Copper(I) Trifluoromethanesulfonate


[42152-44-3]  · CCuF3O3S  · Copper(I) Trifluoromethanesulfonate  · (MW 212.62) (2:1 benzene complex)

[37234-97-2; 42152-46-5]  · C8H6Cu2F6O6S2  · Copper(I) Trifluoromethanesulfonate  · (MW 503.34)

(efficient catalyst for 2p + 2p photocycloadditions and other photoreactions of alkenes,1 and for alkene cyclopropanation and other reactions of diazo compounds; also a selenophilic and thiophilic Lewis acid that enhances the nucleofugacity of selenide and sulfide leaving groups)

Alternate Name: copper(I) triflate.

Physical Data: moisture-sensitive white crystalline solid.

Solubility: sol MeCN, MeCO2H, 2-butanone, alkenes; slightly sol benzene.

Form Supplied in: must be freshly prepared.

Preparative Methods: copper(I) trifluoromethanesulfonate (CuOTf) was first prepared as a solution in acetonitrile by synproportionation of Copper(II) Trifluoromethanesulfonate with copper(0).2 The CuI in these solutions is strongly coordinated with acetonitrile, forming complexes analogous to Tetrakis(acetonitrile)copper(I) Perchlorate.3 A white crystalline solid benzene complex, (CuOTf)2.C6H6, is prepared by the reaction of a suspension of Copper(I) Oxide in benzene with Trifluoromethanesulfonic Anhydride.4 Traces of Trifluoromethanesulfonic Acid apparently catalyze the reaction.5 CuOTf is generated in situ by the reduction of Cu(OTf)2 with diazo compounds.6

Handling, Storage, and Precautions: moisture sensitive.

Cyclopropanation with Diazo Compounds.

Copper(I) triflate is a highly active catalyst for the cyclopropanation of alkenes with diazo compounds.6 In contrast to other more extensively ligated copper catalysts, e.g. Copper(II) Acetylacetonate, that favor cyclopropanation of the most highly substituted C=C bond, cyclopropanations catalyzed by CuOTf show a unique selectivity for cyclopropanation of the least alkylated C=C bond in both intermolecular (eq 1) and intramolecular (eq 2) competitions. The same selectivity is found with Cu(OTf)2 as nominal catalyst. This is because Cu(OTf)2 is reduced by the diazo compound to CuOTf, and CuOTf is the actual cyclopropanation catalyst in both cases.6 Selective cyclopropanation of the least substituted C=C bond is a consequence of the alkene coordinating with the catalyst prior to interaction with the diazo compound, and the increase in stability of CuI-alkene complexes with decreasing alkyl substitution on the C=C bond. For catalysts with more strongly ligated CuI, an electrophilic carbene or carbenoid intermediate reacts with the free alkene, and the preference for cyclopropanation of the more highly substituted C=C bond arises from the enhancement of alkene nucleophilicity with increasing alkyl substitution (see Copper(II) Trifluoromethanesulfonate).

Cyclopropanecarboxylic esters are conveniently available, even from volatile alkenes, because CuOTf promotes cyclopropanations in good yields at low temperatures. Thus trans- and cis-2-butenes, boiling under reflux, react stereospecifically with Ethyl Diazoacetate to produce the corresponding ethyl 2,3-dimethylcyclopropanecarboxylates (eqs 3 and 4),6 and cyclobutene reacts with ethyl diazoacetate at 0 °C to deliver a mixture of exo- and endo-5-ethoxycarbonylbicyclo[2.1.0]pentanes (eq 5).7

CuOTf is an outstandingly effective catalyst for the synthesis of cyclopropyl phosphonates by the reaction of Diethyl Diazomethylphosphonate with alkenes (eq 6).8 The resulting cyclopropylphosphonates are useful intermediates for the synthesis of alkylidenecyclopropanes by Wadsworth-Emmons alkenation with aromatic carbonyl compounds (eq 7).8

A complex of a chiral, nonracemic bis(oxazoline) with CuOTf is a highly effective catalyst for asymmetric cyclopropanation of alkenes.9 Copper(II) triflate complexes do not catalyze the reaction unless they are first converted to CuI by reduction with a diazo compound or with phenylhydrazine. CuOTf complexes are uniquely effective. Thus the observed enantioselectivity and catalytic activity, if any, are much lower with other CuI or CuII salts including halide, cyanide, acetate, and even perchlorate. Both enantiomers of the bis(oxazoline) ligand are readily available. Spectacularly high levels of asymmetric induction are achieved with both mono- (eq 8) and 1,1-disubstituted alkenes (eq 9).

