symmetric Sm complexes S- and R-137a,b exhibited higher enantioselectivity (up to 96%ee) in this reaction; the enantioselectivity was regulated by varying the reaction conditions and the ratio of enantiomers of the catalyst (Scheme 82) [83,183].

The polymerization of pentene-1 in the presence of D₂ and R,p-R,p-R-16 or p-R,p-R-17 and MAO resulted in the formation of deuterated products with the general formula C₅H_12−xD_x (Scheme 53) [140,188]. Through fractional distillation, 1,2-dideuteriopentane (249b) was isolated with an enantiomeric excess of 23–35%ee, and the R-configuration of the product was determined by optical rotation. As a result of styrene deuteration, the R-enantiomer of 249b was obtained with enantiomeric excesses of 65% and 43%ee when using R,p-R,p-R-16 [188] and R-137b [83], respectively (Scheme 82). The mixture of S-137b/R-137b (70/30) allowed for the attainment of a relatively high enantiomeric excess of the S-product at the level of 72%ee.

Trisubstituted olefins 250a–h were hydrogenated under the 2000 psi of H₂ in the presence of Ti complex S,p-S,p-S-12, activated with n-BuLi at 1 atm of H₂ and stabilized by phenylsilane. The reaction afforded products 251a–h with high yields of 70–94% and enantioselectivity up to 99%ee (Scheme 83) [190].

The hydrogenations of tetrasubstituted olefins were conducted at ambient temperature in aromatic hydrocarbons utilizing either p-R,p-R- or p-S,p-S- dimethylzirconocene complex 17 and activator [PhMe₂NH]⁺[B(C₆F₅)₄]⁻, applying hydrogen pressures of 80 or 1000–2000 psig (Scheme 84) [191]. Products 253a–h exhibiting high diastereomeric (>95%, cis) and enantiomeric purity exceeding 90% were achievable at elevated pressures and catalyst concentrations. The reaction mechanism suggests the formation of highly reactive cationic species bearing a Zr-H bond, which engage in reactions with re face of alkene.

3.10. Hydroamination and Cyclization of Amino- and Phosphinoalkenes

The binaphtholate complex S,p-S,p-S-16 was employed in the synthesis of enantiomerically enriched N-heterocycles 255 and 256 (Scheme 85) [192]. In this reaction, the substituent at the double bond of the alkenyl group of the amine significantly influenced the course of the reaction. For example, the reaction of Bu₂Mg with diallylbenzylamine in the presence of R,p-R,p-R-16 produced heterocycle 255a with low yield and enantiomeric excess (up to 13%). The reaction of substituted alkenes 254b–j with BuMgCl, catalyzed by S,p-S,p-S-16, afforded five- and six-membered cyclic products 255a–j and 256c,d with yields of up to 84% (combined cis- and trans-) and enantioselectivity of up to 95%ee.

The successful application of enantiomerically enriched η⁵-complexes of lanthanide metals has been demonstrated in the reaction of asymmetric hydroamination/cyclization of amino- and phosphinoalkenes 257a–e [85,193,194,195,196] (Scheme 86). With a substrate-to-catalyst ratio of (40–200)/1, nearly 100% conversion of N- or P-substituted alkenes was achieved with over 95% regioselectivity. Thus, C₁-symmetric complexes of Sm and La with menthyl- or neomenthyl-substituted ansa-cyclopentadienyl ligands catalyzed the hydroamination/cyclization of aminoalkenes 257a–c, providing the corresponding pyrrolidines and piperidines 258a–c with substrate conversion of up to 100% and enantiomeric purity < 74%ee [193,194,195]. The high diastereoselectivity of the reaction was demonstrated using 2-aminohexene 257d, which was transformed into trans-2,5-dimethylpyrrolidine 258d with >95%de in the presence of complexes 133a and 137a.

Furthermore, complexes with octahydrofluorenyl ligand S-141a–c (Sm, Y, Lu) exhibited activity and stereoselectivity in the hydroamination/cyclization of aminoalkenes 257a–c,e (Scheme 87) [85]. However, these complexes proved to be less active and stereoselective compared to cyclopentadienyl complexes 133 and 137. Enantioselectivity at the level of 67%ee was achieved only in the case of the sterically hindered substrate 257c (Scheme 87). It was noted that the decrease in turnover frequency (N_t) and selectivity for aminoalkenes 257a,b correlated with the decrease in atomic radii of the lanthanide centers Sm³⁺—Y³⁺—Lu³⁺. The hydrophosphination/cyclization of phosphinoalkenes 259a,b catalyzed by S-141a–c under the same conditions produced phospholanes 260a,b with high diastereoselectivity of up to 96%de (Scheme 87).

The cyclization of aminodienes 261a–c in the presence of complexes S-141a, S-133b, or S-134b yielded enantiomerically enriched substituted pyrrolidines and piperidines 262a–c (Scheme 88) [196,197,198,199]. The natural alkaloid (+)-coniine·HCl (264) was obtained from the prochiral aminodiene 261b via a two-step sequence in high isolated yield (S-141a was used as a catalyst). Catalytic hydroamination/cyclization of aminoalkenes 261d–f with an internal double bond was also investigated (Scheme 88) [197,198]. The highest enantioselectivity was observed in the reaction catalyzed by Y complex S-141b.

