Exchange Interactions Switch Tunneling: A Comparative Experimental and Theoretical Study on Relaxation Dynamics by Targeted Metal Ion Replacement

Article abstract of DOI:10.1002/chem.201801523

The magnetic relaxation and magnetization blocking barriers of tailor-made homo- and heterodinuclear compounds [Dy₂(opch)₂(OAc)₂(H₂O)₂]?MeOH (1) and [DyMn(opch)₂(OAc)(MeOH)(H₂O)<

Full text of DOI:10.1002/chem.201801523

A Journal of
Accepted Article
Title: Exchange Interactions Switch Tunneling: A Comparative
Experimental and Theoretical Study on Relaxation Dynamics via
Targeted Metal Replacement
Authors: Haiquan Tian, Liviu Ungur, Lang Zhao, Shuai Ding, Jinkui
Tang, and Liviu Chibotaru
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201801523
Link to VoR: http://dx.doi.org/10.1002/chem.201801523
Supported by

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
Exchange Interactions Switch Tunneling: A Comparative
Experimental and Theoretical Study on Relaxation Dynamics via
Targeted Metal Replacement
Haiquan Tian,*[a, b] Liviu Ungur,*[c] Lang Zhao,[a] Shuai Ding,[d] Jinkui Tang,*[a] and Liviu F. Chibotaru[c]
Abstract: The magnetic relaxation and magnetization blocking
limit for spin-based devices, making them almost immediately
ideal candidate in high-density data storage technologies and
molecular spintronics since its discovery in the early 1990s.[1]
The requisite for such a system is the energy barrier towards
reversal of the magnetization (U), which is derived from a
combination of an appreciable spin ground-state (S) and
magneto-anisotropy (D).[2] Apparently, the success of such
technological applications hinges upon raising the blocking
temperature (TB) as well as the inherent anisotropic barriers (U).
Throughout the development of SMMs,[3] transition-metal
elements have acted as pioneers during a relatively long period
with several remarkable results, such as magnetic hysteresis of
4.5 K and magnetization reversal of 84.6 K in a [Mn6]
barriers of tailor-made homo- and hetero-dinuclear compounds
[Dy2(opch)2(OAc)2(H2O)2]∙MeOH
(1)
and
[DyMn(opch)2(OAc)(MeOH) (H2O)2] (2) were systematically
investigated and the change of SMM behavior originating from the
targeted replacement of one dysprosium site in Dy2 compound with
manganese was successfully elucidated through combined
experimental and theoretical study. A detailed comparative study on
theses close-related model compounds reveal the remarkable
changes of the crystal field splitting and anisotropy of Dy site and the
total exchange spectrum due to the replacement of Dy with Mn. And
the blocking barriers in these two compounds were analyzed which
explain their different blocking behavior. The two Ising doublets
arising from the magnetic interaction in the case of 1 are strongly
uniaxial, with tunneling splittings smaller than 10-6 cm-1, leads to
magnetic relaxation at temperatures exceeding the exchange energy
(2.13 cm-1) which involves transition via the excited states,
corresponding to local transitions on the excited doublet on Dy site.
While the third and fourth exchange doublets in 2 (placed at 2.16
and 3.25 cm-1, respectively) display much larger tunneling splittings
(of 10-4 and 10-3 cm-1 respectively), opening thus an important path
for magnetic relaxation.
compound.[4] In particular,
a two-coordinate cobalt imido
compound was successfully explored based on the highly
covalent Co=N core, where a high effective energy barrier of
magnetization reversal of 594 K can be achieved and displaying
magnetic blocking below 9.5 K.[5] By contrast, lanthanide-SMMs
have developed at an extraordinary pace since the discovery
that a mononuclear TbIII compound shows slow relaxation of the
magnetization in 2003.[6] Accompanied by the establishment of
the most important landmark, the threshold of 1000 K of spin
relaxation barrier has been crossed by several outstanding
monometallic examples in the last two years, including a few
[Dy(bbpen)Br],[7]
[Dy(OtBu)2(py)5][BPh4],[8]
[(Cpttt)2Dy][B(C6F5)4][9] compounds. The latter represents two
champions of the energy barrier and the magnetic hysteresis
temperature with 1837 K and 60 K, respectively. By modulating
the hyperfine interactions non-Kramers HoIII single-ion magnet
(SIM) was able to suppress the fast quantum tunneling of the
magnetization (QTM) at zero field to display a high energy
barrier (341 K).[10] Another striking achievement is that the
exceptionally strong magnetic exchange brought about by the
Introduction
Single-molecular magnets (SMMs) that exhibit magnetic
bistability of purely molecular origin approach the ultimate size
[a]
[b]
Dr. H. Tian, Dr. L. Zhao, Prof. Dr. J. Tang
State Key Laboratory of Rare Earth Resource Utilization,
Changchun Institute of Applied Chemistry, Chinese Academy of
Sciences
Changchun 130022, P. R. China
E-mail: tang@ciac.ac.cn
3-
diffuse spin of an N2 radical-bridge in dinuclear compounds
hinders the QTM, creating a record blocking temperature of 14
K.[11]
Dr. H. Tian
In the aspect of relaxation mechanism, those dinuclear Dy2
compounds have indisputably played a key role in elucidating
the relaxation mechanism.[12] Retrospectively, by virtue of the
diamagnetic Y2 matrix, the magnetic dilution method was
employed in order to elucidate the influence of the neighboring
DyIII ion in a single-ion relaxation mechanism based on a
centrosymmetric dinuclear Dy2 compound.[12b] The blocking
mechanism principally originates from the individual DyIII sites
and the exchange interaction between the sites in an
asymmetric Dy2 SMM was revealed by taking the advantage of
combined experimental and high level ab initio calculations.[12c]
On the other hand, the researches indicate that the utilization of
the stronger magnetic exchange between transition-metal and
lanthanide ions represents a promising route to integrate the
large magnetic anisotropies and high-spin ground states, mainly
because the synthesis of hetero 3d-4f system is relatively simple
Shandong Provincial Key Laboratory of Chemical Energy Storage
and Novel Cell Technology
School of Chemistry and Chemical Engineering
Liaocheng University
Liaocheng 252000, P. R. China
E-mail: tianhaiquan@lcu.edu.cn
[c]
[d]
Dr. L. Ungur, Prof. Dr. L. F. Chibotaru
Department of Chemistry
KU Leuven
Celestijnenlaan 200F, 3001 Leuven, Belgium
E-mail: Liviu.Ungur@kuleuven.be
E-mail: Liviu.Chibotaru@kuleuven.be
S. Ding
Shandong Juye County Coal Board,
Juye 274900, P. R. China
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
compared to that of radical-containing system. It is worth
mentioning that heterometallic [M2Ln2] (M = CoII, CrIII or RuIII, Ln
= GdIII, TbIII or DyIII) model system was successfully explored
where different transition-metal ions were embedded to
lanthanide system to elucidate the relaxation mechanism and to
enhance the blocking temperature.[13] Although the effects of
different trivalent 3d ions on the relaxation dynamics have been
unambiguously identified in above mentioned [M2Ln2] system,
the targeted replacement of lanthanide with 3d ion in elegant
dinuclear lanthanide model in order to probe the effects that 3d
ions (replacing 4f) have on the relaxation dynamics have not
been explored. A comparative investigation into such homo- and
hetero-dinuclear models will provide insight into the influence
that the 3d ions have on the system, and how they affect the
static and dynamic magnetic behavior.
obtained results were correlated with the calculated charges on
neighboring ligand atoms and the exchange interaction and the
exchange spectrum were subsequently described. Through such
a detailed comparative study, we were able to compare the 1)
structural; 2) magnetic; 3) crystal field splitting and anisotropy of
Dy site and 4) total exchange spectrum and exchange
parameters changes due to the replacement of Dy with Mn.
Finally, the blocking barriers in these two compounds were
analyzed which explain their different blocking behavior.
Results and Discussion
Structural Analysis of H2opch. The multidentate Schiff-
base H2opch ligand is obtained from the reaction of pyrazine-2-
carbohydrazide and o-vanillin aldehyde. This ligand provides
N,O,N,O,O-based multichelating sites that are especially
favorable for the formation of lanthanide compounds. The
molecular structure of H2opch ligand determined by sing-crystal
X-ray diffraction is depicted in Figure S1 and the crystal data are
summarized in Table S1. The ligand crystallizes in the
monoclinic space group C2/c having eight molecules in each
unit cell. The free ligand is planar and found to be in keto form
as the C(9)-O(3) distance of 1.214 (5) Å corresponds to a
carbon-oxygen double bond. The adjacent molecules are linked
via intermolecular hydrogen bonds N2—H2···N4#1, C13—
H13···N3#1 and C1—H1C···O1#2 (symmetry codes: #1, 1.5-x,
0.5+y, 1.5-z; #2, 2.5-x, 0.5-y, 1-z), generating a two-dimensional
supramolecular plane, as shown in Figure S2.