Asymmetric Aziridination.

A chiral, nonracemic bis(oxazoline) complex of copper(I) triflate catalyzes asymmetric aziridination of styrene in good yield (eq 10).9 However, enantioselectivity is not as high as the corresponding cyclopropanation (eq 8).


CuOTf is an exceptionally effective catalyst for 2p + 2p photocycloadditions of alkenes.1 Thus while CuBr promotes photodimerization of norbornene in only 38% yield,10 the same reaction affords dimer in 88% yield with CuOTf as catalyst (eq 11).11 A mechanistic study of this reaction revealed that although both 1:1 and 2:1 alkene CuI complexes are in equilibrium with free alkene and both the 1:1 and 2:1 complexes absorb UV light, only light absorbed by the 2:1 complex results in photodimerization. In other words, photodimerization requires precoordination of both C=C bonds with the CuI catalyst. Thus the exceptional ability of CuOTf, with its weakly coordinating triflate counter anion, to form p-complexes with as many as four C=C bonds12 is of paramount importance for its effectiveness as a photodimerization catalyst.

The importance of precoordination is also evident in the CuOTf-promoted 2p + 2p photocycloaddition of endo-dicyclopentadiene. This diene forms an isolable 2:1 complex with CuOTf involving exo-monodentate coordination with the 8,9-C=C bond of two molecules of diene. Consequently, intermolecular 2p + 2p photocycloaddition involving exo addition to the 8,9-C=C bond is strongly favored over intramolecular reaction between the 8,9- and 3,4-C=C bonds (eq 11).11 This contrasts with the intramolecular photocycloaddition that is promoted by high energy triplet sensitizers.12

Especially interesting is the trans,anti,trans stereochemistry of the major cyclobutane product generated in the photodimerization of cyclohexene (eq 12).13 It was noted that the formation of this product may be the result of a preliminary CuOTf promoted cis-trans photoisomerization that generates a trans-cyclohexene intermediate (eq 13).13 Since one face of the trans C=C bond is shielded by a polymethylene chain, the trans-cyclohexene is restricted to suprafacial additions. Although a highly strained trans-cyclohexene intermediate could be stabilized by coordination with CuI, such a complex has not been isolated.

An isolable CuOTf complex of a highly strained alkene, trans-cycloheptene, is produced by UV irradiation of a hexane solution of cis-cycloheptene in the presence of CuOTf (eq 14).14 Photocycloaddition of cycloheptene is also catalyzed by CuOTf. Surprisingly, the major product is not a trans,anti,trans dimer analogous to that formed from cyclohexene (eq 12) but rather a trans,anti,trans,anti,trans trimer (eq 15).15

Dissolution of the trans-cycloheptene-CuOTf complex in cycloheptene and evaporation of the solvent delivers a tris alkene complex of CuOTf containing one trans-cycloheptene and two cis-cycloheptene ligands. Heating trans-cycloheptene-CuOTf in neat cis-cycloheptene delivers the trans,anti,trans,anti,trans trimer (eq 16). Experiments with cis-cycloheptene-d4 show that the cyclotrimerization involves only trans-cycloheptene molecules, although the reaction is accelerated by the presence of cis-cycloheptene.16 A likely explanation for these observations is concerted "template" cyclotrimerization of a tris-trans-cycloheptene-CuOTf complex formed by ligand redistribution (eq 16).16

The involvement of a transient photogenerated trans-cyclohexene-CuOTf intermediate was also adduced to explain CuOTf catalysis of photoinduced 2p + 4p cycloaddition between cis-cyclohexene and 1,3-butadiene (eq 17).17 In contrast to thermal Diels-Alder reactions, this reaction generates trans-D2-octalin rather than the cis cycloadduct expected for a 2ps + 4ps cycloaddition. A mechanism was proposed that involves the 2ps + 4ps cycloaddition of a trans-cyclohexene with 1,3-butadiene in the coordination sphere of CuI (eq 18).17

That CuOTf-catalyzed 2p + 2p photocycloadditions are not restricted to cyclic alkenes was first demonstrated in mixed cycloadditions involving allyl alcohol. To suppress homodimerization of endo-dicyclopentadiene (i.e. eq 11) the diene to CuI ratio is maintained at <=1:1 and allyl alcohol is used as solvent. Under these conditions, a high yield of mixed cycloadduct is generated (eq 19).18

