Mutable Properties of Nonheme Iron(III)−Iodosylarene ComplexesResult in the Elusive Multiple-Oxidant MechanismYiran Kang,†,‡,# Xiao-Xi Li,†,# Kyung-Bin Cho,§ Wei Sun,† Chungu Xia,† Wonwoo Nam,*,†,§

and Yong Wang*,†

†State Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy ofSciences, Lanzhou 730000, China‡University of Chinese Academy of Sciences, Beijing 100049, China§Department of Chemistry and Nano Science, Ewha Womans University, Seoul 03760, Korea

*S Supporting Information

ABSTRACT: Although nonheme iron(III)−iodosylarenecomplexes present amazing oxidative efficiency andselectivity, the nature of such complexes and relatedoxidation mechanism are still unsolved after decades ofexperimental efforts. Density functional calculations wereemployed to explore the structure−reactivity relationshipof the iron(III)−iodosylbenzene complex, [FeIII(tpena−)(PhIO)]2+ (1), in thioanisole sulfoxidation. Our theoreticalwork revealed that complex 1 can evolve into tworesonance valence-bond electronic structures (a high-valent iron−oxo species and a monomeric PhIO species)in thioanisole sulfoxidation to present different reactionmechanisms (the novel bond-cleavage coupled electrontransfer mechanism or the direct oxygen-atom transfermechanism) as a response to different substrate attackorientations.

Iodosylarenes are considered as the most versatile and efficientterminal oxidants used in organic synthesis, biological and

biomimetic oxidation reactions.1 Generally, they are insolubleamorphous powder, in a linear polymeric, asymmetricallybridged zigzag structure (Scheme 1a).2 Catalysts (metalcomplexes, Lewis or Brønsted acids, etc.) are usually employed

to make them soluble before introducing iodosylarenes into thecatalytic systems.3 According to the consensus one-oxidantmechanism (Scheme 1b),4 once iodosylarenes are mixed withmetal complexes, metal−iodosylarene adducts are formed.Subsequently, these complexes are converted to high-valentmetal−oxo species, which are postulated to act as the oxidants toperform various oxidations, such as CH hydroxylation, CCepoxidation, sulfoxidation, etc. For instance, Groves and co-workers reported that when an iron(III) porphyrin complex wasmixed with iodosylbenzene to oxidize substrates, alkenes werepreferentially oxidized to the corresponding epoxides.4a

Similarly, alcohols were obtained as major products in alkanehydroxylation.4a The iron(IV)−oxo porphyrin cation radicalcomplex was therefore proposed as the sole oxidant to do thoseoxidations via the well-known “oxygen-rebound” mechanism.5

Combined with Shaik’s “two-state reactivity” (TSR) mecha-nism,6 most mechanistic problems in P450 chemistry and relatedbiomimetic nonheme chemistry could be well explained andsolved with the one-oxidant mechanism, which makes itpopular.7

However, in the 1990s, Valentine and co-workers reportedthat olefin epoxidation by iodosylbenzene could be significantlypromoted by redox-innocent metal ions, such as zinc(II) andaluminum(III), in which the metal−iodosylbenzene complexescannot be converted to high-valent metal−oxo oxidants.8 Thisexperimental observation casts the one-oxidant mechanism intodoubt. An electrophilic iodine(III)-attacking mechanism, usingthe metal−iodosylbenzene complex instead of the high-valentmetal−oxo species as the oxidant, was postulated.In 2000, Collman, Brauman and co-workers carried out

competitive alkane hydroxylation with various iodosylarenes inan iron porphyrin system and found that the reaction rate ratiosfor each pair of substrates varied when different iodosylareneswere employed as terminal oxidants.9 Further, Nam and co-workers reported that iron(III)-iodosylbenzene species couldreversibly convert to high-valent iron−oxo complex andiodobenzene,10 unambiguously demonstrating that at least onemore oxidant is formed besides the high-valent metal−oxooxidant. Thus, a multiple-oxidant mechanism (Scheme 1c) wasproposed.9,10 According to this mechanism, if the OI bond

