Magnetic Relaxation Studies on Trigonal Bipyramidal Cobalt(II) Complexes

Article abstract of DOI:10.1002/asia.201901511

We report the preparation and the full characterization of a novel mononuclear trigonal bipyramidal Co^II complex [Co(NS₃ ^iPr)Br](BPh₄) (1) with the tetradentate sulfur-containing ligand NS₃ ^iPr

Full text of DOI:10.1002/asia.201901511

CHEMISTRY
AN ASIAN JOURNAL
www.chemasianj.org
Accepted Article
Title: Magnetic relaxation studies in trigonal bipyramidal cobalt(II)
complexes
Authors: Feng Shao, Benjamin Cahier, Yi-Ting Wang, Feng-Lei Yang,
Eric Rivie
̀re, Régis Guillot, Nathalie Guiheŕy, Jia-Ping Tong,
and Talal Mallah
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. Asian J. 10.1002/asia.201901511
Link to VoR: http://dx.doi.org/10.1002/asia.201901511
A Journal of
A sister journal of Angewandte Chemie
and Chemistry – A European Journal

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
Magnetic relaxation studies in trigonal bipyramidal cobalt(II)
complexes
Feng Shao (邵锋),*‡[a,b] Benjamin Cahier,$[a] Yi-Ting Wang (汪毅婷),[a,b] Feng-Lei Yang (杨凤磊),[a,b,c] Eric
Rivière,[a] Régis Guillot,[a] Nathalie Guihéry,[d] Jia-Ping Tong (仝佳平)*[b,e] and Talal Mallah (ح ﻣلﻼ طﻼ)*[a]
Dedication to my supervisor Professor Talal Mallah on the occasion of his 60th birthday.
Abstract: We report the preparation and the full characterization of a
applications in (nano)spintronics, data storage and quantum
computing due to their magnetic bistability.
novel
mononuclear
trigonal
bipyramidal
CoII
complex
[Co(NS3iPr)Br](BPh4) (1) with the tetradentate sulphur containing
The first studied SMM is [MnIII8MnIV4O12(O2C2H3)16(H2O)4],
commonly called Mn12(OAc),[2] that was first prepared by Lis et al.
in 1980.[3] Thanks to its axial symmetry due to the presence of
axially distorted MnIII ions, Mn12(OAc) has a large spin ground
state (S = 10) with ground MS = ±10 sub-levels responsible of its
magnetic bistability.[1c, 1f] However, the control of the molecular
symmetry and therefore magnetic anisotropy that is responsible
for bistability is hard to achieve in polynuclear complexes. In
recent years, research efforts shifted to design mononuclear
SMMs where such a control is easier to reach and where
magneto-structural correlation combining theory and experiment
is accessible.[4] The success of such an approach allowed
reporting large energy barriers and zero-field open magnetic
hysteresis loops at high temperatures in Dy(III) containing
complexes.[5],[6] Chemical control of the crystal field of d7 transition
metal ions in a linear FeI complex, a linear CoII complex, a
tetrahedral CoII complex, and trigonal bipyramidal CoII complexes
led to slow relaxation of the magnetization and to an opening of
the hysteresis loop at zero field, even above liquid helium
temperature in some cases.[7]
iPr
ligand NS3
(N(CH2CH2SCH(CH3)2)3). The comparison of its
magnetic behaviour with those of two previously reported compounds
tBu
[Co(NS3iPr)Cl](BPh4) (2) and [Co(NS3tBu)Br](ClO4) (3) (NS3
=
N(CH2CH2SC(CH3)3)3) with similar structure shows that 1 displays a
single-molecule magnet behaviour with the longest magnetic
relaxation time (0.051 s) at T = 1.8 K, which is almost thirty times
larger than that of 3 (0.0019 s) and more than three times larger than
for 2 (0.015 s), though its effective energy barrier (26 cm–1) is smaller.
1, that contains two crystallographically independent molecules,
presents smaller rhombic parameters (E = 1.45 and 0.59 cm–1) than 2
(E = 2.05 and 1.02 cm–1) and 3 (E = 2.00 and 0.80 cm–1) obtained from
theoretical calculations. 2 and 3 have almost the same axial (D) and
rhombic (E) parameter values, but present a large difference of their
effective energy barrier and magnetic relaxation which may be
–
–
attributed to the larger volume of BPh4 than ClO4 leading to larger
diamagnetic dilution (weaker magnetic dipolar interaction) for 2 than
for 3. The combination of these factors leads to a much slower
magnetic relaxation for 1 than for the two other compounds.
