-
DaltonTransactions
PERSPECTIVE
Cite this: Dalton Trans., 2019, 48,9469
Received 2nd April 2019,Accepted 1st May 2019
DOI: 10.1039/c9dt01402k
rsc.li/dalton
Structure and reactivity of the first-row d-blockmetal-superoxo
complexes
Shunichi Fukuzumi, *a,b Yong-Min Lee *a,c and Wonwoo Nam
*a,d
In the first-row of d-block metals, ten elements are included,
such as scandium (Sc, 3d1), titanium (Ti,
3d2), vanadium (V, 3d3), chromium (Cr, 3d54s1), manganese (Mn,
3d5), iron (Fe, 3d6), cobalt (Co, 3d7),
nickel (Ni, 3d8), copper (Cu, 3d104s1) and zinc (Zn, 3d10). The
synthesis, characterization, and reactivity of
first-row d-block metal-superoxo complexes are discussed
together with the structures of the end-on
(η1) and side-on (η2) metal-superoxo complexes in this review
article. Electron transfer from electrondonors to O2 is enhanced by
binding of Sc
3+ to produce an end-on type Sc(III)-superoxo complex.
Metal-
superoxo complexes such as Ti(IV)-superoxo,
oxovanadium(V)-superoxo, Cr(III)-superoxo, Fe(III)-superoxo,
Co(III)-superoxo, Ni(III)-superoxo and Cu(II)-superoxo species
generally undergo hydrogen atom transfer
reactions. A Cr(III)-superoxo complex undergoes not only
hydrogen atom transfer but also oxygen atom
transfer reactions. In the presence of protons (e.g.,
trifluoromethanesulfonic acid, HOTf), much enhanced
acid catalysis was observed in oxygen atom transfer reactions
from a nonheme Cr(III)-superoxo complex,
[(Cl)(TMC)CrIII(O2)]+, to thioanisole. The enhanced reactivity
of [(Cl)(TMC)CrIII(O2)]
+ by HOTf results from
proton-coupled electron transfer (PCET) from electron donors,
including thioanisole, to [(Cl)(TMC)
CrIII(O2)]+. A manganese(IV)-superoxo complex plays a very
important role in thermal and photoinduced
dioxygen activation by a Mn(III) corrolazine complex. A
metal-superoxide complex using the last element
in the first-row of transition metals, that is a
Zn(II)-superoxide complex, is produced to accelerate the
reduction of O2•− in a SOD (superoxide dismutase) model.
1. Introduction
The first step in the dioxygen activation cycle of
metalloen-zymes is the binding of dioxygen to metal centre by
electrontransfer from the metal centre to O2, forming
metal-superoxocomplexes, in which the nature of metal ions as well
asligands modulates the redox reactivity of the superoxideion.1–15
Thus, metal-superoxo complexes play crucial roles askey
intermediates in various biological redox reactions ofheme enzymes
[e.g., cytochrome c oxidases (COX) and cyto-chrome P450], nonheme
iron enzymes [e.g., isopenicillin Nsynthase (IPNS), myo-inositol
oxygenase (MIOX), and cysteinedioxygenase (CDO)] and copper enzymes
[e.g., dopamineβ-monooxygenase (DβM) and
peptidylglycine-α-amidating
monooxygenase (PHM)].1–15 The first-row d-block metals, suchas
Mn, Fe, Co, Ni, Cu and Zn, are used in various metalloen-zymes.
Among these transition metals, an X-ray diffractionstructure of a
cobalt(III)-superoxo complex supported by bzacenand pyridine
ligands (bzacen = N,N′-ethylene-bis(benzoylaceto-niminide) was
reported for the first time in 1972.16 Theobserved O–O distance of
126(4) pm is comparable to thesuperoxide ion value (128 pm) and is
therefore in accord withthe Co(III)–O2
•− formulation as proposed from EPR studies.17
The Co–O–O bond angle of 126(2)° observed for the end-ontype
coordination of O2
•− agrees with that predicted by Paulingin 1948,18,19 and
differs from the alternative sideways coordi-nation postulated by
Griffith.20 Since then, there have so farbeen many papers and
reviews on the structure and reactivity ofthe first-row d-block
metal-superoxo complexes, mainly focusingon Fe, Co and Cu.21–34
However, no research studies on thestructure and reactivity of
metal-superoxo complexes coveringall the ten first-row metals from
Sc to Zn have yet to be reported.
This review focuses on the structure and reactivity of
metal-superoxo complexes covering all the first-row d-block
metalsfrom Sc to Zn. Firstly, the structure and reactivity of
Sc3+–O2
•−
complexes are discussed, including how Sc3+–O2•− complexes
are produced and characterized. Then, the structure and
redoxreactivity of the other d-block metal-superoxo complexes
are
aDepartment of Chemistry and Nano Science, Ewha Womans
University, Seoul
03760, Korea. E-mail: [email protected],
[email protected],
[email protected] School of Science and Technology,
Meijo University, Nagoya, Aichi 468-
8502, JapancResearch Institute for Basic Sciences, Ewha Womans
University, Seoul 03760, KoreadState Key Laboratory for Oxo
Synthesis and Selective Oxidation, Suzhou Research
Institute of LICP, Lanzhou Institute of Chemical Physics (LICP),
Chinese Academy of
Sciences, Lanzhou 730000, China
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9469
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
.
View Article OnlineView Journal | View Issue
www.rsc.li/daltonhttp://orcid.org/0000-0002-3559-4107http://orcid.org/0000-0002-5553-1453http://orcid.org/0000-0001-8592-4867http://crossmark.crossref.org/dialog/?doi=10.1039/c9dt01402k&domain=pdf&date_stamp=2019-07-03https://doi.org/10.1039/c9dt01402khttps://pubs.rsc.org/en/journals/journal/DThttps://pubs.rsc.org/en/journals/journal/DT?issueid=DT048026
-
discussed in the order of the periodic table, such as Sc, Ti,
V,Cr, Mn, Fe, Co, Ni, Cu and Zn.
2. Scandium-superoxo complexes
Superoxide ions (O2•−) can be produced by photoinduced elec-
tron transfer (ET) from dimeric
1-benzyl-1,4-dihydronicotin-amide [(BNA)2] to O2 as shown in Scheme
1,
35 where the photo-induced ET is followed by a facile cleavage
of the C–C bondof the BNA dimer radical cation to produce
theN-benzylnicotinamide radical (BNA•) and BNA+.36,37 Then, ETfrom
BNA• to O2 occurs rapidly to produce BNA
+ and O2•−
(Scheme 1), because the one-electron oxidation potential ofBNA•
(Eox vs. SCE = −1.08 V)38 is more negative than the one-electron
reduction potential of O2 (Ered vs. SCE = −0.87 V)39 inacetonitrile
(MeCN).35 Thus, two equivalents of O2
•− are pro-duced by photoinduced ET from (BNA)2. In the presence
ofscandium triflate (Sc(OTf)3), Sc
3+ is bound to O2•− to produce
the Sc3+–O2•− complex, which becomes more stable in the
pres-
ence of three equivalents of hexamethylphosphoric triamide(HMPA)
ligand.35 Thus, the EPR spectrum of the (HMPA)3Sc
3+–
O2•− complex was observed even at 60 °C in propionitrile,
exhi-
biting the clear eight lines due to the superhyperfine
couplingof O2
•− with the Sc nucleus (I = 7/2, aSc = 3.82 G).35
When dioxygen is enriched in 17O2, two sets of six lines dueto
the hyperfine splitting of two inequivalent 17O atoms (I =5/2) are
observed as shown in Fig. 1, although the centrelines are
overlapped by the strong eight-line signal of(HMPA)3Sc
3+–16O2•−.35 The two inequivalent a(17O) values are
determined to be 21 and 14 G by comparison of the observedsignal
(Fig. 1a) with the computer simulation lines (Fig. 1b).35
Such inequivalent a(17O) values clearly indicate an
“end-on”coordination of the O2
•− ligand to the Sc3+ centre in the(HMPA)3Sc
3+–O–O•− complex, where the unpaired electron ismore localized
at the terminal oxygen.35
A Sc3+-superoxo (ScO22+) complex is reported to act as an
intermediate for the catalytic two-electron reduction of O2
bydecamethylferrocene (Fc*) with mononuclear copper com-plexes,
[(tmpa)CuII(CH3CN)](ClO4)2 (tmpa = tris(2-pyridyl-methyl)amine) and
[(BzQ)CuII(H2O)2](ClO4)2 (BzQ =
bis(2-quinolinylmethyl)benzylamine), in the presence of Sc(OTf)3
inacetone.40 It was confirmed that the one-electron oxidation
ofScO2
+ afforded Sc3+–O2•− that was characterized by EPR
(vide supra).40 The catalytic cycle is shown in Scheme 2, where
ETfrom Fc* to [LCuII]2+ produces the Cu(II)-superoxo ([LCuIIO2]
+)complex, which is replaced by the Sc3+-superoxo (ScO2
2+)complex, followed by ET from Fc* to ScO2
2+ to produce theSc3+-peroxide (ScO2
+) complex and regenerate [LCuII]2+. The re-placement of
[LCuII]2+ in [LCuIIO2]
+ by Sc3+ to yield ScO22+
results from the stronger Lewis acidity of Sc3+ that binds
withO2
•− much stronger than [LCuIIO2]+. The stronger the Lewis
acidity of metal ions, the stronger the binding of metal ions
toO2
•−.41–43
(HMPA)3Sc3+–O–O•− acts as a one-electron reductant to
reduce p-benzoquinone derivatives (X-Q) to produce(HMPA)3Sc
3+-semiquinone radical anion (X-Q•−) complexes.44
The number of (HMPA)3Sc3+ ion binding to semiquinone
radical anions is changed depending on the type of X-Q.44
In the case of p-benzoquinone derivatives with
electron-with-drawing substituents (Cl2Q, Cl4Q and F4Q), ET
from(HMPA)3Sc
3+–O2•− to X-Q occurs, followed by rapid binding
of (HMPA)3Sc3+ to X-Q•− (Scheme 3), when the ET rate con-
stant exhibited no dependence on the concentration of
Shunichi Fukuzumi
Shunichi Fukuzumi received B.S.,M.S. and Ph.D. degrees
inChemical Engineering andApplied Chemistry at the TokyoInstitute
of Technology in 1973,1975 and 1978, respectively. Heworked as a
postdoctoralresearcher at Indiana Universityin the USA from 1978 to
1981. In1981, he became an AssistantProfessor at Osaka
Universitywhere he was promoted to a FullProfessor in 1994. He
wasselected as a Distinguished
Professor in 2013 and retired from Osaka University in 2015.
Hehas studied electron transfer chemistry, particularly
bioinspiredartificial photosynthesis, metal-ion-coupled electron
transfer(MCET) and redox catalysis. He is now a Distinguished
Professorof Ewha Womans University in the Republic of Korea, a
ProfessorEmeritus of Osaka University and a Designated Professor of
MeijoUniversity in Japan.
Yong-Min Lee
Yong-Min Lee received Bachelor,Master and Ph.D. degrees
inChemistry at Pusan NationalUniversity, Republic of Korea in1990,
1995 and 1999, respect-ively, under the supervision ofProfessor
Sung-Nak Choi. Then,he joined the Centro di Ricercadi Risonanze
Magnetiche(CERM) at Università degli Studidi Firenze (Italy) as a
postdoc-toral researcher under the super-vision of Professors Ivano
Bertiniand Claudio Luchinat from 1999
to 2005. Then, he moved to the Centre for Biomimetic Systems
atEwha Womans University, as a Research Professor (2006–2009).He is
now a Special Appointment Professor at Ewha WomansUniversity from
2009.
Perspective Dalton Transactions
9470 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
(HMPA)3Sc3+.44 In the case of p-benzoquinone derivatives
with electron donating substituents (Me2Q and MeQ), ETfrom
(HMPA)3Sc
3+–O2•− to X-Q is coupled with the binding
of two (HMPA)3Sc3+ molecules to X-Q•− (Scheme 4), when
Scheme 1 Photoinduced generation of O2•− via photoinduced ET
from
(BNA)2 to O2 and binding of Sc3+ to O2
•−.35
Fig. 1 (a) EPR spectrum detected by photoirradiation of an 17O
(40%)oxygen-saturated propionitrile solution containing (BNA)2 (6.9
× 10
−3 M),Sc(OTf)3 (8.1 × 10
−2 M) and HMPA (2.5 × 10−1 M) using a high pressuremercury lamp
at 298 K. (b) Computer simulation spectrum using theparameters, g =
2.0165, a(Sc) = 3.82 G, a(17O1) = 21 G, a(17O2) = 14 G andΔHmsl =
3.5 G. Reprinted with permission from ref. 35. Copyright
1999,American Chemical Society.
