Visible‐Light‐Induced Homolysis of Earth‐Abundant Metal‐Substrate Complexes: A Complementary Activation Strategy in Photoredox Catalysis

Abstract The mainstream applications of visible‐light photoredox catalysis predominately involve outer‐sphere single‐electron transfer (SET) or energy transfer (EnT) processes of precious metal RuII or IrIII complexes or of organic dyes with low photostability. Earth‐abundant metal‐based MnLn‐type (M=metal, Ln=polydentate ligands) complexes are rapidly evolving as alternative photocatalysts as they offer not only economic and ecological advantages but also access to the complementary inner‐sphere mechanistic modes, thereby transcending their inherent limitations of ultrashort excited‐state lifetimes for use as effective photocatalysts. The generic process, termed visible‐light‐induced homolysis (VLIH), entails the formation of suitable light‐absorbing ligated metal–substrate complexes (MnLn‐Z; Z=substrate) that can undergo homolytic cleavage to generate Mn−1Ln and Z. for further transformations.


Introduction
Thee mergence and upsurge of visible-light photoredox catalysis have made an ineradicable impact on contemporary organic synthesis in the last decade,p roviding access to unconventional reactivity profiles of small molecules by the efficient conversion of photonic into chemical energy. [1] To date,t he prevailing external chromophores used in such transformations are heavy transition-metal catalysts with appropriate ligands such as Ru II -o rI r III -polypyridyl complexes or metal-free organic dye sensitizers as they possess long excited-state lifetimes,s trong absorption in the visible region of the electromagnetic spectrum, and the corresponding photoexcited states have high reduction or oxidation potentials. [1c] Them odes of action of the excited states of these photocatalysts are either single-electron transfer (SET) or energy transfer (EnT) processes to generate various radical species complementary to common thermal two-electron processes. [2] However,o rganic dyes and heavy transition metal-based complexes have downsides in terms of their low photostability [2b] and adverse economic,b iological, and environmental impacts, [3] respectively.I nt his context, the exploration and exploitation of earth-abundant and inexpensive 3d transition-metal complexes as the next generation photocatalysts is rewarding from the perspectives of sustainability and large-scale synthetic applicability. [4] However,t he wide application of such earth-abundant metal complexes is greatly limited by their ultrashort excited-state lifetimes (pico-to nanosecond range) compared to iridium-and rutheniumbased photocatalysts (microsecond range), thus making the prospect of initiating bimolecular SET and EnT processes bleak. [5] Afew well-designed copper complexes have been used as alternative photocatalysts that demonstrate distinct mechanisms involving electron transfer within the inner coordination sphere,thereby controlling reactions through their ligand environment. [6] Other first-row transition-metal salts have been used as successful photocatalysts in isolated cases, [5b] but more notably used as co-catalysts in various photochemical coupling reactions. [7] Complexes based on 3d transition metals generally possess ahigh degree of ligand-substitution lability-a feature that impedes the attainment of favorable photoexcited-state properties such as long lifetimes or photoluminescence. [1c, 2a] Nonetheless,t his property can be creatively utilized for developing mechanistically distinct new photocatalytic processes,t ermed visible-light-induced homolysis (VLIH), complementary to the conventional/cooperative processes with coordinatively saturated and substitution-inert heavy-metal-based photocatalysts.
Themechanism of VLIH proceeds through 1) the initial formation of the metal-substrate complex [L n M n (X)-Z] from the electronic ground state of the metal complex and the substrate through ligand transfer/ exchange,o xidative addition, single-electron oxidation, or transmetalation;2 )photoexcitation of the metal-substrate complex to form [L n M n (X)-Z]*; and 3) inner-sphere redox processes through various metal-complex-specific electronic transitions that ultimately result in homolysis of the metalsubstrate (M n -Z) bond to generate the reduced metal species [L n M nÀ1 (X)] and aradical species (ZC)from the substrate that is set for further transformations ( Figure 1A). Thea dvantages of this strategy are the high chemoselectivity and site selectivity of the photochemical processes,a sthe targeted oxidation takes place solely at the transiently ligating atom or functional group,w ith other oxidation-prone functionalities being left intact.
Thek ey mechanistic event in the VLIH process is the homolytic cleavage of the metal-substrate (M À Z) bond, for which various inner-sphere electronic charge-transfer modes can be responsible.T he traditional mononuclear heavymetal-based photoactive complexes display metal-to-ligand charge transfer (MLCT) involving d!p*, ligand-to-metal charge-transfer (LMCT) involving p!d, and intraligand (IL) transitions.Ingeneral, these charge-separating transitions are nondissociative and do not result in the cleavage of the corresponding metal-ligand bonds.T herefore,the complexes can participate in various reversible outer-sphere electron- The mainstream applications of visible-light photoredox catalysis predominately involve outer-sphere single-electron transfer (SET) or energy transfer (EnT) processes of precious metal Ru II or Ir III complexes or of organic dyes with low photostability.Earth-abundant metal-based M n L n -type (M = metal, L n = polydentate ligands) complexes are rapidly evolving as alternative photocatalysts as they offer not only economic and ecological advantages but also access to the complementary inner-sphere mechanistic modes,t herebytranscending their inherent limitations of ultrashort excited-state lifetimes for use as effective photocatalysts.The generic process,termed visiblelight-induced homolysis (VLIH), entails the formation of suitable light-absorbing ligated metal-substrate complexes (M n L n -Z; Z = substrate) that can undergo homolytic cleavage to generate M nÀ1 L n and ZC for further transformations.
transfer processes without losing the integrity of their molecular structure.
In the majority of modern synthetic applications of VLIH, the LMCT electronic transition induces the desired homolysis of the MÀZb onds as ar esult of the ability of the MÀZ complexes to absorb in the visible-light region of the electromagnetic spectrum. However,t hese dissociative LMCT processes are inherently different from the nondissociative ones,asthe dissociative processes tend to alter the electronic population of the s/s * orbitals of covalent M À Zbonds either by depopulating the MÀZ s-molecular orbital or by populating the MÀZ s*-molecular orbital and usually engage metals to participate in the process from their high oxidation states [8] (Cu II ,N i III ,F e III ,Ce IV ,Co III ,etc;F igure 1B). [9] Nevertheless,V LIH does not always exclusively involve LMCT modes.E xcitation of the metal-substrate complexes with visible light can result in other modes of electronic transitions that also induce homolysis of the MÀZb onds ( Figure 1B). In square-planar [Ni II ( t-Bu bpy)(o-Tol)Cl]-type complexes,t he VLIH events involve MLCT/ 3 d-d electronic transitions that result in cleavage of the Ni À aryl bond to generate aryl radicals and the corresponding Ni I species. [10] However,M (Sub)(CO) 3 (diimine)-type complexes can demonstrate an alternate charge-transfer mode from a s-bond to the ligand (SBLCT, s!p*), thereby resulting in ligand reduction along with the generation of the radical (ZC)f rom the substrate ( Figure 1B). [11] Although direct access to sp* states is forbidden by spectroscopic transition rules,itcan be generated by relaxation from the 1 MLCT states.However,in some complexes with non-oxidizable metal centers,access to 1 MLCT states is also prohibited, rendering electrons prone to be directly transferred from the s-orbital of the M À Zbond to the antibonding orbitals of other ligands,which is regarded as aligand-to-ligand charge-transfer mode (L s L p* CT). [12] As mentioned earlier, most of the developments in newer synthetic methods reliant on the VLIH concept have been limited to LMCT transitions.H owever, the ever-expanding development of spectroscopic and analytical techniques has led to the other types of electronic transitions being recognized as the effective cause of VLIH, which might open up opportunities in future developments for organic synthesis.In this Minireview,w ed iscuss the advancements in the field of VLIH, their different mechanistic aspects based on chargetransfer modes,a nd the prospects for its application in synthetic organic chemistry.