That both C=C bonds participating in 2p + 2p photocycloadditions can be acyclic is evident from the photobicyclization reactions of simple diallyl ethers that deliver bicyclic tetrahydrofurans (eq 20).19,20 In conjunction with Ruthenium(VIII) Oxide-catalyzed oxidation by Sodium Periodate, these CuOTf-catalyzed photobicyclizations provide a synthetic route to butyrolactones from diallyl ethers (eq 20).20 The synthetic method is applicable to the construction of multicyclic tetrahydrofurans and butyrolactones from diallyl ethers (eqs 21 and 22) as well as from homoallyl vinyl ethers (eq 23).20

3-Oxabicyclo[3.2.0]heptanes are also produced in the CuOTf-catalyzed photocycloadditions of allyl 2,4-hexadienyl ethers (eq 24).21 The CuOTf-catalyzed photocycloadditions of bis-2,4-hexadienyl ethers are more complex. Thus UV irradiation of 5,5-oxybis[(E)-1,3-pentadiene] in THF for 120 h produces vinylcyclohexene and tricyclo[,6]octane derivatives (eq 25).22 However, shorter irradiations reveal that these products arise by secondary CuOTf-catalyzed rearrangements of 6,7-divinyl-3-oxabicyclo[3.2.0]heptanes that are the primary photoproducts (eq 26). UV irradiation of the divinylcyclobutane intermediates in the presence of CuOTf promotes formal [1,3]- and [3,3]-sigmatropic rearrangements to produce a vinylcyclohexene and a 1,5-cyclooctadiene that is the immediate precursor of the tricyclo[,6]octane.

CuOTf-catalyzed photobicyclization of 1,6-heptadien-3-ols produces bicyclo[3.2.0]heptan-2-ols (eq 27).23 In conjunction with pyrolytic fragmentation of the derived ketones, these CuOTf-catalyzed photobicyclizations provide a synthetic route to 2-cyclopenten-1-ones from 1,6-heptadien-3-ols (eq 28).23 The derived ketones can also be converted into lactones by Baeyer-Villiger oxidation and, in conjunction with pyrolytic fragmentation, CuOTf-catalyzed photobicyclizations provide a synthetic route to enol lactones of glutaraldehydic acid from 1,6-heptadien-3-ols (eq 28).23

Copper(I) triflate-catalyzed photobicyclization of b- and g-(4-pentenyl)allyl alcohols provides a synthetic route to various multicyclic carbon networks in excellent yields (eqs 29-31).24 The reaction was exploited in a total synthesis of the panasinsene sesquiterpenes (eq 32).25 It is especially noteworthy in this regard that attempted synthesis of a key tricyclic ketone intermediate for the panasinsenes by the well-known photocycloaddition of Isobutene to an enone failed to provide any of the requisite cyclobutyl ketone (eq 33).25

In conjunction with carbocationic skeletal rearrangement, photobicyclization of 1,6-heptadien-3-ols provides a synthetic route to 7-hydroxynorbornanes (eq 34).26 Noteworthy is the stereoselective generation of exo-1,2-polymethylenenorbornanes from either the exo or endo epimer of 2,3-polymethylenebicyclo[3.2.0]heptan-3-ol.

N,N-Diallylamides are recovered unchanged when irradiated in the presence of CuOTf.27 This is because the amide chromophore interferes with photoactivation of the CuI-alkene complex. Thus CuOTf-alkene complexes containing one, two, three, or even four coordinated C=C bonds exhibit UV absorption at 235 ± 5 nm (εmax 2950 ± 450).12 The CuOTf complex of ethyl N,N-diallylcarbamate exhibits lmax = 233.4 nm (εmax 2676) but the free ligand is virtually transparent at this wavelength. Consequently, UV irradiation of ethyl N,N-diallylcarbamates in the presence of CuOTf delivers bicyclic (eq 35) or tricyclic (eq 36) pyrrolidines incorporating the 3-azabicyclo[3.2.0]heptane ring system.27

Catalyzed Diels-Alder Reactions.

The uncatalyzed thermal intramolecular Diels-Alder reaction of 5,5-oxybis[(E)-1,3-pentadiene] nonstereoselectively generates four isomeric 4-vinylcyclohexenes (eq 37). The major product has a trans ring fusion, in contrast to the single cis ring-fused isomer generated in the copper(I) triflate-catalyzed photoreaction of the same tetraene (eq 25). Copper(I) triflate also catalyzes a thermal Diels-Alder reaction of 5,5-oxybis[(E)-1,3-pentadiene] that proceeds under milder conditions than the uncatalyzed reaction. The stereoselectivity is remarkably enhanced, generating mainly the major isomer of the uncatalyzed thermal reaction and a single cis-fused isomer (eq 37) that is different than the one favored in the photochemical reaction (eq 25).

