Photophysics and Photochemistry of Transition Metal
Compounds
Département de chimie physique,
Sciences II,
Université de Genève
30, Quai Ernest-Ansermet,
CH-1211
Genève 4,
Switzerland
Tel: +41-22-37-96547, Fax: +41-22-37-96103
Research Activities:
Photophysical and photochemical properties of transition metal compounds are increasingly being made use of in advanced technological applications. It is thus of more than just academic interest to fully understand the fundamental photophysical and photochemical processes, such as laser-induced luminescence, intersystem crossing, internal conversion, excitation energy transfer, light-induced electron transfer etc, and the parameters which govern their rates and quantum efficiencies.
Our research interests are focussed on establishing relationships between structural, electronic and energetic parameters, and the dynamics of elementary radiationless processes at a molecular level. Spin-crossover compounds and three-dimensional metal-tris-oxalate networks serve as model systems for their investigation.
The experimental methods we use are optical spectroscopy in condensed media. This includes straightforward polarised single crystal absorption and luminescence techniques as well as time-resolved methods and advanced high resolution laser spectroscopy at cryogenic temperatures.
Spin-crossover compounds
Spin-crossover compounds are transition metal compounds having a d4 - d7 electron configuration, which show a thermal transition from a low-spin state at cryogenic temperatures to a high-spin state at elevated temperatures. During such a transition the physical properties (magnetic, optical, structural) change dramatically. The discovery of a light-induced spin transition in this class of compounds, during which the high-spin state can be populated as metastable state below the thermal transition temperature, initiated a research project on the photophysical properties of spin-crossover compounds [1]. Even though Fe(II) compounds show no luminescence, it became possible to identify a large number of intersystem crossing processes (as depicted in Figure 1) [2].
The high-spin low-spin relaxation is understood quantitatively as nonadiabatic multiphonon process, showing a thermally activated behaviour at elevated temperatures and a non-vanishing tunnelling rate at cryogenic temperatures. The tunnelling rate constant depends exponentially upon the zero-point energy difference between the high-spin and the low-spin state. It varies from 10-6 s-1 (t = 10 days) for compounds with small values of the zero-point energy difference to 106 s-1 (t = 1 µs) for larger values [3].
Because of the large difference in metal-ligand bond lengths and the concomitant large difference in volume between high-spin and low-spin states, elastic interactions influence both the thermal spin transition as well as the relaxation behaviour [4]. In concentrated spin-crossover systems cooperative effects may lead to very steep thermal transition curves, hysteresis behaviour and even to the phenomenon of a light-induced bistabilty [5]. In some cases the elastic forces become so large that the crystal lattice responds with a crystallographic phase transition [6]. The elastic interactions may be thought of as a changing internal pressure, which increases with increasing low-spin fraction. Thus high-spin low-spin relaxation curves in concentrated systems deviate from single exponential behaviour, showing a typical self-accelerated behaviour [7].
An external pressure, too, influences both the thermal spin transition as well as the HS -> LS relaxation. Whereas the transition curve is being moved to higher temperature by typically 25 K/kbar, the HS -> LS relaxation is accelerated by up to one order of magnitude per kbar [8]. In one spin-crossover compound, the lifetime of the light-induced metastable HS state at 10 K was in fact reduced by a recordbreaking nine orders of magnitude from 10 s at 1 bar to 10 ns at 28 kbar. In Fe(III) compounds the bond length difference is smaller than in Fe(II) systems, resulting in a much larger low-T tunnelling rate constant for the former [9].
Oxalate network structures
The class of chiral 3-dimensional metal-tris-oxalate network structures of compositions [M'II2(ox)3][MII(bpy)3], [M'IM''III(ox)3][MII(bpy)3] and [M'IM''III(ox)3][MIII(bpy)3]ClO4 (ox = C2O42-, bpy = 2,2'-bipyrindine) recently synthesised by Decurtins et al. [10] have interesting structural, magnetic and photophysical properties. Despite the complexity of the nuclear structure, they crystallise in cubic space groups. There is basically no limit to the combination of metal ions, as long as the charges of those on the oxalate network add up to 4+ and the one on the tris-bpy cation is 2+ or 3+ with the incorporation of an additional ClO4- for the latter. Transition metal ions on the oxalate network can interact with each other via the bridging ligands, giving rise to magnetic phenomena such as antiferro- and ferri- and ferromagnetism.
Depending upon the transition metal ions incorporated into the oxalate backbone as well as into the tris-bpy cation, the compounds exhibit a variety of photophysical properties. For instance, the compound with the photophysically inactive Na+ and Al3+ metal ions on the oxalate backbone and [Ru(bpy)3]2+ in the cavities shows the famous red luminescence attributed to this complex cation upon irradiation at 457 nm, i.e. at the maximum of the MLCT absorption band of [Ru(bpy)3]2+. If Al3+ is replaced by Co3+ or Cr3+ the [Ru(bpy)3]2+ luminescence is either partially (Co3+) or completely (Cr3+) quenched. In the latter case sharp luminescence bands characteristic for the 2E -> 4A2 transition of octahedrally coordinated Cr3+ are observed instead. This is a clear indication for very efficient energy transfer from the initially excited [Ru(bpy)3]2+ to [Cr(ox)3]3- [11a]. In the Co3+ case, the [Ru(bpy)3]2+ luminescence is not completely quenched and no additional luminescence is being observed. Since [Ru(bpy)3]2+ in the excited state is sufficiently reducing to reduce [Co(ox)3]3-, the quenching is thought to be due to a photo-electron transfer process [11b].
A particularly interesting phenomenon was recently discovered and is currently under investigation in [NaCr(ox)3][Rh(bpy)3]ClO4. Energy transfer within the 2E state of Cr3+ complexes is a common phenomenon. At low temperatures the inhomogeneous line width usually far exceeds the homogeneous line width. So far phonon-assisted processes have thus dominated this energy transfer. In the above compound, surprisingly, the process which is resonant within the homogeneous linewidth is the dominant process at 1.5 K [12]. This is due to the structure of the compound. As it is a neat material the concentration of [Cr(ox)3]3- chromophores is comparatively high, but as they are separated by Na+ ions, there are no exchange interactions between them, which would destroy the resonant process because of exchange splittings. At temperatures above 4.2 K the common temperature dependent phonon-assisted process takes over.
[Cr(ox)3]3- can also serve as photosensitiser for [Cr(ox)3]3- to [Cr(bpy)3]3+ energy transfer. In the mixed crystal system [Rh1-yCry(bpy)3][NaAl1-xCrx(ox)3]ClO4 it proved possible to clearly distinguish between an exchange mechanism for energy transfer between nearest neighbours and a dipole-dipole mechanism for energy transfer over longer distances [13].
Spin-crossover in oxalate networks
The size of the cavity provided for the [M(bpy)3]n+ cation by the oxalate network can be varied by changing the metal ions on the backbone. Thus the chemical pressure acting upon the [M(bpy)3]n+ cations can be varied. [Co(bpy)3]2+ is normally a classic high-spin complex with a 4T1 ground state. In [NaCr(ox)3][Co(bpy)3] it continues to be so. However, substituting Na by Li increases the internal pressure to such an extent that the high-spin state becomes sufficiently destabilised for [LiCr(ox)3][Co(bpy)3] to show spin-crossover behaviour [14].
Collaborations
Recent Publications:
Publications by Dr. Hagemann
and the correspondings author list
and keyword list
Open positions
Last modified: 17/02/2009 (DL)
![[Figure 1]](figure_1.gif)