Radical‐Induced 1,2‐Migrations of Boron Ate Complexes

Abstract 1,2‐Boron ate rearrangements represent a fundamental class of transformations to establish new C−C bonds while retaining the valuable boron moiety in the product. In established ionic processes, the boron ate complex is activated by an external electrophile to induce a 1,2‐migration from boron to an adjacent sp 3 or sp 2 carbon atom. Recently, two complementary radical polar crossover approaches have been explored for both classes, 1,2‐migrations to sp 2 and sp 3 carbon centers. This review describes the general concepts in this emerging research field and summarizes recent developments of radical‐induced 1,2‐migrations from boron to carbon.


1I ntroduction
Organoboranes and boronic estersa re highly valuable reagents to conduct various cross-coupling reactions. [1] Furthermore,t he boron moiety can be converted into ab road range of functional groups,o ften with high levels of stereospecifity when considering chiral secondary or tertiary alkylboronic esters as substrates. [2] Mechanistically,t hese transformationsu sually proceed by nucleophilic addition to the empty p-orbital of the boron atom followed by as tereospecific 1,2-migrationt oa na djacent electrophilic acceptor atom. In the early 1960s,H illman [3] andM atteson [4] first demonstrated that 1,2-borona te rearrangements allow for C À Cb ond formation. Notably, in these transformations the valuable boron moiety remains in the product. Along these lines,t he introduction of the air-a nd moisture-stable boronic ester entity,i np articular the pinacol boronic ester, opened an ew field of organoboron homologationc hemistry. [5] As af irst class of this important transformation type,1 ,2-boron ate shifts to sp 3 carbons bearing as uitable a-leaving group have to be highlighted (Scheme 1a). In the socalled Matteson reaction, which initiated the research field, [6] ab oron ate complexd erived from ap re-functionalized a-halo boronic ester and an alkyl-or arylmetal compound is activated by aL ewis acid to induce the 1,2-boront oc arbon migration with concomitant substitution of the halide anion. Alternatively,t he boron ate complexc an be generated through the reaction of ab oronic ester with ac hiral carbanion bearing as uitable a-leaving group (carbenoid-type compound). [7] In the second class,t he p-bond of a vinyl boron ate complexi sa ctivated by an external electrophile to trigger a1 ,2-shift to the former sp 2 acarbon center (Scheme 1b). Them ost prominent reaction of this class is the Zweifel olefination, [8] in which the 1,2-shift is inducedb ye lectrophilic halogenation of the vinyl moiety in the B-ate complex. Following Murakamiss tudies [9] on alkynyltrialkyl boron ates, Morkena nd co-workers established in 2015t he use of transitionm etal complexes to activate alkenyl boron ate complexes in conjunctive cross-couplings. [10] Recently,t wo complementary radical approaches have been developed for both classes, 1,2-migrations to sp 2 and sp 3 carbon centers. [11] In the first approach, the 1,2-boront oc arbon migration is induced by an initial radical addition to the b-position of av inylb oron ate complex( Scheme 1c). In the second approach, ar egioselective radical a-C(sp 3 ) À Ha bstraction in ab oron ate complexd erived from an on-pre-functionalized alkyl boronic ester triggers the 1,2-shift (Scheme 1d). In both cases ar educing radical anion [12] is generated, which furtherr eacts via electront ransfer (ET) oxidation followed by 1,2-aryl/alkyl migration.O fn ote, such radical polar crossover reactions usually occur via electronc atalysis [13] and proceed without the need of anyt ransitionm etal catalyst. Published review articles in this field cover 1,2-boron ate shifts that are triggeredt hrough ionic mechanisms. [5,6e,f,7d,8b] In this perspective we will focuse xclusively on radical-induced 1,2-migrations to sp 2 and sp 3 centers that have been discovered only recently.

Vinyl Boron Ate Complexes
Vinyl boronic estersa re highly importantb uilding blocks in organic synthesis. [14] Such boron-substituted alkenes have found attention as radical acceptors for more than 50 years, [15] but the radical chemistry on the corresponding vinyl borona te complexes has been explored only recently.I nc ontrastt ot heir neutral congeners,v inyl boron ate complexes bear an electron-rich double bond and therefore express a high reactivity towards electrophilic alkyl radicals. Theg eneration of such Cr adicals can be achieved from ar adical precursorR À Ib yIabstraction, reductive mesolytic C À Ic leavage or by photolytic homolysis of the C À Ib ond. Addition of the electrophilic Post-doctoral Fellow at the Universityo f Pittsburghw ith Prof.C urran. In 1996h es tarted his independent career at the ETH Zürich.I n2 000, he was appointed as associatep rofessor at the Philipps-University in Marburg, Germany andi n2 004 as full professor of organic chemistry at the Westfälische Wilhelms-University in Münster, Germany.H is researchi nterests focus on the development of new synthetic methods.I n addition,h ea lso works on living radical polymerizations,o nt he preparation of functional polymers and on the development of methods for the chemical modificationo fs urfaces.
