Electronic effects of ligand substitution on spin crossover in a series of diiminoquinonoid-bridged FeII2 complexes

Article abstract of DOI:10.1021/ic5025586

A series of four isostructural Fe^II₂ complexes, [(TPyA)₂Fe₂(^XL)]²⁺ (TPyA = tris(2-pyridylmethyl)amine; ^XL^2-= doubly deprotonated form of 3,6-disubstituted-2,5-dianilin

Full text of DOI:10.1021/ic5025586

Article
pubs.acs.org/IC
Electronic Effects of Ligand Substitution on Spin Crossover in a
Series of Diiminoquinonoid-Bridged FeII Complexes
2
Jesse G. Park,† Ie-Rang Jeon,† and T. David Harris*
Department of Chemistry, Northwestern University, Evanston, Illinois 60208-3113, United States
S
* Supporting Information
ABSTRACT: A series of four isostructural FeII complexes,
2
X
[(TPyA)2Fe2(XL)]2+ (TPyA = tris(2-pyridylmethyl)amine; L2−
= doubly deprotonated form of 3,6-disubstituted-2,5-dianilino-
1,4-benzoquinone; X = H, Br, Cl, and F), were synthesized to
enable a systematic study of electronic effects on spin crossover
behavior. Comparison of X-ray diffraction data for these
complexes reveals the sole presence of high-spin FeII at 225 K
and mixtures of high-spin and low-spin FeII at 100 K, which is
indicative of incomplete spin crossover. In addition, crystal
packing diagrams show that these complexes are well-isolated
from one another in the solid state, owing primarily to the
presence of bulky tetra(aryl)borate counteranions, such that spin crossover is likely not significantly affected by intermolecular
interactions. Variable-temperature dc magnetic susceptibility data confirm the structural observations and reveal that 54(1),
56(1), 62(1), and 84(1)% of FeII centers remain high-spin even below 65 K. Moreover, fits to magnetic data provide crossover
temperatures of T1/2 = 160(1), 124(1), 121(1), and 110(1) K for X = H, Br, Cl, and F, respectively, along with enthalpies of ΔH
= 11.4(3), 8.5(3), 8.3(3), and 7.5(2) kJ/mol, respectively. These parameters decrease with increasing electronegativity of X and
X
thus increasing electron-withdrawing character of L2−, suggesting that the observed trends originate primarily from inductive
effects of X. Moreover, when plotted as a function of the Pauling electronegativity of X, both T1/2 and ΔH undergo a linear
decrease. Further analyses of the low-temperature magnetic data and variable-temperature Mossbauer spectroscopy suggest that
̈
the incomplete spin crossover behavior in [(TPyA)2Fe2(XL)]2+ is best described as a transition from purely [FeHS-FeHS] (HS =
high-spin) complexes at high temperature to a mixture of [FeHS-FeHS] and [FeHS-FeLS] (LS = low-spin) complexes at low
X
temperature, with the number of [FeHS-FeHS] species increasing with decreasing electron-withdrawing character of L2−.
molecular level by judicious modulation of ligand field
represents an important synthetic chemistry challenge. Toward
this end, an ideal system for studying the effects of ligand field
on spin crossover might involve a series of complexes that
feature a common local metal coordination environment and
systematic variation of the ligand substitution pattern.
Complicating such studies, spin crossover in a bulk sample is
also largely governed by solid-state packing effects in which
elastic interactions between molecular species significantly
influence the crossover.8,9 As such, changes in steric and
packing effects associated with changes in a substituent should
be minimized in order to extract a meaningful correlation
between electronic effects and spin crossover. To date, only a
few systematic studies of electronic effects on spin crossover
have been reported, most notably those involving substitution
of the four-position of an axially coordinated pyridyl ligand in
six-coordinate FeII complexes.10−12 Here, incorporation of
electron-withdrawing substituents led to increased π-back-
bonding character of the pyridyl ligands, thereby stabilizing the
low-spin state and increasing T1/2. In addition, a recent report
INTRODUCTION
■
Complexes that undergo a transition in electronic configuration
from a low-spin to a high-spin state as a function of an external
stimulus, such as temperature, light, or pressure, constitute a
class of molecules known as spin crossover complexes.1 These
species have garnered tremendous interest over several decades,
largely owing to their potential application as active
components in molecular memories, displays, and sensors.2−5
The spin crossover phenomenon is most commonly observed
in pseudo-octahedral FeII complexes that feature an Fe
coordination environment comprised by six nitrogen donor
atoms. This metal−ligand combination can give rise to a
situation in which the ligand field splitting energy is comparable
in magnitude to the spin pairing energy.6,7 In this case,
increasing temperature results in an entropy-driven conversion
from a low-spin S = 0 ground state to a high-spin S = 2 excited
state, which is accompanied by corresponding changes in
structural, magnetic, and spectroscopic properties.
The realization of applications associated with thermally
induced spin crossover requires the ability to adjust the
crossover temperature (T1/2), defined as the temperature at
which the low-spin state and the high-spin state are equally
populated. As such, predicting and controlling T1/2 at the
Received: October 24, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
commercial solvent purification system from Pure Process Technology
and stored over 3 or 4 Å molecular sieves. Et2O, THF, benzene, and n-
hexane were tested with a standard purple solution of sodium
benzophenone ketyl in THF to confirm effective oxygen and moisture
removal prior to use. Perdeuterated dimethyl sulfoxide (DMSO-d6)
was purchased from Cambridge Isotope Laboratories and stored over
4 Å molecular sieves prior to use. Diatomaceous earth (Celite 545)
was purchased from Fisher Scientific and dried by heating at 250 °C
under vacuum for 12 h. Sodium tetrafluoroborate was dried at 80 °C
under vacuum before use. The compounds tris(2-pyridylmethyl)amine
(TPyA)27 and sodium bis(trimethylsilyl)amide (Na[N(SiMe3)2])28
were synthesized as previously described. The compounds 3,6-
dihydro-, 3,6-difluoro-, 3,6-dichloro-, and 3,6-dibromo-2,5-dianilino-
1,4-benzoquinone (XLH2; X = H,29 F,30 Cl,31 and Br31), sodium
tetrakis[3,5-bis(trifluoromethyl)phenyl]borate (Na(BArF)),32 Tl-
(BArF),33 and Fe(BArF)2·6MeCN34 were synthesized through
modified literature procedures as described below. All other reagents
were obtained from commercial vendors and used without further
purification.
studied the variation of electronic and steric effects across a
series of four-coordinate pseudotetrahedral FeII complexes
supported by a strongly donating tris(carbene)borate ligand.13
Along with the extensive library of mononuclear FeII
complexes that exhibit thermally induced spin crossover, a
number of dinuclear FeII complexes have also been shown to
2
demonstrate this phenomenon.14−17 In a dinuclear complex,
spin crossover often occurs via an intermediate state comprising
an equimolar mixture of low-spin and high-spin FeII centers,
which can be made up of either a localized mixed-spin state
[FeHS-FeLS] (HS = high-spin, LS = low-spin) or a 1:1 mixture
of pure [FeHS-FeHS] and [FeLS-FeLS] states. Density functional
theory (DFT) calculations have suggested that the exact nature
of spin crossover in these complexes depends on the relative
enthalpies of these three possible compositions, which
determine whether the transition will follow one-step, two-
step, or incomplete spin crossover.18 In addition, the extent of
stabilization of the intermediate state is largely governed by the
ability of the Fe coordination sphere to distort from idealized
coordination geometries in order to undergo the spin crossover.
Here, highly rigid or already distorted coordination environ-
ments act to disfavor full conversion to low-spin FeII with
decreasing temperature, thus often giving rise to incomplete
spin crossover behavior.17,19
2,5-Dianilino-1,4-benzoquinone (HLH2). This compound was
synthesized through a modified literature procedure.29 1,4-Benzoqui-
none (1.08 g, 10.0 mmol) was added as a solid to methanol (35 mL)
to give a yellow mixture. Aniline (2.23 g, 24.0 mmol) was then added
to this suspension, and the resulting dark brown mixture was stirred at
room temperature for 12 h. The brown mixture was then filtered, and
the brown residue was washed with water (3 × 25 mL) and then dried
at 50 °C under vacuum for 12 h to afford 0.637 g (22%) of product as
1
Derivatives of 2,5-dihydroxy-1,4-benzoquinone have long
been employed as bridging ligands between two metal
centers.20,21 Indeed, scores of dinuclear molecular complexes
and extended solids containing both diamagnetic and para-
magnetic derivatives of 2,5-dihydroxy-1,4-benzoquinone as
bridging ligands have been reported and thoroughly inves-
a brown powder. H NMR (DMSO-d6, 500 MHz): δ 9.33 (s, 2H,
NH), 7.44 (dd, 2H, aryl-H), 7.38 (dd, 4H, aryl-H), 7.24 (dd, 2H, aryl-
H), 5.79 (s, 2H, quinone-H). FT-IR (ATR, cm−1): 3224 (m), 3053
(w), 1636 (w), 1563 (s), 1484 (s), 1441 (s), 1355 (m), 1286 (s), 1230
(m), 1186 (s), 1175 (s), 1079 (w), 1023 (w), 1023 (w), 1000 (w), 913
(w), 894 (w), 857 (w), 840 (m), 759 (m), 739 (s), 724 (s), 690 (s).
