Extension of Azine-Triazole Synthesis to Azole-Triazoles Reduces Ligand Field, Leading to Spin Crossover in Tris-L Fe(II)

Article abstract of DOI:10.1021/acs.inorgchem.9b02993

The first examples of azole-Triazole Rat ligands, bidentate L^4NMeIm(3-(1-methyl-1H-imidazol-4-yl)-5-phenyl-4-(p-Tolyl)-4H-1,2,4-Triazole) and L^4SIm (4-(5-phenyl-4-(p-Tolyl)-4H-1,2,4-Triazol-3-yl)thiazole), have been prepared, by extension of the general synthesis used to access many examples of azine-Triazoles. The tris-L Fe^II complexes of the azine-Triazoles are consistently low spin (LS). As intended, these new azole-Triazole ligands provide lower field strengths, resulting in high-spin (HS) [Fe^II(L^4NMeIm)₃](BF₄)₂ (1·4H₂O) and spin crossover (SCO) active [Fe^II(L^4SIm)₃](BF₄)₂ (2·0.5H₂O). Single-crystal structure determinations revealed that at 100 K 1·solvents is HS whereas 2·solvents is LS. Solid-state variable temperature magnetic studies of air-dried crystals showed that the methylimidazole-Triazole complex 1·4H₂O remains HS while the thiazole-Triazole complex 2·0.5H₂O undergoes a two-step gradual SCO (T_1/2 approximately 275 and 350 K). Variable-Temperature Evans method NMR studies of 2, in five different solvents (CD₃NO₂, CD₃CN, CD₃COCD₃, CD₂Cl₂, and CDCl₃) gave T_1/2 values in a relatively narrow range, 214-259 K. These T_1/2 values did not correlate with the solvent polarity index P′ (R² = 0.25) but did correlate with the solvent basicity parameter SB (R² = 0.90). Variable-Temperature UV-vis studies on a golden yellow CH₃CN solution of 2, with monitoring of the d-d transition at 530 nm (? = 39 L mol^-1 cm^-1 at 293 K) while the solution was heated from 253 to 303 K, showed that the high-spin fraction increased from 0.51 to 0.77. Cyclic voltammetry studies in CH₃CN revealed a Fe(III)/Fe(II) redox process that was reversible for 1 and irreversible for 2, with significant tuning of the E_pa value: The methylimidazole-Triazole complex 1 is significantly easier to oxidize (0.46 V) than the thiazole-Triazole complex 2 (0.68 V; both vs 0.01 M Ag/AgNO₃).

Full text of DOI:10.1021/acs.inorgchem.9b02993

Article
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/IC
Extension of Azine-Triazole Synthesis to Azole-Triazoles Reduces
Ligand Field, Leading to Spin Crossover in Tris‑L Fe(II)
Sandhya Singh and Sally Brooker*
Department of Chemistry and the MacDiarmid Institute for Advanced Materials and Nanotechnology, University of Otago, PO Box
5
6, Dunedin 9054, New Zealand
*
S Supporting Information
ABSTRACT: The first examples of azole-triazole Rat ligands,
4
NMeIm
bidentate L
(
(3-(1-methyl-1H-imidazol-4-yl)-5-phenyl-4-
p-tolyl)-4H-1,2,4-triazole) and L4
SIm
(4-(5-phenyl-4-(p-
tolyl)-4H-1,2,4-triazol-3-yl)thiazole), have been prepared, by
extension of the general synthesis used to access many
II
examples of azine-triazoles. The tris-L Fe complexes of the
azine-triazoles are consistently low spin (LS). As intended,
II 4NMeIm
these new azole-triazole ligands provide lower field strengths, resulting in high-spin (HS) [Fe (L
) ](BF ) (1·4H O) and
3 4 2 2
II 4SIm
spin crossover (SCO) active [Fe (L ) ](BF ) (2·0.5H O). Single-crystal structure determinations revealed that at 100 K 1·
3
4 2
2
solvents is HS whereas 2·solvents is LS. Solid-state variable temperature magnetic studies of air-dried crystals showed that the
methylimidazole-triazole complex 1·4H O remains HS while the thiazole-triazole complex 2·0.5H O undergoes a two-step
2
2
gradual SCO (T1/2 approximately 275 and 350 K). Variable-temperature Evans method NMR studies of 2, in five different
solvents (CD NO , CD CN, CD COCD , CD Cl , and CDCl ) gave T values in a relatively narrow range, 214−259 K.
3
2
3
3
3
2
2
3
1/2
2
These T1/2 values did not correlate with the solvent polarity index P′ (R = 0.25) but did correlate with the solvent basicity
2
parameter SB (R = 0.90). Variable-temperature UV−vis studies on a golden yellow CH CN solution of 2, with monitoring of
3
−
1
−1
the d−d transition at 530 nm (ε = 39 L mol cm at 293 K) while the solution was heated from 253 to 303 K, showed that
the high-spin fraction increased from 0.51 to 0.77. Cyclic voltammetry studies in CH CN revealed a Fe(III)/Fe(II) redox
3
process that was reversible for 1 and irreversible for 2, with significant tuning of the Epa value: the methylimidazole-triazole
complex 1 is significantly easier to oxidize (0.46 V) than the thiazole-triazole complex 2 (0.68 V; both vs 0.01 M Ag/AgNO3).
INTRODUCTION
electronic nature of the ligand (induced by changes to
■
1
2−14
substituents, electron withdrawing or donating).
In
The term spin crossover (SCO) refers to the switching
between the low-spin (LS) and high-spin (HS) states that is
contrast, in the solution phase, ligand field strength and
15
4
7
solvent polarity/pH effects dominate. The influence of
possible in octahedral 3d −d transition-metal complexes when
the ligand field is “just right” (not too strong and not too
weak). SCO can be induced by a variety of external stimuli,
such as temperature, pressure, irradiation, and guest mole-
solvent choice on T1/21 values in the solution state is well-
5−19
known in the literature.
For example, a strong correlation
between the solvent polarity index P′ and T value is seen for
1
/2
azine
19
1
[Fe(L ) (NCE) ] (Figure 1, bottom left; E = BH , S, Se).
2 2 3
cules. This phenomenon is mostly studied as a function of
Hydrogen bonding with the solvent also affects the T1/2 value:
temperature and can be easily monitored by many different
analytical techniques, including solid-state magnetic measure-
ments, SC-XRD, P-XRD, DSC, and NMR (Evans), UV−vis,
typically increased hydrogen bonding stabilizes the LS state,
15−21
leading to a higher T1/2
.
2
,3
1,2,4-Triazole-based ligands have been targeted by many
Raman, FTIR, and Mossbauer spectroscopy. This is because
̈
researchers, as 1,2,4-triazoles are readily functionalized at some
the change in spin state, HS ↔ LS, is accompanied by
3
5
4
or all of the C , C , and N positions of the triazole ring
(Figure 1) and have generated many interesting complexes of a
variations in the number of unpaired electrons, color,
5
7
molecular vibrations, Fe isomer shift and quadrupole
splitting, and proton chemical shifts, as well as in Fe−N
bond lengths and hence the volume occupied by the complex.
The bistability of the magnetic, optical, structural, and
dielectric properties of SCO-active complexes
many potential practical applications: on/off switches, displays,
3
,11,22−24
wide range of metal ions.
Of interest herein is that
many such 1,2,4-triazole based ligands, including the bis-
3
,25−27
terdentate PMRT/PSRT (Figure 1, top middle)
and the
4
,5
related thia-/oxa-diazoles PMTD/PMOD (Figure 1, top
leads to
2
8−30
right),
as well as the bis- or monobidentate Rdpt/Rat
1
9,24,31−37
ligand types (Figure 1, bottom),
are reported in the
chemical/guest sensors, molecular spintronics, molecular
electronics, and data storage.
6
−11
literature to form SCO-active iron(II) complexes.
When SCO is monitored in the solid state, crystal packing,
intermolecular cooperativity, and cocrystallized solvent effects
can dominate over the effects of any modifications to the
Received: October 10, 2019
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. Synthesis of the first examples of azole-triazole Rat ligands
4
4
by extension of our protocol for preparing azine-triazoles:
monotopic L4
NMeIm
and L
4SIm
(box).
II 4NMeIm
of the new azole-triazole ligands, [Fe (L
4
) ](BF ) (1·
3 4 2
II 4SIm
H O) and [Fe (L ) ](BF ) (2·0.5H O), are shown to be
2 3 4 2 2
Figure 1. Generic 1,2,4-triazole showing ring numbering (top left),
ditopic PMRT (4-R-substituted-3,5-bis{[(2-pyridylmethyl)amino]-
methyl}-4H-1,2,4-triazole)/PSRT (4-R-substituted-3,5-bis{[(2-
pyridylmethyl)sulfanyl]methyl}-4H-1,2,4-triazole) (top mid-
HS and SCO-active, not LS as all of the analogous complexes
of azine-triazole ligands have been, demonstrating that this
approach to reducing the ligand field has worked.