alkyl radical RC to the b-position of the vinyl boron ate complex A generates the corresponding adduct radical anion B (Scheme 2a). In the following radical polar crossover step,a dduct B undergoes an outersphere electron transfer oxidation to deliver zwitterion C.A ccording to an electron-catalyzed process, [13] the electroni sd irectlyt ransferred to R À I, thereby generating radical RC sustaining the chain. In the subsequent ionic1 ,2-R''-migration the carbon substituent is transferred from boront ot he a-carbon atom. The overallp rocess comprises the formation of twoC À C bonds and deliversaboronic ester of type D.A na lternativei nner-sphere electront ransfer mechanism is presented in Scheme 2b.I nt his scenario,a dduct radical anion B could abstract the Ia tom of R À It og ive the atom transfer product E,w hich could further engageinaMatteson-type 1,2-R''-migration to deliver the same target D.W hether the reaction proceeds by outer-sphere (B to C)o ri nner-sphere (B to E)e lectron transfer depends on the reductionp otential of the alkyl radical precursor ando nt he halogen atom transfer efficiency. [13b] However, this reaction design causes two major challenges.F irstly,e specially b-substitutedv inyl borona te complexes are prone to undergo ac ompetingr adical a-addition leading to the formation of trans-alkenes (Scheme 2c). This reactivity was established by the groups of Buchwalda nd Akita for the transformation of vinyltrifluoro boron ate complexes with the Togni reagent [16] in the presence of Fe(II)salts [17] or an Ru-photoredox catalyst. [18] Later, that approachw as furthere xtended by Leonori and co-workers to the coupling of vinyl borona te compounds with various electrophilic alkyl radicals. [19] Secondly,t he alkyl radical precursors hould be am ild oxidantt hat doesn ot directly react with the starting electron-rich boron ate complex via an electront ransfer process leading to the formation of the ethenyl radical (Scheme2d).
In 2017, the groups of Studer andA ggarwal independently implemented such as trategy( Scheme 3a, b). In both methods the sequence commences with the in situ formationo ft he vinyl boron ate complex by treatment of av inylb oronic pinacol ester with an aryl-or alkyl-lithium compound.A fter solvent exchange, the boron ate complex can be used in the radical sequence without any furtherp urification. Notably,t he three-component coupling tolerates a-a nd bsubstituted vinyl borona te complexes and the substrate scope with respect to the radical precursor R À I includesp erfluoroalkyl iodides, a-iodo sulfonates, aiodo phosphonates, a-iodo nitriles, a-iodo esters, aiodo ketones,p rimary a-iodo amides and diethyl bromomalonates allowing the rapid build-up of molecular complexity.I ntheS tuder protocol, initiation of the radical chain reactioni sa chieved with BEt 3 /O 2 (Scheme 3a). [20] In contrast, the Aggarwal approach initiation proceeds under blue light irradiation (Scheme 3b). [21] Moreover, less reactive alkylb romides could successfully be employed in the presence of 20 mol%s odium iodide or 1mol% of as martR uphotoinitiator.T he term smarti nitiation was first introducedb yS tuder andC urran to describe catalysis of initiation andi se specially valuablei fs hortc hains are involved. In the context of this article,asmart initiator is interpreted as ar edox-active compound that is eligible to bothr eductively initiate ar adical chain reactiona nd to be regeneratedb yo xidation of ar educing species of the innate chain (in this case radical anion B,s ee Scheme 2). Even though in borona te chemistry so faro nly photoredox catalysts have been described as smartinitiators,the concept is readily ap-plicable to other substance classes. One obvious feature of smarti nitiation is displayed by the quantum yield F,w hich is often determined to be greatert han 1. Consequently,o ne photon generates multiplep roduct molecules,s trongly suggesting that an innate radical chain is operative and the photoredox catalyst is mainly involved in the initiation step. [13b] Thep ractical value of the Studerp rotocolw as further documented in an improved ande xperimentally checked Organic Synthesis procedure using only 1.5 equivalents of the radical precursoru nder irradiation with visible light. [22] In 2018, Renaud and coworkers presented ar elatedp rotocol that can be performed without solvent exchange in methyl tert-butyl ether (MTBE). Nevertheless,2 .0 equivalents of BEt 3 and 30 mol%d i-tert-butyl hypodinitrite (DTBHN) are required for initiation( Scheme 3c). [23] In additional mechanistic experiments,t he authors could further support that the mechanism likelyp roceeds via an outer-sphere electron transfer process (see Scheme 2a). More recently,S hi and co-workers demonstrated that in situ generated alkenyl diborona te complexes derived from alkenyl-Grignard reagents and bis(pinacolato)diboron undergo efficient radicalinduced 1,2-boront oc arbon shifts of ab orong roup to give various gem-bis(boryl)alkanes( Scheme 3d). [24] Of note, the quantum yield of this process was determined to be F = 49.8, clearly showing that the Ru complexmainly acts to initiate an innate radical chain reaction.