Wonwoo Nam
Wonwoo Nam received his B.S.(honours) degree in Chemistryfrom
California State University,Los Angeles, and his Ph.D.degree in
Inorganic Chemistryfrom the University of California,Los Angeles
(UCLA) in the USAunder the supervision ofProfessor Joan S.
Valentine in1990. After working as a post-doctoral researcher at
the UCLAfor one year, he became anAssistant Professor at
HongikUniversity in 1991. In 1994, he
moved to Ewha Womans University, where he is currently
aDistinguished Professor. His research interests are on O2
acti-vation, water oxidation, metal–oxygen intermediates, such
asmetal-oxo, metal-superoxo, metal-peroxo and
metal-hydroperoxospecies, and important roles of metal ions in
bioinorganicchemistry.
Scheme 2 Catalytic cycle of the 2e− reduction of O2 by Fc* with
Sc3+
and [LCuII]2+ complexes.
Scheme 3 ET from (HMPA)3Sc3+–O–O•− to Cl4Q, followed by
binding
of Sc(HMPA)33+ to Cl4Q
•−.
Scheme 4 ET from (HMPA)3Sc3+–O–O•− to Q, coupled by binding
of
two (HMPA)3Sc3+ molecules to Q•−.
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9471
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
the ET rate constant exhibited second-order dependence onthe
concentration of (HMPA)3Sc
3+ in addition to first-orderdependence as observed for ET from
(HMPA)3Sc
3+–O2•−
to Q.44
3. Titanium(IV)-superoxo complexes
Two titanium ozonide complexes, OTiIV(η2-O2)(η2-O3)
andOTiIV(η2-O2)(η1-O3), were produced via the reactions of Tiatoms
with O2 in solid Ar.
45 Their geometric structures andchemical bondings were
clarified via matrix isolation infrared(IR) absorption spectroscopy
and density functional theory(DFT) calculations (Fig. 2).45 The
formation of an end-on-bonded OTiIV(η2-O2)(η1-O3) complex is
accompanied by thedecay of a side-on-bonded OTiIV(η2-O2)(η2-O3)
complex undervisible light (λ = 532 nm) irradiation and this is
reversed uponannealing.45
The light yellow Ti(IV)-superoxo (TiIV–O2•−) complex was
prepared by the reaction of H2O2 (50%) on Ti(OR)4 in anhy-drous
methanol at 298 K [eqn (1)]. The TiIV–O2
•− complexwas characterised by EPR, FTIR, Raman, X-ray
diffraction(XRD), thermogravimetric/differential thermal analysis
(TG/DTA), and elemental analysis.46 The TiIV–O2
•− catalystefficiently catalysed the oxidation of primary amines
by H2O2to the corresponding nitro derivatives in high yields.46
Thecatalytic cycle is started by hydrogen atom transfer from
theamine to the TiIV(O2
•−) species to generate the transientRNH• radical, which is
further oxidized by H2O2 to yield thenitro derivatives.46 The
TiIV(O2
•−) species also catalysed theoxidative esterification of
aldehydes with alkylarenes or alco-hols to afford the corresponding
benzyl and alkyl esters inexcellent yields.47 The TiIV–O2
•− species were detected by
EPR in the treatment of titanium silicate-1 with H2O2 in thegas
phase.48
ð1Þ
4. Vanadium(V)-superoxo complexes
Vanadium(V)-superoxo species are believed to be involved inmany
vanadium-catalysed oxidation reactions.49–51 An
oxo-peroxovanadium(V) complex, [V(L-N4Me2)(O)(O2)]
+ (1: seeScheme 5 for the molecular structure of L-N4Me2),
wasobtained from the reaction of the dioxovanadium(V)
complex,[V(L-N4Me2)(O)2]
+ (2), with H2O2 (30% aqueous) in MeCN(Scheme 5).52 The side-on
coordination of the peroxo group tothe V(V) centre in 1 was
characterized by X-ray crystal structureand 1H NMR and IR
spectroscopy.52 The electrochemical oxi-dation of
[V(L-N4Me2)(O)(O2)]
+ was performed by applying apotential of 1.63 V vs. SCE at −30
°C in the cavity of an EPRspectrometer. The solution EPR spectrum
(Fig. 3a) is com-posed of a signal at an isotropic g value (giso =
2.0119), whichis split into eight lines with a hyperfine coupling
constant of2.50 × 10−4 cm−1 due to the vanadium nucleus (I =
7/2).52 Thisresult indicates that one electron is obtained from a
molecularorbital on the peroxo ligand of [V(L-N4Me2)(O)(O2)]
+ (1) duringelectrochemical oxidation to produce the
oxovanadium(V)-superoxo species, [V(L-N4Me2)(O)(O2
•−)]2+ (3), as shown inScheme 5.52
When the temperature of an MeCN solution of 3 was raisedto
higher than −30 °C, an ESR spectrum corresponding to the
Fig. 2 Optimized structures of the OTi(η2-O2)(η2-O3) and
OTi(η2-O2)(η1-O3) complexes (bond distances in Å and bond angles in
degrees) onthe left, the electron density ρ contour maps in the
middle, and thedensity gradient maps ∇2ρ on the right (green lines
denote regions ofelectronic charge connection, and black lines
denote regions of elec-tronic charge depletion). Reprinted with
permission from ref. 45.Copyright 2007, American Chemical
Society.
Scheme 5 Formation of a vanadium(V)-peroxo complex
([VV(L-N4Me2)(O)(O2)]
+) and electrochemical reduction to produce the
[VV(L-N4Me2(O)(O2
•−)]2+ species. Reprinted with permission from ref. 52.
Copyright2001, WILEY-VCH Verlag GmbH.
Perspective Dalton Transactions
9472 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
vanadyl(IV) complex, [VIV(L-N4Me2)(MeCN)(O)](ClO4)2 (4-(ClO4)2),
exhibited eight lines centred at giso = 1.977 with ahyperfine
coupling constant of 92.8 × 10−4 cm−1 (Fig. 3b).52 4was produced by
the release of O2 via intramolecular ET fromthe superoxo moiety to
the V(V) centre of 3.52 4 was preparedindependently from the
vanadyl(IV) complex, [VIV(L-N4Me2)(ClO)](ClO4), by treatment with
AgClO4 (Scheme 5).
52 A quanti-tative conversion of 4 to 1 was observed with O2
using an anhy-drous MeCN–THF mixture (v/v = 1 : 1) as a
solvent(Scheme 5).52 Since the one-electron reduction potential of
3 isrelatively high (1.63 V vs. SCE), the high oxidising ability of
3 isexpected.52
The addition of non-redox metal ions such as Al3+ actingas a
Lewis acid can facilitate dioxygen activation by anoxovanadium(IV)
complex, [VIV(O)(Cl)(TPA)]PF6 (TPA =
tris-[(2-pyridyl)methyl]-amine), leading to efficient hydrogen
atomtransfer from cyclohexadiene at ambient temperature. In
theabsence of a Lewis acid, hydrogen transfer from cyclohexa-diene
to the oxovanadium(IV) complex occurred much moreslowly.53 The
acceleration effect of Al3+ results from the dis-sociation of Cl−
from [VIV(O)(Cl)(TPA)]PF6 to generate avacant site for O2 binding
to produce the vanadium(V)-super-oxo species, which may be
stabilized by binding Al3+
(Scheme 6).53 The generated oxovanadium(V)-superoxospecies
undergo hydrogen atom transfer from cyclohexadieneto produce a
cyclohexadienyl radical and oxovanadium(V)-hydroperoxide, which
reacts with H2O to produce H2O2 andoxovanadium(V)-hydroxo species
(Scheme 6).53 Hydrogenatom transfer from the cyclohexadienyl
radical to theoxovanadium(V)-hydroxo species affords the formation
ofbenzene and the regeneration of the oxovanadium(IV)complex
(Scheme 6).53
5. Chromium(III)-superoxocomplexes
The reaction of Cr2+ with O2 yields a chromium(III)-superoxoion,
Cr(O2
•−)2+, in an aqueous solution.54 Unlike most of theother
transition metal–oxygen adducts, Cr(O2
•−)2+ can behandled at room temperature even under
air-freeconditions.54a,55 The X-ray crystal structure of a
Cr(III)-superoxocomplex, [CrIII(O2
•−)(TptBu,Me)(pz′H)]BARF (TptBu,Me =
hydrotris(3-tert-butyl-5-methylpyrazolyl)borate, pz′H =
3-tert-butyl-5-methylpyrazole and BARF =
tetrakis(3,5-bis(trifluoromethyl)phenyl)borate), which was
generated by the reaction of[CrII(TptBu,Me)(pz′H)]BARF with O2, is
shown in Fig. 4, wherethe two Cr–O distances 1.861(4) and 1.903(4)
Å are essentiallyidentical in the side-on coordination of the
O2
•− ligand.56 TheO–O bond length of 1.327(5) Å is consistent with
the superoxocategory.56 The solid-state IR spectrum of
[CrIII(O2
•−)(TptBu,Me)(pz′H)]BARF exhibited an O–O stretching vibration
at1072 cm−1, which was shifted to 1007 cm−1 when 16O2 wasreplaced
by 18O2.
56 Antiferromagnetic coupling between theCrIII ion (d3, S = 3/2)
and the coordinated O2
•− (S = 1/2) gavethe effective magnetic moment of the
CrIII-superoxo complex(μeff (295 K) = 2.8(1)μB).
56
Fig. 3 (a) EPR spectrum of [VV(L-N4Me2(O)(O2•−)]2+ generated in
elec-
trolysis of [VV(L-N4Me2)(O)(O2)]+ in MeCN containing 0.10 M
Et4NClO4
with an applied potential of 1.63 V vs. SCE at −30 °C in the EPR
cavityand (b) after raising temperature to room temperature.
Reprinted withpermission from ref. 52. Copyright 2001, WILEY-VCH
Verlag GmbH.
Scheme 6 Catalytic mechanism for desaturation of cyclohexadiene
tobenzene by O2 with [V
IV(O)(Cl)(TPA)]PF6. Reprinted with permission fromref. 53.
Copyright 2017, American Chemical Society.
Fig. 4 ChemDraw and X-ray structures of a
chromium(III)-superoxocomplex, [TptBu,MeCrIII(pz’H)(O2)]BARF.
Reprinted with permission fromref. 56. Copyright 2002, WILEY-VCH
Verlag GmbH.
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9473
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
A chromium(II) complex, [(Cl)(TMC)CrII]+ (TMC =
1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane) (Fig. 5a),
reactswith O2 to yield an end-on chromium(III)-superoxo
complex,[(Cl)(TMC)CrIII(O2
•−)]+ (Fig. 5b),57 where the O–O bond length(1.231(6) Å) of the
end-on Cr(III)-superoxo complex is shorterthan that of the side-on
Cr(III)-superoxo complex (1.327 Å inFig. 4b).56–58 The resonance
Raman (rRaman) spectrum of[(Cl)(TMC)CrIII(O2)]
+ exhibited an O–O stretching vibrationband at 1170 cm−1, which
shifts to 1104 cm−1 when 18O2 wasused instead of 16O2.
57 This value is comparable to thosereported for other end-on
Cr(III)-superoxo complexes, such as[(H2O)5Cr
III(O2•−)]2+ (1166 cm−1)59 and [(H2O)(cyclam)
CrIII(O2•−)]2+ (1134/1145 (doublet) cm−1),59 but higher than
that of the side-on Cr(III)-superoxo complex (1072 cm−1),56
agreeing with the shorter O–O bond distance of the
end-onCr(III)-superoxo complex compared to that of the side-on
Cr(III)-superoxo complex.56,57
The structural and vibration data of [(Cl)(TMC)CrIII(O2)]+
are listed in Table 1 together with the data of other
metal-superoxo complexes (vide infra).
