This PhD thesis presents a study of structure-composition correlations of bioactive glasses (BGs) by employing solid-state Nuclear Magnetic Resonance (NMR) spectroscopy. Silicate-based Na2O−CaO−SiO2−P2O5 BGs are utilized clinically and are extensively investigated for bone regeneration purposes. Once implanted in the human body, they facilitate bone regeneration by partially dissolving in the body fluids, followed by the formation of a biomimetic surface-layer of calcium hydroxy-carbonate apatite (HCA). Eventually, the implanted BG totally integrates with the bone. The bioactivity of melt-prepared BGs depends on their composition and structure, primarily on the phosphorus content and the average silicate-network connectivity (NC). We explored these composition-structure relationships for a set of BGs for which the NC and phosphorus contents were varied independently. The short-range structural features of the glasses were explored using 29Si and 31P magic-angle-spinning (MAS) NMR spectroscopy. 31P MAS NMR revealed that the orthophosphate content is directly proportional to the total P content of the glass, with a linear correlation observed between the orthophosphate content and the silicate network connectivity. The bearings of the results for future BG design are discussed. By using multiple-quantum coherence-based 31P NMR experiments, the spatial distribution of orthophosphate groups was probed in the melt prepared BGs, as well as in two mesoporous bioactive glasses prepared by an evaporation-induced self-assembly technique. The results evidence randomly distributed orthophosphate groups in the melt-prepared BGs, whereas the pore-walls of the mesoporous bioactive glasses constitute nanometer-sized clusters of calcium phosphate. The distribution of Na+ ions among the phosphate/silicate groups were studied by heteronuclear dipolar-based 23Na−31P NMR experiments, verifying that sodium is dispersed nearly randomly in the glasses. The phosphorus and proton environments in biomimetically grown HCA were investigated by using 1H and 31P MAS NMR experiments. Our studies revealed that the biomimetic HCA shared many local structural features with synthetic and well-ordered hydroxy-apatite. / <p>At the time of the doctoral defense, the following paper was unpublished and had a status as follows: Paper 4: Accepted.</p>

Made available in DSpace on 2014-12-17T14:57:49Z (GMT). No. of bitstreams: 1 IvanMFL.pdf: 663615 bytes, checksum: a9a612a8d366a8eadb51a7b7c58984a5 (MD5) Previous issue date: 2009-02-20 / Due to advances in the manufacturing process of orthopedic prostheses, the need for better quality shape reading techniques (i.e. with less uncertainty) of the residual limb of amputees became a challenge. To overcome these problems means to be able in obtaining accurate geometry information of the limb and, consequently, better manufacturing processes of both transfemural and transtibial prosthetic sockets. The key point for this task is to customize these readings trying to be as faithful as possible to the real profile of each patient. Within this context, firstly two prototype versions (&#945; and &#946;) of a 3D mechanical scanner for reading residual limbs shape based on reverse engineering techniques were designed. Prototype &#946; is an improved version of prototype &#945;, despite remaining working in analogical mode. Both prototypes are capable of producing a CAD representation of the limb via appropriated graphical sheets and were conceived to work purely by mechanical means. The first results were encouraging as they were able to achieve a great decrease concerning the degree of uncertainty of measurements when compared to traditional methods that are very inaccurate and outdated. For instance, it's not unusual to see these archaic methods in action by making use of ordinary home kind measure-tapes for exploring the limb's shape. Although prototype &#946; improved the readings, it still required someone to input the plotted points (i.e. those marked in disk shape graphical sheets) to an academic CAD software called OrtoCAD. This task is performed by manual typing which is time consuming and carries very limited reliability. Furthermore, the number of coordinates obtained from the purely mechanical system is limited to sub-divisions of the graphical sheet (it records a point every 10 degrees with a resolution of one millimeter). These drawbacks were overcome by designing the second release of prototype &#946; in which it was developed an electronic variation of the reading table components now capable of performing an automatic reading (i.e. no human intervention in digital mode). An interface software (i.e. drive) was built to facilitate data transfer. Much better results were obtained meaning less degree of uncertainty (it records a point every 2 degrees with a resolution of 1/10 mm). Additionally, it was proposed an algorithm to convert the CAD geometry, used by OrtoCAD, to an appropriate format and enabling the use of rapid prototyping equipment aiming future automation of the manufacturing process of prosthetic sockets. / Devido aos avan?os no processo de fabrica??o de pr?teses ortop?dicas, a necessidade de uma melhor qualidade t?cnica na leitura da forma (i.e., com menos incertezas) da parte residual dos amputados tornou-se um desafio. Para superar esses problemas ? necess?rio ser capaz de obter informa??es precisas da geometria do membro e, consequentemente, melhorar os processos de fabrica??o de pr?teses e cartuchos transfemurais e transtibiais. O ponto-chave para esta tarefa ? personalizar essas leituras tentando ser o mais fiel poss?vel ao verdadeiro perfil de cada paciente. Dentro deste contexto, duas vers?es (&#945; e &#946;) de um prot?tipo de um scanner 3D mec?nico para ler forma residual de membros com base em t?cnicas de engenharia reversa foram concebidas. Prot?tipo &#946; ? uma vers?o melhorada do prot?tipo &#945;, apesar de ainda trabalhar em modo anal?gico. Ambos os prot?tipos s?o capazes de produzir uma representa??o CAD do membro atrav?s de apropriada planilha gr?fica e foram concebidos para funcionar exclusivamente atrav?s de meios mec?nicos. Os primeiros resultados foram encorajadores, uma vez que foram capazes de conseguir uma grande redu??o em rela??o ao grau de incerteza das medi??es quando comparadas com as dos m?todos tradicionais, que s?o muito imprecisas e desatualizadas. Por exemplo, n?o ? incomum ver estes m?todos arcaicos sendo usados utilizando fitas m?tricas caseiras para medir a forma da perna do paciente. Embora o prot?tipo &#946; tenha melhorado as leituras, ainda ? necess?rio algu?m para introduzir os pontos plotados (i.e., aqueles marcados na planilha em formato de disco gr?fico) para um software acad?mico CAD chamado OrtoCAD. Esta tarefa ? realizada por digita??o manual, que ? morosa e tem confiabilidade limitada. Al?m disso, o n?mero de coordenadas obtidas a partir do sistema puramente mec?nico ? limitado ?s subdivis?es do disco gr?fico (ele grava um ponto a cada 10 graus com uma resolu??o de um mil?metro). Estes inconvenientes foram resolvidos na concep??o da segunda vers?o do prot?tipo &#946;, na qual foi desenvolvida uma adapta??o eletr?nica, tornando-o capaz de realizar uma leitura autom?tica (i.e., sem interven??o humana no modo digital). Um software de interface (i.e., driver), foi desenvolvido para facilitar a transfer?ncia de dados. Resultados muito melhores foram obtidos significando menor grau de incerteza (ele registra um ponto a cada 2 graus com uma resolu??o de 1/10 mm). Al?m disso, foi proposto um algoritmo para converter a geometria CAD, utilizada pelo OrtoCAD, para um formato adequado e que permita a utiliza??o de equipamento de prototipagem r?pida destinado a uma futura automa??o do processo de fabrica??o dos cartuchos de pr?teses.