We therefore selected the versatile (E)-N'-(2-hyborxy-3-
methoxybenzylidene)pyrazine-2-carbohydrazide
(H2opch)
hydra-zone ligand in order to assemble the aforementioned two
model complexes. In our previous studies, this H2opch ligand
has been successfully applied to construct four polynuclear DyIII-
based SMMs thanks to their flexible coordination modes at
different reaction conditions, as depicted in Scheme 1.
Compounds 3 and 6 exhibit the butterfly-shaped topologies with
peculiar mirror images of each other, and the latter has a larger
anisotropic energy gap of 197 K and a longer characteristic
relaxation time (τ) of 3.5 s.[14] Additionally double- and
2-
quadruple-CO3
bridged dinuclear DyIII cores have been
obtained (compounds 5 and 4) by spontaneous fixation of two
and four atmospheric CO2 molecules, respectively.[15]
Table 1. Selected bond lengths (Å) and angles (º) in compounds 1 and 2.
Compound 1
Dy1–O3
Dy1–O4
Dy1–O5
Dy1–O6
Dy1–O2a
2.324(1)
2.419(1)
2.363(1)
2.362(1)
2.145(1)
Dy1–O3a
2.338(1)
2.566(2)
2.478(2)
3.891(2)
113.1(2)
Dy1–N3
Dy1–N1a
Dy1···Dy1a
Dy1–O3–Dy1a
Compound 2
Scheme 1. Six SMMs originate from the H2opch ligand and different DyIII salts.
Mn1–O3
2.188(1)
2.314(1)
2.313(1)
2.242(1)
112.5(1)
Mn1–O10
Mn1–N4
2.079(1)
2.422(1)
2.354(2)
3.727(2)
106.6(1)
Mn1–O5
Herein we report the isolations of homo- and hetero-
Mn1–O6
Mn1–N5
dinuclear
[Dy2(opch)2(OAc)2(H2O)2]∙MeOH
(1)
and
[DyMn(opch)2(OAc) (MeOH)(H2O)2] (2). These two structurally
close-related model complexes provide unique opportunity to
probe simultaneously the contributions of 3d and 4f ions, as well
as their exchange interaction to the relaxation dynamics. Ab
initio calculations are applied to get insight into the influence that
the individual 3d, 4f ions and their interaction have on the
system, and how they affect the static and dynamic magnetic
behavior. Besides detailed experimental characterization of
structure and magnetism, a thorough description of the local
electronic and magnetic properties of Dy sites, including the
calculated crystal field parameters were given. In addition, the
Mn1–O9
Mn1···Dy1
Mn1–O5–Dy1
Mn1–O3–Dy1
Structural Analysis of 1. The reaction of Dy(OAc)3·6H2O
with H2opch (1:1 ratio) in MeOH/CH2Cl2 (1:1 ratio), in the
presence oftriethylamine (3 equivalents), produces golden
yellow crystals of [Dy2(opch)2(OAc)2(H2O)2]∙MeOH (1) in 46%
yield in one week, whose molecular structure determined by
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
shown in Figure S3. The shortest intermolecular Dy···Dy
distance is 8.563(1) Å.
Structural Analysis of 2. The reaction of Dy(OAc)3·6H2O,
Mn(ClO4)2·6H2O and the solution of H2opch in a mixture of
MeOH/CH2Cl2 (1:1 v/v), in the presence of triethylamine (3
equivalents), leads to the formation of dinuclear heterometallic
DyMn
compound
after
3
weeks,
namely,
[DyMn(opch)2(OAc)(MeOH)(H2O)2] (2). The single-crystal X-ray
diffraction studies revealed that 2 crystallizes in the monoclinic
space group P21/c with Z = 4. A perspective view of the
molecular structure of 2 is represented in Figure 1 bottom and
the crystal data are summarized in Table S1 and selected bond
lengths and angles are shown in Table 1 and S2. In the
asymmetric molecule, two opch2- polydentate Schiff-base
ligands provide N, O, N, O and O, N, O, O-based multi-chelating
sites, respectively, to chelate MnII and DyIII. In doing so two
metal centers are doubly bridged by the phenol oxygen atom
(O3) and the carboxy oxygen atom (O5) with a Mn···Dy distance
of 3.726(1) Å and two Mn-O-Dy angles of 112.5(1)° and
106.6(1)°, respectively. The coordination sphere of Mn1 is
completed by water and methanol molecule, generating a N2O5
environment with nearly perfect pentagonal bipyramidal
coordination geometry, whereas the Dy1 ion adopts a distorted
dodecahedral geometry based on the NO7 environment
resembling the DyIII ions of compound 1 (Figure 1 bottom). Here,
two H2opch ligands show 2.111121222 and 2.112121232
coordination modes in its di-deprotonated form (Scheme S1)
with the latter is first observed for this ligand. Meanwhile, this is
the first example of H2opch ligand coordinating to 3d metal ions.
The Mn-O bond lengths are in the range of 2.01(1)-2.318(1) Å
and the two Mn-N bond lengths are 2.358(2) and 2.425(2) Å,
respectively. Additionally, the Dy-O bond lengths are in the
range of 2.174(1)-2.469(1) Å and only one Dy-N bond length is
2.458(2) Å. Finally, strong intermolecular hydrogen-bonging
interactions produce a one-dimensional zigzag chain of the
molecules (Figure S4). The shortest intermolecular Mn···Dy
distance is 6.901(1) Å.
Comparison of Molecular Structures of 1 and 2. The Dy2
core of compound 1 and the DyMn core of compound 2 are
embedded in two same ligands with the main differences lie in
the following aspects: Firstly, only one kind of N2O6 environment
can be found in compound 1 with only one binding model of the
two H2opch ligands, while the N2O5 environment of the MnII ion
and eight coordination of the DyIII ion are present in compound 2
derived from the two binding models of the ligands. Among the
SMMs field, their magnetic properties are influenced not only by
the coordination environments of the metal ions, but also other
subtle effects induced by their nearest neighbors. Furthermore,
for the lanthanide-SMMs, the geometry of the core is strongly
correlated to the nature or directions of the easy axes of
anisotropic ions. According to this peculiarity, the SHAPE
software was implemented to quantify the geometry of the
coordination center cations.[18] A systematic analysis of the
resulting parameters (Table S3) reveals that an intermediate
geometry between triangular dodecahedron (D2d) and square
antiprism (D4d) can be observed for the octa-coordinated DyIII
ions of compounds 1 and 2. The hepta-coordinated MnII ion
Figure 1. The molecular structures of compounds 1 (top) and 2 (bottom).
Hydrogen atoms and lattice solvents are omitted for clarity. Color scheme:
Turquiose Dy, Dark Rose Mn, red O, blue N.
single-crystal X-ray diffraction is depicted in Figure 1 top.
Compound 1 crystallizes in the triclinic space group P-1 with Z =
1. Details for the structure solution and refinement are
summarized in Table S1, and selected bond lengths and angles
are listed in Tables 1 and S2. In this compound, molecular
structure exhibits a center of symmetry with a Dy···Dy distance
of 3.891(1) Å and two Dy-O-Dy angles of 113.1(2)°. Two opch2-
hydrazone ligands behave as a pair of tetradentate N2O2
ligands to coordinate two DyIII centers, in which the two DyIII ions
are doubly bridged by the carboxy oxygen atoms (O3 and O3a),
with Dy-O distances of 2.324(1) and 2.338(1) Å. Each of the DyIII
centers was further coordinated by one acetate anion in a
bidentate fashion, as well as by one terminal water molecule to
complete the N2O6 environment based on the distorted
dodecahedral geometry (Figure 1 top). A closer look at the
crystal structure of 1 reveals that two water molecules and two
acetate groups coordinate to Dy1 and Dy1a above and below
the opch2--DyIII plane, producing the hula hoop-like coordination
geometry which is believed to be crucial for the observation of
SMM behavior in related Dy2 compounds.[12c] The Dy-O bond
lengths are in the range of 2.145(1)-2.419(1) Å and the two Dy-N
bond lengths are 2.478(2) and 2.565(2) Å, respectively. Only
one 2.111121222 coordination mode (Harris notation[16]) can be
observed for H2opch ligand in its di-deprotonated form (Scheme
S1). Additionally one non-coordinated methanol molecule is
located in the crystal lattice. Finally, strong inter- and
intramolecular hydrogen-bonding interactions produce a two-
dimensional supramolecular plane with the acs net (4,4),[17] as
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
shows a nearly pentagonal bipyramid (D5h) geometry with the
shape measure deviation of 0.647. On the other hand,
scrutinizing these two inter-nuclear distances of the metal sites,
we observed that the Dy···Mn distance (3.728(1) Å) in
compound 2 is much closer than the Dy···Dy distance (3.891(1)
Å) in compound 1 when the transition-metal MnII ion successfully
infiltrated into this Dy2 compound, and which the short
intermolecular distance has been intensely affected with the
values of 8.563(1) Å and 6.901(1) Å for the compounds 1 and 2,
respectively.