Actually, if we want to employ SMMs for potential applications,
mononuclear complexes are excellent candidates because they
can be easily manipulated in solution and assembled on surfaces
keeping their integrity, in particular when chelating ligands are
used as for the complexes reported here.[7d]
Introduction
Single-molecule magnets (SMMs) that have been intensively
studied for almost 30 year,[1] are envisioned to be employed for
Herein, we report the preparation and the full characterization
of a mononuclear CoII complex [Co(NS3iPr)Br](BPh4) (1) with
iPr
NS3 = N(CH2CH2SCH(CH3)2)3 that adopts a trigonal bipyramid
[a]
[b]
[c]
Institut de Chimie Moléculaire et des Matériaux d’Orsay, CNRS,
Université Paris-Sud, Université Paris-Saclay, 91405 Orsay Cedex,
France. E-mail: fengshao@outlook.com, talal.mallah@u-psud.fr
State Key Laboratory of Physical Chemistry of Solid Surfaces and
Department of Chemistry, College of Chemistry and Chemical
Engineering, Xiamen University, Xiamen 361005, P. R. China
School of Chemistry and Material Science and Jiangsu Key
Laboratory of Green Synthetic Chemistry for Functional Materials,
Jiangsu Normal University, Xuzhou, Jiangsu 221116, P. R. China
Laboratoire de Chimie et Physique Quantiques, Université Toulouse
III, Paul Sabatier, 118 route de Narbonne, 31062 Toulouse, France
Lab of Chemical Materials and Devices, Training Base of Army
Logistics University of PLA, Xiangyang 441000, P.R. China;
jiapingtong@hotmail.com
geometry thanks to the tetradentate NS3iPr ligand. We compare its
magnetic behaviour to those of two already reported complexes
with the same geometry [Co(NS3iPr)Cl](BPh4) (2) and
[Co(NS3tBu)Br](ClO4) (3) with NS3tBu = N(CH2CH2SC(CH3)3)3.[7d, 7f]
Magnetic studies reveal that 1 exhibits an Ising-type anisotropy
with the longest magnetic relaxation time (0.051 s at T = 1.8 K)
despite its moderate effective energy barrier Ueff = 26.2 cm–1 and
t0 = 6.4 × 10–9 s due to its inherent smaller transverse anisotropy
and larger counter-ions diamagnetic dilution effect.
[d]
[e]
[‡]
Present address: Department of Chemistry, University of British
Columbia, 2036 Main Mall, Vancouver, British Columbia V6T 1Z1,
Canada
Results and Discussion
Syntheses
[$]
Present address: Max-Planck-Institut für Kohlenforschung, Kaiser-
Wilhelm-Platz 1, 45470 Mülheim an der Ruhr, Germany
iPr
Tris(2-(isopropylthio)ethyl)amine (NS3
)
and compound
1
Supporting information and ORCID(s) from the author(s) for this
article are a given via a link at the end of the document.
(Scheme 1) were synthesized similarly to compounds 2 and 3 as
we previously reported.[7d,
The one-pot reaction of organic
7f]
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
iPr
ligand NS3 and anhydrous cobalt(II) bromide with sodium
trigonal bipyramidal geometry with a pseudo-C3 molecular
tetraphenylborate (NaBPh4) in 1-BuOH afforded the trigonal
bipyramidal topology compound 1, the one positive charge is
counter-balanced by a BPh4– anion in the molecule (Scheme 1).
It should be noted that in the final step of synthesis of NS3iPr, 2-
propanethiol is a poisonous gas with an obnoxious odour, all
procedures must be carried out in an efficient fume hood. The
apparatus must be connected to a series of washing bottles
charged with an absorbing solution, such as chromic acid mixture
or others.
symmetry axis. The compound has two crystallographically
independent molecules (noted 1i and 1ii) in the asymmetric unit).
The comparison of the structural bond lengths and angles of 1i
and 1ii is shown in Table 1 together with those of compounds 2
and 3. For complex 1i, the CoII ion lies 0.375 Å (0.330 for 1ii)
below the equatorial plane of the three sulphur atoms with an
ꢅ
average ꢀꢁꢂꢃꢄ angle of 98.98° (97.90° for 1ii). The Co–N axial
and Co–Br bond lengths are 2.351, 2.402 Å and 2.219, 2.400 Å
ꢅ
for 1i and 1ii respectively. The average Co–S bond lengths, ꢀꢁꢂꢀ
ꢅ
and ꢆꢁꢂꢀ angles are equal to 2.384, 117.62°, 81.04° and 2.388,
OH
Cl
–
Cl
S
a
H+
N
b
c
d
118.14°, 82.10° for complexes 1i and 1ii respectively. There are
two main structural differences between 1i and 1ii: (i) the Co–Nax
bond length is longer for 1i and (ii) the deviation of the ꢀꢁꢂꢀ angle
values in the equatorial plane from their average value is larger
for 1i than for 1ii (Table 1). For the three compounds 1–3, all the
different CoII–ligand bond lengths and angles will affect the zero-
field splitting parameters and therefore their magnetic properties
as analysed in the theoretical calculations part.
S
HO
Cl
Cl
N
N
N
Cl
[Co(NS3iPr)Br]BPh4
(1)
Cl
Cl
ii
S
OH
ꢅ
iPr
i
NS3
a: SOCl2/CHCl3/298 K;
b: NaOH/H2O/298 K;
c: Na/Me2CHOH/298 K, Me2CHSH/Me2CHOH/360 K
d: CoBr2/NaBPh4/BuOH
Scheme 1. Synthesis procedure of ligand of NS3iPr and compound 1.