In 2017, Lovinger and Morken disclosedt heir results on aN i-catalyzed enantioselective conjunctive coupling with C(sp 3 )e lectrophiles. [25] Interestingly,f or non-activated alkyl halides the authors observedh igh enantioselectivity,whereas the reactions with the electron-poor congeners furnished racemic products (Scheme 4). Mechanistic experiments revealed that the origin of this different reactiono utcome is ar adical-ionic mechanistic dichotomy.I nt he case of nonactivated alkyl halides,t he vinyl boron ate complex engages in an enantioselective metal-induced 1,2boron ate shift with subsequent reductivee limination in analogy to the previously describedP d-catalyzed enantioselective conjunctive couplings [10a] with C(sp 2 ) electrophiles.I nc ontrast, activatedC ( sp 3 )e lectrophiles are reducedm ore easily and an Ni-initiated radical polarcrossover chain reactioniso perative.
Highly enantioenriched compounds are readilya ccessible by the radical polar crossover strategy using chiral alkyl boronice sters as substrates for the formation of the corresponding vinyl borona te complexes. Along these lines,t he radical-induced 1,2-borona te rearrangement was successfully used for the preparation of a-chiral ketones and enantioenriched alkanes, which were obtained after oxidation of the initial Bcontainingp roductsw ith sodium perboratea nd Dess-Martin periodinane or by protodeborylation, respectively (Scheme5a). [26] In these processes,v inyl boron ate complexes derived from enantioenricheda lkyl boronice sters and vinyllithium are reacted with various commercially available alkyli odides.N otably, chiral boronic esters are easily accessible by hydrobo-ration, [27] the Matteson approach (substrate control) [6] or by lithiation-borylation using Hoppesc hiral lithiated carbamates (reagent control). [7] Deborylative follow-up chemistry was necessary,s ince the stereocentera tt he a-position to the boron atom could not be controlled in the radical polar crossover cascade.I f a-perfluoroalkylated ketones are formed as initial productsa fter C À Bt oC =Of unctional group transformation, HF elimination was observedd uring purification leading to a,b-unsaturated ketones.T he high synthetic potential of the stereospecific radical-induced 1,2-borona te rearrangement was further demonstrated by its applicationt on ovelf ormal total syntheses of d-(R)-coniceine and indolizidine 209B, starting with the enantiopure Boc-protected pyrrolidine boronic ester (Scheme 5b). [28] Via ar adical polarc rossover reactionu sing a-iodoacetates as C-radical precursors, twoC À C-bonds are formed in the keyr eaction. As equence of lactamization, protodeborylation and reductiono ft he lactams finally gives the desired natural products.

DienylB oron Ate Complexes
Inspired by the radical polarc rossover reactiono f "simple" vinyl borona te complexes (see Section 2.1), the Studer group investigated dienyl boron ate complexesa sr adical acceptors. [29] If the addition of the transient C-radical selectively occurs at the d-position of the borona te complex A,a na llyl radical intermediate B is generated. Tr iggered by electron-transfer oxidation, it was assumed that B might then undergo a1 ,2-boron ate rearrangement via zwitterionic intermediate C to give allyl boronic esterso ft ype D, thereby propagating the chain reactionb yg eneration of the transient C-radical R 5 C.I nf act, this scenario was realized using various borona te complexes derived from dienyl boronic esters anda ryl-or alkyllithium reagents in combination with electron-poor alkyl iodides under visiblel ight irradiation (Scheme 6a). Theg eometryo ft he double bond arising from reactions of the alkenyl boron ate complexes in Scheme 6w as well controlled in most cases with the E-isomer being formed with good to excellent selectivity.N otably, side products formed by the undesired b-addition to the ate complex A were not detected, revealing the high intrinsic d-selectivity of the dienyl boron ate acceptor. Thea llyl boronic ester productsa re valuable building blocks in synthesis and were employed in an allylation/lactonization sequence to afford highly substituted d-lactones with excellent diastereoselectivity as exemplified in Scheme 6b.

