An azophenine radical-bridged Fe2 single-molecule magnet with record magnetic exchange coupling

Article abstract of DOI:10.1021/ja409927v

One-electron reduction of the complex [(TPyA)₂Fe ^II₂(^NPhL^2-)]²⁺ (TPyA = tris(2-pyridylmethyl)amine, ^NPhLH₂ = azophenine = N,N′,N″,Na€-tetraphenyl-2,5-diamino-1, 4-diim

Full text of DOI:10.1021/ja409927v

Communication
pubs.acs.org/JACS
An Azophenine Radical-Bridged Fe2 Single-Molecule Magnet with
Record Magnetic Exchange Coupling
Ie-Rang Jeon,† Jesse G. Park,† Dianne J. Xiao,‡ and T. David Harris*,†
†Department of Chemistry, Northwestern University, Evanston Illinois 60208-3113, United States
‡Department of Chemistry, University of California, Berkeley, California 94720-1460, United States
S
* Supporting Information
one another only weakly through diamagnetic bridging ligands
ABSTRACT: One-electron reduction of the complex
via a superexchange mechanism. As an alternative, incorporation
of a paramagnetic bridging ligand between metal centers enables
strong direct metal-radical exchange and thus much larger values
of J. Despite this potential, very few single-molecule magnets
have incorporated radical bridging ligands.1h,5,6
[(TPyA)2FeII (NPhL2−)]2+ (TPyA = tris(2-pyridylmethyl)-
2
amine, NPhLH2 = azophenine = N,N′,N″,N‴-tetraphenyl-
2,5-diamino-1,4-diiminobenzoquinone) affords the com-
plex [(TPyA)2FeII (NPhL3−•)]+. X-ray diffraction and
2
Mossbauer spectroscopy confirm that the reduction occurs
̈
Benzoquinonoid bridging ligands offer an ideal platform for
the construction of radical-bridged single-molecule magnets with
strong magnetic coupling, as these ligands readily undergo redox
chemistry to form both diamagnetic and paramagnetic electron-
transfer isomers (see Figure 1). Indeed, a number of dinuclear
complexes featuring a radical form of 1,2,4,5-tetraoxolene
(OLH2) and its derivatives have been shown to exhibit strong
metal-radical interactions,7 in some cases so strong that the spin
ground state remains isolated from excited states even at 300 K.8
As an extension, a radical form of the related azophenine
on NPhL2− to give an S = 1/2 radical bridging ligand. Dc
magnetic susceptibility measurements demonstrate the
presence of extremely strong direct antiferromagnetic
exchange between S = 2 FeII centers and NPhL3−• in the
reduced complex, giving an S = 7/2 ground state with an
estimated coupling constant magnitude of |J| ≥ 900 cm−1.
Mossbauer spectroscopy and ac magnetic susceptibility
̈
reveal that this complex behaves as a single-molecule
magnet with a spin relaxation barrier of Ueff = 50(1) cm−1.
To our knowledge, this complex exhibits by far the
strongest magnetic exchange coupling ever to be observed
in a single-molecule magnet.