2,5-Dianilino-3,6-fluoro-1,4-benzoquinone (FLH2). This com-
pound was synthesized through a modified literature procedure.30
2,3,5,6-Tetrafluoro-1,4-benzoquinone (0.108 g, 1.00 mmol) was added
as a solid to a solution of sodium acetate (0.150 g, 1.83 mmol) in a
mixture of ethanol (10 mL), acetic acid (10 mL), and water (5 mL) to
give a purple mixture. Aniline (0.372 g, 4.00 mmol) was then added to
this suspension, and the resulting dark brown mixture was heated at
110 °C for 5 h. The dark brown mixture was then filtered while hot,
and the dark brown residue was washed with water (3 × 15 mL) and
then dried at 50 °C under vacuum for 12 h to afford 0.288 g (88%) of
tigated.22−26 Among these, FeII complexes bridged by 3,6-
2
dihydroxy- and 3,6-bis(tert-butyl)-2,5-dihydroxy-1,4-benzoqui-
none have been shown to exhibit thermally induced spin
crossover.26 However, relative differences in the crystal
structures of these complexes precluded the correlation of
ligand substitution effects on spin crossover.
The observation of spin crossover in these FeII2 complexes, in
conjunction with the dearth of studies that probe electronic
effects on spin crossover, prompted us to investigate how the
substitution pattern of benzoquinonoid-bridging ligands affects
1
product as a dark brown powder. H NMR (DMSO-d6, 500 MHz): δ
9.16 (s, 2H, NH), 7.34 (dd, 4H, aryl-H), 7.13−7.19 (m, 6H, aryl-H).
FTIR (ATR, cm−1): 3240 (m), 3055 (w), 1667 (w), 1594 (m), 1574
(s), 1489 (s), 1440 (s), 1354 (s), 1321 (s), 1305 (m), 1271 (m), 1227
(s), 1176 (m), 1090 (w), 1081 (w), 1064 (w), 1027 (w), 993 (m), 913
(w), 841 (w), 810 (w), 787 (w), 751 (s), 721 (s), 701 (s), 681 (m).
2,5-Dianilino-3,6-dichloro-1,4-benzoquinone (ClLH2). This
compound was synthesized through a modified literature procedure.31
2,3,5,6-Tetrachloro-1,4-benzoquinone (2.46 g, 10.0 mmol) was added
as a solid to a solution of sodium acetate (0.600 g, 7.32 mmol) in a
mixture of ethanol (50 mL), acetic acid (50 mL), and water (25 mL)
to give a yellow mixture. Aniline (3.80 g, 40.8 mmol) was then added
to this suspension, and the resulting dark brown mixture was heated at
110 °C for 5 h. The dark brown mixture was then filtered while hot,
and the dark brown residue was washed with water (3 × 25 mL) and
then dried at 50 °C under vacuum for 12 h to afford 1.93 g (82%) of
spin crossover in FeII complexes. Toward this end, we have
2
targeted the synthesis of FeII complexes containing 3,6-
2
disubstituted-2,5-dianilino-1,4-benzoquinone derivatives as
bridging ligands. Replacement of the hydroxy groups with
aniline groups provides steric bulk to minimize intermolecular
interactions. Herein, we report the synthesis and character-
ization of a series of [(TPyA)2Fe2(XL)]2+ complexes, where
XL2− = dianion of 3,6-disubstituted-2,5-dianilino-1,4-benzoqui-
none and X = H, Br, Cl, and F. The isostructural nature of these
complexes, along with large intermolecular separation in the
solid state afforded by aniline groups and tetra(aryl)borate
counterions, enables a systemic examination of the electronic
effects of ligand substitution on spin crossover using an
1
extensive array of structural, magnetic, and Mossbauer
̈
product as a dark brown powder. H NMR (DMSO-d6, 500 MHz): δ
9.64 (s, 2H, NH), 7.33 (dd, 4H, aryl-H), 7.14−7.19 (m, 6H, aryl-H).
FTIR (ATR, cm−1): 3232 (m), 3038 (w), 1650 (w), 1602 (w), 1562
(s), 1494 (s), 1480 (s), 1443 (s), 1323 (s), 1309 (s), 1247 (m), 1193
(m), 1167 (w), 1150 (w), 1077 (w), 1025 (m), 1002 (w) 909 (m),
886 (m), 831 (w), 780 (m), 752 (m), 745 (m), 713 (m), 691 (s).
2,5-Dianilino-3,6-dibromo-1,4-benzoquinone (BrLH2). This
compound was synthesized through a modified literature procedure.31
2,3,5,6-Tetrabromo-1,4-benzoquinone (2.12 g, 5.00 mmol) was added
as a solid to a solution of sodium acetate (0.600 g, 7.32 mmol) in a
mixture of ethanol (50 mL), acetic acid (50 mL), and water (25 mL)
spectroscopic analyses.
EXPERIMENTAL SECTION
■
General Considerations. Unless otherwise noted, all procedures
were performed under a dinitrogen atmosphere using a Vacuum
Atmospheres Nexus II glovebox. Glassware was oven-dried at 150 °C
for at least 4 h and allowed to cool in an evacuated antechamber prior
to use in the glovebox. Acetonitrile (MeCN), diethyl ether (Et2O),
tetrahydrofuran (THF), benzene, and n-hexane were dried using a
B
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
to give a yellow mixture. Aniline (1.86 g, 20.0 mmol) was then added
to this suspension, and the resulting dark brown mixture was heated at
110 °C for 5 h. The dark brown mixture was then filtered while hot,
and the dark brown residue was washed with water (3 × 25 mL) and
then dried at 50 °C under vacuum for 12 h to afford 2.15 g (96%) of
dark brown mixture. To this mixture was added dropwise with stirring
a solution of Na[N(SiMe3)2] (31.2 mg, 0.170 mmol) in THF (4 mL),
resulting in a dark brown solution. After 12 h of stirring, the resulting
dark green solution was filtered through diatomaceous earth. The
filtrate was concentrated under reduced pressure to a volume of 2 mL.
A portion of n-hexane (8 mL) was carefully layered onto the filtrate to
give a dark green microcrystalline solid after 1 day, which was
subsequently washed with cold EtOH (2 × 0.3 mL). The residue was
dissolved in THF (2 mL) and recrystallized through layering with n-
hexane (8 mL) to afford 172 mg (75%) of 1 as a dark green
microcrystalline solid. FTIR (ATR, cm−1): 2921 (w), 2859 (w), 1604
(w), 1575 (w), 1537 (m), 1485 (w), 1440 (w), 1352 (m), 1273 (s),
1243 (m), 1156 (s), 1116 (m), 1052 (m), 1019 (m), 862.71 (m), 828
(m), 780 (m), 743 (m), 715 (m), 699 (w), 681 (m), 668 (m). Anal.
Calcd for C118H74B2F48Fe2N10O3: C, 52.00; H, 2.74; N, 5.14%. Found:
C, 51.75; H, 2.61; N, 4.76%. Dark green needle-shaped crystals of the
related compound [(TPyA)2Fe2(HL)](BArF)2·4Et2O] (1′), suitable for
single-crystal X-ray diffraction, were obtained through diffusion of
Et2O vapor into a THF solution.
1
product as a dark brown powder. H NMR (DMSO-d6, 500 MHz): δ
9.68 (s, 2H, NH), 7.34 (dd, 4H, aryl-H), 7.17−7.20 (m, 6H, aryl-H).
FTIR (ATR, cm−1): 3229 (m), 3051 (w), 1646 (w), 1600 (w), 1559
(s), 1493 (s), 1475 (s), 1443 (s), 1318 (s), 1308 (m), 1244 (m), 1141
(m), 1164 (m), 1150 (m), 1075 (m), 1016 (m), 1001 (w), 908 (m),
873 (m), 833 (m), 771 (s), 751 (s), 730 (s), 711 (s), 690 (s).
Sodium Tetrakis[3,5-bis(trifluoromethyl)phenyl]borate (Na-
(BArF)). This compound was prepared according to a modified
literature procedure.32 Mg turnings (10.0 g, 413 mmol), NaBF4 (7.03
g, 64.0 mmol), and 1,2-dibromoethane (10.7 g, 57.0 mmol) were
added to a round-bottom flask containing Et2O (1.5 L). The mixture
was heated to reflux in order to initiate the reaction, and once the
mixture was at reflux, external heating was immediately ceased. The
white mixture was stirred for 3 h, and then a solution of 3,5-
bis(trifluoromethyl)bromobenzene (105 g, 360 mmol) in Et2O (500
mL) was added dropwise over 3 h; the resulting mixture was heated at
reflux for 1 h. The mixture was then allowed to cool to ambient
temperature and stirred overnight. A solution of Na2CO3 (160 g) in
water (2 L) was then added, and the resulting mixture was stirred for 1
h and then filtered to remove the remaining white insoluble material.
After separation of the Et2O layer, the remaining aqueous layer was
washed with Et2O (3 × 500 mL). All Et2O layers were then combined,
dried over Na2SO4, treated with decolorizing carbon (20.0 g), and
dried under vacuum at 50 °C for 4 h. Under a dinitrogen atmosphere,
the remaining oily residue was dissolved in benzene (200 mL) to give
a pale yellow solution, and trace amounts of water were removed using
a Dean−Stark apparatus. The resulting pale yellow solution was dried
under vacuum for 12 h to give a pale yellow powder. This powder was
washed with a 2:1 cold CH2Cl2/n-hexane mixture (4 × 10 mL) to
remove the yellow color and then was dried at 110 °C under vacuum
[(TPyA)2Fe2(BrL)](BArF)2·THF·n-Hexane (2). This compound was
synthesized analogously to 1 using TPyA (49.6 mg, 0.170 mmol),
Fe(BArF)2·6MeCN (345 mg, 0.170 mmol), BrLH2 (38.1 mg, 0.0850
mmol), and Na[N(SiMe3)2] (31.2 mg, 0.170 mmol). Layering of the
filtrate with n-hexane (8 mL) yielded a mixture of dark green and
white microcrystalline solid, which was washed with cold EtOH (2 ×
0.3 mL) to remove the white solid. The remaining dark green solid
was dissolved in THF (2 mL) and recrystallized through layering with
n-hexane (8 mL) to afford 194 mg (80%) of 2 as dark green needle-
shaped crystals suitable for single-crystal X-ray diffraction. FTIR (ATR,
cm−1): 2923 (w), 2858 (w), 1605 (w), 1511 (m), 1484 (m), 1440 (w),
1352 (s), 1272 (s), 1155 (s) 1113 (s), 1053 (m), 1020 (w), 886 (m),
874.71 (m), 837 (m), 762 (m), 743 (w), 715 (w), 695 (w), 681 (m),
668 (m). Anal. Calcd for C126H86B2Br2F48Fe2N10O4: C, 50.29; H, 2.88;
N, 4.65%. Found: C, 50.51; H, 2.46; N, 4.67%.