Received: April 2, 2017Published: May 19, 2017

Scheme 1. Schematic Plots of (a) the Zigzag Structure ofIodosylbenzene Polymer, (b) the One-Oxidant Mechanismand (c) the Multiple-Oxidant Mechanism

Communication

pubs.acs.org/JACS

© 2017 American Chemical Society 7444 DOI: 10.1021/jacs.7b03310J. Am. Chem. Soc. 2017, 139, 7444−7447

cleavage occurs fast, then the metal−iodosylarene complex willconvert to a high-valent metal−oxo oxidant to oxidize substrates.On the other hand, if the OI bond cleavage is slow and themetal−iodosylarene complex has enough oxidizing power, thenthe metal−iodosylarene complex becomes the oxidant. Toexplore the elusive nature of metal−iodosylarene complexes,extensive experimental efforts using competitive reaction kineticsand various spectroscopic methods (e.g., ESI-MS, UV−vis,rRaman, X-ray diffraction etc.) were performed.11 However,despite these efforts, the mechanistic mystery of the structure−reactivity relationship of the metal−iodosylarene complexremains elusive.Herein, on the basis of the crystal structure of a nonheme

iron(III)−iodosylarene complex, [FeIII(tpena−)OIPh]2+ (1,tpena− = N,N,N′-tris(2-pyridylmethyl)ethylendiamine-N′-ace-tate), reported by McKenzie and co-workers,11b we haveperformed for the first time a comprehensive theoretical studyon the conversion of 1 and its reaction with thioanisole. Allcalculations were performed in the acetonitrile solvent using theconductor-like polarizable continuummodel. The benchmark onthe reliability and stability of the employed theoretical methods ispresented in the Supporting Information. Surprisingly, we foundthat during the substrate approaches, the iron(III)−iodosylarenecomplex 1 evolves into two resonance valence-bond structures (ahigh-valent iron−oxo species and a monomeric iodosylbenzenespecies). These species then oxidize thioanisole to thecorresponding product.The conversion of 1 to a high-valent iron−oxo species 2 was

studied at the UB3LYP/lanl2dz(Fe)-lanl2dzdp(I)-6-31+G*-(rest) computational level (Figure 1). For 1, the ground state

is a high-spin sextet state (S = 5/2). The excited quartet/doubletspin states lie at 14.3/21.7 kcal mol−1 higher. In 61, the averageFeN distance is 2.293 Å, the FeO(carboxylate) distance is2.080 Å, and the Fe−O(PhIO) distance is 2.001 Å (Figure S2).Thus, the iron core is in a heptacoordinate state, which isconsistent with McKenzie’s experimental results.11b During OI bond elongation, spin reversion occurs and the reaction path

switches from the sextet to the doublet spin state (S = 1/2) onthe transition state; therefore, it is a two-state reactivity (TSR)process.6 The lowest activation energy is 24.5 kcal mol−1. In2TS12, the FeO distance is 1.630 Å, the OI distance is 2.756Å and the FeO(carboxylate) distance is 3.860 Å. Therefore,the iron core becomes hexacoordinated. The formed hexacoordi-nated iron−oxo complex 2 on the quartet ground state lies 18.1kcal mol−1 higher than 1. In 2, the spin of the FeOmoiety is ca.2.3 and the PhI moiety is ca. 0.6, thus the OI bond cleavage is ahomolysis process to form an iron(IV)(O) (tpena−) (PhI cationradical) species 2 (Table S4). Interestingly, further enlogation ofthe OI distance from 4.7 to 7.7 Å in 2 triggers anintermolecular electron transfer from the tpena− ligand to thePhI to form an iron(IV)(O) (tpena radical) (PhI) species 3(Figure S2 and Table S4), which lies 8.6 kcal mol−1 higher than 2.In short, the conversion of 1 to the high-valent iron−oxo species2 and 3 is both kinetically and thermodynamically difficult.When the substrate thioanisole is introduced in the reaction