3
,25−27
dle)
and related thia-/oxazole ligands PMTD (2,5-bis[(2-
RESULTS AND DISCUSSION
■
pyridylmethyl)amino]methyl-1,3,4-thiadiazole)/PMOD (2,5-bis{[(2-
2
8−30
Synthesis. The extension of our general azine-triazole Rat
pyridylmethyl)amino]methyl}-1,3,4-oxadiazole) (top right).
di-
44
topic Rat ligands L2
/4pym‑meta/para
ligand synthesis to access azole-triazole Rat ligands was a step
change, as azines and azoles typically exhibit very different
(1,n-bis(4-isobutyl-5-(pyrimidin-2/
-yl)-4H-1,2,4-triazol-3-yl)benzene (middle), and Rdpt (4-R-
4
1
4
45
substituted 3,5-di(2-pyridyl)triazole) (bottom left) as well as
chemistry. However, given that in this protocol the azole is
introduced as a carbohydrazide (Figure 2), it was reasonable to
anticipate that this would work, as is found to be the case. Both
azine
monotopic Rat ligands (bottom), either azine-substituted L
19,24,31−37
(
bottom middle)
right, box) versions made and used in this study, L
or the new azole-substituted (bottom
4NMeIm
and L4SIm.
4NMeIm
4SIm
ligands, L
and L , are obtained as analytically pure
white powders in good yields (72−76%).
Subsequent 3:1 reactions of these ligands, L
II
4NMeIm
and L4SIm
,
The [Fe (L) (NCE) ] complexes of the Rdpt/Rat ligands
2
2
(
Figure 1, bottom left; E = −BH , −S, −Se) are often SCO-
with iron(II) tetrafluoroborate, in nitromethane or acetonitrile,
respectively, followed by vapor diffusion of diethyl ether into
3
24
active in both solution and the solid state. In contrast, the
tris-L complexes, [Fe (L) ](BF ) , of these azine-triazole
ligands are consistently LS,
include the recently reported pairs of dinuclear helicates,
II
the colorless reaction solution of 1 and orange reaction
3
4 2
3
8−40
II 4NMeIm
a finding which extends to
solution of 2, gave single crystals of [Fe (L
) ](BF ) ·
3 4 2
CH NO ·solvent (1·solvents; colorless needles) and
[Fe (L ) ](BF ) ·0.125H O·solvent (2·solvents; orange
3
2
II
2/4pym‑meta
II 4SIm
[
[
Fe (L
) ](BF ) and tetranuclear cages
2
3
4
4
3 4 2 2
II
2/4pym‑meta
Fe (L
) ](BF ) of new m- and p-phenylene-linked
6
blocks) suitable for X-ray diffraction (see below, Figure 3).
Elemental analysis and TGA on the bulk samples, after drying
in air, are consistent with 1·4H O and 2·0.5H O, formed in 81
4
4 8
ditopic Rat ligands (Figure 1, middle), all four of which are
4
1
LS. While these helicates and cages also feature tris-Rat
binding pockets (which had always generated LS complexes as
noted above), it had been hoped that the ligand field would be
reduced sufficiently by the potential ligand strain inherent in
linking these binding pockets together across the ends of the
helicates or along the edges of the cages and by the use of 2/4-
pyrimidine-triazole pockets (pyrimidine has the weakest ligand
2
2
and 76% yields, respectively.
Crystal Structure Descriptions. In both cases some of
the lattice solvent molecules were highly disordered and so
46
47
were treated with SQUEEZE in PLATON.
The mononuclear tris-ligand complex 1·solvents crystallized
(P1) with the entire complex in the asymmetric unit. The
̅
33,42,43
field of the diazines).
This was not found to be the case;
iron(II) center in these colorless crystals has Fe−N bond
lengths (2.175−2.197 Å) and an octahedral distortion
parameter (∑ = 92.9°; cis N−Fe−N angle range 74.97(7)-
99.12(7)°) consistent with it being in the HS state. This is a
first for such a Rat complex (all previous tris-Rat-ligated Fe(II)
complexes were LS) and is consistent with the replacement of
the azine by an azole having successfully reduced the resulting
ligand field strength.
therefore, in order to develop access to SCO-active supra-
molecular assemblies of such ligands, the ligand field imposed
upon tris-ligand coordination to iron(II) must be reduced
further. This requires a change from the use of azines to azoles.
44
Herein our general synthetic strategy to robust Rat ligands,
all of which have been azine-triazoles to date, is further
extended to enable the preparation of the first examples of
azole-triazole Rat ligands by this method (Figures 1 and 2, in
box), as these should impose a weaker ligand field.
Furthermore, the resulting mononuclear tris-ligand complexes
The SCO-active complex (see later), 2·solvents, crystallized
̅
in the same space group (P1) as 1·solvents, but in this case the
−
asymmetric unit consists of two dicationic FeL units, four BF
3
4
B
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
similar situation arises here. The complex of the diazole ligand,
4NMeIm
L
, is HS at 100 K, whereas replacement of N by S gives a
4SIm
thiazole ligand, L , which must impose a stronger ligand
field, as it results in a complex that is clearly LS at 100 K. This
probably occurs for the same reasons as proposed by
Rentschler for PMAT vs PMTDbut as the structure
determinations are again only available for different spin
states, and packing interactions could yet be the key to the
observed spin states, this issue cannot be clearly resolved
herein either.
Solid-State Magnetic Studies. The temperature depend-
ence of the magnetic susceptibility of the air-dried crystalline
Figure 3. Crystal structures of the dication of 1·solvents (left) and
one of the two dications in the asymmetric unit of 2·solvents (right).
For clarity, hydrogen atoms, anions, and solvent molecules are not
shown.
II
4NMeIm
samples of [Fe (L
) ](BF ) ·4H O (1·4H O) and
3 4 2 2 2
II 4SIm
[
Fe (L ) ](BF ) ·0.5H O (2·0.5H O) is shown in Figure
3 4 2 2 2
4
and Figures S3−S5 in the Supporting Information.
anions, and a 0.25 occupancy H O molecule of crystallization.
2
At 100 K the shorter Fe−N bond lengths (1.958−1.997 Å)
and reduced octahedral distortion parameters (∑ = 58.23 and
5
8
8.20° for Fe1 and Fe2; cis N−Fe−N angle ranges Fe1
0.33(8)−96.29(8)° and Fe2 80.29(8)-95.81(8)°) are con-
3
sistent with both being fully LS at 100 K. Attempts to also
structurally characterize the HS state of this complex, by
collecting the data at higher temperatures, failed due to loss of
crystallinity.
All of the previous structures of Fe(II) complexes with three
bidentate Rat/Rdpt ligands bound were reported to be in the
3
8−40
mer configuration,
which is also seen in 1·solvents and 2·
II
solvents. As expected, the intraligand Ntriazole−Fe −N
bite
azole
angles widen from just 74.97(7)−76.05(8)° for HS complex 1·
solvents to 80.29(8)−80.98(8)° for LS complex 2·solvents
(
Table S3 in the Supporting Information). This occurs due to
the shorter Fe−N bonds in the latter LS complex, causing this
otherwise constrained angle to open up by an average of 5°,
toward the ideal of 90°. The other cis-N−Fe−N angles (Table
S4 in the Supporting Information) for HS 1·solvents
II 4SIm
Figure 4. χ T versus T plot for [Fe (L ) ](BF ) ·0.5H O (2·
M
3
4
2
2
0
.5H O) over three successive cycles, 300−50−400, 400−50−400,
and 400−50−400 K (cooling, filled squares; heating, open squares),
2
(
91.51(8)− 99.12(7)°; average 95.51°) also deviate more
in settle mode.
from 90° than do those of LS 2·solvents (90.18(8)−96.29(8)°;
II
average 93.34°). Similarly, as expected, the trans-N−Fe −N
3
−1
angles (Table S5 in the Supporting Information) are closer to
The χ T value for 1·4H O at 300 K is 3.6 cm K mol and
M 2
3
−1
1
1
80° for the LS complex (169.42(8)−173.53(8)°; average
only drops very slowly and slightly to 3.1 cm K mol at 50 K,
consistent with it remaining fully HS (Figure S3 in the
Supporting Information). This is consistent with the crystal
structure showing that this complex is HS at 100 K and that
the mer coordination of the three diazole-triazole L4
ligands generates only a weak ligand field.
71.30°) than for the HS complex (162.89(8)−166.45(8)°;
average 164.08°).