[(TPyA)2Fe2(ClL)](BArF)2·THF·n-Hexane (3). This compound was
synthesized analogously to 1 using TPyA (49.6 mg, 0.170 mmol),
Fe(BArF)2·6MeCN (345 mg, 0.170 mmol), ClLH2 (30.5 mg, 0.0850
mmol), and Na[N(SiMe3)2] (31.2 mg, 0.170 mmol). Layering of the
filtrate with n-hexane (8 mL) yielded a mixture of dark green and
white microcrystalline solid, which was washed with cold EtOH (2 ×
0.3 mL) to remove the white solid. The remaining dark green solid
was redissolved in THF (2 mL) and recrystallized through layering
with n-hexane (8 mL) to afford 166 mg (70%) of 3 as dark green
needle-shaped crystals suitable for single-crystal X-ray diffraction.
FTIR (ATR, cm−1): 2922 (w), 2858 (w), 1605 (w), 1512 (m), 1484
(w), 1440 (w), 1352 (m), 1272 (s), 1156 (s) 1114 (m), 1053 (m),
1020 (m), 996 (m), 886 (m), 837 (m), 784 (m), 761 (m), 744 (m),
7 1 4 ( w ) , 6 9 7 ( m ) , 6 8 1 ( m ) . A n a l . C a l c d f o r
C122H78B2Cl2F48Fe2N10O3: C, 51.45; H, 2.76; N, 4.92%. Found: C,
51.55; H, 2.84; N, 4.58%.
1
for 12 h to give 47.8 g (84%) of product as an off-white powder. H
NMR (DMSO-d6, 500 MHz): δ 7.66 (s, 4H), 7.61 (br, 8H). FTIR
(ATR, cm−1): 1629 (w), 1612 (w), 1356 (s), 1280 (s), 1115 (br), 945
(w), 933 (w), 887 (m), 837 (m), 743 (w), 709 (m), 680 (m), 669
(m).
Thallium Tetrakis[3,5-bis(trifluoromethyl)phenyl]borate (Tl-
(BArF)). This compound was prepared according to a modified
literature procedure.33 A solution of Na(BArF) (2.57 g, 2.89 mmol) in
EtOH (9 mL) was added to a solution of TlNO3 (0.771 g, 2.89 mmol)
in water (15 mL), and the resulting pale yellow solution was stirred at
ambient temperature for 30 min. The solution was then concentrated
under reduced pressure to a volume of 12 mL, resulting in the
precipitation of an off-white solid. The precipitate was collected by
filtration, washed with cold water (5 × 5 mL), and dried under vacuum
at 110 °C for 12 h to give 2.31 g (75%) of product as a white powder.
FTIR (ATR, cm−1): 1629 (w), 1611 (w), 1356 (s), 1278 (s), 1118
(br), 943 (w), 933 (w), 886 (m), 837 (m), 743 (w), 709 (m), 680
(m), 669 (m).
[(TPyA)2Fe2(FL)](BArF)2·THF·n-Hexane (4). This compound was
synthesized analogously to 1 using TPyA (49.6 mg, 0.170 mmol),
Fe(BArF)2·6MeCN (345 mg, 0.170 mmol), FLH2 (27.7 mg, 0.0850
mmol), and Na[N(SiMe3)2] (31.2 mg, 0.170 mmol). Layering of the
filtrate with n-hexane (8 mL) yielded a mixture of dark green and
white microcrystalline solid, which was washed with cold EtOH (2 ×
0.3 mL) to remove the white solid. The remaining dark green solid
was dissolved in THF (2 mL) and recrystallized through layering with
n-hexane (8 mL) to afford 179 mg (77%) of 4 as a dark green
microcrystalline solid. FTIR (ATR, cm−1): 2921 (w), 2859 (w), 1606
(w), 1542 (m), 1484 (w), 1441 (w), 1352 (m), 1330 (m), 1272 (s),
1224 (s), 1156 (s), 1115 (s), 1052 (m), 1008 (m), 904 (w), 886 (m),
837 (m), 761 (m), 743 (w), 716 (m), 700 (m), 681 (m), 668 (m).
Anal. Calcd for C122H78B2F50Fe2N10O3: C, 52.05; H, 2.82; N, 4.79%.
Found: C, 51.92; H, 2.57; N, 5.11%. Dark green needle-shaped crystals
of the related compound [(TPyA)2Fe2(FL)](BArF)2·4Et2O (4′),
suitable for single-crystal X-ray diffraction, were obtained through
diffusion of Et2O vapor into a THF solution.
Fe(BArF)2·6MeCN. This compound was prepared according to a
modified literature procedure.34 A solution of FeBr2 (63.3 mg, 0.293
mmol) in MeCN (6 mL) was added to a solution of Tl(BArF) (626
mg, 0.586 mmol) of MeCN (10 mL). The resulting yellow mixture
was stirred at ambient temperature for 5 h and then filtered to remove
a white insoluble material. The pale yellow filtrate was concentrated
under vacuum to ∼1 mL and stored at −35 °C overnight to give white
needle-shaped crystals. The crystals were immediately collected by
filtration and washed with cold MeCN to give 145 mg (54%) of
product. FTIR (ATR, cm−1): 1611 (w), 1354 (s), 1276 (s), 1109 (br),
937 (w), 888 (m), 839 (m), 746 (w), 712 (m), 682 (m), 669 (m).
[(TPyA)2Fe2(HL)](BArF)2·THF·n-Hexane (1). TPyA (49.6 mg, 0.170
mmol) was dissolved in THF (2 mL), and this solution was added
dropwise with stirring to a solution of Fe(BArF)2·6MeCN (345 mg,
0.170 mmol) in THF (2 mL). The resulting red solution was stirred
for 10 min at ambient temperature and then was treated with a
suspension of HLH2 (24.7 mg, 0.0850 mmol) in THF (4 mL) to give a
X-ray Structure Determination. Single crystals of 1′, 2, 3, and 4′
were coated with deoxygenated Paratone-N oil, mounted on a
C
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Crystallographic Data for 1′, 2, 3, and 4′
1′
2
3
4′
formula
fw (g/mol)
T (K)
C
134H112B2F48Fe2N10O6
C128H92B2Br2F48Fe2N10O3
3023.25
C124H84B2Cl2F48Fe2N10O3
2934.33
C134H110B2F50Fe2N10O6
3039.63
3003.65
225
100
100
225
100
225
100
225
space group
Z
P1
P1
P1
P1
P1
P1
P1
P1
̅
̅
̅
̅
̅
̅
̅
̅
1
1
1
1
1
1
1
1
a (Å)
12.1891(9)
16.250(1)
17.898 (2)
88.042(5)
74.917(3)
79.386(3)
3364.1(5)
1.483
12.3247(6)
16.9752(6)
17.7140(9)
86.386(2)
75.656(2)
76.244(1)
3487.5(3)
1.430
12.3771(7)
16.844(1)
17.592(1)
85.473(2)
76.051(2)
78.445(2)
3485.5(3)
1.395
12.4619(5)
17.1585(8)
17.9518(8)
84.820(2)
76.042(2)
77.842(2)
3638.5(3)
1.380
12.2621(8)
16.824(1)
17.610(1)
85.809(3)
76.479(3)
77.825(3)
3451.7(4)
1.412
12.3707(5)
17.2025(7)
17.9700(8)
84.892(2)
76.314(1)
77.330(2)
3622.3(3)
1.345
12.1907(7)
16.6435(9)
17.3578(9)
87.671(2)
77.622(2)
77.463(2)
3357.9(3)
1.503
12.2877(5)
16.8312(7)
17.7318(8)
87.114(2)
76.717(1)
77.569(2)
3485.4(3)
1.448
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
dcalc (g/cm3)
a
R1 (I > 2σ(I))
8.79
7.41
5.65
6.63
9.95
7.14
8.57
6.49
b
wR2 (all)
26.62
22.82
17.35
20.79
28.50
22.65
24.27
20.56
GoF
1.078
1.101
1.083
1.079
1.029
1.140
1.084
1.105
a
b
2
2
2
R1 = ∑∥F0| − |FC∥/∑|F0|. wR2 = [∑w(F0 − FC )2/∑w(F0 )2]1/2
.
Scheme 1. Synthesis of Compounds [(TPyA)2Fe2(XL)](BArF)2 (X = H, Br, Cl, and F)
MicroMounts rod, and frozen under a stream of N2 during data
collection. Crystallographic data were collected at 225 K using a
Bruker Kappa Apex II diffractometer equipped with an APEX-II
detector, a Cu Kα microfocus source, and either Quazar Optics (for 1′
and 2) or MX Optics (for 3 and 4′). Crystallographic data were also
collected at 100 K using a Bruker Kappa Apex II diffractometer
equipped with an APEX-II detector, a Cu Kα microfocus source, and
MX Optics for 1′, 2, and 4′, and a Bruker Kappa Apex II diffractometer
equipped with an APEX-II detector and a Mo Kα sealed-tube source
for 3. Raw data were integrated and corrected for Lorentz and
polarization effects with SAINT v8.27B.35 Absorption corrections were
applied using SADABS.36 Space group assignments were determined
by examination of systematic absences, E-statistics, and successive
refinement of the structure. Structures were solved using direct
methods in SHELXS and further refined with SHELXL-201336
operated with the OLEX2 interface.37 All hydrogen atoms were
placed at calculated positions using suitable riding models and refined
using isotropic displacement parameters derived from their parent
atoms. Thermal parameters were refined anisotropically for all non-
hydrogen atoms. Crystallographic data for 1′, 2, 3, and 4′ at 100 and
225 K are given in Table 1. Significant disorder of the −CF3 groups on
the [BArF]− anions was modeled with partial occupancies; however,
some disorder that did not improve the final refinement of the solution
was placed at one position with full occupancy. The interstitial THF
molecules in 2 (100 and 225 K) and 3 (100 K) and the Et2O
molecules in 1′ (100 K) and 4′ (100 K) required distance and
displacement parameter restraints because of disorder, and the
associated hydrogen atoms were not placed but were included in the
final chemical formulas. All solvent molecules in 1′, 3, and 4′ at 225 K
and the n-hexane molecules in 2 (100 and 225 K) and 3 (100 K) were
severely disordered and could not be modeled properly. These species
were therefore treated as a diffuse contribution to the overall scattering
without specific atom positions using the solvent masking procedure
implemented in OLEX2. These molecules were included in the final
molecular formula.
