Synthesis, characterization, and ligand exchange reactivity of a series of first row divalent metal 3-hydroxyflavonolate complexes

Article abstract of DOI:10.1021/ic901405h

A series of divalent metal flavonolate complexes of the general formula [(6-Ph₂TPA)M(3-Hfl)]X (1 -5-X; X = OTf or CIO₄ ^-; 6-Ph₂TPA = N,N-bis((6-phenyl-2-pyridyl)methyl)-A/-((2- pyridyl)methyl)amine; M = Mn(II), Co(II), Ni(II), Cu(II), Zn(II); 3-HfI = 3-hydroxyflavonolate) were prepared and characterized by X-ray crystallography, elemental analysis, FTIR, UV-vis, ¹H NMR or EPR, and cyclic voltammetry. All of the complexes have a bidentate coordinated flavonolate ligand. The difference in M-O distances (△M-o) involving this ligand varies through the series, with the asymmetry of flavonolate coordination increasing in the order Mn(II) ～ Ni(II) < Cu(II) < Zn(II) < Co(II). The hypsochromic shift of the absorption band I (π→π) of the coordinated flavonolate ligand in 1 -5-OTf (relative to that in free anion) increases in the order Ni(II) < Mn(II) < Cu(II) < Zn(II), Co(II). Previously reported 3-HfI complexes of divalent metals fit well with this ordering.¹H NMR studies indicate that the 3-HfI complexes of Co(II), Ni(II), and Zn(II) exhibit a pseudo-octahedral geometry in solution. EPR studies suggest that the Mn(II) complex 1-OTf may form binuclear structures in solution. The mononuclear Cu(II) complex 4-OTf has a distorted square pyramidal geometry. The oxidation potential of the flavonolate ligand depends on the metal ion present and/or the solution structure of the complex, with the Mn(II) complex 1-OTf exhibiting the lowest potential, followed by the pseudo-octahedral Ni(II) and Zn(II) 3-HfI complexes, and the distorted square pyramidal Cu(II) complex 4-0Tf. The Mn(II) complex [(6Ph₂TPA)Mn(3-HfI)]OTf (1 -OTf) is unique in the series in undergoing ligand exchange reactions in the presence of M(CIO₄) ₂-6H₂O (M = CO, Ni, Zn) in CD₃CN to produce [(6-Ph₂TPA)M(CD₃CN)_n](X)₂, [Mn(3-Hfl)₂-0.5H₂O], and MnX₂ (X = OTf or CIO₄ ^-). Under similar conditions, the 3-HfI complexes of Co(II), Ni(II), and Cu(II) undergo flavonolate ligand exchange to produce [(6-Ph₂TPA)M(CD₃CN)_n](X)₂ (M = Co, Ni, Cu; n = 1 or 2) and [Zn(3HfI)₂ ? 2H₂O]. An Fe(II) complex of 3-HfI, [(6-Ph₂TPA)Fe(3-Hfl)]ClO₄ (8), was isolated and characterized by elemental analysis, FTIR, UV-vis, ¹H NMR, cyclic voltammetry, and a magnetic moment measurement. This complex reacts with O₂ to produce the diiron(lll)μ-oxo compound [(6-Ph ₂TPAFe(3Hfl))₂(μ-O)](ClO₄)₂ (6).

Full text of DOI:10.1021/ic901405h

82 Inorg. Chem. 2010, 49, 82–96
DOI: 10.1021/ic901405h
Synthesis, Characterization, and Ligand Exchange Reactivity of a Series
of First Row Divalent Metal 3-Hydroxyflavonolate Complexes
Katarzyna Grubel,† Katarzyna Rudzka,† Atta M. Arif,‡ Katie L. Klotz,§ Jason A. Halfen,§ and Lisa M. Berreau*,†
†Department of Chemistry & Biochemistry, Utah State University, Logan, Utah 84322-0300,
‡Department of Chemistry, University of Utah, Salt Lake City, Utah 84112-0850, and
§Department of Chemistry, University of Wisconsin-Eau Claire, Eau Claire, Wisconsin 54702
Received July 17, 2009
A series of divalent metal flavonolate complexes of the general formula [(6-Ph2TPA)M(3-Hfl)]X (1-5-X; X = OTf- or
ClO4-; 6-Ph2TPA = N,N-bis((6-phenyl-2-pyridyl)methyl)-N-((2-pyridyl)methyl)amine; M = Mn(II), Co(II), Ni(II),
Cu(II), Zn(II); 3-Hfl = 3-hydroxyflavonolate) were prepared and characterized by X-ray crystallography, elemental
analysis, FTIR, UV-vis, 1H NMR or EPR, and cyclic voltammetry. All of the complexes have a bidentate coordinated
flavonolate ligand. The difference in M-O distances (ΔM-O) involving this ligand varies through the series, with the
asymmetry of flavonolate coordination increasing in the order Mn(II) ∼ Ni(II) < Cu(II) < Zn(II) < Co(II). The
hypsochromic shift of the absorption band I (πfπ*) of the coordinated flavonolate ligand in 1-5-OTf (relative to that
in free anion) increases in the order Ni(II) < Mn(II) < Cu(II) < Zn(II), Co(II). Previously reported 3-Hfl complexes of
divalent metals fit well with this ordering. 1H NMR studies indicate that the 3-Hfl complexes of Co(II), Ni(II), and Zn(II)
exhibit a pseudo-octahedral geometry in solution. EPR studies suggest that the Mn(II) complex 1-OTf may form
binuclear structures in solution. The mononuclear Cu(II) complex 4-OTf has a distorted square pyramidal geometry.
The oxidation potential of the flavonolate ligand depends on the metal ion present and/or the solution structure of the
complex, with the Mn(II) complex 1-OTf exhibiting the lowest potential, followed by the pseudo-octahedral Ni(II) and
Zn(II) 3-Hfl complexes, and the distorted square pyramidal Cu(II) complex 4-OTf. The Mn(II) complex [(6-
Ph2TPA)Mn(3-Hfl)]OTf (1-OTf) is unique in the series in undergoing ligand exchange reactions in the presence of
M(ClO4)2 6H2O (M = Co, Ni, Zn) in CD3CN to produce [(6-Ph2TPA)M(CD3CN)n](X)2, [Mn(3-Hfl)2 0.5H2O], and
3
3
MnX2 (X = OTf- or ClO4-). Under similar conditions, the 3-Hfl complexes of Co(II), Ni(II), and Cu(II) undergo
flavonolate ligand exchange to produce [(6-Ph2TPA)M(CD3CN)n](X)2 (M = Co, Ni, Cu; n = 1 or 2) and [Zn(3-
Hfl)2 2H2O]. An Fe(II) complex of 3-Hfl, [(6-Ph2TPA)Fe(3-Hfl)]ClO4 (8), was isolated and characterized by elemental
3
analysis, FTIR, UV-vis, 1H NMR, cyclic voltammetry, and a magnetic moment measurement. This complex reacts
with O2 to produce the diiron(III) μ-oxo compound [(6-Ph2TPAFe(3Hfl))2(μ-O)](ClO4)2 (6).
Introduction
reaction. Fungal quercetinases are known to contain a
mononuclear Cu(II) center and have been extensively in-
vestigated.4-9 Studies of a bacterial quercetinase from Bacil-
lus subtilis (YxaG) showed that when this enzyme is produced
in Escherichia coli it will bind a variety of divalent metals,
with the highest level of reactivity being found for Mn(II).10,11
Flavonoids are polyphenolic compounds that are pro-
duced in plants.1 One of these compounds, quercetin
(Scheme 1), is found in many fruits and vegetables, and is
of considerable current interest for its antioxidant and anti-
microbial properties.2,3 In the soil environment, fungal and
bacterial quercetin dioxygenases catalyze oxidative car-
bon-carbon bond cleavage and CO release from a metal-
coordinated quercetin in a 2,4-dioxygenolytic ring cleavage
(6) Hund, H. K.; Breuer, J.; Lingens, F.; Huttermann, J.; Kappl, R.;
Fetzner, S. Eur. J. Biochem. 1999, 263, 871–878.
(7) Tranchimand, S.; Ertel, G.; Gaydou, V.; Gaudin, C.; Tron, T.;
Iacazio, G. Biochimie 2008, 90, 781–789.
(8) Kooter, I. M.; Steiner, R. A.; Dijkstra, B. W.; van Noort, P. I.;
Egmond, M. R.; Huber, M. Eur. J. Biochem. 2002, 269, 2971–2979.
(9) Fusetti, F.; Schroter, K. H.; Steiner, R. A.; van Noort, P. I.; Pijning,
T.; Rozeboom, H. J.; Kalk, K. H.; Egmond, M. R.; Dijkstra, B. W. Structure
2002, 10, 259–268.
(10) Gopal, B.; Madan, L. L.; Betz, S. F.; Kossiakoff, A. A. Biochemistry
2005, 44, 193–201.
*To whom correspondence should be addressed. E-mail: lisa.berreau@
usu.edu. Phone: (435) 797-1625. Fax: (435) 797-3390.
(1) Iwashina, T. J. Plant Res. 2000, 113, 287–299.
(2) Pietta, P.-G. J. Nat. Prod. 2000, 63, 1035–1042.
(3) Cushine, T. P. T.; Lamb, A. J. Int. J. Antimicrob. Agents 2005, 26, 343–
356.
(4) Oka, T.; Simpson, F. J. Biochem. Biophys. Res. Commun. 1971, 43, 1–5.
(5) Oka, T.; Simpson, F. J.; Krishnamurty, H. G. Can. J. Microbiol. 1972,
18, 493–508.
(11) Schaab, M. R.; Barney, B. M.; Francisco, W. A. Biochemistry 2006,
45, 1009–1016.
r
pubs.acs.org/IC
Published on Web 12/02/2009
2009 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 1, 2010 83
Scheme 1
The quercetinase QueD from Streptomyces sp. FLA is a
monocupin dioxygenase.16 When overexpressed in E. coli,
this enzyme exhibited the highest level of activity when Ni(II)
and Co(II) salts were added to the LB medium.12,17 An
increase in activity was not observed when Mn(II), Fe(II),
Cu(II), or Zn(II) was added. EPR studies of the cobalt-
containing enzyme indicate a high-spin (S = 3/2) Co(II)
center in a trigonal bipyramidal or tetrahedral geometry in
the resting state. EPR experiments performed in the presence
of quercetin and O2 revealed no evidence of a change in
valency of the cobalt ion during substrate turnover.12
Recent investigationsof a quercetinase from Streptomycessp.
FLA expressed in E. coli revealed that this enzyme is most
active in the presence of Ni(II), with the next highest level of
activity being found with Co(II).12
Different mechanistic pathways have been proposed for
the spin-forbidden, O2-dependent quecetin oxidation reac-
tion depending on whether a redox active metal center is
present inthe active site of the enzyme. For Cu(II)-containing
quercentinase enzymes, it is unlikely that O2 will coordinate
to the type II oxidized metal center, and the reaction is
suggested to involvevalencetautomerism between theCu(II)-
quercetin adduct and a Cu(I)-flavonoxy radical species, the
latter of which can act as a one-electron reductant toward O2.
The pyramidalization of the C(2) center identified in the
X-ray structure of the ES adduct of the A. japonicus enzyme
suggests possible stabilization of the radical at this atom.13
For redox-active metal ions such as Mn(II), the metal center
may serve as an electron conduit in an adduct wherein both
the flavonolate and the O2 coordinate to the metal center
prior to oxygen activation. Specifically, electron transfer
from the metal center to coordinated O2, with subsequent
electron transfer from the coordinated quercetin monoanion
to the oxidized metal center, would generate a quercetin
The quercetinase enzymes from Aspergillus japonicus
and B. subtilis have been characterized by X-ray crystallo-
graphy.9,10 Both are members of the cupin superfamily of
proteins (bicupins), with two well-separated active site metal
centers, each having a ligand donor set composed of three
histidine donors and a glutamate ligand. In the structure of
the copper-containing enzyme from A. japonicus, the copper
centers exhibit two different geometries. One is a distorted
tetrahedron composed of three histidine donors and a water
molecule, and in the other, a glutamate ligand (Glu73) is also
coordinated in an axial position to give an overall distorted
trigonal bipyramidal copper center. These geometries are
present in a ∼70:30 ratio. Electron paramagnetic resonance
(EPR) studies of the protein are consistent with this mixture
of geometries also being present in solution. In the tetrahe-
dral geometry, Glu73 acts as a hydrogen bond acceptor for
the metal-bound water molecule. Coordination of quercetin
to the copper center occurs in a monodentate fashion via the
deprotonated C(3)-OH moiety, with Glu73 possibly acting
as the active site base for substrate deprotonation.13 The
overall geometry of the copper center in the enzyme/substrate
adduct is distorted trigonal bipyramidal. A key feature of the
coordinated quercetin is pyramidalization of the C(2) atom
indicating increased sp3 character that may stabilize radical
formation in a reaction involving O2.
The B. subtilis enzyme was initially reported to be an
Fe(II)-containing quercetin 2,3-dioxygenase,14,15 and was
crystallized with Fe(II) present in both active sites of this
bicupin enzyme.10 Both metal centers exhibit a coordination
number of five, with three histidine donors, a glutamate, and
a water molecule. The overall geometry of the Fe(II) center is
distorted trigonal bipyramidal in the N-terminal active site
and distorted square pyramidal in the C-terminal domain of
the protein. Metal ion replacement studies of the B. subtilis
enzyme (via reconstitution of the apoenzyme) indicated
increased levels of activity for Mn(II)- and Co(II)-containing
enzyme (35- and 24-fold, respectively) relative to that found
for Fe(II). On the basis of this data, it has been suggested that
Mn(II) may be the preferred metal cofactor for the B. subtilis
enzyme. EPR studies of the Mn(II)-containing enzyme from
B. subtilis suggest octahedral coordination of the metal
center.11
-
radical-M(II)-O2 species, from which C-C bond cleavage
and CO release could proceed. This proposed mechanism is
similar to the reaction pathway suggested for Fe(II)- and
Mn(II)-containing extradiol catechol dioxygenases.18
Overall, from studies of the A. japonicus, B. subtilis, and
Streptomyces sp. FLA quercetinases, possible roles for the
divalent metal center in substrate oxidation have been sug-
gested to include the following: (1) reduction of the pKa of
quercetin to enable substrate deprotonation, (2) stabilization
of the flavonolate intermediate, (3) influence on the coordi-
nation mode (mono- versus bidentate) and redox potential of
the bound flavonolate, (4) facilitation of the formation of
radical character on the coordinated flavonolate via valence
tautomerism, and (5) acting as a redox conduit for structures
wherein both flavonolate and O2 are coordinated to the metal
center.
Model studies of copper-containing forms of quercetinases
have demonstrated that both Cu(I) and Cu(II) flavonolate
complexes undergo reaction with O2 to produce CO and the
depside.19 Kinetic studies of the reaction involving the Cu(II)
complex [Cu(3-Hfl)2] (3-Hfl = 3-hydroxyflavonolate) with
O2 in DMF indicate a rate law that is second-order overall
(-d[Cu(3-Hfl)2]/dt = kobs[Cu(3-Hfl)2][O2]). In the presence
of pyridine, the observed rate increases by ∼2.5-fold, which
(16) Merkens, H.; Sielker, S.; Rose, K.; Fetzner, S. Arch. Microbiol. 2007,
187, 475–487.
(12) Merkens, H.; Kappl, R.; Jakob, R. P.; Schmid, F. X.; Fetzner, S.
Biochemistry 2008, 47, 12185–12196.
(13) Steiner, R. A.; Kalk, K. H.; Dijkstra, B. W. Proc. Natl. Acad. Sci.
2002, 99, 16625–16630.
(14) Bowater, L.; Fairhurst, S. A.; Just, V. J.; Bornemann, S. FEBS Lett.
2004, 557, 45–48.
(17) Based on comparison of kcat and KM values with those of Fe(II)-
containing enzyme.
(18) Emerson, J. P.; Kovaleva, E. G.; Farquhar, E. R.; Lipscomb, J. D.;
Que, L. Proc. Natl. Acad. Sci. 2008, 105, 7347–7352 and references cited
therein.
(15) Barney, B. M.; Schaab, M. R.; LoBrutto, R.; Francisco, W. A.
Protein Expression Purif. 2004, 35, 131–141.
(19) Kaizer, J.; Balogh-Hergovich, E.; Czaun, M.; Csay, T.; Speier, G.
Coord. Chem. Rev. 2006, 250, 2222–2233.