Heteroaryl Boron AteComplexes
Basedo np reviously reported electrophile-induced enantiospecific coupling of alkyl boronic esters with lithiated electron-rich arenes, [30] Aggarwal disclosed a trifluoromethyl-radical induced three-componentc oupling of boron ate complexes derived from boronic estersa nd lithiated furans (Scheme 7a). [31] In this elegant process,t he Umemoto reagenti su sed as the trifluoromethyl radical precursora nd the chain-carrying Scheme 5. Stereospecificr adical-polar crossover reaction of chiral boronic esters and its application to the synthesiso fd-(R)-coniceine and indolizidine209B. trifluoromethyl radical regioselectively adds to the electron-rich furan moietyo ft he boron ate complex A to givet he corresponding radical anion B.E lectron transfer from intermediate B to the Umemoto reagent generates the trifluoromethyl radical along with the zwitterionic species C that rapidlyu ndergoes a 1,2-borona te rearrangement to provide the substituted dihydrofuran D.S uch furans are isolable but can also be readily converted to the corresponding 2,5functionalized furans upon treatment with iodine under basic conditions.I nt he overall sequence,t he pharmacologically relevant CF 3 -group along with an aryl or an alkyl group can be transferred to the furan moiety,t he latter with complete enantiospecifity. Later, this protocolw as extended to electron-poor alkyl iodidesa ss ubstitutesf or the Umemoto reagent and chain initiation was achieved by irradiationw ith blue LEDs (Scheme 7). [32] This required addition of Ru(bpy) 3 Cl 2 to the reactionm ixture,a ctinga sasmart initiator -ah igh quantum yield of well above1( F = 27.8) suggests ar adical chain mechanism to be operative in these cascades. [13b] Fori ndoles, neither irradiation nor ap hotoredox catalyst was required to achieve product formation. Basedo nm echanistic studies, the authors stated that, in this particularc ase,t he reaction likely followsanon-radicalS N 2p athway. [32] 3R adical-Induced1,2-BoronA te Shifts to sp 3 Carbons

Radical Addition to Bicyclobutyl Boron Ate Complexes
While the generation of sp 3 -centered radicals via radical addition to p-systemsi sacommon approach in synthetic chemistry,r adical generation via homolytic substitution at carbon (sp 3 )i sn ot well established. Due to the high intrinsic strain energy and the partial p-character of the central C À Cb ondo fabicyclobutane (BCB), such amoietyc an act as ar adical acceptor. [33] Aggarwal first employed BCB boron ate complexeso ft ype A in ar adical polar crossover sequence (Scheme 8). [34] Ther equiredB CB boron ate complex A is accessible via lithiation of BCB p-tolyl sulfoxide with t-BuLi and subsequent addition to ab oronic ester. [35] Upon radical addition,t he ring strain is released,l eading to ac yclobutyl radical intermediate B, which can further react via a1 ,2-boron ate rearrangement triggered by oxidation to finally give af unctionalized cyclobutane C.T his structural motif has received growing interest in medicinal chemistry. [36] The strongly reducing properties of the boryl radical anion B allowed the use of various electron-deficient alkyl iodidess uch as perfluoroalkyl iodides, a-iodomethyl sulfonates, a-iodo nitriles, a-iodo acetates and a-iodo amides as C-radical precursors.N on-activated alkyl iodidesw ere found to be inactive.N otably,b oth the radical addition andt he subsequent migration are stereochemicallyw ellc ontrolled leading to a cis-configuration in the products. Form ost substrates,g oodt o excellent diastereoselectivity was found and the migrationo ccurred with complete stereospecifity,a se xpected.