Csp-H Bond Activation.

Hydrogen-deuterium exchange between terminal alkynes and CD3CO2D is catalyzed by CuOTf (eq 38).28 Proton NMR studies revealed that CuOTf and alkynes form p-complexes that rapidly exchange coordinated with free alkyne. A complex of CuOTf with 1,7-octadiene was isolated (eq 39). The complex rapidly exchanges terminal alkynic hydrogen with deuterium from CD3CO2D and undergoes a much slower conversion to a copper alkynide (eq 39).28 Exchange of alkynic hydrogen and deuterium is also catalyzed by CuOTf (eq 40).28

Activation of Aryl Halides.

Ullmann coupling of o-bromonitrobenzene is accomplished under exceptionally mild conditions and in homogeneous solution by reaction with copper(I) triflate in the presence of aqueous NH3 (eq 41).29 Yields are enhanced by the presence of a small quantity of copper(II) triflate. That the reaction is diverted to reductive dehalogenation by Ammonium Tetrafluoroborate is presumptive evidence for an organocopper intermediate that can be captured by protonation.

Biaryl is only a minor product from the reaction of methyl o-bromobenzoate with CuOTf (eq 42). The major product can result from replacement of the halide by NH2, H, or OH, depending on reaction conditions. In the presence of 5% aqueous NH3, methyl anthranilate is the major product.30 More concentrated aqueous NH3 (20%) favors the generation of methyl salicylate, and the yield of this product is enhanced by the presence of a substantial quantity of CuII ion.29 Reductive dehalogenation is favored by the presence of ammonium ions, presumably owing to protonolysis of an arylcopper(III) intermediate.29

Activation of Vinyl Halides.

Under the optimum conditions for reductive coupling of o-bromonitrobenzene (eq 41), diethyl iodofumarate gives very little coupling product; the overwhelming product was diethyl fumarate generated by hydrodehalogenation (eq 43).29 Reductive coupling delivers trans,trans-1,2,3,4-tetraethoxycarbonyl-1,3-butadiene in 95% yield (GLC, or 80% of pure crystalline product) in 2 h if aqueous NH3 is replaced by anhydrous NH3.29 Under the same conditions, diethyl iodomaleate undergoes 45% conversion in 20 h to deliver diethyl maleate, as well as minor amounts of cis,cis- and trans,trans-tetraethoxycarbonyl-1,3-butadiene (eq 44).29 The stereospecificity of the reductive dehalogenations in eqs 43 and 44 is presumptive evidence for the noninvolvement of radicals in these reactions.

Elimination of Thiophenol from Thioacetals.

Conversion of thioacetals to vinyl sulfides is accomplished under exceptionally mild conditions by treatment with (CuOTf)2.C6H6 (eq 45).31 The reaction involves an a-phenylthio carbocation intermediate. Three factors contribute to the effectiveness of this synthetic method: the Lewis acidity of a copper(I) cation that is unencumbered by a strongly coordinated counter anion, the solubility of the copper(I) triflate-benzene complex, and the insolubility of CuSPh in the reaction mixture. An analogous elimination reaction provides an effective route to phenylthio enol ethers from ketones (eq 46).31

This conversion of thioacetals into vinyl sulfides was applied to a C-C connective synthesis of 2-phenylthio-1,3-butadienes from aldehydes (eq 47).32 The key elimination step converts cyclobutanone thioacetal intermediates into 1-phenylthiocyclobutenes that undergo electrocyclic ring opening to deliver dienes.

A different synthesis generates 2-phenylthio-1,3-butadienes directly by elimination of two molecules of thiophenol from b-phenylthio thioacetals that are readily available from the corresponding a,b-unsaturated ketones (eq 48).33

CuOTf-promoted elimination of thiophenol was exploited in two syntheses of 1-phenylthio-1,3-butadiene, one a C-C connective route from allyl bromide31 and bis(phenylthio)methyllithium,34 and another from Crotonaldehyde (eq 49).33 A topologically analogous C-C connective strategy provides 2-methoxy-1-phenylthio-1,3-butadiene from acrolein (eq 50).5,33 That the phenylthio rather than the methoxy substituent in 2-methoxy-1-phenylthio-1,3-butadiene controls the orientation of its Diels-Alder cycloadditions is noteworthy (eq 50).