Copper
Copper(I)-based complexes are rapidly emerging as capable visible-light-mediated photoredox catalysts that offer not only economic and ecological advantages but also otherwise inaccessible inner-sphere mechanisms to enable challenging transformations. [6,9a, b, 13] In contrast, there are only ah andful of reports available for photocatalytic processes using Cu II compounds.Inthe area of radical-mediated organic reactions,c hlorine radicals are attractive reactive species, partly because of their varied reactivity with different organic compounds and partly because of the easy availability of awide array of earth-abundant transition-metal chloride salts as potential precursors.However,generating chlorine radicals from these salts by photochemical means is challenging,asthe oxidation potential of the chloride anion is much higher (E o (ClC/Cl À ) =+2.03 Vvs. SCE in MeCN at 298 K) [14] than the excited-state oxidation potentials of commonly used photocatalysts. [9c] In 1962, Kochi observed the photolysis of cupric chloride (CuCl 2 ;exists as achlorocupric complex in organic media) to cuprous chloride (CuCl) and the chlorine radical under unfiltered radiation from amedium-pressure mercury lamp at ambient temperatures. [15] Theo bservation could only be explained by as equential process of ligand-to-metal charge transfer (LMCT) followed by homolysis of the CuÀCl bond, thereby establishing one of the earliest examples of the VLIH principle with ac opper salt. Subsequently,t he photogenerated chlorine radical could be successfully exploited to perform different organic transformations,s uch as the quantitative oxidation of 2-propanol to acetone or the formation of styrene dichloride from styrene in 87 %y ield [Scheme 1D,E q. a-i].
Capitalizing on Kochisd iscovery,W an and co-workers developed av isible-light-induced vicinal dichlorination of olefins by directly using CuCl 2 as ap hotoactive species without any exogenous ligand [Scheme 1D,E q. a-ii]. [16] Although ac ombination of 20 mol %C uCl 2 and 2.5 equiv hydrochloric acid as the chlorine source was required for effective dichlorination of unactivated olefins,4 .0 equiv CuCl 2 alone were adequate to induce the same transformations for activated olefins upon irradiation with a3 8W white LED (l = 390-760 nm). Very recently,t he Rovis group achieved the selective C(sp 3 )ÀHa lkylation and amination of feedstock alkanes with electron-deficient olefins,s uch as acrylates and vinyl sulfones,i nt he presence of ac atalytic amount of CuCl 2 under irradiation with long-wavelength UV light. [17] Thet ransformation proceeds by VLIH of an intermediate Cu II species by LMCT to generate achlorine radical which acts as ap owerful hydrogen atom transfer reagent capable of abstracting strong electron-rich C(sp 3 In 2019, [Cu II (dap)Cl 2 ]( dap = 2,9-bis(4-methoxyphenyl)-1,10-phenanthroline) was used in ap hotochemical atomtransfer radical addition (ATRA) reaction between sulfonyl chloride and olefins [Scheme 1D,E q. a-iv]. [9b] In line with Kochisp roposal, VLIH of the LCu(II)ÀCl bond generates the catalytically active LCu(I) species that initiates the reduction of sulfonyl chlorides.I mproving on this concept, [Cu II (dmp) 2 Cl]Cl (dmp = 2,9-dimethyl-1,10-phenanthroline, Scheme 1C)c an be utilized as am ore robust and economic photocatalyst compared with its dap variant, as the dmp ligand is inexpensive and commercially available [Scheme 1D,E q. a-v]. [18] Direct spectroscopic evidence obtained from af ollow-up study in collaboration with the Castellano group [8] has proved that cleavage of the L n CuÀCl bond occurs in < 100 fs and requires blue excitation into the Cl!Cu LMCT transition for the photochemical transformation of Cu II to Cu I and the generation of ar eactive chlorine atom radical.
Activation of the LMCT state of Cu II X 2 -type complexes endowed with suitable ligands other than halides by irradiation with visible light could also be expected to produce radicals (XC)byhomolysis and these could initiate productive organic transformations (Scheme 1A). In 2018, Reiser and co-workers developed ap hotocatalyzed method based on Cu(dap)Cl 2 (Scheme 1C)f or the synthesis of azido ketones from vinyl arenes and trimethylsilyl azide in air [Scheme 1D, Eq. b-i]. [9a] Mechanistically,t he Cu II complex undergoes ligand exchange with azide to generate an ew LCu II N 3bridged dimer,which upon VLIH forms an LCu I species and an azido radical. Thei ncipient azido radical can be intercepted by an alkene,f ollowed by trapping of molecular oxygen. Ther ebinding of the O-centered radical with LCu I regenerates the LCu II species,which releases the product and closes the catalytic cycle.
Shortly after this report, Gong and co-workers developed the visible-light-induced copper(II)-catalyzed enantioselective alkylation of imines [Scheme 1D,E q. c-i], [19] wherein achiral Cu II -bisoxazoline complex is alkylated through transmetalation from the corresponding alkyl trifluoroborate salt and, subsequently,VLIH generates an alkyl radical and aCu I intermediate.I nasecond catalytic cycle,t his alkyl radical adds to ap rotected imine,w hich is activated by the same chiral Cu II -bisoxazoline complex. Then ewly generated Ncentered radical is reduced by the previously formed Cu I species of the first cycle to release the alkylated imine with high enantioselectivity.

