Stereochemistry for engineering spin crossover: Structures and magnetic properties of a homochiral vs. racemic [Fe(N3O2)(CN)2] complex

Article abstract of DOI:10.1039/c5dt00357a

The Schiff-base condensation of the R,R-(+)-diamine (2a) with 2,6-diacetyl pyridine in the presence of Fe^II affords the macrocyclic complex [Fe(dpN₃O₂)(CN)₂] (3a) (dp = diphenyl) with ligand centred chirality co

Full text of DOI:10.1039/c5dt00357a

Dalton
Transactions
View Article Online
View Journal
COMMUNICATION
Stereochemistry for engineering spin crossover:
structures and magnetic properties of a
Cite this: DOI: 10.1039/c5dt00357a
homochiral vs. racemic [Fe(N3O2)(CN)2] complex†
Qiang Wang,a,b Shari Venneri,a Niloofar Zarrabi,a Hongfeng Wang,c,d
Cédric Desplanches,c,d Jean-François Létard,c,d Takele Sedae and
Melanie Pilkington*a
Received 26th January 2015,
Accepted 13th March 2015
DOI: 10.1039/c5dt00357a
www.rsc.org/dalton
The Schiff-base condensation of the R,R-(+)-diamine (2a) with 2,6- activity could also be used to monitor the spin-state switching
diacetyl pyridine in the presence of FeII affords the macrocyclic in these materials affording an alternative readout process to
complex [Fe(dpN3O2)(CN)2] (3a) (dp
=
diphenyl) with ligand conventional achiral optical recording media.4 In this context,
centred chirality comprising of a 1 : 1 mixture of LS 6- and HS a breakthrough in the field was made when Ohkoshi et al.
7-coordinate FeII centres. Variable temperature magnetic suscepti- reported that the magnetic properties of a chiral three dimen-
bility and Mössbauer studies reveal that (3a) undergoes an in- sional [Fe(CN)6]4− complex displayed alternate optical switch-
complete thermal SCO transition with a T1/2 = 250 K as well as a ing between the crystallographic and magnetic contributions
LIESST effect. In contrast its racemic counterpart (3c) comprises of to the second-harmonic generation, opening up new and excit-
mostly LS FeII and exhibits no LIESST properties.
ing applications for chiral SCO complexes as magneto-optical
memory and optoelectronic devices.5 Despite these advances,
examples of chiral SCO complexes are quite rare.6 The majority
of chiral SCO complexes reported to-date have been obtained
via spontaneous resolution i.e. fortuitous crystallisation of
crystals of both Λ and Δ configuration rather than generation
of a racemate.5,7 or via a self-assembly process, together with
an auxiliary chiral building block.6a Here we report an alterna-
tive strategy in which we employ chiral centres on the ligand
framework in order to enforce formation of a chiral SCO
complex with full control over the stereochemistry of the bulk
material. Specifically we introduce chirality into the backbone
of the parent [Fe(N3O2)CN)2] SCO complex 1, first reported by
Nelson in the 1970’s.8 Complex 1 remains the only known
example of an FeII SCO system for which the solid state phase
transition involves a significant change in the coordination
geometry of the FeII centre2 and has also been shown to
undergo light induced excited state spin trapping or LIESST,
with a long lived excited state.9 In more recent years this
macrocycle has been exploited for the self-assembly of diverse
structural topologies that include magnetic chains and net-
works.10 Surprisingly no synthetic derivatives have been
studied to-date, most likely due to the lack of commercial
availability of suitably substituted diamine precursors. As part
of our approach we have invested considerable effort in the
preparation and characterization of two chiral diamines 2a
and 2b as the precursors to the novel Schiff-base (N3O2) macro-
cyclic complexes 3a and 3b, Fig. 1.11
The first spin crossover (SCO) complex was discovered by
Cambi et al. in the 1930’s.1 In the years that followed the devel-
opment of the field of coordination chemistry provided insight
into this behaviour which more than eighty years later has
developed into a research intensive field. The large proportion
of SCO complexes reported to-date incorporate octahedral FeII
centres that switch between high-spin (HS) and low-spin (LS)
states via the application of temperature, pressure or light.
