Simmons–Smith reaction

The Simmons–Smith reaction is an organic chemical reaction that effects the stereospecific cyclopropanation of alkenes by adding a methylene group via an organozinc carbenoid intermediate, generated in situ from diiodomethane (CH₂I₂) and a zinc-copper couple in ether solvent.^[1] This mild, concerted process delivers syn addition across the double bond, preserving the alkene's geometry and producing cyclopropanes in high yields with excellent chemoselectivity toward electron-rich olefins.^[2] Discovered in 1958 by Howard E. Simmons Jr. and Ronald D. Smith at the DuPont Company, the reaction builds on earlier work with halomethylzinc halides and has become a cornerstone method for constructing strained three-membered rings due to its compatibility with a wide range of functional groups, including alcohols, ethers, and halides, under neutral conditions.^[1]^[2] Key advantages of the Simmons–Smith reaction include its operational simplicity, minimal side reactions, and ability to achieve high stereocontrol, particularly when directing groups like allylic hydroxyl moieties coordinate to the zinc carbenoid for diastereoselective outcomes.^[2] Over the decades, variants have expanded its scope: the Furukawa modification employs diethylzinc (Et₂Zn) with CH₂I₂ for faster generation of the carbenoid and broader substrate tolerance, while asymmetric versions incorporate chiral ligands such as bis(sulfonamide)s or bipyridine derivatives to enable enantioselective cyclopropanation with ee values often exceeding 90%.^[3] These adaptations, including those using chloromethyl iodide or samarium-based reagents, allow for the introduction of substituted alkylidene groups beyond simple methylene.^[2]^[3] The reaction's utility extends to the total synthesis of complex natural products and pharmaceuticals featuring cyclopropane motifs, such as terpenoids (e.g., dictyopterenes), alkaloids (e.g., lundurine A), and drugs like saxagliptin (a DPP-4 inhibitor for diabetes) and lenvatinib (an anticancer tyrosine kinase inhibitor), where the stereochemical precision is crucial for biological activity.^[3] Its mechanistic insight—a butterfly-like transition state involving electrophilic attack by the zinc-bound carbenoid—has informed ongoing developments in carbenoid chemistry, ensuring its enduring role in synthetic organic chemistry despite competition from transition-metal-catalyzed alternatives.^[3]^[2]

Overview and History

Discovery and Original Conditions

The Simmons–Smith reaction was discovered in 1958 by Howard Ensign Simmons, Jr. and Ronald D. Smith at the Central Research Department of the DuPont Company. While exploring organozinc reagents for carbon-carbon bond formation, they identified a method for stereospecifically synthesizing cyclopropanes from alkenes using diiodomethane as the methylene source. Their preliminary findings were published later that year in a communication to the Journal of the American Chemical Society.^[1]^[1] The original procedure employed diiodomethane (CH₂I₂) and a zinc-copper couple (prepared from zinc dust and a trace of copper sulfate) in anhydrous diethyl ether as the solvent. The alkene substrate was added to the preformed reagent mixture, which was then stirred at room temperature under an inert atmosphere for 1–3 days to allow complete conversion. This setup generated an organozinc carbenoid in situ, enabling mild conditions suitable for unfunctionalized alkenes without the need for high temperatures or pressures. A detailed experimental protocol, submitted by the inventors themselves, appeared in Organic Syntheses in 1966, confirming the reliability of these conditions for laboratory-scale preparations.^[1]^[4] Early experiments highlighted the reaction's stereospecific syn addition across the alkene double bond, preserving the geometry of the starting material in the resulting cyclopropane. For instance, cyclohexene was converted to norcarane (bicyclo[4.1.0]heptane) in a representative demonstration of the method's efficacy with simple cyclic alkenes. Initial yield data for unfunctionalized alkenes, such as terminal and internal olefins, ranged from 50–80%, depending on substrate reactivity and reaction scale, establishing the reaction as a practical tool for cyclopropane synthesis despite moderate efficiency in some cases.^[1]^[4]