A synthesis of 4-alkyl-2-methoxy-1-phenylthio-1,3-butadienes by a simple b-elimination of thiophenol from a thioacetal is not possible owing to skeletal rearrangement that is fostered by stabilization of a cyclopropylcarbinyl carbocation intermediate by the alkyl substituent (eq 51).35 Interconversion of an initial a-phenylthio carbocation to a more stable a-methoxy carbocation intermediate leads to the generation of a 4-alkyl-1-methoxy-2-phenylthio-1,3-butadiene instead.

Syntheses of 1-phenylthio-1,3-butadienes from carboxylic esters (eq 52) and carboxylic acids (eq 53) are achieved by CuOTf-promoted elimination of thiophenol from intermediate thioacetals.36

Heterocyclization of g-Keto Dithioacetals.

A C-C connective synthesis of furans is completed by a CuOTf-promoted heterocyclization of g-keto thioacetals (eq 54).31 Rather than simple b-elimination to generate a vinyl sulfide (eq 46), a presumed g-keto carbocation intermediate is captured intramolecularly by an intimately juxtaposed carbonyl oxygen nucleophile.

Friedel-Crafts Alkylation of Arenes with Thioacetals.

(CuOTf)2.C6H6 promotes a-thioalkylation of anisole by a dithioacetal under mild conditions (eq 55).37

Elimination of Benzylic Phenyl Thioethers.

That C-S bond activation by CuOTf is not limited to substrates that can generate sulfur-stabilized carbocation intermediates is illustrated by a C-C connective synthesis of trans-stilbene (eq 56).31 The elimination of thiophenol under mild conditions is favored by benzylic stabilization of a carbocation intermediate or an E2 transition state with substantial carbocationic character.

Hydrolysis of Vinylogous Thioacetals.

Carbanions prepared by lithiation of g-phenylthioallyl phenyl thioethers can serve as synthetic equivalents of b-acyl vinyl anions.38 Umpölung of the usual electrophilic reactivity of 2-cyclohexenone is achieved by a sequence exploiting electrophilic capture of a lithiated vinylogous thioacetal and subsequent CuOTf-assisted hydrolysis (eq 57).39 Otherwise unfunctionalized vinylogous thioacetals can be hydrolyzed to enones by Mercury(II) Chloride in wet acetonitrile.38 However, the keto-substituted derivative in eq 57 gave only a 25% yield of enone by this method. A superior yield was obtained by CuOTf-assisted hydrolysis.39

Grob Fragmentation of b-[Bis(phenylthio)methyl]alkoxides.

A method for achieving Grob-type fragmentation of five- and six-membered rings depends upon the ability of a thiophenyl group to both stabilize a carbanion and serve as an anionic leaving group. For example, reaction of cyclohexene oxide with lithium bis(phenylthio)methide34 produces a b-[bis(phenylthio)methyl]alkanol that undergoes fragmentation in excellent yield upon treatment with n-Butyllithium followed by CuOTf (eq 58).40 Copper(I) trifluoroacetate is equally effective but salts of other thiophilic metals, e.g. mercury or silver, were ineffective. Treatment of the intermediate b-[bis(phenylthio)methyl]alkanol with CuOTf in the absence of added strong base leads primarily to elimination of thiophenol as expected (see eq 45). Fragmentation does not occur with only one equivalent of CuOTf. This suggests a key intermediate with at least one CuI ion to coordinate with the alkoxide and another to activate the phenylthio leaving group (eq 58).

Ring-Expanding Rearrangements of a-[Bis(phenylthio)methyl]alkanols.

A one-carbon ring-expanding synthesis of a-phenylsulfenyl ketones from homologous ketones depends upon the ability of a thiophenyl group to both stabilize a carbanion and serve as an anionic leaving group. For example, reaction of cyclopentanone with lithium bis(phenylthio)methide34 produces an a-[bis(phenylthio)methyl]alkanol that rearranges to a ring-expanded a-phenylsulfenyl ketone in good yield upon treatment with CuOTf in the presence of Diisopropylethylamine (eq 59).41 Epoxy thioether intermediates are generated from the a-[bis(phenylthio)methyl]alkanols by intramolecular nucleophilic displacement of thiophenoxide.

