Trinuclear first row transition metal complexes of a hexapyridyl, trialkoxy 1,3,5-triarylbenzene ligand

Article abstract of DOI:10.1039/c0cc05608a

Trinuclear complexes of Mn^II, Fe^II, Co^II, Ni^II, Cu^II, and Zn^II were synthesized using a ligand architecture based upon a 1,3,5-triarylbenzene core decorated with six pyridines and three alk

Full text of DOI:10.1039/c0cc05608a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 4189–4191
www.rsc.org/chemcomm
COMMUNICATION
Trinuclear first row transition metal complexes of a hexapyridyl,
trialkoxy 1,3,5-triarylbenzene ligandw
Emily Y. Tsui,z Jacob S. Kanady,z Michael W. Day and Theodor Agapie*
Received 16th December 2010, Accepted 14th February 2011
DOI: 10.1039/c0cc05608a
Trinuclear complexes of MnII, FeII, CoII, NiII, CuII, and ZnII
were synthesized using a ligand architecture based upon a 1,3,5-
triarylbenzene core decorated with six pyridines and three
alkoxide moieties. Characterization via X-ray diffraction,
NMR, and magnetism studies is discussed.
more complicated bridging patterns of up to three metals.4
Although the MII3(m-OR)3 structural motif is commonly
found in higher nuclearity clusters in complexes of 2,20-
dipyridylketone4 and as part of self-assembled tetranuclear
clusters such as cubanes5 and defective dicubanes,6 the motif is
less common in trinuclear complexes.7 To further investigate
the metal coordination potential of H3L and its control over
cluster nuclearity, trinuclear complexes of L containing other
first row transition metals were targeted.
The active sites of several enzymes involved in dioxygen
chemistry (laccase, ascorbate oxidase, the oxygen evolving
center of photosystem II) display three or more first row
transition metal centers.1 Synthetic catalysts for water oxidation
are also proposed to be multinuclear.2
Metallation studies were initiated with the acetate salts of
the first-row metals MnII, FeII, CoII, NiII, CuII, and ZnII in the
presence of base. Addition of three equivalents of solid
MII(OAc)2ꢀxH2O to a suspension of H3L in acetonitrile or a
mixture of acetonitrile–water followed by three equivalents of
a base such as sodium hydroxide or triethylamine resulted
in complete dissolution of insoluble materials within 12 h.
Analytically pure crystals were obtained by vapor diffusion of
diethyl ether into dichloromethane or chloroform solutions of
the reaction products.
In continued efforts to rationally design multinucleating
scaffolds,
a 1,3,5-triarylbenzene framework was utilized
to hold three multidentate binding sites near each other.
1,3,5-tris(2-(di(2-pyridyl)hydroxymethyl)phenyl)benzene (H3L,
Scheme 1) is accessible in two steps from commercially
available starting materials.3 Trinuclear copper complexes
supported by framework L have been synthesized containing
a conserved Cu3(m-OR)3 central moiety; varying the capping
anions from halides, phosphate, tetrafluoroborate, and triflate
causes subtle structural changes that affect the magnetism of
these complexes.3
Single crystal X-ray diffraction (XRD) studies demonstrate
the trinucleating nature of the deprotonated H3L framework
to give complexes generally formulated as LM3(OAc)3
(Fig. 1a). The three metal centers are bridged by three
alkoxides forming a six membered ring, and the pendant
pyridines coordinate with the two pyridines of each dipyridyl
moiety bound to adjacent metal centers. The coordination
environment is completed by acetate counterions.
Protonated and deprotonated dipyridylhydroxymethyl
moieties are known to exhibit an array of coordination modes,
from tridentate N,O,N coordination of a single metal center to
The LM3 core displays pseudo-C3 symmetry induced by a
twist of each dipyridylmethoxide arm. This binding mode renders
the two pyridines of each arm different, which is reflected in
variations in the M–N bond lengths (Table 1). The M–O
(alkoxide) bonds are also differentiated albeit less than the
M–N bonds—the largest difference observed is about 0.05 A.
