Diastereoselective Reformatsky Reaction Mediated by Dichlorocyclopentadienyltitanium(III)

Abstract

The Reformatsky reaction, first reported in 1887, has long been recognized as a fundamental method for carbon–carbon bond construction due to its mild conditions and functional group tolerance. Over the past few decades, this transformation has undergone a notable revival, with modern catalytic variants addressing limitations of stoichiometric protocols and expanding its role in complex molecule synthesis. Yet, despite its versatility, achieving stereoselective control remains a longstanding challenge. Herein we report the use of dichlorocyclopentadienyltitanium(III) (CpTiCl₂), generated in situ from CpTiCl₃ and manganese, as an efficient catalyst for Reformatsky-type couplings of aldehydes with α-haloesters and α-iodonitriles. Under mild conditions, CpTiCl₂ promotes the formation of β-hydroxy esters in high yields and with significant diastereoselective preference for the syn isomer (up to 100:0 syn:anti). This behavior contrasts sharply with the poor or anti-selective outcomes previously observed with titanocene(III) chloride (Cp₂TiCl). Mechanistic analysis suggests that the unique steric and electronic environment of CpTiCl₂—characterized by enhanced Lewis acidity and increased coordination vacancies—favors a Zimmerman–Traxler-type transition state that enforces syn stereocontrol. The methodology tolerates a wide variety of substrates, including aliphatic and aromatic aldehydes as well as α-iodonitriles, extending the scope of titanium-mediated Reformatsky chemistry. These findings establish CpTiCl₂ as a sustainable, selective, and robust organotitanium catalyst for stereoselective carbon–carbon bond formation, providing a promising alternative to the Nugent reagent and paving the way for new applications in complex molecule synthesis.

1. Introduction

The Reformatsky reaction, discovered in 1887 by Serhii Mykolayovych Reformatskyi, stands as one of the most enduring and adaptable transformations in organic chemistry [1]. By introducing a zinc-mediated coupling of ethyl haloacetates with aldehydes or ketones to yield β-hydroxyesters, Reformatsky offered a method of carbon–carbon bond formation that was both milder and more selective than the strongly basic organometallic reagents available at the time [2]. Unlike early nucleophiles, zinc enolates tolerated functional groups and enabled access to sterically hindered products, giving the reaction a distinct role in synthetic methodology. This complementarity to the later-emerging Grignard chemistry ensured its longevity, and more than a century later its core principle —the in situ generation of a metal enolate from an α-halo carbonyl compound, followed by nucleophilic addition to a carbonyl electrophile—continues to inspire innovation [3].

Over the past three decades, the Reformatsky reaction has experienced a revival [4,5]. Classical stoichiometric protocols, though effective, were hampered by metal waste, purification issues, and limited substrate scope. Modern developments have reimagined the process as a versatile catalytic platform [6,7] capable of operating under milder and more sustainable conditions. This shift has redefined the meaning of the “Reformatsky reaction” now understood as a family of transformations in which α-halo carbonyls engage electrophiles in the presence of a wide range of metals. Transition metals such as chromium [8], iridium [9,10], or palladium [11] have been harnessed to tackle challenges of selectivity, reactivity, and sustainability, demonstrating how a traditional transformation can be redefined by contemporary design. Other metals have also attracted considerable attention for their ability to mediate the coupling between α-haloesters and carbonyl compounds while minimizing competing reactions and byproduct formation, like indium [12], germanium [13,14] or tin [15,16]. The modern Reformatsky reaction has become a versatile tool in complex molecule construction, demonstrating advances in stereocontrol and efficiency [17]. Once limited to simple β-hydroxyesters, it now provides powerful strategies for assembling intricate architectures and enantioenriched building blocks, securing its place in contemporary natural product and pharmaceutical synthesis [8,10,18,19,20].

Among these modern advances, titanium(III) reagents hold a particularly prominent place. Titanocene(III) chloride (Cp₂TiCl), known as the Nugent–RajanBabu reagent, has proven to be a versatile tool in organic synthesis [21]. Functioning as a powerful single-electron transfer (SET) agent, it promotes a wide variety of radical and organometallic transformations under mild conditions. In addition to its efficiency, Cp₂TiCl is often considered environmentally friendly, as it can be regenerated from the widely available Cp₂TiCl₂, aligns with the principles of Green Chemistry, and operates without the need for stoichiometric additives [22,23]. These features, combined with its low toxicity and operational simplicity, have contributed to its prominent position in contemporary synthetic methodology, which spans multiple domains of carbon–carbon bond formation [24]. Its applications range from umpolung processes [25,26], to radical-mediated ring-opening and metallation reactions [27], cooperative catalysis [28,29], and even hydrogen-atom transfer and photochemical water splitting [21,30].

The utility of Cp₂TiCl also extends to Reformatsky-type reactions [31,32,33]. Compared to zinc-based protocols, the titanium(III)-catalyzed variant proceeds under milder, homogeneous conditions and typically affords shorter reaction times and cleaner product mixtures. Early work by Little and co-workers in 2003 demonstrated the feasibility of these transformations with α-haloesters and aldehydes, although stoichiometric amounts of titanocene were required (Scheme 1a) [31]. Later, catalytic versions were developed [32], being specially successful the use of the combination of TMSCl and collidine as titanocene regenerating system introduced by Cuerva and co-workers (Scheme 1b) [33].

However, despite these advantages, the diastereoselective outcome remains a critical limitation: most reported cases deliver poor syn/anti ratios or are even biased toward the anti isomer, severely restricting the applicability of this otherwise valuable methodology [31,32,33] (Scheme 1).

