# Fileset

[s41467-025-63515-x-1.pdf](https://mdr.nims.go.jp/filesets/8d275dcf-31ba-48ad-a3a2-7d6b05c64ea0/download)

## Creator

Longhui Yu, Shangzhao Li, Hiroshige Ogawa, Yilin Ma, Qing Chen, Ken Yamazaki, [Yuuya Nagata](https://orcid.org/0000-0001-5926-5845), Hugh Nakamura

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Fe-electrocatalytic deoxygenative Giese reaction](https://mdr.nims.go.jp/datasets/938e322b-4061-459f-8125-855fa10adacc)

## Fulltext

Fe-electrocatalytic deoxygenative Giese reactionArticle https://doi.org/10.1038/s41467-025-63515-xFe-electrocatalytic deoxygenative GiesereactionLonghui Yu1,5, Shangzhao Li1,5, Hiroshige Ogawa1,5, Yilin Ma1, Qing Chen 1,Ken Yamazaki 2, Yuuya Nagata 3,4 & Hugh Nakamura 1A redox-neutral Fe-electrocatalytic deoxygenative Giese reaction is reported.Hydroxyl groups are among the most abundant functional groups, and thus,the development of efficient reactions for their conversion has significantimportance in medicinal and process chemistry. Here, we present a redox-neutral Giese reaction via anodic oxidation to generate phosphonium ions incombination with a cathodic reduction to yield low-valent Fe-catalysts. Thisreaction represents a promising example of a redox-neutral reaction using anFe-catalyst and electrochemistry. The results obtained in this study will facil-itate the exploration of a wide range of novel reactions employing this redoxcycle in the future.The development of efficient transformations of abundant functionalgroups, such as the hydroxyl group, constitutes a central topic incontemporary synthetic organic chemistry. For example, among thenumerous C‒C-bond-formation reactions, the indispensabletransition-metal-catalyzed cross-coupling reactions are generally themost reliable1–3, but typically require the use of halogenated substratesand a two-step process involving a halogenation step and the forma-tion of the C‒C bond. In this context, several efficient C‒C-bond-for-mation reactions based on the conversion of the hydroxyl group haverecently been reported (Fig. 1A)4–6. In 2018 and 2022, Suga and Ukajireported direct conversion reactions of alcohols using titaniumreagents7,8. These reactions are applicable to primary, secondary, andtertiary alcohols and are particularly useful for the transformation ofaliphatic alcohols. In 2020, Wang and Shu reported the first C‒O bondcleavage of tertiary alcohols using Cp*TiCl3 as a catalyst9. Moreover,thedevelopment of sustainable chemical reactions thatdiminishwasteusing photoredox chemistry10–13 and electrochemistry14–33 has recentlygained momentum. In 2021, Li et al. reported a nickel-catalyzeddehydroxylative cross-coupling reaction based on electrochemistry34.This reaction is an excellent way to directly form C(sp2)‒C(sp3) bondsfrom primary and secondary alcohols. From 2021 to 2023, MacMillanet al. reported direct alcohol-conversion reactions using NHCs, pho-toredox, and nickel or iron catalysts35–40. This reaction is suitable forprimary to tertiary alcohols.Nickel catalysts have been exploited intensively recently due totheir abundance and low toxicity. Nickel is a relatively electropositivelate transition metal and readily facilitates oxidative addition, whichallows the use of cross-couplings between less-reactive reactants34–40.However, due to the highly reactive nature of low-valent nickel spe-cies, controlling their reactivity can prove challenging. Furthermore,nickel-catalyzed reactions sometimes require the use of a glove box,which can be a limitation for the development of practicalapplications.As with nickel catalysis, the development of iron-based catalysishas been very active in this field41–59. Iron is the most abundant tran-sition metal, minimally toxic, and has the potential for unique andcomplementary modes of reactivity. Additionally, compared to nickelcatalysis, iron catalysis is easier to handle, which enables more prac-tical applications. Despite these virtues, reports of C-C bond formationreactions using iron are limited compared to other transition metalslike palladium, copper, nickel, and cobalt. Moreover, many C-C bondformation reactions using Fe-catalysts require strong nucleophilessuch as Grignard reagents, which presents challenges in terms offunctional group compatibility41–59. Regarding electrocatalytic reac-tions, only oxidative reactions have been reported so far, which is acurrent limitation16. In this context, we report here the direct forma-tion of C–C bonds using a redox-neutral Fe-electrocatalytic deox-ygenative Giese reaction.Received: 8 February 2024Accepted: 21 August 2025Check for updates1The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong SAR, China. 2Okayama University, Tsushimanaka, Okayama, Japan. 3WPIInstitute forChemical ReactionDesign andDiscovery (WPI-ICReDD),HokkaidoUniversity, Sapporo, Japan. 4AutonomousPolymerDesign andDiscoveryGroupResearch Center for Macromolecules and Biomaterials, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba, Ibaraki, Japan. 5These authorscontributed equally: Longhui Yu, Shangzhao Li, Hiroshige Ogawa. e-mail: NAGATA.Yuuya@nims.go.jp; hnakamura@ust.hkNature Communications |         (2025) 16:8379 11234567890():,;1234567890():,;http://orcid.org/0000-0003-3106-9281http://orcid.org/0000-0003-3106-9281http://orcid.org/0000-0003-3106-9281http://orcid.org/0000-0003-3106-9281http://orcid.org/0000-0003-3106-9281http://orcid.org/0000-0002-2039-4321http://orcid.org/0000-0002-2039-4321http://orcid.org/0000-0002-2039-4321http://orcid.org/0000-0002-2039-4321http://orcid.org/0000-0002-2039-4321http://orcid.org/0000-0001-5926-5845http://orcid.org/0000-0001-5926-5845http://orcid.org/0000-0001-5926-5845http://orcid.org/0000-0001-5926-5845http://orcid.org/0000-0001-5926-5845http://orcid.org/0000-0001-5475-7883http://orcid.org/0000-0001-5475-7883http://orcid.org/0000-0001-5475-7883http://orcid.org/0000-0001-5475-7883http://orcid.org/0000-0001-5475-7883http://crossmark.crossref.org/dialog/?doi=10.1038/s41467-025-63515-x&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-025-63515-x&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-025-63515-x&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-025-63515-x&domain=pdfmailto:NAGATA.Yuuya@nims.go.jpmailto:hnakamura@ust.hkwww.nature.com/naturecommunicationsResultsFirst, an Fe-catalyzed redox-neutral Giese-type reaction was investi-gated using 4-phenyl-2-butanol (1) and 4-tert-butylstyrene (2) as thesubstrates (Fig. 1C). Themost challenging aspect of this reaction is thatthe halogenation of the alcohol at the anode and the reduction of theFe-catalyst at the cathode must proceed at appropriate reaction ratesand potentials. Hundreds