Rozprawy doktorskie na temat „STRUCTURAL INTERFACE PROPERTIES”

The opportunity of using the organic molecules in spintronic devices appeared challenging since these materials, having nominally high spin relaxation times, are suitable for coherent spin manipulation. The spin behaviour in these molecular spintronic devices has been demonstrated to strongly depend on the nature of the chemical bonds between the organic molecules and the magnetic electrodes affecting also the magnetic response of both molecular and metallic sides. In particular, the adsorption of an organic molecule on a ferromagnetic layer has been proved to change the local magnetism of a magnetic substrate. In spite of their technological interest, the investigation of such effect in the case of the polycrystalline magnetic thin film is still lacking. My work contributes to filling this gap by studying the structural, morphological and magnetic properties of ultrathin polycrystalline cobalt films covered by the well-known buckminsterfullerene organic molecule (C60). The combined investigation by AFM, TEM, SQUID magnetometry and anisotropic magnetoresistance allowed to correlate the sample microstructure with the magnetic response and to identify the main mechanism responsible for spin transport in these FM layers. Analysed films are composed of polycrystalline cobalt grains decoupled by non-crystalline amorphous regions. The volume ratio between crystalline grains and amorphous regions increases by increasing the film thickness. As expected, the values of saturation magnetisation decrease as the crystallinity decreases and a typical blocking behaviour is present. The cobalt layers are also subjected to oxidation at the interface with the single crystal oxide substrate. The presence of amorphous phase in polycrystalline cobalt ultrathin film impacts the analysis of transport properties: the anisotropic magnetoresistance slightly depends on the crystalline phase while it is mainly inherent to the amorphous component.

Molten salts attracted the attention of the scientific community several times during the last century. This interest is motivated by the physico-chemical properties of these systems. In fact, usually molten salts show chemical and thermal stability, i.e. they do not easily decompose or react. Furthermore, these compounds remain liquid over an extended range of temperatures, in which they show also a remarkably low volatility. The fact that molten salts are composed solely by ions, and can have a quite wide electrochemical window, make them very interesting as electrolytes[1]. The main disadvantage in the usage of molten salts in any practical process, is their high melting point (for example as high as 800°C for NaCl), which severely limits the number of reactions that can be done in these media and reduces the possibility of industrial scaling, due to the high energy required to maintain those high temperatures. Since the '70s lower temperature molten salts has been synthesised, like chloroaluminate eutectic mixtures, having melting points around 100°C or even lower, but the real turning point that boosted the research field has been the development of the first water-stable low melting point molten salts, that is what are now usually named room temperature ionic liquids, or simply ionic liquids. Ionic liquids are usually composed by a big organic cation and a bulky inorganic, water stable, anion: the bulkiness and the complex asymmetric structure of the ions prevent an efficient packaging, leading to a lowering of the coulombic cohesive energy and so of the melting point. Ionic liquids maintain all the characteristics of the high temperature molten salts, but they are usually liquid at room temperature. This fact induced a renewed interest in the field, as is proved by the several thousand papers published on the topic in 2011. The community of chemists devoted a great effort to the study of ionic liquids, because of the potential use of those liquids as solvents. Ionic liquids are complex systems, that usually are organised in polar and apolar domains, and can dissolve both polar and apolar species. In addition, there are virtually unlimited choices of ions, and each choice changes the physico-chemical characteristics of the systems: this allows to tailor the properties of the ionic liquids (like miscibility, density, viscosity...), in order to match specific tasks. The characteristics of ionic liquids, last but not least their low vapour pressure, promote them as good solvents for the growing field of the Green Chemistry, in substitution of the volatile organic compounds. Ionic