Nickel
Theuse of Ni II - [20] or Ni 0 -based [21] complexes as standalone photocatalysts has only been sporadically reported. [22] However,inthe realm of metallaphotoredox-catalyzed CÀCcrosscoupling reactions,n ickel compounds have been exploited most widely because of the excellent radical-capturing ability (aryl, alkyl, acyl, etc.) and ligand lability of Ni II species (d 8 system). In this case,the formation of the products takes place either by oxidation-induced reductive elimination from the electronic ground state of the Ni III species [23] or from excitation-induced reductive elimination from an electronically excited state of Ni II *s pecies. [24] Halogen radicals can be generated from Ni II complexes either by UV irradiation [25] or through triplet-triplet energy transfer from exogenous photocatalysts [26] and used as HAT catalysts for C(sp 3 )ÀHc ross-coupling reactions.Acounterintuitive mechanistic approach has emerged, wherein direct VLIH of high-valent nickel(III) complexes is exploited to photogenerate halogen radicals (Scheme 2B). In 2015, Nocera and co-workers reported several Ni III trihalide complexes from which homolytic photoextrusion of halogen radicalsintermediately stabilized by an arene-to-halogen-atom charge-transfer interaction in the secondary coordination sphere-could be possible from adissociative LMCT excited state to induce aN i À Cl s!s*t ransition [Scheme 2C, Eq. i] . [27] Thef eature was subsequently exploited by the Doyle group in as eries of cross-coupling reactions involving the generation of alkyl radicals through CÀHa bstraction by the incipient photogenerated chlorine radical from light-Scheme 1. Mechanistic features of the VLIH of Cu II species and selected transformations.

Angewandte
Chemie absorbing Ni III species. [9c, 28,29] Theg eneral mechanistic pathway initiates with the oxidative addition of L n Ni 0 to the halide substrate to generate an intermediary ZL n Ni II Xs pecies, which is oxidized by the excited photocatalyst to give ZL n Ni III Xspecies.Irradiation of this species with visible light results in the homolytic cleavage of the Ni III À Xb ond and generation of the corresponding halogen radical (XC)a nd Z-L n Ni II species.XC can participate in ahydrogen atom transfer (HAT) process by interacting with the substrate (or aH AT mediator) to generate an incipient alkyl radical (RC), which gets trapped by the ZL n Ni II species.T he resulting ZL n Ni III R species can then undergo reductive elimination to furnish the cross-coupled product Z-R and aL n Ni I species,w hich gets reduced by the reduced photocatalyst to L n Ni 0 to complete both of the catalytic cycles (Scheme 2A).
Doyle and co-workers used this concept to develop as uccessful strategy for the (hetero)arylation of cyclic and acyclice thers in the presence of [Ir(dF(CF 3 )ppy) 2 -(dtbbpy)]PF 6 as the exogenous photocatalyst and aN ic ocatalyst, whereby (hetero)aryl chlorides were used as both the cross-coupling partners and the chlorine radical source. [9c] Although the strategy was effective for the abstraction of

