Redox coupled-spin crossover in cobalt β-diketonate complexes: Structural, electrochemical and computational studies

Article abstract of DOI:10.1016/j.poly.2012.05.037

Structural, electrochemical, spectroelectrochemical, magnetic and spectroscopic studies are reported for the octahedral cobalt β-diketonate complexes, [Co(β-diketonate)₂(N-N)] {β-diketonate = 2,2,6,6-tetramethylheptane-3,5-dionate (tmhd); N-N = 1,10-phenanthroline (phen) 1, 2,2′-bipyridine (2,2′-bpy) 2 and dimethylaminoethylamine (dmae) 3; β-diketonate = 1,3-diphenylpropane-1,3-dionate (dbm); N-N = phen 4, 2,2′-bpy 5, dmae 6}. X-ray crystallographic studies of the redox pair [Co(tmhd)₂(2,2′-bipy)]^0/+ 2/2⁺ show a shortening of the Co-ligand bond lengths by between 0.18 and 0.22 A? upon oxidation and a significantly more regular octahedral geometry around the cobalt in the cation consistent with spin crossover in addition to a change in oxidation state. Cyclic voltammetry of 1-6 reveals an irreversible one-electron oxidation to Co^III with large peak separations between the oxidation and reduction peaks, indicative of redox coupled-spin crossover (RCSCO); i.e. [Co(β-diketonate)₂(N-N)] (S = 3/2) ? [Co(β-diketonate) ₂(N-N)]⁺ + e^- (S = 0). Moreover, the complexes represent rare examples of RCSCO species with a CoO₄N₂ coordination sphere. The tmhd complexes are more easily oxidized than the respective dbm analogues with the oxidation peak potentials in the order bipy < phen < dmae. Oxidation of 1-6 with AgBF₄ yields the corresponding Co^III cations, [Co(β-diketonate) ₂(N-N)]BF₄ 1⁺-6⁺ which has been confirmed by ¹H NMR spectroscopy. Spectroelectrochemistry of the redox pairs [Co(β-diketonate)₂(N-N)]^0/+ is consistent with the isolated compounds being identical to the species formed at the electrode. Theoretical studies reveal that the SOMO is essentially metal d-orbital and β-diketonate based, consistent with the strong effect of the β-diketonate ligand on the oxidation potential. In addition, there are substantial changes in the relative stabilities of the various spin states compared with [Co(tacn)₂]^2+/3+ such that the high spin states become more accessible. The above results are consistent with a square scheme mechanism.

Full text of DOI:10.1016/j.poly.2012.05.037

Polyhedron 42 (2012) 291–301
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Redox coupled-spin crossover in cobalt b-diketonate complexes: Structural,
electrochemical and computational studies
a,
a
b
Phimphaka Harding a, , David J. Harding , Rattawat Daengngern , Theeraphol Thurakitsaree ,
⇑
⇑
Brian M. Schutte c, Michael J. Shaw c, Yuthana Tantirungrotechai d
a Molecular Technology Research Unit, School of Science, Walailak University, Thasala, Nakhon Si Thammarat 80161, Thailand
b Department of Physics, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand
c Department of Chemistry, Southern Illinois University Edwardsville, Box 1652, IL, USA
d Department of Chemistry, Faculty of Science, Thammasat University, Pathumthani 12120, Thailand
a r t i c l e i n f o
a b s t r a c t
Article history:
Structural, electrochemical, spectroelectrochemical, magnetic and spectroscopic studies are reported for
the octahedral cobalt b-diketonate complexes, [Co(b-diketonate)2(N–N)] {b-diketonate = 2,2,6,6-tetra-
methylheptane-3,5-dionate (tmhd); N–N = 1,10-phenanthroline (phen) 1, 2,20-bipyridine (2,20-bpy) 2
and dimethylaminoethylamine (dmae) 3; b-diketonate = 1,3-diphenylpropane-1,3-dionate (dbm);
N–N = phen 4, 2,20-bpy 5, dmae 6}. X-ray crystallographic studies of the redox pair [Co(tmhd)2(2,20-
bipy)]0/+ 2/2+ show a shortening of the Co-ligand bond lengths by between 0.18 and 0.22 Å upon oxida-
tion and a significantly more regular octahedral geometry around the cobalt in the cation consistent with
spin crossover in addition to a change in oxidation state. Cyclic voltammetry of 1–6 reveals an irreversible
one-electron oxidation to CoIII with large peak separations between the oxidation and reduction peaks,
indicative of redox coupled-spin crossover (RCSCO); i.e. [Co(b-diketonate)2(N–N)] (S = 3/2) M [Co(b-dike-
tonate)2(N–N)]+ + eꢀ (S = 0). Moreover, the complexes represent rare examples of RCSCO species with a
CoO4N2 coordination sphere. The tmhd complexes are more easily oxidized than the respective dbm ana-
logues with the oxidation peak potentials in the order bipy < phen < dmae. Oxidation of 1–6 with AgBF4
yields the corresponding CoIII cations, [Co(b-diketonate)2(N–N)]BF4 1+–6+ which has been confirmed by
1H NMR spectroscopy. Spectroelectrochemistry of the redox pairs [Co(b-diketonate)2(N–N)]0/+ is consis-
tent with the isolated compounds being identical to the species formed at the electrode. Theoretical stud-
ies reveal that the SOMO is essentially metal d-orbital and b-diketonate based, consistent with the strong
effect of the b-diketonate ligand on the oxidation potential. In addition, there are substantial changes in
the relative stabilities of the various spin states compared with [Co(tacn)2]2+/3+ such that the high spin
states become more accessible. The above results are consistent with a square scheme mechanism.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 21 March 2012
Accepted 30 May 2012
Available online 9 June 2012
Keywords:
Cobalt b-diketonate complexes
Redox coupled-spin crossover
X-ray crystallography
Cyclic voltammetry
DFT calculations
1. Introduction
complexes is rather rare with [Co(bipy)3]2+ and [Co(terpy)2]2+ as
perhaps the best known examples [20–23].
Spin crossover is one of the most interesting and widely studied
phenomena of transition metal complexes in part due to the possi-
ble applications of such complexes in fields such as molecular elec-
tronics, memory storage and display devices [1–4]. In principle, a
coordination complex with any electron configuration from d4–d7
may exhibit spin crossover but the most extensively studied sys-
tems are those of d6 FeII [5–18]. Moreover, in most reported cases
the iron center is surrounded by six nitrogen donors as this pro-
vides the intermediate ligand field required for spin crossover
(e.g. [Fe(phen)2(NCS)2]) [19]. Unlike FeII, spin crossover in cobalt(II)
While spin crossover systems have been widely studied, com-
paratively little attention has been given to reactions in which
the spin crossover is coupled to electron transfer. This is surprising
given that reactions in which electron transfer accompanied by a
change in spin state are widespread in both chemistry and biology
[2–4,24–27]. For instance in cytochrome P450, substrate binding
produces a change in spin state of the FeIII heme active site thereby
shifting the redox potential and triggering an electron transfer
which initiates the catalytic cycle of the enzyme [28,29]. Further
examples include the iron-molybdenum cofactor of nitrogenase
and cytochrome c peroxidases [30–32]. In relation to coordination
chemistry the oxidation of high spin (HS) CoII to low spin (LS) CoIII
is a well known example of redox coupled-spin crossover (RCSCO)
[33]. As with the spin crossover complexes discussed earlier most
⇑
Corresponding authors. Tel.: +66 7567 2100; fax: +66 7567 2004.
E-mail addresses: kphimpha@wu.ac.th (P. Harding), hdavid@wu.ac.th (D. J.
Harding).
0277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.05.037

292
P. Harding et al. / Polyhedron 42 (2012) 291–301
RCSCO complexes studied have N6 coordination environments.
Furthermore, although the mechanism of such reactions has been
debated for many years it is still unclear whether the reaction pro-
ceeds in a concerted fashion or via higher energy spin state inter-
mediates [34–41]. However, recently Schultz and co-workers have
shown that the reduction of [Co(tacn)2]3+ (tacn = 1,4,7-triazacyclo-
nonane) may proceed through a high energy spin state intermedi-
ate [42–44].
In this paper we report the synthesis of a series of octahedral
cobalt b-diketonate complexes, [Co(b-diketonate)2(N–N)] {b-dike-
tonate = 2,2,6,6-tetramethylheptane-3,5-dionate (tmhd); N–N =
1,10-phenanthroline (phen) 1, 2,20-bipyridine (2,20-bpy) 2 and
dimethylaminoethylamine (dmae) 3; b-diketonate = 1,3-diphenyl-
propane-1,3-dionate (dbm); N–N = phen 4, 2,20-bpy 5, dmae 6}
which are shown to undergo redox coupled-spin crossover. Oxida-
tion of these complexes yields the low spin d6 CoIII cations [Co(b-
diketonate)2(N–N)]+. Structural studies reveal a significant change
in Co-ligand bond lengths and a more regular octahedral geometry
for the oxidized species. Electrochemical studies support a square
scheme mechanism. Moreover, unusually for redox coupled-spin
crossover species the complexes possess a CoN2O4 coordination
sphere rather than the more common CoN6 coordination environ-
ment. Finally, through computational studies we attempt to ratio-
nalize the observed behavior in these complexes.
ilar bands are observed for other [Co(b-diketonate)2(N–N)] (b-dike-
tonate = acac, hfac; N–N = en, 2,20-bipy, phen) complexes [50,51].
