[Zurück]

E. Hernández:
"High Spatial Resolution Investigation of Spin Crossover Phenomena using Scanning Probe Microscopies";
Betreuer/in(nen), Begutachter/in(nen): W. Linert, D. Ruiz-Molina, A. Bousseksou, G. Molnar, P. Demont, P. Carron Castro, B. Domenichini; l´Université Toulouse 3 Paul Sabatier, France, 2015; Rigorosum: 21.07.2015.

The research presented in this thesis was motivated by the recent development of nanosized materials (thin films, nanoparticles, etc.) of transition metal complexes displaying molecular spin state switching phenomena. Bistable spin crossover nanoobjects are appealing for a variety of applications, but a lot of efforts are still needed to analyze and understand the different physical and chemical phenomena governing their properties. In this work we demonstrated that spin crossover phenomena can be studied and triggered by means of scanning probe microscopy with nanometric spatial resolution. We showed that scanning probe microscopy can be used to follow the thermal spin transition in nanoscale SCO materials through the analysis of different kinds of tip-sample interactions including the detection (or generation) of evanescent electromagnetic waves by an apertured fiber tip, the analysis of elastic deformation induced by a sharp silicon tip. Additionally, we proved that SCO phenomena in bulk can also be followed by observing changes of features in the sample topography as a function of the temperature. In addition, the possibility to trigger and even finely control the spin transition by local heat exchange between the sample and the probe was also put in evidence. Scanning probe microscopy techniques have been used in the past two decades to investigate phase change phenomena in different materials, such as ferroelectrics, polymers and magnetic materials. In most cases, however, SPM imaging across the phase change phenomena remained qualitative. Quantitative imaging of a material property which can be controlled by independent means has been rarely reported even in other fields. The main obstacle in any quantitative SPM approach is the analysis of the interactions between the SPM probe and the sample, all the more that both can change their properties during the SPM measurements. In addition, the phase change must be triggered by some external stimuli, which can also perturb the SPM analysis. In the case of SCO compounds the most convenient way to induce the spin transition is thermal excitation. We observed at several instances that the wear of the tip and the sample during successive scans as well as various unwanted thermal effects (sample drift, heat exchange between the sample and the probe, etc.) are very difficult to handle when using conventional SPM heating stages. 146 General Conclusions and Perspectives To overcome (or at least minimize) these problems we implemented a novel experimental strategy to study SCO films based on the localized heating of the sample by a Joule-heated metallic wire. This original approach allowed us to achieve very fast and tightly controlled temperature changes as well as to significantly reduce thermal drift and thermal exchange with the probe. Scans over the non-heated area allowed also to correct (or at least to detect) tip and/or sample degradation. These latter were also monitored in most cases in-situ by means of high sensitivity optical microscopy coupled to the SPM. We believe that this nanoheater strategy is not only interesting for the SCO field, but can find a more general application within the SPM community. Perhaps somewhat unexpectedly AFM thermo-mechanical measurements proved to be the most powerful approach for the quantitative imaging of the spin transition. We succeeded to indent SCO thin films in a non-destructive manner and from the temperature dependence of the indentation data we could extract the Young´s modulus of the films in the two spin states. This information is very relevant on its own, but even more importantly it can be used to monitor the spin state change with unparalleled spatial resolution. These measurements were carried out using a relatively recent AFM mode - the Peak Force Tapping, which belongs to the family of fast force mapping techniques. Preliminary tests using other mechanical modes were also made and it was shown that multifrequency AC imaging can provide similar results, but at an even faster imaging rate. Even if the SCO is associated with a significant volume change, surface topography AFM images were not useful to follow the spin transition in thin film samples. (N.B. This simple approach deserves, however, further efforts in the future.) On the other hand, very interesting surface topography changes were observed during the SCO in single crystal samples. The topography differences between the two spin states were analyzed and they correlate well with the change of the crystal shape. Even more importantly, intriguing transient topography changes were also observed during the spin transition in the form of surface undulations around the phase boundary. We suggested these undulations are formed in order to minimize the significant elastic stress and strain near the phase boundary. Nanoscale optical imaging of SCO films provided also useful contrast between the high spin and low spin phases both in fluorescence and reflectance modes. The former uses fluorescent probes incorporated in the film, while the latter probes directly the refractive index change of the SCO sample. Reflectivity measurements proved to be more robust, in particular in constant height NSOM mode. In the case of the investigated samples the quantification of the spin state changes observed in NSOM was difficult (both in fluorescence and in reflectivity). A possible solution to this problem would be to work in transmission mode, but this would request samples with rather specific optical properties. (I.e., a thin sample with a high optical density change upon the SCO in the UV spectral General Conclusions and Perspectives 147 region is needed.) Generally speaking, samples with appropriate properties are crucial not only for NSOM, but also for any other SPM technique and future work must focus more on this issue. Beside the detection of spin state changes in a small sample volume, we have shown that an SPM tip can be also used to trigger the spin transition locally. We tested high resolution photothermal laser writing using NSOM tips and we succeeded in switching the spin state of individual nanoparticles. In another (simple) approach, a cold Si probe was used to induce the high spin to low spin transition in a single crystal. We have shown that the tip can induce the nucleation of the LS phase and allows also the fine control of the position of the phase boundary. This finding opens up interesting possibilities to investigate the properties of the phase boundary. Clearly there are many other ways to use an SPM probe to manipulate the spin state of matter with nanometric resolution, such as using specific thermal probes or mechanical effects, which remain to be explored. These tools are appealing since they can allow to perform nanometric write/errase operations and can therefore give access to new devices, such as nanoscale memories. For example, we have shown in this work that the thermal memory observed in some SCO compounds provides an unprecedented scope for imaging transient thermal events with a high spatial resolution. Finally we must note that other SPM methods provide also exciting perspectives for the SCO field. Notably it would be interesting to explore the possibilities provided by magnetic force microscopy (MFM) and nanoscale vibrational spectroscopy methods, such as tip enhanced Raman and FTIR spectroscopies.

spin-crossover, scanning probe microscopy,