II
2
It is interesting to note that the dinuclear [Fe (PMAT) ]-
2
NMeIm
(
BF ) ·DMF complex of the ditopic pyridine-amine-triazole
4 4
26
ligand (Figure 1, PMRT with R = NH2) was structurally
characterized in both the [HS-HS] state, at 298 K, and the
localized [HS-LS] state, at 123 K. In contrast, the analogous
pyridine-amine-thiadiazole complexes of the PMTD ligand
4SIm
On the other hand, the thiazole analogue, L , provides an
II 4SIm
appropriate ligand field strength in [Fe (L ) ](BF ) ·
3
4 2
0.5H O (2·0.5H O) to enable SCO. First the magnetic
2
2
(Figure 1), reported by Rentschler and co-workers, were
properties of 2·0.5H O were studied in the solid state over
2
shown to be [LS-LS] until they were heated well above 300
five successive cycles of 300−50 K, in settle mode (Figure S4
in the Supporting Information), to confirm a reproducible
response. The sample consistently traced the same magnetic
3
0
K. This demonstrated that changing a noncoordinating
heteroatom can change the ligand field strength of the resulting
azole. They postulated that this is due to the larger sulfur atom
response curve throughout all five cycles. The χ T value at 300
K of 1.76 cm K mol , consistent with the presence of about
M
II
3
−1
in the azole ring causing the intraligand Ntriazole−Fe −N
azole
bite angle to increase toward 90° relative to that seen for the
nitrogen analogue, thereby also increasing the field
1:1 HS:LS (HS fraction γ ≈ 0.5), dropped gradually to 0.08
HS
3
−1
cm K mol at 50 K, which is consistent with a half-SCO to
the fully LS state (γHS = 0).
3
0
strength but it should be noted, as a caution, that in the
solid state the packing interactions can dominate ligand field
considerations in determining the observed spin state. As these
two systems, PMAT and PMTD, were structurally charac-
terized in differing spin states, this issue cannot be resolved. A
Increasing the maximum temperature studied to 400 K was
explored next, to test whether further SCO, to the fully HS
state, was possible for 2·0.5H O: three successive cycles, 300−
2
50−400, 400−50−400, and 400−50−400 K, were carried out
C
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Figure 4). In the first heating cycle, from 50 to 400 K, the
Article
(
response traces almost the same path from 50 to 300 K as it
had in the first cooling cycle, 300−50 K. Above 300 K, SCO
continues to occur but more gradually, with χ T reaching 3.21
M
3
−1
cm K mol at 400 K, consistent with 87% of the Fe(II)
centers being in the HS state (assuming χ T(max) = 3.7 cm
3
M
−
1
K mol ). Two subsequent cycles followed almost the same
response curve, albeit with signs that heating as high as 400 K
does cause some changes, with both SCO steps becoming
somewhat more gradual, probably due to solvent loss (see
Figure S5 and the associated table in the Supporting
Information).
Overall, in the solid state, 2·0.5H O undergoes a gradual
2
two-step SCO, from fully LS to 87% HS (SCO is not yet
complete at 400 K), with T1/2 values of about 275 and 350 K.
Solution-State Magnetic Studies. In the solid state, the
magnetic response can be affected by intermolecular
cooperativity, cocrystallized lattice solvent, and/or crystal
packing. In contrast, in the solution phase these are absent,
enabling the effect of electronic modifications to the ligand on
the ligand field imposed on the iron(II) center to be observed
II 4SIm
Figure 5. χ T vs T for 10.6 mM solutions of [Fe (L ) ](BF ) ·
m
3
4 2
0.5H O (2·0.5H O) in CD NO (brown), CD CN (blue), C D CO
(green), CD Cl (violet) ,and CDCl (sky blue), as determined by the
2
2
3
2
3
2
6
3
2
3
15
48,51
more readilyproviding speciation is not a problem. Hence,
Evans method.
The curves correspond to the best fit found for a
2
7,50
complete one-step SCO using the regular solution model.
1
and 2 were also studied in acetonitrile solution, with
48,49
responses monitored by the Evans NMR method
and
UV−vis spectroscopy as a function of temperature.
Complex 1·4H O is colorless in both the solid state and
2
acetonitrile solution (Figure S6 in the Supporting Informatio-
nI). In contrast, 2·0.5H O is dark orange in the solid state and
2
golden yellow in acetonitrile solution (Figure S7 in the
Supporting Information). The Evans NMR method was used
to determine the magnetic susceptibility of the complexes
versus temperature. As in the solid state, in CD CN solution 1·
3
4
H O remains in the HS state across the measured
2
temperature range (323−243 K, Figure S8 in the Supporting
Information). In contrast, 2·0.5H O undergoes a quite abrupt
2
3
−1
SCO: the χ T value at 343 K of 3.27 cm K mol dropped to
M
3
−1
1
.65 cm K mol at 243 K (Figure 5). Fitting the data to the
50
regular solution model, as a one-step SCO, gave a T1/2 value
of 249 K for 2·0.5H O in MeCN.
2
Recently, some of us reported a strong correlation between
solvent polarity index (P′) and T values for a pair of neutral
1
/2
npyrimidine
19
[
Fe(L
) (NCBH ) ] complexes. Hence the same five
2 3 2
2
Figure 6. Plot showing (green; R = 0.25) lack of correlation between
solvents, ranging from polar to nonpolar (from CD NO ,
CD CN, C D CO, and CD Cl to CDCl ) were also used for
[
53,54
3
2
2
polarity index (P′)
versus (blue; R = 0.90) good correlation
between the solvent basicity parameter (SB) of the solvent used,
52
II 4SIm
3
2
6
2
2
3
SIm
FeL
](BF ) ·0.5H O (2·0.5H O). Unlike the strong
3 4 2 2 2
2
with the solution T1/2 values for [Fe (L ) ](BF ) ·0.5H O (2·
23 4 2 2
correlations seen for those neutral complexes (R = 0.95 and
0
.5H O). Solvents, in order of increasing T : CDCl , CD Cl ,
2 1/2 3 3 2
1
9
0
.97), the T1/2 values for this salt, 2·0.5H O, only varied
CD NO , CD CN, and (CD ) CO.
2
3 2 3 3 2
from 214 to 259 K regardless of the solvent polarity index
2
(
Figure 6, green; R = 0.25). In contrast, the solvent basicity
For 2·0.5H O, the additional band at 530 nm is associated
2
5
2
1
parameter (SB), based on hydrogen bond acceptor ability,
with the Laporte-forbidden, spin-allowed d−d transition, A
1
g
1
shows a strong positive correlation with T1/2 for this salt, 2·
0
→ T , for LS Fe(II). The intensity of this 530 nm band
1g
2
.5H O (Figure 6, blue; R = 0.90). However, given the many
decreased with increasing temperature, consistent with the
2
possible parameters and indices available to describe solvents
see the Supporting Information for more details), and the
fraction of HS Fe(II) centers, γ , increasing, as anticipated
(Figure S36 in the Supporting Information). The extinction
HS
(
narrow T1/2 range observed in this case, this finding should be
viewed with considerable caution.
coefficient (used here as a proxy for the fraction of LS complex
present), ε, decreased from 90 to 34 L mol− cm on
increasing the temperature from 253 to 303 K. No significant
change in extinction coefficient was observed at the weak 870
nm band across all measured temperatures (similar to a
1
−1
At room temperature (293 K) in acetonitrile, the UV−vis
spectrum of 1·4H O showed a d−d band at a λ /nm (ε per
2
max
−
1
−1
mole of complex/L mol cm ) value of 870 (6.2) (Figure
27
S33 in the Supporting Information) and 2·0.5H O showed d−
previous study ).
2
d bands at 530(39) and 870(14) (Figure 7). The weak band in
the 870 nm region is in both cases attributed to the spin-
The χ T values obtained from the Evans method study (also
m
in MeCN) were used to calculate the set of corresponding γHS
values. The γHS values were then plotted against the ε values
5
5
allowed d−d transition, T2 → E , for HS Fe(II).
g
g
D
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 8. Cyclic voltammograms of 1 mM solutions of
Figure 7. Variable-temperature UV−vis study of a 4.8 mM
II 4NMeIm
II 4SIm
[
(
Fe (L
) ](BF ) (1·4H O, green) and [Fe (L ) ](BF )
3 4 2 2 3 4 2
acetonitrile solution of 2·0.5H O. Note: ε is calculated per mole of
−1
2
2·0.5H O, orange), in MeCN (0.1 mol L TBAClO ) versus 0.01
2
4
complex so that it can be used to monitor the spin state as a function
of T.
−1
−1
mol L Ag/AgNO at 200 mV s .