84 Inorganic Chemistry, Vol. 49, No. 1, 2010
Grubel et al.
has been attributed to a change in the coordination mode
of the flavonolate ligand from bidentate to monodentate.
Model studies involving other divalent metal ions are con-
siderably fewer in number. Catalytic oxygenation of 3-hydro-
xyflavone derivatives by bis(salicylidene)ethylenediaminato-
cobalt(II) ([Co(salen]) at room temperature has been
reported.20 This catalysis occurs in the presence of DMSO
and DMF, but not in methanol, THF, or CH2Cl2. The
involvement of a Co-O2 complex in the reaction has been
proposed. A Co(III) complex, [Co(III)(salen)(40MeOflaH)],
was found to be susceptible to oxygenation in pyridine and
DMF, but was found to be stable to oxygen in non-coordi-
nating solvents. This O2 reactivity correlates with dissocia-
tion of the flavonolate ligand from the Co(III) center.21,22
Two recent studies outlined the structural and O2 reactivity
properties of Fe(III) and Mn(II) flavonolate complexes.23,24
Fe(III)(40MeOflaH)3 and Mn(II)(3-Hfl)2(py)2 were found to
undergo reaction with O2 at 100 ꢀC in DMF with second-
tion, and spectroscopic and redox properties. The choice of
the tetradentate 6-Ph2TPA ligand as the supporting scaffold
was based on its relevance to the coordination environment in
enzymes of the cupin superfamily.25 We have previously used
this ligand to study chemistry of relevance to Ni(II)-acireduc-
tone dioxygenase,26-29 another dioxygenase of the cupin
superfamily that produces CO upon substrate oxygenation.
The results presented herein indicate that the nature of the
divalent metal ion influences the coordination mode and redox
potential of a 3-Hfl ligand in complexes of the general formula
[(6-Ph2TPA)M(3-Hfl)]X (M=Mn(II), Co(II), Ni(II), Cu(II),
Zn(II); X = OTf- or ClO4-). In the course of characterizing
the Mn(II) complex, we found evidence for ligand exchange
reactions, which were subsequently further explored. The
Fe(II) flavonolate complex [(6-Ph2TPA)Fe(3-Hfl)]ClO4 was
prepared and characterized, albeit an X-ray structure was not
obtained. This is because the complex is very O2 sensitive and
quickly undergoes reaction to produce a diiron(III) μ-oxo
compound, wherein each iron center has a coordinated flavo-
nolate ligand. The structure of this diiron(III) complex was
determined by X-ray crystallography.
order rate constants of 0.50 M-1 s-1 and 0.08 M-1 s-1
,
respectively. On the basis of comparison of rate constants at
100 ꢀC, these complexes are more reactive than the copper
analogues Cu(3-Hfl)2 (k = 0.0087 M-1 s-1) and Cu(3-
Hfl)2(py)2 (0.04 M-1 s-1), suggesting a metal dependent
reactivity order of Fe(III) > Mn(II)>Cu(II).23 The Fe(III)
complex Fe(III)(salen)(3-Hfl) contains a bidentate flavono-
late ligand and undergoes reaction with O2 at elevated
temperatures (100-120 ꢀC).24 In the presence of excess
carboxylate anion, the rate of reaction with O2 increases,
with bulky carboxylates (e.g., triphenyl acetate) producing
the greatest rate enhancement (∼2 orders of magnitude at
100 ꢀC). This is attributed to the formation of a more reactive
monondentate flavonolatoiron(III) complex. Evidence for
direct electron transfer from the flavonolate ligand to O2 to
Experimental Section
General and Physical Methods. All chemicals were purchased
from commercial sources and used as received unless otherwise
noted. Synthetic reactions were performed in a MBraun Unilab
glovebox under a N2 atmosphere. Solvents for glovebox use
were dried according to published methods and distilled under
N2 prior to use.30 The 6-Ph2TPA ligand was prepared as
previously described.25
1H NMR spectra of 2-5 were obtained in CD3CN solution
on a Bruker ARX-400 spectrometer. Data was collected for
paramagnetic complexes as previously described.26 Chemical
shifts (in ppm) are referenced to the residual solvent peak(s) in
CHD2CN (1H, 1.94 (quintet) ppm). FTIR spectra were collected
using KBr pellets on a Shimadzu FTIR-8400 spectrometer.
UV-vis spectra were recorded at ambient temperature using a
Hewlett-Packard 8453 diode array spectrophotometer. Cyclic
voltammetry data was collected using a BAS-Epsilon system.25
Conditions for the CV experiments are listed in a footnote of
Table 3 and in the text. EPR spectra were collected on a Bruker
EMX-Plus spectrometer fitted with a liquid helium cooled
probe. ESI/APCI and MALDI mass spectral data for complexes
was collected at the Mass Spectrometry Facility, University of
California, Riverside. Room temperature magnetic susceptibil-
ities were determined by the Evans method.31 Elemental ana-
lyses were performed by Atlantic Microlab, Inc., Norcross, GA.
Caution! Perchlorate salts of metal complexes with organic
ligands are potentially explosive. Only small amounts of material
should be prepared, and these should be handled with great care.32
General Procedure for the Synthesis of [(6-Ph2TPA)M-
(3-Hfl)]OTf Complexes. (a)M = Mn (1-OTf), Cu (4-OTf), or
Zn (5-OTf). In a N2-filled glovebox, a methanol solution (2 mL)
-
form O2 was found in the reaction via the use of the
superoxide scavenger nitroblue tetrazolium (NBT).
From the studies described above and investigations of the
O2 reactivity of non-coordinated flavonolate anions, it is
clear that enhancing the electron density within the flavono-
late anion through deprotonation and limiting its coordina-
tion mode to monodentate are key factors in enhancing the
rate of oxygenation. What is currently unclear is how
differences in the divalent metal ion present in quercetinase
enzymes produce differing rates of oxygenation, as has been
found for the B. subtilis and Streptomyces sp. FLA enzymes.
Such differences may result from modulation of the coordi-
nation mode and/or the redox potential of the flavonolate, or
may relate to the ability of the metal center to serve as a redox
conduit. As an approach toward systematically examining
the influence of the divalent metal ion, we report the pre-
paration and characterization of the first extensive series of
structurally related divalent metal flavonolate complexes.
Our initial goal in this research was to examine how differ-
ences in the divalent metal center (Mn(II), Fe(II), Co(II),
Ni(II), Cu(II), and Zn(II)) influence flavonolate coordina-
(25) Makowska-Grzyska, M. M.; Szajna, E.; Shipley, C.; Arif, A. M.;
Mitchell, M. H.; Halfen, J. A.; Berreau, L. M. Inorg. Chem. 2003, 42, 7472–
7488.
(26) Szajna, E.; Dobrowolski, P.; Fuller, A. L.; Arif, A. M.; Berreau, L.
M. Inorg. Chem. 2004, 43, 3988–3997.
(27) Szajna, E.; Arif, A. M.; Berreau, L. M. J. Am. Chem. Soc. 2005, 127,
17186–17187.
(28) Szajna-Fuller, E.; Rudzka, K.; Arif, A. M.; Berreau, L. M. Inorg.
Chem. 2007, 46, 5499–5507.
(29) Szajna-Fuller, E.; Chambers, B. M.; Arif, A. M.; Berreau, L. M.
Inorg. Chem. 2007, 46, 5486–5498.
(30) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory
Chemicals, 4th ed.; Butterworth-Heinemann: Boston, MA, 1996.
(31) Evans, D. F. J. Chem. Soc. 1959, 2003–2005.
(32) Wolsey, W. C. J. Chem. Educ. 1973, 50, A335–A337.
(20) Nishinaga, A.; Tojo, T.; Matsuura, T. J. Chem. Soc., Chem. Com-
mun. 1974, 896–897.
(21) Nishinaga, A.; Numada, N.; Maruyama, K. Tetrahedron Lett. 1989,
30, 2257–2258.
(22) Nishinaga, A.; Kuwashige, T.; Tsutsui, T.; Mashino, T.; Maruyama,
K. J. Chem. Soc., Dalton Trans. 1994, 805–810.
ꢀ
ꢀ
(23) Kaizer, J.; Barath, G.; Pap, J.; Speier, G.; Giorgi, M.; Reglier, M.
Chem. Commun. 2007, 5235–5237.
ꢀ
ꢀ
ꢀ
ꢀ
(24) Barath, G.; Kaizer, J.; Speier, G.; Parkanyi, L.; Kuzmann, E.; Vertes,
A. Chem. Commun. 2009, 3630–3632.