2.2. Synthesis of [Co(b-diketonate)2(N–N)]+
The peak potentials from the electrochemical studies suggested
that oxidation of the neutral complexes could be achieved with Ag+
(vide infra, Table 2). Thus, addition of one equivalent of AgBF4 to
the isolated CoII species affords the cationic complexes
[Co(tmhd)2(N–N)][BF4] (N–N = phen 1+, 2,20-bpy 2+ and dmae 3+)
and [Co(dbm)2(N–N)][BF4] (N–N = phen 4+, 2,20-bpy 5+ and dmae
6+) as green-brown or purple compounds (Table 1). They may also
be prepared in a stepwise, one pot reaction from [Co(b-diketo-
nate)2(H2O)2] followed by addition of the appropriate ligand and
subsequent addition of AgBF4 giving 1+–6+. The mCO of 1+–6+ shift
to lower wavenumbers by ca. 40 or 70 cmꢀ1 for the tmhd and
dbm complexes respectively. Related [Co(acac)2(N–N)]+ (N–N =
2,20-bpy, en) complexes show mCO bands at 1524 and 1528 cmꢀ1
confirming that the CoIII complexes have been formed [52–54].
The cationic complexes despite their different colors exhibit
very similar electronic spectra to 1–6 with a band at ca. 550 nm
which by comparison with [Co(acac)2(N–N)]+ (N–N = en, 2,20-bipy,
phen) is assigned to the 1A1g (F) ? 1T1g transition [52–54]. As with
the neutral complexes additional charge transfer transitions are
evident below 350 nm. The band at ca. 290 nm for 1+, 2+, 4+ and
5+, is assigned to an MLCT transition while the other bands are as-
signed to b-diketonate
p ?
p⁄ transitions [52].
2. Results and discussion
2.3. NMR spectroscopy
2.1. Synthesis of [Co(b-diketonate)2(N–N)]
The 1H NMR spectra of 1+–6+ show sharp, readily identifiable
peaks consistent with a low spin d6 electron configuration (Table
3). The 1H NMR spectra of 1+, 2+, 4+ and 5+ show signals in the aro-
matic region consistent with the presence of coordinated phen and
bipy ligands and are assigned on the basis of coupling constants,
integration values and comparison with the related [Co(acac)2(N–
N)]+ (N–N = 2,20-bipy and phen) complexes [52,53]. As expected
we observe a single resonance for the central b-diketonate proton
indicating that the b-diketonate ligands are magnetically equiva-
lent. In the case of 1+ and 2+ there are two singlets between 1.32
and 0.80 ppm for the t-butyl groups. The phenyl protons for the
dbm ligands in 4+ and 5+ are observed between 6.97 and
8.04 ppm as a series of multiplets.
Reaction of [Co(tmhd)2(H2O)2] or [Co(dbm)2(H2O)2] (tmhd = 2,
2,6,6-tetramethylheptane-3,5-dionate, dbm = 1,3-diphenylpropane-
1,3-dionate) and N–N donor ligands in THF or CH2Cl2 yields the CoII
complexes [Co(b-diketonate)2(N–N)] {b-diketonate = tmhd; N–
N = 1,10-phenanthroline (phen) 1, 2,20-bipyridine (2,20-bpy) 2 and
dimethylaminoethylamine (dmae) 3; b-diketonate = dbm; N–
N = phen 4, 2,20-bpy 5, dmae 6; Scheme 1}. The complexes are red
or orange and air stable. IR spectroscopy of 1–6 shows the mCO band
between 1583 and 1596 cmꢀ1 indicating that the b-diketonate li-
gands adopt a chelating coordination mode (Table 1) [45]. Similar
bands have been observed in the related Ni analogues which are ob-
served between 1582 and 1595 cmꢀ1 [46]. The two dmae complexes,
3 and 6, also exhibit m
NH bands at 3365 and 3362 cmꢀ1, respectively,
The dmae complexes show more complex spectra as a result of
the asymmetry of the dmae ligand. This is most clearly evidenced
by the presence of two singlets for the b-diketonate hydrogen con-
firming the magnetic inequivalence of the b-diketonate ligands.
Moreover, in the case of 3+ there are three resonances for the t-bu-
tyl groups integrating in a ratio of 2:1:1 indicating that two of the
singlets are coincident. As with 4+ and 5+, the phenyl protons for
the dbm ligands of 6+ are split into a number of doublets and mul-
tiplets between 7.37 and 8.10 ppm. The dmae ligands show two
sharp singlets for the methyl groups but four broad resonances
for the individual protons on the ligand’s backbone. The amino pro-
tons are not observed in both cases possibly due to H-bonding [46].
and are again almost identical with those of their Ni counterparts.
Magnetic susceptibility measurements indicate that the complexes
are paramagnetic with leff between 4.41 and 5.03 BM consistent
with high spin CoII [47].
UV–Vis spectroscopy of the compounds in CH2Cl2 supports an
octahedral CoII coordination environment with a low intensity
band at ca. 570 nm which by comparison with [Co(hfac)2(tmeda)]
(hfac = hexafluoroacetylacetonate) is assigned to an overlap be-
4
4
tween the T1g (F) ? 4T2g (P) and T1g (F) ? 4A2g (F) transitions
[48]. Additional, very intense bands are found in the UV region
which are ascribed to ligand centered
p ?
p⁄ transitions [49]. Sim-
2.4. Structural studies of a CoII/III redox pair
Crystals of the redox pair [Co(tmhd)2(2,20-bipy)]0/+ 2/2+ were
grown by allowing diffusion of hexane into a concentrated solution
of the complex in CH2Cl2. The structures are shown in Fig. 1; selected
bond lengths and angles are given in Table 4 and crystallographic
data in Table 5. The structure of 2+ contains two independent mole-
cules in the unit cell, however the bonds lengths and angles in these
two molecules are virtually identical differing from each other by
less than 0.005 Å or 1° and thus the discussion is limited to only
Scheme 1. Synthesis of [Co(b-diketonate)2(N–N)].

P. Harding et al. / Polyhedron 42 (2012) 291–301
293
Table 1
IR and UV–Vis spectroscopic data for [Co(b-diketonate)2(N–N)]0/+ complexes.
a
Complex
Color
IR (cmꢀ1
)
k/nm(loge )
/Mꢀ1 cmꢀ1
mC@O
mN–H
[Co(tmhd)2(phen)] 1
[Co(tmhd)2(2,20-bpy)] 2
[Co(tmhd)2(dmae)] 3
[Co(dbm)2(phen)] 4
[Co(dbm)2(2,20-bpy)] 5
[Co(dbm)2(dmae)] 6
[Co(tmhd)2(phen)]+ 1+
[Co(tmhd)2(2,20-bpy)]+ 2+
[Co(tmhd)2(dmae)]+ 3+
Deep red
Orange
Deep red
Red
Red
Red
Dark red
Maroon
Pastel purple
1589
1587
1583
1595
1596
1594
1543
1545
1560
–
–
3365
–
–
3362
–
–
3310
3264
–
232 (4.60), 268 (4.62), 290 (4.37 sh)
284 (4.49)
282 (4.18)
232 (4.73), 268 (4.71), 352 (4.52)
247 (4.51), 350 (4.49)
249 (4.35), 349 (4.47)
234 (4.74), 268 (4.60), 300 (4.29 sh)
236 (4.73), 298 (4.38), 342 (3.81)
238 (4.65), 266 (4.41 sh), 334 (3.81 sh)
[Co(dbm)2(phen)]+ 4+
[Co(dbm)2(2,20-bpy)]+ 5+
[Co(dbm)2(dmae)]+ 6+
Olive green
Brown green
Olive green
1524
1524
1525
230 (4.69), 274 (4.88), 296 (4.83 sh), 386 (4.20)
230 (4.70), 292 (4.83), 388 (4.23)
230 (4.48), 282 (4.61), 388 (4.11)
–
3320
3264
a
As KBr discs.
Table 2
Cyclic voltammetric data of [Co(b-diketonate)2(N–N)].
and is often used to quantify octahedral distortion [57]. A related
parameter, , is defined as the sum of all the deviations from 60°
H
Peak potential (V)a
of the 24 angles between the triangular planes within the octahe-
dron and as well as quantifying octahedral distortion is also found
to directly correlate with the temperature at which spin crossover
Complex
Epa1
Epc1
Epc2
[Co(tmhd)2(phen)] 1
[Co(tmhd)2(2,20-bpy)] 2
[Co(tmhd)2(dmae)] 3
[Co(dbm)2(phen)] 4
ꢀ0.15
ꢀ0.11
ꢀ0.13
0.03
ꢀ0.27
–
–
ꢀ1.18
ꢀ1.08
ꢀ1.32
ꢀ0.98
ꢀ0.61
ꢀ1.21
ꢀ1.38
ꢀ1.41
ꢀ1.53
ꢀ0.86
ꢀ0.73
ꢀ0.72
occurs [58]. In this redox pair
is 86 and 34 for 2 and 2+, respec-
tively, thus = 52 while DH = 180 consistent with the spin
R
D
R
ꢀ0.37
ꢀ0.38
ꢀ0.43
ꢀ0.34
ꢀ0.38
ꢀ0.31
ꢀ0.22
–
change that occurs upon oxidation. Iron spin crossover compounds
also exhibit significant distortion in the high spin state which is re-
duced upon spin crossover. However, in the [Fe(NCS)2(N–N)2] series
[Co(dbm)2(2,20-bpy)] 5
[Co(dbm)2(dmae)] 6
[Co(tmhd)2(phen)]+ 1+
[Co(tmhd)2(2,20-bpy)]+ 2+
[Co(tmhd)2(dmae)]+ 3+
[Co(dbm)2(phen)]+ 4+
[Co(dbm)2(2,20-bpy)]+ 5+
[Co(dbm)2(dmae)]+ 6+
ꢀ0.07
0.06
ꢀ0.17
ꢀ0.24
ꢀ0.08
0.00
DR = 33 and DH = 79 on average, suggesting that redox coupled-
spin crossover results in an even greater reduction in octahedral
distortion [58]. In addition, while the tmhd ligands bind in an angu-
lar fashion to the cobalt center in 2 they exhibit a planar coordina-
tion mode in 2+ as demonstrated by the change in the b angle (see
Table 4). This profound difference in bond lengths and geometry
around the metal is due to a spin change in addition to an increase
in oxidation state which result in complete depopulation of the
ꢀ0.06
ꢀ0.17
–
a
All measurements were performed at 298 K, in a dried and degassed CH2Cl2
0.1 M [NBun4][PF6] solution; scan rate 100 mV sꢀ1; calibrated with [FeCp2], and
reported relative to the [FeCp2]0/+ couple.