The Cr(III)-superoxo complex ([(Cl)(TMC)CrIII(O2•−)]+) acts
as
an unusual three electron oxidant in the oxidation of anNADH
analogue, 1-benzyl-1,4-dihydronicotinamide (BNAH), asshown in eqn
(2),
ð2Þ
where the 3 : 2 stoichiometry entails the removal of six
elec-trons from BNAH by 2 molecules of [(Cl)(TMC)CrIII(O2
•−)]+ (athree electron oxidant) to produce BNA+ and
[(Cl)(TMC)CrIII(OH)]+.60 The oxidation of BNAH by
[(Cl)(TMC)CrIII(O2
•−)]+
is started by hydride transfer from BNAH to
[(Cl)(TMC)CrIII(O2
•−)]+ to produce BNA+ and [(Cl)(TMC)CrII(OOH)][eqn (3)],60
followed by the O–O bond heterolysis of [(Cl)(TMC)CrII(OOH)] to
produce the Cr(IV)-oxo complex, [(Cl)(TMC)CrIV(O)]+ [eqn (4)].
Then, hydrogen atom transfer from BNAH
to [(Cl)(TMC)CrIV(O)]+ occurs to produce BNA• and
[(Cl)(TMC)CrIII(OH)]+ [eqn (5)]. This is followed by fast ET from
BNA•,which is a strong electron donor (Eox = −1.1 V vs. SCE),38
to[(Cl)(TMC)CrIV(O)]+ to produce BNA+ and [(Cl)(TMC)CrIII(O)]+
[eqn (6)] that reacts with H2O to produce
[(Cl)(TMC)CrIII(OH)]+
and OH− [eqn (7)].60 The overall 3 : 2 stoichiometry of the
reac-tion of BNAH and [(Cl)(TMC)CrIII(O2
•−)]+ in eqn (2) is obtainedby summing up equations of (3) × 2,
(4) × 2, (5), (6) and (7).60
BNAHþ ½ðClÞðTMCÞCrIIIðO2Þ�þ ! BNAþ þ ½ðClÞðTMCÞCrIIðOOHÞ�ð3Þ
½ðClÞðTMCÞCrIIðOOHÞ� ! ½ðClÞðTMCÞCrIVðOÞ�þ þ OH� ð4Þ
BNAHþ ½ðClÞðTMCÞCrIVðOÞ�þ ! BNA• þ ½ðClÞðTMCÞCrIIIðOHÞ�þð5Þ
BNA• þ ½ðClÞðTMCÞCrIVðOÞ�þ ! BNAþ þ ½ðClÞðTMCÞCrIIIðOÞ�þð6Þ
½ðClÞðTMCÞCrIIIðOÞ�þ þH2O! ½ðClÞðTMCÞCrIIIðOHÞ�þ þ OH�ð7Þ
When BNAH was replaced by an acid-stable NADH ana-logue,
10-methyl-9,10-dihydroacridine (AcrH2), four-electronoxidation of
AcrH2 by [(Cl)(TMC)Cr
III(O2•−)]+ occurred to yield
10-methylacridone (AcrvO), which is the four-electron oxi-dized
product of AcrH2,
61,62 and [(Cl)(TMC)CrIII(OH)]+ [eqn(8)].60 The 3 : 4
stoichiometry entails the removal of twelve elec-trons from AcrH2
by four molecules of [(Cl)(TMC)Cr
III(O2•−)]+
(a three electron oxidant) to produce AcrvO and
[(Cl)(TMC)CrIII(OH)]+ [eqn (8)].60 The 18O-labeling experiments
confirmedthat the oxygen atom in AcrvO was derived from
[(Cl)(TMC)CrIII(18O2
•−)]+, which was synthesized by reacting [(Cl)(TMC)CrII]+ with
18O2.
63,64
ð8Þ
In the case of BNAH, hydride transfer from AcrH2
to[(Cl)(TMC)CrIII(O2
•−)]+ also occurs to produce AcrH+ and[(Cl)(TMC)CrII(OOH)] [eqn
(9)].60 The Cr(II)-hydroperoxo complex([(Cl)(TMC)CrII(OOH)])
undergoes the O–O bond heterolysis toproduce the Cr(IV)-oxo
complex, [(Cl)(TMC)CrIV(O)]+, and OH−
[eqn (4)]. Then, OH− is added to AcrH+ to produce
AcrH(OH),61–63,65 which transfers an hydride ion to
[(Cl)(TMC)CrIII(O2
•−)]+ to produce AcrvO, [(Cl)(TMC)CrIV(O)]+ and H2O[eqn (11)].60
This is followed by hydrogen atom transfer fromAcrH2 to
[(Cl)(TMC)Cr
IV(O)]+ to produce AcrOH• and[(Cl)(TMC)CrIII(OH)]+ [eqn (12)].60
Finally, ET from AcrOH• to[(Cl)(TMC)CrIV(O)]+ occurs to yield AcrvO
and [(Cl)(TMC)CrIII(OH)]+ [eqn (13)].60 The overall 3 : 4
stoichiometry of the
Fig. 5 X-ray structures of (a) chromium(II) complex
([(Cl)(TMC)CrII]+) and(b) chromium(III)-superoxo species
([(Cl)(TMC)CrIII(O2)]
+). Reprinted withpermission from ref. 57. Copyright 2010,
American Chemical Society.
Perspective Dalton Transactions
9474 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
reaction of AcrH2 and [(Cl)(TMC)CrIII(O2
•−)]+ in eqn (8) isobtained by summing up 3 × [eqn (9) + eqn
(4)] + eqn (10) +eqn (11) + 2 × [eqn (12) + eqn (13)].60
AcrH2 þ ½ðClÞðTMCÞCrIIIðO2Þ�þ ! AcrHþ þ
½ðClÞðTMCÞCrIIðOOHÞ�ð9Þ
ð10Þ
ð11Þ
AcrHðOHÞ þ ½ðClÞðTMCÞCrIVðOÞ�þ
! AcrOH• þ ½ðClÞðTMCÞCrIIIðOHÞ�þ ð12Þ
AcrOH• þ ½ðClÞðTMCÞCrIVðOÞ�þ ! AcrvOþ
½ðClÞðTMCÞCrIIIðOHÞ�þð13Þ
The rate of hydride transfer from AcrH2 (large excess)
to[(Cl)(TMC)CrIII(O2
•−)]+ [eqn (9)], which is the rate-determiningstep in the
overall reaction [eqn (8)], obeyed the pseudo-first-order kinetics,
and the observed second-order rate constant(k2) was determined to
be 2.0 M
−1 s−1 at 253 K from the slopeof the plot of the
pseudo-first-order rate constant (k1) vs. con-centration of AcrH2.
A very large deuterium kinetic isotopeeffect (KIE = 74) was
observed at 253 K in the oxidation ofAcrH2 and a deuterated
compound (AcrD2) by [(Cl)(TMC)CrIII(O2)]
+, indicating the involvement of hydrogen atom tun-nelling in
the hydride-transfer reaction.66,67
A good linear correlation between the log k2 values ofhydride
transfer from NADH analogues to [(Cl)(TMC)CrIII(O2
•−)]+ and those of hydride transfer from the sameNADH analogues
to p-chloranil (Cl4Q) was obtained as shownin Fig. 6.68,69 Such a
linear correlation indicates that hydridetransfer from NADH
analogues to [(Cl)(TMC)CrIII(O2
•−)]+
occurs via a concerted proton-coupled electron transfer(PCET),
followed by a rapid ET as reported for hydride transferfrom NADH
analogues to Cl4Q.
68,69
[(Cl)(TMC)CrIII(O2•−)]+ can also oxidise thioanisole via
direct oxygen atom transfer to yield methyl phenyl sulfoxideand
[(Cl)(TMC)CrIV(O)]+ [eqn (14)].63 The observed second-order rate
constant of sulfoxidation of thioanisole by [(Cl)(TMC)CrIII(O2
•−)]+ (kox) in the absence of HOTf in MeCN wasdetermined to be
3.6 × 10−4 M−1 s−1 at 233 K, where the reac-tion rate was very
slow.70 However, the addition of HOTf (1equiv.) to an MeCN solution
of [(Cl)(TMC)CrIII(O2
•−)]+ andthioanisole resulted in 104-fold enhancement to afford
the rateconstant of 3.5(3) M−1 s−1 at 233 K.70 Thus, HOTf acts as
aremarkable acid-catalyst to accelerate the sulfoxidation
ofthioanisole by [(Cl)(TMC)CrIII(O2
•−)]+ [eqn (14)]. The kox valueof sulfoxidation of
p-methoxythioanisole by [(Cl)(TMC)CrIII(O2)]
+ increased with increasing concentration of HOTf,exhibiting the
second-order dependence on [HOTf] (Fig. 7).70
ð14Þ
As indicated by the negative one-electron reduction poten-tial
of [(Cl)(TMC)CrIII(O2)]
+ (Ered vs. SCE = −0.52 V), no electrontransfer from
[Fe(bpy)3]
2+ (Eox vs. SCE = 1.06 V) to [(Cl)(TMC)CrIII(O2
•−)]+ occurred because the electron transfer is highly
Fig. 6 Plot of log k2 for hydride-transfer reactions of
Cr(III)-superoxocomplex ([(Cl)(TMC)CrIII(O2)]
+) and NADH analogues in MeCN at 253 Kversus log k2 for
hydride-transfer reactions of p-chloranil (Cl4Q) and thesame NADH
analogues in MeCN at 253 K. Reprinted with permissionfrom ref. 60.
Copyright 2017, WILEY-VCH Verlag GmbH.
Table 1 Binding mode, O–O bond distance and O–O bond stretching
frequencies of metal-superoxo complexes
Metal-superoxo complex Binding mode O–O (Å) ν(O–O)a (cm−1)
Ref.
[CrIII(O2)(TptBu,Me)(pz′H)]+ η2 1.327(5) 1072 (1007) 56
[CrIII(O2)(Cl)(TMC)]+ η1 1.231(6) 1170 (1104) 57
[MnIII(O2•−)(L)(H2O)]
2+ η1 1.249(4) 1124 (1035) 75[FeIII(O2)(N-Meimid)TpivPP] η1
1.23(8) 1150 (1074) 86[(TAML)FeIII(O2
•−)]2− η2 1.323(3) 1260 (1183) 90[CoIII(η1-O2•−)(LOiPr)(TpMe2)]
η1 1.301(5) 1168 (1090) 100[NiII(O2
•−)(L)] (L = β-diketiminato anion) η2 1.347(2) 971 (919)
134[CuII(O2)[HB(3-
tBu-5-iPrpz)3]] η2 1.22(3) 1112 (1060) 150
a ν(18O–18O) in parentheses.