To address this challenge, we turned our attention to the closely related half-sandwich complex CpTiCl₂ [34]. Unlike Cp₂TiCl, this mono-cyclopentadienyl species possesses an 11-electron coordination sphere (vs. 15 for Cp₂TiCl), providing enhanced Lewis acidity, greater coordination flexibility, and reduced steric hindrance. These features are expected to exert a decisive influence on both reactivity and stereocontrol. Indeed, CpTiCl₂, generated in situ by the reduction of CpTiCl₃ with Mn, has already demonstrated excellent performance in Barbier-type allylation and propargylation of aldehydes with high regio- and stereoselectivity [35,36]. Building on this precedent, we envisaged that CpTiCl₂ could overcome the long-standing stereoselectivity issues in titanium(III)-mediated Reformatsky reactions. Herein we report the first systematic study of CpTiCl₂ in Reformatsky-type couplings, demonstrating that this reagent not only matches the reactivity of Cp₂TiCl but also delivers a remarkable diastereoselective preference for the syn isomer. This work highlights how subtle changes in the ligand environment and electron count of titanium complexes can translate into substantial improvements in stereochemical control, thereby expanding the synthetic scope and significance of Reformatsky chemistry (Scheme 1).

2. Results and Discussion

To evaluate the performance of CpTiCl₂ in Reformatsky-type couplings, we first examined the benchmark reaction between undec-10-enal (1a) and methyl 2-bromo-2-methylpropanoate (2) (Scheme 2) under catalytic conditions similar to those previously developed for Barbier-type processes in our group [35,36]. CpTiCl₂ is obtained by reduction with Mn of CpTiCl₃, and as a regenerating agent we used only Me₃SiBr instead of the regenerating mixture of Me₃SiBr and Et₃N·HBr used in previous works [35]. The desired β-hydroxy ester 3 was obtained in excellent yield (87%, Scheme 2).

The overall efficacy of the catalytic method was evaluated with a variety of aldehydes (1a–1d) and α-halo esters (4a–4c), obtaining a collection of β-hydroxy esters with good to high yields (70–89%

Table 1. Titanium catalyzed Reformatsky reaction with different substrates.

Entry	Aldehyde	Bromoester	Hydroxyester	Syn:Anti Ratio	Yield (c)
1 (a)	1a	4a R1 = R2 = Me		82:18	89%
2 (a)		4a R1 = R2 = Me		84:16	72%
3 (a)		4a R1 = R2 = Me		81:19	70%
4 (a)	1a	4b R1 = iPr R2 = Et		83:17	71%
5 (a)	1b	4b R1 = iPr R2 = Et		82:18	79%
6 (a)	1a	4c R1 = Me R2 = tBu		100:0	78%
7 (a)	1b	4c R1 = Me R2 = tBu		100:0	80%
8 (b)		4c R1 = Me R2 = tBu		100:0	78%

^(a) Procedure A: CpTiCl₃ (0.1 equiv), Mn (2 equiv), Me₃SiBr (3 equiv), THF, r.t. ^(b) Procedure B: CpTiCl₃ (1 equiv), Mn (2 equiv), THF, r.t. ^(c) overall yield after purification by column chromatography.

).

The aromatic aldehyde 1d (Table 1, entry 8) required stoichiometric conditions to achieve acceptable conversion, as the reaction was sluggish under catalytic conditions. This outcome aligns with previous observations that such substrates are prone to undergo pinacol coupling in the presence of Cp₂TiCl [35,37]. However, under stoichiometric conditions with CpTiCl₂ no pinacol-type by products were detected, and the desired β-hydroxy ester was obtained in notable yield (70%) with outstanding diastereoselectivity (100% syn).

The key feature of this transformation is its diastereoselective outcome. Across a wide substrate range, the syn isomer predominated in ratios of approximately 8:2. Even more strikingly, reactions involving the bulky tert-butyl ester 4c (Table 1, entries 6–8) yielded exclusively the syn isomer (100:0), underscoring the powerful stereocontrol achievable with CpTiCl₂. This stands in sharp contrast to Cp₂TiCl-mediated Reformatsky reactions, where either poor selectivity or anti-bias is generally reported [31,32,33].

We propose that the observed stereochemical bias originates from a Zimmerman–Traxler-type transition state (Scheme 3). In this chair-like six-membered arrangement, the ester substituent occupies a pseudo-equatorial position, thereby minimizing steric repulsions with the aldehyde substituent. The result is a predictable syn orientation of the new stereocenters. The ability of CpTiCl₂ to enforce this ordered geometry likely arises from its enhanced Lewis acidity, reduced steric hindrance, and greater coordination flexibility compared to Cp₂TiCl. These features facilitate strong coordination to both the ester and the aldehyde, favoring a cyclic transition state and rationalizing the exceptional syn-selectivity observed. In contrast, other titanium(III) systems reported previously probably operate through less ordered, non-cyclic pathways, which could account for their limited or even anti-selective diastereocontrol.

To elucidate the stereochemistry of the hydroxyesters, the acetals 6 and 7 were prepared in two steps starting from 5f and 5g (Scheme 4). First, the reduction with lithium aluminum hydride in THF at 0 °C allows the corresponding diol, and then, the acetal formation with 2,2-dimethoxypropane and camphorsulfonic acid in DCM is achieved.