progressive depopulation of excited Stark sublevels.[2f, 2m, 19] With
decreasing temperature further, the χMT value then increases
sharply to a maximum of 38.4 cm3 K mol-1 at 2 K, which
obviously suggesting the presence of intramolecular
ferromagnetic interactions between the metal centers, as
observed in other DyIII compounds.[20] For compound 2, the χMT
value of 15.6 cm3 K mol-1 at 300 K is lower than the theoretical
value of 18.55 cm3 K mol-1 (MnII: S = 5/2, C = 4.375 cm3 K mol-1
with g = 2; and S = 5/2, L = 5, 6H15/2, C = 14.17 cm3 K mol-1 with
g = 4/3). The χMT product gradually decreases with lowing
temperature, reaching a minimum value of 17.4 cm3 K mol-1 at
40 K, which is a typical decrease, induced by the depopulation
of excited Stark sublevels of the DyIII ions.[21] With the presence
of intramolecular ferromagnetic interactions between the metal
centers, the χMT value exhibits a sharp increase below 20 K (the
maximum of 20.8 cm3 K mol-1 at 2 K).[2n, 2o, 22] Obviously, the
disparity of two curves is most likely due to the nature of the
overall exchange interaction between MnII and DyIII ions for
compound 2, whereas the magnetic behavior of compound 1
depends only on the interaction between two DyIII ions.
The magnetizations of the two complexes from zero dc field
to 70 kOe at different temperatures are shown in the inset of
Figure 2, with the corresponding maximum values reaching at
1.9 K of 10.1 (1) and 10.0 μB (2), where μB is the Bohr
magneton. These values are lower than the expected saturation
value of 20 (1) and 15 μB (2) for two non-interacting DyIII ions (gJ
× J = 4/3 × 15/2 = 10 μB per DyIII), one isolated MnII (5 μB per MnII
with S = 5/2 and g = 2) and one DyIII ion, respectively. This is
most likely due to significant anisotropy and the crystal-field
effect in the system for compounds 1 and 2.[3c] At low fields, the
magnetization in compound 1 increases more rapidly than that
of compound 2. Meanwhile, the non-superposition of the M vs.
H/T data on a single master curve and the high-field non-
saturation suggests the presence of a significant magnetic
anisotropy and/or low lying excited states in compounds 1 and
2.[23]
Figure 2. Temperature dependence of the χMT products at 1000 Oe for
compounds
1 (top) and 2 (bottom). Inset: M vs. H/T plots at different
temperatures below 5 K.
Magnetic properties. Static susceptibility measurements on
polycrystalline samples of compounds 1 and 2 have been
carried out in the temperature range 2 – 300 K under an applied
field of 1000 Oe. The plots of χMT vs. T, where χM is the molar
magnetic susceptibility, are shown in Figure 2. The χMT
observed value at a room temperature of 28.1 cm3 K mol-1 for
compound 1 is in good agreement with the expected value of
28.34 cm3 K mol-1 for two uncoupled DyIII ions (S = 5/2, L =
6
5, H15/2, C = 14.17 cm3 K mol-1 with g = 4/3). The χMT value
decreases slightly down to a minimum value of 27.4 cm3 K mol-1
with decreasing temperature, which is mainly ascribed to the
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
Below 3 K, there is a temperature-independent characteristic
time of τQTM = 0.17 s, as expected in a pure quantum relaxation
regime.
For compound 2, no obvious out-of-phase signals were
observed above 1.9
K under zero dc field (Supporting
Information, Figure S8), indicating a very fast relaxation of the
magnetization. To investigate further the relaxation behavior and
study for quantum tunneling effects, the frequency-dependent ac
susceptibility was measured with the application of the dc fields
up to 2000 Oe at 1.9 K (see Figure S9). Remarkably, the
application of dc field has a strongly influence on their dynamic
behavior of magnetization, suggesting the presence of fast
quantum tunneling of the magnetization.[25] As seen in ac
susceptibility signals, the relaxation mode is slightly moved to
higher frequency with increasing field and the Cole-Cole plot
exhibits a series of evident asymmetric semicircles under
variable static fields (between 600 and 2000 Oe), as depicted in
Figure S10. All the above features indicate that applied field has
an important effect on relaxation process.[26]
Figure 3. Frequency dependence of the in-phase (top) and out-of-phase
(bottom) ac susceptibility of 1 under zero-dc field.
To further characterize the dynamics of magnetization, a
series of ac-susceptibilities measurements at different
frequencies and temperatures were carried out for compound 1
under a zero dc field, as depicted in Figures 3 and S5-S6. The
strong frequency-dependence of both in-phase (χ′) and out-of-
phase (χ″) susceptibilities reveal the onset of slow relaxation of
the magnetization that is typical of SMM behavior. The χ'T value
exhibits a rapid decrease at around 17 K for 1500 Hz, which is
coincident with emerging peak of the χ'' signal when the
magnetization is blocked by the anisotropy barriers. The Cole-
Cole plots of χ″ vs. χ′ were constructed and fitted to a
generalized Debye model to obtain α values and relaxation
times (τ) in the temperature range 3.0 – 19.0 K (Figure 5 top,
inset).
The result indicated a relatively narrow distribution of τ with
small α parameters (0.008–0.17, Figure S7).[11] From frequency
dependencies of the ac susceptibility, the magnetization
relaxation times were deduced in the temperature range of 1.9 –
17.0 K and the lnτ vs T–1 plots obtained from these data is given
in Figure 5 top. At high temperature, the relaxation follows a
thermally activated Orbach mechanism[24] with ∆ = 95.2 cm-1 and
τ0 = 4.2 × 10-8 s based on an Arrhenius law (τ = τ0exp(∆/kT)).
Figure 4. Frequency dependence of the in-phase (top) and out-of-phase ac
susceptibility of compound 2 under a 1000 Oe dc field.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
5
and 1.8 × 10-7 s for the low- and high-temperature domain,
respectively. The observation of multiple relaxation processes is
the result of the presence of distinct anisotropic center of single
DyIII ion and the exchange interaction between the DyIII ion and
MnII ion.[13a, 13b]
Comparison of Magnetic Properties of 1 and 2. The
temperature dependence of the susceptibility for compounds 1
and 2 shows similar intramolecular ferromagnetic interactions
between the metal centers at temperatures below ∼ 40 K, while
the χMT product conspicuously decreases from 300 to 40 K in
compound 2 and slight decline tendency observed along the
same temperature range in compound 1. This variation of χMT at
higher temperatures is directly correlated to the replacement of
DyIII ion with MnII ion. In ac susceptibility aspect, for compound 1,
a single slow relaxation of the magnetization process can be
detected, giving a thermal energy barrier for the reversal of
magnetization of 95.2 cm-1 without the external dc field. In
contrast, compound 2 exhibits a complex behavior of the two-
step relaxations by providing a small dc field. In order to interpret
the origin of their magnetic properties, we performed ab initio
theoretical calculations on 1-2, a method that is very applicable
for the investigation of lanthanide compounds.[2f, 2g, 28]
Table 2. Energies (cm-1) and g tensors of the low-lying Kramers doublets (KD)
of Dy sites in compounds 1 and 2 obtained within the largest computational
model employed.
KD
Dy in Dy2 (1)
Dy in DyMn (2)
E
g
E
g
gx
gy
gz
0.000
0.00656
0.01297
19.47883
0.000
0.00377
0.00746
19.60838
Figure 5. Magnetization relaxation time, lnτ, versus T–1 under 0 Oe (1, top)
and 1000 Oe (2, bottom) dc field, respectively. The solid line is fitted with the
Arrhenius law (see text); inset: Cole-Cole plots. The solid lines are the best fits
to the experimental data.
