Spin-Crossover Films
DOI: 10.1002/anie.201307968
Iron(II) Spin-Crossover Complexes in Ultrathin Films: Electronic
Structure and Spin-State Switching by Visible and Vacuum-UV Light**
E. Ludwig, H. Naggert, M. Kallne, S. Rohlf, E. Krçger, A. Bannwarth, A. Quer, K. Rossnagel,
L. Kipp,* and F. Tuczek*
Abstract: The electronic structure of the iron(II) spin cross-
over complex [Fe(H
2
bpz)
2
(phen)] deposited as an ultrathin
film on Au(111) is determined by means of UV-photoelectron
spectroscopy (UPS) in the high-spin and in the low-spin state.
This also allows monitoring the thermal as well as photo-
induced spin transition in this system. Moreover, the complex is
excited to the metastable high-spin state by irradiation with
vacuum-UV light. Relaxation rates after photoexcitation are
determined as a function of temperature. They exhibit a tran-
sition from thermally activated to tunneling behavior and are
two orders of magnitude higher than in the bulk material.
Miniaturization of electronic devices is an important
driving force towards reducing functional elements to the
size of single molecules. A model system of functional
molecules with potential applications in nano-electronics
and spintronics, the electrical, magnetic, structural, or optical
properties of which can be manipulated by external stimuli,
are spin crossover (SCO) complexes, the largest number of
examples being available for Fe
II
. Depending on the nature
and field strength of the surrounding ligands the central iron
ion can exist in two different electronic configurations, “low
spin” (LS, S = 0) and “high spin” (HS, S = 2).
[1–4]
Herein, the Fe
II
complex [Fe(H
2
bpz)
2
(phen)] is investi-
gated (compound 1;H
2
bpz = 1,2-bis(pyrazolyl)borate, phen =
1,10’-phenanthroline; see Figure 1, top). This compound
undergoes a thermal spin transition between HS and LS at
T
1/2
164 K (Figure 1, bottom). Below T
LIESST
= 43 K, the
complex can be switched to a metastable HS (mHS) state by
irradiation with 532 nm light.
[5]
This effect is known as light-
induced excited spin-state trapping (LIESST).
[6–8]
The life-
time of the metastable HS state is very long at temperatures
well below T
LIESST
but strongly decreases with increasing
temperature, eventually leading to relaxation to the LS
ground state at T = T
LIESST
(Figure 1, bottom).
[2]
In bulk materials, the LIESST effect has been well
investigated.
[9,10]
Successful vacuum deposition of 1 and the
related complex [Fe(H
2
bpz)
2
(bipy)] (2; bipy = 2,2’-bipyri-
dine)
[11,12]
has opened up the possibility to study this effect
in solvent-free high-quality films, using methods that require
ultra-high vacuum conditions, such as ultraviolet
[13]
and X-ray
photoelectron or inverse photoemission spectroscopy.
[14]
In
recent studies, thin films of 1 and 2 have also been
investigated with scanning tunneling microscopy
[15–17]
as well
as X-ray absorption spectroscopy,
[18–20]
and evidence for
electron-induced spin-state switching has been provided for
1.
[15,16]
Detailed studies of the valence electronic structure of
1 and 2 are still missing, however. The same applies to the
dynamics of the photoexcited spin state exhibited by these
systems in ultrathin films. A recent study has shown that the
spin transition temperature and the relaxation dynamics of
the metastable HS state of Fe
II
can vary greatly from
crystalline to nanocrystalline and amorphous samples.
[21]
Figure 1. Representation of 1 and Fe
II
high-spin fraction versus temper-
ature with LIESST effect; the configuration of the Fe d electrons is
indicated.
[*] E. Ludwig, Dr. M. Kallne, S. Rohlf, E. Krçger, A. Quer,
Dr. K. Rossnagel, Prof. Dr. L. Kipp
Institut fr Experimentelle und Angewandte Physik
Christian-Albrechts-Universitt zu Kiel
24098 Kiel (Germany)
E-mail: kipp@physik.uni-kiel.de
Homepage: www.ieap.uni-kiel.de/surface/ag-kipp
H. Naggert, Dr. A. Bannwarth, Prof. Dr. F. Tuczek
Institut fr Anorganische Chemie
Christian-Albrechts-Universitt zu Kiel
24098 Kiel (Germany)
E-mail: ftuczek@ac.uni-kiel.de
Homepage: www.ac.uni-kiel.de/tuczek
[**] T. Riedel and the Petra III staff at DESY are thanked for technical
assistance as well as Dr. L. Carella and Prof. Dr. E. Rentschler
(Mainz) for bulk magnetic measurements under light irradiation.