Article
Inorganic Chemistry, Vol. 49, No. 1, 2010 85
of M(OTf)2 (1.37 ꢀ 10-4 mol) was added to solid 6-Ph2TPA
(1.37 ꢀ 10-4 mol), and the mixture was stirred until all of the
chelate ligand had dissolved. The resulting solution was added
to a methanol solution (2 mL) containing 3-Hfl (1.37 ꢀ 10-4
CH2Cl2, and the solution was filtered through a glass wool/
Celite plug. The filtrate was then brought to dryness under
reduced pressure. Crystals suitable for single crystal X-ray
crystallography were obtained using the following approaches
at ambient temperature: 1-ClO4 (green crystals), diethyl ether
diffusion into a dichloromethane solution; 2-ClO4 (dark red
crystals) and 3-ClO4 (green crystals), diethyl ether diffusion into
a acetonitrile solution; and 4-ClO4 (green crystals), diethyl ether
diffusion into dichloromethane:isopropanol/methanol (1:0.1:1)
solution. For 5-ClO4, yellow crystals were obtained from di-
chloromethane/diethyl ether solution at 4 ꢀC.
mol) and Me4NOH 5H2O (1.37 ꢀ 10-4 mol). The mixture was
3
then allowed to stir overnight at ambient temperature. The
reaction mixture was taken out of the glovebox, and the solvent
was removed under reduced pressure. The residual solid was
suspended on the top of a Celite plug and washed several times
with distilled water. The wet solid was then dissolved in CH2Cl2,
and the filtrate was collected and brought to dryness under
reduced pressure. The residue was dissolved in CH2Cl2, and the
analytically pure microcrystalline flavonolate complex was pre-
cipitated via the addition of excess Et2O and cooling of the
mixture at -30 ꢀC for 12 h.
Note Regarding Experimental Data. Extensive characteriza-
tion data (FTIR, UV-vis, mass spectrometry, magnetic mo-
ment, cyclic voltammetry) for the triflate compounds 1-5-OTf
is provided in Table 3. Selected data for the perchlorate analo-
gues 1-5-ClO4, which were primarily prepared for X-ray crys-
tallographic studies, is given below.
The 1H NMR features of 2-ClO4, 3-ClO4, and 5-ClO4 match
those found for the triflate analogues. 1-ClO4: Anal. Calcd for
C45H35ClMnN4O7 1/4CH2Cl2: C, 63.54; H, 4.18; N, 6.55.
3
Found: C, 63.47; H, 4.11; N, 6.45. 2-ClO4: Anal. Calcd for
C45H35ClCoN4O7 1/7CH2Cl2: C, 63.77; H, 4.18; N, 6.59.
3
Found: C, 64.03; H, 4.19; N, 6.54. 3-ClO4: Anal. Calcd for
C45H35ClN4NiO7: C, 64.50; H, 4.21; N, 6.69. Found: C, 64.18;
H, 3.92; N, 6.55. 4-ClO4: Anal. Calcd for C45H35ClCuN4O7
1/5CH2Cl2: C, 63.14; H, 4.15; N, 6.52. Found: C, 63.45; H, 4.30;
3
N, 6.31. 5-ClO4: Anal. Calcd for C45H35ClN4O7Zn 1/5CH2Cl2:
3
[(6-Ph2TPA)Mn(3-Hfl)]OTf (1-OTf). Yield: 73% (green crys-
tals).
C, 63.01; H, 4.14; N, 6.50. Found: C, 63.15; H, 4.31; N, 6.33.
Ligand Exchange Reactions. (a). To a solution of [(6-Ph2-
TPA)Mn(3-Hfl)]OTf (9.0 ꢀ 10-6 mol) in CD3CN (0.8 mL) solid
[(6-Ph2TPA)Cu(3-Hfl)]OTf (4-OTf). Yield: 74% (dark green
crystals).
M(ClO4)2 6H2O (M = Co, Ni, or Zn; 9.0 ꢀ 10-6 mol) was
3
[(6-Ph2TPA)Zn(3-Hfl)]OTf (5-OTf). Yield: 98% (yellow
crystals). 1H NMR (CD3CN, 400 MHz): δ 8.53 (d, J =
5.2 Hz, 1 H), 8.18 (dt, J = 7.3 Hz, J = 1.5 Hz, 2 H), 7.92 (td,
J = 7.8 Hz, J = 1.7 Hz, 1 H), 7.78 (t, J = 7.7 Hz,
2 H), 7.65 (m, 3 H), 7.39 (m, 11 H), 7.13 (dt, J = 6.8 Hz, J =
1.5 Hz, 4 H), 7.03 (tt, J = 7.4 Hz, J = 1.2 Hz, 2 H), 6.94 (tt, J =
8.0 Hz, J = 1.5 Hz, 4 H), 4.85 (d, J = 14.7 Hz, 2 H), 4.51 (d,
J = 14.8 Hz, 2 H), 4.39 (s, 2 H); 13C{1H} NMR (CD3CN,
400 MHz): δ 187.7, 168.1, 164.0, 163.8, 162.9, 156.5, 154.6,
148.8, 147.6, 146.3, 141.1, 140.5, 137.4, 136.8, 136.1, 135.7,
135.6, 135.1, 134.8, 132.2, 131.9, 131.3, 130.9, 130.2, 130.1,
125.8, 125.2, 64.9, 61.5 (29 signals expected for equivalent
phenyl-appended pyridyl donors; 29 observed).
added. Each reaction mixture was then capped, shaken vigor-
ously, and a 1H NMR spectrum was recorded within 15 min. For
each metal perchlorate salt, the NMR spectrum is consistent
with the formation of [(6-Ph2TPA)M(CD3CN)n](ClO4)2 (M =
Co (n = 1), Ni (n = 2), or Zn (n = 1)). The product [Mn(3-
Hfl)2 0.5H2O] is formed in each reaction, and was isolated from
the Ni(II)-containing reaction mixture and characterized by
3
elemental analysis, FTIR, and UV-vis.
(b). In a NMR tube, a CD3CN solution (0.8 mL) of [(6-
Ph2TPA)M(3-Hfl)]OTf (M = Co, Ni, and Cu; 2-4-OTf; 7.7 ꢀ
10-6 mol) was treated with solid 0.5 equiv of Zn(ClO4)2 6H2O
3
(3.9 ꢀ 10-6 mol). Each reaction mixture was then capped,
shaken vigorously, and a 1H NMR spectrum was recorded
within 15 min. For the reactions involving the Co(II) and Ni(II)
derivatives, the 1H NMR spectroscopic features are consistent
with the formation of [(6-Ph2TPA)M(CD3CN)n](X)2 (M = Ni,
n = 2; M = Co, n = 1; X = OTf- or ClO4-). A poorly soluble,
yellow precipitate is also formed in each reaction mixture.
Spectroscopic analysis (1H NMR (CD3OD) and FTIR (KBr))
of this solid suggested the formation of [Zn(3-Hfl)2]. This
compound was independently synthesized via treatment of Zn-
(b)M = Co (2-OTf) or Ni (3-OTf). Under a nitrogen atmo-
sphere, a methanol (∼2 mL) solution of MCl2 5H2O (1.37 ꢀ
3
10-4 mol) was added to solid 6-Ph2TPA (1.37 ꢀ 10-4 mol), and
the mixture was stirred until all of the chelate ligand had
dissolved. Two equivalents of silver triflate (AgOTf; 2.74 ꢀ
10-4 mol) was then added to the mixture. After stirring for
30 min, the solution was filtered through a Celite/glass wool
plug. The filtrate was added to a methanol solution (∼2 mL)
containing 3-Hfl (1.37 ꢀ 10-4 mol) and Me4NOH 5H2O
3
(ClO4)2 6H2O with 2 equiv each of 3-Hfl and Me4NOH 5H2O
in methanol, which yielded a yellow precipitate. Analysis of this
3
3
(1.37 ꢀ 10-4 mol). The resulting solution was stirred overnight
at ambient temperature. At this time, the reaction was taken out
of the glovebox, and the solvent was removed under reduced
pressure. Using a workup procedure identical to that described
above for 1-OTf, analytically pure microcrystalline products
were obtained.
1
material by H NMR (CD3OD), UV-vis, and elemental ana-
lysis indicated the formulation [Zn(3-Hfl)2 2H2O]. The UV-vis
3
and 1H NMR spectroscopic features of this material match that
of the yellow precipitate generated in the ligand exchange
reaction.
[(6-Ph2TPA)Co(3-Hfl)]OTf (2-OTf). Yield: 54% (dark red
crystals).
[(6-Ph2TPA)Ni(3-Hfl)]OTf 0.25CH2Cl2 (3-OTf). Yield: 82%
Synthesis of [(6-Ph2TPA)Fe(3-Hfl)]ClO4 (7). Under a nitro-
gen atmosphere, a methanol solution (∼2 mL) of Fe-
(ClO4)2 6H2O (1.37 ꢀ 10-4 mol) was added to solid 6-
3
3
Ph2TPA (1.37 ꢀ 10-4 mol), and the resulting mixture was stirred
for 2 h at ambient temperature. The solvent was then removed
under reduced pressure, and the solid was dissolved in a small
amount of methanol (∼1 mL). Addition of Et2O produced a
yellow precipitate, which was dried under vacuum. A 1H NMR
spectrum of the complex in CD3CN was obtained. This spec-
trum matched that of [(6-Ph2TPA)Fe(CH3CN)](ClO4)2, which
has been independently generated and characterized (see Sup-
porting Information). A methanol solution (∼2 mL) of [(6-
Ph2TPA)Fe(CH3CN)](ClO4)2 (1.37 ꢀ 10-4 mol) was treated
(green crystals). The presence of dichloromethane in the elemen-
tal analysis sample was confirmed using 1H NMR spectroscopy.
General Procedure for the Synthesis of [(6-Ph2TPA)M(3-
Hfl)]ClO4 Complexes (1-5-ClO4). In a glovebox, an acetontrile
solution (∼2 mL) of M(ClO4)2 6H2O (M = Mn, Co, Ni, Cu,
3
Zn; 1.37 ꢀ 10-4 mol) was added to solid 6-Ph2TPA (1.37 ꢀ
10-4 mol), and the resulting mixture was stirred until all of the
chelate ligand had dissolved. An acetonitrile slurry (∼2 mL) of
tetramethylammonium hydroxide pentahydrate (Me4NOH
3
5H2O; 1.37 ꢀ 10-4 mol) and 3-hydroxyflavone (3-Hfl; 1.37 ꢀ
10-4 mol) was then added, and the resulting mixture was stirred
overnight at ambient temperature. After removal of the solvent
under reduced pressure, the remaining solid was dissolved in
with 3-Hfl (1.37 ꢀ 10-4 mol) and Me4NOH 5H2O (1.37 ꢀ 10-4
3
mol) dissolved in methanol (∼2 mL). The resulting mixture was
stirred for 15 min, and the solvent was then removed under

