The fullerene cage provides an ideal, isolated environment for trapping spin active atoms such as nitrogen or phosphorous. Alignment of these endohedral fullerenes in linear arrays would have applications in quantum computing as the interactions between spin-active molecules can be easily controlled. Self-assembled molecular networks such as block copolymers, Langmuir-Blodgett films, and self-assembled monolayers are ideal for this purpose as the spacing and geometry can be easily tuned. This dissertation will discuss using each of these methods to achieve alignment or orientation of fullerenes for application in quantum information processing.

Thesis (M.S.)--Oregon State University, 2008. / Printout. Includes bibliographical references (leaves 96-103). Also available on the World Wide Web.

Endohedral fullerenes are closed carbon shells encapsulating molecular or ionic species in their inner space. Obtained for the first time in 1985, endohedral metallofullerenes (EMFs) remain in focus of research for many years with a broad variety of metal atoms, endohedral cluster and cage sizes being reported. Electrochemical studies of endohedral metallofullerenes are of particular interest because of the more complex redox behavior in comparison to empty fullerenes. The EMF molecules can be considered as a combinations of positively charged cluster and negatively charged carbon shell “ligand”, and both constituents can be redox active. A cage-based electrochemical activity is more common, in particular, the most abundant nitride clusterfullerenes generally have redox-active cages. Cluster-based electrochemical activity is less common and can be revealed via unexpected redox behavior (e.g., shifted potential when compared to analogous molecules, potential metal dependence) and with the use of spectroscopic methods. Here we report electrochemical and EPR studies of three EMF families: (i) M2@C82-C3v and M2@C82-Cs dimetallofullerenes with a covalent bonding between two metal atoms, (ii) M2@C80(CH2Ph) dimetallofullerene derivatives with single-occupied metal-bonding orbital, and (iii) M2TiC@C80 EMFs with endohedral Ti(IV) (M is either Sc or Y or a lanthanide). For the first two families, the metal-metal bonding orbital has been found to be redox active: in M2@C82, the double-occupied M-M bonding orbital is involved in the first oxidation process, while in M2@C80(CH2Ph) the unoccupied component of single-occupied metal-bonding orbital acts as the LUMO, accepting one electron during the first reduction step. Thus, single electron transfer reactions in both cases lead to the changes in the magnetic properties of EMFs, which is especially well revealed by EPR spectroscopy. For the series of M2TiC@C80 EMFs, the first reduction predominantly occurs on internal Ti atom and can be described as TiIV/TiIII redox process. Due to the variation of the size of the Ti ion in different oxidation states, reduction changes the inner strain of the cluster, leading to a large variability of the TiIV/TiIII reduction potential in dependence on the size of the formally inert lanthanide metal in M2TiC@C80.

No description available.

No description available.

An experimental apparatus has been constructed to investigate ionisation mechanisms of complex molecules and nanoparticles after femtosecond and/or picosecond laser excitation. The photoproducts are detected by time-of-flight mass spectrometry and velocity-map imaging (VMI) photoelectron spectroscopy. Test measurements on C60 and Xe have successfully reproduced previously published work indicating that the setup is working in a satisfactory manner. New detailed investigations of mass spectra and angle resolved photoelectron spectra (PES) have been carried out as a function of laser intensity, wavelength and pulse duration for C60 and C70, providing new insights into the electronic structure and ionisation mechanisms of these molecules. For 400 nm, 130 fs laser excitation, an isotropic contribution from thermally emitted electrons is found. A series of peaks are seen superimposed on the thermal background with binding energies in agreement with the recently discovered superatom molecular orbitals (SAMOs) of C60 [Feng et.al. Science 320 (2008) p. 359]. Furthermore, the angular dependence of the peak in the PES corresponding to the s-SAMO is in agreement with this assignment. To confirm the assignment of the other observed peaks it is concluded that the measured photoelectron angular distributions (PADs) need to be compared to calculated angular distributions. Measurements have also been made with the same wavelength but with a pulse duration of about 5 ps. Mass spectra, PES and PADs for these measurements show that the main ionisation mechanism for these laser conditions is delayed (thermionic) ionisation. For 800 nm, 130 and 180 fs laser excitation, thermally emitted electrons are observed. In contrast to the 400 nm measurements, the PADs show an asymmetry with higher apparent temperatures along the laser polarisation direction. Measurements were also made for longer pulse durations (1.0 – 3.8 ps). For pulse durations above 1 ps the asymmetry is gradually reduced while the delayed ionisation component in the mass spectrum increases with increasing pulse duration. The asymmetry is compared to calculations made assuming a field-assisted thermal electron emission. Similarly to the 400 nm experiments, a series of peaks are seen superimposed on the thermal background. PADs are presented for these peaks. PADs for peaks with the same binding energy as peaks seen in the 400 nm experiments follow the same trend. Isotropic PADs after ns laser excitation are also presented confirming delayed ionisation for these pulse durations.