General Reaction Scheme

The Simmons–Smith reaction is a stereospecific cyclopropanation of alkenes mediated by a zinc carbenoid, converting the carbon-carbon double bond into a three-membered ring through methylene transfer. The general reaction scheme is represented as:

\ce{>C=C< + CH2I2 + Zn -> >CH-CH2-C< + ZnI2}

where >C=C< denotes a generic alkene, and >CH-CH2-C< indicates the resulting cyclopropane derivative.^[5] The key reagents include diiodomethane (CH₂I₂) as the methylene source and a zinc-copper couple (Zn-Cu), prepared by activating zinc dust with copper sulfate to generate the organozinc carbenoid ICH₂ZnI in situ.^[5] This carbenoid species facilitates the controlled addition of the CH₂ unit to the alkene.^[6] Typical conditions for the standard Simmons–Smith reaction involve anhydrous diethyl ether as the solvent, room temperature operation under an inert atmosphere (e.g., nitrogen), and reaction times of 1–3 days to ensure complete conversion.^[5] These parameters minimize side reactions and promote efficient carbenoid formation and transfer.^[6] The reaction proceeds with syn stereochemistry, preserving the geometry of the starting alkene; for instance, a cis-alkene yields the corresponding cis-substituted cyclopropane.^[5] A representative example is the conversion of cyclohexene to bicyclo[4.1.0]heptane (norcarane) in 70% yield, demonstrating the method's utility for simple cyclic alkenes.^[5]

Mechanism

Carbenoid Formation

The Simmons–Smith reaction begins with the formation of an organozinc carbenoid species from the reaction of a zinc-copper couple with diiodomethane (CH₂I₂). This initial step generates the active reagent, typically represented as iodomethylzinc iodide (ICH₂ZnI), which serves as the methylene transfer agent in subsequent cyclopropanation. The preparation involves activating zinc powder with copper to form the couple, followed by addition of CH₂I₂ in an inert atmosphere.^[1]^[6] The proposed structure of the carbenoid is a zinc-bound methylene species, often denoted as :CH₂-ZnI, where the carbon-zinc bond imparts carbenoid reactivity without generating a free carbene. A simplified equation for the formation is:

\mathrm{Zn} + \mathrm{CH_2I_2} \rightarrow \mathrm{ICH_2ZnI}

This process likely proceeds via oxidative addition of the zinc to the C-I bond of CH₂I₂, yielding the halomethylzinc iodide, which may exist in equilibrium with dimeric or solvated forms depending on the conditions. Structural characterization through NMR spectroscopy and X-ray crystallography has confirmed the tetrahedral coordination around zinc in these species, supporting the bound carbenoid model.^[7]^[6] Copper plays a crucial role in activating the zinc surface, preventing passivation and facilitating the oxidative addition step by promoting electron transfer. Without copper, the reaction with plain zinc is significantly slower or inefficient, as the metal surface becomes coated with zinc iodide. This activation is essential for generating the reactive carbenoid in high yield.^[1]^[6] Evidence for the carbenoid nature includes the absence of free carbene intermediates, as no products typical of singlet carbenes (such as C-H insertions remote from directing groups) are observed under standard conditions. Instead, the reagent exhibits selective reactivity, such as insertions into allylic C-H bonds or reactions with electrophiles like aldehydes, consistent with a coordinated, electrophilic methylene species. Mechanistic studies, including kinetic analyses, further support that the carbenoid transfers methylene in a concerted manner without dissociation.^[7]^[8] Several factors influence carbenoid formation efficiency. Solvent polarity affects solubility and coordination; nonpolar ethers like diethyl ether are traditionally used, though more polar options such as dichloromethane can accelerate the process by stabilizing ionic intermediates. Temperature control is vital, with formation typically occurring at 0 °C to room temperature to avoid decomposition, while higher temperatures (e.g., reflux in ether) may be employed for complete activation. Trace water must be rigorously excluded, as it leads to hydrolysis of the zinc couple and carbenoid quenching, reducing yields.^[1]^[6]