liquids are also promising as lubricants, in particular in micro- and nano- -electromechanical devices (NEMSs and NEMSs)[2, 3] as well as electrolytes in photoelectrochemical devices used for energy storage and energy production such as supercapacitors[4, 5] or Grätzel solar cells[6]. In all these cases, the most relevant processes determining the performance of the devices, take place at the liquid/solid interface between ILs and solid surfaces: this is a region only a few nanometers thick where the properties of ILs can be significantly different from those of the bulk. The investigation of the interfacial properties of ILs is therefore of primary importance for their technological exploitation. To date, the (bulk)liquid-vapour and solid- (bulk)liquid ILs interfaces have been studied, mostly by sum-frequency generation spectroscopy[7, 8] and by X-ray photoemission spectroscopy[9]. For imidazolium-based ILs, ordering of the ions at the solid/liquid or liquid/ vapour interface has been inferred from vibrational spectroscopy data. For example Mezger et al.[10] performed a study of the (bulk)liquid/solid interface between a negatively charged sapphire substrate and imidazolium and pyrrolidinium-based ILs. They found strong interfacial layering, with repeated spacing of 0.7-0.8 nm, decaying exponentially into the bulk liquid. Recently the ionic liquid/solid interface has been studied with the surface force apparatus (SFA)[11, 12, 13]. In those experiments thin layers of ionic liquids are compressed between two approaching sheets of mica, while the normal force is recorded. The force vs. distance curves show a characteristic oscillatory profile, extending for few nanometers from the interface and exponentially decaying into the bulk. Only very recently, in order to explore in details the ionic liquid/solid interfaces, more local approaches has been used, i.e. scanning probe microscopies and numerical simulations. Layering at the solid/(bulk)liquid have been found by Atkin[14, 15] performing force spectroscopy with atomic force microscope (AFM). In these kind of measurements, in which the AFM tip is ramped against the interface while measuring the force acting on it, as in SFA experiments, subsequent ruptures of solvation layers are seen, beginning few nanometers above the surface. The group of Atkin, in collaboration with other groups, performed also scanning tunneling microscopy (STM)/AFM study on ionic liquid/Au(111) interface, finding a dependence of the behaviour of the solvation layers upon potential changes[16]. In simulation studies is again evidenced the formation of strong interfacial layering on different surfaces[17, 18], where moreover, the preferential orientation of the ions at the interface can be analysed. The studies previously cited about the interfacial behaviour of the ionic liquids, are mainly focused to the analysis of the bulk ionic liquid/solid interface. To date, a little effort has been devoted to what happens when thin layers of an ionic liquid are put in contact with a solid substrate. In those kind of situations, where the surface/volume ratio is high, it can been argued that the interaction with the surface can greatly change the physico-chemical and structural properties of the ionic liquid. The study of systems with high surface/volume ratio is directly relevant in all those applications in which a thin _lm of ionic liquid is used, e.g. in tribological applications, as well as in photoelectrochemical devices, where the liquid is soaked into a nanoporous matrix: a change of the properties of the ionic liquid in this case can heavily modify the final performance of the devices. My PhD work has been devoted to improving further extend the understanding of the behaviour of thin films of ionic liquids in contact with solid surfaces. At the time when I begun my work, very few studies regarding such systems were published. A pioneering study on thin _lm of [Bmim][PF6] on mica has been published in 2006 by Liu et al.