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9475
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
endergonic (ΔGet = +1.57 eV).70 In the presence of HOTf,however,
electron transfer from [Fe(bpy)3]
2+ to [(Cl)(TMC)CrIII(O2
•−)]+ was made possible to yield [Fe(bpy)3]3+ and [(Cl)
(TMC)CrIII(H2O2)]+ [eqn (15)].70 The Ered value of
[(Cl)(TMC)
CrIII(O2•−)]+ in the presence of HOTf (2.5 mM) was
determined
to be 1.12 V vs. SCE, which is by 1.64 V more positive than
thevalue without HOTf (−0.52 V vs. SCE).70
½FeðbpyÞ3�2þ½ðClÞðTMCÞCrIIIðO2Þ�þ þ 2Hþ
! Ket ½FeðbpyÞ3�3þ þ ½ðClÞðTMCÞCrIIIðH2OÞ�2þ
ð15Þ
Remarkable acid catalysis in sulfoxidation of
p-methoxy-thioanisole results from binding of two protons to the
super-oxo moiety of [(Cl)(TMC)CrIII(O2
•−)]+ to produce [(Cl)(TMC)CrIII(O2
•−)]+ − (H+)2 (Scheme 7).70 The protonation equilibriumlies far
to the left, when the concentration of [(Cl)(TMC)
CrIII(O2•−)]+ − (H+)2 is proportional to [HOTf]2.70 Although
no electron transfer occurs from p-methoxythioanisole
to[(Cl)(TMC)CrIII(O2
•−)]+ in the absence of HOTf, electrontransfer from
p-methoxythioanisole to [(Cl)(TMC)CrIII(O2
•−)]+ −(H+)2 occurs to produce a p-methoxythioanisole radical
cationand [(Cl)(TMC)CrIII(H2O2)]
+, followed by O•− transfer from[(Cl)(TMC)CrIII(H2O2)]
+ to the p-methoxythioanisole radicalcation to yield
p-methoxyphenyl methyl sulfoxide and[(Cl)(TMC)CrIV(O)]+,
accompanied by the regeneration of twoprotons, indicating that no
HOTf is consumed in the reaction(Scheme 7).70 Thus, HOTf acts as an
efficient acid-catalystrather than a reactant for the sulfoxidation
of p-methoxythio-anisole by [(Cl)(TMC)CrIII(O2
•−)]+ via PCET. Similar catalyticeffects of acids were reported
for sulfoxidation by non-hemeiron(IV)-oxo complexes via
PCET.71–74
6. Manganese(III)-superoxocomplexes
The reaction of [H4L][PF6]4 (H4L =
5,11,17,23-tetrakis-(tri-methylammonium)-25,26,27,28-tetrahydroxy-calix[4]arene)with
four equivalents of Mn(OAc)2·4H2O and O2 in air affordeda
Mn(III)-superoxo complex, [MnIII(O2
•−)(L)(H2O)](PF6)2, whichcontains a bowl-shaped cationic D4d
structure and a linearend-on Mn(III)-O2 structure as shown by the
X-ray crystal struc-ture in Fig. 8.75 The rRaman spectrum (λex =
632.8 nm) of[MnIII(O2
•−)(L)(H2O)]2+ showed the O–O stretching vibration at
1124 cm−1, which was shifted to 1035 cm−1 by 18O-labelling.The
ν(O–O) vibration of the linear end-on Mn(III)-superoxocomplex is
in-between those of the side-on Cr(III)-superoxocomplex (1072
cm−1)56 and the end-on Cr(III)-superoxocomplex (1170 cm−1) (Table
1).57 The magnetic moment for[MnIII(O2
•−)(L)(H2O)]2+ was determined to be 5.8μB, which indi-
cates a high-spin state (S = 5/2) of the Mn(III) species with
theO2
•− ligand.75 Inclusion of the water solvation effect wasrequired
to describe the geometric structure of the linear end-on
[MnIII(O2
•−)(L)(H2O)]2+ complex from a theoretical model.76
Fig. 7 Plot of kox versus [HOTf] for the oxidation of
para-MeO-thio-anisole (0.50 mM) by Cr(III)-superoxo complex
([(Cl)(TMC)CrIII(O2)]
+;0.50 mM) in the presence of HOTf (0–3.0 mM) at 233 K. Inset
shows thesecond-order dependence of kox versus [HOTf]2. Reprinted
with per-mission from ref. 70. Copyright 2018, American Chemical
Society.
Scheme 7 Proposed reaction mechanism of acid-catalysed
oxidationof p-methoxythioanisole by [(Cl)(TMC)CrIII(O2
•−)]+. Reprinted with per-mission from ref. 70. Copyright 2018,
American Chemical Society.
Fig. 8 X-ray crystal structure of a mononuclear
Mn(III)-superoxocomplex, [MnIII(O2
•−)(L)(H2O)]2+. Reprinted with permission from ref. 75.
Copyright 2011, Royal Society of Chemistry.
Perspective Dalton Transactions
9476 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
The Mn(III)-superoxo complex exhibited good reactivity
andselectivity in the catalytic oxidation of alkenes with O2 and
iso-butyraldehyde under mild reaction conditions.75
A manganese(IV)-superoxo complex plays a very importantrole in
photoinduced dioxygen activation by a Mn(III) corrol-azine complex,
[MnIII(TBP8Cz)].
77 Femtosecond laser flashphotolysis of [MnIII(TBP8Cz)] resulted
in the formation of atripquintet excited state (5T1), followed by
the rapid intersys-tem crossing (ISC) to a tripseptet excited state
(7T1), whichreacts with O2 via a diffusion-limited rate constant to
generatethe putative manganese(IV)-superoxo species, [MnIV(O2
•−)-(TBP8Cz)].
77 The MnIV-superoxo complex abstracts a hydrogenatom from
toluene to produce a benzyl radical and the MnIV-hydroperoxo
(MnIV–OOH) complex, both of which further reactvia the O–O bond
cleavage to yield benzyl alcohol and theMnV-oxo complex,
[MnV(O)(TBP8Cz)].
77,78
In the presence of HOTf (1 equiv.), [MnIII(TBP8Cz)] can actas a
photoredox catalyst for the oxygenation of hexamethyl-benzene (HMB)
by O2 as shown in Scheme 8,
79 where[MnIII(TBP8Cz)] is protonated by HOTf to produce the
mono-protonated complex, [MnIII(TBP8Cz(H))(OTf)]. Photoexcitationof
[MnIII(TBP8Cz(H))(OTf)] affords the tripquintet state (
5T1),followed by rapid ISC to produce the tripseptet excited
state(7T1). ET from
7T1 to O2 occurs to produce the MnIV-superoxo
complex, [MnIV(O2•−)(TBP8Cz(H))(OTf)], followed by hydrogen
atom transfer from HMB to produce the MnIV-hydroperoxocomplex,
[MnIV(OOH)(TBP8Cz(H))(OTf)], and pentamethyl-benzyl radical
species, competing with the back ET to regener-ate the ground state
[MnIII(TBP8Cz)] and O2.Pentamethylbenzyl radical species reacts
with [MnIV(OOH)(TBP8Cz(H))(OTf)] via O–O bond cleavage to produce
penta-methylbenzyl alcohol and [MnIV(O)(TBP8Cz(H)
•+)(OTf)], whichis produced by the protonation of
[MnV(O)(TBP8Cz)] via intra-molecular ET from the TBP8Cz ligand to
the Mn
V(O) moiety.79
HMB can also be thermally oxidised by [MnIV(O)(TBP8Cz(H)•+)
(OTf)] via PCET to produce pentamethylbenzyl alcohol,accompanied
by the regeneration of [MnIII(TBP8Cz(H))(OTf)] tocomplete the
catalytic cycle (Scheme 8).79 In the presence ofexcess HOTf,
however, [MnIII(TBP8Cz(H))(OTf)] is further pro-tonated to produce
the diprotonated complex, [MnIII(TBP8Cz(H)2)(OTf)(H2O)](OTf).
The
7T1 excited state of [MnIII(TBP8Cz
(H)2)(OTf)(H2O)](OTf) exhibited no reactivity towards O2
toproduce the MnIV-superoxo complex, because the
one-electronoxidation potential of
[MnIII(TBP8Cz(H)2)(OTf)(H2O)](OTf) ismuch higher than that of
[MnIII(TBP8Cz(H))(OTf)], indicatingthat ET from the 7T1 excited
state to O2 may be too endergonicto occur.80
Manganese porphyrins [MnIII(Porp)], [MnIII(TMP)(OH)](TMP =
5,10,15,20-tetrakis-(2,4,6-trimethylphenyl)porphinatodianion) and
[MnIII(TPFPP)(CH3COO)] (TPFPP =
5,10,15,20-tetrakis(pentafluorophenyl)porphyrinato dianion), also
act aseffective photocatalysts for photodriven oxygenation
of10-methyl-9,10-dihydroacridine by O2 via photoinduced ETfrom
MnIII(Porp) to O2 to produce Mn
IV(O2•−)(Porp) species.62
Manganese(III)-superoxo complexes are often proposed asthe first
intermediates formed upon reaction of Mn(II) withO2.
81,82 For example, a manganese(IV)-superoxo complex is pro-posed
as a putative intermediate for an O2-activation with[MnIII(TPFC)]
(TPFC = 5,10,15-tris(pentafluorophenyl)corrolatotrianion) as shown
in Scheme 9, where ET from [MnIII(TPFC)]with an axially coordinated
OH− (pathway a) to O2 occurs toproduce a putative Mn(IV)-superoxo
complex (pathway b),which abstracts a hydrogen atom from hydrogen
donor mole-cules such as tetrahydrofuran and cyclohexene to
generate theMn(IV)-hydroperoxo complex (pathway c).83 The O–O
bondhomolysis of the Mn(IV)-hydroperoxo complex affords
theMn(V)-oxo complex (pathway d), which can be produced by
thereaction of [MnIII(TPFC)] with PhIO (pathway h).83 In the
pres-ence of excess OH−, the Mn(IV)-peroxo complex can be
formedeither from the reaction of the Mn(V)-oxo complex and OH−
(pathway e) or from the deprotonation of the Mn(IV)-hydro-
Scheme 8 Mechanism of acid-catalysed oxygenation of HMB
by[MnIV(O2
•−)(TBP8Cz(H))(OTf)]. Reprinted with permission from ref.
79.Copyright 2015, American Chemical Society.
Scheme 9 Mechanism of O2-activation and O–O bond formation
reac-tions with [MnIII(corrole)]. Reprinted with permission from
ref. 83.Copyright 2017, American Chemical Society.
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9477
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
peroxo species (pathway f or i).83 The addition of a proton
tothe Mn(IV)-peroxo complex regenerates the
Mn(IV)-hydroperoxocomplex (pathway g).83
7. Iron(III)-superoxo complexes
Iron(III)-superoxo intermediates are known not only in
hemeenzymes but also in the nonheme mononuclear FeII
enzymefamily.84 An Fe(III)–O2
•− intermediate was detected by the reac-tion of FeII-containing
homoprotocatechuate 2,3-dioxygenasewith mutation of the active site
His200 to Asn (H200N).84 AnFe(III)–O2
•− intermediate was also produced under illuminationto an
O2-saturated acetonitrile solution of Fe(bpmcn)Cl2(bpmcn =
(1S,2S)-N,N′-dimethyl-N,N′-bis(2-pyridinylmethyl)cyclohexane-1,2-diamine).85
It was also reported that dioxygenbound with
[FeII(N-Meimid)(α,α,α,α-o-TpivPP)] (N-Meimid =N-methyl imidazole
and α,α,α,α-o-TpivPP =
meso-tetra(α,α,α,α-o-pivalamidephenyl)porphinato dianion)
reversibly intoluene.86 The X-ray crystal structure of the O2
complex,[FeII(O2)(N-Meimid)(α,α,α,α-o-TpivPP)], showed four
pivalamidogroups on one side of the porphyrin to generate a
hydro-phobic pocket of 5.4 Å depth, which encloses “end-on”
coordi-nation of O2 with a bent Fe–O–O bond.
86 The O–O stretchingvibrations were observed as doublets at
1150 and 1155 cm−1
for 16O2 and 1074 and 1079 cm−1 for 18O2.
87 These values werecomparable to those reported for an end-on
Cr(III)-superoxocomplex, [Cr(O2
•−)(cyclam)(H2O)]2+ (1134 and 1145 cm−1).59
Cryoreduction of oxy-hemoglobin (oxy-GMH3) from
Glyceradibranchiata, oxy-ferrous octaethyl porphyrin, and
oxy-ferrouscomplex of the heme model (cyclidene complex) resulted
inthe formation of the ferrous-superoxo complexes with nearlyunit
spin density localized on a superoxo moiety.88 The
end-oncoordination of O2
•− to a low-spin ferrous ion is supported bytheir g tensors and
17O hyperfine couplings, which are charac-teristic of the
superoxide ion coordinated to a diamagneticmetal ion. Upon
annealing to T > 150 K, the ferrous-superoxospecies were
converted to peroxo/hydroperoxo-ferricintermediates.88
The one-electron reduction of the Fe(III)-superoxo specieswas
also carried out by γ-ray irradiation of the sample at 77 Kin
MeCN/2-MeTHF (v/v 1 : 4; 2-MeTHF = 2-methyl-tetrahydrofuran) as
shown in Scheme 10.89 The γ-ray irradiatedsample exhibited an Fe–O
stretching vibration at 459 cm−1 thatshifted to 435 cm−1 by
18O-labelling (16–18Δν = −24 cm−1).Thus, the iron–oxygen stretching
(νFe–O) frequency of theferrous-superoxo is significantly lower
than that of the ferric-superoxo complex observed at 579 cm−1.89
The O–O stretchingvibration of the ferrous-superoxo complex could
not beobserved as this mode is known to be difficult to
detect.89
A side-on iron(III)-superoxo non-heme complex
([(TAML)FeIII(O2
•−)]2−) was successfully synthesized by the reaction ofan
iron(III) complex bearing a tetraamido macrocyclic ligand(TAML)
with solid potassium superoxide (KO2) in the presenceof
2.2.2-cryptand (6 equiv.) in CH3CN at 5 °C.
