The opposite insertion orientation gives a benzylic metal–carbon bond, . C–Si reductive elimination then gives the branched product
A plausible explanation is stabilization of the developing and final benzylic Co–C interaction by the adjacent aromatic pi system, which can outweigh the steric preference for a primary alkyl. That makes an electronically different choice from an ordinary alkyl substituent. This is a mechanistic rationale for the proposed cycle, not proof of its rate-determining step. In the absence of a chiral catalyst or another source of asymmetric induction, the new stereogenic carbon would give a racemic mixture.
Disconnect the new bond between the two six-carbon units. One portion of bromocyclohexane supplies a grignard reagent; another supplies cyclohexene oxide:
A bulky strong base can favour E2 elimination in the alkene-forming step; a peracid such as meta-chloroperoxybenzoic acid provides epoxidation. None of its carbons enters the target.
Treat cyclohexene oxide with cyclohexylmagnesium bromide, then perform acidic aqueous work-up. Backside attack on one epoxide carbon breaks its C–O bond and installs cyclohexyl opposite the retained oxygen. Protonation of the oxygen gives trans-2-cyclohexylcyclohexan-1-ol, the required relative configuration. The two six-carbon reactant portions supply all 12 target carbons. An achiral synthesis produces a racemic mixture trans pair; obtaining only the particular enantiomer represented by the drawing would require resolution or an enantioselective step. The useful stable chair can place the two large trans-1,2 substituents approximately diequatorial.
Figure 1.
Original two-branch synthesis from bromocyclohexane and anti epoxide opening to the trans alcohol
. Details.
B is 4-hydroxy-1-methylpyrrolidin-2-one; C is 1-methylpyrrolidin-3-ol. Their ring connectivities are shown below. Each ring remains five membered; the attached N-methyl group is retained.
Figure 1.
Original cyclic product structures and hydride electron-pair arrows for ketone and lactam reduction
. Details.
For B, a B–H bond of sodium borohydride supplies a hydride equivalent to the ketone carbonyl carbon, and the carbonyl pair moves to oxygen. This gives an alkoxide, whose protonation during the reaction or workup gives the secondary alcohol. The lactam carbonyl remains. Converting one to adds two H atoms and gives .
For C, lithium aluminium hydride reduces that ketone in the same net way and also reduces the lactam. Coordination of the amide oxygen to aluminium assists hydride attack at its carbonyl carbon, giving a tetrahedral intermediate. The N lone pair then forms a C=N bond while the C–O bond breaks to an aluminium-bound oxygen group. A resulting iminium ion accepts a second hydride at carbon, yielding the cyclic tertiary amine. Aqueous workup protonates the ketone-derived alkoxide.
The lactam conversion replaces by ; its oxygen is removed rather than retained as another alcohol. Together with ketone reduction, the net change is , giving . No stereochemistry is specified: the new alcohol stereocentre is formed as a racemic mixture in an achiral environment.
Both side chains have the order ring–, and both printed double bonds have the Z geometry. The two carbon side chains are cis to one another; equivalently the labelled ring H and ring Me are cis on the opposite face. A useful common late precursor therefore has two cis ring– substituents, which can undergo two Z-selective ethylidene Wittig reactions.
Use the starting 2-methylcyclohex-2-en-1-one in a reductive alpha-alkylation. Conjugate reduction under dissolving-metal/enolate-trapping conditions, followed by a primary electrophile , installs a protected two-carbon alcohol side chain at the original methyl-bearing alpha carbon. This gives 2-methyl-2-(2-siloxyethyl)cyclohexanone. The carbon at that position is now quaternary, with its original methyl retained. Alternatively, reduce the enone to 2-methylcyclohexanone and deliberately form/select its more substituted enolate before alkylation; its opposite kinetic enolate would give the wrong regioisomer.
Convert the remaining ring ketone to an exocyclic enoate by Horner–Wadsworth–Emmons olefination with a phosphonoacetate:
Hydrogenate this sole alkene to give ring– at that carbon. Select the diastereomer with this substituent cis to the pre-existing protected hydroxyethyl side chain, so the newly installed ring H is cis to ring Me. Less hindered-face hydrogenation can assist this relationship, but the route should include separation/verification of the desired cis product rather than assert complete stereoselectivity without data.