Angewandte
Chemie hydrogen from ethereal C(sp 3 )ÀHb onds [BDE(C-H-(THF)) = 92 kcal mol À1 ;S cheme 2C,E q. a-i]b yachlorine radical generated through VLIH of [(dtbbpy)Ni III (aryl)(Cl)]type species and led to the formation of the corresponding benzylic ethers in up to 93 %, cyclohexane was only obtained in 41 %yield. Theissue was addressed in asubsequent report by the same group,wherein chloroformates and acid chlorides were used as the cross-coupling partner and the chlorine radical sources to functionalize C(sp 3 ) À Hb onds of unactivated alkanes (BDE = 90-95 kcal mol À1 )f or the syntheses of various carbonyl derivatives [Scheme 2C,E q. b-i]. [29a] A strategy for the alkylation of cyclic ethers has been developed by Kçnig and co-workers,w herein VLIH of [L n Ni III (alkyl)-(Br)] species (Scheme 2Bc) is involved in the generation of nascent bromine radicals that can act as HATmediators and abstract hydrogen from ethereal C(sp 3 ) À Hb onds [Scheme 2C,Eq. c-i]. [30] In an expansionary study of the first report, Doyle and coworkers also developed as elective formylation reaction of aryl chlorides by employing 1,3-dioxolane as the solvent (instead of THF) and apost-reaction mildly acidic workup of the reactions [Scheme 2C,E q. a-ii]. [28] Here also,t he key mechanistic step involves the generation of achlorine radical by VLIH of [L n Ni III (aryl)(Cl)] species (Scheme 2Ba) to abstract hydrogen from the 2-position of 1,3-dioxolane (BDE(C2-H) = 86.8 kcal mol À1 ).
In the same line,the Doyle group reported amethylation strategy of (hetero)aryl chlorides using trimethyl orthoformate as the methyl radical source [Scheme 2C,E q. a-iii]. [29b] Thet ransformation proceeds via the formation and subsequent VLIH of [L n Ni III (aryl)(Cl)] species to generate an incipient chlorine radical that undergoes aHAT process with trimethyl orthoformate and asubsequent homolytic b-scission to form the methyl radical for methylation.
Aphotophysical and photochemical study collaboratively conducted by the Castellano and Doyle groups on aseries of ( R bpy)Ni II (aryl)X-type of complexes using ultrafast UV/Vis and mid-IR transient absorption spectroscopy revealed that, upon irradiation with visible light, an initially formed squareplanar 1 MLCT state of the complex gradually evolves over 5-10 ps into along-lived, tetrahedral 3 d-d (MC) state lying about 0.5 eV above the ground state with alifetime of about 4ns. [10] This transition also results in achange in orbital symmetry to (e) 4 (t 2 *) 4 and thus to ah igher occupancyo fa ntibonding orbitals (t 2 *) that weakens the Ni II ÀAr bond and ultimately leads to its homolysis to generate the aryl radical (probed by spin-trapping experiments with N-tert-butyl-a-phenylnitrone, Scheme 2D)a nd aN i I species.T he study refuted previously assigned long-lived MLCT states [31] and offered an ew mechanistic pathway to initiate catalysis by Ni I .I ndeed, an Ni-catalyzed C-O coupling strategy of (hetero)aryl electrophiles with 18 8 and 28 8 alcohols mediated by long-wavelength UV light (l = 390-395 nm) was subsequently developed by Xue and co-workers,w herein the photoexcited Ni II intermediary complex undergoes Ni À Cb ond homolysis to generate aryl radicals and the Ni I proceeds to take part in aN i I -Ni III catalytic cycle to furnish the corresponding CÀOc rosscoupled products. [32]

Iron
Unlike the precious-element-based (Ru II ,I r III ,O s II ,R e I , etc.)p olypyridyl complexes,i ron(II) complexes have found far fewer applications in organic photoredox catalysis because of their much shorter photoexcited-state lifetimes.This arises because their MLCT excited states can be deactivated extremely rapidly (ca. 50 fs) by energetically lower-lying metal-centered (MC) excited states,w hich results in incompetent electron-transfer reactivity and al ack of photoluminescence. [33] Significant efforts have been expended to prolong the excited-state lifetimes of iron complexes either by the use of chelating ligands that allow robust metal coordination to achieve high symmetry-ideally close to O h coordination-to maximize the overlap between the metal and ligand orbitals or by enhancing the ligand-field strength, thereby raising the energy levels of the metal-centered states by the use of ligands with strong s-donor and p-acceptor properties such as N-heterocyclic carbenes. [34] However, another emerging complementary method entails in situ formation of photoactive iron-substrate complexes that can undergo VLIH to generate radicals that can initiate the desired reactions (Scheme 3A).
Potassium ferrioxalate has been widely used as asensitive chemical actinometer since the discovery of its photoinduced reduction to ferrous oxalate and carbon dioxide under irradiation at l < 490 nm, as first reported by Parker and Bowen in 1953. [35] In 1986, Sugimori and Yamada reported that the alkylation of pyridine rings with alkyl radicals could be performed through the decarboxylation of alkanoic acids in the presence of visible light instead of g-rays by using ferric sulfate as as toichiometric additive [Scheme 3C,E q. a-i]. [36] Thef ormation of aF e III -alkanecarboxylate complex that absorbed near-ultraviolet visible light and could undergo VLIH followed by decarboxylation to generate the desired alkyl radical as well as the potencyo fF e 2 (SO 4 ) 3 to act as an oxidant were postulated to facilitate the homolysis.J in and co-workers brought this transformation into the catalytic domain in 2019 with the successful photoinduced ironcatalyzed decarboxylative alkylation of heteroarenes [Scheme 3C,Eq. a-ii]. [37] With the effective combination of 5mol % FeSO 4 ·7 H 2 O, 10 mol %2 -picolinic acid as the ligand, and sodium bromate as the exogenous oxidant, aw ide range of alkanoic acids and heteroarenes could be employed to furnish the corresponding products in up to 94 %yield. Thekey step involves the VLIH of aFe III -carboxylate complex to generate the Fe II species and the carboxyl radical, which upon CO 2 extrusion produces the nucleophilic alkyl radical (Scheme 3 Ba). Theexogenous oxidant oxidizes Fe II to Fe III ,which then reenters the catalytic cycle.Inasubsequent report, the same group could extend the scope of the radical decarboxylative alkylation strategy by employing ar ange of Michael acceptors,s uch as alkylidenemalononitriles and azodicarboxylates, to furnish the corresponding products with C À Ca nd C À N bonds,r espectively. [38] Of note,t he electron-deficient radical intermediate generated after the initial addition of the alkyl radical to the Michael acceptor could effectively oxidize Fe II back to Fe III to complete the catalytic cycle,t hereby making Angewandte Chemie the process redox-neutral and obviating the use of any exogenous oxidant [Scheme 3C,Eq. a-iii].
After demonstrating the effectivity of iron(II) and iron-(III)-based photocatalysts in decarboxylative alkylation reactions,Lei, Jin, and co-workers developed an intramolecular C À Ho xygenation of 2-biphenylcarboxylic acids in the presence of ac atalytic amount of Fe(NO 3 ) 3 ·9 H 2 O/ 2,2'bipyridine-6,6'-dicarboxylic acid and two equivalents of sodium bromate as an exogenous oxidant under irradiation at l = 427 nm with an LED to synthesize several benzo-3,4coumarins. [39] Ther eaction proceeds through the intermediacy of an aryl carboxylate-iron(III) complex which undergoes VLIH under the reaction conditions to furnish Fe II and aroyloxy radicals that are almost impervious to decarboxylation at ambient temperature and could easily oxygenate aromatic CÀHb onds.S ubsequently,N aBrO 3 can oxidize the Fe II to Fe III to complete the catalytic cycle [Scheme 3C, Eq. b-i].
Direct visible-light-induced homolysis of ferric halides has recently been exploited by Zhu and co-workers when developing as traightforward and nonhazardous synthesis of ahaloketones from activated olefins. [40] Thecatalytic amount of FeX 3 (X = Br,Cl) used in the reactions undergoes homolytic cleavage under irradiation with visible light to generate Fe II X 2 and ah alogen atom radical that gets trapped by the olefin. Theresulting C-centered radical reacts with oxygen and, upon dehydration, furnishes the desired product. An additional halogen source (KX) and TsOH assist in regenerating FeX 3 to complete the catalytic cycle [Scheme 3C,Eq. c-i].