3
Table 1. Comparison of Cyclic Voltammetry Data Obtained
for Iron(II) Complexes 1·4H O and 2·0.5H O with Those
2
2
azine
40
of [Fe(L ) ](BF ) , in MeCN versus 0.01 M Ag/AgNO3
3
4 2
obtained from the UV−vis study, enabling a calibration line to
be fitted and in turn conversion of the ε values into γHS values
−
1
at 200 mV s
(
Figures S34 and S35 in the Supporting Information). From
E (V)
m
complex
[Fe(L4NMeIm)
](BF
Epc (V) Epa (V)
(ΔE/V)
ref
this analysis, heating the MeCN solution of 2·0.5H O from
2
2
53 to 303 K increased the γHS value from 0.51 to 0.77 (Figure
and Table S14 in the Supporting Information). The fitting of
3
)
4
2
(1·
0.36
0.46
0.41 (0.10) this
4
H O)
work
this
work
2
7
4
SIm
27,50
[Fe(L ) ](BF ) (2·
0.68
irrev
3
4 2
these UV−vis data with the regular solution model,
to a
0
.5H O)
2
pyridine
full SCO from [LS] → [HS], gave a T1 value of 245 K for 2·
/2
[Fe(L
[Fe(L
[Fe(L
[Fe(L
[Fe(L
) ](BF )
4 2
0.62
0.77
0.73
0.69
0.87
0.88
0.91
0.99
0.66 (0.07) 40
0.82 (0.10) 40
0.80 (0.15) 40
irrev
irrev
3
0
.5H O in MeCN, which unsurprisingly is within experimental
pyridazine
2
) ](BF )
4 2
3
error the same as that determined from the Evans method data
(
2pyrimidine
4pyrimidine
pyrazine
) ](BF )
4 2
3
249 K).
) ](BF )
4 2
40
40
3
Electrochemical Studies. The step-change modification
) ](BF )
4 2
3
of the Rat ligands described herein, from azine-triazoles to
azole-triazoles, was carried out in order to profoundly reduce
the ligand field experienced by the coordinated Fe(II) center
and has been successful in this aim, as proven above. Clearly
the redox properties of the resulting azole-triazole complexes,
CONCLUSION
■
In summary, we have reported the successful extension of our
general protocol to access azine-triazoles, to include the
preparation of the first examples of the analogous azole-triazole
1
·4H O and 2·0.5H O, are also expected to be very different
2 2
II azine -
based ligands, bidentate L4
NMeIm
and L . From these new
ligands the corresponding tris-ligand Fe(II) complexes
4SIm
from those of the analogous azine-triazoles, [Fe L
3]
4
0
(BF ) .
4 2
II 4NMeIm
II 4SIm
To test this expectation, cyclic voltammetry studies were
[
Fe (L
) ](BF ) (1) and [Fe (L ) ](BF ) (2)
3 4 2 3 4 2
carried out on both complexes as 1 mM solutions in
acetonitrile, with 0.1 M TBAClO , versus 0.01 M Ag/AgNO
were made. Single-crystal X-ray structure determinations at
100 K revealed that both crystallized as the mer isomer and
that 1 was HS while 2 was LS. The high-spin fractions of 1·
4H O and 2·0.5H O were monitored as a function of
4
3
with a platinum working electrode (Figure 8, Table 1, and
Figures S37 and S38 in the Supporting Information). The one-
2
2
electron Fe(II)/Fe(III) process for complex 1·4H O was
temperature in both the solid and solution states.
2
reversible, with E = 0.41 and ΔE = 0.10 V (compared with
In the solid-state study, 1·4H O remained HS whereas 2·
m
2
+
ΔE(Fc /Fc) = 0.08 V). In contrast this process is irreversible
0.5H O underwent a two-step SCO with T values of 275
2
1/2
for the thiazole-triazole complex 2·0.5H O, with E = 0.68 V,
and 350 K. These findings, for this pair of tris-azole-triazole
iron(II) complexes, are in stark contrast with the literature
reports of tris-azine-triazole complexes, all of which remained
LS. These findings confirm that, as expected, the ligand field
strength imposed by the azoles is weaker than that imposed by
the azines.
2
pa
consistent with it being 0.22 V less prone to oxidation than the
methylimidazole-triazole complex 1·4H O (E = 0.46 V, Table
2
pa
1
). This process, for 2·0.5H O, remained irreversible when a
2
glassy-carbon working electrode was employed (Figure S39 in
the Supporting Information). The HS [Fe (L
II 4NMeIm
) ](BF )
3 4 2
has a much lower E value in comparison to those of all of the
Variable-temperature Evans NMR studies of 1·4H O and 2·
pa
2
LS complexes, both the thiazole-triazole complex 2·0.5H O
0.5H O in CD CN solution showed that the former was HS
2
2
3
reported herein and all five of the azine-triazole family
while the latter was SCO active. This mirrors the solid-state
findings, but it is interesting to note that the T1/2 values for 2·
II azine
40
[
Fe (L ) ](BF ) reported earlier (Table 1).
3
4 2
E
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
.5H O are lower in CD CN solution (249 K) than in the
Article
0
temperature being within the smaller value of ±0.5 K or ±0.5% of the
target value) and an applied field of 0.1 T with a Quantum Design
Physical Property Measurement System equipped with a vibrating
sample mount (VersaLab). The data were corrected for the
diamagnetism of the capsule and of the sample ((−M × 0.5) ×
cm mol ).
The single-crystal X-ray diffraction measurements were performed
on an Oxford Diffraction SuperNova diffractometer with an Atlas
CCD apparatus, equipped with a Cryostream N open-flow cooling
device using a mirror monochromated microfocus Cu Kα radiation
source (λ = 1.56 Å) at 100 K. A series of scans was performed in such
a way as to collect a complete set of unique reflections (and many
equivalents) to a maximum θ value of 72°. Raw frame data (including
data reduction, interframe scaling, unit cell refinement, and absorption
corrections) for all structures were processed using CrysAlis Pro.
Structures were solved using ShelXT and using intrinsic phasing and
refined using all data by least-squares techniques with the SHELXL
and olex2 programs. All hydrogen atoms were placed in calculated
positions riding on the attached carbon atom, with U(H) =
1.2[U(attached C)], except for the methyl groups and water molecule,
for which U(H) = 1.5[U(attached atom)]. All non-H atoms were
2
3
solid state (275 and 350 K). The SCO in CH CN was also
3
monitored by following the intensity of the d−d transition at
5
30 nm, from 253 to 303 K, which increased the γ value
HS
from 0.51 to 0.77.
−6
3
−1 50
10
The SCO-active complex 2·0.5H O was then studied in a
2
range of deuterated solvents to test for a possible correlation
between the solution T1/2 and solvent polarity index P′. All of
the T1/2 values fell in a relatively narrow range (214−259 K).
There was no correlation between the T values and the
2
1
/2
2
solvent polarity index P′ (R = 0.25), but there was a good
2
correlation with the solvent basicity parameter SB (R = 0.90).
In this particular case, given the narrow T1/2 range observed as
a function of solvent, this finding should be regarded with
caution. It is clear that more such studies are needed, as there
are a great many parameters and indices available to describe
solvents (see the Supporting Information for more details),
and it is unclear yet which of these are the most robust and
appropriate for this purpose and may in time provide a means
of predictable tuning of solution T1/2 values.
55
56
57
58
59
refined anisotropically.
FeL4
NMeIm
](BF ) ·CH NO ·solvent (1·solvent) crystallized in
[
Finally, cyclic voltammetry studies at 293 K in acetonitrile
solution showed that the methylimidazole-triazole complex 1·
3 4 2 3 2
̅
the space group P1. There was one entire complex, two
tetrafluoroborate anions, and one nitromethane molecule in the
asymmetric unit. The methyl group on one of the three independent
methylimidazole rings is disordered over two positions, C101:C201
4
H O (E = 0.46 V; reversible with E = 0.41 V) is 0.22 V
2 pa m
more prone to oxidation than the thiazole-triazole complex 2·
0
.5H O (E = 0.68 V, irreversible).
2
pa
(occupancies 0.7:0.3). One of the two independent BF anions is
disordered, with three full-occupancy F atoms (shared by both
4
This proof of principle study opens the door to (a) further
fine tuning of both T1/2 and redox potential and (b) the use of
ditopic analogues of these monotopic azole-triazoles in the self-
assembly of new families of SCO-active helicates and cages.
These studies are underway in our laboratories.
positions of the BF ) and 0.7:0.3 occupancy for (B2/F7):(B3/F9).
4
One full-ccupancy nitromethane of crystallization was included in the
refinement, but the remaining solvent molecules of crystallization
were highly disordered and could not be modeled; therefore,
46
47
SQUEEZE in PLATON was applied. The electron density
found by SQUEEZE was 211 electrons/cell: i.e., 105 electrons per
complex (Z = 2). This is in agreement with the presence of about
another three molecules of nitromethane (3 × 32 = 96 electrons) per
EXPERIMENTAL SECTION
Instrumentation and Measurements. Elemental analyses (C,
H, N, S) were measured at the Campbell Microanalytical Laboratory,
University of Otago. TGA was recorded on a TA Instruments Q50
■
4
NMeIm
complex, corresponding to an overall formula of [FeL
4CH NO . For further details, see Table S1 and the cif file.