[19]. In this work, performed using an AFM, the ionic liquid has been found to structure in two different ways on substrate: as droplets and as at layers, qualitatively referred to as solid layers. Inspired by the approach of Liu, I decided to study more quantitatively the thin-layer/solid interface, in particular using atomic force microscopy, because this instrument can acquire morphologies with vertical sub-nanometer and lateral nanometric resolution, so giving access to very localised information. Furthermore other kind of maps regarding different physico-chemical properties can be acquired simultaneously to the topography and moreover, force spectroscopy studies can be performed ramping the AFM tip against the surface. In particular I focused my investigation on the interaction of [Bmim][Tf2N], a hydrophobic and almost water-stable liquid, with different substrates, i.e. mica, amorphous silica, single crystal silicon covered by its native oxide, and HOPG graphite. The main part of the work has been performed on those test substrates, because their properties are well known and because they are at, so very suitable for an accurate AFM investigation. With the experience gained on those systems and the results obtained, on the last part of my PhD, the attention has been moved to surfaces more relevant for applications, as the nanostructured silicon oxide, directly synthesised in our lab with supersonic cluster beam deposition (SCBD). In order to perform my investigation, I first obtained very thin ionic liquid layers on the surface, diluting the liquid into a solvent, (methanol, ethanol, chloroform), and then drop-casting few μl of the solution onto a freshly cleaned substrate, letting the evaporation to proceed in air. On all the insulating surfaces studied, the [Bmim][Tf2N] coexists in 2 forms: as liquid micro- and nano- droplets and in at ordered domains. The melting temperature of this particular ionic liquid is ~-4°C, so a fundamental role in the liquid to solid transition has to be played by the interaction with the solid surface. The solid-like terraces appear as at layers, often growing one on top of the other. The hypothesis is that each layer is composed by several layers of cations and anions. By a statistical analysis of the morphological maps acquired, I extrapolated that the height of the best sub-multiple of the solid-like terraces is δ=0.6nm, in good agreement with the result of the simulations. The AFM has also been used to study the mechanical behaviour of the solid-like structures. If imaged in contact mode, the layers tend to be eroded after repeated scans, and in some cases the terraces are removed one by one, as in a lamellar solid. Investigating the resistance to normal loads, saw tooth profiles in force vs. separation curves have been found, highlighting a sequence of ruptures, separated by 1.8-2nm, where the average rupture pressure is ~2-3KBar, very similar to the maximum pressure found in a MD simulation of a tip penetrating in 4nm of [Bmim][Tf2N] on silica[20]. Moreover, the observation that supported IL islands were not disrupted by intense electric fields up to 108-109V/m (applied biasing the AFM with respect to sample during imaging) and that Scanning Nanoscale Impedance Microscopy[21] measurements highlighted a dielectric insulating character of the ordered domains (εr = 3-5 was measured[22]) is consistent with the idea that IL ordered domains behave as solid materials in which the ions are tightly bound. To understand what is the influence of the nanostructure on the formation of the solid-like layers, I realised [Bmim][Tf2N] depositions on nanostructured silica deposited on oxidised silicon by SCBD in our lab. Of particular interest is the case of IL coating on a sub-monolayer deposition of silica nanoparticles. The preliminary results show that the presence of a dense distribution of nanoparticles on the surface of oxidised silicon actually doesn't prevent the growth of multilayered solid-like domains, that are as thick as on at surfaces and densely distributed. The liquid part of the deposition is pinned to the silica clusters, fact that evidences the strong affinity of [Bmim][Tf2N] with silica. The results suggest that the formation of immobilised, possibly solid-like, layers of ionic liquids in contact with nanoporous matrices, is not unlikely and such structures and can strongly affect the properties of the devices in which those interfaces are present. This possibility is also supported by the fact that a small percentage of silica nanoparticles (5 wt%) is enough to induce the gelation in an IL-based electrolyte used for dye sensitised solar cells[23]. The findings of my PhD work highlight the potentialities of scanning probe techniques for the quantitative investigation of the interfacial properties of thin ionic liquid films. My AFM investigation highlights how heterogeneous can be the IL/solid interfaces and so how is of fundamental importance to deal with local and not with average properties. The results of my work show that the behaviour of thin layers of ionic liquid is greatly modified by the influence of the substrate. In particular, I found a coexistence of liquid domains and terraces with elevated structural order. The