⁄
antibonding eg orbitals. This combined change in spin and oxida-
one of these independent molecules. The structures of 2 and 2+ are
broadly similar to each other possessing an octahedral CoO4N2 coor-
dination environment. The most striking change upon oxidation is a
shortening of the metal ligand bond lengths by between 0.18 and
0.22 Å. Similar differences are observed for the [Co(2,20-bipy)3]2+/
3+, [Co(phen)3]2+/3+ and [Co(en)3]2+/3+ redox pairs where the differ-
ence is 0.19 Å in all cases [55,56].
tion state is known as redox coupled-spin crossover.
2.5. Electrochemistry
The complexes 1–6 have been studied by cyclic voltammetry
(CV) and square wave voltammetry in dry CH2Cl2 under anaerobic
conditions and reveal irreversible one-electron oxidation. The CV
of [Co(tmhd)2(phen)] 1 is shown in Fig. 2a as a representative
example. The cyclic voltammograms (CVs) are broadly similar,
exhibiting a single oxidation peak, Epa1, and in most cases a small
reduction peak, Epc1, followed by a much larger reduction peak,
Interestingly, the degree of distortion around the metal center is
also markedly different with the cation, 2+ exhibiting a much more
regular octahedral environment. The parameter, R, is the sum of the
deviations from 90° of all the cis angles in the coordination sphere
Table 3
1H NMR spectroscopic data of [Co(b-diketonate)2(N–N)]BF4 complexes.
Complex
d (ppm)
[Co(tmhd)2(phen)]+
8.98 (2H, dd, JHH 0.6 Hz, 8.1 Hz, CHPhen), 8.38 (2H, d, JHH 5.4 Hz, CHPhen), 8.34 (2H, s, CHPhen), 8.15 (2H, dd, JHH 8.1 Hz, 5.4 Hz, CHPhen), 5.91 (2H, s,
CHcentral), 1.31 (18H, s, t-Bu), 0.80 (18H, s, t-Bu)
1+
[Co(tmhd)2(2,20-
bpy)]+ 2+
8.90 (2H, d, JHH 7.8 Hz, H6, bpy), 8.42 (2H, t, JHH 7.8 Hz, 6.3 Hz, H5, bpy), 8.12 (2H, d, JHH 5.1 Hz, H3, bpy), 7.70 (2H, t, JHH 6.3 Hz, 5.1 Hz, H4, bpy), 5.86
(2H, s, CHcentral), 1.32 (18H, s, t-Bu), 0.87 (18H, s, t-Bu)
[Co(tmhd)2(dmae)]+
3+
5.87 (1H, s, CHcentral), 5.80 (1H, s, CHcentral), 4.00 (2H, s, N–CH2), 2.80 (2H, s, N–CH2), 2.39 (3H, s, N–CH3), 1.80 (3H, s, N–CH3), 1.21 (18H, s, t-Bu),
1.12 (9H, s, t-Bu), 1.05 (9H, s, t-Bu)
[Co(dbm)2(phen)]+
4+
9.01 (2H, dd, JHH 8.1 Hz, 0.9 Hz, CHPhen), 8.70 (2H, d, JHH 5.1 Hz, CHPhen), 8.40 (2H, s, CHPhen), 8.12 (2H, dd, JHH 8.1 Hz, 5.1 Hz, CHPhen), 8.04 (4H, m,
H
Ph), 7.60–7.55 (6H, m, HPh), 7.50–7.40 (6H, m, HPh), 7.32–7.28 (4H, m, HPh), 7.01 (2H, s, CHcentral
)
[Co(dbm)2(2,20-
bpy)]+ 5+
9.03 (2H, d, JHH 7.50 Hz, H6, bpy), 8.47 (4H, m, H6, bpy, H3, bpy), 7.98 (4H, d, JHH 7.20 Hz, HPh), 7.70 (2H, t, JHH 6.90 Hz, H4, bpy), 7.64 (4H, d, JHH
7.20 Hz, HPh), 7.55–7.34 (12H, m, HPh), 6.97 (2H, s, CHcentral
)
[Co(dbm)2(dmae)]+
6+
8.10 (2H, d, JHH 6.90 Hz, HPh), 7.97 (3H, m, HPh), 7.91 (2H, d, JHH 7.2 Hz, HPh), 7.71 (2H, d, JHH 7.50 Hz, HPh), 7.59–7.50 (9H, m, HPh), 7.42–7.37 (4H,
m, HPh), 6.97 (1H, s, CHcentral), 6.81 (1H, s, CHcentral), 4.70 (1H, broad, s, N–CH2), 4.50 (1H, broad, s, N–CH2), 3.20 (1H, broad, s, N–CH2), 2.80 (1H,
broad, s, N–CH2), 2.56 (3H, s, N–CH3), 2.01 (3H, s, N–CH3)

294
P. Harding et al. / Polyhedron 42 (2012) 291–301
Fig. 1. POV-Ray thermal ellipsoid plot of 2 (left) and 2+ (right) with ellipsoids at 50%. Only selected atoms are labeled and hydrogen atoms are omitted for clarity.
Table 4
Table 5
Selected bond lengths (Å) and angles (°) of [Co(tmhd)2(2,20-bipy)]0/+ 2/2+.
Crystallographic data and structure refinement for complexes [Co(tmhd)2(2,20-
bipy)]0/+ 2/2+.
2a
2+
D
Co–L
D\
2
2+
Co–O1
Co–O2
Co–O3
Co–O4
Co–N1
Co–N2
N(1)–Co–N(2) {N(1)⁄}
N(1)–Co–O(1) {O(2)}
N(1)–Co–O(2) {O(1)}
N(1)–Co–O(3) {O(2)⁄}
N(1)–Co–O(4) {O(1)⁄}
N(2)–Co–O(1)
2.048(1)
2.051(1)
–
–
2.128(1)
–
1.867(5)
1.871(5)
1.862(4)
1.865(5)
1.915(5)
1.911(5)
0.181
0.180
0.186
0.186
0.213
0.218
Formula
C32H46CoN2O4 C64H92B2Co2F8N4O8
Molecular weight (g molꢀ1
)
581.64
monoclinic
C2/c
1336.90
monoclinic
P21/c
Crystal system
Space group
a (Å)
b (Å)
c (Å)
11.2757(2)
30.2006(6)
9.9551(2)
90
25.8255(14)
14.4006(7)
20.2612(10)
90
76.02(5)
80.91(4)
161.69(5)
102.44(4)
94.87(5)
–
–
–
–
85.73(4)
97.96(4)
175.82(4)
91.52(4)
–
83.7(2)
86.6(2)
177.0(2)
93.5(2)
90.4(2)
88.7(2)
93.7(2)
176.5(2)
87.4(2)
94.8(2)
89.2(2)
175.3(2)
89.2(2)
88.1(2)
94.7(2)
7.68
5.69
15.31
8.94
4.47
a
(°)
b (°)
(°)
111.3550(10)
90
91.059(3)
90
c
T (K)
100(2)
100(2)
Cell volume (Å3)
3157.29(11)
4
7533.9(7)
4
N(2)–Co–O(2)
N(2)–Co–O(3)
N(2)–Co–O(4)
O(1)–Co–O(2)
Z
Absorption coefficient (mmꢀ1
Reflections collected
)
0.580
0.508
17040
75018
9.07
8.76
0.52
2.22
Independent reflections (Rint
Maximum and minimum
transmission
Restraints/parameters
Final R indices [I > 2
)
3662, 0.015
0.746, 0.671
9163, 0.085
0.746, 0.637
O(1)–Co–O(3) {O(2)⁄}
O(1) {O(2)}–Co–O(4) {O(2)⁄}
O(1)–Co–O(3) {O(1)⁄}
O(2)–Co–O(4)
0/223
174/917
r
(I)]: R1, wR2
0.032, 0.080
0.081, 0.205
O(3)–Co–O(4)
–
b
23.5
23.5
86
10.3
13.4
34
13.2
10.1
52
R
crossover is best described by the square scheme shown in
Scheme 2.
H
223
43
180
a
The labels for symmetry related atoms in 2 are given in parenthesis.
The two processes 1 and 3 represent the oxidation and reduction
reactions of the individual spin state isomers while processes 2 and
4 represent the spin exchange reactions in the individual oxidation
states. Redox coupled-spin crossover has also been observed and
extensively studied by Schultz et al. in a series of [M(tacn)2]2+/3+
(M = Fe, Co; tacn = 1,4,7-triazacyclononane) complexes although
in this system the redox processes are fully reversible [42–44]. Sim-
ilar redox behavior to 1–6 has recently been reported in the dimeric
complex [(bpbp)Co2(O2P(OPh)2)2]+ (bpbpꢀ = 2,6-bis(N,N0-bis-(2-
picolyl)amino)methyl)-4-tertbutylphenolato) which also shows
wide separation of the oxidation and reduction peaks [61]. A similar
result is also found in the case of [(TpR)2Co]0/+ (TpR = Tp, pzTp and
Tp⁄) where increasing steric bulk results in a corresponding in-
crease in irreversibility [62]. Comparisons between [Co(tacn)2]2+
and complexes 1–6 show that the former is oxidized ca. 0.9 V more
easily than the latter consistent with the greater stability of the CoII
oxidation state for the [Co(b-diketonate)2(N–N)] complexes. In con-
trast, the [(bpbp)Co2(O2P(OPh)2)2]+ dimer oxidizes at a higher po-
tential, 0.42 V (vs. [FeCp2]0/+).