Water monomer and dimer encapsulations into D-2(22)-C-84 fullerene are evaluated. The encapsulation energy is computed at the M06-2X/6-31++G** level, and it is found that the monomer and dimer storage in C-84 yields an energy gain of 10.7 and 17.4kcalmol(-1), respectively. Encapsulation equilibrium constants are computed by using partition functions based on the M06-2X/6-31G** and M06-2X/6-31++G** molecular data. Under high-temperature/high-pressure conditions, similar to that for the encapsulation of rare gases in fullerenes, the computed (H2O)(2)@C-84-to-H2O@C-84 ratio is close to 1:2.

Information processors that harness quantum mechanics may be able to outperform their classical counterparts at certain tasks. Quantum information processing (QIP) can utilize the quantum mechanical phenomenon of entanglement to implement quantum algorithms. Endohedral fullerenes, where atoms, ions or clusters are trapped in a carbon cage, are a class of nanomaterials that show great promise as the basis for a solid-state QIP architecture. Some endohedral fullerenes are spin–active, and offer the potential to encode information in their spin-states. This thesis addresses the challenges of how to engineer the components of a scalable QIP architecture based on endohedral fullerenes. It focuses on the synthesis and characterization of molecules which may, in the future, permit the demonstration of entanglement; the optical read-out of quantum states; and the creation of quasi-one-dimensional molecular arrays. Due to its long spin decoherence time, N@C₆₀ is the selected as the basic molecular unit for ‘coupled’ fullerene pairs, molecular systems for which it may be possible to demonstrate entanglement. To this end, isolated fullerene pairs, in the form of spin-bearing fullerene dimers, are created. This begins with the processing of N@C₆₀ at the macroscale and leads towards the synthesis of ¹⁵N@C₆₀-¹⁵N@C₆₀ dimers at the microscale. High throughput processing is introduced as the most efficient technique to obtain high purity N@C₆₀ on a reasonable timescale. A scheme to produce symmetric and asymmetric fullerene dimers is also demonstrated. EPR spectroscopy of the dimers in the solid-state confirms derivatization, whilst permitting the modelling of spin–spin interactions for 'coupled' fullerene pairs. This suggests that the optimum inter–spin separation for which to observe spin–spin coupling in powders is circa 3 nm. Motivated by the properties of the trivalent erbium ion for the optical detection of quantum states, optically–active erbium–doped fullerenes are also investigated. These erbium metallofullerenes are synthesized and isolated as individual isomers. They are characterized by low temperature photoluminescence spectroscopy, emitting in the infra- red at a wavelength of 1.5 μm. The luminescence is markedly different where a C₂ cluster is trapped alongside the erbium ions in the fullerene cage. Er₂C₂@C₈₂ (isomer I) exhibits emission linewidths that are comparable to those observed for Er³⁺ in crystals. Finally, the discovery of a novel praseodymium-doped fullerene is reported. The balance of evidence favours the structure being assigned as Pr₂@C₇₂. This novel endohedral fullerene forms quasi-one-dimensional arrays in carbon nanotubes, which is a useful proof-of-principle of how a scaled fullerene-based architecture may be achieved.