of combinations of metal catalyst, ligand,base, phosphine, halogen source, electrolyte, electrode, and currentvalues were investigated to optimize the present redox-neutral Giese-type reaction. To determine the optimal conditions for alcohol halo-genation at the anode in the present reaction, the conditions from theNi-catalyzed paired electrolysis approach pioneered by Li et al. wereused in the initial attempt34. Detailed investigations revealed that thebest results were obtained using FeCl2 (15mol%), IPr·HCl (L1) (15mol%), PPh3 (4 eq.), TBAB (4 eq.), DIPEA (4 eq.), and nBu4NBF4 (0.5 eq.) inDMA60,61. In terms of the electrochemical conditions, a current of 6mA(0.2mmol scale, constant current) was effective at room temperaturein an undivided cell, and the redox-neutral Giese reaction of secondaryalcohol 1 and styrene derivative 2 was found to proceed with 60%isolated yield using carbon plates as the anode and Ni foams as thecathode.Direct control experiments revealed that the desired product (3)was not obtained in the absence of an electric current or Fe-catalyst(entries 1–2). Moreover, the yield was significantly reduced to 10%when a sacrificial Zn electrode was used as the anode instead of acarbon plate (entry 3). Based on these results, it can be concluded thatthe oxidation at the anode is necessary for this reaction.When a higherconstant current (9mA) was employed, the yield was dropped to 41%(entry 4). It may be attributed to the discrepancies in reaction rates ofeach subprocess due to the higher current. Similarly, decreased cata-lyst loading (10mol%) gave the inferior yield (entries 5). Different Fe-catalysts were also investigated. When Fe(acac)2 was applied, thereaction proceeded, albeit in only 34% yield (entry 6). This wasattributed to the fact that Fe(acac)2 has two acac ligands coordinatedto Fe,making it veryunfavorable for other ligands to coordinate.WhenFeCl3 was used, the yield decreased to 37% (entry 7), probably becauseFeCl3 is more hygroscopic than FeCl2; the small amount of water in thereaction system may be the cause of the low yield. Another set ofcontrol experiments was performed to investigate the ligand effect ofthis reaction. Only a 20% yield of 3 was obtained under ligand-freeconditions (entry 8). Then, a range of alternative ligands (L2–L7) werescreened, but these proved to be ineffective for this reaction (entries9–14). Several other NHC ligands with varying steric and electronicproperties were also examined, but all gave inferior results (see Sup-plementary information, Figure S16). There appears to be a trendindicating that the use of electron-rich and bulky ligands correlateswith increased yields62. In case of the bipyridine ligand, substituentadjacent to the nitrogen atom seems to have an adverse effect on thecatalytic activity.It is worth noting here that the presence of the electrolytenBu4NBF4 (0.5 eq.) in this reaction promoted a higher yield (entry 15).Next, the halogen source, which also plays a dual role as the electro-lyte, was investigated. The yield decreased when TBAI or NaI was used(entries 16–17). This is probably due to the high reactivity of the alkyliodide generated in the reaction system, which may cause side reac-tions such as reduction and elimination. Furthermore, the reaction didnot proceed when PPh3 was omitted (entry 18). To study the solventeffect, NMP, which is frequently employed inelectrochemical reac-tions, was tested, but caused the yield to drop to 16% (entry 19). Finally,when the reaction was performed in air, the yield fell below 5%(entry 20).Based on the above screening results, it was demonstrated that anFe-catalyst is effective for this electrochemical deoxygenative Giesereaction. To elucidate why the iron catalyst was so effective, attemptswere made to perform this reaction using other metals (Ni, Co, Cu, Ti,etc.). The results showed that metal catalysts other than iron (Ni, Co,Cu, Ti) were not effective for this reaction (see Supplementary infor-mation, Figure S14). This ineffectiveness is attributed to the formationof byproducts such as the reduced form of alcohol 1 and dimers whennon-iron metal catalysts were used. Furthermore, inspired by thereaction conditions reported by Li et al.34 in 2021 for an electro-chemical nickel-catalyzed dehydroxylative cross-coupling reaction,the deoxygenative Giese reaction of alcohol 1 was attempted using aNi-catalyst. However, despite screening various ligands and condi-tions, the reactiondid not proceedwith theNi-catalyst, and the desiredcompound was scarcely obtained (see Supplementary information,Figure S15). One reason is the difficulty in controlling the process, aslow-valent Ni-catalysts generally accelerate the oxidative additionprocess. Indeed, the byproducts when using Ni-catalysts in this reac-tion included the reduced, halogenated, and elimination products ofalcohol 1, while the desired compound was present only in traceamounts (see Supplementary information, Figure S15). Consideringthe redox potential of the nickel complex, the reduction of nickel inthis electrochemical system is indeed feasible. Consequently, thesuboptimal results can be attributed to this excessive reactivity of theNi-complex.With the optimal conditions in hand, we investigated thesubstrate scope. First, we screened different Michael acceptorsusing 1 as the alcohol and found that a variety of styrene deri-vatives were applicable as substrates (Fig. 2A). Substrates withalkyl groups such as tBu and Me on the aromatic ring easilyprovided the desired compounds (3, 4). When styrene was used,the reaction furnished the desired compound (5) in 72% isolatedyield. A gram-scale experiment revealed that this reaction canprovide 5 in 75% isolated yield. The electrochemical deox-ygenative Giese reaction also proceeded using styrene derivativeswith electron-withdrawing groups such as fluorine and chlorine.Good yields of the desired products were obtained using fluorine-substituted styrenes (6, 7), while a slight decrease in yield wasobserved for styrene derivatives with chlorine substituents(8–10). When the reaction was attempted using styrene deriva-tives substituted with bromine and iodine, dehalogenation wasobserved, and the desired products were not obtained. Thereaction proceeded also for styrene derivatives with differentelectron-withdrawing ester groups (11–12). The reaction was alsoapplicable to other styrene derivatives, such as 1-vinyl naphtha-lene, 2-vinyl naphthalene, and 4-vinyl biphenyl (13–15). When thedisubstituted olefin 1,1-diphenylethylene and α-methylstyrenewere used (16–17), the desired products were obtained. As thefunctional-group transformation of