(
NRLH2) should afford even stronger coupling, owing to the
more diffuse orbitals of nitrogen compared to oxygen. In support
of this hypothesis, J has been shown to increase by 32-fold in
moving from a tetraoxolene to an azophenine bridge in Cu2
9
complexes. Moreover, a value of J = −946 cm−1 (H = −2J(SA·
̂
̂
̂
SB)) has been calculated using DFT for a mononuclear nickel(II)
ver the past two decades, a number of molecules, known as
Osingle-molecule magnets, have been shown to retain a net
magnetization upon removal of an applied magnetic field, similar
to the behavior exhibited by classical permanent magnets.1 This
slow magnetic relaxation arises due to a uniaxial magnetic
anisotropy, quantified by the axial zero-field splitting parameter
(D), which breaks the degeneracy of the MS levels comprising the
spin ground state (S) of the molecule. The discovery of single-
molecule magnets has generated much interest and excitement,
as these species could find use in applications such as high-
density spin-based information storage and processing,2
quantum computing,3 and spintronics.4
With the goal of maximizing the relaxation barrier (Ueff) in
single-molecule magnets, a tremendous research effort has
focused on synthesizing molecules with large values of both S and
D. However, as values of Ueff continue to climb, a third critical
parameter must also be considered: the strength of magnetic
exchange between paramagnetic centers, which is quantified by
the exchange coupling constant J. Indeed, the separation between
spin ground state and excited states is directly correlated to the
magnitude of J. When J is insufficiently large, low-lying spin
excited states enable fast relaxation pathways that can shortcut
the overall thermal barrier. As such, the development of single-
molecule magnets with large values of J represents an important
challenge. Nevertheless, the vast majority of single-molecule
magnets feature multiple paramagnetic metal centers coupled to
bis(o-phenylenediamine radical) complex.10 Providing an addi-
tional advantage over tetraoxolene analogues, derivatives of
NRLH2 with bulky aryl substituents (R) bound to the nitrogen
donors can be readily synthesized through palladium-catalyzed
cross-coupling reactions, thereby enabling fine control over
intermolecular metal-metal separation in complexes.11 Herein,
we report the synthesis and magnetic analysis of an azophenine
radical-bridged Fe2 complex that, to the best of our knowledge,
exhibits by far the strongest magnetic exchange coupling yet
observed in a single-molecule magnet.
Reaction of anhydrous Fe(SO3CF3)2 with TPyA in MeCN,
followed by treatment with a mixture of NPhLH2 and Li[N-
(SiMe3)2], afforded a dark-brown solution.12 Subsequent
diffusion of diethyl ether vapor into this solution yielded dark-
yellow, needle-shaped crystals of [(TPyA)2FeII (NPhL2−)]-
2
( S O 3 C F 3 ) 2 · 2 M e C N ( 1 ) . T h e s t r u c t u r e o f
[(TPyA)2FeII (NPhL2−)]2+ (see Figure 2, upper) consists of two
2
crystallographically equivalent [(TPyA)FeII]2+ units connected
by a deprotonated NPhL2− bridging ligand. Each FeII center
resides in a distorted octahedral coordination environment, with
two cis-disposed sites occupied by neighboring nitrogen donor
atoms from NPhL2− and the other four bound by nitrogen atoms
Received: September 25, 2013
Published: October 28, 2013
© 2013 American Chemical Society
16845
dx.doi.org/10.1021/ja409927v | J. Am. Chem. Soc. 2013, 135, 16845−16848

Journal of the American Chemical Society
Communication
this solution at −35 °C afforded the one-electron-reduced
compound [(TPyA)2Fe2(NPhL)](SO3CF3)·3DMF·0.3Na-
(SO3CF3) (2) as a dark-blue microcrystalline powder in 79%
yield. Alternatively, carrying out the above reaction in MeCN
using the reductant (C5Me5)2Co,15 followed by diffusion of
diethyl ether vapor into the resulting solution at −35 °C, yielded
a product mixture containing crystals of 1 and dark-blue, plate-
shaped crystals of [(TPyA)2Fe2(NPhL)](SO3CF3)·6MeCN (2′;
see Figure 2, lower).
Figure 1. Redox series of deprotonated benzoquinonoid ligands. Left to
right: EL4−, EL3−•, EL2− (E = O and NR).