During the SCO transition, the majority of these complexes
undergo very subtle changes in their molecular geometries
and/or shapes, a topic that has been extensively reviewed by
Halcrow.2
Scientists are currently exploring new applications for SCO
compounds that include display and memory storage devices,
temperature sensitive MRI contrast agents, as well as switch-
able liquid crystals, SCO nanoparticles and pressure
sensors.1,3 In recent years, it has been proposed that optical
aDepartment of Chemistry, Brock University, 500 Glenridge Ave, St Catharines,
Ontario L2S 3A1, Canada. E-mail: mpilkington@brocku.ca
bSchool of Chemistry and Chemical Engineering, Wuhan Textile University, Wuhan,
430073, China
cCNRS, ICMCB, UPR 9048, F-33600 Pessac, France
dUniv. Bordeaux, ICMCB, UPR 9048, F-33600 Pessac, France
eDepartment of Physics and Astronomy, Western Washington University, Bellingham,
WA98225, USA
†Electronic supplementary information (ESI) available: Experimental, X-ray
The two new chiral diamines 2a and 2b were prepared in
five steps from commercially available (R,R)-(+)- or (S,S)-
crystallographic, Magnetic, Mössbauer and XRD data. CCDC 1025310 and
1025311. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c5dt00357a
(−)-hydrobenzoin
4
following the strategy presented in
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Communication
Dalton Transactions
Fig. 1 Molecular structure of the parent macrocycle [Fe(N3O2)(CN)2]
(1); the chiral diamine (2) and the chiral macrocycle [Fe(dpN3O2)(CN)2],
where dp = diphenyl (3).
Fig. 2 Left, temperature dependence of the χMT product for 3a. Right,
VT 57Fe Mössbauer spectra of 3a (red line), simulated HS component
(green line) and LS component (blue line).
Variable temperature 57Fe Mössbauer spectroscopy studies
of 3a show a dominant quadrupole doublet at 293 K for the HS
FeII complex (green line, Fig. 2), but on cooling the narrower
quadrupole doublet (blue line, Fig. 2), corresponding to the LS
FeII complex increases in intensity. At 5 K both HS and LS
forms of the complex are still present, consistent with the mag-
netic data. A determination of the percentage of FeII sites as a
function of temperature (Fig. 2) based on both Mössbauer and
magnetic data reveals T1/2 = 250 K, i.e. at 250 K the complex
comprises a 1 : 1 mixture of HS and LS complexes with 25%
HS at 5 K and 61% HS at room temperature (see also S-5,
ESI†). In contrast, VT magnetic susceptibility and Mössbauer
studies on racemic 3c are consistent with a predominantly LS
complex of which less than 10% undergoes a SCO transition
(S-5, ESI†).
Scheme 1 Reagents and Conditions: (i) NaH, THF allyl bromide, reflux;
(ii) OsO4, NaIO4 THF–H2O (iii) NaBH4, DCM–NaOH (iv) phthalimide, tri-
phenylphosphine, di-iso-propylazodicarboxylate, THF, 4 days, (v) hydra-
zine monohydrate, EtOH, reflux.
Photomagnetic studies of 3a reveal a LS to HS conversion
could be obtained by illuminating the sample with green light
(λ = 514 nm) at 10 K. After photo-excitation for 10 hours, χMT
Scheme 1. In the first step, hydrobenzoin was reacted with
allyl bromide to afford the diene 5 as pale yellow oil in 73%
yield. Oxidation of 5 with OsO4/NaIO4 afforded the dialdehyde doubled to a value of 0.86 cm3 K mol−1. A subsequent slow
increase of the sample temperature revealed that the χMT plot
remained almost constant up to 40 K, when it decreases
6 which was then reduced in situ with NaBH4 to yield the
corresponding diol 7 that was isolated as a yellow oil in 84%
yield. Reaction of 7 with phthalimide, triphenylphosphine and slightly and remains more or less constant up to 125 K, the
temperature at which the curve merges with the thermal spin
transition recorded in the dark (Fig. 2, left, open circles). The
di-isopropylazodicarboxylate gave the bis-phthalimide 8 that
after hydrazinolysis afforded the diamine 2 as a white solid in
78% yield. Full experimental details for 2 and its synthetic photoconversion is estimated to be ca. 18% (S-4, ESI†). In con-
trast, photomagnetic studies on 3c under the same conditions
reveal no such LIESST behaviour.