Cyclopropanation Step

The cyclopropanation step of the Simmons–Smith reaction involves the concerted addition of the organozinc carbenoid, such as iodomethylzinc iodide (ICH₂ZnI), to the π-bond of an alkene, resulting in the formation of a zinc-coordinated cyclopropane intermediate. This process is a cheletropic reaction in which the methylene unit is delivered simultaneously to both carbons of the double bond, establishing two new carbon-carbon bonds in a single, synchronous step. The reaction proceeds with complete syn stereochemistry, preserving the geometric configuration of the starting alkene, as cis-alkenes yield cis-cyclopropanes and trans-alkenes yield trans-cyclopropanes with 100% stereospecificity.^[9]^[10] This stereospecificity provides strong evidence for a concerted mechanism, as the absence of rearrangement or isomerization rules out the involvement of carbocation or diradical intermediates that could lead to stereorandomization. Early studies demonstrated this by showing that the reaction on (Z)- and (E)-2-butene substrates produced the corresponding cis- and trans-1,2-dimethylcyclopropanes without crossover products. The overall transformation can be represented as follows, where the initial adduct undergoes hydrolysis to liberate the free cyclopropane:

\text{Alkene} + \ce{ICH2ZnI} \rightarrow \text{cyclopropane-ZnI} \xrightarrow{\ce{H2O}} \text{cyclopropane} + \ce{ZnI2}

^[9]^[11] A notable feature enhancing the efficiency and selectivity of this step is the directing effect observed with allylic alcohols, where the zinc atom coordinates to the oxygen lone pair, orienting the carbenoid for syn delivery from the same face as the hydroxyl group. This coordination accelerates the reaction rate by up to several orders of magnitude compared to non-directed alkenes and ensures high diastereoselectivity in chiral substrates. Mechanistic investigations, including kinetic and structural analyses, confirm that this zinc-oxygen interaction stabilizes a chair-like transition state, facilitating the methylene transfer without altering the core concerted pathway.^[12]^[13]

Scope and Limitations

Substrate Scope

The Simmons–Smith reaction exhibits broad applicability to achiral simple alkenes, including terminal, internal, and cyclic variants. Terminal alkenes such as 1-hexene undergo efficient cyclopropanation to afford the corresponding monosubstituted cyclopropanes in high yields, typically exceeding 80%. Internal alkenes, like cis-2-butene, react stereospecifically to preserve the alkene geometry in the cyclopropane product. Cyclic alkenes, exemplified by norbornene, are particularly well-suited substrates, yielding exo-syn bicyclo[2.2.1]heptane derivatives in up to 95% yield due to the strained and accessible double bond. Electron-rich alkenes, such as styrenes and enol ethers, undergo the reaction readily owing to the electrophilic nature of the zinc carbenoid intermediate. For instance, styrene derivatives provide arylcyclopropanes in good yields, with the reaction proceeding faster than for unactivated alkenes. Enol ethers, like ethyl vinyl ether, form cyclopropyl ethers efficiently, highlighting the enhanced reactivity of electron-donating substituents.^[6] In polyenes, the reaction demonstrates selectivity for mono-cyclopropanation, particularly favoring the more electron-rich or less substituted double bond. Conjugated dienes, such as 1,3-butadiene, undergo regioselective addition at one terminal double bond to produce vinylcyclopropane derivatives in moderate to good yields (50–70%), avoiding over-functionalization of the conjugated system. This selectivity arises from the directing influence of electron density in the π-system. Allenes serve as effective substrates, leading to the formation of spiro[2.2]pentane derivatives through regioselective cyclopropanation at the central double bond. For example, allenamides react to yield amido-spiro[2.2]pentanes in yields up to 67%, with diastereoselectivity improving (up to 20:1) upon α-substitution on the allene, due to steric control over the carbenoid approach. This spirocyclopropanation preserves the orthogonal geometry of the allene.^[14] Limitations in substrate scope include sluggish reactivity with electron-poor alkenes, such as acrylates, which often require excess reagent or modified conditions for acceptable yields due to reduced nucleophilicity of the double bond. Trisubstituted alkenes can react but may suffer from steric hindrance, leading to variable yields compared to monosubstituted counterparts.^[6]