heterocyclic compounds isparticularly important in medicinal chemistry, we applied thisreaction to obtain heterocyclic compounds, and the electro-chemical deoxygenative Giese reaction was found to proceed forpyridine, thiazole, thiophene, and other heterocycles (18–21).Interestingly, the reaction also proceeded for ferrocene andprovided the corresponding product (22) in 23% isolated yield.Next, acrylate derivatives and other Michael acceptors wereinvestigated (Fig. 2B). The results showed that various acrylate deri-vatives, including methyl methacrylate, can be applied in this reaction(23–29). To confirm the applicability of different functional groups,this reaction was also tested using Michael acceptors with aminogroups and found to be applicable to substrates such as 2-(diethyla-mino)ethyl methacrylate and 2-(dimethylamino)ethyl methacrylate(30, 31). Furthermore, the reaction proceeded well with various acry-late derivatives, including cyclic and acyclic acrylates (32–37). Theelectrochemical deoxygenative Giese reaction also works well foramides such as 38. A further investigation of different Michaelacceptors revealed that the reaction also proceeds well using diethylvinyl phosphonate (39), which is of great significance for diversitysynthesis.Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 2www.nature.com/naturecommunicationsVarious primary alcohols were investigated using either tert-butylmethacrylate (40) or styrene (41) as theMichael acceptor (Fig. 3A). Thereaction proceeded well with primary alcohols irrespective of thepresenceof various electron-donating and -withdrawing groups on thearomatic ring (42–52; 62–64). The deoxygenative Giese reaction alsoproceeded well with alcohols that contain heterocycles such as pyr-idine rings and with other aliphatic primary alcohols (53–59). Thereaction was also effective using primary alcohols derived fromimportant pharmaceuticals such as ibuprofen and naproxen (60, 61).The reaction was also applicable to a wide range of differentsecondary alcohols (Fig. 3B). When 2-hydroxyindan was used in thisreaction, the target product (65) wasobtained in 50% isolated yield.Onthe other hand, 1-hydroxyindan and 1,2,3,4-tetrahydro-1-naphthol,which have an alcohol at the benzoic position, gave the target com-pounds in lower yield (66, 67). The deoxygenative Giese reaction alsoproceeded with substrates such as cyclohexanol, cycloheptanol, andcyclooctanol (68–70). A further investigation of the substrate scopefor this reaction demonstrated that the desired compounds could alsobe obtained by coupling secondary alcohols with styrene (41) (71–74).In addition, the reactionwas also applicable to steroidal skeletons,which have a variety of biological activities and are important in drugdevelopment (75, 76). It is worth noting that 75 can be synthesized onthe gram scale. Finally, we confirmed the applicability of the presentreaction to a wide range of heterocyclic compounds (77–83); suchcompounds are extremely important building blocks in medicinalchemistry. These results indicate that this reaction can introduce abroad variety of Michael acceptors to a variety of primary andsecondary alcohols (Figs. 2 and 3). For the low-yield compounds inFigs. 2 and 3, mainly debrominated (reduced) compounds wereobserved as by-products. This is because the Appel reaction at theanode electrode proceeds without problems, but the compounds withslow C-C bond formation by the Giese reaction are gradually reducedby the cathodic reduction. In addition, some compounds containingheteroatoms such as nitrogen and sulfur are easily oxidized, and it ispresumed that anodic competition with Shono oxidation-type reac-tions has resulted in lower reaction yields.Then, mechanistic studies were carried out to elucidate theunderlying reaction mechanism (Fig. 4A). First, cyclopropylmethanol(84) and 4-vinyltoluene (85) were reacted using the standard condi-tions for this reaction, which furnished the radical ring-opening pro-duct 86. The Giese reaction of the chiral compound 87 with 41afforded racemic compound 77. These experimental results suggestthat this reaction follows a radical reaction mechanism.At the same time, control experiments were performed using (3-bromobutyl)benzene (88), a putative intermediate of this reaction(Fig. 4B). First, we attempted the reaction between 88 and 4-tert-butylstyrene (2) under the standard conditions, except that no elec-trodes or current was used and Zn dust (10 eq.) orMndust (10 eq.) wasapplied as a chemical reductant. However, the desired compound (3)was not obtained using Mn and Zn dust. The starting material 88werecompletely recovered and no side reactions such as dehalogenation orelimination were observed. We also attempted this reaction under thestandard conditions with a (+)C plate / (–)Ni foam, but no currentflowed (0mA) and the desired compound (3) was not obtained. BasedOTiO ClClR3Suga & Ukaji (2018, 2022)7-8Li (2021)34MacMillan (2021, 2023)35-40tBubpy (cat.) NiBr2 (cat.) PPh3, LiBrNMP, rt, 4 mA (+)C / (–)NiR3OHR1R2R3R1R2RR3OHR1R2(1° & 2° alcohols)R3R1R2EWG(2.2 eq.)Zn (4 eq.)HCl•Et3N (1 eq.)DMA, 70�140 °C, 24 hEWG++[Ni] or [Fe] (cat.)blue LEDsR3OHR1R2R3R1R2(1°, 2°, & 3° alcohols)+RBrFeCl2 (15 mol%)PPh3 (4 eq.), TBAB (4 eq.), DIPEA (4 eq.)nBu4NBF4 (0.5 eq.)DMA (2.5 ml), rt, 6 mA, 20 h(+)C plate / (–)Ni foam[Fe-electrocatalytic deoxygenative Giese reaction]IPr•HCl (L1) (15 mol%)2 (5 equiv.)N NMeMeMe Me MeMeMe MeClL1N NMeMeMeMeClL6MeMe N NClL7L2entry deviations from above yieldsb12345678910no electricityw/o FeCl2Zn instead of Carbon plate9 mA, 12hcat. loading: 10 mol%Fe(acac)2 instead of FeCl2FeCl3 instead of FeCl2w/o L1L2 instead of L1L3 instead of L10%0%10%41%24%34%37%20%10%34%NNH2HNNNHNL5PhMeOHtBu+1 (1 equiv.) Ph MetBu11121314151617181920L4 instead of L1L5 instead of L1L6 instead of L1L7 instead of L1w/o nBu4NBF4TBAI instead of TBABNaI instead of TBABw/o PPh3NMP instead of DMAopen to air0%41%42%<5%36%44%23%0%16%<5%aReactions were conducted at a 0.2 mmol scale in 2.5 mL of DMA. bYields were determined by 1H NMR with pyridine as the internal standard. cIsolated yield.entry deviations from above yieldsb3, 62%b, 60%cR2OHR1R2R1XO(1° & 2°alcohols)(+)C  / (–)Ni constant current undivided cell, rt[Fe-electrocatalytic deoxygenative Giese reaction]R2R1ArXO+Fe P(lll)90 examplesB. This workN NL3: R1 = OMe, R2 = HL4: R1 = H, R2 = MeR1 R1R2 R2RR rA•mild conditions•room temperature[abundant metal][undivided cell][redox-neutral]RBr(1°, 2°, & 3° alcohols)undivided cellNHC;R1R5R4R4Br R5(1°, 2°)or orR2RPC (cat.)A. Representative deoxygenative C-C bond formation from alcoholsorWang & Shu (2020)9R3OHR1R2R3R1R2EWGCp*TiCl3 (cat.)TESCl, ZnTHF, 60 °C, 12 hEWG+(3° alcohols)C. OptimizationaFig. 