Once the signals of acetals 6 and 7 were assigned using bidimensional NMR experiments such as HSQC and HMBC, the mono and bidimensional NOE spectra (Figure 1) enabled the identification of correlations between H4 and the axial methyl at position 2 (Scheme 4, green arrow), and the equatorial proton H6 with the methyl at position 5 (Scheme 4, blue arrow). The absence of correlation between the methyl at C5 and the axial proton H6 (Figure 1) also supports the syn configuration of the Reformatsky products 5f and 5g.

We have also extended the scope of the reaction to α-halo nitriles (Scheme 5). CpTiCl₂ promoted the Reformatsky reaction between iodoacetonitrile (8) and two electronically different aldehydes, one aliphatic and the other aromatic. The reaction with the aliphatic aldehyde 1b proceeded efficiently, affording the β-hydroxy nitrile 9 in good yield. However, in the case of the aromatic aldehyde 1d, pinacol coupling partially competed with the desired Reformatsky reaction, reflecting substrate-dependent limitations [35].

Taken together, these results establish CpTiCl₂ as a robust, selective, and general catalyst for titanium-mediated Reformatsky reactions. The system overcomes the long-standing issue of poor or anti-selective diastereocontrol associated with Cp₂TiCl, delivering instead a strong and predictable syn bias. This enhanced performance highlights the importance of subtle changes in ligand environment and electron count in dictating the outcome of organotitanium catalysis. From a broader perspective, the predictable syn-selectivity has significant synthetic implications. β-Hydroxy esters and nitriles are versatile building blocks in polyketide and macrolide synthesis, and the ability to access them stereocontrolled under mild, catalytic conditions positions CpTiCl₂ as a valuable addition to the toolbox of sustainable C–C bond-forming methodologies.

3. Materials and Methods

3.1. General Remarks

In all experiments involving Ti(III), reactions were performed under an argon atmosphere, using oven-dried glassware in all cases. THF was distilled from Na/benzophenone under argon and was deoxygenated prior to use. NMR spectra were acquired at ambient temperature on Bruker Nanobay Avance III HD 300 MHz, Avance III HD 500 MHz, and Avance III HD 600 MHz spectrometers (Bruker Corporation, Billerica, MA, USA). ¹H NMR, proton-decoupled ¹³C NMR and DEPT-135 were recorded in all cases. In some instances, quantitative ¹³C NMR was used for the determination of the ratio of syn:anti isomers. When required, edited HSQC was used for signal assignation and disambiguation and NOESY-1D/2D for relative configuration determination. Chemical shifts (δ) are expressed in ppm and coupling constants (J) in hertz (Hz). Chemical shifts are reported using CDCl₃ as internal reference. IR Spectra were recorded on a Bruker Alpha spectrometer (Bruker Corporation, Billerica, MA, USA). Mass spectra were obtained on a Waters Xevo LC-QTof-MS instruments (Waters Corporation, Milford, MA, USA) by electrospray ionization (ESI). Reactions were monitored by thin-layer chromatography (TLC) carried out on 0.2 mm DC-Fertigfolien Alugram^® (Southend-On-Sea, UK) XtraSil G/UV254 silica gel plates. The TLC plates were visualized with UV light and 7% phosphomolybdic acid/heat. Flash chromatography was performed on silica gel 60 (0.04–0.06 mm) using different eluents depending on the compound to be separated. Commercially available reagents were purchased from Aldrich Chemical Co. (Saint Louis, MO, USA), Acros (Fukuoka, Japan), Alfa Aesar (Karlsruhe, Germany), and TCI (Zwijndrecht, Belgium) and used as received.

3.2. General Procedure A for Ti-Catalyzed Reformatsky Reaction

Under an Ar atmosphere, dry THF (8 mL/1.4 mmol of aldehyde) previously deoxygenated was added to a miscellany of CpTiCl₃ (0.1 equiv) and Mn dust (2 equiv) resulting in a dark blue suspension. Then, Me₃SiBr (3 equiv) was added and the mixture turned turquoise. A solution of aldehyde or ketone (1 equiv) and α-halo ester (2 equiv) in THF (2 mL/1.4 mmol of aldehyde) was added dropwise, and the reaction was stirred overnight. The mixture was filtered, diluted with EtOAc, washed with HCl 3% and brine, and dried (anhydrous MgSO₄), and the solvent was removed. Products were purified by silica gel flash column chromatography (hexane/EtOAc mixtures).

Methyl 3-hydroxy-2,2-dimethyltridec-12-enoate (3). Reaction of undec-10-enal (1a) (0.24 mL, 1.19 mmol) and methyl 2-bromo-2-methylpropanoate (2) (0.31 mL, 2.38 mmol), according to the general procedure A, afforded product 3 (279 mg, 87%) isolated as light-yellow oil (hexane/EtOAc, 8:2; Rf: 0.33). IR (film) v (cm⁻¹) 3513, 2925, 2854, 1728, 1641, 1463, 1440, 1391, 1264, 1194, 1138, 1077, 992, 910, 862, 771. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 5.90–5.77 (m, 1H), 5.04–4.93 (m, 2H), 3.72 (s, 3H), 3.67 (s, 1H), 3.62 (d, J = 9.7 Hz, 1H), 2.09–2.02 (m, 2H), 1.61–1.58 (m, 1H), 1.41–1.28 (m, 13H), 1.21 (s, 3H), 1.19 (s, 3H).¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 178.2 (C), 139.2 (CH), 114.1 (CH₂), 76.7 (CH), 51.9 (CH₃), 47.2 (C), 33.8 (CH₂), 31.7 (CH₂), 29.6 (CH₂), 29.4 (CH₂), 29.1 (CH₂), 28.9 (CH₂), 26.7 (CH₂), 22.2 (CH₃), 20.4 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₆H₃₁O₃ 271.2273; found: 271.2258.