Functional Group Compatibility

The Simmons–Smith reaction demonstrates broad functional group tolerance under standard conditions, accommodating ethers, esters, ketones, and halides without significant interference or side reactions. These groups remain intact, allowing the cyclopropanation to proceed selectively at the alkene site, as the zinc carbenoid preferentially targets the π-bond over other functionalities.^[15] This compatibility stems from the mild, non-oxidative nature of the reagent, which avoids disrupting carbonyls or halogen substituents.^[16] A key feature is the directing effect of hydroxyl groups, particularly in allylic alcohols, where the oxygen coordinates to the zinc carbenoid, facilitating stereoselective delivery of the methylene unit to the same face as the OH (syn diastereoselectivity). This coordination enhances reaction rates and selectivity, often achieving diastereomeric ratios greater than 20:1. In homoallylic alcohols, coordination leads to predominant syn diastereoselectivity relative to the hydroxyl and cyclopropane ring, with yields typically exceeding 80% when the unprotected OH is present.^[16] While generally tolerant, free carboxylic acids and amines may interfere by protonating or coordinating to the zinc carbenoid in some cases, though successful examples exist without protection. Similarly, strong Lewis bases like thioethers and phosphines may complex the zinc, reducing efficiency and yields by competing with the substrate for coordination.^[3]

Modifications

Furukawa Modification

The Furukawa modification of the Simmons–Smith reaction, developed by Junji Furukawa, Naohiro Kawabata, and Jiro Nishimura in the late 1960s, utilizes diethylzinc (Et₂Zn) and diiodomethane (CH₂I₂) in the presence of an alkene to generate the iodomethylzinc ethyl carbenoid (ICH₂ZnEt) in situ for cyclopropanation.^[17] This approach marked a key improvement over the original zinc-copper couple method by enabling milder conditions and broader applicability. The carbenoid forms through the initial reaction of the reagents:

\ce{Et2Zn + CH2I2 -> ICH2ZnEt + EtI}

The resulting species then undergoes stereospecific syn addition to the alkene, yielding the cyclopropane product.^[17] This modification offers several advantages, including reactions at room temperature rather than reflux, shortened times from days to hours, and higher yields of 80–95% in many cases.^[6] It is particularly effective for allylic alcohols, where the hydroxyl group directs the carbenoid addition, enhancing regioselectivity and stereocontrol.^[6] The scope extends to electron-deficient alkenes to a moderate degree, allowing cyclopropanation of substrates challenging for the classic variant, and later adaptations enable catalytic use of zinc for efficiency.^[6] Despite these benefits, the procedure shows greater sensitivity to air and moisture, requiring rigorous inert atmosphere handling to prevent decomposition.^[6]

Achiral Zinc-Based Variants

Achiral zinc-based variants of the Simmons-Smith reaction extend the utility of the classical Zn/Cu couple or diethylzinc-mediated procedures by incorporating additives and optimized conditions to address limitations with challenging substrates, such as electron-deficient alkenes, while preserving the inherent syn stereospecificity of the methylene transfer. These modifications avoid chiral ligands, focusing instead on enhancements to carbenoid reactivity and generation. One key approach involves the use of Lewis acid additives to activate the zinc carbenoid or coordinate with electron-withdrawing groups on the alkene. For electron-poor alkenes like α,β-unsaturated carbonyls, the addition of trimethylsilyl chloride (TMSCl) or boron trifluoride diethyl etherate (BF₃·OEt₂) promotes efficient cyclopropanation by increasing the electrophilicity of the carbenoid or by complexing the carbonyl to enhance alkene nucleophilicity.^[6] Ultrasonication represents another effective variant, particularly for activating zinc insertion into diiodomethane and improving yields with recalcitrant substrates. This technique accelerates the formation of the iodomethylzinc iodide species at the metal surface, leading to faster reaction times and higher efficiency. Preformation of the iodomethylzinc iodide carbenoid prior to substrate addition offers precise control over reagent stoichiometry and minimizes side reactions in sensitive systems. This variant, often conducted in ether at low temperature, allows the carbenoid to be isolated or generated cleanly before introducing the alkene, which is advantageous for scale-up and substrates prone to over-cyclopropanation.^[9] Solvent effects also play a crucial role in these achiral variants, with 1,2-dimethoxyethane (DME) providing superior solubility for the zinc species compared to diethyl ether or dichloromethane, thereby enhancing reaction rates and yields for sterically hindered or polar alkenes. These variants collectively expand the substrate scope to include electron-deficient systems like α,β-unsaturated esters and ketones. Unlike the Furukawa modification, which primarily improves homogeneity via Et₂Zn, these tweaks emphasize activation strategies for difficult alkenes. The advantages include retained syn addition stereochemistry—delivering cis-cyclopropanes from (Z)-alkenes with >95:5 selectivity—and operational simplicity without the need for chiral auxiliaries, making them ideal for achiral natural product fragments and pharmaceutical intermediates.^[6]