90 The X-raycrystal structure of [(TAML)FeIII(O2
•−)]2− showed two crystallo-
graphically independent mononuclear side-on 1 : 1 iron
com-plexes of O2 with O–O bond lengths (O1–O2: 1.323(3) Å andO7–O8:
1.306(7) Å) (Fig. 9), which are significantly shorterthan those of
Fe(III)-peroxo species, such as [FeIII(TMC)(O2)]
+
(1.463(6) Å)91 and naphthalene dioxygenase (ca. 1.45 Å),92
butsimilar to that of Fe(II)-superoxo species found in
homoproto-catechuate 2,3-dioxygenase (1.34 Å).93 The infrared
spectrumof [(TAML)FeIII(O2
•−)]2−, which was recorded in CH3CN at−40 °C, exhibits an O–O
stretching at 1260 cm−1, which shiftsto 1183 cm−1 on substitution
of 16O with 18O. The observed16–18Δν value of 77 cm−1 agrees within
an experimental errorwith the calculated value of 72 cm−1.90
The [(TAML)FeIII(O2•−)]2− complex can abstract a hydrogen
atom from 2,4-di-tert-butyl phenol to yield
2,2′-dihydroxy-3,3′,5,5′-tetra-tert-butylbiphenyl, which is a C–C
bond couplingproduct, as a major product (75% yield based on the
initialconcentration of [(TAML)FeIII(O2
•−)]2−).90 The logarithm of therate constants (log k2) of
hydrogen atom transfer from para-substituted
2,6-di-tert-butylphenols (p-X-2,6-t-Bu2-C6H2OH; X =OMe, Me, H, CN)
to [(TAML)FeIII(O2
•−)]2− is linearly correlatedwith the O–H bond dissociation
energies (BDEs) of p-X-2,6-t-Bu2-C6H2OH.
90 A linear Hammett plot of log k2 vs. σp+ was
obtained with a negative slope (ρ) of −2.5, indicating
that[(TAML)FeIII(O2
•−)]2− acts as an electrophile for the hydrogenatom transfer
reactions.90 Singha and Dey recently reportedthat an
iron(III)-superoxo porphyrin with a covalently attachedhydroquinol
group underwent hydrogen atom abstractionfrom the hydroquinol group
to produce the ferric hydroperox-ide, which performed the second
hydrogen abstraction togenerate the ferryl species.94
Scheme 10 Conversion from FeIII-superoxo porphyrin to
FeII-superoxoporphyrin by one-electron reduction (R = mesityl
group). Reprinted withpermission from ref. 89. Copyright 2015,
Royal Society of Chemistry.
Fig. 9 X-ray crystal structures of [(TAML)FeIII(O2•−)]2−,
showing the two
crystallographically independent moieties. Reprinted with
permissionfrom ref. 90. Copyright 2014, Springer Nature.
Perspective Dalton Transactions
9478 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
The FeIII-superoxo complex ([(TAML)FeIII(O2•−)]2−) also acts
as a nucleophile in the reaction with
2-phenylpropionaldehyde(2-PPA) to yield acetophenone as the
deformylated product(90% yield based on the initial concentration
of [(TAML)FeIII(O2
•−)]2−),90 as reported for other nucleophilic reactions
ofmetal-peroxo complexes.95–98 The nucleophilic reaction
of[(TAML)FeIII(O2
•−)]2− with para-substituted benzaldehydederivatives
(p-Y-C6H4CHO; Y = OMe, Me, H, Cl) afforded theHammett plot with a
positive ρ value of 1.4 in contrast to theelectrophilic hydrogen
atom transfer reactions (vide supra).90
The [(TAML)FeIII(O2•−)]2− complex can also oxidise nitric
oxide (NO) to yield [(TAML)FeIII(NO3)]2− and NO2
− via for-mation of Fe(IIII)-peroxynitrite species
([(TAML)FeIII(O–O–NvO−)]2−), which undergo O–O bond homolysis,
followed byrebound of the resulting Fe(IV)-oxo species,
[(TAML)FeIV(O)]2−,and NO2
•.99
A high-spin mononuclear iron(II) complex supported by
afive-azole donor set, [FeII(LPh)(TpMe2)], is reported to react
withO2 to yield the corresponding mononuclear non-heme
iron(III)-superoxo species, [FeIII(O2
•−)(LPh)(TpMe2)], in THF at −60 °C asshown in Scheme 11.100 The
rRaman spectrum of [FeIII(O2
•−)(LPh)(TpMe2)] exhibited an O–O bond stretching vibration
at1168 cm−1, which shifts to 1090 cm−1 on substitution of 16Owith
18O.100 The 16–18Δν value of 78 cm−1 is consistent with
thecalculated value of 72 cm−1. The ν(O–O) value of 1168 cm−1
isnearly the same as that of the end-on Cr(III)-superoxo
complex,[(Cl)(TMC)CrII(O2
•−)]+, with ν(O–O) of 1170 cm−1.57 Thus, thesuperoxo ligand of
[FeIII(O2
•−)(LPh)(TpMe2)] is suggested to becoordinated to the sixth site
of the Fe(III) centre with an end-onmode.100 A somewhat smaller
ν(O–O) value (1125 cm−1) wasreported for a Fe(III)-superoxo
complex, [FeIII(O2
•−)(BDPP)](H2BDPP =
2,6-bis(((S)-2-(diphenylhydroxymethyl)-1-pyrrolidi-
nyl)methyl)-pyridine), in which an end-on coordination ofO2
•− to the Fe(III) centre is also suggested.101 A similar
ν(O–O)value (1120 cm−1) was recently reported for an
Fe(III)-superoxocomplex, [FeIII(O2
•−)TpMe2(2-ATP)] (2-ATP = 2-aminothiopheno-late).102 DFT
calculations suggest the lowest-energy quintetstructure as shown in
Fig. 10, where the O2
•− ligand forms ahydrogen bond with the –NH2 donor of 2-ATP, as
indicated bythe O⋯H distance of 1.81 Å and also by the smaller
Fe–O–Oangle of 122°.102
The [FeIII(O2•−)(LPh)(TpMe2)] complex exhibited a hydrogen
atom abstraction ability from substrates having a weak X–Hbond,
where X = O or N; BDE of X–H < 72.6 kcal mol−1, suchas
2-hydroxy-2-azaadamantane (AZADOL) and phenylhydra-zine.100
Hydrogen atom transfer from AZADOL to [FeIII(O2
•−)(LPh)(TpMe2)] afforded an iron(III)-hydroperoxo
complex,[FeIII(OOH)(LPh)-(TpMe2)] (Scheme 11).100 The [FeIII(O2
•−)(TpMe2)(2-ATP)] complex can also abstract hydrogen atomsfrom
9,10-dihydroanthracene (DHA) to yield anthracene.102
When DHA-d4 was used as a substrate, a deuterium kineticisotope
effect (KIE = 7) was observed, indicating that C–Hbond cleavage is
the rate-determining step.102
An alkyl thiolate-ligated iron(II) complex can also react withO2
to form an Fe(III)-superoxo intermediate, [Fe
III(O2•−)-
(S2Me2N3(Pr,Pr))], in THF at −73 °C.103 The rRaman spectrum
of [FeIII(O2•−)(S2
Me2N3(Pr,Pr))] showed O–O stretchingvibrations at 1093 and 1122
cm−1 that shift to 1022 cm−1 when16O2 was replaced by
18O2.103 The DFT calculated structure of
[FeIII(O2•−)(S2
Me2N3(Pr,Pr))] contains an O2 moiety cis to one ofthe thiolate
sulphurs with an O–O bond length of 1.289 Å.103 Acalculated O–O
stretching vibration of 1154 cm−1 is consistentwith a ferric
superoxo complex.103 The frontier orbitals of[FeIII(O2
•−)(S2Me2N3(Pr,Pr))] as shown in Fig. 11 contain two
unpaired electrons of opposite spin; one on the superoxo π*(O–O)
orbital and the other on the Fe(dxy) orbital.
103
The [FeIII(O2•−)(S2
Me2N3(Pr,Pr))] complex can also abstracthydrogen atoms from
1,4-cyclohexadiene (CHD : BDE =76 kcal mol−1) with the deuterium
kinetic isotope effect(KIE = 4.8) and also from THF (BDE = 92 kcal
mol−1) to producean Fe(III)-hydroperoxo complex, [FeIII(OOH)(S2
Me2N3(Pr,Pr))].103
Scheme 11 Formation of a non-heme iron(III)-superoxo
complex,[FeIII(O2
•−)(LPh)(TpMe2)], and the hydrogen atom transfer reaction
with2-hydroxy-2-azaadamantane to produce
[FeIII(OOH)(LPh)(TpMe2)]species. Reprinted with permission from
ref. 100. Copyright 2015,WILEY-VCH Verlag GmbH.
Fig. 10 DFT-optimized structure of [FeIII(O2•−)(TpMe2)(2-ATP)]
with a
quintet state (S = 2). Black and green texts provide interatomic
distancesin Å and Mulliken spin populations, respectively.
Reprinted with per-mission from ref. 102. Copyright 2018, Royal
Society of Chemistry.
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9479
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
8. Cobalt(III)-superoxo complexes
Mononuclear and dinuclear cobalt(III)-superoxide complexeshave
been known for decades.104–106 It was reported earlierthat
cobalt(II) complexes such as cyanocob(II)alamin (vitaminB12r) and
cobalt(II) porphyrins reacted with O2 to produceCo(III)-superoxo
complexes.18,107,108 The end-on coordinationof O2
•− to a Co(III) complex, [CoIII(bzacen)(pyridine)] (bzacen
=N,N′-ethylene-bis(benzoylacetoniminide)), was shown by theX-ray
crystal structure of [CoIII(O2
•−)(bzacen)(pyridine)].17
Single crystal EPR of the 17O enriched superoxide complexwith
vitamin B12r gave the g tensor and the
17O hyperfinetensors of B12rO2, which are coaxial with the
largest g principalaxis along the Oα–Oβ bond, whereas one of the
principal axesof the 59Co hyperfine tensor is oriented
approximately alongthe corrin normal.109,110 The Co–O–O moiety was
bent with abond angle of 111°.109 A total spin density on the O2 of
ρO2 wasdetermined to be 0.7 ± 0.1.109 The X-ray crystal structure
ofB12rO2 confirmed that the dioxygen molecule is attached to
themetal centre in a bent end-on mode at the β-face of the
cobal-amin molecules.111
As the case of a high-spin iron(II) complex supported by
afive-azole donor set ([FeII(LPh)(TpMe2)]) (vide supra; Scheme
11),cobalt(II) complexes, [CoII(LX)(TpMe2)] (X = Ph and
OiPr),reacted with O2 reversibly to produce the end-on
Co(III)-super-oxo complex, [CoIII(η1-O2•−)(LOiPr)(TpMe2)] (Scheme
12), wherean O–O bond length of 1.301(5) Å is typical of the
superoxocomplex.100,112 The O2 binding affinity of the five
azole-sup-ported cobalt centres is controlled by the structural and
elec-tronic properties of the ligand substituent groups located
inthe secondary coordination sphere.112
Dinuclear cofacial cobalt(II) porphyrin μ-superoxo com-plexes
were generated by the reactions of cofacial dicobalt(II)porphyrins
with O2 in the presence of a bulky base
(1-tert-butyl-5-phenylimidazole) and the subsequent one-electron
oxi-dation of the resulting peroxo species by iodine.113 The
super-hyperfine structure due to two equivalent 59Co nuclei
wasobserved at room temperature in the EPR spectra of the
μ-superoxo species.113 The peroxo-bridged dinuclear
cofacialCo(III) complexes act as key intermediates for the
catalytic four-electron reduction of O2 by one-electron reductants
such asferrocene in the presence of an acid such as HClO4 in
benzo-nitrile via the O–O bond cleavage of the dinuclear
peroxospecies to produce the high-valent cobalt(IV)-oxo complex
thatundergoes further reduction by ferrocene to regenerate
dinuc-lear cobalt(III) complexes.113–115
In contrast to dinuclear cobalt complexes that catalyse
thefour-electron reduction of O2 (vide supra), mononuclear
cobaltcomplexes act as efficient catalysts for two-electron
reductionof O2 to H2O2 via cobalt(III)-superoxo complexes.