Magnetic Measurements. All samples were prepared and
manipulated with the rigorous exclusion of dioxygen, either under a
dinitrogen atmosphere or under vacuum. Magnetic measurements of
1−4 were performed on polycrystalline samples retained in their
original THF/n-hexane mother liquor loaded into quartz tubes. The
quartz tube was fixed to a sealable hose-adapter, degassed through
three successive freeze−pump−thaw cycles, evacuated on a Schlenk
line while frozen in liquid N2, and then flame-sealed. All data were
collected using a Quantum Design MPMS-XL SQUID magnetometer
from 1.8 to 300 K at applied dc fields ranging from 0 to +7 T. Variable-
temperature data were collected at a scan rate of ∼1 K/min at 1 T and
of ∼2 K/min at 0.1 T. dc susceptibility data were corrected for
diamagnetic contributions from the sample holder and for the core
diamagnetism of each sample (estimated using Pascal’s constants38).
M(H) curves constructed from data collected from 0 to 4 T at 100 K
confirmed the absence of ferromagnetic impurities in all samples.
̈
Mossbauer Measurements. Zero-field iron-57 Mossbauer
̈
spectra were obtained between 80 and 250 K with a constant
acceleration spectrometer and a cobalt-57 rhodium source. Prior to
measurements, the spectrometer was calibrated at 295 K with α-iron
foil. Samples were prepared in a dinitrogen-filled glovebox with crystals
of 1−4 that were quickly filtered and inserted into the cryostat prior to
the measurement. Each sample contained approximately 80 mg/cm2 of
compound. All spectra were analyzed using the WMOSS Mossbauer
̈
Spectral Analysis Software (www.wmoss.org). Fits to a minor doublet
in the spectra of 1−4 at 250 K gave an isomer shift of δ = 0.23(1)−
0.25(1) mm/s and a quadrupole splitting of ΔEQ = 0.30(2)−0.39(2)
mm/s, consistent with a small amount of high-spin FeIII-containing
impurity in the sample. To fit the spectra of 4 at 100 and 80 K, the
D
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
upper limit of ΔEQ for the quadrupole doublet assigned to the high-
spin FeII site in the [FeHS-FeLS] species was fixed to 2.93 and 3.02
mm/s, respectively, due to the otherwise unstable fit.
Other Physical Measurements. Elemental analyses of all
compounds were performed by the Midwest Microlab (Indianapolis,
IN). Infrared spectra were recorded on a Bruker Alpha Fourier
transform infrared (FTIR) spectrometer equipped with an attenuated
1
total reflectance accessory. H NMR spectra were collected at 500
MHz on a Varian Inova 500 system at 298 K.
RESULTS AND DISCUSSION
■
Syntheses and Structures. The bridging ligands XLH2
were synthesized by reaction of aniline with 1,4-benzoquinone
(X = H29) or the corresponding 2,3,5,6-tetrahalo-1,4-
benzoquinone (X = F,30 Cl,31 Br31) following modified
literature procedures. The deprotonated benzoquinone deriv-
ative (XL2−) was generated in solution by reaction of XLH2 with
2 equiv of Na[N(SiMe3)2] in THF (see Scheme 1). Metalation
of this species with FeII was then effected by treatment with a
THF solution containing 2 equiv each of tris(2-pyridylmethyl)-
amine (TPyA) and Fe(BArF)2 ([BArF]− = tetrakis[3,5-bis-
(trifluoromethyl)phenyl]borate anion) to give a dark green
solution. Subsequent careful layering of this solution with n-
hexane resulted in the cocrystallization of dark green needle-
shaped crystals and colorless crystals of NaBArF. The NaBArF
side product was then removed by successive washing with
minimal amounts of EtOH that had been cooled to −78 °C,
giving rise to pure dark green crystalline material. Finally, this
material was recrystallized via n-hexane layering on THF
solutions to yield dark green needle-shaped crystals of
[(TPyA)2Fe2(XL)](BArF)2·THF·n-hexane (X = H (1), Br (2),
Cl (3), and F (4)). Note that metal complexes featuring
derivatives of 2,5-dianilino-1,4-benzoquinone39−42 as bridging
ligands are far less common than those of the isoelectronic 2,5-
dihydroxy-1,4-benzoquinone derivatives.20−26 Indeed, to our
knowledge, 1−4 represent the first examples of Fe complexes of
a 2,5-dianilino-1,4-benzoquinone derivative.
Figure 1. Crystal structure of [(TPyA)2Fe2(ClL)]2+ at 100 (bottom)
and 225 K (top), as observed in 3, with ellipsoids shown at the 40 and
20% probability level, respectively. Colored ellipsoids represent Fe
(orange), Cl (green), O (red), N (blue), and C (gray) atoms. H atoms
are omitted for clarity.
At 225 K, the average Fe−NTPyA, Fe−NL, and Fe−O
distances across the series fall in the ranges of 2.185(3)−
2.189(3), 2.148(3)−2.171(2), and 1.997(2)−2.002(2) Å,
respectively, which is consistent with the presence of two
high-spin S = 2 FeII centers (Table 2). Within the bridging
ligand, the O1−C3 and N1−C2 bond distances of 1.288(4)−
1.359(4) and 1.297(4)−1.306(4) Å, respectively, are longer
than a CO double bond and shorter than a C−N single
bond. Moreover, the C1−C2 bond lengths of 1.413(5)−
1.425(4) Å are longer than the C3−C1A bond lengths of
1.361(4)−1.368(5) Å. Finally, the C2−C3 bond distances are
in the range of 1.505(5)−1.516(4) Å across the series, which is
typical for a C−C single bond. These collective distances
suggest that XL2− is comprised of localized 6-π-electron
fragments connected by two C−C single bonds. Furthermore,
these distances are consistent with each FeII center being
ligated by an oxygen atom of a deprotonated hydroxy group
and a nitrogen atom of a neutral imine group, such that XL2− is
best described as a bis(phenolate)diimine derivative. Likely, this
composition results from better stabilization of negative charges
afforded by the more electronegative oxygen atoms. Indeed, a
similar observation was previously reported for CoII and RuII
Single crystals suitable for X-ray diffraction analysis were
obtained for 2 and 3 through the layering method described
above. In contrast, this method gave crystals of 1 and 4 that
were not of sufficient quality to enable collection of a complete
data set. Nevertheless, unit cell determinations confirmed that
these compounds are indeed isostructural to 2 and 3 (Table S1,
Supporting Information). Alternatively, diffusion of diethyl
ether vapor into THF solutions containing 1 and 4 gave dark
green needle-shaped crystals of [(TPyA)2Fe2(XL)](BArF)2·
4Et2O (X = H (1′) and F (4′)) suitable for single-crystal X-
ray diffraction analysis. Note that the relatively high solubility of
2 and 3 in Et2O prohibited crystallization of the analogous
compounds 2′ and 3′.
Single-crystal X-ray diffraction analysis for 1′, 2, 3, and 4′ was
carried out at both 225 and 100 K to probe for structural
changes associated with spin crossover. At both temperatures,
all compounds crystallize in the triclinic space group P1 and are
̅
2
2
isostructural to one another (Table 1). The general structure of
complexes bridged by 2,5-diarylamino-1,4-benzoquinone de-
[(TPyA)2Fe2(XL)]2+ comprises a dinuclear FeII complex in
rivatives.39−42
2
which two crystallographically equivalent [(TPyA)Fe]2+ units
are connected by a dianionic XL2− bridging ligand with a
crystallographic site of inversion located at the center of the
bridging ligand (Figure 1 and Figures S1−S3 of the Supporting
Information). Each FeII center resides in a distorted octahedral
coordination environment made up of four nitrogen donor
atoms from the TPyA capping ligand and cis-disposed nitrogen
and oxygen donor atoms from the bridging ligand.