Epc2, which is widely separated from Epa1 (Table 2). The irrevers-
ibility of the redox processes, contrasts with the Ni analogues
which show either quasi-reversible oxidation or completely
chemically irreversible oxidations with no corresponding
reduction peaks [46], suggests that the oxidation/reduction is
accompanied by a structural and electronic rearrangement. The
electrochemical data may suggest that the oxidation of 1–6 to
CoIII (Epa1) initially produces
a
high spin CoIII intermediate
(Epc1) which can be re-reduced close to the oxidation peak,
Epa1. Epc1 is presumably the cathodic partner to the anodic peak
represented by Epa1. The high spin CoIII species rapidly undergoes
spin crossover to the more stable low spin CoIII isomer which is
reduced at a significantly more negative potential (Epc2) resulting
in a wide separation between the oxidation and reduction peak
potentials, Epa1 and Epc2. The absence of a corresponding peak
for the low spin CoII intermediate (Epa2) may be due to more ra-
pid spin crossover in this case. It is important to note that a con-
certed pathway directly from high spin CoII to low spin CoIII is also
consistent with the above data. The redox behavior of these com-
plexes is in agreement with the structural studies and indicative
of redox coupled-spin crossover [59,60]. Redox coupled-spin
The oxidation peak potentials for 1–6 are ca. 0.54 V more nega-
tive than the related Ni compounds [46]. A similar difference in the
redox potentials has also been observed in the two series
[M(tacn)2]2+/3+ and [(TpR)2M]0/+ (TpR = Tp, pzTp and Tp⁄; M = Co,
Ni) [42,62–64]. Within the two series of 1–6 we find that the tmhd

P. Harding et al. / Polyhedron 42 (2012) 291–301
295
Fig. 2. Cyclic voltammograms of (a) [Co(tmhd)2(phen)] 1 (b) [Co(tmhd)2(phen)]BF4 1+[BF4].
easily oxidized which is more in line with the typical electron
donating properties of diamine and diimine ligands. The difference
in the order of the peak potentials may be the result of the spin
change which occurs upon oxidation. The difference in the oxida-
tion and reduction peak potentials (
0.54 to 1.06 V with 1–3 exhibiting larger
D
E = Epa1 ꢀ Epc2) varies from
D
E values than 4–6,
thereby indicating that the greater steric bulk of the t-butyl groups
results in a reduction in the rate of electron-transfer. Similar obser-
vations in the [(TpR)2Co]0/+ series are also attributed to steric ef-
fects [62]. In the case of [(bpbp)Co2(O2P(OPh)2)2]+
DE = 1.04 and
0.61 V for the two redox reactions which agrees well with our find-
ings for 1–6 [61].
Scheme 2. Square scheme of the redox coupled-spin crossover process showing
individual spin crossover and electron transfer processes.
It is interesting to note that while the [Co(tacn)2]2+/3+ and
[(TpR)2Co]0/+ redox pairs have a CoN6 coordination sphere 1–6
and [(bpbp)Co2(O2P(OPh)2)2]+ possess CoN2O4 and CoN3O3 coordi-
nation spheres, respectively. The presence of the weaker field oxy-
gen donors would be expected to favor the high spin isomers and
may be responsible for the irreversibility seen in these systems.
It may also explain why we are able to observe a reduction peak
for the high spin CoIII intermediate in some of the complexes.
In an attempt to better understand the redox coupled-spin
crossover process in complexes 1–6 we undertook cyclic voltam-
metric studies at high scan rates using a microelectrode. There
complexes are easier to oxidize than the corresponding dbm com-
plexes by ca. 0.14 V indicating that the inductive effects of the t-
butyl groups increase the electron density on the metal thereby
making it easier to oxidize. A similar trend is observed for the Ni
analogues although the difference is larger by ca. 0.30 V [46]. The
Co complexes show oxidation peak potentials in the order bipy
< phen < dmae. This order is surprising given that for the [Ni(b-
diketonate)2(N–N)] complexes the dmae compounds are the most

296
P. Harding et al. / Polyhedron 42 (2012) 291–301
was little change in the reversibility of the complexes even at
moderately fast scan rates (10 V sꢀ1) consistent with a rapid spin
exchange equilibrium and slow electron transfer as is frequently
observed in these systems (Fig. 3) [33,60]. This result seems rea-
sonable since going from the high spin to the low spin state does
not require any bonds to be broken but rather optimization of
the Co-ligand bond lengths. Similar irreversibility at high scans
rates is also observed in the [(bpbp)Co2(O2P(OPh)2)2]+ dimer. In
addition, studies on a wide range of cobalt am(m)ine systems also
reveal slow electron transfer [60].
Despite the fact that redox coupled-spin crossover reactions are
well documented the mechanism still remains the subject of con-
siderable debate with a pathway via high energy spin-state inter-
mediates and a concerted pathway both proposed [34–41]. As
noted earlier the cyclic voltammograms of 1–6 clearly show the
presence of peaks which may be attributable to a high spin CoIII
intermediate and suggest that the mechanism for the oxidation
of 1–6 proceeds via a high spin intermediate. However, it must
be stated that at this time we cannot rule out a concerted pathway.
Moreover, we cannot definitively prove which high spin intermedi-
ate may be involved, although on the basis of computational stud-
ies (vide infra) a high spin triplet seems likely. In addition, at all
applied scan rates (0.02–10 V sꢀ1) we were unable observe a peak
for the potential low spin CoII intermediate further supporting high
spin pathway.
served current [65]. In such a reaction, the HS CoIII produced by
equilibrium 2 in Scheme 2 would react with electrogenerated LS
CoII. Thus, the reduction of the LS CoIII to HS CoII proceeds by the
expected LS CoII intermediate. Similar reasoning leads to the con-
clusion that the oxidation of the HS CoII to LS CoIII proceeds by
the observed HS CoIII intermediate.
To confirm that the products isolated from the chemical oxida-
tion of 1–6 were the same as those observed in the cyclic voltam-
mograms we undertook IR spectroelectrochemical studies. The
results are detailed in Table 6 with a representative difference
spectrum shown in Fig. 4. The oxidation of 1–6 are all accompanied
by the loss (i.e. downward-pointing features) of the b-diketonate
mCO bands at ca. 1590, 1570, 1520 and 1500 cmꢀ1 due to the CoII
starting materials and a corresponding increase in bands at be-
tween ca. 1550 and 1520 cmꢀ1 due to the CoIII products. Bands at
frequencies lower than 1500 cmꢀ1 are lost due to interference from
the electrolyte. Subsequent reduction restores the original spec-
trum showing (as expected) that the chemical oxidation or reduc-
tion is reversible. The shift in the bands to lower wavenumbers
mirrors exactly what is observed in the isolated complexes and is
consistent with the higher CoIII oxidation state. Moreover, the posi-
tions of the bands are almost identical between the isolated redox
pairs and the results from spectroelectrochemistry confirm that
the isolated complexes are the species formed at the electrode.
Similar studies on the cationic species 1+–6+ show an exact oppo-
site trend with bands moving to higher wavenumbers as one
would expect. Consistent with the CV results, no features due to
HS CoIII were observed, consistent with the presence of a very small
equilibrium concentration.
The square wave voltammograms show peak potentials similar
to those found in the CVs. The slight deviations between the two
methods might be due to the dependence of the peak potentials
on scan rate noted in the CVs. The peaks also exhibit slight asym-
metry consistent with the electrochemical irreversibility observed
in the CVs.
The cations 1+–6+ were also studied by cyclic voltammetry and
show a large reduction peak coupled to a smaller redox couple at a
higher potential (Fig. 2b). The peak potentials observed in the cyc-
lic voltammograms are almost identical to those found in the neu-
tral analogues although there are slight deviations. Similar results
are noted in the complex [(bpbp)Co2(O2P(OPh)2)2]3+ where shifts
of up to 0.18 V are observed [61]. As with 1–6 the peaks for the
high spin CoIII intermediate are clearly discernable but no peak is
apparent (i.e. ‘‘Epa2’’) for the low spin CoII intermediate. Such
behavior is typical of square scheme systems which involve first
order isomerization of an initial electrode product to the final
observed form [61]. Furthermore, the apparent reversibility of
the CVs do not change with concentration a result which rules
out significant contribution of the second-order homogeneous
cross reaction (i.e. LS CoII + HS CoIII ? LS CoIII + HS CoII) to the ob-
2.6. Computational studies
In order to better understand the electronic structure of 1–6,
and their respective cations, and in particular the role of the elec-
tronic structure on the redox coupled-spin crossover process we
have undertaken DFT calculations [66]. For a given complex, all
possible spin states were considered.
The fact that we have structurally characterized [Co(tmhd)2
(2,20-bipy)] 2 allows a comparison between the computational
and experimental geometries (see Table 7). The computed bond
lengths match well with the experimental bond lengths with dif-
ferences no more than 0.05 Å. The bonds angles show similar
agreement between experiment and theory. Good agreements be-
tween the X-ray structure and computed geometry of [Co(tmhd)2
(2,20-bipy)] in the quartet spin state indicates that the complex
Fig. 3. Cyclic voltammograms of [Co(tmhd)2(phen)] at scan rates of 50–1600 mV/s.

P. Harding et al. / Polyhedron 42 (2012) 291–301
297
Table 6
IR spectroelectrochemical difference data of [Co(b-diketonate)2(N–N)] complexes (cmꢀ1).