The elongated M–N bonds correspond to the three pyridines
trans to alkoxide ligands. The pyridines located below the plane
of the three metals and displaying shorter M–N distances are
roughly trans to the bridging acetates. M–O (alkoxide) bonds
trans to acetates are slightly shorter than those trans to pyridines.
These variations may be caused by the larger trans influences of
pyridine and alkoxide vs. acetate, but distortions caused by steric
strain in the ligand framework cannot be ruled out. Consistent
with the increase in the ionic radius, the metal–ligand distances
increase from Ni to Mn and the M–M distances increase from
3.182(4) A for Ni to 3.415(1)–3.464(1) A for Mn. The ligand
Scheme 1 Synthesis of first row divalent trinuclear metal complexes
with deprotonated H3L. Acetate binding mode is variable (see text).
Department of Chemistry and Chemical Engineering,
California Institute of Technology, 1200 E. California Blvd MC
127-72, Pasadena CA, USA. E-mail: agapie@caltech.edu
w Electronic supplementary information (ESI) available: Synthetic and
crystallographic details. CCDC 777599, 787163 and 803592–803595. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c0cc05608a
z E.Y.T. and J.S.K. contributed equally to this work.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 4189–4191 4189

View Article Online
pseudo-C3-geometry observed in the solid-state. The single peak
assigned to the acetate counterions is indicative of fluxional
processes that exchange the capping ligands on the NMR
time scale.
The magnetism of triangular clusters has been studied in the
context of spin frustration and molecular magnets.9 Although
several alkoxo-bridged NiII and CuII complexes have been
3
3
studied,7d,10 there have been fewer investigations of MnII
,
3
FeII3, and CoII cores. Triangular clusters of manganese
3
and iron more commonly contain higher oxidation state
metal centers.11 The present LM3(OAc)3 family provides an
opportunity to systematically study the magnetic interactions
of several divalent transition metals in a single trinuclear
system, allowing for better understanding of the magneto-
structural characteristics of trinuclear complexes.
Magnetic susceptibility measurements were performed on
powdered crystalline samples of LM3(OAc)3 (M = Mn, Fe,
Co, Ni, Cu) in the temperature range 4–300 K. At room
temperature, the wMT values approach 12.0, 9.0, 6.7, 3.3, and
1.0 cm3 K molꢁ1, respectively (Fig. 2). The difference between
these and the spin-only values may be due to spin–orbit coupling
and population of excited states.12 Upon cooling, the wMT values
decrease gradually and then drop sharply below 40 K, indicating
the presence of antiferromagnetic exchange interactions. With
the exception of the wMT values of LCu3(OAc)3, which approach
a plateau near the expected value for the spin-only S = 1/2 state
(ca. 0.4 cm3 K molꢁ1), the wMT plots do not approach obvious
limiting values at 4 K.
Fig.
1 (a) Solid-state structure of LCo3(OAc)3. Coordination
environments of (b) LFe3(OAc)3 and (c) LMn3(OAc)3 taken from
the solid-state structures. The MII3(m-OR)3 structural motif is in bold;
hydrogen atoms and solvent are not shown for clarity.
To determine the magnitude of exchange between neighboring
metal centers, the magnetic behavior of the compounds
was analyzed using the isotropic spin Hamiltonian [eqn (1)]
considering the two exchange pathways of an isosceles
triangular arrangement. Application of the Van Vleck
equation according to the Kambe vector method13 yields the
magnetic susceptibility equation [eqn (2)].