1
2
3
4
5
6
7
8
gx
gy
gz
191.727
333.817
353.559
415.434
442.544
513.722
579.318
0.20683
0.24975
16.64449
204.075
365.896
423.700
537.958
585.852
671.784
754.154
0.09764
0.14829
16.68307
In order to obtain quantitative information regarding the
relaxation barrier of compound 2, the frequency dependence of
χ″ susceptibilities under 1000 Oe applied dc field was examined
below 18.0 K, as shown in Figures 4 and S11-S13. This
behavior indicates the field-induced slow relaxation of the
magnetization. The maximum values of the temperatures and
frequencies are observed in the 1.9 – 8.0 K range. From these
gx
gy
gz
1.73682
3.00333
15.63005
2.43880
4.01876
15.24130
gx
gy
gz
1.66865
5.31728
11.80336
8.30491
5.18753
1.34074
data, the Cole-Cole plots (Figure
5
bottom, inset) were
gx
gy
gz
1.44494
3.80855
12.14542
8.37729
6.74977
2.28868
constructed with a series of asymmetric semicircles, moreover,
the generalized Debye model cannot fit these data well and
gives a large α ≈ 0.24 (see Figure S12), indicating a wide
distribution of τ.[27] All the above features suggest the presence
of multiple relaxation pathways. These distinctly larger α values
compared to those of compound 1 may be due to the
substitution of DyIII by MnII ion and the slight changes in the
ligand field caused by the replacement of the acetate with
methanol. To quantify the relaxation barrier, the relaxation time
was then extracted from the frequency-dependent data between
2.2 and 18.0 K and the Arrhenius plot obtained from these data
is given in Figure 5 bottom. It is interesting to note that two
relaxation regimes are clearly visible with a transition between
them corresponding to ∆ of 3.7 and 36.2 cm-1 and τ0 of 4.9 × 10-
gx
gy
gz
1.19377
2.07300
16.24514
1.92027
3.15284
15.15655
gx
gy
gz
0.44372
1.15980
15.03593
0.30107
0.52062
17.15311
gx
gy
gz
0.23857
0.53052
18.07661
0.12276
0.14069
19.33366
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
Computational Details. Fragment ab initio CASSCF
calculations were performed on the compounds 1 and 2 with
MOLCAS program package.[29] Two basis set approximations
have been employed a) a medium-sized DZP-quality and b) a
large TZP-quality basis set corresponding to the following
All BS-DFT calculations were done with ORCA program
package[31] using the B3LYP exchange correlation functional. All
atoms were described by relativistic Split-Valence + Polarization
basis sets (SVP-DKH) or by large relativistic def2-Triple-Zeta
Valence
+
Polarization basis (def2-TZVPP).[32] Relativistic
contractions: for the DZP basis: [7s6p4d2f1g] for DyIII ions, [3s2p] corrections were accounted for in the Douglas-Kroll-Hess
for all N and C, and [2s] for H; for the basis TZP: [8s7p5d4f2g1h]
for DyIII ions, [4s3p1d] for N and C from the first and second
coordination spheres, [3s2p] for distant C, and [2s] for H (see
Table S2). For DyIII, the active space of the CASSCF method
included the 9 electrons from the last shell spanning seven 4f
orbitals, CAS(9,7). All active molecular orbitals contain ~99.8%
contribution from the 4f basis functions of the Dy site. State-
average CASSCF calculations were performed for all possible
spin states. All spin sexted states (6H, 6F and 6P manifolds),
some of the spin quartet states (128 out of 224) and spin doublet
states (130 out of 490) were further mixed by the spin-orbit
interaction within the restricted active space state interaction
(RASSI) method, resulting in a spin-orbital spectrum comprising
898 states, grouped in doublets (Kramers doublets, KD). On the
basis of the resulting spin-orbital multiplets the SINGLE_ANISO
program[30] computed local magnetic properties (g-tensors, main
magnetic axes, local magnetic susceptibility, parameters of the
crystal-field for the ground atomic multiplet, etc). Exchange
spectrum and magnetism of the binuclear compounds was
simulated using the POLY_ANISO software,[30] on the basis of
the ab initio results. More details are given in the Experimental &
Computational Section.
formalism, truncated at the second order (default
implementation). The strictest convergence criteria for the self-
consistent
field
energy
minimization
was
employed
(VeryTightSCF) in combination with the most accurate angular
grids for the grid integration accuracy (Grid7).
Table 3. Exchange interactions between DyIII ions in compounds 1 and 2
obtained within the largest computational modes employed. zJ is the
parameter describing the intermolecular interaction. All values are given in cm-
1
.
Dy2 (1)
DyMn (2)
Jtotal* = Jexch + Jdipolar
Jexch
0.185
-0.023
0.208
-0.0024
-0.06
0.454
0.330
0.124
-0.0146
0.54
Jdipolar
zJ
*
Jexch from BS-DFT
* Only Jexch was fitted to experimental data, while the value of the dipole-
dipole interaction Jdipolar is given here just for illustration, estimated as the
difference Jtotal*-Jexch. In practice, the dipole-dipole magnetic Hamiltonian was
computed using the ab initio results and added to the Lines exchange
Hamiltonian. We mention here that this estimation of Jdipole was possible only
because the dipole-dipole interaction is close to Ising type in these two
particular cases, due to strong Ising anisotropy of the Dy sites in these
compounds.
Ab initio investigation of local electronic and magnetic
properties of Dy sites in 1 and 2. Ab initio calculations on the
individual DyIII magnetic sites were performed in order to find the
local electronic structure and magnetic anisotropy. Tables 2, S5
and S6 show the obtained spectrum of the low-lying Kramers
doublets arising from the crystal field splitting of the free ion J =
15/2 manifold of the DyIII and their magnetic anisotropy. We
notice the large energy separation between the ground and first
excited doublet, of ca 190-200 cm-1, as well as the strong
uniaxial magnetic anisotropy of the ground doublet states (gx,y
<< gz; see Figure S14). The main magnetic axis (gZ) of the
ground doublets on the dysprosium sites are making a small
angle with the shortest Dy-O2 chemical bonds (dashed lines in
Figure 6), lying approximately in the plane formed by the two
bridging oxygen atom and two dysprosium ions. The reason for
this orientation is the electrostatic and covalent effects arising
from the O2 oxygen ligand atom, which apparently produces the
strongest effect on the Dy site. The ab initio calculated LoProp
atomic charges[33] on all atoms are shown in Tables S7-S8. Note
that the O2 oxygen atom is not the one exerting the largest
electrostatic effect, in both compounds. The angle between the
main magnetic axes (gZ) of the ground and first excited Kramers
doublets is larger in 1 than in 2 (20.4° and 8.6°, respectively,
Figure 6. Calculated orientations of the local magnetic axes (dashed lines)
and the magnetic moments in the ground exchange state (green arrows) on
the metal sites in 1 and 2.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
Tables S9-S10), denoting a slighter larger contribution to
tunneling in the first excited state. For both investigated
compounds, the calculated excitation energy between the
ground and first excited state is higher than the value of a single
thermally activated relaxation regime extracted from the ac-
susceptibility data (95.2 cm-1). We can assume that in the high
temperature regime magnetic relaxation of the Dy2 compound
occurs via several relaxation processes, some of them having
lower activation barrier, which results in an effective activation
barrier extracted in ac measurements (Figures 7 top and 8 top).
Ab initio investigation of magnetic coupling in 1 and 2.
In order to get an estimation of the sign and magnitude of
exchange coupling in 1 and 2, we have performed BS-DFT
calculations (as described above). On the other hand, exchange
coupling was found from the fitting of experimental data within
the POLY_ANISO program. On the basis of calculated
exchange spectrum, the temperature-dependent magnetic
susceptibility (see Figure 2) and high-field magnetization at low
temperatures (see Figure 2, inset) are described. Tables 3 and
S11 show a comparison of the extracted exchange coupling
parameters in the investigated compounds.
interaction. BS-DFT calculations also predict ferromagnetic
coupling. Calculated energy spectrum for both compounds is
shown in Table 4. We notice that the two Ising doublets arising
from the magnetic interaction in the case of 1 are strongly
uniaxial, with tunneling splittings smaller than 10-6 cm-1. This
explains why 1 is showing blocking the magnetization. Magnetic
relaxation at temperatures exceeding the exchange energy (2.38
cm-1) must involve transition via the excited states,
corresponding in this case to local transitions on the excited
doublet on Dy site (Figure 7 bottom).
The low-lying exchange spectrum of 2 is formed by six Ising
doublets (Figure 8 bottom), originating from the exchange
interaction of the ground doublet on Dy and the S = 5/2 on Mn
(Table 4). Here we see that the third and fourth exchange
doublets in 2 (placed at 1.94 and 2.91 cm-1, respectively) display
much larger tunneling splittings (of 10-4 and 10-3 cm-1
respectively), opening thus an important path for magnetic
relaxation in this compound. We can associate this energy with
the extracted exchange barrier of magnetic relaxation (3.7 cm-1).
In the case of 1, we notice that both BS-DFT calculations
and the fitting of the experimental data predict quite small
antiferromagnetic exchange coupling (Tables 3 and S12).
However, the total magnetic interaction between Dy sites in 1 is
ferromagnetic due to the strong dipolar magnetic interaction
between large moments on the DyIII sites, which stabilizes their
parallel arrangement (Figure 6). The latter in its turn, originates
from a near-parallel alignment of the local anisotropy axes and
the axis connecting the DyIII ions.
Table 4. The low-lying exchange spectrum (cm-1) arising from the exchange
interaction of the lowest Kramers doublets on magnetic centers in the
compounds 1 and 2.