At 100 K, crystals of 1′, 2, 3, and 4′ maintain their space
group of P1, albeit with slightly decreased unit cell dimensions
̅
relative to the 225 K data (Table 1). Close inspection of the
diffraction images for the crystal of 1′ revealed a splitting of
Bragg peaks along nonspecific crystallographic directions, and
extraction of unit cell parameters for these new peaks was not
possible (Figure S4, Supporting Information). The splitting at
low temperature did not correspond to physical cracking of the
E
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 2. Selected Interatomic Distances (Å) and Angles (deg) for 1′, 2, 3, and 4′
1′
2
3
4′
T (K)
225
100
225
100
225
100
225
Fe−N2
Fe−N3
Fe−N4
Fe−N5
Fe−NTPyA,avg
Fe−N1
Fe−O1
2.141(3)
2.162(3)
2.161(3)
2.290(3)
2.188(3)
2.148(3)
1.999(2)
7.9994(8)
12.325(1)
1.421(5)
1.505(5)
1.361(4)
1.306(4)
1.359(4)
122.9(3)
2.168(3)
2.040(3)
2.056(4)
2.168(3)
2.106(3)
2.108(3)
1.964(2)
7.9680(8)
12.2595(9)
1.419(4)
1.502(4)
1.374(4)
1.313(4)
1.292(4)
98.7(3)
2.169(3)
2.145(4)
2.141(3)
2.283(3)
2.185(3)
2.158(3)
1.997(2)
8.0561(8)
12.3491(8)
1.423(5)
1.510(4)
1.368(5)
1.302(4)
1.288(4)
123.5(3)
2.077(3)
2.053(4)
2.060(3)
2.179(4)
2.093(3)
2.092(3)
1.972(2)
7.9688(8)
12.251(1)
1.413(5)
1.504(4)
1.365(5)
1.324(4)
1.290(4)
102.4(3)
2.167(3)
2.148(3)
2.150(4)
2.276(3)
2.185(3)
2.163(3)
1.997(2)
8.0538(8)
2.152(4)
2.149(4)
2.134(3)
2.261(4)
2.174(4)
2.149(3)
1.997(3)
8.0080(9)
12.191(1)
1.417(6)
1.506(6)
1.362(6)
1.300(5)
1.294(5)
116.9(3)
2.165(3)
2.144(3)
2.160(3)
2.288(3)
2.189(3)
2.171(2)
2.002(2)
8.0360(7)
12.2877(9)
1.415(4)
1.508(4)
1.362(4)
1.303(4)
1.292(3)
b
Fe−Feintra
Fe−Feinter
c
12.3331(8)
1.425(4)
1.516(4)
1.367(4)
1.297(4)
1.292(4)
C1−C2
C2−C3
C3−C1A
N1−C2
O1−C3
d
Σ
123.8(3)
123.8(3)
c
a
b
See Figure 1 and Figures S5−S7 in the Supporting Information for the atomic numbering scheme. Intramolecular Fe···Fe distance. Shortest
intermolecular Fe···Fe distance. Octahedral distortion parameter (Σ) = sum of the absolute deviations from 90° of the 12 cis angles.
d
Figure 2. Portion of the crystal structure of 3 at 225 K, as viewed normal to the cb (left) and ab (right) planes, highlighting the large intermolecular
separation of dicationic Fe2 complexes afforded by bulky [BArF]− anions. Colored spheres represent Fe (orange), Cl (green), F (turquoise), O (red),
N (blue), C (gray), and B (magenta) atoms. For clarity, H atoms and solvent molecules are omitted, and [BArF]− anions are shown as translucent.
Black dotted lines denote the shortest intermolecular Fe···Fe distances.
crystal; it possibly stemmed from long-range structural disorder
of the low-spin state.43 A structure solution obtained from the
averaged peak assignment confirmed the connectivity of a
cationic complex; however, determination of interatomic
distances and angles was not possible. As such, the structure
of 1′ at 100 K will not be considered in future discussion.
Comparison of the bond distances at 100 and 225 K reveals
several key differences. First, at 100 K, the average Fe−NTPyA
distances were found to decrease substantially to 2.106(3),
2.093(3), and 2.174(4) Å for 2−4′, respectively. In addition,
the Fe−NL and Fe−O bonds also exhibited considerably
shorter distances of 2.108(3) and 1.964(2), 2.092(3) and
1.972(2), and 2.149(3) and 1.997(3) Å for 2−4′, respectively
(Table 2). The decreased bond distances in 2 and 3, while too
short to correspond to high-spin FeII, are nevertheless longer
than those expected for two low-spin FeII centers per dinuclear
complex.39 This observation suggests the presence of a mixture
of high- and low-spin FeII centers at 100 K, possibly resulting
from incomplete spin crossover. Along these lines, the average
Fe−NTPyA, Fe−NL, and Fe−O bond distances in 2 and 3 are
quite similar, both intermediate between those expected for
low-spin and high-spin states, suggesting the presence of a near
equimolar population of low-spin and high-spin FeII centers. In
contrast, the distances observed in 4′ fall much closer to the
high temperature values, suggesting a majority of high-spin FeII
centers in 4′ at 100 K. This comparative observation is not
surprising given the electron-withdrawing nature of the more
electronegative fluorine substituent, which apparently serves to
F
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
stabilize the high-spin state. Although the bond distances in 2−
4′ suggest a mixture of low-spin and high-spin FeII centers in
the crystal at 100 K, the presence of only one [(TPyA)Fe]2+
moiety in the asymmetric unit complicates further differ-
entiation between the presence of mixed-spin-state dinuclear
[FeHS-FeLS] complexes and a mixture of single-spin-state [FeHS-
FeHS] and [FeLS-FeLS] complexes (further discussion below).
Compounds 1′, 2, 3, and 4′ feature intramolecular Fe···Fe
distances in the range of 7.999(1)−8.056(1) Å at 225 K and
7.968(1)−8.008(1) Å at 100 K. In the overall structures, the 2+
charge of each Fe2 is compensated for by two noncoordinating
[BArF]− counteranions (Figure 2 and Figures S5−S7 in the
Supporting Information). These bulky anions interact with the
Fe2 complexes through a number of hydrogen-bonding
interactions, evident from contacts of XL2− phenyl- and
TPyA-based carbon atoms with −CF3 groups of [BArF]−,
with the shortest intermolecular C···F distances in the range of
2.91(1)−3.14(1) Å across the series at 225 K. The presence of
such bulky anions engenders a large separation between Fe2
complexes in the solid state, with the shortest intermolecular
Fe···Fe distances across the series of 12.2877(5)−12.349(1) Å
at 225 K and 12.191(1)−12.259(3) Å at 100 K. Importantly,
this large intermolecular separation should enable the
evaluation of electronic effects on spin crossover with minimal
complications from intermolecular cooperative interactions that
are likely to be similar across the series.
Compounds 1′, 2, 3, and 4′ contain rare examples of spin
crossover FeII complexes that feature an [N5O] local
coordination environment.5b,45 At each Fe center, the presence
of an Fe−O distance that is significantly shorter than the Fe−N
distances leads to a severely distorted coordination environ-
ment. This distortion from an ideal octahedron can be
quantified through the parameter Σ, which is defined as the
sum of the absolute deviations of each of the 12 cis-disposed
L−Fe−L angles from 90°.46 Analysis of the Fe centers in 1′, 2,
3, and 4′ at 225 K gives values of Σ = 122.9(3), 123.5(3),
123.8(3), and 123.8(3), respectively. As discussed above, the
pathway of spin crossover in dinuclear complexes has been
proposed to be highly dependent on the degree of distortion at
the FeII center.17,19 Indeed, a recent report summarized the
correlation of spin crossover and values of Σ for a number of
known dinuclear FeII complexes.17 This study demonstrated a
general trend that complexes with Σ < 40 invariably undergo
complete spin crossover from high-spin to low-spin states
without the presence of any intermediate state. In contrast,
those complexes with Σ > 120 were found to exhibit
incomplete spin crossover involving an intermediate state of
mixed-spin [FeHS-FeLS] composition. Accordingly, the high
values of Σ in 1′, 2, 3, and 4′ are consistent with the hypothesis
that these complexes undergo incomplete spin crossover
involving an intermediate state.
Figure 3. Variable-temperature dc magnetic susceptibility data for
solid samples of 1−4 collected under an applied field of 1 T. Circles
represent experimental data, whereas black lines correspond to fits
based on an ideal solution model as described in the text.
of χMT near 50 K. Although the S-shaped χMT curve is a
signature of spin crossover, the non-zero minimum values of
χMT at low temperature indicate that a considerable number of
FeII centers remain trapped in the high-spin state. The overall
shape of the data and the minimum values of χMT suggest an
incomplete spin crossover, and this conversion may be
kinetically slower or highly impeded with increasing electro-
negativity of X on XL2−. Indeed, similar partial trapping of high-
spin Fe centers was previously observed for a number of
mononuclear47 and some dinuclear complexes.48 The temper-
ature at which a discrepancy from the high-temperature plateau
appears decreases across the series from 1 to 4 and therefore
demonstrates larger stabilization of the high-spin state by a
more electronegative substituent on the bridging ligand. Finally,
note that these observations are consistent with the bond
distances of Fe ascertained from single-crystal X-ray diffraction
analysis.
To quantify the spin crossover phenomena in 1−4, the data
were fit to the ideal solution model49,50 to give crossover
temperatures (T1/2) of 160(1), 124(1), 121(1), and 110(1) K
and percentages of high-spin states of FeII centers at low
temperature of 54(1), 56(1), 62(1), and 84(1) for 1, 2, 3, and
4, respectively (see Table 3). Importantly, the observed trend
Table 3. Summary of Parameters Obtained from Fits to
a
Magnetic Data
ΔH (kJ/mol)
ΔS (J/mol·K)
T1/2 (K)
FeHS,trapped (%)
1
2
3
4
11.4(3)
8.5(3)
8.3(3)
7.5(2)
71(2)
69(3)
69(2)
68(2)
160(1)
124(1)
121(1)
110(1)
54(1)
56(1)
62(1)
84(1)
a
Fits obtained using an ideal solution model as described in the text.
Magnetic Properties. To probe and compare the presence
of spin crossover in compounds 1−4, variable-temperature dc
magnetic susceptibility data were collected in the temperature
range of 1.8−300 K on crystalline samples retained in their
original THF/n-hexane mother liquor to prevent partial
desolvation. The resulting plots of χMT vs T are shown in
Figure 3 (see also Figure S8 in the Supporting Information). At
300 K, compounds 1−4 exhibit values ranging from χMT = 6.74
(for 4) to 6.83 (for 1) cm3 mol−1 K, corresponding to two non-
interacting high-spin S = 2 FeII centers with g = 2.12−2.13. As
the temperature is decreased, the data initially remain relatively
constant before undergoing a gradual decline to minima values
X
scales with electronegativity of X on L2−, with increasingly
electronegative substituents providing better stabilization for
high-spin FeII centers. Indeed, the presence of these relatively
remote substituents remarkably decreases the crossover
temperature by 50 K, or more than 31%, in moving from X
= H to X = F.