3.11. Other Applications of Enantiomerically Enriched ansa-Complexes in Asymmetric Catalysis

Epoxides are valuable starting materials in organic synthesis due to their easy availability and high reactivity of the three-membered ring. Asymmetric catalytic epoxidation of alkyl- and aryl-substituted alkenes 265a–g using t-BuOOH in toluene at 40–80 °C was conducted in the presence of ansa complexes 38 and R,p-R,p-R-41 (Scheme 89). The enantiomeric purity of epoxides 266a–g did not exceed 22%ee, with substrate conversion of 6–25% [62,200].

Enantiomerically pure ansa-titanocene dichloride p-S,p-S-11 catalyzed the ring-opening of epoxides 267a–c, producing the corresponding alcohols 268a–c with moderate enantioselectivity (56%ee), compared to conformationally labile neomenthylcyclopentadienyl titanium complexes 269a–c (more than 90%ee) (Scheme 90) [201,202,203].

Chiral bis-(indenyl)-(52) and bis-(tetrahydroindenyl)-(53) titanocenes were tested as catalysts in the homocoupling of benzaldehyde 270. As a result, pinacol-type compounds 271 were obtained (Scheme 91) [66]. It was found that the steric hindrance in tetrahydroindenyl complex 53 provided a lower yield but a higher enantiomeric excess of rac-1,2-diphenylethane-1,2-diol (32%ee).

The catalytic properties of the Ti complex p-R,p-R-11 were demonstrated in the synthesis of γ-lactol (+)-274. The product was isolated with a yield of 49% and enantiomeric purity of 33%ee (Scheme 92) [204].

The potential application of chiral titanium ansa-complexes 52 or 73 in enantioselective isomerization of trans- or cis-4-tert-butyl-1-vinylcyclohexanes 275 to 4-tert-butyl-1-(ethylidene)cyclohexanes 276 was studied (Scheme 93) [71,205]. Complex 52 showed higher activity and enantioselectivity compared to the catalyst 73. The reaction rate and the enantiomeric purity of the product were highly influenced by the temperature of the reaction. The decrease in enantiomeric purity observed at higher temperatures (180 °C) is believed to be caused by racemization through product equilibration, a phenomenon that is likely to be overcome with longer reaction times (12–120 h).

4. Conclusions

Chiral metal complexes of transition metals have occupied significant positions in organometallic chemistry and catalysis. Moreover, the potential applications for enantiomerically pure complexes include their use as sensors, antitumor agents, and other medicinal drugs, as well as templates for constructing 2D and 3D supramolecular structures. The introduction of ansa-η⁵-ligands significantly alters the electronic and spatial structure of the complexes, as well as their dynamics. Enantiomerically pure ansa-η⁵-complexes can be obtained through several approaches: synthesis from optically active precursors of η⁵-ligands, separation of diastereomers of the complexes utilizing enantiomerically pure agents, and the reaction of chiral σ-complexes with prochiral anions of η⁵-ligands. In organometallic synthesis, kinetic resolution of racemic complexes through interaction with derivatizing enantiomerically pure reagents is most commonly employed. However, the approach involving the fixation of ligands with a relatively rigid chiral bridge provides excellent results, allowing for the immediate production of enantiomers of metallocenes, including those that are less stable, such as indenyl complexes, which typically do not withstand separation using derivatizing reagents. The yield of the reaction and stereoselectivity in the synthesis of metal complexes is predetermined by a combination of numerous factors: the nature of the transition metal, the structure of the ligand, and the conditions of the synthesis (solvent, temperature, reagent ratio, etc.). As literature analysis shows, there is currently no universal solution to the problem of selecting these factors in the synthesis of desired enantiomerically pure metallocene structures.

The ansa-complexes of IVB and IIIB group metals demonstrate exceptional activity and stereoselectivity in the formation of metal–carbon, carbon–carbon, carbon–hydrogen, and carbon–heteroatom bonds. These complexes have proven to be effective catalysts for enantioselective alkene carbometalation, oligomerization, and polymerization. Notably, the stability of the spatial structure of ansa-complexes enhances stereoselectivity as the oligomeric chain grows (chain control). In contrast, chiral conformationally labile complexes can achieve high stereoselectivity at the initial stages of alkene insertion into catalytically active sites (site control); however, they struggle to maintain adequate control over stereoinduction as the chain lengthens. This area of research holds significant promise, particularly in advancing diastereo- and enantioselective oligomerization and polymerization of alkenes using conformationally stable complexes, facilitating the stereoselective synthesis of biologically active compounds and optically active polymers, which are in demand for the development of new materials.

Examples of asymmetric cycloaddition, cyclization, hydrosilylation, hydroamination, hydrogenation, etc., catalyzed by enantiomerically pure ansa-complexes of Group IIIB and IVB metals demonstrate the potential of these complexes as effective catalysts for practically important processes. Further advancement in this area necessitates an understanding of the stereoinduction and chiral recognition mechanisms that enable comprehension of the structure and dynamics of catalytically active centers, as well as the processes of substrate coordination and transformations. These studies are highly relevant, as their realization will facilitate a more optimal approach to the development of catalytic systems and processes based on them.