This work was supported by DFG via SFB 677 and BMBF via project
No. 05K10FK1.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201307968.
A
ngewandte
Chemi
e
3019Angew. Chem. Int. Ed. 2014, 53, 3019 –3023 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Herein we present an ultraviolet photoelectron spectros-
copy (UPS) study of the electronic structure and the light-
induced switching behavior of 1 in vacuum-deposited ultra-
thin films. The dynamic population of the metastable HS state
is investigated under permanent irradiation for different laser
intensities to determine mHS-to-LS relaxation rates at
temperatures below, around, and well above T
LIESST
(1) =
43 K.
The electronic structure of a thin layer of 1 determined by
UPS, and a comparison to spectra obtained from DFT
calculations, is shown in Figure 2 (for experimental details,
see the Supporting Information). The upper part of Figure 2
shows the valence electronic structure of an about 7 nm-thick
layer of 1 on Au(111), corresponding to 5–6 molecular layers,
if a size of 1.4 nm 0.76 nm of the individual molecule is
assumed.
[16]
The spectra are composed of several broad
features at binding energies of 2.4 eV, 4.8 eV, 6.8 eV, and
a very broad feature centered at about 9.6 eV. Comparison of
spectra at near room temperature, at 56 K, and at 28 K shows
no significant changes of the overall electronic structure.
Additional irradiation with 532 nm at 28 K seems to have
little influence as well. Detailed investigation of the feature at
a binding energy of about 2.4 eV, however, reveals an
intensity dependence on temperature as well as on 532 nm
irradiation (shown in the inset of Figure 2; see below). For
better visibility and to account for sample charging, the
energy axis of the spectra has been calibrated to the binding
energy of the feature located at 4.8 eV.
The bottom part of Figure 2 shows the spectra derived
from the results of the DFT calculations at the BP86/tzvp
level.
[15,22]
The calculated ionization energies are indicated as
vertical lines below the respective spectra. The theoretical
spectra were obtained by assigning a Gaussian profile of equal
intensity and width to each transition. Polarization- and
matrix-element related effects were not taken into account.
The overall shape of the calculated spectra matches the
photoemission very well. A rough correlation between
experimental and theoretical spectral features is indicated
by dark gray lines.
A clear difference of the theoretical spectra between the
two spin states can be observed at ionization energies of about
4.5 eV. In the LS state, the three doubly occupied t
2g
orbitals
give rise to a distinct feature at 4.5–5 eV, which will be
referred to as a “low-spin feature”. Owing to the spin
polarization effect, the d orbitals are distributed over
a much larger energy range in the HS state, and in the
spectral region of the low-spin feature much less intensity is
predicted. Experimentally, cooling of the sample from 275 K
to 56 K is accompanied by an intensity increase of the LS
feature (Figure 2 inset). Further cooling to 28 K, however,
leads to an intensity decrease, and irradiation with 532 nm
light at 28 K further decreases the intensity of this feature.
A more detailed investigation of these phenomena is
shown in Figure 3, where the intensity of the LS feature is
plotted versus the sample temperature during sample cooling
and additional irradiation of the sample with 532 nm at 28 K.
A fit function consisting of an exponentially decaying back-
ground and a Gaussian profile is used to determine the
temperature dependence of the feature intensity. The inset in
Figure 3 illustrates the Gaussian fit component as a shaded
area.
Figure 2. Top: Valence band structure of a circa 7 nm-thick layer of
1 on Au(111) at 275 K (filled gray area), 56 K (black line), 28 K (blue
line), and 28 K with additional irradiation with 532 nm (green line).
Photon energy for photoemission was 21.22 eV. The inset emphasizes
the feature at a binding energy of 2.4 eV. Bottom: Vertical bars
illustrate the ionization energies results of molecular orbitals as
obtained from the DFT calculations for the LS and HS case. Spectra
composed from the DFT results are shown as blue (LS) and red lines
(HS). Dark gray lines serve as guides to the eye to roughly assign
features of the theoretical data to the measured valence band
structure.