Suitable single crystals of 3a were obtained via slow
diffusion of diethyl ether into an acetonitrile solution of 3a. At
250 K, 3a crystallizes in the chiral space group P21, with two
intermediates are presented in S-1 of the ESI.†
The metal-templated Schiff-base condensation of 2a and 2b
with 2,6-diacetyl pyridine afforded the new chiral complexes 3a
(R,R) and 3b (S,S) as dark blue solids in 59% yield.8 The race-
mate 4c was prepared from a 1 : 1 mixture of 2a and 2b in 56%
yield following the same strategy. The opposite chiralities of 3a independent molecules (A and B) in the unit cell (Fig. 3). Inter-
estingly, the two crystallographically independent macrocycles
A and B possess different coordination geometries. For A, the
and 3b were confirmed by CD spectroscopy where as expected,
the racemic complex 3c showed no CD spectrum (S-2, ESI†).
Magnetic susceptibility studies on 3a reveal that at 370 K, the three nitrogen atoms N(1), N(2) and N(3) and the O(2) atom
from the macrocycle coordinate the FeII cation in the equa-
torial plane with the second Fe(1)–O(1) distance (3.097(5) Å)
χMT value (2.78 cm3 K mol−1) is less than that expected for
100% HS FeII (3.3 cm3 K mol−1, g = 2.1), reflecting an in-
complete SCO transition. Upon cooling, a smooth decrease in longer than the sum of the van der Waals radii. Two axial
cyanide ligands complete a distorted octahedron around the
FeII cation (Fig. 3, left), reminiscent of the structure of the LS
χMT is observed down to 50 K where it remains constant before
decreasing again to 0.43 cm3 K mol−1 at 10 K which can be
attributed to the zero field splitting of the remaining HS form of the parent complex 1.12
In contrast, for B, three nitrogen atoms N(6), N(7) and N(8)
fraction. In contrast to the parent macrocycle 1, no thermal
hysteresis was observed.
and two oxygen atoms O(3) and O(4) coordinate the FeII cation
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Communication
Fig. 4 View down the c-axis of the unit cell. Showing the head-to-tail
arrangement of macrocycles in 3a. H-atoms are omitted for clarity.
Fig. 3 ORTEP plot of the two independent molecules A and B in
complex 3a with partial labelling scheme. Atomic displacement para-
meters are plotted at 50%. Solvent molecules are omitted for clarity.
Crystallographic data for 3a is available as (S-3, ESI†).
izing the hydrogen bonding interactions in the crystal packing
at both 250 and 100 K is presented in the S-3 of the ESI.†
Although there are subtle changes to these intermolecular
interactions during the SCO propagation, the D⋯A contacts
are in the range of 3.389(15)–3.541(15) at 250 K and 3.221(16)–
3.522(15) at 100 K, much longer than those reported in the
crystal structure of 1 affording much weaker cooperativity in
the crystal lattice leading to the observed gradual spin cross-
over transition in 3a with no hysteresis.
Small single crystals of the S,S enantiomer 3b measured at
100 K afforded an identical unit cell to that of 3a confirming
that the two enantiomers are isostructural (S-3. ESI†). No
single crystals of 3c have been grown to-date but VT powder
data clearly reveal changes in the intensities, but no changes
to the positions of the peaks in the powder patterns on
cooling, (Fig. 5, left), consistent with a second order phase
transition where subtle changes of atomic positions indicate
some of the molecules are undergoing a SCO. Moreover the
PXRD pattern of 3c at 250 K is different from that simulated
from the single crystal data for the chiral complex 3a, confirm-
ing that the molecule crystallises in a different unit cell and
space group (Fig. 5, right).
in the equatorial plane and the two cyanide groups complete a
pentagonal bipyramidal geometry that closely resembles the
HS structure of the parent complex 1.9,13 These observations
are consistent with the 1 : 1 mixture of HS and LS FeII centres
obtained from Mössbauer and magnetic measurements at
250 K. The coordination geometries of A and B were further
examined by calculating their continuous shape measures
(CSMs)14 which reveal that A is best described as 6-coordinate,
but the large CSM value of 3.06 confirms that the octahedron
is significantly distorted. In contrast, B is close to ideal D5h
symmetry, reflected in the small CSM value of 0.33. At 100 K,
the P21 space group is retained, but there are subtle changes to
the molecular structure; whilst A remains 6-coordinate with no
significant changes in the CSM value, the coordination geome-
try of B undergoes more marked changes, with both the Fe–O
bond lengths increasing by 0.1 and 0.15 Å and the Fe–N and
Fe–CN bond lengths decreasing by 0.02–0.27 and 0.12 Å
respectively, affording a more distorted D5h polyhedron with a
CSM of 0.99. This structural change, in which both O atoms
move away from the metal centre, appears to be an alternative
strategy for this macrocycle to stabilize the LS complex, not
previously observed. In contrast, the first order SCO transition
in the parent macrocycle 1, is accompanied by a dramatic
change in coordination geometry from 7 to 6 due to a revers-
ible Fe–O bond break.12 The difference in the two mechanisms
to achieve a LS configuration is attributed to the larger steric
effects of the phenyl substituents at the stereogenic centres of
3; a large displacement of the O atom positions (as for 1)
would lead to significant motion of the phenyl groups leading
to considerable strain and disruption in the crystal lattice.