Complex
Bands of species produced
Bands of starting material consumed
[Co(tmhd)2(phen)] 1a
[Co(tmhd)2(2,20-bpy)] 2a
[Co(tmhd)2(dmae)] 3a
[Co(dbm)2(phen)] 4a
1552, 1543, 1523
1545, 1527
1557, 1547, 1533
1587, 1528
1530
1526
1589, 1573, 1506
1587, 1572, 1506
1596, 1581, 1569
1596, 1557, 1552, 1512
1597, 1579, 1554, 1513
1596, 1557, 1516
1587, 1572, 1533, 1516, 1505
1588, 1581, 1570, 1518, 1504
1592, 1582, 1570
1596, 1564, 1554, 1514
1597, 1563, 1549
1563, 1553
1558, 1548, 1525
1552 (sh), 1544, 1531, 1496
1546, 1534
1538 (sh), 1529
[Co(dbm)2(2,20-bpy)] 5a
[Co(dbm)2(dmae)] 6a
[Co(tmhd)2(phen)]+ 1+b
[Co(tmhd)2(2,20-bpy)]+ 2+b
[Co(tmhd)2(dmae)]+ 3+b
[Co(dbm)2(phen)]+ 4+b
[Co(dbm)2(2,20-bpy)]+ 5+b
[Co(dbm)2(dmae)]+ 6+b
1530, 1588
1588, 1540, 1528
a
Oxidized at ca. Epa1.
Reduced at Epc2.
b
Table 7
Selected computed and X-ray crystallographically determined bond lengths (Å) and
bond angles (°) for [Co(tmhd)2(2,20-bipy)] 2. See Fig. 1 for the numbering scheme.
2
X-ray
2.048
Doublet
Diff.a
Quartet
2.0589
Diff.a
Co–O1
Co–O2
Co–N1
C1–O1
C7–O2
C1–C6
C6–C7
O1–Co–O2
O2–Co–O2⁄
O1–Co–O2⁄
O1–Co–N1
O1–Co–N1⁄
O2–Co–N1
O1–Co–O1⁄
N1–Co–N1⁄
2.144
1.948
1.946
1.308
1.293
1.412
1.425
88.40
176.99
89.43
93.53
175.40
89.37
90.44
82.53
0.10
0.10
0.18
0.04
0.03
0.01
0.02
2.67
1.17
2.09
1.32
13.79
13.07
7.52
6.51
0.01
0.02
0.01
0.03
0.04
0.01
0.01
0.16
1.02
1.01
0.98
4.70
9.31
0.34
0.43
2.051
2.128
1.269
1.261
1.403
1.407
85.73
175.82
91.52
94.87
161.69
102.44
97.96
76.02
2.0680
2.1339
1.3027
1.3002
1.4159
1.4201
85.57
176.84
92.53
95.85
166.39
93.31
97.62
76.45
Fig. 4. IR spectroelectrochemical difference spectra of [Co(dbm)2 (phen)]BF4
4+[BF4].
a
Difference between the X-ray crystallographic and optimized structure.
exists in a high spin state. It also confirms that the B3LYP/SDD
model can be used to describe these types of complexes.
associated with spin crossover calculations are ca. 3 kcal/mol, even
for the state-of-the-art TPSSh functional [67], we can only conclude
that the triplet and quintet states lie close to each other in energy.
The differences between E(HS)–E(LS) of 1+–6+ and [Co(tacn)2]3+ are
simply explained by the presence of the lower field b-diketonate
ligands which are better able to stabilize the HS states. The thmd
ligand stabilizes the HS state compared to the LS state to a greater
extent than the dbm ligand. Interestingly, the phen ligand stabi-
lizes the HS state most in comparison with the other N–N ligands.
This might be due to a d–p⁄ orbital interaction from a low-lying p⁄
orbital of the phen ligand. The smaller difference between the LS
and HS states in 1+–6+ compared with [Co(tacn)2]3+ might also ex-
plain why we are able to observe the reduction peak for the HS CoIII
species in some of the electrochemical studies. These studies fur-
ther suggest that the mechanism for the redox coupled-spin cross-
over in this system may involve a HS CoIII intermediate.
In comparing the different functionals, the TPSSh functional
actually yields the largest E(HS)–E(LS) gap for the CoII complexes
despite it having the lowest exchange energy contribution (10% ex-
act exchange), while the B3LYP⁄ functional gives a much smaller
gap. Friesner and Hughes have recently compared B3LYP and
B3LYP⁄ and concluded that while the B3LYP⁄ functional makes
the relative energies of all the spin states more positive it doesn’t
affect the spin states equally and this can lead to B3LYP⁄ actually
performing less well than B3LYP [70].
Despite this agreement between the computed and observed
bond lengths and angles the stability of the high spin states are
known to be routinely overestimated by B3LYP. This is largely
due to the fact that B3LYP significantly overestimates the exchange
energy [67,68]. One approach to correct this is to reduce the
amount of exchange energy in the hybrid functional. Two particu-
larly effective functionals which have been used primarily with
spin crossover compounds are B3LYP⁄ and TPSSh where the degree
of exchange is 15% and 10%, respectively [68,69]. The results for all
three functionals are reported in Table 8.
All calculations confirm the observation that the HS state is the
most stable state for the CoII complexes and the LS state is the most
stable state for the CoIII complexes and as such are consistent with
the magnetic susceptibility measurements. In contrast, with the
previously studied [Co(tacn)2]2+/3+ redox pair the gas-phase energy
difference between the HS and LS states, E(HS)-E(LS), for the
neutral CoII complexes is between ꢀ8.7 and ꢀ17.3 kcal/mol (see
Table 7) compared with ꢀ0.09 kcal/mol for [Co(tacn)2]2+. In the
case of the CoIII cations the gap is significantly decreased to be-
tween 15.0 and 21.4 kcal/mol (cf. 45.3 kcal/mol for [Co(tacn)2]3+).
As expected reducing the amount of exchange energy results in a
re-ordering of the quintet and triplet state of the CoIII cations in
the case of the B3LYP⁄ and TPSSh functionals compared to the
B3LYP functional, although the difference between the quintet
and triplet state is still very small. It follows that as the errors often
Thus, in this system with a high spin ground state the quartet
state undergoes a larger change in energy than the doublet state

298
P. Harding et al. / Polyhedron 42 (2012) 291–301
Table 8
The high-spin and low-spin complex energy difference, E(HS)–E(LS), at B3LYP/SDD,
B3LYP⁄/SDD and TPSSh/SDD level for [Co(b-diketonate)2(N–N)] complexes. For 1+–6+,
the quintet and triplet (in parenthesis) HS states were considered (kcal molꢀ1).
Complex
E(HS)–E(LS)a
E(HS)–E(LS)b
E(HS)–E(LS)c
[Co(tmhd)2(phen)] 1
[Co(tmhd)2(2,20-bpy)] 2
[Co(tmhd)2(dmae)] 3
[Co(dbm)2(phen)] 4
[Co(dbm)2(2,20-bpy)] 5
[Co(dbm)2(dmae)] 6
[Co(tmhd)2(phen)]+ 1+
[Co(tmhd)2(2,20-bpy)]+ 2+
[Co(tmhd)2(dmae)]+ 3+
[Co(dbm)2(phen)]+ 4+
[Co(dbm)2(2,20-bpy)]+ 5+
[Co(dbm)2(dmae)]+ 6+
ꢀ15.63
ꢀ12.40
ꢀ17.34
ꢀ14.55
ꢀ10.32
ꢀ15.16
ꢀ12.67
ꢀ10.44
ꢀ16.84
ꢀ13.84
ꢀ9.79
ꢀ14.60
ꢀ12.80
ꢀ8.66
ꢀ13.50
ꢀ13.26
ꢀ10.11
ꢀ15.45
16.12 (16.38)
17.15 (17.19)
17.41 (17.60)
16.93 (17.31)
18.01 (18.16)
17.93 (16.61)
19.59 (17.84)
20.66 (18.67)
20.71 (19.05)
20.32 (18.69)
21.44 (19.53)
21.17 (17.92)
15.01 (13.74)
16.10 (14.61)
16.23 (14.60)
15.82 (14.85)
16.92 (15.84)
16.70 (14.68)
a
b
c
B3LYP/SDD.
B3LYP⁄/SDD.
TPSSh/SDD.
resulting in a greatly reduced E(HS)–E(LS) gap. For instance, in
[Co(tmhd)2(phen)] the relative energy of the doublet is
1
730 kcal/mol higher than that of the B3LYP functional while that
of the quartet is 733 kcal/mol higher yielding a difference of
3 kcal/mol. However, for the TPSSh functional the difference is only
1 kcal/mol resulting in a E(HS)–E(LS) gap comparable to that pre-
dicted by B3LYP. Similar results are found for 2–6. The tendency
of the B3LYP⁄ functional to affect the different spin states un-
equally is also manifest in the cations with differences in the
changes of the relative energies of the triplet and quintet states
Fig. 6. The LUMO orbital of complexes 1–6.
typically 4–5 kcal/mol compared to 1–2 kcal/mol for the TPSSh
functional. This results in the B3LYP⁄ functional predicting the larg-
est E(HS)–E(LS) gap while the TPSSh functional gives the smallest
gap with B3LYP in between. Overall, our results indicate that the
state-of-the-art TPSSh functional performs best among the three
functionals. The rather poor performance of the B3LYP⁄ functional
especially for the CoII complexes is probably due to the fact that
unlike most systems where B3LYP⁄ performs well, 1–6 do not ex-
hibit thermal spin crossover and contain principally oxygen donor
ligands rather than aromatic nitrogen ligands.