Figure 3. Evolution of the low-spin feature intensity during sample
cooling, as obtained from a Gaussian fit function. The Gaussian fit
component is illustrated in the inset for spectra recorded at 52 K und
at 28 K with illumination as a shaded area. The gray shaded area
indicates the error margins obtained from the fit procedure. Dark gray
lines illustrate the overall intensity evolution. Intensity variations due
to the VUV-induced and LIESST effects are indicated by arrows. The
shaded area on the right indicates measurements under irradiation
with 532 nm.
.
Angewandte
Communications
3020
www.angewandte.org 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2014, 53, 3019 –3023
Upon cooling the sample, we first observe an increase of
the LS feature starting at about 180 K down to about 60 K.
This behavior reflects a thermal spin transition from the HS to
the LS state. Further lowering of the temperature then leads
to an intensity decrease to about room temperature level with
a transition temperature of about 50 K, indicating a popula-
tion of the mHS state. Presumably, this transition is induced
by the vacuum-UV (VUV) irradiation of the sample, similar
to the optical excitation of 1 to the mHS state, which occurs at
about the same temperature.
[5]
The VUV-induced process,
however, does not lead to full conversion from the LS to the
HS state, as additional irradiation of the sample at 28 K with
the 532 nm laser causes a further intensity decrease of the LS
feature, as indicated by the dark gray area in Figure 3. We
assume that now all molecules with active centers are in the
HS state (low-spin fraction g
LS
= 0). Importantly, the thermal
spin transition in our film of 1 appears to occur at a much
lower temperature than in the bulk material (165 K),
[5,9]
and
the spin crossover behavior is much more gradual. This seems
to agree with the rule that the spin transition generally is less
steep and occurs at lower temperatures for nanocrystalline or
amorphous samples as compared to the crystalline bulk
material.
[21]
The observation that the intensity of the LS feature at
28 K is lowered by illumination of the sample with laser light
prompted us to explore the influence of a continuous laser
irradiation at 532 nm on the spin equilibrium in ultrathin films
of 1 in more detail. The results of these measurements are
shown in Figure 4 where the low-spin fraction g
LS
observed
under irradiation of the sample with different laser intensities
is plotted versus the temperature. The laser was also switched
off for short periods at 28 K. A representative spin transition
curve without laser irradiation is given as a reference (gray
filled trace in Figure 4). The intensity of the LS feature was
determined as shown in Figure 3; for the determination of the
LS fraction, the residual intensity of the LS feature in the high
spin state was accounted for (see the Supporting Informa-
tion). The temperatures at which a stationary population of
the mHS state of 50% occurs (g
LS
= 0.5) are defined as
steady-state spin transition temperatures T
1/2
SS
(SS = steady
state). Without any additional irradiation a value of T
1/2
SS
=
37 K is observed. Importantly, permanent irradiation of the
sample with 532 nm light shifts T
1/2
SS
to higher temperatures,
depending on the intensity of the irradiation. Averaging spin
transition temperatures obtained for several measurements
with the same laser power density results in 69 K for
7.7 mWcm
2
, 78 K for 19.1 mWcm
2
, and 99 K for
153.1 mWcm
2
(Table 1). A plot of the steady-state transition
temperature T
1/2
SS
versus the laser power indicates a logarith-
mic dependence (Supporting Information, Figure S1). Note
that the signal-to-noise ratio decreases for low laser power
densities. When switching off the laser at low temperatures,
the intensity of the low-spin feature relaxes to about room
temperature level within the recording time of one photo-
electron spectrum (about 130 s). Resuming the laser irradi-
ation leads to an intensity decrease again, as observed before.
The shift of the spin-transition temperatures under
permanent laser irradiation can be explained by a steady-
state population of the metastable state above T
LIESST
owing to
a high photon flux. A light-induced shift of T
LIESST
towards
higher temperature has previously been observed for crystal-
line Fe
II
spin crossover complexes in conjunction with a light-
induced thermal hysteresis (LITH).
[25,26]
Susceptibility meas-
urements under permanent irradiation show that this effect
exists for bulk samples of 1 as well (Supporting Information,
Figure S2). In the thin film of 1, however, no hysteresis is
observed, as curves of the LS feature intensity versus T
obtained upon heating the sample are similar to those
determined in the cooling mode (Supporting Information,
Figure S3).