86 Inorganic Chemistry, Vol. 49, No. 1, 2010
Grubel et al.
vacuum. The remaining green-yellow solid was dissolved
CH2Cl2 (∼2 mL) and passed through a Celite/glass wool plug.
The filtrate was brought to dryness under reduced pressure, and
the resulting solid (yield: 96%) was analyzed by 1H NMR,
FTIR, UV-vis, a magnetic moment measurement, and elemen-
tal analysis. FTIR (KBr, cm-1) 1558 (νCdO); UV-vis (CH3CN)
nm (ε, M-1 cm-1) 415 (14700); μeff = 4.7 μB. Anal. Calcd for
C45H35ClFeN4O7: C, 64.72; H, 4.22; N, 6.71. Found: C, 65.23;
H, 4.28; N, 6.71.
labeled with (A)) have very minor differences in bond lengths/
angles involving the Zn(II) center, and both exhibit a distorted
square pyramidal geometry (τ = 0.35).36 One CH2Cl2 solvate
exhibits disorder over two positions (50:50) for each heavy
atom. Complex 6 crystallizes in the space group P1 with
2.5CH3CN solvate molecules per asymmetric unit. Two oxygen
atoms and the chlorine atom of a perchlorate anion are dis-
ordered over two positions (90:10).
Reactivity of [(6-Ph2TPA)Fe(3-Hfl)]ClO4 in Air; Isolation of
[(6-Ph2TPAFe(3-Hfl))2(μ-O)](ClO4)2 (6). A methanol solution
(∼2 mL) of [(6-Ph2TPA)Fe(3-Hfl)]ClO4 was prepared by mixing
equimolar amounts (1.37 ꢀ 10-4 mol) of 6-Ph2TPA, Fe-
(ClO4)2 6H2O, Me4NOH 5H2O and 3-Hfl and stirring under
Results
The goal of this investigation was to prepare and compre-
hensively characterize a structurally similar set of divalent
metal flavonolate complexes as a prelude to O2 reactivity
studies of these complexes. The metal ions employed are
those of relevance to bacterial and fungal quercetin dioxy-
genases (Mn(II), Fe(II), Co(II), Ni(II), Cu(II)). For spectro-
scopic and redox behavior comparisons, the d10 Zn(II)
flavonolate analogue complex was also prepared.
Synthesis. Divalent metal flavonolate complexes of the
general formula [(6-Ph2TPA)M(3-Hfl)]X were prepared
by the synthetic routes outlined in Scheme 2. Triflate
derivatives were isolated as analytically pure polycrystal-
line materials. Perchlorate analogues were found to be
highly crystalline materials suitable for single crystal
X-ray crystallography. All syntheses were performed
under a N2 atmosphere.
X-ray Crystallography. The perchlorate analogues
1-5-ClO4 were characterized by single crystal X-ray
crystallography. A summary of the data acquisition and
refinement parameters are given in Table 1. Selected bond
distances and angles are given in Table 2 and Supporting
Information, Table S1, respectively.
Thermal ellipsoid drawings of the Mn(II) (1-ClO4) and
Ni(II) (3-ClO4) complexes are shown in Figure 1. Each
metal center exhibits a pseudo-octahedral geometry with
the ketone oxygen (O(2)) of the flavonolate positioned
trans to the tertiary amine nitrogen of the chelate ligand.
In both complexes, the flavonolate ligand is located
between the two hydrophobic phenyl appendages of the
chelate ligand.
3
3
a nitrogen atmosphere for 2 h. The solvent was then removed
under reduced pressure and the remaining solid was dissolved in
CH2Cl2 (∼2 mL) and filtered through a Celite/glass wool plug.
The filtrate was then brought to dryness. A portion of the solid
(8.45 ꢀ 10-5 mol) was dissolved in CH3CN (∼ 5 mL) and this
solution was exposed to air for 24 h. Diethyl ether was then
diffused into the CH3CN solution, resulting in the deposition of
dark-brown crystals suitable for single crystal X-ray crystal-
lography. Yield: 58%. UV-vis (MeOH), nm (ε, M-1 cm-1) 388
(24400), 490 (7300); ESI/APCI-MS, m/z (relative intensity)
743.2 ([M - 2ClO4]2þ, 69%). Anal. Calcd for: C90H70Cl2Fe2-
N8O15 2H2O: C, 62.77; H, 4.33; N, 6.51. Found: C, 62.96; H,
3
4.19; N, 6.70.
X-ray Crystallography. For each compound, 1-5-ClO4
and 6, a single crystal was mounted on a glass fiber with traces
of viscous oil and then transferred to a Nonius KappaCCD
˚
diffractometer equipped with Mo KR radiation (λ = 0.71073 A)
for data collection. For unit cell determination, 10 frames of
data were collected at 150(1) K with an oscillation range of
1 deg/frame and an exposure time of 20 s/frame. Final cell
constants were determined from a set of strong reflections from
the actual data collection. Reflections were indexed, integrated,
and corrected for Lorentz, polarization, and absorption effects
using DENZO-SMN and SCALEPAC.33 The structures were
solved by a combination of direct and heavy-atom methods
using SIR 97.34 All of the non-hydrogen atoms were refined with
anisotropic displacement coefficients. Unless otherwise stated,
all hydrogen atoms were assigned isotropic displacement co-
efficients U(H) = 1.2U(C) or 1.5U(Cmethyl), and their coordi-
nates were allowed to ride on their respective carbons using
SHELXL97.35
To our knowledge, complex 1-ClO4 is the first Mn(II)
complex of the 6-Ph2TPA ligand to exhibit coordination
of both phenyl-appended pyridyl donors. Complexes
having an additional bidentate ligand, such as the hydro-
xamate complex [(κ3-6-Ph2TPAMn)2(μ-ONHC(O)CH3)2]-
(ClO4)225 and the oxalate derivative [(κ3-6-Ph2TPAMn)2-
(μ-C2O4)](ClO4)2,37 have previously been found to exhibit
κ3-coordination (facial and meridional, respectively) of
the chelate ligand, with one non-coordinated phenylpyr-
idyl appendage. In 1-ClO4, the Mn(1)-O(1) and Mn-
Structure Solution and Refinement. Complex 1-ClO4-crystal-
lizes in the space group P21/n, with a disordered ClO4 in the
asymmetric unit. Complex 2-ClO4 crystallizes in the space group
C2/c with two cation/anion pairs per asymmetric unit along with
two molecules of CH3CN. The differences in the cations (with
the second being labeled with (A)) are subtle, with the most
noticeable differences being in one Co-NPhPy bond distance
and in O-Co-N bond angles involving the O(1) atom of the
flavonolate and nitrogen atoms of the chelate ligand. One of the
two CH3CN solvate molecules is composed of two 50% occu-
pied positions for all heavy atoms. Complex 3-ClO4 crystallizes
in the space group P21/a. Two oxygen atoms of the perchlorate
anion exhibit disorder (80:20) over two positions. Complex 4-
ClO4 crystallizes in the space group P21/c. One molecule of
diethyl ether is also present in the asymmetric unit. Complex 5-
ClO4 crystallizes in the space group P1. Two independent
cation/anion pairs and two CH2Cl2 solvent molecules are found
in the asymmetric unit. The two cations (with the second being
˚
˚
(1)-O(2) distances (2.121(3) A and 2.143(3) A, respe-
ctively) are similar. In the only other Mn(II) complex of
3-Hfl reported to date, [(Mn(3-Hfl)2(py)2],23 the Mn-O
˚
distances differ by ∼0.06 A, with the bond involving the
ketone oxygen being longer. The average Mn-O distance
in 1-ClO4 (2.13 A) is shorter than that found in [(Mn(3-
˚
23
˚
Hfl)2(py)2] (2.16 A). The average Mn-N distance in
3
˚
1-ClO4 is ∼2.33 A, which is similar to that found in [(κ -6-
Ph2TPAMn)2(μ-ONHC(O)CH3)2](ClO4)2 and [(κ3-6-
25
(33) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307–326.
(34) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Gia-
covazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. J.
Appl. Crystallogr. 1999, 32, 115–119.
Ph2TPAMn)2(μ-C2O4)](ClO4)2.37
(36) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor, G.
C. J. Chem. Soc., Dalton Trans. 1984, 1349–1356.
(37) Fuller, A. L.; Arif, A. M.; Berreau, L. M. unpublished result.
€
€
(35) Sheldrick, G. M. SHELXL97; University of Gottingen: Gottingen,
Germany, 1997.

Article
Inorganic Chemistry, Vol. 49, No. 1, 2010 87
Scheme 2
The mononuclear 3-ClO4 is structurally similar to
[(Ni(3-Hfl)2(py)2],38 theNi-O distances are slightly longer
(2.023(2) and 2.067(2) A).
˚
Ni(II) acetohydroxamate and enolate complexes of the
6-Ph2TPA ligand.25,28 For example, the average Ni-O
and Ni-NPhPy, as well as the NPy and NAmine distances
(Table 2) are similar in these complexes. The Ni(1)-O(1)
The structure of the Co(II) complex [(6-Ph2TPA)Co(3-
Hfl)]ClO4 (2-ClO4 2CH3CN) contains two independent
3
cation/anion pairs per asymmetric unit (the second la-
beled with an “A” designation). The two cations are
structurally similar, each having a pseudo-octahedral
geometry (Figure 2). Unlike the Mn(II) and Ni(II) analo-
gues (1-ClO4 and 3-ClO4), this complex has the deproto-
nated oxygen atom of the 3-Hfl ligand positioned trans to
the tertiary amine nitrogen atom of the chelate ligand and
˚
distance (1.996(2) A) is slightly shorter than the distance
˚
involving the ketone oxygen atom (Ni(1)-O(2), 2.010(3) A).
In the only other Ni complex of 3-Hfl reported to date,
(38) Farina, Y.; Yamin, B. M.; Fun, H.-K.; Yip, B.-C.; Teoh, S.-G. Acta
Crystallogr. 1995, C51, 1537–1540.