In addition to T1/2, fits to the data also provide estimates of
thermodynamic parameters for 1−4 of ΔH = 11.4(3), 8.5(3),
8.3(3), and 7.5(2) kJ/mol and ΔS = 71(2), 68(3), 69(2), and
68(2) J/mol·K, respectively; these values are similar in
G
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
magnitude to others previously reported for spin crossover FeII
complexes (Table 3).51 Across the series, ΔH decreases by
34%, whereas ΔS decreases by only 4%. This comparison
suggests that the observed differences in T1/2 can be attributed
primarily to changes in enthalpy rather than entropy. This trend
is indicative of a direct correlation between T1/2 of the spin
crossover and differences in the ligand field at FeII stemming
Plots of both T1/2 vs χP and ΔH vs χP follow nearly linear
trends, with linear fits giving coefficients of determination of R2
= 0.93 and 0.92, respectively. This observation suggests that the
ligand field at Fe in this series of complexes is determined
primarily by inductive effects, propagated through bonds and/
or through space, of the different substituents. Note that this
trend is inversely related to that previously observed for six-
coordinate spin crossover FeII complexes featuring substituted
pyridyl ligands, which behave primarily as π acceptors.10−12
Finally, as temperature is further decreased, the χMT data
eventually undergo an increase and then a precipitous decrease,
with maximum values of χMT = 3.53, 4.29, 5.07, and 7.05 cm3 K
mol−1 at 10 K for 1−4, respectively. This increase in the χMT
data with decreasing temperature is almost certainly the result
of intramolecular ferromagnetic exchange coupling between
two high-spin FeII centers. Moreover, the maximum value of
χMT at 10 K increases with decreasing T1/2 across the series
from 1 to 4. This trend perhaps lends insight into the nature of
the incomplete spin crossover, in particular whether the
intermediate state primarily comprises a mixed-spin [FeHS-
FeLS] composition or a mixture of [FeHS-FeHS] and [FeLS-FeLS]
species. Indeed, the data for 1 show almost no increase below
50 K. This observation strongly suggests that 1 features nearly
exclusively [FeHS-FeLS] complexes below 50 K, as the
alternative would most likely lead to a significant upturn in
the data. Moreover, if compounds 2−4 behave similarly, then
they should be comprised solely of [FeHS-FeLS] and [FeHS-
FeHS] complexes.
X
from substituent variation in L2−. Overall, the difference in
ΔH across the series is 3.9 kJ/mol, or 330 cm−1, which can be
considered as the relative decrease in the energy gap between S
= 0 and S = 2 spin states in moving from 1 to 4.
The correlation between thermodynamics of spin crossover
X
and electronic character of L2− can be further examined by
plotting T1/2 and ΔH as a function of the Pauling electro-
negativity parameter (χP) of each X substituent (Figure 4).52
To test this hypothesis, the number of high-spin FeII centers
corresponding to [FeHS-FeHS] complexes, assuming the absence
of any [FeLS-FeLS] complexes, can be calculated from the total
number of high-spin centers at low temperature as extracted
from fits to the magnetic data (Table 3). This treatment reveals
that 7, 11, 23, and 67% of all high-spin Fe centers correspond to
molecules of [FeHS-FeHS] composition in 1−4, respectively.
Assuming an S = 4 ground state with g = 2.00 for each [FeHS-
Figure 4. Plot of T1/2 (red circles) and ΔH (blue circles) as a function
of Pauling electronegativity (χP) for 1−4. Solid lines correspond to
linear fits to the data with equations of the best linear fit providing R2
= 0.93 and 0.92, respectively.
Figure 5. Zero-field Mossbauer spectra for crystalline samples of 1−4 (left to right, respectively) at selected temperatures. Black crosses represent
̈
experimental data, and gray solid lines correspond to overall fits. Colored solid lines correspond to individual components of the overall fits assigned
to [FeHS-FeHS] (red), the high-spin center in [FeHS-FeLS] (green), and the low-spin center in [FeHS-FeLS] (blue). The dashed purple line corresponds
to a fit assigned to a small amount of FeIII-containing impurity. Each vertical scale bar represents an absorption of 1%.
H
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Table 4. Summary of Parameters Obtained from Fits to Mossbauer Spectra
Article
̈
high-spin site in [FeHS-FeHS
]
high-spin site in [FeHS-FeLS]
low-spin site in [FeHS-FeLS]
T (K)
δ (mm/s)
ΔEQ (mm/s)
area (%)
δ (mm/s)
ΔEQ (mm/s)
area (%)
δ (mm/s)
ΔEQ (mm/s)
area (%)
1
2
3
4
80
150
250
80
0.99(1)
0.97(1)
3.06(1)
49(2)
0.46(1)
0.44(1)
0.45(1)
0.45(1)
51(1)
34(1)
0.98(1)
0.91(1)
3.10(7)
2.74(2)
21(20)
84(4)
2.89(13)
44(27)
0.98(1)
0.98(1)
3.18(1)
51(1)
0.46(1)
0.44(1)
0.59(1)
0.61(2)
49(1)
33(2)
120
250
80
0.96(1)
0.91(1)
0.98(1)
0.97(1)
0.91(1)
0.98(1)
0.98(1)
0.91(1)
3.24(5)
2.87(2)
3.27(3)
3.18(3)
2.88(2)
3.23(3)
3.18(1)
2.83(2)
34(20)
79(6)
31(10)
63(10)
87(1)
36(7)
63(3)
85(3)
3.01(10)
33(19)
0.98(1)
0.96(1)
3.02(4)
2.90(4)
35(11)
19(10)
0.45(1)
0.45(1)
0.60(1)
0.62(2)
33(1)
19(2)
130
250
80
0.99(1)
0.98(1)
3.02(2)
2.93(2)
33(8)
18(3)
0.46(1)
0.45(1)
0.55(1)
0.54(2)
31(1)
19(1)
100
250
FeHS] complex arising from ferromagnetic coupling between
two S = 2 centers, these percentages correspond to expected
values of χMT = 3.49, 3.77, 4.61, and 7.69 cm3 K mol−1 for 1−4,
respectively, which are close to the observed maximum values
of χMT at low temperature. Although rudimentary, this analysis
provides strong evidence that the incomplete spin crossover in
1−4 is best described as a conversion from [FeHS-FeHS]
complexes at high temperature to a mixture of [FeHS-FeHS] and
[FeHS-FeLS] complexes at low temperature, with the number of
trapped [FeHS-FeHS] species increasing with increasing
spectrum, we assign this second site to an [FeHS-FeLS] complex.
This assignment is supported by previous reports wherein the
two high-spin sites in a mixture of [FeHS-FeHS] and [FeHS-FeLS]
complexes could be resolved by zero-field Mossbauer spectros-
̈
copy.55 Here, the site with a smaller ΔEQ was shown to
correspond to the [FeHS-FeLS] complex, stemming from the
distortion of the high-spin FeII coordination sphere imposed by
spin crossover of the second FeII center.55a,56 Finally, the third
doublet, which accounts for 34(1), 33(2), 19(2), and 19(1)% of
the spectra for 1−4, respectively, exhibits parameters of δ =
0.44(1)−0.45(1) mm/s and ΔEQ = 0.45(1)−0.62(2) mm/s
(blue lines in Figure 5). These isomer shifts and quadrupole
splittings are consistent with low-spin FeII, and as such we
assign this site to the low-spin center of an [FeHS-FeLS]
species.56
X
electron-withdrawing character of L2−.
̈
Mossbauer Spectroscopy. To further probe the spin
crossover in 1−4, zero-field Mossbauer spectra were collected
̈
for microcrystalline samples at selected temperatures. The
Mossbauer spectra of all compounds at 250 K exhibit a major
̈
Upon further decreasing temperature to 80 K, we observed
the continuation of this trend, revealing the additional decrease
of the high-spin FeII site in [FeHS-FeHS] to 31(10) and 36(7)%
for 3 and 4, respectively, whereas both the high- and low-spin
FeII sites in [FeHS-FeLS] increased to 33(1) and 35(11)% for 3
and 31(1) and 33(8)% for 4, respectively. Moreover, in the
cases of 1 and 2, the data were best fit considering only two
quadrupole doublets each, corresponding to the high- and low-
spin centers of an [FeHS-FeLS] species, with no evidence for
[FeHS-FeHS]. Indeed, this low-temperature behavior is con-
sistent with that observed in magnetic measurements, which
showed only a minor presence of [FeHS-FeHS] at low
symmetric quadrupole doublet and a second minor doublet
that is assigned to a small amount of FeIII-containing impurity
(Figure 5). Fits of the major doublet across the series gave an
isomer shift of δ = 0.91(1) mm/s and a quadrupole splitting of
ΔEQ = 2.74(2)−2.88(2) mm/s (red lines in Figure 5), which is
in good agreement with two high-spin FeII ions in a similar
coordination environment.53 We assign this doublet to the
[FeHS-FeHS] species. As temperature was decreased to 150, 120,
130, and 100 K for 1−4, respectively, this doublet developed
shoulders. In addition, a new doublet with a much smaller
quadrupole splitting appeared at lower velocity.