113,116–123 Inthe case of cobalt(III) corrole complexes, such as
(TPFCor)CoIII
(TPFCor = 5,10,15-tris(
pentafluorophenyl)corrole),(F5PhMes2Cor)Co
III (F5PhMes2Cor = 10-penta-fluorophenyl-5,15-dimesitylcorrole),
(Mes3Cor)Co
III (Mes3Cor = 5,10,15-trismesityl-corrole) and (tpfcBr8)Co
III (tpfcBr8
=2,3,7,8,12,13,17,18-octabromo-5,10,15-tris(pentafluorophenyl)-corrole),
the Co(III)/(IV) couple is responsible for the
catalytictwo-electron reduction of O2.
124 The best catalytic performancewas obtained for
[(tpfcBr8)Co
III], which exhibited the onsetpotential as positive as 0.81 V
vs. RHE.125 Cobalt(II) corrolesare known to react with O2 to
produce the end-on cobalt(III)-superoxo corroles, which are well
characterised by rRaman andEPR spectroscopy, together with DFT
analyses.126
A mononuclear cobalt(II) complex, [CoII(Pytacn)(CH3CN)2]2+,
is also reported to react with O2 to form initially the
mono-nuclear cobalt(III)-superoxo complex, followed by
stabilisationwith another molecule of [CoII(Pytacn)(CH3CN)2]
2+ to affordthe peroxo-bridged dicobalt(III) complex (Scheme
13).127 TheCW X-band EPR spectrum of [CoIII(O2
•−)(Pytacn)(CH3CN)2]2+
recorded in acetone at 100 K exhibited an axial S = 1/2
signal,which is assigned to a CoIII-superoxo species based on
thecomparison of the corresponding EPR parameters (A∥ = 15.9 Gdue
to 59Co (I = 7/2) nucleus, g∥ = 2.0808, g⊥ = 1.99) with those
Fig. 11 Singly occupied molecular orbitals (SOMO) of
[FeIII(O2•−)
(S2Me2N3(Pr,Pr))]. Reprinted with permission from ref. 103.
Copyright
2019, American Chemical Society.
Scheme 12 Formation of cobalt(III)-superoxo complexes,
[CoIII(η1-O2•−)(LX)(TpMe2)] (X = Ph and OiPr), and X-ray crystal
structure of [CoIII(η1-O2
•−)(LOiPr)(TpMe2)]. Reprinted with permission from ref. 100.
Copyright2015, WILEY-VCH Verlag GmbH.
Perspective Dalton Transactions
9480 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
reported for cobalt(III)-superoxo complexes.128 A
dithiolate-ligated cobalt(III)-superoxo complex,
CoIII(O2)(Me3TACN)(S2SiMe2), was also produced by the reaction of
Co
II(Me3TACN)(S2SiMe2) with O2 and characterized structurally by
X-rayabsorption spectroscopy (XAS) and spectroscopically by
elec-tron paramagnetic resonance (EPR) and resonance Raman
(rR)spectroscopies.129
A CoII triazacorrole complex, [CoII(TBP8Cz)(py)]− (TBP8Cz =
octa(4-tert-butylphenyl)corrolazine), produced by the
one-elec-tron reduction of [CoIII(TBP8Cz)(py)2], also reacts with
O2 rever-sibly to generate the Co(III)-superoxo complex,
[CoIII(O2
•−)(TBP8Cz)(py)]
−, in CH2Cl2/py (Scheme 14).130,131 At 240 K > T >
200 K, a clear eight-line signal with Aiso(59Co) = 40 MHz (14
G)
and giso = 2.026 was observed, whereas the EPR spectrum
dis-appeared at T > 240 K (Fig. 12), indicating a shift to the
rightin the equilibrium between [CoIII(O2
•−)(TBP8Cz)(py)]− and
{[CoII(TBP8Cz)(py)]− + O2}.
131
The [CoIII(O2•−)(TBP8Cz)(py)]
− complex can abstract ahydrogen atom from TEMPOH
(2,2,6,6-tetramethylpiperidin-1-ol, BDE(O–H) = 72.1 kcal mol−1),
phenylhydrazine (BDE(N–H)= 75.0 kcal mol−1) and diphenylhydrazine
(BDE(N–H) =71.7 kcal mol−1) to produce a cobalt(III)-hydroperoxo
complexthat rapidly loses hydroperoxide via displacement by
theexcess pyridine (Scheme 15).130 The [CoIII(O2
•−)(BDPP)]complex, which contains a low-spin cobalt(III) ion
bound to asuperoxo ligand, can also abstract a hydrogen atom
from
TEMPOH to form a structurally characterized
cobalt(III)-hydro-peroxo complex, [CoIII(OOH)(BDPP)], and
TEMPO•.132
Besides cobalt(III)-superoxo complexes, a
mononuclearcobalt(II)-superoxo complex, (Et4N)[Co
II(O2•−)(L•2−)] (L3− = (N(o-
PhNC(O)iPr)2)3−), was reported to be produced by the
reaction
of a bimetallic (Et4N)2[CoII2 (L)2] complex with two
equivalentsof O2 (Fig. 13).
133 The cobalt(II)–cyanide complex,(Et4N)2[Co
II(CN)(L)], was also produced by the reaction of a bi-metallic
(Et4N)2[Co2(L)2] complex with two equivalents of CN
−.The X-ray crystal structure of (Et4N)2[Co
II(CN)(L)] is shown inFig. 13.133 Magnetic measurements indicate
[CoII(CN)(L)]2− tobe a high-spin CoII-cyano species (S = 3/2),
whereas IR, EXAFS,and EPR spectroscopies indicate [Co(O2)(L)]
− to be an end-onsuperoxo complex with an S = 1/2 ground
state.133 X-ray spec-troscopy and calculations indicate that
[Co(O2)(L)]
− features ahigh-spin CoII centre. The S = 1/2 spin state
results from theCo electrons coupled to both O2
•− and the aminyl radical onredox noninnocent L•2−.133
Scheme 14 Formation of [CoIII(O2•−)(TBP8Cz)(py)]
− by O2-activation.Reprinted with permission from ref. 130.
Copyright 2019, Royal Societyof Chemistry.
Fig. 12 EPR spectra of [CoIII(O2•−)(TBP8Cz)(py)]
− in fluid pyridine/EtOH(v/v 1 : 1) solution at various
temperatures (160–239 K). Reprinted withpermission from ref. 131.
Copyright 2004, American Chemical Society.
Scheme 15 Mechanism for the reaction between the CoIII(O2•−)
complex and H-atom donors. Reprinted with permission from ref.
130.Copyright 2019, Royal Society of Chemistry.
Scheme 13 Reversible reaction of Co(II) with O2 to form
CoIII-superoxo
and CoIII-peroxo-CoIII species depending on temperature.
Reprintedwith permission from ref. 127. Copyright 2017, Royal
Society ofChemistry.
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9481
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
9. Nickel(II)-superoxo complexes
The first nickel(II)-superoxo complex, which was
structurallycharacterized by X-ray diffraction analysis, was
prepared by thereaction of the nickel(I) precursor
[(NiI(β-diketiminato))2-(μ–η3:η3-C6H5Me)] with dry O2 in toluene
and the structure isshown in Fig. 14, where O2
•− is coordinated to the Ni(II) centrein a side-on fashion.134
The O–O bond length of 1.347(2) Å istypical of the superoxo species
as compared with those ofperoxo ligands, which are longer than 1.40
Å.135 The 16O–16Ostretching vibration at 971 cm−1 (ν(18O–18O) = 919
cm−1) deter-mined by IR spectroscopy is also typical of the
superoxospecies.134 The low O–O bond stretching vibration may
resultfrom the side-on coordination as compared with the value ofan
end-on Ni(II)-superoxo complex (vide infra). The EPR spec-trum of
[NiII(O2
•−)(L)] (L = β-diketiminato) in frozen toluene at
50 K showed a signal at g = 2.138, 2.116, and 2.067, and
theaverage gavg value of 2.107 agrees with the effective
magneticmoment determined for a solid sample in the
temperaturerange of 20–300 K (μeff = 1.8 B.M. that corresponds to
gavg =2.08).135 The [NiII(O2
•−)(L)] complex can oxygenate PPh3 toproduce Ph3PvO. The [Ni
II(O2•−)(L)] complex can also abstract
a hydrogen atom from 2,4,6-tri-tert-butylphenol, demonstrat-ing
the dioxygenase-like reactivity.135
A side-on Ni(II)-superoxo complex was also reported
for[NiII(O2
•−)(PhTtAd)] (PhTtAd =
phenyltris((1-adamantylthio)-methyl)borate), which showed a rhombic
EPR spectrum at g =2.24, 2.19 and 2.01, containing a
five-coordinate NiII centre.136
There was no 17O hyperfine broadening of the EPR
spectrum,indicating that the unpaired electron in O2
•− is coupled anti-ferromagnetically with that in the Ni dx2−y2
orbital and theremaining unpaired electron resides primarily in the
Ni dz2orbital.136
The first end-on Ni(II)-superoxo complex was reported
for[NiII(O2
•−)(14-TMC)](OTf) (14-TMC =
1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane), which was
generated by thereaction of [NiI(14-TMC)](OTf) with excess O2 in
THF at−78 °C.137 The end-on coordination of O2•− to the Ni(II)
centreis indicated by the extended X-ray absorption fine
structure(EXAFS) data, which exhibited an O/N scatterer at 1.98 Å
andfour N/O scatterers at 2.17 Å.137 The DFT calculations
indicatethat the end-on low-spin state (S = 1/2) is much more
favouredover both the end-on high-spin (S = 3/2) and side-on
low-spinstates.137 The [NiII(O2
•−)(14-TMC)]+ complex as well as the[NiII(O2
•−)(L)] complex as shown in Fig. 14 can oxygenate PPh3to Ph3PvO
and it can also abstract a hydrogen atom fromxanthene and
cyclohexadiene.137
When 14-TMC was replaced by 12-TMC (=
1,4,7,10-tetra-methyl-1,4,7,10-tetraazacyclododecane), a
Ni(III)-peroxocomplex was formed with the 12-TMC ligand as shown
inFig. 15.138,139 The X-ray crystal structure of
[NiIII(O2)(12-TMC)](ClO4)·CH3CN showed the mononuclear side-on 1 :
1 nickelcomplex with the O2 moiety in a distorted octahedral
geome-try.138 The O–O bond length (1.386(4) Å) of the
Ni(III)-peroxocomplex, [NiIII(O2
2−)(12-TMC)]+, is longer than that of theNi(II)-superoxo
complex, [NiII(O2
•−)(14-TMC)]+.138 The observedO–O stretching frequency of
[NiIII(O2
2−)(12-TMC)]+
Fig. 13 Formation of a cobalt(II)-superoxo complex,
(Et4N)[CoII(O2
•−)(L•2−)], and a cobalt(II)-cyano species (Et4N)2[Co(CN)(L)] by
the reactionof (Et4N)2[Co2(L)2] with 2 equiv. of O2 and CN
−, respectively. Reprintedwith permission from ref. 133.
Copyright 2016, American ChemicalSociety.
Fig. 14 X-ray crystal structure of nickel(II)-superoxo complex,
[NiII(O2•−)
(L)] (L = β-diketiminato anion). Reprinted with permission from
ref. 134.Copyright 2008, WILEY-VCH Verlag GmbH.
Fig. 15 The macrocyclic ring-size effect of TMC ligands on the
gene-ration of Ni–O2 species. Reprinted with permission from ref.
139.Copyright 2013, Royal Society of Chemistry.