Cobalt
Vitamin B 12 (cobalamin), an aturally occurring organocobalt complex, has been utilized extensively in organic synthesis for its ability to undergo homolytic cleavage of the

Angewandte Chemie
CoÀCbond to generate C-centered radical species. [41] Different cobalt salts and complexes have been used in conjunction with exogenous photocatalysts to perform various dehydrogenative and C À Co rC -heteroatom bond-forming transformations. [42] Ligand photodissociation of CoH[PPh(OR) 2 ] 4type complexes is aw ell-known feature [43] that has been leveraged in various transformations. [44] Nevertheless,V LIH of CoÀRbonds is observed only when LMCT transitions are within the visible region and depends on the Rm oiety.T he Rovis group has reported that in situ formed photoactive Co II -acetylide species can undergo LMCT excitation upon irradiation (l % 380 nm) to generate an aryl radical cation and aC o I complex without any bond cleavage to assist in as ubsequent oxidative cyclization process (Scheme 4). [45] Thed istinctive catalytic activities of the three oxidation states of four-coordinated cobalt complexes ("supernucleophilic" Co I ,m etalloradical Co II ,o rgano-, and hydro-Co III species) possessing substantial ligand field stabilization energy (LFSE) are characteristic features and are also involved in VLIH events (Scheme 5A). Of note,s everal organo-Co III species are critical intermediates in cobalt-catalyzed transformations involving photoinduced b-hydride elimination [46] as well as the VLIH process. [47] In 2011, the Carreira group reported an intramolecular Heck-type coupling of aryl iodides with olefins catalyzed by ac obaloxime complex, wherein the use of am ild base can deprotonate ah ydridocobalt [Co III -H] intermediate to regenerate the catalytically active Co I species. [48] Photoinduced homolysis of Co III Àalkyl bonds was observed in alkylcobalamins and alkylcobaloximes under irradiation with a1 00 Wh igh-pressure mercury lamp. [49] In 2018, Soper and co-workers carried out the trifluoromethylation of (hetero)arenes with [( S OCO)Co III -(CF 3 )(MeCN)]-type complexes supported by redox-active [ S OCO] pincer ligands. [50] Thet rifluoromethylcobalt(III) complexes could undergo facile VLIH of the Co III ÀCF 3 bond to release the corresponding Co II species and aCF 3 radical for further reactions.T he resulting Co II species can trap HC to generate an unobserved [( S OCO)Co III (H)] intermediate that could produce H 2 and regenerate the Co I species to close the catalytic cycle.In2019, Martin and co-workers demonstrated an efficient approach for the activation of CÀOb onds of alcohols by carbonylating them with Co II -porphyrins to generate alkoxycarbonyl cobalt(III) complexes that could undergo VLIH of the CoÀCb onds (BDE = 39.8 kcal mol À1 ) and subsequent decarboxylation to furnish the corresponding alkyl radicals for trapping. [51] Af ew common Co complexes are shown in Scheme 5C.
In 2016, Gryko and co-workers reported ac obalestercatalyzed olefinic C(sp 2 )ÀHalkylation with diazo reagents as the carbene source. [52] Thek ey mechanistic step involves the VLIH of the alkylcobalester(III) species-formed by the reaction of Co I with ethyl diazoacetate-to generate the Co II and the corresponding a-ester alkyl radical species for further transformation [Scheme 5D,E q. a-i]. Co I is regenerated by the reduction of the hydridocobalester (Co III -H) intermediate.I n2 019, the same group demonstrated the reductive dimerization of 1,1-diphenylethylene in the presence of cobalamin-like catalysts through VLIH of Co III À Cb onds [Scheme 5D,E q. a-ii]w hile studying the role of the nucleotide loop in general cobalamine-catalyzed reactions. [53] Theg eneration of acyl radicals [54] by the VLIH of Co IIIacyl complexes was also achieved by Gryko and co-workers [55] wherein the heptamethyl cobyrinate [(CN)(H 2 O)Cby-(OMe) 7 ], av itamin B 12 derivative,i si nitially reduced to the corresponding supernucleophilic Co I complex that undergoes addition-elimination with 2-S-pyridyl thioesters,t he acyl radical precursors,t of orm the acyl-vitamin B 12 complex. Afterwards,V LIH of the CoÀCb ond furnishes the Co II complex and the acyl radical that participates in the Giesetype acylationo fa ctivated olefins [Scheme 5D,E q. b-i]. In as ubsequent study,t he merger of the alkyl and acyl radical generation capabilities of the same Co III catalyst through VLIH of Co À Cb onds of in situ generated alkylcobalt(III) and acylcobalt(III) complexes was demonstrated, which allowed the consecutive Giese-type alkylation and acylation of electron-deficient olefins to synthesize highly functionalized molecules in as ingle step [Scheme 5D,E q. b-ii]. [56] Of note,the in situ formation of the alkylcobalt(III) complex was faster than that of the acylcobalt(III) complex, which is reflected in the order of the two successive VLIH events in the reaction pathway.
Ar ecent report from the Gryko group involves the visible-light-driven heptamethylcobyrinate-catalyzedG iesetype addition and Co/Ni-catalyzed reductive cross-coupling radical reactions of spring-loaded cyclic reagents. [57] The mechanistic pathway involves the initial formation of the Co III -alkyl complex intermediate by the reaction of the "supernucleophilic" Co I form of the catalyst and the electrophilic bicyclic reagents.The Co III -alkyl complex subsequently undergoes visible-light-induced homolysis to generate the Co II complex and the corresponding alkyl radicals that further engage in different radical transformations,s uch as the addition to SOMOphiles or transition-metal-catalyzed radical cross-coupling reactions [Scheme 5D,Eq. a-iii].
Ar egioselective coupling reaction of epoxides and aziridines with alkenes in the presence of as imple cobalt dimethylglyoximate complex has been developed by Prina Cerai and Morandi to synthesize value-added homoallylic alcohols and amines [Scheme 5D,E q. a-iv]. [58] Thek ey mechanistic steps of the transformation involve the nucleophilic opening of the epoxide/aziridine ring with Co I and the VLIH of the Co III ÀCb ond from the corresponding Co III Scheme 4. Photo-and Co-acetylide-catalyzed [2+ +2+ +2] cycloaddition reaction.

