Magnetic properties of 1:2 mixed cobalt(II) salicylaldehyde schiff-base complexes with pyridine ligands carrying high-spin carbenes (S car = 2/2, 4/2, 6/2, and 8/2) in dilute frozen solutions: Role of organic spin in heterospin single-molecule magnets

Article abstract of DOI:10.1021/ic403074d

The 1:2 mixtures of Co(p-tolsal)₂, p-tolsal = N-p-tolylsalicylideniminato, and diazo-pyridine ligands, DXpy; X = 1, 2, 3l, 3b, and 4, in MTHF solutions were irradiated at cryogenic temperature to form the corresponding 1:2 cobalt-carbene complexes Co(p-tolsal)₂(CXpy) ₂, with S_total = 5/2, 9/2, 13/2, 13/2, and 17/2, respectively. The resulting Co(p-tolsal)₂(CXpy)₂, X = 1, 2, 3l, 3b, and 4, showed magnetic behaviors characteristic of heterospin single-molecule magnets with effective activation barriers, U _eff/k_B, of 40, 65, 73, 72, and 74 K, for reorientation of the magnetic moment and temperature-dependent hysteresis loops with a coercive force, H_c, of ～0, 6.2, 10, 6.5, and 9.0 kOe at 1.9 K, respectively. The relaxation times, τ_Q, due to a quantum tunneling of magnetization (QTM) were estimated to be 1.6 s for Co(p-tolsal)₂(C1py)₂, ～2.0 × 10³ s for Co(p-tolsal)₂(C2py)₂, and >10⁵ s for Co(p-tolsal)₂(CXpy)₂; X = 3b, 3l, and 4. In heterospin complexes, organic spins, carbenes interacted with the cobalt ion to suppress the QTM pathway, and the τ_Q value increased with increasing the S_total values.

Full text of DOI:10.1021/ic403074d

Article
pubs.acs.org/IC
Magnetic Properties of 1:2 Mixed Cobalt(II) Salicylaldehyde Schiff-
Base Complexes with Pyridine Ligands Carrying High-Spin Carbenes
(Scar = 2/2, 4/2, 6/2, and 8/2) in Dilute Frozen Solutions: Role of
Organic Spin in Heterospin Single-Molecule Magnets
Satoru Karasawa,† Kimihiro Nakano,† Daisuke Yoshihara,† Noriko Yamamoto,† Jun-ichi Tanokashira,†
Takahito Yoshizaki,† Yuji Inagaki,‡ and Noboru Koga*,†
†Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka, 812-8582 Japan
‡Department of Applied Quantum Physics, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka, 812-0395 Japan
S
* Supporting Information
ABSTRACT: The 1:2 mixtures of Co(p-tolsal)2, p-tolsal = N-p-tolylsalicylideniminato,
and diazo-pyridine ligands, DXpy; X = 1, 2, 3l, 3b, and 4, in MTHF solutions were
irradiated at cryogenic temperature to form the corresponding 1:2 cobalt−carbene
complexes Co(p-tolsal)2(CXpy)2, with Stotal = 5/2, 9/2, 13/2, 13/2, and 17/2,
respectively. The resulting Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, showed
magnetic behaviors characteristic of heterospin single-molecule magnets with effective
activation barriers, Ueff/kB, of 40, 65, 73, 72, and 74 K, for reorientation of the magnetic
moment and temperature-dependent hysteresis loops with a coercive force, Hc, of ∼0,
6.2, 10, 6.5, and 9.0 kOe at 1.9 K, respectively. The relaxation times, τQ, due to a
quantum tunneling of magnetization (QTM) were estimated to be 1.6 s for Co(p-
tolsal)2(C1py)2, ∼2.0 × 103 s for Co(p-tolsal)2(C2py)2, and >105 s for Co(p-
tolsal)2(CXpy)2; X = 3b, 3l, and 4. In heterospin complexes, organic spins, carbenes
interacted with the cobalt ion to suppress the QTM pathway, and the τQ value increased with increasing the Stotal values.
reported heterospin SMMs taking advantage of the magnetic
coupling between the anisotropic metal ion and organic
radicals.10 In our heterospin systems, the combination of a
high-spin cobalt(II) ion and the pyridine-carbene11 and
-aminoxyl11e,12 ligands provided unique heterospin SMMs
having slow magnetic relaxation with a large Ueff/kB and
showing the hysteresis loop of magnetization, M, with a large
coercive force (Hc) above 1.9 K. The values of Ueff/kB and Hc at
1.9 K for heterospin SMMs, especially cobalt−carbene
complexes, were considerably large compared with those for
metal clusters containing 3d metals reported as SMMs up to
now. The schematic structures designed according to the
strategy of our heterospin SMM are shown in Scheme 1.
INTRODUCTION
■
Single-molecule magnets (SMMs)1−6 exhibiting hysteresis
loops of magnetization similar to bulk magnets and the spin
quantum tunneling effect at low temperatures have been
attracting the interest of many chemists in the fields of both
new functional materials and fundamental chemistry. In an
SMM, the characteristic slow magnetic relaxation for
reorientation of the magnetic moment takes place via various
pathways due to the direct,1a,7 Orbach,1a,7 Raman,1a,7 and
quantum tunneling of the magnetization (QTM)2−6 process.
The thermodynamic activation barrier, U/kB, corresponds to |
D|S2 and |D|(S2 − 1/4) for integer spin and noninteger spin,
respectively, where |D| (D < 0) is a uniaxial anisotropic
parameter and S is a spin quantum number. Therefore, the
effective activation barriers (Ueff/kB: Ueff/kB < U/kB) depend on
the U/kB value and the relaxation pathways, especially QTM
relaxation time, τQ. To obtain a large Ueff/kB in SMMs, an
increasing U/kB value and/or a suppressing QTM pathway
would be required.
Notably, the monometallic 1:4 cobalt complexes Co-
11a,b,e−g
(NCO)2(C1py)4
and the cyclic 2:2 cobalt complex-
es11c,d [Co(I-hfpip)2(C2py2)]2, I-hfpip = 1,1,1,5,5,5-hexafluoro-
4-(4-iodophenylimino)-2-pentanonate, showed SMM behaviors
with extra large values of Ueff/kB = 130 and 139 K and Hc = 20
and 26 kOe at 1.9 K, respectively, in addition to the
temperature-independent QTM relaxation times (τQ) of 1.0
× 105 and 1.1 × 105 s, respectively.
For the construction of SMMs, we used a heterospin
system8,9 consisting of the 3d spins of the anisotropic metal
ions and the 2p spins of the organic spins. In our heterospin
systems, carbene9b,d−f and a stable aminoxyl9a,c were used as
organic spins, and they magnetically interacted with 3d metal
ions through aromatic ligands by the large delocalization of π-
spin of organic unpaired electrons. Recently, other groups also
The cobalt complexes for heterospin SMMs had similar
compressed or elongated octahedral structures.11a,b In those
Received: December 15, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
ation of the mixed solution of Co(p-tolsal)2 and cyclic
pentadiazopyridine, cD5py, has already been reported to
function as an SMM with Stotal = 21/2, Ueff/kB = 72 K, and
Hc = 7.1 kOe in frozen solution.11f
Scheme 1. Schematic Structures of Heterospin SMMs
We report here the relationship between the SMM magnetic
properties and the ground-state spin multiplicity in a series of
Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, together with
Co(p-tolsal)2(cC5py)2. In addition, the magnetic properties of
Co(p-tolsal)2(py)2 having no carbene in the absence and
presence of a dc field were also reported. In the 1:4 complex
CoCl2(py)4,11a the dynamic magnetic behavior showing slow
magnetic relaxation with Ueff/kB = 16 K in the presence of a 3
kOe dc field was observed.