Description of the structures
Compound 1 crystallized in the monoclinic space group P21/n,
(Table S1). The cation structure comprised a central CoII ion
iPr
surrounded by three sulphur atoms from ligand NS3 in the
equatorial plane and a nitrogen atom from the ligand, with a
bromide ion in the axial positions (Figure 1), the ligand adopts a
Figure 1. Molecular structure of compounds 1–3 obtained from X-ray diffraction.
Violet sphere, Co; tan sphere, S; blue, N; brown, Br; green, Cl; black, C; snow,
H. The most hydrogen atoms and counter-ion molecules were omitted.
Table 1. Relevant CoII–Ligand bond distances and angles and anisotropy parameters for compounds
1, 2 and 3.
1
(this work)
2
(ref 7d)
3 (ref 7f)
Complexes
i
ii
2.219
i
ii
i
ii
a dꢇꢈꢉ
a dꢇꢈꢊ
a dꢇꢈꢋ
a dꢇꢈꢌꢊꢊꢊ
b
2.351
2.263
2.408
2.231
2.274
2.384; 2.386; 2.382
2.402
2.384; 2.392; 2.389
2.400
2.412; 2.391; 2.388
2.255
2.362; 2.365; 2.378
2.265
2.435; 2.416; 2.385
2.361
2.412; 2.411; 2.407
2.383
0.375
0.330
0.352
0.415
0.331
0.357
121.36; 118.15;
113.34
120.95; 117.15;
116.32
124.02; 122.08;
108.25
121.40; 116.35;
113.40
122.81; 122.28;
109.71
122.81; 116.83;
114.07
ꢅ
SCoS
b,c
ꢅ
3.74; 0.53; –4.28
2.81; –0.99; –1.82
5.88; 3.94; -9.89
3.26; –1.79; –4.74
4.67; 4.14; –8.34
4.67; –1.31; –4.07
98.66; 97.24; 99.29
SCoS
100.05; 98.38;
98.50
103.49; 94.84;
95.86
103.82; 99.40;
96.98
93.85; 101.17;
98.19
b
ꢅ
98.18; 102.35; 93.18
SCoBr
b
ꢅ
NCoBr
177.89
174.23
173.63
175.78
175.01
178.92
b
ꢅ
NCoS
82.06; 80.39; 80.76
81.90; 82.68; 81.72
82.87; 81.94; 81.54
79.60; 81.29; 79.18
81.82; 83.31; 82.21
81.76; 81.69; 81.42
dUeff
gexp
26.2
32.0
2.28
2.33
21.0
2.29
2.30
2.50
2.46
gcalc
dDexp
–20.0
–15.8
–19.9
–21.0
–20.2
–19.8
dDcalc(av)
dDcalc
–13.7
1.45
–17.9
0.59
–23.0
2.05
–19.1
1.02
–21.7
2.00
–18.0
0.80
dEcalc
a In angstroms; b in degrees. c deviation of the ꢍꢎꢏꢍ angles from their average value; d in wavenumbers; (av) stands for average.
ꢅ
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
and 1ii corresponding to the energy difference between the
excited Ms = ±1/2 and the ground Ms = ±3/2 Kramers doublets
calculated using the theoretical D values are equal to 28 and 36
cm–1 (2|D|), while the fit of the magnetization data gives 40 cm–1.
The main result here is that 1 similarly to the two reported
compounds 2 and 3 has a negative ZFS axial parameter leading
to the MS = ±3/2 sub-levels lying lower than the ±1/2 ones, which
is expected to lead to an Ising type anisotropy and a slow
relaxation of the magnetization. Based on the ZFS parameters
obtained from ab initio calculations (Table 1), we observe that
compound 1 has the smallest axial ZFS parameter value.
Direct current magnetic studies
Direct-current (dc) magnetic susceptibilities were measured on
polycrystalline samples with a temperature range of 2–300 K
(Figure S2). For 1, cMT (cM is the molar susceptibility) is constant
between room temperature and 45 K with a value of 2.61 cm3 K
mol–1 (Figure S2). Below 40 K, it decreases indicating the
presence of zero-field (ZFS) splitting of the S = 3/2 manifold (Ms
= ±1/2 and ±3/2 sublevels). The quality of the cMT data in the high
temperature range is not good enough because the relatively
large amount of eicosane used during the measurement that
generates a diamagnetic signal that has been corrected. We,
therefore, focused on the analysis of the low temperature
magnetization data where the diamagnetic contribution of
eicosane is negligible. The magnetization (M) versus B plots at
different temperatures (Figure 2) were fitted considering the
following spin Hamiltonian and using a homemade software:
Alternating-current magnetic studies
Ac susceptibility measurements were carried out on
microcrystalline samples of compound 1 with an ac oscillating-
drive field of 3 Oe. We measured the ac susceptibility at T = 2 K
with different frequencies and at variable applied dc magnetic field
in order to find the optimum measurement dc field (Figures S3,
S4) where the magnetic relaxation is slower. The optimal dc
magnetic field is the one where the direct mechanism is minimized
and quantum tunnelling of magnetization (QTM) is suppressed. In
Figures S3, S4, we can hardly observe obvious out-of-phase
susceptibility (c’’) under an applied zero dc magnetic field,
however, upon increasing the dc magnetic field of 200 Oe, a
maximum appears around 26.80 Hz, it shifts to low frequency
when increasing the dc magnetic field up to 2600 Oe. At dc
magnetic field of 1400 Oe, the maximum of the c” = f(n) is around
4.31 Hz. If the dc magnetic field is further increased, the frequency
of the maximum of the curve keeps constant till a value of 2600
Oe. The frequency variation of the maximum of c” vs. B (Figure
S4) indicates that the optimal applied dc magnetic field is a range
of 1400–2600 Oe, which also indicate the presence of QTM for
the compound 1, and thus decreases the effective anisotropy
barrier and speeds up the magnetic relaxation.