While the cationic complex in 2′ exhibits an overall structure
similar to that in 1, close comparison of the bond lengths in the
two complexes reveals several key features (see Table 1). First,
the average benzoquinone C−C distance decreases slightly by
0.9%, from 1.431(5) to 1.418(5) Å in moving from 1 to 2′, while
the average benzoquinone C−N distance increases by 2.3%, from
1.331(4) to 1.361(4) Å. These differences highlight a net
increase in C−C bond order and net decrease in C−N bond
order, consistent with the additional electron occupying a
molecular orbital of primarily ligand character. In further support
of this inference, the average Fe−NL distance decreases by 1.8%,
from 2.110(3) to 2.073(3) Å, consistent with a stronger Fe−N
interaction resulting from a trianionic bridging ligand. These
structural changes are similar to those previously observed for a
ligand-centered reduction in a chloranilate radical-bridged Co2
complex.7bc Finally, the average Fe−NTPyA distance changes only
slightly, increasing from 2.233(3) to 2.249(3) Å (0.7%),
supporting a ligand-centered redox event. Taken together,
these observations suggest a configuration for the complex in 2
Figure 2. Reduction of [(TPyA)2FeII (NPhL2−)]2+ (upper) to give
2
[(TPyA)2FeII (NPhL3−•)]+ (lower). Orange, blue, and gray ellipsoids,
2
shown at the 50% probability level, represent Fe, N, and C atoms,
respectively; H atoms are omitted for clarity.
and 2′ of [(TPyA)2FeII (NPhL3−•)]+. To our knowledge, 2′
2
represents the first structurally characterized example of a
molecule incorporating a radical bridging ligand derived from
azophenine.
from the TPyA capping ligand. The average Fe−N distance of
2.110(3) Å is consistent with a high-spin S = 2 electronic
configuration. The complex features an intramolecular Fe···Fe
distance of 8.154(1) Å and a closest intermolecular Fe···Fe
distance of 9.999(1) Å.
The cyclic voltammogram of an acetonitrile solution
containing 1, shown in Figure 3, reveals the presence of three
reversible redox processes, centered at +0.32, −0.19, and
−1.65 V vs [Cp2Fe]0/1+. Considering precedent in tetraox-
olene-bridged Fe2 complexes, we tentatively assign the first two
processes to metal-centered events and the third process to a
ligand-based event.7,8,13 Indeed, the process at −1.65 V suggests
that a [(TPyA)2Fe2(NPhL)]+ complex should be chemically
accessible. Toward that end, a solution of 1 in DMF was treated
at −35 °C with the strong reductant Na(C10H8)14 to give an
intense blue solution.12 Subsequent addition of diethyl ether into
In order to confirm the assignment of a primarily ligand-based
reduction, Mossbauer spectra were collected for 1 and 2. The
̈
Mossbauer spectrum collected on a solid sample of 1 at 77 K (see
̈
Figure 4, left) exhibits a major symmetric quadrupole doublet
and a second minor doublet assigned to a small amount of FeIII-
containing impurity. A fit to the data provides an isomer shift of δ
= 1.027(3) mm/s and a quadrupole splitting of ΔEQ = 2.775(5)
mm/s for the major doublet, consistent with two equivalent high-
spin FeII ions. The spectrum of 2 exhibits a similar quadrupole
doublet, with an identical isomer shift of δ = 1.024(4) mm/s and
a considerably larger quadrupole splitting of ΔEQ
=
3.119(7) mm/s. The identical isomer shifts in 1 and 2 provide
strong evidence that the reduction is indeed ligand centered, and
the larger quadrupole splitting in 2 may stem from the change in
ligand field at the FeII center associated with ligand reduction and
the greater distortion from an octahedral coordination environ-
ment at iron in 2 compared to 1. Finally, as temperature is
a
Table 1. Selected Bond Distances (Å) in 1 and 2′
1
2′
1
2′
C1−C2
C2−C3
C3−C1A
C−C
N1−C2
N2−C3
N−C
1.410(5)
1.498(5)
1.385(4)
1.431(5)
1.318(4)
1.344(4)
1.331(4)
1.411(5)
1.461(6)
1.381(4)
1.418(5)
1.350(4)
1.372(4)
1.361(4)
Fe−N1
Fe−N2
Fe−NL
Fe−N3
Fe−N4
Fe−N5
Fe−N6
2.142(3)
2.078(2)
2.110(3)
2.297(2)
2.172(3)
2.159(3)
2.305(3)
2.103(3)
2.042(3)
2.073(3)
2.278(3)
2.314(3)
2.177(4)
2.226(3)
2.249(3)
Fe−NTPyA 2.233(3)
Figure 3. Cyclic voltammogram obtained for an acetonitrile solution of
1 using a platinum electrode, 50 mV/s scan rate, and 0.1 M (Bu4N)PF6
supporting electrolyte.
a
Average distances are shown in bold.