complexes, the z axis of the axial ligands might be anisotropic
axes, and the organic spins were located in the x−y plane of the
cobalt complex. To understand heterospin SMMs, the effects of
an axial ligand and organic spin (Y and S, respectively, in
Scheme 1) in the cobalt complexes on the heterospin SMM
property were systematically investigated in frozen solution. In
the 1:4 cobalt−aminoxyl and −carbene complexes,11b,12b,c
Co(Y)2(4NOpy)4 and Co(Y)2(C1py)4, Y = NCS−, Cl− (Br−
for 4NOpy complex), and NCO−, respectively, the axial ligand
(ion) was revealed to strongly affect the Ueff/kB (or U/kB) for
SMMs, and the Ueff/kB values increased in the order Y = NCS−,
Cl− (Br−), and NCO−. This result might be due to the increase
of the |D| value caused by the change of axial ligand. In the
heterospin SMM, on the other hand, the role of organic spin
located in the x−y plane of the cobalt complex has not yet been
sufficiently clarified. However, according to the experimental
facts before irradiation, the diazo-Co complex showed no SMM
behavior, while after irradiation, the carbene-Co complexes
generated by photolysis of the diazo moieties showed SMM
behavior, obviously indicating that the carbenes affected the
SMM properties of carbene−Co complexes. Furthermore, since
the U/kB value of cobalt complexes corresponded to |D|(S2 −
1/4) for noninteger spin, the increase of the S value in the
ground state by using triplet carbene spins would be expected
to lead to a large U/kB value. Recently, the 1:4 complexes of
CoCl2 and DXpy, X = 1, 2, 3l, 3b, and 4, which were precursors
of high-spin carbene−pyridine ligands with spin multiplicities,
Scar, of 2/2, 4/2, 6/2, 6/2, and 8/2, respectively, were employed
to change the Scar of carbenes systematically, and the influence
of the carbenes on heterospin SMM properties11a was
investigated. In the experimental results, all 1:4 cobalt
complexes showed similar SMM properties with a constant
Ueff/kB of ca. 90 K, and no dependence of Ueff/kB on the Stotal
value was observed. Furthermore, the τQ values were suggested
to increase with increasing the Stotal value. In a series of 1:4
complexes CoCl2(DXpy)4, the complex with the smallest Stotal
value of 9/2, CoCl2(C1py)4, already had a τQ value long
enough not to affect the activation barrier, U/kB.
RESULTS AND DISCUSSION
■
Preparations of DXpy, X = 1, 2, 3l, 3b, and 4, and
Their Complexes with Co(p-tolsal)2. Monodiazo-, didiazo-,
tridiazo-, and tetradiazopyridine ligands, DXpy, X = 1, 2, 3l, 3b,
and 4, were prepared by the procedure reported previously.11a
D2py and D3bpy were obtained as single crystals and
characterized by X-ray crystallography (S2). The photo-
products after irradiation of DXpy were confirmed to be
high-spin polycarbene in the ground state by the field
dependence of the magnetization in frozen solution.11a,b For
the preparation of the 1:2 cobalt complexes, solutions of Co(p-
tolsal)2 and DXpy in degassed CH2Cl2 were mixed in a ratio of
1:2 at 4 °C under a nitrogen atmosphere. No single crystals
amenable to X-ray structural analysis were obtained with any
combination of Co(p-tolsal)2 and DXpy at the present stage.
The solutions of Co(p-tolsal)2(DXpy)2 were relatively unstable
in solution, and therefore the solutions were used as samples
for the magnetic measurements immediately after mixing Co(p-
To reduce the spin multiplicities of the cobalt complexes, this
time, bidentate cobalt complex Co(p-tolsal)2, p-tolsal = N-p-
tolylsalicylideniminato, was used as the heterospin complex
with an Stotal value smaller than 1:4 complexes, and the
magnetic properties before and after irradiation of the 1:2
complexes of Co(p-tolsal)2 and DXpy in frozen solution were
investigated. The 1:2 complexes Co(p-tolsal)2(CXpy)2, X = 1,
2, 3l, 3b, and 4, were expected to produce the high-spin ground
state with Stotal = 5/2, 9/2, 13/2, 13/2, and 17/2, respectively.
The 1:2 complex Co(p-tolsal)2(cC5py)2 formed after irradi-
B
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
tolsal)2 and DXpy in a ratio of 1:2. In contrast, the 1:2 cobalt
complexes [Co(p-tolsal)2(py)2] and [Co(Phsal)2(py)2] were
obtained as single crystals by mixing the solutions of Co(p-
tolsal)2 (or Co(Phsal)2) and pyridine.
X-ray Crystallography. Single crystals of [Co(p-tol-
sal)2(py)2] and [Co(Phsal)2(py)2] were analyzed by X-ray
crystallography. The complexes were crystallized in space
DXpy coordinated to the cobalt ion as the fifth and the sixth
ligand.
Magnetic Properties in Frozen Solution. The solutions
(15, 10, 10, 5, and 5 mM, 50 μL) of the 1:2 mixture of Co(p-
tolsal)2 and DXpy, X = 1, 2, 3l, 3b, and 4, respectively, in
MTHF were employed as samples of Co(p-tolsal)2(DXpy)2 for
dc and ac magnetic susceptibility measurements. A solution (50
mM, 50 μL) of [Co(p-tolsal)2(py)2] dissolved in MTHF was
used as the solution sample.
groups P2 /c and P1 with Z = 4 for [Co(p-tolsal) (py) ] and
̅
1
2
2
[Co(Phsal)2(py)2], respectively. In the molecular structures of
[Co(p-tolsal)2(py)2] and [Co(Phsal)2(py)2], the oxygen atoms
of phenoxide in Schiff-base ligands are located at the apical
position and two pyridines are coordinated to the cobalt ion in
cis and trans configurations, respectively. Both complexes
formed compressed octahedral structures, in which the bond
lengths between the cobalt ion and the phenoxide oxygen
atoms were shorter by 0.15−0.23 Å than those between the
cobalt ion and the imine and pyridine nitrogen atoms. The
bond angles of O−Co−O were 179° and 178° for [Co(p-
tolsal)2(py)2] and [Co(Phsal)2(py)2], respectively. Dihedral
angles between the pyridine rings and the x−y plane defined by
the nitrogen atoms of Schiff-base and pyridine ligands were
53−71° and 72−84° for [Co(p-tolsal)2(py)2] and [Co-
(Phsal)2(py)2], respectively. The observed molecular structures
were distorted octahedral, similar to those reported previ-
ously.11,12 In the crystal packings, the shortest intermolecular
distances between the cobalt ions were 8.38 and 8.80 Å for
[Co(p-tolsal)2(py)2] and [Co(Phsal)2(py)2], respectively.
ORTEP drawings of molecular structures for [Co(p-
tolsal)2(py)2], [Co(Phsal)2(py)2], D2py, and D3bpy are
shown in Figure S1, and their selected bond lengths, bond
angles, and dihedral angles are summarized in Tables S1−S3.
Formation of Co(p-tolsal)2(DXpy)2 Complexes in
Solution. To confirm the formation of Co(p-tolsal)2(DXpy)2
in solution, the spectra of vis−NIR and CSI mass for the
solution samples were measured.
Photolysis of Co(p-tolsal)2(DXpy)2, X = 1, 2, 3l, 3b, and
4, in Frozen Solution. Photolysis of the sample was
performed by the procedures reported previously.11a,b The M
vs irradiation time plot for the sample of 1:2 mixture of Co(p-
tolsal)2(D3lpy)2 in MTHF is shown in Figure S4. Quantitative
photolysis of the sample was confirmed by the disappearance of
the absorption at 2055 cm−1 due to the diazo moiety in the IR
spectra after SQUID measurements.
A. Static Magnetic Properties in Frozen Solution. A-
1. Temperature Dependence of dc Magnetic Susceptibility.
The values of dc molar magnetic susceptibility, χmol, before and
after irradiation of Co(p-tolsal)2(DXpy)2, X = 1, 2, 3l, 3b, and
4, in frozen solution were collected at a constant field of 5 kOe
below 20 K. The obtained χmol values were plotted as a function
of temperature. The plots of χmolT vs T before and after
irradiation of Co(p-tolsal)2(DXpy)2, X = 1, 2, 3l, 3b, and 4, are
shown in Figure 1 together with Co(p-tolsal)2(py)2.
A. Vis−NIR Spectra of Co(p-tolsal)2(DXpy)2 Complexes.
The vis−NIR spectrum of the 1:2 mixture (10 mM) of Co(p-
tolsal) and DXpy in MTHF at room temperature showed a set
of weak absorptions at 900 and 1300 nm. On cooling from 288
to 140 K, these absorptions gradually decreased and a new
absorption at 950 nm appeared below 190 K. The absorptions
at rt and 140 K are characteristic of the d−d transitions of the
cobalt(II) ion in the tetrahedral and octahedral coordination
structure, respectively,13,14 indicating that Co(p-tolsal)2 and
DXpy are mostly dissociated at rt but associate to form the 1:2
complex Co(p-tolsal)2(DXpy)2 at temperatures lower than 140
K. The spectral changes for Co(p-tolsal)2(D3lpy)2 in the
temperature range 288−140 K are shown in Figure S2. The
other combination of Co(p-tolsal)2 and DXpy showed similar
spectral changes under similar conditions.