1 | Evolution of deoxygenative C-C bond formation from traditionalapproaches to redox-neutral Fe-electrocatalytic deoxygenativeGiese reaction:context, concept, and optimization. A Overview of existing strategies fordeoxygenative C–C bond formation from alcohol. B Concept of the Fe-electrocatalytic deoxygenative Giese reaction. C Optimization of the reactionconditions.Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 3www.nature.com/naturecommunicationson these results and the aforementioned control experiments (Fig. 1C,entry 1), it can be concluded that the application of electrochemistry isessential for this reaction. A further control experimentwas performedto investigate the effect of FeCl2 on this reaction (Fig. 4B, bottom).Here, sacrificial anodes ((+)Zn plates) were used instead of (+)C platesunder the standard reaction conditions, and 3was isolated in 19% yieldin the absence of FeCl2 and IPr·HCl. However, the yield was improved(61% isolated yield) under these conditions when FeCl2 and IPr·HClwere used. These experimental results indicate that FeCl2 and IPr·HClare crucial in this reaction system.In their entirety, themechanistic studies and control experiments,allow proposing a feasible reaction mechanism, which is shown inFig. 4C. First, as reported by Li et al.34, X- (X = Br or I) is oxidized to Br2or I2 at the anode. Subsequently, the resulting Br2 or I2 reactswith PPh3,XYOP OEtOOEtSMe MeFe18, 62% 21, 38%27, 53%(dr = 1:1.3)N NMeMeMeMeMeMeMe MeClMeMe2NNEt2OEtOMe33, 65%MeOOOOMe MeMeN19, 59%NOO MeMeOEtOOEtPOMeOOMeMeOMeMeOOOMeMeOOOMeOOMeMeMeMeOtBuOMeOOMeMe23, 80%(dr = 1:1)MeOMeOMe28, 46%(dr = 1:1.4)36, 45%30, 41%(dr = 1:1.3)37, 60%F FF24, 83%(dr = 1:1.1)31, 31%(dr = 1:1.7)OtBuOMe34, 59%MeMe MeOOMe26, 63%MeMe MeOO25, 76%(dr = 1:1)38, 44%29, 43%(dr = 1:1.2)35, 60%(dr = 1:1.3)39, 40%OMeOMe32, 47%• abundant metal• room temperature• heteroarenes• redox-neutralMeNMe15, 48%NSMeMe20, 64%Me22, 23%14, 53%FeCl2 (15 or 20 mol%) IPr•HCl (L1) (15 or 20 mol%)PPh3, TBAB, DIPEA nBu4NBF4 (0.5 eq.)DMA (2.5 ml), rt, 6 mA, 20h(+)C plate / (–)Ni foamundivided cell[Fe-electrocatalyticdeoxygenative Giese reaction]L1ArXOR1 (1 equiv.)OHMe+MeR rAMeY(X = OR, NR2)(Y = alkyl)or(5 eq.)B. Scope of acrylate derivatives & vinylphosphonatedA. Scope of styrene derivativesbMe13, 65%(X = OR, NR2)(Y = alkyl)0.2 mmol scale3, R = tBu, 60%4, R = Me, 50%5, R = H, 72% (75% (gram scale))MeMeRF6, meta-F, 75%7, para-F, 61%MeCl8, ortho-Cl, 35%c9, meta-Cl, 34%c10, para-Cl, 38%c11, R = CO2Me, 39%12, R = OAc, 61%MeRRMe16, R = Ph, 48%17, R = Me, 33%Fig. 2 | Substrate scope of Fe-electrocatalyzed deoxygenative Giese reaction:reactivity across styrene and acrylate derivatives. A Scope of styrene deriva-tives. B Scope of acrylate derivatives. aIsolated yields. bFeCl2 (15mol%), IPr·HCl(15mol%), PPh3 (4 eq.), TBAB (4 eq.), and DIPEA (4 eq.) were used. cTBAI (4 eq.) wasused instead of TBAB. dFeCl2 (20mol%), IPr·HCl (20mol%), PPh3 (6 eq.), TBAB(4 eq.), and DIPEA (2 eq.) were used.Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 4www.nature.com/naturecommunicationsand the Appel reaction proceeds in the presence of the alcohol sub-strate. The alcohol substrate is then converted to the Mitsunobuintermediate 89, which gives the alkyl halide 90. The resulting alkylhalide 90 undergoes halogen-atom transfer (XAT) through the Fe-complex to form radical intermediate 91. 91 reacts readily with theMichael acceptor to form 92, which is then derivatized by single-electron transfer (SET) to give the desired product (93; path A). TheFe(II) species produced after XAT and SET are expected to be con-verted to Fe(I) by cathodic reduction and used in the next catalyticcycle. The transformation of 92 to 93 may also be mediated by aOtBuOMeMeMeMe MeMeMe MeOtBuOMe R2R1MeMe HMe72, 59%71, 40%F3CF79, 47%77, 30% 82, 31%NBoc81, 32%78, 35%NBocA. Scope of 1° alcoholsOOtBuOtBuOMeHH74, 35% (dr = 1:0)68, 32%76, 22% (dr = 1:1)Me HMeH HMeMeMeHNBocOOtBuMe55, 45%(dr = 3:1)80, 32%51, 51%MeOOtBuMe50, 61%NMeOtBuO53, 50 %58, 51% 59, 46%61, 33% (dr = 1:1)from Naproxen derivativeOOB. Scope of 2° alcohols(1 equiv.)R2OHR10.2 mmol scale(1° & 2° alcohols)OROtBuMeOtBuOMeMeOtBuOOOMeMeOtBuOMeOtBuOMeOtBuOOOtBuOtBuOMeMeMeMeOMeOtBuOMe MeMeOtBuOMe66, 20%(dr = 1:1.5)52, 55%70, 33%OtBuOMeMeOOtBu54, 65%65, 50% 69, 40%NBocBocNNBocNBoc83, 31% (dr = 1:0)73, 62%N NClL1+OtBuOR2R1or(5 eq.)H HMeHH75, 74% (dr = 2:1)(3.66 g synthesized)OHOOMeMeOOtBu67, 20%(dr = 1:1)60, 40% (dr = 1:1)from Ibuprofen derivative57, 67%56, 65%62, 50% 63, 46% 64, 24%• 1° & 2° alcohols• aliphatic & benzylic• medicinally important compounds• N-containing  compoundsMe4041FeCl2 (15 or 20 mol%) IPr•HCl (L1) (15 or 20 mol%)PPh3, TBAB, DIPEA nBu4NBF4 (0.5 eq.)DMA (2.5 ml), rt, 6 mA, 20h(+)C plate / (–)Ni foamundivided cell[Fe-electrocatalyticdeoxygenative Giese reaction]0.2 mmol scale reactions (HSPY-120-01)HHH42, R = OMe, 80%43, R = Me, 48%OOtBuMe44, para-F, 58%45, meta-F, 50%FOOtBuMe46, para-Cl, 49%47, meta-Cl, 41%ClOROtBuMe48, R = CF3, 58%49, R = CN, 31%Fig. 3 | Substrate scope of Fe-electrocatalyzed deoxygenative Giese reaction:reactivity across 1° alcohols and 2° alchohols.a,b. A Scope of 1° alcohols. B Scopeof 2° alcohols. aIsolated yields. bFeCl2 (15mol%), IPr·HCl (15mol%), PPh3 (4 eq.),TBAB (4 eq.), and DIPEA (4 eq.) were used for 40. FeCl2 (20mol%), IPr·HCl(20mol%), PPh3 (6 eq.), TBAB (4 eq.), and DIPEA (2 eq.) were used for 41.Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 5www.nature.com/naturecommunicationshydrogen-atom transfer (HAT) mechanism (path B) in addition to theSETmechanism (path A). The potential hydrogen atom transfer (HAT)process can be explained as follows. The HBr generated through theAppel reaction, which proceeds via anodic oxidation, is captured byDIPEA. Consequently, the resulting HBr salt of DIPEA serves as a pro-ficient hydrogen atom source. Therefore, it is hypothesized that thepotential HAT process from intermediate 92 is mediated by an excessof DIPEA or its HBr salt present in the reaction system.NBocH HMeHHOHNBocOH61%PPh3 (4 eq.), TBAB (4 eq.)DIPEA (4.0 eq.)nBu4NBF4 (0.5 eq.)DMA (2.5 ml), rt6 mA, 20h(+)Zn plate / (–)Ni foamMe86, 30%(isolated yield)C. Proposed reaction mechanismHO60% (dr = 1:1)[gram scale][2.90 g]3.9 g (10 mmol)FeCl2 (15 mol%)IPr•HCl (L1) (15 mol%)PPh3 (4 eq.), TBAI (4 eq.)DIPEA (4.0 eq.)nBu4NBF4 (0.5 eq.)DMA (125 ml), rt, 100 mA, 20h(+)C rod / (–)Ni foamundivided cell87, >99%ee(5 eq.)MeA. Mechanistic studies77, 30%[radical ring-opening][Fe-electrocatalytic deoxygenative Giese reaction]D. Application to gram scale and diversificationMe HMeH HMeHHOHNHOSHOO OOHRelyvrioMe HMeMeOOMe HMeH HMeHHMeOO2022 FDA-approved drug for ALS(Amylyx Pharmaceuticals Inc.)9594Ni foamC rodelectrodes S.M., PPh3, TBAI, TBABF4, styrenevacuum pumpingnitrogen replacementundivided cellstandard conditionsas in table 1(5 eq.)