system, we found two orientation modes of the reagent complex(Figure 2). In the halogen-bonding mode RC (Figure 2a),

thioanisole lies trans to the oxygen of 1. The SIO angle isnearly 180° and the phenyl ring of PhIO forms a T-shape withthis linear SIOmoiety. In the other normal substrate attackmodeRC′ (Figure 2b), thioanisole is oriented toward the oxygenof 1. RC is more stable than RC′ by 4.1 kcal mol−1. For thehalogen-bonding case, surprisingly, the activation barrier of OIbond cleavage dramatically decreases to 19.4 kcal mol−1 (Figure3), compared to that of OI cleavage without the substrate(24.5 kcal mol−1 in Figure 1). In 6TSBCCET, the OI distance is2.628 Å and the FeO distance is 1.659 Å. One pyridine liganddecoordinates from the iron core (the FeN distance is 4.325Å), thereby generating a hexacoordinate complex. The halogenbond interaction between the sulfur atom of thioanisole and theleaving iodine atom still persists with an SI distance of 3.184 Å.Molecular orbital analysis also shows the strong conjugationinteraction between the PhSMePhIOFe triad (FigureS8), which stabilizes the electronic structure to lower the reactionenergy barrier. Inspection of the electronic structure of TSBCCETand IM reveals that during the OI bond cleavage, anintermolecular electron transfer from thioanisole to 1 occurs toform an iron(IV)−oxo species with an one-electron oxidizedthioanisole cation radical. Thus, this step is a novel halogen-bondpromoted electron transfer process. The lowerest energy state ofthe intermediate IM is now a degenerated doublet/quartet pair.For 2/4IM, The FeO distance is 1.624 Å, and the iron is in ahexacoordinate state with one pyridine free. Subsequently, 2IM

Figure 1. Energy profiles for the conversion of the ferric−iodosylbenzene complex 1 to the iron(IV)(O)(tpena radical) species3 and PhI via the iron(IV)(O) (tpena−) (PhI cation raidcal) species 2.The key geometric information on the transition state 2TS12 ispresented. Hydrogen atoms are omitted for clarity. Energies are inkcal mol−1 units, lengthes are in Å units, angles are in degree units, andthe imaginary frequency is in cm−1 unit.

Figure 2. Geometric information on (a) the halogen-bonding reagentcomplex RC and (b) the normal reagent complex RC′. Hydrogen atomsare omitted for clarity. Lengths are in Å units, angles are in degree units.

Journal of the American Chemical Society Communication

DOI: 10.1021/jacs.7b03310J. Am. Chem. Soc. 2017, 139, 7444−7447

7445

undergoes a transition state 2TSOT with a tiny barrier of 0.7 kcalmol−1, and transfers the oxygen to the one-electron oxidizedthioanisole cation radical to yield the iron(III)(tpena−) complexand sulfoxide as products. At the product complex state, the freepyridine ligand recoordinates to the iron core and generates thesix-coordinate iron(III)(tpena−) complex in preparation for thenext catalytic cycle. In short, the reaction proceeds stepwise viaan initially concerted but asynchronous process of halogen-bond-promoted OI bond cleavage, coupled to an electrontransfer from the substrate thioanisole to the iron−oxo moiety.This is followed by an oxygen rebound step to form the sulfoxideproduct. Despite the role of iodine, this mechanism is not thesame as Valentine’s mechanism,8b which was deemed notpossible in this reaction due to a large barrier (>40 kcal mol−1,Figure S17). Such a novel intermolecular electron transfer eventis similar to the metal-ion coupled electron transfer (MCET)mechanism proposed by Nam, Fukuzumi and co-workers in thethioanisole sulfoxidation by the Sc3+-binding nonheme iron-(IV)−oxo complexes12 and we term this mechanism as the bond-cleavage coupled electron transfer (BCCET) mechanism.For thioanisole sulfoxidation by the heme/nonheme high-