B. Cold-Spray Ionization Mass Spectra. Cold-spray
ionization mass spectra15 for the solution samples were
measured at −10 °C using positive mode. The 1:2 mixtures
of Co(p-tolsal)2 with pyridine and DXpy in CH3CN were used
as solution samples. The spectra showed the positive molecular
ion peaks with isotope patterns, which were in accordance with
the corresponding molecular weight, M+. The CSI mass spectra
for the 1:2 mixtures of Co(p-tolsal)2 and DXpy are shown in
Figure S3 together with the simulation values.
Figure 1. χmolT vs T plots of Co(p-tolsal)2(D1py)2 (black ×), Co(p-
□
▽
△
tolsal)2(CXpy)2, X = 1 (purple ), 2 (green ), 3l (blue ), 3b
◇
(pink ), and 4 (red circle), and Co(p-tolsal)2(py)2 (red #) below 20
K in frozen solution.
The χmolT values before irradiation of Co(p-tolsal)2(DXpy)2
were nearly constant (1.5 cm3 K mol−1) in the temperature
range 1.9−10 K and were close to those for Co(p-tolsal)2(py)2.
After irradiation, the χmolT values for Co(p-tolsal)2(CXpy)2
were nearly constant depending on the multiplicity of the
complexes in the range 5−20 K and gradually decreased below
it. The constant χmolT values were 6.2, 12.8, 23.8, 24.2, and 32.1
for Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, respectively,
and were much larger than the theoretical values (3.5, 7.5, 13.5,
13.5, and 21.5 cm3 kmol−1 for Co(p-tolsal)2(CXpy)2, X = 1, 2,
3l, 3b, and 4, respectively) calculated by a spin-only equation
with two isolated high-spin carbenes (1.0, 3.0, 6.0, 6.0, and 10.0
cm3 K mol−1 × 2) and one high-spin cobalt(II) ion with
effective spin S′eff = 1/216,17 (χmolT value of 1.5 cm3 K mol−1
before irradiation). These results suggested that the carbenes
and the cobalt ions in the complex interacted ferromagnetically
to form a high-spin ground state with Stotal = 5/2, 9/2, 13/2,
13/2, and 17/2 for Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and
4, respectively. The decrease in the χmolT values below 10 K
From the results of vis−NIR and CSI mass spectrometry, the
complexes formed from the 1:2 mixture of Co(p-tolsal)2 and
DXpy at cryogenic temperature studying the magnetic
properties were safely considered to be octahedral, in which
C
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
might be due to the effect of the zero-field splitting caused by
spin−orbit coupling in the cobalt ion.
the U/kB values calculated from |D|(S2 − 1/4) by using the
obtained D values were 216 and 108 K for Co(p-tolsal)2(py)2
and Co(p-tolsal)2(C1py)2, respectively, which were largely
different from the following experimental Ueff/kB value. The
observed discrepancy was not clear at the present stage.
B. Dynamic Magnetic Properties in Frozen Solution. The
dynamic magnetic properties for Co(p-tolsal)2(CXpy)2, X = 1,
2, 3l, 3b, and 4, together with a parent complex, Co(p-
tolsal)2(py)2, were investigated by ac and dc magnetic
susceptibility measurements in frozen solution. The ac
magnetic susceptibility data were collected with 3.9 or 5.0 Oe
ac fields at various frequencies in the temperature range 10−1.9
K. The dc magnetization decays for Co(p-tolsal)2(CXpy)2, X =
2, 3l, 3b, and 4, were also measured in the range 3.0−1.9 K.
B-1. The ac Magnetic Susceptibility Measurements.
i. Co(p-tolsal)2(py)2 and Co(p-tolsal)2(C1py)2. The ac
magnetic susceptibilities for parent complex Co(p-tolsal)2(py)2
were measured with a 3.9 Oe ac field at various frequencies in
the temperature range 1.9−10 K. Although both nonzero
signals of χ′mol and χ″mol (in-phase and out-of-phase
components of ac magnetic susceptibilities, respectively) were
observed, no maxima of the signals in the χ′mol and χ″mol vs T
plots were observed above 1.9 K (Figure S7-1). By applying the
dc field, the maxima of the signals dependent on the frequency
appeared in the χ′mol and χ″mol vs T plots. To optimize the
applied dc field, the temperature dependencies of χ″mol signals
with the frequency of 1.0 Hz were investigated in the presence
of dc fields of 1.0, 3.0, and 5.0 kOe. In the χ″mol vs T plots, the
dc field of 1.0 kOe showing the maximum intensity of the χ″mol
signal was determined as the optimized dc field (Figure S7-2).
In the presence of a dc field of 1.0 kOe, the peak-top
temperature of χ″mol signals shifted to lower temperature with
decreasing frequency (Figure 3a, left) and were below 1.9 K for
frequencies lower than 10 Hz. This observation in the absence
and presence of a dc field was typical of SMM (or single-ion
magnet; SIM) behaviors affected by the QTM pathway. In the
absence of a dc field, the QTM relaxation counteracted the slow
magnetic relaxation for SMM (SIM), while in the presence of a
dc field for suppression of QTM relaxation, the SMM (SIM)
behavior was observable. In Co(p-tolsal)2(C1py)2, in contrast,
both χ′mol and χ″mol signals were clearly observed in the absence
of a dc field and showed frequency dependence (Figure S8-1
and Figure 3a, right, respectively). As observed in the χ″mol vs T
plot, the peak of χ″mol signals had a shoulder below 2.5 K. The
χ″mol values at 1.9 K increased with decreasing frequency and
eventually were dominant at frequencies less than 1.0 Hz.
When the dc field of 1 kOe was applied, the shoulder for the
peak of χ″mol signals disappeared (Figure S8-2b), suggesting
that the χ″mol signals below 2.5 K might be affected by the
relaxation due to the QTM pathway.
A-2. Hysteresis Loops below 3.0 K. The dc magnetization
for Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, was measured
in the range −50 to 50 kOe with a field-sweep rate of 0.35
kOe/s at various temperatures in the range 1.9−3.5 K. The
Mmol values gradually increased on applying the field and kept
on increasing even at 50 kOe, for which the Mmol values were
2.0 × 104, 3.8 × 104, 5.2 × 104, 5.6 × 104, and 6.8 × 104 cm3 Oe
mol−1 for Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4,
respectively. These values at 50 kOe were lower than those for
the values of saturation magnetization, Ms, suggesting that the
effect of large spin−orbit coupling operated. In all complexes,
hysteresis loops relating to the field appeared below ca. 3.0 K,
and the width of the loop increased upon cooling. The
hysteresis loops at 1.9 K for Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l,
3b, and 4, are summarized in Figure 2, and those at given
temperatures together with the Hc vs T plots are shown in
Figure S5.
Figure 2. Plots of Mmol vs H of Co(p-tolsal)2(CXpy)2, X = 1 (purple),
2 (green), 3l (blue), 3b (pink), and 4 (red), at 1.9 K in frozen solution
with a sweep rate of 0.35 kOe/s.
In Co(p-tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4, the coercive
force (Hc) and the remnant magnetization (Mr) depended on
the temperature. On cooling from 3.0 K, both values gradually
increased until 1.9 K. The observed thermal behaviors of Hc
and Mr were characteristic of an SMM. The values of Hc and Mr
at 1.9 K were ∼0, 6.0, 10, 6.2, and 9.0 kOe and 0, 1.6 × 104, 2.6
× 104, 1.7 × 104, and 2.9 × 104 cm3 Oe mol−1 for Co(p-
tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, respectively. The Hc
values at 1.9 K increased in the order Co(p-tolsal)2(CXpy)2, X
= 1, 2, 3b, 3l, and 4, in which those for Co(p-tolsal)2(CXpy)2,
X = 3l and 4, were close. It was noted that the Hc value for
Co(p-tolsal)2(3bpy)2, having a branched structure, was differ-
ent from that for the linear isomer, and the value of 6.2 kOe
was rather close to that for Co(p-tolsal)2(C2py)2 and that (7.1
kOe) for Co(p-tolsal)2(cC5py)2, which had a partial structure
of Co(p-tolsal)2(3bpy)2.