To further examine these spectral changes with decreasing
temperature, the data for each compound were fit considering
three distinct quadrupole doublets, two assigned to high-spin
FeII ions and one assigned to low-spin FeII. This fit revealed
that the doublet assigned to the high-spin FeII centers of the
[FeHS-FeHS] species decreased considerably in intensity, now
comprising only 21(20), 34(20), 63(10), and 63(3)% of the
total spectrum for 1−4, respectively, with slightly different
parameters of δ = 0.96(1)−0.98(1) mm/s and ΔEQ = 3.10(7)−
3.24(5) mm/s (Table 4). This small increase in ΔEQ with
decreasing temperature is commonly observed for high-spin
FeII centers and has been attributed to the increased population
of low-lying excited states.54 The fit also revealed the presence
of an additional doublet corresponding to a second high-spin
FeII center, with an isomer shift δ = 0.96(1)−0.98(1) mm/s
and a quadrupole splitting of ΔEQ = 2.89(13)−3.01(10) mm/s,
along with relative areas of 44(27), 33(19), 19(10), and
18(3)% for 1−4, respectively (green lines in Figure 5). Given
the presence of two distinct high-spin FeII sites in each
temperature. Taken together, this Mossbauer analysis further
̈
supports the hypothesis suggested by the magnetic measure-
ments, which is that the spin crossover in 1−4 involves a
mixed-spin [FeHS-FeLS] complex, with no evidence of an [FeLS-
FeLS] species.
SUMMARY AND OUTLOOK
■
The foregoing results demonstrate that subtle, remote variation
of the ligand field can lead to significant modulation of the
crossover temperature in spin crossover FeII complexes, as
2
exemplified in complexes of the isostructural compounds
[(TPyA)2Fe2(XL)]2+ (X = H, Br, Cl, and F). Indeed, the
transition temperature (T1/2) and enthalpy (ΔH) of spin
crossover across this series were found to decrease from 160(1)
to 110(1) K and from 11.4(3) to 7.5(2) kJ/mol in moving from
X = H to F, respectively. While these parameters vary
significantly with substituent, ΔS was found to remain relativity
constant. This observation suggests that the trend is directly
I
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(10) Nakano, K.; Suemura, N.; Yoneda, K.; Kawata, S.; Kaizaki, S.
related to the associated change in enthalpy rather than
entropy. Across the series, both T1/2 and ΔH decrease with
increasing electronegativity of X and consequently increasing
Dalton Trans. 2005, 740−743.
(11) Martínez, V.; Gaspar, A. B.; Munoz, M. C.; Levchenko, G.; Real,
̃
X
electron-withdrawing character of L2−, and both parameters
J. A. Chem.Eur. J. 2009, 15, 10960−10971.
(12) Rodríguez-Velamazan
́
, J. A.; Carbonera, C.; Castro, M.; Palacios,
scale linearly with the Pauling electronegativity of X. These
trends suggest that the ligand field at Fe is dominated by
contributions from inductive, rather than resonance, effects
across the series. Finally, the low-temperature magnetic data, in
E.; Let
(13) Lin, H.-J.; Siretanu, D.; Dickie, D. A.; Subedi, D.; Scepaniak, J.
J.; Mitcov, D.; Clerac, R.; Smith, J. M. J. Am. Chem. Soc. 2014, 136,
13326−13332.
(14) Real, J. A.; Gaspar, A. B.; Munoz, M. C.; Gutlich, P.;
́
ard, J.-F.; Burriel, R. Chem.Eur. J. 2010, 16, 8785−8796.
́
conjunction with variable-temperature Mossbauer spectrosco-
̈
̃
̈
py, suggest that the incomplete spin crossover behavior in
[(TPyA)2Fe2(XL)]2+ is best described as a conversion from
purely [FeHS-FeHS] (HS = high-spin) complexes at high
temperature to a mixture of [FeHS-FeHS] and [FeHS-FeLS] (LS
= low-spin) complexes at low temperature, with the number of
[FeHS-FeLS] species increasing with decreasing electron-with-
drawing character of XL2−. Future efforts will take advantage of
the chemical tunability of XL2− to increase T1/2 by further
modifying both electronic and steric properties of X. In
addition, we will seek to engender magnetic bistability in
related spin crossover compounds through replacement of
[BArF]− with smaller counteranions in an effort to introduce
large degrees of cooperativity.
Ksenofotov, V.; Spiering, H. Top. Curr. Chem. 2004, 233, 167−193.
(15) Bousseksou, A.; Molnar, G.; Real, J. A.; Tanaka, K. Coord. Chem.
́
Rev. 2007, 251, 1822−1833.
(16) Gaspar, A. B.; Munoz, M. C.; Real, J. A. J. Mater. Chem. 2006,
̃
16, 2522−2533.
(17) Olguín, J.; Brooker, S. Spin-Crossover in Discrete Polynuclear
Complexes. In Spin-Crossover Materials: Properties and Applications;
Halcrow, M. A., Ed.; John Wiley & Sons Ltd., Oxford, U.K., 2013; pp
77−120.
(18) Zein, S.; Borshch, S. A. J. Am. Chem. Soc. 2005, 127, 16197−
16201.
(19) (a) Verat, A. Y.; Ould-Moussa, N.; Jeanneau, E.; Le Guennic, B.;
Bousseksou, A.; Borshch, S. A.; Matouzenko, G. S. Chem.Eur. J.
2009, 15, 10070−10082. (b) Matouzenko, G. S.; Jeanneau, E.; Verat,
A. Y.; Bousseksou, A. Dalton Trans. 2011, 40, 9608−9618.
(20) Kitagawa, S.; Kawata, S. Coord. Chem. Rev. 2002, 224, 11−34.
(21) Miller, J. S.; Min, K. S. Angew. Chem., Int. Ed. 2009, 48, 262−
272.
ASSOCIATED CONTENT
* Supporting Information
■
S
Crystal structures of 1′, 2, and 4′, X-ray diffraction images of 1′,
additional magnetic data for 1−4, and crystallographic
information files (CIF) for 1′, 2, 3, and 4′. This material is
available free of charge via the Internet at http://pubs.acs.org.
(22) Magnetic coupling measurements: (a) Pierpont, C. G.;
Francesconi, L. C.; Hendrickson, D. N. Inorg. Chem. 1977, 16,
2367−2376. (b) Tinti, F.; Verdaguer, M.; Kahn, O.; Savariault, J.-M.
́
Inorg. Chem. 1987, 26, 2380−2384. (c) Folgado, J. V.; Ibanez, R.;
̃
Coronado, E.; Beltran
27, 19−26. (d) Chatterjee, P. B.; Bhattacharya, K.; Kundu, N.; Choi,
K.-Y.; Cler
́
, D.; Savariault, J. M.; Galy, J. Inorg. Chem. 1988,
AUTHOR INFORMATION
Corresponding Author
*E-mail: dharris@northwestern.edu.
■
́
ac, R.; Chaudhury, M. Inorg. Chem. 2009, 48, 804−806.
(e) Schweinfurth, D.; Klein, J.; Hohloch, S.; Dechert, S.; Demeshko,
S.; Meyer, F.; Sarkar, B. Dalton Trans. 2013, 42, 6944−6952.
Author Contributions
(23) Extended solids: (a) Nakabayashi, K.; Ohkoshi, S.-i. Inorg. Chem.
2009, 48, 8647−8649. (b) Abrahams, B. F.; Lu, K. D.; Moubaraki, B.;
Murray, K. S.; Robson, R. J. Chem. Soc., Dalton Trans. 2000, 1793−
1797. (c) Nagayoshi, K.; Kabir, M. K.; Tobita, H.; Honda, K.;
Kawahara, M.; Katada, M.; Adachi, K.; Nishikawa, H.; Ikemoto, I.;
Kumagai, H.; Hosokoshi, Y.; Inoue, K.; Kitagawa, S.; Kawata, S. J. Am.
Chem. Soc. 2003, 125, 221−232. (d) Luo, T.-T.; Liu, Y.-H.; Tsai, H.-L.;
Su, C.-C.; Ueng, C.-H.; Lu, K.-L. Eur. J. Inorg. Chem. 2004, 4253−
4258. (e) Atzori, M.; Benmansour, S.; Espallargas, G. M.; Clemente-
†These authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was funded by National Science Foundation
Grant DMR-1351959, Northwestern University, the Chemistry
of Life Processes Institute, and the International Institute for
Nanotechnology. We thank J. A. DeGayner and J. Martinez for
experimental assistance and Prof. J. M. Smith for helpful advice
on the synthesis of Fe(BArF)2·6MeCN.
́ ́ ́
Leon, M.; Abherve, A.; Gomez-Claramunt, P.; Coronado, E.; Artizzu,
F.; Sessini, E.; Deplano, P.; Serpe, A.; Mercuri, M. L.; Gomez García,
C. J. Inorg. Chem. 2013, 52, 10031−10040.
(24) Radical-bridged complexes: (a) Dei, A.; Gatteschi, D.; Pardi, L.;
Russo, U. Inorg. Chem. 1991, 30, 2589−2594. (b) Min, K. S.;
Rheingold, A. L.; DiPasquale, A.; Miller, J. S. Inorg. Chem. 2006, 45,
6135−6137. (c) Guo, D.; McCusker, J. K. Inorg. Chem. 2007, 46,
3257−3274. (d) Min, K. S.; DiPasquale, A. G.; Golen, J. A.; Rheingold,
A. L.; Miller, J. S. J. Am. Chem. Soc. 2007, 129, 2360−2368. (e) Baum,
A. E.; Lindeman, S. V.; Fiedler, A. T. Chem. Commun. (Cambridge,
U.K.) 2013, 49, 6531−6533.
REFERENCES
■
(1) (a) Gutlich, P.; Goodwin, H. A. Top. Curr. Chem. 2004, 233, 1−
̈
47. (b) Halcrow, M. A. Spin-Crossover Materials: Properties and
Applications; John Wiley and Sons, Ltd.: Oxford, U.K., 2013.
(2) Kahn, O.; Martinez, C. J. Science 1998, 279, 44−48.
́
(3) Letard, J.-F.; Guionneau, P.; Goux-Capes, L. Top. Curr. Chem.
́
(25) Valence tautomerism: (a) Carbonera, C.; Dei, A.; Letard, J.-F.;
2004, 235, 221−249.
