Alexander Woesz

Production and characterisation of bioceramic-nanoparticle reinforced biopolymers as bone replacement materials:

Biodegradable biopolymers are currently used in many applications in the human body, such as in craniomaxillofacial and dental surgery, sports medicine, orthopaedic trauma treatment and spine surgery. Although their mechanical properties are, compared to metals like stainless steel or titanium which are used in the same sites, quite low, their usage is recommended due to the fact that they are slowly degrading within the body, and finally vanish or are replaced by body's own bone material, thus avoiding the psychological problems associated with a foreign material in the patients body. Unfortunately, the increased cross-sectional area necessary to bear comparable loads sometimes causes problems, for instance if the implant applies pressure onto the surrounding organic material. Increasing the strength and stiffness of biopolymers by introduction of a hard and stiff inorganic filler is therefore a possibility to widen the applicability of bioresorbable implants. In addition to that, a ceramic filler which is itself slowly dissolved in the body can change the degradation properties of the biopolymer, allowing to adjust the implant's delay time within the body within a larger period of time.

However, the usage of micrometer-sized bioceramic particles has so far not yielded the desired results, as usually just the stiffness of the composite increases with increasing filler content, whereas the strength is either kept constant or even decreased by addition of the filler. Although the usage of nanometre-sized bioceramic particles is associated with some problems too, it could avoid this difficulty, as the natural material bone with its magnificent mechanical properties suggests.

In this project we utilised several methods to produce composites from biopolymers and bioceramic nanoparticles, in order to adjust the mechanical as well as the degradation properties of the material to the desired applications.


Production and characterisation of hydroxyapatite scaffolds as future bone replacement materials:

Architecture strongly influences the mechanical properties as well as the biocompatibility and cell ingrowth behaviour of highly porous scaffolds. In order to optimise the architecture, we used Rapid Prototyping to produce casting moulds for ceramic gelcasting. The three dimensional shape of the mould, which determines the architecture of the later scaffold, was designed using a computer aided design (CAD) program. The moulds were produced from resin layer by layer in the RP process and then filled with a water based thermosetting ceramic slurry in vacuum. During the following temperature treatment, the ceramic slurry was first solidified and then the mould was burnt and the ceramic particles were sintered. This method is capable of producing structures of almost arbitrary architecture and has a resolution sufficient to built individual "trabeculae" of 300 �m thickness. In a first attempt, we constructed structures made of interconnected layers of parallel struts, oriented into two orthogonal directions. 10 struts were used per layer and 20 layers were superimposed. The porosity of the structure was 50 % of the volume (figure 1a).

HA-scaffolds

Fig. 1: a) virtual structure designed with a CAD program, b)hydroxylapatite scaffold, c)a strut of a HA scaffold covered with osteoblasts embedded in a collagenous matrix.

The ceramic powder used for gelcasting was artificial hydroxylapatite (HA). Different material qualities were produced by varying the sintering temperature as well as the sintering atmosphere. The scaffolds (figure 1b) were characterised with scanning electron microscopy, scanning force microscopy and x-ray diffraction to determine surface roughness and phase composition. Biocompatibility was derived using cells from the preosteoblastic cell line MC3T3-E1, derived from mouse calvariae, which were seeded on the scaffolds immersed in the culture medium. During the culture period, cell proliferation and ingrowth was observed, after embedding in PMMA and staining with Giemsa the cell ingrowth behaviour was histologically investigated. For some of the scaffolds, cells were found to cover the totality of the internal surface and even to form several cell layers and a collagenous matrix (fig. 1c), Most interestingly, the cell proliferation on the scaffolds depends strongly on their thermal pre-treatment, which is determining surface roughness and phase composition of the ceramic material


Mechanical Properties of Cellular Solids:

The mechanical properties of cellular solids strongly depend on the apparent density and on the materials properties the cellular solid is made of. In addition, the geometry of the cellular structure is of great importance. We investigated the influence of the geometry of regular, unit-cell based cellular solids on the mechanical properties independently from the apparent density and the dense materials properties. Unit-cell based structures were designed using a computer aided design program and produced in polyamide with a rapid prototyping method called selective laser sintering (compare figure 2). In compression tests the mechanical< properties were derived and it was shown that the strength and stiffness of the structures as well as their defect tolerance can be varied independently from each other within a wide range at constant apparent density, just by changing the architecture.

RP-structures

Fig. 2: Unit cell based structures produced from polyamide using selective laser sintering

In a continuation of this first set of experiments we used a digital light processing based rapid prototyping method to produce structures from resin. We at first varied the orientation of the structures against the loading direction when tested in compression, and furthermore produced structures with increasing degree of irregularity, which resulted in lower stiffness and strength, but higher tolerance against defects and deviation from the optimal loading direction. In addition, the experimental results were compared with results from finite element simulations, which were performed using beam elements as well as continuum elements, the former with and without an adaptation of stiffness in the vicinity of the vertices.