To estimate the D and E values for Co(p-tolsal)2(py)2 and
Co(p-tolsal)2(C1py)2, which showed no hysteresis and an
extremely weak hysteresis loop at 1.9 K, respectively, the
reduced magnetization data in the Mmol vs H/T plots were
analyzed by using the program ANISOFT 2.0.18 The best
fitting results for Co(p-tolsal)2(py)2 and Co(p-tolsal)2(C1py)2
were −108 and −18 K for D and −0.0008 and −0.0001 K for E
values, respectively (Figure S6). The negative D values were
obtained, and the large reduction of the |D| value from Co(p-
tolsal)2(py)2 to Co(p-tolsal)2(C1py)2 was observed. However,
The observed thermal profile of the χ″mol signals for Co(p-
tolsal)2(py)2 and Co(p-tolsal)2(C1py)2 might suggest that the
magnetic relaxations through thermodynamic and QTM
pathways were observed at high and low temperatures,
respectively. To estimate the relaxation time at low temper-
ature, the χ″mol values in the χ″mol vs T plot were analyzed by
the Debye model19 and were plotted as a function of frequency
(Figure 3b). The τ values obtained from the plots of χ″mol vs T
and χ″mol vs frequency were combined and plotted as a function
of inverse T (Figure 3c). In the τ vs T−1 plots, on cooling, the τ
values almost linearly increased in the high-temperature region
and deviated from the straight line at low temperature. The
relaxation in the high-temperature region obeyed the Arrhenius
D
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Plots of (a) χ″mol vs T, (b) χ″mol vs frequency, and (c) τ vs T−1 for Co(p-tolsal)2(py)2 (left) and Co(p-tolsal)2(C1py)2 (right) in the
presence and absence of a dc field of 1 kOe, respectively. The solid lines in (a) and (b) are visual guides and fitted curves by the Debye model,
respectively. In (c), the red circles and blue squares are τ values obtained from the plots of χ″mol vs T and χ″mol vs frequency, respectively, and the
straight lines are obtained by the Arrhenius law using the data at 1000, 750, 500, 250, and 200 Hz in the χ″mol vs T plot.
law; τ = 1/(2πν) = τ0 exp(Ueff/kBT). From the straight line in
the τ vs T−1 plots, the effective activation energy, Ueff/kB, and
the pre-exponential factor, τ0, were estimated. In Co(p-
tolsal)2(py)2, the values of Ueff/kB and τ0 in the presence of
1.0 kOe were estimated to be 38 K and 8.0 × 10−7 s,
respectively. The τ values obtained by the Debye model
deviated downward from the straight line below 4.0 K and
reached a near-plateau (0.05 s) at 1.9 K. In the absence of a dc
field, on the other hand, values of Ueff/kB of 40 K and τ0 of 8.0
× 10−8 s were obtained for Co(p-tolsal)2(C1py)2. The τ values
below 2.5 K obtained from the χ″mol vs frequency plot deviated
from the straight line for the Arrhenius plot and approached a
constant value. The relaxation time, τ1.9, of 1.6 s at 1.9 K might
be close to the relaxation time, τQ, due to QTM, which was
temperature independent. Applying a dc field of 1 kOe, the
deviation below 2.5 K in the τ vs T−1 plot disappeared and
values of Ueff/kB of 42 K and τ0 of 5.2 × 10−8 s were obtained
(Figure S8-2d). The values obtained in the absence and
presence of a dc field were ca. 40 K and similar to each other.
This result indicated that the contribution of QTM relaxation
to the Ueff/kB value for Co(p-tolsal)2(C1py)2 was insignificant
and the relaxation below 2.5 K was controlled by QTM.
The χ″mol vs T, χ″mol vs frequency, and Arrhenius plots for
Co(p-tolsal)2(py)2 and Co(p-tolsal)2(C1py)2 in the presence
and absence of a dc field (1 kOe), respectively, are shown in
Figure 3, and those for Co(p-tolsal)2(C1py)2 in the presence of
a dc field (1 kOe) are shown in Figure S8-2.
The microcrystalline samples of [Co(p-tolsal)2(py)2] and
[Co(Phsal)2(py)2], which had a cis and trans configuration,
respectively, also showed similar magnetic behavior in the
absence and presence of a dc of 1.0 kOe, which are shown in
Figure S9-1 and S9-2 for [Co(p-tolsal)2(py)2] and in Figure S9-
3 and S9-4 for [Co(Phsal)2(py)2]. The values of Ueff/kB and τ0
of 35 and 34 K and 1.2 × 10−6 and 3.3 × 10−7 s for [Co(p-
tolsal)2(py)2] and [Co(Phsal)2(py)2], respectively, in the
presence of a dc of 1.0 kOe were obtained. These results
E
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
suggest that the coordination geometry of pyridine in these
cobalt complexes did not largely affect the Ueff/kB values.
The parent complex Co(p-tolsal)2(py)2 is a high-spin
cobalt(II) complex with noninteger spin. In the easy-axis (D
< 0) noninteger spin system, the transverse anisotropy should
not affect the QTM pathway according to Kramers’ theorem. In
addition, the measuring conditions, which were the presence of
a dc field and in frozen solution, might eliminate the
contribution of the dipole interaction and the hyperfine
coupling to QTM. However, the observed magnetic behaviors
in the absence and presence of a dc field suggested that Co(p-
tolsal)2(py)2 had a non-negligible QTM pathway within a
molecule. Recently, similar results for complexes containing
high-spin cobalt(II) ions have been reported.6 In contrast,
Co(p-tolsal)2(C1py)2 showed a slow magnetic relaxation with
the Ueff/kB value in the absence of dc field, indicating that the
generating carbene magnetically coupled with the cobalt ion to
suppress the relaxation through the QTM pathway.
tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4, respectively, and
remained essentially constant on cooling to ∼5 K. Below 5 K,
the χ′molT values with frequency dependence decreased, while
the χ″mol signals appeared at the same temperature. In the χ″mol
vs T plot, the χ″mol signals with frequency dependence were
also observed above 1.9 K, and the maximum χ″mol signals
shifted to lower temperature with decreasing frequency. From
each frequency (1/4πτ) at the peak-top temperature for χ″mol
,
the values for Ueff/kB and τ0 were obtained in terms of the
Arrhenius equation. The values of Ueff/kB and τ0 for Co(p-
tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4, were estimated to be 65,
73, 72, and 74 K, and 4.4 × 10−9, 1.5 × 10−9, 1.1 × 10−9, and
1.6 × 10−9 s, respectively (Figure 7). The Ueff/kB and τ0 values
ii. Co(p-tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4. The ac
magnetic susceptibilities of Co(p-tolsal)2(CXpy)2, X = 2, 3l, 3b,
and 4, were measured under conditions similar to those for
Co(p-tolsal)2(C1py)2. Both χ′mol and χ″mol signals depending
on the frequency were clearly observed in the absence of a dc
field. The χ′molT vs T plots are shown in Figure S10. The χ″mol
vs T plots are shown in Figure 4 for Co(p-tolsal)2(CXpy)2, X =
2, 3l, and 4, and Figure S11 for Co(p-tolsal)2(C3bpy)2.
In the plot of χ′molT vs T, the values of χ′molT were 13.5,
23.7, 22.7, and 33.3 cm3 K mol−1 at 10 K for Co(p-
□
○
Figure 5. Plot of χ′mol vs χ″mol at 4.5 (blue ), 4.9 (red ), and 5.3
△
(green ) K for Co(p-tolsal)2(C3lpy)2. Solid lines are a least-squares
fitting of the data to the distribution of a single relaxation process.
of the isomeric complexes Co(p-tolsal)2(C3bpy)2 and Co(p-
tolsal)2(C3lpy)2 were close to each other, indicating that the
structural difference did not affect the Ueff/kB value. Those for
Co(p-tolsal)2(cC5Py)2 were reported11f to be 72 K and 2.0 ×
10−9 s, respectively. The obtained τ0 values, which were smaller
than those of typical SMMs,2 were close to those of SMMs with
large Stotal values reported previously.2c,d,20
In order to investigate the relaxation process due to the
cobalt complexes in frozen solutions in more detail, further ac
magnetic susceptibility measurements were carried out. The ac
magnetic susceptibility data after irradiation of the 1:2 mixture,
Co(p-tolsal)2(D3lpy)2, in frozen MTHF solution were
collected by holding at various constant temperatures and
varying the frequency of the ac field (5 Oe) from 1000 to 1 Hz.
The plots of χ′mol vs frequency and χ″mol vs frequency at 4.5,
4.9, and 5.3 K are shown in Figure S12, and that of χ′mol vs
χ″mol (Cole−Cole diagram) is shown in Figure 5.
The frequency dependences of χ′mol and χ″mol at constant
temperatures were analyzed by a Debye model, and the fitted
curves are shown as solid lines in Figure S12. The Cole−Cole
plot21 shown in Figure 5, which is nearly symmetric with
distribution widths α = 0.23, 0.28, and 0.34 at 5.3, 4.9, and 4.5
K, respectively, indicated that a single relaxation process was
present in the Co(p-tolsal)2(C3lpy)2. The obtained α values
were relatively large, which suggested that there was a
distribution in the single relaxation process. This distribution
would be mainly due to the frozen solution condition used for
the formation of SMM.