Methyl 3-hydroxy-2-methyltridec-12-enoate (5a). Reaction of undec-10-enal (1a) (0.12 mL, 0.60 mmol) and methyl 2-bromopropanoate (4a) (0.27 mL, 2.38 mmol), according to the general procedure A, afforded product 5a (137 mg, 89%) as a light-yellow oil (hexane/EtOAc, 8:2; Rf: 0.30) and an inseparable mixture of isomers (82:18 syn:anti). IR (film) v (cm⁻¹) 3471, 2925, 2854, 1736, 1720, 1640, 1459, 1436, 1256, 1197, 1170, 1043, 992, 909. ¹H NMR (600 MHz, CDCl₃) δ (ppm) common signals: 5.82 (ddt, J = 17.4, 10.2, 7.2 Hz, 1H), 5.02–4.98 (m, 1H), 4.95–4.92 (m, 1H), 3.72 (s, 3H), 2.57–2.52 (m, 1H), 2.06–2.03 (m, 2H), 1.52–1.45 (m, 2H), 1.41–1.36 (m, 3H), 1.29 (br s, 9H); characteristic signals of 5a syn: 3.89 (dt, J = 8.4, 4.2 Hz, 1H), 1.19 (d, J = 7.2 Hz, 3H); characteristic signals of 5a anti: 3.67–3.65 (m, 1H), 1.22 (d, J = 7.2 Hz, 3H). ¹³C NMR (150 MHz, CDCl₃, DEPT) δ (ppm) common signals: 139.2 (CH), 114.1 (CH₂), 33.8 (CH₂), 29.5 (CH₂), 29.5 (CH₂), 29.4 (CH₂), 29.1 (CH₂), 28.9 (CH₂); signals of 5a syn: 176.6 (C), 71.8 (CH), 51.8 (CH₃), 44.2 (CH), 33.8 (CH₂), 26.0 (CH₂), 10.6 (CH₃); signals of 5a anti: 176.5 (C), 73.4 (CH), 51.7 (CH₃), 45.2 (CH), 34.7 (CH₂), 25.5 (CH₂), 14.3 (CH₃).

Methyl 3-hydroxy-2-methyloctanoate (5b). Reaction of hexanal (1b) (0.25 mL, 1.96 mmol) and methyl 2-bromopropanoate (4a) (0.45 mL, 3.92 mmol), according to general procedure A, afforded compound 5b (266 mg, 72%) as a light-yellow oil (hexane/EtOAc, 9:1; Rf: 0.26) and an inseparable mixture of isomers (84:16 syn:anti). IR (film) v (cm⁻¹) 3455, 2931, 2862, 1723, 1459, 1371, 1346, 1259, 1198, 1121, 1040, 935. ¹H NMR (600 MHz, CDCl₃) δ (ppm) common signals: 3.64 (s, 3H), 2.47 (qd, J = 7.2, 3.6 Hz, 1H), 1.44–1.37 (m, 2H), 1.36–1.19 (m, 7H), 0.82 (t, J = 6.6 Hz, 3H); characteristic signals of 5b syn: 3.82 (dt, J = 8.4, 4.2 Hz, 1H), 1.11 (d, J = 7.2 Hz, 3H); characteristic signals of 5b anti: 3.60–3.57 (m, 1H), 1.14 (d, J = 7.2 Hz, 3H). ¹³C NMR (150 MHz, CDCl₃, DEPT) δ (ppm) common signals: 22.6 (CH₂), 14.0 (CH₃); signals of 5b syn: 176.7 (C), 71.8 (CH), 51.8 (CH₃), 44.2 (CH), 33.8 (CH₂), 31.8 (CH₂), 25.7 (CH₂), 10.6 (CH₃); signals of 5b anti: 176.5 (C), 73.4 (CH), 51.7 (CH₃), 45.2 (CH), 34.7 (CH₂), 31.8 (CH₂), 25.2 (CH₂), 14.3 (CH₃). IR, ¹H NMR and ¹³C NMR spectral data of 5b syn and 5b anti are in agreement with literature values [38,39].

Reaction of hexanal (1b) (0.13 mL, 0.98 mmol) and methyl (R)-2-bromopropanoate (0.23 mL, 1.96 mmol), according to general procedure A, afforded compound 5b (123 mg, 65%) as a light-yellow oil and an inseparable mixture of isomers (79:21 syn:anti).