88 Inorganic Chemistry, Vol. 49, No. 1, 2010
Table 1. Summary of X-ray Data Collection and Refinementa
1-ClO4
Grubel et al.
2-ClO4 2CH3CN
3-ClO4
4-ClO4 Et2O
3
5-ClO4 2CH2Cl2
3
3
empirical formula
Mr
C45H35ClMnN4O7
834.16
monoclinic
P21/n
11.6175(8)
16.3708(14)
20.6689(11)
90
94.219(5)
90
3920.3(5)
4
1.413
150(1)
yellow
prism
0.38 ꢀ 0.13 ꢀ 0.13
Nonius KappaCCD
0.464
C94H76Cl2Co2N10O14
1758.41
monoclinic
C2/c
43.8551(6)
18.2721(2)
21.2336(3)
90
101.7509(8)
90
16658.4(4)
8
1.402
150(1)
brown
plate
0.38 ꢀ 0.38 ꢀ 0.20
Nonius KappaCCD
0.536
C45H35ClN4NiO7
837.93
monoclinic
P21/a
18.5881(4)
19.6427(4)
10.4897(2)
90
90.4537(11)
90
C49H45ClCuN4O8
916.88
monoclinic
P21/c
17.7565(6)
12.6011(2)
19.0305(6)
90
90.0709(12)
90
4258.1(2)
4
1.430
150(1)
green
prism
0.25 ꢀ 0.23 ꢀ 0.05
Nonius KappaCCD
0.638
C92H74Cl6N8O14Zn2
1859.04
triclinic
crystal system
space group
P1
˚
a /A
11.6633(2)
18.3461(3)
20.3958(3)
103.0861(8)
90.5675(9)
98.3257(9)
4201.95(12)
2
1.469
150(1)
yellow
prism
˚
b /A
˚
c /A
R /deg
β /deg
γ /deg
3
˚
V /A
Z
3829.88(13)
4
1.453
Dc/Mg m-3
T /K
color
150(1)
yellow/green
dichroic prism
0.35 ꢀ 0.18 ꢀ 0.10
Nonius KappaCCD
0.636
crystal habit
crystal size/ mm
diffractometer
0.28 ꢀ 0.28 ꢀ 0.15
Nonius KappaCCD
0.833
μ/ (mm-1
)
2θmax /deg
50.02
97.8
10543
6770
0.0623
552
0.0640/0.1367
1.043
54.96
99.9
36402
19074
0.0519
1084
0.0562/0.1411
1.027
54.98
99.2
15907
8729
0.0410
532
0.0576/0.1087
1.036
50.74
98.8
14181
7731
0.0377
568
0.0646/0.1564
1.029
52
99.2
30941
16397
0.0519
1127
0.0458/0.1011
1.027
completeness to θ (%)
reflections collected
independent reflections
Rint
variable parameters
R1/wR2b
goodness-of-fit (F2)
-3
˚
ΔFmax/min/e A
0.415/-0.304
1.181/-0.760
0.744/-0.672
1.093/-0.772
0.709/-0.726
P
P
P
P
a
b
Radiation used: Mo KR (λ = 0.71073 A). R1 = ||Fo| - |Fc||/ |Fo|; wR2 = [ w(Fo2 - Fc2)2/ w(Fo2)2]1/2, where w = 1/[σ2(Fo2) þ (aP)2 þ bP].
˚
˚
Table 2. Selected Bond Distances (A) for Complexes 1-5-ClO4
a
1-ClO4
2-ClO4 2CH3CNa
3-ClO4
4-ClO4 Et2O
3
5-ClO4 2CH2Cl2
3
3
M-N(1)
M-N(2)
M-N(3)
M-N(4)
M-O(1)
M-O(2)
ΔM-O
2.284(4)
2.340(3)
2.390(3)
2.338(3)
2.121(3)
2.143(3)
0.022
2.111(3)
2.175(2)
2.244(2)
2.331(3)
1.956(2)
2.172(2)
0.22
2.061(2)
2.067(3)
2.278(2)
2.231(2)
1.996(2)
2.031(2)
0.035
1.993(3)
2.029(3)
2.340(3)
2.085(2)
2.144(2)
2.107(2)
1.921(3)
2.010(3)
0.089
1.951(2)
2.1175(19)
0.17
C(31)-O(1)
C(32)-O(2)
1.313(4)
1.262(5)
1.317(4)
1.260(4)
1.315(3)
1.267(3)
1.349(5)
1.252(5)
1.315(3)
1.255(3)
a Data for one of two cations present in the asymmetric unit.
Table 3. Analytical, Spectroscopic, Magnetic, and Cyclic Voltammetry Data for 1-5-OTf
complex
1-OTf
2-OTf
3-OTf 0.25CH2Cl2
3
4-OTf
5-OTf
Anal. Calcd. (found):
C:
H:
N:
62.51 (62.05)
3.99 (4.06)
6.34 (6.16)
431 (17600)
734.2059 (100)b
1550
5.90
þ365
62.23 (61.94)
3.97 (3.89)
6.31 (6.11)
422 (17100)
738.2037 (16)
1557
61.11 (61.06)
3.97 (4.04)
6.16 (5.86)
443 (20000)
737.2064 (15)
1548
3.34
þ506
61.91 (62.02)
3.95 (3.98)
6.28 (6.21)
428 (22200)
742.2002 (16)
1542
1.93
þ660h
61.78 (61.74)
3.94 (4.01)
6.28 (6.30)
420 (20100)
743.1984 (2)
1552
UV-vis, nm (ε, M-1 cm-1 a
)
ESI-MS m/z (rel intensity) [M-OTf]þ
FTIRc, cm
ν
CdO
-1
d
μ
eff, μB
4.38
g
e
Epaf versus Fc/Fcþ
þ534
a Spectra collected in dry CH3CN. b Obtaind by MALDI. c Spectra collected as dilute KBr pellets. d Determined by Evans method (Evans, D. F.
J. Chem. Soc. 1959, 2003). e Diamagnetic. All cyclic voltammetry data was obtained under argon in CH2Cl2 with a complex concentration of 1 mM and
f
Bu4NClO4 (0.1 M) as the supporting electrolyte. The scan rate was 50-100 mV/s. The experimental setup consisted of a platinum button working
electrode, a silver wire reference electrode, and a platinum wire auxiliary electrode. All potentials are reported relative to an internal Fc/Fcþ standard.
Under these conditions, the Fc/Fcþ couple is at þ460 mV versus silver wire. g No electrochemical behavior between þ1.4 V and -1.0 V. hA quasi-
reversible reduction peak at -1088 mV vs Fc/Fcþ has been tentatively assigned to the Cu(II)/Cu(I) couple as no other compound of the group has waves
at potentials lower than Fc/Fcþ.
exhibits notably different Co-O distances (Co(1)-O(1)
ohydroxamate complex [(6-Ph2TPA)Co(ONHC(O)CH3)]-
ClO4, which exhibits Co-O bond lengths of 1.935(2) and
˚
˚
1.956(2) A; Co(1)-O(2), 2.172(2) A). This is similar to the
bidentate ligand coordination present in the Co(II) acet-
2.142(2) A.25 The average Co-NPhPy distance is notably
˚

Article
Inorganic Chemistry, Vol. 49, No. 1, 2010 89
Figure 2. Thermal ellipsoid drawing (50% probability) of the cationic
portion of one of the two cations present in the asymmetric unit of
2-ClO4 2CH3CN. Hydrogen atoms have been omitted for clarity.
3
Figure 1. Thermal ellipsoid drawings (50% probability) of the cationic
portions of 1-ClO4 (top) and 3-ClO4 (bottom). Hydrogen atoms have
been omitted for clarity.
˚
longer in 2-ClO4 2CH3CN (av 2.39 A) versus the hydro-
3
˚
xamate complex (av 2.27 A), while the Co-NAmine and
Co-NPy distances are similar. A search of the Cambridge
Crystallographic Database (version 5.30 (November))
revealed that 2-ClO4 is the first structurally characterized
Co(II) flavonolate complex. One Co(III) flavonolate
complex has been previously characterized by X-ray
crystallography.39
The cationic portions of the Cu(II) and Zn(II) analo-
gues 4-ClO4 Et2O and 5-ClO4 2CH2Cl2 are shown in
3
3
Figure 3. The Cu(II) center in 4-ClO4 Et2O is distorted
3
square pyramidal (τ = 0.11).36 For the Zn(II) complex,
there are two cation/anion pairs per asymmetric unit. The
cations have only minor differences in bond lengths/
angles involving the Zn(II) center and both exhibit a
distorted square pyramidal geometry (τ = 0.35).36 The
overall features of 6-Ph2TPA chelate ligand coordination
in the cationic portions of 4-ClO4 Et2O and 5-ClO4
3
3
2CH2Cl2 differ in terms of the M-NPhPy distance, which
˚
is ∼0.24 A longer in the Cu(II) complex, and the M-NPy
˚
and M-NAmine distances which are >0.05 A longer in the
Zn(II) complex. In both 4-ClO4 Et2O and 5-ClO4
3
3
2CH2Cl2 the coordination of the flavonolate ligand in-
volves a shorter bonding interaction between the metal
and the deprotonated hydroxyl donor (1.921(3) and
˚
1.951(2) A, respectively) and a longer bond with the
˚
ketone oxygen (2.010(3) and 2.1175(19) A, respectively).
Several Cu(II) complexes having a 3-Hfl ligand have been
(39) One Co(III) flavonolate complex supported by a salen ligand is listed
in the Cambridge Crystallographic Database: Hiller, W.; Nishinaga, A.;
Rieker, A. Z. Naturforsch., B: Chem. Sci. 1992, 47, 1185.
(40) Kaizer, J.; Pap, J.; Speier, G.; Parkanyi, L. Eur. J. Inorg. Chem. 2004,
2253–2259.
(41) Lippai, I.; Speier, G.; Huttner, G.; Zsolnai, L. Chem. Commun. 1997,
741–742.
Figure 3. Thermal ellipsoid drawings (50% probability) of the cationic
portions of 4-ClO4 Et2O (top) and 5-ClO4 2CH2Cl2 (bottom). Hydro-
gen atoms have been omitted for clarity.
3
3
previously characterized by X-ray crystallography.40-46
A portion of these compounds have a bidentate or