(
)
ꢀꢚ ꢀꢚ + 1
ꢑ
ꢔ
ꢕ⃗
ꢔꢙ
ꢔꢙ ꢔꢙ
ꢝ ꢠ
ꢗ
ꢛ
ꢐ = ꢒµꢓꢀ ⋅ ꢃ + ꢖ ꢀꢘ –
+ ꢜ ꢀꢞ − ꢀꢟ
(1)
3
ꢔ ꢔ
ꢔ
ꢔ
ꢕ⃗
where ST = 3/2, ꢀ, ꢀꢞ, ꢀꢟ, ꢀꢘ are spin operators, ꢃ is the applied dc
magnetic field vector, g is Landé-factor that was assumed to be
scalar, µB is Bohr Magneton, D and E are the axial and rhombic
ZFS parameters respectively.
Figure 2. Field dependent magnetization at variable temperatures for
compound 1; (™) experimental data; (—) theoretical fit with the best D and g
parameters; (q) an average of the calculated magnetization considering D
values from ab initio calculations.
The fit of the magnetization data gives the following values g =
2.51, D = –20.0 cm–1 and E = 2.0 cm–1. Theoretical calculations
give slightly smaller D (in absolute value) and E parameters for 1i
and 1ii (Table 1). We calculated the magnetization curves from
the theoretical parameters, made their average and obtained a
reasonably good agreement with the experimental data but with a
slightly different g value (2.46 instead of 2.51) (Figure 2). The
theoretical energy barriers for the two independent molecules 1i
Figure 3. Frequency-dependent out-of-phase susceptibilities at 1.80 K with
applied dc magnetic fields of 2000, 2000 and 3000 Oe for compounds 1, 2 and
3 respectively.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
The ac susceptibility data were collected under an optimal field of
2000 Oe for compound 1 (Figures S13, S14), which revealed that
1
displayed frequency-dependence of the out-of-phase
susceptibility and slow relaxation of magnetization. The
comparison of the frequency-dependent out-of-phase signals at T
= 1.8 K for compounds 1–3 (Figure 3) showed that the magnetic
relaxation time for 1 is 0.051 s (Figure 3, calculated from
experimental data and the same below), almost thirty times larger
than for 3 (0.0019 s) and more than three times than for 2 (0.015
s), though the effective energy barrier of 1 (Figure 4 and Tables
1, 2) is smaller than 2, probably because 1 presents much smaller
rhombic parameters E (1.453 and 0.589 cm–1) than 2 (2.055 and
1.021 cm–1) and 3 (2.00 and 0.80 cm–1, Table 1) as shown by
theoretical calculations. In addition, we found that compounds 2
and 3 have almost the same Ising-type anisotropy parameter D
values (D = –21.05 and –19.80 cm–1), but the effective energy
barrier and magnetic relaxation of 2 are better than 3, though the
rhombic parameter E value of 3 (Table 1) is a little smaller than 2,
it should be due to the dilution effect (to some degree) of the
–
counter-ions, because the volume of BPh4 is much larger than
–
ClO4 . Furthermore, the relaxation time at 1.8 K of 1 (0.051 s) also
presented much longer relaxation time than two popular trigonal
Figure 4. Ln(t) as a function of 1/T from the data of Cole-Cole fitting. Blue line,
Arrhenius fit according to the thermally activated region (Orbach process); red
line, fit includes direct, QTM, Raman, and Orbach processes for compounds 1–
3, 2f, 2s represented the deconvoluted fast and slow relaxation of compound 2.
[7g]
bipyramidal complexes [Co(Me6tren)Cl]ClO4
(0.0035 s) and
[Co(TPMA)(Cl)]Cl[8] (0.016 s) at the same temperature.
The effective energy barrier (Ueff) and relaxation time (t; t0
is the relaxation time at infinite temperature) can be extracted by
fitting the plot of ln(t) vs. 1/T (Figure 4) of the high temperatures
(namely Orbach relaxation) using Arrhenius equation ln(t) = ln(t0)
+ Ueff/kBT, which elicited a value of Ueff = 35.0 K and t0 = 1.2 × 10–
Table 2. Fit parameters for compounds
1–3. 2f, 2s represented the
deconvoluted fast and slow relaxation of compound
2
.