Methyl 3-hydroxy-2-methyl-5-phenylpentanoate (5c). Reaction of hydrocinnamaldehyde (1c) (0.22 mL, 1.49 mmol) and methyl 2-bromopropanoate (4a) (0.34 mL, 2.98 mmol), according to general procedure A, afforded compound 5c (232 mg, 70%) as a yellow oil (hexane/EtOAc, 1:1; Rf: 0.30) as an inseparable mixture of isomers (81:19 syn:anti). IR (film) v (cm⁻¹) 3442, 3027, 2944, 1721, 1453, 1199, 1165, 1040, 933, 745, 698, 496. ¹H NMR (600 MHz, CDCl₃) δ (ppm) common signals: 7.20–7.18 (m, 2H), 7.12–7.10 (m, 3H), 2.80–2.75 (m, 1H), 2.57 (ddd, J = 13.8, 9.7, 6.8 1H), 2.48–2.46 (m, 1H), 1.75–56 (m, 2H); characteristic signals of 5c syn: 3.82 (dt, J = 9.4, 3.6 Hz, 1H), 3.60 (s, 3H), 1.10 (d, J = 7.2 Hz, 3H); characteristic signals of 5c anti: 3.61 (s, 3H), 3.56–3.59 (m, 1H), 1.12 (d, J = 7.2 Hz, 3H). ¹³C NMR (150 MHz, CDCl₃, DEPT) δ (ppm) common signals: 176.5 (C), 141.9 (C), 128.5 (CH), 128.4 (CH), 125.9 (CH); signals of 5c syn: 71.1 (CH), 51.8 (CH₃), 44.5 (CH), 35.7 (CH₂), 32.3 (CH₂), 10.8 (CH₃); signals of 5c anti: 72.7 (CH), 51.8 (CH₃), 45.3 (CH), 36.6 (CH₂), 31.9 (CH₂), 14.3 (CH₃). IR, ¹H NMR and ¹³C NMR spectral data of 5c syn and 5c anti are in agreement with literature values [40].

Ethyl 3-hydroxy-2-isopropyltridec-12-enoate (5d). Reaction of undec-10-enal (1a) (0.24 mL, 1.19 mmol) and ethyl 2-bromo-3-methylbutanoate (4b) (0.39 mL, 2.38 mmol), according to the general procedure A, afforded product 5d syn (209 mg, 59%) and 5d anti (43 mg, 12%) (hexane/EtOAc, 9:1; Rf syn: 0.20, Rf anti: 0.30). Compound 5d syn, isolated as yellow oil: IR (film) v (cm⁻¹) 3469, 2925, 2854, 1729, 1710, 1641, 1465, 1372, 1178, 1026, 909. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 5.81 (ddt, J = 17.1, 10.2, 6.6 Hz, 1H, H12), 5.02–4.90 (m, 2H, H13), 4.16 (dq, J = 7.2, 3.6 Hz, 2H, OCH₂CH₃), 3.90–3.85 (m, 1H, H3), 2.34 (t, J = 6.8 Hz, 1H, H2), 2.23–2.11 (m, 2H, OH, CH(CH₃)₂), 2.04 (q, J = 7.2 Hz, 2H, H11), 1.46–1.28 (m, 17H), 0.99 (d, J = 6.3 Hz, 3H, CH(CH₃)₂), 0.97 (d, J = 6.3 Hz, 3H, CH(CH₃)₂). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 173.9 (C, C1), 139.2 (CH, C12), 114.1 (CH₂, C13), 70.8 (CH, C3), 60.1 (CH₂, OCH₂CH₃), 57.6 (CH, C2), 34.5 (CH₂), 33.8 (CH₂, C11), 29.5 (CH₂), 29.4 (CH₂), 29.1 (CH₂), 28.9 (CH₂), 26.6 (CH, CH(CH₃)₂), 26.0 (CH₂), 21.6 (CH₃, CH(CH₃)₂), 19.4 (CH₃, CH(CH₃)₂), 14.4 (CH₃, OCH₂CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₈H₃₅O₃ 299.2586; found: 299.2598. Compound 5d anti, isolated as yellow oil: IR (film) v (cm⁻¹) 3522, 2926, 2855, 1712, 1463, 1377, 1176, 1027, 910. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 5.83 (ddt, J = 16.8, 10.2, 6.6 Hz, 1H, H12), 5.04–4.92 (m, 2H, H13), 4.21 (q, J = 7.2 Hz, 2H, OCH₂CH₃), 3.80–3.72 (m, 1H, H3), 2.74 (d, J = 10.2 Hz, 1H, OH), 2.22–2.02 (m, 4H, H2, H11, CH(CH₃)₂), 1.46–1.25 (m, 17H), 1.04 (d, J = 6.4 Hz, 3H, CH(CH₃)₂), 0.94 (d, J = 6.4 Hz, 3H, (CH(CH₃)₂). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 175.7 (C, C1), 139.2 (CH, C12), 114.1 (CH₂, C13), 70.1 (CH, C3), 60.3 (CH₂, OCH₂CH₃), 57.3 (CH, C2), 36.6 (CH₂), 33.8 (CH₂, C11), 29.5 (CH₂), 29.5 (CH₂), 29.4 (CH₂), 29.1 (CH₂), 28.9 (CH₂), 27.8 (CH, CH(CH₃)₂), 26.0 (CH₂), 20.9 (CH₃, CH(CH₃)₂), 20.6 (CH₃, CH(CH₃)₂), 14.4 (CH₃, OCH₂CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₈H₃₅O₃ 299.2586; found: 299.2602.

