Single-molecule magnet behavior with a single metal center enhanced through peripheral ligand modifications

Article abstract of DOI:10.1021/ja204562m

Bis(imino)pyridine pincer ligands in conjunction with two isothiocyanate ligands have been used to prepare two mononuclear Co(II) complexes. Both complexes have a distorted square-pyramidal geometry with the Co(II) centers lying above the basal plane. This leads to significant spin-orbit coupling for the d⁷ Co(II) ions and consequently to slow relaxation of the magnetization that is characteristic of Single-Molecule Magnet (SMM) behavior.

Full text of DOI:10.1021/ja204562m

Supporting Information for
Inducing Single-Molecule Magnet Behavior with a Single Metal Center Derived from
Peripheral Ligand Modifications
Titel Jurca, Ahmed Farghal, Po-Heng Lin, Ilia Korobkov, Muralee Murugesu and Darrin S. Richeson
Centre for Catalysis Research and Innovation and Department of Chemistry, University of Ottawa, Ottawa, Ontario, K1N 6N5
Experimental Details
page 2-3
page 4
page 5
page 6
page 6
Computational Model
Figure S1. UV-Vis spectra of compounds 3 and 4
Figure S2. Intermolecular interactions for compound 4
Figure S3. The field-dependent magnetization measured at 100 K for 3 and 4.
Figure S4. Field dependence of the magnetization, M,
at 2.5, 3, 5 and 8 K for 3.
page 7
Figure S5. Field dependence of the magnetization, M,
at 2.2, 3, 5 and 8 K for 4.
page 7
page 8
page 9
page 9
page 10
Figure S6. Field dependence of the magnetization, M, at 2.5, 3, 5 and 8 K for 4-S
Figure S7. M vs. H/T plots at 2.5, 3, 5 and 8 K for 3.
Figure S8. M vs. H/T plots at 2.2, 3, 5 and 8 K for 4.
Figure S9. M vs. H/T plots at 2.5, 3, 5 and 8 K for 4-S.
Figure S10: Temperature dependence of the χT product at 1000 Oe
for complex 3 and 4.
page 11
Figure S11: Temperature dependence of the in-phase χ' ac susceptibility signals 3 and 4. page 12
Figure S12. Field dependence of the characteristic frequency (maximum of χ")
as a function of the applied dc field for 3, 4 and 4-S at 3K.
Figure S13. Relaxation time of the magnetization ln(τ) vs. T-1
(Arrhenius Plot using ac data) of 3.
page 13
page 13
Figure S14. Relaxation time of the magnetization ln(τ) vs. T-1
(Arrhenius Plot using ac data) of 4 (triangle) and 4-s (circle).
Figure S15. Cole-Cole plots for 3 (top), 4 (middle) and 4-s ( bottom)
page 14
page 15
1

Experimental Details
General:
Reactions were performed in a glovebox under a nitrogen atmosphere, with the exception of
ligand synthesis, which was performed using standard Schlenk techniques under a N2
atmosphere. All solvents were sparged with nitrogen and then dried by passage through a column
of activated alumina using an apparatus purchased from Anhydrous Engineering. Cobalt(II)
thiocyanate was purchased from Strem Chemicals and used as received. All other chemicals
were purchased from Aldrich and used without further purification. Compounds 1 and 2 were
synthesized according to literature procedures.1 Elemental analyses for 3, and 4 were performed
by Midwest Microlab LLC, Indianapolis IN.
Magnetic Meaurements:
The magnetic susceptibility measurements were obtained using a Quantum Design SQUID
magnetometer MPMS-XL7 operating between 1.8 and 300 K for dc-applied fields ranging from
−7 to 7 T. Direct current (dc) susceptibility measurements were performed on freshly filtered
crushed polycrystalline sample of 3 (19.9 mg) and 4 (10.1 mg), wrapped in a
polyethylenemembrane. Sample preparation for compound 3 was carried out rapidly in order to
avoid any solvent loss. Alternating current (ac) susceptibility measurements were carried out
under an oscillating ac field of 3 Oe and ac frequencies ranging from 1 to 1500 Hz. In order to
eliminate intermolecular interactions in 4, crystals (18.2mg) were fully dissolved in THF in a
sealed tube and dc and ac susceptibility measurements were carried out on the resulting frozen
solution below 50 K. A diamagnetic correction was applied for the sample holder and Pascal
constants were used for the diamagnetic correction of the sample. The magnetization data were
collected at 100 K to check for ferromagnetic impurities that were absent in all samples.
Synthesis:
[Co[2,6-Bis{1-[(2,6-diisopropylphenyl)imino]ethyl}pyridine](NCS)2](CH2Cl2] (3):
Co(SCN)2 powder (0.099 g, 0.57 mmol) was added to a clear yellow solution of 1 (0.300 mg,
0.623 mmol) in 5 mL of toluene. The reaction mixture was sealed and allowed to stir for 6 h. A
color change from translucent yellow to translucent green was observed. The solution was then
cooled to -20oC for 48 h, to produce a light green precipitate. The reaction mixture was filtered;
the solid was washed with 3 x 5 mL hexanes, and dried under vacuum. A light green powder of 3
was isolated in 97% yield. Crystals suitable for X-ray analysis were grown by slow vapor
diffusion of hexanes into a moderately saturated solution of 3 in CH2Cl2 at -20oC for several
days. Elemental analysis performed on crystals grown from a saturated solution of CH2Cl2 in
hexanes [C35H43CoN5S2]3[CH2Cl2]: Calcd. C 61.94, H 6.42, N 10.22, Found C 62.28 H 6.40, N
10.25.
1 T. Jurca, K. Dawson, I. Mallov, T. Burchell, G. P. A. Yap, and D. S. Richeson Dalton Trans, 2010, 39, 1266; R.-
Q. Fan, D.-S. Zhu, Y. Mu, G.-H. Li, Y.-L. Yang, Q. Su, and S.-H. Feng Eur. J. Inorg. Chem. 2004, 4891.
2