90 Inorganic Chemistry, Vol. 49, No. 1, 2010
Grubel et al.
tridentate supporting chelate ligand,40-43 whereas others
are species such as [Cu(3-Hfl)2(py)2] and [Cu(3-Hfl)2]. In
these two types of complexes, the Cu-O bond distances
which indicate that each complex has a high-spin divalent
metal center, and redox features identified in the range
of þ1.4 V to -1.0 V. 1H NMR spectra were also collected
for the Co(II), Ni(II), and Zn(II) complexes.
˚
are generally in the range of 1.90-2.21 A, and the
47
˚
ΔM-O varies from 0.04 to 0.29 A. For Zn(II), three
Infrared Spectroscopy. In the solid state, each complex
exhibits a νCdO vibration for the coordinated flavonolate
ketone moiety. The highest energy νCdO is found for the
cobalt complex 2-OTf, which is consistent with the fact
that the perchlorate analogue has the longest M(1)-O(2)
distance of the series of complexes and therefore should
polarize the carbonyl to the weakest extent. That being
said, we note that the C(32)-O(2) bond lengths through-
out the series 1-5-ClO4 are the same within experimental
error. The Cu(II) complex 4-OTf exhibits the lowest
energy νCdO vibration, which is consistent with the
structural data for 4-ClO4, which has the shortest M(1)-
O(2) distance of the series of complexes.
3-Hfl complexes, having either a bidentate or triden-
tate supporting chelate ligand, have been previously
reported.48-50 These complexes exhibit a range of Zn-O
˚
distances of 1.98-2.24 A and ΔM-O values of 0.16-
˚
0.26 A. The bond lengths and ΔM-O values for 4-ClO4
Et2O and 5-ClO4 2CH2Cl2 fall within the noted ranges.
3
3
Considering the entire series of 6-Ph2TPA-supported
complexes, the asymmetry in terms of the M-O interac-
˚
tions in 5-ClO4 2CH2Cl2 (ΔM-O ∼ 0.17 A) is slightly less
3
than that found in the Co(II) derivative 2-ClO4 (ΔM-O
˚
∼ 0.21 A), considerably greater than that found in the
˚
Mn(II) and Ni(II) analogues (ΔM-O < 0.04 A), and
approximately double that found in the Cu(II) complex
UV-vis Spectroscopy. Neutral flavonol compounds
exhibit an absorption feature in the range of 350-
380 nm, which is termed band I and is assigned to the
πfπ* transition.52 When dissolved in CH3CN under
anaerobic conditions, each complex 1-5-OTf exhibits
an intense absorption feature in the range of 420-443
nm. For comparison, we have measured the absorption
features of a flavonolate salt ([Me4Nþ][3Hfl-]) in CH3CN
(band I, λmax = 458 nm).53 For each metal complex, a
hypsochromic shift of the absorption band is found
relative to the 3-Hfl salt, with the exact shift being
influenced by the nature of the metal ion. For example,
the Mn(II) complex 1-OTf exhibits band I at 431 nm. This
matches well with the band I reported for Mn(3-Hfl)2-
(py)2 (432 nm).23 To our knowledge, absorption spectra
of an enzyme/substrate complex a Mn(II)-containing
quercetinase enzyme have not been reported.11 Notably,
the Ni(II) complex 3-OTf exhibits the smallest shift with
band I at 443 nm, and the zinc and cobalt complexes
5-OTf and 2-OTf exhibit the largest shifts, with band I at
420 and 422 nm, respectively. For the Ni(II)-containing
form of QueD from Streptomyces sp. FLA, under anae-
robic conditions, addition of quercetin at pH = 8 pro-
duces an absorption feature at 385 nm, which represents a
hypsochromic shift relative to the absorption feature of
free quercetin at pH = 8 (λmax = 391 nm). Similarly,
addition of quercetin to anaerobic cobalt-containing
QueD produces a band I absorption at 378 nm.12 Thus,
in both the synthetic and enzyme systems, the presence of
Co(II) produces a more significant hypsochromic shift in
band I. The absorption maximum found for the Cu(II)
complex 4-OTf (428 nm) is within the reported range for
Cu(II) flavonolate complexes (416-434 nm).19 The band
I absorption of [Zn(3-Hfl)(idpa)]ClO4 (idpa = 3,30-
iminobis(N,N-dimethylpropylamine) has been reported
as 419 nm, which matches well with that found for
5-ClO4.54 Overall, from the limited set of well-character-
ized divalent metal 3-Hfl complexes reported in the
literature to date (including those reported herein), the
hypsochromic shift of the band I πfπ* transition
˚
(ΔM-O ∼ 0.09 A). Hence, in the series of 6-Ph2TPA
supported complexes, the asymmetry of flavonolate co-
ordination increases in the order Mn (II) ∼ Ni(II) <
Cu(II) < Zn(II) < Co(II).
Examination of the bond lengths within the coordi-
nated flavonolate ligands of 1-5-ClO4 revealed that the
Mn(II), Co(II), Ni(II), and Zn(II) complexes exhibit only
˚
a slight elongation (0.02-0.03 A) of the C(32)-O(2) bond
involving the ketone moiety, and a slight contraction
˚
(∼0.04 A) of the C(31)-O(1) bond of the hydroxyl donor,
51
relative to the distances found in uncoordinated flavonol
˚
(1.232(3) and 1.357(3) A, respectively). The Cu(II)
analogue 4-ClO4 Et2O exhibits C-O distances (C(32)-
3
˚
˚
O(2) 1.252(5) A; C(31)-O(1), 1.349(5) A) that are quite
close to those of the free flavonol.51 Notably, the Cu(II)
complex exhibits a short C(31)-C(32) bond (1.399(6) A)
relative to that found in the other 6-Ph2TPA-supported
˚
˚
complexes (av 1.45 A). A similar short C-C bond length
was found in [Cu(II)(bpy)(3-Hfl)(2-HOC6H4COCO2)]
41
(1.369(14) A), but other known Cu(II) flavonolate
˚
42-46
˚
derivatives exhibit a distance of ∼1.44 A.
Overall,
the bond lengths within the flavonolate ligands of
1-5-ClO4 are minimally affected by the identity of the
metal ion present.
Spectroscopic, Magnetic, and Redox Properties of 1-
5-OTf. Presented in Table 3 are selected spectroscopic
features of 1-5-OTf. Also given are magnetic moments,
(42) Balogh-Hergovich; Kaizer, J.; Speier, G.; Huttner, G.; Zsolnai, L.
Inorg. Chim. Acta 2000, 304, 72–77.
(43) Balogh-Hergovich, E.; Kaizer, J.; Speier, G.; Huttner, G.; Jacobi, A.
Inorg. Chem. 2000, 39, 4224–4229.
(44) Okabe, N.; Yamamoto, E.; Yasunori, M. Acta Crystallogr., Sect. E:
Struct. Rep. Online 2003, 59, m715–m716.
(45) Balogh-Hergovich, E.; Speier, G.; Argay, G. J. Chem. Soc., Chem.
Commun. 1991, 551–552.
ꢀ
ꢀ
(46) Balogh-Hergovich, E.; Kaizer, J.; Speier, G.; Argay, G.; Parkanyi, L.
J. Chem. Soc., Dalton Trans. 1999, 3847–3854.
(47) The Cu(II) complex [Cu(II)(bpy)(fla)(2-HOC6H4COCO2)] has been
˚
reported to have Cu-O bond lengths of 2.190(9) and 1.760(11) A and
41
˚
Δ
M-O = 0.43 A.
(48) Balogh-Hergovich, E.; Kaizer, J.; Speier, G.; Huttner, G.; Rutsch, P.
Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1999, 55, 557–558.
(49) Annan, T. A.; Peppe, C.; Tuck, D. G. Can. J. Chem. 1990, 68, 423–
430.
(52) Jurd, L.; Geissman, T. A. J. Org. Chem. 1956, 21, 1395–1401.
ꢀ
(53) Barhacs, et al. report the absorption maxium of [K][3-Hfl] to be 465
ꢀ
ꢀ
(50) Kaizer, J.; Kupan, A.; Pap, J.; Speier, G.; Reglier, M.; Michel, G. Z.
ꢀ
Kristallogr. -New Cryst. Struct. 2000, 215, 571–572.
(51) Etter, M. C.; Urbanczyk-Lipkowska, Z.; Baer, S.; Barbara, P. F. J.
Mol. Struct. 1986, 144, 155–167.
nm. Barhacs, L.; Kaizer, J.; Speier, G. J. Org. Chem. 2000, 65, 3449-3452.
ꢀ
(54) Barhacs, L.; Kaizer, J.; Speier, G. J. Mol. Catal. A 2001, 172, 117–
125.

Article
Inorganic Chemistry, Vol. 49, No. 1, 2010 91
Figure 5. 1H NMR spectra of 2-OTf (top) and 3-OTf (bottom) in
CD3CN at ambient temperature.
Figure 4. EPR spectra of 1-OTf and 4-OTf. Spectra were recorded at
4.7 K, 1.002 mW microwave power, and 9.39 GHz. The samples were
e1 mM in CH3CN.
1H NMR Spectroscopy. The 1H NMR spectra of analy-
tically pure 2-OTf, 3-OTf, and 5-OTf were measured in
CD3CN at ambient temperature. The spectra for the
paramagnetic Co(II) and Ni(II) complexes are shown in
Figure 5. While full assignment of the resonances of 2-OTf
and 3-OTf have not been made, the spectra are clearly
distinct from those of other complexes of bidentate ligands
(e.g., hydroxamate derivatives) and can be used toevaluate
complex purity and reactivity.25 There are no concentra-
tion dependent changes in these NMR spectra. Attempts
were made to collect two-dimensional COSY spectra to
assign the pyridyl ring proton resonances of the 6-Ph2TPA
ligands in 2-OTf and 3-OTf.26 Unfortunately these experi-
ments did not reveal any crosspeaks. It is important to note
that the total number of paramagnetically shifted reso-
nances in the spectra of 2-OTf and 3-OTf is consistent with
the presence of an effective plane of symmetry containing
the pyridyl donor, which gives equivalent phenyl-ap-
pended pyridyl donors. This indicates that the solution
structure of these compounds is generally similar to that
(relative to the free 3-Hfl anion) generally increases in the
order Ni(II) < Mn(II) < Cu(II) < Zn(II), Co(II).
EPR Spectroscopy. EPR spectra were obtained for the
Mn(II) and Cu(II) complexes of the series (1-OTf and
4-OTf) and are shown in Figure 4. The spectrum collected
for 1-OTf is similar to the spectrum collected for the
hydroxamate-bridged dimanganese complex [(6-Ph2-
TPAMn)(μ-ONHC(O)CH3)2](ClO4)2.25 No hyperfine
coupling is discernible in the spectrum of either the
1-OTf or the Mn(II) hydroxamate complex. The similar-
ity of these spectra suggested to us the possible formation
of a flavonolate-bridged dimanganese complex in solution.
Reactivity studies outlined below provide additional evi-
dence for the chemical similarities of 1-OTf and the hydro-
xamate complex in CH3CN. We note that the Mn(II)-
containing form of the B. subtilis quercetin dioxygenase in
the presence of quercetin under anaerobic conditions ex-
hibits a six-line EPR signal at g = 2, with a hyperfine
coupling constant of approximately 93 G.11 A second, less
intense six-line signal is also present at g = 9 and exhibits a
similar hypefine coupling constant. This data is consistent
with the enzyme containing a mononuclear pseudo-octahe-
dral Mn(II) center having oxygen/nitrogen ligands.55
The axial EPR spectrum of 4-OTf is consistent with the
distorted square pyramidal geometry of the Cu(II) center
and a dx2-y2 ground state (g ∼ 2.24; g^ ∼ 2.06; A ∼
180 G). The anaerobic enzyme/substrate complex of the
quercetinase enzyme from A. japonius exhibits an EPR
signal with g = 2.366 and A ∼ 110 G.56
1
found in the solid state. The features of the H and 13C
NMR spectra of 5-OTf (see Experimental Section) are
consistent the formation of a six-coordinate Zn(II) center
in solution via coordination of all donors of the supporting
6-Ph2TPA ligand. Most notably, the total number of
carbon signalsisconsistentwiththepresenceofaneffective
plane of symmetry that makes the phenyl-appended pyr-
idyl donors equivalent.
Cyclic Voltammetry. The redox properties of 1-5-OTf
were examined by cyclic voltammetry. An irreversible oxi-
dation wave was identified for 1-OTf and 3-5-OTf, and the
Epa values are given in Table 3. The Co(II) complex 2-OTf
did not show any electrochemical behavior in the range
examined. A quasi-reversible reduction peak at -1088 mV
versus Fc/Fcþ (Supporting Information, Figure S1) for
(55) Reed, G. H.; Markham, G. D. Biological Magnetic Resonance;
Berliner, L. J., Rueben, J., Eds.; Plenum: New York, 1978; Vol. 6, pp 73-142.
(56) Spectrum obtained at 77 K.