In Table 1, excitation rates k
ex
calculated from the rate of
incident photons are also shown. These values are calculated
from a cross section of 1.37 10
17
cm
2
for excitation with
532 nm light
[12]
and a quantum efficiency of 1.0 for the light-
induced population of the metastable HS state (see the
Supporting Information).
[27]
The steady-state high-spin frac-
tion under continuous irradiation as a function of k
ex
and the
HS!LS relaxation rate k
HL
is given by Equation (1):
[21]
g
SS
HS
¼ 1 g
SS
LS
¼
k
ex
k
ex
þ k
HL
ð1Þ
Therefore k
HL
k
ex
at T
1/2
SS
, where g
HS
SS
= g
LS
SS
0.5. The
LS!HS excitation rates k
ex
of Table 1 thus can be identified
with the HS!LS relaxation rates k
HL
. A plot of the relaxation
rates k
HL
determined this way versus 1/T is shown in Figure 5.
Figure 4. Low-spin fraction versus temperature as a function of
permanent irradiation with varying intensities of 532 nm light. Steady-
state spin transition temperatures T
1/2
SS
are indicated by dotted vertical
arrows. Gray shaded areas illustrate measurements without additional
irradiation.
Table 1: Steady-state spin transition temperatures and light-induced
SCO excitation rates calculated from the laser intensities.
Spin transition temperature
T
1/2
SS
[K]
Laser intensity
[mWcm
2
]
Excitation rate
k
ex
[s
1
]
37 0 (VUV exc. only) 5.810
3
69 7.7 0.3
78 19.1 0.7
99 153.1 5.6
A
ngewandte
Chemi
e
3021Angew. Chem. Int. Ed. 2014, 53, 3019 –3023 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org
For comparison, relaxation rates derived from optical meas-
urements of a 480 nm-thick film of 1 (Supporting Informa-
tion, Figure S5) are also given. Importantly, the decay curves
obtained for these films after excitation with 532 nm light
exclusively exhibit mono-exponential behavior (Supporting
Information, Figure S6), in contrast to those recorded for the
bulk material of 1, which reflect the presence of cooperative
interactions.
[5]
As evident from Figure 5, the high-temperature relaxa-
tion rates derived from UPS on ultrathin films of 1 (filled
circles) can be described with an Arrhenius behavior, and
extrapolation to low temperatures leads to the values
determined for the 480 nm thick film by UV/Vis absorption
spectroscopy (k
HL
10
3
–10
4
s
1
; open circles). A fit curve
simulating a transition from thermally activated to tunneling
behavior is given by a gray line, leading to a low-temperature
limit for k
HL
in vacuum-deposited films of 1 on the order of
about 10
4
s
1
. Notably, this value is about one order of
magnitude higher than the low-temperature limit of k
HL
in the
crystalline bulk material of 1 (ca. 10
5
s
1
).
[5]
On the other
hand, the relaxation rate in the thin film of 1 at 28 K must be
significantly higher than the thick-film value of 10
4
s
1
,as
after switching off irradiation with 532 nm light (having
converted the LS to the HS state to 100%) the LS fraction is
found to return to a level similar to the room temperature
value (ca. 0.25) within the acquisition time of one photo-
electron spectrum (130 s; Figure 4). Based on a value of
g
HS
SS
= 0.75 and the corresponding relaxation kinetics (see the
Supporting Information), we conclude that k
HL
= 1.9
10
3
s
1
and k
ex
= 5.8 10
3
s
1
. The given k
HL
value can be
considered as the lower limit of relaxation rates attainable to
the ultrathin films of 1 under our experimental conditions
(that is, in the presence of continuous VUV irradiation), one
and two orders of magnitude higher than in the thick film and
in the crystalline sample of 1,
[5]
respectively. The value of k
ex
=
5.8 10
3
s
1
corresponds to excitation from the LS to the HS
state owing to VUV irradiation. This value of k
ex
equals k
HL
at
T
1/2
SS
= 37 K, where a dynamic spin equilibrium with g
HS
SS
=
0.5 under continuous VUV irradiation is observed (see
Figure 4 “dark sample” and Table 1). The apparently close
agreement between this spin-transition temperature and
T
LIESST
= 43 K determined for 1 after optical excitation in
the bulk (see below) is therefore accidental. As a matter of
fact, the excitation mechanisms and the relaxation times are
quite different for these two cases.