[racemization]standard conditionsas in table 1100 mA, rt, 20h(�) (+)848541[gram scale procedure](racemic)3, 0%[no desired product][chemical reductant]Zn dust (10 eq.)Mn dust (10 eq.)or0 mA, 20h(+)C plate / (–)Ni foamundivided cellFeCl2 (15 mol%)IPr•HCl (L1) (15 mol%)PPh3 (4 eq.), TBAB (4 eq.)DIPEA (4.0 eq.)nBu4NBF4 (0.5 eq.)DMA (2.5 ml), rt, 20h[no electricity]Ph MetBuPh MeBr+88 (1 eq.)2 (5 eq.)tBu3[sacrificial anode]withoutFe-catalyst & IPrPh MetBuPh MeBr+88 (1 eq.)2 (5 eq.)tBuFeCl2 (15 mol%)IPr•HCl (L1) (15 mol%)19%B. Control experiments[no catalyst]0.4 mmol scale(1 eq.)(1 eq.)lithocholic acid methyl ester(5 eq.)R rA999896103 10510010437%(dr = 1.2:1)MeHMeHFMeHSMeHtBuFeMeHMeHCO2Me53%(dr = 1.5:1)34%(dr = 1:1)16%(dr = 1.3:1)55%(dr = 1.2:1)40% (dr = 1:1)29%(dr = 1.3:1)97MeHMe23%(dr = 1:1)101MeH20%(dr = 1.5:1)102MeH41%(dr = 1.3:1)MeHN NiPriPriPr iPrFeIClClN NiPriPriPr iPrFeIIClCl XN NiPriPriPr iPrFeIIClCl XXAT cycleSET cycleedohtac+gA/gA  svV 2 .1- = derE  :edohtace–e–90(X = Br or I)91+ X– –�X�– X–+ X–92desired productX+ H+932X�X2OPPh3X�[Mitsunobuintermediate]89– OPPh3OHPPh3, X2– H+anode+gA/gA  svV 9.1 =  xoE :edo na– 2e–HAT: hydrogen atom transferXAT: halogen atom transferSET: single electron transfer[Fe-electrocatalytic deoxygenative Giese reaction][redox-neutral]  [paired electrolysis](B) (A)HAT + H•Fig. 4 | Mechanistic studies and application to the gram-scale diversification. A Mechanistic studies. B Control experiments. C Proposed reaction mechanism.D Application to gram-scale synthesis and diversification.Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 6www.nature.com/naturecommunicationsTo demonstrate the utility of this reaction, a gram-scale synthesisand diversification studywere conducted using lithocholic acidmethylester (94), which contains the RelyvrioTM scaffold (Fig. 4D)63. Asexpected, the reaction proceeded well at the 10-mmol scale (3.9 g of94), and the target compound 95 was successfully obtained in 60%yield (dr = 1:1; 2.9 g). Further attempts weremade to diversify 94 at the0.4mmol scale. We found that various Michael acceptors could beintroduced into 94 in a single step (96–105). This diversificationmethod could be applied to various bioactive andmedically importantcompounds with hydroxyl groups.For a better understanding of the unprecedented XAT processcatalyzed by the iron complex, a mechanistic investigation wasconducted. Initially, the use of the stoichiometric amount of theFe(II)-IPr complex was examined without electrolysis. However, evenupon adding one equivalent of the Fe(II)-IPr complex, the desiredreaction did not proceed, and the alkyl halide was quantitativelyrecovered (see Supplementary information, Figure S17). This obser-vation suggests that the catalytic cycle in this reaction does notinvolve Fe(II)/ Fe(III). Then, DFT calculations on the XAT processmediated by Fe(II)/ Fe(III) were performed. However, the energy ofthe product state is significantly higher than that of the initial state.Additionally, the activation energy was nearly comparable with theproduct state energy. These results also indicates that the XAT pro-cess involving Fe(II)/ Fe(III) catalytic cycle is energeticallyunfavourable.Subsequent DFT calculations were conducted to investigate thehalogen atom transfer (XAT) process mediated by the Fe(I)/Fe(II)catalytic cycle (Fig. 5). In the DFT calculations, the high-spin state ofiron was assumed based on the study by Nakamura and co-workers64,65.　Ethyl bromide (EtBr) and ethyl iodide (EtI) wereemployed as substrates, with separate calculations performed foreach. The XAT process is initiated by the in-situ generated Fe(I)species, Int-1. The computed activation barriers for the halogenatom transfer transition states (TS) are 14.4 kcal/mol for EtBr and8.3 kcal/mol for EtI, respectively, relative to Int-1 as the ground state.This transformation is highly exergonic, with reaction free energiesexceeding 10 kcal/mol, and leads to the formation of an ethyl radicaland the Fe(II) species, Int-2. Computational results revealed that, inboth EtBr and EtI cases, the rate-determining step is the processleading to the transition state of XAT. Comparison of the two sub-strates showed that the pathway involving EtI proceeds through amore thermodynamically stable transition state, rendering it a morefavorable process.Furthermore, Nakamura and coworkers have reported a similarmechanistic investigation of the Fe-catalyzed halogen atom transfer,involving DFT calculations64,65. These studies suggested that the XATprocess is more likely to be mediated by Fe(I)/ Fe(II) catalytic cycle,rather than Fe(II)/ Fe(III) cycle, based on a comparative analysis ofcalculation results. These experimental and computational results,along with reported studies64,65, indicate that the XAT process in thisreaction is likely mediated by the Fe(I)/ Fe(II) catalytic cycle.Further detailed experiments were then conducted to investigatethe mechanism of this reaction (Fig. 6). Cyclic voltammetry measure-ments were performed to investigate the electrochemical propertiesof various chemical species. (Fig. 6A) The redox potentials of theFeCl2·IPr complex have already been reported66. The CV spectrum ofthe FeCl2·IPr complex, synthesized in the same way as that used in thereaction, was measured (Fig. 6A). As a result, it was revealed that thereduction potential of FeCl2·IPr complex was within the redox poten-tial range of our reaction system. Considering the electrode potentials(cathode: −1.2 V vs. Ag/AgCl, anode: +1.9 V vs. Ag/AgCl), the FeCl2·IPrcomplex can be electrochemically reduced in the reaction system. Itsupports the Fe-catalyst regeneration process in the catalytic cycle(Fig. 4C). Additionally, an oxidation peak was observed at Eox = +0.8 Vvs. Ag/AgCl, which is similar to the oxidation peak of DIPEA. Stoi-chiometric amounts of DIPEA were used to synthesize FeCl2·IPr com-plex. The observed peak is presumed to be attributed to theresidual DIPEA.An irreversible oxidation peak was observed for TBAB atEox = +0.6 V vs. Ag/AgCl. This result supports the proposedmechanism     (kcal mol–1)DMA�GMeHXH ‡ = 8.3 (X = I)‡ = 14.4 (X = Br)MeHHrxn = –20.7 (X = I)[high-spin Fe(I)]Int1Int2XMeHHN NiPriPriPr iPrFeCl[XAT][radical generation]N NiPriPriPr iPrFeIClXN NiPriPriPr iPrFeIICl0.08.3 (X = I)–9.2 (X = Br)14.4 (X = Br)(X = Br or I)–20.7 (X = I)rxn = –9.2 (X = Br)Int2Energy profile for the Fe-catalyzed halogen atom transfer (XAT)TS (X = Br) Int2  (X = Br)TS (X = I) Int2  (X = I)TSInt12.478 Å(Fe-Br)2.324 Å(Fe-Cl)2.313 Å(Br-C)2.714 Å(Fe-I)2.355 Å(Fe-Cl)3.214 Å(I-C)2.459 Å(Fe-Br)2.315 Å(Fe-Cl)2.355 Å(Fe-cL)2.793 Å(Fe-i)[in situgenerated]XAT by in situ-generated Fe(I) active spieces.Strong Fe-X bond formationdrives XAT step.