Angewandte
Chemie intermediate to generate Co II and carbon-centered radical species.T he catalyst is regenerated with the help of the basic intermediate (e.g.alkoxide), which can deprotonate the Co III -Hs pecies.T he method successfully addressed the inefficiencies of the previous method reported by Harrowven and Pattenden by obviating the use of stoichiometric amounts of cobalt, base,and reductant. [59]

Cerium
Besides the ever-increasing use of first-row 3d transitionmetal complexes,interest has mounted substantially in recent times in the use of earth-abundant lanthanide complexes.
Being the 26th most abundant element, cerium has found extensive use in photocatalysis and warrants discussion in the context of VLIH. In aseries of seminal studies conducted by Schelter and co-workers,s everal luminescent cerium(III)based complexes have been used as both inner-sphere [60] and outer-sphere potent single-electron photoreductants. [61] Their ability to absorb in the visible-light region and undergo interconfigurational doublet-to-doublet, parity,a nd spin-allowed 4f!5d metal-centered electronic transitions,t hereby minimizing the loss of energy from the long-lived 2 De xcited states,p rovides au nique profile for application as photocatalysts. [60][61][62] Whereas the excited-state Ce III metalloradical (5d 1 )has been exploited for the abstraction of achlorine atom from benzyl chlorides to generate benzyl radicals,t he Scheme 5. Mechanistic features of the VLIH of Co III species and selected transformations.