B-2. The dc Magnetization Decay for Co(p-tolsal)2(CXpy)2,
X = 2, 3l, 3b, 4, and Co(p-tolsal)2(cC5py)2. To characterize
the slow magnetization relaxation at even lower temperatures,
the dc magnetization decay experiments for Co(p-tol-
Figure 4. Plots of χ″mol vs T of Co(p-tolsal)2(CXpy)2, X = (a) 2, (b)
3l, and (c) 4, in frozen MTHF solution with a 5 Oe ac field oscillating
at 1 ( ), 10 ( ), 100 ( ), 500 ( ), and 1000 ( ) Hz. The solid
◇
■
▼
▲
●
lines are visual guides.
F
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
times were due to QTM. To obtain the relaxation times for
Co(p-tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4, and Co(p-
tolsal)2(cC5py)2, the experimental decay data were fitted to a
stretched exponential decay (eq 1).11a,b,22
ln(M) = ln(M0) − (t/τ)B
(1)
where M0 is the initial magnetization, τ is the average relaxation
time, and B is the width of the distribution; B = 1 is a single-
exponential decay. Taking the magnitude of the M0 value
(M0(T); M0 at T, > 0.5M0(1.9)) into account, in which the
values of M0(1.9) for Co(p-tolsal)2(CXpy)2, X = 2, 3l, 3b, and
4, and Co(p-tolsal)2(cC5py)2 were 31, 46, 31, 47, and 35%,
respectively, of the M value at 50 kOe, the experimental decay
data were selected and the data in the temperature ranges 1.9−
2.2, 1.9−2.6, 1.9−2.6, 2.0−2.6, and 2.1−2.8 K for Co(p-
tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4, and Co(p-
tolsal)2(cC5py)2 were used for the fitting experiments. The τ
values in the range 103−105 s with B values in the range 0.4−
0.6 were obtained. The values of τ and B obtained by a
stretched exponential decay are listed in Table 1 together with
the relaxation times for Co(p-tolsal)2(C1py)2 obtained by the
Debye model, and the fitting curves are shown in Figure 6 as
solid lines.
Magnetic Relaxations in Co(p-tolsal)2(CXpy)2, X = 2,
3l, 3b, and 4. In SMM, the characteristic slow magnetic
relaxation for the reorientation of the magnetization takes place
in two pathways due to U/kB and QTM. The relaxations due to
the former are temperature-dependent and obtained by the ac
magnetic susceptibility technique, while those for the latter are
temperature-independent and determined by the temperature
dependence of the dc magnetization decay at extremely low
temperature. For Co(p-tolsal)2(CXpy)2, X = 2, 3l, and 4, and
Co(p-tolsal)2(cC5py)2, the dc magnetization decays are
combined with those given by means of the ac magnetic
susceptibility technique and are shown as τ vs T−1 plots in
Figure 7. The τ vs T−1 plot for Co(p-tolsal)2(C3bpy)2 is shown
in Figure S14.
In the higher temperature region, the decay obeyed the
Arrhenius law, while in the lower temperature region, the τ
values for Co(p-tolsal)(CXpy)2 collected by the dc magnet-
ization decay deviated downward from the extrapolation of the
linear Arrhenius plot obtained by the ac magnetic susceptibility
technique. The observed deviation suggests that the τ values in
the low-temperature region (<3.0 K) were affected by the τQ. In
addition, the degree of deviation became smaller in the order
Co(p-tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4, suggesting that the
contribution of τQ due to RQT relaxation became smaller with
increasing Stotal value. Accordingly, the τQ values might become
larger with increasing Stotal value. Actually, the τ values at a
given temperature below 3 K for Co(p-tolsal)2(cC5py)2, which
had the largest Stotal value, were traced on a straight line, and no
deviation was observed until 1.9 K. Although the τQ values for
Co(p-tolsal)2(CXpy)2, X = 2, 3l, 3b, and 4, could not be
determined, they were estimated to be ∼2 × 103 s for Co(p-
tolsal)2(C2py)2 and >105 s for Co(p-tolsal)2(CXpy)2, X = 3l,
3b, and 4, from the τ values at 1.9 or 2.0 K.
Figure 6. The dc magnetization decays for (a) Co(p-tolsal)2(C2py)2,
(b) Co(p-tolsal)2(C3lpy)2, and (c) Co(p-tolsal)2(C4py)2, at given
temperatures, after applying a field of 50 and reducing to 0 kOe. Solid
lines show the fitting results by the stretched exponential equation.
sal)2(CXpy)2, X = 2, 3l, 3b, and 4, were carried out in the
temperature range 1.9−3.0 K. In addition, the decay experiment
for Co(p-tolsal)2(cC5py)2, which had the largest Scar value of
10/2, was also carried out under a similar condition. After the
cycle of applying the field of 50 kOe at 10 K, cooling to the
desired temperature, and then reducing to 0 kOe, the
magnetizations were measured as a function of time. The
measurements were performed at various temperatures in the
range 2.8−1.9 K. For each measurement, the temperature was
raised to 10 K once and then cooled to the desired temperature.
The magnetization data, M, were normalized by that, M0, at t =
0 at each temperature and were plotted as M/M0 vs time. The
decay curves at each given temperature are shown in Figure 6
for Co(p-tolsal)2(CXpy)2, X = 2, 3l, and 4, and in Figure S13
for Co(p-tolsal)2(C3bpy)2 and Co(p-tolsal)2(cC5py)2.
Role of Carbene in SMM Behaviors of 1:2 Complexes
of Co(p-tolsal)2(CXpy)2 in Frozen Solution. The 1:2 cobalt
complexes with Stotal = 3/2−21/2 showed heterospin SMM
behaviors, and their thermal activation barriers partially
depended on the Stotal values due to carbene multiplicity. All
magnetic physical values for Co(p-tolsal)2(py)2 and Co(p-
The dc magnetization decays for Co(p-tolsal)2(CXpy)2, X =
2, 3l, 3b, and 4, and Co(p-tolsal)2(cC5py)2 became slower on
cooling, and their decay times at the same temperature
decreased in the order Co(p-tolsal)2(CXpy)2, X = 2, 3, and
4, and Co(p-tolsal)2(cC5py)2. However, the decay times above
1.9 K did not reach constant values, determining that the decay
G
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 7. Plots of τ vs T−1 for Co(p-tolsal)2(CXpy)2, X = (a) 2, (b) 3l, (c) 4, and (d) Co(p-tolsal)2(cC5py)2. The symbols of red circles and blue
squares indicate the τ data collected by ac magnetic susceptibility technique and by dc magnetization decay, respectively. The solid lines are the least-
squares fits of the ac data (red circles) according to the Arrhenius equation.
Table 1. Values of Average Relaxation Time, τ, and Distribution Parameter, B, in Parentheses for Co(p-tolsal)2(CXpy)2, X = 2,
3l, 3b, and 4, and Co(p-tolsal)2(cC5py)2, Together with Those for Co(p-tolsal)2(C1py)2 at Various Temperatures below 2.8 K
τ/s
T/K
Co(p-tolsal)2(C1py)2
Co(p-tolsal)2(C2py)2
Co(p-tolsal)2(C3lpy)2
Co(p-tolsal)2(C3bpy)2
2.8 × 105 (0.50)
Co(p-tolsal)2(C4py)2
Co(p-tolsal)2(cC5py)2
a
b
1.9
2.0
2.1
2.2
2.3
2.4
2.6
2.8
1.6
2.2 × 103 (0.62)
1.4 × 103 (0.66)
3.6 × 105 (0.44)
1.8 × 105 (0.45)
5.1 × 104 (0.57)
2.8 × 104 (0.58)
n.e.
n.e.
a
1.4
2.5 × 105 (0.42)
7.7 × 104 (0.45)
3.7 × 104 (0.48)
n.e.
a
1.0
6.0 × 104 (0.53)
2.0 × 104 (0.57)
3.9 × 105 (0.51)
7.3 × 102 (0.61)
3.8 × 104 (0.60)
n.e.
n.e.
2.3 × 103 (0.54)
1.2 × 103 (0.51)
n.e.
3.6 × 103 (0.54)
8.2 × 102 (0.50)
2.9 × 104 (0.59)
4.2 × 103 (0.54)
n.e.
2.7 × 103 (0.62)
7.2 × 102 (0.54)
n.e.
a
b
Value obtained by a least-squares method using the data for the Debye plot. n.e., not estimated.
tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, are summarized in
Table 2 together with those for Co(p-tolsal)2(cC5py)2.