Dy2 (1)
DyMn (2)
E
∆
gz
E
∆
gz
t
t
0.0000
0.0000
0.0000
0.0000
5.7377E-07 38.96
1.0403E-06 0.00
3.6103E-06 35.56
2.8020E-05 35.56
1.9685E-05 0.00
1.5421E-05 0.00
1.5107E-03 27.30
6.900E-11
29.54
2.1389
2.1389
1.0822
1.0822
1.1333E-08
3.5167E-04
1.2282E-03
3.5175E-07
7.4000E-11
4.2800E-10
25.56
21.59
17.63
13.69
9.81
191.9828
191.9828
2.1646
2.1649
192.1815
192.1815
3.2469
3.2481
193.4588
193.4588
4.3305
4.3305
193.5632
193.5632
5.4138
5.4138
334.2725
334.2740
204.6030
204.6030
26.56
Figure 7. Top: the magnetization blocking barrier of Dy in Dy2 (1). The thick
black lines represent the Kramers doublets in accordance with the value of its
magnetic moment. The green dashed lines correspond to diagonal quantum
tunneling of magnetization (QTM); the blue dashed lines represent Orbach
In the case of 2, both exchange interaction is much stronger
than the magnetic dipolar coupling, which is natural for a 4f-3d
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
relaxation processes; the red arrows show the most probable path for
magnetic relaxation. The numbers at each arrow stand for the mean absolute
value of the corresponding matrix element of transition magnetic moment.
Bottom: the low-lying exchange spectrum arising from the exchange
interaction of the lowest Kramers doublets on magnetic center (∆tun are
corresponding tunneling gaps).
crystal X-ray diffraction demonstrates that the molecular
structures of compounds 1 and 2 are priori comparable.
Compound 1 shows typical single-molecule magnet behavior
with an energy gap (∆) of 95.2 cm-1 and a pre-exponential factor
(τ0) of 4.2 × 10-8 s under zero dc field, while no out-of-phase
alternating-current signal was observed for compound 2.
Through a detailed comparative study on these close-related
model compounds, the structural and magnetic properties,
especially the changes of the crystal field splitting and
anisotropy of Dy site and the total exchange spectrum due to the
replacement of Dy with Mn were investigated and finally the
blocking barriers in these two compounds were rationalized
which explain their different blocking behavior. The Ising
interaction in 1 suppress the tunneling efficiently and the
magnetic relaxation at temperatures exceeding the exchange
energy (2.13 cm-1) must involve transition via the excited states,
corresponding to local transitions on the excited doublet on Dy
site. While in the case of 2, the third and fourth exchange
doublets in placed at 2.16 and 3.25 cm-1, respectively, display
much larger tunneling splittings (of 10-4 and 10-3 cm-1
respectively), switching on thus an important path for magnetic
relaxation in this compound.
This analysis was made possible by taking advantage of
deliberate design of dinuclear model compounds with targeted
metal replacement. Such a deep exploration enhances the
understanding of magnetic interactions in elucidating the
relaxation dynamics of lanthanide-containing system and will
then direct the rational design of new SMMs with high
performance.
Experimental & Computational Section
General Considerations. All reagents employed in the experiments were
analytical grade and used without further purification. The elemental
analyses for C, H, and N were performed with a Perkin-Elmer 2400
analyzer. The IR measurements were recorded on a VERTEX 70 Fourier
transform infrared (FTIR) spectrophotometer using the reflectance
technique (4000-300 cm-1); the samples were prepared as KBr disks.
X-ray Crystallographic Analysis and Data Collection. Crystallographic
data and refinement details are given in Table 1. Suitable single crystals
with dimensions of 0.30 × 0.30 × 0.20 mm3, 0.30 × 0.25 × 0.21 mm3 and
0.25 × 0.25 × 0.30 mm3 for H2opch, 1 and 2, respectively, were selected
for single-crystal X-ray diffraction analysis. Crystallographic data were
collected at temperatures of 293 K (H2opch and 1) and 273 K (2) on a
Bruker ApexII CCD diffractometer with graphite monochromated Mo Kα
radiation (λ = 0.71073 Å). Data processing was accomplished with the
SAINT processing program. The structure was solved by the direct
methods and refined on F2 by full-matrix least squares using
SHELXTL97.[34] All non-hydrogen atoms were refined with anisotropic
thermal parameters. The H atoms were introduced in calculated positions
and refined with fixed geometry with respect to their carrier atoms. CCDC
1030999, 960070 and 960071 contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
Figure 8. Top: the magnetization blocking barrier of Dy in DyMn (2). The thick
black lines represent the Kramers doublets in accordance with the value of its
magnetic moment. The green dashed lines correspond to diagonal quantum
tunneling of magnetization (QTM); the blue dashed lines represent Orbach
relaxation processes; the red arrows show the most probable path for
magnetic relaxation. The numbers at each arrow stand for the mean absolute
value of the corresponding matrix element of transition magnetic moment.
Bottom: the low-lying exchange spectrum arising from the exchange
interaction of the lowest Kramers doublets on magnetic center (∆tun are
corresponding tunneling gaps).
Conclusions
A particularly telling pair of tailor-made centrosymmetric
dinuclear Dy2 compound and its heterometallic derivative DyMn
compound has been successfully developed to probe
systematically the magnetic relaxation mechanism. Single-
Magnetic Measurements. Magnetic susceptibility measurements were
performed in the temperature range 2–300 K using a Quantum Design
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
MPMS XL-7 SQUID magnetometer equipped with a 7 T magnet. The dc
measurements were collected from 300 to 2 K and the ac measurements
were carried out in a 3.0 Oe ac field oscillating at various frequencies
from 1 to 1500 Hz. The diamagnetic corrections for the compounds were
estimated using Pascal’s constants,[35] and magnetic data were corrected
for diamagnetic contributions of the sample holder.
individual metal sites, the resulting exchange matrix describes in fact the
anisotropic exchange interaction. To this matrix we also added the
dipolar-magnetic coupling between local magnetic moments.
Diagonalization of the total interaction matrix (exchange + dipolar)
provides the exchange (coupled) spectrum. These coupled eigenstates
are further used to compute all measurable properties (magnetic
susceptibility, molar magnetization, g-tensors of the low-lying coupled
eigenstates, etc.) of the entire binuclear systems within the
POLY_ANISO program.[30] The exchange parameters Jij are usually
extracted from the comparison between the calculated and measured
magnetic data (magnetic susceptibility, molar magnetization) through
minimization of the standard deviation, and are the only fitting
parameters used in this approach. In the present case, this is just one
fitting parameter describing the interaction between metal sites.
Intermolecular interaction is accounted for in the mean field
approximation, using one parameter describing the average exchange
interaction between the considered molecule and neighboring ones (zJ).
Computational Methodology. Electronic and magnetic properties of
individual metal centers in the Dy2 and DyMn compounds have been
studied by fragment ab initio calculations using the MOLCAS-8.0
program package.[29] We have employed the fragment approach because
of the limitations the current ab initio methods and our computers have,
being not yet suitable to treat at this moment such binuclear complexes
entirely ab initio. In this connection, the question arises to build suitable
mononuclear fragments from the molecules, and which would not change
significantly the energy structure on the magnetic centre. In order not to
introduce additional errors, the full molecular structure of both molecules
was employed. Since we aim at direct comparing the calculated
magnetism with experimental measured one on crystalline samples, no
geometry optimization on the fragments have been done, all atomic
coordinates being taken from the X-ray analysis. Magnetic interaction is
cancelled by substituting neighboring metal sites (containing unpaired
electrons) by their diamagnetic equivalents. In 1, the neighboring DyIII ion
has been simulated by the closed shell LuIII ion while in 2 the MnII site
An alternative approach to extract the exchange coupling constants
employs the broken-symmetry DFT calculations.[41] Since the orbitally
near-degenerate ground state of the DyIII sites complicates much the
DFT calculations, it has to be replaced computationally by the closest
atom with non-degenerate ground state. It was reported recently[13b] that
broken-symmetry calculations performed on the same molecular
structure as initial compounds involving computational substitution of the
Dy by Gd provides accurate values for the exchange couplings. In this
approach, the exchange values obtained for the isostructural Gd-Gd
compound (two interacting spins of 7/2) have to be rescaled for the Dy-
Dy (two interacting spins of 5/2) by a factor of 7/5 × 7/5 =49/25. For the
Dy-Mn the rescaling involves only the spin of one metal site, therefore,
the rescaling coefficient is 7/5.
has been simulated by the closed shell ZnII. In other words,
a
mononuclear Dy fragment is in fact the entire molecule where all
neighboring metal sites are substituted by their diamagnetic equivalents.