Perspective Dalton Transactions
9482 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
(1002 cm−1)138 is significantly lower than that of
[NiII(O2•−)
(14-TMC)]+ (1131 cm−1).137 The nickel(III)-peroxo
complex([NiIII(O2
2−)(12-TMC)]+) is not reactive in electrophilic reac-tions, but
is capable of conducting nucleophilic reactions.138
When 13-TMC (=
1,4,7,10-tetramethyl-1,4,7,10-tetraaza-cyclotridecane) was employed
as a ligand for generation of theNi–O2 complex, both the
nickel(II)-superoxo complex([NiII(O2
•−)(13-TMC)]+) and the nickel(III)-peroxo complex([NiIII(O2
2−)(13-TMC)]+) were formed in the reaction of
[NiII(13-TMC)(CH3CN)]
2+ and H2O2 in the presence of tetramethyl-ammonium hydroxide
(TMAH) and triethylamine (TEA),respectively.139 The superoxo ligand
in [NiII(O2
•−)(13-TMC)]+ isbound in an end-on fashion to the nickel(II)
centre, whereasthe peroxo ligand in [NiIII(O2
2−)(13-TMC)]+ is bound in a side-on fashion to the nickel(III)
centre.139 The observed O–Ostretching frequency [ν(O–O)] of
[NiII(O2
•−)(13-TMC)]+ at1130 cm−1 is virtually the same as that reported
for [NiII(O2
•−)(14-TMC)]+ (1131 cm−1).139 The ν(O–O) value of [NiIII(O2
2−)(13-TMC)]+ (1008 cm−1) is lower than those of
Ni(II)-superoxocomplexes, such as [NiII(O2
•−)(13-TMC)]+ (1130 cm−1) and[NiII(O2
•−)(14-TMC)]+ (1131 cm−1),137–139 but similar to that ofthe
Ni(III)-peroxo complex, [NiIII(O2
2−)(12-TMC)]+
(1002 cm−1).138 The [NiII(O2•−)(13-TMC)]+ complex exhibited
electrophilic reactivity, whereas the [NiIII(O22−)(13-TMC)]+
complex showed nucleophilic reactivity.139
A side-on nickel(II)-superoxo species was also produced bythe
reaction of nickel(I) dispersed inside the nanopores of theZSM-5
zeolite with O2.
140 The side-on η2-coordination of O2•−
to the Ni(II) centre was indicated by detailed analysis of
theEPR spectra of both 16O2 and
17O2 species. The computersimulations of the spectra and
relativistic DFT calculations ofthe EPR signatures with the g and
A(17O) tensors (gxx = 2.0635,gyy = 2.0884, gzz = 2.1675; |Axx| ≈ 10
G, |Ayy| = 56.7 G, |Azz| ≈ 13G) indicate a mixed metalloradical
with two supporting oxygendonor ligands and even triangular
spin-density redistributionwithin the η2-{NiO2} unit.140 This shows
sharp contrast to thecase of the side-on Ni(II)-superoxo complex
([NiII(O2
•−)(PhTtAd)]), which shows no 17O hyperfine when the
unpairedelectron resides primarily in the Ni dz2 orbital (vide
supra).
136
A pyrazolate-based dinickel(II) dihydride complex [KL(Ni–H)2]
reacted with O2 to produce the μ-1,2-peroxo Ni(II)
complex[KLNi2(O2
2−)], which reacted further with excess O2 to affordthe
μ-1,2-superoxo dinickel(II) complex ([LNi2(O2•−)]).141 TheX-ray
crystal structure of [LNi2(O2
•−)] is shown in Fig. 16, wherethe O–O bond distance is typical
of a superoxo ligand (1.326(2)Å) that is significantly shorter than
that of the peroxo ligand(1.465(2) Å) in the corresponding
μ-1,2-peroxo dinickel(II)complex.141 A Raman spectrum of
[LNi2(O2
•−)] exhibited theO–O stretching frequency at 1007 cm−1, which
was shifted to951 cm−1 when 16O2 was replaced by
18O2 (Δν(18O2–16O2) =−56 cm−1).141 This is in the range that is
typical of superoxocomplexes and has significantly higher energy
than the O–Ostretches of the peroxo complex ([KLNi2(O2
2−)]: 680 cm−1 andΔν(18O2–16O2) = −40 cm−1).141 The one-electron
reductionpotential of the μ-1,2-superoxo dinickel(II)
complex([LNi2(O2
•−)]) was determined by the cyclic voltammogram,
which showed the reversible redox couple at E1/2 = −1.22 V
vs.Fc+/Fc in THF.141
10. Copper(II)-superoxo complexes
Mononuclear CuII-superoxo complexes are the first intermedi-ates
produced at several Cu protein active sites, such as
pepti-dylglycine α-hydroxylating mono-oxygenase,
dopamineβ-monooxygenase, Cu/Zn superoxide dismutase and copperamine
oxidase.10,34,142,143 CuII-Superoxo complexes are also thefirst
intermediates for the catalytic O2 reduction.
144–148 Thefirst copper(II)-superoxo species, which were
characterizedstructurally by X-ray diffraction analysis, were
formed by react-ing a copper(I) species with a sterically demanding
ligand, HB(3-tBu-5-iPrpz)3, to prevent the dimerization.
149,150 The X-raycrystal structure of [HB(3-tBu-5-iPrpz)3]Cu
II(O2•−) is shown in
Fig. 17, where O2•− is bound to the Cu(II) centre in a
side-on
fashion with the O–O bond distance of 1.22 Å. The O–O
bondstretching frequency at 1112 cm−1 observed in the IR
spectrum
Fig. 16 (a) X-ray crystal structure of a pyrazolate-based
μ-1,2-superoxodinickel(II) complex ([(L)Ni2(O2
•−)]). (b) Front view of the structure of[LNi2(O2
•−)] (iPr groups and hydrogen atoms are omitted for
clarity).Reprinted with permission from ref. 141. Copyright 2018,
AmericanChemical Society.
Fig. 17 X-ray structure of [HB(3-tBu-5-iPrpz)3]CuII(O2
•−). Reprinted withpermission from ref. 150. Copyright 2003,
American Chemical Society.
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9483
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
of [HB(3-tBu-5-iPrpz)3]CuII(O2
•−) shifted to 1060 cm−1 when16O2 was replaced by
18O2 (Δν(18O2–16O2) = −52 cm−1).150 Theground state is a singlet
due to the covalent interactionbetween the superoxide π*σ and Cu
dxy orbitals.150
The X-ray crystal structure of the Cu–O2 complex in theenzyme
peptidylglycine-α hydroxylating mono-oxygenase (PHM)was
successfully obtained as shown in Fig. 18, where the CuBsite
reveals a four-coordinate distorted tetrahedral (Td) geometryand
O2
•− binds to the CuB centre with an end-on η1 geometry(O–O
distance = 1.23 Å and the CuB–O–O angle = 110°).
151 Thisgeometry is compatible with the O2 molecule or O2
•− speciesbound to the copper centre, but not with Cu-peroxo
species,which have typical O–O bond lengths of >1.4
Å.152–155
Cu(II)-Superoxo complexes are capable of abstracting ahydrogen
atom to cleave phenol O–H bonds and weak C–Hbonds.156–163 The
hydrogen abstracting reactivity of Cu(II)-superoxo complexes is
enhanced by hydrogen-bonding moi-eties in a series of TMPA-based
Cu(L) (L = TMPA, BA, F5BA andMPPA; see the structures A–D in Fig.
19) complexes, whichinhibit the formation of the corresponding
binuclear trans-μ-1,2-peroxo[{CuII(L)}2(O22−)]2+ complex, in favour
of mono-copper [CuII(O2
•−)(L)]+ species.164 [CuII(O2•−)(TMPA)]+ (A), which
has no H-bonding moiety, exhibited the lowest O–O
stretchingfrequency at 1119 cm−1 with Δν(18O2–16O2) = −61 cm−1
andthe ν(O–O) values increased with enhancing H-bonding abilitywith
an increase in the N–H dipoles in order of B–D.164
[CuII(O2•−)(TMPA)]+ (A) exhibited no reactivity toward
4-methoxyphenol with BDE(O–H) = 87.6 kcal mol−1 in DMSOat −135
°C.164 In contrast, [CuII(O2•−)(BA)]+ (B) exhibited slug-gish
reactivity, which took >6 h for completion. [CuII(O2
•−)(F5BA)]
+ (C) reacted much faster than [CuII(O2•−)(BA)]+ (B).
[CuII(O2•−)(MPPA)]+ (D) exhibited the highest reactivity
with
the second-order rate constant of 2.33 × 10−2 M−1 s−1.
Thehydrogen bonding is proposed to occur to the proximalO-atom in
[CuII(O2
•−)(L)]+ complexes.164
A mononuclear copper(II)-(end-on)superoxide complex sup-ported
by a N-[(2-pyridyl)methyl]-1,5-diazacyclooctane triden-tate ligand
was recently reported to induce a catalytic C–C
bond formation reaction between carbonyl compounds (sub-strate)
and the solvent molecule (acetone), giving β-hydroxy-ketones
(aldol).165
Besides mononuclear copper-superoxo complexes (videsupra), a
dinuclear Cu(II)-superoxo complex was prepared bythe reaction of a
mixed-valence phenolate-bridged Cu(I)Cu(II)complex,
[(UN-O−)CuICuII]2+ (UN-O− = phenol-containing binu-cleating
ligand), with O2 in a reversible manner, under cryo-genic
conditions.166 The dinuclear Cu(II)-peroxo complex,[(UN-O−)CuII2
(O2
2−)]+, was produced by the reversible reactionof the dicopper(I)
precursor complex [(UNO−)CuI2]
+ with O2.166
A standard reduction potential for the
Cu(II)-superoxo/Cu(II)-peroxo pair was determined to be E° vs. SCE
= +130 mV.166 Akinetic study using a stopped-flow technique
revealed anouter-sphere ET process with a total reorganization
energy (λ)of 1.1 eV between the superoxo and peroxo complexes,
whichwas evaluated in light of the Marcus theory of ET.166
A pyrazolate-based μ-1,2-peroxo dicopper(II) complex under-goes
clean one-electron oxidation at a low potential (−0.22 vs.SCE) to
produce the μ-1,2-superoxo dicopper(II) complex,which was
characterized by the O–O bond stretching vibrationfrequency at 1070
cm−1 with −59 cm−1 of Δν(18O2–16O2).167
The μ-1,2-superoxo dicopper(II) complex can abstract
hydrogenatoms from weak X–H bonds such as TEMPO-H to produce
thehydroperoxide complex.167
The formation of a mixed-valence Cu(II)–Cu(I)-superoxocomplex,
[CuII(O2
•−)CuI]2+ (λmax = 685–740 nm), was madepossible upon femtosecond
laser photoexcitation of an end-ontrans-μ-1,2-peroxodicopper(II)
complex [(tmpa)2CuII2 (O2
2−)]2+
(λmax = 525 and 600 nm), followed by fast intramolecular
elec-tron transfer to yield an “O2-caged” dicopper(I) adduct,
CuI2–O2, and a barrierless stepwise back ET to
regenerate[(tmpa)2CuII2 (O2
2−)]2+.168 Femtosecond laser excitation of side-on
μ–η2:η2-peroxodicopper(II) complexes, [(N5)CuII2 (O2
2−)]2+ and[(N3)CuII2 (O2
2−)]2+, also resulted in the generation of [CuII(O2•−)
CuI]2+, but followed by O2 release to produce the
CuI2complexes.168,169
Fig. 18 X-ray crystal structure of the binding site of O2 (red
rod) to theCuB centre (green ball) in an end-on manner. Reprinted
with permissionfrom ref. 151. Copyright 2004, the American
Association for theAdvancement of Science.
Fig. 19 (Left panel) [CuII(O2•−)(L)]+ complexes without (A) and
with
internal hydrogen-bonding substituents (B–D) used in this study.
(Rightpanel) Resonance Raman spectra (λex = 413.1 nm) of complexes
A–D(blue for [CuII(16O2
•−)(L)]+ and red for [CuII(18O2•−)(L)]+) in frozen
2-methyltetrahydrofuran. Reprinted with permission from ref.
164.Copyright 2018, American Chemical Society.
Perspective Dalton Transactions
9484 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
11. Zinc(II)-superoxo complexes
The important role of Zn2+ ions in accelerating both the
oxi-dation and reduction of O2
•− has been indicated by an imid-azolate-bridged CuII–ZnII
heterodinuclear complex containinga dinucleating Hbdpi ligand
(Hbdpi = 4,5-bis(di(2-pyridyl-methyl)aminomethyl)imidazole), which
is well-characterizedas a SOD model.170 A large positive shift
(0.21 V) in the one-electron reduction potential (Ered) of the
copper(II) moiety ofthe CuII–ZnII complex is observed as compared
to the corres-ponding mononuclear CuII complexes without Zn2+
ions.170
Such a positive shift of the Ered value of the CuII–ZnII
complex
results from the electron-withdrawing effect of the
imidazo-late-bound Zn2+ ion, which leads to a decrease in the
electrondensity on the copper ion.170 Thus, an important role of
theZn2+ ion in the imidazolate-bridged CuII–ZnII complex is
toaccelerate an outer-sphere electron transfer from O2
•− toproduce the CuI–ZnII complex, when the free energy change
ofelectron transfer becomes thermodynamically more favourableas
compared to that without Zn2+ ions as shown in Scheme 16,where ET
from the CuI–ZnII complex to O2
•− is also acceleratedby binding of O2
•− to the Zn2+ centre.170 The binding of O2•− to
the Zn2+ centre of [ZnII(MeIm(Me)2)]2+ was confirmed by the
reaction of the [ZnII(MeIm(Me)2)]2+ complex with the O2
•−
anion to produce [ZnII(O2•−)(MeIm(Me)2)]
2+ (Scheme 17),which was detected by the EPR spectrum.171
12. Conclusions
Superoxide anions (O2•−) bind with metal complexes of the
first row d-block elements to produce metal-superoxo com-plexes.