The molecular orbital analysis reveals that in the case of 1–3
and 6 the SOMO is composed of an antibonding interaction be-
tween the b-diketonate oxygen p-orbitals and the cobalt dz2 orbital
(see Fig. 5). The dmae complexes also exhibit an additional anti-
bonding interaction between the metal orbital and a hybrid
r do-
nor orbital on the NMe2 nitrogen of the dmae ligand. The SOMO of
the related bipy and phen complexes, 4 and 5 are different from the
others with a metal dxz orbital involved in a weaker antibonding
interaction with the b-diketonate ligands. The significant electron
density on the b-diketonate ligands in 1–6 is consistent with the
strong influence of the b-diketonate upon the oxidation potentials
of the above complexes. In the case of 1, 2, 4 and 5 the LUMO is
essentially a low lying phen or bipy p⁄ orbital while for 3 and 6
the absence of such p⁄ orbitals results in a LUMO which is domi-
nated by a b-diketonate p⁄ orbital (Fig. 6). Consequently, a much
larger SOMO–LUMO gap is observed in the case of the dmae com-
plexes compared with 1, 2, 4 and 5.
3. Conclusions
In summary, six redox-active Co b-diketonate complexes have
Fig. 5. The SOMO orbital of complexes 1–6.
been prepared in two oxidation states which undergo a redox

P. Harding et al. / Polyhedron 42 (2012) 291–301
299
coupled-spin crossover from high spin d7 to low spin d6 which has
been confirmed by cyclic voltammetry and spectroelectrochemical
studies. Chemical oxidation of the Co b-diketonate compounds per-
mits isolation of the CoIII cations which are low spin as shown by
NMR spectroscopy. Structural studies of the redox pair show that
the cobalt-ligand bond lengths are substantially shorter and the
geometry around the cobalt center much less distorted upon oxi-
dation. Electrochemical and computational studies on the redox
pairs strongly suggest that the redox coupled-spin crossover in
the above complexes proceeds via a square scheme mechanism
which does not involve a significant amount of homogenous
cross-reaction.
auxiliary electrode was a platinum rod and the working electrode
was a platinum disc (2.0 mm diameter). The reference electrode
was a Ag–AgCl electrode. Solutions were 5 ꢁ 10ꢀ4 M in the test
compound and 0.1 M in [NBun4][PF6] as the supporting electrolyte.
Under these conditions, E°0 for the one-electron oxidation of [Fe(
g-
C5H5)2] added to the test solutions as an internal calibrant is
0.52 V. For the electrochemical studies at high scan rates a 50
l
m
platinum disc microelectrode was used as the working electrode
with a silver and platinum wire used as the reference and auxiliary
electrode, respectively. Fiber-optic difference IR spectroelectro-
chemical measurements were obtained as previously described
[79]. The cyclic voltammograms of [Co(tmhd)2(H2O)2] and
[Co(dbm)2(H2O)2] are presented in Figures S1 and S2 in the sup-
porting information.’
4. Experimental
4.5. Calculations
4.1. Materials
All calculations were performed by using the Gaussian 03 pack-
age [66]. The B3LYP, B3LYP⁄ and TPSSh hybrid functional with the
Stuttgart/Dresden SDD effective core potential basis set was used
in all calculations [80–83]. The geometries of complexes for a given
spin state were optimized and verified by performing Hessian cal-
culations. The molecular orbital analyses were then conducted at
those geometries [66]. The SOMO and LUMO three-dimensional
isosurface plots were generated using the Avogadro program
[84]. The atomic coordinates for all computed structures may be
found in the supporting information.
All reactions were conducted in air using HPLC grade solvents.
[Co(tmhd)2(H2O)2], and [Co(dbm)2(H2O)2] were prepared by litera-
ture methods [71,72]. [Co(dbm)2(phen)] 4 has previously been
made although our synthesis differs from that reported [73]. All
other chemicals were purchased from Fluka Chemical Company
and used as received. Elemental analyses and ESI–MS were carried
out by the staff of the School of Chemistry, University of Bristol, UK.
Elemental analyses were carried out on a Eurovector EA3000 ana-
lyser. ESI–MS were carried out on a Bruker Daltonics 7.0T Apex 4
FTICR Mass Spectrometer. Magnetic susceptibilities were deter-
mined using the Evan’s method at 297 K [74].
4.5.1. [Co(tmhd)2(phen)] 1
To a stirred purple solution of [Co(tmhd)2] (0.0826 g, 0.2 mmol)
in CH2Cl2 (15 cm3), phen (0.0396 g, 0.2 mmol) was added. The or-
ange solution was stirred for 1 hour at room temperature then fil-
tered through celite. The solution was left to slowly evaporate at
room temperature yields deep red needle crystals 0.0772 g (64%).
4.2. Spectroscopy
Infrared spectra, as KBr discs, were recorded on a Perkin-Elmer
Spectrum One infrared spectrophotometer in the range 400–
4000 cmꢀ1. Electronic spectra were recorded in CH2Cl2 on a Uni-
cam UV300 UV–Vis spectrometer. 1H NMR spectra were recorded
on a Bruker 300 MHz FT-NMR spectrometer at 298 K in CDCl3 with
SiMe4 added as an internal standard.
Magnetic moment (leff/lB, 297 K): 4.51. C34H46N2O4Co; Anal. Calc.
C 67.4, H 7.7, N 4.6; found C 67.5, H 7.4, N 5.6.
4.5.2. [Co(tmhd)2(2,20-bpy)] 2
To
a stirred purple solution of [Co(tmhd)2] (0.0638 g,
0.15 mmol) in CH2Cl2 (5 cm3), 2,20-bpy (0.0240 g, 0.15 mmol) was
added. The orange solution was stirred for 1 h then filtered through
celite. The solution was left to slowly evaporate at room tempera-
ture yields orange needle crystals 0.0451 g (53%). Magnetic mo-
4.3. X-ray crystallography
Crystal data and data processing parameters for the structures
of 2 and 2+ are given in Tables 4 and 5. X-ray quality crystals of
2 and 2+ were grown by allowing hexane to diffuse into a concen-
trated solution of the complex in CH2Cl2. Crystals were mounted on
a glass fiber using perfluoropolyether oil and cooled rapidly to
100 K in a stream of cold nitrogen. All diffraction data were col-
lected on a Bruker APEX II area detector with graphite monochro-
ment (leff/lB, 297 K): 4.53. C32H46N2O4Co; Anal. Calc. C 66.1, H
8.0, N 4.8; found C 66.3, H 7.8, N 5.1.
4.5.3. [Co(tmhd)2(dmae)] 3
To
a
stirred purple solution of [Co(tmhd)2] (0.0637 g,
L, 0.15 mmol) was
0.15 mmol) in CH2Cl2 (10 cm3), dmae (150
l
mated Mo Ka(k = 0.71073 Å). After data collection, in each case an
empirical absorption correction (SADABS) was applied [75], and the
structures were then solved by direct methods and refined on all
F2 data using the SHELX suite of programs [76]. In all cases non-
hydrogen atoms were refined with anisotropic thermal parame-
ters; hydrogen atoms were included in calculated positions and
refined with isotropic thermal parameters which were ca.
1.2 ꢁ (aromatic CH) or 1.5 ꢁ (Me) the equivalent isotropic thermal
parameters of their parent carbon atoms. In both structures there
was disorder in some of the t-butyl groups and this was modeled
by splitting the carbon atoms into two parts. X-SEED was used as
a graphical interface with SHELX and pictures were generated using
POV-Ray [77,78].
added. The orange solution was stirred for 1 h at room temperature
then filtered through celite. The solution was left to slowly evapo-
rate at room temperature yields deep red needle crystals 0.0313 g
(54%). Magnetic moment (leff/lB, 297 K): 4.53. C26H50N2O4Co;
Anal. Calc. C 60.8, H 9.8, N 5.4; found C 52.6, H 8.9, N 5.9. Electro-
spray ionization (ESI) mass data: m/z 513 [M]+.
4.5.4. [Co(dbm)2(phen)] 4
To
a stirred red-orange solution of [Co(dbm)2] (0.0816 g,
0.15 mmol) in THF (20 cm3) at 70 °C, phen (0.0301 g, 0.15 mmol)
was added. The deep red solution was stirred for 1 h at room tem-
perature then filtered through celite. The solution was left to
slowly evaporate at room temperature yields deep red needle crys-
4.4. Electrochemistry
tals 0.0914 g (88%). Magnetic moment (leff/lB, 297 K): 5.03.
C
42H34N2O4Co; Anal. Calc. C 73.1, H 5.0, N 4.1; found C 73.4, H
Electrochemical studies were carried out using a palmsensPC Vs
2.11 potentiostat in conjunction with a three electrode cell. The
4.6, N 4.3. Electrospray ionization (ESI) mass data: m/z 686 [M]+,
462 [M ꢀ dbm]+.

300
P. Harding et al. / Polyhedron 42 (2012) 291–301
4.5.5. [Co(dbm)2(2,20-bpy)] 5
4.5.11. [Co(dbm)2(2,20-bpy)][BF4] 5+
To
a
stirred red-orange solution of [Co(dbm)2] (0.0811 g,
To a stirred orange suspension of [Co(dbm)2(H2O)2] (0.1586 g,
0.3 mmol) in CH2Cl2 (20 cm3), 2,20-bpy (0.0470 g, 0.3 mmol) was
added. The orange solution was stirred for 1 h then AgBF4
(0.0646 g, 0.33 mmol) added. The olive green solution was stirred
overnight. The dark green solution was filtered through celite then
washed with CH2Cl2 (3 ꢁ 5 cm3). The solvent was removed to dry-
ness. The solid was purified using CH2Cl2-n-hexane to give green
brown microcrystals yield 0.1097 g (51%). C40H30BCoF4N2O4; Anal.