Asymmetric Variants

Charette Modification

The Charette modification represents a landmark advancement in asymmetric Simmons-Smith cyclopropanation, developed by André B. Charette and colleagues in the mid-1990s. This variant builds on the Furukawa conditions, employing diethylzinc and diiodomethane to generate bis(iodomethyl)zinc in situ, but incorporates chiral amino alcohol ligands—such as derivatives of (S)-prolinol—to enable ligand-controlled stereoselectivity. These ligands, often in the form of chiral dioxaborolanes, coordinate to the zinc carbenoid, directing the reaction toward high enantiomeric excess without requiring covalent attachment to the substrate. The method was first reported in 1994, demonstrating efficient enantioselective cyclopropanation of allylic alcohols under mild conditions, typically at room temperature in dichloromethane, with ligand loadings of 110-120 mol%.^[18] In the mechanism, the chiral ligand binds to the zinc center of the iodomethylzinc species, forming a chiral zinc-alkoxide complex with the allylic alcohol substrate through directed coordination of the hydroxyl group. This coordination induces facial selectivity during the carbenoid addition to the alkene, favoring syn delivery of the methylene unit relative to the hydroxyl. Computational studies support a concerted, chair-like transition state where the ligand's steric bulk shields one face of the alkene, leading to predictable stereocontrol and enantioselectivities up to 99% ee for allylic alcohols. The ligand can be recovered post-reaction via aqueous extraction, enhancing practicality.^[19] The general reaction scheme mirrors the Furukawa modification but includes the chiral ligand (L*) for asymmetry:

\text{Allylic alcohol} + \ce{CH2I2} + \ce{Et2Zn} \xrightarrow{\text{L* (110 mol\%)}} \text{enantiopure cyclopropane} + \ce{EtI} + \ce{ZnI2}

This produces the corresponding cyclopropylmethanol with high enantiopurity, typically as the (R)-enantiomer when using (S)-prolinol-derived ligands. The scope of the Charette modification is particularly suited to electron-rich alkenes, such as those in allylic alcohols, where the directing hydroxyl group enhances reactivity and selectivity; it performs best with substrates bearing primary or secondary alcohols, delivering yields of 80-95% and diastereomeric excesses (de) exceeding 95% in most cases. Functional group compatibility includes tolerance for ethers, esters, and protected amines, though electron-withdrawing groups may reduce efficiency. A key example is the enantioselective cyclopropanation of cinnamyl alcohol, affording the (R)-cyclopropane product in >98% yield and 93% ee, showcasing the method's utility for aryl-substituted systems. This transformation has been pivotal in synthesizing chiral building blocks for complex molecules.^[19]