Ethyl 3-hydroxy-2-isopropyloctanoate (5e). Reaction of hexanal (1b) (0.25 mL, 1.96 mmol) and ethyl 2-bromo-3-methylbutanoate (4b) (0.64 mL, 3.92 mmol), according to general procedure A, afforded compounds 5e syn (293 mg, 65%) and 5e anti (63 mg, 14%) (hexane/EtOAc, 9:1; Rf syn: 0.20, Rf anti: 0.27). Compound 5e syn, isolated as yellow oil: IR (film) v (cm⁻¹) 3469, 2958, 2932, 2873, 1728, 1709, 1465, 1373, 1228, 1177, 1025, 924. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 4.16 (dq, J = 7.2, 3.5 Hz, 2H, OCH₂CH₃), 3.91–3.86 (m, 1H, H3), 2.36–2.32 (m, 1H, H2), 2.12–215 (m, 1H, CH(CH₃)₂), 1.54–1.40 (m, 3H), 1.36–1.24 (m, 5H), 1.28 (t, J = 7.2 Hz, 3H, OCH₂CH₃), 0.99 (d, J = 6.3 Hz, 3H, CH(CH₃)₂), 0.97 (d, J = 6.3 Hz, 3H, CH(CH₃)₂), 0.89 (t, J = 6.9 Hz, 3H, H8). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 173.9 (C, C1), 70.7 (CH, C3), 60.1 (CH₂, OCH₂CH₃), 57.7 (CH, C2), 34.5 (CH₂), 31.7 (CH₂), 26.6 (CH, CH(CH₃)₂), 25.6 (CH₂), 22.6 (CH₂), 21.5 (CH₃, CH(CH₃)₂), 19.3 (CH₃, CH(CH₃)₂), 14.3 (CH₃, C8), 14.0 (CH₃, OCH₂CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₃H₂₇O₃ 231.1960; found 231.1945. Compound 5e anti, isolated as yellow oil: IR (film) v (cm⁻¹) 3523, 2959, 2932, 2873, 1709, 1465, 1376, 1178, 1126, 1025, 921. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 4.20 (q, J = 7.2 Hz, 2H, OCH₂CH₃), 3.80–3.71 (m, 1H, H3), 2.75 (d, J = 10.2 Hz, 1H, OH), 2.21–2.07 (m, 2H, H2, CH(CH₃)₂), 1.46–1.27 (m, 8H), 1.30 (t, J = 7.2 Hz, 3H, OCH₂CH₃), 1.02 (d, J = 6.3 Hz, 3H, CH(CH₃)₂), 0.93 (d, J = 6.3 Hz, 3H, CH(CH₃)₂), 0.89 (t, J = 6.9 Hz, 3H, H8). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 175.7 (C, C1), 70.0 (CH, C3), 60.3 (CH₂, OCH₂CH₃), 57.3 (CH, C2), 36.5 (CH₂), 31.7 (CH₂), 27.7 (CH, CH(CH₃)₂), 25.7 (CH₂), 22.6 (CH₂), 20.8 (CH₃, CH(CH₃)₂), 20.6 (CH₃, CH(CH₃)₂), 14.3 (CH₃, C8), 14.0 (CH₃, OCH₂CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₃H₂₇O₃ 231.1960; found 231.1950.

tert-Butyl (2S*,3R*)-3-hydroxy-2-methyltridec-12-enoate (5f). Reaction of undec-10-enal (1a) (0.24 mL, 1.19 mmol) and tert-butyl 2-bromopropanoate (4c) (0.41 mL, 2.38 mmol), according to the general procedure A, afforded product 5f (274 mg, 78%) isolated as light-yellow oil (hexane/EtOAc, 9:1; Rf: 0.22). IR (film) v (cm⁻¹) 3465, 2976, 2926, 2855, 1722, 1458, 1367, 1254, 1151, 909, 847. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 5.83 (ddt, J = 17.1, 10.2, 6.6 Hz, 1H, H12), 5.04–4.92 (m, 2H, H13), 3.88–3.82 (m, 1H, H3), 2.68 (d, J = 4.5 Hz, 1H, OH), 2.43 (dq, J = 7.2, 3.6 Hz, 1H, H2), 2.05 (m, 2H, H11), 1.48 (s, 9H, C(CH₃)₃), 1.41–1.30 (m, 14H), 1.15 (d, J = 7.2 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 175.8 (C, C1), 139.2 (CH, C12), 114.1 (CH₂, C13), 80.9 (C, C(CH₃)₃), 71.8 (CH, C3), 44.9 (CH, C2), 33.8 (CH₂), 33.7 (CH₂, C11), 29.6 (CH₂), 29.5 (CH₂), 29.4 (CH₂), 29.1 (CH₂), 28.9 (CH₂), 28.1 (CH₃, C(CH₃)₃), 26.0 (CH₂), 10.8 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₈H₃₅O₃ 299.2586; found: 299.2564.

tert-Butyl (2S*,3R*)-3-hydroxy-2-methyloctanoate (5g). Reaction of hexanal (1b) (0.25 mL, 1.96 mmol) and tert-butyl 2-bromopropanoate (4c) (0.67 mL, 3.92 mmol), according to general procedure A, afforded compound 5g (363 mg, 80%), isolated as yellow oil (hexane/EtOAc, 9:1; Rf: 0.22). IR (film) v (cm⁻¹) 3437, 2931, 2865, 1719, 1459, 1367, 1255, 1152, 1024, 847. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 3.88–3.82 (m, 1H, H3) 2.69 (br s, 1H, OH), 2.43 (dq, J = 7.2, 3.6 Hz, 1H, H2), 1.48 (s, 9H, C(CH₃)₃), 1.44–1.27 (m, 8H), 1.15 (d, J = 7.2 Hz, 3H), 0.90 (t, J = 6.9 Hz, 3H, H8). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 175.8 (C, C1), 80.9 (C, C(CH₃)₃), 71.8 (CH, C3), 44.9 (CH, C2), 33.7 (CH₂), 31.8 (CH₂), 28.1 (CH₃, C(CH₃)₃), 25.7 (CH₂), 22.6 (CH₂), 14.0 (CH₃, C8), 10.8 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₃H₂₇O₃ 231.1960; found.231.1974.