Angewandte
Chemie chloride-Ce IV LMCT excitation has also been leveraged in parallel photo-oxidation processes involving C À Ca nd C À heteroatom bond-forming reactions. [62d] Thephotocatalytic properties of Ce III chloride complexes in aC ÀCb ond-cleavage and amination reaction of cycloalkanols was first reported by Zuo and co-workers in 2016. [63] However,amechanistically different and complementary catalytic manifold of cerium photocatalysis was unveiled in aseries of reports by the same group, [9d,64] wherein the general mechanistic archetype involves:a )initial single-electron oxidation of the Ce III L n complex by an oxidant to generate an intermediary Ce IV L n complex;b )coordination of an alcoholic ligand to form aL n Ce IV -OR complex;a nd finally c) VLIH of the L n Ce IV -OR species through photoinduced LMCT excitation and subsequent homolysis to generate oxygen-centered radicals and Ce III L n species to complete the catalytic cycle. Theincipient, reactive alkoxide radical can then participate in av ariety of transformations,s uch as intramolecular or intermolecular hydrogen atom abstraction (HAT), addition to another functional group,orhomolytic bscission (Scheme 6A). [65] Based on these mechanistic features,Zuo and co-workers developed am ethod for the efficient d-selective C À Hb ond functionalization of protecting-group-free primary alcohols in the presence of 1mol %C eCl 3 and 5mol % nBu 4 NCl. [9d] In this process,t he VLIH of the Ce IV -OR complex under irradiation with visible light was carried out to generate atransient alkoxy radical that undergoes athermodynamically favored intramolecular 1,5-HATtoform ahighly nucleophilic alkyl radical. Subsequent addition of this radical to DBAD and aS ET reduction of the N-centered radical furnishes the desired product and regenerates the Ce IV catalyst [Scheme 6C,Eq. a-i]. Thestrategy of combining the VLIH of Ce IV -OR complexes and intermolecular HATwas later expanded by the same group to valorize low molecular hydrocarbon feedstocks (C n H 2n+2 , n = 1-4, Cy;BDE(C-H) = 105 kcal mol À1 for CH 4 )b ys uccessfully achieving CÀHa mination with DBAD,C ÀHa lkylation with electron-deficient alkenes,a nd Minisci-type CÀHh eteroarylation [Scheme 6C,E q. a-ii]. [64a] In 2020, Zuo and co-workers extended this strategy to the C(sp 3 ) À Hf unctionalization of hydrocarbons [Scheme 6C, Eq. a-iii]. [64c] Of note,s teady-state homolysis experiments and transient absorption spectroscopic studies revealed that the VLIH event involving the Ce IV -OMe complex was not the rate-determining step in CÀHa mination and alkylation processes.
In ad ifferent mechanistic approach, an effective merger of the VLIH of Ce IV -alkoxide complexes with asubsequent b-C À Cscission of the alkoxy radical species has been achieved to develop arange of useful transformations.In2018, Zuo and co-workers reported atom-and step-economic formal cycloadditions of cycloalkanols with alkenes to afford bridged lactone scaffolds. [64b] Thek ey step of the developed transformation entails the VLIH of aC e IV -OR complex to generate as econdary alkoxy radical, which subsequently undergoes ar apid b-scission process to form an ucleophilic alkyl radical that adds to an electron-deficient alkene.T hen, aS ET process for the reduction of the generated a-acyl radical by photoexcited 9,10-diphenylanthracene (DPA; E 1/2 = À1.77 Vv ersus SCE in CH 3 CN), an intramolecular aldol reaction of the enolate,a nd acidification furnishes the desired bridged lactone product [Scheme 6C,Eq. b-i]. When 1,2-diols were employed as the substrate instead of alcohols, oxidative cleavage of the CÀCb ond was observed and the corresponding aldehydes were obtained in very high yields [Scheme 6C,E q. b-ii]. [66] Following this approach, Zuo and co-workers developed ad ihydroxymethylation strategy wherein primary alcohols were converted into alkyl radicals with the loss of one molecule of formaldehyde,w hich underwent 1,4-conjugate additions with Michael acceptors [Scheme 6C, Eq. b-iii]. [64d] Notably,adouble-excitation mechanism was proposed for the transformation, as it was observed that irradiation with LEDs at l = 365 nm could induce excitation of L n Ce III -OR as well as ultraviolet-induced homolysis (UVLIH) of L n Ce IV -OR complexes,w hereas ac erium/DPA dual photocatalytic system had to be employed under irradiation with LEDs at l = 400 nm as it could only effect the VLIH of the intermediary L n Ce IV -OR complex.
Recently,Zhang and co-workers successfully achieved the selective cleavage of C a À C b bonds in various lignin model compounds in the presence of 2mol %C eCl 3 and 5mol % nBu 4 NCl under irradiation with visible light (l = 460 nm). [67] Them echanistic pathway entails VLIH of the L n Ce IV -lignin species coordinated through the benzylic a-hydroxy group (a-OH). TheV LIH-generated alkoxy radical intermediate enables cleavage of the C a À C b bond to ultimately furnish the corresponding aldehydes (up to 97 %) and the hydrazinium derivatives (up to 95 %) by amination with DBAD [Scheme 6C,E q. b-iv].Z uo and co-workers extended the activation strategy from alcohols to ketones through an effective merger of Ce IV -VLIH and Lewis acid catalysis to selectively cleave C À Cb onds of various acyclic and cyclic ketones and install different functional groups at the incipient acyl and alkyl radicals. [68] Ther eaction proceeds through the activation of the carbonyl group by TiCl 4 and nucleophilic addition of TMSCN to form the corresponding cyanohydrin derivative.T hen, VLIH of the coordination complex formed between Ce IV and cyanohydrin results in the formation of Ce III species and an O-centered radical species that undergoes bscission of the C À Cbond to form adistal C-centered radicalaprocess facilitated by the release of ring strain in the case of small cyclic ketones.F inally,o rthogonal selective functionalization of both acyl and alkyl radicals with diisopropyl azodicarboxylate (DIAD) in the case of small and mediumsized cyclic as well as acyclic ketones followed by aP ET process with DPAfurnish the desired products in high yields [Scheme 6C,E q. b-v].T he same group has developed as traightforward ring-expansion strategy of cyclic alkoxyketones to synthesize 9-to 19-membered macrolactones in the presence of acerium salt and cyanoanthracene under aerobic conditions through irradiation with visible light [Scheme 6C, Eq. b-vi]. [69]