Reduce the ester with lithium aluminium hydride to ring– and remove the TBS group from the other side chain. Both appendages are now two-carbon primary alcohols. Oxidize them to aldehydes under non-overoxidizing conditions, for example Dess–Martin oxidation, obtaining the cis dialdehyde with two ring– groups. Treat with excess salt-controlled, nonstabilized ethylidenetriphenylphosphorane under conditions favouring kinetic Z olefination; isolate the Z,Z product. Each aldehyde carbon becomes the inner vinylic carbon of its crotyl group, while the ylide supplies .
This order postpones both stereodefined alkenes until after hydrogenation, so they are not accidentally saturated. It also avoids assuming that a crotyl grignard reagent adds without allylic transposition or that direct alcohol deoxygenation fixes the required ring stereochemistry. An achiral synthesis gives a racemic mixture pair with the specified relative configuration.
Figure 1.
Original stereochemical plan through a cis diol/dialdehyde and late double Z-selective Wittig olefination
. Details.
The target is 5-Aminohexanoic acid, . Disconnect its terminal amine/carboxylic-acid relationship through hydrolysis of 6-methylpiperidin-2-one. This six-membered lactam follows from a regioselected Beckmann ring expansion of 2-methylcyclopentanone oxime.
Start with cyclopentanone. Form its enolate with lithium diisopropylamide in dry THF at low temperature, add methyl iodide, then quench to obtain 2-methylcyclopentanone. Control the methyl iodide/base stoichiometry to suppress a second alkylation. React the ketone with hydroxylamine to form its oxime and select the geometrical oxime isomer in which the bond to the methyl-substituted adjacent ring carbon is anti to the departing oxime oxygen group.
Activate the oxime, for example under appropriate acid-promoted Beckmann conditions. The anti methyl-bearing C–C bond migrates onto nitrogen as the N–O-derived leaving group departs, inserting nitrogen between the original carbonyl carbon and the methyl-substituted ring carbon. Hydration/tautomerization gives 6-methylpiperidin-2-one. The opposite oxime geometry would insert nitrogen on the other side and ultimately place methyl at a different position in the amino acid; oxime geometry is therefore an essential part of this proposal.
Hydrolyse the lactam with aqueous acid and heat, then adjust the work-up to the desired acid/ammonium form. Cleavage of its acyl–nitrogen bond opens the ring to after suitable neutralization. All five original cyclopentanone carbons are retained, and the one methyl carbon comes from methyl iodide. A racemic mixture product is appropriate to the stereochemically unspecified target. The reaction sequence is an organic ring-expansion synthesis and requires no biological transformation.
Figure 1.
Original carbon-numbered route through the anti oxime and ring-expanded lactam to the amino acid
. Details.
Disconnect the tertiary amide into cyclohex-3-ene-1-carboxylic acid and the secondary amine . The acid-derived ring is an immediate Diels–Alder disconnection: buta-1,3-diene plus the indicated methyl acrylate gives methyl cyclohex-3-ene-1-carboxylate. Heating, or suitable Lewis-acid activation of the dienophile, forms the two new sigma bonds and leaves the diene's internal double bond as the ring alkene.
Hydrolyse this ester to the acid. Convert it to its acid chloride with oxalyl chloride or thionyl chloride, and react with the secondary amine in the presence of a base to capture HCl. The tertiary amide now contains the complete cyclohexene acyl fragment; direct conjugate addition of the amine to methyl acrylate before ring construction would give the wrong connectivity.
Prepare the amine by reductive amination of 4-phenylbutan-2-one, , with ethylamine, using sodium triacetoxyborohydride or another iminium-selective reducing system. This places nitrogen on , not on a terminal methylene. If a route from simpler carbon compounds is wanted, crossed aldol condensation of acetone with benzaldehyde gives , and selective C=C reduction gives the required ketone before reductive amination.
The acid branch retains the acrylic carbonyl as the amide carbonyl; the amine branch supplies both N-ethyl and N-(4-phenylbutan-2-yl) substituents. The unlabelled stereogenic centres may be racemic mixture. This sequence separates ring formation from amide formation and avoids invoking uncontrolled alkylation of a tertiary amide nitrogen.
Figure 1.
Original two-branch synthesis combining a Diels–Alder acid fragment with a reductively aminated secondary amine
. Details.
An individual bipyramid is chiral. Matching helicities at the two iron centres give the homochiral and enantiomers, with the three spanning connections winding in opposite senses in the two structures. The idealized molecular framework has a threefold axis but no mirror plane, inversion centre or improper rotation that converts it into itself. Achiral synthesis can give a racemic mixture of these chiral cages; racemic bulk composition does not make each cage achiral. The corresponding experimentally characterized methyl-substituted Fe–Pd bipyramid contains both homochiral enantiomers in its crystal.[ref]