An analogous synthesis of a,a-bis(methylsulfenyl) ketones from homologous ketones by one-carbon ring expansion depends on copper(I)-promoted rearrangement of an a-tris(methylthio)methyl alkoxide intermediate (eq 60). Both Tetrakis(acetonitrile)copper(I) Perchlorate42 and Tetrakis(acetonitrile)copper(I) Tetrafluoroborate43 are effective in promoting the rearrangement but (CuOTf)2.C6H6, HgCl2, or Hg(TFA)2 are not. Apparently, the MeCN ligand is crucial. Furthermore, treatment of the intermediate a-[tris(methylthio)methyl] alcohol with CuOTf and EtN(i-Pr)2 in toluene followed by aqueous workup delivers an a-hydroxy methylthio ester (eq 61),43 in contrast to the ring-expanding rearrangement of the analogous a-[bis(phenylthio)methyl]alkanol (eq 59).41

The a-[bis(phenylthio)methyl]alkanol derived from cycloheptanone does not undergo ring expansion upon treatment with CuOTf and EtN(i-Pr)2 in benzene. Instead, 1,3-elimination of thiophenol delivers an epoxy thioether intermediate that undergoes a rearrangement involving 1,2-shift of a phenylsulfenyl group to produce an a-phenylsulfenyl aldehyde (eq 62).41

The a-[bis(phenylthio)methyl]alkanol derived from cyclohexanone, upon treatment with CuOTf and EtN(i-Pr)2 in benzene, undergoes both ring-expanding rearrangement to deliver a-phenylsulfenylcycloheptanone as the major product, as well as rearrangement involving 1,2-shift of a phenylsulfenyl group to produce an a-phenylsulfenylcyclohexanecarbaldehyde (eq 63).41 In contrast, neither ring expansion nor 1,2-shift of a methylsulfenyl group occurs upon treatment of a-[tris(methylthio)methyl]cyclohexanol with n-butyllithium followed by (MeCN)4CuBF4. Rather, after aqueous workup, an a-hydroxy methylthio ester is obtained (eq 64).43

Chain-Extending Syntheses of a-Phenylsulfenyl Ketones.

A C-C connective, chain-extending synthesis of a-phenylsulfenyl ketones from aldehydes (eq 65) or acyclic ketones (eq 66)41 can be accomplished by a CuOTf-promoted activation of the a-[bis(phenylthio)methyl]alkanols generated by addition of lithium bis(phenylthio)methide.34 Preferential migration of hydride generates phenylsulfenylmethyl ketones from aldehydes (eq 65). Regioselective insertion of a phenylsulfenylmethylene unit occurs owing to a preference for migration of the more highly substituted alkyl group of dialkyl ketones (eq 66).

Cyclopropanation of Enones.

Conjugate addition of lithium tris(phenylthio)methide44 to a,b-unsaturated ketones produces enolates that cyclize to bis(phenylthio)cyclopropyl ketones at -78 °C upon treatment with nearly one equivalent of (CuOTf)2.C6H6, i.e. 1.9 equivalents of CuI (eq 67).45 The mild conditions that suffice to bring about nucleophilic displacement of thiophenoxide in the presence of CuOTf are especially noteworthy. In view of the requirement for more than one equivalent of CuI to achieve Grob-type fragmentation of b-[bis(phenylthio)methyl]alkoxides (see eq 58), it seems likely, although as yet unproven, that one equivalent of CuI coordinates strongly with the enolate oxygen and that a second equivalent of CuI is required to activate the thiophenoxide leaving group.

Vinylcyclopropanation of Enones.

Conjugate addition of sulfur-stabilized allyl carbanions to a,b-unsaturated ketones produces enolates that cyclize to vinylcyclopropyl ketones upon treatment with nearly one equivalent of (CuOTf)2.C6H6, i.e. 1.9 equivalents of CuI (eq 68).45

Friedel-Crafts Acylation with Thio- or Selenoesters.