3.3. General Procedure B for Ti-Induced Reformatsky Reaction

Under an Ar atmosphere, dry THF (8 mL/1.4 mmol of aldehyde) previously deoxygenated was added to a miscellany of CpTiCl₃ (1 equiv) and Mn dust (2 equiv) resulting in a green suspension. A solution of aldehyde (1 equiv) and α-halo ester or 2-iodoacetonitrile (2 equiv) in THF (2 mL/1.4 mmol of aldehyde) was added dropwise, and the reaction was stirred overnight. The mixture was filtered, diluted with EtOAc, washed with HCl 3% and brine, and dried (anhydrous MgSO₄), and the solvent was removed. Products were purified by silica gel flash column chromatography (hexane/EtOAc mixtures).

tert-Butyl (2R*,3R*)-3-hydroxy-3-(3-methoxyphenyl)-2-methylpropanoate (5h). Reaction of 3-methoxybenzaldehyde (1d) (0.18 mL, 1.47 mmol) and tert-butyl 2-bromopropanoate (4c) (0.50 mL, 2.94 mmol), according to the general procedure B, afforded compound 5h (308 mg, 78%) isolated as colorless oil (hexane/EtOAc, 7:3; Rf: 0.40). ¹H NMR spectral data are in agreement with literature values [41]. IR (film) v (cm⁻¹) 3479, 2976, 2937, 2838, 1714, 1596, 1485, 1458, 1254, 1151, 1032, 845, 786, 700.¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 175.4 (C), 159.6 (C), 143.3 (C), 129.2 (CH), 118.5 (CH), 113.0 (CH), 111.7 (CH), 81.1 (C), 73.6 (CH), 55.2 (CH₃), 47.0 (CH), 28.0 (CH₃), 11.0 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₄H₂₁O₃ 237.1491; found: 237.1509.

3-Hydroxyoctanenitrile (9). Reaction of hexanal (1b) (0.13 mL, 0.98 mmol) and 2-iodoacetonitrile (8) (0.33 g, 1.96 mmol), according to general procedure B, afforded compound 9 (131 mg, 95%) isolated as yellow oil (hexane/EtOAc, 8:2; Rf: 0.22). ¹H NMR and ¹³C NMR spectral data are in agreement with literature values [42]. IR (film) v (cm⁻¹) 3446, 2930, 2863, 2253, 1714, 1461, 1415, 1267, 1078, 1041, 911, 731.

3-Hydroxy-3-(3-methoxyphenyl)propanenitrile (10) and 1,2-bis(3-methoxyphenyl)ethane-1,2-diol. Reaction of 3-methoxybenzaldehyde (1d) (0.10 g, 0.74 mmol) and 2-iodoacetonitrile (8) (0.25 g, 1.48 mmol), according to the general procedure B, afforded products 10 (32 mg, 32%) and 1,2-bis(3-methoxyphenyl)ethane-1,2-diol (36 mg, 18%). Compound 10 was isolated as a light-yellow oil (hexane/EtOAc, 1:1; Rf: 0.40). ¹H NMR and ¹³C NMR spectral data are in agreement with literature values [43]. IR (film) v (cm⁻¹) 3434, 3005, 2924, 2842, 2255, 1711, 1594, 1460, 1259, 1154, 1039, 788, 698. 1,2-bis(3-methoxyphenyl)ethane-1,2-diol was isolated as a pale-yellow solid (dl/meso mixture), and spectral data are in agreement with literature values [35].

3.4. Reduction of the β-Hydroxy Esters

A solution of 5f or 5g (1 equiv) in dry THF (15 mL/1.5 mmol) at 0 °C was added dropwise to a solution of LiAlH₄ (4 equiv) in dry THF (15 mL/1.5 mmol). The mixture was stirred under N₂ at room temperature for 7 h. Then, Na₂SO₄·10H₂O was added at 0 °C and diluted with Et₂O. The solution was filtered, and the two phases were separated. The organic phase was dried over MgSO₄, filtered and concentrated in vacuum. The products were purified by flash chromatography (hexane/EtOAc mixtures).

(2R*,3R*)-2-methyltridec-12-ene-1,3-diol (269 mg, 77%) was isolated as colorless oil (hexane/EtOAc, 6:4; Rf: 0.30). IR (film) v (cm⁻¹) 3331, 2924, 2853, 1641, 1461, 1089, 1028, 911, 908, 722. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 5.79 (ddt, J = 17.1, 10.2, 6.9 Hz, 1H), 4.99–4.89 (m, 2H), 3.79–3.73 (m, 1H), 3.62–3.60 (m, 2H), 2.05–1.98 (m, 2H), 1.76–1.70 (m, 1H), 1.48–1.27 (m, 14H), 0.85 (d, J = 7.0 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 139.1 (CH), 114.1 (CH₂), 73.7 (CH), 66.4 (CH₂), 38.9 (CH), 33.9 (CH₂), 33.8 (CH₂), 29.7 (CH₂), 29.6 (CH₂), 29.5 (CH₂), 29.1 (CH₂), 28.9 (CH₂), 26.3 (CH₂), 10.0 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₄H₂₉O₂ 229.2168; found: 229.2154.

(2R*,3R*)-2-methyloctane-1,3-diol (82 mg, 49%) was isolated as colorless oil (hexane/EtOAc, 6:4; Rf: 0.36). IR (film) v (cm⁻¹) 3351, 2928, 2863, 1460, 1370, 1227, 1026, 979. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 3.80–3.75 (m, 1H), 3.63–3.62 (m, 2H), 1.77–1.70 (m, 1H), 1.49–1.21 (m, 8H), 0.89–0.84 (m, 6H). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 74.0 (CH), 66.6 (CH₂), 39.0 (CH), 33.9 (CH₂), 31.9 (CH₂), 25.9 (CH₂), 22.6 (CH₂), 14.0 (CH₃), 10.0 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₉H₂₁O₂ 161.1542; found 161.1558.