All atoms were described by all-electron relativistic ANO-RCC basis sets
available in MOLCAS package.[36] Relativistic basis sets are suitable for
the description of scalar relativistic effects by means of Douglas-Kroll-
Hess formalism.[37] These basis functions were used to optimize the
electronic spin states of the individual metal sites within the active space
of the complete active space self-consisting field (CASSCF) method.[38]
The active space of the CASSCF calculation of the Dy included the 4f
orbitals (CAS (9 in 7)) since we are interested in the ligand field states
only. The 4f orbitals are well localized on the Ln site, which allows us to
consider the metal-to-metal or metal-to-ligand charge transfer states
much higher in energy, thus being not relevant for the magnetism. The
optimized spin states for ground and excited manifolds within CASSCF
Synthesis of H2opch. The pyrazine-2-carboxylic acid methyl ester was
prepared by a literature procedure described elsewhere.[15a] A mixture of
pyrazine-2-carboxylic acid methyl ester (2.07 g, 15 mmol) and hydrazine
hydrate (85%, 20 ml) in methanol (30 ml) was refluxed overnight, at a
temperature somewhat below 80 °C. The resulting pale-yellow solution
set aside 15 h. During this period, a colorless product, i.e. pyrazine-2-
carbohydrazide, was precipitated from the reaction mixture as crystalline
solid (yield = 1.35 g, 65%). Pyrazine-2-carbohydrazide (0.276 g, 2 mmol)
was suspended together with o-vanillin (0.304 g, 2 mmol) in methanol (20
ml), and the resulting mixture was stirred at the room temperature
overnight. The pale yellow solid was collected by filtration. Then 0.054 g
product (0.2 mmol) was dissolved into 25 ml CH3OH/CH2Cl2 (1:1 v/v)
giving a pale yellow solution. Pale yellow single crystals, suitable for X-
calculations were further mixed by the spin-orbit interaction within the
39]
restricted active space state interaction (RASSI) method.[37b,
In this
approach, the spin-orbit interaction is included by means of the atomic
mean field approach (AMFI).[37b] Matrix elements of the orbital
momentum over the spin states are also computed within RASSI method.
Ab initio calculated spectrum and orbital momentum are further employed
by the SINGLE_ANISO module[30] in MOLCAS to calculate the powder
susceptibility, molar magnetization and the g-tensors for the ground and
several excited Kramers doublets of isolated metal fragments. In
calculations of magnetic properties, all spin-orbit multiplets included
within spin-orbit coupling in RASSI are taken into account. Therefore,
including a large number of spin states in the spin-orbit mixing in RASSI
method is important for an accurate description of magnetism, in
particular in cases of strong magnetic anisotropy.
1
ray diffraction analysis, were formed after 2 h (yield = 0.037 g, 68%). H
NMR (400 MHz, DMSO) δ 12.62 (s, 1H), 10.97 (s, 1H), 9.28 (d, J = 1.6
Hz, 1H), 8.94 (d, J = 2.8 Hz, 1H ), 8.85 (s, 1H), 8.80-8.81 (m, 1H), 7.12
(dd, J1 = 1.2 Hz, J2 = 8.0 Hz, 1H), 7.05 (dd, J1 = 1.2 Hz, J2 = 8.0 Hz, 1H),
6.87 (t, J = 8.0 Hz, 1H), 3.82 (s, 3H). Elemental analysis (%) calcd for
C
13H12N4O3: C, 57.35, H, 4.44, N, 20.58: found C, 57.64, H, 4.59, N,
20.39. IR (KBr, cm-1): 3415(w), 3258(w), 1677(vs), 1610(s), 1579(m).
1530(s), 1464(s), 1363(m), 1255(vs), 1153(s), 1051(w), 1021(s), 986(w),
906(m), 938(w), 736(s), 596(m), 498(w).
For the simulation of magnetic properties of binuclear compounds we
used an approach combining the calculated electronic and magnetic
properties of individual metal fragments with the model description of the
anisotropic exchange interaction between metal sites, achieved within
the Lines model.[40] In this model, the isotropic Heisenberg exchange
interaction is included between the true spins on neighboring metal sites
Synthesis of [Dy2(opch)2(OAc)2(H2O)2]∙MeOH (1). The solution of
Dy(OAc)3∙6H2O (0.357 g, 0.1 mmol) and H2opch (0.0405 g, 0.15 mmol)
in 20 ml CH3OH/CH2Cl2 (1:1 v/v) was stirred with Et3N (0.014 ml, 0.3
mmol) for 5 h. The resultant golden yellow solution was left unperturbed
to allow the slow evaporation of the solvent. Golden yellow single crystals,
suitable for X-ray diffraction analysis, were formed after one week. Yield:
18 mg (46%, based on metal salt). Elemental analysis (%) calcd for
�
̂ ̂
(ꢀꢁꢂꢃℎ = −ꢄꢁꢂꢃℎꢅ1ꢅ2, where S12 = 5/2 for Dy and Mn) in the absence of
spin-orbit coupling. Considering explicitly (non-perturbatively) the spin-
orbit interaction on the individual metal sites and expanding the
interaction matrix in terms of product of localized spin-orbitals on
C
32H38Dy2N8O14: C, 35.47, H, 3.53, N, 10.34: found C, 34.96, H, 3.41,
N, 10.42. IR (KBr, cm-1): 3389(br), 3057(w), 2841(w), 1607(vs), 1561(m),
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
[3] a) G. Christou, D. Gatteschi, D. N. Hendrickson, R. Sessoli, Mrs Bull 2000,
25, 66-71; b) J. M. Zadrozny, D. J. Xiao, M. Atanasov, G. J. Long, F.
Grandjean, F. Neese, J. R. Long, Nat Chem 2013, 5, 577-581; c) C. M.
Zaleski, E. C. Depperman, J. W. Kampf, M. L. Kirk, V. L. Pecoraro, Angew.
Chem. Int. Ed. 2004, 43, 3912-3914; d) E. Coronado, P. Day, Chem. Rev.
2004, 104, 5419-5448; e) L. Bogani, W. Wernsdorfer, Nat Mater 2008, 7,
179-186; f) A. Dei, D. Gatteschi, Angew. Chem. Int. Ed. 2011, 50, 11852-
11858; g) S. V. Eliseeva, J. C. G. Bunzli, New Journal of Chemistry 2011,
35, 1165-1176; h) D. N. Woodruff, R. E. P. Winpenny, R. A. Layfield,
Chem. Rev. 2013, 113, 5110-5148; i) D. Q. Wu, D. Shao, X. Q. Wei, F. X.
Shen, L. Shi, D. Kempe, Y. Z. Zhang, K. R. Dunbar, X. Y. Wang, J. Am.
Chem. Soc. 2017, 139, 11714-11717; j) J. K. Tang, I. Hewitt, N. T. Madhu,
G. Chastanet, W. Wernsdorfer, C. E. Anson, C. Benelli, R. Sessoli, A. K.
Powell, Angew. Chem. Int. Ed. 2006, 45, 1729-1733; k) S. G. McAdams,
A. M. Ariciu, A. K. Kostopoulos, J. P. S. Walsh, F. Tuna, Coordin Chem
Rev 2017, 346, 216-239; l) B. S. Dolinar, D. I. Alexandropoulos, K. R.
Vignesh, T. James, K. R. Dunbar, J. Am. Chem. Soc. 2018, 140, 908-911;
m) P. Abbasi, K. Quinn, D. I. Alexandropoulos, M. Damjanovic, W.
Wernsdorfer, A. Escuer, J. Mayans, M. Pilkington, T. C. Stamatatos, J.
Am. Chem. Soc. 2017, 139, 15644-15647.
1519(w), 1478(s), 1465(w), 1454(s), 1417(m), 1381(m), 1341(w),
1303(m), 1285(w), 1245(m), 1216(s), 1159(m), 1087(w), 1031(w),
1019(w), 971(w) 872(w), 741(s), 644(w), 589(w), 431(m).
Synthesis of [DyMn(opch)2(OAc)(MeOH)(H2O)2] (2). The solution of
Dy(OAc)3∙6H2O (0.0357 mg, 0.1 mmol) and H2opch (0.0405 mg, 0.15
mmol) in 20 ml CH3OH/CH2Cl2 (1:1 v/v) was stirred with Et3N (0.014 ml,
0.3 mmol). After 3 h, Mn(ClO4)2∙6H2O (36.2 mg, 0.1 mmol) was added to
the solution. Then the mixture was stirred for another 2 h, followed by
filtration. The solution was left unperturbed to allow the slow evaporation
of the solvent. Dark red single crystals, suitable for X-ray diffraction
analysis, were formed after three weeks. Yield: 24 mg (52%, based on
Dy(OAc)3∙6H2O). Elemental analysis (%) calcd for C29H32DyMnN8O11
:
C, 39.31, H, 3.64, N, 12.65: found C, 39.47, H, 3.56, N, 12.81. IR (KBr,
cm-1): 3352(br), 3047(w), 2944(w), 2861(w), 1605(m), 1578(s), 1562(m),
1477(s), 1463(w), 1455(s), 1417(m), 1389(m), 1346(w), 1301(m),
1289(w), 1247(m), 1217(s), 1162(m), 1079(w), 1033(w), 972(w), 921(w),
869(w), 781(w), 739(s), 689(w), 665(w), 627(w), 587(w), 457(w), 429(m).