valent metal−oxo species, the widely accepted mechanism is thedirect oxygen-atom transfer (DOT) mechanism.13 Thus,thioanisole sulfoxidation by 1 via the DOT mechanism wasinvestigated. As shown in Figure 4, for the normal reagentcomplex (RC′ in Figure 2), the activation energy barrier of theDOT process is 18.7 kcal mol−1, which is competing with that ofthe BCCET process (19.4 kcal mol−1, taking the halogen-bonding reaction complex RC as the reference point).Investigation of the geometry of 6TSDOT shows that the FeObond is 3.097 Å, the OI bond is 1.886 Å and the OS bond is3.003 Å. This TS geometry demonstrates that the direct DOTprocess proceeds in a concerted way, i.e., initially substrate-induced iodosylbenzene deligation from the iron(III) core tosome extent, then this iodosylbenzene transfers its oxygen atom

to thioanisole to form sulfoxide. This indicates that themonomeric iodosylbenzene may be a potent oxidant. Toinvestigate this, the thioanisole sulfoxidation oxidized by oneiodosylbenzene molecule was investigated. The activation barrieris only 11.0 kcal mol−1, thus the iodosylbenzene monomer is arobust oxidant, which is consistent with previous reports.1,14

However, after taking the PhIOFe bond dissociation energyinto account (16.0 kcal mol−1, Table S13), the overall barrier ofsuch stepwise PhIO-deligation/sulfoxidation process is 27.0 kcalmol−1, which is much higher than that of the concerted process(18.7 kcal mol−1).Based on our theoretical findings and the previous postulated

multiple-oxidant mechanism (Scheme 1c), we propose thefollowing mechanism (Scheme 2). When the metal catalyst ismixed with the iodosylarene polymer, a metal−iodosylarenecomplex 1 is formed. If the OI bond cleavage is faster, the

Figure 3. Energy profiles for thioanisole sulfoxidation via the bond-cleavage-coupled electron transfer mechanism. Key geometric informa-tion on transition states 6TSBCCET and 2TSOT is presented. Hydrogenatoms are omitted for clarity. Energies are in kcal mol−1 units, lengths arein Å units, angles are in degree units, and imaginary frequencies are incm−1 units.

Figure 4. Energy profiles for thioanisole sulfoxidation via the directoxygen-atom transfer mechanism. The top-right inset shows the energyprofile for thioanisole sulfoxidation by monomeric PhIO. Key geometricinformation on transition states is presented. Hydrogen atoms areomitted for clarity. Energies are in kcal mol−1 units, lengths are in Åunits, angles are in degree units, and imaginary frequencies are in cm−1

units.

Scheme 2. Proposed Catalytic Cycle of Oxidation Involvingthe Metal−Iodosylarene Complexes

Journal of the American Chemical Society Communication

DOI: 10.1021/jacs.7b03310J. Am. Chem. Soc. 2017, 139, 7444−7447

7446

metal−iodosylarene complex preferentially self-decays to form ahigh-valent metal−oxo species 2 (oxidant 1) to oxidize thesubstrate to form the product. If the OI bond cleavage isslower and the oxidizing power of the metal−iodosylarenecomplex is robust, the metal−iodosylarene complex 1 acts as theoxidant (oxidant 2). It will evolve into two resonance valence-bond structures (a high-valent metal−oxo species and amonomeric PhIO species shown in Scheme 2)15 as a responseto different substrate orientation. If the substrate forms ahalogen-bond with 1, 1 will evolve into the high-valent iron−oxospecies to trigger an international electron transfer from thesubstrate to the high-valent metal−oxo species (e.g., the BCCETprocess in Figure 3). Thus, the oxidation occurs in a stepwisefashion; if on the other hand the substrate attacks directly theoxygen of 1, then 1will evolve into the monomeric PhIO species,which will directly transfer the oxygen atom to the substrate.Thus, this process is concerted. After the oxidation, the metalcatalyst is regenerated and joins in the next catalytic cycle.Obviously, the substrate-induced process in Scheme 2 is moreinteresting, because we can potentially modulate the nature ofthe metal−iodosylarene complex to generate the specificoxidation species at will. For instance, we can employ moreelectron-rich nonheme ligands and electron-deficient iodosylar-enes to facilitate the OI bond cleavage via the “push−pull”mechanism.16