16846
dx.doi.org/10.1021/ja409927v | J. Am. Chem. Soc. 2013, 135, 16845−16848

Journal of the American Chemical Society
Communication
magnetically isolated S = 2 FeII centers and one S =1/2 NPhL3−•
.
Instead, this value is in agreement with an S = 7/2 ground state (g
= 2.12) resulting from exceptionally strong antiferromagnetic
exchange between FeII centers and NPhL3−• radical. Indeed, the
invariance of χMT in the temperature range 130−300 K indicates
that the S = 7/2 ground state remains well isolated from excited
states even up to 300 K. While the temperature independence of
χMT precludes a definitive quantitation of J, a lower bound of the
magnitude can be estimated. The inset of Figure 5 shows an
expanded view of the high-temperature data, along with
simulations of the data for selected values of J according to the
spin Hamiltonian H = −2J[Srad·(SFe1 + SFe2)].19 These
simulations provide a value of J ≤ −900 cm−1 for 2. This value
should be regarded as an estimate, as subtle changes to the
diamagnetic correction of the data and/or introduction of
additional parameters to the simulation, such as zero-field
splitting or intra/intermolecular Fe···Fe interactions, can lead to
significant variation of J. Nevertheless, this estimated magnitude
of J is nearly 50× larger than the J = +19 cm−1 observed in a
̂
̂
̂
̂
Figure 4. Left: Zero-field 57Fe Mossbauer spectra for 1 (upper) and 2
̈
(lower), collected at 77 K. Black circles correspond to experimental data,
and solid lines correspond to fits of the data. Right: Variable-
temperature spectra for 2, revealing the onset of slow magnetic
relaxation with lowering temperature.
chloranilate radical-bridged FeII complex with ferromagnetic
2
decreased, the doublet in 2 undergoes spectral broadening,
ultimately reaching an unresolved multiplet at 5 K (see Figure 4,
right). This phenomenon indicates the onset of slow magnetic
relaxation with decreasing temperature, suggesting that 2 exhibits
single-molecule magnet behavior at low temperature on the
metal-ligand radical exchange.7c Most importantly, to the best of
our knowledge, this value is by far the largest ever observed in a
single-molecule magnet, eclipsing the previous record of J =
−133 cm−1, held by a recently reported nindigo radical-bridged
CoII2 complex.6b Finally, we note that a similar energy separation
from excited states has been observed for the S = 9/2 ground
state of a mixed-valence [Fe2]V complex, however this behavior
arises through a double-exchange mechanism due to electron
delocalization.17
Mossbauer time scale.16,17
̈
To probe and compare magnetic interactions in 1 and 2,
variable-temperature dc magnetic susceptibility data were
collected for solid samples under an applied field of 1000 Oe.