[reduced on cathode](kcal mol-1)   HBDEFe-Br 59.0[Giesereaction]∆G∆G∆G∆G�Fig. 5 | Energy profile for the Fe-catalyzed halogen atom transfer. Schematic representation of XAT transition state, computed at the SMD(DMA)-(U)B3LYP-D3BJ/BS1level of theory. Energies (kcal mol–1) and bond lengths (Å) are provided in the insert.Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 7www.nature.com/naturecommunications(Fig. 4C) in which the anodic oxidation of TBAB generates reactivespecies through the reaction of Br2 and triphenylphosphine. Triphe-nylphosphine exhibited an irreversible oxidation peak (Eox = +0.7 V vs.Ag/AgCl). It can undergo anodic oxidation and be consumed in thesystem. An excess amount of triphenylphosphine is essential in thisreaction, presumably due to competing anodic oxidation.Similarly, 4-tBu-styrenedisplayed an irreversible oxidationpeak atEox = +1.3 V vs. Ag/AgCl. DIPEA showed an irreversible oxidationpeakatEox = +0.7 V vs. Ag/AgCl, suggesting DIPEA can be oxidized in thereaction system. The resulting DIPEA radical cation may mediate theHAT process, as shown in Fig. 4C. This mechanism is speculatedto compete with the single electron transfer process mediated by thePh MeBr88Me5 (1 equiv.)0.2 mmol scaleMe MeA. Cyclic voltammetryC. NMR traceOHMe Me106 (5 eq.)standard conditionsas in table 141109, 0%[no desired product]OH107108, 0% Fe(IPr)Cl2 (Ered = -0.5 V, Eox = +0.2 V) PPh3 (Eox = +0.7 V) TBAB (Eox = +0.6 V)4-tBu-stylene (2) (Eox = +1.3 V)4-phenylbutan-2-ol (1)Eox = +0.7 V)blankPotential (V) vs Ag/AgClCurrent (mA)[Fe-electrocatalyticdeoxygenativeGiese reaction]tert-alcoholB. Limitation of the substrates[limitation]N NMeMeMeMe MeMeMe MeClFeCl2 (15 mol%) IPr•HCl (L1) (15 mol%)PPh3, TBAB, DIPEA nBu4NBF4 (0.5 eq.)DMA (2.5 ml), rt, 6 mA, 20h(+)C plate / (–)Ni foamundivided cell[Fe-electrocatalyticdeoxygenative Giese reaction]L11 (1 equiv.)OHMe+0.2 mmol scale41 (5 eq.)[predictedintermediate]crude 5(after the purification)standard conditionsas in table 1FeCl2 (15 mol%) IPr•HCl (L1) (15 mol%)PPh3, TBAB, DIPEA nBu4NBF4 (0.5 eq.)DMA (2.5 ml), rt, 6 mA, 20h(+)C plate / (–)Ni foamundivided cell (1 equiv.)0.2 mmol scale1 (S.M.)hP HOMe1.23 ppm(d, J = 6.2 Hz, 3H)0.99 ppm(d, J = 6.1 Hz, 3H)0.15.10.20.10.20.30.4)mpp( tfihs lacimehc)mpp( tfihs lacimehcPh Me Ph5 (desired)1:5 = 94:61:5 = 52:481:5 = 15:851:5 = 0:100 1H NMR (400 MHz, CDCl3)H3.83 ppm(qt, J = 6.2, 6.2 Hz, 1H)1:5 = 100:0crude 4crude 3crude 2crude 1desired product (5)[potential range of the reaction system]Ered = -1.2 V Eox = 1.9 VPh MeBr88[plausibleintermediate]fastS.M. (1)NMR ratio72% (super dry DMA)33% (normal DMA)A. ――――　―　― DIPEA ―　Fig. 6 | Further mechanistic studies and 1H NMR reaction monitoring. A Cyclic voltammetry. B limitation of the substrates. C NMR trace.Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 8www.nature.com/naturecommunicationsFe-catalyst. On the other hand, 4-phenylbutan-2-ol (1) was confirmedto be stable in the reaction system, as no significant redox peak wasobserved. In other words, the activation of the alcohol through theAppel reaction is speculated to be essential in this reaction.The control experiments of this reaction to tertiary alcohols wereperformed (Fig. 6B)66. When the optimal conditions (Fig. 1C) wereapplied to 2-methyl-4-phenyl-2-butanol (106) and adamantan-1-ol(107), only the starting materials were recovered, and the desiredproducts were not observed. The limitation suggested here is con-sistent with that reported by Li et al.34.The reaction using 4-phenylbutan-2-ol (1) and styrene (41) wasmonitored by 1H NMR spectroscopy (Fig. 6C). After two hours, a sig-nificant amount of the starting material 1 remained, and the ratio ofstartingmaterial 1 to the desired product 5was 94:6 (determined by 1HNMR). After six hours, the reaction had partially proceeded, and thedesired product 5was observed at a ratio of 52:48. Ten hours after, thestarting material 1 was consumed, and the desired product 5 wasobserved at a ratio of 15:85. After 20hours, the starting material 1 hadcompletely disappeared, giving the desired product 5 in 72% yield.Noteworthy, the brominated compound 88was not observed in any 1HNMR spectra. This observation suggests that the brominated com-pound 88 promptly underwent the XAT process mediated by Fe-catalyst and was converted to compound 5. As a control experiment,the reaction was conducted using normal DMA instead of dehydratingDMA, and the yield dropped to 33%. This is assumed to be due to thedecomposition of Mitsunobu intermediate 89 by water present in thereaction system during the Appel reaction.DiscussionThe redox-neutral Fe-electrocatalytic deoxygenative Giese reactionreported herein represents a powerful approach for the one-stepinstallation of Michael acceptors into both primary and secondaryalcohols. The key feature of this reaction is that it allows the use ofreadily available commercial reagents to effortlessly form C‒C bondsfromalcohols at ambient temperaturewithout the use of scarcemetalsor highly toxic reagents. In addition, an unprecedented cathodicreduction of the Fe-complex has been realized, which effectively pro-motes this Giese reaction. Coupled with the Appel reaction via anodicoxidation, this redox-neutral Giese reaction is characterized by highlevels of efficiency for small-molecule diversification. This methodol-ogy represents a pioneering investigation into the fusion of Fe-catalysis and electrochemistry in the field of redox-neutral reactions.Further exploration of a broad range of innovative reactions using thisredox cycle promises significant potential. Moreover, this approach isexpected to lead to further advances in the field of medicinalchemistry.Data availabilityThe authors declare that all experimental and computational datagenerated in this study, including experimental procedures and com-pound characterization, NMR, and DFT-optimized structure coordi-nates, are provided in the Supplementary Information/Source Datafile. Should any raw data files be needed in another format they areavailable from the corresponding author upon request. Source dataare provided with this paper.References1. Tsuji, J. Palladium Reagents and Catalysts: New Perspectives for the21st Century 2nd ed. (Wiley, New York, 2004).2. Hartwig, J. F. Organotransition Metal Chemistry: From Bonding toCatalysis (University Science Books, New York, 2009).3. Gladysz, J. A. Introduction