Notably, on warming above 250 K, crystals of 3a collapse
suggesting the transition of molecule A from LS to HS requires
a major reorganisation of the crystal lattice. Furthermore, the
breadth of the SCO transition in 3a reflects a lack of coopera-
tivity in the solid. A close examination of the crystal packing
reveals that molecules of 3a are packed in a head-to-tail
arrangement along the c-axis of the unit cell (Fig. 4). In con-
trast to 1, the axial cyanide ligands of 3a do not participate in
short intermolecular H-bonding interactions. A table summar-
In conclusion, we have reported a new strategy for the
preparation of an FeII SCO macrocycle employing a chiral
ligand that enables full control over the stereochemistry of the
final complex. This is the first study where the SCO properties
of a chiral and racemic complex are compared side-by-side. To
Fig. 5 Left, VT PXRD for the racemic complex 3c; right, comparison of
experimental PXRD data for 3c (blue) vs. simulated PXRD for chiral 3a
(red).
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Communication
Dalton Transactions
the best of our knowledge this methodology has not previously
been reported and therefore constitutes the first proof-of-
concept work in this area. In essence we show that crystal
structures of chiral complexes that crystallize in polar space
groups can be quite different from their racemic counterparts
which have a marked impact on their magnetic behaviour.
These stark differences augers well for the development of new
families of magneto-optical materials, in contrast to the most
common previously reported methodology which employs the
spontaneous (i.e. entirely fortuitous) resolution of a chiral
material during the crystallization process which affords chiral
single crystals for study, but the bulk material remains a
racemic mixture, comprising a 50 : 50-mix of ‘left-handed’ and
‘right handed’ crystals. Although the magnetic properties of 3a
are modest, it is the first chiral mononuclear FeII complex to
display a LIESST. The phenyl substituents shed more light on
the correlation between the structural changes and SCO pro-
perties within this unique N3O2 system where a 7- to 6-coordi-
nation change is not the only pathway for this macrocycle.
Interestingly, the coordination equilibria between a 7- and
5-coordinate FeII N5 complex was recently demonstrated by
spectroscopic studies in solution, but has yet to be observed in
a single crystal.15 Work is currently in progress to study and
prepare new chiral FeII N3O2 derivatives and employ them as
building blocks for the self-assembly of cyanide bridged clus-
ters, chains and networks with magneto-optical and/or ferro-
electric properties. This work was financially supported by
NSERC-DG, CFI, CRC, ORF, ERA and Brock University.
F. Dahan and J.-P. Tuchagues, Angew. Chem., Int. Ed., 2003,
42, 1614; (b) C. Train, M. Grusselle and M. Verdaguer,
Chem. Soc. Rev., 2011, 40, 3297.
5 S.-I. Ohkoshi, S. Takano, K. Imoto, M. Yoshiikiyo, A. Namai
and H. Tokoro, Nat. Photonics, 2014, 8, 65.
6 (a) N. Hoshino, F. Iijima, G. N. Newton, N. Yoshida,
T. Shiga, H. Nojiri, A. Nakao, R. kumai, Y. Murakami and
H. Oshio, Nat. Chem., 2012, 4, 921.
7 (a) W. Liu, L.-L. Mao, J. Tucek, R. Zboril, J.-L. Liu, F.-S. Guo,
Z.-P. Ni and M.-L. Tong, Chem. Commun., 2014, 50, 4059;