Methylseleno esters are readily available in excellent yields by the reaction of Dimethylaluminum Methylselenolate with O-alkyl esters.37 These selenoesters will acylate reactive arenes (eq 69) and heterocyclic compounds (eq 70) when activated by CuOTf, a selenophilic Lewis acid.37 Of the potential activating metal salts tested, (CuOTf)2.C6H6 is uniquely effective. Mercury(II) or copper(I) trifluoroacetates that are partially organic-soluble, as well as the corresponding chlorides, silver nitrate, and copper(I) oxide that are not organic-soluble, all failed to promote any acylation. The highly reactive CuOTf-benzene complex, in dramatic contrast, was found to readily promote the acylations in benzene solution within minutes at room temperature. The presence of vinyl and keto groups is tolerated by the reaction, and while the alkyl- and vinyl-substituted derivatives afford para substitution only, a 2:1 mixture of para and ortho substitution occurs with the methylseleno ester of levulinic acid. Acylation of toluene is sluggish. Excellent yields of 2-acylfurans, -thiophenes, and -pyrroles are generated by this new variant of the Friedel-Crafts acylation reaction (eq 70). An intramolecular version of this reaction was shown to generate 1-tetralone from the methylseleno ester of g-phenylbutyric acid (eq 71).37

Notwithstanding a prior claim that methylthio esters react only sluggishly under these conditions,37 such a variant proved effective for a short synthesis of the 4-demethoxy-11-deoxyanthracycline skeleton (eq 72).46 This is especially significant because methylthio esters are available by an efficient C-C connective process involving C-acylation of ketone lithium enolates with Carbon Oxysulfide (COS) followed by S-methylation with Iodomethane.46 For the deoxyanthracycline synthesis, the requisite enolate was generated by 1,4-addition of a silyl-stabilized benzyllithium derivative to 2-cyclohexenone. Treatment of the methylseleno ester with (CuOTf)2.C6H6 in benzene, according to the method employed with analogous seleno esters,37 results in efficient cyclization to deliver a tetracyclic diketone in good yield.

O-Acylation with Thioesters.

Activation of a thioester with (CuOTf)2.C6H6 was exploited as a key step in the synthesis of a macrocyclic pyrrolizidine alkaloid ester (eq 73).47 Since thioesters are relatively unreactive acylating agents, a highly functionalized imidazolide containing acetate and t-butyl thioester groups selectively acylated only the primary hydroxyl in the presence of the secondary hydroxyl group in (+)-retronecine. Completion of the synthesis required activation of the t-butylthio ester. Mercury(II) Trifluoroacetate, that had proven effective for the synthesis of several natural products by lactonization,48,49 failed to promote any lactonization in the present case.47 Similarly, Mercury(II) Chloride and Cadmium Chloride, that have proven effective for promoting lactonizations,49 had no effect in the present case. Even copper(I) trifluoroacetate failed to induce the crucial lactonization. In contrast, CuOTf was uniquely effective for inducing the requisite macrolactonization by activating the thioester.