to frontiers in transition metal cat-alyzed reactions (and a brief adieu). Chem. Rev. 111, 1167–1169(2011).4. Villo, P., Shatskiy, A., Kärkäs, M. D. & Lundberg, H. ElectrosyntheticC‒O Bond Activation in Alcohols and Alcohol Derivatives. Angew.Chem. Int. Ed. 62, e202211952 (2023).5. Wang, Y., Xu, J., Pan, Y. & Wang, Y. Recent advances in electro-chemical deoxygenation reactions of organic compounds. Org.Biomol. Chem. 21, 1121–1133 (2023).6. Anwar, K., Merkens, K., Aguilar Troyano, F. J. & Gómez-Suárez, A.Radical deoxyfunctionalisation strategies. Eur. J. Org. Chem. 2022,e202200330 (2022).7. Suga, T., Shimazu, S. & Ukaji, Y. Low-Valent titanium-mediatedradical conjugate addition using benzyl alcohols as benzyl radicalsources. Org. Lett. 20, 5389–5392 (2018).8. Suga, T., Takahashi, Y., Miki, C. & Ukaji, Y. Direct and unified accessto carbon radicals fromaliphatic alcohols by cost-efficient titanium-mediated homolytic C-OH bond cleavage. Angew. Chem. Int. Ed.61, e202112533 (2022).9. Shu, X.-H. et al. Radical dehydroxylative alkylation of tertiary alco-hols by Ti catalysis. J. Am. Chem. Soc. 142, 16787–16794 (2020).10. Wang, J., Lv, X. & Jiang, Z. Visible-light-mediated photocatalyticderacemization. Chem. Eur. J. 29, e202204029 (2023).11. Yin, Y., Zhao, X., Qiao, B. & Jiang, Z. Cooperative photoredox andchiral hydrogen-bonding catalysis. Org. Chem. Front. 7,1283–1296 (2020).12. Beil, S. B., Chen, T. Q., Intermaggio, N. E. & MacMillan, D. W. C.Carboxylic acids as adaptive functional groups in metallaphotor-edox catalysis. Acc. Chem. Res. 55, 3481–3494 (2022).13. Macmillan, D. W. C. et al. Chem. Rev. 122, 1485–1542 (2022).14. Maiti, D. et al. Non-directed Pd-catalysed electrooxidative olefina-tion of arenes. Chem. Sci. 13, 9432–9439 (2022).15. Qiu, Y., Scheremetjew, A. & Ackermann, L. Electro-oxidative C‒Calkenylation by rhodium(III) catalysis. J. Am. Chem. Soc. 141,2731–2738 (2019).16. Mei, R., Sauermann, N., Oliveira, J. C. A. & Ackermann, L. Electro-removable traceless hydrazides for cobalt-catalyzed electro-oxi-dative C‒H/N‒H activation with internal alkynes. J. Am. Chem. Soc.140, 7913–7921 (2018).17. Lundberg, H. et al. Electroreductive desulfurative transformationswith thioethers as alkyl radical precursors. Angew. Chem. Int. Ed.62, e202304272 (2023).18. Lin, S. et al. Electrochemically driven cross-electrophile coupling ofalkyl halides. Nature 604, 292–297 (2022).19. Mei, T.-S. et al. Electrochemistry-enabled Ir-catalyzed vinylic C‒Hfunctionalization. J. Am. Chem. Soc. 141, 18970–18976 (2019).20. Mei, T.-S. et al. TEMPO-Enabled Electrochemical EnantioselectiveOxidative Coupling of Secondary Acyclic Amines with Ketones. J.Am. Chem. Soc. 143, 15599–15605 (2021).21. Xiong, P., Xu, H.-H., Song, J. & Xu, H.-C. Electrochemical difluor-omethylarylation of alkynes. J. Am. Chem. Soc. 140, 2460–2464(2018).22. Lin, S. et al. Anodically Coupled Electrolysis for the Hetero-difunctionalization of Alkenes. J. Am. Chem. Soc. 140, 2438–2441(2018).23. Xu, H.-C. et al. Photoelectrochemical asymmetric catalysis enablessite- and enantioselective cyanation of benzylic C–H bonds. Nat.Catal. 5, 943–951 (2022).24. Baran, P. S. et al. Ni-electrocatalytic Csp3–Csp3 doubly decarbox-ylative coupling. Nature 606, 313–318 (2022).25. Yan, M., Kawamata, Y. & Baran, P. S. Synthetic Organic Electro-chemical Methods Since 2000: On the Verge of a Renaissance.Chem. Rev. 117, 13230–13319 (2017).26. Baran, P. S. et al. A Survival Guide for the “Electro-curious. Acc.Chem. Res. 53, 72–83 (2020).27. Hupp, J. T. et al. Charge transport in zirconium-basedmetal–organic frameworks. Acc. Chem. Res. 53, 1187–1195 (2020).Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 9www.nature.com/naturecommunications28. Moeller, K. D. Using physical organic chemistry to shape the courseof electrochemical reactions. Chem. Rev. 118, 4817–4833 (2018).29. Little, R. D. A perspective on organic electrochemistry. J. Org.Chem. 85, 13375–13390 (2020).30. Little, R. D. et al. Versatile tools for understanding electrosyntheticmechanisms. Chem. Rev. 122, 3292–3335 (2022).31. Little, R. D. & Moeller, K. D. Introduction: Electrochemistry: Tech-nology, synthesis, energy, and materials. Chem. Rev. 118,4483–4484 (2018).32. Wirtanen, T., Prenzel, T., Tessonnier, J.-P. & Waldvogel, S. R.Cathodic corrosion of metal electrodes-how to prevent it in elec-troorganic synthesis. Chem. Rev. 121, 10241–10270 (2021).33. Liu, Y., Li, P., Wang, Y. & Qiu, Y. Electroreductive Cross-ElectrophileCoupling (eXEC) Reactions. Angew. Chem. Int. Ed.62, e202306679(2023).34. Li, C. et al. Electrochemically Enabled, Nickel-Catalyzed Dehy-droxylative Cross-Coupling of Alcohols with Aryl Halides. J. Am.Chem. Soc. 143, 3536–3543 (2021).35. Lyon, W. L. & MacMillan, D. W. C. Expedient Access to Under-explored Chemical Space: Deoxygenative C(sp3)‒C(sp3) Cross-Coupling. J. Am. Chem. Soc. 145, 7736–7742 (2023).36. Gould, C. A., Pace, A. L. & MacMillan, D. W. C. Rapid and ModularAccess to Quaternary Carbons from Tertiary Alcohols via Bimole-cular Homolytic Substitution. J. Am. Chem. Soc. 145, 16330–16336(2023).37. Wang, J. Z., Sakai, H. A. & MacMillan, D. W. C. Alcohols asAlkylating Agents: Photoredox-Catalyzed Conjugate Alkylationvia in Situ Deoxygenation. Angew. Chem. Int. Ed. 61,e202207150 (2022).38. Intermaggio, N. E., Millet, A., Davis, D. L. & MacMillan, D. W. C.Deoxytrifluoromethylation of alcohols. J. Am. Chem. Soc. 144,11961–11968 (2022).39. Sakai, H. A. & MacMillan, D. W. C. Nontraditional Fragment Cou-plings of Alcohols and Carboxylic Acids: C(sp3)‒C(sp3) Cross-Coupling via Radical Sorting. J. Am. Chem. Soc. 144, 6185–6192(2022).40. Dong, Z. & MacMillan, D. W. C. Metallaphotoredox-enabled deox-ygenative arylation of alcohols. Nature 598, 451–456 (2021).41. Shang, R., Ilies, L. & Nakamura, E. Iron-Catalyzed Directed C(sp2)‒Hand C(sp3)‒H Functionalization with Trimethylaluminum. J. Am.Chem. Soc. 137, 7660–7663 (2015).42. Shang, R., Ilies, L. & Nakamura, E. Iron-Catalyzed Ortho C‒HMethylation of Aromatics Bearing a Simple Carbonyl Group withMethylaluminum and Tridentate Phosphine Ligand. J. Am. Chem.Soc. 138, 10132–10135 (2016).43. Doba, T., Matsubara, T., Ilies, L., Shang, R. & Nakamura, E.Homocoupling-free iron-catalysed twofold C–H activation/cross-couplings of aromatics via transient connection of reactants. Nat.Cat. 2, 400–406 (2019).44. Norinder, J., Matsumoto, A., Yoshikai, N. & Nakamura, E. Iron-catalyzed direct arylation through directed C‒H bond activation. J.Am. Chem. Soc. 130, 5858–5859 (2008).45. Yoshikai, N., Matsumoto, A., Norinder, J. & Nakamura, E. Iron-catalyzed chemoselective ortho arylation of aryl imines by directedC-H