[FeL
space group P1
](BF ) ·
3
4 2
3
2
−
1
4SIm
apparatus with the samples in a platinum pan heated at 5 °C min .
3
](BF
) ·0.125H O·solvent (2·solvent) crystallized in the
4 2 2
−
1
The nitrogen gas flow was set to 30 mL min over the balance and
̅
. Two entire complexes, four tetrafluoroborate anions,
−
1
1
13
3
0 mL min over the sample. H and C NMR spectra were
and a quarter-occupancy water molecule were present in the
asymmetric unit. The F atoms of one of the four independent BF4
anions, B4, were disordered over two positions with occupancies of
0.6 (F41, F42, F43, and F44) and 0.4 (F45, F46, F47, and F48). The
other solvent molecules of crystallization were highly disordered and
1
5
recorded on a Varian 400 MHz NMR spectrometer at 298 K. N and
1
H VT-NMR ranging from −30 to 90 °C in CD CN, CD NO ,
3
3
3
C D CO, CD Cl , and CDCl were recorded on a Varian 500 MHz
2
6
2
2
3
4
NMeIm
NMR spectrometer. ESI-MS measurements of ligands L
and
4
SIm
46
47
L
were recorded, in chloroform at 293 K, in house using a Bruker
could not be modeled; therefore, SQUEEZE in PLATON was
applied. The electron density found by SQUEEZE was 76 electrons/
cell: i.e., 19 electrons per complex (Z = 4). This is in agreement with
the presence of about one molecule of acetonitrile (1 × 22 = 22
MicrOTOF-Q mass spectrometer; the m/z values have a standard
error of ±10 ppm. High-resolution ESI-MS measurements of 1·
4
H Oand 2·0.5H O were recorded, in acetonitrile at 293 K (with
2 2
heating of the source and nebulizer gas switched off), at Bio21
University of Melbourne), using a Shimadzu LCMS_9030 mass
electrons) per complex (or two per asymmetric unit), corresponding
to an overall formula of [FeL4
SIm
](BF ) ·0.125H O·CH CN.
3 4 2 2 3
(
spectrometer; the m/z values have a standard error of ±1 ppm.
For further details, see Table S1 and the cif file. Crystallographic
data for the structures have been deposited with the Cambridge
Crystallographic Data Center: CCDC 1956926 and 1956927.
Synthesis. All chemicals were used as commercially available
without further purification. N-(p-Tolyl)benzothioamide was pre-
Variable-temperature UV−vis spectra were recorded from −20 to 30
°
C on a PerkinElmer Lambda 950 UV−vis/NIR spectrometer.
Electrochemistry measurements were performed with a two-electrode
electrochemical cell using an Iviumstat.XRe potentostat, on 1 mM
samples of 1·4H O and 2·0.5H O, in degassed, freshly distilled
40
pared as previously described. Dry ethanol was prepared by distilling
absolute ethanol from Mg/I
Ethyl 1-Methyl-1H-imidazole-4-carboxylate. This compound was
2
2
−
1
MeCN (over CaH ), with TBAClO (0.1 mol L ) as the supporting
electrolyte and 0.01 mol L Ag/AgNO . Both the counter electrode
.
2
2
4
−
1
3
60
and working electrode were Pt (in response to a helpful suggestion
from the reviewers, a glassy-carbon working electrode was trialed for
synthesized as reported in the literature. To a dry ethanol (30 mL)
solution of 1-methyl-1H-imidazole-4-carboxylic acid (3.3 g, 26 mmol)
cooled in an ice bath was added thionyl chloride (6 mL, 90 mmol)
dropwise with continuous stirring. The resulting light brown solution
was stirred at 80 °C overnight. After it was cooled to room
temperature, the dark brown reaction mixture was taken to dryness
2
·0.5H O; see Figure S39 in the Supporting Information). The
2
+
potential values are referenced to Fc/Fc redox couple at E = 0.08 ±
m
−
1
0.01 V with ΔE = 0.08 V at a 200 mV s scan rate.
The solid-state magnetic susceptibilities were measured from 300−
5
0−400 K for 1·4H O in sweep mode (measuring continuously as the
under vacuum. Saturated aqueous NaHCO (50 mL) was added to
2
3
−1
temperature was swept at 5 K min ) and for 2·0.5H O in settle
mode(measuring in 5 K steps, ramping between steps at 5 K min ;
the resulting dark brown solid, with stirring, to neutralize it. The
product was then extracted with dichloromethane (2 × 20 mL) and
the combined organic layers were dried over MgSO4 and then
2
−
1
the instrument considers the temperature “settled” after 1 min of the
F
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
concentrated under vacuum to give a pale yellow oil that was kept in
the freezer overnight, giving an off-white solid (3.9 g, 97%). Anal.
Calcd for C H N O (M = 154): C, 54.54; H, 6.54; N, 18.17.
°C, and then N-(p-tolyl)benzothioamide (1 g, 4.4 mmol) was added
and the reaction mixture stirred in the ice bath for 15 min, before the
addition of EtBr (400 μL, 5.4 mmol). The reaction mixture was
heated at 50 °C for 5 h and then cooled to room temperature. The
solvent was evaporated under vacuum, and the product was dissolved
in dichloromethane (30 mL) and then washed with water (2 × 20
mL). The organic fraction was dried over MgSO4 and then
concentrated under vacuum to give a yellow oil (1.1 g, 98%). The
oil was promptly dissolved in n-butanol (15 mL) and solid thiazole-4-
carbohydrazide (977 mg, 6.4 mmol) added. The reaction mixture was
refluxed for 3 days at 108 °C, resulting in a dark yellow solution with a
white precipitate. After the mixture was cooled to room temperature,
the solid was collected by filtration, washed with diethyl ether (4 × 4
7
10
2
2
r
1
Found: C, 54.53; H, 6.69; N, 18.00. NMR (CDCl , H) ppm: 7.55 (s,
3
1
H), 7.43 (s, 1H), 4.34 (q, 2H), 1.35 (t, 3H).
1
-Methyl-1H-imidazole-4-carbohydrazide. This compound was
61
synthesized as reported in the literature. To an ethanol (8 mL)
solution of ethyl 1-methyl-1H-imidazole-4-carboxylate (1 g, 6.4
mmol) was added 80% hydrazine hydrate (4 mL, 80 mmol). The
reaction mixture was refluxed for 4 h and then cooled to room
temperature, which caused a white precipitate to form. This was
collected by filtration, washed with chilled ethanol, and dried for 2 h
at room temperature to give the crude product as a white solid (990
4
SIm
mg, 88%). Anal. Calcd for C H N O (M = 140): C, 42.85; H, 5.75;
5
8
4
r
mL) and water (2 × 6 mL), and then air-dried overnight to give L
as a white solid (1.04 g, 76% yield). Anal. Calcd for C H N S·
1
N, 39.98. Found: C, 42.77; H, 5.39; N, 39.23. NMR (DMSO-d , H)
6
1
8
14
4
ppm: 8.96 (b, 1H); 7.61 (d, 2H); 4.30 (b, 2H); 3.67 (s, 3H).
Ethyl Thiazole-4-carboxylate. To a dry ethanol (30 mL) solution
of thiazole-4-carboxylic acid (2.4 g, 18 mmol) cooled in an ice bath
was added thionyl chloride (6 mL, 90 mmol) dropwise with
continuous stirring. The colorless solution was stirred at 80 °C
overnight. After it was cooled to room temperature, the pale yellow
reaction mixture was taken to dryness under vacuum. The oily crude
was dissolved in 10 mL of water followed by addition of saturated
0
6
.1H O (M = 320): C, 67.52; H, 4.47; N, 17.50; S, 10.01. Found: C,
2 r
+
7.32; H, 4.49; N, 17.75; S, 10.02. MS (m/z): [(C H N S)H]
1
8
14
4
+
calcd 319.1017, found 319.1024; [(C H N S)Na] calcd 341.0837,
found 341.0800; [(C H N S) Na] calcd 659.1776, found
6
1
8
14
4
+
1
8
14
4
2
1
59.1782. H NMR (CDCl , 400 MHz) ppm: 8.67 (s, 1H), 7.69
3
(
(
d, 1H), 7.42 (d, 2H), 7.30 (t, 1H), 7.25 (t, 2H), 7.18 (d, 2H), 7.10
13
d, 2H), 2.38 (s, 3H). C NMR (CDCl , 100 MHz): 154.89 (trzC7),
3
1
1
1
2
52.91 (azoC16), 150.47 (trzC9), 143.61 (tolc20), 139.77 (azoC13),
32.52 (tolC12), 130.19 (tolC21), 129.67 (bezC6), 128.68 (bezC4),
aqueous NaHCO (50 mL) with stirring, to neutralize it. The product
3
was then extracted with dichloromethane (3 × 30 mL), and the
combined organic layers were dried over MgSO4 and then
concentrated under vacuum to give a dark brown oil that was kept
in a freezer overnight to afford a light brown solid (2.89 g, 96%). Anal.