formation of those structures, not present in the bulk liquid, is clearly induced by the contact with the solid surface. The solid-like terraces are very resistant to normal loads and to intense electric fields and, differently from the bulk, they tend to behave as insulating layers: the development of those structures can then have a crucial influence in the performances of photoelectrochemical devices. The importance of this field of research and the validity of my work, are witnessed by the increasing number of papers studying thin ionic liquid layers appeared just before, but especially during the course of my PhD[24, 25, 26, 27, 28, 15, 16]; in many of these works, the AFM is the instrument of choice for the interfacial investigation because it allows to access to the physico-chemical properties of the system with nanometric or sub-nanometric resolution in all the three dimensions. In the next future the structural behaviour of the ionic liquids in contact with nanostructured surfaces will be further studied, making use of the experience of our group in synthesise the nanostructured oxides and metals. Moreover, we will explore in more details the dielectric properties of thin fillm of ionic liquids, in particular selecting those ILs directly used in supercapacitors and solar cells. Another interesting field that to date is still poorly explored is the interaction of ionic liquids with biological tissue. For this reason, we are going to begin to study the effects of ionic liquids on supported lipid bilayers. Bibliography [1] Wilkis, J. S., Green Chem 4, 73-80, (2002). [2] Qu, J., Truhan, J. J., Dai, S., Luo, H., Blau, P., J. Tribol. Lett. 22, 207-214, (2006). [3] Bhushan, B., Palacio, M., Kinzig, B., J. Coll. and Inter. Sci. 317, 275-287, (2008). [4] Simon, P. and Gogotsi, Y., Nature Materials 7, 845-854, (2008). [5] Appetecchi, G. B., Montanino, M., Carewska, M., Moreno, M., Alessandrini and F., Passerini, S., Electrochimica Acta 56, 1300-1307, (2011). [6] Grätzel, M., Journal of Photochemistry and Photobiology A: Chemistry 164, 3-14, (2004). [7] Santos, C. S., Baldelli, S., J. Phys. Chem. B 111, 4715-4723, (2007). [8] Rollins, J. B., Fitchett, B. D. and Conboy, J. C., J. Phys. Chem. B 111, 4990-4999, (2007). [9] Lovelock, K. R. J., Villar-Garcia, I. J., Maier, F., Steinrück, H. and Licence, P., Chem. Rev. 110, 5158-5190, (2010). [10] Mezger, M., Schröder, H., Reichert, H., Schramm, S., Okasinski, J. S., Schöder, S., Honkimäki, V., Deutsch, M., Ocko, B. M., Ralston, J. And Rohwerder, M., Science 322, 424428, (2008). [11] Ueno, K., Kasuya, M., Watanabe, M., Mizukami, M. and Kurihara, K., Phys. Chem. Chem. Phys. 12, 4066-4071, (2010). [12] Perkin, S., Albrecht, T. and Klein, J., Phys. Chem. Chem. Phys. 12, 1243-1247 (2010). [13] Min, Y., Akbulut, M., Sangoro, J. R., Kremer, F., Prudhomme, R. K. and Israelachvili, J., J. Phys. Chem. C 37, 16445{16449, (2009). [14] Atkin, R. and Warr, G. G., J. Phys. Chem. C, 111, 5162-5168, (2007). [15] Hayes, R., Warr, G. G. and Atkin, R., Phys. Chem. Chem. Phys. 12, 1709-1723, (2010). [16] Atkin, R., Borisenko, N., Drüschler, M., El Abedin, S. Z., Endres, F., Hayes, R., Huber, B. and Roling, B., Phys. Chem. Chem. Phys. 13, 6849-6857, (2011). [17] Sieffert, N., and Wip_, G., J. Phys. Chem. C 112, 19590-19603 (2008). [18] Sha, M., Zhang, F., Wu, G., Fang, H., Wang, C., Chen, S., Zhang, Y., and Hu, J., J. Chem. Phys. 128, 134504, (2008). [19] Liu, Y., Zhang, Y., Wu, G. and Hu, J., J. Am. Chem. Soc. 128, 7456-7457, (2006). [20] Ballone, P., Del Pópolo, M. G., Bovio, S., Podestà, A., Milani, P. and Manini, N., Phys. Chem. Chem. Phys. in press (2011) (arXiv:1101.5424v1). [21] Cassina, V., Gerosa, L., Podestà, A., Ferrari, G., Sampietro, M., Fiorentini, F., Mazza, T., Lenardi, C. and Milani, P., Phys. Rev. B 79, 115422, (2009). [22] M. Galluzzi, Study of morphological and dielectric properties of thin ionic liquid films by Atomic Force Microscopy, Master Thesis, University of Milan, (2010). [23] Wang, P., Zakeeruddin, S. M., Compte, P., Exnar, I. and Grätzel, M., J. Am. Chem. Soc., 125, 1166{1167, (2003). [24] Cremer, T., Killian, M., Gottfried, J. M., Paape, N., Wasserscheid, P., Maier, F. and Steinrück, H.-P., Chem. Phys. Chme. 9, 2185-2190, (2008). [25] Cremer, T., Stark, M., Deyko, A., Steinrück, H.-P. and Maier, F., Langmuir 27, 3662-3671, (2011). [26] Yokota, Y., Harada, T. and Fukui, K., Chem. Commun. 46, 8627-8629, (2010). [27] Zhang, F., Sha, M., Ren, X., Wu, G., Hu, J. and Zhang, Y., Chin. Phys. Lett. 27, 086101, (2010). [28] Kaisei, K., Kobayashi, K., Matsushige, K., Yamada, H., Ultramicroscopy 110, 733-736, (2010).