Neste trabalho, propomos o uso do átomo do silício da molécula C59Si como uma maneira possível de ancorar moléculas de fulerenos em uma superfície de silício, devido à formação de uma ligação forte com um dos átomos de silício da superfície. Consideramos uma superfície Si(100) monohidrogenada com um H removido de um dos átomos de silício da superfície e neste local uma molécula de C59Si é adsorvida. Através de cálculos ab initio baseados na Teoria do Funcional da Densidade, obtivemos uma distância de ligação de 2,37 Å e uma energia de ligação de aproximadamente 2,10 eV. Diferentes possíveis rotas para obter a adsorção dessa molécula são discutidas. Em particular, propomos que uma possível rota seria através da molécula, ainda não detectada, (C59Si)2. A partir de cálculos ab initio mostramos que as duas moléculas de C59Si ((C59Si)2) estão ligadas através de seus átomos de silício. Obtivemos uma distância final de ligação de 2,36 Å e uma energia de ligação com respeito à dissociação em dois monômeros C59Si de 1,6 eV. Uma análise da estrutura eletrônica e do espectro vibracional da molécula (C59Si)2 é apresentada e comparada com o monômero C59Si. Investigamos também a ligação do fulereno C60N com a superfície de silício através do átomo de nitrogênio, assim como a adsorção do fulereno endoédrico N@C59Si na superfície Si(100):H. Finalmente, investigamos a interação da molécula de C59Si com uma superfície de Au(111) mediada por um átomo do enxôfre. / In this work we propose the use of the Si atom in the C59Si molecule as a possible way to controllably anchor fullerene molecules on a Si surface, due to the formation of a strong bond to one of the Si surface atoms. We consider a monohydride Si(100) surface with a H removed from one of the Si surface atoms and the C59Si adsorbed at this site. Through ab initio calculations based on Density Functional Theory, we obtained a final Si_super-Si_C59Si bond distance of 2.37 Å, and a binding energy of approximately 2.1 eV. Different possible routes to obtain the adsorption of such a molecule will be discussed. In particular, we propose that a as yet undetected (C59Si)2 molecule could be used for such a reaction. Through ab initio calculations we show that the two C59Si molecules will be bonded via their Si atoms, forming a dumbbell-like structure. We obtain a final Si-Si bond distance of 2.36 Å, and a binding energy with respect to dissociation into two C59Si monomers of 1.6 eV. An analysis of the (C59Si)2 electronic structure and vibrational spectra will be presented and compared with the C59Si monomer. We also investigate the binding of the C60} fullerene on the Si surface via nitrogen atoms, as well as the endohedral fullerene N@C59Si at the Si(100) surface. Finally, we also investigate the interaction of the C59Si molecule with Au(111) surface via sulfur atoms.

Fullerenes have highly excited electronic states with interesting properties for possible wide ranging applications including in electronics. These highly excited, Rydberg-like states, so-called superatom molecular orbitals (SAMOs), are diffuse low-angular momenta states with molecular orbitals centred on the hollow fullerene core. The SAMOs can be detected by femtosecond photoelectron spectroscopy (PES) and characterised by photoelectron angular distributions (PADs) combined with time-dependent density functional theory (TD-DFT) calculations. The photoelectron spectra of C60 and C70 show a peak structure below kinetic energies corresponding to the photon energy, superimposed on a thermal electron background. This peak structure was assigned to one-photon ionisation of the SAMO states based on PAD and TD-DFT. In this thesis, studies of the fullerene species C82 and Sc3N@C80 revealed PES and PAD with similar features to C60 and C70. The SAMO peaks became less prominent compared to the thermal electron background for increasing molecular size and decreasing symmetry, and were almost absent for the endohedral species. To provide more information about the influence of encapsulated atoms in the fullerene cage on the SAMO states, experiments on Li@C60 have been carried out. A lower thermal electron emission temperature and a splitting of the SAMO peaks has been observed for Li@C60 compared to C60. Nevertheless the binding energies are remarkably similar in all investigated fullerenes, which is important for any applications. Since the binding energies are about the same, but the ionisation potentials of the fullerenes are different, the excitation energy to the SAMOs scales with the ionisation energy. The reasons for the well-pronounced peak structure of the SAMO states in the PES of C60 could be explained by the similarity of the SAMOs to Rydberg states along with the higher photoionisation probabilities compared to valence states which were modelled by Benoît Mignolet and Françoise Remacle. As the SAMOs are highly excited electronic states, like Rydberg states, the potential energy surface of the neutral molecule and the ionised molecule are similar. Therefore the vibrational energy is conserved in the molecule during the photoionisation on the femtosecond time scale. The TD-DFT calculations on C60, carried out by Benoît Mignolet and Françoise Remacle, revealed the photoionisation probabilities of the SAMOs to be at least three orders of magnitude higher than for non-SAMOs for the applied experimental conditions. To test the prediction of the model, the relative photoionisation probabilities of the s-SAMO to p-SAMO and the s-SAMO to d-SAMO were obtained experimentally from the PES at various photon energies (2-3.5 eV) within this work. The analysis indicates remarkable agreement between the experiment and the theoretical values. Further quantum chemical calculations on a series of polycyclic aromatic hydrocarbons (PAHs) were carried out within this thesis, which revealed similar Rydberg-like molecular orbitals in analogy to the SAMOs in fullerenes. The first series included benzene, naphthalene, anthracene, tetracene, pentacene and hexacene. The second series consisted of phenanthrene, pyrene and coronene. Finally, the third series covered cubane, adamantane and dodecahedral C20. All modelled molecules showed diffuse, excited electronic states similar to the SAMOs. Within each series the binding energies of these states decrease with increasing molecular size as well as the ionisation energies, except for the 3rd series. A comparison between all series shows that the binding energies of the states for the 3rd series (the 3-D series) are slightly higher than for the 1st and 2nd series in relation to similar molecular size. The results of the coronene calculations are compared to experimental photoelectron spectra and are shown to be in good agreement with the experiments.