bond activation. Angew. Chem. Int. Ed. 48, 2925–2928 (2009).46. Yoshikai, N., Mieczkowski, A., Matsumoto, A., Ilies, L. &Nakamura, E.Iron-catalyzed C‒C bond formation at alpha-position of aliphaticamines via C‒H bond activation through 1,5-hydrogen transfer. J.Am. Chem. Soc. 132, 5568–5569 (2010).47. Mazzacano, T. J. & Mankad, N. P. Base metal catalysts for photo-chemical C‒H borylation that utilize metal-metal cooperativity. J.Am. Chem. Soc. 135, 17258–17261 (2013).48. Ackermann, L. et al. Iron-Catalyzed C‒H Activation with PropargylAcetates: Mechanistic Insights into Iron(II) by Experiment, Kinetics,Mössbauer Spectroscopy, andComputation.Angew. Chem. Int. Ed.58, 12874–12878 (2019).49. Mo, J., Müller, T., Oliveira, J. C. A. & Ackermann, L. 1,4-IronMigrationfor Expedient Allene Annulations through Iron-CatalyzedC‒H/N‒H/C‒O/C‒H Functionalizations. Angew. Chem. Int. Ed. 57, 7719–7723(2018).50. Zhu, C., Stangier, M., Oliveira, J. C. A., Massignan, L. & Ackermann,L. Iron-Electrocatalyzed C‒H Arylations: Mechanistic Insights intoOxidation-Induced Reductive Elimination for Ferraelectrocatalysis.Chem. Eur. J. 25, 16382–16389 (2019).51. Green, S. A., Vásquez-Céspedes, S. & Shenvi, R. A. Iron-nickel dual-catalysis: A new engine for olefin functionalization and the forma-tion of Quaternary centers. J. Am. Chem. Soc. 140, 11317–11324(2018).52. Shenvi, R. A. et al. Iron-Catalyzed Hydrobenzylation: Stereo-selective Synthesis of (-)-Eugenial C. J. Am. Chem. Soc. 145,15714–15720 (2023).53. Mako, T. L. & Byers, J. A. Recent advances in iron-catalysed crosscoupling reactions and their mechanistic underpinning. Inorg.Chem. Front. 3, 766–790 (2016).54. Scheiper, B., Bonnekessel, M., Krause, H. & Fürstner, A. Selectiveiron-catalyzed cross-coupling reactions of grignard reagents withenol triflates, acid chlorides, and dichloroarenes. J. Org. Chem. 69,3943–3949 (2004).55. Fürstner, A., Leitner, A., Méndez, M. & Krause, H. Iron-catalyzedcross-coupling reactions. J. Am. Chem. Soc. 124, 13856–13863(2002).56. Tong, X., Yang, Z.-P., Del Angel Aguilar, C. E. & Fu, G. C. Iron-Catalyzed Reductive Cross-Coupling of Alkyl Electrophiles withOlefins. Angew. Chem. Int. Ed. 62, e202306663 (2023).57. Jin, M., Adak, L. & Nakamura, M. Iron-Catalyzed EnantioselectiveCross-Coupling Reactions of α-Chloroesters with Aryl GrignardReagents. J. Am. Chem. Soc. 137, 7128–7134 (2015).58. Kessler, S. N. & Bäckvall, J.-E. Iron-catalyzed Cross-Coupling ofPropargyl Carboxylates and Grignard Reagents: Synthesis of Sub-stituted Allenes. Angew. Chem. Int. Ed. 55, 3734–3738 (2016).59. Shaik, S. Iron opens up to high activity. Nat. Chem. 2, 347–349(2010).60. Liang, Q. & Song, D. Iron N-heterocyclic carbene complexes inhomogeneous catalysis. Chem. Soc. Rev. 49, 1209–1232 (2020).61. Kühn, F. E. et al. Chemistry of iron N-heterocyclic carbene com-plexes: syntheses, structures, reactivities, and catalytic applica-tions. Chem. Rev. 114, 5215–5272 (2014).62. Jones, G.O., Liu, P., Houk, K. N. & Buchwald, S. L. J. Am. Chem. Soc.132, 6205–6213 (2010).63. Aschenbrenner, D. S. New drug approved for ALS. Am. J. Nurs. 123,22–23 (2023).64. Sharma, A. K. et al. DFT and AFIR Study on the Mechanism and theOrigin of Enantioselectivity in Iron-Catalyzed Cross-CouplingReactions. J. Am. Chem. Soc. 139, 16117–16125 (2017).65. Sharma, A. K. & Nakamura, M. A DFT study on FeI/FeII/FeIIImechanism of the cross-coupling between haloalkane and arylgrignard reagent catalyzed by iron-SciOPP complexes. Molecules25, 3612 (2020).66. Przyojski, J. A., Arman, H. D. & Tonzetich, Z. J. Complexes of Iron(II)and Iron(III) Containing Aryl-Substituted N-Heterocyclic CarbeneLigands. Organometallics 31, 3264–3271 (2012).AcknowledgementsFinancial support for this work was provided by a grant from the RGC ofthe Hong Kong SAR, China (ECS, HKUST 26302024), and start-up fundsfrom HKUST (Project No. R9820) to H.N. The computation was partlyperformed using Research Center for Computational Science, Okazaki,Japan (Projects: 23-IMS-C119, 24-IMS-C114, and 25-IMS-C115 Y.N.).Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 10www.nature.com/naturecommunicationsAuthor contributionsL.Y., S.Li., H.O., H.N. conducted the experiments. H.N. conceptualizedand designed the catalytic strategy. Y.N. conducted the computationalstudy. The manuscript was prepared by H.N. with the feedback of allother authors. L.Y., S.Li., H.O. prepared the supplementary informationunder the guidance of H.N. For the mechanistic study, Y.M. and Q.C.contributed to the cyclic voltammetry. K.Y. contributed to the design ofthe mechanistic study.Competing interestsThe authors declare no competing interests.Additional informationSupplementary information The online version containssupplementary material available athttps://doi.org/10.1038/s41467-025-63515-x.Correspondence and requests for materials should be addressed toYuuya Nagata or Hugh Nakamura.Peer review information Nature Communications thanks the anon-ymous reviewers for their contribution to the peer review of this work. Apeer review file is available.Reprints and permissions information is available athttp://www.nature.com/reprintsPublisher’s note Springer Nature remains neutral with regard to jur-isdictional claims in published maps and institutional affiliations.Open Access This article is licensed under a Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License,which permits any non-commercial use, sharing, distribution andreproduction in any medium or format, as long as you give appropriatecredit to the original author(s) and the source, provide a link to theCreative Commons licence, and indicate if you modified the licensedmaterial. Youdonot havepermissionunder this licence toshare adaptedmaterial derived from this article or parts of it. The images or other thirdparty material in this article are included in the article’s CreativeCommons licence, unless indicated otherwise in a credit line to thematerial. If material is not included in the article’s Creative Commonslicence and your intended use is not permitted by statutory regulation orexceeds the permitted use, you will need to obtain permission directlyfrom the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.© The Author(s) 2025Article https://doi.org/10.1038/s41467-025-63515-xNature Communications |         (2025) 16:8379 11https://doi.org/10.1038/s41467-025-63515-xhttp://www.nature.com/reprintshttp://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/www.nature.com/naturecommunications Fe-electrocatalytic deoxygenative Giese reaction Results Discussion Data availability References Acknowledgements Author contributions Competing interests Additional information