Angewandte
Chemie ketone/lactol tautomeric equilibrium. In ad ifferent set of transformations,effective decarboxylation of alkyl carboxylic acids has been achieved with Ce photocatalysis,w herein the key mechanistic step involves the VLIH of cerium-carboxylate complexes of the Ce IV -O(CO)R type to form Ce III and highly reactive alkylcarboxyl radical species that undergo facile decarboxylation to generate the corresponding alkyl radicals for further transformations. [70] In 2019, Kçnig and coworkers utilized this property for the decarboxylative hydrazination of 18 8,2 8 8,a nd 38 8 carboxylic acids in the presence of 10 mol %C eCl 3 ·7 H 2 Oa nd 20 mol %C s 2 CO 3 under irradiation with blue LEDs,w herein the corresponding VLIHgenerated alkyl radicals were trapped with DBAD to furnish the hydrazine derivatives in 28-90 %yield [Scheme 6C,Eq. ci] . [70a] In 2020, Tsurugi, Satoh, Mashima, and co-workers performed the decarboxylative oxygenation of aliphatic carboxylic acids in the presence of 5mol %Ce(O t Bu) 4 under air at atmospheric pressure to obtain products containing CÀ Obonds such as aldehydes and ketones in up to quantitative yields. [70b] Thet ransformation proceeds via the formation of ah exanuclear oxocerium(IV) carboxylate complex cluster, Ce 6 O 4 (OH) 4 (OCOR) 12 ,f rom the reaction between Ce-(O t Bu) 4 and carboxylic acids.T his hexanuclear Ce IV species undergoes VLIH under irradiation with blue light and forms Ce III species and ac arboxyl radical that further engages in decarboxylation and oxygenation to form the corresponding alkyl peroxyl radical (RCH 2 OOC). This radical forms alkylperoxo-Ce IV species,f rom which alkyl hydroperoxides are formed that undergo dehydration to finally afford the corresponding aldehydes as the terminally oxidized major products along with minor amounts of the corresponding alcohols [Scheme 6C,E q. c-ii].
Recently,S ong, Xu, and co-workers have brought the decarboxylative alkylation of heteroarenes with aliphatic carboxylic acids and cerium into the electrophotocatalytic domain, [70c] wherein the reaction is initiated by the anodic oxidation of Ce III to Ce IV ,w hich coordinates with the carboxylic acid. As ubsequent VLIH of the Ce IV -carboxylate complex and decarboxylation forms the corresponding alkyl radical that adds to the heteroarene in aMinisci-type reaction to afford the alkylated product in good to high yields [Scheme 6C,E q. c-iii]. Ap hotocatalytic method for the dehydrogenative lactonization of 2-arylbenzoic acids in the presence of CeCl 3 as the photocatalyst and O 2 as the terminal oxidant has been developed by Yatham and co-workers. [71] In this process,aCe IV -aryl carboxylate complex is formed by the coordination of the aryl carboxylic acid with Ce IV ,a nd subsequent VLIH generates aC e III species and the corresponding aryl carboxyl radical, which gets trapped by the aryl substituent without undergoing decarboxylation and eventually furnishes the lactonized product in very high yields [Scheme 6C,E q. c-iv].

Miscellaneous Examples
Although not explored for synthesis in as much detail as the specific cases discussed previously,t here is increasing recognition of the potential to develop photocatalytic trans-formations with other transition-metal-based photocatalysts such as vanadium, chromium, manganese,o rp alladium by capitalizing on their ability to undergo VLIH. VLIH of Mnalkyl bonds has been studied with various Mn(CO) 5 R-type complexes. [72] Photoinduced homolytic cleavage of arange of paramagnetic Cr III monohydrocarbyl complexes with the general formula CpCr[(ArNCMe) 2 CH](R) has also been studied in detail. [73] Wang and co-workers have recently reported vanadium-(V)-catalyzed visible-light-driven selective C a À C b bond cleavage of b-1 interlinkages of lignin models to afford valuable aromatic products (Scheme 7a). [74] Theproposed mechanistic pathway for this transformation entails the initial coordination of the benzylic hydroxy group to the vanadium center. Then, excitation of the resulting complex with visible light induces an LMCT process and reduction of the vanadium center, which in turn causes the homolytic cleavage of the C a À C b bond to produce benzaldehyde and ab enzyl radical for further reaction. Torres et al. have reported av isible-lightdriven palladium-catalyzed carbonylation reaction to synthesize acid chlorides from aryl halides,w herein the irradiation with light assists in the initial radical-induced oxidative addition of Pd 0 as well as excitation of the Pd II intermediate that subsequently undergoes photoinduced reductive elimination. [75] Interestingly,o ne of the possible pathways for the last mechanistic step involves VLIH of aP d Àacyl bond to generate the incipient acyl radical in the reaction medium, which has been probed by trapping experiments.
TheV LIH concept has also been successfully applied to organic catalyst-substrate complexes.M elchiorre and coworkers recently used an ucleophilic dithiocarbamate anion catalyst with an attached chromophoric unit to activate various alkyl electrophiles bearing different leaving groups through an S N 2p athway.T he resulting photon-absorbing intermediate undergoes VLIH to generate C-centered radicals which can, thereafter,p articipate in various C À Cb ondforming reactions (Scheme 7b). [76]

Summary and Outlook
As can be gleaned from the examples discussed in this Minireview,b yu tilizing the VLIH concept with earthabundant transition-metal-based photocatalysts it is possible to transcend the limits of traditional photocatalysts that demonstrate only outer-sphere electron-transfer and energytransfer processes from their excited, populated, and emissive MLCT states.T he concept has also been successfully applied within the wavelength range of 350-400 nm, where the event could be termed as UVLIH. Although the majority of the VLIH processes that have found successful applications in organic synthesis involve dissociative LMCT of different metal-substrate complexes to generate targeted radicals, evermore variants of electronic transitions such as dissociative 1 MLCT/ 3 d-d or dissociative SBLCT transitions are also being recognized, together with the continuous advancement in the field of sophisticated spectroscopic methods and computational studies to determine the intermediate radicals species and complexes.W ith the creative exploitation of the VLIH activation mode,new synthetic processes are possible, wherein the reaction pathways will be directed by the innersphere mechanism of the sustainable photocatalysts and should, in turn, allow the development of enantioselective approaches and the generation of new radical species by the selective homolysis of metal-substrate bonds.W ea re confident that alternative modes and applications will be discovered through the effective collaboration of synthetic organic chemists and spectroscopists,a nd that the so far discovered methods will find wide applications in both academic and industrial set-ups.