The resulting plot of χMT vs T for 1 is shown in Figure 5. At
300 K, χMT = 6.38 cm3 K/mol, corresponding to two
magnetically isolated high-spin FeII centers with g = 2.06. As
temperature is decreased, the data undergo a gradual then rapid
decline, reaching a minimum value of 0.14 cm3 K/mol at 2 K that
corresponds to an S = 0 ground state. The decrease in χMT with
decreasing temperature is indicative of weak antiferromagnetic
superexchange coupling between high-spin FeII centers through
the diamagnetic NPhL2− ligand. To quantify this interaction, the
data were fit to the Van Vleck equation according to the spin
In order to assess the presence of magnetic anisotropy in 2,
low-temperature magnetization data were collected at selected dc
fields (see Figure 6, lower right). The splitting of the resulting
isofield curves, along with their saturation well below the
expected M = 7 μB for an S = 7/2 ground state with g = 2,
demonstrates qualitatively the presence of significant magnetic
anisotropy. To quantify this effect, the data were fit using the
program ANISOFIT 2.020,21 to give parameters of D =
−8.4 cm−1, |E| = 0.6 cm−1, and g = 2.11. To our knowledge,
this value of D is the largest yet observed in a multinuclear single-
̂
̂
̂
Hamiltonian H = −2J(SFe1·SFe2) to give an exchange constant of J
= −2.90(2) cm−1.18
molecule magnet, surpassing a cyano-bridged CuII FeIII2 cluster
3
The plot of χMT vs T obtained for 2 (see Figure 5) shows a
dramatically different profile than that for 1. At 300 K, χMT =
8.84 cm3 K/mol, much higher than the value expected for two
with an S = 5/2 ground state and D = −5.7 cm−1,22 and likely
stems from a combination of high-anisotropy high-spin FeII
centers and relatively small spin ground state. This value of D
Figure 5. Variable-temperature dc magnetic susceptibility data for 1
(blue) and 2 (red), collected under an applied field of 1000 Oe. The
black line corresponds to a fit of the data. Inset: Expanded view of data
for 2; lines correspond to simulations with selected values of −J (cm−1).
Figure 6. Left: Variable-frequency ac susceptibility data for 2. Upper
right: Arrhenius plot of relaxation time, with Ueff = 50(1) cm−1. Lower
right: Low-temperature magnetization data for 2 at selected fields.
16847
dx.doi.org/10.1021/ja409927v | J. Am. Chem. Soc. 2013, 135, 16845−16848

Journal of the American Chemical Society
Communication
implies that 2 could behave as a single-molecule magnet with a
maximum relaxation barrier of U = (S2 − 1/4)|D| = 101 cm−1.
Following the observation of slow magnetic relaxation from
REFERENCES
■
(1) (a) Caneschi, A.; Gatteschi, D.; Sessoli, R.; Barra, A. L.; Brunel, L.
C.; Guillot, M. J. Am. Chem. Soc. 1991, 113, 5873. (b) Sessoli, R.; Tsai,
H.-L.; Schake, A. R.; Wang, J. B.; Folting, K.; Gatteschi, D.; Christou, G.;
Hendrickson, D. N. J. Am. Chem. Soc. 1993, 115, 1804. (c) Sessoli, R.;
Gatteschi, D.; Caneschi, A.; Novak, M. A. Nature 1993, 365, 141.
variable-temperature Mossbauer spectroscopy and presence of a
̈
large negative D, variable-frequency ac susceptibility data under
zero applied dc field were collected in order to further probe
single-molecule magnet behavior. The frequency-dependent out-
of-phase component (χM″) of the ac susceptibility (see Figure 6,
left) demonstrates that 2 is indeed a single-molecule magnet. The
corresponding Arrhenius plot of relaxation time (see Figure 6,
upper right) exhibits a linear region at high temperature,
indicative of a thermally activated relaxation process (see SI). A
fit to the data in the temperature range 5.0−6.8 K provides a
relaxation barrier of Ueff = 50(1) cm−1, with τ0 = 2.7(2) × 10−10 s.
Considering the value of D = −8.4 cm−1 obtained from
magnetization data, this relaxation barrier is in exact agreement
with the energetic separation between ground MS = 7/2 levels
and first-excited MS = 5/2 levels, given as 6|D| = 51 cm−1. This
observation suggests that thermally assisted quantum tunneling
́ ́
(d) Berlinguette, C. P.; Vaughn, D.; Canada-Vilalta, C.; Galan-Mascaros,
̃
J. R.; Dunbar, K. R. Angew. Chem., Int. Ed. 2003, 42, 1523. (e) Ishikawa,
N.; Sugita, M.; Ishikawa, T.; Koshihara, S.-y.; Kaizu, Y. J. Am. Chem. Soc.