92 Inorganic Chemistry, Vol. 49, No. 1, 2010
Grubel et al.
4-OTf has been tentatively assigned to the Cu(II)/Cu(I)
couple, as no other compound of the group has waves at
potentials lower than Fc/Fcþ. These data for 1-5-OTf
indicate that the nature of the divalent metal influences
the redox potential of the coordinated flavonolate over a
range of ∼300 mV. The differences in oxidation potential
found for this series of complexes depend on the metal and/
or the solution structure of the metal complex. The Mn(II)
derivative 1-OTf, which based on EPR, may form binuclear
species in solution, has the lowest potential, indicating that
the flavonolate is a better reductant toward O2 than the
same ligand in the Ni(II), Cu(II), and Zn(II) analogue
complexes. For the Ni(II) and Zn(II) complexes 3-ClO4
and 5-ClO4, irreversible oxidation of the flavonolate ligand
occurs at þ0.506 and þ0.536 V versus Fc/Fcþ, respectively.
For the Cu(II) complex, which is distorted square pyramidal
in acetonitrile, the potential is more positive (þ0.660 V).
Potentials previously reported for 3-Hfl complexes of the
generalformula[M(3-Hfl)(H2O)n]Cl(M=Cu, n=2;M=
Fe, n = 4) are in the same range as those reported herein.57
Ligand Exchange Reactivity. The results of initial elec-
trospray ionization mass spectrometry experiments sug-
gested that the Mn(II) flavonolate complex 1-OTf would
undergo reaction with available Zn(II) ion to produce a
ligand exchange product, [(6-Ph2TPA)Zn(OTf)]þ. To ex-
amine this reactivity, in independent NMR tube experi-
ments, [(6-Ph2TPA)Mn(3-Hfl)]OTf was treated with an
the compounds may have similar solution structures. We
propose that these structures include μ-η1:η2-coordina-
tion of the chelate anion (3-Hfl or acetohydroxamate) as
is found in the X-ray structure of [(6-Ph2TPAMn)(μ-
ONHC(O)CH3)2](ClO4)2.25 In this coordination motif,
each Mn(II) center coordinates the anionic oxygen of
each bridging ligand. We propose that from this structure
an equilibrium mixture is formed, and is composed of
0.5 equiv of [Mn(3-Hfl)2 0.5H2O], 0.5 equiv of free
3
chelate ligand, and 0.5 equiv of [(6-Ph2TPA)Mn(CD3-
CN)2](X)2 (Scheme 3). Formation of the final product
mixture, including 1 equiv of [(6-Ph2TPA)M(CD3CN)n]-
(X)2 (M = Co, Ni, Zn), could result from coordination of
the free divalent metal ion to the free chelate ligand, and
displacement of Mn(II) (as MnX2) from [(6-Ph2TPA)Mn-
(CD3CN)2](X)2. The latter reaction has been indepen-
dently investigated by 1H NMR and found to occur
rapidly in CD3CN for M = Co(II), Ni(II), and Zn(II).
In another type of ligand exchange reaction, we found
that treatment of the non-manganese flavonolate com-
plexes [(6-Ph2TPA)M(3-Hfl)]OTf (M = Co (2-OTf), Ni
(3-OTf), and Cu (4-OTf)) with 0.5 equiv of Zn(ClO4)2
6H2O in CD3CN results in the formation of [(6-Ph2TPA)-
3
-
M(CD3CN)n](X)2 (n = 1 or 2; X = OTf- or ClO4
)
1
species as indicated by H NMR. Hence, in these reac-
tions flavonolate ligand exchange occurs, but no change
in the metal bound by the 6-Ph2TPA ligand takes place.
equimolar amount of M(ClO4)2 6H2O (M = Co(II),
The other product in the reaction is [Zn(3-Hfl)2 2H2O]
3
3
Ni(II), and Zn(II)) in CD3CN. For each reaction, well
resolved 1H NMR spectra consistent with the quantitative
formation of a divalent metal solvate complex of the added
metal, [(6-Ph2TPA)M(CD3CN)n](X)2 (n = 1 or 2; X =
OTf- or ClO4-) were obtained.25,26 A yellow-green pre-
cipitate formed in each reaction mixture and was identified
(eq 3), which is a poorly soluble yellow precipitate. This
material was identified by independent synthesis and
characterization, followed by comparison of UV-vis
1
and H NMR properties between the reaction product
and the independently generated material.
M ¼ Co, Ni, Cu; X ¼ OTf -or ClO4
-
as [Mn(3-Hfl)2 0.5H2O] by elemental analysis. This com-
3
pound was isolated from the reaction mixture involving
Ni(II) by selective crystallization in 39% yield (for
0.5 equiv). In addition to the stoichiometric formation of
[(6-Ph2TPA)M(CD3CN)n](X)2 and 0.5 equiv of [Mn-
½ð6-Ph2TPAÞMð3-HflÞꢁOTf þ 0:5ZnðClO4Þ2 3 6H2O
þ nCD3CN f ½ð6-Ph2TPAÞMðCD3CNÞnꢁðXÞ2
þ 0:5½Znð3-HflÞ2 3 2H2Oꢁ þ 2H2O
ð3Þ
(3-Hfl)2 0.5H2O], mass balance requires the formation of
3
0.5 equiv of MnX2 salt (eq 1), which could not be isolated
from the reaction mixture. Notably, we found that the
Mn(II) hydroxamate complex [(6-Ph2TPAMn)(μ-ONHC-
What about Fe(II)? The quercetinase enzyme from B.
subtilis was originally described as a non-heme iron
enzyme.10 However, the turnover number for this enzyme
was reported to be 2 orders of magnitude lower than for
the Cu(II)-containing A. flavus, which led researchers to
suggest that Fe(II) might not be the correct cofactor for
the enzyme. Follow-up studies suggested that Mn(II) was
probably the correct cofactor for the B. subtilis enzyme.11
To date, only two iron flavonolate complexes, both Fe-
(III) derivatives, have been crystallographically charac-
terized.23,24
25
(O)CH3)2](ClO4)2 also reacts with M(ClO4)2 6H2O to
3
produce [(6-Ph2TPA)M(CD3CN)](X)2 (M = Co, Ni, Zn;
X = OTf- or ClO4-; eq 2), as determined by 1H NMR.
The similarity in EPR spectral features (vida supra)
and reactivity of [(6-Ph2TPA)Mn(3-Hfl)]OTf and [(6-
Ph2TPAMn)(μ-ONHC(O)CH3)2](ClO4)2 suggest that
-
M ¼ Co, Ni, Zn; X ¼ OTf -or ClO4
½ð6-Ph2TPAÞMnð3-HflÞꢁOTf þ MðClO4Þ2 6H2O
3
To gauge the chemistry of Fe(II) relative to the other 3d
metal ions investigated herein, we initiated attempts to
prepare a Fe(II) flavonolate complex supported by the
6-Ph2TPA ligand. In our first approach, treatment of
þ nCD3CN f ½ð6-Ph2TPAÞMðCD3CNÞnꢁðXÞ2
þ 0:5MnX2 þ 0:5½Mnð3-HflÞ2 3 0:5H2Oꢁ þ 5:75H2O ð1Þ
½ð6-Ph2TPAÞMnðONHCðOÞCH3ÞꢁClO4 þ MðClO4Þ2 3
6H2O þ nCD3CN f ½ð6-Ph2TPAÞMðCD3CNÞnꢁðXÞ2
þ 0:5MnX2 þ 0:5½MnðONHCðOÞCH3Þ2ꢁ þ 6H2O ð2Þ
6-Ph2TPA with equimolar amounts of Fe(ClO4)2 6H2O,
3
3-hydroxyflavonol, and Me4NOH 5H2O under a N2 atmo-
3
sphere, followed by stirring overnight, workup, and crystal-
lization from CH3CN/Et2O resulted in the deposition of
dark brown single crystals. X-ray crystallographic analysis
of these crystals revealed the formation of the diiron(III)
μ-oxo compound [(6-Ph2TPAFe(3-Hfl))2(μ-O)](ClO4)2
(57) de Souza, R. F. V.; Sussuchi, E. M.; De Giovani, W. F. Synth. React.
Inorg. Met.-Org. Chem. 2003, 33, 1125–1144.

Article
Inorganic Chemistry, Vol. 49, No. 1, 2010 93
Scheme 3. Proposed Reaction Pathway for Ligand Exchange in the Reaction of 1-OTf with Divalent Metal Perchlorate Salts
(6 2.5CH3CN, Figure 7). The structural and spectroscopic
spin Fe(II) center.58 The 1H NMR spectrum of 8 contains
3
properties of this complex are discussed in detail below.
However, as an approach toward understanding the reac-
tion pathway leading to the formation of 6, we further
explored the synthetic conditions. Treatment of 6-Ph2TPA
several paramagnetically shifted resonances over a range of
∼110 ppm (Figure 6). Investigation of the redox properties
of 8 by cyclic voltammetry (CH2Cl2, [8] = 1 mM, Bu4-
NClO4 (0.1 M) supporting electrolyte, scan rate 50-
100 mV/s) revealed a quasi-reversible couple at ∼-0.05 V
versus ferrocene/ferrocenium, which is assigned as the Fe(II)/
Fe(III) couple. No redox activity was found at more positive
potentials, indicating that upon oxidation of the iron center,
the redox potential for the coordinated flavonolate is shifted
from that observed for the other divalent metal 3-Hfl com-
plexes. This redox behavior is consistent with the observed
metal-centered O2 reactivity described below.
Complex 8 is very air sensitive, which complicated our
efforts to characterize it, especially in solution. Exposure
of a CH3CN solution of 8 to air results in a rapid
darkening of the color, which corresponds to the forma-
tion of the μ-oxo compound 6. This complex was char-
acterized by X-ray crystallography, elemental analysis,
with Fe(ClO4)2 6H2O in CH3CN, followed by recrystalli-
3
zation from CH3CN/Et2O, yielded [(6-Ph2TPA)Fe(CH3-
CN)](ClO4)2 (7) as pale yellow crystals. Characterization
details for this Fe(II) complex are given in the Supporting
Information. Treatment of 7 with equimolar amounts of
3-hydroxyflavonol and Me4NOH 5H2O in methanol, and
3
stirring for 15 min under a freshly purged N2 atmosphere,
gave a green-yellow solution from which, after workup, a
green-yellow precipitate was isolated. The analytical and
spectroscopic properties of this complex are consistent
with the formulation [(6-Ph2TPA)Fe(3-Hfl)]ClO4 (8). For
this complex, the flavonolate νCdO vibration is at
1558 cm-1, and the band I π fπ* transition is at 415 nm
(ε ∼ 14700 M-1 cm-1). These features are similar to those
found for the Co(II) derivative 2-OTf, suggesting that the
flavonolate ligand may be coordinated with a ΔM-O similar
to that found in the cobalt complex. The magnetic moment
for 8 (μeff = 4.7 μB) is consistent with the presence of a high-
(58) Huheey, J. E.; Keiter, E. A.; Keiter, R. L. Inorganic Chemistry:
Principles of Structure and Reactivity; Harper Collins: New York, 1993.