1. (a) Salomon, R. G. Adv. Chem. Ser. 1978, 168, 174. (b) Salomon, R. G. T 1983, 39, 485. (c) Salomon, R. G.; Kochi, J. K. TL 1973, 2529.
2. Jenkins, C. L.; Kochi, J. K. JACS 1972, 94, 843.
3. (a) Hathaway, B. J.; Holah, D. G.; Postlethwaite, J. D. JCS 1961, 3215. (b) Kubota, M.; Johnson, D. L. J. Inorg. Nucl. Chem. 1967, 29, 769.
4. (a) Salomon, R. G.; Kochi, J. K. CC 1972, 559. (b) Salomon, R. G.; Kochi, J. K. JACS 1973, 95, 1889. (c) Dines, M. B. Separ. Sci. 1973, 8, 661.
5. Cohen, T.; Ruffner, R. J.; Shull, D. W.; Fogel, E. R.; Falck, J. R. OS 1980, 59, 202; OSC 1988, 6, 737.
6. Salomon, R. G.; Kochi, J. K. JACS 1973, 95, 3300.
7. Wiberg, K. B.; Kass, S. R.; Bishop, III, K. C. JACS 1985, 107, 996.
8. Lewis, R. T.; Motherwell, W. B. TL 1988, 29, 5033.
9. (a) Evans, D. A.; Woerpel, K. A.; Hinman, M. M.; Faul, M. M. JACS 1991, 113, 726. (b) Evans, D. A.; Woerpel, K. A.; Scott, M. J. AG(E) 1992, 31, 430.
10. Trecker, D. J.; Foote, R. S. In Organic Photochemical Synthesis; Srinivasan, R., Ed.; Wiley: New York, 1971; Vol. 1, p 81.
11. Salomon, R. G.; Kochi, J. K. JACS 1974, 96, 1137.
12. Salomon, R. G.; Kochi, J. K. JACS 1973, 95, 1889.
13. Salomon, R. G.; Folting, K.; Streib, W. E.; Kochi, J. K. JACS 1974, 96, 1145.
14. Evers, J. T. M.; Mackor, A. RTC 1979, 98, 423.
15. Evers, J. T. M.; Mackor, A. TL 1980, 21, 415.
16. Spee, T.; Mackor, A. JACS 1981, 103, 6901.
17. Evers, J. T. M.; Mackor, A. TL 1978, 2317.
18. Salomon, R. G.; Sinha, A. TL 1978, 1367.
19. Evers, J. T. M.; Mackor, A. TL 1978, 821.
20. (a) Raychaudhuri, S. R.; Ghosh, S.; Salomon, R. G. JACS 1982, 104, 6841. (b) Ghosh, S.; Raychaudhuri, S. R.; Salomon, R. G. JOC 1987, 52, 83.
21. Avasthi, K.; Raychaudhuri, S. R.; Salomon, R. G. JOC 1984, 49, 4322.
22. Hertel, R.; Mattay, J.; Runsink, J. JACS 1991, 113, 657.
23. (a) Salomon, R. G.; Coughlin, D. J.; Easler, E. M. JACS 1979, 101, 3961. (b) Salomon, R. G.; Ghosh, S. OS 1984, 62, 125; OSC 1990, 7, 177. (c) Salomon, R. G.; Coughlin, D. J.; Ghosh, S.; Zagorski, M. G. JACS 1982, 104, 998.
24. Salomon, R. G.; Ghosh, S.; Zagorski, M. G.; Reitz, M. JOC 1982, 47, 829.
25. McMurry, J. E.; Choy, W. TL 1980, 21, 2477.
26. Avasthi, K.; Salomon, R. G. JOC 1986, 51, 2556.
27. Salomon, R. G.; Ghosh, S.; Raychaudhuri, S.; Miranti, T. S. TL 1984, 25, 3167.
28. Hefner, J. G.; Zizelman, P. M.; Durfee, L. D.; Lewandos, G. S. JOM 1984, 260, 369.
29. (a) Cohen, T.; Cristea, I. JOC 1975, 40, 3649. (b) Cohen, T.; Cristea, I. JACS 1976, 98, 748.
30. Cohen, T.; Tirpak, J. TL 1975, 143.
31. Cohen, T.; Herman, G.; Falck, J. R.; Mura, Jr., A. J. JOC 1975, 40, 812.
32. Kwon, T. W.; Smith, M. B. SC 1992, 22, 2273.
33. Cohen, T.; Mura, A. J.; Shull, D. W.; Fogel, E. R.; Ruffner, R. J.; Falck, J. R. JOC 1976, 41, 3218.
34. Corey, E. J.; Seebach, D. JOC 1966, 31, 4097.
35. Cohen, T.; Kosarych, Z. TL 1980, 21, 3955.
36. Cohen, T.; Gapinski, R. E.; Hutchins, R. R. JOC 1979, 44, 3599.
37. (a) Kozikowski, A. P.; Ames, A. JACS 1980, 102, 860. (b) Kozikowski, A. P.; Ames, A. T 1985, 41, 4821.
38. (a) Corey, E. J.; Noyori, R. TL 1970, 311. (b) Corey, E. J.; Erickson, B. W.; Noyori, R. JACS 1971, 93, 1724.
39. Cohen, T.; Bennett, D. A.; Mura, A. J. JOC 1976, 41, 2506.
40. Semmelhack, M. F.; Tomesch, J. C. JOC 1977, 42, 2657.
41. Cohen, T.; Kuhn, D.; Falck, J. R. JACS 1975, 97, 4749.
42. Knapp, S.; Trope, A. F.; Ornaf, R. M. TL 1980, 21, 4301.
43. Knapp, S.; Trope, A. F.; Theodore, M. S.; Hirata, N.; Barchi, J. J. JOC 1984, 49, 608.
44. Seebach, D. AG(E) 1967, 6, 442.
45. Cohen, T.; Meyers, M. JOC 1988, 53, 457.
46. Vedejs, E.; Nader, B. JOC 1982, 47, 3193.
47. Huang, J.; Meinwald, J. JACS 1981, 103, 861.
48. (a) Masamune, S. Aldrichim. Acta 1978, 11, 23. (b) Masamune, S.; Yamamoto, H.; Kamata, S.; Fukuzawa, A. JACS 1975, 97, 3513.
49. Masamune, S.; Kamata, S.; Schilling, W. JACS 1975, 97, 3515.

Robert G. Salomon

Case Western Reserve University, Cleveland, OH, USA

Copyright 1995-2000 by John Wiley & Sons, Ltd. All rights reserved.