Side-on (η2) or end-on (η1) binding of the superoxideligand to the
metal centres has been observed depending onthe types of metals and
ligands. Binding of metal ions to O2
•−
results in the enhancement of the radical reactivity of
metal-superoxo complexes, which generally undergo
hydrogen-atomabstraction reactions. An end-on Cr(III)-superoxo
complexundergoes not only hydrogen-atom abstraction reactions
butalso oxygen atom transfer reactions. The sulfoxidation of
thio-anisole by a Cr(III)-superoxo complex is much enhanced in
thepresence of HOTf by the PCET pathway from thioanisole to
aCr(III)-superoxo complex. ET from one-electron donors to
theCr(III)-superoxo complex is much enhanced by PCET. Hydrogenatom
transfer from phenol derivatives to the Cr(III)-superoxocomplex is
also much enhanced in the presence of HOTf bythe PCET pathway. The
introduction of hydrogen bonding moi-eties into Cu(II)-superoxo
complexes resulted in the enhance-ment of the hydrogen atom
transfer (HAT) reaction probablyvia PCET processes. Such PCET
processes may also be appliedto other metal-superoxo complexes
although the effects ofacids on the reactivity of metal-superoxo
complexes have yet tobe studied well. Not only PCET pathways but
also metal ion-coupled electron transfer (MCET) pathways of other
metalsuperoxo complexes to produce dinuclear metal-superoxo
com-plexes may be exploited further to provide new insights
intometal–oxygen chemistry.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We are grateful to the collaborators and co-workers whosenames
are presented in the references for their contributionsto the work
described herein. Financial support for the workdescribed herein
was provided by the JSPS KAKENHI (GrantNumbers 16H02268 to S. F.)
from MEXT, Japan and by theNRF of Korea through CRI
(NRF-2012R1A3A2048842 to W. N.),GRL (NRF-2010-00353 to W. N.), and
the Basic ScienceResearch Program (2017R1D1A1B03029982 to Y. M. L.
and2017R1D1A1B03032615 to S. F.).
Notes and references
1 X. Huang and J. T. Groves, Chem. Rev., 2018,
118,2491–2553.
2 S. M. Adam, G. B. Wijeratne, P. J. Rogler, D. E. Diaz,D. A.
Quist, J. J. Liu and K. D. Karlin, Chem. Rev., 2018,118,
10840–11022.
Scheme 16 Proposed catalytic cycle of the imidazolate-bridged
CuII–ZnII heterodinuclear SOD model complex.170
Scheme 17 Reaction of [ZnII(MeIm(Me)2)(MeCN)]2+ with O2
•− toproduce [ZnII(O2
•−)(MeIm(Me)2)]2+ complex.171
Dalton Transactions Perspective
This journal is © The Royal Society of Chemistry 2019 Dalton
Trans., 2019, 48, 9469–9489 | 9485
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
3 E. I. Solomon and S. S. Stahl, Chem. Rev., 2018, 118,
2299–2862.
4 M. L. Pegis, C. F. Wise, D. J. Martin and J. M. Mayer,Chem.
Rev., 2018, 118, 2340–2391.
5 A. J. Jasniewski and L. Que Jr., Chem. Rev., 2018,
118,2554–2592.
6 S. Fukuzumi, Y.-M. Lee and W. Nam, ChemCatChem, 2018,10,
9–28.
7 W. Nam, Y.-M. Lee and S. Fukuzumi, Acc. Chem. Res.,2018, 51,
2014–2022.
8 R. A. Baglia, J. P. T. Zaragoza and D. P. Goldberg, Chem.Rev.,
2017, 117, 13320–13352.
9 W. Zhang, W. Lai and R. Cao, Chem. Rev., 2017, 117,
3717–3797.
10 C. E. Elwell, N. L. Gagnon, B. D. Neisen, D. Dhar,A. D.
Spaeth, G. M. Yee and W. B. Tolman, Chem. Rev.,2017, 117,
2059–2107.
11 S. Hong, Y.-M. Lee, K. Ray and W. Nam, Coord. Chem.Rev.,
2017, 334, 25–42.
12 M. Guo, T. Corona, K. Ray and W. Nam, ACS Cent. Sci.,2019, 5,
13–28.
13 F. Nastri, M. Chino, O. Maglio, A. Bhagi-Damodaran, Y. Luand
A. Lombardi, Chem. Soc. Rev., 2016, 45, 5020–5054.
14 W. Nam, Acc. Chem. Res., 2015, 48, 2415–2423.15 S. Fukuzumi
and K. D. Karlin, Coord. Chem. Rev., 2013,
257, 187–195.16 J. Cho, R. Sarangi and W. Nam, Acc. Chem. Res.,
2012, 45,
1321–1330.17 G. A. Rodley and W. T. Robinson, Nature, 1972, 235,
438–
439.18 A. L. Crumbliss and F. Basolo, J. Am. Chem. Soc., 1970,
92,
55–60.19 L. Pauling, Stanford Med. Bull., 1948, 6, 215–222.20 J.
S. Griffiths, Proc. R. Soc. London, Ser. A, 1956, 235, 23–36.21 B.
Boitrel and S. Le Gac, New J. Chem., 2018, 42, 7516–
7521.22 A. D. Ure and A. R. McDonald, Synlett, 2015, 26,
2060–2066.23 H. Noh and J. Cho, Coord. Chem. Rev., 2019, 382,
126–144.24 M. Sankaralingam, Y.-M. Lee, W. Nam and S. Fukuzumi,
Coord. Chem. Rev., 2018, 365, 41–59.25 S. Sahu and D. P.
Goldberg, J. Am. Chem. Soc., 2016, 138,
11410–11428.26 K. Ray, F. F. Pfaff, B. Wang and W. Nam, J. Am.
Chem. Soc.,
2014, 136, 13942–13958.27 S. P. de Visser, J.-U. Rohde, Y.-M.
Lee, J. Cho and W. Nam,
Coord. Chem. Rev., 2013, 257, 381–393.28 M. R. Tiné, Coord.
Chem. Rev., 2012, 256, 316–327.29 A. Bakac, Adv. Inorg. Chem.,
2004, 55, 1–59.30 S. Hikichi, M. Akita and Y. Moro-oka, Coord.
Chem. Rev.,
2000, 198, 61–87.31 J. A. Kovacs, Acc. Chem. Res., 2015, 48,
2744–2753.32 S. Itoh, T. Abe, Y. Morimoto and H. Sugimoto,
Inorg.
Chim. Acta, 2018, 481, 38–46.33 S. Itoh, Acc. Chem. Res., 2015,
48, 2066–2074.34 D. A. Quist, D. E. Diaz, J. J. Liu and K. D.
Karlin, J. Biol.
Inorg. Chem., 2017, 22, 253–288.
35 S. Fukuzumi, M. Patz, T. Suenobu, Y. Kuwahara andS. Itoh, J.
Am. Chem. Soc., 1999, 121, 1605–1606.
36 M. Patz, Y. Kuwahara, T. Suenobu and S. Fukuzumi,Chem. Lett.,
1997, 567–568.
37 S. Fukuzumi, T. Suenobu, M. Patz, T. Hirasaka, S. Itoh,M.
Fujitsuka and O. Ito, J. Am. Chem. Soc., 1998, 120,8060–8068.
38 S. Fukuzumi, S. Koumitsu, K. Hironaka and T. Tanaka,J. Am.
Chem. Soc., 1987, 109, 305–316.
39 D. T. Sawyer, G. Chlerlcato Jr., C. T. Angells, E. J. Nannl
Jr.and T. Tsuchlya, Anal. Chem., 1982, 54, 1720–1724.
40 S. Kakuda, C. J. Rolle, K. Ohkubo, M. A. Siegler,K. D. Karlin
and S. Fukuzumi, J. Am. Chem. Soc., 2015,137, 3330–3337.
41 S. Fukuzumi, H. Ohtsu, K. Ohkubo, S. Itoh andH. Imahori,
Coord. Chem. Rev., 2002, 226, 71–80.
42 S. Fukuzumi and K. Ohkubo, Chem. – Eur. J., 2000,
6,4532–4535.
43 S. Fukuzumi and K. Ohkubo, J. Am. Chem. Soc., 2002,
124,10270–10271.
44 T. Kawashima, K. Ohkubo and S. Fukuzumi, Phys. Chem.Chem.
Phys., 2011, 13, 3344–3352.
45 Y. Gong, M. Zhou, S. X. Tian and J. Yang, J. Phys. Chem.
A,2007, 111, 6127–6130.
46 G. K. Dewkar, M. D. Nikalje, I. Sayyed Ali, A. S. Paraskar,H.
S. Jagtap and A. Sudalai, Angew. Chem., Int. Ed., 2001,40,
405–408.
47 S. Dey, S. K. Gadakh and A. Sudalai, Org. Biomol. Chem.,2015,
13, 10631–10640.
48 S. Heinrich, M. Plettig and E. Klemm, Catal. Lett., 2011,141,
251–258.
49 R. R. Langeslay, D. M. Kaphan, C. L. Marshall, P. C. Stair,A.
P. Sattelberger and M. Delferro, Chem. Rev., 2019,
119,2128–2191.
50 M. R. Maurya, Coord. Chem. Rev., 2019, 383, 43–81.51 J. She,
X. Lin, Z. Fu, J. Li, S. Tang, M. Lei, X. Zhang,
C. Zhang and D. Yin, Catal. Sci. Technol., 2019, 9, 275–285.
52 H. Kelm and H.-J. Krüger, Angew. Chem., Int. Ed., 2001,
40,2344–2348.
53 J. Zhang, H. Yang, T. Sun, Z. Chen and G. Yin, Inorg.Chem.,
2017, 56, 834–844.
54 (a) S. L. Scott, A. Bakac and J. H. Espenson, Inorg.
Chem.,1991, 30, 4112–4117; (b) F. Schax, S. Suhr, E. Bill,B. Braun,
C. Herwig and C. Limberg, Angew. Chem., Int.Ed., 2015, 54,
1352–1356; (c) M.-L. Wind, S. Hoof,C. Herwig, B. Braun-Cula and C.
Limberg, Chem. – Eur. J.,2019, 25, 5743–5750.
55 A. Bakac, Coord. Chem. Rev., 2006, 250, 2046–2058.56 K. Qin,
C. D. Incarvito, A. L. Rheingold and K. H. Theopold,
Angew. Chem., Int. Ed., 2002, 41, 2333–2335.57 J. Cho, J. Woo
and W. Nam, J. Am. Chem. Soc., 2010, 132,
5958–5959.58 C. J. Cramer, W. B. Tolman, K. H. Theopold and
A. L. Rheingold, Proc. Natl. Acad. Sci. U. S. A., 2003,
100,3635–3640.
Perspective Dalton Transactions
9486 | Dalton Trans., 2019, 48, 9469–9489 This journal is © The
Royal Society of Chemistry 2019
Publ
ishe
d on
01
May
201
9. D
ownl
oade
d by
Ew
ha W
oman
s U
nive
rsity
on
7/4/
2019
3:5
7:40
AM
. View Article Online
https://doi.org/10.1039/c9dt01402k
-
59 A. Bakac, S. L. Scott, J. H. Espenson and K. R. Rodgers,J.
Am. Chem. Soc., 1995, 117, 6483–6488.
60 T. Devi, Y.-M. Lee, J. Jung, M. Sankaralingam, W. Namand S.
Fukuzumi, Angew. Chem., Int. Ed., 2017, 56, 3510–3515.
61 S. Fukuzumi, M. Ishikawa and T. Tanaka, J. Chem. Soc.,Perkin
Trans. 2, 1989, 1037–1045.
62 J. Jung, K. Ohkubo, D. P. Goldberg and S. Fukuzumi,J. Phys.
Chem. A, 201