Table 1 Metal–metal and average metal–nitrogen distances
M–N trans to M–N trans to
Compound
alkoxide (A)
acetate (A)
M–M (A)
LMn3(OAc)3 2.336(3)
2.232(6)
2.150(3)
2.091(2)
2.037(3)
2.081(5)
2.056(7)
3.415(1)–3.464(1)
3.238(6)–3.456(6)
3.228(2)
3.182(4)
3.1822(7)–3.3282(7)
3.1975(9)–3.4199(9)
LFe3(OAc)3 2.232(3)
LCo3(OAc)3 2.213(2)
LNi3(OAc)3 2.127(3)
LCu3(OAc)3 2.027(5)
LZn3(OAc)3 2.229(7)
H = ꢁ2J[(S1S2) + (S2S3)] ꢁ 2J13(S3S1)
(1)
P
ꢀ
ꢁ
NAb2g2
¼
3kT
S0ðS0 þ 1Þð2S0 þ 1ÞOðS0ÞexpðꢁWðS0Þ=kTÞ
P
framework accommodates the different size metals by allowing
for twist around the aryl-aryl bonds and of the C–O vector vs. the
plane of the pendant arene.
wM
ð2S0 þ 1ÞOðS0ÞexpðꢁWðS0Þ=kTÞ
ð2Þ
The Curie–Weiss parameter y was included to account for
Systematically changing the nature of the metal centers
from MnII to ZnII does not disrupt the trinuclear core, but
changes the binding mode of the acetates. Three capping
acetates are present for LM3(OAc)3 (M = Mn–Ni); two
acetates bind in monodentate and one in bidentate fashion.
The bidentate acetate bridges two or three metal centers via a
m2- or m3-oxygen atom. For M = Cu and Zn, single crystal
XRD studies show two acetates bound to the trimetallic core
(see ESIw). However, a third outer sphere acetate required for
charge balance could not be located due to disorder. This
change in coordination mode may be due to the smaller size of
CuII and ZnII hindering the binding of a third acetate.
1H NMR spectroscopy and mass spectrometry studies confirm
that the trinuclear cores of the complexes are maintained in
possible intermolecular interactions.14
1
solution. The H NMR spectra of LM3(OAc)3 (M = Fe–Zn)
display fourteen resonances, with chemical shifts between
ꢁ20 and 160 ppm for the paramagnetic species.8 Thirteen signals
correspond to protons on framework L, consistent with the
Fig. 2 Plots of wMT vs. T. Solid lines show the best fits obtained.
c
4190 Chem. Commun., 2011, 47, 4189–4191
This journal is The Royal Society of Chemistry 2011

View Article Online
Table 2 Magnetic susceptibility parameters17
Notes and references
1 (a) R. H. Holm, P. Kennepohl and E. I. Solomon, Chem. Rev.,
1996, 96, 2239–2314; (b) J. P. McEvoy and G. W. Brudvig, Chem.
Rev., 2006, 106, 4455–4483.
2 (a) M. W. Kanan, J. Yano, Y. Surendranath, M. Dinca,
V. K. Yachandra and D. G. Nocera, J. Am. Chem. Soc., 2010, 132,
13692–13701; (b) M. Risch, V. Khare, I. Zaharieva, L. Gerencser,
P. Chernev and H. Dau, J. Am. Chem. Soc., 2009, 131, 6936–6937.
3 E. Y. Tsui, M. W. Day and T. Agapie, Angew. Chem., Int. Ed.,
2011, 50, 1668–1672.
Compound
J (cmꢁ1
)
g
y (K)
R (ꢂ10ꢁ4
10
1.6
1.9
0.4
12
)
LMn3(OAc)3
LFe3(OAc)3
LCo3(OAc)3
LNi3(OAc)3
LCu3(OAc)3
ꢁ1.1
ꢁ1.4
ꢁ1.2
ꢁ1.2
ꢁ13.7
1.97
1.99
2.30
2.11
2.01
0.53
2.35
0.23
0.74
0.75
4 T. C. Stamatatos, C. G. Efthymiou, C. C. Stoumpos and
S. P. Perlepes, Eur. J. Inorg. Chem., 2009, 3361–3391.
5 G. S. Papaefstathiou, A. Escuer, F. A. Mautner, C. Raptopoulou,
A. Terzis, S. P. Perlepes and R. Vicente, Eur. J. Inorg. Chem., 2005,
879–893.