Calcd for C H NO S (M = 157): C, 45.85; H, 4.49; N, 8.91; S,
28.37 (tolC22), 127.61 (bezC5), 126.71 (bezC3), 120.48 (azoC14),
15
1.31 (tolC23). N NMR (CDCl , 500 MHz) ppm (indirect
3
detection): 285 (Nthiazole), 145 (Ntriazole).
4NMeIm
[
FeL
.03 mmol) was added to a nitromethane (4 mL) solution of L
.2H O (30 mg, 0.08 mmol). The resultant colorless mixture was
](BF ) ·4H O (1·4H O). Solid Fe(BF ) ·6H O (10 mg,
3 4 2 2 2 4 2 2
4NMeIm
6
7
2
r
0
2
·
2
0.40. Found: C, 45.93; H, 4.32; N, 8.16; S, 20.31. NMR (CDCl3,
1
2
H) ppm: 8.83 (s, 1H), 8.23 (s, 1H), 4.42 (q, 2H), 1.40 (t, 3H).
Thiazole-4-carbohydrazide. To an ethanol (4 mL) solution of
ethyl thiazole-4-carboxylate (2.6 g, 16 mmol) was added 80%
hydrazine hydrate (8 mL, 160 mmol). The reaction mixture was
stirred at room temperature overnight. The resulting white solid was
collected by filtration, washed with chilled ethanol, and dried for 2 h
in air (1.5 g, 65%). Anal. Calcd for C H N OS (M = 143): C, 33.56;
stirred for 2 h at room temperature. The resulting clear colorless
solution was vapor-diffused with diethyl ether. After 1 month,
colorless crystals of 1·solvents, suitable for X-ray crystallography, were
obtained. These were collected by filtration and dried overnight in air
to give the off-white solid 1·4H O (30 mg, 81%). Anal. Calcd for
2
C H N B F Fe·4H O (M = 1247.6); C, 54.87; H, 4.77; N, 16.84.
5
7
51 15
2
8
2
r
4
5
3
r
Found: C, 54.79; H, 4.81; N, 16.94. TGA: calcd 6.17%, found 5.78%.
H, 3.52; N, 29.35; S, 22.39. Found: C, 33.77; H, 3.54; N, 29.30; S,
2+
Cryo MS (m/z): [Fe(C H N ) (CH CN)] calcd. 206.054393,
1
19 17
5
1
3
2
2.47. NMR (DMSO-d , H) ppm: 9.63 (s, 1H); 9.14 (d, 1H); 8.27
2+
6
found 206.05505; [Fe(C H N )(CH CN) ] calcd. 226.56766,
1
9
17
5
3
2
(
d, 1H), 4.49 (d, 2H).
2+
found 226.56729; [Fe(C H N ) ] : calcd. 343.11531, found
1
9
17
5 2
3
-(1-Methyl-1H-imidazol-4-yl)-5-phenyl-4-(p-tolyl)-4H-1,2,4-tria-
2+
4NMeIm
343.11671; [Fe(C19
H
17
N
5
)
2
(CH CN)Cl]+H : calcd 382.12084,
3
zole (L
dry ethanol at 0 °C, and then N-(p-tolyl)benzothioamide (800 mg,
.5 mmol) was added and the reaction mixture stirred in an ice bath
). Sodium (300 mg) was cautiously dissolved in 30 mL of
2
+
found 382.12410; [Fe(C H N ) ] calcd. 500.68951, found
00.69141; [Fe(C H N ) ] calcd. 658.26371, found 658.26580;
Fe(C H N ) ](BF ) calcd. 773.23409, found 773.23324; [Fe-
1
9
17
5 3
2
+
5
19 17 5 4
3
+
[
(
(
(
1
9
17
5
2
4
for 15 min, before the addition of EtBr (400 μL, 5.4 mmol). The
reaction mixture was heated at 50 °C for 5 h and then cooled to room
temperature. The solvent was evaporated under vacuum and the
residue dissolved in dichoromethane (30 mL) and then washed with
2
+
C H N ) ]
19 17
calcd. 816.33940, found 816.33880; [Fe-
1
9
17
5 5
+
C H N ) ](BF ) calcd 1088.38249, found 1088.38863. UV−vis
5
3
4
−
1
−1
293 K in MeCN): 870 nm (ε = 6.2 L mol cm ).
[
FeL4 ](BF ) ·0.5H O (2·0.5H O). Solid Fe(BF ) ·6H O (30 mg,
3 4 2 2 2 4 2 2
4SIm
SIm
water (2 × 20 mL). The organic fraction was dried over MgSO and
4
0
0
.089 mmol) was added to an acetonitrile (10 mL) solution of L
·
then concentrated under vacuum to give a light yellow oil (886 mg,
.1H O (90 mg, 0.28 mmol). The resultant orange-yellow clear
9
1
8%). The oil was promptly dissolved in n-butanol (8 mL), and then
-methyl-1H-imidazole-4-carbohydrazide (550 mg, 3.9 mmol) was
2
solution was stirred for 2 h at room temperature, then vapor diffused
with diethyl ether. After few weeks, orange block shaped crystals of 2·
solvents formed, suitable for X-ray crystallography. These were
collected by filtration, washed with diethyl ether, and then air-dried
added and the reaction mixture refluxed for 3 days at 108 °C,
resulting in a pale yellow precipitate which, after the mixture was
cooled to room temperature, was collected by filtration, washed with
overnight to obtain 2·0.5H O as a dark orange-brown solid (80 mg,
diethyl ether (4 × 5 mL) and water (4 × 3 mL), then air-dried
2
overnight, to give L4
NMeIm
as a white solid (710 mg, 72%). Anal. Calcd
7
5
6%). Anal. Calcd for C H N S B F Fe·0.5H O (M = 1193.6): C,
54 42 12 3 2 8 2 r
4.34; H, 3.63; N, 14.08; S, 8.06. Found: C, 54.35; H, 3.65; N, 13.87;
for C H N ·2.2H O (M = 355): C, 64.28; H, 6.08; N, 19.73.
1
9
17
5
2
r
+
S, 8.10. TGA: calcd 1.2%, found 0.75%. Cryo MS (m/z):
Found: C, 64.11; H, 5.60; N, 19.28. MS (m/z): [(C H N ) Na]
calcd 338.1382, found 338.1357; [(C H N ) Na] calcd 653.2866,
found 653.2742. H NMR (CDCl , 400 MHz) ppm: 7.45 (d, 2H),
1
9
17
5
+
2+
[
Fe(C H N S)(CH CN)] calcd 207.5271, found 207.52765;
18 14
14
14
4
3
1
9
17
5 2
1
]2+ calcd 228.0404, found 228.03925;
calcd 346.06123, found 346.06123; [Fe-
CN)Cl]+H
[Fe(C18
[Fe(C18
(C18
H
H
N
N
S)(CH
CN)
3
4
3
2
2
+
7
7
.35 (s, 1H), 7.30 (d, 2H), 7.28 (s, 1H), 7.21 (m, 2H), 7.16 (s, 1H),
4
S)
(CH
385.06951; [Fe(C18
2
]
1
3
2+
.14 (d, 2H), 3.65 (s, 3H), 2.41 (s, 3H). C NMR (CDCl , 100
H
14
N
4
S)
2
3
H
calcd 385.066364, found
3
2
+
MHz): 154.37 (trzC4), 150.60 (trzC1), 139.55 (azoC10), 138.30
N
14
4
S)
3
]
calcd 505.1077, found 505.1108;
+
(
(
(
(
tolC20), 132.73 (tolC17), 130.15 (tolC19), 129.92 (bezC7), 129.42
bezC15 and C13), 128.58 (azoC8), 128.31 (bezC14), 127.75
tolC18), 127.01 (bezC16 and C12), 121.34 (azoC6), 33.64
azoC23), 21.38 (tolC24).
[Fe(C18
(C18
(C18
(C18
H
N
14
N
4
S)
](BF
4
]2 calcd 664.15475, found 664.15348; [Fe-
H
S)
)+ calcd 779.1256, found 779.12748; [Fe-
14
4
2
4
2
+
H
N
S)
]
calcd 823.7032, found 823.70332; [Fe-
) calcd 1097.2208, found 1097.22332. UV−vis
(293 K in MeCN): 530 nm (ε = 39 L mol cm ) and 870 nm (ε =
14
4
5
+
H
14
N
4
S)
](BF
3
4
-(5-Phenyl-4-(p-tolyl)-4H-1,2,4-triazol-3-yl)thiazole (L4SIm). So-
−1
−1
4
−
1
−1
dium (300 mg) was cautiously dissolved in 30 mL of dry ethanol at 0
14 L mol cm ).
G
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
11) Senthil Kumar, K.; Ruben, M. Emerging trends in spin
ASSOCIATED CONTENT
■
crossover (SCO) based functional materials and devices. Coord. Chem.