τQ values depended on the Stotal value. The τQ value for Co(p-
tolsal)2(C1py)2 with Stotal = 5/2 was 1.6 s, which was
sufficiently longer than τ used to determine the activation
barrier (Figure 3c, right). Furthermore, the τQ value for Co(p-
tolsal)2(C2py)2 with Stotal = 9/2 was ca. 2 × 103 s and much
longer than the time scale for ac measurements. In Co(p-
tolsal)2(CXpy)2, X = 3 and 4, and Co(p-tolsal)2(cC5py)2 with
Stotal = 13/2, 17/2, and 21/2, respectively, although accurate
values of τQ for 3, 4, and Co(p-tolsal)2(cC5py)2 could not be
determined and were extremely slow (>105 s), a tendency for
the τQ values to increase with increasing Stotal value was
observed; the deviation of the data obtained from the dc decay
became small in the order 3, 4, and Co(p-tolsal)2(cC5py)2
(Figure 6). In Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4,
and Co(p-tolsal)2(cC5py)2, therefore, the τQ increased with
increasing Stotal and the contribution of QTM relaxation to the
obtained Ueff/kB value could be eliminated.
The Ueff/kB value in SMM is mainly determined by the
relative ratio for the thermodynamic and the QTM relaxation
times (τ and τQ, respectively) in both pathways, which are
temperature dependent and independent, respectively. Co(p-
tolsal)2(py)2, having no carbene, showed no SMM behavior in
the absence of a dc field, suggesting that the τQ was faster than τ
and counteracted the SMM behavior. Applying the dc field (1
kOe) for depression of the QTM pathway elicited an SMM
behavior with a Ueff/kB of 38 K. In contrast, in the absence of a
dc field, Co(p-tolsal)2(C1py)2 exhibited an SMM behavior with
a Ueff/kB of 40 K, whose value was close to that for Co(p-
tolsal)2(py)2 in the presence of a dc field. This result suggested
that carbene magnetically interacted with the cobalt ion to
suppress the QTM pathway and make the τQ value large.
Actually, in Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, the
H
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
and Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, and Co(p-
tolsal)2(cC5py)2. This order was consistent with that of the |D|
values calculated from |D|(S2 − 1/4). Below Sc, the increase of
Ueff/kB for Co(p-tolsal)2(CXpy)2, X = 1, 2, and 3, might be due
to the increase of Stotal accompanied by the decrease of |D|. In
contrast, above Sc, the constant values of ca. 70 K for Ueff/kB of
Co(p-tolsal)2(CXpy)2, X = 3l, 3b, and 4, and Co(p-
tolsal)2(cC5py)2 could be explained in a similar way; the
constant Ueff/kB value of ca. 70 K could be obtained by the
increase of Stotal with a decrease of |D| using the equation U/kB
= |D|(S2 − 1/4). However, in the large spin system with
relatively small exchange coupling,25 generally, the contribution
of the low-lying thermally excited state to the ground state
might be large and increase with increasing Stotal value. In Co(p-
tolsal)2(CXpy)2, X = 3l, 3b, 4, and Co(p-tolsal)2(cC5py)2,
being above Sc, the observed spin multiplicities in the ground
state did not correspond to the Stotal value. The large
participations of the low-lying thermally excited state in the
ground state might lead the Ueff/kB values of the complexes
over Stotal = 13/2 to become constant (ca. 70 K). In the
polymetal cluster systems with high-spin ground states, actually,
similar participations in the ground state have been
reported.2c,d,20 Similar results were also observed in the 1:4
Co−carbene complexes.11a,b In the experiments using Co(Cl2)-
(CXpy)4, X = 1, 2, 3l, 3b, and 4, with Stotal = 9/2−33/2, the
obtained Ueff/kB values for all complexes were nearly constant
at 90 K. This result suggested the possibility that the Ueff/kB
value did not depend on the multiplicity, Scar, of carbenes in
heterospin SMM. In this study using the 1:2 complexes,
however, the result that, when the Stotal value was smaller than
Sc = 13/2, the Ueff/kB value depended on Stotal and increased
with increasing Stotal value, eliminated such a possibility. The
Stotal value of 9/2 for Co(Cl)2(C1py)4, which was the smallest
value in Co(Cl)2(CXpy)4, might already be above the value of
Sc for the 1:4 complexes.
Table 2. Magnetic Physical Data for Co(p-tolsal)2(py)2,
Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, and Co(p-
tolsal)2(cC5py)2 in Frozen Solution
a
Hc
kOe
/
1:2 complexes
Stotal
Ueff/K
τ0/s
τQ/s
b
bc
,
bc
,
Co(p-tolsal)2(py)2
3/2
38
40
8.0 × 10−7 ∼0.05
0
d
c
Co(p-
tolsal)2(C1py)2
5/2
8.4 × 10−8 1.6
∼ 0
d
d
d
d
d
Co(p-
9/2
65
73
72
74
72
4.4 × 10−9 ∼2 × 103
1.5 × 10−9 >105
1.1 × 10−9 >105
1.6 × 10−9 >105
6.2
10
tolsal)2(C2py)2
Co(p-
tolsal)2(C3lpy)2
13/2
13/2
17/2
21/2
Co(p-
tolsal)2(C3bpy)2
6.0
9.0
7.1
Co(p-
tolsal)2(C4py)2
Co(p-
tolsal)2(cC5py)2
2.0 × 10−9 >105
e
a
b
c
At 1.9 K. In the presence of a 1.0 kOe dc field. τ value at 1.9 K
d
obtained by the Debye model. Calculated as effective spin (S′ = 1/2)
for the cobalt ion. Ref 11f.
e
In contrast, the dependence of Ueff/kB on Stotal showed a
profile different from the relationship between τQ and Stotal. The
Ueff/kB vs Stotal plot is shown in Figure 8.
CONCLUSION
■
A parent complex Co(p-tolsal)2(py)2 showed no SMM
behavior in the absence of a dc field and showed an SMM
behavior in the presence of a dc field. This result could be
explained by the SMM behavior for Co(p-tolsal)2(py)2 being
counteracted by the fast relaxations through the QTM pathway
and the QTM relaxation being suppressed by applying a static
dc field. In contrast, the 1:2 complexes Co(p-tolsal)2(DXpy)2,
X = 1, 2, 3, and 4, showed SMM behaviors after irradiation.
This result suggested that the magnetic coupling between
carbene and the cobalt ion suppressed the QTM pathway,
leading to SMM behavior being observable. In the generated
high-spin cobalt−carbene complexes with Stotal = 5/2, 9/2, 13/
2, and 17/2 after irradiation, the QTM relaxation time, τQ,
depended on the Stotal of the complex, and the τQ value
increased with increasing Stotal value. On the other hand, the
obtained effective activation barrier, Ueff/kB, also depended on
the Stotal value, which had a critical value, Sc. Below Sc, the Ueff/
kB value increased with increasing Stotal value according to the
equation Ueff/kB = |D|(S2 − 1/4), in which |D| decreased with
increasing Stotal. Above Sc, the high-spin ground state was
affected by the large contribution of the low-lying excited state
to provide heterospin SMM complexes with constant Ueff/kB
values.
Figure 8. Plot of Ueff/kB vs Stotal for Co(p-tolsal)2(CXpy)2, X = 1, 2, 3l,
3b, 4 (open circles), and Co(p-tolsal)2(cC5py)2 (filled circles).
The dependence of Ueff/kB on Stotal for Co(p-tolsal)2(CXpy)2
and Co(p-tolsal)2(cC5py)2 had a critical Stotal value, Sc, of 13/2
and showed a different dependence below and above Sc, as
shown in Figure 8. Below Sc, the Ueff/kB values increased in the
order 40, 65, and 74 K at Stotal values of 5/2, 9/2, and 13/2,
respectively, while above Sc values of 13/2, the Ueff/kB values
were constant (ca. 70 K). Since in these 1:2 cobalt−carbene
complexes the contribution of QTM relaxation to the
thermodynamic activation barrier, U/kB, for the orientation of
magnetic moments could be ignored, the Ueff/kB (∼U/kB)
values could be considered as the equation |D|(S2 − 1/4) for
noninteger spin. Therefore, the Ueff/kB values for Co(p-
tolsal)2(CXpy)2, X = 1, 2, 3l, 3b, and 4, and Co(p-
tolsal)2(cC5py)2 corresponded to 6|D|, 20|D|, 42|D|, 42|D|,
72|D|, and 110|D|, respectively, and the |D| values were
estimated to be 6.6, 3.2, 1.5, 1.5, 1.0, and 0.65 K, respectively,
indicating that the |D| values decreased with increasing the Stotal
value in Co(p-tolsal)2(CXpy)2 and Co(p-tolsal)2(cC5py)2.