8
1
2f
2s
3
s for compound 1. Meanwhile, we also attempted to fit the
A / s–1 T–2 K–1
B1 / s–1
1.20 × 103
1.21 × 105
2.08 × 106
1.12
3.01 × 103
1.10 × 103
1.20 × 104
0.5
3.01 × 103
1.10 × 103
1.20 × 104
17.0
8.11 × 103
2.00 × 104
4.60 × 102
4.10
dependence of the relaxation time considering all relaxation
processes with the following general expression:
B2 /T–2
C / s–1 K–n
n
ꢩꢪꢫꢫ
ꢬꢭꢥ
ꢃꢢ
1 + ꢃꢙꢐꢙ
ꢡ–ꢢ = ꢣꢐꢤꢥ +
+ ꢁꢥꢦ + ꢡꢧ–ꢢ exp ꢨ–
ꢮ
(2)
5
5
5
5
H / T
0.2
0.2
0.2
0.3
6.4 × 10–9
1.2 × 10–8
26.3
1.5 × 10–xxx
4.5 × 10–9
21.3
1.6 × 10–11
6.4 × 10–9
32.6
2.9 × 10–9
4.5 × 10–8
30.5
t0all /s
ꢭ
t0orbach /s
Ueffall / cm–1
Uefforbach / cm–1
ꢯ
where ꢣꢐꢤꢥ
,
,
ꢁꢥꢦ and ꢡꢧ–ꢢexp (– ꢩꢪꢫꢫ/ꢬꢓꢥ)
,
ꢱ
ꢢꢰꢭ ꢲ
ꢱ
24.3
20.8
32.0
21.1
corresponding to the direct, the QTM, the Raman and the Orbach
processes respectively; A, B1, B2, C, and n are coefficients, H is
the applied dc magnetic field in Tesla, T is the temperature in
Kelvin, Ueff is the thermal barrier of the Orbach relaxation
mechanism in wavenumbers, t0 is the attempt time in seconds,
and kB is the Boltzmann constant.[1k, 9] As shown in Figure 4, it is
possible to obtain an excellent fit (Table 2, H = 0.2 T) with A =
1200.5 s–1 T–4 K–1, B1 = 120650 s–1, B2 = 2.08 × 106 T–2, C = 1.12
s–1 K–5, n = 5 (dimensionless), Ueff = 26.5 cm–1 (38.2 K) and t0 =
6.4 × 10–9 s, where the energy barrier is similar to the value
obtained from Arrhenius linear fit (35.0 K). It should be mentioned
that the parameters A, B1, and B2 were obtained from the fit of the
dependence of the relaxation time with the external applied dc
magnetic field (t–1 vs B), we did this for all the three compounds
1–3 (Equation S1, Figure S3-S12). We also performed the fit of
the dependence of the relaxation time only considering Orbach
process and considering all relaxation processes in a similar way
for compounds 2 and 3 to get their effective energy barriers
(Figure 4, Table 2 and Table S3).
Figure 5. Cole-Cole plot of compound 1 under an applied dc magnetic field of
2000 Oe, experimental data (™) and fit (—).
The ac data in the form of Cole-Cole plots for compound 1 (Figure
5) were fitted to the generalized Debye model[7d, 10] that allowed
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
extracting the relaxation times at different temperatures (t) and
their distribution (a) (Table S2).[11] The resulting a values present
with a range of 0.405–0.147 in the superparamagnetic distribution
range for 1 (Table S2). The a values increase up to 0.41 at 1.80
K and decrease to 0.15 and further approaching zero in high
temperatures, which indicates the presence of a closely uniformly
distributed relaxation process in the measured temperature range
of 1.80–3.75 K.
Seq and Co-Br bond distances in the coordination sphere of the
two isomers are almost identical (Table 1). The main difference is
the Co-Nax bond length that is larger for 1i (2.351 Å) than for 1ii
(2.219 Å) leading to a smaller orbital energy separation (DE2)
ꢱ
between the ꢳꢞꢟ, ꢳꢟꢘ and the ꢳꢘ for the former than for the latter.
The positive contribution to D is, therefore, larger for 1i than for
1ii leading to an overall less negative D value for 1i. This is the
origin of the difference of the axial ZFS parameter between 1i and
1ii.
The difference in the ZFS rhombic parameter values
between 1i and 1ii can also be qualitatively rationalized. We have
already carried out theoretical calculations on model complexes
Theoretical Calculations
In a trigonal bipyramid the splitting of the d orbitals by the ligand
field leads to the scheme depicted in Figure 6 right, the ꢳꢞꢟ, ꢳꢟꢘ
orbitals have the lowest energy, the ꢳꢞꢟ, ꢳꢞ
intermediate energies and the ꢳꢘ one has the highest energy. All
calculation results are reported in Tables 1 and S4. Our goal here
is to perform a simple analysis of the relationships between the
structure of the complexes and the anisotropy. From the
theoretical results we will then seek to identify general ideas on
the effect of the axial and equatorial ligands on the magnitude of
the ZFS parameters.