To conclude, ultraviolet photoelectron spectroscopy has
been shown to be a suitable method to investigate the
electronic structure as well as the thermal and light-induced
spin-crossover behavior of 1 in ultrathin films on Au(111).
Our results confirm that spin switching of Fe
II
complexes is
possible in such systems, but with significantly larger relax-
ation rates than in the bulk. The photoelectron spectra of
a circa 7 nm-thick film of 1 can be reproduced with DFT
calculations of the complex in the gas phase, indicating that
the molecules deposited on the Au(111) surface are intact.
Evaluation of the intensity of the LS feature in the photo-
electron spectra yields transition temperatures T
1/2
for the
thermal HS–LS spin crossover that are significantly lower
than in the bulk (164 K). For the first time, quantitative light-
induced conversion of the low-spin to the metastable high-
spin state has been achieved for a Fe
II
complex deposited as
a thin layer on gold. Furthermore, mHS!LS relaxation rates
have been determined for such a system. Without additional
laser irradiation, a VUV-induced excited spin state trapping
(VUVIESST) effect is observed; the corresponding steady-
state spin transition temperature T
1/2
SS
is determined to 37 K.
Permanent irradiation with a 532 nm laser shifts T
1/2
SS
toward
higher temperatures, which is explained by a competition
between (temperature-dependent) HS!LS relaxation and
(power-dependent) LS!HS excitation through optical irra-
diation. The temperature dependence of k
HL
shows a transi-
tion from Arrhenius behavior to tunneling. A low-temper-
ature limit for the relaxation rates in the thin films systems
seems to be a value of about 10
3
s
1
, which is one to two
orders of magnitude higher than in thick films or the bulk.
Population of the metastable high-spin state in the
absence of laser illumination is induced by irradiation of the
sample with VUV photons of 21.22 eV energy. Similar to the
spin-state-trapping effects by exposure to hard and soft X-ray
radiation,
[24,25]
secondary electrons created during the photo-
emission process can transfer energy to the Fe complex by
inelastic scattering processes, leading to population of the
metastable high spin state. In our case, however, this effect is
much weaker, because the photon flux of a laboratory VUV
source is several orders of magnitude lower in comparison to
synchrotron sources.
Received: September 10, 2013
Revised: November 20, 2013
Published online: February 12, 2014
.
Keywords: molecular switches · spin crossover · thin films ·
ultraviolet photoelectron spectroscopy
Figure 5. Arrhenius plot of the HS-to-LS relaxation constants obtained
from measurements under permanent irradiation. Filled circles show
data obtained from UPS measurements, open circles those determined
with UV/Vis absorption spectroscopy in a 480 nm film. The full gray
line shows a fit resulting from superposition of a tunneling process
and an activated process with E
A
= 469 cm
1
and A= 4.710
3
s
1
). The
dashed gray curve serves as guide to the eye. Dotted gray lines
emphasize the minimum relaxation rate for the thin film system under
VUV irradiation (1.9 10
3
s
1
) and the 480 nm film (ca. 10
4
s
1
).
.
Angewandte
Communications
3022
www.angewandte.org 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2014, 53, 3019 –3023
[1] P. Gtlich, H. A. Goodwin, Spin Crossover in Transition-Metal
Compounds, Vol. I – III, Springer, Berlin-Heidelberg-New York,
2004.
[2] P. Gtlich, A. Hauser, H. Spiering, Angew. Chem. 1994, 106,
2109 – 2141; Angew. Chem. Int. Ed. Engl. 1994, 33, 2024 – 2054.
[3] O. Kahn, C. J. Martinez, Science 1998, 279, 44 – 48.
[4] T. Mahfoud, G. Molnr, S. Cobo, L. Salmon, C. Thibault, C.
Vieu, P. Demont, A. Bousseksou, Appl. Phys. Lett. 2011, 99,
053307.
[5] N. Moliner, L. Salmon, L. Capes, M. C. MuÇoz, J.-F. Ltard, A.
Bousseksou, J.-P. Tuchagues, J. J. McGarvey, A. C. Dennis, M.
Castro, R. Burriel, J. A. Real, J. Phys. Chem. B 2002, 106, 4276 –
4283.
[6] S. Decurtins, P. Gtlich, C. P. Kçhler, H. Spiering, A. Hauser,
Chem. Phys. Lett. 1984, 105,1–4.
[7] S. Decurtins, P. Gtlich, K. M. Hasselbach, A. Hauser, H.