2003, 125, 8694. (f) Schelter, E. J.; Prosvirin, A. V.; Dunbar, K. R. J. Am.
Chem. Soc. 2004, 126, 15004. (g) Freedman, D. E.; Harman, W. H.;
Harris, T. D.; Long, G. J.; Chang, C. J.; Long, J. R. J. Am. Chem. Soc. 2010,
132, 1224. (h) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. Nat.
Chem. 2011, 3, 538.
(2) (a) Bogani, L.; Wernsdorfer, W. Nat. Mater. 2008, 7, 179.
(b) Ardavan, A.; Blundell, S. J. J. Mater. Chem. 2009, 19, 1754.
(3) (a) Leuenberger, M. N.; Loss, D. Nature 2001, 410, 789. (b) Stamp,
P. C. E.; Gaita-Arino, A. J. Mater. Chem. 2009, 19, 1718.
(4) Rocha, A. R.; García-suarez, V. M.; Bailey, S. W.; Lambert, C. J.;
́
Ferrer, J.; Sanvito, S. Nat. Mater. 2005, 4, 335.
(5) Lanthanide complexes: (a) Poneti, G.; Bernot, K.; Bogani, L.;
Caneschi, A.; Sessoli, R.; Wernsdorfer, W.; Gatteschi, D. Chem.
Commun. 2007, 1807. (b) Xu, J.-X.; Ma, Y.; Liao, D.-Z.; Xu, G.-F.;
Tang, J.; Wang, C.; Zhou, N.; Yan, S.-P.; Cheng, P.; Li, L.-C. Inorg. Chem.
2009, 48, 8890. (c) Tian, H.-X.; Liu, R.-N.; Wang, X.-L.; Yang, P.-P.; Li,
Z.-X.; Li, L.-C.; Liao, D.-Z. Eur. J. Inorg. Chem. 2009, 4498. (d) Rinehart,
J. D.; Fang, M.; Evans, W. J.; Long, J. R. J. Am. Chem. Soc. 2011, 133,
14236. (e) Demir, S.; Zadrozny, J. M.; Nippe, M.; Long, J. R. J. Am.
Chem. Soc. 2012, 134, 18546. (f) Liu, R.; Zhang, C.; Li, L.; Liao, D.;
Sutter, J.-P. Dalton. Trans. 2012, 41, 12139.
(6) Transition-metal complexes: (a) Yoshihara, D.; Karasawa, S.; Koga,
N. J. Am. Chem. Soc. 2008, 130, 10460. (b) Fortier, S.; Le Roy, J. J.; Chen,
C.-H.; Vieru, V.; Murugesu, M.; Chibotaru, L. F.; Mindiola, D. J.;
Caulton, K. G. J. Am. Chem. Soc. 2013, 135, 14670.
between MS =
5/2 levels acts as the dominant relaxation
pathway for 2 in this temperature range. As temperature is further
decreased, the data begin to deviate from linearity, denoting the
presence of additional fast relaxation processes, such as quantum
tunneling and/or spin−spin relaxation, that shortcut the barrier.
The foregoing results demonstrate that a radical form of
azophenine can be employed as a bridging ligand to synthesize
single-molecule magnets with magnetic exchange coupling of
unprecedented strength. Indeed, the one-electron reduction of a
[FeII (NPhL2−)]2+ complex gives a [FeII (NPhL3−•)]+ complex that
2
2
behaves as a single-molecule magnet with a spin ground state of S
= 7/2 and a relaxation barrier of Ueff = 50(1) cm−1. The metal-
ligand radical exchange coupling in this complex, with an
estimated magnitude of |J| ≥ 900 cm−1, is to our knowledge by far
the strongest magnetic exchange yet observed in a single-
molecule magnet. Work is underway to incorporate metal centers
with stronger magnetic anisotropy and more radially diffuse d-
orbitals to this system, append substituents onto NPhLH2 in order
to modulate ligand-centered redox processes, and synthesize
related bridging ligands with even more diffuse donor atom
orbitals.