Relating to the |D| value,23 the magnetic couplings between
the cobalt ion and the carbene (|D|/hc = 0.434 and |E|/hc =
0.020 cm−1 for C1py)24 would reduce the |D| value. Therefore,
the |D| values would be smaller in the order Co(p-tolsal)2(py)2
In conclusion, the organic spin in a heterospin complex was
considered to be a useful tool for suppressing the QTM
pathway for the construction of an SMM with a relatively large
I
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(2) (a) Martínez-Lillo, J.; Dolan, N.; Brechin, E. K. Dalton Trans.
2013, 12824−12827. (b) Mitsumoto, K.; Nishikawa, H.; Newton, G.
N.; Oshio, H. Dalton Trans. 2012, 13601−13608. (c) Milios, C. J.;
Vinslava, A.; Wood, P. A.; Parsons, S.; Wernsdorfer, W.; Christou, G.;
Perlepes, S. P.; Brechin, E. K. J. Am. Chem. Soc. 2007, 129, 8−9.
(d) Milios, C. J.; Vinslava, A.; Wernsdorfer, W.; Moggach, S.; Parsons,
S.; Perlepes, S. P.; Christou, G.; Brechin, E. K. J. Am. Chem. Soc. 2007,
129, 2754−2755. (e) Chakov, N. E.; Lee, S.-C.; Harter, A. G.; Kuhns,
P. L.; Reyes, A. P.; Hill, S. O.; Dalal, N. S.; Wernsdorfer, W.; Abboud,
K. A.; Christou, G. J. Am. Chem. Soc. 2006, 128, 6975−6989.
(f) Lecren, L.; Wernsdorfer, W.; Li, Y. G.; Roubeau, O.; Miyasaka, H.;
activation barrier for reorientation of the magnetic moment.
The observed heterospin SMM properties might be due to the
location of the organic spin in the cobalt complex. In Co(p-
tolsal)2(CXpy)2, the anisotropic axes were anticipated to be the
z axis (O−Co−O), and the organic spins located on the x−y
plane affect the transverse anisotropy. Attempts to construct
the heterospin system with organic spin at the anisotropic axis
are in progress.
EXPERIMENTAL SECTION
■
́
Clerac, R. J. Am. Chem. Soc. 2005, 127, 11311−13117.
General Procedures. Infrared spectra were recorded on a JASCO
420 FT-IR spectrometer. 1H NMR spectra were measured on a Varian
UNITY-400 using CDCl3 as solvent and referenced to TMS. Visible
spectra were recorded on a JASCO V570 spectrometer, and an NACC
cryo-system LTS-22X was attached for the low-temperature measure-
ments. The irradiation system and the samples for vis−near IR spectra
measurements are described in detail in the Supporting Information
(SI). Cold-spray ionization mass spectra were recorded on a JEOL
JMS-T100CS spectrometer. Elemental analyses were performed at the
Analytical Center of the Faculty of Science in Kyushu University.
Magnetic Measurements. The ac and dc magnetic susceptibility
data were obtained on Quantum Design MPMS2 (0 10 kOe) and
(3) (a) Meihaus, K. R.; Long, J. R. J. Am. Chem. Soc. 2013, 135,
17952−17957. (b) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R.
Nat. Chem. 2011, 3, 538−542. (c) Rinehart, J. D.; Fang, M.; Evans, W.
J.; Long, J. R. J. Am. Chem. Soc. 2011, 133, 14236−14239. (d) Guo, Y.-
N.; Xu, G.-F.; Wernsdorfer, W.; Ungur, L.; Guo, Y.; Tang, J. K.; Zhang,
H.-J.; Chibotaru, L. F.; Powell, A. K. J. Am. Chem. Soc. 2011, 133,
11948−11951.
(4) (a) Rinck, J.; Novitchi, G.; Van den Heuvel, W.; Ungur, L.; Lan,
Y. H.; Wernsdorfer, W.; Anson, C. E.; Chibotaru, L. F.; Powell, A. K.
Angew. Chem., Int. Ed. 2010, 49, 7583−7587. (b) Baskar, V.; Gopal, K.;
Helliwell, M.; Tuna, F.; Wernsdorfer, W.; Winpenny, R. E. P. Dalton
Trans. 2010, 4747−4750. (c) Osa, S.; Kido, T.; Matsumoto, N.; Re,
N.; Pochaba, A.; Mrozinski, J. J. Am. Chem. Soc. 2004, 126, 420−421.
(5) (a) Jiang, S.-D.; Wang, B.-W.; Su, G.; Wang, Z.-M.; Gao, S.
Angew. Chem., Int. Ed. 2010, 49, 7448−7451. (b) Lin, P.-H.; Burchell,
T. J.; Ungur, L.; Chibotaru, L. F.; Wernsdorfer, W.; Murugesu, M.
Angew. Chem., Int. Ed. 2009, 48, 9489−9492. (c) Pointillart, F.; Bernot,
K.; Sessoli, R.; Gatteschi, D. Chem.−Eur. J. 2007, 13, 1602−1609.
(d) Ishikawa, N.; Sugita, M.; Wernsdorfer, W. J. Am. Chem. Soc. 2005,
127, 3650−3651. (e) Ishikawa, N.; Sugita, M.; Ishikawa, T.; Koshihara,
S.; Kaizu, Y. J. Am. Chem. Soc. 2003, 125, 8694−8695.
MPMS-5S (0
50 kOe) SQUID magneto/susceptometers,
respectively. The irradiation system and the sample for magnetic
susceptibility measurements are described in detail in the SI.
Materials. 2-Methyltetrahydrofuran and diethyl ether were distilled
from sodium benzophenone ketyl. The cobalt complex Co(p-tolsal)2
and diazo-pyridine derivatives D1py, D2py, D3lpy, D3bpy, D4py, and
cD5py were prepared by the procedure reported previously.11a
ASSOCIATED CONTENT
* Supporting Information
■
S
(6) (a) Zadrozny, J. M.; Liu, J. J.; Piro, N. A.; Chang, C. J.; Hill, S.;
Long, J. R. Chem. Commun. 2012, 3927−3929. (b) Jurca, T.; Farghal,
A.; Lin, P.-H.; Korobkov, I.; Murugesu, M.; Richeson, D. S. J. Am.
Chem. Soc. 2011, 133, 15814−15817. (c) Zadrozny, J. M.; Long, J. R. J.
Am. Chem. Soc. 2011, 133, 20732−20734. (d) Harman, W. H.; Harris,
T. D.; Freedman, D. E.; Fong, H.; Chang, A.; Rinehart, J. D.;
Ozarowski, A.; Sougrati, M. T.; Grandjean, F.; Long, G. J.; Long, J. R.;
Chang, C. J. J. Am. Chem. Soc. 2010, 132, 18115−18126. (e) Freedman,
D. E.; Harman, W. H.; Harris, T. D.; Long, G. J.; Chang, C. J.; Long, J.
R. J. Am. Chem. Soc. 2010, 132, 1224−1225.
Experimental details, vis−NIR spectra changes, CSI mass
spectra of Co(p-tolsal)2(DXpy)2, X = 1, 2, 3l, and 4, and
[Co(p-tolsal)2(py)2] in solution, and supplementary magnetic
data for their photoproducts. These materials are available free
of charge via the Internet at http://pubs.acs.org. Full
crystallographic data (CCDC No. 966174, 996590, 966175,
and 966176) for [Co(p-tolsal)2(py)2], [Co(Phlsal)2(py)2],
D2py, and D3bpy have been deposited at the Cambridge
Crystallographic Database Center and are available on request
from the Director, CCDC, 12 Union Road, Cambridge, CB2
1EZ, UK (fax: +44-1223-336-033; e-mail: deposit@ccdc.cam.
ac.uk or http://www.ccdc.cam.ac.uk).
(7) Abragam, A.; Bleaney, B. Electron Paramagnetic Resonance of
Transition Ions; Clarendon Press: Oxford, 1970.
(8) (a) Zhao, H. H.; Heintz, R. A.; Dunbar, K. R.; Rogers, R. D. J.
Am. Chem. Soc. 1996, 115, 12844−12845. (b) Manriquez, J. M.; Yee,
G. T.; McLean, R. S.; Epstein, A. J.; Miller, J. S. Science 1991, 252,
1415−1417. (c) Caneschi, A.; Gatteschi, D.; Laugier, J.; Rey, P.;
Sessoli, R. Inorg. Chem. 1988, 27, 1553−1557.