6 (a) Z. E. Serna, L. Lezama, M. K. Urtiaga, M. I. Arriortua,
´
M. G. Barandika, R. Cortes and T. Rojo, Angew. Chem., Int.
Ed., 2000, 39, 344–347; (b) C. G. Efthymiou, C. P. Raptopoulou,
A. Terzis, R. Boca, M. Korabic, J. Mrozinski, S. P. Perlepes and
E. G. Bakalbassis, Eur. J. Inorg. Chem., 2006, 2236–2252.
The fits were not appreciably improved when modeling two J
values instead of one; as a result, the magnetism data were
simulated for an equilateral triangle arrangement of spins,
corresponding to the approximate C3-symmetry of the
M3(OR)3 cores (without acetates).15 It should be noted that the
modeled parameters approximate the spins of each compound as
isotropic and do not account for the lowered symmetry of each
complex induced by the coordinated acetates. Nevertheless, the
simulated magnetic susceptibility parameters (Table 2) show a
good fit to the experimental data (RB10ꢁ4). In accordance with
the wMT plots, the simulated parameters show that compounds
LM3(OAc)3 display weak antiferromagnetic exchange coupling
(Table 2). Except for LCu3(OAc)3 (J = ꢁ13.7 cmꢁ1), the best fits
were obtained with |J| o 2 cmꢁ1. Although the ground states are
predicted to be S = 0 or S = 1/2 for an equilateral triangle of
antiferromagnetically coupled ions, such small J values indicate
that higher spin states are thermally populated even at low
temperatures.16 For these complexes, the presence of spin
equilibria between these states is consistent with the observation
that no limiting values of wMT are reached at 4 K.
7 (a) B. F. Abrahams, T. A. Hudson and R. Robson, Chem.–Eur. J.,
2006, 12, 7095–7102; (b) R. D. Kohn, M. Haufe, G. Kociok-Kohn
¨
¨
and A. C. Filippou, Inorg. Chem., 1997, 36, 6064–6069;
(c) S. G. Telfer, R. Kuroda, J. Lefebvre and D. B. Leznoff, Inorg.
Chem., 2006, 45, 4592–4601; (d) M. Kodera, Y. Tachi, T. Kita,
H. Kobushi, Y. Sumi, K. Kano, M. Shiro, M. Koikawa, T. Tokii,
M. Ohba and H. Okawa, Inorg. Chem., 2000, 39, 226–234;
(e) E. Labisbal, L. Rodrı
´
J. A. Garcıa-Vazquez, J. Romero, A. Sousa, M. Ya
guez, O. Souto, A. Sousa-Pedrares,
nez,
F. Orallo and J. A. Real, Dalton Trans., 2009, 8644–8656.
´
´
´
8 The resonances in the 1H NMR spectrum of LMn3(OAc)3 were
broadened such that only four resonances were visible.
9 (a) L.-L. Wang, Y.-M. Sun, Z.-Y. Yu, Z.-N. Qi and C.-B. Liu,
J. Phys. Chem. A, 2009, 113, 10534–10539; (b) S. Ferrer, F. Lloret,
I. Bertomeu, G. Alzuet, J. Borra
Gonzalez and J. G. Haasnoot, Inorg. Chem., 2002, 41, 5821–5830.
s, S. Garcıa-Granda, M. Liu-
´ ´
´
10 M. Inoue, C. Ikeda, Y. Kawata, S. Venkatraman, K. Furukawa
and A. Osuka, Angew. Chem., Int. Ed., 2007, 46, 2306–2309.
11 A CSD search for oxygen atom-bridged trinuclear clusters with two
N-donors per metal centers resulted in only one example of a complex
with three MnII centers and no results for three FeII centers. The
remaining search results contained MnIV3 and FeIII3 clusters.
12 R. S. Drago, Physical Methods for Chemists, Surfside Scientific
Publishers, Gainesville, 1992.