Calc. C 64.2, H 4.0, N 3.7; found C 65.2, H 4.3, N 3.9.
0.15 mmol) in THF (20 cm3) at 70 °C, 2,20-bpy (0.0234 g,
0.15 mmol) was added. The deep red solution was stirred for 1 h
at room temperature then the solution was left to slowly evaporate
at room temperature yields deep red needle crystals 0.1014 g
(100%). Magnetic moment (leff/lB, 297 K): 4.41. C40H30N2O4Co;
Anal. Calc. C 72.6, H 4.6, N 4.2; found C 72.6, H 5.3, N 4.1. Electro-
spray ionization (ESI) mass data: m/z 661 [M]+, 438 [M ꢀ dbm]+.
4.5.6. [Co(dbm)2(dmae)] 6
4.5.12. [Co(dbm)2(dmae)][BF4] 6+
To
a
stirred red-orange solution of [Co(dbm)2] (0.0811 g,
L, 0.15 mmol)
0.15 mmol) in THF (20 cm3) at 70 °C, dmae (150
l
To a stirred orange suspension of [Co(dbm)2(H2O)2] (0.1718 g,
0.3 mmol) in CH2Cl2 (20 cm3), dmae (34.7
lL, 0.3 mmol) was
was added. The orange solution was stirred for 1 h at room temper-
ature then filtered through celite. The solution was left to slowly
evaporate at room temperature yields deep red needle crystals
added. The orange solution was stirred for 1 h then AgBF4
(0.0710 g, 0.36 mmol) added. The olive green solution was stirred
overnight. The dark green solution was filtered through celite then
washed with CH2Cl2 (3 ꢁ 5 cm3). The solvent was removed to dry-
ness. The solid was purified using CH2Cl2-n-hexane to give olive
green microcrystals yield 0.1072 g (49%). C34H34BCoF4N2O4; Anal.
Calc. C 60.0, H 5.0, N 4.1; found C 62.7, H 5.3, N 3.7.
0.0646 g (73%). Magnetic moment
(
leff/lB
,
297 K): 4.98.
C
34H34N2O4Co; Anal. Calc. C 68.8, H 5.7, N 4.7; found C 68.8, H
5.7, N 5.0. Electrospray ionization (ESI) mass data: m/z 593 [M]+,
370 [M ꢀ dbm]+.
4.5.7. [Co(tmhd)2(phen)][BF4] 1+
Acknowledgments
To a stirred purple solution of [Co(tmhd)2(H2O)2] (0.2303 g,
0.5 mmol) in CH2Cl2 (20 cm3), phen (0.0992 g, 0.5 mmol) was
added. The orange solution was stirred for 1 h then AgBF4
(0.110 g, 0.56 mmol) added. The brown-orange solution was stir-
red overnight. The dark green solution was filtered through celite
then washed with CH2Cl2 (3 ꢁ 5 cm3). The solvent was removed
to dryness. The solid was purified using CH2Cl2-diethyl ether to
give dark red block crystals yield 0.2120 g (61%). C34H46BCoF4N2O4;
Anal. Calc. C 59.0, H 6.7, N 4.0; found C 58.7, H 6.9, N 4.3.
We thank the Walailak University Research Fund (Grant No. 5/
2549) and the Thailand Research Fund (RSA5080007) for partial
financial support. Computational resources: Sila Cluster at Ramk-
hamhang University and HPC service at NECTEC. The University
Mobility in Asia and the Pacific scholarship (UMAP) is thanked
for travelling grant (to P.H.). Additional support from the Royal
Society of Chemistry in the form of a J.W.T. Jones Travelling Fellow-
ship is also acknowledged (to D.J.H.).
4.5.8. [Co(tmhd)2(2,20-bpy)][BF4] 2+
Appendix A. Supplementary data
To a stirred purple solution of [Co(tmhd)2(H2O)2] (0.2330 g,
0.5 mmol) in CH2Cl2 (20 cm3), 2,20-bpy (0.0785 g, 0.5 mmol) was
added. The orange solution was stirred for 1 h then AgBF4
(0.110 g, 0.56 mmol) added. The brown-orange solution was stir-
red overnight. The dark green solution was filtered through celite
then washed with CH2Cl2 (3 ꢁ 5 cm3). The solvent was removed
to dryness. The solid was purified using CH2Cl2-diethyl ether to
give maroon microcrystals yield 0.2138 g (64%). C32H46BCoF4N2O4;
Anal. Calc. C 57.5, H 6.9, N 4.2; found C 57.5, H 6.9, N 4.6.
Crystallographic data (excluding structure factors) has been
deposited with the Cambridge Crystallographic Data Centre (CCDC)
with CCDC 829130 for 2 and CCDC 829131 for 2+. These data can be
obtained free of charge at www.ccdc.cam.ac.uk/data_request/cif.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.poly.2012.05.037.
References
4.5.9. [Co(tmhd)2(dmae)][BF4] 3+
[1] J.-F. Létard, P. Guionneau, L. Goux-Capes, Top. Curr. Chem. 235 (2004) 221.
[2] O. Kahn, C.J. Martinez, Science 279 (1998) 44.
[3] L. Fabbrizzi, M. Licchelli, P. Pallavicini, Acc. Chem. Rev. 32 (1999) 846.
[4] M. Venturi, A. Credi, V. Balzani, Coord. Chem. Rev. 185–186 (1999) 233.
[5] H.A. Goodwin, Coord. Chem. Rev. 18 (1976) 293.
[6] (a) J.A. Real, A.B. Gaspar, V. Niel, M.C. Muñoz, Coord. Chem. Rev. 236 (2003)
121;
(b) Y. Garcia, V. Niel, M.C. Muñoz, J.A. Real, Top. Curr. Chem. 233 (2004) 229.
[7] K.S. Murray, C.J. Kepert, Top. Curr. Chem. 233 (2004) 195.
[8] A.B. Gaspar, V. Ksenofontov, J.A. Real, P. Gütlich, Chem. Phys. Lett. 373 (2003)
385.
To a stirred purple solution of [Co(tmhd)2(H2O)2] (0.2298 g,
0.5 mmol) in CH2Cl2 (20 cm3), dmae (54.4
lL, 0.5 mmol) was
added. The orange solution was stirred for 1 h then AgBF4
(0.110 g, 0.56 mmol) added. The dark green solution was stirred
overnight. The deep purple solution was filtered through celite
then washed with CH2Cl2 (3 ꢁ 5 cm3). The solvent was removed
to dryness. The solid was purified using CH2Cl2-diethyl ether to
give pastel purple microcrystals yield 0.1841 g (61%). C26H50BCoF4-
N2O4; Anal. Calc. C 52.0, H 8.4, N 4.7; found C 52.7, H 8.6, N 5.2.
[9] (a) H. Spiering, T. Kohlhaas, H. Romstedt, A. Hauser, C. Bruns-Yilmaz, P.
Gütlich, Coord. Chem. Rev. 190–192 (1999) 629;
(b) P. Gütlich, A. Hauser, H. Spiering, Angew. Chem., Int. Ed. Engl. 33 (1994)
2024.
[10] (a) J.A. Real, in: J.P. Sauvage (Ed.), Transition Metals in Supramolecular
Chemistry, John Wiley and Sons Ltd., Chichester, 1999, p. 53;
(b) J.A. Real, A.B. Gaspar, M.C. Muñoz, P. Gütlich, V. Ksenofontov, H. Spiering,
Top. Curr. Chem. 233 (2004) 167.
[11] J.G. Haasnoot, Coord. Chem. Rev. 200 (2000) 131.
[12] E. Breuning, M. Ruben, J.M. Lehn, F. Renz, Y. Garcia, V. Ksenofontov, P. Gütlich,
E. Wegelius, K. Rissanen, Angew. Chem., Int. Ed. 39 (2000) 2504.
[13] J.A. Real, E. Andres, M.C. Muñoz, M. Julve, T. Granier, A. Bousseksou, F. Varret,
Science 268 (1995) 265.
4.5.10. [Co(dbm)2(phen)][BF4] 4+
To a stirred orange suspension of [Co(dbm)2(H2O)2] (0.1678 g,
0.3 mmol) in CH2Cl2 (20 cm3), phen (0.0620 g, 0.3 mmol) was
added. The orange solution was stirred for 1 h then AgBF4
(0.0700 g, 0.36 mmol) added. The olive green solution was stirred
overnight. The dark green solution was filtered through celite then
washed with CH2Cl2 (3 ꢁ 5 cm3). The solvent was removed to dry-
ness. The solid was purified using CH2Cl2-n-hexane to give olive
green microcrystals yield 0.1295 g (56%). C42H34BCoF4N2O4; Anal.
Calc. C 65.0, H 4.4, N 3.6; found C 66.5, H 4.2, N 3.9.
[14] G.J. Halder, C.J. Kepert, B. Moubaraki, K.S. Murray, J.D. Cashion, Science 298
(2002) 1762.
[15] A.M. Greenaway, E. Sinn, J. Am. Chem. Soc. 100 (1978) 8080.

P. Harding et al. / Polyhedron 42 (2012) 291–301
301
[16] A.W. Addison, S. Burman, C.G. Wahlgren, O.A. Rajan, T.M. Rowe, E. Sinn, J.
Chem. Soc., Dalton Trans. (1987) 2621.
[56] A. Wheeler, C. Brouty, P. Spinat, P. Herpin, Acta Crystallogr., Sect.B Struct.
Crystallogr. Cryst. Chem. 831 (1975) 2069.
[17] D. Collison, C.D. Garner, C.M. McGrath, J.F.W. Mosselmans, M.D. Roper, J.M.W.
Seddon, E. Sinn, N.A. Young, J. Chem. Soc., Dalton Trans. (1997) 4371.