[4] C. J. Milios, A. Vinslava, W. Wernsdorfer, S. Moggach, S. Parsons, S. P.
Perlepes, G. Christou, E. K. Brechin, J. Am. Chem. Soc. 2007, 129, 2754-
+.
[5] X. N. Yao, J. Z. Du, Y. Q. Zhang, X. B. Leng, M. W. Yang, S. D. Jiang, Z.
X. Wang, Z. W. Ouyang, L. Deng, B. W. Wang, S. Gao, J. Am. Chem. Soc.
2017, 139, 373-380.
[6] N. Ishikawa, M. Sugita, T. Ishikawa, S. Koshihara, Y. Kaizu, J. Am. Chem.
Soc. 2003, 125, 8694-8695.
Acknowledgements
[7] J. Liu, Y. C. Chen, J. L. Liu, V. Vieru, L. Ungur, J. H. Jia, L. F. Chibotaru,
Y. H. Lan, W. Wernsdorfer, S. Gao, X. M. Chen, M. L. Tong, J. Am. Chem.
Soc. 2016, 138, 5441-5450.
[8] Y. S. Ding, N. F. Chilton, R. E. P. Winpenny, Y. Z. Zheng, Angew. Chem.
Int. Ed. 2016, 55, 16071-16074.
[9] a) F. S. Guo, B. M. Day, Y. C. Chen, M. L. Tong, A. Mansikkamaki, R. A.
Layfield, Angew. Chem. Int. Ed. 2017, 56, 11445-11449; b) C. A. P.
Goodwin, F. Ortu, D. Reta, N. F. Chilton, D. P. Mills, Nature 2017, 548,
439-442.
[10] Y. C. Chen, J. L. Liu, W. Wernsdorfer, D. Liu, L. F. Chibotaru, X. M. Chen,
M. L. Tong, Angew. Chem. Int. Ed. 2017, 56, 4996-5000.
This work was supported by the National Natural Science
Foundation of China (21525103, 21521092 and 21331003). L.U.
is a post-doc of the Fonds Wetenschappelijk Onderzoek-
Vlaanderen, and also gratefully acknowledges INPAC and
Methusalem grants of K.U. Leuven. The computational
resources and services used for the ab initio calculations were
provided by the VSC (Flemish Supercomputer Center) and
funded by the Hercules Foundation and the Flemish
Government department EWI.
[11]J. D. Rinehart, M. Fang, W. J. Evans, J. R. Long, J. Am. Chem. Soc. 2011,
133, 14236-14239.
[12] a) F. Habib, M. Murugesu, Chem. Soc. Rev. 2013, 42, 3278-3288; b) F.
Habib, P. H. Lin, J. Long, I. Korobkov, W. Wernsdorfer, M. Murugesu, J.
Am. Chem. Soc. 2011, 133, 8830-8833; c) Y. N. Guo, G. F. Xu, W.
Wernsdorfer, L. Ungur, Y. Guo, J. K. Tang, H. J. Zhang, L. F. Chibotaru, A.
K. Powell, J. Am. Chem. Soc. 2011, 133, 11948-11951; d) P. Zhang, M.
Perfetti, M. Kern, P. P. Hallmen, L. Ungur, S. Lenz, M. R. Ringenberg, W.
Frey, H. Stoll, G. Rauhut, J. van Slageren, Chem. Sci. 2018, 9, 1221-
1230; e) E. M. Pineda, N. F. Chilton, R. Marx, M. Dorfel, D. O. Sells, P.
Neugebauer, S. D. Jiang, D. Collison, J. van Slageren, E. J. L. McInnes,
R. E. P. Winpenny, Nature Communications 2014, 5; f) R. A. Layfield, J. J.
W. McDouall, S. A. Sulway, F. Tuna, D. Collison, R. E. P. Winpenny,
Chem. Eur. J. 2010, 16, 4442-4446; g) F. Habib, G. Brunet, V. Vieru, I.
Korobkov, L. F. Chibotaru, M. Murugesu, J. Am. Chem. Soc. 2013, 135,
13242-13245.
Keywords: single-molecule magnet • ab initio calculations •
magnetization blocking barrier • exchange interaction •
relaxation dynamics
[1] a) A. Caneschi, D. Gatteschi, R. Sessoli, A. L. Barra, L. C. Brunel, M.
Guillot, J. Am. Chem. Soc. 1991, 113, 5873-5874; b) R. Sessoli, H. L.
Tsai, A. R. Schake, S. Y. Wang, J. B. Vincent, K. Folting, D. Gatteschi, G.
Christou, D. N. Hendrickson, J. Am. Chem. Soc. 1993, 115, 1804-1816; c)
R. Sessoli, D. Gatteschi, A. Caneschi, M. A. Novak, Nature 1993, 365,
141-143.
[13] a) K. C. Mondal, A. Sundt, Y. H. Lan, G. E. Kostakis, O. Waldmann, L.
Ungur, L. F. Chibotaru, C. E. Anson, A. K. Powell, Angew. Chem. Int. Ed.
2012, 51, 7550-7554; b) S. K. Langley, D. P. Wielechowski, V. Vieru, N. F.
Chilton, B. Moubaraki, B. F. Abrahams, L. F. Chibotaru, K. S. Murray,
Angew. Chem. Int. Ed. 2013, 52, 12014-12019; c) S. K. Langley, D. P.
Wielechowski, V. Vieru, N. F. Chilton, B. Moubaraki, L. F. Chibotaru, K. S.
Murray, Chem. Commun. 2015, 51, 2044-2047; d) A. Upadhyay, C. Das,
S. Vaidya, S. K. Singh, T. Gupta, R. Mondol, S. K. Langley, K. S. Murray,
G. Rajaraman, M. Shanmugam, Chem. Eur. J. 2017, 23, 4903-4916.
[14]H. Q. Tian, L. Zhao, H. F. Lin, J. K. Tang, G. S. Li, Chem. Eur. J. 2013, 19,
13235-13241.
[15] a) H. Q. Tian, L. Zhao, Y. N. Guo, Y. Guo, J. K. Tang, Z. L. Liu, Chem.
Commun. 2012, 48, 708-710; b) H. Q. Tian, M. Wang, L. Zhao, Y. N. Guo,
Y. Guo, J. K. Tang, Z. L. Liu, Chem. Eur. J. 2012, 18, 442-445.
[16] R. A. Coxall, S. G. Harris, D. K. Henderson, S. Parsons, P. A. Tasker, R.
E. P. Winpenny, J. Chem. Soc. Dalton 2000, 2349-2356.
[2] a) D. Gatteschi, R. Sessoli, J. Villain, Molecular nanomagnets, Oxford
University Press, Oxford ; New York, 2006; b) D. E. Freedman, W. H.
Harman, T. D. Harris, G. J. Long, C. J. Chang, J. R. Long, J. Am. Chem.
Soc. 2010, 132, 1224-+; c) M. N. Leuenberger, D. Loss, Nature 2001, 410,
789-793; d) S. Hill, R. S. Edwards, N. Aliaga-Alcalde, G. Christou,
Science 2003, 302, 1015-1018; e) M. Mannini, F. Pineider, C. Danieli, F.
Totti, L. Sorace, P. Sainctavit, M. A. Arrio, E. Otero, L. Joly, J. C. Cezar, A.
Cornia, R. Sessoli, Nature 2010, 468, 417-421; f) M. L. Kahn, R. Ballou, P.
Porcher, O. Kahn, J. P. Sutter, Chem. Eur. J. 2002, 8, 525-531; g) L. F.
Chibotaru, L. Ungur, C. Aronica, H. Elmoll, G. Pilet, D. Luneau, J. Am.
Chem. Soc. 2008, 130, 12445-12455; h) H. Oshio, M. Nihei, S. Koizumi, T.
Shiga, H. Nojiri, M. Nakano, N. Shirakawa, M. Akatsu, J. Am. Chem. Soc.
2005, 127, 4568-4569; i) J. R. Friedman, M. P. Sarachik, J. Tejada, R.
Ziolo, Phys Rev Lett 1996, 76, 3830-3833; j) C. Aronica, G. Pilet, G.
Chastanet, W. Wernsdorfer, J. F. Jacquot, D. Luneau, Angew. Chem. Int.
Ed. 2006, 45, 4659-4662; k) R. Sessoli, A. K. Powell, Coordin Chem Rev
2009, 253, 2328-2341; l) M. A. AlDamen, J. M. Clemente-Juan, E.
Coronado, C. Marti-Gastaldo, A. Gaita-Arino, J. Am. Chem. Soc. 2008,
130, 8874-8875; m) M. L. Kahn, J. P. Sutter, S. Golhen, P. Guionneau, L.