Before performing the analysis of the relationship linking the
energy difference between the states and the magnitude of the
axial and rhombic zero-field splitting parameters, it is important to
note that in C3v symmetry (Figure 6) the first excited state 4A1
couples with the ground 4A2 state via the spin-orbit operator as we
have already demonstrated[7g,
others.[8, 13] This excited state brings a negative contribution to the
overall D value because of its multideterminantal nature as we
have already shown for CoII complexes in C3v symmetry and in
particular for complexes 2 and 3.[7f] While the E2 excited states
contribute positively to D. It is, therefore, possible to analyse the
magnitude of the overall D values of the different complexes by
ꢅ
demonstrating that the more the SCoS angles deviates from their
average values, the larger the rhombic term is.[7f] We can,
therefore, analyse the contribution of the rhombic terms in light of
our previous results. The presence of a rhombic parameter
different from zero is mainly due to the distortion in the equatorial
plane (deviation from the strict C3v symmetry) that induces the
splitting of the 4E excited state and concomitantly lift the
ꢱ
ꢱ
orbitals have
ꢌꢟ
ꢱ
ꢱ
ꢱ
degeneracy of the ꢳꢞꢟ, ꢳꢞ
orbitals. This splitting can be
ꢌꢟ
ꢅ
quantified by the deviation of the SCoS angles from their average
values that is much larger for 1i than for 1ii (Table 1) and is,
therefore, responsible for the largest rhombic E value for the
former than for the latter. Indeed, the degree of splitting of the 4E
electronic term that is responsible for the magnitude of the
rhombic parameter E is directly related to the distortion in the
equatorial plane. The examination of this distortion for the isomers
of the three compounds (Table S4) supports this correlation.
12]
and that was confirmed by
ꢅ
Indeed, the largest deviation of the SCoS (9.89°) is found for
4
isomer 2i where the energy splitting of the E term is the largest
(718.9 cm–1) leading to the largest E parameter (2.06 cm–1). While
4
ꢅ
for 1ii, the SCoS angle deviation of 2.8° leads to an energy splitting
of 218 cm–1 and to an E value of 0.59 cm–1 (Table S4).
4
4
analysing the contributions coming from the excited A1 and E2
states via second order spin-orbit coupling (SOC).
1(i)
1(ii)
-
-
-
-
-
6000
4500
3000
1500
0
The magnitude of |D| is inversely proportional to the energy
difference between the ground and the excited states interacting
through SOC at the second order of perturbations.[7d, 8, 14] For CoII
(d7, S = 3/2) ions in a trigonal bipyramidal environment with a
symmetry close to C3v (Figure 6), the first excited quadruplet
state (4A1) brings a negative contribution to D, while the 4E state
4E2
4A2
d
z2
–1
m
c
/
y
[7g]
g
r
d
d
xy, dx2-y2
e
4A1
En
ꢀE2
(splitted in the distorted geometry) brings a positive one; the
ꢀE1
4
excited A2 does not couple to the ground state by SOC.[12] The
4A1 state results from excitations involving the ꢳꢞꢟ, ꢳꢟꢘ and
4A2
ꢱ
ꢱ
ꢱ
ꢳꢞꢟ, ꢳꢞ
orbitals only and does not involve ꢳꢘ . When including
xz, dyz
ꢌꢟ
SOC this state interacts with the ground state through the
part of the SO operator and therefore contributes negatively to D.
On the contrary, the quadruplets coming from the splitting of the
Figure 6. left, Calculated electronic energy (CASSCF) of the lowest quadruplets
for complexes 1(i) and 1(ii); right, orbital energy diagram for a distorted trigonal
bipyramidal Co(II) complex. The electronic states are labelled by the irreducible
representation of the C3v point group for simplicity even though the symmetry is
lower due to distortion.
4E state involve mainly (but not only) the ꢳꢘ orbital (Figure 6) and
ꢱ
interacts with the ground state through the
part of the
operator that generates a positive contribution. In order to
increase the negative contribution to D, the 4A1-4A2 energy
difference must decrease and the 4E-4A2 difference must increase
(Figure 6 left, Table S4). This translates for the energy of the
orbitals in a decrease of DE1 and an increase of DE2 (Figure 6
right) that can be made by longer equatorial Co–L distances and
shorter axial Co–L bonds (larger axial s-donating effect). The Co-
Conclusions
We prepared and fully characterized a novel sulphur-containing
trigonal bipyramidal CoII compound 1 that behaves as a SMM
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
[1]
a) L. Bogani, W. Wernsdorfer, Nat. Mater. 2008, 7, 179-186; b) R.
Sessoli, Angew. Chem. Int. Ed. 2012, 51, 43-45; c) R. Sessoli, H. L.
Tsai, A. R. Schake, S. Wang, J. B. Vincent, K. Folting, D. Gatteschi,
G. Christou, D. N. Hendrickson, J. Am. Chem. Soc. 1993, 115,
1804-1816; d) A. Caneschi, D. Gatteschi, R. Sessoli, A. L. Barra, L.