(7) (a) Dei, A.; Gatteschi, D.; Pardi, L.; Russo, U. Inorg. Chem. 1991,
30, 2589. (b) Min, K. S.; Rheingold, A. L.; DiPasquale, A. G.; Miller, J. S.
Inorg. Chem. 2006, 45, 6135. (c) Min, K. S.; DiPasquale, A. G.; Golen, J.
A.; Rheingold, A. L.; Miller, J. S. J. Am. Chem. Soc. 2007, 129, 2360.
(8) Guo, D.; McCusker, J. K. Inorg. Chem. 2007, 46, 3257.
(9) Schweinfurth, D.; Khusniyarov, M. M.; Bubrin, D.; Hohloch, S.; Su,
C.-Y.; Sarkar, B. Inorg. Chem. 2013, 52, 10332.
(10) Chłopek, K.; Bothe, E.; Neese, F.; Weyhermuller, T.; Wieghardt,
̈
K. Inorg. Chem. 2006, 45, 6298.
(11) (a) Wenderski, T.; Light, K. M.; Ogrin, D.; Bott, S. G.; Harlan, C.
J. Tetrahedron Lett. 2004, 45, 6851. (b) Khramov, D. M.; Boydston, A. J.;
Bielawski, C. W. Org. Lett. 2006, 8, 1831.
(12) See Supporting Information for absorption spectra of 1 and 2.
(13) Min, K. S.; DiPasquale, A. G.; Rheingold, A. L.; White, H. S.;
Miller, J. S. J. Am. Chem. Soc. 2009, 131, 6229.
(14) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877.
(15) Gennett, T.; Milner, D. F.; Weaver, M. J. J. Phys. Chem. 1985, 89,
2787.
(16) Abbas, G.; Lan, Y.; Mereacre, V.; Wernsdorfer, W.; Clerac, R.;
́
Buth, G.; Sougrati, M. T.; Grandjean, F.; Long, G. J.; Anson, C. E.;
Powell, A. K. Inorg. Chem. 2009, 48, 9345.
(17) Hazra, S.; Sasmal, S.; Fleck, M.; Grandjean, F.; Sougrati, M. T.;
Ghosh, M.; Harris, T. D.; Bonville, P.; Long, G. J.; Mohanta, S. J. Chem.
Phys. 2011, 134, 174507/1.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental details and characterization data. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
dharris@northwestern.edu
Notes
The authors declare no competing financial interest.
̌
(18) Boca, R. Theoretical Foundations of Molecular Magnetism; Elsevier:
Amsterdam, 1999.
ACKNOWLEDGMENTS
(19) Susceptibility data were simulated using the program
■
́
MAGPACK: Borras-Almenar, J. J.; Clemente-Juan, J. M.; Coronado,
Research was funded by Northwestern University, the Interna-
tional Institute for Nanotechnology, and the Initiative for
Sustainability and Energy at Northwestern (Booster Award
E.; Tsukerblat, B. S. J. Comput. Chem. 2001, 22, 985.
(20) Shores, M. P.; Sokol, J. J.; Long, J. R. J. Am. Chem. Soc. 2002, 124,
2279.
10033416). We thank Prof. C. J. Chang for use of his Mossbauer
̈
(21) Data were fit using the following zero-field splitting Hamiltonian:
2
2
2
̂
̂
̂
̂
spectrometer, Prof. E. A. Weiss for use of her UV/vis/near-IR
spectrometer, and K. Du, J. Martinez, V. L. Weidner, and G. Yang
for experimental assistance.
H = DSz + E(Sx − Sy ) + gisoμBS·H.
(22) Wang, C.-F; Zuo, J.-L.; Bartlett, B. M.; Song, Y.; Long, J. R.; You,
X.-Z. J. Am. Chem. Soc. 2006, 128, 7162.
16848
dx.doi.org/10.1021/ja409927v | J. Am. Chem. Soc. 2013, 135, 16845−16848