(9) (a) Murashima, K.; Watanabe, T.; Kanegawa, S.; Yoshihara, D.;
Inagaki, Y.; Karasawa, S.; Koga, N. Inorg. Chem. 2012, 51, 4982−4993.
(b) Karasawa, S.; Koga, N. Inorg. Chem. 2011, 50, 2055−2057.
(c) Zhu, Z.; Karasawa, S.; Koga, N. Inorg. Chem. 2005, 44, 6004−6011.
(d) Morikawa, H.; Imamura, F.; Tsurukami, Y.; Itoh, T.; Kumada, H.;
Karasawa, S.; Koga, N.; Iwamura, H. J. Mater. Chem. 2001, 493−502.
(e) Karasawa, S.; Kumada, H.; Koga, N.; Iwamura, H. J. Am. Chem. Soc.
2001, 123, 9685−9686. (f) Karasawa, S.; Sano, Y.; Akita, T.; Koga, N.;
Itoh, T.; Iwamura, H. J. Am. Chem. Soc. 1998, 120, 10080−10087.
(10) Jeon, I.-R.; Park, J. G.; Xiao, D. J.; Harris, T. D. J. Am. Chem. Soc.
2013, 135, 16845−16848.
(11) (a) Karasawa, S.; Nakano, K.; Tanokashira, J.; Yamamoto, N.;
Yoshizaki, T.; Koga, N. Dalton Trans. 2012, 13656−13667.
(b) Karasawa, S.; Koga, N. Inorg. Chem. 2011, 50, 5186−5195.
(c) Yoshihara, D.; Karasawa, S.; Koga, N. Polyhedron 2011, 30, 3211−
3217. (d) Yoshihara, D.; Karasawa, S.; Koga, N. J. Am. Chem. Soc.
2008, 130, 10460−10461. (e) Karasawa, S.; Yoshihara, D.; Watanabe,
N.; Nakano, M.; Koga, N. Dalton Trans. 2008, 1418−1421.
(f) Tobinaga, H.; Suehiro, M.; Itoh, T.; Zhou, G.; Karasawa, S.;
AUTHOR INFORMATION
Corresponding Author
*Tel: 81-92-642-6590. Fax: 81-92-642-6590. E-mail: koga@fc.
phar.kyushu-u.ac.jp.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Grants-in-Aid for Scientific
Research (B)(2) (Nos. 17350070 and 25288038) from the
Ministry of Education, Science, Sports and Culture, Japan, and
by the “Nanotechnology Support Project” of the Ministry of
Education, Culture, Sports, Science and Technology (MEXT),
Japan.
REFERENCES
■
(1) (a) Gatteschi, D.; Sessoli, R.; Villain, J. Molecular Nanomagnets;
Oxford University Press: New York, 2006. (b) Christou, G.; Gatteschi,
D.; Hendrickson, D. N.; Sessoli, R. MRS Bull. 2000, 25, 66−71.
J
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Koga, N. Polyhedron 2007, 26, 1905−1911. (g) Karasawa, S.; Zhou, G.;
Morikawa, H.; Koga, N. J. Am. Chem. Soc. 2003, 125, 13676−13677.
(h) Koga, N.; Karasawa, S. Bull. Chem. Soc. Jpn. 2005, 78, 1384−1400.
(12) (a) Kanegawa, S.; Karasawa, S.; Maeyama, M.; Nakano, M.;
Koga, N. J. Am. Chem. Soc. 2008, 130, 3079−3094. (b) Kanegawa, S.;
Karasawa, S.; Nakano, M.; Koga, N. Bull. Chem. Soc. Jpn. 2006, 79,
1372−1382. (c) Kanegawa, S.; Karasawa, S.; Nakano, M.; Koga, N.
Chem. Commun. 2004, 3590−3591.
(13) (a) Kuzniarska-Biernacka, I.; Bartecki, A.; Kurzak, K. Polyhedron
2003, 22, 997−1007. (b) Holland, J. M.; Kilner, C. A.; Thornton-Pett,
M.; Halcrow, M. A. Polyhedron 2001, 20, 2829−2840.
(14) (a) Lever, A. B. P. Inorganic Electronic Spectroscopy, 2nd ed.;
Elsevier Publishers: Amsterdam, 1984. (b) Cotton, F. A.; Goodgame,
D. M. L.; Goodgame, D. J. Am. Chem. Soc. 1961, 83, 4690−4699.
(15) Yamanoi, Y.; Sakamono, Y.; Kusukawa, T.; Fujita, M.;
Sakamoto, S.; Yamaguchi, K. J. Am. Chem. Soc. 2001, 123, 980−981.
(16) Kahn, O. Molecular Magnetism; Wiley-VCH Publishers:
Weinheim, 1993.
(17) (a) Sarkar, M.; Aromí, G.; Cano, J.; Bertolasi, V.; Ray, D.
Chem.−Eur. J. 2010, 16, 13825−13833. (b) Pardo, E.; Ruiz-García, R.;
Lloret, F.; Faus, J.; Julve, M.; Journaux, Y.; Novak, M. A.; Delgado, F.
́
S.; Ruiz-Perez, C. Chem.−Eur. J. 2007, 13, 2054−2066. (c) Rizzi, A. C.;
Brondino, C. D.; Calvo, R.; Baggio, R.; Garland, M. T.; Rapp, R. E.
Inorg. Chem. 2003, 42, 4409−4416. (d) Caneschi, A.; Gatteschi, D.;
Lalioti, N.; Sessoli, R.; Sorace, L.; Tangoulis, V.; Vindigni, A. Chem.−
Eur. J. 2002, 8, 286−292.
(18) Shores, M. P.; Sokol, J. J.; Long, J. R. J. Am. Chem. Soc. 2002,
124, 2279−2292.
(19) (a) Miyasaka, H.; Clerac, R.; Mizushima, K.; Sugiura, K.;
́
Yamashita, M.; Wernsdorfer, W.; Coulon, C. Inorg. Chem. 2003, 42,
8203−8213. (b) Aubin, S. M. J.; Sun, Z. M.; Pardi, L.; Krzystek, J.;
Folting, K.; Brunel, L. C.; Rheingold, A. L.; Christou, G.; Hendrickson,
D. N. Inorg. Chem. 1999, 38, 5329−5340.
(20) (a) Langley, S. K.; Chilton, N. F.; Moubaraki, B.; Murray, K. S.
Dalton Trans. 2011, 12201−12209. (b) Moushi, E. E.; Stamatatos, T.
C.; Wernsdorfer, W.; Nastopoulos, V.; Christou, G.; Tasiopoulos, A. J.
Inorg. Chem. 2009, 48, 5049−5051. (c) Murugesu, M.; Takahashi, S.;
Wilson, A.; Abboud, K. A.; Wernsdorfer, W.; Hill, S.; Christou, G.
Inorg. Chem. 2008, 47, 9459−9470. (d) Stamatatos, T. C.; Abboud, K.
A.; Wernsdorfer, W.; Christou, G. Angew. Chem., Int. Ed. 2007, 46,
884−888.
(21) (a) Cole, K. S.; Cole, R. H. J. Chem. Phys. 1941, 9, 341−351.
(b) Dekker, C.; Arts, A. F. M.; De Wijn, H. W.; Van Duyneveldt, A. J.;
Mydosh, J. A. Phys. Rev. B 1989, 40, 11243−11251.
(22) (a) Hendrickson, D. N.; Christou, G.; Ishimoto, H.; Yoo, Y.;
Brechin, E. K.; Yamaguchi, A.; Rumberger, E. M.; Aubin, S. M. J.; Sun,
Z. M.; Aromi, G. Polyhedron 2001, 20, 1479−1488. (b) Thomas, L.;
Caneschi, A.; Barbara, B. Phys. Rev. Lett. 1999, 83, 2398−2401.
(23) (a) Cirera, J.; Ruiz, E.; Alvarez, S.; Neese, F.; Kortus, J. Chem.−
Eur. J. 2009, 15, 4078−4087. (b) Ruiz, E.; Cirera, J.; Cano, L.; Alvarez,
S.; Loose, C.; Kortus, J. Chem. Commun. 2008, 52−54. (c) Waldmann,
O. Inorg. Chem. 2007, 46, 10035−10037.
(24) Morikawa, H.; Karasawa, S.; Koga, N. Appl. Magn. Reson. 2003,
23, 507−515.
(25) In the 1:4 complex Co(Cl)2(C1py)4, the exchange coupling
parameter, JCo‑carbene, between the cobalt ion and the carbene was
estimated to be 51 K.11e
K
dx.doi.org/10.1021/ic403074d | Inorg. Chem. XXXX, XXX, XXX−XXX