C. Brunel, M. Guillot, J. Am. Chem. Soc. 1991, 113, 5873-5874; e)
R. Sessoli, H. L. Tsai, A. R. Schake, S. Wang, J. B. Vincent, K.
Folting, D. Gatteschi, G. Christou, D. N. Hendrickson, J. Am. Chem.
Soc. 1993, 115, 1804-1816; f) R. Sessoli, D. Gatteschi, A. Caneschi,
M. A. Novak, Nature 1993, 365, 141-143; g) A. Caneschi, D.
Gatteschi, R. Sessoli, A. L. Barra, L. C. Brunel, M. Guillot, J. Am.
Chem. Soc. 1991, 113, 5873-5874; h) F. Shao, J. J. Zhuang, M. G.
Chen, N. Wang, H. Y. Shi, J. P. Tong, G. Luo, J. Tao, L. S. Zheng,
Dalton Trans. 2018, 47, 16850-16854; i) J.-P. Tong, F. Shao, J. Tao,
R.-B. Huang, L.-S. Zheng, Inorg. Chem. 2011, 50, 2067-2069; j) K.
S. Pedersen, L. Ungur, M. Sigrist, A. Sundt, M. Schau-Magnussen,
V. Vieru, H. Mutka, S. Rols, H. Weihe, O. Waldmann, L. F. Chibotaru,
J. Bendix, J. Dreiser, Chem. Sci. 2014, 5, 1650-1660; k) J. M.
Zadrozny, M. Atanasov, A. M. Bryan, C.-Y. Lin, B. D. Rekken, P. P.
Power, F. Neese, J. R. Long, Chem. Sci. 2013, 4, 125-138; l) P.
Zhang, L. Zhang, C. Wang, S. Xue, S.-Y. Lin, J. Tang, J. Am. Chem.
Soc. 2014, 136, 4484-4487.
thanks to the trigonal pyramid environment imposed by the
tetradente ligand that leads to a uniaxial ZFS and Ising-type
anisotropy. Magnetic studies reveal that, despite a smaller energy
barrier compound 1 has a magnetic relaxation time three times
slower than that of 2. This was attributed to the difference in the
rhombic parameters of the two compounds that was found to be
correlated to the difference in the structural parameters. The
much slower (thirty times) relaxation time of 1 compared to 3 is
mainly due to the smaller energy barrier of 3, that may be due to
two effects: the larger rhombic parameter and the smaller
distance between the paramagnetic centres in 3 that may induce
larger dipolar magnetic interactions. This last effect stems from
the smaller size of ClO4– (3) compared to BPh4– (1). Therefore, in
order to design CoII complexes with slow magnetic relaxation and
large barrier to the reorientation of the magnetization one needs
not only to perform a large axial anisotropy but also to decrease
the rhombic term, preferably up to zero to eliminate the transverse
process. Besides, large diamagnetic counter-ion(s), would
increase diamagnetic dilution and slow the magnetic relaxation. It
is important to note that other processes (Raman for example)
involving molecular and lattice vibrations are active and contribute
to speed up the relaxation process, but experimental
determination of the vibrational modes are necessary to draw
some qualitative conclusions on their effect.
[2]
D. Gatteschi, R. Sessoli, J. Villain, Molecular Nanomagnets, Oxford
University Press: New York, 2006.
[3]
[4]
T. Lis, Acta Crystallographica Section B 1980, 36, 2042-2046.
S. Gomez-Coca, E. Cremades, N. Aliaga-Alcalde, E. Ruiz, J. Am.
Chem. Soc. 2013, 135, 7010-7018.
[5]
[6]
[7]
C. A. P. Goodwin, F. Ortu, D. Reta, N. F. Chilton, D. P. Mills, Nature
2017, 548, 439.
F.-S. Guo, B. M. Day, Y.-C. Chen, M.-L. Tong, A. Mansikkamäki, R.
A. Layfield, Science 2018, 362, 1400-1403.
Experimental Section
a) J. M. Zadrozny, D. J. Xiao, M. Atanasov, G. J. Long, F. Grandjean,
F. Neese, J. R. Long, Nat. Chem. 2013, 5, 577-581; b) J. M.
Zadrozny, J. Liu, N. A. Piro, C. J. Chang, S. Hill, J. R. Long, Chem.
Commun. 2013, 48, 3927; c) P. C. Bunting, M. Atanasov, E.
Damgaard-Møller, M. Perfetti, I. Crassee, M. Orlita, J. Overgaard, J.
van Slageren, F. Neese, J. R. Long, Science 2018, 362, eaat7319;
d) F. Shao, B. Cahier, N. Guihery, E. Riviere, R. Guillot, A. L. Barra,
Y. Lan, W. Wernsdorfer, V. E. Campbell, T. Mallah, Chem. Commun.
2015, 51, 16475-16478; e) D. Schweinfurth, M. G. Sommer, M.