[Co[2,6-Bis{1-[(2,6-diisopropylphenyl)imino]benzyl}pyridine](NCS)2], (4):
Co(NCS)2 powder (0.056 mg, 0.32 mmol) was added to a clear yellow solution of 2 (0.200 mg,
0.330 mmol) in 5 mL of toluene. The reaction mixture was sealed and allowed to stir for 6 h. A
color change from translucent yellow to translucent brown was observed. The solution was then
cooled to -20oC for 48 h, to produce a light brown precipitate. The reaction mixture was filtered,
the solid was washed with 3 x 5 mL hexanes, and dried under vacuum. A light brown powder of
4 was isolated in 95% yield. Crystals suitable for X-ray analysis were grown by slow vapor
diffusion of hexanes into a saturated solution of 4 in THF at -20oC for several days. Elemental
analysis for C45H47CoN5S2: Calcd. C 69.21, H 6.07, N 8.97, Found C 68.95, H 5.99, N 8.88.
3

Computational Model
The distortion of the ligand environment around the Co center, as discussed in the manuscript,
were modeled using DFT computations that were carried out using the Gaussian 09 program
suite and employing the B3LYP functional and DGDZVP basis set. The model compound
Co(NH3)52+ was examined using two different geometric ligand arrays; a square pyramidal
ligand geometry with the Co center in the basal plane(“ideal”) and a geometry with the Co
center raised by 0.5 Å out of the basal plane (“elevated”). The metal ligand bond lengths (Co-
NH3 = 1.86 Å) were kept constant for both “ideal” and “elevated” geometries.
The following relative orbital arrangements were obtained:
The orbital energies are in atomic units and have been listed relative to the lowest energy orbital
having a zero value. These results were used for the model proposed in the manuscript text and
represented by Scheme 2.
4

Figure S1. UV-Vis spectrum Co compounds 3 and 4 obtained in dichloromethane (DCM) and THF. The
THF recorded spectra of compound 3 is shown in red and compound 4 in blue. The DCM recorded
spectrum of 3 is shown in green and compound 4 in purple. The maximum wavenumber value was
determined by background THF absorbance (shown below). The inset provides a magnification of the low
energy bands.
Table S1. Data for compounds 3 and 4.
ν
max (cm-1) (εmolar
)
3 in THF
4 in THF
3 in DCM
4 in DCM
26800 (3.25) 27100 (2.3) 25900 (3.9) 26600(2.9)
14700 (0.26) 14400 (0.10) 14500 (0.12) 14600(0.21)
5

Figure S2. The intermolecular interactions for compound 4 as discussed in the manuscript.
6

Figure S3. The field-dependent magnetization measured at 100 K for 3 and 4 in order to detect the
presence of any bulk ferromagnetic impurities.
7

Figure S4. Field dependence of the magnetization, M, at 2.5, 3, 5 and 8 K for 3
Figure S5. Field dependence of the magnetization, M, at 2.2, 3, 5 and 8 K for 4
8

Figure S6. Field dependence of the magnetization, M, at 2.5, 3, 5 and 8 K for 4-S
9

Figure S7. M vs. H/T plots at 2.5, 3, 5 and 8 K for 3.
Figure S8. M vs. H/T plots at 2.2, 3, 5 and 8 K for 4.
10

Figure S9. M vs. H/T plots at 2.5, 3, 5 and 8 K for 4-S.
11

Figure S10. Temperature dependence of the χT product at 1000 Oe for complexes 3 and 4 (with χ
being the molar susceptibility per mononuclear complex defined as M/H). Pale green and blue data
points are for solid samples of 3 and 4, respectively. Solid lines represent the obtained fits using
following equation, which includes axial zero-field splitting parameter.
where x=D/kT
12

Figure S11: Temperature dependence of the in-phase χ' ac susceptibility signals 3 (top) and 4 (middle:
solid state, bottom: solution), collected over the temperature range 2.5-10 K at under an applied dc field
of 2000 Oe.
13

.
Figure S12. Field dependence of the characteristic frequency (maximum of χ") as a function of the
applied dc field for 3 (black circle), 4 (red square) and 4-S (green triangles) at 3K. Line is guide for
the eyes.
Figure S13. Relaxation time of the magnetization ln(τ) vs. T-1 (Arrhenius Plot using ac data) of 3.
The solid line corresponds to the fit.
14

Figure S14. Relaxation time of the magnetization ln(τ) vs. T-1 (Arrhenius Plot using ac data) of 4
(triangle) and 4-S (circle). The solid red line corresponds to the fit.
15

Figure S15. Cole-Cole plots for 3 (top), 4 (middle) and 4-s ( bottom)
16