Sangregorio, C.; Sorace, L. Angew. Chem., Int. Ed. 2004, 43, 3136−
3138. (b) Tao, J.; Maruyama, H.; Sato, O. J. Am. Chem. Soc. 2006, 128,
1790−1791. (c) Li, B.; Tao, J.; Sun, H.-L.; Sato, O.; Huang, R.-B.;
Zheng, L.-S. Chem. Commun. (Cambridge, U.K.) 2008, 2269−2271.
(d) Min, K. S.; DiPasquale, A. G.; Rheingold, A. L.; White, H. S.;
Miller, J. S. J. Am. Chem. Soc. 2009, 131, 6229−6236.
́
(4) Bousseksou, A.; Molnar, G.; Demont, P.; Menegotto, J. J. Mater.
Chem. 2003, 13, 2069−2071.
(5) (a) Muller, R. N.; van der Elst, L.; Laurent, S. J. Am. Chem. Soc.
2003, 125, 8405−8407. (b) Jeon, I-R.; Park, J. G.; Haney, C. R.;
Harris, T. D. Chem. Sci. 2014, 5, 2461−2465.
(6) Halcrow, M. A. Polyhedron 2007, 26, 3523−3576.
(26) Spin crossover: (a) Min, K. S.; DiPasquale, A. G.; Rheingold, A.
L.; Miller, J. S. Inorg. Chem. 2007, 46, 1048−1050. (b) Min, K. S.;
Swierczek, K.; DiPasquale, A. G.; Rheingold, A. L.; Reiff, W. M.; Arif,
A. M.; Miller, J. S. Chem. Commun. (Cambridge, U.K.) 2008, 317−319.
(7) Gaspar, A. B.; Seredyuk, M.; Gutlich, P. J. Mol. Struct. 2009, 9,
̈
924−926.
(8) Weber, B. Coord. Chem. Rev. 2009, 253, 2432−2449.
(9) Halcrow, M. A. Chem. Soc. Rev. 2011, 40, 4119−4142.
J
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(27) Tyeklar
́
, Z.; Jacobson, R. R.; Wei, N.; Murthy, N. N.; Zubieta, J.;
(53) Jeon, I-R.; Park, J. G.; Xiao, D. J.; Harris, T. D. J. Am. Chem. Soc.
2013, 135, 16845−16848.
Karlin, K. D. J. Am. Chem. Soc. 1993, 115, 2677−2689.
(28) Mansell, S. M.; Perandones, B. F.; Arnold, P. L. J. Organomet.
Chem. 2010, 695, 2814−2821.
(29) Bayen, S.; Barooah, N.; Sarma, R. J.; Sen, T. K.; Karmakar, A.;
Baruah, J. B. Dyes Pigm. 2007, 75, 770−775.
(30) Nield, E.; Tatlow, J. C. Tetrahedron 1960, 8, 38−41.
(31) Mital, R. L.; Jain, S. K. J. Chem. Soc. C 1971, 1875−1878.
(32) Reger, D. L.; Little, C. A.; Lamba, J. J. S.; Brown, K. J. Inorg.
Synth. 2004, 34, 5−8.
(33) Cullinane, J.; Jolleys, A.; Mair, F. S. Dalton Trans. 2013, 42,
11971−11975.
(34) Syukri, S.; Radhakrishnan, N.; Herdtweck, E.; Voit, B.; Kuhn, F.
(54) Merrithew, P. B.; Rasmussen, P. G.; Vencent, D. H. Inorg. Chem.
1971, 10, 1401−1406.
(55) (a) Ingalls, R. Phys. Rev. A 1964, 787−795. (b) Grunert, C. M.;
Reiman, S.; Spiering, H.; Kitchen, J. A.; Brooker, S.; Gultlich, P. Angew.
̈
Chem., Int. Ed. 2008, 47, 2997−2999. (c) Verat, A. Y.; Ould-Moussa,
N.; Jeanneau, E.; Le Guennic, B.; Bousseksou, A.; Borshch, S. A.;
Matouzenko, G. S. Chem.Eur. J. 2009, 15, 10070−10082.
(56) Gutlich, P.; Bill, E.; Trautwein, A. X. Mossbauer Spectroscopy and
̈
̈
Transition Metal Chemistry: Fundamentals and Applications; Springer-
Verlag: Heidelberg, Germany, 2010.
̈
E. Eur. J. Inorg. Chem. 2008, 18, 2892−2898.
(35) SAINT: Software for the Integration of CCD Detector System,
version 8.27B; Bruker AXS Inc.: Madison, WI, 2007.
(36) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
2008, 64, 112−122.
(37) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;
Puschmann, H. J. Appl. Crystallogr. 2009, 42, 339−341.
(38) Bain, G. A.; Berry, J. F. J. Chem. Educ. 2008, 85, 532−536.
(39) (a) Das, H. S.; Weisser, F.; Schweinfurth, D.; Su, C.-Y.; Bogani,
L.; Fiedler, J.; Sarkar, B. Chem.Eur. J. 2010, 16, 2977−2981.
(b) Sommer, M. G.; Schweinfurth, D.; Weisser, F.; Hohloch, S.;
Sarkar, B. Organometallics 2013, 32, 2069−2078. (c) Das, H. S.;
Schweinfurth, D.; Fiedler, J.; Khusniyarov, M. M.; Mobin, S. M.;
Sarkar, B. Chem.Eur. J. 2014, 20, 4334−4346.
(40) (a) Schweinfurth, D.; Khusniyarov, M. M.; Bubrin, D.; Hohloch,
S.; Su, C.-Y.; Sarkar, B. Inorg. Chem. 2013, 52, 10332−10339.
(b) Schweinfurth, D.; Rechkemmer, Y.; Hohloch, S.; Deibel, N.;
Peremykin, I.; Fiedler, J.; Marx, R.; Neugebauer, P.; van Slageren, J.;
Sarkar, B. Chem.Eur. J. 2014, 20, 3475−3486.
(41) Zhang, D.; Jin, G.-X. Organometallics 2003, 22, 2851−2854.
(42) Jia, W.-G.; Han, Y.-F.; Lin, Y.-J.; Weng, L.-H.; Jin, G.-X.
Organometallics 2009, 28, 3459−3464.
(43) (a) Kusz, J.; Gutlich, P.; Spiering, H. Top. Curr. Chem. 2004,
̈
234, 129−153. (b) Lakhloufi, S.; Guionneau, P.; Lemee
́
-Cailleau, M.
H.; Rosa, P.; Let
́
ard, J.-F. Phys. Rev. B 2010, 82, 132104-1−132104-4.
(c) Kusz, J.; Zubko, M.; Neder, R. B.; Gutlich, P. Acta Crystallogr., Sect.
̈
B: Struct. Sci. 2012, 68, 40−56.
́
(44) Guionneau, P.; Marchivie, M.; Bravic, G.; Letard, J.-F.;
Chasseau, D. Top. Curr. Chem. 2004, 234, 97−128.
(45) Kruger, H.-J. Coord. Chem. Rev. 2009, 253, 2450−2459.
̈
(46) Drew, M. G. B.; Harding, C. J.; McKee, V.; Morgan, G. G.;
Nelson, J. J. Chem. Soc., Chem. Commun. 1995, 1035−1038.
(47) (a) Ksenofontov, V.; Levchenko, G.; Spiering, H.; Gultlich, P.;
̈
́
Letard, J.-F.; Bouhedja, Y.; Kahn, O. Chem. Phys. Lett. 1998, 294, 545−
553. (b) Stassen, A. F.; de Vos, M.; van Koningsbruggen, P. J.; Renz,
F.; Ensling, J.; Kooijman, H.; Spek, A. L.; Haasnoot, J. G.; Gultlich, P.;
̈
Reedijk, J. Eur. J. Inorg. Chem. 2000, 2231−2237. (c) Moliner, N.;
Gaspar, A. B.; Munoz, M. C.; Niel, V.; Cano, J.; Real, J. A. Inorg. Chem.
̃
2001, 40, 3986−3991. (d) Stassen, A. F.; Grunert, M.; Dova, E.;
Muller, M.; Weinberger, P.; Wiesinger, G.; Schenk, H.; Linert, W.;
̈
Haasnoot, J. G.; Reedijk, J. Eur. J. Inorg. Chem. 2003, 2273−2282.
(e) Money, V. A.; Carbonera, C.; Elhaïk, J.; Halcrow, M. A.; Howard,
J. A. K.; Let
́
ard, J.-F. Chem.Eur. J. 2007, 13, 5503−5514. (f) Let
́
ard,
J.-F.; Asthana, S.; Shepherd, H. J.; Guionneau, P.; Goeta, A. E.;
Suemura, N.; Ishikawa, R.; Kaizaki, S. Chem.Eur. J. 2012, 18, 5924−
́
5934. (g) Paradis, N.; Chastanet, G.; Varret, F.; Letard, J.-F. Eur. J.
Inorg. Chem. 2013, 968−974.
(48) Ksenofontov, V.; Gaspar, A. B.; Real, J. A.; Gultlich, P. J. Phys.
̈
Chem. B 2001, 105, 12266−12271.
(49) (a) Watt, K.; Baye, L. J.; Drummond, F. O., Jr. J. Am. Chem. Soc.
1966, 88, 1138−1140. (b) Atkins, P.; de Paula, J. Physical Chemistry,
8th ed.; Oxford University Press: Oxford, U.K., 2006; Chapter 5.
(50) The data were fit in the temperature range of 65−300 K.
(51) Gultlich, P.; Hauser, A.; Spiering, H. Angew. Chem., Int. Ed.
̈
1994, 33, 2024−2054.
(52) Pauling, L. J. Am. Chem. Soc. 1932, 54, 3570−3582.
K
dx.doi.org/10.1021/ic5025586 | Inorg. Chem. XXXX, XXX, XXX−XXX