In summary, the structure−reactivity relationship of theiron(III)−iodosylbenzene complex, [FeIII(tpena−)OIPh]2+ (1),was investigated by means of density functional calculations. Ourtheoretical work demonstrated for the first time that the metal−iodosylarene complex 1 can evolve into two resonance valence-bond structures (a high-valent iron−oxo species and amonomeric PhIO species) in thioanisole sulfoxidation to presentdifferent reaction mechanisms (the novel bond-cleavage coupledelectron transfer mechanism or the direct oxygen-atom transfermechanism) as a response to different substrate attackorientations. Based on these findings and the multiple-oxidantmechanism, the whole catalytic cycle of oxidation involving themetal−iodosylarene complex is postulated (Scheme 2) topresent a profound understanding on the elusive metal−iodosylarene chemistry, which will greatly stimulate the develop-ment of rational nonheme catalyst design.

■ ASSOCIATED CONTENT

*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/jacs.7b03310.

The detailed computational methodology, energetic andgeometric data and Cartesian coordinates (PDF)

■ AUTHOR INFORMATION

Corresponding Authors*wwnam@ewha.ac.kr*wangyong@licp.cas.cn

ORCIDKyung-Bin Cho: 0000-0001-6586-983XWei Sun: 0000-0003-4448-2390Wonwoo Nam: 0000-0001-8592-4867Yong Wang: 0000-0002-7668-2234Author Contributions#These authors contributed equally to this work.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThe authors acknowledge financial support from NationalNatural Science Foundation of China to Y.W. (grants21003116 and 21173211) and to W.S. (grant no. 21473226),the NRF of Korea through the CRI (NRF-2012R1A3A2048842)and GRL (NRF-2010-00353) to W.N and MSIP(2013R1A1A2062737) to K.-B.C.).