Shi Modification

The Shi modification of the Simmons-Smith reaction, developed by the group of Yian Shi, integrates the Furukawa variant (using Et₂Zn and CH₂I₂) with chiral dipeptide ligands at low catalytic loadings of 10–20 mol% to coordinate the zinc center and induce asymmetry. This approach addresses the limitations of traditional Simmons-Smith methods for unfunctionalized olefins, providing high enantioselectivities, typically in the range of 80–95% ee, while maintaining excellent diastereocontrol in syn addition to the alkene.^[20]^[21] Chiral ligands in this modification are often tartrate-derived esters or bisoxazoline derivatives, but primarily dipeptides such as N-Boc-L-Val-L-Pro-OMe. These ligands form a chiral pocket around the carbenoid, facilitating stereoselective methylene transfer in a concerted fashion. This variant contrasts with ligand-directed approaches like Charette's by enabling asymmetry for non-directed, unfunctionalized substrates, broadening the scope to include simple alkenes without requiring proximal coordinating groups. A representative example is the asymmetric cyclopropanation of trans-β-methylstyrene, where the Shi protocol delivers the corresponding cyclopropane product in 85% yield with 91% ee, demonstrating the method's utility for styrenes and other unfunctionalized systems.^[20]

Applications in Synthesis

Natural Product Syntheses

The Simmons–Smith reaction plays a pivotal role in natural product total synthesis by introducing cyclopropane rings that confer structural rigidity and enhance bioactivity, such as in pheromones, terpenoids, and alkaloids.^[3] These strained motifs often require stereocontrol, where asymmetric variants like the Charette modification are employed to access enantiopure intermediates.^[22] The reaction's mild conditions and functional group tolerance make it ideal for late-stage installations in complex scaffolds derived from biological sources. A notable application is the 2021 total synthesis of peyssonnoside A, a diterpenoid glycoside isolated from the red alga Peyssonnelia sp., using a modified Simmons–Smith reaction with iodoform (CHI₃) and diethylzinc. This step constructed the key three-membered ring with precise stereocontrol, yielding a single isomer in 72% yield and enabling completion of the synthesis in 15 steps from a known precursor.^[23] Similarly, in the 2016 synthesis of (+)-omphadiol, a sesquiterpene from the liverwort Omphalanthus filiformis, the reaction performed stereoselective cyclopropanation on the C2–C4 double bond, achieving 70% overall yield for that fragment and facilitating the assembly of the tricyclic core.^[24] More recently, the Furukawa variant was crucial in the 2015 total synthesis of (-)-lundurine A, an indole alkaloid from the plant Lundia cuneifolia, where it installed the cyclopropane to dictate the C2 and C7 stereocenters in 63% yield, supporting the formation of fused six- and seven-membered rings essential for the natural product's architecture.^[25] These examples illustrate the reaction's impact in enabling stereocontrolled access to strained cyclopropane motifs in terpenes and alkaloids, advancing the synthesis of bioactive marine and plant-derived compounds.^[3]

Pharmaceutical Syntheses

The incorporation of cyclopropane rings into pharmaceutical agents often improves metabolic stability by blocking susceptible sites to oxidative metabolism and enhances target binding affinity through rigidification of molecular conformations or optimal geometric presentation of pharmacophores.^[26] A prominent application is in the commercial-scale synthesis of saxagliptin (Onglyza), a dipeptidyl peptidase-4 (DPP-4) inhibitor for type 2 diabetes management, where a diastereoselective Simmons-Smith cyclopropanation constructs the critical aminocyclopropane core from a dihydropyrrole precursor. This step employs diethylzinc and diiodomethane with additives such as ZnI₂ or trifluoroacetic acid to achieve >99:1 diastereomeric ratio (dr) and 70% yield on decagram scales, enabling efficient progression to the bicyclic proline derivative.^[27] The Charette modification of the Simmons-Smith reaction has been utilized in the synthesis of the cyclopropane core for GSK-1360707F, a triple reuptake inhibitor developed as an antipsychotic agent, facilitating access to the fused-ring system in an eight-step racemic route.^[28] In medicinal chemistry, Simmons-Smith cyclopropanation of allenamides affords amido-spiro[2.2]pentanes with moderate to good diastereoselectivity, depending on substrate geometry and reaction conditions.^[29]^[3] The Furukawa-modified Simmons-Smith reaction is also applied to generate γ-keto esters from β-keto esters, yielding versatile precursors for pharmaceutical intermediates that incorporate cyclopropane motifs to modulate lipophilicity and stability.^[30] In pharmaceutical contexts, these reactions are frequently adapted for scale-up, delivering key steps in 70-90% yields while maintaining functional group tolerance for downstream elaboration.^[27]