Ouahab, O. Kahn, D. Chasseau, J. Am. Chem. Soc. 2000, 122, 3413-
3421; n) G. Karotsis, S. Kennedy, S. J. Teat, C. M. Beavers, D. A. Fowler,
J. J. Morales, M. Evangelisti, S. J. Dalgarno, E. K. Brechin, J. Am. Chem.
Soc. 2010, 132, 12983-12990; o) V. Mereacre, Y. H. Lan, R. Clerac, A. M.
Ako, I. J. Hewitt, W. Wernsdorfer, G. Buth, C. E. Anson, A. K. Rowell,
Inorg. Chem. 2010, 49, 5293-5302.
[17] S. R. Batten, S. M. Neville, D. R. Turner, Coordination Polymers: Design,
Analysis and Application 2009, 19-58.
[18] D. Casanova, M. Llunell, P. Alemany, S. Alvarez, Chem. Eur. J. 2005, 11,
1479-1494.
[19] O. Kahn, Molecular magnetism, VCH, New York, NY, 1993.
[20] a) I. J. Hewitt, Y. H. Lan, C. E. Anson, J. Luzon, R. Sessoli, A. K. Powell,
Chem. Commun. 2009, 6765-6767; b) K. Katoh, T. Kajiwara, M. Nakano,
Y. Nakazawa, W. Wernsdorfer, N. Ishikawa, B. K. Breedlove, M.
Yamashita, Chem. Eur. J. 2011, 17, 117-122; c) B. Hussain, D. Savard, T.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
J. Burchell, W. Wernsdorfer, M. Murugesu, Chem. Commun. 2009, 1100-
1102.
M. Stenrup, D. G. Truhlar, L. Ungur, A. Valentini, S. Vancoillie, V.
Veryazov, V. P. Vysotskiy, O. Weingart, F. Zapata, R. Lindh, J. Comput.
Chem. 2016, 37, 506-541.
[21] a) V. Mereacre, M. N. Akhtar, Y. H. Lan, A. M. Ako, R. Clerac, C. E.
Anson, A. K. Powell, Dalton. Trans. 2010, 39, 4918-4927; b) C. M. Liu, D.
Q. Zhang, D. B. Zhu, Dalton. Trans. 2010, 39, 11325-11328; c) S. K.
Langley, B. Moubaraki, K. S. Murray, Dalton. Trans. 2010, 39, 5066-5069;
d) C. M. Zaleski, J. W. Kampf, T. Mallah, M. L. Kirk, V. L. Pecoraro, Inorg.
Chem. 2007, 46, 1954-1956.
[30] L. Ungur, L. F. Chibotaru, University of Leuven, 2006-2017.
[31] F. Neese, Wires Comput. Mol. Sci. 2012, 2, 73-78.
[32] D. A. Pantazis, F. Neese, J. Chem. Theory Comput. 2009, 5, 2229-2238.
[33] L. Gagliardi, R. Lindh, G. Karlstrom, J. Chem. Phys. 2004, 121, 4494-
4500.
[34] G. M. Sheldrick, Acta Crystallogr A 2008, 64, 112-122.
[35] G. A. Bain, J. F. Berry, J. Chem. Educ. 2008, 85, 532-536.
[36] a) B. O. Roos, R. Lindh, P. A. Malmqvist, V. Veryazov, P. O. Widmark, J.
Phys. Chem. A 2004, 108, 2851-2858; b) B. O. Roos, R. Lindh, P. A.
Malmqvist, V. Veryazov, P. O. Widmark, J. Phys. Chem. A 2005, 109,
6575-6579; c) B. O. Roos, R. Lindh, P. A. Malmqvist, V. Veryazov, P. O.
Widmark, A. C. Borin, J. Phys. Chem. A 2008, 112, 11431-11435.
[37] a) M. Douglas, N. M. Kroll, Ann Phys-New York 1974, 82, 89-155; b) B. A.
Hess, C. M. Marian, U. Wahlgren, O. Gropen, Chem. Phys. Lett. 1996,
251, 365-371; c) M. Reiher, Wires Comput. Mol. Sci. 2012, 2, 139-149.
[38] a) B. O. Roos, Int. J. Quantum. Chem. 1980, 17, 175-189; b) B. O. Roos,
P. R. Taylor, P. E. M. Siegbahn, Chem. Phys. 1980, 48, 157-173; c) P. E.
M. Siegbahn, J. Almlof, A. Heiberg, B. O. Roos, J. Chem. Phys. 1981, 74,
2384-2396.
[22] V. Mereacre, A. M. Ako, R. Clerac, W. Wernsdorfer, I. J. Hewitt, C. E.
Anson, A. K. Powell, Chem. Eur. J. 2008, 14, 3577-3584.
[23] N. F. Chilton, S. K. Langley, B. Moubaraki, K. S. Murray, Chem. Commun.
2010, 46, 7787-7789.
[24] a) A. H. Cooke, C. B. P. Finn, B. W. Mangum, R. L. Orbach, J Phys Soc
Jpn 1962, 17, 462-&; b) C. B. Finn, R. Orbach, W. P. Wolf, P Phys Soc
Lond 1961, 77, 261-&.
[25] a) S. Nayak, O. Roubeau, S. J. Teat, C. M. Beavers, P. Gamez, J.
Reedijk, Inorg. Chem. 2010, 49, 216-221; b) A. Okazawa, T. Nogami, H.
Nojiri, T. Ishida, Chem Mater 2008, 20, 3110-3119.
[26] a) M. Ferbinteanu, T. Kajiwara, K. Y. Choi, H. Nojiri, A. Nakamoto, N.
Kojima, F. Cimpoesu, Y. Fujimura, S. Takaishi, M. Yamashita, J. Am.
Chem. Soc. 2006, 128, 9008-9009; b) V. Chandrasekhar, B. M. Pandian,
R. Boomishankar, A. Steiner, J. J. Vittal, A. Houri, R. Clerac, Inorg. Chem.
2008, 47, 4918-4929.
[27]A. Bhunia, M. T. Gamer, L. Ungur, L. F. Chibotaru, A. K. Powell, Y. H. Lan,
P. W. Roesky, F. Menges, C. Riehn, G. Niedner-Schatteburg, Inorg.
Chem. 2012, 51, 9589-9597.
[39] a) P. A. Malmqvist, B. O. Roos, Chem. Phys. Lett. 1989, 155, 189-194; b)
P. A. Malmqvist, B. O. Roos, B. Schimmelpfennig, Chem. Phys. Lett.
2002, 357, 230-240.
[40] M. E. Lines, J. Chem. Phys. 1971, 55, 2977-&.
[28] a) L. Ungur, W. Van den Heuvel, L. F. Chibotaru, New Journal of
Chemistry 2009, 33, 1224-1230; b) D. Visinescu, A. M. Madalan, M.
Andruh, C. Duhayon, J. P. Sutter, L. Ungur, W. Van den Heuvel, L. F.
Chibotaru, Chem. Eur. J. 2009, 15, 11808-11814.
[29] F. Aquilante, J. Autschbach, R. K. Carlson, L. F. Chibotaru, M. G. Delcey,
L. De Vico, I. F. Galvan, N. Ferre, L. M. Frutos, L. Gagliardi, M. Garavelli,
A. Giussani, C. E. Hoyer, G. Li Manni, H. Lischka, D. X. Ma, P. A.
Malmqvist, T. Muller, A. Nenov, M. Olivucci, T. B. Pedersen, D. L. Peng, F.
Plasser, B. Pritchard, M. Reiher, I. Rivalta, I. Schapiro, J. Segarra-Marti,
[41]a) E. Ruiz, P. Alemany, S. Alvarez, J. Cano, J. Am. Chem. Soc. 1997, 119,
1297-1303; b) E. Ruiz, J. Cano, S. Alvarez, P. Alemany, J. Comput.
Chem. 1999, 20, 1391-1400; c) P. J. Hay, J. C. Thibeault, R. Hoffmann, J.
Am. Chem. Soc. 1975, 97, 4884-4899; d) M. Shoji, K. Koizumi, Y.
Kitagawa, T. Kawakami, S. Yamanaka, M. Okumura, K. Yamaguchi,
Chem. Phys. Lett. 2006, 432, 343-347.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801523
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
A comparative experimental
and theoretical study on the
relaxation dynamics of Dy2
and DyMn compounds via
targeted metal replacement
rationalizes their different
blocking behavior.
Haiquan Tian,* Liviu Ungur,*
Lang Zhao, Shuai Ding,
Jinkui Tang,* and Liviu F.
Chibotaru
Page No. – Page No.
Exchange Interactions
Switch Tunneling: A
Comparative Experimental
and Theoretical Study on
Relaxation Dynamics via
Targeted Metal
Replacement
This article is protected by copyright. All rights reserved.