The fabrication of microfluidic channels with complex three-dimensional (3D) geometries presents a major challenge to the field of microfluidics, because conventional lithography methods are mainly limited to rectangular cross-sections. In this paper, we demonstrate the use of mechanical micromachining to fabricate microfluidic channels with complex cross-sectional geometries. Micro-scale milling tools are first used to fabricate semi-circular patterns on planar metallic surfaces to create a master mold. The micromilled pattern is then transferred to polydimethylsiloxane (PDMS) through a two-step reverse molding process. Using these semi-circular PDMS channels, circular cross-sectioned microchannels are created by aligning and adhering two channels face-to-face. Straight and serpentine-shaped microchannels were fabricated, and the channel geometry and precision of the metallic master and PDMS molds were assessed through scanning electron microscopy and non-contact profilometry. Channel functionality was tested by perfusion of liquid through the channels. This work demonstrates that micromachining enabled soft lithography is capable of fabricating non-rectangular cross-section channels for microfluidic applications. We believe that this approach will be important for many fields from biomimetics and vascular engineering to microfabrication and microreactor technologies.

A sharp-interface numerical approach is developed for modeling the electrochemical environment in crevices and pits due to galvanic corrosion in aqueous media. The concentration of chemical species and the electrical potential in the crevice or pit solution environment is established using the steady state Nernst–Planck equations along with the assumption of local electroneutrality (LEN). The metal-electrolyte interface fluxes are defined in terms of the cathodic and anodic current densities using Butler–Volmer kinetics. The extended finite element method (XFEM) is employed to discretize the nondimensionalized governing equations of the model and a level set function is used to describe the interface morphology independent of the underlying finite element mesh. Benchmark numerical studies simulating intergranular crevice corrosion in idealized aluminum–magnesium (Al–Mg) alloy microstructures in two dimensions are presented. Simulation results indicate that corrosive dissolution of magnesium is accompanied by an increase in the pH and chloride concentration of the crevice solution environment, which is qualitatively consistent with experimental observations. Even for low current densities the model predicted pH is high enough to cause passivation, which may not be physically accurate; however, this model limitation could be overcome by including the hydrolysis reactions that potentially decrease the pH of the crevice solution environment. Finally, a mesh convergence study is performed to establish the accuracy of the XFEM and a sensitivity study examining the relationship between crevice geometry and species concentrations is presented to demonstrate the robustness of the XFEM formulation in handling complex corrosion interface morphologies.

This study demonstrates a novel model generation methodology that addresses several limitations of conventional finite element head models. By operating chiefly in imagespace, new structures can be incorporated or merged, and the mesh either decimated or refined both locally and globally. This methodology is employed in the development of a highly bio-fidelic finite element head model from high resolution scan data. The model is adaptable and presented here in a form optimised for impact and blast (Schneiderman et al. 2008). Considerable research has been devoted to investigating the mechanisms which generate TBI in cases where the head is subjected to impact or blast. Of the various experimental methods available (in vivo human experiments, or tests on cadaveric, animal, physical, or mathematical models) the increasing availability of computing power has seen numerical simulation, and in particular the finite element (FE) method, come to the forefront of this research. The current study details the development of a 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 Finite element modellingFE simulation is an invaluable tool in the exploration of complex trauma mechanisms resulting from dynamic insults such as impacts and blasts. While sophisticated FE models of the head offer the prospect of providing accurate and repeatable experimental analyses of all categories of mechanical head injury, the validity of these models is heavily influenced by the geometric accuracy of the structures within the model. Traditionally computational models are created manually using computer aided design (CAD) tools, but when concerned with the complex anatomy of the head and its internal structures this manual approach has significant disadvantages; there is a large scope for subjectivity and user error, as well as of becoming rapidly computationally intractable as more geometric fidelity is introduced. A novel model generation approach which can reduce the inaccuracies associated with modelling a highly complex structure is known as 'image based meshing'. This refers to the conversion of volume scan data, generally from computed tomography (CT) or magnetic resonance imaging (MRI), directly into an FE mesh by way of fully-and semi-automated processes with minimal user input. This increases not only the accuracy, but also the speed at which computational models of complex geometries can be constructed, thereby allowing a greater number of anatomical features to be distinguished.In 1997, Mehta et al. selected 14 cross-sectional image 'slices' from an MRI scan of the head and highlighted the skull in these images using an image processing tool. These highlighted outlines of the bone could then be read by a C++ computer code and converted into CAD coordinate and spline data, from which a model could be constructed that was, at least partially, based on the ...

Anisotropy of workpiece crystals has a significant effect in micromachining since the uncut chip thickness values used in micromachining are commensurate with characteristic dimensions of crystals in crystalline materials. This paper presents an experimental investigation on orthogonal micromachining of single-crystal aluminum at different crystallographic orientations for varying uncut chip thicknesses and cutting speeds using a diamond tool. Micromachining forces, specific energies, effective coefficient of friction, shear angles, shear stresses, and chip morphology were examined for six crystallographic orientations at uncut chip thicknesses ranging from 5 �m to20�m and cutting speeds ranging from 5 mm/s to15mm/s. Three distinct types of forces were observed, including steady (Type-I), bistable (Type-II), and fluctuating (Type-III) force signatures. The forces were seen to vary by as much as threefold with crystallographic orientation. Although the effect of cutting speed was small, the uncut chip thickness was seen to have a significant orientation-dependent effect on average forces. Chip morphology, analyzed under scanning electron microscopy, showed shear-front lamella, the periodicity of which was seen to vary with crystallographic orientations and uncut chip thicknesses. �DOI: 10.1115/1.2917268�

No abstract