Due to the presence of multiple types of bridging ligands, it
is difficult to definitively assign the exchange pathways in these
LM3(OAc)3 complexes.18 Since there are few alkoxo-bridged
trinuclear complexes containing metals other than CuII—and
none with FeII to our knowledge—there is yet no clear
correlation between the J constants and common structural
parameters such as M–M distances or M–O–M angles.19
Previously studied acetate-bridged trinuclear clusters of
divalent metals have been shown to have similar intra-
molecular exchange interactions.20 Alkoxo- and phenoxo-
bridged tricobalt(II), trinickel(II), and triiron(III) clusters all
show small antiferromagnetic coupling.7e,21 While there are no
examples of MnII bridged by alkoxides, amido-bridged22 or
carboxylate-bridged23 MnII clusters also demonstrated anti-
ferromagnetic coupling of magnitudes similar to LMn3(OAc)3.
In summary, the trinucleating ligand described above is a
scaffold capable of supporting different first-row transition
metals in a conserved trinuclear core geometry. These trime-
tallic complexes have been structurally and spectroscopically
characterized. Current efforts are underway to explore
multielectron reactivity and small molecule activation with
these clusters.
13 (a) K. Kambe, J. Phys. Soc. Jpn., 1950, 5, 48–51; (b) E. Sinn,
Coord. Chem. Rev., 1970, 5, 313.
14 V. H. Crawford, H. W. Richardson, J. R. Wasson, D. J. Hodgson
and W. E. Hatfield, Inorg. Chem., 1976, 15, 2107–2110.
15 Magnetic susceptibility parameters were determined by minimizing
P
P
T
T
T
2
R = |(wM
)
ꢁ (wM
)
calcd|2/ (wM
)
obs
.
obs
16 (a) F. E. Mabbs and D. J. Machin, Magnetism and Transition
Metal Complexes, Dover Publications, Inc., Mineola, 2008;
´
(b) S. Haddadpour, H. Niedermeyer, R. Clerac and S. Dehnen,
Dalton Trans., 2009, 8162–8164.
17 For LMn3(OAc)3, the discrepancy between the expected g-value of
2.00 and the fitted value of 1.97 may be accounted for by small
amounts of solvent in the sample detected by 1H NMR spectroscopy.
18 O. Kahn, Molecular Magnetism, VCH Publishers, Inc., New York,
1993.
19 Y.-B. Jiang, H.-Z. Kou, R.-J. Wang, A.-L. Cui and J. Ribas, Inorg.
Chem., 2005, 44, 709–715.
20 R. A. Reynolds, W. O. Yu, W. R. Dunham and D. Coucouvanis,
Inorg. Chem., 1996, 35, 2721–2722.
21 (a) H. Adams, D. E. Fenton, L. R. Cummings, P. E. McHugh,
M. Ohba, H. Okawa, H. Sakiyama and T. Shiga, Inorg. Chim.
Acta, 2004, 357, 3648–3656; (b) C. Boskovic, E. Rusanov,
We thank Lawrence M. Henling for crystallographic
assistance. We are grateful to Caltech, Chemistry Department
Dow Fellowship (JSK), and NSF GRFP (EYT) for funding.
The Bruker KAPPA APEXII X-ray diffractometer was
purchased via an NSF CRIF:MU award to Caltech,
CHE-0639094. SQUID data were collected at the MMRC of
the Beckman Institute of the California Institute of Technology.
H. Stoeckli-Evans and H. U. Gudel, Inorg. Chem. Commun.,
2002, 5, 881–886.
¨
22 J. A. Hatnean, R. Raturi, J. Lefebvre, D. B. Leznoff, G. Lawes and
S. A. Johnson, J. Am. Chem. Soc., 2006, 128, 14992–14999.
23 P. Christian, G. Rajaraman, A. Harrison, M. Helliwell,
J. J. W. McDouall, J. Raftery and R. E. P. Winpenny, Dalton
Trans., 2004, 2550–2555.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 4189–4191 4191