Spiering, Inorg. Chem. 1985, 24, 2174 – 2178.
[8] A. Hauser, Chem. Phys. Lett. 1986, 124, 543 – 548.
[9] J. A. Real, A. B. Gaspar, M. C. MuÇoz, Dalton Trans. 2005,
2062 – 2079.
[10] J.-F. Ltard, J. Mater. Chem. 2006, 16, 2550– 2559.
[11] S. Shi, G. Schmerber, J. Arabski, J.-B. Beaufrand, D. J. Kim, S.
Boukari, M. Bowen, N. T. Kemp, N. Viart, G. Rogez, E.
Beaurepaire, H. Aubriet, J. Petersen, C. Becker, D. Ruch,
Appl. Phys. Lett. 2009, 95, 043303.
[12] H. Naggert, A. Bannwarth, S. Chemnitz, T. von Hofe, E. Quandt,
F. Tuczek, Dalton Trans. 2011, 40, 6364 – 6366.
[13] M. Kamada, K. Takahashi, Y. Doi, K. Fukui, T. Tayagaki, K.
Tanaka, Phase Transitions 2002, 75, 847 – 853.
[14] X. Zhang, T. Palamarciuc, P. Rosa, J.-F. Ltard, B. Doudin, Z.
Zhang, J. Wang, P. A. Dowben, J. Phys. Chem. C 2012, 116,
23291 – 23296.
[15] T. G. Gopakumar, F. Matino, H. Naggert, A. Bannwarth, F.
Tuczek, R. Berndt, Angew. Chem. 2012, 124, 6367 – 6371; Angew.
Chem. Int. Ed. 2012, 51, 6262 – 6266.
[16] T. G. Gopakumar, F. Matino, H. Naggert, A. Bannwarth, F.
Tuczek, R. Berndt, Angew. Chem. 2013, 125, 3884; Angew.
Chem. Int. Ed. 2013, 52, 3796.
[17] T. Miyamachi, M. Gruber, V. Davesne, M. Bowen, S. Boukari, L.
Joly, F. Scheurer, G. Rogez, T. K. Yamada, O. Ohresser, E.
Beaurepaire, W. Wulfhekel, Nat. Commun. 2012, 3, 938.
[18] B. Warner, J. C. Oberg, T. G. Gill, F. El Hallak, C. F. Hirjibehe-
din, M. Serri, S. Heutz, M.-A. Arrio, P. Sainctavit, M. Mannini,
G. Poneti, R. Sessoli, P. Rosa, J. Phys. Chem. Lett. 2013, 4, 1546 –
1552.
[19] J.-J. Lee, H.-S. Sheu, C.-R. Lee, J.-M. Chen, J.-F. Lee, C.-C. Wang,
C.-H. Huang, Y. Wang, J. Am. Chem. Soc. 2000, 122, 5742 – 5747.
[20] M. Bernien, D. Wiedemann, C. F. Hermanns, A. Krger, D. Rolf,
W. Kroener, P. Mller, A. Grohmann, W. Kuch, J. Phys. Chem.
Lett. 2012, 3, 3431– 3434.
[21] P. Chakraborty, M.-L. Boillot, A. Tissot, A. Hauser, Angew.
Chem. 2013, 125, 7279 – 7282; Angew. Chem. Int. Ed. 2013, 52,
7139 – 7142.
[22] In Ref. [16], the functional/basis set has erroneously been given
as tppsh/tzvp.
[23] 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 – 4376.
[24] G. Vank, F. Renz, G. Molr, T. Neisius, S. Krpti, Angew.
Chem. 2007, 119, 5400 – 5403; Angew. Chem. Int. Ed. 2007, 46,
5306 – 5309.
[25] A. Desaix, O. Roubeau, J. Jeftic, J. G. Haasnoot, K. Boukhed-
daden, E. Codjovi, J. Linars, M. Nogus, F. Varret, Eur. Phys. J.
B 1998, 6, 183 – 193.
[26] J.-F. Ltard, P. Guionneau, L. Rabardel, J. A. K. Howard, A. E.
Goeta, D. Chasseau, O. Kahn, Inorg. Chem. 1998, 37, 4432 –
4441.
[27] A. Hauser, Top. Curr. Chem. 2004, 234, 155– 198.
A
ngewandte
Chemi
e
3023Angew. Chem. Int. Ed. 2014, 53, 3019 –3023 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org