Rev. 2017, 346, 176−205.
(
complex with a 70 K wide thermal hysteresis loop. Angew. Chem., Int.
Ed. 2008, 47, 10098−10101.
(
Dudley, S.; Barrett, S. A.; Little, M. A.; Deeth, R. J.; Halcrow, M. A. A
unified treatment of the relationship between ligand substituents and
spin state in a family of iron(II) complexes. Angew. Chem., Int. Ed.
*
S
Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.inorgchem.9b02993.
12) Weber, B.; Bauer, W.; Obel, J. An iron(II) spin-crossover
Additional crystallographic, magnetic, and cyclic voltam-
metry data and figures, as well as UV−vis, NMR, and
mass spectra (PDF)
13) Kershaw Cook, L. J.; Kulmaczewski, R.; Mohammed, R.;
Accession Codes
2
(
016, 55, 4327−4331.
CCDC 1956926−1956927 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
14) Phonsri, W.; Macedo, D. S.; Vignesh, K. R.; Rajaraman, G.;
Davies, C. G.; Jameson, G. N. L.; Moubaraki, B.; Ward, J. S.; Kruger,
P. E.; Chastanet, G.; Murray, K. S. Halogen Substitution Effects on
N2O Schiff Base Ligands in Unprecedented Abrupt FeII Spin
Crossover Complexes. Chem. - Eur. J. 2017, 23, 7052−7065.
(
15) Shores, M. P.; Klug, C. M.; Fiedler, S. R. Spin-state switching in
solution. In Spin-Crossover Materials: Properties and Applications;
Wiley: Chichester, U.K., 2013; pp 281−301.
AUTHOR INFORMATION
■
(16) Turner, J. W.; Schultz, F. A. Solution characterization of the
Corresponding Author
E-mail for S.B.: sbrooker@chemistry.otago.ac.nz.
iron(II) bis(1,4,7-triazacyclononane) spin-equilibrium reaction. Inorg.
Chem. 2001, 40, 5296−5298.
*
ORCID
(17) Barrett, S. A.; Kilner, C. A.; Halcrow, M. A. Spin-crossover in
Sandhya Singh: 0000-0002-9123-8182
Sally Brooker: 0000-0002-5878-8238
Notes
[
Fe(3-bpp) ][BF ] in different solvents - A dramatic stabilisation of
2 4 2
the low-spin state in water. Dalton Trans. 2011, 40, 12021−12024.
(18) Kershaw Cook, L. J.; Mohammed, R.; Sherborne, G.; Roberts,
T. D.; Alvarez, S.; Halcrow, M. A. Spin state behavior of iron(II)/
dipyrazolylpyridine complexes. New insights from crystallographic
and solution measurements. Coord. Chem. Rev. 2015, 289, 2−12.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(19) Rodríguez-Jimenez, S.; Barltrop, A. S.; White, N. G.; Feltham,
́
We thank the University of Otago (including a doctoral
scholarship to S.S. and the 2015 purchase of a Versalab
magnetometer). S.S. is very thankful to Dr. Humphrey Feltham
H. L. C.; Brooker, S. Solvent polarity predictably tunes spin crossover
1/2 in isomeric iron(II) pyrimidine triazoles. Inorg. Chem. 2018, 57,
266−6282.
20) Strauß, B.; Gutmann, V.; Linert, W. Spin-crossover complexes
in solution, II solvent effects on the high spin-low spin-equilibrium of
Fe(bzimpy)2](ClO4)2. Monatsh. Chem. 1993, 124, 515−522.
21) Linert, W.; Enamullah, M.; Gutmann, V.; Jameson, R. F. Spin-
T
6
(
(
Otago) for training in variable-temperature UV−vis studies.
We are grateful to Chris Bowen (Shimadzu Scientific
Australasia) for kindly running the ESI MS of complexes 1
and 2 at Bio21 (Melbourne), Mr. Ian Stewart (Otago) for
[
(
1
5
1
crossover complexes in solution. III. Substituent- and solvent effects
on the spin-equilibrium of 4-substituted iron(II)-2,6-bis-(benzimida-
zol-2′-yl)-pyridine systems. Monatsh. Chem. 1994, 125, 661−670.
measuring the N and VT H NMR spectra, and Assoc. Prof.
Aaron Marshall (Canterbury) for advice on error analysis.
(22) Haasnoot, J. G. Mononuclear, oligonuclear and polynuclear
REFERENCES
■
metal coordination compounds with 1,2,4-triazole derivatives as
ligands. Coord. Chem. Rev. 2000, 200−202, 131−185.
(
1) Gu
̈
tlich, P.; Goodwin, H. A. Top. Curr. Chem. 2004, 233, 1−47.
tlich, P.; Gaspar, A. B.; Garcia, Y. Spin state switching in iron
(
2) Gu
̈
(23) Aromí, G.; Barrios, L. A.; Roubeau, O.; Gamez, P. Triazoles and
tetrazoles: Prime ligands to generate remarkable coordination
materials. Coord. Chem. Rev. 2011, 255, 485−546.
coordination compounds. Beilstein J. Org. Chem. 2013, 9, 342−391.
(
3) Hogue, R. W.; Singh, S.; Brooker, S. Spin crossover in discrete
polynuclear iron(II) complexes. Chem. Soc. Rev. 2018, 47, 7303−7338
and inside front cover.
(24) Feltham, H. L. C.; Barltrop, A. S.; Brooker, S. Spin crossover in
iron(II) complexes of 3,4,5-tri-substituted-1,2,4-triazole (Rdpt), 3,5-
(
4) Gu
̈
tlich, P.; Goodwin, H. A. Spin Crossover in Transition Metal
−
di-substituted-1,2,4-triazolate (dpt ), and related ligands. Coord.
Compounds I-III. Topics in Current Chemistry; Springer: Berlin
Chem. Rev. 2017, 344, 26−53.
Heidelberg, 2004; Vol. 233-235.
(25) Kulmaczewski, R.; Olguín, J.; Kitchen, J. A.; Feltham, H. L. C.;
(
5) Halcrow, M. A. Spin-Crossover Materials: Properties and
Applications, 1st ed.; Wiley: Chichester, U.K., 2013.
6) Kahn, O.; Martinez, C. J. Spin transition polymers: from
molecular materials toward memory devices. Science 1998, 279, 44−
8.
7) Halcrow, M. A. Structure:function relationships in molecular
spin-crossover complexes. Chem. Soc. Rev. 2011, 40, 4119−4142.
8) Molnar, G.; Salmon, L.; Nicolazzi, W.; Terkib, F.; Bousseksou, A.
Jameson, G. N. L.; Tallon, J. L.; Brooker, S. Remarkable scan rate
dependence for a highly constrained dinuclear iron(II) spin crossover
complex with a record thermal hysteresis loop. J. Am. Chem. Soc.
(
2
(
014, 136, 878−881.
4
(
26) Klingele, M. H.; Moubaraki, B.; Cashion, J. D.; Murray, K. S.;
Brooker, S. The first X-ray crystal structure determination of a
II
dinuclear complex trapped in the [low spin-high spin] state: [Fe
(
́
2(
cover feature.
27) Hogue, R. W.; Feltham, H. L. C.; Miller, R. G.; Brooker, S. Spin
PMAT) ](BF ) ·DMF. Chem. Commun. 2005, 987−989 and front
2
4 4
Emerging properties and applications of spin crossover nanomaterials.
J. Mater. Chem. C 2014, 2, 1360−1366.
́
9) Lefter, C.; Davesne, V.; Salmon, L.; Molnar, G.; Demont, P.;
Rotaru, A.; Bousseksou, A. Charge transport and electrical properties
of SCO materials: towards nanoelectronic and spintronic devices.
Magnetochemistry 2016, 2, 18.
10) Khusniyarov, M. M. How to Switch Spin-Crossover Metal
Complexes at Constant Room Temperature. Chem. - Eur. J. 2016, 22,
5178−15191.
(
(
crossover in dinuclear N4S iron(II) thioether-triazole complexes:
2
access to [HS-HS], [HS-LS] and [LS-LS] states. Inorg. Chem. 2016,
55, 4152−4165.
(28) Herold, C. F.; Carrella, L. M.; Rentschler, E. A Family of
Dinuclear Iron(II) SCO Compounds Based on a 1,3,4-Thiadiazole
Bridging Ligand. Eur. J. Inorg. Chem. 2015, 2015, 3632−3636.
(
1
H
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
29) Ko
̈
hler, C.; Rentschler, E. The First 1,3,4-Oxadiazole Based
(47) Spek, A. PLATON SQUEEZE: a tool for the calculation of the
disordered solvent contribution to the calculated structure factors.
Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 9−18.
(48) Evans, D. F. The determination of the paramagnetic
susceptibility of substances in solution by nuclear magnetic resonance.
J. Chem. Soc. 1959, 2003−2005.
(49) Schubert, E. M. Utilizing the Evans Method with a
Superconducting NMR Spectrometer in the Undergraduate Labo-
ratory. J. Chem. Educ. 1992, 69, 62.
Dinuclear Iron(II) Complexes Showing Spin Crossover Behavior with
Hysteresis. Eur. J. Inorg. Chem. 2016, 2016, 1955−1960.
30) Herold, C. F.; Shylin, S. I.; Rentschler, E. Solvent-dependent
SCO Behavior of Dinuclear Iron(II) Complexes with a 1,3,4-
Thiadiazole Bridging Ligand. Inorg. Chem. 2016, 55, 6414−6419.
31) Sheu, C.-F.; Chen, S.-M.; Wang, S.-C.; Lee, G.-H.; Liu, Y.-H.;
Wang, Y. The commensurate modulated structure of the metastable
state in spin crossover complex [Fe(abpt) (NCS) ]. Chem. Commun.
(
(
2
2
2
(
009, 7512−7514.
32) Schneider, C. J.; Cashion, J. D.; Chilton, N. F.; Etrillard, C.;
Fuentealba, M.; Howard, J. A. K.; Letard, J.-F.; Milsmann, C.;
(50) Kahn, O. Molecular Magnetism; VCH: New York, 1993.
(51) Yatsunyk, L. A.; Walker, F. A. Structural, NMR, and EPR
studies of S = 1/2 and S = 3/2 Fe(III) bis(4-cyanopyridine)
complexes of dodecasubstituted porphyrins. Inorg. Chem. 2004, 43,
757−777.
́
Moubaraki, B.; Sparkes, H. A.; Batten, S. R.; Murray, a. K. S. Spin
Crossover in a 3,5-Bis(2-pyridyl)-1,2,4-triazolate-Bridged Dinuclear
Iron(II) Complex [{Fe(NCBH3)(py)}2-(μ-L1)2] - Powder versus
Single Crystal Study. Eur. J. Inorg. Chem. 2013, 2013, 850−864.
(52) Catalan, J.; Díaz, C.; Lopez, V.; Perez, P.; De Paz, J.-L. G.;
́ ́ ́
Rodríguez, J. G. A generalized solvent basicity scale: The
solvatochromism of 5-nitroindoline and its homomorph 1-methyl-5-
nitroindoline. Liebigs Ann. Chem. 1996, 1996, 1785−1794.
(
33) Rodríguez-Jimenez, S.; Yang, M.; Stewart, I.; Garden, A. L.;
́
Brooker, S. A simple method of predicting spin state in solution. J.
Am. Chem. Soc. 2017, 139, 18392−18396.
(
(53) Ahuja, S. Chromatography and Separation Science; Academic
Press: 2003; Vol. 4.
54) Schirmer, R. E. Modern Methods of Pharmaceutical Analysis, 2nd
ed.; Taylor & Francis: 1990.
55) Sheldrick, G. M. SADABS: Empirical absorption correction
program for area detector data; University of Gottingen: Gottingen,
Germany, 1996.
34) Mason, H. E.; Hamilton, M. L.; Howard, J. A. K.; Sparkes, H. A.
(
[
Fe(abpt)2(NCSe)2] polymorph A: structural studies into the spin
crossover behaviour. New J. Chem. 2018, 42, 18028−18037.
35) Lada, Z. G.; Andrikopoulos, K. S.; Chrissanthopoulos, A.;
(
(
̈
̈
Perlepes, S. P.; Voyiatzis, G. A. A Known Iron(II) Complex in
Different Nanosized Particles: Variable-Temperature Raman Study of
Its Spin-Crossover Behavior. Inorg. Chem. 2019, 58, 5183−5195.
(
(
56) CryAlisPRO; Oxford Diffraction: 2011.
57) Sheldrick, G. M. SHELXT - Integrated space-group and crystal-
(
36) Yan, Z.; Liu, W.; Peng, Y.-Y.; Chen, Y.-C.; Li, Q.-W.; Ni, Z.-P.;
structure determination. Acta Crystallogr., Sect. A: Found. Adv. 2015,
1, 3−8.
58) Sheldrick, G. M. Crystal structure refinement with SHELXL.
Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8.
59) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.
Tong, M.-L. Spin-Crossover Phenomenon in a Pentanuclear Iron(II)
7
(
Cluster Helicate. Inorg. Chem. 2016, 55, 4891−4896.
(37) Yan, Z.; Ni, Z.-P.; Guo, F.-S.; Li, J.-Y.; Chen, Y.-C.; Liu, J.-L.;
Lin, W.-Q.; Aravena, D.; Ruiz, E.; Tong, M.-L. Spin-Crossover
Behavior in Two New Supramolecular Isomers. Inorg. Chem. 2014,
(
K.; Puschmann, H. OLEX2: a complete structure solution, refinement
5
(
3, 201−208.
and analysis program. J. Appl. Crystallogr. 2009, 42, 339−341.
38) Klingele, J.; Scherer, H.; Klingele, M. H. Mononuclear Low-
(60) Myles, L.; Gore, R. G.; Gathergood, N.; Connon, S. J. A new
Spin 3:1-Type Iron(II) Complexes of 4-Substituted 3,5-Bis(2-
pyridyl)- and 3-Phenyl-5-(2-pyridyl)-4H-1,2,4-triazoles. Z. Anorg.
Allg. Chem. 2009, 635, 2279−2287.
generation of aprotic yet Bronsted acidic imidazolium salts: low
toxicity, high recyclability and greatly improved activity. Green Chem.
2
(
013, 15, 2740−2746.
(39) Klingele, J.; Kaase, D.; Hilgert, J.; Steinfeld, G.; Klingele, M. H.;
61) Shafiee, A.; Naimi, E.; Mansobi, P.; Foroumadi, A.; Shekari, M.
Lach, J. Triazolopyridines as ligands: structural diversity in iron(II),
Syntheses of substituted-oxazolo-1,3,4-thiadiazoles, 1,3,4-oxadiazoles,
and 1,2,4-triazoles. J. Heterocycl. Chem. 1995, 32, 1235−1239.
cobalt(II), nickel(II) and copper(II) complexes of 3-(2-pyridyl)-
1
0
[
[
1,2,4]triazolo[4,3-a]-pyridine (L ) and spin crossover in
II 10
Fe (L ) (NCS) ]. Dalton Trans. 2010, 39, 4495−4507.
40) (a) Kitchen, J. A.; White, N. G.; Boyd, M.; Moubaraki, B.;
Murray, K. S.; Boyd, P. D. W.; Brooker, S. Iron(II) tris-[N -
substituted-3,5-di(2-pyridyl)-1,2,4-triazole] complexes: structural,
magnetic, NMR and density functional theory studies. Inorg. Chem.
2
2
(
4
2
009, 48, 6670−6679. (b) Rodríguez-Jimenez, S.; Bondì, L.; Yang,
́
M.; Garden, A. L.; Brooker, S. Predictable electronic tuning by choice
of azine substituent in five iron(II) triazoles: redox properties and
DFT calculations. Chem. - Asian J. 2019, 14, 1158−1166.
(
41) Singh, S.; Hogue, R. W.; Feltham, H. L. C.; Brooker, S.
Dinuclear helicate and tetranuclear cage assembly using appropriately
designed ditopic triazole-azine ligands. Dalton Trans. 2019, 48,
1
5435−15444.
42) Wiberg, K. B.; Nakaji, D.; Breneman, C. M. Azines. A
theoretical study of π-electron delocalization. J. Am. Chem. Soc. 1989,
11, 4178−4190.
43) Steel, P. J. Aromatic nitrogen heterocycles as bridging ligands; a
survey. Coord. Chem. Rev. 1990, 106, 227−265.
44) Klingele, M. H.; Brooker, S. From N-substituted thioamides to
(
1
(
(
symmetrical and unsymmetrical 3,4,5-trisubstituted 4H-1,2,4-tria-
zoles: synthesis and characterisation of new chelating ligands. Eur. J.
Org. Chem. 2004, 2004, 3422−3434.
(45) Joule, J. A.; Mills, K.; Smith, G. F. Heterocyclic Chemistry, 3rd
ed.; Chapman & Hall: Padstow, 1995.
46) van der Sluis, P.; Spek, A. L. BYPASS: an effective method for
(
the refinement of crystal structures containing disordered solvent
regions. Acta Crystallogr., Sect. A: Found. Crystallogr. 1990, 46, 194−
201.
I
DOI: 10.1021/acs.inorgchem.9b02993
Inorg. Chem. XXXX, XXX, XXX−XXX