■ REFERENCES(1) (a) Stang, P. J.; Zhdankin, V. V. Chem. Rev. 1996, 96, 1123.(b) Zhdankin, V. V.; Stang, P. J. Chem. Rev. 2002, 102, 2523.(c) Yoshimura, A.; Zhdankin, V. V. Chem. Rev. 2016, 116, 3328.(2) (a) Carmalt, C. J.; Crossley, J. G.; Knight, J. G.; Lightfoot, P.;Martın, A.; Muldowney, M. P.; Norman, N. C.; Orpen, A. G. J. Chem.Soc., Chem. Commun. 1994, 2367. (b) Wegeberg, C.; Frankær, C. G.;McKenzie, C. J. Dalton Trans. 2016, 45, 17714.(3) Zhdankin, V. V.; Stang, P. J. Chem. Rev. 2008, 108, 5299.(4) (a) Groves, J. T.; Nemo, T. E.; Myers, R. S. J. Am. Chem. Soc. 1979,101, 1032. (b) Groves, J. T.; Nemo, T. E. J. Am. Chem. Soc. 1983, 105,5786. (c) Groves, J. T. J. Chem. Educ. 1985, 62, 928.(5) (a) Groves, J. T.; McClusky, G. A. J. Am. Chem. Soc. 1976, 98, 859.(b) Groves, J. T.; McClusky, G. A.; White, R. E.; Coon, M. J. Biochem.Biophys. Res. Commun. 1978, 81, 154.(6) (a) Shaik, S.; Filatov, M.; Schroder, D.; Schwarz, H. Chem. - Eur. J.1998, 4, 193. (b) Ogliaro, F.; Harris, N.; Cohen, S.; Filatov, M.; deVisser, S. P.; Shaik, S. J. Am. Chem. Soc. 2000, 122, 8977. (c) Shaik, S.; deVisser, S. P.; Ogliaro, F.; Schwarz, H.; Schroder, D. Curr. Opin. Chem.Biol. 2002, 6, 556.(7) (a) Ortiz de Montellano, P. R.; De Voss, J. J. Nat. Prod. Rep. 2002,19, 477. (b) Shaik, S.; Kumar, D.; de Visser, S. P.; Altun, A.; Thiel, W.Chem. Rev. 2005, 105, 2279.(8) (a) Nam, W.; Valentine, J. S. J. Am. Chem. Soc. 1990, 112, 4977.(b) Yang, Y.; Diederich, F.; Valentine, J. S. J. Am. Chem. Soc. 1990, 112,7826.(9) (a) Collman, J. P.; Chien, A. S.; Eberspacher, T. A.; Brauman, J. I. J.Am. Chem. Soc. 2000, 122, 11098. (b) Collman, J. P.; Zeng, L.; Decreau,R. A. Chem. Commun. 2003, 2974.(10) (a) Nam, W.; Choi, S. K.; Lim, M. H.; Rohde, J.-U.; Kim, I.; Kim,J.; Kim, C.; Que, L., Jr. Angew. Chem., Int. Ed. 2003, 42, 109.(11) (a) Song, W. J.; Sun, Y. J.; Choi, S. K.; Nam, W. Chem. - Eur. J.2006, 12, 130. (b) Lennartson, A.; McKenzie, C. J.Angew. Chem., Int. Ed.2012, 51, 6767. (c) Hong, S.; Wang, B.; Seo, M. S.; Lee, Y.-M.; Kim, M.J.; Kim, H. R.; Ogura, T.; Garcia-Serres, R.; Clemancey, M.; Latour, J.-M.; Nam, W. Angew. Chem., Int. Ed. 2014, 53, 6388. (d) Wang, B.; Lee,Y.-M.; Seo, M. S.; Nam, W. Angew. Chem., Int. Ed. 2015, 54, 11740.(e) Wang, C.; Kurahashi, T.; Fujii, H. Angew. Chem., Int. Ed. 2012, 51,7809.(12) (a) Park, J.; Morimoto, Y.; Lee, Y.-M.; Nam, W.; Fukuzumi, S. J.Am. Chem. Soc. 2011, 133, 5236. (b) Park, J.; Morimoto, Y.; Lee, Y.-M.;Nam, W.; Fukuzumi, S. J. Am. Chem. Soc. 2012, 134, 3903.(13) (a) Li, C.; Zhang, L.; Zhang, C.; Hirao, H.; Wu, W.; Shaik, S.Angew. Chem., Int. Ed. 2007, 46, 8168. (b) Shaik, S.; Wang, Y.; Chen, H.;Song, J.; Meir, R. Faraday Discuss. 2010, 145, 49.(14) (a) Kim, S. J.; Latifi, R.; Kang, H. Y.; Nam, W.; de Visser, S. P.Chem. Commun. 2009, 1562. (b) Barea, G.;Maseras, F.; Lledos, A.New J.Chem. 2003, 27, 811.(15) (a) Wang, Y.; Janardanan, D.; Usharani, D.; Han, K.; Que, L., Jr.;Shaik, S. ACS Catal. 2013, 3, 1334. (b) Oloo, W. N.; Meier, K. K.; Wang,Y.; Shaik, S.; Munck, E.; Que, L., Jr. Nat. Commun. 2014, 5, 3046.(16) (a) Poulos, T. L.; Finzel, B. C.; Gunsalus, I. C.; Wagner, G. C.;Kraut, J. J. Biol. Chem. 1985, 260, 16122. (b) Sono, M.; Roach, M. P.;Coulter, E. D.; Dawson, J. H. Chem. Rev. 1996, 96, 2841.

Journal of the American Chemical Society Communication

DOI: 10.1021/jacs.7b03310J. Am. Chem. Soc. 2017, 139, 7444−7447

7447