3.5. Synthesis of the Ketal Derivatives

To a solution of the diol ((2R,3R)-2-methyltridec-12-ene-1,3-diol or (2R,3R)-2-methyloctane-1,3-diol (1 equiv)) and 2,2-dimethoxypropane (3 equiv) in dry DCM (10 mL/0.3 mmol), camphor sulfonic acid (0.3 equiv) was added at 0 °C. The mixture was stirred under N₂ at room temperature overnight. Then, an aqueous solution of NaHCO₃ sat. was added. The aqueous layers were extracted with EtOAc, and the combined organic layers were dried over MgSO₄, filtered and concentrated in vacuum. The products were purified by flash chromatography (hexane/EtOAc mixtures).

(4R*,5S*)-4-(dec-9-en-1-yl)-2,2,5-trimethyl-1,3-dioxane (6) (58 mg, 72%) isolated as colorless oil (hexane/EtOAc, 8:2; Rf: 0.26). IR (film) v (cm⁻¹) 2991, 2925, 2854, 1641, 1460, 1378, 1271, 1242, 1196, 1103, 1006, 907, 850. ¹H NMR (500 MHz, CDCl₃) δ (ppm) 5.82 (ddt, J = 17.0, 10.2, 6.7 Hz, 1H, H9′), 5.00 (ddt, J = 17.0, 2.2, 1.6 Hz, 1H, H10′a), 4.94 (ddt, J = 10.2, 2.2, 1.3 Hz, 1H, H10′b), 4.11 (dd, J = 11.5, 2.8 Hz, 1H, H6_ax), 3.91 (ddd, J = 7.6, 5.5, 2.5 Hz, 1H, H4), 3.59 (dd, J = 11.5, 1.6 Hz, 1H, H6_eq), 2.07–2.03 (m, 2H, H8′), 1.45 (s, 3H, CH_3ax), 1.40 (s, 3H, CH_3eq), 1.29–1.20 (m, 15H), 1.06 (d, J = 6.9 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 139.2 (CH, C9′), 114.1 (CH₂, C10′), 98.5 (C, C2), 71.5 (CH, C4), 67.0 (CH₂, C6), 33.8 (CH₂, C8′), 32.9 (CH₂), 31.6 (CH_, C5), 29.8 (CH₃, CH_3eq), 29.6 (CH₂), 29.5 (CH₂), 29.4 (CH₂), 29.1 (CH₂), 28.9 (CH₂), 25.2 (CH₂), 19.1 (CH₃, CH_3ax), 10.5 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₇H₃₃O₂ 269.2481; found: 269.2475.

(4R*,5R*)-2,2,5-trimethyl-4-pentyl-1,3-dioxane (7) (51 mg, 46%) isolated as colorless oil (hexane/EtOAc, 8:2; Rf: 0.30). IR (film) v (cm⁻¹) 2932, 2861, 1460, 1378, 1270, 1240, 1196, 1105, 1006, 860. ¹H NMR (300 MHz, CDCl₃) δ (ppm) 4.12 (dd, J = 11.4, 3.0 Hz, 1H, H6_ax), 3.95–3.90 (m, 1H, H4), 3.60 (dd, J = 11.4, 1.7 Hz, 1H, H6_eq), 1.46 (s, 3H, CH_3ax), 1.41 (s, 3H, CH_3eq), 1.51–1.24 (m, 9H), 1.07 (d, J = 6.9 Hz, 3H, CH₃), 0.93–0.88 (m, 3H, H5′). ¹³C NMR (75 MHz, CDCl₃, DEPT) δ (ppm) 98.5 (C, C2), 71.5 (CH, C4), 67.0 (CH₂, C6), 32.8 (CH₂), 31.8 (CH₂), 31.6 (CH, C5), 29.8 (CH₃, CH_3eq), 24.9 (CH₂), 22.6 (CH₂), 19.1 (CH₃, CH_3ax), 14.1 (CH₃, C5′), 10.5 (CH₃). HRMS (ESI/Q-TOF) m/z: [M+H]⁺ calcd for C₁₂H₂₅O₂ 201.1855 found 201.1849.

4. Conclusions

In summary, we have demonstrated that the half-sandwich titanium(III) complex CpTiCl₂ represents a highly effective alternative to the Nugent–RajanBabu reagent in Reformatsky-type couplings. The methodology affords β-hydroxy esters and nitriles in consistently high yields and, more importantly, with a remarkable diastereoselective bias towards the syn isomer. This outcome clearly contrasts with the limited or anti-selective stereocontrol previously reported for Cp₂TiCl-mediated systems. The improved performance of CpTiCl₂ can be attributed to its lower steric hindrance, greater Lewis acidity, and increased coordination flexibility, which favor a Zimmerman–Traxler-type transition state that enforces syn selectivity.

The scope of the reaction, encompassing aliphatic, unsaturated, and aromatic aldehydes as well as α-bromoesters and α-iodonitriles, underscores the robustness and synthetic utility of this titanium system. Taken together, these results reveal CpTiCl₂ as a powerful, sustainable, and stereoselective catalyst for Reformatsky reactions, with potential implications for the construction of complex molecular architectures.