94 Inorganic Chemistry, Vol. 49, No. 1, 2010
Grubel et al.
Scheme 4
Figure 6. 1H NMR spectrum of [(6-Ph2TPA)Fe(3-Hfl)]ClO4 (8) in
CD3CN at ambient temperature. In addition to the resonances shown,
broad peaks are also present at 112 and 92 ppm.
Figure 7. Thermal ellipsoid drawings (50% probability) of the cationic
portions of 6 2.5CH3CN. Hydrogen atoms have been omitted for clarity.
UV-vis, FTIR, and mass spectrometry. A summary of
the iron coordination chemistry is shown in Scheme 4.
The cationic portion of 6 2.5CH3CN is shown in
3
5-ClO4 CH2Cl2. Similarly, the Fe(III) complexes [(salen)-
3
3
Fe(III)(3-Hfl)] and [Fe(40-MeOflaH)3] have ΔM-O
=
Figure 7. Details of the X-ray data collection of this
complex are given in Table 4. Selected bond distances
and angles are given in Table 5 and Supporting Informa-
tion, Table S2, respectively. Each Fe(III) center has a κ3-
6-Ph2TPA ligand, a bidentate flavonolate ligand, and the
μ-oxo bridge. The flavonolate ligands are coordinated to
0.18 and 0.15 A, respectively.23,24 Thus, all of the Fe(III)
flavonolate complexes that have been structurally char-
acterized to date exhibit very similar levels of asymmetry
with respect to flavonolate coordination. The Fe-
O(flavonolate) bond distances are also very similar in
all three compounds. We note that the flavonolate ketone
oxygen donor is positioned trans to the μ-oxo bridge in
˚
˚
Fe(1) and Fe(2) with ΔM-O = 0.17 and 0.14 A, respec-
tively. This is similar to the coordination found in the
Co(II) and Zn(II) flavonolate complexes 2-ClO4 and
6 2.5CH3CN, whereas the deprotonated oxygen donor is
3

Article
Inorganic Chemistry, Vol. 49, No. 1, 2010 95
Table 4. Summary of X-ray Data Collection and Refinement for 6 2.5CH3CNa
3
empirical formula
Mr
C95H77.5Cl2Fe2N10.5O15
1788.78
triclinic
crystal system
space group
P1
˚
a/ A
14.02370(10)
15.2054(2)
21.6638(3)
70.1168(7)
74.8473(7)
81.5826(8)
4184.67(9)
2
1.420
150(1)
red
prism
˚
b/ A
˚
c/ A
R/ deg
β/ deg
γ/ deg
3
˚
V/ A
Z
Dc/Mg m-3
T/ K
color
crystal habit
crystal size/ mm
diffractometer
0.28 ꢀ 0.18 ꢀ 0.15
Nonius KappaCCD
0.486
μ/ (mm-1
)
Figure 8. UV-vis absorption features of [(6-Ph2TPAFe(3-Hfl))2(μ-O)]-
(ClO4)2 (6) and [(6-Ph2TPA)Fe(3-Hfl)]ClO4 (8).
2θmax/ deg
54.92
99.0
35639
18968
0.0393
1151
0.0460/0.1010
1.019
completeness to θ (%)
reflections collected
independent reflections
Rint
nances in the range of 0 to ∼40 ppm (Supporting Infor-
mation, Figure S3). Several features of this spectrum are
significantly overlapped in the aromatic range (-10 to
40ppm), consistent with the presence of several types of aryl
protons. The dramatically different features of this spec-
trum versus that found for [(6-Ph2TPA)Fe(3-Hfl)]ClO4 (8,
Figure 6) is likely due to antiferromagnetic coupling of the
S = 5/2 metal centers via the oxo bridge.59,61,62
variable parameters
R1/wR2b
goodness-of-fit (F2)
-3
˚
ΔFmax/min/e A
0.397/-0.487
P
P
a
b
˚
Radiation used: Mo KR (λ = 0.71073 A). R1 = ||Fo| - |Fc||/
|
P
P
Fo|; wR2 = [ w(Fo - Fc2)2/ w(Fo2)2]1/2, where w = 1/[σ2(Fo2) þ
2
(aP)2 þ bP].
Final Comments
˚
Table 5. Selected Bond Distances (A) for 6 2.5CH3CN
3
The chemistry of metal complexes of flavonolate ligands is
of significant current interest. Metal flavonolate complexes
have been prepared and investigated for their antioxidant
and DNA cleavage reactivity.63-74 Because of their biologi-
cal relevance, metal flavonolate complexes have also been the
subject of several recent computational investigations.75-78
As noted in the introduction, fungal and bacterial querceti-
nase enzymes have been shown to exhibit varying levels of
activity as a function of the divalent metal ion present. While
Fe(1)-O(1)
Fe(1)-O(2)
Fe(1)-O(4)
Fe(1)-N(1)
Fe(1)-N(2)
Fe(1)-N(3)
O(1)-C(31)
O(2)-C(32)
1.9712(14)
2.1460(15)
1.7890(14)
2.1679(17)
2.2489(17)
2.2241(17)
1.330(3)
Fe(2)-O(1A)
Fe(2)-O(2A)
Fe(2)-O(4)
Fe(2)-N(1A)
Fe(2)-N(2A)
Fe(2)-N(3A)
O(1A)-C(31A)
O(2A)-C(32A)
1.9655(14)
2.1054(15)
1.7992(14)
2.1822(18)
2.2180(18)
2.2109(18)
1.326(3)
1.271(3)
1.274(2)
trans to the tertiary amine nitrogen. This results in a
meridional donor array from the κ3-bound chelate ligand.
The Fe-O bond distances involving the oxo bridge in
(61) Wu, F.-J.; Kurtz, D. M., Jr. J. Am. Chem. Soc. 1989, 111, 6563–6572.
(62) Norman, R. E.; Yan, S.; Que, L., Jr.; Backes, G.; Ling, J.; Sanders-
Loehr, J.; Zhang, J. H.; O’Connor, C. J. J. Am. Chem. Soc. 1990, 112, 1554–
1562.
(63) Ryan, P.; Hynes, M. J. J. Inorg. Biochem. 2008, 102, 127–136.
(64) Kazazic, S. P.; Butkovic, V.; Srzic, D.; Klasinc, L. J. Agric. Food
Chem. 2006, 54, 8391–8396.
(65) de Souza, R. F. V.; De Giovani, W. F. Spectrochim Acta A 2005, 61,
1985–1990.
(66) Bukhari, S. B.; Memon, S.; Mahroof-Tahir, M.; Bhanger, M. I.
Spectrochim. Acta A 2009, 71, 1901–1906.
6 2.5CH3CN are within experimental error identical to
3
those found in [(6-C6H4O-TPAFe)2(μ-O)](BPh4)2.59 The
˚
average Fe-N distance in 6 2.5CH3CN (∼2.20 A) is
3
slightly longer than what is found in [(6-C6H4-TPAFe)2-
˚
(μ-O)](BPh4)2 (∼2.18 A). However, both distances are
consistent with a high-spin (S = 5/2) state for each iron
center.60
A vibration at 1549 cm-1 in the solid state infrared
spectrum of 6 is assigned as the νCdO vibration of the
flavonolate, based on comparison to the spectral features
of [(salen)Fe(III)(3-Hfl)] (1549 cm-1).24 Compound 6 is
slightly soluble in methanol and exhibits absorption
bands at 388 and 490 nm. Unlike its Fe(II) flavonolate
precursor 8, an absorption feature for the flavonolate
ligand in the region of 400-450 nm could not be readily
identified. The absorption features of 6 and 8 are com-
pared in Figure 8. For [(salen)Fe(III)(3-Hfl)], an absorp-
tion maximum at 407 nm was reported.24 The 1H NMR
spectrum of 6 in d4-methanol consists of broad reso-
(67) Guo, M.; Perez, C.; Wei, Y.; Rapoza, E.; Su, G.; Bou-Abdallah, F.;
Chasteen, N. D. Dalton Trans. 2007, 4951–4961.
(68) El Hajji, H.; Nkhili, E.; Tomao, V.; Dangles, O. Free Radical Res.
2006, 40, 303–320.
(69) de Souza, R. F. V.; De Giovani, W. F. Redox Rep. 2004, 9, 97–104.
(70) Mira, L.; Fernandez, M. T.; Santos, M.; Rocha, R.; Florencio, M.
H.; Jennings, K. R. Free Radical Res. 2002, 36, 1199–1208.
(71) Tan, J.; Wang, B.; Zhu, L. J. Biol. Inorg. Chem. 2009, 14, 727–739.
(72) Tan, J.; Wang, B.; Zhu, L. Bioorg. Med. Chem. 2009, 17, 614–620.
(73) Tan, J.; Zhu, L.; Wang, B. Dalton Trans. 2009, 4722–4728.
ꢀ
ꢀ
ꢀ
(74) El Amrani, F. B. A.; Perrello, L.; Real, J. A.; Gozalez-Alvarez, M.;
ꢀ
Alzuet, G.; Borras, J.; Garcia-Granda, S.; Montejo-Bernardo, J. J. Inorg.
Biochem. 2006, 100, 1208–1218.
(75) Lekka, C. E.; Ren, J.; Meng, S.; Kaxiras, E. J. Phys. Chem. B 2009,
113, 6478–6483.
(76) Ren, J.; Meng, S.; Lekka, C. E.; Kaxiras, E. J. Phys. Chem. B 2008,
112, 1845–1850.
(59) Jensen, M. P.; Lange, S. J.; Mehn, M. P.; Que, E. L.; Que, L., Jr. J.
Am. Chem. Soc. 2003, 125, 2113–2128.
(60) Zang, Y.; Kim, J.; Dong, Y.; Wilkinson, E. C.; Appelman, E. H.;
Que, L., Jr. J. Am. Chem. Soc. 1997, 119, 4197–4205.
(77) Leopoldini, M.; Russo, N.; Chiodo, S.; Toscano, M. J. Agric. Food
Chem. 2006, 54, 6343–6351.
(78) Cornard, J. P.; Dangleterre, L.; Lapouge, C. Chem. Phys. Lett. 2006,
419, 304–308.

96 Inorganic Chemistry, Vol. 49, No. 1, 2010
Grubel et al.
a few initial investigations directed at evaluating the effect of
the metal ion on flavonolate/O2 chemistry of relevance to
quercetin dioxygenase enzymes have been reported,23,24 these
studies lack a systematic approach. Specifically, the studies
reported to date have included comparisons between com-
plexes having different supporting chelate ligands and/or
metal oxidation states. The research described herein is the
first detailed systematic mapping of how metal ion content
influences the structural, spectroscopic, redox properties, and
ligand exchange reactivity of structurally related flavonolate
complexes. These complexes are relevant to the metal/flavo-
nolate adducts proposed to form in quercetinase enzymes
of varying metal ion content. The results presented herein
lay the groundwork for detailed O2 reactivity studies as a
function of the divalent metal ion present. These investiga-
tions are currently in progress.
Acknowledgment. The authors thank the National
Science Foundation (CHE-0848858 (L.M.B.), CHE-
0615479 (J.A.H.)) for financial support of this research.
We also thank Ewa Szajna-Fuller for producing the
X-ray crystals of [(6-Ph2TPA)Fe(CH3CN)]ClO4 (7).
Supporting Information Available: Tables of bond angles for
1-5-ClO4 and 6 2.5CH3CN; cyclic voltammogram for 4-OTf;
3
synthetic and characterization details for [(6-Ph2TPA)Fe-
(NCCH3)](ClO4)2 (7); room temperature 1H NMR spectrum
of 6 in CD3OD. This material is available free of charge via the
Internet at http://pubs.acs.org.