Amorphous silicon carbide (a-SiC) and silicon carbonitride thin films have been deposited onto a variety of substrates by Polymer-Source Chemical Vapor Deposition (PS-CVD). The interfacial interaction between the a-SiC films and several substrates including silicon, SiO[subscript 2], Si[subscript 3]N[subscript 4], Cr, Ti and refractory metal-coated silicon has been studied. The effect of thermal annealing on the structural and mechanical properties of the prepared films has been discussed in detail. The composition and bonding states are uniquely characterized with respect to the nitrogen atomic percentage introduced into the a-SiCN:H films. Capacitance-voltage (C-V) measurements were systematically used to evaluate the impurity level of the deposited a-SiC films. The chemical bonding of the films was systematically examined by means of Fourier transform infrared spectroscopy (FTIR). In addition, elastic recoil detection (ERD) and X-ray photoelectron spectroscopy (XPS) techniques were used to determine the elemental composition of the films and of their interface with substrates, while X-ray reflectivity measurements (XRR) were used to account for the film density. Spectral deconvolution was used to extract the individual components of the FTIR and XPS spectra. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) were also employed to characterize the surface morphology of the films. In addition, their mechanical properties [(hardness (H) and Young's modulus (E)] were investigated by using the nanoindentation technique. The impurity levels of the a-SiC films were found to be clearly correlated with the nature of the underlying substrates. The Pt-Rh and TiN-coated Si substrates were shown to lead to the lowest impurity level (~ 1×10[superscript 13] cm[superscript -3]) in the PS-CVD grown a-SiC films, while Cr and Ti-coated Si substrates induced much higher impurity concentrations. Such high impurity levels were shown to be a consequence of a strong metallic diffusion of the metallic species (Cr or Ti). In contrast, no diffusion was observed at the interface of a-SiC with either Pt-Rh or TiN. Our results pinpointed TiN-coated Si as the electrode material of choice that satisfied best all criteria required for the integration of a-SiC into opto-electronic devices. FTIR measurements revealed that not only the intensity of a-SiC absorption band linearly increased, but also its position is found to shift to a higher wave number as a result of annealing. In addition, the bond density of Si-C is found to increase from (101.6-224.5)×10[superscript 21] bond[dot]cm[superscript -3] accompanied by a decrease of Si-H bond density from (2.58-0.46)× 10[superscript 21] bond[dot]cm[superscript -3] as a result of increasing the annealing temperature (T[subscript a]) from 750 to 1200 [degrees]C. Annealing-induced film densification is confirmed by the XRR measurements, as the a-SiC film density is found to increase from 2.36 to ~ 2.75 g/cm[superscript -3] when T[subscript a] is raised from 750 to 1200 [degrees]C. In addition, as annealing temperature T[subscript a] is increased from 750 to 1200 [degrees]C, both hardness and Young's modulus are found to increase from 15.5 to 17.6 GPa and 155 to 178 GPa, respectively. On the microstructural level, the increase incorporation of N in the a-SiCN:H films is found not only to lead to C atoms substitution by N atoms in the local Si-C-N environment but also to the formation of a complex structure between Si, C and N. For instance, the FTIR spectra show a remarkable drop in the intensity of Si-C vibration accompanied by the formation of further bonds including Si-N, C-N, C=N, C[identical to]N and N-H with increasing NH[subscript 3]/Ar ratio. Moreover, the XPS spectra showed the existence of different chemical bonds in the a-SiCN:H films such as Si-C, Si-N, C-N, C=N and C=C. Both FTIR and XPS data demonstrate that the chemical bonding in the amorphous matrix is more complicated than a collection of single Si-C, Si-N, or Si-H bonds. Furthermore, the increase incorporation of N in the a-SiCN:H films resulted in an increase of the average R[subcsript rms] surface roughness from 4 to 12 nm. Moreover, the films became porous with pore size and density increase as a result of increasing N at.%. Ultimately, both H and E of the a-SiCN:H films were found to be sensitive to their N content, as they decreased (from ~17 GPa and 160 GPa to ~13 GPa and 136 GPa, respectively) when the N content was increased from 0 to 27 at.%. The formation of Si-N, Si-H, and N-H bonds at the detriment of the more stiffer Si-C bonds are thought to account for the observed lowering of the mechanical properties of the a-SiCN:H films such as their N content increased. Our results confirmed the previously-established constant-plus-linear correlation between the mechanical properties of the a-SiC films and their bond densities.