(b) M. Yamada, H. Hagiwara, H. Torigoe, N. Matsumoto,
M. Kojima, F. Dahan, J. P. Tuchagues, N. Re and S. Iijima,
Chem. – Eur. J., 2006, 12, 4536; (c) Y. Ikuta, M. Ooidemizu,
Y. Yamahata, M. Yamada, S. Osa, N. Matsumoto, S. Iijima,
Y. Sunatsuki, M. Kojima, F. Dahan and J. P. Tuchagues,
Inorg. Chem., 2003, 42, 7001; (d) T. Sato, S. Iijima,
M. Kojima and N. Matsumoto, Chem. Lett., 2009, 178;
(e) H. Hagiwara, N. Matsumoto, S. Iijima and M. Kojima,
Inorg. Chim. Acta, 2011, 366, 283; (f) Y. Sunatsuki, Y. Ikuta,
N. Matsumoto, H. Ohta, M. Kojima, S. Ijima, S. Hayami,
Y. Maeda, S. Kaizaki, F. Dahan and J. P. Tuchagues, Angew.
Chem., Int. Ed., 2003, 42, 1614; (g) Z.-G. Gu, C.-Y. Pang,
D. Qiu, J. Zhang, J.-L. Huang, L. F. Qin, A.-Q. Sun and Z. Li,
Inorg. Chem. Commun., 2013, 35, 164.
8 (a) M. G. B. Drew, A. Hamid bin Othman, S. G. McFall,
P. D. Mcllroy and S. M. Nelson, J. Chem. Soc., Dalton Trans.,
1977, 1173; (b) S. M. Nelson, P. D. A. McIlroy,
C. S. Stevenson, E. König, G. Ritter and J. Waigel, J. Chem.
Soc., Dalton Trans., 1986, 991.
9 S. Hayami, Z.-Z. Gu, Y. Einaga, Y. Kobayasi, Y. Ishikawa,
Y. Yamada, A. Fujishima and O. Sato, Inorg. Chem., 2001,
40, 3240.
Notes and references
10 (a) Y.-Z. Zhang, B.-W. WNG, O. Sato and S. Gao, Chem.
Commun., 2010, 46, 6959; (b) Y.-Z. Zhang and O. Sato, Inorg.
Chem., 2010, 49, 1271; (c) F. Bonadio, M.-C. Senna,
J. Ensling, A. Sieber, A. Neels, H. Stoeckli-Evans and
S. Decurtins, Inorg. Chem., 2005, 44, 969; (d) S. Hayami,
G. Juhasz, Y. Maeda, T. Yokoyama and O. Sato, Inorg.
Chem., 2005, 44, 7289.
11 Q. Wang, S. Venneri, E. Gavey and J. Regier, 2014, manu-
script in preparation.
12 P. Guionneau, F. Le Gac, A. Kaiba, J. Sanchez Costa,
D. Chasseau and J. F. Letard, Chem. Commun., 2007,
3723.
13 P. Guionneau, J. S. Costa and J.-F. Letard, Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun., 2004, m588.
14 H. Zabrodsky, S. Peleg and D. Avnir, J. Am. Chem. Soc.,
1992, 114, 7843–7851.
15 M. Grau, J. England, R. Torres Martin de Rosales,
H. S. Rzepa, A. J. White and G. J. P. Briovsek, Inorg. Chem.,
2013, 52, 11867.
1 L. Cambi and L. Szego, Chem. Ber. Dtsch. Ges., 1931, 64,
2591.
2 (a) M. A. Halcrow, Spin-crossover Materials: Properties and
Applications, John Wiley & Sons, 2013; (b) M. A. Halcrow,
Chem. Soc. Rev., 2011, 40, 4119.
3 (a) J.-F. Letard, P. Guionneau and L. Goux-Capes, Spin
Crossover in Transition Metal Complexes III, Springer, 2004,
pp. 221–249; (b) A. Bosseksou, G. Molnar, L. Salmon and
W. Nicolazzi, Chem. Soc. Rev., 2011, 40, 3313; (c) O. Kahn
and C. J. Martinez, Science, 1998, 279, 44; (d) I. Salitros,
N. Madhu, R. Boca, J. Pavlik and M. Ruben, Monatsh.
Chem., 2009, 140, 695; (e) S. Hayami, M. R. Karim and
Y. H. Lee, Eur. J. Inorg. Chem., 2013, 683; (f) J. Linares,
E. Codjovi and Y. Garcia, Sensors, 2012, 12, 4479;
(g) V. Ksenofontov, A. B. Gaspar, G. Levchenko,
B. Fitzsimmons and P. Gütlich, J. Phys. Chem. B, 2004, 108,
8196.
4 (a) Y. Sunatsuki, Y. Ikuta, N. Matsumoto, H. Ohta,
M. Kojima, S. Iijima, S. Hayami, Y. Maeda, S. Kaizaki,
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015