[18] E.J. MacLean, C.M. McGrath, C.J. O’Connor, C. Sangregorio, J.W.M. Seddon, E.
Sinn, F.E. Sowery, S.J. Teat, G.B.M. Vaughan, N.A. Young, Chem. Eur. J. 9 (2003)
5314.
[57] P. Guionneau, M. Marchivie, G. Bravic, J.-F. Létard, D. Chasseau, J. Mater. Chem.
12 (2002) 2546.
[58] M. Marchivie, P. Guionneau, J.-F. Létard, D. Chasseau, Acta Crystallogr., Sect. B
61 (2005) 25.
[59] J.W. Turner, F.A. Schultz, Coord. Chem. Rev. 219–221 (2001) 81.
[19] K. Madeja, E. König, J. Inorg. Nucl. Chem. 25 (1963) 377.
[20] R. Sieber, S. Decurtins, H. Stoeckli, C. Evans, D. Wilson, J.A.K. Yufit, S.C. Howard,
A. Capelli, A. Hauser, Chem. Eur. J. 6 (2000) 361.
[21] (a) M. Zerara, A. Hauser, Chem. Phys. Chem. 5 (2004) 395;
(b) A. Hauser, N. Amstutz, S. Delahaye, S. Schenker, A. Sadki, R. Sieber, M.
Zerara, Struct. Bond. 106 (2004) 81.
[22] (a) R. Hogg, R.G. Wilkins, J. Chem. Soc. (1962) 341;
(b) J.S. Judge, W.A. Baker, Inorg. Chim. Acta 1 (1967) 68.
[23] B.N. Figgis, E.S. Kucharski, A.H. White, Aust. J. Chem. 36 (1983) 1537.
[24] W.R. Scheidt, C.A. Reed, Chem. Rev. 81 (1981) 543.
[25] I. Bertani, H.B. Gray, S.J. Lippard, J.S. Valentine, Bioinorganic Chemistry,
University Science Books, Sausalito, CA, 1994.
[60] (a) For examples of other Co complexes exhibiting RCSCO see E. Buhks, M.
Dixon, J. Jortner, G. Navon, Inorg. Chem. 18 (1979) 2014;
(b) J.F. Endicott, G.R. Brubaker, T. Ramasami, K. Kumar, K. Dwarakanath, J.
Cassel, D. Johnson, Inorg. Chem. 22 (1983) 3754;
(c) A. Hammershøi, D.A. Geselowitz, H. Taube, Inorg. Chem. 32 (1984) 979;
(d) P.V. Bernhardt, L.A. Jones, P.C. Sharpe, Inorg. Chem. 36 (1997) 2420.
[61] F.B. Johansson, A.D. Bond, U.G. Nielson, B. Moubaraki, K.S. Murray, K.J. Berry,
J.A. Larrabee, C.J. McKenzie, Inorg. Chem. 47 (2008) 5079.
[62] D.C.L. De Alwis, F.A. Schultz, Inorg. Chem. 42 (2003) 3616.
[63] K. Wieghardt, W. Schmidt, W. Herrmann, H.J. Kueppers, Inorg. Chem. 22 (1983)
2953.
[64] H.J. Kueppers, A. Neves, C. Pomp, D. Ventur, K. Wieghardt, B. Nuber, J. Weiss,
Inorg. Chem. 25 (1986) 2400.
[26] S.J. Lippard, J.M. Berg, Principles of Bioinorganic Chemistry, University Science
Books, Mill Balley, CA, 1995.
[65] T.C. Richards, W.E. Geiger, J. Am. Chem. Soc. 116 (1994) 2028.
[66] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox,
H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E.
Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y.
Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J.
Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03, Revision D.01, Gaussian Inc., Wallingford, CT, 2004.
[67] K.P. Jensen, J. Cirera, J. Phys. Chem. A 113 (2009) 10033.
[68] M. Reihner, Inorg. Chem. 41 (2002) 6928.
[27] W. Kaim, B. Schwederski, Bioinorganic Chemistry: Inorganic Elements in the
Chemistry of Life, Wiley, New York, 1995.
[28] S.C. Sligar, Biochemistry 15 (1976) 5399.
[29] E.J. Mueller, P.J. Lioda, S.C. Sligar, in: P.R. Ortiz de Montellano (Ed.),
Cytochrome P450: Structure, second ed., Mechanism and Biochemistry,
Plenum Press, New York, 1995 (Chap. 3).
[30] D.C. Rees, M.K. Chan, J. Kim, Adv. Inorg. Chem. 40 (1993) 89.
[31] B.K. Burgess, D.J. Lowe, Chem. Rev. 96 (1996) 2983.
[32] H. Lopes, G.W. Pettigrew, I. Moura, J.J.G. Moura, J. Biol. Inorg. Chem. 3 (1998)
632.
[33] P. Hendry, A. Ludi, Adv. Inorg. Chem. 35 (1990) 117.
[34] H.C. Stynes, J.A. Ibers, Inorg. Chem. 10 (1971) 2304.
[35] E. Buhks, M. Bixon, J. Jortner, G. Navon, Inorg. Chem. 18 (1979) 2014.
[36] D.A. Geselowitz, H. Taube, Adv. Inorg. Bioinorg. Mech. 1 (1982) 391.
[37] S. Larsson, K. Ståhl, M.C. Zerner, Inorg. Chem. 25 (1986) 3033.
[38] D.A. Geselowitz, Inorg. Chim. Acta 154 (1988) 225.
[39] M.D. Newton, J. Phys. Chem. 95 (1991) 30.
[40] R.D. Shalders, T.W. Swaddle, Inorg. Chem. 34 (1995) 4815.
[41] T.W. Swaddle, Can. J. Chem. 74 (1996) 631.
[42] J.W. Turner, F.A. Schultz, Inorg. Chem. 38 (1999) 358.
[43] J.W. Turner, F.A. Schultz, J. Phys. Chem. B 2002 (2009) 106.
[44] R.L. Lord, F.A. Schultz, M.-H. Baik, J. Am. Chem. Soc. 131 (2009) 6189.
[45] K. Nakamoto, In Infrared and Raman Spectra of Inorganic and Coordination
Compounds: Part B, sixth ed., Wiley, New York, 2009. pp. 96.
[46] P. Harding, D.J. Harding, W. Phonsri, S. Saithong, H. Phetmung, Inorg. Chim.
Acta 362 (2009) 78.
[47] F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochmann, in: Advanced Inorganic
Chemistry, sixth ed., John Wiley & Sons, New York, 1999, p. 817.
[48] G. Bandoli, D. Barreca, A. Gasparotto, C. Maccato, R. Seraglia, E. Tondello, A.
Devi, R.A. Fischer, M. Winter, Inorg. Chem. 48 (2009) 82.
[49] N. Nagashima, S. Kudoh, M. Nakata, Chem. Phys. Lett. 374 (2003) 59.
[50] D.S. Sankhla, R.C. Mathur, S.N. Misra, J. Inorg. Nucl. Chem. 42 (1980) 489.
[51] P. Chassot, F. Emmenegger, Inorg. Chem. 35 (1996) 5931.
[52] R.D. Archer, B.P. Cotsoradis, Inorg. Chem. 4 (1965) 1584.
[53] R.J. York, W.D. Bonds Jr., B.P. Cotsoradis, R.A. Archer, Inorg. Chem. 8 (1969) 789.
[54] F. Izumi, R. Kurosawa, H. Kawamoto, H. Akaiwa, Bull. Chem. Soc. Jpn. 48 (1975)
3188.
[69] J.P. Perdew, J. Tao, V.N. Staroverov, G.E. Scuseria, J. Chem. Phys. 120 (2004)
6898.
[70] T.F. Hughes, R.A. Friesner, J. Chem. Theory Comput. 7 (2011) 19.
[71] D. Barreca, C. Massignan, S. Daolio, M. Fabrizio, C. Piccirillo, L. Armelao, E.
Tondello, J. Chem. Mater. 13 (2001) 588.
[72] L.A. Barrios, J. Ribas, G. Aromí, J. Ribas-Ariño, P. Gamez, O. Roubeau, S.J. Teat,
Inorg. Chem. 46 (2007) 7154.
´
[73] E. Meštrovic, B. Kaitner, J. Chem. Crystallogr. 36 (2006) 599.
[74] D.F. Evans, J. Chem. Soc. (1959) 2003.
[75] Bruker APEX, SAINT and SADABS 2008, Bruker AXS Inc., Madison, Wisconsin, USA.
[76] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[77] L.J. Barbour, J. Supramol. Chem. 1 (2001) 189.
[78] J.A. Atwood, L.J. Barbour, Cryst. Growth Des. 3 (2003) 3.
[79] M.J. Shaw, R.L. Henson, S.E. Houk, J.W. Westhoff, G.B. Richter-Addo, M.W.
Jones, J. Electroanal. Chem. 534 (2002) 47.
[80] A.D. Becke, J. Chem. Phys. 98 (1993) 1372.
[81] P.J. Stephens, F.J. Devlin, M.J. Frisch, C.F. Chabalowski, J. Phys. Chem. 98 (1994)
11623.
[82] T.H. Dunning Jr., P.J. Hay, in: H.F. Schaefer III (Ed.), Modern Theoretical
Chemistry, vol. 3, Plenum, New York, 1976, p. 1.
[83] M. Dolg, U. Wedig, H. Stoll, H. Preuss, J. Chem. Phys. 86 (1987) 866.
[84] http://avogadro.openmolecules.net/ (accessed 12/22/2009).
[55] B.S. Brunschwig, C. Creutz, D.H. Macartney, T.K. Sham, N. Sutin, Faraday
Discuss. Chem. Soc. 74 (1982) 113.