Atanasov, S. Demeshko, S. Hohloch, F. Meyer, F. Neese, B. Sarkar,
J. Am. Chem. Soc. 2015, 137, 1993-2005; f) F. Shao, B. Cahier, E.
Rivière, R. Guillot, N. Guihéry, V. E. Campbell, T. Mallah, Inorg.
Chem. 2017, 56, 1104-1111; g) R. Ruamps, L. J. Batchelor, R.
Guillot, G. Zakhia, A.-L. Barra, W. Wernsdorfer, N. Guihéry, T.
Mallah, Chem. Sci. 2014, 5, 3418-3424; h) R. Ruamps, R. Maurice,
L. Batchelor, M. Boggio-Pasqua, R. Guillot, A.-L. Barra, J. Liu, E.-E.
Bendeif, S. Pillet, S. Hill, T. Mallah, N. Guihéry, J. Am. Chem. Soc.
2013, 135, 3017-3026.
Generally, all starting materials were obtained commercially and
were used without further purification unless otherwise stated.
Elemental analyses for C, H, N were performed on a Thermo
Scientific Flash analyser. Infrared (IR) spectra were recorded on
a Bruker TENSOR-27 FT-IR spectrometer equipped with an
attenuated total reflectance (ATR) sample holder in a range of
4000−500 cm−1. Nuclear magnetic resonance (NMR) spectra
were recorded on a Bruker Aspect 300 NMR spectrometer. X-ray
diffraction data were collected by using a Kappa X8 APPEX II
Bruker
diffractometer
with
graphite-monochromated
MoKa radiation (l = 0.71073 Å). Magnetic data were collected
using a Quantum Design MPMS XL7 SQUID magnetometer.
Acknowledgements
We appreciate the financial supported by a public grant overseen
by the French National Research Agency (ANR) as part of the
“Investissements d’Avenir” program “Labex NanoSaclay”
(Reference: ANR-10-LABX-0035), partially financed by ANR-
project MolNanoSpin 13-BS10-0001-03 and NSF of Xiamen
University (Grant 2014013). F.S. and Y.-T.W. thank the
collaborative program between the China Scholarship Council
(No. 201306310014; 201506310023) and Université Paris-Sud,
Université Paris-Saclay. T.M. thanks the Institut Universitaire de
France for financial support.
[8]
[9]
T. J. Woods, M. F. Ballesteros-Rivas, S. Gómez-Coca, E. Ruiz, K.
R. Dunbar, J. Am. Chem. Soc. 2016, 138, 16407-16416.
a) V. V. Novikov, A. A. Pavlov, Y. V. Nelyubina, M.-E. Boulon, O. A.
Varzatskii, Y. Z. Voloshin, R. E. P. Winpenny, J. Am. Chem. Soc.
2015, 137, 9792-9795; b) Y. Rechkemmer, F. D. Breitgoff, M. van
der Meer, M. Atanasov, M. Hakl, M. Orlita, P. Neugebauer, F. Neese,
B. Sarkar, J. van Slageren, Nat. Commun. 2016, 7, 10467; c) X.-F.
Jiang, M.-G. Chen, J.-P. Tong, F. Shao, New J. Chem. 2019, 43,
8704-8710.
[10]
a) K. S. Cole, R. H. Cole, J. Chem. Phys. 1941, 9, 341-351; b) M.
Perovic, V. Kusigerski, V. Spasojevic, A. Mrakovic, J. Blanusa, M.
Zentkova, M. Mihalik, J. Phys. D: Appl. Phys. 2013, 46, 165001.
K. S. Cole, R. H. Cole, J. Chem. Phys. 1941, 9, 341-351.
B. Cahier, R. Maurice, H. Bolvin, T. Mallah, N. Guihéry,
Magnetochemistry 2016, 2, 31.
[11]
[12]
Keywords: magnetic anisotropy • magnetic relaxation • single
molecule magnets • molecular magnetism
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
[13]
A. K. Mondal, J. Jover, E. Ruiz, S. Konar, Chem. Eur. J. 2017, 23,
Charron, E. Rivière, R. Guillot, A. L. Barra, M. D. Serrano, J. van
12550-12558.
Slageren, T. Mallah, Chem. Eur. J. 2008, 14, 1169-1177.
[14]
a) F. E. Mabbs, D. Collison, Electron paramagnetic resonance of d
transition metal compounds, Elsevier, 2013; b) J. N. Rebilly, G.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201901511
Chemistry - An Asian Journal
FULL PAPER
Entry for the Table of Contents
Layout 1:
FULL PAPER
Counter anion together with subtle
structural changes in coordination
sphere slow down the magnetic
relaxation time in trigonal bipyramidal
Co(II) complexes.
Feng Shao,* Benjamin Cahier, Yi-Ting
Wang, Feng-Lei Yang, Eric Rivière,
Régis Guillot, Nathalie Guihéry, Jia-Ping
Tong* and Talal Mallah*
Page No. – Page No.
Magnetic relaxation studies in
trigonal bipyramidal cobalt(II)
complexes
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.