H-Transfer-Induced Coupling of Alkylboronic Esters andO rganometallicReagents
IntermolecularH AT is av aluables trategyf or the direct functionalization of C(sp 3 ) À Hb onds,a lbeit sometimes difficult to control. Although this approach for C À Hf unctionalization has been known for decades,m ost preparatively useful protocols in this field have appeared in the last ten years in the context of photoredox catalysis. [37,38] In contrast to heteroatomcentered radicals, C-radicals usuallyl ack thermodynamic drivingf orce to engage as H-acceptors in intermolecular HATs.T he highly reactive CF 3 radical is an exception along those lines. [39] TheS tuder group recently disclosed highly regioselective generation of sp 3 -centeredc arbon radicals via intermolecular HATf rom alkyl boron ate complexes to CF 3 radicals (Scheme 9). [40] Instead of the a-prefunctionalized organometallic reagents used in the classicalM atteson reaction, unfunctionalized alkyl boronice stersc an be applied as substrates.I nt hese transformations, the target a-C À Hb ond of the boron ate complex is directly activated to trigger the 1,2boron ate rearrangement, thus significantly increasing step economy.I nt he keys tep,ahydrogena tomi s transferred from the a-C(sp 3 ) À Hb ondo fapreviously formed borona te complex A to ar eactiveC F 3 radical. Subsequent oxidation of the intermediately generated radical B to the zwitterionic species C inducesa 1,2-borona te rearrangement to affordt he desired rearrangedp roduct D.C F 3 Ii su sed as the oxidant and in the oxidation of radical B,t he chain-carryingt rifluoromethyl radical is generated along with C.O ne obvious challenge in this sequence is the realization of ar egioselective HAT, where the trifluoromethyl radical turns out to be ideally suited as active species. Likely,p olar effects play ak ey rolei nt he selective a-H-abstraction of these anionic boron atec omplexes by the electrophilic CF 3 radical (see below). Moreover, it is well established that CF 3 Ii sa ne xcellent terminal oxidant for intermediates of type B, [20,26,29] rendering this cheap iodide to be the reagento f choice to realize this interesting sequence. [39] Ir(ppy) 3 was employed as as mart initiator uponi rradiation with blue LEDs (F = 8.8). In selected cases,i nitiation was achieved with blue LED irradiationi nt he absence of the Ir complex, underlining the apparent chain-type reactionm echanism. [13b] This methodology enables botht he a-C À Ha rylation (Scheme 9a) and a-C À Ha lkylation of alkylboronic esters (Scheme9b).
Considering the latter transformation,t he starting dialkyl borona te complexes offer two different C À H sites at the a-position to boron that have to be differentiatedb yt he CF 3 radical. It was found that HAT selectivity follows ac leart rend in which generally the weaker andl ess sterically hinderedC À Hb ond is preferably activated. Interestingly,t he isopropyl group out-competest he a-methoxyalkyl group,e mphasizing that conformation likelya lso plays ac rucial role in determining the regioselectivity of the intermolecular HAT. Due to the stereospecific nature of the 1,2boron ater earrangement,e nantioenriched boronic estersa re also accessible via the radical induced a-C À Ha lkylation reaction (Scheme9b). Thek ey HAT transfer stepw as furtheri nvestigated by density functional theory (DFT) calculations. These theoretical studies revealed that the high HATs electivity is caused by polare ffects leading to low energy barriers for the transfer of the a-H-atom of the borona te complextot he electrophilic CF 3 radical.

4S ummaryand Outlook
In recent years,r adical chemistry has experienced a renaissance and accordingly also boron-based radical chemistry has matured. Va riousm ethods for C-radical borylation to access boronic esters have been developed. [41] In these reactions,t he boron moiety gets introducedi ntoa no rganic compound. There are also radical transformations in which the boronic esters are used as substrates.F or example,v inylb oronic estersh ave been shown to react as acceptors with various C-radicalsw ith pioneering studies being published decades ago.I nc ontrast, radical transformations on boron ate complexes represents av ery young researchf ield that hase volved during the last two years.A sd iscussed in this perspective, in radical cascades involving borona te complexes,t he radical bond-forming step can be combined with an ionic1 ,2metallate rearrangement from boron to carbon culminating in ar adical polarc rossover process.T his is meanwhile well documented for C-radical addition to vinyl borona te complexes.S ince ionic1 ,2-boron to carbon migrationso fa lkylg roups are stereospecific, such crossover chemistry allows the preparation of enantiomerically pure compounds. Alkyl, aryl and even borylg roups can act as the migrating moieties in these 1,2-rearrangements.S uch cascades are usually conducted under mild conditions,t olerate many functional groups and are highly modular. Theb oron moiety remains in the product and thus such compounds can serve as valuables ubstrates for follow-up chemistry.
Despite greatr ecent achievementsi nt his emerging researcha rea, some limitations have become apparent, especially considering enantioselective radical Scheme 9. HAT-induced coupling of alkyl boronic esters with organometallic reagents. polar crossover processes.I na ddition,c urrent cascades are restricted to ar ather small set of C-radicals. Fore xample,h eteroatom-centered radicals have not yet been usedi nc ombination with borona te complexesasreactionpartners.T ofurtherexpand the synthetic utility of the radical-induced 1,2-boron ate rearrangement, electrochemistry or extended combination with transitionm etal catalysism ight provide answers. In anyc ase,t he authors of this perspectivea re confident that manye xciting findings will be published in this highly active research field in future.